Microbial Diversity For Biotechnology

BioMed Research International
Microbial Diversity for Biotechnology
Guest Editors: George Tsiamis, Dimitrios Karpouzas, Ameur Cherif,

and Konstantinos Mavrommatis
Guest Editors: George Tsiamis, Dimitrios Karpouzas,

Ameur Cherif, and Konstantinos Mavrommatis
Copyright © 2014 Hindawi Publishing Corporation. All rights reserved.
This is a special issue published in “BioMed Research International.” All articles are open access articles distributed under the Creative
Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original
work is properly cited.
Contents
Microbial Diversity for Biotechnology, George Tsiamis, Dimitrios Karpouzas, Ameur Cherif,
and Konstantinos Mavrommatis
Volume 2014, Article ID 845972, 3 pages
Cultivation-Dependant Assessment, Diversity, and Ecology of Haloalkaliphilic Bacteria in Arid Saline

Systems of Southern Tunisia, Darine El Hidri, Amel Guesmi, Afef Najjari, Hanen Cherif, Besma Ettoumi,
Chadlia Hamdi, Abdellatif Boudabous, and Ameur Cherif
Microbial Diversity in the Era of Omic Technologies, Sofia Nikolaki and George Tsiamis
The World Bacterial Biogeography and Biodiversity through Databases: A Case Study of NCBI
Nucleotide Database and GBIF Database, Okba Selama, Phillip James, Farida Nateche,
Elizabeth M. H. Wellington, and Hocine Hacène
The Microbiology of Olive Mill Wastes, Spyridon Ntougias, Kostas Bourtzis, and George Tsiamis
Diversity and Antimicrobial Properties of Lactic Acid Bacteria Isolated from Rhizosphere of Olive Trees
and Desert Truffles of Tunisia, Imene Fhoula, Afef Najjari, Yousra Turki, Sana Jaballah,
Abdelatif Boudabous, and Hadda Ouzari
An Insight into the “-Omics” Based Engineering of Streptomycetes for Secondary Metabolite
Overproduction, Amit Kumar Chaudhary, Dipesh Dhakal, and Jae Kyung Sohng
Genetic and Biochemical Diversity of Paenibacillus larvae Isolated from Tunisian Infected Honey Bee
Broods, Chadlia Hamdi, Jihène Essanaa, Luigi Sansonno, Elena Crotti, Khaoula Abdi, Naima Barbouche,
Annalisa Balloi, Elena Gonella, Alberto Alma, Daniele Daffonchio, Abdellatif Boudabous, and Ameur Cherif
Characterization of the Bacterial Community Associated with Larvae and Adults of Anoplophora
chinensis Collected in Italy by Culture and Culture-Independent Methods, Aurora Rizzi, Elena Crotti,
Luigimaria Borruso, Costanza Jucker, Daniela Lupi, Mario Colombo, and Daniele Daffonchio
Population Abundance of Potentially Pathogenic Organisms in Intestinal Microbiome of Jungle Crow

(Corvus macrorhynchos) Shown with 16S rRNA Gene-Based Microbial Community Analysis,
Isamu Maeda, Mohammad Shohel Rana Siddiki, Tsutomu Nozawa-Takeda, Naoki Tsukahara, Yuri Tani,
Taki Naito, and Shoei Sugita
Microbial Adhesion and Biofilm Formation on Microfiltration Membranes: A Detailed

Characterization Using Model Organisms with Increasing Complexity, L. Vanysacker, C. Denis,
P. Declerck, A. Piasecka, and I. F. J. Vankelecom
Microbial Inoculants and Their Impact on Soil Microbial Communities: A Review, Darine Trabelsi and
Ridha Mhamdi
Plant Growth Promotion Potential Is Equally Represented in Diverse Grapevine Root-Associated

Bacterial Communities from Different Biopedoclimatic Environments, Ramona Marasco, Eleonora Rolli,
Marco Fusi, Ameur Cherif, Ayman Abou-Hadid, Usama El-Bahairy, Sara Borin, Claudia Sorlini,
and Daniele Daffonchio
Seasonal and Spatial Variability of Virioplanktonic Abundance in Haihe River, China, Lili Ma, Rui Sun,
Guannan Mao, Hui Yu, and Yingying Wang
The Occurrence of Two Species of Entomophthorales (Entomophthoromycota), Pathogens of Sitobion

avenae and Myzus persicae (Hemiptera: Aphididae), in Tunisia, Ibtissem Ben Fekih,
Sonia Boukhris-Bouhachem, Jørgen Eilenberg, Mohamed Bechir Allagui, and Annette Bruun Jensen
Effects of Systemic Pesticides Imidacloprid and Metalaxyl on the Phyllosphere of Pepper Plants,
Constantinos Moulas, Christos Petsoulas, Konstantina Rousidou, Chiara Perruchon, Panagiotis Karas,
and Dimitrios G. Karpouzas
Combinatorial Mutagenesis and Selection to Understand and Improve Yeast Promoters, Laila Berg,
Trine Aakvik Strand, Svein Valla, and Trygve Brautaset
Cyanobacterial Toxin Degrading Bacteria: Who Are They?, Konstantinos Ar. Kormas and
Despoina S. Lymperopoulou
Potential for Plant Growth Promotion of Rhizobacteria Associated with Salicornia Growing in Tunisian
Hypersaline Soils, Francesca Mapelli, Ramona Marasco, Eleonora Rolli, Marta Barbato, Hanene Cherif,
Amel Guesmi, Imen Ouzari, Daniele Daffonchio, and Sara Borin
Streptomyces rochei ACTA1551, an Indigenous Greek Isolate Studied as a Potential Biocontrol Agent
against Fusarium oxysporum f.sp. lycopersici, Grammatiki S. Kanini, Efstathios A. Katsifas,
Alexandros L. Savvides, and Amalia D. Karagouni
A Phylogenetic Analysis of Greek Isolates of Aspergillus Species Based on Morphology and Nuclear and
Mitochondrial Gene Sequences, Antonios Krimitzas, Ioanna Pyrri, Vassili N. Kouvelis,
Evangelia Kapsanaki-Gotsi, and Milton A. Typas
Unraveling the Lipolytic Activity of Thermophilic Bacteria Isolated from a Volcanic Environment,
Panagiota M. Stathopoulou, Alexander L. Savvides, Amalia D. Karagouni, and Dimitris G. Hatzinikolaou
Characterization and Dynamic Behavior of Wild Yeast during Spontaneous Wine Fermentation in Steel
Tanks and Amphorae, Cecilia Dı́az, Ana Marı́a Molina, Jörg Nöhring, and Rainer Fischer
Engineering Microbial Cells for the Biosynthesis of Natural Compounds of Pharmaceutical

Significance, Philippe Jeandet, Yann Vasserot, Thomas Chastang, and Eric Courot
An Internalin A Probe-Based Genosensor for Listeria monocytogenes Detection and Differentiation,

Laura Bifulco, Angela Ingianni, and Raffaello Pompei
Brazilian Cerrado Soil Actinobacteria Ecology, Monique Suela Silva, Alenir Naves Sales,
Karina Teixeira Magalhães-Guedes, Disney Ribeiro Dias, and Rosane Freitas Schwan
Hindawi Publishing Corporation
http://dx.doi.org/10.1155/2014/845972
Editorial
George Tsiamis,1 Dimitrios Karpouzas,2 Ameur Cherif,3 and Konstantinos Mavrommatis4

1
Department of Environmental and Natural Resources Management, University of Patras, 2 Seferi Street, 30100 Agrinio, Greece
2
Department of Biochemistry and Biotechnology, University of Thessaly, Ploutonos 26 and Aiolou Street, 41221 Larisa, Greece
3
Higher Institute for Biotechnology, University of Manouba, Biotechpole of Sidi Thabet, Sidi Thabet, 2020 Ariana, Tunisia
4
DOE Joint Genome Institute, 2800 Mitchell Drive, Walnut Creek, CA 94598, USA
Correspondence should be addressed to George Tsiamis; gtsiamis@upatras.gr
Received 24 December 2013; Accepted 24 December 2013; Published 23 January 2014
Copyright © 2014 George Tsiamis et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
The aim of this special issue is to emphasize on the ecology S. Nikolaki and G. Tsiamis presented a comprehensive
of microorganisms, the most diverse and abundant group overview of the advanced tools that became available in the
of organisms on Earth, and to infer on biotechnological last few years to the microbial ecologist and describe recent
applications. Even though we live in a microbially dominated examples and future applications of such tools including
planet only the past decade the study of the microbial phylogenetic and functional microarrays, next generation
diversity has entered a period of considerable importance sequencing, and single cell genomics. These tools will enable
to science in general, industry, protection of the environ- exploitation of the previously unknown microbial diversity in
ment, and public policy making. Environmental microbes are biotechnology.
immensely diverse and have numerous metabolic activities O. Selama et al. attempt to correlate the presence of
and products that could have industrial applications. This common bacterial phyla to geographical regions (coun-
treasured reservoir is largely unexploited since more than tries/continents) using the NCBI nucleotide database to get
99% of environmental microbes cannot be cultured under microbial data accessions with a country qualifier set and
current laboratory conditions, leaving their potential largely compare with the global biodiversity information facility
unused. Understanding the unculturable fraction of Earth’s (GBIF) database data set. This work emphasizes the impor-
microbiome is essential to understand the evolution, sus- tance of metadata tracking and submission in a standard-
tainability of life on Earth, and the development of various ized and enforced fashion and highlights the limitations of
industrial products that have potential applications across all resources such as NCBI and GBIF as well as the biased
major industries. sampling throughout the years, which limits an unbiased
This special issue contains seven reviews and eighteen understanding of bacterial biodiversity and geographical
research articles that provide a better understanding (a) of the abundance.
hidden microbial diversity and (b) the use and development S. Ntougias et al. provided a detailed review of the
of new technologies that can lead to potential biotechnologi- microbial communities identified over the past 20 years in
cal applications. olive mill wastes (OMW) using both culture-dependent and
K. A. Kormas and D. S. Lymperopoulou presented a culture-independent approaches. They also discuss OMW-
review article on the meta-analysis of the 16S rRNA and induced toxicity and the effects that the OMW have on
mlrA (microcystinase) gene diversity of isolates that are soil microbiota. Finally, it is presented in an elegant way
known to degrade cyanobacterial toxins. Members of Pro- the biotechnological importance of the OMW microbiota
teobacteria, Arthrobacter, Bacillus, and Lactobacillus have with respect to (a) the biodegradation of OMWs, (b) the
been found as potential degraders of the toxins produced by bioconversion aspects of the OMW microbiota, and (c) the
Cyanobacteria. plant-disease-suppressive properties of OMW.
2 BioMed Research International
D. Trabelsi and R. Mhamdi report on the effect that the 70–80∘ C (pH 8-9). The isolated lipases revealed exceptional
release of microbial agents could have on indigenous soil thermostability with high optimum activity temperatures,
microbial communities and provide an extensive review on and they represent very promising candidate enzymes for a
the most significant studies addressing the impact of inocu- variety of high temperature industrial lipolytic applications.
lation on soil microbial communities. The authors conclude The third paper by I. Fhoula et al. examined the biotech-
that the assessment of the observed impacts depends largely nological properties of lactic acid bacteria isolated from the
on the techniques used to address the dynamics of the soil rhizosphere of olive trees and desert truffles. Rhizospheric
microbial communities. strains that belong to Enterococcus and Weissella exhibited
Three papers by L. Berg et al., A. K. Chaudhary et al., strong antibacterial activity against plant pathogenic and
and P. Jeandet et al. add new insights into the concept of food-borne bacteria.
microbial cell factories and their exploitation as platforms The papers by I. Maeda et al., C. Hamdi et al., and A. Rizzi
for the biosynthesis of numerous molecules. The first paper et al. examined the diversity of gut microbes. More specifi-
describes the application of a combinatorial mutagenetic cally I. Maeda et al. present a study of gut microorganisms
approach, which resulted in the development of variants from three jungle crows using 16S rRNA-targeted sequencing
of the PAOX1 promoter of the yeast Pichia pastoris with which reveals the presence of potentially pathogenic bacterial
increased expression levels and abolished glucose repression. genera. These findings indicate the importance of crows in
The second paper provides an overview of the application of public health and their role as natural reservoirs of pathogenic
omic approaches into the engineering of Streptomyces aiming microorganisms. C. Hamdi et al. examined the diversity
to augment their industrial exploitation in the production of Paenibacillus larvae, the causative agent of American
of secondary metabolites. Pitfalls that have hampered the foulbrood (AFB). BOX-PCR fingerprints indicated relatively
use of Streptomyces as cell factories for the production of high intraspecific diversity among the P. larvae isolates with
antibiotics are described and current examples and future six genotypes being identified. A. Rizzi et al. characterized the
trends on the use of omics in the genetic improvement of this gut microbial communities that are associated with the wood-
Actinobacteria group are presented. The third paper gives an boring beetle Anoplophora chinensis using culture-dependent
overview of the recent progresses that have been made for and culture-independent approaches. Mainly Proteobacteria,
the last 10 years in metabolic engineering of microorganisms Actibobacteria, and Firmicutes dominated the bacterial diver-
for the biosynthesis of natural products of pharmaceutical sity with the most dominant taxa being the Enterobacteriaceae
significance. family of Gammaproteobacteria.
Two papers report on the complexity of the microbial Three papers by D. El Hidri et al., C. Moulas et al., and
interactions. L. Vanysacker et al. following a multiphasic L. Ma et al. studied the diversity of microbes in different
approach demonstrated the complexity of the microbial ecosystems. D. El Hidri et al. deployed a culture-dependent
interactions controlling biofilm formation on microfiltration approach to study the diversity and ecology of haloalka-
membranes used in wastewater treatment facilities. Their liphilic bacteria in arid saline ecosystems from Southern
study elegantly demonstrated that it is mostly the com- Tunisia. 122 haloalkaliphilic strains were isolated with thir-
position of the microbial community and not the type of teen genera and twenty distinct species being identified. In
membrane that drives the biofouling process. C. Dı́az et al. the second paper C. Moulas et al. examined the impact of
studied the behaviour of yeasts during spontaneous wine fer- two systemic pesticides on the epiphytic fungal and bacterial
mentation using culture-dependent and culture-independent communities via DGGE and cloning. The fungal community
approaches. They report that the yeast diversity in musts was dominated by putative plant pathogenic ascomycetes,
from red grape was greater than the one recorded from white with yeast isolates of Cryptococcus sp. being stimulated
grape varieties. Interestingly, yeast quantification indicated by the application of imidacloprid, suggesting a potential
that non-Saccharomyces yeasts were present during the entire role in its degradation. In the third paper the composition
fermentation process with R. mucilaginosa and P. anomala and dynamics of planktonic viruses and their relationship
being the most prominent species. with environmental parameters in natural freshwater were
Three papers by L. Bifulco et al., M. Stathopoulou et examined. Correlation analysis indicated that the main factor
al., and I. Fhoula et al. report on the exploitation of bac- influencing viral abundance is bacteria.
terial and gene diversity for the development of potential Two papers by M. S. Silva et al. and G. S. Kanini et al.,
biotechnological applications. The first paper presents a first addressed the Actinobacterial diversity and their potential
evidence for the development and preliminary validation of biotechnological applications. In the first paper the diversity
an electrochemical genosensor, which could be used as a of Actinobacteria spp from three Brazilian Savannah soils was
rapid diagnostic tool for the detection of pathogenic Listeria examined. All sites exhibited a high Streptomyces diversity
monocytogenes over nonpathogenic Listeria strains. This is and interestingly enough there were no significant differ-
based on the detection of the inlA gene encoding an 80 ences in the concentrations of phosphorus, magnesium, and
kDa surface protein which allows Listeria to enter the cells. organic matter in the soils. Samples from the rainy season
The second paper by P. Stathopoulou et al. examined the exhibited higher Actinobacterial diversity than the samples
lipolytic profile of 101 bacterial strains isolated from the from the dry season. In the second paper the bacterial
volcanic area of Santorini, Aegean Sea, Greece. Nine bacterial strain ACTA155, member of Streptomyces genus isolated from
strains exhibiting extracellular lipase activities were charac- diverse Greek habitats, exhibited strong antifungal activity
terized as Aneurinibacillus sp. with an optimum activity at against the phytopathogenic fungus Fusarium oxysporum.
BioMed Research International 3
Interestingly, the metabolites involved in the antagonistic

action of the isolate showed to be more than one and
produced independently of the presence of the pathogen
which further highlighted the biotechnological potential of
this isolate.
Two papers by A. Krimitzas et al. and I. B. Fekih et al.
examined fungal diversity. In the first paper the diversity of
Aspergillus species using morphological and molecular crite-
ria is presented. In very diverse genera like Aspergillus the use
of single gene-based analyses does not solve all ambiguities
and does not always represent the evolutionary history of
the species. Utilization of nuclear and mitochondrial genes
enabled the authors to fully characterize more than thirty-five
Aspergillus strains in sixteen sections. In the second paper the
diversity of Entomophthoralean fungi was investigated using
morphological and molecular (ITS) markers in the aphid
species of Sitobion avenae and Entomophthora planchoniana,
with the fungal species of Pandora neoaphidis and Entomoph-
tra planchoniana being the most dominant.
The role of plant growth promoting (PGP) bacteria was
examined in two papers. R. Marasco et al. examined the
environmental factors that influence the PGP potential of
the root-associated bacteria with the grapevine root sys-
tem from Egypt, Tunisia, and Northern Italy. PCR-DGGE
analysis indicated that the structure of endospheric and
rhizospheric bacterial communities was highly diverse and
was associated with a cultivar/latitudinal/climatic effect. The
microbial profile of the Tunisian grapevines was more sim-
ilar with the Tunisian than those cultivated in Northern
Italy. In the second paper F. Mapelli et al. characterized
halophilic/halotolerant bacteria from the rhizosphere of
Salicornia plants and bulk soil collected from hypersaline
systems in Tunisia. The biotechnological potential of these
isolated bacterial strains was further highlighted when at least
twenty Halomonas strains displayed PGP features in vitro
and were able to colonize Salicornia roots under laboratory
conditions.
Collectively, these papers give a comprehensive view on
the immense potential that the study of microbial diversity
can provide towards the development of biotechnological
applications. Current and ongoing advances in the field
of omic approaches will certainly propel our capacity to
understand and further exploit microbial diversity. In this
regard, these articles provide a prospective into the future of
this field.
Acknowledgment
We thank the authors of the submitted papers for their
contribution. The preparation of this special issue would
not have been possible without the generous support and
dedication of experts that evaluated the papers submitted.
George Tsiamis
Dimitrios Karpouzas
Ameur Cherif
Konstantinos Mavrommatis
http://dx.doi.org/10.1155/2013/648141
Research Article
Cultivation-Dependant Assessment, Diversity, and
Ecology of Haloalkaliphilic Bacteria in Arid Saline Systems of
Southern Tunisia
Darine El Hidri,1,2 Amel Guesmi,1 Afef Najjari,2 Hanen Cherif,1 Besma Ettoumi,2
Chadlia Hamdi,1 Abdellatif Boudabous,1 and Ameur Cherif1,2
1
Laboratory of Microorganisms and Active Biomolecules, Faculty of Sciences of Tunis, University of Tunis El Manar,
2092 Tunis, Tunisia
2
LR Biotechnology and Bio-Geo Resources Valorization, Higher Institute for Biotechnology, Biotechpole Sidi Thabet,
University of Manouba, 2020 Ariana, Tunisia
Correspondence should be addressed to Ameur Cherif; cherif.ameur@gmail.com
Received 30 April 2013; Revised 31 August 2013; Accepted 14 September 2013
Academic Editor: George Tsiamis
Copyright © 2013 Darine El Hidri et al. This is an open access article distributed under the Creative Commons Attribution License,
Haloalkaliphiles are polyextremophiles adapted to grow at high salt concentrations and alkaline pH values. In this work, we isolated
122 haloalkaliphilic bacteria upon enrichments of 23 samples from 5 distinct saline systems of southern Tunisia, growing optimally
in media with 10% salt and at pH 10. The collection was classified into 44 groups based on the amplification of the 16S–23S rRNA
internal transcribed spacers (ITS-PCR). Phylogenetic analysis and sequencing of the 16S rRNA genes allowed the identification
of 13 genera and 20 distinct species. Three gram-positive isolates showing between 95 and 96% of 16S rRNA sequence homology
with Bacillus saliphilus could represent new species or genus. Beside the difference in bacterial diversity between the studied sites,
several species ecological niches correlations were demonstrated such as Oceanobacillus in salt crust, Nesterenkonia in sand, and
Salinicoccus in the rhizosphere of the desert plant Salicornia. The collection was further evaluated for the production of extracellular
enzymes. Activity tests showed that gram-positive bacteria were mostly active, particularly for protease, lipase, DNase, and amylase
production. Our overall results demonstrate the huge phenotypic and phylogenetic diversity of haloalkaliphiles in saline systems
of southern Tunisia which represent a valuable source of new lineages and metabolites.
1. Introduction To adapt in such conditions, haloalkaliphilic microorganisms

have developed various physiological strategies to maintain
Extreme environments are distributed on Earth which were their cell structure and function [4, 5]. These bacteria have
thought to prevent the existence of life. These habitats are largely been identified and studied from the hypersaline
characterized by extreme conditions including physical (tem- environments, soda lakes, solar saltern, salt brines, carbonate
perature and pressure) and chemical parameters (salinity and
springs, and Dead Sea [6]. Their existence clearly indicated
pH) [1]. Major categories of extremophiles include halophiles,
the widespread distribution of such organisms in natural
thermophiles, acidophiles, alkaliphiles, and haloalkaliphiles.
The microflora of high salinity ecosystems has attracted a saline environments [4, 7].
great deal of attention from researchers in this last decade, During the past years special attention has been focused
especially haloalkaliphiles bacteria. In 1982, the term haloal- on the distribution of haloalkaliphilic bacteria diversity in
kaliphile was used for the first time to describe bacteria that different hypersaline and hyperalkaline environments [8].
are both halophilic and alkaliphilic [2]. This group of bacteria Culture based methods are usually used to investigate bio-
is able to grow optimally or very well at pH values at or chemical and genetic diversity by selecting a particular pop-
above 10 along with high salinity (up to 25% (w/v) NaCl) [3]. ulation of microorganisms [9]. Molecular microbial surveys
based on 16S rRNA gene have been adopted to study the 2.2. Enrichment and Isolation of Haloalkaliphilic Bacteria.
phylogenetic diversity in different extreme environments [10– Enrichment was performed on Soap lake Basal Medium
14]. Generally, saline systems are dominated by representa- (SLBM) [25], an enrichment medium for moderately haloal-
tives of the domain Bacteria [15–21]. They possess special kaliphilic bacteria, containing (L−1 ): CaSO4 4 mg; FeSO4
adaptation mechanisms to survive, grow, and thrive under 1 mg; NaCl 10 g; SiO2 5 mg; MgCl2 4 mg; MnSO4 4 mg;
high salinity and alkaline pH. This dual extremity of halophile NH4 O3 50 mg; Na2 SO4 13 g; KH2 PO4 3 g; K2 HPO4 3 g;
and alkaliphile makes these microorganisms very interesting Na2 CO3 1 g, and 1 mL trace element stock solution consist-
from the fundamental and biotechnological research sides ing of (L−1 ) sodium nitriloacetate 1.5 g; MgSO4 ⋅7H2 O 3 g;
[22]. MnSO4 ⋅7H2 O 0.5 g; NaCl 1 g; FeSO4 ⋅7H2 O 0.1 g; CaCl2 ⋅2H2 O
The interest in haloalkaliphilic microorganisms is due 0.1 g; CoCl2 ⋅6H2 O 0.1 g; ZnCl2 0.13 g; CuSO4 ⋅5H2 O 0.01 g;
not only to the necessity for understanding the mechanisms AlK (SO4 )2 ⋅12H2 O 0.01 g; H3 BO3 0.01 g; Na2 MoO4 ⋅2H2 O
of adaptation to multiple stresses and detecting their diver- 0.025 g; NiCl2 ⋅6H2 O 0.024 g, and Na2 WO4 ⋅2H2 O 0.025 g. The
sity, but also to their possible application in biotechnology. final pH of the medium was adjusted to 10 by adding 5 M
Research efforts focused on the discovery of industrial NaOH before autoclaving. One g or 1 mL of each sample
enzymes capable of performing their function under harsh was added to 20 mL of SLBM and incubated in a shaking
conditions have greatly increased over the past decade [7, 22, incubator (200 rpm) at 30∘ C for 5 days. Serial dilutions of the
23]. These enzymes include proteases, lipases, amylases, and enriched cells were plated on solid SLBM [25]. Plates were
DNase, viewed as important candidates for various industries incubated at 30∘ C for 5 days. Colonies growing on the plates
such as food, detergent, chemical, pharmaceutical, paper, were selected based on morphological features, considering
and pulp or waste treatment [4]. Southern Tunisia features pigmentation and size. Each isolate was subjected to suc-
numerous ecosystems including coastal and inland salt lakes, cessive streak plating until a pure colony was obtained. The
respectively, named Sabkha or Sabkhet, and Chotts [24]. isolates were stored in glycerol stocks (25% v/v) at −80∘ C.
These environments are characterized by unstable climatic
conditions, due to the periodic flooding by the subsurface 2.3. DNA Extraction and PCR Conditions. Genomic DNA of
ground water associated with high salt conditions during bacteria was extracted by sodium dodecyl sulfate-proteinase
dry phases, making them fascinating ecosystems to study the K treatment [26]. The 16S rRNA gene from pure cultures
diversity and the ecological adaptations of microorganisms was amplified as a 1.5 kb DNA fragment by PCR using the
thriving in saline systems. universal primers S-D-Bact-0008-a-S-20 (5󸀠 -CTA CGG CTA
CCT TGT TAC GA-3󸀠 ) and S-D-Bact-1495-a-S-20 (5󸀠 -AGA
To our knowledge, no studies have been carried out
GTT TGA TCC TGG CTC AG-3󸀠 ) [26]. 16S–23S rRNA ITS
in order to describe the diversity of haloalkaliphilic bacte-
were amplified using the universal primers S-D-Bact-1494-
ria from North African arid and hypersaline systems. The
a-20 (5󸀠 -GTC GTA ACA AGG TAG CCG TA-3󸀠 ) and L-
present work aimed to evaluate the diversity of haloalka-
D-Bact-0035-a-15 (5󸀠 -CAA GGC ATC CAC CGT-3󸀠 ) [27].
liphilic strains isolated from the inland Chotts and the coastal
PCR amplification was carried out according to the procedure
Sebkha hypersaline systems in Tunisian Sahara, based on
described previously [26]. The presence of specific PCR
different phylogenetic markers and biochemical patterns.
products was verified by electrophoresis on 1.5% and 2% (w/v)
agarose gels for 16S rRNA and ITS amplicons, respectively.
2. Materials and Methods 2.4. Sequencing and Phylogenetic Analysis of 16S rRNA
2.1. Sample Collections. All enrichments and strains Sequences. The 16S rRNA gene sequencing has been car-
described here were isolated from twenty-three samples ried with an automated capillary ABI Biosystem 3130. The
collected from arid saline systems in southern Tunisia obtained sequences were identified by comparison with those
during February 2008 and 2010: salt crust, hypersaline water, available at the National Centre for Biotechnology Informa-
thermomineral water, sand, sediment (with or without salt), tion (NCBI) database (http://www.ncbi.nlm.nih.gov) using
bulk soil, algal biofilm, and the rhizosphere of the desert the BLAST program [28]. The sequences were aligned using
plant Salicornia when present. The sampling sites include Clustal W version 1.8 [29]. Evolutionary distances were
three continental ephemeral salt lakes: Chott el Djerid (9 computed using Jukes and Cantor method [30]. Phyloge-
samples from 4 sites: BDV17, N 33∘ 59󸀠 558󸀠󸀠 , E 08∘ 39󸀠 212󸀠󸀠 ; netic dendrograms were constructed by the neighbor-joining
BDV18, N 33∘ 58󸀠 736󸀠󸀠 , E 08∘ 20󸀠 632󸀠󸀠 ; BDV19, N 33∘ 57󸀠 252󸀠󸀠 , method and trees topology was evaluated by performing
E 08∘ 24󸀠 507󸀠󸀠 ; BDV20, N 33∘ 57󸀠 252󸀠󸀠 , E 08∘ 24󸀠 508󸀠󸀠 ), Chott bootstrap analysis of 1000 data sets using MEGA 4.1 (Molec-
el Douz (3 samples from site BDV6: N 33∘ 28󸀠 204󸀠󸀠 , E ular Evolutionary Genetics Analysis) [31]. The sequences
08∘ 56󸀠 733󸀠󸀠 ), and Sabkhet Ennaouel (2 samples from BDV4: reported in this study have been submitted to NCBI GenBank
N 34∘ 26󸀠 951󸀠󸀠 , E 09∘ 54󸀠 102󸀠󸀠 ); one coastal salt lake, Sabkhet El and the accession numbers are listed in Table 1.
Melah (4 samples from BDIII-11: N 33∘ 25󸀠 119󸀠󸀠 , E 11∘ 00󸀠 523󸀠󸀠 ),
and one nonsaline system; Ksar Ghilane Oasis (5 samples 2.5. Morphological and Physiological Characterization of Iso-
from 2 sites: BDV1, N 32∘ 59󸀠 012󸀠󸀠 , E 09∘ 38󸀠 072󸀠󸀠 ; BDV2, N lates. Gram staining of all isolates was performed according
32∘ 59󸀠 293󸀠󸀠 , E 09∘ 38󸀠 374󸀠󸀠 ) (Figure 1). Samples were collected to the method of Murray and colleagues [32]. Growth of
into sterile flasks and kept aseptically at 4∘ C until analyzed. strains at different pH values was determined in solid SLBM,
France
Italy
Spain
Tunisia
Mediterranean
Morocco Sea
Algeria
Libya
Sabkhet Ennaouel
Site BDV4
Chott el Djerid sites
BDV17, 18, 19 and 20
Chott Douz
Site BDV6
Sabkhet El Melah
Site BDIII-11
N
W E Ksar Ghilane sites

BDV1 and BDV2
S
Figure 1: Location of the sampled sites: BDV1 and BDV2 (Oasis Ksar Ghilane), BDV4 (Sabkhet Ennaouel), BDV6 (Chott el Douz), BDV17,
BDV18, BDV19, BDV20 (Chott el Djerid), and BDIII-11 (Sabkhet El Melah).
in which the pH was adjusted to 7.0, 10, and 11. The ability around the growth indicated the hydrolysis of starch [35].
of strains to grow at different range of salinity at pH 10 and DNase activity of the strains was determined using DNase
pH 7 was performed in solid SLBM plates supplemented with test agar medium. After incubation at 30∘ C for 5 days, the
0, 5, 10, 15, 20, and 25% NaCl (w/v). Growth behaviors were plates were flooded with toluidine blue (0.1%) (w/v). A pink
observed after 5 days of incubation at 30∘ C. halo around the colonies showed the secretion of DNase
[36]. Lipase screening was achieved based on the method of
Gutiérrez and González [37] using Tween 20 as a substrate.
2.6. Screening of Strains for Extracellular Hydrolytic Activities. The presence of lipase activity was demonstrated by the
A qualitative screening was performed to detect the ability formation of white halo due to the formation of precipitates of
of the isolated bacteria to produce extracellular enzymes calcium laurate around the growth after 5 days of incubation
responsible for hydrolytic activities. The tests were performed at 30∘ C.
on different solid media containing 10% NaCl at pH 10.
For alkaline protease detection, SLBM agar medium sup-
plemented with 1% (w/v) skim milk was used as described 3. Results and Discussion
previously [4, 33]. A clear zone around the colony after 5 days
of incubation was taken as evidence of proteolytic activity. 3.1. Isolation and Characterization of Haloalkaliphilic Bacte-
Amylase activity was performed according to the method ria. The diversity of cultivable haloalkaliphilic bacteria was
described by Amoozegar and colleagues [34]. The presence evaluated using culture enrichment followed by isolation on
of amylolytic activity on plates was determined qualitatively haloalkaliphile medium. A total of 23 samples collected from
using SLBM agar medium supplemented with 0.5% (w/v) 4 distinct saline stations (Sabkhas and Chotts) and one desert
soluble starch. After incubation at 30∘ C for 5 days, the plates station were processed. The morphological characteristics of
were flooded with 0.3% I2 -0.6% KI solution. A clear zone the isolates showed a wide variability including size, color,
4
Table 1: Phylogenetic characterization of the haloalkaliphiles (ITS haplotype and 16S rRNA identification) and samples (code, nature, and location) collected from arid saline systems in
southern Tunisia during February 2008 and 2010.
Representative
strains of ITS Sampling sites
Accession Phylogenetic Closet described species Accession Sampling Sample Type of
haplotypes (number of isolates
number group and identity (%) number date code matrix
(number of isolates per site)
per haplotype)
H1-BMG G12 (1) KF179184 Proteobacteria Halomonas boliviensis (98%) EU 308325.1 February 2010 BDV1.8.A Ksar Ghilane Mud
H2-BMG F5 (2) KF179185 Firmicutes Bacillus saliphilus (96%) HM 811185.1 February 2010 BDV1.8.B Ksar Ghilane Black sediment
H3-BMG D30 (3) KF179190 Proteobacteria Halomonas janggokensis (100%) EU 308361.1 February 2010 BDV1.8.C Ksar Ghilane Thermomineral water
H4-BMG D39 (3) KF179189 Firmicutes Bacillus saliphilus (95%) HM 811185.1 February 2010 BDV1.8.C Ksar Ghilane Thermomineral water
BDV1.4 Ksar Ghilane (1) Sand with vegetation
H5-BMG D32 (3) KF179187 Actinobacteria Nesterenkonia lacusekhoensis (98%) GQ 064877.1 February 2010
BDV6.2 Chottel Douz (2) Salt crust
H6-BMG G3 (1) KF179188 Firmicutes Bacillus saliphilus (95%) HM 811185.1 February 2010 BDV1.8.B Ksar Ghilane Black sediment
H7-BMG F7 (1) KF179205 Proteobacteria Halomonas gomseomensis (99%) EU 308352.1 February 2010 BDV4.2 Sabkhet Ennaouel Saline water
H8-BMG ED18 (4) KF179213 Firmicutes Salinicoccus alkaliphilus (98%) GU 363531.1 February 2008 BDV6.3 Chottel Douz Algal biofilm
BDV1.8.A Ksar Ghilane (1) Mud
H9-BMG D26 (6) KF179191 Proteobacteria Halomonas janggokensis (100%) EU 308361.1 February 2010
BDV6.3 Chottel Douz (5) Algal biofilm
BDV4.1 Sabkhet Ennaouel (1) Sediment with salt crust
H10-BMG F8 (3) KF179206 Proteobacteria Halomonas taeanensis (98%) HQ 190038 February 2010
BDV17.3/18.3 Chottel Djerid (2) Sediment with salt
H11-BMG G6 (1) KF179174 Proteobacteria Halomonas elongata (99%) NR 029227.1 February 2010 BDV18.3 Chottel Djerid Black sediment
BDIII-11.A1/B1 Sabkhet El Melah (6) Sediment with salt
H12-BMG ED65 (7) KF179199 Firmicutes Salimicrobium luteum (100%) FJ 157154.1 February 2008
H13-BMG ED6 (3) KF179211 Firmicutes Oceanobacillus iheyensis (98%) GU 326361.1 February 2008 BDV6.2 Chottel Douz Salt crust
H14-BMG F4 (1) KF179175 Actinobacteria Nesterenkonia halobia (99%) EF 153433.1 February 2010 BDV20.1 Chottel Djerid Sand
H15-BMG E7 (1) KF179194 Firmicutes Salinicoccus hispanicus (99%) NR 025645.1 February 2008 BDIII-11.C3 Sabkhet El Melah Salicornia rhizosphere
H16-BMG E6 (1) KF179210 Firmicutes Salinicoccus alkaliphilus (97%) GU 363531.1 February 2008 BDV6.3 Chottel Douz Algal biofilm
BDIII-11.C3 Sabkhet El Melah (7) Salicornia rhizosphere
H17-BMG ED88 (11) KF179193 Firmicutes Salinicoccus hispanicus (99%) NR 025645.1 February 2008
BDV6.3/BDV6.2 Chottel Douz (4) Algal biofilm/salt crust
H18-BMG G2 (1) KF179183 Firmicutes Piscibacillus salipiscarius (98%) HM 222702.1 February 2010 BDV19.6 Chottel Djerid Sediment and salt
BDIII-11.B1/A2 Sabkhet El Melah (3) Sediment with salt crust
H19-BMG E11 (4) KF179202 Firmicutes Halobacillus litoralis (100%) HM 636928.1 February 2008
BDV20.2 Chottel Djerid (1) Sediment (sandy soil)
H20-BMG D102 (1) KF179176 Proteobacteria Halomonas elongata (99%) NR 029227.1 February 2010 BDV18.3 Chottel Djerid Sediment
H21-BMG ED25 (1) KF179200 Actinobacteria Arthrobacter gangotriensis (99%) FR 749771.1 February 2008 BDIII-11.A1 Sabkhet El Melah Sediment with salt
BDV17.2 Chottel Djerid (1) Salt crust
H22-BMG D91 (5) KF179178 Proteobacteria Halomonas janggokensis (100%) EU 308361.1 February 2010
BDV1.8.A/1.8.C Ksar Ghilane (4) Mud/thermomineral water
BDV20.1/19.6 Chottel Djerid (4) Sand
H24-BMG D115 (15) KF179177 Actinobacteria Nesterenkonia halobia (99%) EF 153433.1 February 2010 BDV1.4 Ksar Ghilane (7) Sand with vegetation
BDV4.2/4.1 Sabkhet Ennaouel (4) Saline water/salt crust
Table 1: Continued.
Representative
strains of ITS Sampling sites
Accession Phylogenetic Closet described species Accession Sampling Sample Type of

haplotypes (number of isolates
number group and identity (%) number date code matrix
(number of isolates per site)
per haplotype)
H26-BMG G7 (2) KF179182 Proteobacteria Halomonas elongata (99%) NR 029227.1 February 2010 BDV20.2 Chottel Djerid Sediment
H27-BMG E9 (6) KF179209 Firmicutes Oceanobacillus iheyensis (98%) HM 854234.1 February 2008 BDV6.2 Chottel Douz Salt crust
Sediment with water and
H28-BMG G8 (1) KF307740 Actinobacteria Leucobacter chromiireducens (99%) EF 153433.1 February 2008 BDIII-11.A2 Sabkhet El Melah
salt crust
H29-BMG E4 (1) KF179215 Proteobacteria Halomonas ventosae (99%) FM 210950.1 February 2008 BDV6.1 Chottel Douz Saline water
H30-BMG G11 (1) KF179180 Proteobacteria Halomonas subterranea (99%) EU 308353.1 February 2010 BDV17.3 Chottel Djerid Sediment
H31-BMG ED15 (3) KF179214 Firmicutes Oceanobacillus picturae (98%) HM 179167.1 February 2008
BDIII-11.B1 Sabkhet El Melah (1) Sediment with salt
H32-BMG E2 (2) KF179198 Firmicutes Brevibacillus agri (99%) HM 355618.1 February 2008 BDIII-11.A1 Sabkhet El Melah Sediment with salt
Sediment with water and
H33-BMG E1 (1) KF179203 Firmicutes Halobacillus litoralis (100%) HM 636928.1 February 2008 BDIII-11.A2 Sabkhet El Melah
salt crust
H34-BMG E5 (2) KF179197 Firmicutes Piscibacillus salipiscarius (98%) HM 222702.1 February 2008 BDIII-11.A1 Sabkhet El Melah Sediment with salt
H35-BMG ED37 (1) KF179195 Firmicutes Salinicoccus hispanicus (99%) NR 025645.1 February 2008 BDIII-11.C3 Sabkhet El Melah Salicornia rhizosphere
H36-BMG D109 (1) KF179179 Firmicutes Halobacillus profundi (99%) EU 482426.1 February 2010 BDV19.6 Chottel Djerid Sediment and salt
H37-BMG E3 (1) KF179207 Proteobacteria Halomonas ventosae (99%) FM 210950.1 February 2008 BDV6.1 Chottel Douz Saline water
H38-BMG G4 (2) KF179181 Firmicutes Halobacillus litoralis (100%) HM 636928.1 February 2010 BDV17.3 Chottel Djerid Sediment
H39-BMG F11 (1) KF179186 Proteobacteria Halomonas ventosae (99%) AB 617544.1 February 2010 BDV1.8.B Ksar Ghilane Black sediment
H40-BMG D16 (1) KF179204 Proteobacteria Halomonas gomseomensis (99%) EU 308352.1 February 2010 BDV4.2 Sabkhet Ennaouel Saline water
H41-BMG E8 (3) KF307741 Firmicutes Marinococcus halophilus (99%) FJ 887949.1 February 2008 BDIII-11.C3 Sabkhet El Melah Salicornia rhizosphere
H42-BMG F2 (1) KF179201 Proteobacteria Halomonas boliviensis (98%) EU 308325.1 February 2008 BDIII-11.C3 Sabkhet El Melah Salicornia rhizosphere
Sediment with salt
Virgibacillus halodenitrificans BDIII-11.B1/C3 Sabkhet El Melah (6) crust/salicornia
H43-BMG ED33 (8) KF179196 Firmicutes AM 950296.1 February 2008
(99%) rhizosphere
H44-BMG D12 (1) KF179212 Firmicutes Oceanobacillus iheyensis (98%) HM 854234.1 February 2008 BDV6.2 Chottel Douz Salt crust
5
and margin, with 76.15% of them being Gram-positive. The 3.2. Bacterial Collection Dereplication and Identification. ITS
versatility to grow in different range of NaCl concentrations fingerprinting method is a molecular tool based on the
and pH values is reported in Table 2. On the basis of their sequence and length heterogeneity of the bacterial rRNA
salt tolerance, the collection could be classified into three operon 16S–23S intergenic spacer and provides a high phy-
groups: extremely halotolerant (growing at NaCl concentra- logenetic resolution. It can discriminate bacterial isolates up
tion ranging from 0 to 25%), moderate halotolerant (growing to the subspecies level [26, 49, 50]. To manage the large set of
between 0 and 10% NaCl), and strict halophilic bacteria (i.e., isolates in our collection, ITS-PCR fingerprinting was applied
that cannot be cultured without salt). Similarly, depending on as a first screening method. Among the 122 isolates, 44
their tolerance to pH, strains can be divided into two groups: distinct haplotypes (H1–H44) were detected. All profiles were
facultative alkaliphile which represent the dominant fraction composed by 1 up to 8 reproducible bands of approximate
of the collection (81.15%) and obligate alkaliphile bacteria sizes ranging from 180 to 800 bp (Figure 2).
(18.85%).
The most encountered haplotype was ITS-H24 revealed
Extremely halotolerant bacteria, in which their salt toler- in 15 strains isolated from sand and sediment samples
ance ranged between 0 and 15, 20, or 25% (w/v) represented collected from Ksar Ghilane, Chott el Djerid, and Sabkhet
the major part of this collection (71.3%). In similar studies, Ennaouel. Strains belonging to this haplotype were classified
strains isolated from alkaline Lonar lake in India [7] and as strict halophile (at pH 7) and extremely halotolerant at
from mineral pool in Campania (southern Italy) [38] were alkaline pH (pH 10-11). The second most frequent represented
shown to be extremely haloalkalitolerant, tolerating high pattern was haplotype ITS-H17 present in 11 strains isolated
concentrations of NaCl up to 25% and different pH values (7– from Salicornia plants rhizosphere and algal biofilm collected
10). from Sabkhet El Melah and Chott el Douz, respectively. These
Combining the salt and pH requirements and their effects strains were found to be able to grow in media with 15%
on the growth, the group of bacteria that could be considered NaCl and pH ranging from 7 to 11. Other ITS haplotypes
as obligate haloalkaliphiles represent 24.5% (𝑛 = 30) of were frequently encountered like ITS-H43 in 8 isolates and
the collection. They were mainly isolated from the extreme ITS-H9, and H12 shown by 6 strains. The remaining ITS
saline systems of Chott el Djerid, Sabkhet El Melah, and haplotypes were shown to be, in the major part, strains
Chott el Douz (Tables 1 and 2). The ability of haloalkaliphilic specific haplotypes (Table 1).
strains to grow at a wide range of salinities and pH could be Partial 16S rRNA gene sequencing was performed for
assigned to their adaptation to the changing levels of salinity representative isolates of each distinct haplotype (𝑛 = 44) and
and by evolving typical strategies to cope with salt stress: analyzed using BLAST. Phylogenetic analysis revealed that
osmoregulation and modification in cell morphology and the isolates were allocated into thirteen different genera with
structure [39, 40]. It is interesting to note that a subcollection an uneven distribution: Halomonas, Salinicoccus, Nesterenko-
(23 isolates) of the obligate haloalkaliphiles showed variability nia, Oceanobacillus, Virgibacillus, Halobacillus, Salimicro-
in their salt tolerance with different pH values. At alkaline pH bium, Bacillus, Piscibacillus, Marinococcus, Brevibacillus,
(10-11), they were able to cope with the absence of salt, but at Leucobacter, and Arthrobacter. They were placed into the
neutral pH 7, they require an amount of NaCl higher than three major bacteria phyla Firmicutes, Actinobacteria and
1%. They were thus considered as strict halophilic bacteria at Gammaproteobacteria. In a similar work, microbial diversity
neutral pH. Only 7 isolates (5.73%) from Sabkhet El Melah analysis in water and sediment of lake Chaka, a hypersaline
and Chott el Douz were shown to be strict halophiles at all lake on Tibetan plateau, permitted the assignation of bac-
pH values (Tables 1 and 2). The exact relation between the terial community into the same three groups of Firmicutes,
salt requirement and tolerance and the pH homeostasis in Gammaproteobacteria, and Actinobacteria [51].
the cell, raises several questions and represents an interesting The Firmicutes phyla including Bacillus, Halobacillus,
issue to be studied [7]. Studies on aerobic alkaliphilic bacteria Piscibacillus, Oceanobacillus, Virgibacillus, Salimicrobium,
thriving in alkaline Lonar Lake in India showed that obligate Marinococcus and Salinicoccus were more abundant and
haloalkaliphles related to the genus Alkalibacillus could be diverse. They constitute also the obligate haloalkaliphiles
isolated only in specific medium containing 2% NaCl and at fraction at neutral pH (Halobacillus, Piscibacillus, and
pH 10 [7]. Marinococcus) and at all pH values (Salimicrobium that needs
A fraction of 4.1% of our collection was classified as at least 5% NaCl to grow) (Table 2). Compared to similar
moderate haloalkaliphiles (0–10% of NaCl growth range), studies carried out on salt lake [21, 44, 52], marine habitat
a proportion similarly isolated from other different saline [20, 53], and other hypersaline sediments [47] where limited
and alkaline environments [41–47]. Occurrence of haloalka- number of genera were identified, arid saline systems of
litolerant, obligate, and moderate haloalkaliphiles bacteria, Tunisia revealed a highly diverse community. The isolates
in different sampling locations, highlighted the diversity obtained from Alkaline Lonar lake in India were associated
and the widespread distribution of these microorganisms in with the members of diverse Bacillus related genera (Paeni-
arid-saline systems of southern Tunisia. This versatility of bacillus, Bacillus, and Alkalibacillus) [7]. While in deep-
growth characteristics could be explained by their ability of sea hypersaline lakes, taxonomic analyses showed that two-
osmoregulation, in relation with alkaline pH, through which thirds of 89 isolates were mostly representative of the genus
they maintain an internal osmotic potential that equals their Bacillus and the related genera Halobacillus, Virgibacillus, and
external environment [48]. Pontibacillus [54]. In comparison to these reports, Bacillus
Table 2: Salt and pH tolerance levels and hydrolytic activities of ITS haplotype representatives isolates.
Representative strains NaCl tolerance pH tolerance Production of extracellular enzymes

Identification Gram
of ITS haplotypes range (%) range Protease Lipase DNase Amylase
H1-BMG G12 Halomonas boliviensis − 0–20 ± 0.1 7–11 ± 0.2 − + + −
H2-BMG F5 Bacillus saliphilus + 0–25 ± 0.1 7–11 ± 0.2 − − + +
H3-BMG D30 Halomanas janggokensis − 0–20 ± 0.1 7–11 ± 0.2 − + − −
H4-BMG D39 Bacillus saliphilus + 0–25 ± 0.1 7–11 ± 0.2 + − + +
H5-BMG D32 Nesterenkonia lacusekhoensis + 0–25 ± 0.1 7–11 ± 0.2 − − − −
H6-BMG G3 Bacillus saliphilus + 0–25 ± 0.1 7–11 ± 0.2 + − + +
H7-BMG F7 Halomonas gomseomensis − 0–20 ± 0.1 7–11 ± 0.2 − + + −
H8-BMG ED18 Salinicoccus alkaliphilus + 0–25 ± 0.1 7–11 ± 0.2 + – − −
H10-BMG F8 Halomonas taeanensis − 0–20 ± 0.1 7–11 ± 0.2 − – − −
H11-BMG G6 Halomonas elongata − 0–25 ± 0.1 7–11 ± 0.2 − + − +
H12-BMG ED65 Salimicrobium luteum + 5–20 ± 0.1 7–11 ± 0.2 − + − −
H13-BMG ED6 Oceanobacillus iheyensis + 0–15 ± 0.1 7–11 ± 0.2 + – + –
H14-BMG F4 Nesterenkonia halobia + 0–25 ± 0.1 7–11 ± 0.2 + + − +
H15-BMG E7 Salinicoccus hispanicus + 0–15 ± 0.1 7–11 ± 0.2 − − − −
H16-BMG E6 Salinicoccus alkaliphilus + 0–25 ± 0.1 7–11 ± 0.2 − − − −
H17-BMG ED88 Salinicoccus hispanicus + 0–15 ± 0.1 7–11 ± 0.2 − − − −
1–20 ± 0.1 at pH 7
H18-BMG G2 Piscibacillus salipiscarius + 7–11 ± 0.2 − + − +
0–20 ± 0.1 at pH 10-11
H19-BMG E11 Halobacillus litoralis + 0–20 ± 0.1 7–11 ± 0.2 + − + +
H20-BMG D102 Halomonas elongata − 0–25 ± 0.1 7–11 ± 0.2 + + + +
H21-BMG ED25 Arthrobacter gangotriensis + 0–10 ± 0.1 7–11 ± 0.2 − − − −
7
8
Table 2: Continued.
Representative strains NaCl tolerance pH tolerance Production of extracellular enzymes
Identification Gram
of ITS haplotypes range (%) range Protease Lipase DNase Amylase
1–25 ± 0.1 at pH 7
H24-BMG D115 Nesterenkonia halobia + 7–11 ± 0.2 − + − +
0–25 ± 0.1 at pH 10-11
H25-BMG ED60 Salinicoccus hispanicus + 0–15 ± 0.1 7–11 ± 0.2 + − − −
H26-BMG G7 Halomonas elongata − 0–20 ± 0.1 7–11 ± 0.2 − + − +
H27-BMG E9 Oceanobacillus iheyensis + 0–20 ± 0.1 7–11 ± 0.2 + − + −
H28-BMG G8 Leucobacter chromiireducens + 0–10 ± 0.1 7–11 ± 0.2 − + − +
H29-BMG E4 Halomonas ventosae − 0–15 ± 0.1 7–11 ± 0.2 + − − −
H30-BMG G11 Halomonas subterranea − 0–20 ± 0.1 7–11 ± 0.2 − + − −
H31-BMG ED15 Oceanobacillus picturae + 0–15 ± 0.1 7–11 ± 0.2 + − − −
H32-BMG E2 Brevibacillus agri + 0–10 ± 0.1 7–11 ± 0.2 − − − −
H33-BMG E1 Halobacillus litoralis + 0–20 ± 0.1 7–11 ± 0.2 + − + +
1–15 ± 0.1 at pH 7
H34-BMG E5 Piscibacillus salipiscarius + 7–11 ± 0.2 + + − +
0–15 ± 0.1 at pH 10-11
H36-BMG D109 Halobacillus profundi + 0–20 ± 0.1 7–11 ± 0.2 + − + +
H37-BMG E3 Halomonas ventosae − 0–15 ± 0.1 7–11 ± 0.2 − − − −
1–20 ± 0.1 at pH 7
H38-BMG G4 Halobacillus litoralis + 7–11 ± 0.2 − − + +
0–20 ± 0.1 at pH 10-11
H39-BMG F11 Halomonas ventosae − 0–25 ± 0.1 7–11 ± 0.2 − − − −
H40-BMG D16 Halomonas gomseomensis − 0–20 ± 0.1 7–11 ± 0.2 − + + −
1–25 ± 0.1 at pH 7
H41-BMG E8 Marinococcus halophilus + 7–11 ± 0.2 − − + −
0–25 ± 0.1 at pH 10-11
H42-BMG F2 Halomonas boliviensis − 0–25 ± 0.1 7–11 ± 0.2 − + + −
H43-BMG ED33 Virgibacillus halodenitrificans + 0–15 ± 0.1 7–11 ± 0.2 + − − −
H44-BMG D12 Oceanobacillus iheyensis + 0–10 ± 0.1 7–11 ± 0.2 − − + −
Isolates (number of strains/16S rRNA(%) similarity, [salt tolerance]) Assignation ITS-haplotypes
M (bp)
200
300
500
800
100 Oceanobacillus picturae (HM179167)
BMG ED15 (3/98%, [0–15%]) H 31
56
Oceanobacillus iheyensis (NR 028001.1) Oceanobacillus
100 BMG E9 (6/98%, [0–20%]) H 27
48 BMG D12 (1/98%, [0–10%]) H 44
71 BMG ED6 (3/98%, [0–15%]) H 13
44 BMG ED33 (8/99%, [0–15%]) Virgibacillus H 43
100 Virgibacillus halodenitrificans (NR 042967.1)
Piscibacillus salipiscarius (NR 041260.1) Piscibacillus H 34
61 100 BMG E5 (2/98%, [1–15%] at pH 7) H 18
66 BMG G2 (1/98%, [1–20%] at pH 7)
100 BMG ED65 (7/100%, [5–20%]) Salimicrobium H 12
Salimicrobium luteum (NR 043659.1)
BMG E1 (1/100%, [0–20%]) H 33
97 Halobacillus profundi (NR 041246.1)
53 Halobacillus litoralis (NR 029304.1) Halobacillus
H 19
100 BMG E11 (4/100%, [0–20%]) H 36
Firmicutes
BMG D109 (1/99%, [0–20%]) H 38
BMG G4 (2/100%, [1–20%] at pH 7) H 41
100 BMG E8 (3/99%, [1–25%] at pH 7)
Marinococcus halophilus (FJ887949.1) Marinococcus
Bacillus saliphilus (NR 025554.1)
42 49 BMG F5 (2/96%, [0–25%]) H2
98 Bacillus H6
100 BMG G3 (1/95%, [0–25%]) H4
BMG D39 (3/95%, [0–25%])
92 BMG E6 (1/97%, [0–25%]) H 16
99 BMG ED18 (4/98%, [0–25%]) H8
60 Salinicoccus kekensis (GU363531.1)
100 Salinicoccus alkaliphilus (NR 025108.1)
100 Salinicoccus hispanicus (NR 025645.1) Salinicoccus
BMG ED60 (2/99%, [0–15%]) H 25
H 17
99 BMG ED88 (11/99%, [0–15%]) H 15
BMG E7 (1/99%, [0–15%]) H 35
67 BMG ED37 (1/99%, [0–15%]) H 23
BMG ED46 (2/99%, [0–15%])
BMG E2 (2/99%, [0–10%]) H 32
100 Brevibacillus agri (HM355618.1) Brevibacillus
99 BMG G8 (1/99%, [0–10%]) H 28
Leucobacter chromiireducens (NR 042287.1) Leucobacter
Actinobacteria
100 BMG ED25 (1/99%, [0–10%]) H 21

100 Arthrobacter gangotriensis (NR 029026.1) Arthrobacter
82 BMG D32 (3/98%, [0–25%]) H5
85 Nesterenkonia lacusekhoensis (NR 028928.1)
99 Nesterenkonia halobia (NR 026197.1) Nesterenkonia H 24
99 BMG D115 (15/99%, [1–25%] at pH 7) H 14
76 BMG F4 (1/99%, [0–25%])
87 BMG E3 (1/99%, [0–15%]) H 37
91 BMG E4 (1/99%, [0–15%]) Halomonas H 29
Halomonas ventosae (NR 042812.1) group I
98 BMG F11 (1/99%, [0–25%]) H 39
BMG F8 (3/98%, [0–20%]) H 10
100
Gammaproteobacteria
Halomonas taeanensis (NR 043087.1)

51 Halomonas elongata (NR 029227.1)
99 BMG G7 (2/99%, [0–20%]) Halomonas H 26
BMG G6 (1/99%, [0–25%]) group II H 11
36 BMG D102 (1/99%, [0–25%]) H 20
98 BMG F2 (1/98%, [0–25%]) H 42
81 71 BMG G12 (1/98%, [0–20%]) H1
Halomonas boliviensis (NR 029080.1)
Halomonas gomseomensis (NR 042488.1)
98 BMG F7 (1/99%, [0–20%]) Halomonas H7
76
BMG D16 (1/99%, [0–20%]) group III H 40
60 BMG G11 (1/99%, [0–20%]) H 30
BMG D30 (3/99%, [0–20%]) H3
78 BMG D91 (5/99%, [0–20%]) H 22
84 BMG D26 (6/99%, [0–20%]) H9
21 Halomonas janggokensis (NR 042489.1)
22
69 Halomonas subterranea (NR 044116.1)
0.05
(a) (b)
Figure 2: Phylogenetic diversity of haloalkaliphilic bacteria. (a) Unrooted phylogenetic tree of 44 partial 16S rRNA sequences (500 bp) of
the arid saline system isolates with the 24 closest phylogenetic relatives. The method of Jukes and Cantor was used to calculate evolutionary
distances and tree topology was constructed using MEGA 4.0. Bootstrap values (𝑛 = 1000 replicates) were indicated at the nodes. The number
of isolates per ITS haplotype, the 16S rRNA similarity percentage (refseq rna database), and NaCl range for growth at pH 11 (or at pH 7 where
mentioned), are indicated in parenthesis. (b) 16S–23S rRNA ITS haplotypes of 44 representative isolates as resolved on 2% agarose gels. ITS
haplotype numbers are indicated. Lane M corresponds to a 100 bp ladder.
species are among the most commonly found aerobic, bac- [64]. Interestingly, the moderate halophile isolates Leucobac-
terial alkaliphiles, both in Soda lakes and in less selective ter chromiireducens and Arthrobacter gangotriensis are not
environments [44, 55–58]. The same result was observed known to be natural inhabitant of arid-saline systems.
in other arid saline systems such as the Golea Salt lake in Gram-negative bacteria were represented by a unique
Algeria Sahara [52], Chott el Djerid [21, 44, 59] and Tunisian genus Halomonas counting 23.87% of the whole collection,
multipond solar saltern [18, 58, 59]. This high occurrence in accordance with the recent work of Mapelli et al. [21].
and the ability of Bacillus and Bacillus related genera to Halomonas isolates were retrieved from all sample types,
tolerate salt and alkaline stress prove that they are well assigned to 7 distinct species and clustered within 3 phylo-
adapted to arid-saline environments being physiologically genetic groups: (i) Halomonas group I including H. ventosae
active and not only present as dormant spores. Indeed, recent (𝑛 = 3), and H. taeanensis (𝑛 = 3); (ii) Halomonas
report indicated that they contribute to the system biological group II represented by H. elongata (𝑛 = 4); and (iii)
robustness and function [60]. Three Bacillaceae strains (BMG Halomonas group III constituted by H. boliviensis (𝑛 = 2), H.
F5, BMG D39, and BMG G3) isolated from sediments and gomseomensis (𝑛 = 2) and the related species H. janggokensis
(𝑛 = 14) and H. subterranea (𝑛 = 1) (Table 1, Figure 2).
thermomineral water from Ksar Ghilane BDV1.8 site (a
Considering the nonmonophyletic status of the Halomonas
thermomineral natural pool) showed a very low 16S rRNA
genus and the need of a deep taxonomic revision [65],
sequence homology (95-96%) with Bacillus saliphilus that was
the number of recovered species indicates high intragenus
previously isolated from mineral pool in southern Italy [38]. diversity. In addition, there was no clear correlation between
The Ksar Ghilane strains could represent new alkaliphilic the recovered Halomonas species with their isolation origin,
and extremely halotolerant species related to B. saliphilus, pointing out their adaptation capabilities to harsh conditions.
particularly adapted to high mineral concentrations in desert Indeed, members of this genus have been isolated from
environment. diverse saline environments, including athalassohaline and
Other species-microniche correlations are noteworthy. thalassohaline Lakes and marine waters [20, 66]. However,
Oceanobacillus iheyensis strains (𝑛 = 10) were all isolated by applying culture dependent and independent approaches
from salt crust samples, whereas Halobacillus (𝑛 = 8), [18, 20, 58], more diverse communities including bacteria
Piscibacillus (𝑛 = 3), and Salimicrobium (𝑛 = 7) isolates from the Alpha-, Beta-, Gamma-, and Deltaproteobacteria
were recovered from salty sediments and soils (Table 1). subclasses were revealed in similar ecosystems like the Inner
On the other hand, the 17 isolates identified as Salinicoccus Mongolian Soda Lake [17] and the hyperalkaline spring
hispanicus, 5 isolates of Salinicoccus alkaliphilus, and 3 isolates waters in Jordan [67]. The limited number of Gram-negative
of Marinococcus halophilus were clearly associated with the bacteria detected in our hypersaline samples may be due
rhizosphere of the desert plant Salicornia and algal biofilm. to the enrichment and culturing procedure that favor the
Whilst Marinococcus halophilus was recently described as growth of Gram-positive bacteria, as reported earlier [7], and
a plant-growth promoting rhizospheric bacterium isolated where fast-growing alkalitolerant Halomonas sp. outcompete
from the same environment [21], this work constitutes the other Gram negative microorganisms at different NaCl con-
first report on the capabilities of haloalkaliphilic Salinicoccus centration and pH values [51].
species to colonize and thrive into the plant rhizosphere in
desert environment.
The phylum Actinobacteria was represented by 4 species 3.3. Geographic Distribution and Microdiversity. Arid envi-
that belong to the Micrococcaceae family: Nesterenkonia ronment and saline systems in southern Tunisia are charac-
halobia (16 isolates from Chott el Djerid, Sabkhet Ennaouel, terized by unstable climatic conditions, due to the periodic
and Ksar Ghilane; 99% of 16S rRNA sequence identity), flooding by the subsurface ground water associated with high
Nesterenkonia lacusekhoensis (3 isolates from Ksar Ghilane salt during dry phases. These specific conditions make such
and Chott el Douz; 98% of identity), Leucobacter chromi- environment fascinating ecosystems to study the diversity
ireducens (isolate BMG G8 from Sabkhet El Melah; 99% of and the ecological adaptations of thriving microorganisms. In
identity), and Arthrobacter gangotriensis (isolate BMG ED25 the current study, cultivation approach showed a particular
from Sabkhet El Melah; 99% of identity). Species of the genus distribution of haloalkaliphilic bacteria according to their
Nesterenkonia were previously reported as halotolerant and sampling origin (Table 1, Figure 3). The general distribution
were isolated from different saline ecosystems like Brazilian of the genera was very similar in Ksar Ghilane, Sabkhet
Mangrove sediment [61] and hypersaline Ekho lake in East Ennaouel, and Chott el Djerid with low bacterial diversity
Antarctica [62]. Nesterenkonia halobia was also found as and the dominance of Halomonas and Nesterenkonia species
the unique Actinobacteria representative in Salicornia rhi- (Figure 3). Beside the specific occurrence of Nesterenkonia
zosphere [21]. In the current prospection, N. lacusekhoensis species in these stations, Bacillus saliphilus and Halobacillus
and particularly N. halobia were recovered mainly from profundi were exclusively isolated from Ksar Ghilane and
sand samples and showed changing halotolerance behavior Chott el Djerid, respectively.
at neutral and alkaline pH indicating a specific fine-tuned Sabkhet El Melah showed the most diverse commu-
adaptation of these species to sand and salty sediments as nity displaying a mixture of strains affiliated into 11 gen-
ecological niche. With regard to the Arthrobacter species, era. Among them, Arthrobacter, Leucobacter, Brevibacillus,
they were previously reported as halotolerant and were and Marinococcus were exclusively detected in this site. In
isolated from east African soda lakes [63] and Antarctica contrast, we noted the absence of Nesterenkonia strains,
Oceanobacillus
5% Salimicrobium
Virgibacillus 16%
16%
Nesterenkonia
29%
Halomonas
3% Halomonas
50%
Marinococcus
8% Salinicoccus
29%
Piscibacillus
5%
Brevibacillus Bacillus
5% 21%
Leucobacter
3% Arthrobacter Halobacillus
2% 8%
Sabkhet El Melah Ksar Ghilane
(1 site, 4 samples, and 38 isolates) (2 sites, 7 samples, and 24 isolates)
Nesterenkonia
7%
Virgibacillus
7% Halobacillus
Salimicrobium
3% 23%
Halomonas Salinicoccus Halomonas

10% 36% Piscibacillus 45%
5%
Oceanobacillus Nesterenkonia
37% 27%
Chott Douz Chott Djerid

(1 site, 3 samples, and 35 isolates) (4 sites, 9 samples, and 18 isolates)
Nesterenkonia
50% Halomonas
50%
Sabkhet Ennaouel
(1 site, 2 samples, and 7 isolates)
Bacillus Salinicoccus
Virgibacillus Marinococcus
Halobacillus Halomonas
Oceanobacillus Nesterenkonia
Brevibacillus Leucobacter
Piscibacillus Arthrobacter
Salimicrobium
Figure 3: Geographic distribution of haloalkaliphilic bacteria isolated from natural saline systems of southern Tunisian Sahara.
frequently isolated from all the other sites (Figure 3). The high may indicate that the observed diversity is of marine origin
diversity detected in Sabkhet El Melah could be explained by rather than terrestrial. Besides, Leucobacter chromiireducens
its geographic location (a coastal saline system) that allows was first isolated from activated sludge of a waste water treat-
water exchange with the open sea. Indeed, the occurrence of ment plant contaminated with chromium and was shown
Marinococcus halophilus (BMG E8 and two other isolates), a to be halotolerant and able to tolerate up to 5 mM Cr(VI)
marine bacterium shown to be strict halophilic at neutral pH, [68]. Likewise, Arthrobacter gangotriensis is closely related to
A. sulfurous isolated from oil contaminated sludge and able (Table 2). It is interesting to note that combined hydrolytic
to achieve desulphurization [69]. The presence in Sabkhet activities were also detected in many strains. One strain,
El Melah of L. chromiireducens and A. gangotriensis related BMG D102, affiliated to Halomonas elongata showed all four
species may indicate anthropogenic and industrial pollution enzyme activities (PGPR strain as Mapelli et al). Strains affili-
due to their vicinity to an offshore oil field and oil harbor ated to Bacillus saliphilus, Nesterenkonia halobia, Halobacillus
terminal. litoralis, Piscibacillus salipiscarius, and Halobacillus profundi
Chott el Douz is most similar to Sabkhet El Melah in were able to produce 3 hydrolytic activities. Sanchez-Porro
terms of diversity with 6 distinct detected genera: Halomonas, and colleagues [73] showed the abundance of these hydrolytic
Virgibacillus, Salimicrobium, and Nesterenkonia and a marked enzymes produced by moderately halophilic bacteria. It
dominance of bacteria assigned to Oceanobacillus (37%) and is worth noting that Gram-positive bacteria showed more
Salinicoccus (36%). Interestingly, all the isolates assigned to hydrolytic activities. Similar variations in the production of
Salinicoccus alkaliphilus (𝑛 = 5, ITS haplotypes H8 and H16) these enzymes were reported among the bacteria isolated
and to Oceanobacillus iheyensis (𝑛 = 10, ITS haplotypes from Howz Soltan lake in Iran and Pulicat Lake in India
H13, H27, and H44) occurred specifically in this site (Table 1, [15, 23].
Figure 3). In similar studies, S. alkaliphilus was isolated from Two hydrolytic activities were demonstrated by 13
salt lakes; however, O. iheyensis is a deep-sea bacterium isolates affiliated to Halomonas, Halobacillus, Piscibacillus,
with original genomic futures and adaptive capabilities to Oceanobacillus, and Bacillus genera. However, unique
changing environments [5, 70, 71]. The high prevalence of hydrolytic activity was detected in 12 strains assigned
O. iheyensis species in salt crust samples of Chott el Douz to Halomonas, Salinicoccus, Piscibacillus, Virgibacillus,
confirms its adaptation potential to such extreme ecosystem. Oceanobacillus, and Marinococcus genera. On the other
The adaptive capabilities of the dominating haloalka- hand, 11 isolates, members of Nesterenkonia, Halomonas,
liphile species detected in the current study could be, in part, Salinicoccus, and Arthrobacter genera, did not show any
inferred to their intraspecific microdiversity. This micro- activity. This absence may be due to the released hydrolase
diversity is highlighted by the number of ITS haplotypes quantity, not sufficiently enough to cause visible clearing
displayed by a single or a complex of bacterial species. zone on the plates.
Salinicoccus hispanicus isolates, shown to thrive in plant The majority of the enzyme producers were affiliated to
rhizosphere and algal biofilm, were clustered in 5 ITS hap- the Bacillus and Halomonas genera. Lipase was produced
lotypes (H15, H17, H23, H25, and H35). As well, Halomonas by 38.6% of the isolates; DNase was shown by 36.3% of the
isolates recovered from all the sites were allocated into seven strains. For protease and amylase, 34% of the selected strains
different species and 15 ITS haplotypes. Within this genus, were able to release these enzymes. Similar results were
Halomonas group III includes the 3 closely related species observed for species isolated from saline alkaline systems
with 6 distinct ITS haplotypes: H. gomseomensis (H7 and affiliated to Halobacillus sp. [74], Nesterenkonia sp. [75],
H40), H. janggokensis (H2, H9, and H22) and H. subterranea Virgibacillus sp. [76], and Bacillus sp. [77]. The most active
(H30). Isolates of these species that could be considered strains are able to produce at least 3 hydrolases, were isolated
as a single one [65, 72] were recovered from all the sites from Chott el Djerid, Ksar Ghilane, and Sabkhet El Melah,
except from Sabkhet El Melah (Table 1, Figure 3). Their high and were all extremely haloalkalitolerant bacteria.
level of microdiversity could contribute to their ecological
fitness and their ability to adapt to desert and saline environ-
ments. Overall, the microdiversity is attributed to different 4. Conclusion
combinations of DNA sequence blocks making the genome
more competent to accumulate mutations, insertions, and Our overall results indicate that haloalkaliphilic bacteria
deletions due to selective pressure. The exact contribution constitute an important part of the microbiota that inhabits
of the microdiversity to microbial adaptive strategies is arid and saline systems in southern Tunisia. A huge phe-
not clearly elucidated. However, high extent of intraspecific notypic and phylogenetic diversity was observed. Extremely
polymorphism is usually shown by bacterial species that are haloalkalitolerant bacteria were the most dominant group
well adapted and thriving in extreme environments [20, 26, and were affiliated to Bacillus, Nesterenkonia, Salinicoccus,
59]. and Marinococcus genera, of which several isolates could rep-
resent putative new species. A clear correlation between some
species with specific ecological niches was also demonstrated.
3.4. Hydrolytic Activities of Isolates. Beside the bacterial Besides, difference in the bacterial diversity rates between
diversity of the southern Tunisia ecosystem, the current study the studied sites was shown. The heterogeneity of haloal-
assesses the biotechnological potential of desert isolates. The kaliphilic bacteria was confirmed by their hydrolytic enzy-
occurrence of hydrolytic enzymes could be used as biochem- matic patterns variability including protease, lipase, DNase,
ical marker to judge the microbial heterogeneity among the and amylase. These enzymes are generally haloalkaliphilic
selected haloalkaliphilic bacteria. The ability of producing which makes them interesting candidates to be employed
four different hydrolytic enzymes was tested qualitatively for in different industrial processes. The detected phenotypic
44 identified strains in the optimum growth conditions (10% and phylogenetic diversity points out that saline systems of
NaCl and pH 10). A total of 15, 17, 16, and 15 isolates were able southern Tunisia could represent a valuable source of new
to produce protease, lipase, DNase, and amylase, respectively lineages and metabolites.
Acknowledgments [15] H. Sahay, S. Singh, R. Kaushik, A. K. Saxena, and D. K. Arora,

“Characterization of halophilic bacteria from environmental
The authors acknowledge the financial support from the samples from the brackish water of Pulicat Lake, India,” Biolo-
European Union in the ambit of the Project BIODESERT (EU gia, vol. 66, no. 5, pp. 741–747, 2011.
FP7-CSA-SA REGPOT-2008-2, Grant agreement no. 245746) [16] H. Sahay, S. Mahfooz, A. K. Singh et al., “Exploration and
and the Tunisian Ministry of Higher Education and Scientific characterization of agriculturally and industrially important
research in the ambit of the laboratory Projects LR MBA206 haloalkaliphilic bacteria from environmental samples of hyper-
and LR11ES31. saline Sambhar Lake, India,” World Journal of Microbiology and
Biotechnology, vol. 28, no. 11, pp. 3207–3217, 2012.
[17] Y. Ma, W. Zhang, Y. Xue, P. Zhou, A. Ventosa, and W. D. Grant,
References “Bacterial diversity of the Inner Mongolian Baer Soda Lake as
revealed by 16S rRNA gene sequence analyses,” Extremophiles,
[1] L. J. Rothschild and R. L. Mancinelli, “Life in extreme environ-
vol. 8, no. 1, pp. 45–51, 2004.
ments,” Nature, vol. 409, no. 6823, pp. 1092–1101, 2001.
[18] H. Baati, S. Guermazi, R. Amdouni, N. Gharsallah, A. Sghir, and
[2] G. S. H. Soliman and H. G. Trueper, “Halobacterium pharaonis
E. Ammar, “Prokaryotic diversity of a Tunisian multipond solar
sp.nov., a new, extremely haloalkaliphilic archaebacterium with
saltern,” Extremophiles, vol. 12, no. 4, pp. 505–518, 2008.
low magnesium requirement,” Zentralblatt für Bakteriologie
Mikrobiologie und Hygiene: I. Abt. Originale C, vol. 3, no. 2, pp. [19] R. Mwirichia, A. W. Muigai, B. Tindall, H. I. Boga, and E.
318–329, 1982. Stackebrandt, “Isolation and characterisation of bacteria from
the haloalkaline Lake Elmenteita, Kenya,” Extremophiles, vol. 14,
[3] K. Horikoshi, Alkaliphiles, Wiley Online Library, 2008.
no. 4, pp. 339–348, 2010.
[4] K. Horikoshi, “Alkaliphiles: some applications of their prod-
[20] B. Ettoumi, E. Bouhajja, S. Borin, D. Daffonchio, A. Boud-
ucts for biotechnology,” Microbiology and Molecular Biology
abous, and A. Cherif, “Gammaproteobacteria occurrence and
Reviews, vol. 63, no. 4, pp. 735–750, 1999.
microdiversity in Tyrrhenian Sea sediments as revealed by
[5] H. Takami, Y. Takaki, and I. Uchiyama, “Genome sequence cultivation-dependent and -independent approaches,” System-
of Oceanobacillus iheyensis isolated from the Iheya Ridge and atic and Applied Microbiology, vol. 33, no. 4, pp. 222–231, 2010.
its unexpected adaptive capabilities to extreme environments,” [21] F. Mapelli, R. Marasco, E. Rolli et al., “Potential for plant growth
Nucleic Acids Research, vol. 30, no. 18, pp. 3927–3935, 2002. promotion of Rhizobacteria associated with Salicornia growing
[6] S. P. Singh, “Extreme environments and extremophiles,” in in Tunisian hypersaline soils,” BioMed Research International,
National Science Digital Library (CSIR): E- Book, Environmental vol. 2013, Article ID 248078, 13 pages, 2013.
Microbiology, pp. 1–35, CSIR India, 2006. [22] R. Margesin and F. Schinner, “Potential of halotolerant and
[7] A. A. Joshi, P. P. Kanekar, A. S. Kelkar et al., “Cultivable bacterial halophilic microorganisms for biotechnology,” Extremophiles,
diversity of alkaline Lonar lake, India,” Microbial Ecology, vol. vol. 5, no. 2, pp. 73–83, 2001.
55, no. 2, pp. 163–172, 2008. [23] R. Rohban, M. A. Amoozegar, and A. Ventosa, “Screening
[8] M. Foti, D. Y. Sorokin, B. Lomans et al., “Diversity, activity, and isolation of halophilic bacteria producing extracellular
and abundance of sulfate-reducing bacteria in saline and hyper- hydrolyses from Howz Soltan Lake, Iran,” Journal of Industrial
saline soda lakes,” Applied and Environmental Microbiology, vol. Microbiology and Biotechnology, vol. 36, no. 3, pp. 333–340,
73, no. 7, pp. 2093–2100, 2007. 2009.
[9] J. L. Kirk, L. A. Beaudette, M. Hart et al., “Methods of studying [24] N. Stivaletta, R. Barbieri, C. Picard, and M. Bosco, “Astrobiolog-
soil microbial diversity,” Journal of Microbiological Methods, vol. ical significance of the sabkha life and environments of southern
58, no. 2, pp. 169–188, 2004. Tunisia,” Planetary and Space Science, vol. 57, no. 5-6, pp. 597–
[10] K. Ohta, A. Kiyomiya, N. Koyama, and Y. Nosoh, “The basis 605, 2009.
of the alkalophilic property of a species of Bacillus,” Journal of [25] P. A. Dimitriu, S. K. Shukla, J. Conradt et al., “Nitrincola
General Microbiology, vol. 86, no. 2, pp. 259–266, 1975. lacisaponensis gen. nov., sp. nov., a novel alkaliphilic bacterium
[11] O. Shida, H. Takagi, K. Kadowaki, and K. Komagata, “Proposal isolated from an alkaline, saline lake,” International Journal of
for two new genera, Brevibacillus gen. nov. and Aneurinibacillus Systematic and Evolutionary Microbiology, vol. 55, no. 6, Article
gen. nov,” International Journal of Systematic Bacteriology, vol. ID 63647, pp. 2273–2278, 2005.
46, no. 4, pp. 939–946, 1996. [26] A. Cherif, S. Borin, A. Rizzi, H. Ouzari, A. Boudabous, and D.
[12] J. Lu, Y. Nogi, and H. Takami, “Oceanobacillus iheyensis gen. Daffonchiol, “Bacillus anthracis diverges from related clades of
nov., sp. nov., a deep-sea extremely halotolerant and alkaliphilic the Bacillus cereus group in 16S–23S ribosomal DNA intergenic
species isolated from a depth of 1050 m on the Iheya Ridge,” transcribed spacers containing tRNA genes,” Applied and Envi-
FEMS Microbiology Letters, vol. 205, no. 2, pp. 291–297, 2001. ronmental Microbiology, vol. 69, no. 1, pp. 33–40, 2003.
[13] S. Spring, W. Ludwig, M. C. Marquez, A. Ventosa, and K.- [27] E. W. Alm, D. B. Oerther, N. Larsen, D. A. Stahl, and L.
H. Schleifer, “Halobacillus gen. nov., with descriptions of Raskin, “The oligonucleotide probe database,” Applied and
Halobacillus litoralis sp. nov. and Halobacillus trueperi sp. nov., Environmental Microbiology, vol. 62, no. 10, pp. 3557–3559, 1996.
and transfer of Sporosarcina halophila to Halobacillus halophilus [28] S. F. Altschul, W. Gish, W. Miller, E. W. Myers, and D. J. Lipman,
comb. nov,” International Journal of Systematic Bacteriology, vol. “Basic local alignment search tool,” Journal of Molecular Biology,
46, no. 2, pp. 492–496, 1996. vol. 215, no. 3, pp. 403–410, 1990.
[14] H. Schlesner, P. A. Lawson, M. D. Collins et al., “Filobacillus [29] J. D. Thompson, D. G. Higgins, and T. J. Gibson, “CLUSTAL
milensis gen. nov., sp. nov., a new halophilic spore-forming W: improving the sensitivity of progressive multiple sequence
bacterium with Orn-D-Glu-type peptidoglycan,” International alignment through sequence weighting, position-specific gap
Journal of Systematic and Evolutionary Microbiology, vol. 51, no. penalties and weight matrix choice,” Nucleic Acids Research, vol.
2, pp. 425–431, 2001. 22, no. 22, pp. 4673–4680, 1994.
[30] T. H. Jukes and C. R. Cantor, “Evolution of protein molecules,” spirochaete isolated from soda Mono Lake in California,” Inter-
in Mammalian Protein Metabolism, H. N. Munro, Ed., pp. 21– national Journal of Systematic and Evolutionary Microbiology,
123, Academic Press, New York, NY, USA, 1969. vol. 53, no. 3, pp. 815–821, 2003.
[31] S. Kumar, K. Tamura, and M. Nei, “MEGA3: integrated software [46] B. Nowlan, M. S. Dodia, S. P. Singh, and B. K. C. Patel, “Bacillus
for molecular evolutionary genetics analysis and sequence okhensis sp. nov., a halotolerant and alkalitolerant bacterium
alignment,” Briefings in Bioinformatics, vol. 5, no. 2, pp. 150–163, from an Indian saltpan,” International Journal of Systematic and
2004. Evolutionary Microbiology, vol. 56, no. 5, Article ID 63861, pp.
[32] R. G. E. Murray, R. N. Doetsch, and C. F. Robinow, “Determi- 1073–1077, 2006.
native and cytological light microscopy,” in Methods for General [47] A. Ventosa, M. C. Márquez, M. J. Garabito, and D. R. Arahal,
and Molecular Bacteriology, P. Gerhardt, R. G. E. Murray, W. A. “Moderately halophilic gram-positive bacterial diversity in
Wood, and N. R. Krieg, Eds., pp. 21–41, American Society for hypersaline environments,” Extremophiles, vol. 2, no. 3, pp. 297–
Microbiology, Washington, DC, USA, 1994. 304, 1998.
[33] K. Horikoshi, “Production of alkaline enzyme by alkalophilic [48] Y. V. Boltyanskaya, E. N. Detkova, A. N. Shumskii, L. E. Dulov,
microorganisms. Part I alkaline protease produced by Bacillus,” and M. A. Pusheva, “Osmoadaptation in representatives of
Agricultural Biological Chemistry, vol. 35, no. 221, pp. 1407–1414, haloalkaliphilic bacteria from soda lakes,” Mikrobiologiya, vol.
1971. 74, no. 6, pp. 738–744, 2005, English Translation in Microbiol-
[34] M. A. Amoozegar, F. Malekzadeh, and K. A. Malik, “Production ogy, vol. 74, no. 6, pp. 640–645, 2005.
of amylase by newly isolated moderate halophile, Halobacillus [49] V. Gürtler and V. A. Stanisich, “New approaches to typing
sp. strain MA-2,” Journal of Microbiological Methods, vol. 52, no. and identification of bacteria using the 16S–23S rDNA spacer
3, pp. 353–359, 2003. region,” Microbiology, vol. 142, no. 1, pp. 3–16, 1996.
[35] K. Horikoshi, “Production of alkaline enzyme by alkalophilic [50] D. Daffonchio, S. Borin, A. Consolandi, D. Mora, P. L. Man-
microorganisms. Part III alkaline amylase produced by Bacillus achini, and C. Sorlini, “16S-23S rRNA internal transcribed
no. A-40-2,” Agricultural Biological Chemistry, vol. 35, pp. 1783– spacers as molecular markers for the species of the 16S rRNA
1791, 1971. group I of the genus Bacillus,” FEMS Microbiology Letters, vol.
[36] H. Onishi, T. Mori, and S. Takeuchi, “Halophilic nuclease of a 163, no. 2, pp. 229–236, 1998.
moderately halophilic Bacillus sp.: production, purification, and [51] H. Jiang, H. Dong, B. Yu et al., “Microbial response to salinity
characterization,” Applied and Environmental Microbiology, vol. change in Lake Chaka, a hypersaline lake on Tibetan plateau,”
45, no. 1, pp. 24–30, 1983. Environmental Microbiology, vol. 9, no. 10, pp. 2603–2621, 2007.
[37] C. Gutiérrez and C. González, “Method for simultaneous [52] H. Hacěne, F. Rafa, N. Chebhouni et al., “Biodiversity of
detection of proteinase and esterase activities in extremely prokaryotic microflora in El Golea Salt lake, Algerian Sahara,”
halophilic bacteria,” Applied Microbiology, vol. 24, no. 3, pp. 516– Journal of Arid Environments, vol. 58, no. 3, pp. 273–284, 2004.
517, 1972.
[53] J. L. Siefert, M. Larios-Sanz, L. K. Nakamura et al., “Phylogeny
[38] I. Romano, L. Lama, B. Nicolaus, A. Gambacorta, and A.
of marine Bacillus isolates from the Gulf of Mexico,” Current
Giordano, “Bacillus saliphilus sp. nov., isolated from a mineral
Microbiology, vol. 41, no. 2, pp. 84–88, 2000.
pool in Campania, Italy,” International Journal of Systematic and
Evolutionary Microbiology, vol. 55, no. 1, pp. 159–163, 2005. [54] A. M. Sass, B. A. McKew, H. Sass, J. Fichtel, K. N. Timmis, and T.
J. McGenity, “Diversity of Bacillus-like organisms isolated from
[39] V. Torsvik and L. Overas, “Microbial diversity, life strategies, and
deep-sea hypersaline anoxic sediments,” Saline Systems, vol. 4,
adaptation to life in extreme soils,” in Microbiology of Extreme
article 8, 2008.
Soils, Soil Biology, P. Dion and C. S. Nautiyal, Eds., vol. 13,
Springer, Berlin, Germany, 2008. [55] K. Horikoshi and T. Akiba, Alkalophilic Microorganisms: A New
Microbial World, Springer, Heidelberg, Germany, 1982.
[40] H. H. Zahran, “Diversity, adaptation and activity of the bacterial
flora in saline environments,” Biology and Fertility of Soils, vol. [56] T. A. Krulwich and A. A. Guffanti, “Physiology of Acidophilic
25, no. 3, pp. 211–223, 1997. and Alkalophilic Bacteria,” Advances in Microbial Physiology,
[41] M. S. Dodia, Stability and folding of extracellular enzymes from vol. 24, pp. 173–214, 1983.
haloalkaliphilic bacteria [Ph.D. thesis], Saurashtra University, [57] H. Takami and K. Horikoshi, “Analysis of the genome of an
Rajkot, India, 2005. alkaliphilic Bacillus strain from an industrial point of view,”
[42] M. S. Dodia, H. G. Bhimani, C. M. Rawal, R. H. Joshi, and S. P. Extremophiles, vol. 4, no. 2, pp. 99–108, 2000.
Singh, “Salt dependent resistance against chemical denaturation [58] H. Baati, R. Amdouni, N. Gharsallah, A. Sghir, and E. Ammar,
of alkaline protease from a newly isolated haloalkaliphilic “Isolation and characterization of moderately halophilic bacte-
Bacillus sp,” Bioresource Technology, vol. 99, no. 14, pp. 6223– ria from tunisian solar saltern,” Current Microbiology, vol. 60,
6227, 2008. no. 3, pp. 157–161, 2010.
[43] M. S. Dodia, C. M. Rawal, H. G. Bhimani, R. H. Joshi, S. K. [59] A. Guesmi, B. Ettoumi, D. El Hidri et al., “Uneven distribution
Khare, and S. P. Singh, “Purification and stability characteristics of Halobacillus trueperi species in arid natural saline systems of
of an alkaline serine protease from a newly isolated Haloalka- southern Tunisian Sahara,” Microbial Ecology, 2013.
liphilic bacterium sp. AH-6,” Journal of Industrial Microbiology [60] B. A. Lomstein, A. T. Langerhuus, S. D’Hondt, B. B. Jørgensen,
and Biotechnology, vol. 35, no. 2, pp. 121–131, 2008. and A. J. Spivack, “Endospore abundance, microbial growth and
[44] A. Hedi, N. Sadfi, M.-L. Fardeau et al., “Studies on the biodiver- necromass turnover in deep sub-seafloor sediment,” Nature, vol.
sity of halophilic microorganisms isolated from El-Djerid salt 484, no. 7392, pp. 101–104, 2012.
lake (Tunisia) under aerobic conditions,” International Journal [61] A. C. F. Dias, F. D. Andreote, F. Dini-Andreote et al., “Diversity
of Microbiology, vol. 2009, Article ID 731786, 17 pages, 2009. and biotechnological potential of culturable bacteria from
[45] R. B. Hoover, E. V. Pikuta, A. K. Bej et al., “Spirochaeta Brazilian mangrove sediment,” World Journal of Microbiology
americana sp. nov., a new haloalkaliphilic, obligately anaerobic and Biotechnology, vol. 25, no. 7, pp. 1305–1311, 2009.
[62] M. D. Collins, P. A. Lawson, M. Labrenz, B. J. Tindall, N. Weiss, Nesterenkonia sp. AL20,” Enzyme and Microbial Technology, vol.
and P. Hirsch, “Nesterenkonia lacusekhoensis sp. nov., isolated 37, no. 5, pp. 534–540, 2005.
from hypersaline Ekho Lake, East Antarctica, and emended [76] S. Sinsuwan, S. Rodtong, and J. Yongsawatdigul, “Characteriza-
description of the genus Nesterenkonia,” International Journal tion of Ca2+ -activated cell-bound proteinase from Virgibacillus
of Systematic and Evolutionary Microbiology, vol. 52, no. 4, pp. sp. SK37 isolated from fish sauce fermentation,” LWT—Food
1145–1150, 2002. Science and Technology, vol. 41, no. 10, pp. 2166–2174, 2008.
[63] A. W. Duckworth, W. D. Grant, B. E. Jones, and R. Van [77] M. Kamekura and H. Onishi, “Protease formation by a moder-
Steenbergen, “Phylogenetic diversity of soda lake alkaliphiles,” ately halophilic Bacillus strain,” Journal of Applied Microbiology,
FEMS Microbiology Ecology, vol. 19, no. 3, pp. 181–191, 1996. vol. 27, no. 4, pp. 809–810, 1974.
[64] P. Gupta, G. S. N. Reddy, D. Delille, and S. Shivaji, “Arthrobacter
gangotriensis sp. nov. and Arthrobacter kerguelensis sp. nov. from
Antarctica,” International Journal of Systematic and Evolution-
ary Microbiology, vol. 54, no. 6, pp. 2375–2378, 2004.
[65] R. R. De La Haba, D. R. Arahal, M. C. Márquez, and A. Ventosa,
“Phylogenetic relationships within the family Halomonadaceae
based on comparative 23S and 16S rRNA gene sequence
analysis,” International Journal of Systematic and Evolutionary
[66] D. R. Arahal and A. Ventosa, “The family Halomonadaceae,”
in The Prokaryotes: An Evolving Electronic Resource for the
Microbial Community, M. Dworkin, S. Falkow, E. Rosenberg, K.
H. Schleifer, and E. Stackebrandt, Eds., Spring, New York, NY,
USA, 2005.
[67] K. Pedersen, E. Nilsson, J. Arlinger, L. Hallbeck, and A.
O’Neill, “Distribution, diversity and activity of microorganisms
in the hyper-alkaline spring waters of Maqarin in Jordan,”
Extremophiles, vol. 8, no. 2, pp. 151–164, 2004.
[68] P. V. Morais, R. Francisco, R. Branco, A. P. Chung, and
M. S. Da Costa, “Leucobacter chromiireducens sp. nov, and
Leucobacter aridicollis sp. nov., two new species isolated from a
chromium contaminated environment,” Systematic and Applied
[69] S. Labana, G. Pandey, and R. K. Jain, “Desulphurization of
dibenzothiophene and diesel oils by bacteria,” Letters in Applied
[70] G.-J. Chee and H. Takami, “Alternative splicing by participation
of the group II intron ORF in extremely halotolerant and alka-
liphilic Oceanobacillus iheyensis,” Microbes and Environments,
vol. 26, no. 1, pp. 54–60, 2011.
[71] N. Toor, K. S. Keating, S. D. Taylor, and A. M. Pyle, “Crystal
structure of a self-spliced group II intron,” Science, vol. 320, no.
5872, pp. 77–82, 2008.
[72] K. K. Kim, L. Jin, H. C. Yang, and S.-T. Lee, “Halomonas gom-
seomensis sp. nov., Halomonas janggokensis sp. nov., Halomonas
salaria sp. nov. and Halomonas denitrificans sp. nov., moderately
halophilic bacteria isolated from saline water,” International
Journal of Systematic and Evolutionary Microbiology, vol. 57, no.
4, pp. 675–681, 2007.
[73] C. Sánchez-Porro, E. Mellado, C. Bertoldo, G. Antranikian, and
A. Ventosa, “Screening and characterization of the protease CP1
produced by the moderately halophilic bacterium Pseudoal-
teromonas sp. strain CP76,” Extremophiles, vol. 7, no. 3, pp. 221–
228, 2003.
[74] H. R. Karbalaei-Heidari, M. A. Amoozegar, M. Hajighasemi,
A.-A. Ziaee, and A. Ventosa, “Production, optimization and
purification of a novel extracellular protease from the moder-
ately halophilic bacterium Halobacillus karajensis,” Journal of
Industrial Microbiology and Biotechnology, vol. 36, no. 1, pp. 21–
27, 2009.
[75] S. Bakhtiar, R. J. Estiveira, and R. Hatti-Kaul, “Substrate speci-
ficity of alkaline protease from alkaliphilic feather-degrading
http://dx.doi.org/10.1155/2013/958719
Review Article
Microbial Diversity in the Era of Omic Technologies
Sofia Nikolaki and George Tsiamis

Received 30 April 2013; Revised 26 August 2013; Accepted 26 August 2013
Academic Editor: Dimitrios Karpouzas
Copyright © 2013 S. Nikolaki and G. Tsiamis. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.
Human life and activity depends on microorganisms, as they are responsible for providing basic elements of life. Although microbes
have such a key role in sustaining basic functions for all living organisms, very little is known about their biology since only a
small fraction (average 1%) can be cultured under laboratory conditions. This is even more evident when considering that >88%
of all bacterial isolates belong to four bacterial phyla, the Proteobacteria, Firmicutes, Actinobacteria, and Bacteroidetes. Advanced
technologies, developed in the last years, promise to revolutionise the way that we characterize, identify, and study microbial
communities. In this review, we present the most advanced tools that microbial ecologists can use for the study of microbial
communities. Innovative microbial ecological DNA microarrays such as PhyloChip and GeoChip that have been developed for
investigating the composition and function of microbial communities are presented, along with an overview of the next generation
sequencing technologies. Finally, the Single Cell Genomics approach, which can be used for obtaining genomes from uncultured
phyla, is outlined. This tool enables the amplification and sequencing of DNA from single cells obtained directly from environmental
samples and is promising to revolutionise microbiology.
1. Introduction contain a vast array of biochemical transformations, and the

microbial cells have accumulated DNA changes over a period
Microbes are essential for every part of the human life on of billion years of environmental change and evolution [3].
Earth as they are responsible for converting the key elements Human civilization has been greatly improved by the
of life—carbon, nitrogen, oxygen, and sulfur—into forms development of numerous technologies that have their source
accessible to all other living things [1]. Even more interest- in microbes. For instance, they are being used to produce a
ingly, the majority of the photosynthetic capacity of the planet vast array of antibiotics and drugs for clinical use, to
does not depend on plants but on microbes [2]. Microbial remediate pollutants in soil and water, to produce biofuels,
communities are closely associated with plants and animals to enhance and protect agricultural crops, and to ferment
making necessary nutrients, metals, and vitamins available to human foods and even they are used as markers for the
their hosts. For humans, the billions of gut microbes assist detection of diseases [2, 3].
us to digest food, break down toxins, and fight off pathogens Comparative analysis of ribosomal RNA (rRNA) seq-
[2]. Humanity not only depends on microbes for nutritional uences implied that all of the cellular life belonged to one of
and health reasons but also for cleaning up pollutants in the the three domains, namely, Bacteria, Archaea, and Eukarya
environment, such as oil and chemical spills [2]. Amazingly, [4]. This enabled the definition of the major lineages (phyla
these activities are not carried out by individual microbes or divisions) within the three primary domains [5]. Microbes
but by microbial communities that can adapt and excel even are the most diverse group on Earth on the basis of phy-
under extreme environmental changes. These communities logeny and functionality, occupying every conceivable niche.
can live under extreme conditions, at pH level, pressure, The vast majority of these organisms are characterized
and temperatures, in which no other organism can survive. through culture-independent molecular surveys using con-
This has been achieved through numerous strategies that served marker genes like the small subunit ribosomal RNA or
have been developed by microbes for survival. Their genomes more recently the shotgun sequencing (metagenomics) [6,
7]. Although microbes are such an important group, the Table 1: Comparison of the main phylogenetic oligonucleotide and
characterization, identification, and quantification remain an functional gene arrays.
immense challenge even with the conventional molecular
Gene array Probe type No. of probes Analysis provided
tools [8, 9].
The most important revolution in microbial ecology was PhyloChip G2 25-mer oligos 297,851 8,935 OTUs
the use of DNA sequencing in phylogenetic studies and the PhyloChip G3 25-mer oligos 1,100,000 59,959 OTUs
application of this technology to uncultured organisms in the GeoChip 3.0 50-mer oligos 27,812 292 gene families
1980s and the 1990s. This transformed microbiology revealed GeoChip 4.0 50-mer oligos 120,054 539 gene familiesa
that the prokaryotic diversity was vastly underestimated a
GeoChip contains genes targeting human microbiomes in 139 functional
with the current classical cultivation-based techniques [4, gene families with 36,062 probes.
6, 7, 10, 11]. In the last ten years, metagenomic projects
have been combined with next generation sequencing (NGS)
technologies. This has boosted microbial ecology forward in a of life sciences, including environmental microbiology and
very fast pace [1, 12, 13]. Current NGS technologies provide a microbial ecology [30]. Microbial ecological microarrays
throughput, which is at least 100 times that of classical Sanger have been developed for investigating the composition and
sequencing, and the technologies are quickly improving [14– functions of microbial communities in environmental niches
16]. This makes NGS one of the hottest topics in biological [31]. This tool is valuable in bacterial diversity studies since a
sciences. With the assistance of the newly developed disci- single array can contain thousands of DNA sequences with a
pline of metagenomics and the high-throughput sequencing high degree of specificity [32].
technologies, scientists can now unravel the mysteries of the The small subunit ribosomal RNA gene (16S rRNA) is
life of the still uncultured microorganisms [6, 17]. The use the biomarker of choice for characterizing complex microbial
of metagenomic approaches led to the discovery of a large communities [12, 33]. This biomarker is extensively used for
array of new genes and enabled the genome sequence of phylogenetic analysis since it contains highly conserved and
various uncultured microbes [18–20]. This is true for low variable regions that allow a reliable and detailed microbial
to medium complex ecosystems. In highly diverse environ- classification. The most comprehensive POA is the Phy-
ments, metagenomic approaches have not been so successful loChip, which uses the Affymetrix format (Santa Clara, CA)
since assembly is extremely challenging due to the highly [34–36]. The PhyloChip is updated regularly, and currently
heterogeneity of these samples. A way to overcome this two versions of the DNA microarray are available. The devel-
bottleneck is to use a single cell genomics approach that has opment of the second generation PhyloChip (G2) started
been recently developed and allows the genome analysis of in 2002 and it became available in 2006. The PhyloChip
individual community members [21, 22]. G3 is currently available through the Second Genome Inc.
Furthermore, powerful high-throughput tools can be PhyloChip G2 contains 297,851 perfect match (PM) and
provided with the use of phylogenetic oligonucleotide and mismatch (MM) 16S rRNA gene probes for the detection
functional gene arrays [23–25]. The so-called phyloge- of 842 subfamilies or 8,741 taxa, covering 121 bacterial and
netic oligonucleotide arrays (POAs-PhyloChip) use a short archaeal orders [37]. The remaining 209,093 probes are
oligonucleotide design against a phylogenetic marker gene control probes or pathogen detection probes (Table 1) [25, 34,
(such as the 16S rRNA gene). They target polymerase chain 38].
reaction (PCR) amplified rRNA gene fragments, or directly The design of PhyloChip G2 was based on over 30,000
retrieved community rRNA (genes) and can, at least in 16S rRNA gene sequences retrieved from the “Greengenes”
principle, be designed to detect any microorganism [26]. In database in March 2002 [26]. Chimeric sequences were
contrast, functional gene arrays (FGAs) detect selected genes filtered out and, subsequently, they were aligned resulting
or gene families that encode key enzymes that are diagnostic in 8,935 clusters (OTUs), which contained approximately
for a certain metabolic pathway [27–29]. Therefore, these 0–3% sequence divergence. The region selected for probe
arrays are confined to diversity analysis of selected microbial design was flanked on both sides by universally conserved
guilds, while PhyloChips are best suited for detecting changes segments that are used as PCR priming sites. These sites
in the taxonomic composition of microbial populations. can be used to amplify bacterial and/or archaeal genomic
In this review, we present a synopsis on (a) the functional materials. The probe selection strategy was to obtain an
gene arrays and phylogenetic oligonucleotide arrays that are effective set of probes capable of correctly categorizing mixed
currently available to the scientific community, (b) the next amplicons into their proper OTUs. To correctly identify each
generation sequencing technologies with an emphasis on 16S OTU, a set of 11 or more specific 25-mers (probes) was
rRNA amplicon pyrosequencing and, (c) single cell genomics, designed. These probes were prevalent in members of a given
a new approach to study the microbial dark matter. OTU but dissimilar from sequences outside the given OTU.
Probes that were complementary to target sequences were
selected and termed perfect match (PM) probes. Each PM
2. PhyloChip and Functional Gene Arrays probe was paired with a control 25-mer termed mismatching
(MM) probe, identical in all positions except the 13th base.
DNA microarrays are able to detect microbial sequences from The mismatching probe did not contain a central 17-mer
any sample in a parallel and very fast, high throughput way. complementary to sequences in any OTU. The target probe
This technology has found applications across most sectors and MM probe constitute a probe pair analyzed together.
The PhyloChip G2 is arranged in a grid of 732 columns and Functional Gene Arrays (FGAs) are a special type of
rows allowing the placement of 506,944 features. For the DNA microarrays containing probes for key genes involved
design of PhyloChip G3, a similar approach was adapted in microbial functional processes. FGAs are composed of
[23]. The Greengenes database was used and the filtered probes for key genes, involved in microbial functional pro-
rRNA gene sequences were clustered to enable selection of cesses of interest [29, 46, 47]. This type of array allows
perfectly complementary probes representing each sequence the simultaneous examination of many functional genes
of a cluster. Amplicons that were containing 17-mers with unlike PCR-based techniques that limit the number of genes
sequence identity to a cluster were positioned in that cluster. that can be examined at one time [29, 46–48]. FGAs are
This analysis gave rise to 59,959 clusters, each capturing an especially useful for the study of environmental samples
average of 0.5% sequence divergence (Table 1). These clusters since the precise functions are not known due to the lack
were considered as operational taxonomic units (OTUs). The of cultured microorganisms and the high degree of diversity
PhyloChip G3 provides an analysis spanning in 2 domains, and metabolic flexibility that exists in microbial communities
147 phyla, 1,123 classes, 1,219 orders, and 10,993 subfamilies [49].
[23]. GeoChip 3.0 is the most comprehensive DNA microarray
One of the main advantages that the PhyloChip provides currently available for studying microbial communities asso-
is the great sensitivitythat it delivers. A typical 16S PCR ciated with biogeochemical cycling, global climate change,
reaction with a yield of approximately 500 ng of amplicons bioenergy, agriculture, land use, ecosystem management,
provides more than 600 billion sequences, which allows even environmental cleanup and restoration, bioreactor systems,
the less abundant populations to be tracked in addition to the and human health. The design of GeoChip 3.0 involved the
dominant ones. Also, the PhyloChip has been shown to reveal use of 56,990 gene sequences from 292 functional genes
greater diversity within a community when compared with utilizing 27,812 probes. Eight degenerate probes for the 16S
rRNA Sanger sequencing clone libraries due to the placement rRNA gene were used for positive controls, while 672 unique
of the entire gene product on the microarray compared with probes designed from hypothetical genes of seven sequenced
the analysis of up to thousands of individual molecules by genomes of hyperthermophiles were used as negative con-
traditional sequencing methods [39]. The main disadvantage trols [50]. In GeoChip 3.0, 292 key enzymes/genes were
of this technology is that the design of the PhyloChip, and in used to target a variety of microbial mediated processes.
general of the DNA microarrays, does not allow the discovery In brief, a total of 41 enzymes/genes are selected to detect
and the characterization of novel taxa. This is true for the different functional processes of the carbon cycle, and 16
functional gene arrays as well since novel functions cannot enzymes/genes are targeting the nitrogen cycling processes.
be identified through this approach. Four enzymes/genes are used to detect the sulfur cycling
The PhyloChips G2 and G3 have been shown to provide of microbial communities, while three enzymes were tar-
identification resolution at the family to subfamily levels geting the phosphorus cycling of microbial communities.
[34, 37, 40], and they have been used in over 80 publications. The enzymes hydrogenase and cytochrome detect energy
This technology has been used to successfully describe the metabolism processes of microbial communities, while a total
microbial profile in a vast spectrum of complex ecosystems of 41 genes/enzymes cover resistance mechanisms for Ag, Al,
like solar salterns, industrial waste, olive-mill waste marine As, Cd, Co, Cr, Cu, Hg, Ni, Pb, Se, Te and Zn [49].
environments, coral reefs, air craft particulate air, soil, plant In GeoChip, the ability to monitor degradation pathways
tissues, and various human microbiota [23, 25, 35, 36, 39, 41– is also incorporated with a total of 173 genes/enzymes being
44]. used to detect and monitor the degradation of 86 organic
Utilization of the PhyloChip in olive-mill waste revealed contaminants commonly found in the environment [24,
a cultivar-dependent microbial profile [36]. With the imple- 51]. Finally, eleven genes for antibiotic resistance were also
mentation of the PhyloChip, a broader diversity was iden- included.
tified dominated by members of all classes of Proteobacte- Based on GeoChip 3.0, the latest generation, GeoChip
ria, Firmicutes, Bacteroidetes, Chloroflexi, Cyanobacteria and 4.0 in the NimbleGen format has been developed, which not
Actinobacteria, while members of the phyla Acidobacteria, only contains functional categories from GeoChip 3.0 but
Planctomycetes, Gemmatimonadetes, and Verrucomicrobia also includes additional functional categories, such as genes
and the candidate divisions OP3 (Omnitrophica [45]), TM7, involved in stress responses, bacterial phages, and virulence
AD3, marine group A (Marinimicrobia [45]), and SPAM were [50, 52].
minor constituents of the bacterial biota. It can be postulated that future FGAs will be more com-
The PhyloChip was used to associate microbial commu- prehensive for the survey of diverse microbial communities,
nities in aerosols [34]. Samples were collected over a 17- and at the same time they will be more specific for the
week period in San Antonio and Austin. Both sites were detection and identification of microbial communities for
a part of a biosurveillance effort to detect bioterrorism particular ecosystems or functional processes of interest [50].
threads. A diverse group of microorganisms associated with The microbial communities of a Gulf of Mexico coastal
aerosol was detected with the PhyloChip. Sequences similar salt marsh were recently examined during and after the
to or related to potential pathogens including Campylobac- influx of petroleum hydrocarbons following the Deepwater
teraceae, Helicobacteraceae, and Francisella-like and bacteria Horizon oil spill using PhyloChip and GeoChip microar-
related to Bacillus anthracis, Rickettsia, and Clostridium were ray analyses [53]. The abundance of phyla containing
identified. previously described hydrocarbon-degrading bacteria like
Proteobacteria, Bacteroidetes, and Actinobacteria, increased is proportional to the number of nucleotides incorporated.
in hydrocarbon-contaminated sediments and decreased once The FLX instrument provides 100 flows of each nucleotide
the hydrocarbon was below detection. Interestingly, the func- during an 8 h run. This produces an average read length of 250
tional genes involved in the hydrocarbon degradation were nucleotides. Analysis software examines the raw reads using
enriched in hydrocarbon-contaminated sediments. Once the various quality filters for removing poor quality sequences
hydrocarbon concentration was reduced, the detection of and mixed sequences that contain more than one initial
functional genes involved in degrading alkanes, cycloalkanes, DNA fragment per bead. Sequences that do not contain the
aromatic carboxylic acids, chlorinated aromatics, polycyclic initiating TCGA sequence are removed through the quality
aromatics, and other aromatics decreased significantly. control test. The filtered reads yield approximately 100 Mb of
In conclusion, DNA microarrays, that use multiple quality data. It is accepted that FLX reads are of adequate
rRNAs or other phylogenetic markers, can be deployed to length to assemble small genomes such as bacterial and
track variations in population structure and community viral genomes to high quality and contiguity [63]. For the
function over time and space. The use of DNA microarrays specifications of the 454 technology please see Table 2.
based on selected genes (and gene variants) that are involved Solexa developed the second commercial NGS platform.
in interesting processes can be used to assess a community’s Solexa was subsequently acquired by Illumina and is now
ability to perform a collective function, such as biodegra- known by the name Illumina. Roche/454 and Illumina engage
dation of contaminants, and monitor, for example, during the principle of sequencing by synthesis. Illumina uses a solid
bioremediation changes over relevant periods. glass surface that is very similar to a microscope slide, for
capturing individual molecules and bridge PCR to amplify
DNA into small clusters of identical molecules. These clusters
3. Next Generation Sequencing Technologies at the end are sequenced with a strategy that is equivalent
to Sanger sequencing. The difference lies in the use of dye-
The beginning of the modern DNA sequencing era began labelled terminators. 3󸀠 -O-Fluorophore-labeled nucleotides
with the completion of the first draft of the human genome are used as reversible terminators of DNA polymerization.
[54–56]. This was the turning point that led to further inno- This reversible terminator ensures that, in one step, only one
vation and improved development of new advanced strategies nucleotide can be incorporated. After the template is flooded
of high-throughput DNA sequencing. These technologies are with nucleotides and the binding step is accomplished, the
called the next generation sequencing (NGS), and this is unincorporated reagents are washed away and another round
the term that we are going to use throughout this review. of dye-labelled terminators are added [64]. When compared
The main principle in NGS involves DNA molecules that are with 454 sequencing, the Illumina sequencing technology
being sequenced in a massively parallel fashion in a flow cell. achieves (a) much higher throughput (∼1.5 Gbp/run) at the
The sequencing is conducted either in a continuous real-time cost of significantly smaller read lengths and (b) high accu-
manner or in a stepwise iterative process. During this highly racy with error rates of less than 1% (Table 2). The sequencing
parallel process, each clonal template or single molecule is approach is not affected by homopolymer runs to the same
independently sequenced and can be counted among the total extent as the 454 technology [65].
sequences generated [57]. The third commercial NGS technology was developed by
At the moment, six platforms from the second and SOLiD, and it is using ligation to determine sequences. Until
the third generation sequencing technologies are available recently, SOLiD was producing more reads than Illumina.
with most platforms requiring short template DNAs (200– Read lengths for SOLiD are user defined and range between
1000 bp) and with each template containing a forward and 25 and 35 bp, while each sequencing run yields between 2 and
reverse primer binding site [58]. 4 Gb of DNA sequence data [66, 67].
The GS FLX Pyrosequencer utilizes next generation Ion Torrent uses a sequencing strategy similar to the 454,
sequencing technology known as pyrosequencing. The tech- except that (i) hydrogen ions (H+ ) are detected (instead of a
nique was first developed by Pal Nyren and his student pyrophosphatase cascade) and (ii) sequencing chips conform
Mostafa Ronhaghi at the Royal Institute of Technology in to common design and manufacturing standards used for
1996 [59, 60] and is now available through Roche 454 Life commercial microchips [68]. No cameras, lasers or fluores-
Technologies. Pyrosequencing uses beads and starts with a cent dyes are required with the Ion Torrent technology, while
single template molecule, which is amplified with emulsion the common microchip design standards means that low-
PCR (emPCR). Millions of beads after the emPCR are cost manufacturing can be used. Ion Torrent was purchased
loaded onto a picolitre plate, which is specially designed by Life Technologies in 2010. The first early instruments,
so that each well can hold only a single bead. The beads the Ion Personal Genome Machine (PGM), was deployed
are sequenced in a parallel way by flowing pyrosequencing in late 2010, while in September 2012, the Ion Proton was
reagents across the plate. During pyrosequencing, the DNA launched which is capable of producing larger outputs. Field
is synthesized under a complex reaction that includes ATP effect transistors (FETs) are used to measure a change in
sulfurylase, luciferase enzymes, adenosine 5󸀠 phosphosulfate pH in a microwell structure. To increase the throughput, the
and luciferin substrates. These are incorporated in such a way Ion Torrent sequencing chip makes use of a highly dense
that the pyrophosphate group releases upon addition of a microwell array in which each well acts as an individual
nucleotide that results in the production of detectable light DNA polymerization reaction chamber. These chambers
[61, 62]. The amount of light produced with pyrosequencing contain a DNA polymerase and a sequencing fragment.
Table 2: Technical specifications of Next Generation Sequencing platforms.
Life
Helicos biosciences
Platform 454 Illumina technologies Ion torrent Pacific biosciences
heliscope
ABI/SOLID
Year of availability 2005 2006 2006 2007 2010 2010
Sequencing length 200–700 bp Up to 150 bp 35–50 bp 25–55 bp ∼200 bp 1500 bp
20–50 Mb on 314 chip
Sequence yield
700 Mb 2–600 Gba 120 Gb 35 Gb 100–200 Mb on 316 100 Mb
per run
chip Gb on 318 chip
Run time 23 h 27 h–11 days 7-8 days 3–6 days 2h 2h
Single Molecule Real
Time (SMRT)
True Single Molecule sequencing dyes that
emPCR,
Polonies, Sequencing (tSMS) are phospholinked
emPCR, ligation with
Technology cleavable dye Single base, reversible emPCR, H+ detection to the nucleotide,
pyrosequencing cleavable dye
terminators dye terminator very sensitive
terminators
extension reactions fluorescent
detection in zero
mode waveguides
a
2 Gb for the MiSeq and 600 Gb for the HiSeq2000.
Below this layer of microwells an ion-sensitive layer is a short distance and excite the fluorophores attached to
present, followed by a sublayer composed of a highly dense those nucleotides that are in the vicinity of the polymerase
FET array aligned with the microwell array. Sequential at the bottom of the well. As each base is incorporated, a
cycling of the four nucleotides into the microwells enables distinctive pulse of fluorescence is detected in real time. The
primary sequence resolution since the FET detector senses first instruments were deployed in late 2010. The low cost
the change in pH created during nucleotide incorporation per experiment and fast run times have generated much
and converts this signal to a recordable voltage change. While enthusiasm for this platform, especially among investors.
this method of ion sensing-based sequencing by synthesis Although high accuracy can be achieved through circular
offers great potential to reduce the cost of sequencing, there consensus sequencing, which involves sequencing shorter
are several limitations with regards to sequencing complete templates multiple times, this instrument generates single-
genomes. The Ion PGM was mainly targeting small genomes pass reads that average less than 85% nucleotide accuracy
given the output capability of the instrument (currently up [69].
to 1 Gb). The newly launched Ion Proton uses larger chips For each technology, there is a trade-off between advan-
with higher densities and is said to be able to generate 10 Gb tages and disadvantages. The 454 technology delivers the
per run (Table 2). These characteristics make the technology longest read length but with the lowest throughput (8 MB/h
suitable for exome and whole genome sequencing. Currently, during a 9 h run—Table 2) and suffers from errors in
the short read lengths place a burden on the reassembly homopolymeric tracts, even when assembles are at high
process and limit the assembly of de novo sequencing projects coverage. MiSeq (Illumina) generates the highest throughput
due to an inability to read through long repetitive regions in per run and lowest error rate of the instruments but delivers
the genome. Finally, error accumulation can occur if reaction shorter read lengths than those of the 454. Ion Torrent cur-
wells are not properly purged between reaction steps, with an rently produces short reads and the worst performance with
error rate of 1.78% being reported for Ion Torrent [63, 69]. homopolymers, although the new chemistry has improved
The first commercial single-molecule sequencer (3rd performance. Ion Torrent delivers the fastest throughput (80–
generation) has been developed by Helicos. The high cost of 100 Mb/h—Table 2) and shortest run time of an approx. 3 h.
the instruments and short read lengths unfortunately limited This platform has also shown the greatest improvement in
adoption of this platform, and at the moment Helicos no performance in recent months [69, 71].
longer sells instruments; instead it conducts sequencing via
a service centre model.
PacBio has developed an instrument that sequences
individual DNA molecules in real time [70]. Individual 4. 16S rRNA Pyrosequencing and
DNA polymerases are attached to the bottom of 50 nm wide Hypervariable Regions
wells that are termed zero-mode waveguides (ZMWs). Each
polymerase is allowed to carry out second strand DNA syn- Using the 16S ribosomal RNA gene as a phylogenetic marker
thesis in the presence of 𝛾-phosphate fluorescently-labeled was a real breakthrough for microbial ecology studies, with
nucleotides. The width of the ZMW is such that light cannot several culture-independent methods being developed since
proliferate through the waveguide, but energy can penetrate Pace et al. [72] proposed the direct cloning of environmental
Table 3: Oligonucleotide primers that can be used for 16S rRNA
530F–805R
120 variable region PCR amplification and sequencing of bacterial 16S
rRNA genes.
Number of polymorphic sites
100
Primer Sequence 5󸀠 to 3󸀠 Reference
338F–530R
1046F–1220R
967F–1046R
80 8F AGAGTTTGATCCTGGCTCAG [133]
27F AGAGTTTGATCMTGGCTCAG [134]
60 338R GCTGCCTCCCGTAGGAGT [135]
338F ACTCCTACGGGAGGCAGC [136, 137]
40
530R AATACGGAGGGTGCAAGCGT [136, 137]
20 967F–1220R 530F ACGCTTGCACCCTCCGTATT [136, 137]
8F or 27F–338R 805F–1046R
347F–803R 1046F–1392R 805R GGATTAGATACCCTGGTAGTC [136, 137]
0 805F GACTACCAGGGTATCTAATCC [136, 138, 139]
V1 V2 V3 V4 V5 V6 V7 V8
967F CAACGCGAAGAACCTTACC [138, 139]
16S rRNA gene
1046F ACAGCCATGCAGCACCT [138]
Figure 1: Graphical presentation of the variable regions within the 1046R AGGTGCTGCATGGCTGT [138, 139]
16S rRNA gene and location of corresponding primer pairs that 1220R GTAGCRCGTGTGTMGCCC [138, 139]
can be deployed for specific region amplification. Variable regions 1392R ACGGGCGGTGTGTRC [134]
presented exclude poorly supported areas, and for this reason the
V9 region is not presented.
(Figure 1). Hypervariable regions of the 16S rRNA gene are

DNA. PCR-based molecular techniques enabled the descrip- flanked by conserved sequences (Figure 1) and this enables
tion of microbial taxonomic diversity: (a) by means of finger- the design of “universal” PCR primers that can amplify 16S
printing methods, which separate rDNA fragments according rRNA hypervariable regions from a large number of different
to their length and/or nucleotide composition like denatur- bacteria species [39, 80, 81] (see Table 3).
ing/temperature gradient gel electrophoresis (DGGE/TGGE) The most critical step for accurate characterization of
[73], restriction fragment length polymorphisms (RFLP) bacterial and archaeal communities using rDNA amplicon
[74], terminal restriction fragment length polymorphism analysis is the choice of primers. Using suboptimal primers
(T-RFLP) [75], single-strand conformation polymorphism pairs will lead to underrepresentation or underselection
(SSCP) [76], and automated rRNA intergenic spacer analysis against single species or even whole groups which can lead
(ARISA); (b) by microscopy using FISH (fluorescence in situ to questionable biological conclusions [40]. Different hyper-
hybridization) and derived methods (CARD-FISH, MAR- variable regions evolve at different rates, and different species
FISH); and (c) by cloning 16S rRNA gene fragments and of the same genus may be similar in some hypervariable
subsequently sequencing the clones following the Sanger regions and more divergent in others [82]. Primer bias occurs
sequencing method. It is true that fingerprinting technologies when the selected primers do not anneal to the DNA from
enable the processing of many samples, but they are inad- all members of the community equally, but preferentially
equate for taxonomic identification and suffer from a lack amplify certain taxonomic groups [6]. This can lead to the
of resolution. Finally, cloning/sequencing and FISH are not failure in detecting some bacterial/archaeal species since in
compatible with high-throughput approaches. rare biospheres bacteria and archaea can never be identified
Cloning and sequencing of the 16S ribosomal RNA gene if the employed primers are not applicable to them. This will
using conserved broad-range PCR primers was and still is the lead to incomplete surveys in metagenomic studies [83].
most common molecular approach for estimating the micro- It has been shown that sequences of 500–700 bp are
bial diversity. But, with the development of NGS technologies, required for phylogenetic discrimination at the species levels
direct sequencing of PCR amplicons became feasible [77, [84, 85]. With the NGS technologies, fragments of up to
78]. It is true that the rapid development of sequencing 700 bp are being sequenced regularly with investigations
technologies has opened a new dimension in biodiversity supporting that use of the V1, V2, and V3 regions for deep
analysis, but the most critical step for an accurate rDNA sequencing and characterization of bacterial and archaeal
amplicon analysis remains the correct choice of primers and sequences [86, 87]. Others suggest that regions generated
the hypervariable regions that will be targeted [78, 79]. using primer pairs 8F-338R and 967F-1046R for V6 overes-
The 16S rRNA gene in bacteria is comprised of inter- timate species richness and promote the V4–V6 generated
spersed conserved and variable sequences including eight (8) using primer pairs 530F-805R, 805F-1046R and 967F-1220R
hypervariable regions (V1–V8) (Figure 1). The eight hyper- as the most appropriate [82, 88] (Table 3). Also, frag-
variable regions spanned nucleotides 69–99, 137–242, 433– ments encompassing the V3, V7, and V7+V8 hypervariable
497, 576–682, 822–879, 986–1043, 1117–1173, and 1243–1294 regions (generated using primer pairs 338F-530R, 1046F-
for V1 through V8, respectively [23, 80]. These hypervariable 1220R, and 1046F-1392R) underestimated species richness
regions range in size from approximately 50 to 100 bases in [82] (Table 3). Recent studies demonstrated that the V7-
length, while sequences differ with respect to variation and V8 fragments achieve better microbial community coverage
corresponding utility for universal microbial identification from a complex ecosystem [88]. It is highly recommended to
use primers that are targeting two regions of the 16S rRNA the structuring of understudied or highly divergent pop-
gene in all deep-sequencing efforts when trying to charac- ulations. For instance, new putative clades belonging to
terize highly heterogeneous microbial communities [81, 88], Mamiellophyceae, Foraminifera, Dictyochophyceae, and Eu-
although good representation of a microbial community has glenida were recently detected in eight freshwater ecosystems
been achieved by targeting single hypervariable regions [89]. using rDNA pyrotag data [102].
Air microbial diversity has been recently studied using
pyrosequencing technologies in the New York City subway
5. NGS Amplicon Sequencing in platforms and associated sites [103]. Eukaryotic diversity
Microbial Ecology was mainly fungal, dominated by organisms of types asso-
ciated with wood rot. Bacterial diversity was dominated
In the recent years, mass sequencing of environmental sam- by human skin bacterial species including Staphylococcus
ples has been the leading approach for microbial ecology epidermidis (the most abundant and prevalent commensal of
studies. Irrespective of the ecosystem studied, the vast major- the human integument), S. hominis, S. cohnii, S. caprae, and
ity of the studies deployed the 454 pyrosequencing platform, S. haemolyticus, while no organisms of public health concern
although Illumina-based studies are in the increase. were identified [103].
Soil bacterial diversity was examined using NGS tech-
nologies, and this approach revealed that the agricultural
management of the soil can influence the diversity of bacteria 6. Single Cell Genomics
and archaea [90–92]. The pH is the principal diversity driver
for both Bacteria and Archaea [92–94] with the Archaea Recent estimates predict that the number of microbial species
being highly correlated only with pH. The fungal community in the world are well into millions, and based on the rRNA
composition was less strongly affected by pH [93]. Soil fungal phylogeny, these species fall within approximately 60 major
diversity was the focus of other studies in forest and agricul- lines of descent within the bacterial and archaeal domains
tural ecosystems. Analysis using ITS amplicons revealed that [33, 104]. From the 60 major lines, at least half have no
in forests most species belong to the Dikarya subkingdom cultivated representatives, and they are called “candidate”
(Ascomycota and Basidiomycota), with the Agaricomycetes phyla. Even from the phyla with culturable representatives,
being the most dominant fungal class [95]. In agricultural 88% belong to only four bacterial phyla, the Proteobacteria,
ecosystems, it has been revealed that the diversity of the Firmicutes, Actinobacteria, and Bacteroidetes. One approach
fungal community declines with soil depth with communities for sequencing candidate phyla is by deploying a metage-
forming distinct groups among the strata [96]. nomic approach, thus obtaining genome sequences from the
Marine environments have been used in studies with microbial dark mater through direct sequencing of DNA
NGS technologies. Analysis of the 18S rRNA gene identi- from microbial communities [3]. The use of such approach
fied members from all six eukaryotic supergroups. It also enabled the draft to complete genome recovery from candi-
revealed that the eukaryotic microbiota was dominated by date divisions: (a) WWE1/Cloacimonetes (Wastewater Evry
dinoflagellates and close relatives which demonstrates the 1) with the Cloacamonas acidaminovorans, (b) NC10 with
importance of this group to marine ecosystems [97]. The use the Methylomirabilis oxyfera, (c) OP1/Acetothermia with the
of 18S rDNA pyrosequencing enabled the characterization of Acetothermum autotrophicum, and (d) from Korarchaeota
uncultured eukaryotes like flagellates, which are known as the Korachaeum cryptofilum [105–108]. With the develop-
MArine STramenopiles (MAST) [98]. Finally, the use of 16S ment of new bioinformatic tools in combination with deep
rRNA tag pyrosequencing analysis from a temperate marine metagenomic sequencing low abundant genomes have been
coastal site over a period of 6 years, suggested that seasonal recovered including members of candidate phyla like OP11
changes in environmental variables are more important than (Microgenomates), OD1 (Parcubacteria), and GNO2 (Gra-
trophic interactions [99]. cilibacteria) [109].
NGS technologies have been applied in freshwater envi- Another approach that can be deployed in order to
ronmental samples. Monchy et al. [100] used cloning/ obtain genomic data from candidate divisions is the single
sequencing and SSU tag pyrosequencing to study the fungal cell genomics technique. With this approach, cells from any
diversity in freshwater lake ecosystems. This study indi- environmental sample can be isolated, and after amplification
cated that geographical, physical, and chemical factors of the DNA can be sequenced [110]. In more detail, almost
the biotope influence the species community structure and any environmental sample can be processed immediately
spatial variability. In a very interesting study by Loga- or stored in the presence of betaine or glycerol so that
res et al. [101], the bacterioplankton communities in a the integrity of the cells be preserved [111]. The next step
unique system of coastal Antarctic lakes exposed to pro- involves cell separation and this is currently achieved with
gressive long-term environmental change was examined the use of fluorescence-activated cell sorting (FACS) [111–
using 454 pyrosequencing of the 16S rDNA gene (V3- 113]. In comparison to micromanipulation, FACS minimizes
V4 regions). Progressive long-term salinity change appears the risk of contamination since a few picoliters of sample
to have promoted the diversification of bacterioplankton are sorted each time. Cell lysis follows (Figure 2) with the
communities by modifying the composition of ancestral most effective method being the alkaline lysis [114]. Whole
communities and by allowing the establishment of new Genome Amplification can be achieved through the Multiple
taxa [101]. NGS technologies are more robust in describing Displacement Amplification (MDA) approach producing
Complex environmental sample

Metabolic profile
Functional capacity
Evolution
Isolation of single cells

using flow cytometry or Assembly, annotation
microfluids
Genomic sequencing
Lysis
Multiple displacement PCR and 16S rRNA

amplification sequencing
Figure 2: General overview of the single cell genomics approach.
long, overlapping amplicons that can be used with the NGS In a recent study by Rinke et al. [45], an SCG approach
technologies. In silico DNA normalization and specialized was deployed targeting 201 uncultivated archaeal and bacte-
software have been developed to counteract the drawbacks of rial cells that belong to 29 major mostly uncharted branches
MDA [111, 115]. Before the genome sequencing of the Single of the tree of life, the so-called “microbial dark matter.”
Amplified Genomes (SAGs) using NGS technologies, a PCR Sequencing of 201 single amplified genomes enabled the reso-
step can be included for screening purposes. Amplification lution of many intra- and interphylum-level relationships and
and Sanger sequencing of the 16S rRNA enables phylogenetic enabled the characterization of new superphyla. The first one,
characterization of the SAGs (Figure 2). The recovery of Terrabacteria, comprises terrestrial bacterial phyla of Acti-
genomic information from single cells varies from 0% up to nobacteria, Cyanobacteria, Thermi (Deinococcus-Thermus),
complete genomes and depends on the intrinsic properties of Chloroflexi, Firmicutes, and Armatimonadetes. This super-
the cell and on the components of the SCG pipeline deployed. phylum comprises monoderm (single membrane) and atyp-
For instance, the above described approach has been suc- ical lineages. The second superphylum is the Patescribacteria
cessful to target members of the candidate phyla TM7, OP11 (patesco (Latin), meaning bare), which reflects the reduced
(Microgenomates) and Poribacteria [116–118] and led to the metabolic capacities of these lineages. Finally, the superphy-
recovery of genomic sequences of microorganisms from lum DPANN was identified composed by the Diapherotrites
several deep-branching phylogenetic groups with no cultured (pMC2A384), Parvarchaeota, Aenigmarchaeota (DSEG), and
representatives. Other examples include Picobiliphytes and Nanohaloarchaeota. Finally, substantive genomic data for 11
divergent groups of aquatic Proteobacteria, Flavobacteria, and bacterial candidate divisions and several highly divergent
Archaea [111, 112, 119–122]. archaeal groups related to Nanoarchaeota were resolved
With the single cell genomics (SCG) approach, for, first as monophyletic groups. This enabled the proposition of
time, a direct link between phylogenetic and metabolic names for these candidate divisions based on their inferred
markers of uncultured bacteria and archaea is possible. A physiology and distinguishing properties (Table 4).
recent example is the discovery of chemolithoautotrophical SCG generates a whole new way for exploiting bacterial
pathways in uncultured Proteobacteria [111], which reconcil- diversity since, to date, biotechnological applications rely
iate the current discrepancies in dark ocean’s carbon budget. almost exclusively on the part of the microbial world that can
SCG is also capable of producing reference genomes of the be cultured, that is, less than 1% of the microbial diversity.
uncultured microorganisms, enabling the study of complex Although metagenomic-based bioprospecting provides an
ecosystems. For instance, Single Cell Genomics have been alternative [124–126], the main advantage that the single
deployed to investigate biogeographic distribution of uncul- cell genomics approach offers is that rather than individual
tured marine Flavobacteria, and marine bacterioplankton genes of the uncultured microorganisms the whole genome
that were involved in the degradation of hydrocarbons during is sequenced. This approach enables the construction of
the Deepwater Horizon oil spill [112, 123]. complex metabolic pathways, ensuring that all discovered
Table 4: Proposed names for candidate phyla and associated superphyla (adapted from Rinke et al. [45]).
Superphylum Candidate phylum Proposed name Etymology

PVC Omnitrophica Omnitrophus, eating all Om.ni.tro’phi.ca. A higher taxonomic unit
OP3
comprising the genus Omnitrophus
Marinimicrobium, a marine microbe
SAR406 (Marine Marinimicrobia Ma.ri.ni.mi.cro’.bi.a. A higher taxonomic unit comprising the genus
Group A) Marinimicrobium
Latescibacter a hiding small rod

FCB
Latescibacteria La.tes.ci.bac.te’ri.a. A higher taxonomic unit comprising the genus
WS3
Latescibacter
Cloacimonas a unit from a sewer

WW1 Cloacimonetes Clo.a.ci.mo.ne’tes. A higher taxonomic unit comprising the genus
Cloacimonas.
Aminicenans a (bacterium) degrading amino acids
OP8 Aminicenantes A.mi.ni.ce.nan’tes. A higher taxonomic unit comprising the genus
Aminicenans.
Microgenomatus; an organism with a small genome size (∼1 Mbp)
OP11 Microgenomates A higher taxonomic unit comprising the genus Microgenomatus
Paceibacter Pace’s bacterium

norman’i.i. N.L. gen. N. of Norman, referring to Norman Pace (Norman
Patescibacteria Richard Pace, Jr. is an American biochemist, Distinguished Professor of
Parcubacteria Molecular, Cellular and Developmental Biology at the University of Colorado,
OD1
and principal investigator at the Pace lab)
-parcus (lat.), thrifty
A higher taxonomic unit comprising the genus Paceibacter
Gracilibacteria -gracilis (lat.), slim, slender, slight, meager, simple

GN02 (BD1-5)
A higher taxonomic unit comprising the genus Altimarinus
A.tri.bac.te’ri.a. N.L. n.
OP9 Atribacteria A higher taxonomic unit comprising the genus Caldatribacterium
Ca.lesc.a.man’tes. L. v. calesco, to become warm, grow hot; L. v. amo, to love,

Calescamantes N.L. n. Calescamantes heat lovers
EM19
A higher taxonomic unit comprising the genus Calescibacterium
A.er.o.pho’bus. Gr. n. aer, air; Gr. adj. phobos, fear. N.L. n. Aerophobus,
Aerophobetes fearing of air (i.e., oxygen).
CD12 (BHI80-139)
A higher taxonomic unit comprising the genus Aerophobus
Hydrogenedens a hydrogen consumer te.re.phtha’li.cus. N.L. n. (acidum)

terephthalicum, terephthalic acid. N.L. adj. terephthalicus, referring to the
NKB19 Hydrogenedentes environment of isolation, a terephthalate reactor.
A higher taxonomic unit comprising the genus Hydrogenedens.
Acetothermus indicates a vinegar organism living in hot places and

“autotrophicum” (au.to.tro’phi.cum. Gr. pron.autos self; Gr. adj. trophikos
Acetothermia nursing, tending or feeding; N.L. neut. adj. autotrophicum selfnursing or
OP1
self-feeding).
A.ce.to.ther’mi.a. A highe taxonomic unit comprising the genus Acetothermum
Fer.vi.do.bac’ter. L. adj. fervidus, hot, steaming; -i-connecting vowel; N.L. n.

bacter, a rod; N.L. n. Fervidibacter a hot rod. sac.cha’ri. N.L. n. saccharum,
Oct-Spa1-106 Fervidibacteria sugar; N.L. gen. n. sacchari, of sugar
Fer.vi.di.bac.te’ri.a. A higher taxonomic unit comprising the genus
Fervidibacter
Table 4: Continued.
Superphylum Candidate phylum Proposed name Etymology
(Dia.phe.ro.tri’tes. Gr. v. diaphero, to differ; Gr. adj. trı́tos the third; N.L. n.
Diapherotrites Diapherotrites the 3rd HSM group discovered)
pMC2A384
A higher taxonomic unit comprising the genus Iainarchaeum
Parv.ar.chae’a. A higher taxonomic unit comprising the genera Parvarchaeum

ARMAN group Parvarchaeota and Micrarchaeum
DPANN
Ae.nig.mar.chae’a. A higher taxonomic unit comprising the genus
DSEG Aenigmarchaeota Aenigmarchaeum
Na.no.ha.lo.ar.chae.o’ta. Gr. n. nanos, dwarf; Gr. n. hal, -los, salt; Gr. adj.
Nanohaloarchaeota archaios, old; N.L. suffix -ota ending to design a phylum; N.L. n.
Nanohaloarchaeota, small salt-loving Archaeota. A higher taxonomic unit
comprising the genus Nanosalina
genes are originating from the same cell. Some early examples technique, preliminary to more detailed metagenomic stud-
of biotechnological exploitations through the use of the ies, rather than the final stage in ecological analysis.
single cell genomic technique include recoveries of polyketide SCG provides the ability to read genetic information at
biosynthesis pathways from sponge symbionts [118, 127] and the basic level of biological organization. The capacity to
the discovery of uncultured microorganisms that degrade sequence any genomic region of an uncultured cell provides
specific macromolecules and fix CO2 through chemoautotro- for the first time a direct link between phylogenetic and
phy [111, 128, 129]. metabolic markers. The power of SCG has been demonstrated
by revealing metabolic features and in situ interactions that
have not been able to be characterized before with any
other molecular approach. SCG offers a unique opportunity
7. Conclusion to obtain genomic information from major uncultivated
microbial lineages.
The study of the microbial diversity is important for under- Finally, we believe that new approaches exploiting NGS
standing the link between diversity, community structure, technologies and Single Cell Genomics jointly will be devel-
and function. The most advanced technologies in this quest oped targeting genomes and associating them with quan-
are DNA microarrays, NGS, and single cell genomics. DNA titative measurements. This approach can target all active
microarrays provide a fast and high-throughput approach members (abundant and rare) of a given ecosystem, measure
for the parallel detection of microbes from any sample. the transcribed genes, and obtain the full genome.
The most comprehensive phylogenetic oligonucleotide array
is the PhyloChip, which uses the Affymetrix format with
a current analysis of 59,959 OTUs. GeoChip 3.0 is an Acknowledgments
advanced Functional Gene Array with a resolution of 292 This work was partially supported by EU PEOPLE-2012-
key enzymes/genes containing more than 28,000 probes. IAPP 324349, by the Hellenic Ministry of Education ARCHI-
These DNA microarrays revolutionized the way that complex MEDES 409-12, and by intramural funds of the University of
microbial communities are being studied. The development Patras to George Tsiamis.
of the new-generation sequencing technologies challenged
the use of DNA microarrays in microbial community studies.
It appears that in low to medium complexity ecosystems use References
of the NGS technologies is the most promising approach. In
highly complex ecosystems, the sequence-based technologies [1] J. C. Venter, K. Remington, J. F. Heidelberg et al., “Environmen-
suffer from random sampling, under sampling, and rRNA tal genome shotgun sequencing of the Sargasso Sea,” Science,
vol. 304, no. 5667, pp. 66–74, 2004.
interference [130–132]. DNA microarrays like PhyloChip and
GeoChip are excellent tools when highly complex ecosys- [2] National Research Council (US) Committee on Metagenomics:
tems are examined due to their unique features like (a) Challenges and Functional Applications, The New Science of
rapid output, (b) quick sample preparation, (c) community Metagenomics: Revealing the Secrets of Our Microbial Planet,
comparison, (d) quick data analysis, and (d) resistance to National Academies Press, Washington, DC, USA, 2007.
contaminants. [3] J. Handelsman, “Metagenomics: application of genomics to
NGS technologies will continue to improve both accuracy uncultured microorganisms,” Microbiology and Molecular Biol-
and throughput with benchtop sequencers becoming the ogy Reviews, vol. 68, no. 4, pp. 669–685, 2004.
standard equipment in individual labs. We believe that ampli- [4] C. R. Woese and G. E. Fox, “Phylogenetic structure of the
con analysis will become in the near future a quick screening prokaryotic domain: the primary kingdoms,” Proceedings of the
National Academy of Sciences of the United States of America, polymerase and multiply-primed rolling circle amplification,”
vol. 74, no. 11, pp. 5088–5090, 1977. Genome Research, vol. 11, no. 6, pp. 1095–1099, 2001.
[5] C. R. Woese, O. Kandler, and M. L. Wheelis, “Towards a [22] R. S. Lasken, “Single-cell genomic sequencing using multiple
natural system of organisms: proposal for the domains Archaea, displacement amplification,” Current Opinion in Microbiology,
Bacteria, and Eucarya,” Proceedings of the National Academy of vol. 10, no. 5, pp. 510–516, 2007.
Sciences of the United States of America, vol. 87, no. 12, pp. 4576– [23] T. C. Hazen, E. A. Dubinsky, T. Z. DeSantis et al., “Deep-sea oil
4579, 1990. plume enriches indigenous oil-degrading bacteria,” Science, vol.
[6] J. Rajendhran and P. Gunasekaran, “Microbial phylogeny and 330, no. 6001, pp. 204–208, 2010.
diversity: small subunit ribosomal RNA sequence analysis and [24] Z. He, Y. Deng, J. D. Van Nostrand et al., “GeoChip 3.0 as
beyond,” Microbiological Research, vol. 166, no. 2, pp. 99–110, a high-throughput tool for analyzing microbial community
2011. composition, structure and functional activity,” ISME Journal,
[7] J. A. Gilbert and C. L. Dupont, “Microbial metagenomics: vol. 4, no. 9, pp. 1167–1179, 2010.
beyond the genome,” Annual Review of Marine Science, vol. 3, [25] C. A. Kellogg, Y. M. Piceno, L. M. Tom, T. Z. DeSantis, D. G.
pp. 347–371, 2011. Zawada, and G. L. Andersen, “PhyloChip microarray compar-
[8] D. Wu, P. Hugenholtz, K. Mavromatis et al., “A phylogeny- ison of sampling methods used for coral microbial ecology,”
driven genomic encyclopaedia of Bacteria and Archaea,” Journal of Microbiological Methods, vol. 88, no. 1, pp. 103–109,
Nature, vol. 462, no. 7276, pp. 1056–1060, 2009. 2012.
[9] N. C. Kyrpides, “Fifteen years of microbial genomics: meeting [26] T. Z. DeSantis, I. Dubosarskiy, S. R. Murray, and G. L. Andersen,
the challenges and fulfilling the dream,” Nature Biotechnology, “Comprehensive aligned sequence construction for automated
vol. 27, no. 7, pp. 627–632, 2009. design of effective probes (CASCADE-P) using 16S rDNA,”
Bioinformatics, vol. 19, no. 12, pp. 1461–1468, 2003.
[10] D. J. Lane, B. Pace, and G. J. Olsen, “Rapid determination
of 16S ribosomal RNA sequences for phylogenetic analyses,” [27] L. Wu, D. K. Thompson, G. Li, R. A. Hurt, J. M. Tiedje, and J.
Proceedings of the National Academy of Sciences of the United Zhou, “Development and evaluation of functional gene arrays
States of America, vol. 82, no. 20, pp. 6955–6959, 1985. for detection of selected genes in the environment,” Applied and
Environmental Microbiology, vol. 67, no. 12, pp. 5780–5790, 2001.
[11] J. L. Stein, T. L. Marsh, K. Y. Wu, H. Shizuya, and E. F. Delong,
“Characterization of uncultivated prokaryotes: isolation and [28] J. Adamczyk, M. Hesselsoe, N. Iversen et al., “The isotope array,
analysis of a 40-kilobase-pair genome fragment from a plank- a new tool that employs substrate-mediated labeling of rrna for
tonic marine archaeon,” Journal of Bacteriology, vol. 178, no. 3, determination of microbial community structure and function,”
pp. 591–599, 1996. Applied and Environmental Microbiology, vol. 69, no. 11, pp.
6875–6887, 2003.
[12] S. G. Tringe and P. Hugenholtz, “A renaissance for the pioneer-
ing 16S rRNA gene,” Current Opinion in Microbiology, vol. 11, no. [29] P. Dennis, E. A. Edwards, S. N. Liss, and R. Fulthorpe, “Monitor-
5, pp. 442–446, 2008. ing gene expression in mixed microbial communities by using
DNA microarrays,” Applied and Environmental Microbiology,
[13] D. B. Rusch, A. L. Halpern, G. Sutton et al., “The Sorcerer II vol. 69, no. 2, pp. 769–778, 2003.
Global Ocean Sampling expedition: northwest Atlantic through
[30] L. Bodrossy and A. Sessitsch, “Oligonucleotide microarrays in
eastern tropical Pacific,” PLoS Biology, vol. 5, no. 3, article e77,
microbial diagnostics,” Current Opinion in Microbiology, vol. 7,
2007.
no. 3, pp. 245–254, 2004.
[14] R. Drmanac, A. B. Sparks, M. J. Callow et al., “Human genome
[31] T. Majtán, G. Bukovská, and J. Timko, “DNA microarrays—
sequencing using unchained base reads on self-assembling
techniques and applications in microbial systems,” Folia Micro-
DNA nanoarrays,” Science, vol. 327, no. 5961, pp. 78–81, 2010.
biologica, vol. 49, no. 6, pp. 635–664, 2004.
[15] E. R. Mardis, “The impact of next-generation sequencing
[32] V. Torsvik and L. Øvreås, “Microbial diversity and function in
technology on genetics,” Trends in Genetics, vol. 24, no. 3, pp.
soil: from genes to ecosystems,” Current Opinion in Microbiol-
133–141, 2008.
ogy, vol. 5, no. 3, pp. 240–245, 2002.
[16] M. L. Metzker, “Sequencing technologies the next generation,” [33] P. Hugenholtz and N. C. Kyrpides, “A changing of the guard:
Nature Reviews Genetics, vol. 11, no. 1, pp. 31–46, 2010.
genomics update,” Environmental Microbiology, vol. 11, no. 3, pp.
[17] J. Handelsman, M. R. Rondon, S. F. Brady, J. Clardy, and R. 551–553, 2009.
M. Goodman, “Molecular biological access to the chemistry of [34] E. L. Brodie, T. Z. DeSantis, J. P. Moberg Parker, I. X. Zubietta, Y.
unknown soil microbes: a new frontier for natural products,” M. Piceno, and G. L. Andersen, “Urban aerosols harbor diverse
Chemistry and Biology, vol. 5, no. 10, pp. R245–R249, 1998. and dynamic bacterial populations,” Proceedings of the National
[18] G. W. Tyson, J. Chapman, P. Hugenholtz et al., “Community Academy of Sciences of the United States of America, vol. 104, no.
structure and metabolism through reconstruction of microbial 1, pp. 299–304, 2007.
genomes from the environment,” Nature, vol. 428, no. 6978, pp. [35] G. Tsiamis, K. Katsaveli, S. Ntougias et al., “Prokaryotic com-
37–43, 2004. munity profiles at different operational stages of a Greek solar
[19] D. Chivian, E. L. Brodie, E. J. Alm et al., “Environmental ge- saltern,” Research in Microbiology, vol. 159, no. 9-10, pp. 609–627,
nomics reveals a single-species ecosystem deep within earth,” 2008.
Science, vol. 322, no. 5899, pp. 275–278, 2008. [36] G. Tsiamis, G. Tzagkaraki, A. Chamalaki et al., “Olive-mill
[20] T. Woyke, H. Teeling, N. N. Ivanova et al., “Symbiosis insights wastewater bacterial communities display a cultivar specific
through metagenomic analysis of a microbial consortium,” profile,” Current Microbiology, vol. 64, no. 2, pp. 197–203, 2012.
Nature, vol. 443, no. 7114, pp. 950–955, 2006. [37] E. L. Brodie, T. Z. DeSantis, D. C. Joyner et al., “Application
[21] F. B. Dean, J. R. Nelson, T. L. Giesler, and R. S. Lasken, “Rapid of a high-density oligonucleotide microarray approach to study
amplification of plasmid and phage DNA using Phi29 DNA bacterial population dynamics during uranium reduction and
reoxidation,” Applied and Environmental Microbiology, vol. 72, [54] G. Yamey, “Scientists unveil first draft of human genome,”
no. 9, pp. 6288–6298, 2006. British Medical Journal, vol. 321, no. 7252, p. 7, 2000.
[38] D.-Y. Lee, K. Shannon, and L. A. Beaudette, “Detection of [55] J. Craig Venter, M. D. Adams, E. W. Myers et al., “The sequence
bacterial pathogens in municipal wastewater using an oligonu- of the human genome,” Science, vol. 291, no. 5507, pp. 1304–1351,
cleotide microarray and real-time quantitative PCR,” Journal of 2001.
Microbiological Methods, vol. 65, no. 3, pp. 453–467, 2006. [56] E. S. Lander, L. M. Linton, B. Birren et al., “Initial sequencing
[39] K. Katsaveli, D. Vayenas, G. Tsiamis, and K. Bourtzis, “Bac- and analysis of the human genome,” Nature, vol. 409, no. 6822,
terial diversity in Cr(VI) and Cr(III)-contaminated industrial pp. 860–921, 2001.
wastewaters,” Extremophiles, vol. 16, no. 2, pp. 285–296, 2012.
[57] C. S. Pareek, R. Smoczynski, and A. Tretyn, “Sequencing tech-
[40] T. Z. DeSantis, E. L. Brodie, J. P. Moberg, I. X. Zubieta, Y. M. nologies and genome sequencing,” Journal of Applied Genetics,
Piceno, and G. L. Andersen, “High-density universal 16S rRNA vol. 52, no. 4, pp. 413–435, 2011.
microarray analysis reveals broader diversity than typical clone
[58] T. P. Niedringhaus, D. Milanova, M. B. Kerby, M. P. Snyder, and
library when sampling the environment,” Microbial Ecology, vol.
A. E. Barron, “Landscape of next-generation sequencing tech-
53, no. 3, pp. 371–383, 2007.
nologies,” Analytical Chemistry, vol. 83, no. 12, pp. 4327–4341,
[41] T. M. Korves, Y. M. Piceno, L. M. Tom et al., “Bacterial com- 2011.
munities in commercial aircraft high-efficiency particulate air
(HEPA) filters assessed by PhyloChip analysis,” Indoor Air, vol. [59] M. Ronaghi, S. Karamohamed, B. Pettersson, M. Uhlén, and P.
23, no. 1, pp. 50–61, 2013. Nyrén, “Real-time DNA sequencing using detection of pyro-
phosphate release,” Analytical Biochemistry, vol. 242, no. 1, pp.
[42] F. Reith, J. Brugger, C. M. Zammit et al., “Influence of geogenic
84–89, 1996.
factors on microbial communities in metallogenic Australian
soils,” The ISME Journal, vol. 6, pp. 2107–2118, 2012. [60] M. Ronaghi, M. Uhlén, and P. Nyrén, “A sequencing method
based on real-time pyrophosphate,” Science, vol. 281, no. 5375,
[43] U. S. Sagaram, K. M. Deangelis, P. Trivedi, G. L. Andersen, S.-E.
pp. 363–365, 1998.
Lu, and N. Wang, “Bacterial diversity analysis of huanglongbing
pathogen-infected citrus, using phyloChip arrays and 16S rRNA [61] M. Margulies, M. Egholm, W. E. Altman et al., “Genome seq-
gene clone library sequencing,” Applied and Environmental uencing in microfabricated high-density picolitre reactors,”
Microbiology, vol. 75, no. 6, pp. 1566–1574, 2009. Nature, vol. 437, no. 7057, pp. 376–380, 2005.
[44] M. J. Cox, Y. J. Huang, K. E. Fujimura et al., “Lactobacillus casei [62] E. R. Mardis, “A decade’s perspective on DNA sequencing tech-
abundance is associated with profound shifts in the infant gut nology,” Nature, vol. 470, no. 7333, pp. 198–203, 2011.
microbiome,” PLoS One, vol. 5, no. 1, Article ID e8745, 2010. [63] O. Morozova, M. Hirst, and M. A. Marra, “Applications of new
[45] C. Rinke, P. Schwientek, A. Sczyrba et al., “Insights into the sequencing technologies for transcriptome analysis,” Annual
phylogeny and coding potential of microbial dark matter,” Review of Genomics and Human Genetics, vol. 10, pp. 135–151,
Nature, vol. 499, no. 7459, pp. 431–437, 2013. 2009.
[46] C. Palmer, E. M. Bik, M. B. Eisen et al., “Rapid quantitative [64] H. L. Harris, L. J. Brennan, B. A. Keddie, and H. R. Braig, “Bac-
profiling of complex microbial populations,” Nucleic Acids terial symbionts in insects: balancing life and death,” Symbiosis,
Research, vol. 34, no. 1, p. e5, 2006. vol. 51, no. 1, pp. 37–53, 2010.
[47] J.-C. Cho and J. M. Tiedje, “Bacterial species determination [65] N. Whiteford, T. Skelly, C. Curtis et al., “Swift: primary data
from DNA-DNA hybridization by using genome fragments and analysis for the Illumina Solexa sequencing platform,” Bioinfor-
DNA microarrays,” Applied and Environmental Microbiology, matics, vol. 25, no. 17, pp. 2194–2199, 2009.
vol. 67, no. 8, pp. 3677–3682, 2001.
[66] F. Tang, C. Barbacioru, Y. Wang et al., “mRNA-Seq whole-
[48] E. A. Greene and G. Voordouw, “Analysis of environmental transcriptome analysis of a single cell,” Nature Methods, vol. 6,
microbial communities by reverse sample genome probing,” no. 5, pp. 377–382, 2009.
Journal of Microbiological Methods, vol. 53, no. 2, pp. 211–219,
[67] N. Cloonan, A. R. R. Forrest, G. Kolle et al., “Stem cell transcrip-
2003.
tome profiling via massive-scale mRNA sequencing,” Nature
[49] J. D. Van Nostrand, Z. He, and J. Zhou, “Dynamics of microbes Methods, vol. 5, no. 7, pp. 613–619, 2008.
in the natural setting: development of the Geochip,” in Envi-
ronmental Microbiology, K. Sen and N. J. Ashbolt, Eds., Caister [68] J. M. Rothberg, W. Hinz, T. M. Rearick et al., “An integrated
Academic, Norfolk, UK, 2011. semiconductor device enabling non-optical genome sequenc-
ing,” Nature, vol. 475, no. 7356, pp. 348–352, 2011.
[50] Z. He, Y. Deng, and J. Zhou, “Development of functional
gene microarrays for microbial community analysis,” Current [69] M. A. Quail, M. Smith, P. Coupland et al., “A tale of three next
Opinion in Biotechnology, vol. 23, no. 1, pp. 49–55, 2012. generation sequencing platforms: comparison of Ion Torrent,
Pacific Biosciences and Illumina MiSeq sequencers,” BMC
[51] J. Xie, Z. He, X. Liu et al., “GeoChip-based analysis of the
Genomics, vol. 13, article 341, 2012.
functional gene diversity and metabolic potential of microbial
communities in acid mine drainage,” Applied and Environmen- [70] J. Eid, A. Fehr, J. Gray et al., “Real-time DNA sequencing from
tal Microbiology, vol. 77, no. 3, pp. 991–999, 2011. single polymerase molecules,” Science, vol. 323, no. 5910, pp.
[52] Z. He, J. D. Van Nostrand, and J. Zhou, “Applications of func- 133–138, 2009.
tional gene microarrays for profiling microbial communities,” [71] N. J. Loman, R. V. Misra, T. J. Dallman et al., “Performance com-
Current Opinion in Biotechnology, vol. 23, no. 3, pp. 460–466, parison of benchtop high-throughput sequencing platforms,”
2012. Nature Biotechnology, vol. 30, no. 5, pp. 434–439, 2012.
[53] M. J. Beazley, R. J. Martinez, S. Rajan et al., “Microbial commu- [72] N. R. Pace, D. A. Stahl, D. J. Lane, and G. J. Olsen, “The analysis
nity analysis of a coastal salt marsh affected by the Deepwater of natural microbial populations by ribosomal RNA sequences,”
Horizon oil spill,” PLoS One, vol. 7, no. 7, Article ID e41305, 2012. ASM News, vol. 51, no. 4, pp. 4–12, 1985.
[73] G. Muyzer, “DGGE/TGGE a method for identifying genes from [89] S. Vasileiadis, E. Puglisi, M. Arena, F. Cappa, P. S. Cocconcelli,
natural ecosystems,” Current Opinion in Microbiology, vol. 2, no. and M. Trevisan, “Soil bacterial diversity screening using single
3, pp. 317–322, 1999. 16S rRNA gene V regions coupled with multi-million read
[74] G. Laguerre, M.-R. Allard, F. Revoy, and N. Amarger, “Rapid generating sequencing technologies,” PLoS One, vol. 7, no. 8,
identification of rhizobia by restriction fragment length poly- Article ID e42671, 2012.
morphism analysis of PCR-amplified 16S rRNA genes,” Applied [90] L. F. W. Roesch, R. R. Fulthorpe, A. Riva et al., “Pyrosequencing
and Environmental Microbiology, vol. 60, no. 1, pp. 56–63, 1994. enumerates and contrasts soil microbial diversity,” ISME Jour-
[75] J. Dunbar, L. O. Ticknor, and C. R. Kuske, “Assessment of nal, vol. 1, no. 4, pp. 283–290, 2007.
microbial diversity in four Southwestern United States soils by [91] V. Acosta-Martı́nez, S. Dowd, Y. Sun, and V. Allen, “Tag-
16S rRNA gene terminal restriction fragment analysis,” Applied encoded pyrosequencing analysis of bacterial diversity in a
and Environmental Microbiology, vol. 66, no. 7, pp. 2943–2950, single soil type as affected by management and land use,” Soil
2000. Biology and Biochemistry, vol. 40, no. 11, pp. 2762–2770, 2008.
[76] D.-H. Lee, Y.-G. Zo, and S.-J. Kim, “Nonradioactive method [92] S. Vasileiadis, E. Puglisi, M. Arena et al., “Soil microbial
to study genetic profiles of natural bacterial communities by diversity patterns of a lowland spring environment,” FEMS
PCR-single-strand-conformation polymorphism,” Applied and Microbiology Ecology, 2013.
Environmental Microbiology, vol. 62, no. 9, pp. 3112–3120, 1996. [93] J. Rousk, E. Bååth, P. C. Brookes et al., “Soil bacterial and fungal
[77] D. Medini, D. Serruto, J. Parkhill et al., “Microbiology in the communities across a pH gradient in an arable soil,” ISME
post-genomic era,” Nature Reviews Microbiology, vol. 6, no. 6, Journal, vol. 4, no. 10, pp. 1340–1352, 2010.
pp. 419–430, 2008. [94] H. Nacke, A. Thürmer, A. Wollherr et al., “Pyrosequencing-
[78] A. Fabrice and R. Didier, “Exploring microbial diversity using based assessment of bacterial community structure along dif-
16S rRNA high-throughput methods,” Journal of Computer ferent management types in German forest and grassland soils,”
Science and Systems Biology, pp. 074–092, 2009. PLoS One, vol. 6, no. 2, Article ID e17000, 2011.
[79] P. D. Schloss, D. Gevers, and S. L. Westcott, “Reducing the effects [95] M. Buée, M. Reich, C. Murat et al., “454 Pyrosequencing
of PCR amplification and sequencing Artifacts on 16s rRNA- analyses of forest soils reveal an unexpectedly high fungal
based studies,” PLoS One, vol. 6, no. 12, Article ID e27310, 2011. diversity,” New Phytologist, vol. 184, no. 2, pp. 449–456, 2009.
[80] B. J. Baker and J. F. Banfield, “Microbial communities in acid [96] A. Jumpponen, K. L. Jones, and J. Blair, “Vertical distribution of
mine drainage,” FEMS Microbiology Ecology, vol. 44, no. 2, pp. fungal communities in tallgrass prairie soil,” Mycologia, vol. 102,
139–152, 2003. no. 5, pp. 1027–1041, 2010.
[81] C. W. Nossa, W. E. Oberdorf, L. Yang et al., “Design of 16S rRNA [97] T. Stoeck, D. Bass, M. Nebel et al., “Multiple marker parallel
gene primers for 454 pyrosequencing of the human foregut tag environmental DNA sequencing reveals a highly complex
microbiome,” World Journal of Gastroenterology, vol. 16, no. 33, eukaryotic community in marine anoxic water,” Molecular
pp. 4135–4144, 2010. Ecology, vol. 19, no. 1, pp. 21–31, 2010.
[82] N. Youssef, C. S. Sheik, L. R. Krumholz, F. Z. Najar, B. A. [98] R. Logares, S. Audic, S. Santini, M. C. Pernice, C. de Vargas,
Roe, and M. S. Elshahed, “Comparison of species richness esti- and R. Massana, “Diversity patterns and activity of uncultured
mates obtained using nearly complete fragments and simulated marine heterotrophic flagellates unveiled with pyrosequenc-
pyrosequencing-generated fragments in 16S rRNA gene-based ing,” ISME Journal, vol. 6, no. 10, pp. 1823–1833, 2012.
environmental surveys,” Applied and Environmental Microbiol- [99] J. A. Gilbert, J. A. Steele, J. G. Caporaso et al., “Defining seasonal
ogy, vol. 75, no. 16, pp. 5227–5236, 2009. marine microbial community dynamics,” ISME Journal, vol. 6,
[83] Y. Wang and P.-Y. Qian, “Conservative fragments in bacterial no. 2, pp. 298–308, 2012.
16S rRNA genes and primer design for 16S ribosomal DNA [100] S. Monchy, G. Sanciu, M. Jobard et al., “Exploring and quantify-
amplicons in metagenomic studies,” PLoS One, vol. 4, no. 10, ing fungal diversity in freshwater lake ecosystems using rDNA
Article ID e7401, 2009. cloning/sequencing and SSU tag pyrosequencing,” Environmen-
[84] B. J. Paster, S. K. Boches, J. L. Galvin et al., “Bacterial diversity tal Microbiology, vol. 13, no. 6, pp. 1433–1453, 2011.
in human subgingival plaque,” Journal of Bacteriology, vol. 183, [101] R. Logares, E. S. Lindström, S. Langenheder et al., “Biogeogra-
no. 12, pp. 3770–3783, 2001. phy of bacterial communities exposed to progressive long-term
[85] J. E. Clarridge III, “Impact of 16S rRNA gene sequence analysis environmental change,” The ISME Journal, vol. 7, no. 5, pp. 937–
for identification of bacteria on clinical microbiology and 948, 2013.
infectious diseases,” Clinical Microbiology Reviews, vol. 17, no. [102] N. Taib, J. F. Mangot, I. Domaizon, G. Bronner, and D. Debroas,
4, pp. 840–862, 2004. “Phylogenetic affiliation of SSU rRNA genes generated by
[86] B. Wang, J. Wang, W. Zhang, and D. R. Meldrum, “Application massively parallel sequencing: new insights into the freshwater
of synthetic biology in cyanobacteria and algae,” Frontiers in protist diversity,” PLoS One, vol. 8, article e58950, no. 3, 2013.
Microbiology, vol. 3, article 344, 2012. [103] C. E. Robertson, L. K. Baumgartner, J. K. Harris et al., “Culture-
[87] S. Chakravorty, D. Helb, M. Burday, N. Connell, and D. Alland, independent analysis of aerosol microbiology in a metropolitan
“A detailed analysis of 16S ribosomal RNA gene segments for subway system,” Applied and Environmental Microbiology, vol.
the diagnosis of pathogenic bacteria,” Journal of Microbiological 79, no. 11, pp. 3485–3493, 2013.
Methods, vol. 69, no. 2, pp. 330–339, 2007. [104] C. Pedrós-Alió, “Marine microbial diversity: can it be deter-
[88] P. S. Kumar, M. R. Brooker, S. E. Dowd, and T. Camerlengo, mined?” Trends in Microbiology, vol. 14, no. 6, pp. 257–263,
“Target region selection is a critical determinant of community 2006.
fingerprints generated by 16S Pyrosequencing,” PLoS One, vol. [105] E. Pelletier, A. Kreimeyer, S. Bocs et al., “‘Candidatus Cloa-
6, no. 6, Article ID e20956, 2011. camonas acidaminovorans’: genome sequence reconstruction
provides a first glimpse of a new bacterial division,” Journal of [122] P. C. Blainey, A. C. Mosier, A. Potanina, C. A. Francis, and
Bacteriology, vol. 190, no. 7, pp. 2572–2579, 2008. S. R. Quake, “Genome of a low-salinity ammonia-oxidizing
[106] K. F. Ettwig, M. K. Butler, D. Le Paslier et al., “Nitrite-driven archaeon determined by single-cell and metagenomic analysis,”
anaerobic methane oxidation by oxygenic bacteria,” Nature, vol. PLoS One, vol. 6, no. 2, Article ID e16626, 2011.
464, no. 7288, pp. 543–548, 2010. [123] O. U. Mason, T. C. Hazen, S. Borglin et al., “Metagenome, meta-
[107] J. G. Elkins, M. Podar, D. E. Graham et al., “A korarchaeal transcriptome and single-cell sequencing reveal microbial
genome reveals insights into the evolution of the Archaea,” response to Deepwater Horizon oil spill,” The ISME Journal, vol.
Proceedings of the National Academy of Sciences of the United 6, no. 9, pp. 1715–1727, 2012.
States of America, vol. 105, no. 23, pp. 8102–8107, 2008. [124] M. Hess, A. Sczyrba, R. Egan et al., “Metagenomic discovery
[108] H. Tamaki, Y. Tanaka, H. Matsuzawa et al., “Armatimonas rosea of biomass-degrading genes and genomes from cow rumen,”
gen. nov., sp. nov., of a novel bacterial phylum, Armatimo- Science, vol. 331, no. 6016, pp. 463–467, 2011.
nadetes phyl. nov., formally called the candidate phylum OP10,” [125] P. Lorenz and J. Eck, “Metagenomics and industrial applica-
International Journal of Systematic and Evolutionary Microbiol- tions,” Nature Reviews Microbiology, vol. 3, no. 6, pp. 510–516,
ogy, vol. 61, no. 6, pp. 1442–1447, 2011. 2005.
[109] K. C. Wrighton, B. C. Thomas, I. Sharon et al., “Fermentation, [126] P. D. Schloss and J. Handelsman, “Biotechnological prospects
hydrogen, and sulfur metabolism in multiple uncultivated from metagenomics,” Current Opinion in Biotechnology, vol. 14,
bacterial phyla,” Science, vol. 337, no. 6102, pp. 1661–1665, 2012. no. 3, pp. 303–310, 2003.
[110] T. Kalisky and S. R. Quake, “Single-cell genomics,” Nature Meth- [127] A. Siegl and U. Hentschel, “PKS and NRPS gene clusters from
ods, vol. 8, no. 4, pp. 311–314, 2011. microbial symbiont cells of marine sponges by whole genome
[111] B. K. Swan, M. Martinez-Garcia, C. M. Preston et al., “Potential amplification,” Environmental Microbiology Reports, vol. 2, no.
for chemolithoautotrophy among ubiquitous bacteria lineages 4, pp. 507–513, 2010.
in the dark ocean,” Science, vol. 333, no. 6047, pp. 1296–1300, [128] M. Hess, A. Sczyrba, R. Egan et al., “Metagenomic discovery of
2011. biomass-degrading genes and genomes from cow rumen,”
[112] T. Woyke, G. Xie, A. Copeland et al., “Assembling the marine Science, vol. 331, no. 6016, pp. 463–467, 2011.
metagenome, one cell at a time,” PLoS One, vol. 4, no. 4, Article [129] M. Martinez-Garcia, D. M. Brazel, B. K. Swan et al., “Capturing
ID e5299, 2009. single cell genomes of active polysaccharide degraders: an
[113] T. Woyke, D. Tighe, K. Mavromatis et al., “One bacterial cell, unexpected contribution of verrucomicrobia,” PLoS One, vol. 7,
one complete genome,” PLoS One, vol. 5, no. 4, Article ID e10314, no. 4, Article ID e35314, 2012.
2010. [130] M. A. Moran, “Metatranscriptomics: eavesdropping on com-
[114] A. Raghunathan, H. R. Ferguson Jr., C. J. Bornarth, W. Song, M. plex microbial communities,” Microbe, vol. 4, no. 7, pp. 329–335,
Driscoll, and R. S. Lasken, “Genomic DNA amplification from a 2009.
single bacterium,” Applied and Environmental Microbiology, vol. [131] J. Zhou, L. Wu, Y. Deng et al., “Reproducibility and quantitation
71, no. 6, pp. 3342–3347, 2005. of amplicon sequencing-based detection,” ISME Journal, vol. 5,
[115] K. Leung, H. Zahn, T. Leaver et al., “A programmable droplet- no. 8, pp. 1303–1313, 2011.
based microfluidic device applied to multiparameter analysis of [132] J. Zhou, S. Kang, C. W. Schadt, and C. T. Garten Jr., “Spatial
single microbes and microbial communities,” Proceedings of the scaling of functional gene diversity across various microbial
National Academy of Sciences of the United States of America, taxa,” Proceedings of the National Academy of Sciences of the
vol. 109, no. 20, pp. 7665–7670, 2012. United States of America, vol. 105, no. 22, pp. 7768–7773, 2008.
[116] M. Podar, C. B. Abulencia, M. Walcher et al., “Targeted access to [133] S. Turner, K. M. Pryer, V. P. W. Miao, and J. D. Palmer, “Inves-
the genomes of low-abundance organisms in complex microbial tigating deep phylogenetic relationships among cyanobacteria
communities,” Applied and Environmental Microbiology, vol. 73, and plastids by small subunit rRNA sequence analysis,” Journal
no. 10, pp. 3205–3214, 2007. of Eukaryotic Microbiology, vol. 46, no. 4, pp. 327–338, 1999.
[117] N. H. Youssef, P. C. Blainey, S. R. Quake, and M. S. Elshahed, [134] D. J. Lane, “16S/23S rRNA sequencing,” in Nucleic Acid Tech-
“Partial genome assembly for a candidate division OP11 single niques in Bacterial Systematics, E. Stackebrandt and M. Good-
cell from an anoxic spring (Zodletone spring, Oklahoma),” fellow, Eds., pp. 115–175, John Wiley & Son, New York, NY, USA,
Applied and Environmental Microbiology, vol. 77, no. 21, pp. 1991.
7804–7814, 2011.
[135] M. T. Suzuki and S. J. Giovannoni, “Bias caused by template
[118] A. Siegl, J. Kamke, T. Hochmuth et al., “Single-cell genomics annealing in the amplification of mixtures of 16S rRNA genes
reveals the lifestyle of Poribacteria, a candidate phylum symbi- by PCR,” Applied and Environmental Microbiology, vol. 62, no.
otically associated with marine sponges,” ISME Journal, vol. 5, 2, pp. 625–630, 1996.
no. 1, pp. 61–70, 2011.
[136] J. C. Makemson, N. R. Fulayfil, W. Landry et al., “Shewanella
[119] H. S. Yoon, D. C. Price, R. Stepanauskas et al., “Single-cell woodyi sp. nov., an Exclusively respiratory luminous bacterium
genomics reveals organismal interactions in uncultivated ma- isolated from the Alboran sea,” International Journal of System-
rine protists,” Science, vol. 332, no. 6030, pp. 714–717, 2011. atic Bacteriology, vol. 47, no. 4, pp. 1034–1039, 1997.
[120] R. Ghai, L. Pašić, A. B. Fernández et al., “New abundant micro- [137] N. Boon, W. De Windt, W. Verstraete, and E. M. Top, “Eval-
bial groups in aquatic hypersaline environments,” Scientific uation of nested PCR-DGGE (denaturing gradient gel elec-
Reports, vol. 1, 135, 2011. trophoresis) with group-specific 16S rRNA primers for the
[121] H. Chitsaz, J. L. Yee-Greenbaum, G. Tesler et al., “Efficient de analysis of bacterial communities from different wastewater
novo assembly of single-cell bacterial genomes from short-read treatment plants,” FEMS Microbiology Ecology, vol. 39, no. 2, pp.
data sets,” Nature Biotechnology, vol. 29, no. 10, pp. 915–922, 2011. 101–112, 2002.
[138] H. Heuer, K. Hartung, G. Wieland, I. Kramer, and K. Smalla,

“Polynucleotide probes that target a hypervariable region of
16S rRNA genes to identify bacterial isolates corresponding to
bands of community fingerprints,” Applied and Environmental
[139] J. Jonasson, M. Olofsson, and H.-J. Monstein, “Classification,
identification and subtyping of bacteria based on pyrosequenc-
ing and signature matching of 16S rDNA fragments,” APMIS,
vol. 110, no. 3, pp. 263–272, 2002.
http://dx.doi.org/10.1155/2013/240175
Review Article
The World Bacterial Biogeography and
Biodiversity through Databases: A Case Study of
NCBI Nucleotide Database and GBIF Database
Okba Selama,1 Phillip James,2 Farida Nateche,1

Elizabeth M. H. Wellington,2 and Hocine Hacène1
1
Microbiology Group, Laboratory of Cellular and Molecular Biology, Faculty of Biological Sciences, USTHB, BP 32,
EL ALIA, Bab Ezzouar, Algiers, Algeria
2
Environmental Microbiology, School of Life Sciences, University of Warwick, Coventry CV4 7AL, UK
Correspondence should be addressed to Hocine Hacène; h hacene@yahoo.fr
Received 14 March 2013; Revised 11 July 2013; Accepted 13 August 2013
Academic Editor: Konstantinos Mavrommatis
Copyright © 2013 Okba Selama et al. This is an open access article distributed under the Creative Commons Attribution License,
Databases are an essential tool and resource within the field of bioinformatics. The primary aim of this study was to generate an
overview of global bacterial biodiversity and biogeography using available data from the two largest public online databases, NCBI
Nucleotide and GBIF. The secondary aim was to highlight the contribution each geographic area has to each database. The basis
for data analysis of this study was the metadata provided by both databases, mainly, the taxonomy and the geographical area origin
of isolation of the microorganism (record). These were directly obtained from GBIF through the online interface, while E-utilities
and Python were used in combination with a programmatic web service access to obtain data from the NCBI Nucleotide Database.
Results indicate that the American continent, and more specifically the USA, is the top contributor, while Africa and Antarctica
are less well represented. This highlights the imbalance of exploration within these areas rather than any reduction in biodiversity.
This study describes a novel approach to generating global scale patterns of bacterial biodiversity and biogeography and indicates
that the Proteobacteria are the most abundant and widely distributed phylum within both databases.
1. Introduction environments on scales ranging from 0.002 km to 20,000 km

[1] and from scale of a nation [8] to intercontinental scale [9].
Biogeography aims to explain spatial patterns of diversity Data from many of these biodiversity studies are stored
in the context of evolutionary events such as speciation, in databases, a structured and organized collection of infor-
dispersal, extinction, and species interactions [1]. Macroecol- mation where the storage of and the access to information
ogists have long studied the biogeography of higher plants are facilitated to users. In biosciences, the introduction of
and animals in various habitats [2, 3]. In contrast, there computer processing and computer databases has opened up
is very little information available on the biogeography of the potential for further investigation of combined existing
prokaryotes. This stemmed from the difficulty of assessing data sets [10]. These include the study of specie distributions
microbial communities by cultivation methods, which only through both time and space and their use as an educational
sampled 0.1% to 10% of the microbial community [4]. How- resource (both formal and public), for conservation and
ever, with the advent of cultivation-independent sequencing scientific research, use in medicine and forensic studies, in
techniques, microbial communities of many environments natural resource management and climate change, in art,
have been characterized, including soil [5], the Arctic and history, and recreation, and for social and political use. Uses
Antarctic Oceans [6], and the Sargasso Sea [7]. This, in turn, are many and varied and may well form the basis of much of
facilitated prokaryotic biogeography studies in a number of what we do as people every day [11].
In our study, we used the concept of species occurrence a public database along with 52 others that belong to The
data, mainly, observational data, and environmental survey National Center of Biotechnology Information (NCBI),
data. In general, the data are what we term “point based,” which is a division of the National Library of Medicine
although line (transect data from environmental surveys, (NLM) at National Institutes of Health (NIH). The database
collections along a river), polygon (observations from within is formed of a collection of nucleotide sequences from
a defined area such as a national park), and grid data several sources, including GenBank, which is part of the
(observations or survey records from a regular grid) are also International Nucleotide Sequence Database Collaboration
included. The majority of point-based data used here are (INSDC), which is comprised of the DNA DataBank of
georeferenced; that is, records with geographic references tie Japan (DDBJ), the European Molecular Biology Laboratory
them to a particular place in space—whether with a georefer- (EMBL), and GenBank at NCBI. These three organiza-
enced coordinate (e.g., latitude and longitude, UTM) or not tions exchange data on a daily basis—the NCBI Nucleotide
(textual description of a locality, altitude, depth)—and time Database also includes sequences from NCBI Reference
(date, time of day). Often, the data are also tied to a taxonomic Sequences (RefSeq), Third Party Annotation (TPA), and from
name, but unidentified collections may also be included Protein Data Bank (PDB). At the time of writing, the NCBI
[12]. We retrieved bacterial records for different worldwide Nucleotide Database included 78,756,144 records (March 3,
geographical areas, countries/islands, which were stored in 2013 at 04:30) [14].
NCBI Nucleotide Database and GBIF Database [13, 14] and
then assigned them to their respective phyla. This was in 2.2.2. List of Geographical Areas. The list of geographical
order to describe the world bacterial biogeography at a broad areas used in this study was obtained from the Interna-
taxonomic scale in terms of taxa proportional abundance tional Nucleotide Sequence Database Collaboration (INSDC)
by contributed records from each geographic region. Since through controlled vocabulary for “/country qualifier” [16].
databases are growing fast, we limited our search to a The study also included the distribution of bacteria among
determined period, data published on/before December 25, the seven continents.
2012.
2.2.3. List of Phyla. Common phyla were selected from the
2. Material and Methods NCBI Taxonomy (number of species: 11,364 with 31 phyla)
[17, 18] and the catalogue of life taxonomic classification
2.1. Hardware. One personal computer was used having a (number of species: 9,072 with 25 phyla) [19], used respec-
Dual Core CPU E5800 @ 3.20 GHz processor and 2 GB RAM. tively by NCBI Nucleotide and GBIF databases. The final list
Internet connection was tested as 1.36 Mbps download and included 24 common phyla, listed as follows:
5.55 Mbps upload [15]. bacteria main groups = [“Acidobacteria”, “Actinobacte-
ria”, “Aquificae”, “Bacteroidetes”, “Chlamydiae”, “Chlorobi”,
2.2. The Approach. The approach used in this study for both “Chloroflexi”, “Chrysiogenetes”, “Cyanobacteria”, “Deferrib-
databases is divided into three parts: acteres”, “Deinococcus-Thermus”, “Dictyoglomi”, “Fibrobac-
teres”, “Firmicutes”, “Fusobacteria”, “Gemmatimonadetes”,
(i) database query →
“Lentisphaerae”, “Nitrospirae”, “Planctomycetes”, “Proteobac-
(ii) data subset retrieval (bacterial records verifying the teria”, “Spirochaetes”, “Thermodesulfobacteria”, “Thermoto-
query structure) in standardized response format for gae”, “Verrucomicrobia”].
each geographical area →
2.2.4. Access Databases
(iii) analyze data and save the information summary for
each geographical area. GBIF Database. The number of records with geographic
coordinates from the GBIF Database is displayed through the
2.2.1. Databases GBIF species portal [20]. The bacterial records were retrieved
from GBIF Database for each of the geographical areas of the
GBIF Database. The Global Biodiversity Information Facility study through the occurrence search webpage. The keywords
(GBIF) was established as a global megascience initiative to used in “Add search filter” were “Bacteria” for the Taxonomy
address one of the great challenges of the 21st century— (Scientific Name) filter and the respective “geographical
harnessing knowledge of the Earth’s biological diversity. GBIF area’s name” for the Geospatial filter. The generated results
envisions a world in which biodiversity information is freely were downloaded as spreadsheet zipped files [21]. Once
and universally available for science, society, and a sustainable downloaded, a Python script (version 2.7.3) [22] (see
future. GBIF’s mission is to be the foremost global resource Supplementary Materials: GBIF Filter.py available online at
for biodiversity information and engender smart solutions for http://dx.doi.org/10.1155/2013/240175) was used to filter files
environmental and human well-being. At the time of writing, and to retrieve the occurrences of bacterial records for each
the GBIF Database include 396,026,747 records, 345,561,101 geographical area based on a simple algorithm (see
of which have associated georeference data (March 3, 2013 at Algorithm 1: Biodiversity and Biogeography—GBIF Filter).
10:32) (Version 1.2.6) [10–13].
NCBI Nucleotide Database. The general way (simple, direct,
The NCBI Nucleotide Database. The National Center of and manual) to query NCBI Nucleotide Database (save/
Biotechnology Information (NCBI) Nucleotide Database is extract data) is by using web services through a web browser
Definition part:
Bacteria phyla (bacteria main groups)
// all variables are set at zero (0) or an empty list
Define treatments and operations:
Retrieve and set the classification used from the directory
“Classification 2000 Plus”, see supplementary materials directory.
Retrieve data from each geographical area fond in the directory
“GBIF Plus”, see supplementary materials and filter and assign them to their
respective phyla.
Write the occurrences in the file “gbif Classification 2000 Plus.txt”, see
supplementary materials.
Unclassified taxa are saved in the file.
“absent taxa Classification 2000 Plus.txt” and
“absent taxa Classification 2000 Plus ex All.txt” see supplementary
materials.
Algorithm 1: Biodiversity and biogeography—GBIF Filter.
[14]. However, this method is not adapted for automatic word, and the standard search in this case would be for
multitask queries—that is, for the search of information about every researchable field for the combination of both the
few organisms, the user has to introduce queries, one by one, geographical area’s names and the word “country” without
for each organism and to retrieve records each time. Thus, distinguishing between the origin of the sequence and the
the search would be time consuming, and for a large number collaborating country(ies). To verify this, using an additional
of organisms would be manually impossible. Similarly to the name of a geographical area, for instance “Italy”, in the query
two other INSDC partners, EMBL and DDBJ, NCBI provides structure of the search as “country France Italy”, will result in
a programmatic access to various data resources and analysis giving overestimated records where both countries are men-
tools via web services technologies. tioned although the sequences are registered to only one
geographical area.
Programmatic Retrieval System for NCBI Nucleotide Database As there is no direct method to access the “quali-
Records. The programmatic access for NCBI records passes fier/country” by a simple query structure, and to be more
through the Entrez Programming Utilities (NCBI E-utilities), restrictive and more accurate, additional computer process-
a set of eight server-side programs that provide a stable ing to return the desired sample location using the “quali-
interface into the Entrez query and database system at the fier/country” should be applied.
NCBI [23] and a computer language. In this study, Python For each of the retrieved records, where the “geographical
(version 2.7.3) was used with Biopython package (version area’s name” and the word “country” were used as keywords
1.60) [22, 24]. First, Python posts an E-utility URL to NCBI for the filter, we extract the whole information value included
and then retrieves the results of this request, after which it in the “qualifier/country” field when it exists [16]. Then, for
processes the data as required [23]. each record, we match the information to the geographical
When using the geographical area’s name directly as a area’s name of interest; if it matches, we count the record and
search term, for instance “France”, the results retrieved would we consider its phylum.
give all sequences where the word “France” is mentioned. This A Python script was written; see supplementary
is problematic as, for example, results returned would include materials: NCBI Nucleotide Tracker.py, based on an algo-
those where authors institutions are in France rather than the rithm (Algorithm 2: Biodiversity and Biogeography—NCBI
country of origin of the sample, which is required. Nucleotide Tracker) which encompasses three main parts as
A new qualifier has been added since December 15, 1998; below.
this is about the “qualifier/country”, which would “restrict”
the search to records that include the geographical origin of (1) Define the query structure:
the sequence [16].
Using the word “country” or “/country” as an additional (i) the query structure: “country AND geographical
word for the search will restrict the search. Yet, similar prob- area’s name AND Bacteria[Organism] AND date
lems are encountered when using records generated from of publication”
collaborative international work. The result would include (a) country: to limit the search to records that
overlap records since “country” is considered as an ordinary may have the qualifier/country;
Definition part:
Connection variables (undertaken by Biopython package)
Bacteria phyla (bacteria main groups)
List of geographical areas (list from file: countries list all.txt) see supplementary
materials.
The query structure (term = “country AND Geographical area’s name AND
Bacteria [Organism] AND Date of publication”)
gi list (list of records verifying the query structure)
listWC (number of records with the existence of the qualifier/country)
lisV (number of records with a real/country qualifier attributed to the right
geographical area)
// all variables are set at zero (0) or an empty list.
Define treatments and operations:
For every geographical area form the list found in “countries list all.txt”:
(i) Query the NCBI database, using the query structure.
(ii) Retrieve the count of gi list
(iii) Retrieve all the records (Genbank format) one by one
(iv) Access each record:
If the qualifier/country exists then:
listWC ← listWC + 1
If the qualifier value matches the geographical area of
interest:
lisV ← lisV + 1
Check for the taxonomy:
Count the sequence regarding the appropriate phylum.
If there is not taxonomy for the sequence (no

bacteria) then register the GI in
file “geographical area Absence Bact.txt”, see
supplementary materials.
Save results for all records of the geographical area on a row in the result file
(country all.txt) see supplementary materials.
Remove the geographical area from the list of geographical areas.
If any errors occurred, save the error type in “error.txt”, see supplementary materials.
Algorithm 2: Biodiversity and Biogeography—NCBI Nucleotide Tracker.
(b) geographical area’s name: to precise the geo- joint part of the subset results of the other
graphical area in the search, and this with words in the query.
respect to the INSDC list;
(2) Connect the script to the NCBI Nucleotide Database:
(c) Bacteria[Organism]: to limit the search to query the database and retrieve the data as a standard
bacteria domain; format (GenBank format, so the real qualifier/coun-
(d) date of publication: to limit the search to a try can be accessed), and this is mainly handled by
time period; Biopython package.
(e) AND: Boolean operator, the intersection, (3) Analyze data: filter the data, access the “qualifier/
used to narrow the search results to the country”, and match the qualifier value to the searched
Table 1: Occurrences overview of records with coordinates from GBIF Database.

Kingdom Link Records with coordinates Percentages
Archaea http://data.gbif.org/species/2 26,501 0,008
Bacteria http://data.gbif.org/species/3 1,593,278 0,479
Animalia http://data.gbif.org/species/1 238,944,036 71,785
Chromista http://data.gbif.org/species/4 1,539,408 0,462
Fungi http://data.gbif.org/species/5 6,176,944 1,856
Plantae http://data.gbif.org/species/6 80,044,950 24,048
Protozoa http://data.gbif.org/species/7 3,916,926 1,177
Incertae sedis http://data.gbif.org/species/0 616,822 0,185
Table 2: Results of general queries using different filters through the NCBI Nucleotide Database webpage.
Query Records
Search “all” [Filter] Limits: Published between: 1986/1/1 and 2012/11/25 72,020,824
Search “ddbj” [Filter] Limits: Published between: 1986/1/1 and 2012/11/25 10,323,758
Search “embl” [Filter] Limits: Published between: 1986/1/1 and 2012/11/25 11,259,765
Search “genbank” [Filter] Limits: Published between: 1986/1/1 and 2012/11/25 44,146,674
Search Bacteria [organism] Limits: Published between: 1986/1/1 and 2012/11/25 7,156,037
Search Archaea [organism] Limits: Published between: 1986/1/1 and 2012/11/25 306,675
Search Eukaryota [organism] Limits: Published between: 1986/1/1 and 2012/11/25 46,489,750
Search “country” Limits: Published between: 1986/1/1 and 2012/11/25 11,994,306
Search ((country AND Bacteria [organism])) Limits: Published between: 1986/1/1 and 2012/11/25 2,276,928
Search (((country AND Archaea [organism]))) Limits: Published between: 1986/1/1 and 2012/11/25 130,882
Search (((country AND Eukaryota [organism]))) Limits: Published between: 1986/1/1 and 2012/11/25 8,346,238
∗
The Nucleotide Advanced Search Builder was used to construct the queries.
geographical area’s name of interest; if it matches from GBIF Database through the GBIF Species Portal, is sum-
and then the record is counted and the taxonomy is marized in Table 1. It is clear from the results that Eukaryota,
recorded. Finally, the summary of this analysis for mostly animals and plants with nearly 95%, are the dominant
each geographical area is saved. registered records, whereas bacteria represent less than 0.5%
of all records.
Since the computer processing used here is word processing,
particular geographic areas were analyzed independently,
differentiating certain ambiguities; for instance, “Republic of 3.1.2. The NCBI Nucleotide. Data in Table 2 show the results
the Congo” and “Democratic Republic of Congo” are different for general queries using different filters through the NCBI
countries but both contain “Republic of Congo” within the Nucleotide Database webpage. GenBank is the most used
qualifier. A third Python script, modified from the previous database to register sequences compared with the INSDC
NCBI Nucleotide Tracker.py, was used in combination with partners (DDJB) and (EMBL). We also observed that most
an exception list to circumvent this problem, (see supple- records were found to be nucleotide sequences of Eukaryota
mentary materials: NCBI Nucleotide Exception.py) results 64%, while bacteria represent just nearly 10%. Additionally,
are registered in a file (see exception.txt supplementary from the 72,020,824 records found in the NCBI Nucleotide
materials). Database, only 17% as 11,994,306 would be tied to a particular
geographical area.
2.2.5. The World Biogeography Maps. Data from this study
was used to generate world bacterial biogeography maps. The 3.2. Bacterial Biogeography and Biodiversity. While the
package “rworldmap”, available on CRAN, was used for the INSDC’s list contains 275 geographical areas and an addi-
mapping and visualization of global data working under the tional 12 historical country names, the final list of this study
environment “R language-version 2.15.1” [25, 26]. includes only 208 common geographical areas. This was
either because some geographical areas do not appear in both
3. Results databases, for example, Borneo and Taiwan or there were no
bacterial records for these in the GBIF Database, for example,
3.1. General Queries Bahrain, Swaziland, and Jersey.
From the 208 geographical areas of this study, for
3.1.1. GBIF Database. The occurrences overview for records the GBIF Database, using filters as described above, and
with coordinates for the seven kingdoms of life, extracted after downloading files, 1,222,216 records were recovered. In
data from NCBI Nucleotide Database 50% than data from

Abundance (%)
70
60
50 GBIF Database 36%. Eleven phyla had a similar degree
40 of distribution among the two databases with less than
30
20 5% difference in terms of record numbers. A difference
10
0 between databases in terms of phyla global distribution was
Thermodesulfobacteria
Chlamydiae
Acidobacteria
Actinobacteria
Aquificae
Bacteroidetes
Chrysiogenetes
Cyanobacteria
Gemmatimonadetes
Deferribacteres
Deinococcus-Thermus
Dictyoglomi
Fibrobacteres
Firmicutes
Fusobacteria
Lentisphaerae
Nitrospirae
Planctomycetes
Proteobacteria
Spirochaetes
Thermotogae
Verrucomicrobia
Chlorobi
Chloroﬂexi noted for the Acidobacteria, Chloroflexi, Plactomycetes and
Spirochaetes, which were more widely distributed in the NCBI
Nucleotide database, while Deferribacteres, Fibrobacteres,
Fusobacteria, and Lentisphaerae were more widely distributed
in the GBIF database. Those with less than 5% of coverage
and coming from less than 10 geographical areas in both
databases were the Thermodesulfobacteria, Dictyoglomi and
Phyla Chrysiogenetes which are considered to be really restricted to
NCBI
certain geographical areas.
GBIF Finally, considering GBIF Database alone, we also
observe that 12 of the 24 phyla were distributed with nearly
Figure 1: The relative abundance of the 24 common phyla in NCBI 20% coverage for the whole 208 geographical areas nearly 40
Nucleotide Database and GBIF Database. geographical areas.
3.2.3. Occurrences of Records in Different Geographical Areas.

total, using the Catalogue of Life Taxonomic Classification, Table 4 shows the occurrences of records by continent for
88% of all retrieved records were assigned to one of the both NCBI Nucleotide and GBIF databases. The American
24 phyla common with NCBI Taxonomy; see supplementary continent has the largest number of records submitted, rep-
materials: gbif Classification 2000 Plus.txt and NCBI GBIF resenting 39% of all registered records in GBIF Database and
overall data.xlsx. more than 50% in the NCBI Nucleotide Database, yet only
Conversely, using the programmatic access approach to half 634,225 of these NCBI Nucleotide records are assigned
query the NCBI Nucleotide Database, we could retrieve to one of the 24 phyla. Europe with 27% and Australia-
information on 3,232,147 records which satisfied the query Oceania with 16% are second and third, respectively, for the
structure with: the name of the geographical area, the word contribution of the GBIF data input, while Asia is more likely
“country”, and bacteria as organism, of those which were to contribute records in the NCBI Nucleotide Database with
assigned to the right geographical area was 2,322,339, 56% 21%, ranking second than to the GBIF Database 11%. Antarc-
−1,311,049 of those which were assigned to one of the 24 tica is less involved with 1% and 4% of the world bacterial
phyla common to Catalogue of Life Taxonomic Classification. biodiversity being registered for GBIF or NCBI Nucleotide
Moreover, 1,233,118 records were retrieved as environmental databases, respectively. Finally, there is nearly 3% of data
samples in NCBI Nucleotide Database using this method. registration from Africa in each database. The world maps for
These could also be environmental samples within already- bacterial biogeography regarding continents are illustrated in
assigned phyla see supplementary materials: country all.txt Figures 2(a1) and 2(a2).
and NCBI GBIF overall data.xlsx. For a close look at the top ten countries for both NCBI
Nucleotide and GBIF databases recovered records and their
3.2.1. The Relative Abundance of Different Phyla. Records assignment to the 24 phyla, Table 5 reveals that USA occupies
retrieved from both NCBI Nucleotide and GBIF databases the first place for both databases. The number of records from
summarized in Figure 1 and Table 3 show that Proteobacteria GBIF would be greater than this since the GBIF maximum
are the most abundant phylum in both databases with records number returned per file is 250,000. Two countries,
64% and 49%, respectively, Firmicutes 13% and Actinobac- Germany and India, ranked in this list for both databases.
teria (8%) were the second most abundant phyla for For the rest of the geographical areas, we observed different
NCBI Nucleotide Database, and Bacteroidetes (11%) and patterns for the two databases. The world maps for bacterial
then Cyanobacteria (9%) and Planctomycetes (7%) for GBIF biogeography regarding countries are presented in Figures
Database. The remaining phyla represented less than 5% each. 2(b1) and 2(b2).
In the last position, we may find Chrysiogenetes and Dictyo- We also observed from Table 5 that while the continents
glomi with less than 0,004% of records for both databases. and the top ten countries bacterial records occurrences
assignments were close to the overall assignment average
3.2.2. Overall Geographical Occurrences of Different Phyla. (88%) for the GBIF Database, the continents and the top
Records retrieved from both databases summarized in ten countries assignments vary enormously from the average
Table 3 show that the most distributed phylum was Pro- assignment (57%) of NCBI Nucleotide Database.
teobacteria, covering 83% of records for GBIF Database and
90% for NCBI Nucleotide Database for all geographical areas 4. Discussion
in this study. Actinobacteria, Cyanobacteria, and Firmicutes
had more than 50% coverage each in both databases. Bac- The study reveals that most bacterial biodiversity was
teroidetes distribution seems to be more important using retrieved from developed countries and USA, particularly.
Table 3: The relative abundance and the overall geographical occurrences of the 24 common phyla in NCBI Nucleotide Database and GBIF Database.
Chlorobi
Aquificae
Firmicutes
Chloroflexi
Nitrospirae
Chlamydiae
Dictyoglomi
Spirochaetes
Fusobacteria
Thermotogae
Bacteroidetes
Fibrobacteres
Acidobacteria
Lentisphaerae
Cyanobacteria
Proteobacteria
Actinobacteria
Chrysiogenetes
Deferribacteres
Planctomycetes
Verrucomicrobia
Gemmatimonadetes
Deinococcus-Thermus
Thermodesulfobacteria
NCBI
Ab 15043 106127 1417 35795 1489 858 4762 9 43896 152 710 37 821 167616 375 1819 79 3118 9734 841254 59943 80 381 15534
% 1,147 8,095 0,108 2,730 0,114 0,065 0,363 0,001 3,348 0,012 0,054 0,003 0,063 12,785 0,029 0,139 0,006 0,238 0,742 64,166 4,572 0,006 0,029 1,185
GBIF
Ab 65711 75156 688 193454 14418 3799 46328 2 167900 26883 2201 4 1889 71480 2375 12896 22999 8337 117391 841535 11087 211 272 48806
% 3,897 4,146 0,032 11,136 0,899 0,202 2,446 0,000 8,999 1,600 0,124 0,000 0,102 3,967 0,132 0,766 1,303 0,464 6,848 49,449 0,583 0,010 0,013 2,878
NCBI
Oc 55 144 20 112 37 32 62 5 114 21 47 6 16 139 26 39 14 40 64 189 92 9 21 56
% 26,442 69,231 9,615 53,846 17,788 15,385 29,808 2,404 54,808 10,096 22,596 2,885 7,692 66,827 12,500 18,750 6,731 19,231 30,769 90,865 44,231 4,327 10,096 26,923
GBIF
Oc 42 124 14 75 41 35 44 1 177 40 40 2 28 112 40 41 41 41 46 173 47 8 29 47
% 20,192 59,615 6,731 36,058 19,712 16,827 21,154 0,481 85,096 19,231 19,231 0,962 13,462 53,846 19,231 19,712 19,712 19,712 22,115 83,173 22,596 3,846 13,942 22,596
Oc: the overall geographical occurrence of a phylum was calculated as the occurrence of at least one record per geographical area. Ab: relative abundance of phyla.
7
Table 4: Occurrences of records by continent for both NCBI Nucleotide and GBIF databases.
Continents GBIF % Assigned % assigned NCBI % Assigned % assigned

AMERICA 481976 39.435 421526 87,458 1200669 51.701 634225 52,823
AFRICA 42289 3.460 37972 89,792 55796 2.403 39723 71,193
EUROPE 335373 27.440 306014 91,246 371561 15.999 214725 57,790
ASIA 143984 11.781 126967 88,181 504874 21.740 341823 67,705
AUSTRALIA-OCEANIA 204615 16.741 182665 89,273 96073 4.137 72257 75,211
ANTARCTICA 13979 1.144 13363 95,593 93366 4.020 8296 8,885
Total 1222216 1088507 2322339 1311049
Table 5: Top ten countries list for NCBI Nucleotide and GBIF databases recovered records and their assignment to the 24 phyla.
Countries Records (GBIF) Assigned % Countries Records (NCBI) Assigned %

USA 250000 89.224 USA 689988 60.753
New Zealand 132127 87.822 China 185045 66.324
United Kingdom 88823 95.639 Brazil 173997 57.799
Germany 90153 83.569 India 82663 89.801
Chile 84339 82.525 Germany 74444 67.535
Netherlands 53903 94.308 Mexico 40678 84.972
Russia 49308 83.563 Japan 87861 39.219
Northern Mariana Islands 33624 90.519 Australia 48788 66.400
Portugal 30651 93.064 Spain 48057 62.861
India 31981 88.840 France 52411 56.563
234 128900 421 426100

(a1) (a2)
(a)
1 1357 5420 14480 24870 33620 53900 90150 132100 250000 0 3072 11000 23310 40680 52410 93370 162200 185000 690000
(b1) (b2)
(b)
Figure 2: The world biogeography (a) by continent in (a1). GBIF Database. (a2). NCBI Nucleotide Database. (b) By country in (b1). GBIF
Database and (b2). NCBI Nucleotide Database.
The bias seen in these databases toward developed countries controlled by differences in environmental variables in some
may be attributed to several reasons: these countries encom- cases [32], and geographical distance in others [35, 36],
pass technological platforms, especially, for the massive of while the few abundant organisms were more likely to be
both sequencing and registration of data and are engaged in widely distributed [34], and those may form a common
a number of biodiversity exploration projects, and yet the diversity structure within soil bacterial communities around
most important reason is research and development funding the globe [37]. Other works investigating overall community
budget. To maintain its position as a world leader in science composition support the role of environmental gradients in
and research, USA has invested a huge budget over the two structuring both lake and soil bacterial communities [38, 39].
last decades, and this is continuously increasing. The forecast Biotic interactions may also be important in determining
for the 2014 USA budget is $142.8 billion; it calls for a federal microbial community composition; a recent study showed
basic and applied research investment totaling $68.1 billion, that microbial communities exhibit more segregation of
up to $4.8 billion or 7.5 percent increase compared to the taxa than would be predicted by chance, suggesting that
2012 enacted level [27]. On the other hand, less biodiversity competitive interactions and/or niche specialization may
is observed in many areas, particularly countries in Africa be important in structuring bacterial biogeography [40].
and in Asia (the Middle East and Central Asia); we do not Similar to Nemergut et al. [34] and within our study of
suggest that less real biodiversity is present in these countries, both databases, although it only involved the phylum rather
but rather that less microbial biodiversity targeted research is than the inferior taxonomy ranks, we have shown that the
performed, and thus less of the generated data are submitted abundant phyla (Proteobacteria, Actinobacteria, Cyanobacte-
to the different databases. ria, Firmicutes, and Bacteroidetes) are the most distributed,
While we could retrieve information on 3,232,147 records whereas the majority of less abundant taxa are predominantly
from the NCBI Nucleotide Database as they satisfy the query located in particular regions. Yet, these results have to be
structure, it is obvious that if compared with a simple general taken with care especially for geographical regions where few
query used through the NCBI Nucleotide Database website records are registered which would not reflect the bacterial
as “Country AND Bacteria”, we would notice a difference diversity within those regions.
of additional 955,219 records. This may be explained, as In terms of data quality, the collector and then the
stated before, by the overestimation of records. Moreover, submitter of the record(s) have the primary responsibility for
the registered records do not reflect the exact number of data quality in both databases [12]. While the submission of
strains isolated or observed in a geographical area, since it record(s) is possible by anyone to NCBI, the GBIF accepts
is possible to find many sequences belonging to the same only credited organisms already registered and approved by
strain, for a redundancy or the fact that they are fragments of the latter. In our study, we have found that NCBI Nucleotide
one genome (example: Streptomyces globisporus C-1027 from Database seems to cover a larger area and would be the only
China is registered as 557 times for whole genome shotgun available resource for bacterial diversity in some regions, for
sequencing). instance, Andorra, Bahrain, and Equatorial Guinea. However,
Forces shaping the biogeography of macroorganisms— it is more likely to be influenced by the biomedical research
including dispersal limitations, habitat differentiation, com- policy of the leading country and its National Institutes
petition, and adaptive radiation—have been a central focus of Health (NIH) this observation is not only toward this
of ecology for more than a century [28]. Yet, while microor- database but also toward many of the generated data in
ganisms are the most abundant and diverse organisms on several research projects of life sciences; this may be also
Earth [29], relatively little is known about the patterns of, or understood when we examine the annual budget that has
controls over, microbial distribution within and between the been invested in research and development awarded to the
planet’s major habitat types. One common theory holds that National Institutes of Health (NIH) which was of $30 billion
the tremendous dispersal potential of microbes will lead to for the year 2012. This was nearly half of the expenditure for
everything being everywhere (i.e., no dispersal limitations), the nondefense R & D budget [14, 28], so it is obvious to see
with environmental selection determining which species a certain preference for the exploration and the registration
are abundant [1]. However, until recently, methodological of a particular category of microorganisms than others, for
limitations have prevented large-scale tests of ideas about example, microorganisms interfering with health, inducing
where certain microorganisms exist and why [30, 31]. diseases, or producing active biomolecules (antibiotics, anti-
Over the last decades, however, molecular phylogenetic tumoral . . .).
approaches have revolutionized microbiology, expanding While the queries were submitted on November 25, 2012,
our view of microbial diversity and our appreciation of submitting the same queries and readying this paper would
the complexity of microbial communities [30]. While these generate slightly different results, and this is due to the update
techniques do not provide an exhaustive sampling of any process for both databases.
but the simplest microbial assemblages, they do provide
information on the dominant members of the community, 5. Conclusion
allowing ecologically meaningful questions to be addressed
about the distribution of these lineages. These methods New technological advances and approaches are emerging
have been used to reveal that some microorganisms exhibit from sampling to data analysis, and this is to cope with the
distinct biogeographical patterns [1, 32, 33] and are demon- diversity and complexity of life. Therefore, data generated in
strated to be the vast majority [34] which appear to be biosciences are growing exponentially. Analysis software and
methods must also keep up with this rapidly expanding field differences on fundamental aspects such as in taxonomical
so that the most can be made of current studies within this classification which was one example encountered in our
field. It is unknown how the patterns that we observed today study; where phyla: Synergistetes, Caldiserica, Elusimicrobia,
may change with the upcoming “daily results”; our study is Armatimonadetes, Ignavibacteria, Tenericutes, Thermomicro-
considered to be the first attempt to catch the first snapshot bia, and the newly established Nitrospinae phylum are consid-
of a particular moment on the world bacterial biogeography ered either different or completely absent in one or another
and biodiversity through the usage of both NCBI Nucleotide database used in this study. All of these points and others are
and GBIF databases. more and more being discussed worldwide by the scientific
Despite these constraints, our approach may be extended community [17, 41].
to other domains of life (Archaea, Eukaryota) or even for a While the web interface is easier to deal with databases,
more restrictive group of taxa (example: Actinobacteria and the programmatic access seems to be more interesting, more
all subtaxa within this group). flexible, offers more choices, and returns more personalized
For the NCBI Nucleotide Database, the same approach results; however, it needs some basic knowledge on the
could generate more information on the retrieved sequence, database structure, its database management system, and
such as: length, type DNA or ARN, single sequence, complete computer languages.
genome or shotgun sequencing, and function of the gene: Finally, while the study gives a preliminary overview of
16S RNA gene or other genes. Almost all information from the world’s bacterial biogeography, reflecting a part of the
any qualifier of a record would be extractable, which may real biodiversity, other more upcoming efforts to determine
answer some of the questions that we may ask: who is doing Earth microbial biogeography and biodiversity are indeed
what? How and why study these strains? Is it perhaps for in progress, we could mention “Earth Microbiome Project”.
producing active biomolecules (antibiotics, antitumor . . .), or The project already processed over 200,000 samples from
for diversity studies, and so forth, and this would be possible across the globe for these microbial communities using
by adding few lines regarding the qualifier in need. metagenomics, metatranscriptomics, and amplicon sequenc-
Moreover, we suggest that the registration of informa- ing and started to generate huge amount of data to produce
tion regarding the qualifier “/country” should be obligatory. a global Gene Atlas describing protein space, environmental
Again, as it has been mentioned by NCBI Nucleotide, it metabolic models for each biome, approximately 500,000
has to be clear for the submitter that this qualifier is to reconstructed microbial genomes, a global metabolic model,
indicate the origin of the sequence. The geographical area’s and a data-analysis portal for visualization of all information
name indicated by the INSDC should be respected when [42].
registering or searching for data. We also suggest that regions
have to be defined to avoid ambiguity with a different
format, for example: uppercase, or put in another field. Abbreviations
Besides, the search for the qualifier “/country” should be NCBI: National Center of Biotechnology Information
facilitated by simple search word structure, for instance, GBIF: Global Biodiversity Information Facility.
CountryName[country] as applied for other search qualifiers,
for example: OrganismName[Organism] for organisms. The
methodology used in this study would also retrieve the Acknowledgments
diversity in particular regions within a geographical area The authors would like to thank Dr. Porter D. and Mr. Triki
of interest either by declaring it as previously described or M. for the help and the critical reading of this paper. The
adding it as a subcondition after the search. While the new authors would also like to thank the anonymous reviewers
qualifier “/lat lon available as 2005”, which indicates the GPS for the analysis and the enrichment of this paper. They would
coordinates for the location at which a specimen, from which like also to thank the Algerian Ministry of Higher Education
the sequence was obtained, was collected, it would be very and Scientific Research and the University of Warwick for
useful and more accurate to determine the strain origin. This supporting this work.
biogeography search for a particular region is much easier in
GBIF, simply by either using bounding box or introducing
coordinates (latitude, longitude, altitude, and depth) in the References
occurrence webpage as filters.
It would be also possible and interesting to associate [1] J. B. H. Martiny, B. J. M. Bohannan, J. H. Brown et al., “Microbial
biogeography: putting microorganisms on the map,” Nature
this biogeography study to ecological keywords which would
Reviews Microbiology, vol. 4, no. 2, pp. 102–112, 2006.
highly be recommended to be completed by users. This
association would be used in biodiversity informatics which [2] C. B. Cox and P. D. Moore, Biogeography: An Ecological and
surely generates worthwhile knowledge not only about the Evolutionary Approach, Blackwell Science, Oxford, UK, 2000.
presence of the microorganism but also about its probable [3] J. H. Brown, B. R. Riddle, and M. V. Lomolino, Biogeography,
involvement in the ecosystem function and its different Sinauer Associates, Sunderland, Mass, USA, 2005.
interactions. [4] I. M. Head, J. R. Saunders, and R. W. Pickup, “Microbial
One big challenge to the comparison of different data- evolution, diversity, and ecology: a decade of ribosomal RNA
bases is to cope with many different standards: for the regis- analysis of uncultivated microorganisms,” Microbial Ecology,
tration and the retrieval system, data structure, and even the vol. 35, no. 1, pp. 1–21, 1998.
[5] A. E. Mccaig, L. A. Glover, and J. I. Prosser, “Molecular [27] White House Office of Science and Technology Policy, “The
analysis of bacterial community structure and diversity in 2014 budget: a world-leading commitment to science &
unimproved and improved upland grass pastures,” Applied and research,” 2013, www.whitehouse.gov/ostp, http://www.white-
Environmental Microbiology, vol. 65, no. 4, pp. 1721–1730, 1999. house.gov/sites/default/files/microsites/ostp/2014 R&Dbudget
[6] N. Bano, S. Ruffin, B. Ransom, and J. T. Hollibaugh, “Phylo- overview.pdf.
genetic composition of arctic ocean archaeal assemblages and [28] J. H. Brown and M. V. Lomolino, Biogeography, Sinauer,
comparison with antarctic assemblages,” Applied and Environ- Sunderland, Mass, USA, 2nd edition, 1998.
mental Microbiology, vol. 70, no. 2, pp. 781–789, 2004. [29] W. B. Whitman, D. C. Coleman, and W. J. Wiebe, “Prokaryotes:
[7] J. C. Venter, K. Remington, J. F. Heidelberg et al., “Environmen- the unseen majority,” Proceedings of the National Academy of
tal genome shotgun sequencing of the sargasso sea,” Science, vol. Sciences of the United States of America, vol. 95, no. 12, pp. 6578–
304, no. 5667, pp. 66–74, 2004. 6583, 1998.
[8] R. I. Griffiths, B. C. Thomson, P. James, T. Bell, M. Bailey, and [30] P. Hugenholtz, B. M. Goebel, and N. R. Pace, “Impact of culture-
A. S. Whiteley, “The bacterial biogeography of British soils,” independent studies on the emerging phylogenetic view of
Environmental Microbiology, vol. 13, no. 6, pp. 1642–1654, 2011. bacterial diversity,” Journal of Bacteriology, vol. 180, no. 18, pp.
4765–4774, 1998.
[9] E. Pagaling, H. Wang, M. Venables et al., “Microbial biogeog-
raphy of six salt lakes in Inner Mongolia, China, and a salt lake [31] J. I. Prosser, B. J. M. Bohannan, T. P. Curtis et al., “The role
in Argentina,” Applied and Environmental Microbiology, vol. 75, of ecological theory in microbial ecology,” Nature Reviews
no. 18, pp. 5750–5760, 2009. Microbiology, vol. 5, no. 5, pp. 384–392, 2007.
[10] A. D. Chapman, “Quality control and validation of point- [32] M. C. Horner-Devine, M. Lage, J. B. Hughes, and B. J. M.
sourced environmental resource data,” in Spatial Accuracy Bohannan, “A taxa-area relationship for bacteria,” Nature, vol.
Assessment: Land Information Uncertainty in Natural Resources, 432, no. 7018, pp. 750–753, 2004.
K. Lowell and A. Jaton, Eds., pp. 409–418, Ann Arbor Press, [33] J. Green and B. J. M. Bohannan, “Spatial scaling of microbial
Chelsea, Mich, USA, 1999. biodiversity,” Trends in Ecology and Evolution, vol. 21, no. 9, pp.
[11] A. D. Chapman, “Uses of Primary Species-Occurrence Data, 501–507, 2006.
version 1.0,” Report for the Global Biodiversity Information [34] D. R. Nemergut, E. K. Costello, M. Hamady et al., “Global
Facility, Copenhagen, Denmark, 2005. patterns in the biogeography of bacterial taxa,” Environmental
[12] A. D. Chapman, “Principles of Data Quality, version 1.0,” Report Microbiology, vol. 13, no. 1, pp. 135–144, 2011.
for the Global Biodiversity Information Facility, Copenhagen, [35] J. C. Cho and J. M. Tiedje, “Biogeography and degree of
Denmark, 2005. endemicity of fluorescent Pseudomonas strains in soil,” Applied
[13] The Global Biodiversity Information Facility, http://data.gbif and Environmental Microbiology, vol. 66, no. 12, pp. 5448–5456,
.org/welcome.htm. 2000.
[36] R. J. Whitaker, D. W. Grogan, and J. W. Taylor, “Geographic
[14] NCBI Nucleotide Database, http://www.ncbi.nlm.nih.gov/nuc-
barriers isolate endemic populations of hyperthermophilic
core.
archaea,” Science, vol. 301, no. 5635, pp. 976–978, 2003.
[15] Bandwidth place, http://www.bandwidthplace.com/.
[37] H. Chu, N. Fierer, C. L. Lauber, J. G. Caporaso, R. Knight,
[16] Controlled vocabulary for /country qualifier, http://www.insdc and P. Grogan, “Soil bacterial diversity in the Arctic is not
.org/country.html. fundamentally different from that found in other biomes,”
[17] S. Federhen, “The NCBI taxonomy database,” Nucleic Acids Environmental Microbiology, vol. 12, no. 11, pp. 2998–3006, 2010.
Research, vol. 40, Database issue, pp. D136–D143, 2012. [38] N. Fierer and R. B. Jackson, “The diversity and biogeography
[18] NCBI Taxonomy, http://www.ncbi.nlm.nih.gov/taxonomy. of soil bacterial communities,” Proceedings of the National
Academy of Sciences of the United States of America, vol. 103, no.
[19] M. Ruggiero, D. Gordon, N. Bailly, P. Kirk, and D. Nicolson,
3, pp. 626–631, 2006.
“The catalogue of life taxonomic classification,” in Species 2000
& ITIS Catalogue of Life, 2011 Annual Checklist, F. A. Bisby, Y. R. [39] K. van der Gucht, K. Cottenie, K. Muylaert et al., “The power
Roskov, A. Culham et al., Eds., Part A, Species 2000, Reading, of species sorting: local factors drive bacterial community
UK, 2nd edition, 2011. composition over a wide range of spatial scales,” Proceedings of
the National Academy of Sciences of the United States of America,
[20] The Global Biodiversity Information Facility, The species portal, vol. 104, no. 51, pp. 20404–20409, 2007.
http://data.gbif.org/species/.
[40] M. C. Horner-Devine, J. M. Silver, M. A. Leibold et al., “A
[21] GBIF Database occurrence search webpage, http://data.gbif.org/ comparison of taxon co-occurrence patterns for macro- and
occurrences/. microorganisms,” Ecology, vol. 88, no. 6, pp. 1345–1353, 2007.
[22] Python, http://www.python.org/about/. [41] D. Remsen, B. Ko, V. Chavan, and M. Raymond, Getting Started:
[23] E. Sayers, “A general introduction to the E-utilities,” 2009, An Overview of Data Publishing in the GBIF Network, Global
http://www.ncbi.nlm.nih.gov/books/NBK25497/#chapter2.In- Biodiversity Information Facility, Copenhagen, Denmark, 2011,
troduction. http://links.gbif.org/getting started publishing en v1.
[24] Biopython, http://biopython.org/wiki/Biopython. [42] The Earth Microbiome Project, http://www.earthmicrobiome
[25] A. South, “rworldmap: new R package for mapping global data,” .org/.
The R Journal, vol. 3, no. 1, pp. 35–43, 2011.
[26] The R Project for Statistical Computing, http://www.r-project
.org/.
http://dx.doi.org/10.1155/2013/784591
Review Article
The Microbiology of Olive Mill Wastes
Spyridon Ntougias,1 Kostas Bourtzis,2 and George Tsiamis2

1
Department of Environmental Engineering, Democritus University of Thrace, Vas. Sofias 12, 67100 Xanthi, Greece
2
Received 30 March 2013; Revised 18 July 2013; Accepted 22 July 2013
Academic Editor: Ameur Cherif
Copyright © 2013 Spyridon Ntougias et al. This is an open access article distributed under the Creative Commons Attribution
cited.
Olive mill wastes (OMWs) are high-strength organic effluents, which upon disposal can degrade soil and water quality, negatively
affecting aquatic and terrestrial ecosystems. The main purpose of this review paper is to provide an up-to-date knowledge
concerning the microbial communities identified over the past 20 years in olive mill wastes using both culture-dependent and
independent approaches. A database survey of 16S rRNA gene sequences (585 records in total) obtained from olive mill waste
environments revealed the dominance of members of Alphaproteobacteria, Betaproteobacteria, Gammaproteobacteria, Firmicutes,
and Actinobacteria. Independent studies confirmed that OMW microbial communities’ structure is cultivar dependant. On the
other hand, the detection of fecal bacteria and other potential human pathogens in OMWs is of major concern and deserves further
examination. Despite the fact that the degradation and detoxification of the olive mill wastes have been mostly investigated through
the application of known bacterial and fungal species originated from other environmental sources, the biotechnological potential
of indigenous microbiota should be further exploited in respect to olive mill waste bioremediation and inactivation of plant and
human pathogens. The implementation of omic and metagenomic approaches will further elucidate disposal issues of olive mill
wastes.
1. Introduction named as alpeorujo in Spanish [4] and abbreviated herein as

TPOMW (two-phase olive mill waste). Both types of wastes
Disposal of olive mill wastes is an important environmental are characterized by undesirable color and odor, acidic pH,
problem in olive-oil producing countries since they are high salt concentration, and total polyphenolics content. In
generated in huge quantities in a short period of time. addition, OMWs are characterized by high chemical oxygen
Olive oil is mainly produced in the Mediterranean countries, demand (COD) values, whereas TPOMW possesses high
although other producers, such as Argentina, Australia and organic matter and low water activity (𝑎𝑤 ). The physico-
Chile, are facing the toxic effects of olive mill wastes [1]. chemical properties of OMW and TPOMW are presented in
The major olive oil producing countries are Spain, Italy, and Table 1. Due to the high organic load and the elevated salt and
Greece, with a production of 1150, 560, and 370 thousand tons polyphenols content, olive mill wastes are significant sources
annually, respectively, followed by Tunisia and Turkey, with of environmental pollution. Their effective management is
an annual production of 160 thousand tons each [2]. Two and negatively affected by the seasonal operation and the high
three-phase centrifugal decanters are commonly used in the territorial scattering of the olive mill sites [5].
extraction of the olive oil, with two-phase extraction systems Olive mill wastes inhibit seed germination and early plant
gaining ground due to the lower amount of water consumed growth [16], alter soil characteristics [17], and create reduc-
for the malaxation of the olive paste [3]. ing conditions, affecting microbial diversity in soil [18]. In
Three-phase extraction systems result in the production contrast, olive mill waste phenolics may be used in food and
of olive oil, olive press cake, and a liquid waste, commonly chemical industries as natural antioxidants and disinfectants
known as olive mill wastewater (OMW), while two-phase [19, 20]. Olive mill wastes may be used for biopolymer and
decanters produce olive oil and a viscous sludge-like waste, biogas production as well as being fertilizer/compost and
Table 1: Basic OMW and TPOMW physicochemical characteristics. Members of the phyla Acidobacteria, Planctomycetes, Gem-
matimonadetes, Verrucomicrobia and the candidate divisions
Characteristic OMW1 Characteristic TPOMW2 OP3, TM7, AD3, marine group A, and SPAM were minor
(values range) (values range)
constituents of the bacterial biota. DNA microarray devel-
pH 4.01–5.93 pH 4.86–6.45 opment enabled researchers to detect microbial sequences
BOD (g L−1 ) 8.0–38.7 OM (%) 49.5–98.5 from any sample in a parallel and very fast high throughput
Total phenolics manner. The studies of Tsiamis et al. [30] and Goberna et
COD (g L−1 ) 28.6–186 0.5–2.4
(%) al. [31] pioneered the use of such approaches in studying
TOC (g L−1 ) 1.89–38.0 Total N (g kg−1 ) 7.0–18.5 in depth bacterial (PhyloChip) and archaeal (methanogenic)
TS (g L−1 ) 3.13–78.2 Total P (g kg−1 ) 0.5–2.2 communities (ANAEROCHIP) in OMW and anaerobically
Total phenolics digested TPOMW, respectively.
0.03–18.9 Total K (g kg−1 ) 6.3–29.7 Cultivation and harvesting practice highly affect bacterial
(g L−1 )
Total N (g L−1 ) 0.02–2.10 community structure in OMW. Bacterial diversity in OMW
−1
from “Mastoidis” variety was dominated by fermentative
Total P (g L ) 0.01–1.00
members of Bacteria, such as lactic acid (Lactobacillus and
−1
Total K (g L ) 0.17–7.81 Oenococcus spp.) and acetic acid (Acetobacter and Gluconace-
1 tobacter spp.) bacteria as well as fecal bacteria related to the
Based on data reported in Aktas et al. [6], Ammary [7], Zenjari et al. [8],
Amaral et al. [9], Eroǧlu et al. [10], Aviani et al. [11], and Ntougias et al. [12]. family Prevotellaceae and the Ruminococcus-Eubacterium-
2
Based on data reported in Alburquerque et al. [13], Vlyssides et al. [14], and Clostridium (REC) cluster [29]. The proliferation of such
Baeta-Hall et al. [15].
community structure is attributed to the accumulation of
olives in the harvest net which can lead to the establishment of
anaerobic/microaerophilic niches, while their processing in
substrate for growing edible fungi [21–24]. Indeed, these olive mills can increase the oxygen level, favoring the growth
wastes consist of a wide range of valuable resources, such as a of acetic acid bacteria [29]. Members of Betaproteobacteria
high proportion of organic and inorganic nutrients, that can (families Comamonadaceae, Gallionellaceae, Hydrogenophi-
be recycled [20, 25]. laceae, Methylophilaceae, Oxalobacteraceae and Rhodocy-
Several physicochemical and biological treatment ap- claceae), Gammaproteobacteria (families Pasteurellaceae and
proaches have been applied for the degradation and detox- Xanthomonadaceae) and Firmicutes (families Bacillaceae,
ification of olive mill wastes, for example, implementation Paenibacillaceae, Peptococcaceae and Sporolactobacillaceae)
of advanced oxidation systems, aerobic biotreatment, and were identified in O. europaea var. koroneiki-generated OMW
anaerobic digestion [21, 26–28], although the review of treat- [29]. As a result of the early collection and harvesting practice
ment technologies for both OMW and TPOMW is beyond (collection by hand), fermentative bacteria in O. europaea var.
the scope of this paper. koroneiki-generated OMW were restricted to a few represen-
In this review, the microbiology of olive mill wastes is tatives of the families Peptococcaceae and Sporolactobacillace-
examined in depth, and special focus is given to: (a) the ae. The identification of fecal bacteria in Olea europaea var.
microbial ecology of olive mill wastes, (b) OMW-induced mastoidis-generated OMW, due to the prolonged harvesting
toxicity, (c) the effects of olive mill wastes on soil microbial period, is of concern [29].
communities, (d) the microbial ecology in bioreactors treat- Vivas et al. [32] found that TPOMW was dominated by
ing olive mill effluents, and (e) the potential biotechnological members of the phylum Proteobacteria, followed by Acti-
application of olive mill waste microbiota. nobacteria (Streptomyces), Firmicutes (Staphylococcus) and
uncultured Acidobacteria strains as minor constituents of
2. Microbial Ecology of Olive Mill Wastes olive waste microbiota. Members of Hydrocarboniphaga,
Pseudoxanthomonas and Stenotrophomonas (Gammaproteo-
2.1. Bacterial Diversity in Olive Mill Wastes. The majority of bacteria) were identified, while Comamonas (Betaproteobac-
OMW microbiota are originated from soil and freshwater teria) was the main microbial group detected. Moreover,
environments, while fecal bacteria have been also identified a Brevundimonas sp. was the only representative within
[29, 30]. Bacterial community structure is greatly influenced Alphaproteobacteria [32].
by the specific cultivar from which OMWs are generated In order to face disposal problems related to acidic pH
[30]. Bacterial communities in OMW generated from various and undesirable odor, addition of Ca(OH)2 to TPOMW
olive-fruit varieties had only 15% of the OTUs identified in results in the formation of an alkaline secondary waste (alka-
common, indicating a cultivar-dependent microbial profile line TPOMW) [4]. This pH change favors alkalitolerant
[30]. In all OMW samples examined by Tsiamis et al. and alkaliphilic bacteria with some degree of halophilicity
[30], the cultured bacterial diversity consisted of members [4]. Halotolerant alkaliphiles related to the genera Bacillus,
of Firmicutes, Actinobacteria, Alphaproteobacteria, Betapro- Idiomarina, Halomonas, and Nesterenkonia as well as alkali-
teobacteria, Gammaproteobacteria and Bacteroidetes, while tolerant and/or halotolerant bacteria associated phylogenet-
the implementation of a high density DNA microarray ically with the genera Corynebacterium, Novosphingobium,
(PhyloChip) revealed a broader diversity, which was domi- Ochrobactrum, Pseudomonas, Rhodobacter, and Serratia have
nated by members of all classes of Proteobacteria, Firmicutes, been identified in alkaline TPOMW [4]. The majority of those
Bacteroidetes, Chloroflexi, Cyanobacteria and Actinobacteria. isolates could effectively utilize phenolic compounds as the
1.9
0.7 Alphaproteobacteria, Actinobacteria and Bacteroidetes
account for approximately 17, 12, 9 and 7% of the bacterial
6.8 11.6
phylotypes identified in olive mill wastes, respectively
(Figure 1). Phylotypes’ distribution in Alphaproteobacteria,
9.2 Betaproteobacteria, Gammaproteobacteria, Firmicutes, Actin-
obacteria, and Bacteroidetes is presented in Figure 2.
Based on the above analysis (Figure 2), Enterobacteri-
21.5
aceae, Moraxellaceae, Xanthomonadaceae and Pseudomon-
17.3 adaceae spp. are the main representatives of Gammapro-
teobacteria. In addition, Oxalobacteraceae and Comamon-
adaceae strains are the predominant Betaproteobacteria in
olive mill wastes, while Acetobacteraceae is the dominant
1.2 taxon within Alphaproteobacteria. In olive mill wastes, Bacil-
29.7
laceae, Clostridiaceae, Lactobacillaceae and Paenibacillaceae
Alphaproteobacteria Actinobacteria are the most abundant taxa within the phylum Firmicutes,
Betaproteobacteria Bacteroidetes while actinobacterial phylotypes were mainly placed in the
Gammaproteobacteria Chloroflexi families Micrococcaceae, Microbacteriaceae and Propionibac-
Deltaproteobacteria Synergistetes teriaceae. Bacteroidetes phylotypes identified in olive mill
Firmicutes wastes were associated with the families Prevotellaceae, Por-
phyromonadaceae and Sphingobacteriaceae. Indeed, olive mill
Figure 1: Distribution of bacterial phylotypes identified in olive mill
waste environments.
waste environments are dominated by bacterial taxa that are
specialized in degrading the recalcitrant components of olive
mill wastes [29, 30].
Approximately 20% of the bacterial phylotypes identified
sole carbon source [4]. Olive pomace microbiota appear to in olive mill wastes and olive mill waste-related environments
alter their membrane lipids in an atypical manner in order are associated with coliforms (e.g., Citrobacter, Escherichia,
to be adapted to stressful conditions, such as the lowered Klebsiella, and Serratia spp.) and other enteric bacteria [36],
water activity (𝑎𝑤 ), the low acidity, and the high polyphenolic like Porphyromonadaceae, Prevotellaceae, Lachnospiraceae,
content of olive mill wastes [33]. Eubacteriaceae, Peptococcaceae, Peptostreptococcaceae, and
Staphylococcus spp. in olive mill waste should be con- Ruminococcaceae spp. The above findings necessitate the need
sidered as potential infectious agents. Based on microbial for safe disposal of OMWs.
counts and API identification, Enterobacter cloacae strains
were frequently detected in the raw effluent, followed by 2.2. Fungal Diversity in Olive Mill Wastes. Yeast population
members of the species Aeromonas hydrophila, Pseudomonas appears to be high in olive mill wastes [17]. Yeasts related
aeruginosa, and Serratia odorifera [34], while Citrobacter to Geotrichum (G. candidum), Candida (C. membranifaciens,
braakii was predominant during acidogenesis. Representa- C. michaelii, C. inconspicua, and C. tropicalis), Pichia (P.
tives of the species Burkholderia cepacia, Enterobacter cloacae, fermentans and P. holstii), Rhodotorula (R. mucilaginosa),
and Photobacterium damselae were also detected. Moreover, and Saccharomyces (S. cerevisiae) have been recently isolated
high counts of Acinetobacter, Pseudomonas, and Enterobacter from OMW [37]. In accordance, Candida boidinii, Pichia
spp. have been determined in OMW [34]. Interestingly, holstii (syn. Nakazawaea holstii), P. membranifaciens, and
Enterobacter spp. are potential human pathogens. Saccharomyces cerevisiae were the predominant yeasts in
It is concluded that members of Alphaproteobacteria, OMW from Apulia (Italy), exhibiting high pectolytic and
Betaproteobacteria and Gammaproteobacteria as well as Fir- xylanolytic activities. These yeast isolates could effectively
micutes and Actinobacteria are the main bacterial repre- reduce total phenolics, resulting in the reduction of several
sentatives in olive mill wastes (both OMW and TPOMW). phenolic compounds, in particular p-coumaric, vanillic and
Despite the fact that the distribution analysis of the 16S caffeic acids [38]. Pichia (P. guilliermondii–syn. Meyerozyma
rRNA gene sequences deposited in international databases guilliermondii) and Candida (C. diddensiae and C. ernobii)
may be subjected to bias due to PCR amplification and/or spp. were also the main yeast biota in OMW from Moroccan
the specific focus of each research work, for example, olive mills [39].
examination of tannase-expressing communities [35], survey Pichia caribbica (syn. Meyerozyma caribbica), P. holstii
of 16S rRNA gene databases revealed the presence of 585 (syn. Nakazawaea holstii), and Zygosaccharomyces fermentati
deposited sequences of bacteria identified in olive mill waste (syn. Lachancea fermentati) were the predominant yeast
environments. Analysis of these phylotypes confirms the taxa in TPOMW, while Z. florentinus (syn. Zygotorulaspora
placement of the majority of olive mill waste microbiota in florentina), Lachancea thermotolerans (syn. Kluyveromyces
Alphaproteobacteria, Betaproteobacteria, Gammaproteobacte- thermotolerans), Saccharomyces cerevisiae, and S. rosinii
ria, Firmicutes and Actinobacteria (Figure 1). (syn. Kazachstania rosinii) were minor constituents of the
The representatives of Betaproteobacteria and Gamma- yeast community [40]. Some of the yeast isolates from
proteobacteria exceed 50% of the respective 16S rRNA TPOMW exhibited cellulase, 𝛽-glucanase, 𝛽-glucosidase,
gene sequences deposited in GenBank, while Firmicutes, peroxidase, and polygalacturonase activities which could
0.6 Gammaproteobacteria 0.8 Betaproteobacteria

0.6
0.6 3.2 4.0
4.6 1.6
3.2
4.8
12.1
41.4
16.1
54.0
28.6
24.1
Enterobacteriaceae Halomonadaceae Oxalobacteraceae Rhodocyclaceae

Moraxellaceae Pasteurellaceae Comamonadaceae Burkholderiaceae
Xanthomonadaceae Idiomarinaceae Alcaligenaceae Methylophilaceae
Pseudomonadaceae Sinobacteraceae Hydrogenophilaceae Unclassified
Betaproteobacteria
Alphaproteobacteria 1.0 1.0 1.0 Firmicutes

1.5
2.9 1.5 1.5 1.5 2.0 2.0
2.9 2.0 2.0
1.0
4.4 3.0
3.0 24.8
4.4
3.0
5.9 4.0
5.9 8.9
67.6 20.8
20.8
Acetobacteraceae Brucellaceae Bacillaceae Planococcaceae

Caulobacteraceae Bradyrhizobiaceae Clostridiaceae Eubacteriaceae
Methylobacteriaceae Erythrobacteraceae Lactobacillaceae Peptococcaceae
Rhodobacteraceae Rhizobiaceae Paenibacillaceae Veillonellaceae
Rhodospirillaceae Unclassified Lachnospiraceae Aerococcaceae
Sphingomonadaceae Alphaproteobacteria Leuconostocaceae Peptostreptococcaceae
Sporolactobacillaceae Ruminococcaceae
Staphylococcaceae Unclassified Clostridiales
Actinobacteria Bacteroidetes
1.9
1.9 1.9
3.7 5
3.7 5
10
5.6
5.6
7.4 51.9 12.5
67.5
16.7
Micrococcaceae Dermabacteraceae Prevotellaceae Unclassified

Microbacteriaceae Streptomycetaceae Porphyromonadaceae Sphingobacteriales
Propionibacteriaceae Conexibacteraceae Sphingobacteriaceae Unclassified Bacteroidetes
Brevibacteriaceae Corynebacteriaceae
Nocardiaceae Nocardioidaceae
Figure 2: Distribution within Alphaproteobacteria, Betaproteobacteria, Gammaproteobacteria, Firmicutes, Actinobacteria and Bacteroidetes
of bacterial phylotypes identified in olive mill waste environments.
pH of olive mill wastes may be advantageous for this micro-

bial group to outcompete bacteria. Filamentous fungi, such
17% 19% as Aspergillus and Penicillium spp., are common habitants of
olive mill wastes [43, 45], while white-rot fungi have been
3%
isolated to a lesser extent. It appears that the high salt and
sugars concentrations of olive mill wastes as well as the acidic
pH favor the growth of osmotolerant yeasts in olive mill
wastes [46].
3. OMW-Induced Toxicity
Olive mill wastes are toxic to both microorganisms and
61%
aquatic organisms. OMWs from traditional mills appear
more toxic since their effluents are more concentrated than
Glomeromycota Ascomycota those from continuous extraction systems [17, 39]. Olive
Basidiomycota Unclassified fungi mill effluents negatively affect the aquatic fauna of fluvial
ecosystems as a consequence of both high organic load
Figure 3: Distribution of fungal phylotypes identified in olive mill and fecal contamination [47]. Based on Thamnotox kit and
waste environments. zebrafish embryo tests, raw OMW can be characterized
as “extremely toxic” and they can retain a significant part
of their toxicity even after biotreatment [34]. Moreover,
OMW toxicity can reach maximum levels in Aliivibrio fis-
contribute to the degradation of complex compounds, includ- cheri bioluminescence assays [48]. The phenolic fraction of
ing olive pomance phenolics [40]. Based on the data provided olive mill wastes has been reported to exhibit antimicrobial
by Romo-Sánchez et al. [40], yeast diversity in olive pomance activity against nonindigenous Bacillus subtilis, Escherichia
appears to be variety dependent. coli, Pseudomonas aeruginosa, and Staphylococcus aureus
Survey of the National Center for Biotechnology Infor- strains, which is even greater than the respective activities
mation (NCBI) revealed the presence of 106 deposited induced by the individual phenolic compounds, indicating
sequences of fungi identified in olive mill waste environ- the synergistic action of olive mill waste phenolics [49].
ments. Analysis of these sequences confirms the placement Experimental data indicate that individual phenolic com-
of the majority of fungal phylotypes from olive mill wastes pounds at low concentrations cannot inhibit the growth of
in Glomeromycota, Basidiomycota, Ascomycota, and fungi the human pathogens Escherichia coli, Klebsiella pneumoniae,
that have not been assigned (Figure 3). The representatives Staphylococcus aureus, and Streptococcus pyogenes, while
within the Basidiomycota occupy more than 60% of the OMWs exhibit strong inhibitory effects against both Gram-
fungal phylotypes deposited in GenBank. Members of Glom- positive and Gram-negative bacteria [50]. OMW extracts
eromycota and unclassified fungi correspond to 19 and 17% in combination with gallic acid could effectively inhibit the
of the respective records, while members of Ascomycota growth of the above human pathogens at concentrations
comprise only 3% of the total representatives. However, lower than 100 𝜇g mL−1 [50].
Ascomycota population is underestimated since Basidiomy- In earlier studies, OMW toxicity was attributed to
cota and Glomeromycota specific primers were applied for low molecular weight phenolics, in particular monomeric
the majority of the phylotypes identified [41, 42]. phenolic compounds [51]. However, recent findings have
Members of the fungal genera Acremonium, Alternaria, showed that other factors also contribute to OMW acute
Aspergillus, Chalara, Fusarium, Lecythophora, Paecilomyces, toxicity, since reduction of monomeric phenolics cannot
Penicillium, Phoma, Phycomyces, Rhinocladiella, and Scop- necessarily lead to mitigation of toxicity [52]. Despite the
ulariopsis have been identified in OMW disposal ponds, fact that the mechanism of OMW-induced toxicity remains
possessing the ability to detoxify olive mill effluents [43]. unclear, several OMW compounds, including phenolics, may
However, the identification of these fungi was based on their cause a narcotic action to seeds and early plants, as the
morphology and not on molecular techniques. Members of result of a noncovalent membrane interaction [52]. Bioactive
the fungal genera Cerrena, Byssochlamys (syn. Paecilomyces), intermediate compounds derived from the transformation
Lasiodiplodia, and Bionectria are indigenous microbiota of phenolics may also be toxic [52]. OMW has been also
(identified by molecular techniques) which have the ability reported to decrease the phosphorylation efficiency of mito-
to degrade OMW phenolics [44]. chondria, probably as the result of structural changes induced
Based on the above-mentioned studies, Pichia, Candida, in the inner mitochondrial membrane by OMW organic
and Saccharomyces-like species are the predominant yeasts in compounds (e.g., fatty acids) [53]. Phenolic compounds,
olive mill wastes. Reduction of both phenolics and sugars is such as p-coumaric acid and ferulic acid, can also affect the
the main metabolic function of yeasts in olive mill wastes, physiology of both prokaryotic and eukaryotic organisms
while they appear to contribute less in OMW decolorization [54]. Moreover, other OMW components, for example, lipids,
in comparison to white-rot fungi [39]. Moreover, the acidic cannot be excluded. In addition, the low pH and the osmotic
stress caused by the presence of high Na+ and Cl− concentra- of spore-forming bacteria reported in the above studies,
tions may play a role in OMW acute toxicity [54]. lower Gram-positive:Gram-negative FAME quotients were
Microbial communities in OMW may be directly observed during consecutive OMW application [66]. This
involved in acute toxicity mainly against aquatic biota. may indicate a metabolically active Gram-negative bacterial
Venieri et al. [34] reported that specific indigenous microbial population in comparison to a dormant spore-forming pop-
taxa, that is, Aeromonas hydrophila and Enterobacter cloacae, ulation. The rise in total soil microbial population is also
negatively affected the aquatic crustacean Thamnocephalus evidenced by the increased soil respiratory activity deter-
platyurus, linking the microbial activity of certain OMW mined, which is linked to organic substrates degradation and
indigenous microbiota with the toxicity on aquatic assimilation [60, 67]. Besides, the relative abundance of fungi
organisms. This emphasizes the necessity of assessing over bacteria in soils receiving long-term olive mill waste
microbial communities in OMW, not only for bioremediation applications can be attributed to the early decomposition of
purposes but also for safe disposal. the labile organic matter [68]. In addition, Actinobacteria and
On the other hand, olive mill waste total phenolics yeasts as well as several Gram-negative taxa, for example,
content can be used to control and inactivate plant and Pseudomonas spp., are considered as effective degraders
human pathogens. Yangui et al. [55] revealed the fungi- of recalcitrant compounds [69, 70]. Similarly, prolonged
cidal action of hydroxytyrosol-rich OMW extract on the storage of OMW in evaporation ponds increases fungi-to-
soil-borne plant pathogen Verticillium dahliae. Moreover, bacteria ratio [71]. However, apart from the increase in
OMW- and TPOMW-phenolic compounds can be used in mesophilic population, elevated total coliforms counts are
the inactivation of pathogenic bacteria and their toxins. observed during successive OMW applications onto soils
Administration of the phenolic substrate 4-hydroxytyrosol [59].
can inactivate Staphylococcus aureus without being cytotoxic Application of TPOMW onto soils results in an increase
to spleen cells, reducing in parallel the biological activity in fungal diversity and in a consequent decrease in bacterial
of the staphylococcal enterotoxin A [56]. In addition, the diversity [72]. Moreover, phospholipid fatty acid (PLFA) anal-
molluscicidal activity of olive mill waste phenolics has been ysis in TPOMW-amended soils revealed a gradual decrease
reported [49], while TPOMW extracts have been shown to of Gram-positive to Gram-negative bacteria, while microbial
exhibit suppressive properties against common weeds and activity, as determined by dehydrogenase and fluorescein
nematodes [57]. diacetate hydrolase assays, is stimulated in olive mill waste
Several studies have proven the negative effects of these amendments [72].
wastes on soil microbial populations, on aquatic ecosystems, Extended OMW dose applications can increase the abun-
and even in air quality [58]. This realization enforces the dance of the soil denitrifying communities, although the
need for microbial risk assessment during disposal of olive nitrifying population is suppressed as a result of the reducing
mill wastes. There is, therefore, a necessity for guidelines power of OMW phenolics [60]. In particular, ammonia-
to manage these wastes through technologies that minimize oxidizing bacteria (AOB) are highly suppressed in the pres-
the environmental impact and lead to a sustainable use of ence of OMW [59]. At the same time, members of cluster 3 of
resources. Nitrosospira are proliferated [18]. Consequently, inhibition of
nitrification process due to OMW applications on soils affects
the soil nitrogen cycle [59].
4. Effects of Olive Mill Wastes and OMW applications highly influence soil basidiomycete
Olive Mill Wastes-Derived Composts on communities, although N fertilization alleviates these effects
Soil Microbiota [41]. Changes in basidiomycete community structure are
attributed to organic matter addition and N immobilization
4.1. Effects of Olive Mill Waste Spreading on Soil Microbiota. occurring during OMW spreading [41].
Soil microbial activity appears to be enhanced during olive OMW application to soils can lead to transient changes in
mill waste land application. Controlled OMW spreading the arbuscular mycorrhizal colonization of Vicia faba plants
can increase the total soil microbial population, which can by Glomus spp. [42]. However, their population is restored
be accompanied by respective increase in the population after long-term plant growth. At dose application greater
abundance of spore-forming bacteria, Actinobacteria, and
than 30 m3 ha−1 , OMW spreading on olive tree rhizosphere
yeasts [59–63]. OMW applications in loamy soils can affect
soils has been reported to decrease soil and root 16:1𝜔5
bacterial community structure, due to the high availability
FAME biomarker as well as olive tree photosynthetic rates,
of OMW phenolics in this type of soil [64]. By using
fatty acids methyl esters (FAME) analysis, Mechri et al. indicating that arbuscular mycorrhizal population may be
[65, 66] examined the relative abundance of Actinobacte- suppressed under high OMW dose applications [65]. This
ria and fungi under successive OMW dose applications. decrease in the relative proportion of 16:1𝜔5 biomarker was
OMW application influences the actinobacterial community attributed to the increased C/N ratio, total P, and phenolics
structure only in loamy sand soils, indicating soil-dependent concentrations determined after long-term spread of olive
effects [18]. Lowered oxidative conditions, availability of mill effluents [65]. Sampedro et al. [73] stated that the input
phenolics, and N immobilization are the main environmen- of arbuscular mycorrhizas to the plants growing in olive mill
tal factors responsible for shifts in microbial communities dry residue amendments depended on the type of the plant
during OMW spreading [18]. In contrast to the increase and the arbuscular mycorrhizal species.
4.2. Microbial Diversity in Olive Mill Wastes-Based Com- 5. Microbial Community Structure
posts and the Effects of Their Amendments on Soil Micro- in Bioreactor Systems Treating Olive
biota. Functional diversity in olive mill wastes has been Mill Wastes
reported to be relatively low, although increasing during
aerobic treatment [32]. In comparison with untreated olive Bacterial diversity has been investigated during acidogenesis
wastes, composting or vermicomposting showed higher of OMW in anaerobic packed-bed biofilm reactors supported
dehydrogenase, 𝛽-glucosidase, and urease activities [32] as with either granular activated carbon or ceramic cubes.
a result of the transformation of phenolics. Composted Development of acidogenic biofilm in granular activated car-
olive mill wastes and byproducts commonly exhibit high bon resulted in a bacterial community structure, which con-
extracellular enzyme activities, although lower activities have sisted mainly of Betaproteobacteria and Gammaproteobacte-
been determined for enzymes, which function is linked ria (Acinetobacter, Comamonas and Massilia spp.), followed
directly to metabolically active microbial cells. Indeed, by representatives of Clostridiales, Alphaproteobacteria and
activities of enzymes that can be extracellularly secreted, Chloroflexi. On the other hand, ceramic cubes favored the
such as esterases and ureases, are relatively high [22, 32, biofilm formation from Bacillus, Clostridium, Paenibacillus
74, 75]. Carbon and nitrogen content in olive mill waste- and Pasteuriaceae strains [80]. The dominance of Lacto-
based composts appears to influence the functional and bacillus and Acetobacter spp. in the OMW used [80] indi-
catabolic diversity of indigenous microbiota [75]. More- cated that olive mill effluent was naturally fermented during
storage. Removal of OMW phenolics through a resin and
over, Fernández-Gómez et al. [75] showed that TPOMW
acidogenesis of permeate in mesophilic anaerobic packed-
microbiota possess the ability to oxidize plethora of C
bed biofilm reactor resulted in the proliferation of bacterial
substrates.
communities consisting almost exclusively of members of
Composting or vermicomposting can shift bacterial di-
Firmicutes, that is, Clostridium, Anaerotruncus colihominis,
versity, resulting in the abundance of Alphaproteobacteria
Ethanoligenens harbinense, and Syntrophobotulus glycolicus
and Actinobacteria in relation to Betaproteobacteria [32]. strains [81]. Actinomyces suimastitidis and Staphylococcus felis
Fernández-Gómez et al. [76] reported that olive-mill waste were minor constituents of the biofilm formed [81]. Investi-
and biosolids-based vermicomposts were dominated by Mi- gation of microbial dynamics in a granular activated carbon
crobacterium, Pseudomonas, Streptomyces and Sphingobac- packed-bed anaerobic bioreactor fed with OMW resulted
terium spp. Both Actinobacteria and Sphingobacteria are in the identification of members of Gammaproteobacteria,
involved in the biostabilization of complex compounds. Deltaproteobacteria, and Bacteroidetes as well as fermentative
A significant part of indigenous microbiota with degrad- bacteria of the genera Clostridium (Firmicutes) and Anaer-
ing ability is involved in the initial attack of recalcitrant obaculum (Synergistetes) [82]. Syntrophus spp., which were
components of olive mill wastes [35]. Olive mill waste- in syntrophic relationship with methanogenic Archaea, were
based composts appear to favor growth of bacterial com- detected, while sulphate-reducing bacteria were identified as
munities, which are specialized in organic matter decom- the result of the high sulphate concentration in the OMW
position, in particular of phenolics, tannins and lipids. digested. Next to Clostridia, which are predominant during
For instance, tannase-expressing bacterial communities, acidogenesis, Syntrophus and Chloroflexi-like bacteria appear
which consisted of members of the phyla Actinobacte- to be also common inhabitants of phenolic wastewaters [83].
ria (Kocuria, Microbacterium, Micrococcus and Rhodococ- In fact, fermentative Clostridia convert phenolic derivatives
cus spp.), Firmicutes (Bacillus, Lysinibacillus, and Staphylo- to benzoate, which is subsequently transformed by Syntro-
coccus spp.), and Proteobacteria (Acinetobacter, Advenella, phus-like strains to acetate and H2 /CO2 [84]. In syntrophic
Pseudomonas and Pusillimonas spp.), were identified in association, methanogens then form methane from acetate
TPOMW-based composts [35]. Moreover, various Bacillus and H2 /CO2 . Despite the fact that acetate was the main
spp. isolated from OMW exhibit strong lipolytic activities volatile fatty acid identified, Bertin et al. [82] reported that
[77]. the hydrogenotrophic Methanobacterium formicicum was
Olive mill waste-based composts have been proposed to the predominant archaeon detected. Moreover, Rizzi et al.
be beneficial in the bioremediation of contaminated soils [85] reported a relative increase/decrease in Methanomi-
due to the presence of microbial consortia with degradation crobiaceae/Methanobacterium population by increasing the
ability. Olive mill waste-based vermicompost has been used organic loading rate in an upflow anaerobic filter treat-
ing OMW, while no effect on Methanosaeta-like popula-
for the bioremediation of trichloroethylene-contaminated
tion was observed. This indicates that hydrogenotrophic
soils through the dominance of bacterial communities related
methanogenic population is highly affected by the organic
to the phyla Proteobacteria and Acidobacteria, followed by
loading rate applied during anaerobic digestion of olive
members of Bacteroidetes, Actinobacteria and Gemmatimon- mill wastes, while acetoclastic methanogens remain almost
adetes [78]. TPOMW-based vermicompost has been also unaffected.
applied as organic amendment for the bioremediation of The microbial diversity during treatment of TPOMW-
PAH (polycyclic aromatic hydrocarbons)-contaminated soils, based mixtures in aerated and nonaerated bioreactors has
resulting in changes in bacterial communities which led to been also investigated [1], with the microbial community
enhanced naphthalene dioxygenase activity [79]. structure being studied in a limited number of aerobic
treatment systems. Nutrients (N and P) addition could 6. Features of Biotechnological Importance in

decrease the phenolic content in the aerated bioreactors Olive Mill Wastes Microbiota
[1]. The increase in the relative fungal/bacterial ratio was
accompanied by high polyphenolic and organic matter reduc- 6.1. Biodegradation of Olive Mill Wastes Using Indigenous and
tion [1], indicating that the stimulation of indigenous fungi Selected Microbial Strains. Basidiomycetous and ascomyce-
can effectively detoxify olive mill wastes. This microbial tous yeasts, white-rot fungi, and Aspergillus and Penicillium
shift permits direct use of the OMW without the need for spp. are commonly used for the in vitro dephenolization
inclusion of external degraders. Predominant fungi during and/or decolorization of olive mill wastes, whereas bacterial
aerobic treatment of TPOMW-based mixtures were Peni- inocula have been also applied. Most of the microorganisms,
which are used in the degradation and detoxification of
cillium roqueforti, Candida norvegica and Geotrichum sp.
OMW and TPOMW, have been isolated from other environ-
Nonaerated olive mill waste-derived mixtures were dom-
mental sources, although only a few degraders belong to the
inated by fungi, which were related to the species Pichia
indigenous microbiota of olive mill wastes.
membranifaciens and Cladosporium herbarum as well as the OMW indigenous bacteria are capable of degrading
genera Ascochyta and Geotrichum [1]. Aerated olive mill various single-ring aromatic components of olive mill efflu-
waste amendments are commonly dominated by members ents. Members of Alphaproteobacteria, Betaproteobacteria
of Gammaproteobacteria, such as Stenotrophomonas mal- and Gammaproteobacteria, such as Comamonas, Ralstonia
tophilia and Luteibacter sp., whereas anaerobic Clostridia and Sphingomonas spp., have been reported to degrade OMW
can be also identified [1]. Bacterial diversity in nonaer- phenolics. Ring-cleavage and o-demethylation are among the
ated olive mill waste-based mixtures consisted mainly of mechanisms involved in the reduction of olive mill waste
fermentative bacteria belonging to the phylum Firmicutes, phenolics [88]. Bacterial strains isolated from other industrial
for example, Clostridium tyrobutyricum, Lactobacillus vacci- wastewaters or contaminated sites have been also applied for
nostercus, Leuconostoc mesenteroides, and Sporolactobacillus the detoxification of olive mill waste phenolic compounds.
inulinus, followed by Actinobacteria (Rhodococcus fascians, Di Gioia et al. [89] exploited the degradation potential of
Clavibacter michiganensis, Curtobacterium albidum and Frig- Ralstonia sp. LD35 and Pseudomonas putida DSM 1868 in
oribacterium sp.) and Gammaproteobacteria (Erwinia per- the detoxification of OMW phenolics. Furthermore, Pseu-
sicina, Stenotrophomonas maltophilia and Luteibacter sp.) domonas putida and Pediococcus pentosaceus strains which
[1]. were previously isolated from activated sludge and forest
The microbial diversity in an anaerobic continuous litter could effectively decolorize OMW and remove total
stirred tank reactor (CSTR) treating TPOMW was examined phenolics from TPOMW, respectively [90, 91]. Azotobacter
under low and high organic loading conditions [86, 87]. vinelandii with nitrogen-fixing capacity has been widely
At low organic loading rate (OLR), bacterial diversity was used in the bioremediation of OMW [92, 93]. Moreover,
indigenous Klebsiella oxytoca strains can effectively alleviate
dominated by representatives of low G + C Gram-positive
the phytotoxic effects of OMW [94].
bacteria, in particular Clostridium spp., although bacterial
Candida cylindracea, C. rugosa, C. tropicalis, Geotrichum
community structure consisted of members of Actinobacte-
candidum, Rhodotorula glutinis, R. mucilaginosa, Trichospo-
ria, Gammaproteobacteria, Bacteroidetes and Deferribacteres ron cutaneum, and Yarrowia lipolytica are yeasts which have
under high OLR operational conditions [87]. Moreover, been used widely in the bioremediation of olive mill wastes
Chloroflexi spp. were involved in the anaerobic digestion [95–102]. For instance, a Geotrichum candidum strain, which
process of TPOMW [87]. was isolated from an aerated pilot-scale bubble column fed
Methanogenic diversity in TPOMW-fed bioreactors with OMW, could dephenolize and decolorize olive mill efflu-
under mesophilic conditions appears to be composed by ace- ents [103]. Moreover, Rhodotorula mucilaginosa, which was
toclastic Methanosaeta and Methanosarcina species. Investi- isolated from an OMW evaporation pond, could effectively
gation of methanogenic communities during codigestion of reduce phenolics and COD [97].
TPOMW and cattle excreta at 37∘ C and 55∘ C was carried The white-rot fungi Coriolopsis polyzona (syn. Funalia
out by Goberna et al. [31]. Acetoclastic Methanosarcina spp. polyzona), Coriolopsis rigida (syn. Coriolopsis floccosa), Gan-
were predominant under mesophilic conditions, although oderma australe, G. carnosum, Lentinula edodes, Panus ti-
a shift in methanogenic diversity was observed at ther- grinus (syn. Lentinus tigrinus), Phanerochaete chrysosporium
mophilic conditions as the result of increased H2 pres- (syn. Phanerodontia chrysosporium), Phanerochaete flavid-
sure, which favored members of the hydrogenotrophic oalba (syn. Phlebiopsis flavidoalba), Phlebia radiata, Pleuro-
tus ostreatus, P. eryngii, P. pulmonarius, P. sajor-caju (syn.
genera Methanoculleus, Methanobacterium, and Methan-
Lentinus sajor-caju), Poria subvermispora (syn. Ceriporiopsis
othermobacter, together with the acetoclastic thermophile
subvermispora), Pycnoporus cinnabarinus, P. coccineus, Ri-
Methanosarcina thermophila. Methanoculleus thermophilicus gidoporus lignosus (syn. Rigidoporus microporus), and Tra-
and Methanosarcina thermophila dominated the anaerobic metes versicolor have been widely used in the detoxification
sludge under thermophilic conditions. Moreover, methane of olive mill wastes [104–113]. In white-rot fungi, laccases
production was exclusively carried out by Methanosaeta and peroxidases as well as radical oxygen species are
spp. during anaerobic digestion of TPOMW in CSTRs [86, involved in the significant decrease of olive mill waste
87]. phenolics [114, 115]. The ascomycetes Aspergillus ibericus, A.
oryzae (syn. Aspergillus flavus var. oryzae), A. niger, A. terreus, Clavibacter michiganensis subsp. michiganensis, and Pseu-
Fusarium graminearum (syn. Gibberellazeae), F. lateritium domonas syringae pv. tomato [133]. Extract from TPOMW-
(syn. Gibberella baccata), F. oxysporum, Paecilomyces derived composts can suppress the plant pathogen Pythium
farinosus (syn. Isaria farinosa), Penicillium chrysogenum and aphanidermatum [134].
P. citrinum as well as the zygomycetes Mucor racemosus and Some bacteria isolated from OMW have been reported
Rhizopus arrhizus have been also reported to bioremediate to exhibit antagonistic effects against soil-borne pathogens.
olive mill wastes [102, 116–120]. Bacterial strains related to the genus Bacillus and the
Despite the fact that several selected microbial strains species Burkholderia caryophylli and Pseudomonas fluorescens
have been used for the degradation of olive mill waste recal- induced in planta disease suppressiveness against Fusarium
citrant compounds, the examination of those bioremediation and/or Rhizoctonia damping-off of tomato [130]. Moreover,
agents in sterile OMW- and TPOMW-based media does not bacterial strains, which were isolated from OMW and related
necessary mean that they can be effective degraders under to the species Bacillus subtilis, B. pumilis, Pseudomonas
ambient conditions. In fact, white-rot fungi are slow growers, putida, and Stenotrophomonas maltophilia, exerted in planta
and their ability to in vivo dominate over other microbial antimicrobial activity against Agrobacterium tumefaciens
groups is doubted [121]. Indigenous microorganisms, which [135]. Serratia marcescens strain BR2.1 is a biocontrol agent
have been adapted to the adverse conditions of olive mill with in planta antimicrobial activity against Fusarium oxyspo-
wastes, are more likely to effectively colonize the effluent. rum f.sp. radicis-lycopersici which was isolated from the rhi-
Indeed, in vivo experimentation of indigenous microbiota zosphere of tomato plants growing in OMW-derived compost
and other selected microorganisms is needed to guarantee amendments [136]. TPOMW and TPOMW-derived com-
their effectiveness in degrading olive mill wastes [93]. Besides, post extracts appear to exert a general suppression against
genetic engineering of extracellular oxidases [122], for exam- soil-borne oomycete Phytophthora capsici [137]. However,
ple, laccases, manganese peroxidases, versatile peroxidases TPOMW-induced suppression against Pythium ultimum and
and lignin peroxidases, and/or implementation of enzyme Botrytis cinerea was weak and depended on the specific olive
technology approaches, for example, application of enzyme mill waste tested, while only mature compost elicited pro-
immobilization techniques, can overcome the limitation for tection against the above-mentioned plant pathogens [137].
effective colonization of olive mill wastes. TPOMW and TPOMW-based compost extracts could not
suppress Rhizoctonia solani [137]. In contrast, Bonanomi et al.
[138] reported that the radial growth and the hyphal density
6.2. Bioconversion Aspects of Olive Mill Waste Microbiota. A
of the plant pathogens Fusarium oxysporum f.sp. lycopersici,
biotechnological application of OMW microbiota is the con-
Sclerotinia minor, and Botrytis cinerea were increased in olive
version of tyrosol to phenolic compounds of high antioxidant
mill dry residue-amended soils.
activity. The biotransformation of tyrosol to hydroxytyrosol
Actinobacterial strains, which were related to the genera
and 3,4-dihydroxyphenylacetate by a Halomonas strain has
Streptomyces and Lechevalieria and isolated from TPOMW-
been stated by Liebgott et al. [123]. Oleuropein present in olive
derived compost, exerted suppressive action against fungal
mill wastes can be also transformed into hydroxytyrosol [124,
and oomycete pathogens, that is, Fusarium oxysporum f.
125]. Furthermore, phenol-tolerant Enterobacteriaceae strains
sp. melonis, Phytophthora cinnamomi, Pythium debaryanum,
isolated from OMW possess the ability to bioconvert xylose
Sclerotinia sclerotiorum, and Thanatephorus cucumeris, and
to ethanol [126]. In addition, Clostridium bifermentans TYR6
the bacterial strain Agrobacterium tumefaciens CECT 4119
is an anaerobic bacterium which was isolated from OMW
[139].
and can covert cinnamic acid to 3-phenylpropionic acid [127].
Competition for nutrients and ecological niches, antibio-
Enhanced 𝛽-glucan synthase activities have been exhibited
sis (e.g., secretion of volatile metabolites or other antimi-
by mushroom species growing in OMW [128]. Moreover,
crobial agents), spore germination, germ tube elongation
Paracoccus thiocyanatus and Halothiobacillus neapolitanus
inhibition, and lysis via hydrolytic enzymes can be involved
strains are sulfur-oxidizing bacteria which have been isolated
in the suppression of soil-borne pathogens by olive mill
from alkaline TPOMW-based compost and can be used in
waste-derived compost amendments [130, 140, 141]. Indeed, a
compost acidification [129].
microbe-induced suppresion associated with the dominance
of copiotrophs and/or the proliferation of certain microbial
6.3. Plant Disease-Suppressive Properties of Olive Mill Wastes. groups appears to contribute significantly to OMW and
OMWs exert antifungal activity against plant pathogens. TPOMW suppressive effects [28, 67].
OMW can suppress the soil-borne pathogens Rhizoctonia
solani and Fusarium solani at low dose applications [130] as 7. Conclusions
the result of the antimicrobial action of OMW phenolics.
In addition, control of Botrytis fruit rot on strawberries Monitoring of microbial communities is one of the most fun-
and peppers by raw OMW has been also reported [131]. damental tasks to understand any bioremediation process.
Moreover, OMW application on fruits and vegetables can Although the importance of monitoring microbial diversity
inhibit the sporulation of Penicillium and Botrytis spp. and has been extensively stated, only a few studies focusing on the
suppress the phytopathogenic effects of Fusarium oxysporum identification of microbial communities in olive mill wastes
f.sp. lycopersici on tomato plants [132]. OMW can also inhibit have been performed. Indeed, such research studies enable
the growth of the seed-borne pathogens of tomato plants, an in-depth analysis of olive mill waste biotrasformations.
CD studies: yes CD studies: no CD studies: yes

Media used: LB, YPG, MEA, YMA, Media used:
PDA, NA+20% or 40% OMW, LB, King’s B [136]
phenol- and acetate-based media
[30, 34, 37, 38, 39, 43, 44, 77, 135] CI studies: yes CI studies: yes
CI studies: yes DGGE: [80, 81] DGGE: [18, 41, 42, 64,
Clone libraries: [29] T-RFLP/ 78, 79]
PhyloChip: [30] Clone library: [82] FAME: [65, 66]
SSCP: [85]
OMW treated in
Field application
Soil amendments
Raw OMW bioreactors and biofilters
Olive mills
Raw TPOMW TPOMW treated in Soil amendments

bioreactors
CD studies: yes CD studies: no CD studies: yes

Media used: Media used:
10% w/v TPOMW (pH 7 and 11), CI studies: yes ThCl, SCA, AGSA, GAA
NA, YPG [4, 33, 35, 40] TGGE: [1] [129, 139]
DGGE: [86, 87] CI studies: yes
CI studies: yes FAME: [1] FAME: [72]
DGGE: [32] AnaeroChip: [31] DGGE: [32, 35, 72, 75, 76,
qPCR: [31] 78, 79]
COMPOCHIP: [76]
qPCR: [78, 79]
Figure 4: Schematic representation of the olive mill waste route in relation to the studies completed and the level of analysis provided. LB:
Luria Bertani medium; YPG: yeast extract-peptone-glucose medium; MEA: malt extract agar; YMA: yeast-malt agar; PDA: potato dextrose
agar; NA: nutrient agar; ThCl: medium 152 for Thiobacillus (ATCC); SCA: starch casein agar; AGSA: arginine glycerol salts agar; GAA: glycerol
asparagine agar. Research studies, considering less than 5 isolates, are not included; CD: culture-dependent, CI: culture-independent.
For instance, the implementation of 16S rRNA gene clone in olive mill wastes and their adaptive mechanisms to olive
libraries and high density DNA microarray (PhyloChip) not mill waste phenolics. Interestingly, the genome sequence
only enhanced our knowledge on OMW indigenous micro- analyses of Olivibacter sitiensis and Clostridium methoxy-
biota but also established the presence of a cultivar-specific benzovorans are ongoing, and novel fundamental results
effect [29, 30]. As shown in Figure 4, a limited number of are expected. Isolation of novel biocontrol agents with sup-
molecular studies, which permit the identification of both pressive properties against the major plant pathogens is
cultured and uncultured microbial communities, have been another experimental task. Indeed, experimentation on new
carried out, and no studies have been performed by applying approaches for the cultivation of novel microorganisms and
high-throughput techniques, such as pyrotag sequencing new biocontrol agents deserve further examination.
and metagenomic approaches. Indeed, implementation of We propose that the focus of the research should shift
omic approaches, such as high density 16S rRNA microarray from the simple characterization of the microbial commu-
(PhyloChip) and 16S rRNA pyrotags, in the fruits of the nities that are present in OMWs to their functional role,
most important olive-tree varieties and their olive mill- starting with the genome sequencing of important isolates
generated wastes, in combination with the examination of that have been identified in OMWs and/or they have been
the complex nature of olive mill waste phenolics and other characterized as degraders. This effort should not only be
physicochemical and environmental features, will elucidate restricted to strains that have been isolated, but also to
through canonical correspondence analysis the parameters uncultured ones that can be characterized through the use
affecting microbial ecology in olive mill effluents. of single cell genomic (SCG) approach. Genomes obtained
The biotechnological potential of olive mill wastes has not through a SCG approach have been reported for marine
been fully exploited since novel bacterial taxa are still being bacterial strains and for insect endosymbionts [144–147], and
identified in olive mill wastes [142, 143], indicating that there this presents a unique opportunity to identify the metabolic
are still unexhausted sources for biotechnology. Genome features, the timeline of species evolution, and the inter-
sequence analyses of indigenous microbiota will reveal the organismal interactions of the uncultured microbial groups
biodegradation pathways of recalcitrant compounds present that dominate the olive mill wastes. Once this is achieved,
in-depth metagenomic approaches could be used to charac- [12] S. Ntougias, F. Gaitis, P. Katsaris, S. Skoulika, N. Iliopoulos, and
terize the gene repertoire present in OMW, while the genomes G. I. Zervakis, “The effects of olives harvest period and pro-
sequenced would provide the backbone for mapping the duction year on olive mill wastewater properties—evaluation
majority of these genes. Metatranscriptomic approaches will of Pleurotus strains as bioindicators of the effluent’s toxicity,”
further reveal the active microbial communities and the genes Chemosphere, vol. 92, no. 4, pp. 399–405, 2013.
that characterize these environments. [13] J. A. Alburquerque, J. Gonzálvez, D. Garcı́a, and J. Cegarra,
“Agrochemical characterisation of “alperujo”, a solid by-product
of the two-phase centrifugation method for olive oil extraction,”
Acknowledgments Bioresource Technology, vol. 92, no. 2, pp. 195–200, 2004.
[14] A. G. Vlyssides, M. Loizides, and P. K. Karlis, “Integrated
This study received support from the European Community’s strategic approach for reusing olive oil extraction by-products,”
Seventh Framework Programme PEOPLE-2012-IAPP under Journal of Cleaner Production, vol. 12, no. 6, pp. 603–611, 2004.
Grant agreement 324349. The present address of Kostas [15] L. Baeta-Hall, M. Céu Sàágua, M. Lourdes Bartolomeu, A. M.
Bourtzis is: Insect Pest Control Laboratory, Joint FAO/IAEA Anselmo, and M. Fernanda Rosa, “Bio-degradation of olive oil
Programme of Nuclear Techniques in Food and Agriculture, husks in composting aerated piles,” Bioresource Technology, vol.
1220 Vienna, Austria. 96, no. 1, pp. 69–78, 2005.
[16] L. Saez, J. Perez, and J. Martinez, “Low molecular weight phe-
nolics attenuation during simulated treatment of wastewaters
References from olive oil mills in evaporation ponds,” Water Research, vol.
[1] J. A. Morillo, M. Aguilera, B. Antı́zar-Ladislao et al., “Molecular 26, no. 9, pp. 1261–1266, 1992.
microbial and chemical investigation of the bioremediation of [17] A. Ben Sassi, A. Boularbah, A. Jaouad, G. Walker, and A. Bous-
two-phase olive mill waste using laboratory-scale bioreactors,” said, “A comparison of Olive oil Mill Wastewaters (OMW) from
Applied Microbiology and Biotechnology, vol. 79, no. 2, pp. 309– three different processes in Morocco,” Process Biochemistry, vol.
317, 2008. 41, no. 1, pp. 74–78, 2006.
[2] International Olive Oil Council, 2008, http://www.internation- [18] D. G. Karpouzas, S. Ntougias, E. Iskidou et al., “Olive mill
aloliveoil.org/. wastewater affects the structure of soil bacterial communities,”
Applied Soil Ecology, vol. 45, no. 2, pp. 101–111, 2010.
[3] M. Uceda, A. Jiménez, and G. Beltrán, “Olive oil extraction and
quality,” Grasas y Aceites, vol. 57, no. 1, pp. 25–31, 2006. [19] T. Yangui, A. Dhouib, A. Rhouma, and S. Sayadi, “Potential of
hydroxytyrosol-rich composition from olive mill wastewater as
[4] S. Ntougias, G. I. Zervakis, C. Ehaliotis, N. Kavroulakis, and a natural disinfectant and its effect on seeds vigour response,”
K. K. Papadopoulou, “Ecophysiology and molecular phylogeny Food Chemistry, vol. 117, no. 1, pp. 1–8, 2009.
of bacteria isolated from alkaline two-phase olive mill wastes,”
[20] S. Ntougias and N. Kavroulakis, Natural Antioxidants from Olive
Research in Microbiology, vol. 157, no. 4, pp. 376–385, 2006.
Mill Wastes, Nova Publishers, New York, NY, USA, 2010.
[5] M. Hafidi, S. Amir, and J.-C. Revel, “Structural characterization
[21] I. Ntaikou, C. Kourmentza, E. C. Koutrouli et al., “Exploitation
of olive mill waster-water after aerobic digestion using elemen-
of olive oil mill wastewater for combined biohydrogen and
tal analysis, FTIR and 13C NMR,” Process Biochemistry, vol. 40,
biopolymers production,” Bioresource Technology, vol. 100, no.
no. 8, pp. 2615–2622, 2005.
15, pp. 3724–3730, 2009.
[6] E. S. Aktas, S. Imre, and L. Ersoy, “Characterization and lime [22] S. Ntougias, C. Ehaliotis, K. K. Papadopoulou, and G. Zervakis,
treatment of olive mill wastewater,” Water Research, vol. 35, no. “Application of respiration and FDA hydrolysis measurements
9, pp. 2336–2340, 2001. for estimating microbial activity during composting processes,”
[7] B. Y. Ammary, “Treatment of olive mill wastewater using an Biology and Fertility of Soils, vol. 42, no. 4, pp. 330–337, 2006.
anaerobic sequencing batch reactor,” Desalination, vol. 177, no. [23] G. Zervakis, P. Yiatras, and C. Balis, “Edible mushrooms
1–3, pp. 157–165, 2005. from olive oil mill wastes,” International Biodeterioration and
[8] B. Zenjari, H. El Hajjouji, G. Ait Baddi et al., “Eliminating Biodegradation, vol. 38, no. 3-4, pp. 237–243, 1996.
toxic compounds by composting olive mill wastewater-straw [24] G. Ouzounidou, G. I. Zervakis, and F. Gaitis, “Raw and microbi-
mixtures,” Journal of Hazardous Materials, vol. 138, no. 3, pp. ologically detoxified olive mill waste and their impact on plant
433–437, 2006. growth,” Terrestrial and Aquatic Environmental Toxicology, vol.
[9] C. Amaral, M. S. Lucas, J. Coutinho, A. L. Crespı́, M. do Rosário 4, no. 1, pp. 21–38, 2010.
Anjos, and C. Pais, “Microbiological and physicochemical [25] R. Capasso, A. Evidente, L. Schivo, G. Orru, M. A. Marcialis,
characterization of olive mill wastewaters from a continuous and G. Cristinzio, “Antibacterial polyphenols from olive oil mill
olive mill in Northeastern Portugal,” Bioresource Technology, waste waters,” Journal of Applied Bacteriology, vol. 79, no. 4, pp.
vol. 99, no. 15, pp. 7215–7223, 2008. 393–398, 1995.
[10] E. Eroǧlu, I. Eroǧlu, U. Gündüz, and M. Yücel, “Compari- [26] E. Athanasoulia, P. Melidis, and A. Aivasidis, “Optimization of
son of physicochemical characteristics and photofermentative biogas production from waste activated sludge through serial
hydrogen production potential of wastewaters produced from digestion,” Renewable Energy, vol. 47, pp. 147–151, 2012.
different olive oil mills in Western-Anatolia, Turkey,” Biomass [27] P. Baldrian, G. I. Zervakis, V. Merhautová, S. Ntougias, C. Ehali-
and Bioenergy, vol. 33, no. 4, pp. 706–711, 2009. otis, and F. Nerud, “The use of hydroxyl-radical-generating
[11] I. Aviani, M. Raviv, Y. Hadar et al., “Effects of harvest date, irri- systems for the treatment of olive mill wastewaters,” Folia
gation level, cultivar type and fruit water content on olive mill Microbiologica, vol. 51, no. 4, pp. 337–341, 2006.
wastewater generated by a laboratory scale ‘Abencor’ milling [28] S. Ntougias, K. K. Papadopoulou, G. I. Zervakis, N. Kavroulakis,
system,” Bioresource Technology, vol. 107, pp. 87–96, 2012. and C. Ehaliotis, “Suppression of soil-borne pathogens of
tomato by composts derived from agro-industrial wastes abun- on the structure and function of the community of arbuscular
dant in Mediterranean regions,” Biology and Fertility of Soils, vol. mycorrhizal fungi,” Soil Biology and Biochemistry, vol. 41, no. 12,
44, no. 8, pp. 1081–1090, 2008. pp. 2466–2476, 2009.
[29] N. Kavroulakis and S. Ntougias, “Bacterial and 𝛽-proteobacte- [43] B. Millán, R. Lucas, A. Robles, T. Garcı́a, G. A. de Cienfue-
rial diversity in Olea europaea var. mastoidis- and O. europaea gos, and A. Gálvez, “A study on the microbiota from olive-
var. koroneiki-generated olive mill wastewaters: influence of mill wastewater (OMW) disposal lagoons, with emphasis on
cultivation and harvesting practice on bacterial community filamentous fungi and their biodegradative potential,” Microbi-
structure,” World Journal of Microbiology and Biotechnology, vol. ological Research, vol. 155, no. 3, pp. 143–147, 2000.
27, no. 1, pp. 57–66, 2011. [44] J. Mann, J. L. Markham, P. Peiris, N. Nair, R. N. Spooner-
[30] G. Tsiamis, G. Tzagkaraki, A. Chamalaki et al., “Olive-mill Hart, and P. Holford, “Screening and selection of fungi for
wastewater bacterial communities display a cultivar specific bioremediation of olive mill wastewater,” World Journal of
profile,” Current Microbiology, vol. 64, no. 2, pp. 197–203, 2012. Microbiology and Biotechnology, vol. 26, no. 3, pp. 567–571, 2010.
[31] M. Goberna, M. Gadermaier, C. Garcı́a, B. Wett, and H. Insam, [45] H. Aissam, M. J. Penninckx, and M. Benlemlih, “Reduction of
“Adaptation of methanogenic communities to the cofermenta- phenolics content and COD in olive oil mill wastewaters by
tion of cattle excreta and olive mill wastes at 37∘ C and 55∘ C,” indigenous yeasts and fungi,” World Journal of Microbiology and
Applied and Environmental Microbiology, vol. 76, no. 19, pp. Biotechnology, vol. 23, no. 9, pp. 1203–1208, 2007.
6564–6571, 2010. [46] E. P. Giannoutsou, C. Meintanis, and A. D. Karagouni, “Iden-
[32] A. Vivas, B. Moreno, S. Garcia-Rodriguez, and E. Benitez, tification of yeast strains isolated from a two-phase decanter
“Assessing the impact of composting and vermicomposting on system olive oil waste and investigation of their ability for its
bacterial community size and structure, and microbial func- fermentation,” Bioresource Technology, vol. 93, no. 3, pp. 301–
tional diversity of an olive-mill waste,” Bioresource Technology, 306, 2004.
vol. 100, no. 3, pp. 1319–1326, 2009. [47] I. Karaouzas, N. T. Skoulikidis, U. Giannakou, and T. A.
[33] C. E. Jones, P. J. Murphy, and N. J. Russell, “Diversity and Albanis, “Spatial and temporal effects of olive mill wastewaters
osmoregulatory responses of bacteria isolated from two-phase to stream macroinvertebrates and aquatic ecosystems status,”
olive oil extraction waste products,” World Journal of Microbiol- Water Research, vol. 45, no. 19, pp. 6334–6346, 2011.
ogy and Biotechnology, vol. 16, no. 6, pp. 555–561, 2000. [48] A. Mekki, A. Dhouib, F. Feki, and S. Sayadi, “Assessment of
[34] D. Venieri, A. Rouvalis, and J. Iliopoulou-Georgudaki, “Micro- toxicity of the untreated and treated olive mill wastewaters
bial and toxic evaluation of raw and treated olive oil mill and soil irrigated by using microbiotests,” Ecotoxicology and
wastewaters,” Journal of Chemical Technology and Biotechnol- Environmental Safety, vol. 69, no. 3, pp. 488–495, 2008.
ogy, vol. 85, no. 10, pp. 1380–1388, 2010. [49] H. K. Obied, D. R. Bedgood Jr., P. D. Prenzler, and K. Robards,
[35] E. Federici, M. Pepi, A. Esposito et al., “Two-phase olive mill “Bioscreening of Australian olive mill waste extracts: biophenol
waste composting: community dynamics and functional role of content, antioxidant, antimicrobial and molluscicidal activi-
the resident microbiota,” Bioresource Technology, vol. 102, no. 23, ties,” Food and Chemical Toxicology, vol. 45, no. 7, pp. 1238–1248,
pp. 10965–10972, 2011. 2007.
[36] J.-F. Brugère, A. Mihajlovski, M. Missaoui, and P. Peyret, “Tools [50] A. Tafesh, N. Najami, J. Jadoun, F. Halahlih, H. Riepl, and
for stools: the challenge of assessing human intestinal micro- H. Azaizeh, “Synergistic antibacterial effects of polypheno-
biota using molecular diagnostics,” Expert Review of Molecular lic compounds from olive mill wastewater,” Evidence-Based
Diagnostics, vol. 9, no. 4, pp. 353–365, 2009. Complementary and Alternative Medicine, vol. 2011, Article ID
[37] G. Bleve, C. Lezzi, M. A. Chiriatti et al., “Selection of non- 431021, 9 pages, 2011.
conventional yeasts and their use in immobilized form for [51] A. Fiorentino, A. Gentili, M. Isidori et al., “Environmental
the bioremediation of olive oil mill wastewaters,” Bioresource effects caused by olive mill wastewaters: toxicity comparison of
Technology, vol. 102, no. 2, pp. 982–989, 2011. low-molecular-weight phenol components,” Journal of Agricul-
[38] M. Sinigaglia, N. Di Benedetto, A. Bevilacqua, M. R. Corbo, tural and Food Chemistry, vol. 51, no. 4, pp. 1005–1009, 2003.
A. Capece, and P. Romano, “Yeasts isolated from olive mill [52] G. Greco Jr., M. L. Colarieti, G. Toscano, G. Iamarino, M. A. Rao,
wastewaters from southern Italy: technological characterization and L. Gianfreda, “Mitigation of olive mill wastewater toxicity,”
and potential use for phenol removal,” Applied Microbiology and Journal of Agricultural and Food Chemistry, vol. 54, no. 18, pp.
Biotechnology, vol. 87, no. 6, pp. 2345–2354, 2010. 6776–6782, 2006.
[39] A. Ben Sassi, N. Ouazzani, G. M. Walker, S. Ibnsouda, M. El [53] F. Peixoto, F. Martins, C. Amaral, J. Gomes-Laranjo, J. Almeida,
Mzibri, and A. Boussaid, “Detoxification of olive mill wastew- and C. M. Palmeira, “Evaluation of olive oil mill wastewater
aters by Moroccan yeast isolates,” Biodegradation, vol. 19, no. 3, toxicity on the mitochondrial bioenergetics after treatment with
pp. 337–346, 2008. Candida oleophila,” Ecotoxicology and Environmental Safety, vol.
[40] S. Romo-Sánchez, M. Alves-Baffi, M. Arévalo-Villena, J. Úbeda- 70, no. 2, pp. 266–275, 2008.
Iranzo, and A. Briones-Pérez, “Yeast biodiversity from oleic [54] S. Hanifi and I. El Hadrami, “Olive mill wastewaters: diversity
ecosystems: study of their biotechnological properties,” Food of the fatal product in olive oil industry and its valorisation
Microbiology, vol. 27, no. 4, pp. 487–492, 2010. as agronomical amendment of poor soils: a review,” Journal of
[41] D. G. Karpouzas, C. Rousidou, K. K. Papadopoulou et al., Agronomy, vol. 8, no. 1, pp. 1–13, 2009.
“Effect of continuous olive mill wastewater applications, in the [55] T. Yangui, S. Sayadi, A. Gargoubi, and A. Dhouib, “Fungici-
presence and absence of nitrogen fertilization, on the structure dal effect of hydroxytyrosol-rich preparations from olive mill
of rhizosphere-soil fungal communities,” FEMS Microbiology wastewater against Verticillium dahliae,” Crop Protection, vol. 29,
Ecology, vol. 70, no. 3, pp. 388–401, 2009. no. 10, pp. 1208–1213, 2010.
[42] I. Ipsilantis, D. G. Karpouzas, K. K. Papadopoulou, and C. [56] M. Friedman, R. Rasooly, P. M. Do, and P. R. Henika, “The olive
Ehaliotis, “Effects of soil application of olive mill wastewaters compound 4-hydroxytyrosol inactivates Staphylococcus aureus
bacteria and Staphylococcal Enterotoxin A (SEA),” Journal of [70] A. Krastanov, Z. Alexieva, and H. Yemendzhiev, “Microbial
Food Science, vol. 76, no. 8, pp. M558–M563, 2011. degradation of phenol and phenolic derivatives,” Engineering in
[57] M. L. Cayuela, C. Mondini, M. A. Sánchez-Monedero, and A. Life Sciences, vol. 13, no. 1, pp. 76–87, 2013.
Roig, “Chemical properties and hydrolytic enzyme activities for [71] R. Jarboui, F. Sellami, C. Azri, N. Gharsallah, and E. Ammar,
the characterisation of two-phase olive mill wastes composting,” “Olive mill wastewater evaporation management using PCA
Bioresource Technology, vol. 99, no. 10, pp. 4255–4262, 2008. method. Case study of natural degradation in stabilization
[58] M. DellaGreca, P. Monaco, G. Pinto, A. Pollio, L. Previtera, ponds (Sfax, Tunisia),” Journal of Hazardous Materials, vol. 176,
and F. Temussi, “Phytotoxicity of low-molecular-weight phenols no. 1–3, pp. 992–1005, 2010.
from olive mill waste waters,” Bulletin of Environmental Con- [72] I. Sampedro, M. Giubilei, T. Cajthaml et al., “Short-term impact
tamination and Toxicology, vol. 67, no. 3, pp. 352–359, 2001. of dry olive mill residue addition to soil on the resident
[59] A. Mekki, A. Dhouib, and S. Sayadi, “Changes in microbial microbiota,” Bioresource Technology, vol. 100, no. 23, pp. 6098–
and soil properties following amendment with treated and 6106, 2009.
untreated olive mill wastewater,” Microbiological Research, vol. [73] I. Sampedro, E. Aranda, R. Dı́az, M. Garcı́a-Sanchez, J. A.
161, no. 2, pp. 93–101, 2006. Ocampo, and I. Garcı́a-Romera, “Saprobe fungi decreased the
[60] M. G. Di Serio, B. Lanza, M. R. Mucciarella et al., “Effects of olive sensitivity to the toxic effect of dry olive mill residue on
mill wastewater spreading on the physico-chemical and micro- arbuscular mycorrhizal plants,” Chemosphere, vol. 70, no. 8, pp.
biological characteristics of soil,” International Biodeterioration 1383–1389, 2008.
and Biodegradation, vol. 62, no. 4, pp. 403–407, 2008. [74] E. Benitez, H. Sainz, and R. Nogales, “Hydrolytic enzyme activi-
[61] F. Z. El Hassani, A. Zinedine, S. Mdaghri Alaoui, M. Merzouki, ties of extracted humic substances during the vermicomposting
and M. Benlemlih, “Use of olive mill wastewater as an organic of a lignocellulosic olive waste,” Bioresource Technology, vol. 96,
amendment for Mentha spicata L,” Industrial Crops and Prod- no. 7, pp. 785–790, 2005.
ucts, vol. 32, no. 3, pp. 343–348, 2010. [75] M. J. Fernández-Gómez, R. Nogales, H. Insam, E. Romero, and
[62] B. Mechri, M. Issaoui, A. Echbili et al., “Olive orchard amended M. Goberna, “Role of vermicompost chemical composition,
with olive mill wastewater: effects on olive fruit and olive oil microbial functional diversity, and fungal community structure
quality,” Journal of Hazardous Materials, vol. 172, no. 2-3, pp. in their microbial respiratory response to three pesticides,”
1544–1550, 2009. Bioresource Technology, vol. 102, no. 20, pp. 9638–9645, 2011.
[63] L. Nasini, G. Gigliotti, M. A. Balduccini, E. Federici, G. Cenci, [76] M. J. Fernández-Gómez, R. Nogales, H. Insam, E. Romero,
and P. Proietti, “Effect of solid olive-mill waste amendment and M. Goberna, “Use of DGGE and COMPOCHIP for
on soil fertility and olive (Olea europaea L.) tree activity,” investigating bacterial communities of various vermicomposts
Agriculture, Ecosystems and Environment, vol. 164, pp. 292–297, produced from different wastes under dissimilar conditions,”
2013. Science of the Total Environment, vol. 414, pp. 664–671, 2012.
[64] C. Rousidou, K. Papadopoulou, G. Zervakis, B. K. Singh, C. [77] S. Ertuğrul, G. Dönmez, and S. Takaç, “Isolation of lipase pro-
Ehaliotis, and D. G. Karpouzas, “Repeated application of diluted ducing Bacillus sp. from olive mill wastewater and improving its
olive mill wastewater induces changes in the structure of the soil enzyme activity,” Journal of Hazardous Materials, vol. 149, no. 3,
microbial community,” European Journal of Soil Biology, vol. 46, pp. 720–724, 2007.
no. 1, pp. 34–40, 2010. [78] B. Moreno, R. Nogales, C. Macci, G. Masciandaro, and E.
[65] B. Mechri, F. B. Mariem, M. Baham, S. B. Elhadj, and M. Benitez, “Microbial eco-physiological profiles to estimate the
Hammami, “Change in soil properties and the soil microbial biological restoration of a trichloroethylene-contaminated soil,”
community following land spreading of olive mill wastewa- Ecological Indicators, vol. 11, no. 6, pp. 1563–1571, 2011.
ter affects olive trees key physiological parameters and the [79] P. Di Gennaro, B. Moreno, E. Annoni, S. Garcı́a-Rodrı́guez, G.
abundance of arbuscular mycorrhizal fungi,” Soil Biology and Bestetti, and E. Benitez, “Dynamic changes in bacterial com-
Biochemistry, vol. 40, no. 1, pp. 152–161, 2008. munity structure and in naphthalene dioxygenase expression
[66] B. Mechri, H. Chehab, F. Attia, F. B. Mariem, M. Braham, and in vermicompost-amended PAH-contaminated soils,” Journal
M. Hammami, “Olive mill wastewater effects on the microbial of Hazardous Materials, vol. 172, no. 2-3, pp. 1464–1469, 2009.
communities as studied in the field of olive trees by analysis of [80] L. Bertin, S. Lampis, D. Todaro et al., “Anaerobic acido-
fatty acid signatures,” European Journal of Soil Biology, vol. 46, genic digestion of olive mill wastewaters in biofilm reactors
no. 5, pp. 312–318, 2010. packed with ceramic filters or granular activated carbon,” Water
[67] M. Kotsou, I. Mari, K. Lasaridi, I. Chatzipavlidis, C. Balis, and Research, vol. 44, no. 15, pp. 4537–4549, 2010.
A. Kyriacou, “The effect of olive oil mill wastewater (OMW) [81] A. Scoma, L. Bertin, G. Zanaroli, S. Fraraccio, and F. Fava, “A
on soil microbial communities and suppressiveness against physicochemical-biotechnological approach for an integrated
Rhizoctonia solani,” Applied Soil Ecology, vol. 26, no. 2, pp. 113– valorization of olive mill wastewater,” Bioresource Technology,
121, 2004. vol. 102, no. 22, pp. 10273–10279, 2011.
[68] Y. Laor, I. Saadi, M. Raviv, S. Medina, D. Erez-Reifen, and H. [82] L. Bertin, M. C. Colao, M. Ruzzi, and F. Fava, “Performances
Eizenbergc, “Land spreading of olive mill wastewater in Israel: and microbial features of a granular activated carbon packed-
current knowledge, practical experience, and future research bed biofilm reactor capable of an efficient anaerobic digestion
needs,” Israel Journal of Plant Sciences, vol. 59, no. 1, pp. 39–51, of olive mill wastewaters,” FEMS Microbiology Ecology, vol. 48,
2011. no. 3, pp. 413–423, 2004.
[69] R. Kirby, “Actinomycetes and lignin degradation,” in Advances [83] C.-L. Chen, J.-H. Wu, I.-C. Tseng, T.-M. Liang, and W.-T. Liu,
in Applied Microbiology, J. W. B. Allen, I. Laskin Geoffrey, M. “Characterization of active microbes in a full-scale anaerobic
Gadd, and S. Sariaslani, Eds., vol. 58, pp. 125–168, Academic fluidized bed reactor treating phenolic wastewater,” Microbes
Press, 2005. and Environments, vol. 24, no. 2, pp. 144–153, 2009.
[84] S.-Z. Ke, Z. Shi, T. Zhang, and H. H. P. Fang, “Degradation of [97] R. Jarboui, H. Baati, F. Fetoui, A. Gargouri, N. Gharsallah, and
phenol in an upflow anaerobic sludge blanket (UASB) reactor E. Ammar, “Yeast performance in wastewater treatment: case
at ambient temperature,” Journal of Environmental Sciences, vol. study of Rhodotorula mucilaginosa,” Environmental Technology,
16, no. 3, pp. 525–528, 2004. vol. 33, no. 8, pp. 951–960, 2012.
[85] A. Rizzi, M. Zucchi, S. Borin, M. Marzorati, C. Sorlini, and D. [98] A. Karakaya, Y. Laleli, and S. Takaç, “Development of process
Daffonchio, “Response of methanogen populations to organic conditions for biodegradation of raw olive mill wastewater
load increase during anaerobic digestion of olive mill wastewa- by Rhodotorula glutinis,” International Biodeterioration and
ter,” Journal of Chemical Technology and Biotechnology, vol. 81, Biodegradation, vol. 75, pp. 75–82, 2012.
no. 9, pp. 1556–1562, 2006. [99] M. Lopes, C. Araújo, M. Aguedo et al., “The use of olive mill
[86] B. Rincón, F. Raposo, R. Borja, J. M. Gonzalez, M. C. Portillo, wastewater by wild type Yarrowia lipolytica strains: medium
and C. Saiz-Jimenez, “Performance and microbial communities supplementation and surfactant presence effect,” Journal of
of a continuous stirred tank anaerobic reactor treating two- Chemical Technology and Biotechnology, vol. 84, no. 4, pp. 533–
phases olive mill solid wastes at low organic loading rates,” 537, 2009.
Journal of Biotechnology, vol. 121, no. 4, pp. 534–543, 2006. [100] G. Martinez-Garcia, A. C. Johnson, R. T. Bachmann, C. J.
[87] B. Rincón, R. Borja, J. M. González, M. C. Portillo, and C. Williams, A. Burgoyne, and R. G. J. Edyvean, “Anaerobic treat-
Sáiz-Jiménez, “Influence of organic loading rate and hydraulic ment of olive mill wastewater and piggery effluents fermented
retention time on the performance, stability and microbial with Candida tropicalis,” Journal of Hazardous Materials, vol.
communities of one-stage anaerobic digestion of two-phase 164, no. 2-3, pp. 1398–1405, 2009.
olive mill solid residue,” Biochemical Engineering Journal, vol. [101] A. D’Annibale, G. G. Sermanni, F. Federici, and M. Petruccioli,
40, no. 2, pp. 253–261, 2008. “Olive-mill wastewaters: a promising substrate for microbial
[88] D. Di Gioia, C. Barberio, S. Spagnesi, L. Marchetti, and F. Fava, lipase production,” Bioresource Technology, vol. 97, no. 15, pp.
“Characterization of four olive-mill-wastewater indigenous 1828–1833, 2006.
bacterial strains capable of aerobically degrading hydroxylated [102] I. Garcı́a Garcı́a, P. R. Jiménez Peña, J. L. Bonilla Vences-
and methoxylated monocyclic aromatic compounds,” Archives lada, A. Martı́n Martı́n, M. A. Martı́n Santos, and E. Ramos
of Microbiology, vol. 178, no. 3, pp. 208–217, 2002. Gómez, “Removal of phenol compounds from olive mill
wastewater using Phanerochaete chrysosporium, Aspergillus
[89] D. Di Gioia, F. Fava, L. Bertin, and L. Marchetti, “Biodegrada-
niger, Aspergillus terreus and Geotrichum candidum,” Process
tion of synthetic and naturally occurring mixtures of mono-
Biochemistry, vol. 35, no. 8, pp. 751–758, 2000.
cyclic aromatic compounds present in olive mill wastewaters by
two aerobic bacteria,” Applied Microbiology and Biotechnology, [103] N. Assas, L. Marouani, and M. Hamdi, “Scale down and
vol. 55, no. 5, pp. 619–626, 2001. optimization of olive mill wastewaters decolorization by
Geotrichum candidum,” Bioprocess Engineering, vol. 22, no. 6, pp.
[90] B. de Felice, G. Pontecorvo, and M. Carfagna, “Degradation of 503–507, 2000.
waste waters from olive oil mills by Yarrowia lipolytica ATCC
20255 and Pseudomonas putida,” Acta Biotechnologica, vol. 17, [104] S. Ntougias, P. Baldrian, C. Ehaliotis et al., “Biodegradation and
no. 3, pp. 231–239, 1997. detoxification of olive mill wastewater by selected strains of the
mushroom genera Ganoderma and Pleurotus,” Chemosphere,
[91] N. Ben Othman, L. Ayed, N. Assas, F. Kachouri, M. Hammami, vol. 88, no. 5, pp. 620–626, 2012.
and M. Hamdi, “Ecological removal of recalcitrant phenolic
[105] S. M. Alaoui, M. Merzouki, M. J. Penninckx, and M. Benlemlih,
compounds of treated olive mill wastewater by Pediococcus
“Relationship between cultivation mode of white rot fungi
pentosaceus,” Bioresource Technology, vol. 99, no. 8, pp. 2996–
and their efficiency for olive oil mill wastewaters treatment,”
3001, 2008.
Electronic Journal of Biotechnology, vol. 11, no. 4, pp. 1–8, 2008.
[92] M. Papadelli, A. Roussis, K. Papadopoulou et al., “Biochemical [106] M. T. Cambria, S. Ragusa, V. Calabrese, and A. Cambria,
and molecular characterization of an Azotobacter vinelandii “Enhanced laccase production in white-rot fungus Rigidoporus
strain with respect to its ability to grow and fix nitrogen lignosus by the addition of selected phenolic and aromatic
in olive mill wastewater,” International Biodeterioration and compounds,” Applied Biochemistry and Biotechnology, vol. 163,
Biodegradation, vol. 38, no. 3-4, pp. 179–181, 1996. no. 3, pp. 415–422, 2011.
[93] C. Ehaliotis, K. Papadopoulou, M. Kotsou, I. Mari, and C. Balis, [107] A. D’Annibale, R. Casa, F. Pieruccetti, M. Ricci, and R.
“Adaptation and population dynamics of Azotobacter vinelandii Marabottini, “Lentinula edodes removes phenols from olive-
during aerobic biological treatment of olive-mill wastewater,” mill wastewater: impact on durum wheat (Triticum durum
FEMS Microbiology Ecology, vol. 30, no. 4, pp. 301–311, 1999. Desf.) germinability,” Chemosphere, vol. 54, no. 7, pp. 887–894,
[94] A. El Asli, F. Errachidi, R. Bennisse, A.-I. Qatibi, and M. Errami, 2004.
“Effect of cell immobilization on the treatment of olive mill [108] M. Fenice, G. Giovannozzi Sermanni, F. Federici, and A.
wastewater by a total phenols, acetic acid and formic acid D’Annibale, “Submerged and solid-state production of laccase
degrading bacterium strain,” Grasas y Aceites, vol. 56, no. 2, pp. and Mn-peroxidase by Panus tigrinus on olive mill wastewater-
116–120, 2005. based media,” Journal of Biotechnology, vol. 100, no. 1, pp. 77–85,
[95] M. Chtourou, E. Ammar, M. Nasri, and K. Medhioub, “Isolation 2003.
of a yeast, Trichosporon cutaneum, able to use molecular weight [109] C. I. Justino, K. Duarte, F. Loureiro et al., “Toxicity and organic
phenolic compounds: application to olive mill waste water content characterization of olive oil mill wastewater undergoing
treatment,” Journal of Chemical Technology and Biotechnology, a sequential treatment with fungi and photo-Fenton oxidation,”
vol. 79, no. 8, pp. 869–878, 2004. Journal of Hazardous Materials, vol. 172, no. 2-3, pp. 1560–1572,
[96] C. Gonçalves, M. Lopes, J. P. Ferreira, and I. Belo, “Biological 2009.
treatment of olive mill wastewater by non-conventional yeasts,” [110] G. Ouzounidou, S. Ntougias, M. Asfi, F. Gaitis, and G. I.
Bioresource Technology, vol. 100, no. 15, pp. 3759–3763, 2009. Zervakis, “Raw and fungal-treated olive-mill wastewater effects
on selected parameters of lettuce (Lactuca sativa L.) growth— [124] M. M. Santos, C. Piccirillo, P. M. L. Castro, N. Kalogerakis, and
the role of proline,” Journal of Environmental Science and Health, M. E. Pintado, “Bioconversion of oleuropein to hydroxytyrosol
Part B, vol. 47, no. 7, pp. 728–735, 2012. by lactic acid bacteria,” World Journal of Microbiology and
[111] T. de la Rubia, M. Lucas, and J. Martı́nez, “Controversial role of Biotechnology, vol. 28, no. 6, pp. 2435–2440, 2012.
fungal laccases in decreasing the antibacterial effect of olive mill [125] H. Jemai, A. El Feki, and S. Sayadi, “Antidiabetic and antioxidant
waste-waters,” Bioresource Technology, vol. 99, no. 5, pp. 1018– effects of hydroxytyrosol and oleuropein from olive leaves in
1025, 2008. alloxan-diabetic rats,” Journal of Agricultural and Food Chem-
[112] I. Sampedro, S. Marinari, A. D’Annibale, S. Grego, J. A. Ocampo, istry, vol. 57, no. 19, pp. 8798–8804, 2009.
and I. Garcı́a-Romera, “Organic matter evolution and partial [126] A. E. Asli, E. Boles, C. P. Hollenberg, and M. Errami, “Con-
detoxification in two-phase olive mill waste colonized by white- version of xylose to ethanol by a novel phenol-tolerant strain
rot fungi,” International Biodeterioration and Biodegradation, of Enterobacteriaceae isolated from olive mill wastewater,”
vol. 60, no. 2, pp. 116–125, 2007. Biotechnology Letters, vol. 24, no. 13, pp. 1101–1105, 2002.
[113] A. Jaouani, S. Sayadi, M. Vanthournhout, and M. J. Penninckx, [127] M. Chamkha, B. K. C. Patel, J.-L. Garcia, and M. Labat,
“Potent fungi for decolourisation of olive oil mill wastewaters,” “Isolation of Clostridium bifermentans from oil mill wastewaters
Enzyme and Microbial Technology, vol. 33, no. 6, pp. 802–809, converting cinnamic acid to 3-phenylpropionic acid and emen-
2003. dation of the species,” Anaerobe, vol. 7, no. 4, pp. 189–197, 2001.
[114] A. Goudopoulou, A. Krimitzas, and M. A. Typas, “Differential [128] M. Reverberi, F. Di Mario, and U. Tomati, “𝛽-glucan synthase
gene expression of ligninolytic enzymes in Pleurotus ostreatus induction in mushrooms grown on olive mill wastewaters,”
grown on olive oil mill wastewater,” Applied Microbiology and Applied Microbiology and Biotechnology, vol. 66, no. 2, pp. 217–
Biotechnology, vol. 88, no. 2, pp. 541–551, 2010. 225, 2004.
[115] J. Pérez, T. de La Rubia, O. Ben Hamman, and J. Martı́nez, [129] R. Garcı́a-de-la-Fuente, G. Cuesta, E. Sanchı́s-Jiménez, S.
“Phanerochaete flavido-alba laccase induction and modification Botella, M. Abad, and F. Fornes, “Bacteria involved in sulfur
of manganese peroxidase isoenzyme pattern in decolorized amendment oxidation and acidification processes of alkaline
olive oil mill wastewaters,” Applied and Environmental Micro- ‘alperujo’ compost,” Bioresource Technology, vol. 102, no. 2, pp.
biology, vol. 64, no. 7, pp. 2726–2729, 1998. 1481–1488, 2011.
[116] M. Garcı́a-Sánchez, I. Garrido, I. J. Casimiro et al., “Defence [130] T. Yangui, A. Rhouma, M. A. Triki, K. Gargouri, and J. Bouzid,
response of tomato seedlings to oxidative stress induced by “Control of damping-off caused by Rhizoctonia solani and
phenolic compounds from dry olive mill residue,” Chemosphere, Fusarium solani using olive mill waste water and some of its
vol. 89, no. 6, pp. 708–716, 2012. indigenous bacterial strains,” Crop Protection, vol. 27, no. 2, pp.
[117] L. Abrunhosa, F. Oliveira, D. Dantas, C. Gonçalves, and I. 189–197, 2008.
Belo, “Lipase production by Aspergillus ibericus using olive mill [131] I. Vagelas, A. Papachatzis, H. Kalorizou, and E. Wogiatzi, “Bio-
wastewater,” Bioprocess and Biosystems Engineering, vol. 36, no. logical control of Botrytis fruit rot (Gray mold) on strawberry
3, pp. 285–291, 2013. and red pepper fruits by olive oil mill wastewater,” Biotechnology
[118] E. Aranda, I. Sampedro, J. A. Ocampo, and I. Garća-Romera, and Biotechnological Equipment, vol. 23, no. 4, pp. 1489–1491,
“Contribution of hydrolytic enzymes produced by saprophytic 2009.
fungi to the decrease in plant toxicity caused by water-soluble [132] I. Vagelas, H. Kalorizou, A. Papachatzis, and M. Botu, “Bioac-
substances in olive mill dry residue,” Applied Microbiology and tivity of olive oil mill wastewater against plant pathogens
Biotechnology, vol. 64, no. 1, pp. 132–135, 2004. and post-harvest diseases,” Biotechnology and Biotechnological
[119] A. D’Annibale, V. Brozzoli, S. Crognale, A. M. Gallo, F. Federici, Equipment, vol. 23, no. 2, pp. 1217–1219, 2009.
and M. Petruccioli, “Optimisation by response surface method- [133] Z. Özdemir, “Growth inhibition of Clavibacter michiganensis
ology of fungal lipase production on olive mill wastewater,” subsp. michiganensis and Pseudomonas syringae pv. tomato
Journal of Chemical Technology and Biotechnology, vol. 81, no. by olive mill wastewaters and citric acid,” Journal of Plant
9, pp. 1586–1593, 2006. Pathology, vol. 91, no. 1, pp. 221–224, 2009.
[120] I. Sampedro, T. Cajthaml, S. Marinari, M. Petruccioli, S. Grego, [134] R. K. B. Jenana, R. Haouala, M. A. Triki et al., “Composts,
and A. D’Annibale, “Organic matter transformation and detox- compost extracts and bacterial suppressive action on Pythium
ification in dry olive mill residue by the saprophytic fungus aphanidermatum in tomato,” Pakistan Journal of Botany, vol. 41,
Paecilomyces farinosus,” Process Biochemistry, vol. 44, no. 2, pp. no. 1, pp. 315–327, 2009.
216–225, 2009. [135] T. Yangui, A. Rhouma, K. Gargouri, M. A. Triki, and J. Bouzid,
[121] J. A. Libra, M. Borchert, and S. Banit, “Competition strategies “Efficacy of olive mill waste water and its derivatives in the
for the decolorization of a textile-reactive dye with the white- suppression of crown gall disease of bitter almond,” European
rot fungi Trametes versicolor under non-sterile conditions,” Journal of Plant Pathology, vol. 122, no. 4, pp. 495–504, 2008.
Biotechnology and Bioengineering, vol. 82, no. 6, pp. 736–744, [136] N. Kavroulakis, S. Ntougias, M. I. Besi et al., “Antagonistic
2003. bacteria of composted agro-industrial residues exhibit antibio-
[122] P. Chiaiese, F. Palomba, F. Tatino et al., “Engineered tobacco and sis against soil-borne fungal plant pathogens and protection
microalgae secreting the fungal laccase POXA1b reduce phenol of tomato plants from Fusarium oxysporum f.sp. radicis-
content in olive oil mill wastewater,” Enzyme and Microbial lycopersici,” Plant and Soil, vol. 333, no. 1, pp. 233–247, 2010.
Technology, vol. 49, no. 6-7, pp. 540–546, 2011. [137] M. L. Cayuela, P. D. Millner, S. L. F. Meyer, and A. Roig,
[123] P.-P. Liebgott, M. Labat, L. Casalot, A. Amouric, and J. “Potential of olive mill waste and compost as biobased pesti-
Lorquin, “Bioconversion of tyrosol into hydroxytyrosol and cides against weeds, fungi, and nematodes,” Science of the Total
3,4-dihydroxyphenylacetic acid under hypersaline conditions Environment, vol. 399, no. 1–3, pp. 11–18, 2008.
by the new Halomonas sp. strain HTB24,” FEMS Microbiology [138] G. Bonanomi, V. Giorgi, D. S. Giovanni, D. Neri, and F. Scala,
Letters, vol. 276, no. 1, pp. 26–33, 2007. “Olive mill residues affect saprophytic growth and disease
incidence of foliar and soilborne plant fungal pathogens,”

Agriculture, Ecosystems and Environment, vol. 115, no. 1–4, pp.
194–200, 2006.
[139] G. Cuesta, R. Garcı́a-de-la-Fuente, M. Abad, and F. Fornes,
“Isolation and identification of actinomycetes from a compost-
amended soil with potential as biocontrol agents,” Journal of
Environmental Management, vol. 95, pp. S280–S284, 2012.
[140] H. A. J. Hoitink and M. J. Boehm, “Biocontrol within the
context of soil microbial communities: a substrate-dependent
phenomenon,” Annual Review of Phytopathology, vol. 37, no. 1,
pp. 427–446, 1999.
[141] N. Kavroulakis, S. Ntougias, G. I. Zervakis, C. Ehaliotis, K.
Haralampidis, and K. K. Papadopoulou, “Role of ethylene in the
protection of tomato plants against soil-borne fungal pathogens
conferred by an endophytic Fusarium solani strain,” Journal of
Experimental Botany, vol. 58, no. 14, pp. 3853–3864, 2007.
[142] S. Ntougias, G. I. Zervakis, and C. Fasseas, “Halotalea alkalilenta
gen. nov., sp. nov., a novel osmotolerant and alkalitolerant bac-
terium from alkaline olive mill wastes, and emended descrip-
tion of the family Halomonadaceae Franzmann et al. 1989,
emend. Dobson and Franzmann 1996,” International Journal
of Systematic and Evolutionary Microbiology, vol. 57, no. 9, pp.
1975–1983, 2007.
[143] S. Ntougias, C. Fasseas, and G. I. Zervakis, “Olivibacter sitiensis
gen. nov., sp. nov., isolated from alkaline olive-oil mill wastes
in the region of Sitia, Crete,” International Journal of Systematic
and Evolutionary Microbiology, vol. 57, no. 2, pp. 398–404, 2007.
[144] R. Stepanauskas, “Single cell genomics: an individual look at
microbes,” Current Opinion in Microbiology, vol. 15, no. 5, pp.
613–620, 2012.
[145] T. Woyke, G. Xie, A. Copeland et al., “Assembling the marine
metagenome, one cell at a time,” PLoS One, vol. 4, no. 4, Article
ID e5299, 2009.
[146] T. Woyke, D. Tighe, K. Mavromatis et al., “One bacterial cell,
one complete genome,” PLoS One, vol. 5, no. 4, Article ID e10314,
2010.
[147] O. U. Mason, T. C. Hazen, S. Borglin et al., “Metagenome,
metatranscriptome and single-cell sequencing reveal microbial
response to Deepwater Horizon oil spill,” The ISME Journal, vol.
6, no. 9, pp. 1715–1727, 2012.
http://dx.doi.org/10.1155/2013/405708
Research Article
Diversity and Antimicrobial Properties of Lactic Acid
Bacteria Isolated from Rhizosphere of Olive Trees and Desert
Truffles of Tunisia
Imene Fhoula, Afef Najjari, Yousra Turki, Sana Jaballah,

Abdelatif Boudabous, and Hadda Ouzari
Université de Tunis El Manar, Faculté des Science de Tunis, LR03ES03 Laboratoire Microorganismes et Biomolécules Actives,
2092 Tunis, Tunisia
Correspondence should be addressed to Hadda Ouzari; imene.ouzari@fst.rnu.tn
Received 30 April 2013; Revised 30 July 2013; Accepted 10 August 2013
Copyright © 2013 Imene Fhoula et al. This is an open access article distributed under the Creative Commons Attribution License,
A total of 119 lactic acid bacteria (LAB) were isolated, by culture-dependant method, from rhizosphere samples of olive trees and
desert truffles and evaluated for different biotechnological properties. Using the variability of the intergenic spacer 16S-23S and 16S
rRNA gene sequences, the isolates were identified as the genera Lactococcus, Pediococcus, Lactobacillus, Weissella, and Enterococcus.
All the strains showed proteolytic activity with variable rates 42% were EPS producers, while only 10% showed the ability to
grow in 9% NaCl. In addition, a low rate of antibiotic resistance was detected among rhizospheric enterococci. Furthermore,
a strong antibacterial activity against plant and/or pathogenic bacteria of Stenotrophomonas maltophilia, Pantoea agglomerans,
Pseudomonas savastanoi, the food-borne Staphylococcus aureus, and Listeria monocytogenes was recorded. Antifungal activity
evaluation showed that Botrytis cinerea was the most inhibited fungus followed by Penicillium expansum, Verticillium dahliae,
and Aspergillus niger. Most of the active strains belonged to the genera Enterococcus and Weissella. This study led to suggest that
environmental-derived LAB strains could be selected for technological application to control pathogenic bacteria and to protect
food safety from postharvest deleterious microbiota.
1. Introduction products but also the environment safeguarding [4, 5]. Such
bacteria are known by “biological control agents” [6].
Given the world’s growing demand for food, more attention Lactic acid bacteria form an ecologically heterogeneous
is needed for food preservation, postharvest, and agricultural group of Gram-positive bacteria, nonspore forming, immo-
product preservation from different harmful factors such as bile, and catalase negative, that excretes lactic acid as
the contamination caused by microbial spoilage and toxic major end product and generally recognized as safe (GRAS)
metabolites produced by yeast, mold, and/or bacteria [1, 2], as organisms [7]. They are also selected as probiotic, which
well as the extensive use of synthetic chemicals and pesticides are able to promote health and prevent infections against
in food and agriculture. These factors may pose a health risk enteropathogenic bacteria [8, 9]. LAB are usually harbor
for human and animals and affect the ecological equilibrium carbohydrate-rich environments and found in various food
of the environment [3]. products such as milk, plant, meat, intestinal mucosa of
Therefore, there is growing interest to establish alternative human, and animals [8, 10] but especially proliferate in
bioproducts to replace chemicals and toxic pesticides. For this different fermented foods [11]. Owing to particular physi-
purpose, using bacteria or natural compounds which exhibit ological and biochemical traits, such as exopolysaccharide
the same inhibitory effect on phytopathogenic and spoilage production, organic acids, aromatic compounds, tolerance
microbes was not only shown to be efficient in storage life to low water activity, and antimicrobial production [5, 12,
extension and nutritive and safety value retention, of food 13], LAB found different industrial applications, either by
their biopreservatives or techno-functional properties [14]. containing 5 mL of MRS broth (Biolife) and incubated in
In fact, many authors reported that some LAB strains are anaerobic candle jars at 30∘ C for 3 days. After incubation,
able to inhibit food-borne pathogens such as Staphylococcus samples were serially diluted in 0.75% NaCl solution. Frac-
aureus, Salmonella typhimurium, Escherichia, coli and Listeria tions of 0.1 mL of the dilutions ranging between 10−5 and 10−8
monocytogenes [15, 16]. In addition, LAB are efficient to were plated in duplicate on the surface of MRS agar (Biolife)
inhibit mycotoxicogenic fungi (Penicillium expansum, Botry- [24] supplemented with 0.0025% of bromocresol green (MP
tis cinerea, Aspergillus niger, Aspergillus flavus, and Fusarium Biomedicals) and 0.01% cycloheximide (MP Biomedicals)
graminarum) [17, 18] as well as phytopathogenic bacteria to inhibit fungal growth. The plates were incubated in the
(such as Xanthomonas campestris and Erwinia carotovora) same conditions. The different colonies of acid-producing
[17]. bacteria, determined by a yellow zone in the media around
Data reporting LAB isolation from soils and plants each colony, were picked and purified on MRS agar. Gram-
remain scarce. However, environmental and wild LAB strains positive and catalase-negative isolates were selected and
are theoretically good competitors for different growth fac- maintained in broth with 25% glycerol at −80∘ C for further
tors and production of antagonistic compounds but often identification. The isolates were also tested for gas production
undervalued. Olive tree is one of the most important crops in from D-glucose (Bio Basic) (using inverted Durham tubes in
Tunisia, from North to the South but also accompanied all the MRS broth), growth at different temperatures (10 and 45∘ C),
Mediterranean civilizations. It is recognized for its beneficial different pH (4.0 and 9.6), and different concentration of
effects on human health, even by olive oil or by different NaCl (3, 6.5, 8, and 9%) in MRS broth.
derived products [19]. Moreover, truffles are ectomycorrhizal
consumable tuber, which are typical of semiarid land and 2.3. DNA Extraction and PCR Amplification of 16S-23S rDNA
are known by their important economical income for local Internal Transcribed Spacer and the 16S rDNA Gene. DNA
population and their good taste [20, 21]. The specificity of was extracted by using a CTAB/NaCl method described
rhizospheric samples for bacterial isolation is their direct by Wilson [25] and modified by using 1 mg/mL lysozyme
contact with both plant and soil, but especially because the (BIOMATIK) for cell wall digestion. DNA electrophoresis
associated bacteria have coevolved with plant pathogenic was performed on a 0.8% agarose gel and visualized under
bacteria and fungi. This study aimed to isolate LAB from olive UV light according to the standard procedure of [26]. The
tree and desert-truffle rhizospheric soils and to evaluate their PCR amplifications were performed using a thermal cycler
biotechnological properties. (Thermal Cycler; Bio-Rad).
The bacterial collection was dereplicated by fingerprint-
2. Materials and Methods ing analysis of the rRNA 16S-23S intergenic transcribed
spacer (ITS) region, using universal primers, s-d-bact-1494-
2.1. Samples and Microbial Strains Origin. LAB strains were a-20 and s-d-bact-0035-a-15 [27]. The ITS-PCR amplification
obtained from rhizospheric samples (49) which were col- consisted of 1X PCR reaction buffer, 1.5 mM MgCl2 , 0.2 mM
lected from 8 sites located in the following regions in Tunisia: of dNTPs mixture, 0.5 𝜇M of each primer, 1 U Taq polymerase
Jendouba, Ben Arous, Tunis, Kairouan, Gafsa, kebeli, Gabes, (Fermentas), and 150 ng of total DNA, using the following
and Mednine. Samples were collected in sterilized bags, kept program: 94∘ C for 3 min, followed by 35 cycles of 94∘ C
in cool box (<10∘ C) containing ice packs during the trans- for 45∘ C, 55∘ C for 1 min and 72∘ C for 2 min, and a final
port to laboratory, and processed within 7 days. Different extension step at 72∘ C for 7 min. Strains exhibiting the same
other microbial species were used in antimicrobial testing. band patterns were grouped in the same ITS-haplotype. One
Pseudomonas savastanoi knW2, Pantoea agglomerans kn45, or two representative strains from each group have been
and Stenotrophomonas maltophilia KnT2 were previously selected for subsequent identification using 16S rRNA genes
isolated from olive knots [22]. Listeria monocytogenes L15 sequencing. The 16S rRNA amplification was performed
and Botrytis cinerea were obtained from the Laboratory of using the describe primers s-d-bact-0008-a-S-20 and s-d-
Microorganisms and Active Biomolecules (LMBA), Faculty bact-1495-a-A20 [27] and the thermal profile as mentioned
of Sciences of Tunis. Penicillium expansum and Aspergillus previously. The ITS-PCR amplification and 16S products were
niger from Laboratory of Microbial Ecology and Biotech- migrated, respectively, on 2 and 1.5% agarose gels in 0.5 ×
nology, University of Paul Cézanne, France and Verticillium Tris-borate-EDTA buffer (reagents from Fluka-Biochemika)
dahliae from the National Institute of Agronomic Research of and stained with ethidium bromide (Sigma-Aldrich).
Tunis (INRAT). Reference strains from the American Type
Culture Collection (ATCC) were also used including Ente-
2.4. 16S rDNA Gene Sequencing and Phylogenetic Analysis.
rococcus faecium ATCC 19434, Enterococcus faecalis ATCC
The 16S rDNA PCR amplicons were purified with
29212, Staphylococcus aureus ATCC 25923, and S. aureus
Exonuclease-I and Shrimp Alkaline Phosphatase (Exo-
ATCC 6538.
Sap, Fermentas, Life Sciences) following the manufacturer’s
standard protocol. Sequence analyses of the purified DNAs
2.2. Lactic Acid Bacteria Isolation Procedure. The LAB from were performed using a Big Dye Terminator cycle sequencing
rhizospheres were isolated by the accumulation method as kit V3.1 (Applied Biosystems) and an Applied Biosystems
described by Chen et al. [23], with some modifications. 3130XL Capillary DNA Sequencer machine. Sequence
Samples of 1 g were aseptically transferred into tubes of 15 mL similarities were found by BLAST analysis [28] using the
GenBank DNA databases (http://www.ncbi.nih.gov) and the development of a mucoid colony on agar medium or long
Ribosomal Database Project (RDP). Phylogenetic analysis of filaments (when the colony is extended with an inoculation
the 16S rRNA gene sequences were conducted with Molecular loop) indicated the production of exopolysaccharides [33].
Evolutionary Genetics Analysis (MEGA) software, version As well, LAB were assessed for proteolytic activity by agar-
5 [29]. Trees were constructed by using neighbor-joining well-diffusion test in MRS containing 4% of skimmed milk
method [30]. (Scharlau). The diameter of the proteolysis zone was deter-
mined after incubation under anaerobic conditions at 30∘ C
2.5. Nucleotide Sequence Accession Numbers. The sequences for 72 h and examined for clear zone around the wells.
of the 16S rDNA gene of rhizospheric LAB isolates samples
have been submitted to the GenBank databases under acces- 2.9. Antibiotic Susceptibility. The antibiotic susceptibility was
sion numbers KC568531 to KC568560. tested by disk diffusion method on BHIA as recommended by
the standard criteria (CLSI, 2010). The antibiotics used (Bio-
2.6. Antibacterial Activity of LAB against Pathogen, Food- Rad Laboratoires, Hercules, CA, USA) for susceptibility of
Borne, and Phytopathogenic Bacteria. The antibacterial activ- enterococci were ampicillin (AM; 10 𝜇g), ciprofloxacin (CIP;
ity test was performed using the agar-well-diffusion method 5 𝜇g), chloroamphenicol (C; 30 𝜇g), erythromycin (E; 15 𝜇g),
described by Tagg and McGiven [31]. Five bacterial strains gentamycin (GM; 120 𝜇g), streptomycin (S; 300 𝜇g), tetracy-
were used as indicators to evaluate the antibacterial activity cline (TE; 30 𝜇g), teicoplanin (TEI; 30 𝜇g), and vancomycin
of LAB, involving S. aureus ATCC6538, L. monocytogenes (VAN; 30 𝜇g). Antibiotic discs were placed on solid media
L15, St. maltophilia, Ps. savastanoi, and Pa. agglomerans. The and incubated at 37∘ C for 24 h. Based on the inhibition zone
cell-free supernatants (CFS) of LAB culture (48 h) in MRS size, the results were interpreted as resistant (R), intermediate
broth were tested. All indicator strains were grown in BHI resistant (IR), or susceptible to the antimicrobial agents (S).
broth at 37∘ C. Trypticase soy agar plates were overlaid with
5 mL of soft agar (0.75%) containing 50 𝜇L of freshly grown 3. Results and Discussion
culture. The wells were made in agar and filled with 100 𝜇L
of the tested strain CFS. After incubation at 37∘ C for 18 h, 3.1. Lactic Acid Bacteria Isolation. LAB isolates were initially
the diameter of the inhibition zones was measured. The selected based on their ability to produce lactic acid by
spectrum of inhibitory effect of LAB was than evaluated on the presence of yellow halo surrounding the colonies on
indicator bacteria: E. faecium ATCC19434, E. faecalis ATCC MRS-bromocresol green plates. Only Gram-positive strains
29212, S. aureus ATCC 6538, and S. aureus ATCC 25923. All exhibiting the absence of catalase and oxidase activity were
antibacterial tests were performed in triplicate. kept on MRS agar for further identification. In total, 119 LAB
strains were isolated from rhizospheric samples of desert truf-
2.7. Antifungal Activity of LAB. The LAB isolates were tested fles (4) and olive trees (49) from diverse geographic regions
against four phytopathogenic fungi of Aspergillus niger, Peni- in Tunisia (Table 1). LAB are usually isolated from fermented
cillium expansum, Botrytis cinerea, and Verticillium dahliae products of animal and vegetable origin. However, low rate
using the method described by Whipps [32] with some of “somnicells” of LAB [34] are naturally found in different
modifications. A dual culture of the tested pathogen fungi and environments which are close to these biota, such as floor of
the presumed antagonist LAB was established in MRS agar henhouse, rhizosphere of fruit trees, and around horse barn
without sodium acetate (MRS-SA). A mycelium plug of 5 mm [23, 35]. Although LAB isolation from soil and water remains
was taken from the peripheral edge of old cultures (5 days) scarce [23, 36], their presence in rhizospheric samples seems
on PDA plates of fungal pathogens and each plug was placed to be more supported by the abundance of root exudates [37].
at the centre of three replicate MRS-SA plates. Bacteria were
inoculated at 2 cm line from the edge of plates and allowed to 3.2. Ribotyping and Identification of Isolates by 16S rRNA
grow at 30∘ C for 48 h. Untreated control plates were plated Gene Sequence Analysis. Length polymorphism analysis of
with pathogen plugs only. Particularly for V. dahliae, the amplified 16S-23S internal transcribed spacer (ITS) was used
fungus was placed on MRS-SA five days in advance, due to select representative strains of the different taxonomic
to its relatively slower mycelium growth. All plates were units issued from the LAB collection. In fact, different studies
incubated on adequate growth temperature of the fungi, and have previously reported the usefulness of ITS dereplication
the percentage of growth inhibition was calculated by using for inter- and intradifferentiation at the genus/species level
the formula of Whipps [32]: [(R1 − R2)/R1] ∗ 100, where R1 is [38, 39] due to the high variability of these internal spacers.
the radial distance (mm) grown by phytopathogenic fungi in Based on this method, 16 different ITS-haplotypes designated
direction of the antagonist and R2 is the radial distance grown from A to L were distinguished. ITS-PCR patterns showed 1
by phytopathogenic fungi. to 4 reproducible bands ranging from 275 to about 600 bp
(Figure 1). The representative isolates of each ITS-type (30
2.8. Exopolysaccharide Production and Proteolytic Activity. isolates) were identified at species level by 16S rDNA gene
The exopolysaccharide production (EPS) was evaluated by sequence analysis and compared to the known sequences
streaking fresh culture of LAB isolates on MRS agar sup- in GenBank. Phylogenetic relationship between LAB was
plemented with 2% (w/v) of sucrose (Sigma, Life science). constructed based on the 16S rDNA sequences from evolu-
After incubation at 30∘ C under anaerobic condition for 72 h, tionary distances by the neighbor-joining method (Figure 2).
ITS-M
ITS-H
ITS-N
ITS-D
ITS-O
ITS-O
ITS-O
ITS-G
ITS-K
ITS-A
ITS-C
ITS-C
ITS-E
ITS-B
ITS-P
ITS-E
ITS-L
ITS-F
ITS-I
ITS-J
Mw
Mw
NC
1000 bp
300 bp
(a) (b)
Figure 1: Different ITS-haplotypes (A–P) of representative rhizospheric lactic acid bacteria. Mw, Molecular weight (100 bp); NC, negative
control.
Phylogenetic analysis revealed the differentiation of 5 clus- from soil [19, 35, 47, 48] a higher number of species diversity
ters (I–V) and 11 subclusters that include members of the is recorded in olive tree and truffle rhizospheric samples
genera Enterococcus, Lactobacillus, Pediococcus, Lactococcus, such as Lb. sakei, Pc. acidilactici, and W. halotolerans. These
Weissella, and Leuconostoc. The cluster I formed by the strains species are naturally found on several raw fermented food
of Enterococcus genus was divided into 2 groups. The first products of plant and animal origin [49–51]; moreover, Pc.
group included three species of E. faecium, corresponding to acidilactici is emerging as a potential probiotic in animal and
the ITS-types (A, B, and C), E. durans (ITS-type D), and E. human [52].
hirae (ITS-type E). The second group was only represented
by the strain (FS11) of E. faecalis being the most closely 3.3. Physiological and Technological Properties of LAB. The
related species in 100% of bootstrap analyses. The cluster II physiological and biochemical characteristics including salt
grouped strains of the genera Lactobacillus and Pediococcus tolerance, growth at different temperatures, and gas produc-
and presented four subclusters (3 to 6) of Lb. sakei, Lb. tion from glucose of all the strains are presented in Table 2.
plantarum, Pc. acidilactici, and Pc. pentosaceus based on the The majority of isolated strains were coccoid and coccoid-
ITS-types G, H, I, and J, respectively. The cluster III formed rods and only 5.9% showed rod shape. The majority of
by the strains of the genus Lactococcus was divided into two isolates were homofermentative, and only 13 (10.9%) were
subclusters (7 and 8) of L. lactis (ITS-type K) and L. garvieae heterofermentative. From the total isolates (𝑛 = 79) 66.4%
(ITS-type L). Furthermore, strains of the genus Weissella were and (𝑛 = 10) 8.4% of bacterial isolates grew well in low
grouped in the cluster IV, including three subclusters (9 to activity water, 8 and 9% NaCl, respectively. The LAB strains
11) of W. halotolerans (ITS-types M), W. paramesenteroides with high tolerance to 9% NaCl belonged to W. halotolerans
(ITS-types N), and W. confusa (ITS-types O). The cluster (FS58), W. confusa (FS66, FS44, FS53, FS54, and FS63), L.
V was represented by one strain of Ln. mesenteroides (ITS- lactis (LFS20), Lb. sakei (FS62), and E. faecium (FS77 and
type P). The used typing method showed that almost all the FS103). All LAB isolates grew well in pH 4.0 and 10∘ C. A
identified species were represented by one ITS-type, except total of 28 (23%) and 14 (11.5%) of isolates were not able
for E. faecium, which showed an intraspecies heterogeneity to grow in pH 9.6 and 45∘ C, respectively. Besides, LAB
with three major ITS-types (A, B, and C). In fact, E. faecium were screened for proteolytic activity and EPS production on
genome is extremely diverse [40] showing a high plasticity, MRS medium containing sucrose and skim milk, respectively.
due to the abundance of mobile genetic elements [41]. This Results showed that all LAB isolates exhibited proteolytic
result is in accordance with data published by Naı̈mi et activities in the cell-free supernatants as revealed by a clear
al. [42], Park et al. [43], and Brtkova et al. [44], which halo surrounding the wells. However, the proteolytic activity
reports the ITS region variability for E. faecium species. varied among the strains according to the halo diameters. In
Together with W. confusa, this species was found to be the fact, the more proteolytic strains (58.8%) exhibited a diameter
most isolated bacterium from rhizospheric samples (Table 1). greater than 15 mm. The most proteolytic activity (19 mm
Although enterococci are normal inhabitants of the human of diameter) was recorded for the strain Pc. acidilactici
and animal gastrointestinal tract [44, 45], they are widely FS46. The exopolysaccharide production was detected in
distributed in nature due to their high adaptation to various 42.8% of the isolates (Table 2). The good EPS-LAB producers
environmental conditions such as food, plants, water, and belonged mainly to the species W. confusa (12 strains), W.
soil [46]. Moreover, the isolation of bacteria belonging to the paramesenteroides (FS60 and FS45), and Ln. mesenteroides
genus Weissella was already reported either from soil [19] or FS13. The recorded physiochemical properties of the isolated
plants [17]. With regard to others studies on LAB recovery rhizospheric LAB, for instance proteolytic activity, tolerance
ITS- Phylogenetic
haplotypes group
FS065 C
FS026 A
63 FS118 B
FS019 B
FS010 A
99 Enterococcus faecium (EU418442)
67 FS029 D
Enterococcus durans (FJ917730) Enterococcus
100 63 FS039 E
Enterococcus hirae (GQ337029)
63 E
FS88
FS011 F
81 100 Enterococcus faecalis (HM462412)
100 FS033 G
Lactobacillus sakei (EU794737)
H Lactobacillus
FS119
100 Lactobacillus plantarum (AB755630)
97
FS035 H
71 77 88 FS046 I
Pediococcus acidilactici (FJ538581)
99 FS024 J
J Pediococcus
100 FS111
29 FS037 J
30 Pediococcus pentosaceus (AB481102)
100 FS075 K
Lactococcus lactis (EU84135) Lactococcus
98 FS031 L
100 Lactococcus garvieae (AY699289)
97 FS058 M
100 FS008 M
Weissella halotolerans (AB022926)
98 100 FS64 N
Weissella paramesenteroides (JQ806703)
FS044 O
99 FS027 O
O Weissella
FS066
97
FS052 O
77 67 Weissella confusa (EU807756)
FS025 O
FS004 O
7 FS061 O
49 FS053 O
Leuconostoc mesenteroides (GQ253514)
100 FS013 P Leuconostoc
Bordella pertussis (BPU04950)
0.05
Figure 2: Phylogenetic tree showing the relative position of lactic acid bacteria isolates based on 16S rDNA partial sequences, using the
neighbor-joining method. Bordetella pertussis was used as an out group. Bootstrap values for a total of 1000 replicates are shown at the nodes
of the tree, using MEGA-5. The scale bar corresponds to 0.05 units of the number of base substitutions per site.
to high NaCl concentration, and the EPS production could species, since they are the predominant isolates in the collec-
explain their survival in such oligotrophic environments. tion (Table 3) and could present a risk for antibiotic resistance
In particular, EPSs are typically correlated with bacterial gene dissemination. Rhizospheric enterococci showed low
resistance and protection against different stress conditions percentage of resistance to chloramphenicol (3.75%), ery-
such as desiccation, salt stress, and UV radiations [53, 54]. But thromycin (3.75%), streptomycin (7.5%), and tetracycline
it also generally implicated in their adherence to biological (8.75%). Nevertheless, all the strains were susceptible to
surface and sodium toxicity reduction [55]. teicoplanin, ampicillin, and gentamicin. Furthermore, some
strains (3.7%) exhibited intermediate resistance to van-
comycin and a high frequency of resistance to ciprofloxacin
3.4. Antibiotic Susceptibility Testing. The antibiotic suscep- (36.2%). In summary, the rhizospheric enterococci showed
tibility of the isolated was mainly checked for enterococcal a low frequency of resistance to the Gram-positive target
6
Table 1: Origin and identification of rhizospheric LAB isolates.

Geographical Source/number of soil Number of Closest 16S rDNA % of sequence
Sampling point Strains Strains (access number)
position samples isolates sequence similarity
FS01 Enterococcus hirae
R. of olive tree/03 FS02 Lactococcus lactis
Ben Arous 4
FS03 Enterococcus faecalis
SSR of olive tree 01 FS04 Weissella confusa FS04 ( KC568542) 99
Northeast FS07 Lactococcus lactis
Tunisia FS08 Weissella halotolerans FS08 (KC568554) 99
R. of olive tree/03 FS11 Enterococcus faecalis FS11 (KC568559) 99
Tunis 11 FS13 Leuconostoc mesenteroides FS13 (KC568533) 97
FS14 Enterococcus durans
FS09, FS10, FS12, FS15 Enterococcus faecium FS10 (KC568539) 99
SSR of olive tree/01 FS05, FS06 Enterococcus faecium
FS25 (KC568541), FS19
FS16, FS17, FS19, FS21, FS25, FS26, FS30,
(KC568549)
50 FS40, FS48, FS50, FS51, FS55, FS56, FS57, Enterococcus faecium 99
FS26 (KC568553), FS65
FS65
(KC568552)
FS18 Enterococcus faecalis
FS20 Lactococcus lactis
FS24 (KC568551), FS37
FS24, FS37, FS38, FS41 Pediococcus pentosaceus 99
R. of olive tree/21 (KC568550)
Northwest FS46 Pediococcus acidilactici FS46 (KC568555) 98
Tunisia
FS29, FS32, FS42, FS49 Enterococuus durans FS29 (KC568547) 98
Jendouba FS22, FS31 Lactococcus garviae FS31 (KC568548) 99
FS33, FS34, FS62 Lactobacillus sakei FS33 (KC568535) 100
FS35, FS59 Lactobacillus plantarum FS35 (KC568557) 99
FS39, FS43, Enterococcus hirae FS39 (KC568531) 99
Weissella
FS45, FS60, FS64 FS64 (KC568556) 99
paramesenteroides
FS58 Weissella halotolerans FS58 (KC568532) 99
FS61 (KC568543), FS53
FS36, FS44, FS52, FS53, FS54, FS61, FS63 Weissella confusa (KC568544), FS52 99
(KC568545),
SSR of olive tree/04 FS23, FS27, FS28, Weissella confusa FS 27 (KC568537) 99
FS47 Enterococcus faecium
FS69 Lactobacillus sakei
The middle FS66 Weissella confusa FS66 (KC568540) 99
Kairouan R. of olive tree/03 9
Tunisia FS73 Pediococcus pentosaceus
FS67, FS68, FS70, FS71, FS72, FS74 Enterococcus faecium
Table 1: Continued.
Geographical Source/number of soil Number of Closest 16S rDNA % of sequence
Sampling point Strains Strains (access number)
position samples isolates sequence similarity
FS76 Weissella confusa
R. of olive tree/02
3 FS75 Lactococcus lactis FS75 (KC568534) 99
Gafsa
SSR/01 FS77 Weissella confusa
FS81, FS86, FS100 Enterococcus hirae
R. of olive tree/06 FS78, FS79, FS80, FS82, FS83, FS84, FS85,
23 FS88 (KC568538) 99
FS87, FS88 Enterococcus faecium
South Tunisia Kebili
FS89, FS90, FS91, FS92, FS93, FS94, FS95,
FS98, FS99
SSR/01 FS96, FS97 Enterococcus faecium
FS101, FS102, FS103, FS105, FS106, FS107 Enterococcus faecium
Mednine R. of olive tree/03 7
FS104 Pediococcus pentosaceus
R. of truffle/04 FS111 Pediococcus pentosaeus FS111 (KC568560) 98
(Terfezia
12 FS119 Lactobacillus plantarum FS119 (KC568558) 99
boudieri/Pichoa)
FS110 Enterococcus durans
FS112 Enterococccus hirae
FS108, FS109, FS113, FS114, FS115, FS116,
Enterococcus faecium FS118 ( KC568536) 99
FS117, FS118
R: Rhizosphere samples; SSR: soil surrounding rhizosphere; the underlined strains refer to LAB isolated from the SSR.
7
8
Table 2: Phenotypic characteristics of representative Gram-positive rhizospheric-LAB isolates.

Lb. Pc.
E. E. E. Lb. Pc. pen- Lc. W. W. parame-
E. hirae plan- acidi- Lc. lactis W. confuse Ln.mesenteroides
faecium durans faecalis sakei tosaceus garvieae halotolerans senteroides
tarum lactici
ITS types A, B, C D E F G H I J K L M N O P
Number of
64 06 07 03 04 03 01 07 03 03 02 03 12 01
strains
Shape cocci cocci cocci cocci rods rods cocci cocci cocci cocci coccobacilli coccobacilli coccobacilli cocci
Fermentation
Homo Homo Homo Homo Homo Homo Homo Homo Homo Homo Hetero Hetero Hetero Hetero
type
Catalase − − − − − − − − − − − − − −
Growth at pH
4 + + + + + + + + + + + + + +
9.6 + + + + + + − − − − + + + +
Growth in
NaCl
3% + + + + + + + + + + + + + +
6.5% + +(03) +(04) +(02) +(03) + + +(06) +(03) + + + + +
8% +(42)∗ +(03) +(01) +(01) +(03) + + +(04) +(02) +(01) + + + +
9% +(02) − − − +(01) − − − +(01) − +(01) − +(05) −
Growth at
temperature
10∘ C + + + + + + + + + + + + + +
45∘ C + + + + − + + + − − − − − −
EPS
+(28) − − − − + + − +(02) +(02) − +(02) + +
production
Proteolytic
+ + + + + + + + + + + + + +
activity
(𝑥): number of strains; +: positive; −: negative; Homo: homofermentative; Hetero: heterofermentative.
Table 3: Antimicrobial susceptibility of the enterococci isolated from the rhizosphere soils.
E. faecium E. faecalis E. durans E. hirae

Antibiotics % resistance
(𝑛 = 64) (𝑛 = 3) (𝑛 = 6) (𝑛 = 7)
Penicillins AM 0 0 0 0 0
Aminoglycosides GM 0 0 0 0 0
TE 6 0 1 0 8,75
S 6 0 0 0 7,5
Chloramphenicols CH 3 0 0 0 3,75
Macrolides E 2 1 0 0 3,75
Glycopeptides VA 3 0 0 0 3,75
TEI 0 0 0 0 0
Fluoroquinolones CIP 28 1 0 0 36,25
AM: ampicillin, GM: gentamicine, TE: teteracyclin, S: streptomycin, E: erythromycin, C: chloramphenicol, VA: vancomycin, TEI: teicoplanin, and CIP:
ciprofloxacin. 𝑛: total number of strains; numbers indicated resistant strains within species.
Table 4: Antibacterial activity spectrum of neutralized cell-free supernatant of three LAB rhizospheric isolates.
Strains L. monocytogenes S. aureus ATCC S. aureus ATCC E. faecium E. faecalis ATCC

L15 6538 25923 ATCC 19129 29212
Leuconostoc mesenteroides FS013 21 ± 1.00 14 ± 1.00 12 ± 1.00 13 ± 1.00 10 ± 0.00
Weissella halotolerans FS008 17 ± 1.00 15 ± 1.00 20 ± 1.00 13.5 ± 1.00 11 ± 1.00
Enterococcus faecium FS071 16 ± 1.00 15 ± 1.05 19 ± 1.73 14.5 ± 0.80 15.5 ± 1.32
Numbers indicated the diameter of the inhibition zone in mm; each value represents the mean value standard deviation (SD) from three trials; values in the
same column differ significantly (𝑃 < 0.05).
antibiotics compared to food, clinical, and animal isolates infections [58, 59] and among the emergent multidrug
[56, 57]. This result indicates the safety of these bacteria for resistant Gram-negative bacteria [59]. With regard to the
a potential technological application. food-borne pathogen, efficient inhibition was recorded for W.
confusa FS054 strain (28 mm) against S. aureus, for E. faecium
FS106 against L. monocytogenes (20 mm) and for the strain W.
3.5. In Vitro Screening of the Antagonistic Activity of LAB
confusa FS036 against both S. aureus (20 mm), and L. mono-
against Human, Plant, and Food-Borne Pathogenic Bacteria.
cytogenes (24 mm). It is of interest to note that the genera of
LAB isolates were screened for antibacterial activity against
Enterococcus and Weissella may become potential biopreser-
human and plant pathogens, including St. maltophilia, Pa.
vation agents of food-poisoning and plant-borne species.
agglomerans, and Ps. savastanoi and food-borne bacteria of
This antibacterial activity exhibited by the majority of
S. aureus and L. monocytogenes (Figure 3(a)). According to
strains especially toward Gram-negative bacteria may be due
their inhibitory effects on pathogens, LAB were differentiated
to the organic acid effect or to other compounds active in
into three classes: strong inhibitor (with growth inhibition
acidic conditions. For this purpose, the inhibitory effect was
diameter (𝑑) ≥ 19 mm), medium (14 ≤ 𝑑 < 19 mm), and with
checked after supernatant neutralization. By this way, only
no significant inhibitory effect for a diameter less than 14 mm
three strains of E. faecium FS071, Ln. mesenteroides FS013,
(Figure 4(a)). The results showed that 64 strains (53.8%)
and W. halotolerans FS008 retained the inhibition ability
have significant inhibition against St. maltophilia, among
against the tested pathogens (Table 4), leading to suggest the
them 12 strains (10%) with strong inhibitory activity. This
presence of bacteriocin-like substances. This result was also
activity was recorded for the species of Lb. plantarum, Lb.
supported by the broad spectrum known for the majority of
sakei, Lc. garvieae, Ln. mesenteroides, and Pc. pentosaceus
the identified enterocins [60, 61]. Further studies should be
and mostly for the genera Enterococcus and Weissella. Five
conducted to elucidate the nature of the antibacterial metabo-
isolates (4%) including three E. faecium (FS70, FS01, and
lites produced by selected LAB, especially by W. halotolerans
FS03) and two Pc. pentosaceus (FS73 and FS24) showed
FS008 strain. Moreover, different studies proposed that ente-
strong inhibitory activity against Pa. agglomerans. Eleven
rococci may have a prospectively useful role in some dairy
isolates belonging to species W. confusa, Lc. Lactis, Lb.
products, due to their proteolytic and lipolytic activities, and
plantarum, Ln. mesenteroides, E. durans, and E. faecium
may then contribute to the development of the organoleptic
showed also strong inhibitory activity against Ps. savastanoi
properties of fermented foods, and also due to the production
(diameter varied between 19 to 28 mm). The recorded high
of enterocins with anti-Listeria activity [62, 63].
level of inhibition highlights the biotechnological potential of
rhizospheric-LAB to control phytopathogens, particularly for
Pa. agglomerans and St. maltophilia, which are also increas- 3.6. Antifungal Activity of LAB. LAB isolates were screened
ingly identified as important cause of nosocomial human for antifungal activity against soil-borne fungi of B. cinerea
(A) (B) (C)
(D) (E)
(a)
(A)
(B)
(C)
(D)
(b)
Figure 3: Antimicrobial activity of some rhizospheric LAB against pathogenic bacteria (a) Pa. agglomerans (A), St. maltophilia (B), Ps.
savastanoi (C), L. monocytogenes (D), and S. aureus (E) by-agar well-diffusion method [31] and phytopathogen fungi (b) P. expansium (A), A.
niger (B), B. cinerea (C), and V. dahliae (D) by dual culture [32].
90 80
80 70
Inhibition growth (%)

Inhibition growth (%)
70 60
60 50
50 40
40 30
30 20
20 10
10 0
0
Botrytis cinerea
Verticillium dahliae
Penicillium expansum
Aspergillus niger
S. aureus
St. maltophilia
Ps. savastanoi
Pa. agglumerans
L. monocytogenes
6538
knW 2
knT2
kn45
L15
Indicator strains
Strong activity (+++) Plant pathogenic fungi
Medium activity (++) Strong activity (x > 70%)
No significant inhibition (+) Weak activity (40 ≤ x < 70%)
(a) (b)
Figure 4: Histograms showing percentage of LAB having in vitro inhibitory effect on pathogenic and spoilage bacterial species (a) and plant
pathogenic fungi (b). The experiments were repeated at least three times.
and V. dahliae and postharvest contaminants of A. niger the emergence/selection of these bacteria as biocontrol agent
and P. expansum on MRS-SA agar medium (Figure 3(b)). and potential probiotic. Regarding the vascular wilt fungi
According to their degree of mycelium growth reduction, V. dahlia, Enterococcal strains were also shown to be the
active LAB were classified into two main groups: low most efficient inhibitor strains. The highest activity was
(reduction of mycelium growth between 40 and 70%) and observed for E. faecium FS82 with 75% of mycelium reduction
high antifungal activity (>70% of inhibition) (Figure 4(b)). (Table 5). This result constitute a first report on the strong
The results showed that the maximum growth inhibition inhibition of this soil-born-fungus, which is responsible of
rate (28% of the strains) was registered for Botrytis cinerea Verticillium wilt, a serious worldwide disease that affects
(Figure 4(b)). The group of LAB strains with strong inhibition many crops including fruits, vegetables, and oilseed rape and
activity (75.3 to 92.6% inhibition) belonged to the species leads to dramatically yield losses [67]. Biological compounds
of W. paramesenteroides, W. confusa, E. durans, E. faecium investigation to control this pathogen is of great significance
and E. hirae. Besides, the most inhibitor strains toward the [68], as it persists in the soil and resists different chemical
fungus A. niger were E. durans FS29 and four E. faecium, treatments. The present study showed that selected envi-
(FS50, FS06, FS48, and FS87) exhibiting an inhibition rate
ronmental LAB could offer an excellent source for active
between 76.7 and 90%. Furthermore, we noted that 16% of
metabolite to control different pathogenic bacteria and fungi.
the strains belonging to the species E. faecium, E. durans, W.
As it was reported by many authors [5, 13, 69], different
halotolerans, and Lb. plantarum showed a strong inhibition
substances, such as organic acids, hydrogen peroxide, cyclic
rate (75.3 to 87.8%) toward P. expansum.
dipeptides, and phenolic and proteinaceus compounds could
It is worth mentioning that strains of Enterococcus genus
be responsible for the detected antifungal activity. Indeed,
confirmed their antimicrobial efficacy by strong inhibition identification of the issued rhizospheric-LAB metabolites is
of most of the tested postharvest fungi (P. expansum, B. needed for a more target application.
cinerea, and A. niger) and highlight the potential use of
rhizospheric-LAB as biocontrol agents to prevent postharvest
deterioration caused by these fungi. In fact, most of these 4. Conclusion
fungi produce allergenic spores and mycotoxins which are
responsible of the spoilage and poisoning of foods leading In this study, we reported for the first time the isolation
to serious potential health hazards [64]. It is also the case of and characterization of LAB from rhizosphere samples of
Lb. plantarum FS119 which was isolated from desert truffle olive trees and desert truffles. The results showed a high rate
rhizosphere, and that could be considered as a potential of antimicrobial activity among the isolates, indicating that
candidate inhibitor of P. expansum, the agent of blue mold rhizosphere may be a common source for the selection of
in apples. In addition strains of the genus Weissella have LAB with important technological potential, which are useful
showed an efficient inhibition toward either pathogenic for the biocontrol of food-, plant-, and soil-borne pathogenic
bacteria, or the different tested fungi. This result is in accor- bacteria and fungi. Further investigations to elucidate the
dance with Valerio et al. [65] and Lee et al. [66] reporting nature of inhibiting compounds should be considered.
Table 5: Mycelium growth inhibition of four pathogenic fungi by selected potent antifungal rizospheric-LAB isolates using confrontation
assay.
Strains Species A. niger P. expansum B. cinerea V. dahlia

FS29 E. durans ++ (76.7)∗ ++ 80.2∗ ++ (85.1)∗ ++ (70.3)∗
FS50 E. faecium ++ (79.1)∗ + ++ (82.7)∗ +
FS06 E. faecium ++ (79)∗ ++ (75.3)∗ + +
FS87 E. faecium ++ (79.1)∗ + ++ (80.2)∗ +
FS48 E. faecium ++ (90)∗ ++ (72.8)∗ +++ (82.7)∗ +
FS82 E. faecium + + +++ (85.2)∗ +++ (75)∗
FS68 E. faecium + + ++ (75.3)∗ +
FS05 E. faecium + ++ (80.2)∗ ++ (72.8)∗ +
FS14 E. durans + ++ (77.7)∗ ++ (82.7)∗ −
FS21 E. faecium + ++ (77.7)∗ + ++ (70.9)∗
FS101 E. faecium + ++ (75.3)∗ ++ (79)∗ +
FS32 E. durans + ++ (72.8)∗ ++ (82.7)∗ +
FS107 E. faecium + − ++ (80.2)∗ +
FS45 W. paramesenteroides − + ++ (81.4)∗ +
FS61 W. confusa. − + ++ (85)∗ −
FS19 E. faecium + + ++ (77.7)∗ +
FS94 E. faecium + ++ (87.6)∗ ++ (82.7)∗ +
FS119 Lb. plantarum − ++ (83.8)∗ ++ (70)∗ +
FS49 E. durans − ++ (75.3)∗ + +
FS58 W. halotolerans + ++ (80.2)∗ + −
FS74 E. faecium − ++ (72.8)∗ ++ (85.2)∗ ++ (74.5)∗
FS51 E. faecium + ++ (72.8)∗ ++ (82.7)∗ ++ (70.9)∗
FS102 E. faecium − + ++ (81.2)∗ +
FS16 E. faecium − − ++ (92.6)∗ −
FS99 E. faecium − + ++ (91.3)∗ +
FS42 E. durans + + ++ (85.1)∗ +
FS12 E. faecium + + ++ (75.3)∗ +
FS65 E. faecium + + ++ (77.7)∗ +
FS15 E. faecium + + + ++ (70.3)∗
FS106 E. faecium + ++ (70.3)∗ + +
FS53 W. confusa − ++ (71)∗ + +
FS77 E. faecium − ++ (78.4)∗ + +
A. niger: Aspergillus niger, P. expansum: Penicillium expansum, B. cinerea: Botrytis cinerea, V. dahlia: Vercticillium dahliae. (+): weak antifungal activity having
an inhibition rate between 40 and 70%; (++): strong activity with an inhibition rate ≥ 70%; the strains characterized with a broad range against different
fungi appear in bold. Data were obtained at least three replicates. ∗ Means within column show statistically significant difference (𝑃 < 0.05) with a control
(nonexposed to the bacteria).
Acknowledgments [2] J. I. Pitt and A. D. Hocking, Fungi and Food Spoilage, A

Chapman and Hall Food Science Book, Aspen, Gaithersburg,
The authors thank financial support of the European Union Md, USA, 2nd edition, 1999.
in the ambit of Project 20 BIODESERT (EU FP7-CSA- [3] J. K. Huang, F. B. Qiao, L. X. Zhang, and S. Rozelle, “Farm pes-
SA REGPOT-2008-2, Grant agreement no. 245746) and the ticide, rice production, and human health,” EEPSEA Working
21 Tunisian Ministry of Higher Education and Scientific Paper, EEPSEA, Singapore, 2000.
research in the ambit of the laboratory 22 projects LR [4] J. Huang, R. Hu, C. Pray, F. Qiao, and S. Rozelle, “Biotechnology
MBA206 and LR11ES31. as an alternative to chemical pesticides: a case study of Bt cotton
in China,” Agricultural Economics, vol. 29, no. 1, pp. 55–67, 2003.
[5] M. E. Stiles, “Biopreservation by lactic acid bacteria,” Antonie
References van Leeuwenhoek, vol. 70, no. 4, pp. 331–345, 1996.
[1] G. J. Guynes and E. O. Bennett, “Bacterial deterioration of [6] N. Pastor, E. Carlier, J. Andrés, S. B. Rosas, and M. Rovera,
emulsion oils. I. Relationship between aerobes and sulfate- “Characterization of rhizosphere bacteria for control of phy-
reducing bacteria in deterioration,” Applied Microbiology, vol. topathogenic fungi of tomato,” Journal of Environmental Man-
7, no. 2, pp. 117–121, 1959. agement, vol. 95, pp. S332–S337, 2012.
[7] W. N. Konings, J. Kok, O. P. Kuipers, and B. Poolman, “Lactic [23] Y.-S. Chen, F. Yanagida, and I. Shinohara, “Isolation and iden-
acid bacteria: the bugs of the new millennium,” Current Opinion tification of lactic acid bacteria from soil using an enrichment
in Microbiology, vol. 3, no. 3, pp. 276–282, 2000. procedure,” Letters in Applied Microbiology, vol. 40, no. 3, pp.
[8] M. F. Fernández, S. Boris, and C. Barbés, “Probiotic properties 195–200, 2005.
of human lactobacilli strains to be used in the gastrointestinal [24] J. C. de Man, M. Rogosa, and M. E. Sharpe, “A medium for the
tract,” Journal of Applied Microbiology, vol. 94, no. 3, pp. 449– cultivation of lactobacilli,” Journal of Applied Microbiology, vol.
455, 2003. 23, no. 1, pp. 130–135, 1960.
[9] J. Skjermo and O. Vadstein, “Techniques for microbial control [25] K. Wilson, “Preparation of genomic DNA from bacteria,” in
in the intensive rearing of marine larvae,” Aquaculture, vol. 177, Current Protocols in Molecular Biology, F. M. Ausubel, R. Brent,
no. 1–4, pp. 333–343, 1999. R. E. Kingston et al., Eds., pp. 2. 4. 1–2. 4. 5, 1987.
[10] M. Amin, M. Jorfi, A. D. Khosravi, A. R. Samarbafzadeh, and [26] J. Sambrook, E. F. Fritsch, and T. Maniatis, Molecular Cloning:
A. F. Sheikh, “Isolation and identification of Lactobacillus casei A Laboratory Manual, Cold Spring Harbor Laboratory, Cold
and Lactobacillus plantarum from plants by PCR and detection Spring Harbor, NY, USA, 2nd edition, 1989.
of their antibacterial activity,” Journal of Biological Sciences, vol. [27] D. Daffonchio, S. Borin, G. Frova, P. L. Manachini, and C.
9, no. 8, pp. 810–814, 2009. Sorlini, “PCR fingerprinting of whole genomes: the spacers
[11] A. H. Soomro, T. Masud, and K. Anwaar, “Role of lactic acid between the 16s and 23S rRNA genes and of intergenic tRNA
bacteria in food preservation and human health—a review,” gene regions reveal a different intraspecific genomic variability
Pakistan Journal of Nutrition, vol. 1, no. 1, pp. 20–24, 2002. of Bacillus cereus and Bacillus licheniformis,” International Jour-
[12] G. Giraffa, “Microbial polysaccharides produced by lactic acid nal of Systematic Bacteriology, vol. 48, no. 1, pp. 107–116, 1998.
bacteria in the dairy industry,” Industrie Alimentari, vol. 33, no. [28] S. F. Altschul, W. Gish, W. Miller, E. W. Myers, and D. J. Lipman,
324, pp. 295–298, 1994. “Basic local alignment search tool,” Journal of Molecular Biology,
[13] M. Mataragas, E. H. Drosinos, and J. Metaxopoulos, “Antagonis- vol. 215, no. 3, pp. 403–410, 1990.
tic activity of lactic acid bacteria against listeria monocytogenes [29] K. Tamura, D. Peterson, N. Peterson, G. Stecher, M. Nei, and
in sliced cooked cured pork shoulder stored under vacuum or S. Kumar, “MEGA5: molecular evolutionary genetics analysis
modified atmosphere at 4 ± 2∘ C,” Food Microbiology, vol. 20, no. using maximum likelihood, evolutionary distance, and max-
2, pp. 259–265, 2003. imum parsimony methods,” Molecular Biology and Evolution,
[14] R. Bizzarro, G. T. Tarelli, G. Giraffa, and E. Neviani, “Phenotypic vol. 28, no. 10, pp. 2731–2739, 2011.
and genotypic characterization of lactic acid bacteria isolated [30] N. Saitou and M. Nei, “The neighbor-joining method: a new
from Pecorino Toscano cheese,” International Journal of Food method for reconstructing phylogenetic trees,” Molecular Biol-
Science, vol. 12, no. 3, pp. 303–316, 2000. ogy and Evolution, vol. 4, no. 4, pp. 406–425, 1987.
[15] R. K. Darsanaki, M. L. Rokhi, M. A. Aliabadi, and K. Issazadeh, [31] J. R. Tagg and A. R. McGiven, “Assay system for bacteriocins,”
“Antimicrobial activities of Lactobacillus strains isolated from Applied Microbiology, vol. 21, no. 5, p. 943, 1971.
fresh vegetables,” Middle-East Journal of Scientific Research, vol. [32] J. M. Whipps, “Effect of media on growth and interactions
11, no. 9, pp. 1216–1219, 2012. between a range of soil-borne glasshouse pathogens and antag-
[16] M. Jamuna and K. Jeevaratnam, “Isolation and partial char- onistic fungi,” New Phytology, vol. 107, no. 1, pp. 127–142, 1987.
acterization of bacteriocins from Pediococcus species,” Applied [33] P. Ruas-Madiedo and C. G. de los Reyes-Gavilán, “Invited
Microbiology and Biotechnology, vol. 65, no. 4, pp. 433–439, review: methods for the screening, isolation, and characteri-
2004. zation of exopolysaccharides produced by lactic acid bacteria,”
[17] R. Trias, L. Bañeras, E. Montesinos, and E. Badosa, “Lactic Journal of Dairy Science, vol. 88, no. 3, pp. 843–856, 2005.
acid bacteria from fresh fruit and vegetables as biocontrol [34] U. Behrendt, T. Müller, and W. Seyfarth, “The influence of
agents of phytopathogenic bacteria and fungi,” International extensification in grassland management on the populations of
Microbiology, vol. 11, no. 4, pp. 231–236, 2008. micro-organisms in the phyllosphere of grasses,” Microbiologi-
[18] P. Lavermicocca, F. Valerio, A. Evidente, S. Lazzaroni, A. cal Research, vol. 152, no. 1, pp. 75–85, 1997.
Corsetti, and M. Gobbetti, “Purification and characterization of [35] F. Yanagida, Y. Chen, and T. Shinohara, “Isolation and charac-
novel antifungal compounds from the sourdough Lactobacillus terization of lactic acid bacteria from soils in vineyards,” The
plantarum strain 21B,” Applied and Environmental Microbiology, Journal of General and Applied Microbiology, vol. 51, no. 5, pp.
vol. 66, no. 9, pp. 4084–4090, 2000. 313–318, 2005.
[19] L. Laitman, “Le marché et la production de l’huile d’olive en [36] F. Yanagida, Y. Chen, and M. Yasaki, “Isolation and charac-
Tunisie,” Annales de Géographie, vol. 62, no. 332, pp. 271–286, terization of lactic acid bacteria from lakes,” Journal of Basic
1953. Microbiology, vol. 47, no. 2, pp. 184–190, 2007.
[20] A. Morte, M. Zamora, A. Gutiérrez, and M. Honrubia, “Desert [37] P. D. Kiely, J. M. Haynes, C. H. Higgins et al., “Exploiting new
truffle cultivation in semiarid Mediterranean areas,” in Mycor- systems-based strategies to elucidate plant-bacterial interac-
rhizas Functional Processes and Ecological Impact, C. Azcón- tions in the rhizosphere,” Microbial Ecology, vol. 51, no. 3, pp.
Aguilar et al., Ed., chapter 15, pp. 221–233, Springer, Berlin, 2009. 257–266, 2006.
[21] A. Slama, Z. Fortas, M. Neffati, L. Khabar, and A. Boudabous, [38] V. Gürtler and V. A. Stanisich, “New approaches to typing
“Etude taxinomique de quelques Ascomycota hypogés (Ter- and identification of bacteria using the 16S-23S rDNA spacer
feziaceae) de la Tunisie méridionale,” Bulletin de la Société region,” Microbiology, vol. 142, no. 1, pp. 3–16, 1996.
Mycologique de France, vol. 122, no. 2-3, pp. 187–195, 2006. [39] D. Daffonchio, A. Cherif, and S. Borin, “Homoduplex and
[22] H. Ouzari, A. Khsairi, N. Raddadi et al., “Diversity of auxin- heteroduplex polymorphisms of the amplified ribosomal 16S-
producing bacteria associated to Pseudomonas savastanoi- 23S internal transcribed spacers describe genetic relationships
induced olive knots,” Journal of Basic Microbiology, vol. 48, no. in the ‘Bacillus cereus group’,” Applied and Environmental
5, pp. 370–377, 2008. Microbiology, vol. 66, no. 12, pp. 5460–5468, 2000.
[40] H. L. Leavis, R. J. L. Willems, W. J. B. van Wamel, F. H. Schuren, [56] H. Abriouel, N. B. Omar, A. C. Molinos et al., “Comparative
M. P. M. Caspers, and M. J. M. Bonten, “Insertion sequence- analysis of genetic diversity and incidence of virulence factors
driven diversification creates a globally dispersed emerging and antibiotic resistance among enterococcal populations from
multiresistant subspecies of E. faecium,” PLoS Pathogens, vol. 3, raw fruit and vegetable foods, water and soil, and clinical
no. 10, p. 37, 2007. samples,” International Journal of Food Microbiology, vol. 123,
[41] W. van Schaik, J. Top, D. R. Riley et al., “Pyrosequencing- no. 1-2, pp. 38–49, 2008.
based comparative genome analysis of the nosocomial pathogen [57] P. Poeta, D. Costa, J. Rodrigues, and C. Torres, “Antimicrobial
Enterococcus faecium and identification of a large transferable resistance and the mechanisms implicated in faecal enterococci
pathogenicity island,” BMC Genomics, vol. 11, no. 1, article R239, from healthy humans, poultry and pets in Portugal,” Interna-
18 pages, 2010. tional Journal of Antimicrobial Agents, vol. 27, no. 2, pp. 131–137,
[42] A. Naı̈mi, G. Beck, and C. Branlant, “Primary and secondary 2006.
structures of rRNA spacer regions in enterococci,” Microbiology, [58] A. T. Cruz, A. C. Cazacu, and C. H. Allen, “Pantoea agglomer-
vol. 143, no. 3, pp. 823–834, 1997. ans, a plant pathogen causing human disease,” Journal of Clinical
[43] Y. Park, E. Oh, B. K. Kim, S. M. Kim, and S. I. Shim, “Phenotypic Microbiology, vol. 45, no. 6, pp. 1989–1992, 2007.
characteristics of Enterococcus faecium variants confirmed by [59] L. Wjohn, N. Masashi, and M. Kathrin, “Stenotrophomonas
intergenic ribosomal polymerase chain reaction and E. faecium maltophilia: an emerging opportunist human pathogen,” The
polymerase chain reaction,” Diagnostic Microbiology and Infec- Lancet Infectious Diseases, vol. 9, no. 5, pp. 312–323, 2009.
tious Disease, vol. 34, no. 4, pp. 269–273, 1999. [60] G. Giraffa, D. Carminati, and G. T. Tarelli, “Inhibition of listeria
[44] A. Brtkova, M. Filipova, H. Drahovska, and H. Bujdakova, innocua in milk by bacteriocin-producing Enterococcus faecium
“Characterization of enterococci of animal and environmental 7C5,” Journal of Food Protection, vol. 58, no. 6, pp. 621–623, 1995.
origin using phenotypic methods and comparison with PCR [61] R. S. Kumar, P. Kanmani, N. Yuvaraj, K. A. Paari, V. Pattukumar,
based methods,” Veterinarni Medicina, vol. 55, no. 3, pp. 97–105, and V. Arul, “Purification and characterization of enterocin
2010. MC13 produced by a potential aquaculture probiont Entero-
[45] G. Giraffa, “Enterococci from foods,” FEMS Microbiology coccus faecium MC13 isolated from the gut of Mugil cephalus,”
Reviews, vol. 26, no. 2, pp. 163–171, 2002. Canadian Journal of Microbiology, vol. 57, no. 12, pp. 993–1001,
[46] J. O. Mundt, “Occurrence of enterococci on plants in a wild 2011.
environment,” Applied Microbiology, vol. 11, pp. 141–144, 1963. [62] A. Ahmadova, S. Dimov, I. Ivanova et al., “Proteolytic activities
[47] M. Zamudio-Maya, J. Narváez-Zapata, and R. Rojas-Herrera, and safety of use of Enterococci strains isolated from traditional
“Isolation and identification of lactic acid bacteria from sedi- Azerbaijani dairy products,” European Food Research and Tech-
ments of a coastal marsh using a differential selective medium,” nology, vol. 233, no. 1, pp. 131–140, 2011.
Letters in Applied Microbiology, vol. 46, no. 3, pp. 402–407, 2008. [63] S. Maisnier-Patin, E. Forni, and J. Richard, “Purification, partial
[48] M. P. Lutz, V. Michel, C. Martinez, and C. Camps, “Lactic characterisation and mode of action of enterococcin EFS2, an
acid bacteria as biocontrol agents of soil-borne pathogens,” antilisterial bacteriocin, produced by a strain of Enterococcus
Biological Control of Fungal and Bacterial Plant Pathogens, vol. faecalis isolated from a cheese,” International Journal of Food
78, pp. 285–288, 2012. Microbiology, vol. 30, no. 3, pp. 255–270, 1996.
[49] W. P. Hammes, A. Bantleon, and S. Min, “Lactic acid bacteria [64] L. Nickelsen and M. Jakobsen, “Quantitative risk analysis of
in meat fermentation,” FEMS Microbiology Reviews, vol. 87, no. aflatoxin toxicity for the consumers of “kenkey”—a fermented
1-2, pp. 165–173, 1990. maize product,” Food Control, vol. 8, no. 3, pp. 149–159, 1997.
[50] O. Kandler, U. Schillinger, and N. Weiss, “Lactobacillus halotol- [65] F. Valerio, M. Favilla, P. de Bellis, A. Sisto, S. de Candia, and
erans sp. nov., nom. rev. and Lactobacillus minor sp. nov., nom. P. Lavermicocca, “Antifungal activity of strains of lactic acid
rev,” Systematic and Applied Microbiology, vol. 4, no. 2, pp. 280– bacteria isolated from a semolina ecosystem against Penicillium
285, 1983. roqueforti, Aspergillus niger and Endomyces fibuliger contami-
[51] F. Leroy and L. de Vuyst, “Lactic acid bacteria as functional nating bakery products,” Systematic and Applied Microbiology,
starter cultures for the food fermentation industry,” Trends in vol. 32, no. 6, pp. 438–448, 2009.
Food Science and Technology, vol. 15, no. 2, pp. 67–78, 2004. [66] K. W. Lee, J. Y. Park, H. R. Jeong, H. J. Heo, N. S. Han, and J.
[52] N. P. Guerra, A. T. Agrasar, C. L. Macı́as, P. F. Bernárdez, and H. Kim, “Probiotic properties of Weissella strains isolated from
L. P. Castro, “Dynamic mathematical models to describe the human faeces,” Anaerobe, vol. 18, no. 1, pp. 96–102, 2012.
growth and nisin production by Lactococcus lactis subsp. lactis
[67] G. Berg, A. Fritze, N. Roskot, and K. Smalla, “Evaluation of
CECT 539 in both batch and re-alkalized fed-batch cultures,”
potential biocontrol rhizobacteria from different host plants of
Journal of Food Engineering, vol. 82, no. 2, pp. 103–113, 2007.
Verticillium dahliae Kleb,” Journal of Applied Microbiology, vol.
[53] L. Chen, G. Wang, S. Hong, A. Liu, C. Li, and Y. Liu, “UV-B- 91, no. 3, pp. 963–971, 2001.
induced oxidative damage and protective role of exopolysaccha-
[68] V. Mandal, S. K. Sen, and N. C. Mandal, “Detection, isola-
rides in desert cyanobacterium Microcoleus vaginatus,” Journal
tion and partial characterization of antifungal compound(s)
of Integrative Plant Biology, vol. 51, no. 2, pp. 194–200, 2009.
produced by Pediococcus acidilactici LAB 5,” Natural Product
[54] A. S. Ferreira, I. N. Silva, V. H. Oliveira, R. Cunha, and L. M. Communications, vol. 2, pp. 671–674, 2007.
Moreira, “Insights into the role of extracellular polysaccharides
[69] J. Magnusson, K. Ström, S. Roos, J. Sjögren, and J. Schnürer,
in Burkholderia adaptation to different environments,” Frontiers
“Broad and complex antifungal activity among environmental
in Cellular and Infection Microbiology, vol. 1, p. 16, 2011.
isolates of lactic acid bacteria,” FEMS Microbiology Letters, vol.
[55] A. W. Qurashi and A. N. Sabri, “Osmoadaptation and plant 219, no. 1, pp. 129–135, 2003.
growth promotion by salt tolerant bacteria under salt stress,”
African Journal of Microbiology Research, vol. 5, no. 21, pp. 3546–
3554, 2011.
http://dx.doi.org/10.1155/2013/968518
Review Article
An Insight into the ‘‘-Omics’’ Based Engineering of
Streptomycetes for Secondary Metabolite Overproduction
Amit Kumar Chaudhary, Dipesh Dhakal, and Jae Kyung Sohng

Department of Pharmaceutical Engineering, Institute of Biomolecule Reconstruction, SunMoon University,
100 Kalsan-ri, Tangjeongmyeon, Asan-si, Chungnam 336-708, Republic of Korea
Correspondence should be addressed to Jae Kyung Sohng; sohng@sunmoon.ac.kr
Received 28 January 2013; Revised 26 July 2013; Accepted 28 July 2013
Copyright © 2013 Amit Kumar Chaudhary et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Microorganisms produce a range of chemical substances representing a vast diversity of fascinating molecular architectures not
available in any other system. Among them, Streptomyces are frequently used to produce useful enzymes and a wide variety of
secondary metabolites with potential biological activities. Streptomyces are preferred over other microorganisms for producing
more than half of the clinically useful naturally originating pharmaceuticals. However, these compounds are usually produced
in very low amounts (or not at all) under typical laboratory conditions. Despite the superiority of Streptomyces, they still lack
well documented genetic information and a large number of in-depth molecular biological tools for strain improvement. Previous
attempts to produce high yielding strains required selection of the genetic material through classical mutagenesis for commercial
production of secondary metabolites, optimizing culture conditions, and random selection. However, a profound effect on the
strategy for strain development has occurred with the recent advancement of whole-genome sequencing, systems biology, and
genetic engineering. In this review, we demonstrate a few of the major issues related to the potential of “-omics” technology
(genomics, transcriptomics, proteomics, and metabolomics) for improving streptomycetes as an intelligent chemical factory for
enhancing the production of useful bioactive compounds.
1. Introduction [2, 3] (Table 1). Besides, many of these secondary metabolites

function as signaling molecules to control the metabolism
Natural products are chemical compounds with pharmaco- of their own producer [4]. This vast reservoir of diverse
logical characteristics produced by living organisms that can products makes streptomycetes the most important indus-
be utilized during pharmaceutical drug discovery, agricul- trial microbial genus. Based on the Waksman and Henrici
ture, and in the food industry. They are called “secondary classification of 1943, these organisms are classified in
metabolites,” as they can be synthesized by microorganisms the family Streptomycetaceae[5]. They are Gram-positive
and plants but are not essential for their own metabolic mycelial soil bacteria containing about 70% G-C DNA
processes [1]. Microbial fermentation is widely applied to content and undergo a complex process of morphologi-
industrially produce these valuable compounds. Usually, cal development that usually involves secondary metabolite
these compounds are produced in very low amounts (or biosynthesis under depleted nutrient conditions [6].
not at all) by natural strains under typical laboratory con- Genome mining of several Streptomyces [7–10] has
ditions to meet commercial requirements, demonstrating revealed numerous cryptic novel secondary metabolite
the need for heterologous expression of these biosynthetic biosynthetic gene clusters, which encode the potential to
gene clusters. Of the thousands of secondary metabolites synthesize a large diversity of compounds that have never
documented, more than half are produced by Streptomyces been observed before, including polyketides, aminoglyco-
(antibiotics, antitumor agents, immunosuppressants, anti- sides, bacteriocins, terpenoids, shikimate-derived metabo-
helminthics, antifungals, herbicides, and insecticides) and lites, nonribosomal peptides, anthracyclines, macrolides,
have offered decades of interest to industry and academia beta-lactams, and other natural products [11]. The core unit
Table 1: List of bioactive compounds produced by streptomycetes.
Drugs Strains Applications

Aclacinomycin A S. galilaeus Antitumor
Actinorhodin S. coelicolor Antibacterial
Alnumycin Streptomyces sp. CM020 Antitumor; gyrase inhibitor; topoisomerase inhibitor
Alpha-lipomycin S. aureofaciens Antibacterial
Amphotericin B S. nodosus Antifungal
Apramycin S. tenebrarius Antibacterial
Aranciamycin S. echinatus Antibacterial; Collagenase inhibitor
Ascomycin S. hygroscopicus var. ascomyceticus Immunosuppressive; antifungal
Asukamycin S. nodosus subsp. asukaensis Antitumor
Aureothin S. thioluteus Antitumor; antifungal; insecticidal
Avermectin S. avermitilis Anthelmintic
Benastatin Streptomyces sp. A2991200 Antibacterial; apoptosis inducer; glutathione-S-transferase (GST) inhibitor
Bleomycin S.verticillus Antitumor
Borrelidin S. parvulus Tü 4055 Angiogenesis inhibitor; antibacterial; antiviral; Antiproliferative
Chalcomycin S. bikiniensis Antibacterial
Chartreusin S. chartreusis Antibacterial; antitumor
Chlorothricin S. antibioticus Antibacterial
Chloramphenicol S. venezuelae Antibacterial
Chromomycin S. griseus Antibacterial; antitumor; antiviral
Coumermycin S. rishiriensis Antibacterial
Concanamycin A S. neyagawaensis Antifungal; antiprotozoal; antitumor; antiviral
Clavulanic acid S. clavuligerus Antibacterial
Cosmomycin S. olindensis Antitumor
Daptomycin S. roseosporus Antibacterial
Daunorubicin S. peucetius subsp. caesius Antitumor
Doxorubicin S. peucetius subsp. caesius Antitumor
Dunaimycins S. diastatochromogenes Immunosuppressive
Elloramycin S. olivaceus Antibacterial; antitumor
Enterocin S. maritimus Antibacterial
Formycin S. lavendulae Antitumor
Fredericamycin S. griseus; S. chattanoogensis Antibacterial; antifungal; antitumor
Frenolicin S. roseofulvus Antibacterial; antiprotozoal
Gilvocarcin S. griseoflavus; S. anandii Antibacterial; antitumor; antiviral
Granaticin S. violaceoruber Antibacterial
Griseorhodin A Streptomyces sp. JP95 Antibacterial; human telomerase inhibitor
Geldanamycin S. hygroscopicus Antitumor
Griseusin S. griseus Antibacterial
Halstoctacosanolide S. halstedii Antibacterial
Hedamycin S. griseoruber Antitumor
Herbimycin A S. hygroscopicus Antitumor
Herboxidiene S. chromofuscus Herbicidal
Hygromycin B S. hygroscopicus Antibacterial; antifungal
Indanomycin S. antibioticus Antibacterial; antiprotozoal; insecticidal; Ionophore
Istamycin S. tenjimariensis Antibacterial
Jadomycin B S. venezuelae Antibacterial
Kanamycin S. kanamyceticus Antibacterial
Kirromycin S. collinus Antibacterial
Landomycin S. cyanogenus Antitumor
Lasalocid S. lasaliensis Antibacterial
Lymphostin Streptomyces sp. KY11783 Immunosuppressive
Medermycin Streptomyces sp. AM7161 Antibacterial; antitumor
Table 1: Continued.
Drugs Strains Applications
Meilingmycin S. nanchangensis Anthelmintic
Meridamycin S. violaceusniger Neuroprotective
Mitomycin C S. caespitosus Antibacterial; antineoplastic; immunosuppressive
Mithramycin S. argillaceus Antibacterial; antitumor
Monensin S. cinnamonensis Antibacterial; ionophore; antiprotozoal
Nanchangmycin S. nanchangensis Antibacterial; insecticidal; ionophore
Naphthocyclinone S. arenae Antibacterial
Neomycin S. faradae Antibacterial
Niddamycin S. caelestis Antibacterial
Neocarzinostatin S. carzinostaticus Antibacteria; antineoplastic
NFAT-133 Streptomyces sp. PM324667 Antidiabetic
Nigericin S. violaceusniger Antibacterial; ionophore
Nogalamycin S. nogalater Antibacterial; antitumor
Nikkomycin X S. ansochromogenes Antibacterial
Nystatin S. noursei Antifungal
Novobiocin S. niveus Antibacterial
Oligomycin S. avermitilis Antifungal
Oviedomycin S. antibioticus Antitumor
Oxazolomycin S. albus Antibacterial; antitumor; antivirus; ionophore
Oxytetracycline S. rimosus Antibacterial
Pactamycin S. pactum Antibacterial; antiprotozoal; antitumor; antiviral
Paromycin S. rimosus Antiamoebal, antibiotics
Phoslactomycin Streptomyces sp. HK803 Antitumor
Pikromycin S. venezuelae Antibacterial
Pimaricin S. natalensis Antifungal
Pladienolide S. platensis Antitumor
Pristinamycin S. pristinaespiralis Antibacterial
Polyketomycin S. diastatochromogenes Antibacterial; antitumor
Pikromycin S. venezuelae Antibacterial
Rapamycin S. hygroscopicus Immunosuppressive; antitumor; neuroprotective; antiaging
Ribostamycin S. ribosidificus Antibacterial
Resistomycin S. resistomycificus Antibacterial; antiviral
Rimocidin S. diastaticus Antifungal
Rubradirin S. achromogenes var. rubradiris Antibacterial
Saframycin A S. lavendulae Antitumor
Steffimycin S. steffisburgensis Antitumor
Streptolydigin S. lydicus Antibacterial
Sparsomycin S. sparsogenes Antitumor
Spiramycin S. ambofaciens. Antibacterial
Spectinomycin S. spectabilis Antibacterial
Streptomycin S. greseus Antibacterial
Tautomycetin Streptomyces sp. CK4412 Antifungal; antitumor; immunosuppressive
Tautomycin S. spiroverticillatus Antibacterial; antifungal; antitumor
Tetronomycin Streptomyces sp. NRRL11266 Antibacterial; ionophore
Thiostrepton S. azureus Antibacterial
Tetracycline S. aureofaciens Antibacterial
Urdamycin S. fradiae Antibacterial; antitumor
Vicenistatin S. halstedii Antitumor
Virginiamycin S. virginiae Antibacterial
Overproduction
Glucose-6-phosphate
NADP+ Glucose-6-phosphate
NADPH dehydrogenase
6-Phosphogluconolactone
6-Phosphogluconate
NADP+ 6-Phosphogluconate
NADPH dehydrogenase
Ribulose-5-phosphate
Nonoxidative reactions
Engineered metabolic networks (2)
Overproduction
Overproduction
Engineered sigma factors

Microarrays analysis
and ribosomes
p
Secondary metabolites gene cluster

mRNA
Engineered proteomics (4)
Engineered transcriptomics (5)
eered genomics
Engineered geno (3)
Over- or heterologous
O h l expression
(3A) Recombinations
R bi i
(3B)
Overproduction
Figure 1: The approaches used for overproduction of secondary metabolite in Streptomycetes.
responsible for secondary metabolite production is specif- of most Streptomyces, but an extensive review on primary
ically termed a biosynthetic gene cluster, which encom- metabolism reported that the Embden-Meyerhof-Parnas,
passes biosynthetic enzymes, resistance determinants, and the pentose phosphate (PPP), and tricarboxylic acid cycle
regulatory proteins. The use of these cryptic microbial pathways are present in a number of Streptomyces species.
secondary metabolic processes has attracted the attention [18]. Based on the huge progress in “-omics” technologies,
of synthetic microbiologists who exploit recent advances new approaches are aimed at ensuring optimal engineering
in DNA sequencing and synthesis procedures to achieve of the cell factory to achieve optimized metabolite production
unprecedented control over metabolic pathways. In partic- [19] (Figure 1). In this review, we illustrate the pivotal roles of
ular, rare Streptomyces are promising sources for new drugs; genomics, transcriptomics, proteomics, and metabolomics as
therefore, genetic manipulations of these microorganisms are research tools in systems biology of secondary metabolism
crucial for drug discovery and development. for enhancing the production of secondary metabolites in
Recent advances in “-omics” technologies introduced streptomycetes. We discuss the potential of exploiting “-
during the last two decades have allowed the establishment of omics” tools to enhance the production of naturally origi-
various research areas pertaining to Streptomyces. In general, nating pharmaceuticals by circumventing major bottlenecks
“-omics” considers versatile genes and their products such to overproduce compounds of interest (Tables 2, 3, 4 and
as transcripts, proteins, or metabolites. Genomics [12] deals 5). Furthermore, we conclude the paper by highlighting
with genes, their variation, and function, whereas transcrip- the future perspectives of recent technological advances
tomics considers information at the mRNA (transcript) level yet to be applied to Streptomyces species for further stain
at a particular time depending on environmental signals and improvement.
biophysiological parameters. Proteomics considers expres-
sion, function, and regulation of an entire set of proteins [13].
Moreover, as the proteins within cells are the functional units,
their expression is strongly influenced by environmental 2. Engineered Metabolic Networks
signals and physiological conditions, and, thus, proteomics is
a complementary technology to genomic and transcriptomic Among the “-omics” technologies [12–15], one important tool
research [14]. Metabolomics encompasses detailed metabolic in the system biology toolbox is metabolomics, which cata-
analysis [15]. Since the genome of the first cellular organism logues all small metabolites in a biological sample [20, 21]. It is
Haemophilus influenza was sequenced [16], the availability of expected to play a significant role in bridging the phenotype-
metabolic network models has helped develop several com- genotype gap, as it amplifies changes in the proteome and
putational approaches for flux balance analyses [17]. How- provides a better representation of an organism’s phenotype.
ever, relatively little is known about the metabolic pathways Moreover, knowledge of the complete set of metabolites
provides the intracellular fluxes required for a compre- with optimizing glucose concentrations enhances the FK506
hensive characterization of metabolic networks and their titer by approximately 150% in comparison to that of the
operation. Therefore, intracellular fluxes can help in high original S. tsukubaensis strain [31]. Similarly, overexpressing
level production of pharmaceuticals requiring precursors potential biosynthetic sugar genes such as desIII (glucose-
and cofactors from primary metabolism; hence, engineered 1-phosphate thymidylyl-transferase) and desIV (TDP-D-
primary metabolism is a prerequisite for biosynthesis of any glucose 4,6-dehydratase) from S. venezuelae and glycosyl-
secondary metabolite. transferase DnrS-DnrQ transferring TDP-L-daunosamine to
One of the most important primary metabolic pathways 𝜀-rhodomycinone from S. peucetius leads to significantly
is the oxidative PPP, which provides essential cofactors enhanced doxorubicin (DXR) production [32]. These studies
and intermediates for cell growth. The physiological effect confirmed that engineering of rate limiting steps can be a
of glucose-6-phosphate dehydrogenase (G6PDH) encoding robust strategy for efficient flow of intermediates to enhance
isozymes in the PPP has been investigated in a variety production of secondary metabolites.
of bacteria. Greater oxytetracycline (OTC) production is
achieved by increasing the pool of malonyl-CoA, an OTC
precursor, by deleting the zwf1 and zwf2 genes encoding
for two G6PDH isozymes of S. rimosus M4018 [22, 23]. 3. Engineered Genomics
However, the case was the opposite with 16.5% increase
in daptomycin concentration when zwf2 was overexpressed 3.1. Genome Guided Overexpression of Gene Clusters in Native
in S. roseosporus due to greater availability of daptomycin and Heterologous Hosts. Amplifiable units of DNA (AUDs)
precursors by conversion to G6P in the PPP pathway. exist in Streptomyces with multiple copies per chromo-
Deletion of zwf1 or zwf2 also improves actinorhodin and some in tandem. They often lie in unstable chromosomal
undecylprodigiosin production in S. lividans [24]. Significant regions, such as ends of the linear chromosome, and a
changes in central carbon metabolism of S. coelicolor after large deletion frequently accompanies amplification [33].
deleting G6PDH and phosphoglucomutase (pgm) indicate Consequently, inducible amplification of specific regions
that the carbon storage metabolism plays a significant role of microbial genomes helps to improve a wide variety of
in precursor supply for actinorhodin production and over- complex multigene processes in strains that biosynthesize
production [25]. These studies presumed that the lower flux various important secondary metabolites [34]. Yanai et al.
of carbon through the PPP in each of the mutants allows for [35] reported that amplifying the entire kanamycin (Km)
more efficient glucose utilization via glycolysis, resulting in biosynthetic gene cluster in S. kanamyceticus 12-6 results in
higher levels of antibiotic production [24, 25]. Nevertheless, a disparity in antibiotic production. A comparison of Km
as G6PDH is the first enzyme in the PPP pathway and the key production from 12-6 (containing an average of three copies
enzyme for generating NADPH, the increased biomass and of the Km gene cluster) and 12-6-4 (containing one copy
NADPH regeneration would be another factor for favorable of the Km gene cluster) indicated that the titer of strain
cell growth and precursor synthesis in both cases [26, 27]. 12-6 strain (376 𝜇g/mL) was about twice that of strain 12-
Similarly, deleting the key glycolytic enzyme leads to a 2–6- 6-4 (196 𝜇g/mL). Furthermore, integrating cosmid pMJ20-
fold enhancement of actinorhodin and undecylprodigiosin 10-1 (containing the entire Km gene cluster) into strain 12-
production in S. coelicolor A3 (2) on minimal and rich medim 6-4 (results in two copies of the Km gene cluster) results
(2). The pfkA2 (SCO5426, phosphofructokinase) deleted in a lower Km titer, indicating that integration of the
strain shows increased carbon flux through the PPP due to cosmid suppresses Km production. This finding supports
the accumulation of G6P and fructose-6-phosphate (F6P), the notion that Km production level depends on the gene
which help to increase NADPH supply [28] and enhance cluster copy number and that introducing an extra copy of
antibiotic production as suggested by Gunnarsson et al. [26]. the biosynthetic gene cluster into a parent strain may be
Supplementing the culture broth with a metabolic precursor an effective approach to improve antibiotic production [35].
to enhance poly-𝜀-lysine production by S. noursei NRRL 5126 Furthermore, an extra copy of the nikkomycin (a competitive
is another robust example of an engineered metabolic net- inhibitor of chitin synthase with fungicidal, insecticidal, and
work. Poly-𝜀-lysine is a strong inhibitor of a wide spectrum acaricidal activities) biosynthetic gene cluster (35 kb) into
of microorganisms and is used as a food preservative. As an S. ansochromogenes 7100 leads to enhanced production of
economically feasible process, supplementation with 5 mM nikkomycins (880 mg/L, 4-fold nikkomycin X and 210 mg/L,
citric acid and 2 mM L-aspartate increases S. noursei poly-𝜀- 1.8-fold nikkomycin Z) in the resulting exconjugants com-
lysine production significantly in a time-dependent manner pared with that of the parent strain (220 mg/L nikkomycin X
from 97.08 to 497.67 mg/L in 108 h [29]. FK506 (tacrolimus), a and 120 mg/L nikkomycin Z) [36]. Similarly, engineering S.
potent immunosuppressive agent, interacts with the FKBP12 coelicolor utilizing the oriT-like recombination sites RsA and
receptor further targeting calcineurin by inhibiting Ser/Thr RsB and ZouA, a site-specific relaxase flanking actinorhodin
phosphatase activity leading to the arrest of T cell prolifera- gene cluster, results in 4–12 tandem copies of the complete
tion [30]. Engineering the pathway-specific building blocks gene cluster, averaging nine repeats per genome leading to
for biosynthesis occasionally improves production of some a 20-fold increase in actinorhodin production [34]. Thus,
polyketide or nonribosomal peptide natural products. For amplification of the entire gene cluster has direct positive
example, enhancing biosyntheses of methoxymalonyl-ACP effects on enzymatic yield and precursor flow leading to
and allylmalonyl-CoA, the extender units of FK506, together enhanced secondary metabolite production.
Table 2: Methodologies used to overproduce drugs using engineered metabolic networks approach.
Strains Drugs Approach Methodologies

Deletion of zwf1 and zwf2 genes improve the
S. rimosus M4018 Oxytetracycline Engineered metabolic networks
production of oxytetracycline
Over-expression of zwf2 gene improve the
S. roseosporus Daptomycin Engineered metabolic networks
production of daptomycin
Actinorhodin and Deletion of zwf1 or zwf2 improved actinorhodin
S. lividans Engineered metabolic networks
undecylprodigiosin and undecylprodigiosin production
Deletion of pfkA2 (phosphofructokinase) gene
Actinorhodin and
S. coelicolor A3 (2) Engineered metabolic networks improve the production of actinorhodin and
undecylprodigiosin
undecylprodigiosin
Supplementation of citric acid and L-Asp
S. noursei NRRL 5126 𝜀-Poly-l-Lysine Engineered metabolic networks
increases poly-𝜀-lysine production
Enhancing the biosyntheses of
methoxymalonyl-ACP and allylmalonyl-CoA
S. tsukubaensis FK506 (tacrolimus) Engineered metabolic networks
together with optimized glucose concentrations
enhances the FK506 production
Over-expression of potential biosynthetic sugar
S. peucetius ATCC
Doxorubicin Engineered metabolic networks genes and glycosyltransferase enhanced
27952
doxorubicin production
Table 3: Methodologies used to overproduce drugs using engineered genomics approach.

Genome guided overexpression of gene Overexpression of extra copy of the gene
S. kanamyceticus Kanamycin
clusters in native and heterologous hosts cluster enhanced kanamycin production
Overexpression of extra copy of the gene
Genome guided overexpression of gene
S. ansochromogenes Nikkomycin cluster enhanced nikkomycins
clusters in native and heterologous hosts
production
Genome guided overexpression of gene Tandem copies of the gene cluster
S. coelicolor Actinorhodin
clusters in native and heterologous hosts increased actinorhodin production
Streptomycin, cephamycin Genome guided overexpression of gene Heterologous expression in
S. avermitilis
C, and pladienolide clusters in native and heterologous hosts genome-minimized strain
Tylosin, kanamycin,
spectinomycin, Genome guided overexpression of gene Heterologous expression in pikromycin
S. venezuelae YJ003
spectinamine, gentamicin, clusters in native and heterologous hosts gene cluster deleted strain
and epothilones
S. lividans TK-23, Daptomycin and Genome guided overexpression of gene
Heterologous expression
TK-24, and TK-63 paromamine clusters in native and heterologous hosts
S. lividans Capreomycin Heterologous expression
S. albus J1074 Thiocoraline Heterologous expression
Genome shuffling guided enhancement
S. fradiae Tylosin Two rounds of genome shuffling
of secondary metabolites
S. gilvosporeus SG1 Natamycin Four rounds of genome shuffling
Four rounds of genome shuffling to
Genome shuffling guided enhancement increase the resistivity against
S. pristinaespiralis Pristinamycin
of secondary metabolites pristinamycin enhanced pristinamycin
production
Generating resistance mechanism for
S. sp. U121 (2S, 3R)-HCA transepoxyaconitic acid by three rounds
of shuffling
S. padanus,
S. griseofuscus, Through glucose, sulfa guanidine, and
S. graminearus, 𝜀-Poly-l-lysine succinic acid tolerance and genome
S. hygroscopicus, shuffling
and S. albulus
Table 4: Methodologies used to overproduce drugs using engineered proteomics approach.

Proteomics facilitated reverse engineering
Overexpression of the metK gene encoding
S. venezuelae Pikromycin to enhance secondary metabolite
SAM synthetase
production
S. griseus IFO13189 Proteomics facilitated reverse engineering Exogenous feeding of S-Adenosyl
Spectinomycin and
and S. griseoflavus to enhance secondary metabolite methionine (SAM) results in enhanced
bicozamycin
FERM1805 production production
S. sp. FR-008, S.
avermitilis, S. Candicin D,
Proteomics facilitated reverse engineering Coexpression of metK or exogenous
coelicolor A3 (2), S. avermectin,
to enhance secondary metabolite feeding of SAM enhanced antibiotic
lividans TK23, and S. actinorhodin, and
production production
antibioticus oleandomycin
ATCC11891
Proteomics facilitated reverse engineering Overexpression of mutant library of sigma
S. avermitilis Avermectins to enhance secondary metabolite factor 𝜎ℎ𝑟𝑑𝐵 enhanced antibiotic
production production
Proteomics facilitated reverse engineering
S. peucetius ATCC Overexpression of efflux protein DrrA
Doxorubicin to enhance secondary metabolite
27952 enhanced antibiotic production
production
Resistance against streptomycin causes
Ribosome engineering to enhance production of pigmented antibiotic
S. lividans TK24 Actinorhodin
secondary metabolite production actinorhodin not produced in normal
laboratory conditions
Ribosome engineering to enhance Resistance against streptomycin causes
S. chattanoogensis Fredericamycin
secondary metabolite production enhanced production
Actinorhodin,
By introducing rifampicin mutations into
undecylprodigiosin, Ribosome engineering to enhance
S. lividans 66 the rpoB (encoding the RNA polymerase
and calcium secondary metabolite production
subunit) gene
dependent antibiotics
By introducing double and triple
Ribosome engineering to enhance mutations using gentamicin rifampicin
S. coelicolor A3 (2) Actinorhodin
secondary metabolite production and streptomycin increases actinorhodin
production
Enhanced expression of ribosome
Ribosome engineering to enhance
S. coelicolor A3 (2) Actinorhodin recycling factor by mutation increases
secondary metabolite production
production
Ribosome engineering to enhance Overexpression of ribosome recycling
S. avermitilis Avermectin
secondary metabolite production factor increases production
Introducing rpsL and rpoB mutations
Chloramphenicol and Ribosome engineering to enhance
S. coelicolor A3 (2) enhanced chloramphenicol and
congocidine secondary metabolite production
congocidine production
Ribosome engineering to enhance Introducing mutations in rsmG gene
Streptomycetes Antibiotics
secondary metabolite production encoding for 16S rRNA methyltransferase
Table 5: Methodologies used to overproduce drugs using engineered transcriptomics approach.

TetR family transcriptional regulator as a global
Streptomycetes Antibiotics Engineered Transcriptomics
upregulator for enhanced antibiotic production
Disruption of wblA from S. peucetius OIM
S. peucetius OIM Doxorubicin and Daunorubicin Engineered Transcriptomics resulted in increase in the production of both
doxorubicin and daunorubicin
Extensive focus on the genetics of Streptomyces has [7– advantage of multiparental crossing facilitated by recombi-
10] revealed numerous silent gene clusters probably biosyn- nation of an entire genome associated with conventional
thesizing unknown complex natural products [37]. As an breeding applications and, thus, acts as a combined method
alternative, heterologous expression of these gene clusters to improve phenotype [56]. Genome shuffling is a novel
in a suitable strain is a technique of pivotal importance to and promising technique discovered to enhance secondary
exploit drug discovery programs [38]. Normally, heterolo- metabolite production [57, 58]. Desired phenotypes can be
gous expression is preferably carried out in fully sequenced obtained using this technique after several rounds of genome
strains, such as S. coelicolor A3 (2) [8], S. avermitilis [9], or S. recombination of key genes responsible for production [54].
venezuelae (unpublished data, our group) due to unrestricted Genome shuffling basically incorporates (1) construction of
metabolic engineering for proper flux of precursors and reg- diverse parental strains in several rounds of mutagenesis
ulatory networks. Furthermore, active secondary metabolic using chemical agents such as ethyl methanesulfonate and
gene clusters of such strains may be silent to prevent diversion nitrosoguanidine as well as physical agents such as ultraviolet
of precursors into competing secondary metabolic pathways, and 𝛾 irradiations; (2) recursive protoplast fusion of mutants
thus, facilitating enhanced production of the desired com- with a multitude of phenotypes, and (3) intensive screening
pounds [38]. A genetically engineered “clean” host strain, and selection based on product yields or other desired
S. coelicolor CH999, has been constructed in which the characteristics [59, 60].
entire actinorhodin gene cluster is surgically deleted [39]. Genome shuffling has been widely used to enhance
The most intensively studied strains are S. coelicolor M145, secondary metabolite production in streptomycetes. Its first
M512, M1146, and M1154 [40–42], in which extensive deletion use was reported in S. fradiae, where significant phenotypic
of different gene clusters has been performed to prevent improvement was observed in just two rounds of genome
background noise. A genome-minimized strain of S. aver- shuffling [58]. Sixfold higher tylosin production was achieved
mitilis represents a suitable host for efficient production from a hybrid strain, which was equivalent to achieving 20
of secondary metabolites, as demonstrated by heterologous rounds of classical strain improvement by random mutation
expression of the antibiotics streptomycin, cephamycin C, that would probably require 20 years [61]. Nevertheless, pro-
and pladienolide [43]. S. venezuelae YJ003, bearing a dele- duction of about 3.5 g/L natamycin has been reported from
tion of the pikromycin gene cluster, is also a widely used S. gilvosporeus SG1, which was 153% of that of the parental
strain in our lab with the advantages of fast growth, good strain and 1.17 times greater than that of the starting strain
transformation efficiency, and rapid production of tylosin, [62]. Luo et al. [63] also reported similar results, in which
kanamycin, spectinomycin, spectinamine, gentamicin, and 4.7 g/L natamycin was produced in a shaking flask after a 96 h
epothilones (unpublished data) [44–47]. Similarly, other culture. This was 97.1% and 379% of the amount produced by
strains of Streptomyces, such as S. lividans TK23, TK24, and the highest producing parental strain and the initial strain,
TK63, are also used as heterologous hosts to produce dapto- respectively, after four rounds of genome shuffling.
mycin and paromamine [48, 49]. A plasmid cured strain of S. S. pristinaespiralis produces pristinamycin (Ptr), an active
clavuligerus has also been suggested as a heterologous host for drug against various multidrug-resistant pathogenic strains
secondary metabolites [50]. The heterologous expression of [64–67]. However, pristinamycin itself inhibits biosynthesis
cryptic pathways in heterologous hosts suitable for expressing and mycelial growth [68]. Although antibiotic-producing
otherwise silent secondary metabolite gene clusters opens streptomycetes have developed mechanisms to protect them-
new avenues for the production of secondary metabolites selves against their own antibiotics, many antibiotics are toxic
that are not produced under normal laboratory conditions at elevated concentrations. This toxicity could be particu-
in native hosts. For example, capreomycin was produced at larly problematic in the quest for antibiotic overproducing
50 mg/L from S. lividans without any modifications, whereas strains. However, it is not surprising that increased antibiotic
it was produced less by the native strain of Saccharothrix resistance has often been used to select for mutants with
mutabilis and was not amenable to genetic studies [51]. increased antibiotic production levels. As a consequence,
In contrast, the cryptic gene cluster encoding thiocoraline genome shuffling was used for S. pristinaespiralis to increase
biosynthesis from a marine Micromonospora sp. ML1 pro- the resistivity against its own product from 20 to 100 𝜇g/mL,
duces a significant amount of thiocoraline in S. albus J1074 and production was increased from 0.47 g/L to 0.89 g/L after
[52]. Thus, a rationale approach for addressing the expression four rounds of shuffling [54]. A quantitative real-time poly-
of cryptic pathways unexpressed or repressed in native hosts merase chain reaction (qRT-PCR) analysis by Jin et al. [69]
includes identifying novel pathways by bioinformatics and revealed the involvement of snbA and snaB (encoding the
cloning and expressing them in well-characterized hosts with two subunits of pristinamycin IIA synthase catalyzing the last
known secondary metabolomics [53]. step in the biosynthesis of the pristinamycin IIA component)
with higher expression in the recombinant than that in the
ancestor at 24–60 h of fermentation, indicating that their
expression changes might be a key factor during antibiotic
3.2. Genome Shuffling Guided Enhancement of Secondary biosynthesis. Similarly, the ptr resistance gene maintains a
Metabolites. Genome shuffling is an amalgamation of classi- high expression level during the entire fermentation process
cal breeding with modern high-throughput screening based of the recombinant strain, whereas it is expressed at a low
on genome recombination even in the absence of detailed level at 24–48 h of fermentation in the ancestor. These results
genetic knowledge [54, 55]. Genome shuffling combines the indicate that the discrepancy in expression changes might
be a key factor during antibiotic biosynthesis [69]. Similarly, synthetic and nutrient media, respectively, and a 2-fold
amplified fragment length polymorphism analysis of a high- increase in bicozamycin production from S. griseoflavus
pristinamycin-producing strain revealed that a homolog of FERM1805 [84]. Coexpression and standalone expression
the afsR regulatory gene, a global regulator of secondary of metK or exogenous feeding of SAM results in enhanced
metabolism in S. coelicolor A3 (2) [40], and a homolog of the antibiotic production in various streptomycetes, such as can-
transposase gene, belonging to the validamycin biosynthetic dicin D from Streptomyces sp. FR008 [85], avermectin from
gene cluster from S. hygroscopicus [70], are responsible S. avermitilis [85], actinorhodin from S. coelicolor A3 (2) [82],
for yield improvement in S. pristinaespiralis [69]. Similarly, actinorhodin from S. lividans TK23 [86], and oleandomycin
genome shuffling of Streptomyces sp. U121, the producer of from S. antibioticus ATCC11891 [87]. Furthermore, Zhuo et
(2S, 3R)-HCA [71, 72] with potent pancreatic 𝛼-amylase and al. [80] also implemented the same technique to increase
intestinal 𝛼-glucosidase inhibitory activities [73, 74], has been the production of avermectin in S. avermitilis in a round
targeted to achieve rapid improvement in HCA production ofmicroarray studies confirming the overexpression of the
using the resistance mechanism for transepoxyaconitic acid pathway specific regulatory gene aveR in a high-producing
(antibiotic HCA analog), resulting in fivefold higher HCA strain. Based on the assumption that the promoter region of
production than that in the wildtype after three rounds of the aveR gene is recognized by sigma factor 𝜎ℎ𝑟𝑑𝐵 , a mutant
shuffling [75]. This technique has also been suggested to library of the hrdB gene was generated and overexpressed
increase 𝜀-poly-l-lysine productivity in wildtype strains of resulting in >50% improvement in avermectin B1 [80]. This
S. padanus, S. griseofuscus, S. graminearus, S. hygroscopicus, example suggests that manipulating important genes revealed
and S. albulus through glucose tolerance, sulfaguanidine by reverse engineering can effectively improve the yield of
tolerance, and using succinic acid as the sole carbon source, target metabolites.
respectively [76, 77]. Furthermore, S. rimosus was also sub- Advances in proteomics have made it possible to identify
jected to genome shuffling for higher oxytetracycline yielding proteins that show significant changes in expression levels
strains [60]. In combination with genome shuffling, ribosome on 2D gel electrophoresis under certain conditions [88].
engineering has been used in S. viridochromogenes, an avil- These approaches can also be used for engineering secondary
amycin producer and an effective antimicrobial agent against metabolite target genes to enhance antibiotic production. For
multidrug-resistant Gram-positive bacteria [59]. A mutant example, an attempt was made to increase the coenzyme A
strain obtained after 𝛾-irradiation was used for genome shuf- (CoA) pool using the pantothenate kinase (panK) gene to
fling with ribosome engineering mediated by streptomycin enhance production of DXR in S. peucetius ATCC 27952;
resistance. After five rounds of genome shuffling, 300 𝜇g/mL however, the opposite occurred due to increased aglycone
streptomycin resisting strain produced 1.4 g/L avilamycin, polyketide 𝜀-rhodomycinone (RHO) [89]. To understand
which was 4.8-fold and 36.8-fold greater than the shuffling these results in detail, 2D gel electrophoresis was used to show
starter and ancestor, respectively [59]. that the efflux protein DrrA was overexpressed, resulting
in 9.4-fold higher DXR production than that of a panK
integrated strain, which showed that the proteomic approach
4. Engineered Proteomics is quite useful for host development and understanding the
physiology of antibiotic production.
4.1. Proteomics Facilitates Reverse Engineering to Enhance
Secondary Metabolite Production. Since the 1970s, recom-
binant DNA technology has revolutionized the ability to 4.2. Ribosome Engineering to Enhance Secondary Metabolite
engineer microorganisms by modifying specific genes and Production. Ribosomes are the fundamental organelles con-
pathways for optimized production of commercially signif- trolling the protein-RNA complex expression machinery that
icant metabolites [78]. In contrast, the concept of reverse synthesizes proteins using genetic instructions encoded in
engineering has evolved as a powerful tool in which two the mRNA template. Hence, engineering ribosomes to fine
processes are utilized to genetically characterize existing tune protein expression and secondary metabolite produc-
overproducing strains, and a second generation of information is a highly utilized approach. One conventional method
tion is used for more efficient engineering of new strains that to modulate ribosomes is to introduce mutations conferring
synthesize high yields of natural products [79]. This approach resistance to drugs that attack ribosomes, which frequently
elucidates the interrelationships between physiological traits have mutations within ribosomal components (ribosomal
and more efficiently directs the engineering of target com- protein, rRNA, or translation factors) [90–92]. For example,
pound producing strains to synthesize high yields of these generating a point mutation in the ribosomal protein rpsL
natural products [80]. For example, reverse engineering of (str-6) of the S. lividans TK24 strain against Streptomycin (Str)
the S. coelicolor overproducer using two-dimensional (2D) causes production of the pigmented antibiotic actinorhodin,
gel electrophoresis recently identified S-adenosyl methionine which is not produced under normal laboratory conditions,
(SAM) synthetase as an antibiotic overproducing enzyme which could be due to significant changes in translational
[81, 82]. Based on this observation, expression of the metK machinery [82]. Nearly half of the str mutants in S. chat-
gene encoding SAM synthetase has been utilized to enhance tanoogensis exhibit a significant increase in fredericamycin
pikromycin by 1.6-fold in S. venezuelae [83]. Similarly, exoge- production (>five-fold), with one strain showing 26-fold
nous feeding of SAM results in enhanced spectinomycin higher antibiotic production than that of the wild type
production by 3.6, and 3-fold in S. griseus IFO13189 using [91]. The frequency of such antibiotic overproducing strains
among the str mutants is 3–46%, as shown with several strains for functional studies of the basic unit of life; however,
in the genera Streptomyces, Bacillus, and Pseudomonas [91]. transcriptomic analysis was developed using microarray chip
Moreover, biosynthesis of actinorhodin, undecylprodigiosin, technology and mutational analysis and focuses on identi-
and calcium-dependent antibiotics is markedly activated fying the genes/regulators/regulons involved in the growth
by introducing specific types of rifampicin (Rif) mutations phase transition from primary to secondary metabolism
into the rpoB (encoding the RNA polymerase subunit) in S. coelicolor [105]. Since then, immense interest has
gene in S. lividans 66 [93]. Furthermore, generating double developed in transcriptome profiling of various Streptomyces,
mutants using gentamicin (Gen) or rif in str mutant further and studies have concluded that the expression of antibiotic
increases actinorhodin production by 1.7–2.5-fold, whereas biosynthetic genes is tightly controlled through multiple
triple mutants (str, rif and gen) produce almost 48 times more regulatory networks [106, 107]. Exploring the role of aveI
actinorhodin than that of the wild-type strain of S. coelicolor (negative regulator) by microarray in combination with real-
A3 (2) [94] and 2.3-fold higher salinomycin (10 mg/mL) in time reverse transcription PCR in S. avermitilis not only
S. albus [95]. These single, double, and triple mutants display showed a negative effect on the avermectin biosynthetic gene
in hierarchical order a remarkable increase in the production cluster but also affected expression of the oligomycin and
of actII-ORF4, a pathway-specific regulatory protein that filipin biosynthetic clusters. In addition, the genes involved
increases actinorhodin production [94]. Similarly, mutations in precursor biosyntheses for avermectin or other antibiotics,
conferring resistance to geneticin, fusidic acid, thiostrepton, such as crotonyl-CoA reductase and methylmalonyl-CoA
and lincomycin generate quintuple, sextuple, septuple, and decarboxylase, were also upregulated in the aveI mutant.
octuple mutants (C5, C6, C7, and C8, resp.) that produce
Genes of several key primary metabolic pathways were
1.63 g/L and 1.22 g/L actinorhodin, which is 180, and 136-
downregulated in the mutant, suggesting that the aveI gene
fold higher than that of the wild-type strain S. coelicolor A3
may function as a global regulator involved in directing
(2) in GYM33 media [96]. This dramatic overproduction
carbon flux from primary to secondary metabolism [108].
of valuable drugs was the reason that ribosomal mutations
and increased accumulation of bacterial alarmone (ppGpp) Similarly, comparative transcriptome analysis between the
were found to play a pivotal role in the onset of antibiotic low and high producer S. avermitilis using a whole-genome
production in bacteria. Nevertheless, mutations in the RNA chip revealed the tetR family transcriptional regulator as
polymerase beta-subunit circumvent dependence on ppGpp a global upregulator for enhanced antibiotic production
production or increase stability of the 70S complex resulting in Streptomyces species [109]. Moreover, the wblA gene is
in a higher translation level [95–97] and overproduction. a pleiotropic downregulator of antibiotic biosynthesis in
However, the fundamental mechanism by which ribosomal Streptomyces species based on a transcriptomics study using
engineering affects antibiotic production has been summa- DNA microarray analysis for analyzing the discrepancy in
rized in earlier reviews [98, 99]. mRNA abundance associated with DXR in S. peucetius
Boosting translation during the stationary phase is overproducing industrial mutant (OIM) [100]. Furthermore,
another way to enhance secondary metabolite production in disruption of wblA from the S. peucetius OIM resulted in an
streptomycetes. Enhanced expression of the frr gene by muta- additional 1.7-fold increase in the production of both DXR
tions in the rpsL gene, which encodes a ribosome recycling and daunorubicin (DNR) [107]. These results suggest that
factor (RRF), results in greater production of actinorhodin transcriptome based studies provide a comparative profile of
due to enhanced S. coelicolor protein synthesis [100], whereas gene expression at the molecular level and help to assess the
overexpression of the frr gene increases avermectin yield key regulators for manifesting designer strains with enhanced
(by 3 to 3.7-fold) in S. avermitilis strains due to the “copy secondary metabolite production.
number effect” of the frr gene [101]. Moreover, introducing Recent “-omics” guided reverse engineering approaches,
rpsL and rpoB mutations in S. coelicolor enhances production including comparative transcriptomics and proteomics, have
of chloramphenicol and congocidine by 40-fold and 30- been successfully used to identify alterations in gene expres-
fold, respectively [42]. Furthermore, mutations in rsmG gene sion associated with overproduction of secondary metabo-
encoding for 16S rRNA methyltransferase [102] eventually lites in industrial Streptomyces strains [79, 110–113]. The
lead to increase of the intracellular pool of SAM [103] and strategy to “reverse engineer” a reference organism (with a
overproduction of antibiotics in streptomycetes[81, 82]. A desirable property such as higher yield) is carried out by
combination of ribosomal engineering and reporter guided identifying the genetic or molecular basis of the property
mutant selection helped to generate a daptomycin overpro- and subsequently reengineering the property into target
ducing strain that produces twice as much A21978C (acidic organisms of interest by considering the key genes involved in
cyclic lipopeptide antibiotic) as that of the parental strain of complex mechanisms controlling microbial metabolism [80].
S. roseosporus [104]. Successful reverse engineering depends on reproducibility of
the overproducing mechanism in new target strains by using
organisms whose genomic information is already available.
5. Engineered Transcriptomics An overproduction mutation is identified and a similar
genetic manipulation is introduced into the same or closely
Many successful stories of genome sequencing have been related species, which is useful for achieving higher product
generated by efficient mining of genome data either in silico titers without additional knowledge of concrete overproduc-
or in wet lab experiments. Several tools have been developed tion mechanisms [79]. New microarray and proteomic tools
[105, 114, 115] as well as new tools for mutagenesis and mutant and increases the local concentration of intermediates, is
construction [116–118] are handy to characterize overproduc- a promising solution for this problem because scaffolds
ing strains and thus bioengineer new target organisms for help to (1) increase the local concentration of intermediates
enhanced production. around the enzymes on the scaffold, (2) prevent the loss
Similarly, precision engineering is a new approach that of intermediates by diffusion or by competing reactions
has been investigated to optimize existing biotechnology and (3) overcome feedback inhibition on other pathways
processes to improve desirable cell properties [119]. Aske- due to the rapid conversion of feedback inhibitors [130–
nazi et al. [120] described an approach to decipher the 132]. Several successful examples of “-omics” technologies for
complex interrelationships between metabolite production drug production have already appeared, and this trend will
and gene expression events to develop improved produc- continue at an accelerated pace. It is expected that microbial
tion strains. These advancements include transcriptional metabolic engineering will become an essential platform for
profiling using DNA microarrays, proteome profiling by developing and producing drugs in the near future.
2D gel electrophoresis, and metabolite profiling by high-
performance liquid chromatography. The cumulative infor- Acknowledgments
mation from these sources enables a more precise iden-
tification of key genetic targets and pathways engineered This study was supported by the Intelligent Synthetic Biology
for strain improvement [121]. Single “-omic” analyses are Center of Global Frontier Project funded by the Ministry of
not sufficient to fully unravel the complexities of microbial Education, Science and Technology (2011-0031960) and by
physiology and molecular biology associated with the pro- the Technology Development Program for Agriculture and
duction of secondary metabolites; thus, integrating different Forestry, Ministry of Agriculture and Forestry (20100368),
layers of information that is, multi “-omics” approaches, Republic of Korea.
is essential to acquire precise insight into microorganisms
and the mechanism of secondary metabolite overproduction.
Thus, transcriptomics enables quantitative measurements of References
dynamic mRNA expression and variations between different [1] A. L. Demain and A. Fang, “The natural functions of
states, reflecting the genes that are being overexpressed or secondary metabolites,” Advances in Biochemical Engineer-
downregulated at particular times and conditions. Hence, ing/Biotechnology, vol. 69, pp. 1–39, 2000.
knowledge of transcriptomics is crucial for designing a ratio- [2] J. Bérdy, “Bioactive microbial metabolites,” The Journal of
nal integrated approach to enhance secondary metabolite Antibiotics, vol. 58, pp. 1–26, 2005.
production. [3] D. J. Newman and G. M. Cragg, “Natural products as sources of
new drugs over the last 25 years,” Journal of Natural Products,
vol. 70, no. 3, pp. 461–477, 2007.
6. Future Perspectives [4] N. Dufour and R. P. Rao, “Secondary metabolites and other
small molecules as intercellular pathogenic signals,” FEMS
Secondary metabolite production by streptomycetes can be Microbiology Letters, vol. 314, no. 1, pp. 10–17, 2011.
efficiently enhanced by a number of approaches described [5] S. A. Waksman and A. T. Henrici, “The nomenclature and
herein; it is reasonable to expect that these techniques cannot classification of the Actinomycetes,” Journal of Bacteriology, vol.
be the endpoint. There are many such techniques developed 46, no. 4, pp. 337–341, 1943.
for other bacterial genera to enhance secondary metabo- [6] H. Wildermuth and D. A. Hopwood, “Septation during sporula-
lite production, as engineered microbes typically require tion in Streptomyces coelicolor,” Journal of General Microbiology,
a high level of genetically stable expressing heterologous vol. 60, no. 1, pp. 51–59, 1970.
genes and pathways for genetic stability. For example, Tyo [7] S. Omura, H. Ikeda, J. Ishikawa et al., “Genome sequence of
et al. developed a technique called chemically inducible an industrial microorganism Streptomyces avermitilis: deducing
chromosomal evolution (CIChE), which is a plasmid-free the ability of producing secondary metabolites,” Proceedings of
system for engineering E. coli with reduced allele segregation
vol. 98, no. 21, pp. 12215–12220, 2001.
and enables roughly 2 to 4-fold increases in the yields of
[8] S. D. Bentley, K. F. Chater, A.-M. Cerdeño-Tárraga et al., “Com-
lycopene and the polymer poly-3-hydroxybutyrate [122, 123].
plete genome sequence of the model actinomycete Streptomyces
Similarly, reducing the number of plasmids to overcome coelicolor A3(2),” Nature, vol. 417, no. 6885, pp. 141–147, 2002.
differential gene expression by assembling a large construct
[9] H. Ikeda, J. Ishikawa, A. Hanamoto et al., “Complete genome
from small fragments is becoming a popular technology sequence and comparative analysis of the industrial microor-
in synthetic biology [124]. Manipulating gene clusters into ganism Streptomyces avermitilis,” Nature Biotechnology, vol. 21,
monocistronic or pseudooperons has led to engineered no. 5, pp. 526–531, 2003.
biosynthesis of many natural products [125–128]. A tandemly [10] Y. Ohnishi, J. Ishikawa, H. Hara et al., “Genome sequence of the
placed repetitive promoter is another powerful technique for streptomycin-producing microorganism Streptomyces griseus
gene overexpression and enhanced metabolite production IFO 13350,” Journal of Bacteriology, vol. 190, no. 11, pp. 4050–
[129]. Strain development is still hampered by the intrinsic 4060, 2008.
inefficiency of metabolic reactions caused by simple diffusion [11] M. Nett, H. Ikeda, and B. S. Moore, “Genomic basis for natural
and random collisions of enzymes and metabolites. A scaffold product biosynthetic diversity in the actinomycetes,” Natural
system, which promotes the proximity of metabolic enzymes Product Reports, vol. 26, no. 11, pp. 1362–1384, 2009.
[12] D. A. Hopwood, “Soil to genomics: the Streptomyces chromo- [30] J. Liu, J. D. Farmer Jr., W. S. Lane, J. Friedman, I. Weissman, and
some,” Annual Review of Genetics, vol. 40, pp. 1–23, 2006. S. L. Schreiber, “Calcineurin is a common target of cyclophilin-
[13] D. C. Liebler, Introduction to Proteomics: Tools for the New cyclosporin A and FKBP-FK506 complexes,” Cell, vol. 66, no. 4,
Biology, Humana Press, Totowa, NJ, USA, 2002. pp. 807–815, 1991.
[14] M. R. Wilkins, R. D. Appel, K. L. Williams, and D. F. [31] D. Chen, Q. Zhang, Q. Zhang, P. Cen, Z. Xu, and W. Liu,
Hochstrasser, Proteome Research: Concepts, Technology and “Improvement of FK506 production in Streptomyces tsukubaen-
Application, Springer, New York, NY, USA, 2nd edition, 2007. sis by genetic enhancement of the supply of unusual polyketide
[15] J. Nielsen and J. D. Keasling, “Synergies between synthetic extender units viautilization of two distinct site-specific recom-
biology and metabolic engineering,” Nature Biotechnology, vol. bination systems,” Applied and Environmental Microbiology, vol.
29, no. 8, pp. 693–695, 2011. 78, no. 15, pp. 5093–5103, 2012.
[16] R. D. Fleischmann, M. D. Adams, O. White et al., “Whose- [32] S. Malla, N. P. Niraula, K. Liou, and J. K. Sohng, “Enhancement
genome random sequencing and assembly of Haemophilus of doxorubicin production by expression of structural sugar
influenzae Rd,” Science, vol. 269, no. 5223, pp. 496–521, 1995. biosynthesis and glycosyltransferase genes in Streptomyces
[17] A. Varma and B. O. Palsson, “Stoichiometric flux balance peucetius,” Journal of Bioscience and Bioengineering, vol. 108, no.
models quantitatively predict growth and metabolic by-product 2, pp. 92–98, 2009.
secretion in wild-type Escherichia coli W3110,” Applied and [33] J.-N. Volff and J. Altenbuchner, “Genetic instability of the
Environmental Microbiology, vol. 60, no. 10, pp. 3724–3731, 1994. streptomyces chromosome,” Molecular Microbiology, vol. 27, no.
[18] D. A. Hodgson, “Primary metabolism and its control in strep- 2, pp. 239–246, 1998.
tomycetes: a most unusual group of bacteria,” Advances in [34] T. Murakami, J. Burian, K. Yanai, M. J. Bibb, and C. J. Thompson,
Microbial Physiology, vol. 42, pp. 47–238, 2000. “A system for the targeted amplification of bacterial gene
[19] C. Bro and J. Nielsen, “Impact of “ome” analyses on inverse clusters multiplies antibiotic yield in Streptomyces coelicolor,”
metabolic engineering,” Metabolic Engineering, vol. 6, no. 3, pp. Proceedings of the National Academy of Sciences of the United
204–211, 2004. States of America, vol. 108, no. 38, pp. 16020–16025, 2011.
[20] Q.-T. Nguyen, M. E. Merlo, M. H. Medema, A. Jankevics, [35] K. Yanai, T. Murakami, and M. Bibb, “Amplification of the entire
R. Breitling, and E. Takano, “Metabolomics methods for the kanamycin biosynthetic gene cluster during empirical strain
synthetic biology of secondary metabolism,” FEBS Letters, 2012. improvement of Streptomyces kanamyceticus,” Proceedings of the
[21] A. Aharoni and G. Galili, “Metabolic engineering of the plant National Academy of Sciences of the United States of America, vol.
primary-secondary metabolism interface,” Current Opinion in 103, no. 25, pp. 9661–9666, 2006.
Biotechnology, vol. 22, no. 2, pp. 239–244, 2011. [36] G. Liao, J. Li, L. Li, H. Yang, Y. Tian, and H. Tan, “Cloning,
[22] Z. Liu, M. Guo, J. Qian, Y. Zhuang, and S. Zhang, “Disruption reassembling and integration of the entire nikkomycin biosyn-
of zwf2 gene to improve oxytetraclyline biosynthesis in Strepto- thetic gene cluster into Streptomyces ansochromogenes lead to
myces rimosus M4018,” Wei Sheng Wu Xue Bao, vol. 48, no. 1, pp. an improved nikkomycin production,” Microbial Cell Factories,
21–25, 2008. vol. 9, p. 6, 2010.
[23] Z. Tang, C. Xiao, Y. Zhuang et al., “Improved oxytetracycline [37] C. Corre and G. L. Challis, “New natural product biosynthetic
production in Streptomyces rimosus M4018 by metabolic engi- chemistry discovered by genome mining,” Natural Product
neering of the G6PDH gene in the pentose phosphate pathway,” Reports, vol. 26, no. 8, pp. 977–986, 2009.
Enzyme and Microbial Technology, vol. 49, no. 1, pp. 17–24, 2011.
[38] K. Flinspach, L. Westrich, L. Kaysser et al., “Heterologous
[24] M. J. Butler, P. Bruheim, S. Jovetic, F. Marinelli, P. W. Postma,
expression of the biosynthetic gene clusters of coumermycina1,
and M. J. Bibb, “Engineering of primary carbon metabolism
clorobiocin and caprazamycins in genetically modified Strepto-
for improved antibiotic production in Streptomyces lividans,”
myces coelicolor strains,” Biopolymers, vol. 93, no. 9, pp. 823–832,
Applied and Environmental Microbiology, vol. 68, no. 10, pp.
2010.
4731–4739, 2002.
[39] R. McDaniel, S. Ebert-Khosla, D. A. Hopwood, and C. Khosla,
[25] Y.-G. Ryu, M. J. Butler, K. F. Chater, and K. J. Lee, “Engineering
“Engineered biosynthesis of novel polyketides,” Science, vol.
of primary carbohydrate metabolism for increased production
262, no. 5139, pp. 1546–1550, 1993.
of actinorhodin in Streptomyces codicolor,” Applied and Environ-
mental Microbiology, vol. 72, no. 11, pp. 7132–7139, 2006. [40] B. Floriano and M. Bibb, “afsR is a pleiotropic but conditionally
[26] N. Gunnarsson, A. Eliasson, and J. Nielsen, “Control of fluxes required regulatory gene for antibiotic production in Strepto-
towards antibiotics and the role of primary metabolism in myces coelicolor A3(2),” Molecular Microbiology, vol. 21, no. 2,
production of antibiotics,” Advances in Biochemical Engineer- pp. 385–396, 1996.
ing/Biotechnology, vol. 88, pp. 137–178, 2004. [41] T. Hosaka, M. Ohnishi-Kameyama, H. Muramatsu et al.,
[27] D. Huang, J. Wen, G. Wang, G. Yu, X. Jia, and Y. Chen, “In “Antibacterial discovery in actinomycetes strains with muta-
silico aided metabolic engineering of Streptomyces roseosporus tions in RNA polymerase or ribosomal protein S12,” Nature
for daptomycin yield improvement,” Applied Microbiology and Biotechnology, vol. 27, no. 5, pp. 462–464, 2009.
Biotechnology, vol. 94, no. 3, pp. 637–649, 2012. [42] J. P. Gomez-Escribano and M. J. Bibb, “Engineering Strep-
[28] I. Borodina, J. Siebring, J. Zhang et al., “Antibiotic overproduc- tomyces coelicolor for heterologous expression of secondary
tion in Streptomyces coelicolor A3(2) mediated by phosphofruc- metabolite gene clusters,” Microbial Biotechnology, vol. 4, no. 2,
tokinase deletion,” Journal of Biological Chemistry, vol. 283, no. pp. 207–215, 2011.
37, pp. 25186–25199, 2008. [43] M. Komatsu, T. Uchiyama, S. Omura, D. E. Cane, and H.
[29] S. B. Bankar and R. S. Singhal, “Metabolic precursors enhance Ikeda, “Genome-minimized Streptomyces host for the heterol-
the production of poly-𝜀-lysine by Streptomyces noursei NRRL ogous expression of secondary metabolism,” Proceedings of the
5126,” Engineering in Life Sciences, vol. 11, no. 3, pp. 253–258, National Academy of Sciences of the United States of America,
2011. vol. 107, no. 6, pp. 2646–2651, 2010.
[44] W. S. Jung, S. K. Lee, J. S. J. Hong et al., “Heterologous [59] X. A. Lv, Y. Y. Jin, Y. D. Li, H. Zhang, and X. L. Liang, “Genome
expression of tylosin polyketide synthase and production of a shuffling of Streptomyces viridochromogenes for improved pro-
hybrid bioactive macrolide in Streptomyces venezuelae,” Applied duction of avilamycin,” Applied Microbiology and Biotechnology,
Microbiology and Biotechnology, vol. 72, no. 4, pp. 763–769, vol. 97, no. 2, pp. 641–648, 2013.
2006. [60] L. Zhitang, Z. Dawei, L. Zhi, and L. Ye, “Nitrosoguanidine muta-
[45] L. P. Thapa, T.-J. Oh, H. C. Lee et al., “Heterologous expres- genesis and genome shuffling enhanced the oxytetracycline
sion of the kanamycin biosynthetic gene cluster (pSKC2) production of Streptomyces rimosus,” IOSR Journal of Pharmacy
in Streptomyces venezuelae YJ003,” Applied Microbiology and and Biological Sciences, vol. 4, no. 6, pp. 23–27, 2013.
Biotechnology, vol. 76, no. 6, pp. 1357–1364, 2007. [61] J. Gong, H. Zheng, Z. Wu, T. Chen, and X. Zhao, “Genome shuf-
[46] W. P. Je, J. S. J. Hong, N. Parajuli et al., “Genetic dissection of the fling: progress and applications for phenotype improvement,”
biosynthetic route to gentamicin A2 by heterologous expression Biotechnology Advances, vol. 27, no. 6, pp. 996–1005, 2009.
of its minimal gene set,” Proceedings of the National Academy [62] H. Zhu, Z.-H. Jin, and P.-L. Cen, “Natamycin-producing strain
of Sciences of the United States of America, vol. 105, no. 24, pp. breeding by genome shuffling,” Chinese Journal of Antibiotics,
8399–8404, 2008. vol. 31, no. 12, pp. 739–742, 2006.
[47] S. R. Park, J. W. Park, W. S. Jung et al., “Heterologous production [63] J. M. Luo, J. S. Li, D. Liu et al., “Genome shuffling of Streptomyces
of epothilones B and D in Streptomyces venezuelae,” Applied gilvosporeus for improving natamycin production,” Journal of
Microbiology and Biotechnology, vol. 81, no. 1, pp. 109–117, 2008. Agricultural and Food Chemistry, vol. 60, no. 23, pp. 6026–6036,
[48] J. Penn, X. Li, A. Whiting et al., “Heterologous production 2012.
of daptomycin in Streptomyces lividans,” Journal of Industrial [64] V. Paquet, G. Goma, and P. Soucaille, “Induction of pristi-
Microbiology and Biotechnology, vol. 33, no. 2, pp. 121–128, 2006. namycins production in Streptomyces pristinaespiralis,” Biotech-
nology Letters, vol. 14, no. 11, pp. 1065–1070, 1992.
[49] K. K. Nepal, T.-J. Oh, and J. K. Sohng, “Heterologous pro-
duction of paromamine in Streptomyces lividans TK24 using [65] R. Leclercq, C. J. Soussy, P. Weber, N. Moniot-Ville, and C.
kanamycin biosynthetic genes from Streptomyces kanamyceti- Dib, “In vitro activity of the pristinamycin against the isolated
cus ATCC12853,” Molecules and Cells, vol. 27, no. 5, pp. 601–608, Staphylococci in the french hospitals in 1999-2000,” Pathologie
2009. Biologie, vol. 51, no. 7, pp. 400–404, 2003.
[50] M. H. Medema, A. Trefzer, A. Kovalchuk et al., “The sequence [66] J. Ng and I. B. Gosbell, “Successful oral pristinamycin therapy
of a 1.8-Mb bacterial linear plasmid reveals a rich evolutionary for osteoarticular infections due to methicillin-resistant Staphy-
reservoir of secondary metabolic pathways,” Genome Biology lococcus aureus (MRSA) and other Staphylococcus spp,” Journal
and Evolution, vol. 2, no. 1, pp. 212–224, 2010. of Antimicrobial Chemotherapy, vol. 55, no. 6, pp. 1008–1012,
2005.
[51] E. A. Felnagle, M. R. Rondon, A. D. Berti, H. A. Crosby, and M.
G. Thomas, “Identification of the biosynthetic gene cluster and [67] W. Witte, K. G. Naber, B. Pasemann, C. Cuny, and I. Klare,
an additional gene for resistance to the antituberculosis drug “In vitro sensitivity of Staphylococci against the pristinamycin
capreomycin,” Applied and Environmental Microbiology, vol. 73, combination RP59500,” Chemotherapie Journal, vol. 5, no. 1, pp.
no. 13, pp. 4162–4170, 2007. 17–22, 1996.
[68] B. Jia, Z.-H. Jin, Y.-L. Lei, L.-H. Mei, and N.-H. Li, “Improved
[52] F. Lombó, A. Velasco, A. Castro et al., “Deciphering the
production of pristinamycin coupled with an adsorbent resin
biosynthesis pathway of the antitumor thiocoraline from a
in fermentation by Streptomyces pristinaespiralis,” Biotechnology
marine actinomycete and its expression in two Streptomyces
Letters, vol. 28, no. 22, pp. 1811–1815, 2006.
species,” ChemBioChem, vol. 7, no. 2, pp. 366–376, 2006.
[69] Q. Jin, Z. Jin, L. Zhang, S. Yao, and F. Li, “Probing the
[53] R. H. Baltz, “Streptomyces and Saccharopolyspora hosts for
molecular mechanisms for pristinamycin yield enhancement in
heterologous expression of secondary metabolite gene clusters,”
Streptomyces pristinaespiralis,” Current Microbiology, vol. 65, no.
Journal of Industrial Microbiology and Biotechnology, vol. 37, no.
6, pp. 792–798, 2012.
8, pp. 759–772, 2010.
[70] L. Bai, L. Li, H. Xu et al., “Functional analysis of the validamycin
[54] B. Xu, Z. Jin, H. Wang, Q. Jin, X. Jin, and P. Cen, “Evolution of biosynthetic gene cluster and engineered production of vali-
Streptomyces pristinaespiralis for resistance and production of doxylamine A,” Chemistry and Biology, vol. 13, no. 4, pp. 387–
pristinamycin by genome shuffling,” Applied Microbiology and 397, 2006.
[71] H. Hida, T. Yamada, and Y. Yamada, “Production of hydrox-
[55] Z. H. Jin, B. Xu, S. Z. Lin, Q. C. Jin, and P. L. Cen, “Enhanced pro- ycitric acid by microorganisms,” Bioscience, Biotechnology and
duction of spinosad in saccharopolyspora spinosa by genome Biochemistry, vol. 69, no. 8, pp. 1555–1561, 2005.
shuffling,” Applied Biochemistry and Biotechnology, vol. 159, no.
[72] H. Hida, T. Yamada, and Y. Yamada, “Absolute configuration
3, pp. 655–663, 2009.
of hydroxycitric acid produced by microorganisms,” Bioscience,
[56] G. Stephanopoulos, “Metabolic engineering by genome shuf- Biotechnology and Biochemistry, vol. 70, no. 8, pp. 1972–1974,
fling,” Nature Biotechnology, vol. 20, no. 7, pp. 666–668, 2002. 2006.
[57] R. Patnaik, S. Louie, V. Gavrilovic et al., “Genome shuffling of [73] C. Hansawasdi, J. Kawabata, and T. Kasai, “𝛼-amylase inhibitors
Lactobacillus for improved acid tolerance,” Nature Biotechnol- from roselle (Hibiscus sabdariffa Linn.) tea,” Bioscience, Biotech-
ogy, vol. 20, no. 7, pp. 707–712, 2002. nology and Biochemistry, vol. 64, no. 5, pp. 1041–1043, 2000.
[58] Y.-X. Zhang, K. Perry, V. A. Vinci, K. Powell, W. P. C. Stemmer, [74] T. Yamada, H. Hida, and Y. Yamada, “Chemistry, physiological
and S. B. Del Cardayré, “Genome shuffling leads to rapid properties, and microbial production of hydroxycitric acid,”
phenotypic improvement in bacteria,” Nature, vol. 415, no. 6872, Applied Microbiology and Biotechnology, vol. 75, no. 5, pp. 977–
pp. 644–646, 2002. 982, 2007.
[75] H. Hida, T. Yamada, and Y. Yamada, “Genome shuffling of [90] J. Shima, A. Hesketh, S. Okamoto, S. Kawamoto, and K. Ochi,
Streptomyces sp. U121 for improved production of hydroxycitric “Induction of actinorhodin production by rpsL (encoding
acid,” Applied Microbiology and Biotechnology, vol. 73, no. 6, pp. ribosomal protein S12) mutations that confer streptomycin
1387–1393, 2007. resistance in Streptomyces lividans and Streptomyces coelicolor
[76] S. Li, F. Li, X.-S. Chen et al., “Genome shuffling enhanced A3(2),” Journal of Bacteriology, vol. 178, no. 24, pp. 7276–7284,
𝜀-poly-l-lysine production by improving glucose tolerance of 1996.
Streptomyces graminearus,” Applied Biochemistry and Biotech- [91] Y. Hosoya, S. Okamoto, H. Muramatsu, and K. Ochi, “Acquisi-
nology, vol. 166, no. 2, pp. 414–423, 2012. tion of certain streptomycin-resistant (str) mutations enhances
[77] S. Li, X. Chen, C. Dong, F. Zhao, L. Tang, and Z. Mao, “Com- antibiotic production in bacteria,” Antimicrobial Agents and
bining genome shuffling and interspecific hybridization among Chemotherapy, vol. 42, no. 8, pp. 2041–2047, 1998.
Streptomyces improved 𝜀-Poly-l-Lysine production,” Applied [92] Y. Okamoto-Hosoya, T.-A. Sato, and K. Ochi, “Resistance to
Biochemistry and Biotechnology, vol. 169, no. 1, pp. 338–350, paronomycin is conferred by rpsl. mutations, accompanied by
2013. an enhanced antibiotic production in Streptomyces coelicolor
[78] D. L. Crawford, “Development of recombinant Streptomyces A3(2),” Journal of Antibiotics, vol. 53, no. 12, pp. 1424–1427, 2000.
for biotechnological and environmental uses,” Biotechnology [93] H. Hu, Q. Zhang, and K. Ochi, “Activation of antibiotic biosyn-
Advances, vol. 6, no. 2, pp. 183–206, 1988. thesis by specified mutations in the rpoB gene (encoding the
[79] A. M. Lum, J. Huang, C. R. Hutchinson, and C. M. Kao, RNA polymerase 𝛽 subunit) of Streptomyces lividans,” Journal
“Reverse engineering of industrial pharmaceutical-producing of Bacteriology, vol. 184, no. 14, pp. 3984–3991, 2002.
actinomycete strains using DNA microarrays,” Metabolic Engi- [94] H. Hu and K. Ochi, “Novel approach for improving the pro-
neering, vol. 6, no. 3, pp. 186–196, 2004. ductivity of antibiotic-producing strains by inducing combined
[80] Y. Zhuo, W. Zhang, D. Chen et al., “Reverse biological engi- resistant mutations,” Applied and Environmental Microbiology,
neering of hrdB to enhance the production of avermectins in vol. 67, no. 4, pp. 1885–1892, 2001.
an industrial strain of Streptomyces avermitilis,” Proceedings of
[95] N. Tamehiro, T. Hosaka, J. Xu, H. Hu, N. Otake, and K.
Ochi, “Innovative approach for improvement of an antibiotic-
vol. 107, no. 25, pp. 11250–11254, 2010.
overproducing industrial strain of Streptomyces albus,” Applied
[81] D.-J. Kim, J.-H. Huh, Y.-Y. Yang et al., “Accumulation of S- and Environmental Microbiology, vol. 69, no. 11, pp. 6412–6417,
adenosyl-L-methionine enhances production of actinorhodin 2003.
but inhibits sporulation in Streptomyces lividans TK23,” Journal
[96] G. Wang, T. Hosaka, and K. Ochi, “Dramatic activation of
of Bacteriology, vol. 185, no. 2, pp. 592–600, 2003.
antibiotic production in Streptomyces coelicolor by cumulative
[82] S. Okamoto, A. Lezhava, T. Hosaka, Y. Okamoto-Hosoya, and drug resistance mutations,” Applied and Environmental Micro-
K. Ochi, “Enhanced expression of S-adenosylmethionine syn- biology, vol. 74, no. 9, pp. 2834–2840, 2008.
thetase causes overproduction of actinorhodin in Streptomyces
coelicolor A3(2),” Journal of Bacteriology, vol. 185, no. 2, pp. 601– [97] J. Xu, Y. Tozawa, C. Lai, H. Hayashi, and K. Ochi, “A
609, 2003. rifampicin resistance mutation in the rpoB gene confers ppGpp-
independent antibiotic production in Streptomyces coelicolor
[83] S. Maharjan, T.-J. Oh, H. C. Lee, and J. K. Sohng, “Heterologous
A3(2),” Molecular Genetics and Genomics, vol. 268, no. 2, pp.
expression of metK1-sp and afsR-sp in Streptomyces venezuelae
179–189, 2002.
for the production of pikromycin,” Biotechnology Letters, vol. 30,
no. 9, pp. 1621–1626, 2008. [98] K. Ochi, S. Okamoto, Y. Tozawa et al., “Ribosome engineering
and secondary metabolite production,” Advances in Applied
[84] N. Saito, K. Kurosawa, J. Xu, S. Okamoto, and K. Ochi, “Effect of
Microbiology, vol. 56, pp. 155–184, 2004.
s-adenosylmethionine on antibiotic production in Streptomyces
griseus and Streptomyces griseoflavus,” Actinomycetologica, vol. [99] K. Ochi, “From microbial differentiation to ribosome engineer-
17, no. 2, pp. 47–49, 2003. ing,” Bioscience, Biotechnology and Biochemistry, vol. 71, no. 6,
[85] T. Wang, L. Bai, D. Zhu et al., “Enhancing macrolide production pp. 1373–1386, 2007.
in Streptomyces by coexpressing three heterologous genes,” [100] T. Hosaka, J. Xu, and K. Ochi, “Increased expression of
Enzyme and Microbial Technology, vol. 50, no. 1, pp. 5–9, 2012. ribosome recycling factor is responsible for the enhanced
[86] X. Zhang, M. Fen, X. Shi, L. Bai, and P. Zhou, “Overexpression protein synthesis during the late growth phase in an antibiotic-
of yeast S-adenosylmethionine synthetase metK in Streptomyces overproducing Streptomyces coelicolor ribosomal rpsL mutant,”
actuosus leads to increased production of nosiheptide,” Applied Molecular Microbiology, vol. 61, no. 4, pp. 883–897, 2006.
Microbiology and Biotechnology, vol. 78, no. 6, pp. 991–995, [101] L. Li, J. Guo, Y. Wen, Z. Chen, Y. Song, and J. Li, “Overexpression
2008. of ribosome recycling factor causes increased production of
[87] X.-Q. Zhao, Y.-Y. Jin, H.-J. Kwon, Y.-Y. Yang, and J.-W. Suh, “S- avermectin in Streptomyces avermitilis strains,” Journal of Indus-
Adenosylmethionine (SAM) regulates antibiotic biosynthesis in trial Microbiology and Biotechnology, vol. 37, no. 7, pp. 673–679,
Streptomyces spp. in a mode independent of its role as a methyl 2010.
donor,” Journal of Microbiology and Biotechnology, vol. 16, no. 6, [102] S. Okamoto, A. Tamaru, C. Nakajima et al., “Loss of a conserved
pp. 927–932, 2006. 7-methylguanosine modification in 16S rRNA confers low-level
[88] J. R. Yates III, “Mass spectral analysis in proteomics,” Annual streptomycin resistance in bacteria,” Molecular Microbiology,
Review of Biophysics and Biomolecular Structure, vol. 33, pp. vol. 63, no. 4, pp. 1096–1106, 2007.
297–316, 2004. [103] K. Nishimura, T. Hosaka, S. Tokuyama, S. Okamoto, and K.
[89] E. Song, S. Malla, Y.-H. Yang et al., “Proteomic approach to Ochi, “Mutations in rsmG, encoding a 16S rRNA methyltrans-
enhance doxorubicin production in panK-integrated Strepto- ferase, result in low-level streptomycin resistance and antibiotic
myces peucetius ATCC 27952,” Journal of Industrial Microbiology overproduction in Streptomyces coelicolor A3(2),” Journal of
and Biotechnology, vol. 38, no. 9, pp. 1245–1253, 2011. Bacteriology, vol. 189, no. 10, pp. 3876–3883, 2007.
[104] L. Wang, Y. Zhao, Q. Liu, Y. Huang, C. Hu, and G. [117] A. M. Gehring, J. R. Nodwell, S. M. Beverley, and R. Losick,
Liao, “Improvement of A21978C production in Streptomyces “Genomewide insertional mutagenesis in Streptomyces coeli-
roseosporus by reporter-guided rpsL mutation selection,” Jour- color reveals additional genes involved in morphological differ-
nal of Applied Microbiology, vol. 112, no. 6, pp. 1095–1101, 2012. entiation,” Proceedings of the National Academy of Sciences of the
[105] J. Huang, C.-J. Lih, K.-H. Pan, and S. N. Cohen, “Global analysis United States of America, vol. 97, no. 17, pp. 9642–9647, 2000.
of growth phase responsive gene expression and regulation of [118] B. Gust, G. L. Challis, K. Fowler, T. Kieser, and K. F. Chater,
antibiotic biosynthetic pathways in Streptomyces coelicolor using “PCR-targeted Streptomyces gene replacement identifies a pro-
DNA microarrays,” Genes & Development, vol. 15, no. 23, pp. tein domain needed for biosynthesis of the sesquiterpene soil
3183–3192, 2001. odor geosmin,” Proceedings of the National Academy of Sciences
[106] N. P. Niraula, S.-H. Kim, J. K. Sohng, and E.-S. Kim, “Biotech- of the United States of America, vol. 100, no. 4, pp. 1541–1546,
nological doxorubicin production: pathway and regulation 2003.
engineering of strains for enhanced production,” Applied Micro- [119] R. Patnaik, “Engineering complex phenotypes in industrial
biology and Biotechnology, vol. 87, no. 4, pp. 1187–1194, 2010. strains,” Biotechnology Progress, vol. 24, no. 1, pp. 38–47, 2008.
[107] J.-H. Noh, S.-H. Kim, H.-N. Lee, S. Y. Lee, and E.-S. Kim, [120] M. Askenazi, E. M. Driggers, D. A. Holtzman et al., “Integrating
“Isolation and genetic manipulation of the antibiotic down- transcriptional and metabolite profiles to direct the engineering
regulatory gene, wblA ortholog for doxorubicin-producing of lovastatin-producing fungal strains,” Nature Biotechnology,
Streptomyces strain improvement,” Applied Microbiology and vol. 21, no. 2, pp. 150–156, 2003.
Biotechnology, vol. 86, no. 4, pp. 1145–1153, 2010. [121] H. Gao, X. Zhou, Z. Gou et al., “Rational design for over-
[108] L. Chen, J. Chen, Y. Jiang, W. Zhang, W. Jiang, and Y. Lu, production of desirable microbial metabolites by precision
“Transcriptomics analyses reveal global roles of the regulator engineering,” Antonie van Leeuwenhoek, vol. 98, no. 2, pp. 151–
AveI in Streptomyces avermitilis,” FEMS Microbiology Letters, 163, 2010.
vol. 298, no. 2, pp. 199–207, 2009. [122] K. E. J. Tyo, P. K. Ajikumar, and G. Stephanopoulos, “Stabilized
[109] C. T. P. Duong, H.-N. Lee, S.-S. Choi, S. Y. Lee, and E.-S. gene duplication enables long-term selection-free heterologous
Kim, “Functional expression of SAV3818, a putative TetR-family pathway expression,” Nature Biotechnology, vol. 27, no. 8, pp.
transcriptional regulatory gene from Streptomyces avermitilis, 760–765, 2009.
stimulates antibiotic production in Streptomyces species,” Jour- [123] A. H. Kachroo, M. Jayaram, and P. A. Rowley, “Metabolic
nal of Microbiology and Biotechnology, vol. 19, no. 2, pp. 136–139, engineering without plasmids,” Nature Biotechnology, vol. 27,
2009. no. 8, pp. 729–731, 2009.
[110] S.-H. Kang, J. Huang, H.-N. Lee, Y.-A. Hur, S. N. Cohen, and E.- [124] L. Zhang, G. Zhao, and X. Ding, “Tandem assembly of the
S. Kim, “Interspecies DNA microarray analysis identifies WblA epothilone biosynthetic gene cluster by in vitro site-specific
as a pleiotropic down-regulator of antibiotic biosynthesis in recombination,” Scientific Reports, vol. 1, article 141, 2011.
Streptomyces,” Journal of Bacteriology, vol. 189, no. 11, pp. 4315– [125] B. A. Pfeifer, S. J. Admiraal, H. Gramajo, D. E. Cane, and C.
4319, 2007. Khosla, “Biosynthesis of complex polyketides in a metabolically
[111] Y. J. Kim, J. Y. Song, M. H. Moon, C. P. Smith, S.-K. Hong, and Y. engineered strain of E. coli,” Science, vol. 291, no. 5509, pp. 1790–
K. Chang, “pH shock induces overexpression of regulatory and 1792, 2001.
biosynthetic genes for actinorhodin productionin Streptomyces [126] K. Watanabe, K. Hotta, A. P. Praseuth et al., “Total biosynthesis
coelicolor A3(2),” Applied Microbiology and Biotechnology, vol. of antitumor nonribosomal peptides in Escherichia coli,” Nature
76, no. 5, pp. 1119–1130, 2007. Chemical Biology, vol. 2, no. 8, pp. 423–428, 2006.
[112] W. Lian, K. P. Jayapal, S. Charaniya et al., “Genome-wide tran- [127] P. Xu, A. Vansiri, N. Bhan, and M. A. G. Koffas, “ePathBrick: a
scriptome analysis reveals that a pleiotropic antibiotic regulator, synthetic biology platform for engineering metabolic pathways
AfsS, modulates nutritional stress response in Streptomyces in E. coli,” ACS Synthetic Biology, vol. 1, no. 7, pp. 256–266, 2012.
coelicolor A3(2),” BMC Genomics, vol. 9, article 56, 2008. [128] A. K. Chaudhary, J. W. Park, Y. J. Yoon, B. G. Kim,
[113] H.-N. Lee, J.-H. Im, M.-J. Lee, S. Y. Lee, and E.-S. Kim, “A and J. K. Sohng, “Re-engineering of genetic circuit for 2-
putative secreted solute binding protein, SCO6569 is a possible deoxystreptamine (2-DOS) biosynthesis in Escherichia coli BL21
AfsR2-dependent down-regulator of actinorhodin biosynthesis (DE3),” Biotechnology Letters, vol. 35, no. 2, pp. 285–293, 2013.
in Streptomyces coelicolor,” Process Biochemistry, vol. 44, no. 3, [129] M. Li, J. Wang, Y. Geng et al., “A strategy of gene overexpression
pp. 373–377, 2009. based on tandem repetitive promoters in Escherichia coli,”
[114] J. Vohradsky, X.-M. Li, G. Dale et al., “Developmental control of Microbial Cell Factories, vol. 11, article 19, 2012.
stress stimulons in Streptomyces coelicolor revealed by statistical [130] J. M. Baek, S. Mazumdar, S. W. Lee et al., “Butyrate production
analyses of global gene expression patterns,” Journal of Bacteri- in engineered Escherichia coli with synthetic scaffolds,” Biotech-
ology, vol. 182, no. 17, pp. 4979–4986, 2000. nology and Bioengineering.
[115] A. R. Hesketh, G. Chandra, A. D. Shaw et al., “Primary and [131] R. J. Conrado, G. C. Wu, J. T. Boock et al., “DNA-guided
secondary metabolism, and post-translational protein modi- assembly of biosynthetic pathways promotes improved catalytic
fications, as portrayed by proteomic analysis of Streptomyces efficiency,” Nucleic Acids Research, vol. 40, no. 4, pp. 1879–1889,
coelicolor,” Molecular Microbiology, vol. 46, no. 4, pp. 917–932, 2012.
2002. [132] J. H. Lee, S. C. Jung, L. M. Bui, K. H. Kang, J. J. Song, and S. C.
[116] P. R. Herron, M. C. Evans, and P. J. Dyson, “Low target Kim, “Improved production of L-threonine in Escherichia coli
site specificity of an IS6100-based mini-transposon, Tn1792, by use of a DNA scaffold system,” Applied and Environmental
developed for transposon mutagenesis of antibiotic-producing Microbiology, vol. 79, no. 3, pp. 774–782, 2013.
Streptomyces,” FEMS Microbiology Letters, vol. 171, no. 2, pp.
215–221, 1999.
http://dx.doi.org/10.1155/2013/479893
Research Article
Genetic and Biochemical Diversity of Paenibacillus larvae
Isolated from Tunisian Infected Honey Bee Broods
Chadlia Hamdi,1,2 Jihène Essanaa,1 Luigi Sansonno,3 Elena Crotti,3 Khaoula Abdi,2
Naima Barbouche,4 Annalisa Balloi,3 Elena Gonella,5 Alberto Alma,5 Daniele Daffonchio,3
Abdellatif Boudabous,1 and Ameur Cherif1,2
1
Laboratory of Microorganisms and Active Biomolecules, Faculty of Sciences of Tunis,
University of Tunis El Manar, 2092 Tunis, Tunisia
2
LR Biotechnology and Bio-Geo Resources Valorization, Higher Institute for Biotechnology,
Biotechpole Sidi Thabet, University of Manouba, 2020 Ariana, Tunisia
3
Department of Food Environmental and Nutritional Sciences (DeFENS), University of Milan, 20133 Milan, Italy
4
Laboratoire de Zoologie et Apicultures, Institut National Agronomique de Tunis,
43 avenue Charles Nicolle, 1082 Tunis-Mahrajène, Tunisia
5
Dipartimento di Scienze Agrarie, Forestali e Alimentari, University of Turin, I-10095 Grugliasco, Italy
Correspondence should be addressed to Ameur Cherif; cherif.ameur@gmail.com
Received 30 April 2013; Accepted 27 July 2013
Copyright © 2013 Chadlia Hamdi et al. This is an open access article distributed under the Creative Commons Attribution License,
Paenibacillus larvae is the causative agent of American foulbrood (AFB), a virulent disease of honeybee (Apis mellifera) larvae.
In Tunisia, AFB has been detected in many beekeeping areas, where it causes important economic losses, but nothing is known
about the diversity of the causing agent. Seventy-five isolates of P. larvae, identified by biochemical tests and 16S rRNA gene
sequencing, were obtained from fifteen contaminated broods showing typical AFB symptoms, collected in different locations in
the northern part of the country. Using BOX-PCR, a distinct profile of P. larvae with respect to related Paenibacillus species was
detected which may be useful for its identification. Some P. larvae-specific bands represented novel potential molecular markers
for the species. BOX-PCR fingerprints indicated a relatively high intraspecific diversity among the isolates not described previously
with several molecular polymorphisms identifying six genotypes on polyacrylamide gel. Polymorphisms were also detected in
several biochemical characters (indol production, nitrate reduction, and methyl red and oxidase tests). Contrary to the relatively
high intraspecies molecular and phenotypic diversity, the in vivo virulence of three selected P. larvae genotypes did not differ
significantly, suggesting that the genotypic/phenotypic differences are neutral or related to ecological aspects other than virulence.
1. Introduction the world [6–9], and it is classified on list B of the World

Organization for Animal Health [10]. In many countries, an
American foulbrood (AFB), a severe and highly contagious eradication strategy exists with isolation and destruction of
disease affecting the larval and pupal stages of honeybee infected colonies and burning of contaminated equipments
(Apis mellifera), is caused by the bacterium Paenibacillus [10, 11].
larvae [1, 2]. AFB is one of the few diseases capable of Different genotypes of P. larvae have been identified in
killing the honeybee colony [3]. Prevention and control different regions. By using BOX-PCR three genotypes (A,
of AFB are very difficult because the pathogen produces B, and C) have been identified within a worldwide isolate
spores that are resistant to heat and chemical agents and collection [12]. In Germany, four different genotypes of P.
can remain viable for more than 35 years [4, 5]. AFB is larvae, named AB, Ab, ab and 𝛼B, have been described by
causing considerable economic loss to beekeepers all over combining BOX A1R and MBO REP1 primers [13, 14]. Using
N
Mediterranean Sea
France
Bizerte
S Italy
Ariana
N ́ Tunis
Beja
Jendouba
W E Zaghouan
Siliana
Mediterranean Sea S
Sousse
Algeria Tunisia
Kasserine
500 km
Libya 100 km
Figure 1: Location of the 15 sampled AFB contaminated hives in the northern area of Tunisia (grey-shaded circle).
the same combination (BOX A1R and MBO REP 1 primers), of the African continent [26], very limited knowledge is
Loncaric et al. [15] described five different genotypes (ab, aB, available on AFB and the genetic diversity of P. larvae.
Ab, AB, and 𝛼b). After the reclassification of P. larvae was The aim of the present work was to characterize a
proposed, as one species without subspecies separation, the collection of P. larvae isolated from Tunisian diseased brood
use of other techniques as ERIC-PCR for subtyping P. larvae and to study the genetic and biochemical diversity related to
and four different genotypes (ERIC I–IV) were identified these isolates.
[1]. The genotypes ERIC I and II correspond to the former
Paenibacillus larvae subsp. larvae and ERIC III and IV to the
former Paenibacillus larvae subsp. pulvifaciens. 2. Materials and Methods
AFB is readily disseminated by honeybees robbing honey
2.1. P. larvae Isolation. Seventy-five isolates of P. larvae were
from neighboring hives and the larval feeding of spores-
obtained between 2003 and 2005 from diseased honeybee
contaminated pollen and honey [16] or the reuse of contam-
larvae originating from 15 different hives in the northern
inated beekeeping equipments [11]. A role in the spread of P.
part of Tunisia (Figure 1). The isolates were obtained on
larvae has been also attributed to Varroa destructor [17] and
Columbia blood agar containing 5% horse blood for 48 h
the hive beetle Aethina tumida [18]. The spores ingested by
at 37∘ C. This step was preceded by a heat treatment at
the newly hatched larvae germinate in the midgut lumen. The
80∘ C for 10 min to eliminate the quick growing bacteria
vegetative forms of P. larvae penetrate the gut epithelium and
that may outcompete P. larvae on the plates. Nine reference
spread into the larval tissues [19, 20].
strains of seven Paenibacillus species phylogenetically related
Although some studies have investigated the pathogenic-
to P. larvae were obtained from the Bacillus Genetic Stock
ity of P. larvae and the virulence factors involved in the
Center (BGSC), USA: Paenibacillus alvei 33A3 and 33A4,
infection, the picture of the P. larvae virulence mechanisms
Paenibacillus polymyxa ATCC842T, Paenibacillus popilliae
is not yet complete. Dancer and Chantawannakul [21] asso-
2525 and B2519, Paenibacillus vorticalis 30A1, Paenibacillus
ciated the pathogenicity of P. larvae with the secretion of
thiaminolyticus NRRLB-4156T, Paenibacillus dendritiformis
metalloproteases. Antúnez et al. [22] reported the production
T168, and Paenibacillus macerans BKM B-51. All these refer-
by P. larvae of an enolase that could have a role in the
ence strains were routinely cultivated on nutrient broth and
virulence of the pathogen. Recently, P. larvae virulence has
agar at 30∘ C for 24 h.
been associated with an S layer protein [23] whose presence
determined the difference in the virulence between ERIC I
and ERIC II genotypes [24] with the former showing a weaker 2.2. Phenotypic and Biochemical Characterization. Cell and
virulence due to the absence of the specific S-layer [23]. This colony morphologies of all isolates were described, and their
study evidenced the importance of P. larvae genetic diversity biochemical profile was determined according to Gordon
in relation to virulence and highlighted the need for assessing et al. [27] with the following tests: catalase test, nitrate
the intraspecies diversity in areas of intensive apiculture. reduction, gelatin, starch, and casein hydrolysis, tyrosine
AFB disease has been reported in Arab countries includ- and urea degradation, acid from glucose, oxidase test, VP
ing North Africa [25] and in Tunisia; it has been detected in test, production of dihydroxyacetone, and indol and citrate
many beekeeping areas, where it causes important economic test. The growth was tested at different temperatures (4∘ C,
losses. Even though it has been shown that the economic 30∘ C, 37∘ C, and 50∘ C) and in media containing 2% and 5%
value of pollination in North Africa is among the highest of NaCl. All phenotypic tests were made in triplicate and
repeated when inconsistent results were observed. Positive by Forsgren et al. [39], with few modifications. The sporulated
and negative results were coded as 1 and 0, respectively, cultures were centrifuged at 3000 rcf for 15 min, and the
and cluster analysis was carried out by the unweighted pair spores were washed twice with sterile distilled water. The
group method with arithmetic averages (UPGMA) using the number of spores in the final suspensions was determined
Jaccard coefficient [28]. by plate count after 80∘ C heat treatment. The spore solutions
were further diluted in larval diet to give final concentrations
2.3. DNA Extraction and PCR Conditions. DNA was of approximately 5 × 103 CFU mL−1 and 105 CFU mL−1 .
extracted from bacteria using the TE solution (10 mM Tris Honey bee larvae of <24 h (based on body size) were col-
HCl, pH 7.4; 1 mM EDTA, pH 8), lysozyme (35 mg mL−1 ), lected from a healthy beehive and reared in U-shaped 96-well
and proteinase K (10 mg mL−1 ) [29]. The P. larvae strains plates according to the method of Peng et al. [40]. The grafted
were identified by 16S rRNA gene sequencing and typed by larvae were fed with an artificial liquid diet containing 50%
BOX-PCR, a technique widely used for the strain typing of of royal jelly, 50% of an aqueous solution of yeast extract, and
bacteria [30] including Bacillus species [12, 29, 31]. 12% each of D-glucose and D-fructose, both filtered at 0.2 𝜇m
PCR amplification of the 16S rRNA gene and the BOX [41]. The diet was provided to the larvae with micropipette
gene was performed using the universal primers, S-D-Bact- once a day for six days. For experimental infection and
0008-a-S-20/S-D-Bact-1495-a-A-20 and BOX A1R, respec- before grafting, each well of the plate was filled with 20 𝜇L
tively [29]. PCRs were performed in a final volume of 25 𝜇L of artificial liquid diet supplemented with a final P. larvae
containing 0.5 𝜇M of each oligonucleotide primer for the spore concentrations of 5 × 103 CFU mL−1 or 105 CFU mL−1
16S rRNA PCR and 1 𝜇M for the BOX PCR primer, 200 𝜇M for the exposed groups and without P. larvae spores for the
dNTPs, 2.5 mM MgCl2 , and 1U of DNA Taq polymerase. PCR control group. The larvae were exposed to P. larvae spores for
was performed for 35 cycles of 45 s at 94∘ C, 45 s at 55∘ C/42∘ C, 24 h after grafting. Forty-eight larvae per group were used in
respectively, for 16S rRNA PCR and BOX-PCR and 60 s at this exposure test, and the experiments were performed three
72∘ C. BOX-PCR products were separated in standard 1.5% times. After grafting, the plates containing young larvae were
agarose gel and in 6% polyacrylamide gel, visualized under incubated at 35∘ C in presence of a saturated solution of K2 SO4
UV light, and photographed with a gel doc digital image to keep the humidity at 96% [41]. During each of the eight
capture system (Bio-Rad). days of rearing, the larvae were examined for their vitality,
Numerical analysis of BOX patterns was performed using and dead and symptomatic individuals were noted both for
the MVSP 3.1 software [32]. Bands from all the gels were the larvae exposed to P. larvae and the nontreated ones.
manually detected using as markers the 100 bp (Fermentas)
or the 50 bp ladders (Promega), allowing the identification of 2.6. Statistical Analysis. Statistics were calculated by using
the different BOX genotypes. Microsoft Excel software [42]. Mean and standard deviations
were determined for three independent experiments, and
2.4. 16S rRNA Gene Sequencing and Phylogenetic Analysis. results were presented as mean ± SD. The Student’s t-test
The 16S rRNA gene sequencing was performed at the Primm was used to test for statistical significance of the difference
Biotech (Milano, Italy). Partial 16S rRNA gene sequences (E. between the mortality of the three groups of infected larvae
coli coordinates nt 52 to 787) of the isolates were compared with three different strains of P. larvae. A 𝑃 value of less than
with 16S rRNA gene sequences available by the BLAST search 0.05 was considered statistically significant.
[33], in the National Centre for Biotechnology Information
(NCBI) database (http://www.ncbi.nlm.nih.gov/). Multiple 3. Results
sequence alignments were performed using ClustalW version
1.8 [34]. The method of Jukes and Cantor [35] was used 3.1. Biochemical, Physiological, and Morphological Characters.
to calculate evolutionary distances. Phylogenetic tree was P. larvae colonies were small (3 mm in diameter), regular,
constructed by the neighbor-joining method [36], and the buttery, and greyish. Cells were examined, and all isolates
reliability of the tree topology was evaluated by bootstrap were Gram-positive rods with a width of about 1 𝜇m and a
analysis of 500 resampled data sets using MEGA 4.1 software length of 3–5 𝜇m. Bacteria appeared as single cells or pairs
[37, 38]. and sometimes as short chains.
16S rRNA gene sequences of thirteen P. larvae isolates, All isolates were catalase negative, grew at 30 and 37∘ C
BMG 93, BMG 184, BMG 189, BMG 191, BMG 192, BMG and in 2% NaCl media, but not in nutrient broth, at 4∘ C,
194, BMG 198, BMG 201, BMG 232, BMG 235, BMG 245, at 50∘ C, and 5% NaCl. Citrate was not utilized. Isolates
BMG 250, and BMG 259, were deposited under GenBank were positive for degradation of casein and gelatin and for
accession numbers FJ649367, FJ649355, FJ649365, FJ649362, acid production from glucose and starch. Tyrosine was not
FJ649358, FJ649363, FJ649356, FJ649357, FJ649361, degraded. Most of strains reduced nitrate to nitrite. Variable
FJ649359, FJ649364, FJ649360, and FJ649366, respectively. results were obtained for oxidase and methyl red tests, and
the strains did not form dihydroxyacetone and indol and were
2.5. Exposure Bioassays for Investigating the Virulence of Three negative for the Voges-Proskauer test (Table 1).
P. larvae Isolates. The P. larvae strains (BMG 93, BMG 184,
and BMG 259) used for the artificial larval infection were cul- 3.2. Numerical Analysis. The dendrogram of the biochemical
tivated on MYPGP agar, at 37∘ C for 10 to 14 days as described results of the isolates and the reference strains discriminated
4
Table 1: Biochemical characteristics of P. larvae isolates and Paenibacillus reference strains.

P. larvae Paenibacillus reference strains (BGSC)
75 isolates of P. alvei P. alvei P. popilliae P. popilliae P. dendritiformis P. polymyxa P. vorticalis P. macerans
Biochemical tests P. thiaminolyticusa
P. larvae 46-c-3 2771 2525 B2519 T168b ATCC 842T 31A1 BKM B-51
Catalase activity − + + − − + + + + +
Oxidasec v − + − − + + + + +
Starch hydrolysis − − + − − + − + + +
Casein hydrolysis + + + − − + − + − −
Dihydroxyacetone − + + − − − − + − −
Tyrosine
− − − − − + − − − −
decomposition
Gelatin
+ + + − − + − + + +
liquefaction
Methyl redd v − + − − + − + − +
Indole productione v − − − − + + − − −
Urea − − − + + + + + − −
DNase − − − − − − − − − −
Nitrate reductionf v − − − − + − + + +
Voges-Proskauer
− + + − − − − + + +
test
Glucose + + + − − + + + + +
Lactose + + + + + + + + + +
Gaze − − − − − − − − − −
H2 S − − − − − + + − − −
Citrate utilization − − − − − + − − − −
2% NaCl + + + + + + + + + +
5% NaCl − − + − − + − − − −
4∘ C and 50∘ C − − − − − − − − − −
30∘ C and 37∘ C + + + + + + + + + +
Nutrient broth − + + + + + + + + +
a
P. thiaminolyticus NRRLB-4156T; b P. dendritiformis subsp. dendron; BGSC: Bacillus Genetic Stock Center; +: 100% of the strains positive; −: 100% of strains negative; v: variation between strains; c 7% (−) and 93%
(+); d 1% (+) and 99% (−).
Jaccard's coefficient Species
0,4 0,5 0,6 0,7 0,8 0,9 1

P. popilliae B 2519
P. popilliae 2525
A
Paenibacillus reference strains

P. dendritiformis T 168
P. thiaminolyticus
B1
P. alvei 33A3
P. alvei 33A4
P. polymyxa
P. macerans
P. vorticalis
B 1
BMG 232 Ox (+) 4
P. larvae isolates
2
BMG 198 Ox (+), MR (+) 1
B2
3 Ox (−), Nit (+)
BMG 259 66
Ind (−), MR (−)
4
BMG 194 Ind (+) 2
5
BMG 189 Nit (−) 2
Characteristics N. strains
Figure 2: Dendrogram showing the biochemical profile relationship between P. larvae isolates and Paenibacillus reference strains. Ox: oxidase;
Nit: nitrate reduction; MR: methyl red; Ind: indol; +: positive response; −: negative response.
two groups (Figure 2). The first group (A) contained reference 198, BMG 201, and BMG 189). A2.3 included only the isolate
strains P. popilliae 2525, P. popilliae B2519, and P. dendriti- BMG 259.
formis T168. The second group (B) was subdivided into two
subgroups. The first (B1) included the reference strains P. 3.4. BOX-PCR Analysis of P. larvae Isolates. BOX-PCR dis-
thiaminolyticus NRRLB-4156T, P. alvei 46-c-3, P. alvei 2771, tinguished three genotypes out of 75 P. larvae isolates named
P. polymyxa ATCC 842T, P. macerans BKM B-51, and P. vor- A, B, and C (Figure 4(a)). P. larvae isolates presented a
ticalis 31A1. The second sub-group (B2) contained exclusively specific banding pattern clearly different from the other
the local P. larvae isolates (75 strains) well separated from the Paenibacillus species. The presence or absence of bands
Paenibacillus species reference strains. around 300 and 350 bp distinguished the three genotypes.
Genotype A showed six bands of approximate sizes: 280, 300,
3.3. 16S rRNA Gene Sequencing. 16S rRNA gene sequences 350, 650, 700, and 800 bp. Genotype B was characterized by
of thirteen P. larvae isolates (BMG 93, BMG 184, BMG 189, the absence of the 350 bp band, and the genotype C showed
BMG 191, BMG 192, BMG 194, BMG 198, BMG 201, BMG only four bands of 280 bp, 650 bp, 700 bp, and 800 bp.
232, BMG 235, BMG 245, BMG 250, and BMG 259) showed Eleven polymorphic bands in the 200–1000 bp range
99% identity with those of P. larvae in Genbank. 16S rRNA were detected within the BOX-PCR profiles separated by
gene sequences of 480 bp were used for the construction of 6% polyacrylamide gel electrophoresis (Figure 4(b)), some
the phylogeny of the isolates and standard strains of P. larvae of which could not be seen on agarose gel (Figure 4(a)). Six
available in Genbank. BOX-PCR genotypes (G1 to G6) were distinguished for the
The phylogenetic tree of partial 16S rRNA gene sequences 75 isolates (Table 2). Genotypes G2 and G4 represented the
(480 bp) grouped all P. larvae isolates and strains in branch most frequent in the collection, including 50% and 20% of the
A that showed two sub-groups (Figure 3). Subgroup A1 strains, respectively, while the remaining 30% of the strains
contained the reference strain, P. larvae DSM 7030. Sub- were distributed among the other four genotypes (G1, G3, G5,
group A2 showed three branches A2.1, A2.2, and A2.3. Branch and G6).
A2.1 represented two isolates BMG 194 and BMG 93. A2.2
grouped the reference strains 03-183 (DQ079623) P. larvae 3.5. Exposure Bioassays for Investigating the Virulence of
(AY030079), and the Tunisian isolates (BMG 191, BMG 235, P. larvae Isolates. One P. larvae isolate for each of the
BMG 184, BMG 192, BMG 245, BMG 232, BMG 250, BMG three different branches of the 16S rRNA gene phylogenetic
Phenotypes
Genotypes
Strains
66 pbl9 (BMG 194) P4 G3
A2.1
BMG 93 P3 G2
∗
03-189 (DQ079623)
pbl4 (BMG 192) P4 G2
pbl10 (BMG 245) P3 G5

Local P. larvae
A2.2 pbl8 (BMG 191) P1 G6
A2 isolates
69

P. larvae (AY030079)∗
A
pbl11 (BMG 189) P5 G5
pbl12 (BMG 259) P3 G1

A2.3
A1 Paenibacillus larvae DSM 7030∗ Reference
B Paenibacillus polymyxa EU982546 strains
0.005
Figure 3: Neighbour-joining phylogenetic tree of partial 16S rRNA genes sequences of 13 local isolates of P. larvae (BMG 93, BMG 192, BMG
194, BMG 198, BMG 201, BMG 232, BMG 245, BMG 235, BMG 250, BMG 259, BMG 184, BMG 189, and BMG 191) and three of their closest
relatives (indicated by stars). P. polymyxa (EU982546) was used as an out-group. The method of Jukes and Cantor was used to calculate
evolutionary distances. Bootstrap values (𝑛 = 500 replicates) were indicated at the nodes.
A B C 1 2 3 4 5 6 7 m M G6 G2 M G5 G4 G3 G1 M
800 bp 1000 bp 800 bp

700 bp
650 bp
500 bp 500 bp
350 bp
300 bp
300 bp
280 bp 250 bp
200 bp
100 bp 100 bp
(a) (b)
Figure 4: REP-PCR using BOX primer. (a) The relationship between the isolates of P. larvae and other Paenibacillus species detected on
agarose gels: lane 1: P. macerans; lane 2: P. alvei A4; lane 3: P. thiaminolyticus; lane 4: P. alvei A3; lane 5: P. dendritiformis; lane 6: P. vorticalis;
lane 7: P. polymyxa. A, B, and C: three BOX haplotypes detected on agarose gel. (b) BOX-PCR profile of P. larvae isolates detected on 6%
polyacrylamide gels; six BOX haplotypes were detected for 75 isolates (G1 to G6). m: marker 50 bp; M: marker 100 bp; the additional bands
detected on polyacrylamide gel were indicated with arrowheads.
tree was selected for testing its virulence against honeybee and 75 ± 1.5 mortalities, resp.). No significant differences
larvae (Figure 5). All the three isolates, BMG 93, BMG were observed between the three isolates based on the t-test
184, and BMG 259, determined high mortality rates at 5 × (for the treatment with 5 × 103 CFU mL−1 , BMG 93 versus
103 CFU mL−1 (50.3 ± 2.05%, 47.33 ± 3.5%, and 49 ± 2.6% BMG 184, 𝑃 = 0.29; BMG 93 versus BMG 259, 𝑃 = 0.56;
mortalities, resp.) and at 105 CFU mL−1 (79 ± 3.8%, 73 ± 1%, BMG 184 versus BMG 259, 𝑃 = 0.54; for the treatment with
Table 2: Identification of six distinct BOX genotypes for seventy-five isolates of P. larvae, based on the combination of bands size and number
on polyacrylamide gel.
Genotypes
Bands (bp) Total
G1 G2 G3 G4 G5 G6
200 − − + − − −
280 + + + + + +
300 + + + + + +
350 − + + + + −
400 + − − − + −
450 + − − − − −
500 + − − − − −
650 + + + + + +
700 + + + + + +
800 + + + + + +
1000 − − + + + −
Number of strains 1 38 2 20 5 9 75
+: presence of band; −: absence of band.
Cumulative larval mortality (%)
90 isolates in the collection, presented typical characteristics of P.

80 larvae [27] being Gram, casein, and gelatin positive, catalase,
70 oxidase and starch negative, capable of using citrate, reducing
60
50
nitrates to nitrites, and acidifying the medium from glucose
40 without gas and H2 S production and incapable of growing in
30 media containing 5% NaCl or in nutrient broth. The other
20 branches (12% of the collection) presented variability in four
10 tests: nitrates reduction, methyl red test, and oxidase and
0 indol production. The isolates in branch 1 (BMG 191, BMG
5 × 103 105
−1
232, BMG 250, and BMG 257) were oxidase positive while
Treatment (CFU mL ) isolate BMG 198 in branch 2 was double positive for oxidase
BMG 93 and methyl red. The positive response of P. larvae to methyl
BMG 184 red and oxidase was not described previously. Isolates BMG
BMG 259 192 and BMG 194 in branch 4 were able to produce indol,
Figure 5: Larval mortality rate after exposure to the pathogen P. and isolates BMG 184 and BMG 189 in branch 5 contained
larvae. Graphical representation of cumulative mortality percentage isolates unable to reduce nitrates. These results obtained with
of larvae (± SD), fed with artificial diet supplemented with P. larvae isolates retrieved from a relatively small area of northern
spores at 5 × 103 CFU mL−1 or 105 CFU mL−1 , during 8 days. In Y- Tunisia show that P. larvae is not a monoclonal species like
axis the mortality percentage of larvae reported, and in X axis the several other pathogens supporting previous observations
different P. larvae strains used of larval infection tested at the two [4, 14] of a certain phenotypic variability highlighted in the
spore concentrations are reported. former subspecies P. larvae subsp. larvae and P. larvae subsp.
pulvifaciens.
16S rRNA gene sequencing confirmed the assignment of
105 CFU mL−1 , BMG 93 versus BMG 184, 𝑃 = 0.053; BMG all the strains to P. larvae but highlighted certain sequence
93 versus BMG 259, 𝑃 = 0.09; BMG 184 versus BMG 259, variability among the isolates confirming the lack of a strict
𝑃 = 0.18). The mortality rate of the uninfected control group clonality in the species according to the biochemical study.
was less than 20% in all the three experiments. However, it was not possible to identify a clear correspon-
dence in the isolate grouping between the phenotypes and the
4. Discussion 16S rRNA gene sequence variability.
A relative intraspecific diversity within the 75 Tunisian
In the dendrogram resuming the P. larvae isolates relation- isolates was further confirmed by BOX-PCR typing which
ships according to the biochemical features (Figure 2), five allowed the distinction of P. larvae from the related Paeni-
branches corresponding to five biochemical phenotypes (P1 bacillus species. In addition, BOX profiles showed polymor-
to P5) could be distinguished. This clustering was based on phic bands specific for P. larvae that could be useful for its
the detected polymorphism in several biochemical properties identification as in the case of other pathogenic bacilli like
(nitrate reduction, oxidase production, and indol and methyl B. anthracis [29]. Using BOX-PCR, an unexpected genetic
red tests). The isolates in branch 3, representing 88% of the variability was revealed for isolates derived from a relatively
small region northern Tunisia. Alippi and Aguilar [12], by the Tunisian Ministry of Higher Education and Scientific
typing by BOX-PCR a collection of 100 P. larvae originating Research in the ambit of the Laboratory Projects LR MBA206
from a geographic area much larger than northern Tunisia, and LR11ES31. They are grateful to Dr. Ziegler from BGSC for
detected only three genotypes. BOX-PCR combined to REP- providing them with the reference strains of Paenibacillus.
PCR revealed four genotypes within a collection of 105 strains
of P. larvae isolated from Germany [13, 14]. Similarly, within
References
a collection of 214 P. larvae isolates from Austria only five
genotypes were identified by PCR typing using BOX A1R [1] E. Genersch, E. Forsgren, J. Pentikäinen et al., “Reclassification
and MBO REP1 primers [15]. The results obtained with the of Paenibacillus larvae subsp. pulvifaciens and Paenibacillus
present Tunisian isolate collection suggest that the genetic larvae subsp. larvae as Paenibacillus larvae without subspecies
and phenotypic variability of P. larvae can be larger than that differentiation,” International Journal of Systematic and Evolu-
previously estimated. tionary Microbiology, vol. 56, no. 3, pp. 501–511, 2006.
However, despite the combination of the three [2] A. M. Alippi, A. C. López, F. J. Reynaldi, D. H. Grasso, and O. M.
approaches, the biochemical, phylogenetic, and molecular Aguilar, “Evidence for plasmid-mediated tetracycline resistance
typing methods highlighted a relatively high intraspecific in Paenibacillus larvae, the causal agent of American Foulbrood
diversity of the Tunisian P. larvae collection; a clear (AFB) disease in honeybees,” Veterinary Microbiology, vol. 125,
no. 3-4, pp. 290–303, 2007.
correlation and grouping of the isolates according to the
three methods was not evidenced. This may indicate slightly [3] A. M. Alippi, G. N. Albo, F. J. Reynaldi, and M. R. De Giusti,
distinct evolutionary pathways within the species that “In vitro and in vivo susceptibility of the honeybee bacterial
pathogen Paenibacillus larvae subs P. larvae to the antibiotic
apparently remain neutral and not yet clearly evident in
tylosin,” Veterinary Microbiology, vol. 109, no. 1-2, pp. 47–55,
distinct coherent phenotypes. 2005.
The attempt to search for a possible effect of the different
[4] M. Heyndrickx, K. Vandemeulebroecke, B. Hoste et al.,
observed phenotypes/genotypes on the level of virulence “Reclassification of Paenibacillus (formerly Bacillus) pulvifa-
supports the considerations that the observed differences ciens (Nakamura 1984) Ash et al. 1994, a later subjective
have no apparent effects on the pathogenicity against the synonym of Paenibacillus (formerly Bacillus) larvae (White
honeybee larvae, at least in the conditions adopted in the 1906) Ash et al. 1994, as a subspecies of P. larvae, with emended
study to test the virulence. Our results showed that the three descriptions of P. larvae as P. larvae subs P. larvae and P.
tested isolates of P. larvae, BMG 93, BMG 184, and BMG 259 larvae subsp. pulvifaciens,” International Journal of Systematic
representing three 16S rRNA gene phylotypes and two BOX- Bacteriology, vol. 46, no. 1, pp. 270–279, 1996.
PCR genotypes, presented the same virulence level against [5] W. Dobbelaere, D. C. De Graaf, W. Reybroeck, E. Desmedt, J.
honeybee larvae. Such lack of correlation could be due to the E. Peeters, and F. J. Jacobs, “Disinfection of wooden structures
procedure adopted, and we cannot exclude that, for instance, contaminated with Paenibacillus larvae subs P. larvae spores,”
the low number of isolates tested in the virulence assays or the Journal of Applied Microbiology, vol. 91, no. 2, pp. 212–216, 2001.
limited period (8 days) for observing the mortality may have [6] K. Antúnez, B. D’Alessandro, C. Piccini, E. Corbella, and P.
prevented the observation of virulence differences among the Zunino, “Paenibacillus larvae larvae spores in honey samples
different Tunisian genotypes of P. larvae. Also, we do not from Uruguay: a nationwide survey,” Journal of Invertebrate
know the ERIC type of our isolates since all the PCR attempts Pathology, vol. 86, no. 1-2, pp. 56–58, 2004.
to get clear fingerprints with the Tunisian isolates failed. For [7] J. Kilwinski, M. Peters, A. Ashiralieva, and E. Genersch,
instance, we could be in presence of a collection of isolates “Proposal to reclassify Paenibacillus larvae subsp. pulvifaciens
representing a single ERIC type and hence a single virulence DSM 3615 (ATCC 49843) as Paenibacillus larvae subs P. larvae.
Results of a comparative biochemical and genetic study,” Veteri-
type [24]. Similarly we cannot exclude that in the beehive the
nary Microbiology, vol. 104, no. 1-2, pp. 31–42, 2004.
virulence behavior of the Tunisian isolates may vary [43].
[8] C. Hamdi, A. Balloi, J. Essanaa et al., “Gut microbiome dysbiosis
and honeybee health,” Journal of Applied Entomology, vol. 135,
5. Conclusion no. 7, pp. 524–533, 2011.
By keeping in mind all the above considerations related to [9] E. Crotti, A. Balloi, C. Hamdi et al., “Microbial symbionts:
a resource for the management of insect-related problems,”
the limitations of the adopted experimental conditions, the
Microbial Biotechnology, vol. 5, no. 3, pp. 307–317, 2012.
present data indicate a relatively high biological variability
[10] D. C. De Graaf, P. De Vos, M. Heyndrickx, S. Van Trappen, N.
of P. larvae in northern Tunisia and suggest that the variable
Peiren, and F. J. Jacobs, “Identification of Paenibacillus larvae to
phenotypic and genotypic traits observed in the isolate collec- the subspecies level: an obstacle for AFB diagnosis,” Journal of
tion apparently have a neutral effect in relation to virulence or Invertebrate Pathology, vol. 91, no. 2, pp. 115–123, 2006.
affect other ecological aspects of P. larvae nondetectable with
[11] T. S. Thompson, S. F. Pernal, D. K. Noot, A. P. Melathopoulos,
the experimental approaches used here. and J. P. van den Heever, “Degradation of incurred tylosin
to desmycosin-Implications for residue analysis of honey,”
Acknowledgments Analytica Chimica Acta, vol. 586, no. 1-2, pp. 304–311, 2007.
[12] A. M. Alippi and O. M. Aguilar, “Characterization of isolates
The authors acknowledge financial support from the Euro- of Paenibacillus larvae subsp. larvae from diverse geographical
pean Union in the ambit of Project BIODESERT (EU FP7- origin by the polymerase chain reaction and BOX primers,”
CSA-SA REGPOT-2008-2, Grant Agreement no. 245746) and Journal of Invertebrate Pathology, vol. 72, no. 1, pp. 21–27, 1998.
[13] E. Genersch and C. Otten, “The use of repetitive element PCR [29] A. Cherif, S. Borin, A. Rizzi, H. Ouzari, A. Boudabous,
fingerprinting (rep-PCR) for genetic subtyping of German field and D. Daffonchio, “Characterization of a repetitive ele-
isolates of Paenibacillus larvae subs P. larvae,” Apidologie, vol. ment polymorphism-polymerase chain reaction chromoso-
34, no. 3, pp. 195–206, 2003. mal marker that discriminates Bacillus anthracis from related
[14] S. Neuendorf, K. Hedtke, G. Tangen, and E. Genersch, “Bio- species,” Journal of Applied Microbiology, vol. 93, no. 3, pp. 456–
chemical characterization of different genotypes of Paenibacil- 462, 2002.
lus larvae subs P. larvae, a honey bee bacterial pathogen,” [30] B. Chouaia, P. Rossi, M. Montagna et al., “Molecular evidence
Microbiology, vol. 150, no. 7, pp. 2381–2390, 2004. for multiple infections as revealed by typing of asaia bacterial
symbionts of four mosquito species,” Applied and Environmen-
[15] I. Loncaric, I. Derakhshifar, J. T. Oberlerchner, H. Köglberger,
tal Microbiology, vol. 76, no. 22, pp. 7444–7450, 2010.
and R. Moosbeckhofer, “Genetic diversity among isolates of
Paenibacillus larvae from Austria,” Journal of Invertebrate [31] C. Velezmoro, E. Ramos, C. Garcı́a, and D. Zúñiga, “Genotypic
Pathology, vol. 100, no. 1, pp. 44–46, 2009. identification of Bacillus sp. isolated from canned white aspara-
gus (Asparagus officinalis) during the production/processing
[16] J. D. Evans, “Transcriptional immune responses by honey bee chain in northern Peru,” Annals of Microbiology, vol. 62, no. 3,
larvae during invasion by the bacterial pathogen, Paenibacillus pp. 1207–1217, 2012.
larvae,” Journal of Invertebrate Pathology, vol. 85, no. 2, pp. 105–
[32] W. L. Kovach, MVSP-A Multivariate Statistical Package for
111, 2004.
Windows, ver. 3.1. Kovach, Computing Services, Pentraeth,
[17] P. H. De Rycke, J. J. Joubert, S. H. Hosseinian, and F. J. Jacobs, Wales, UK, 1998.
“The possible role of Varroa destructor in the spreading of [33] S. F. Altschul, W. Gish, W. Miller, E. W. Myers, and D. J. Lipman,
American foulbrood among apiaries,” Experimental and Applied “Basic local alignment search tool,” Journal of Molecular Biology,
Acarology, vol. 27, no. 4, pp. 313–318, 2002. vol. 215, no. 3, pp. 403–410, 1990.
[18] M. O. Schäfer, W. Ritter, J. Pettis, and P. Neumann, “Small hive [34] J. D. Thompson, D. G. Higgins, and T. J. Gibson, “CLUSTAL
beetles, Aethina tumida, are vectors of Paenibacillus larvae,” W: improving the sensitivity of progressive multiple sequence
Apidologie, vol. 41, no. 1, pp. 14–20, 2010. alignment through sequence weighting, position-specific gap
[19] E. W. Davidson, “Ultrastructure of American foulbrood disease penalties and weight matrix choice,” Nucleic Acids Research, vol.
pathogenesis in larvae of the worker honey bee, Apis mellifera,” 22, no. 22, pp. 4673–4680, 1994.
Journal of Invertebrate Pathology, vol. 21, no. 1, pp. 53–61, 1973. [35] T. H. Jukes and C. R. Cantor, “Evolution of protein molecules,”
[20] D. Yue, M. Nordhoff, L. H. Wieler, and E. Genersch, “Fluores- in Mammalian Protein Metabolism, N. Munro, Ed., pp. 21–132,
cence in situ hybridization (FISH) analysis of the interactions Academic Press, New York, NY, USA, 1969.
between honeybee larvae and Paenibacillus larvae, the causative [36] N. Saitou and M. Nei, “The neighbor-joining method: a new
agent of American foulbrood of honeybees (Apis mellifera),” method for reconstructing phylogenetic trees,” Molecular Biol-
Environmental Microbiology, vol. 10, no. 6, pp. 1612–1620, 2008. ogy and Evolution, vol. 4, no. 4, pp. 406–425, 1987.
[21] B. N. Dancer and P. Chantawannakul, “The proteases of Ameri- [37] K. Tamura, J. Dudley, M. Nei, and S. Kumar, “MEGA4: molec-
can Foulbrood scales,” Journal of Invertebrate Pathology, vol. 70, ular evolutionary genetics analysis (MEGA) software version
no. 2, pp. 79–87, 1997. 4.0,” Molecular Biology and Evolution, vol. 24, no. 8, pp. 1596–
1599, 2007.
[22] K. Antúnez, M. Anido, D. Arredondo, J. D. Evans, and P.
Zunino, “Paenibacillus larvae enolase as a virulence factor in [38] B. Ettoumi, N. Raddadispi, S. Borin, D. Daffonchio, A. Boud-
honeybee larvae infection,” Veterinary Microbiology, vol. 147, no. abous, and A. Cherif, “Diversity and phylogeny of culturable
1-2, pp. 83–89, 2011. spore-forming Bacilli isolated from marine sediments,” Journal
of Basic Microbiology, vol. 49, no. 1, pp. S13–S23, 2009.
[23] L. Poppinga, B. Janesch, A. Funfhaus et al., “Identification and
[39] E. Forsgren, T. C. Olofsson, A. Vásquez, and I. Fries, “Novel
functional analysis of the S-layer protein SplA of Paenibacillus
lactic acid bacteria inhibiting Paenibacillus larvae in honey bee
larvae, the causative agent of American Foulbrood of honey
larvae,” Apidologie, vol. 41, no. 1, pp. 99–108, 2010.
bees,” PLOS Pathogens, vol. 8, no. 5, Article ID e1002716, 2012.
[40] Y.-S. C. Peng, E. Mussen, A. Fong, M. A. Montague, and T. Tyler,
[24] E. Genersch, A. Ashiralieva, and I. Fries, “Strain- and genotype- “Effects of chlortetracycline of honey bee worker larvae reared
specific differences in virulence of Paenibacillus larvae subs in vitro,” Journal of Invertebrate Pathology, vol. 60, no. 2, pp. 127–
P. larvae, a bacterial pathogen causing American foulbrood 133, 1992.
disease in honeybees,” Applied and Environmental Microbiology,
vol. 71, no. 11, pp. 7551–7555, 2005. [41] P. Aupinel, D. Fortini, H. Dufour et al., “Improvement of
artificial feeding in a standard in vitro method for rearing Apis
[25] M. H. Hussein, “A review of beekeeping in Arab countries,” Bee mellifera larvae,” Bulletin of Insectology, vol. 58, no. 2, pp. 107–111,
World, vol. 81, no. 2, pp. 56–71, 2000. 2005.
[26] N. Gallai, J.-M. Salles, J. Settele, and B. E. Vaissière, “Economic [42] N. Millar, “Biology statistics made simple using excel,” School
valuation of the vulnerability of world agriculture confronted Science Review, vol. 83, no. 303, pp. 23–24, 2001.
with pollinator decline,” Ecological Economics, vol. 68, no. 3, pp. [43] S. Rauch, A. Ashiralieva, K. Hedtke, and E. Genersch, “Nega-
810–821, 2009. tive correlation between individual-insect-level virulence and
[27] R. E. Gordon, W. C. Haynes, and C. H. N. Pang, The Genus colony-level virulence of Paenibacillus larvae, the etiological
Bacillus, Handbook No. 427, United States Department of agent of american foulbrood of honeybees,” Applied and Envi-
Agriculture, Washington, DC, USA, 1973. ronmental Microbiology, vol. 75, no. 10, pp. 3344–3347, 2009.
[28] P. H. A. Sneath and R. R. Sokal, Numerical Taxonomy: The
Principles and Practice of Numerical Classification, Freeman
WH and Co., San Francisco, Calif, USA, 1973.
http://dx.doi.org/10.1155/2013/420287
Research Article
Characterization of the Bacterial Community Associated with
Larvae and Adults of Anoplophora chinensis Collected in Italy
by Culture and Culture-Independent Methods
Aurora Rizzi,1 Elena Crotti,1 Luigimaria Borruso,1,2 Costanza Jucker,1 Daniela Lupi,1
Mario Colombo,1 and Daniele Daffonchio1
1
Department of Food, Environmental and Nutritional Sciences (DEFENS), University of Milan, Via Celoria 2, 20133 Milan, Italy
2
Faculty of Science and Technology, Free University of Bozen-Bolzano, Piazza Università 5, 39100 Bolzano, Italy
Correspondence should be addressed to Daniele Daffonchio; daniele.daffonchio@unimi.it
Received 16 April 2013; Accepted 9 July 2013
Copyright © 2013 Aurora Rizzi et al. This is an open access article distributed under the Creative Commons Attribution License,
The wood-boring beetle Anoplophora chinensis Forster, native to China, has recently spread to North America and Europe causing
serious damage to ornamental and forest trees. The gut microbial community associated with these xylophagous beetles is of
interest for potential biotechnological applications in lignocellulose degradation and development of pest-control measures. In
this study the gut bacterial community of larvae and adults of A. chinensis, collected from different host trees in North Italy,
was investigated by both culture and culture-independent methods. Larvae and adults harboured a moderately diverse bacterial
community, dominated by Proteobacteria, Actinobacteria, and Firmicutes. The gammaproteobacterial family Enterobacteriaceae
(genera Gibbsiella, Enterobacter, Raoultella, and Klebsiella) was the best represented. The abundance of such bacteria in the insect
gut is likely due to the various metabolic abilities of Enterobacteriaceae, including fermentation of carbohydrates derived from
lignocellulose degradation and contribution to nitrogen intake by nitrogen-fixing activity. In addition, bacteria previously shown
to have some lignocellulose-degrading activity were detected at a relatively low level in the gut. These bacteria possibly act
synergistically with endogenous and fungal enzymes in lignocellulose breakdown. The detection of actinobacterial symbionts could
be explained by a possible role in the detoxification of secondary plant metabolites and/or protection against pathogens.
1. Introduction symbiotic microorganisms are typically vertically transmitted

during early stages of oogenesis or embryogenesis, whereas
Insects have complex associations with a wide variety of facultative symbionts can colonize native hosts through hor-
microorganisms. Many bacteria contribute to various phys- izontal transmission between individuals or acquisition from
iological functions, including nutrition, development, repro- the diet or the environment [7–11]. All these properties and
duction, resistance to pathogens, production of pheromones, the important roles that symbionts have in host biology have
and immunity [1]. Some symbionts can play essential roles been proposed for exploitation in novel control strategies of
in the insect gut, compensating for diets deficient in certain insect pests or for the management of insect-related problems
nutrients or containing recalcitrant organic compounds. For [12–14].
instance, in xylophagous termites the gut microflora enables The longhorned beetles (Coleoptera: Cerambycidae) are
the host to digest cellulose and fix atmospheric nitrogen xylophagous insects which feed on healthy or dead woody
[2, 3], and in phytophagous aphids the endocellular symbiont plants causing damage of forest and ornamental trees.
Buchnera aphidicola synthesizes essential amino acids that are Many beetles establish a strict association with fungi that
absent in phloem sap [4, 5]. These gut-microbe interactions naturally colonize their galleries and provide nutrients by
are diverse and include antagonism, commensalism, and lignocellulose degradation and synthesis of other essentials
mutualism and range from obligate to facultative [6]. Obligate compounds. The genus Anoplophora includes xylophagous
longhorned beetles, native to eastern Asia, that live on numer- Table 1: Anoplophora chinensis collection and detection strategies
ous woody plant species. Since its accidental introduction used in this study.
through wood-packing materials and live plants from Asia,
Detection
it has become an important invasive pest both in Europe Host tree Insects (no.)
strategies
and North America. In the United States, the species A.
Isolation and
glabripennis is spread, whereas in Europe the species A.
Alnus Larvae (9) library clones
chinensis, form malasiaca, is mostly present [15, 16]. The (pool of 2 guts)
lifecycle of A. chinensis lasts 12–24 months, and larvae
Liquidambar Larvae (8) DGGE
develop by feeding on cambium, phloem, and subsequently
xylem, forming tunnels into the inner bark of the tree and Salix caprea Larvae (10) DGGE
causing death of the host. Oviposition occurs in the bark of Acer Isolation and
Adults (3)
the host tree, and eggs, larvae, and pupae can overwinter. In saccharinum DGGE
late spring adults emerge and feed on the bark of tender twigs. Isolation, library
Due to their specific diet, comprising highly lignified low- clones (pool of 2
Alnus Adults (2)
nitrogen wood tissues, gut symbionts may play important guts), and
roles in the digestive tract of these xylophagous insects, DGGE
contributing to lignocellulose degradation and synthesis of
essential amino acids or vitamins [17]. Studies conducted
on larvae of A. glabripennis collected in USA and China tubes with 500 𝜇L of saline, and homogenized using a
documented the wide diversity of bacterial taxa harboured sterile plastic pestle. Homogenates were used for culture-
in larval guts [18, 19]. The bacterial communities of animals independent methods and stored at −20∘ C until use.
reared on different host trees were extremely variable, with a
significant impact on cellulase activity [18]. Larval guts of A.
2.2. Bacteria Isolation. The gut homogenates were 10-fold
glabripennis were also found to be associated with the soft-
diluted and directly plated on tryptic soy agar (TSA) and
rot fungus Fusarium solani, capable of degrading proteins,
1/10 strength TSA (Difco, Milan, Italy). Fifty 𝜇L of gut
cellulose, hemicelluloses, and other woody carbohydrate
homogenates were also used for the enrichment of nitrogen-
polymers [20]. However, the recent discovery of an endoge-
fixing bacteria in LGI liquid medium (5% sucrose, 0.06%
nous exocellulase from A. malasiaca [21] raises the question
KH2 PO4 , 0.02% K2 HPO4 , 0.02% MgSO4 , 0.002% CaCl2 ,
of the contribution of gut microorganisms to lignocellulose
0.001% FeCl3 , and 0.0002% NaMoO4 , pH 6 [22]). After
degradation and, more extensively, their contribution to the
growth, the enriched cultures were plated on LGI agar
beetle’s physiology and biochemistry. Further research to
plates containing 20 g/L noble agar (Difco). All media were
characterize the microbial communities of related species,
supplemented with 100 𝜇g/mL cycloheximide. Plates were
investigating the variation in communities in relation to
incubated for 3–5 days at 30∘ C. The colonies obtained by
geography and/or different life stages, could contribute to a
plating were differentiated based on morphological features
better understanding of the complex symbiotic relationships
including shape, colour, margins, elevation, and texture. Two
of beetles with microorganisms and the impact of microor-
or more isolates representative of each colony morphology
ganisms on the host lifecycle.
were transferred to fresh agar plates, and pure colonies were
The aim of this study was to investigate the bacterial
stored at −80∘ C in 15% glycerol.
community associated with both larvae and adults of A.
chinensis, collected in Italy, using both culture-dependent
and independent methods, namely, PCR-DGGE (denaturant 2.3. DNA Isolation, PCR, and Cloning. Total DNA from
gradient gel electrophoresis) and clone library analysis. dissected organs was isolated as previously reported [23].
DNA was extracted by enzymatic and chemical treatment and
purified using the Wizard DNA purification resin (Promega,
2. Materials and Methods Milan, Italy). PCR amplification of the 16S rRNA gene from
bacterial isolates was performed using the universal primers
2.1. Insect Collection and Dissection. Larvae and adults of 27F and 1492R [24]. The reaction mixture (50 𝜇L) contained
A. chinensis were collected from April to November 2008 1× PCR buffer, 1.5 mM MgCl2 , 0.5 𝜇M of each primer, 0.2 mM
at different sites within the infested area in Lombardy, Italy of dNTPs, and 1.5 U of Taq DNA polymerase. The DNA
(Table 1). After collection, larvae and adults were maintained template was obtained by transferring a small portion of a
separately in sterile containers at 10∘ C and processed the pure colony into a PCR tube. The thermal cycling program
following day. The insects were surface disinfected with 60% consisted of 5 min at 95∘ C, followed by 30 cycles of 45 s at
ethanol and rinsed twice in sterile water. Each larva was 95∘ C, 1 min at 55∘ C, and 1 min at 72∘ C, with a final extension
dissected near a Bunsen burner using sterilized dissection of 10 min at 72∘ C.
scissors, and the entire gut was extracted from the insect Two 16 s rRNA gene libraries, one from larvae and one
body. The same procedure was followed for adult individuals; from adults, were constructed using two pooled guts per
in addition, male gonads, and eggs inside the female abdomen each library (Table 2). The DNA isolated from the pooled
were also extracted. The individual guts, gonads and eggs guts was amplified using the primer pair 27F and 1492R,
were washed in 4 mL of sterile water, transferred to 1.5 mL as previously described. The resulting 1.5 kb fragments were
Table 2: Bacterial taxa identified in larvae and adults of Anoplophora chinensis by culture and culture-independent methods.
Larval gut Adult gut Testicles and eggsd
Phylum Class Family Genusa Detection strategy Detection strategy Detection strategy
Isolation Library DGGEb Isolation Library DGGEb Isolation DGGEb
(al)c (al)c (li, sa)c (ac, al)c (al)c (ac, al)c (ac, al)c (ac, al)c
Caulobacteraceae Brevundimonas 1 1 ac ac
Alfaproteobacteria Sphingomonas ac
Sphingomonadaceae
Novosphingobium 1
Phyllobacteriaceae Unclassified 1
Methylobacteriaceae Methylobacterium li ac
Comamonadaceae Comamonas 1T ac
Betaproteobacteria Neisseriaceae Neisseriae 2
Massilia li ac
Oxalobacteraceae
Ralstonia li ac
Proteobacteria 3T ac, E ac, T
Enterobacter 4 3 ac, 6 al∗ 80 ac, al
3E ac ac, al
1T ac, E ac, T
Klebsiella 8 ac, 2 al ac, al
4T al ac, al
Enterobacteriaceae E ac, T
Raoultella 1 sa 5 ac, 1 al 58 ac, al 3T al
ac, al
Gammaproteobacteria Gibbsiella 7 58 li, sa
Rahnella li
Erwinia 2 ac 1
Buttiauxella T ac
Unclassified 3 16 li, sa 2 ac 2 ac 3T ac T ac
1T al, 1E
Stenotrophomonas 1
Xanthomonadaceae ac
Dyella 1
Sinobacteraceae Nevskia ac
Pseudomonadaceae Pseudomonas 1 sa 1 1E ac
Moraxellaceae Acinetobacter 2 1 ac 3
Microbacterium 1 3 li, sa 2 ac 1 ac
Microbacteriaceae
Curtobacterium 1T ac
Dermabacteraceae Brachybacterium 1
Actinobacteria Actinobacteria Brevibacteriaceae Brevibacterium 1 ac, 1 al 1
Propionibacteriaceae Propionibacterium 1
Rothia 1
Micrococcaceae
Kocuria 1
Tsukamurellaceae Tsukamurella 2 ac 1T ac
3
4
Table 2: Continued.
Larval gut Adult gut Testicles and eggsd
a
Phylum Class Family Genus Detection strategy Detection strategy Detection strategy
Isolation Library DGGEb Isolation Library DGGEb Isolation DGGEb
(al)c (al)c (li, sa)c (ac, al)c (al)c (ac, al)c (ac, al)c (ac, al)c
Bacillaceae Bacillus 3 1 2
Paenibacillaceae Paenibacillus 1 ac
Firmicutes Bacilli Enterococcaceae Enterococcus li 1T al
Streptococcaceae Streptococcus 2
Staphylococcaceae Staphylococcus 1 1E ac
Planococcaceae Lysinibacillus 1E ac
Flavobacteriia Flavobacteriaceae Chryseobacterium 2
Bacteroidetes
Sphingobacteriia Chitinophagaceae Chitinophaga T ac
a
Identification based on NCBI (95% limit) and RDP Classifier (80% confidence threshold) results is given, when possible, at genus level.
b
Individuals positive for the presence of the specific band in the DGGE analysis.
c
The host tree from which insects were sampled are indicated in brackets: ac: Acer; al: Alnus; li: Liquidambar; sa: Salix.
d
T: testicles; E: immature eggs.
∗
1 out of 3 ac, and 3 out of 6 al were recovered on LGI medium.
cloned into pCRII-TOPO vector (Invitrogen Life Technolo- In addition, PCR-DGGE analysis was used to further
gies, Milan, Italy) following the manufacturer’s protocol. investigate the dominant microbial species of multiple
Individual colonies were picked up using sterile pipette tips individuals (Figure 2). The gut bacterial profiles obtained
and used directly for PCR amplification. Insert DNA from 16S from eighteen larvae grown in Liquidambar and Salix trees
rRNA clones was amplified by standard PCR amplification differed markedly but were highly similar to larvae collected
using the primers M13F/M13R [25] and sequenced. from the same site/tree (Figure 2). Sequences of dominant
intense bands showing tight affiliationto Raoultella (bands 14,
15, 18–21, 23, 24) and unclassified Enterobacteriaceae (band 17)
2.4. PCR-DGGE Analysis. Bacterial 16S rRNA gene frag- were detected in larvae from Salix, whereas sequences of faint
ments were amplified by PCR using the primer pair GC-357- bands, also related to unclassified Enterobacteriaceae (band
F/907-R [26–28]. PCR reactions were performed as previ- 4), were found in larvae from Liquidambar. Gibbsiella (bands
ously described [29]. Briefly, PCR products (approx. 300 ng) 12 and 16) and Rahnella (band 11) were occasionally detected,
were loaded onto 7% (w/v) polyacrylamide gels (0.75 mm) independently of the host tree.
with a denaturant gradient of 40–60% (100% denaturant The microbial communities of guts of five adults (four
contained 7 M urea and 40% formamide). Electrophoresis males and one female) fed on Alnus or Acer were analyzed
was run in 1× TAE buffer using a D-Code electrophoresis by both culturing and culture-independent methods. The
system (BioRad, Milan, Italy) at 90 V and 60∘ C for 17 h. Gels sequences of 16S rRNA genes from 38 isolates, 151 clones, and
were stained with SYBR Green I Nucleic Acid Gel Stain (Invit- 28 DGGE bands were obtained (Tables 2 and 3). Overall, the
rogen Life Technologies) and documented with GelDoc 2000 results of the different analyses indicated that the Enterobac-
apparatus (BioRad) using the Diversity Database software teriaceae were the dominant bacteria also in the adult gut.
(BioRad). Relevant DNA bands were excised from the gels In particular, DGGE analysis, in accordance with culturing,
and eluted in 50 𝜇L of Tris-HCl 10 mM. Five microlitres of indicated that Enterobacter (bands 29, 32, 33, 46) was detected
DNA was used for 16S DNA fragment reamplification using in all five individuals tested, whereas Klebsiella (25–28, 42–
nonclamped primers and the obtained amplicons sequenced. 45) and Raoultella (47–51) were found only in some indi-
viduals (3 and 1 out of 5 individuals, resp.). One adult indi-
2.5. Sequencing and Data Analysis. Sequencing of the 16S vidual presented Enterobacter (bands 32 and 33) (Figure 2)
rRNA gene fragments was performed using the primer 27F at and microorganisms strictly affiliated to the genus Erwinia
Primm (Milan, Italy). Partial sequences from clones and bac- (Figure 1). Microorganisms affiliated to Enterobacter were
terial isolates were compared against the National Center for also identified when performing enrichment of nitrogen-
Biotechnology Information (NCBI) genomic database with fixing bacteria. Data from library cloning performed on bee-
the BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi) search tles fed on Alnus (pool of two guts) indicated an abundance of
alignment tool. A collection of phylogenetically related Enterobacter and Raoultella in the gut microbial community,
sequences was obtained from the NCBI database. Sequences with percentages over the total sequenced clones of 52% (𝑛 =
from clones were taxonomically classified by the RDP-II 80) and 38% (𝑛 = 58), respectively. This result is consistent
Naive Bayesian Classifier (http://rdp.cme.msu.edu/classifier/ with the high intensities of the bands relative to these bacteria
classifier.jsp) using an 80% confidence threshold. Sequence in DGGE gels. However, no library clones related to the genus
alignment was carried out and phylogenetic trees constructed Klebsiella were detected, maybe due to differences in PCR
using MEGA software, version 5.1 [30]. The trees were amplificability of DNA extracted from the two guts and/or to
constructed using the maximum likelihood algorithm and the different amount of template DNA in PCR reactions due
Tamura Nei parameter correction and were bootstrapped to the different sizes of adult guts.
1000 times. All the analytical methods used revealed a rather diverse
community generally characterized by the dominance of
Enterobacteriaceae in both larvae and adult stages and
3. Results the occurrence of several species encompassing different
taxa (Proteobacteria, Actinobacteria, Firmicutes, and Bac-
3.1. Bacterial Community in Larval and Adult Guts. Guts of teroidetes). In particular, some other Gammaproteobacteria,
larvae fed on Alnus were investigated by both culturing such as Acinetobacter and Pseudomonas, were detected in
(seven individual guts) and library clones (pool of two larval and adult guts by culture-independent analyses. Mem-
guts). The sequences of bacterial 16S rRNA genes from 23 bers of Alpha- and Betaproteobacteria groups were found
isolates and 91 clones were obtained. Most isolates were by DGGE and isolation methods in both larvae and adults.
strictly affiliated to the Gibbsiella genus (Figure 1 and Table 2). Interestingly, Ralstonia, Massilia, and Methylobacterium were
Similarly, using library cloning, the majority of gut-derived found in all larval individuals that fed on Liquidambar. It
clones were represented by the Gibbsiella genus (𝑛 = 58, 63%) can be speculated that the aromatic resin produced from this
and bacteria strictly affiliated to Gibbsiella and uncultured host tree species had an impact on the microbial composition
clones previously identified in larval guts from other wood- of the larval gut communities. The gut bacterial micro-
boring beetles (Agrilus planipennis, Saperda vestita, and biome of Anoplophora comprised additional representatives
Apriona germari). The genera Enterobacter and Raoultella of Actinobacteria and Firmicutes. In particular, the genera
were represented in low proportions (approximately 5 and 1%, Microbacterium and Bacillus were detected in both larvae
resp.). and adults using the majority of methods. Some species were
100
Enterobacteriaceae
54 90 3.4.Ea
97 Z76667 Pseudomonas putida type strain DSM 291
100 AF468452 Pseudomonas koreensis type strain Ps 9-14 Gammaproteobacteria
5L, 146A
97 3.11.a
80 X81661 Acinetobacter calcoaceticus, type strain DSM 30006
92 3L, 143A, 144A, 145A
100 AF509825 Acinetobacter tjernbergiae type strain 7N16
4L
94
X81662 Acinetobacter haemolyticus type strain DSM 6962
100 1L, 2L
99 L06168 Neisseria flavescens type strain ATCC 13120 Betaproteobacteria
97 AJ430342 Comamonas terrigena type strain LMG 1253
100 3.7.Ta
100 7.6.l
AY884571 Dyella koreensis type strain BB4
89 3.19.Ea
86 72 EF540516 Stenotrophomonas sp. strain 4.O.7 Gammaproteobacteria
100 AB008509 Stenotrophomonas maltophilia type strain ATCC 13637
4.13.Ta
73
56 147A
100 3.2.l
EF029110 Novosphingobium resinovorum type strain NCIMB 8767
100 7.5.l
57
84 DQ335215 Brevundimonas terrae type strain KSL-145 Alphaproteobacteria
57 7.3.l
100 AY785323 Phyllobacterium leguminum type strain ORS 1419

46 AY770423 Bacterium from Saperda vestita larval gut
100 88L
CP003877 Propionibacterium acnes strain C1
2.18.a
100 X92981 Tsukamurella pulmonis type strain DSM 44142
1.43.a, 2.16.Ta
100
63 5.7.l, 83L, 84L, 85L, 150A
X77443 Microbacterium arborescens type strain DSM 20754
76 AJ698726 Microbacterium hydrocarbonoxydans type strain DSM 16089 Actinobacteria
55 99 1.3.a, 1.8.a
1.32.Ta
100 X77436 Curtobacterium citreum type strain DSM 16089
100 86L
53 98 M59055 Rothia dentocariosa type strain ATCC 17931
87L
99 Y16263 Kocuria palustris type strain DSM 11925
100 148A
AJ415376 Brachybacterium rhamnosum type strain LMG 19848
5.8.a, 149A
DQ344485 Brevibacterium samyangense type strain SST-8
100
2.4.a
73
99 X76565 Brevibacterium epidermidis type strain NCDO 2286
100 3.12.a
D78470 Paenibacillus glucanolyticus type strain DSM5162
100 3.7.Ea
FJ844477 Lysinibacillus sphaericus strain HytAP-B60
98 91L
100
AB363738 Bacillus simplex type strain NBRC 15720
1.9.l, 4.2.l
65
99 1.15.l
100 AB008315 Streptococcus infantis type strain GTC849
5.7.l
43 Bacilli
89 AF003928 Streptococcus sanguinis type strain ATCC 10556
4.3.Ta
99 AJ420804 Enterococcus casseliflavus type strain CECT969
3.1.l
99 AF302118 Bacillus sonorensis type strain NRRL B-23154
59A, 151A
99 4.5.l
D83361 Staphylococcus cohnii subsp. cohnii type strain ATCC 29974
100 3.8.Ea
69 AP008934 Staphylococcus saprophyticus subsp. sap. type strain ATCC 15305
NR074375 Pyrococcus furiosus strain DSM 3638
0.05
(a)
Figure 1: Continued.
Cluster 1-100A (19)

Cluster 2-120A (39)
83
AF129443 Raoultella planticola type strain ATCC 33531
1.12.a, 131.a, 129.a, 5.1.a, 5.3.a, 5.1.Ta
70
95
6L
4.16.Ta, 4.8.a
AF129441 Raoultella ornithinolytica strain ATCC 31898
2.2.Ta, 2.16.Ta, 3.6.Ta
AJ251468 Enterobacter aerogenes type strain NCTC10006
2.1.Ta, 2.12.Ta, 2.14.Ta
62
AM940417 Un bacterium clone from Rhagium inquisitor larval gut
DQ279642 Un bacterium clone from Saperda vestita larval gut
48
46
13L, 14L, 15L
42 AF129442 Raoultella terrigena type strain ATCC 33257
AM940411 Un bacterium clone from Rhagium inquisitor larval gut

3.3.l
7.4.l
3.5.Ea, 3.6.Ea
AY082447 Enterobacter sp. 253a
10L, 11L, 82L
1.5.a, 2.3.a, 4.2.a

AJ853891 Enterobacter ludwigii type strain EN-119
AJ251469 Enterobacter cloacae type strain ATCC 13047
46
Cluster 3-1A (14)
Cluster 4-80A (66)
2.1.a, 2.15.a, 2.20.a, 3.1.a, 5.12.a
40 AJ508302 Enterobacter hormaechei type strain CIP 103441
U78183 Klebsiella oxytoca
96 1.6.a, 1.9.a, 1.11.a, 1.16.a, 1.26.a, 1.27.a, 3.2.a , 3.13.a , 4.3.a, 4.6.a
1.11.Ta, 4.2.Ta, 4.5.Ta, 4.6.Ta, 4.11.Ta
45 2.22.a
2.14.a, 2.24.a
DQ163936 Un bacterium clone from ant lion Myrmeleon mobilis larval gut
80 2.5.1
GU562337 Gibbsiella quercinecans type strain FRB 97
Cluster 5-60L (23)
GU562340 Gibbsiella quercinecans strain FRB 185
56 1.2.1, 5.2.1, 6.1.1, 6.2.1, 6.3.1, 7.1.1, 7.2.1
100 43 AB566415 Gibbsiella dentisursi type strain NUM 1720

41L
46
EU560774 Un bacterium clone from Apriona germari larval gut
83 Cluster 6-38L (35)

1.4.1, 4.1.1, 4.3.1
EU153078 Bacterium from Agrilus planipennis larval gut

72 2.5.a
NR041976 Erwinia rhapontici strain DSM 4484
141A
61
86 3.1.a
EU681952 Erwinia persicina strain WD 1608
0.05
Larval gut isolate Testes isolate
Larval gut clone Eggs isolate
Adult gut isolate ( ) N. sequences
Adult gut clone
(b)
Figure 1: Phylogenetic tree of partial bacterial 16S rRNA sequences retrieved from culturing and clone library. Bacterial sequences fell mainly
into five classes (a) and most belonged to the Enterobacteriaceae family (b). The category of origin in which each species was identified is
indicated by symbols. Groups of sequences are compressed into clusters, and the number of sequences is provided in brackets. “Un” indicates
an uncultured bacterium. Numbers at nodes represent bootstrap values and are indicated when values were >40%. The scale bar represents
0.05 substitutions per nucleotide position.
Insect organ Larval guts Larval guts Adult Guts Eggs Testes
Plant host Liquidambar Salix Acer Alnus Alnus Alnus
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 2 3 4 5 1 2 3 4
60
6
7 56
57
8 22 61
9 18 34 47
1 19 25 35 48
20 26 36 42 49
50
62
2 21 23 37 43 51
14 24 52 53 63
10 11 15 27 44 58 64
54 59
28 32 45 65
3 4 17 29 46 66 67
33 55
30
5 38
39
40
12 16 31
13
41
(a) (b) (c)
Figure 2: Bacterial DGGE profiles of the 16S rRNA gene PCR products amplified from DNA extracted from guts (a, b) and eggs and testes
(b, c) of larvae (a) and adults (b, c) of A. chinensis collected from different host trees. Triangles and numbers indicate the bands sequenced
(Table 3).
detected occasionally in larvae or adults by culture or culture- of A. glabripennis from China, and a range of 5–31 genera,
independent analysis. Rothia, Kocuria, Propionibacterium, depending on the host tree, were found in the larval gut of
Enterococcus, Streptococcus, Staphylococcus, and Chryseobac- field-collected A. glabripennis from USA. The bacterial com-
terium were found in larvae, while Brevibacterium, Tsuka- munities, especially in the case of larvae, showed significant
murella, and Paenibacillus were found in adults. differences as a function of host tree, site of sampling, and, to
a lesser extent, specific individuals. The influence of host tree
was particularly evident in the case of larvae. Consistent with
3.2. Bacterial Community in Adult Testicles and Immature
previous results, the complexity of the bacterial community
Eggs. Data from isolation and DGGE detection methods
was higher in larvae fed on the host trees preferred by
revealed an abundance of Enterobacteriaceae also associated
the insects, which in this study were Acer and Salix. In
with testicles and eggs (Table 2 and Figure 2). Similar to
addition, we observed that larvae from these trees contained
microbial gut investigation, the DGGE patterns indicated
a higher proportion of Enterobacteriaceae. According to Geib
that the occurrence of diverse microbial genera (Enterobac-
et al. [18], the plasticity that characterizes the Anoplophora
ter, Raoultella, Klebsiella, and Buttiauxella) and unclassified
bacterial community is probably the reason for the broad
bacteria varied among the samples. Lysinibacillus sphaericus
host range of this beetle. However, regardless of differences
and Stenotrophomonas maltophilia isolates were detected in
in the insect species analyzed and geographic location of
both testicles and eggs. Staphylococcus was also found in eggs,
sampling (USA, China, Italy), the bacterial communities
whereas other microorganisms, generally of environmental
found in studies of the larvae of Anoplophora spp. are quite
origin, were identified in testicles.
similar. Interestingly, the studies investigating the taxonomy
and diet of bark-beetles related to Anoplophora spp. found
4. Discussion the majority of these xylophagous insects to have a lower
bacterial diversity than Anoplophora spp., ranging from four
Overall, the results indicated that the gut microbiota of larvae taxa identified in Tetropium castaneum [31] to an average of
and adults of Anoplophora chinensis was relatively complex about ten taxa in Dendroctonus species [32–35]. The toxic
being constituted by bacteria placed in six different bacterial activity of certain tree chemicals, such as terpenes in pine
classes. A total of 23 and 32 bacterial genera were found in resin, may be one of the factors determining the relatively
larvae and adults (19 in the gut), respectively, by both culture- scarce species diversity in the gut of these beetles with respect
dependent and independent methods. This moderately high to Anoplophora.
diversity is in accordance with previous data reported for In this study, for the first time we showed that the bacterial
larval forms of the species Anoplophora glabripennis [18, 19]. community was rather conserved also in adults regardless
Twenty-three bacterial taxa were harboured in the larval gut of the shift in diet occurring after the metamorphosis, with
Table 3: Closest relatives of bacterial 16S rRNA gene sequences of DGGE bands obtained from larvae and adults of Anoplophora chinensis.
Band Closest relative (accession no.) Identity (%) Bacterial division

1 Ralstonia solanacearum (JQ655458) 100 Betaproteobacteria
34 Ralstonia sp. (JN714979) 99.1–99.7 Betaproteobacteria
2 Uncultured Massilia sp. (EF075289) 99.3 Betaproteobacteria
35 Uncultured Massilia sp. (JN648276) 99.3 Betaproteobacteria
37 Massilia sp. (AB623119) 98.4 Betaproteobacteria
Raoultella terrigena (AY292875) 100 Gammaproteobacteria
3, 17
Enterobacter sp. (AB673457) 99.2–99.6 Gammaproteobacteria
Uncultured Raoultella sp. (FJ467399) 99.2–99.8 Gammaproteobacteria
14, 15, 23, 24
Raoultella terrigena (JN815233) 99.0–99.6 Gammaproteobacteria
18, 19, 20, 21 Raoultella ornithinolytica (HE578796) 98.4–99.8 Gammaproteobacteria
47, 48, 49, 50, 51 Raoultella planticola (JN835545) 99.2–99.6 Gammaproteobacteria
52 Raoultella ornithinolytica (HQ242732) 98.4 Gammaproteobacteria
53 Raoultella planticola (HE610795) 99.6 Gammaproteobacteria
56 Uncultured Raoultella sp. (FJ467399) 99.6 Gammaproteobacteria
58 Raoultella terrigena (GQ169108) 100 Gammaproteobacteria
62 Raoultella ornithinolytica (HQ242729) 99.8 Gammaproteobacteria
10 Gammaproteobacterium (EF111244) 99.8 Gammaproteobacteria
11 Rahnella sp. (JQ864392) 99.7 Gammaproteobacteria
12, 16 Gibbsiella dentisursi (AB566415) 99.7 Gammaproteobacteria
22 Pseudomonas sp. (JQ522968) 99.0 Gammaproteobacteria
25, 42, 45, 28 Klebsiella sp. (GU301269) 99.6–99.8 Gammaproteobacteria
27, 44, 43, 26 Klebsiella oxytoca (JF772070) 99.4–99.8 Gammaproteobacteria
54, 59 Klebsiella sp. CPK (GU301269) 99.6 Gammaproteobacteria
57 Klebsiella oxytoca (JX196648) 99.2 Gammaproteobacteria
29, 46 Enterobacter sp. JJDP1 (JQ726698) 100 Gammaproteobacteria
32 Enterobacter sp. ZYXCA1 (JN107752) 100 Gammaproteobacteria
33 Enterobacter sp. IICDBZ6 (JN836923) 99.8 Gammaproteobacteria
55 Enterobacter ludwigii (KC139450) 98.7 Gammaproteobacteria
67 Enterobacter sp. (JN129489) 98.9 Gammaproteobacteria
36 Nevskia sp. (JQ710439) 100 Gammaproteobacteria
63 Enterobacteriaceae bacterium (HM235485) 99.8 Gammaproteobacteria
65 Buttiauxella sp. (JF281151) 99.8 Gammaproteobacteria
66 Buttiauxella sp. (JX406856) 99.8 Gammaproteobacteria
4 Methylobacterium sp. (FJ225120) 100 Alphaproteobacteria
5, 38 Methylobacterium populi (JQ660234) 99.7–100 Alphaproteobacteria
39 Brevundimonas sp. S2U9 (HE814668) 100 Alphaproteobacteria
40 Sphingomonas sp. D40y (HE962513) 100 Alphaproteobacteria
6, 8 Enterococcus sp. (JF813181) 99.0–100 Firmicutes
7 Enterococcus gallinarum (JQ805717) 100 Firmicutes
9 Enterococcus casseliflavus (JX035954) 100 Firmicutes
61 Lysinibacillus sphaericus (JN377788) 98.7 Firmicutes
13, 41 Microbacterium sp. (EU584504) 100 Actinobacteria
30 Uncultured bacterium (JN394024) 99.8 Unclassified
64 Uncultured bacterium (GQ411142) 99.6 Unclassified
60 Uncultured Chitinophaga sp. (KC110981) 100 Bacteroidetes
31 Uncultured plastid (HM270514) 100 Eucariote plastid
the larvae fed on cambium, phloem, and xylem while the the overall function of the community is achieved despite
adults on foliage and tender bark. An analogous finding was variations in its bacterial members. This may indicate that
observed in the case of another wood-boring beetle Agrilus though most symbionts are environmentally-derived tran-
planipennis [36], which was similar to Anoplophora in the sient bacteria, at least some may play a key role in the
complexity of the larval gut community [37]. physiology of this beetle. In particular, the dominance of
The observed stability in the composition of the bacterial Enterobacteriaceae and Gammaproteobacteria in both larval
community at a high taxonomic level may indicate that and adult forms suggests that they are a constant fraction
of the gut bacterial community and may be beneficial to [46, 48]. For example, several bacterial genera affiliated to
host fitness because of their various abilities to hydrolyze Actinobacteria, Gammaproteobacteria, Betaproteobacteria,
and ferment carbohydrates, catalyze nitrogen fixation, and and Firmicutes that are contained in the oral secretions of the
produce vitamins and pheromones. It should be noted bark beetle Dendroctonus rufipennis were demonstrated to
that this phylogenetic group of microorganisms has been significantly inhibit the growth of antagonistic fungi [49]. In
commonly detected in the gut of diverse insect orders and particular, recent findings suggest that symbiotic associations
host diet, with the exception of detritivorous, pollenivorous, between insects and Actinobacteria could play a crucial
and dead wood xylophagous insects [36]. In Anoplophora, role in the protection of the insect host, or its nutritional
such microorganisms might act as facultative mutualistic resources, against parasitoids or predators [50]. In this study,
bacteria recurrently acquired during feeding by ingestion various actinobacterial genera were detected, though they
and possibly horizontally transmitted between individuals. represented a small fraction of the microbiome associated
In particular, the recurrent detection of the diazotrophs with Anoplophora spp.; further research is necessary to
Enterobacter sp., Klebsiella sp., Raoultella sp., Rahnella sp., elucidate their potential functions.
and S. maltophilia suggests that their contribution to beetle A preliminary characterization of the bacterial communi-
nitrogen requirements may be noteworthy [38], as observed ties associated with testicles and eggs of Anoplophora chinen-
in other insect orders [39]. In addition, considering that sis, despite being limited by the low number of individuals
no obligate anaerobic bacteria were identified, facultative analyzed, allowed us to obtain initial information regarding
anaerobic bacteria may work as oxygen scavengers and could the microorganisms potentially associated with these organs.
have a significant role in creating the microsite anaerobic con- Taken together, the results showed that the same microbial
ditions necessary to allow nitrogen fixation [34]. Members species identified in the insect gut were present in these
of the Enterobacteriaceae are also known to be involved in tissues. In accordance with a previous study, it is noteworthy
pheromone production; for example, common gut isolates to mention the occurrence of a Xanthomonadaceae family
in locusts, E. cloacae, K. pneumonia, and P. agglomerans, are member associated with immature eggs that may be vertically
responsible for the production of components of a locust transmitted from the mother to the offspring [51].
cohesion pheromone [40].
Interestingly, the presence of Gammaproteobacteria and 5. Conclusions
more generally the composition of bacterial communities,
are rather similar in different xylophagous beetles and The bacterial gut community of A. chinensis is relatively
significantly distinguished from those of insects feeding diverse and this diversity is maintained throughout different
on dead lignocellulose tissues, such as termites. The diet life stages and geographic locations. The community does not
and, consequently, mechanisms of digestion evolved in the appear to be primarily involved in lignocellulose degradation,
host, including those related to the host gut anatomy, are but conservation of its members at high ranks suggests that
thought to play an important role in structuring the bacterial these bacteria are beneficial to the host fitness and may
community [36]. In view of recent reports identifying a novel contribute to insect nutrition, presumably by providing a
endogenous exo/endocellulase from A. malasiaca, together fixed nitrogen source. Further studies are needed to elucidate
with the characteristic anatomy of this beetle which harbours the specific functions of gut-associated bacteria. Similarly,
a relatively small hindgut, it is likely that the bacterial further investigation is necessary to clarify the role and mode
community associated with Anoplophora spp. is more closely of transmission of bacteria associated with the reproductive
related to host fitness rather than being primarily involved in systems of Anoplophora spp.
wood degradation, though several lignocellulose-degrading
microbes can be harboured [41, 42]. Considering the bacteria Acknowledgment
identified in this study, Pseudomonas putida, Kocuria, and
Acinetobacter were previously shown to have lignin degra- This study was supported by the project “Anoplophora chinen-
dation activity [43, 44]; Bacillus, Paenibacillus, Staphylococ- sis (Forster): nuove acquisizioni di biologia, fisiologia, diffu-
cus (Firmicutes), Sphingomonas (Alfaproteobacteria), Ral- sione e possibilità di contenimento—ANOCHI” by Regione
stonia, Comamonas (Betaproteobacteria), Dyella ginsengisoli, Lombardia, Italy.
Stenotrophomonas (Gammaproteobacteria), Kocuria, Bre-
vibacterium (Actinobacteria), and Chryseobacterium (Bac-
References
teroidetes) were shown to have cellulose and/or aromatics
degradation capabilities [45, 46]. In addition to the bacterial [1] R. J. Dillon and V. M. Dillon, “The gut bacteria of insects:
lignocellulose-degrading activities, the nutrient-extracting nonpathogenic interactions,” Annual Review of Entomology, vol.
capacities exerted by fungi strictly associated with the host are 49, pp. 71–92, 2004.
thought to contribute to host nutrition, as recently indicated [2] J. A. Breznak and A. Brune, “Role of microorganisms in
by enzymatic proprieties of the A. glabripennis isolate F. the digestion of lignocellulose by termites,” Annual Review of
solani [47]. Moreover, some of these bacteria with specific Entomology, vol. 39, pp. 453–487, 1994.
enzymatic degrading activities are thought to play important [3] M. Ohkuma, “Termite symbiotic systems: efficient bio-recycling
roles in the detoxification of plant compounds, production of of lignocellulose,” Applied Microbiology and Biotechnology, vol.
metabolites against pathogens, and plant-insect interactions 61, no. 1, pp. 1–9, 2003.
[4] A. E. Douglas, “Nutritional interactions in insect-microbial [20] E. D. Scully, K. Hoover, J. Carlson, M. Tien, and S. M. Geib,
symbioses: aphids and their symbiotic bacteria Buchnera,” “Proteomic analysis of Fusarium solani isolated from the asian
Annual Review of Entomology, vol. 43, pp. 17–37, 1998. longhorned beetle, Anoplophora glabripennis,” PLoS ONE, vol.
[5] S. Shigenobu, H. Watanabe, M. Hattori, Y. Sakaki, and H. 7, no. 4, Article ID e32990, 2012.
Ishikawa, “Genome sequence of the endocellular bacterial [21] C.-J. Chang, C. P. Wu, S.-C. Lu et al., “A novel exo-cellulase from
symbiont of aphids Buchnera sp. APS,” Nature, vol. 407, no. white spotted longhorn beetle (Anoplophora malasiaca),” Insect
6800, pp. 81–86, 2000. Biochemistry and Molecular Biology, vol. 42, no. 9, pp. 629–636,
[6] A. E. Douglas, “The microbial dimension in insect nutritional 2012.
ecology,” Functional Ecology, vol. 23, no. 1, pp. 38–47, 2009. [22] V. A. Cavalcante and J. Dobereiner, “A new acid-tolerant
[7] E. Crotti, C. Damiani, M. Pajoro et al., “Asaia, a versatile acetic nitrogen-fixing bacterium associated with sugarcane,” Plant and
acid bacterial symbiont, capable of cross-colonizing insects Soil, vol. 108, no. 1, pp. 23–31, 1988.
of phylogenetically distant genera and orders,” Environmental [23] A. Rizzi, L. Panebianco, D. Giaccu, C. Sorlini, and D. Daf-
Microbiology, vol. 11, no. 12, pp. 3252–3264, 2009. fonchio, “Stability and recovery of maize DNA during food
[8] C. Damiani, I. Ricci, E. Crotti et al., “Paternal transmission of processing,” Italian Journal of Food Science, vol. 15, no. 4, pp.
symbiotic bacteria in malaria vectors,” Current Biology, vol. 18, 499–510, 2003.
no. 23, pp. R1087–R1088, 2008. [24] E. F. DeLong, “Archaea in coastal marine environments,” Pro-
[9] G. Favia, I. Ricci, C. Damiani et al., “Bacteria of the genus Asaia ceedings of the National Academy of Sciences of the United States
stably associate with Anopheles stephensi, an Asian malarial of America, vol. 89, no. 12, pp. 5685–5689, 1992.
mosquito vector,” Proceedings of the National Academy of [25] J. Messing, “New M13 vectors for cloning,” Methods in Enzymol-
Sciences of the United States of America, vol. 104, no. 21, pp. ogy, vol. 101, pp. 20–78, 1983.
9047–9051, 2007.
[26] A. M. Sass, H. Sass, M. J. L. Coolen, H. Cypionka, and J.
[10] E. Gonella, E. Crotti, A. Rizzi et al., “Horizontal transmission of
Overmann, “Microbial communities in the chemocline of a
the symbiotic bacterium Asaia sp. in the leafhopper Scaphoideus
hypersaline deep-sea Basin (Urania Basin, Mediterranean Sea),”
titanus Ball (Hemiptera: Cicadellidae),” BMC Microbiology, vol.
12, no. 1, article S4, 2012.
5392–5402, 2001.
[11] R. Koga, X.-Y. Meng, T. Tsuchida, and T. Fukatsu, “Cellular
[27] E. Gonella, I. Negri, M. Marzorati et al., “Bacterial endosym-
mechanism for selective vertical transmission of an obligate
biont localization in Hyalesthes obsoletus, the insect vector
insect symbiont at the bacteriocyte-embryo interface,” Proceed-
of bois noir in Vitis vinifera,” Applied and Environmental
ings of the National Academy of Sciences of the United States of
America, vol. 109, no. 20, pp. 1230–1237, 2012.
[12] E. Crotti, A. Balloi, C. Hamdi et al., “Microbial symbionts: [28] N. Raddadi, E. Gonella, C. Camerota et al., ““Candidatus
a resource for the management of insect-related problems,” Liberibacter europaeus” sp. nov. that is associated with and
Microbial Biotechnology, vol. 5, no. 3, pp. 307–317, 2012. transmitted by the psyllid Cacopsylla pyri apparently behaves
as an endophyte rather than a pathogen,” Environmental Micro-
[13] I. Ricci, C. Damiani, P. Rossi et al., “Mosquito symbioses: from
biology, vol. 13, no. 2, pp. 414–426, 2011.
basic research to the paratransgenic control of mosquito-borne
diseases,” Journal of Applied Entomology, vol. 135, no. 7, pp. 487– [29] G. Merlino, A. Rizzi, F. Villa et al., “Shifts of microbial com-
493, 2011. munity structure during anaerobic digestion of agro-industrial
[14] C. Hamdi, A. Balloi, J. Essanaa et al., “Gut microbiome dysbiosis energetic crops and food industry byproducts,” Journal of
and honeybee health,” Journal of Applied Entomology, vol. 135, Chemical Technology and Biotechnology, vol. 87, no. 9, pp. 1302–
no. 7, pp. 524–533, 2011. 1311, 2012.
[15] M. Colombo and L. Limonta, “Anoplophora malasiaca Thom- [30] K. Tamura, D. Peterson, N. Peterson, G. Stecher, M. Nei, and
son (Coleoptera Cerambycidae Lamiinae Lamiini) in Europe,” S. Kumar, “MEGA5: Molecular evolutionary genetics analysis
Bollettino di Zoologia Agraria e di Bachicoltura, no. 33, Ser. 2, pp. using maximum likelihood, evolutionary distance, and max-
65–68, 2001. imum parsimony methods,” Molecular Biology and Evolution,
vol. 28, no. 10, pp. 2731–2739, 2011.
[16] R. A. Haack, F. Hérard, J. Sun, and J. J. Turgeon, “Managing
invasive populations of asian longhorned beetle and citrus [31] S. Grünwald, M. Pilhofer, and W. Höll, “Microbial associations
longhorned beetle: a worldwide perspective,” Annual Review of in gut systems of wood- and bark-inhabiting longhorned
Entomology, vol. 55, pp. 521–546, 2010. beetles [Coleoptera: Cerambycidae],” Systematic and Applied
[17] K. D. Klepzig, A. S. Adams, J. Handelsman, and K. F. Raffa, Microbiology, vol. 33, no. 1, pp. 25–34, 2010.
“Symbioses: a key driver of insect physiological processes, eco- [32] A. Vasanthakumar, I. Delalibera, J. Handelsman et al., “Char-
logical interactions, evolutionary diversification, and impacts acterization of gut-associated bacteria in larvae and adults of
on humans,” Environmental Entomology, vol. 38, no. 1, pp. 67–77, the southern pine beetle, Dendroctonus frontalis Zimmermann,”
2009. Environmental Entomology, vol. 35, no. 6, pp. 1710–1717, 2006.
[18] S. M. Geib, M. D. M. Jimenez-Gasco, J. E. Carlson, M. Tien, [33] H. Yilmaz, K. Sezen, H. Kati, and Z. Demirbaǧ, “The first
and K. Hoover, “Effect of host tree species on cellulase activity study on the bacterial flora of the European spruce bark beetle,
and bacterial community composition in the gut of larval asian Dendroctonus micans (Coleoptera: Scolytidae),” Biologia, vol. 61,
longhorned beetle,” Environmental Entomology, vol. 38, no. 3, no. 6, pp. 679–686, 2006.
pp. 686–699, 2009. [34] J. Morales-Jiménez, G. Zúñiga, L. Villa-Tanaca, and C.
[19] P. D. Schloss, I. Delalibera Jr., J. Handelsman, and K. F. Raffa, Hernández-Rodrı́guez, “Bacterial community and nitrogen
“Bacteria associated with the guts of two wood-boring beetles: fixation in the red turpentine beetle, Dendroctonus valens
Anoplophora glabripennis and Saperda vestita (Cerambycidae),” LeConte (Coleoptera: Curculionidae: Scolytinae),” Microbial
Environmental Entomology, vol. 35, no. 3, pp. 625–629, 2006. Ecology, vol. 58, no. 4, pp. 879–891, 2009.
[35] J. Morales-Jiménez, G. Zúñiga, H. C. Ramı́rez-Saad, and C. to investigate transmission of bacteria between life stages of
Hernández-Rodrı́guez, “Gut-associated bacteria throught the the wood-boring beetle, Anoplophora glabripennis,” Microbial
life cycle of the bark beetle Dendroctonous rhizophagus Thomas Ecology, vol. 58, no. 1, pp. 199–211, 2009.
and Bright (Curculionidae: Scolytinae) and their cellulolytic
activities,” Microbial Ecology, pp. 1–11, 2012.
[36] D. R. Colman, E. C. Toolson, and C. D. Takacs-Vesbach, “Do
diet and taxonomy influence insect gut bacterial communities?”
Molecular Ecology, vol. 21, no. 20, pp. 5124–5137, 2012.
[37] A. Vasanthakumar, J. O. Handelsman, P. D. Schloss, L. S. Bauer,
and K. F. Raffa, “Gut microbiota of an invasive Subcortical
Beetle, Agrilus planipennis fairmaire, across various life stages,”
Environmental Entomology, vol. 37, no. 5, pp. 1344–1353, 2008.
[38] J. B. Nardi, R. I. Mackie, and J. O. Dawson, “Could microbial
symbionts of arthropod guts contribute significantly to nitrogen
fixation in terrestrial ecosystems?” Journal of Insect Physiology,
vol. 48, no. 8, pp. 751–763, 2002.
[39] A. Behar, B. Yuval, and E. Jurkevitch, “Enterobacteria-mediated
nitrogen fixation in natural populations of the fruit fly Ceratitis
capitata,” Molecular Ecology, vol. 14, no. 9, pp. 2637–2643, 2005.
[40] R. Dillon and K. Charnley, “Mutualism between the desert
locust Schistocerca gregaria and its gut microbiota,” Research in
[41] H. Watanabe and G. Tokuda, “Cellulolytic systems in insects,”
Annual Review of Entomology, vol. 55, pp. 609–632, 2010.
[42] C.-J. Chang, C. P. Wu, S.-C. Lu et al., “A novel exo-cellulase from
white spotted longhorn beetle (Anoplophora malasiaca),” Insect
Biochemistry and Molecular Biology, vol. 42, pp. 629–636, 2012.
[43] D. C. Kalyani, S. S. Phugare, U. U. Shedbalkar, and J. P. Jadhav,
“Purification and characterization of a bacterial peroxidase
from the isolated strain Pseudomonas sp. SUK1 and its applica-
tion for textile dye decolorization,” Annals of Microbiology, vol.
61, no. 3, pp. 483–491, 2011.
[44] G. K. Parshetti, S. Parshetti, D. C. Kalyani, R.-A. Doong, and S. P.
Govindwar, “Industrial dye decolorizing lignin peroxidase from
Kocuria rosea MTCC 1532,” Annals of Microbiology, vol. 62, pp.
217–223, 2012.
[45] T. D. Bugg, M. Ahmad, E. M. Hardiman, and R. Singh, “The
emerging role for bacteria in lignin degradation and bio-
product formation,” Current Opinion in Biotechnology, vol. 22,
no. 3, pp. 394–400, 2011.
[46] N. Calderon-Cortes, M. Quesada, H. Watanabe, H. Cano-
Camacho, and K. Oyama, “Endogenous plant cell wall diges-
tion: a key mechanism in insect evolution,” Annual Review of
Ecology, Evolution, and Systematics, vol. 43, pp. 45–71, 2012.
[47] E. D. Scully, K. Hoover, J. Carlson, M. Tien, and S. M. Geib,
“Proteomic analysis of Fusarium solani isolated from the asian
longhorned beetle, Anoplophora glabripennis,” PLoS ONE, vol.
7, no. 4, Article ID e32990, 2012.
[48] V. Popa, E. Déziel, R. Lavallée, E. Bauce, and C. Guertin,
“The complex symbiotic relationships of bark beetles with
microorganisms: a potential practical approach for biological
control in forestry,” Pest Management Science, vol. 68, no. 7, pp.
963–975, 2012.
[49] Y. J. Cardoza, K. D. Klepzig, and K. F. Raffa, “Bacteria in oral
secretions of an endophytic insect inhibit antagonistic fungi,”
Ecological Entomology, vol. 31, no. 6, pp. 636–645, 2006.
[50] M. Kaltenpoth, “Actinobacteria as mutualists: general health-
care for insects?” Trends in Microbiology, vol. 17, no. 12, pp. 529–
535, 2009.
[51] S. M. Geib, M. del Mar Jimenez-Gasco, J. E. Carlson, M. Tien,
R. Jabbour, and K. Hoover, “Microbial community profiling
http://dx.doi.org/10.1155/2013/438956
Research Article
Population Abundance of Potentially
Pathogenic Organisms in Intestinal Microbiome of Jungle Crow
(Corvus macrorhynchos) Shown with 16S rRNA Gene-Based
Microbial Community Analysis
Isamu Maeda, Mohammad Shohel Rana Siddiki, Tsutomu Nozawa-Takeda,

Naoki Tsukahara, Yuri Tani, Taki Naito, and Shoei Sugita
Faculty of Agriculture, Utsunomiya University, 350 Minemachi, Utsunomiya 321-8505, Japan
Correspondence should be addressed to Isamu Maeda; i-maeda@cc.utsunomiya-u.ac.jp
Received 11 March 2013; Revised 26 June 2013; Accepted 23 July 2013
Academic Editor: Konstantinos Mavrommatis
Copyright © 2013 Isamu Maeda et al. This is an open access article distributed under the Creative Commons Attribution License,
Jungle Crows (Corvus macrorhynchos) prefer human habitats because of their versatility in feeding accompanied with human food
consumption. Therefore, it is important from a public health viewpoint to characterize their intestinal microbiota. However, no
studies have been involved in molecular characterization of the microbiota based on huge and reliable number of data acquisition. In
this study, 16S rRNA gene-based microbial community analysis coupled with the next-generation DNA sequencing techniques was
applied to the taxonomic classification of intestinal microbiome for three jungle crows. Clustering of the reads into 130 operational
taxonomic units showed that at least 70% of analyzed sequences for each crow were highly homologous to Eimeria sp., which
belongs to the protozoan phylum Apicomplexa. The microbiotas of three crows also contained potentially pathogenic bacteria with
significant percentages, such as the genera Campylobacter and Brachyspira. Thus, the profiling of a large number of 16S rRNA gene
sequences in crow intestinal microbiomes revealed the high-frequency existence or vestige of potentially pathogenic microorgan-
isms.
1. Introduction which might harbor pathogenic organisms to humans and

livestock, remains to be insufficient or obscure. Therefore, it
The Jungle Crows (Corvus macrorhynchos) are very adaptable is important for public health to characterize the intestinal
to human habitats and are able to survive on a wide range pathogens within the gut of Jungle Crow, and more efforts
of food sources. They are extremely versatile in their feeding should be given for understanding their profiles and trans-
and prefer environments with an abundant supply of large mission of diseases to humans and animals.
garbage or carcasses in cities and coastal areas [1]. Microor- Gut microorganisms have been isolated and character-
ganisms’ populations that collectively comprise microbiota in ized by a variety of methods such as culture-based assays,
crow’s digestive system are associated with rearing conditions culture-independent DNA sequencing approaches using the
and environment. The intestinal microbial communities do conventional Sanger sequencing, and the fluorescence in situ
not only have an important role in host metabolism [2] but hybridization targeting the 16S rRNA gene [5]. Most of such
also relate with zoonosis and food poisoning to human, ani- traditional approaches are limited in scope. Culture-based
mals, and birds [3]. Crows are considered as potential sources assays bias bacterial profiling towards culturable fractions.
for fecal contamination of water supplies and foods, and the DNA-based traditional approaches have targeted to phyloge-
pathogens are transmitted through the fecal-oral route when netic genes, which provided limited information on the pub-
humans would consume those contaminated foods or water lic health relevance, especially when a large amount of micro-
[4]. However, information on microbiotas of crow intestine, bial groups must be classified [5]. That weakness becomes
obvious in conventional Sanger’s sequencing, which has limi- analyzed spectrophotometrically using the SmartSpec plus
tation in terms of amount of reads produced and affordability. (BioRad, Tokyo, Japan). The extracted 10−17 𝜇g DNA was
In recent years, metagenomics using the next-generation stored at −20∘ C until use.
sequencing has been chosen in analyzing complex microbial
communities and functions. The next-generation sequencers 2.3. PCR Conditions. Twenty-five to fifty nanograms of iso-
are capable of massively parallel sequencing of thousands or lated DNA were transferred to a PCR tube containing 50 𝜇L of
millions of amplified DNA molecules in a single run and a polymerase chain reaction (PCR) mixture containing
do not require the conventional cloning and amplification in LA Taq DNA polymerase (TaKaRa Bio, Shiga, Japan). For
bacteria when compared to the traditional Sanger capillary DNA evaluation, the eubacterial 16S rRNA gene sequence
sequencers [6]. These technologies have enabled metagenom- intervened between UNIV341F (5󸀠 -CCTACGGGAGGC-
ics to become widespread, routine, and inexpensive, rather AGCAG-3󸀠 ) and UNIV907R (5󸀠 -CCCCGTCAATTCCTT-
than requiring significant production-scale efforts [5, 7]. Met- TGAGTTT-3󸀠 ) was amplified. For pyrosequencing, the
agenomic study offers a powerful lens for viewing the com- eubacterial 16S rRNA gene sequence intervened between
prehensive and unbiased assessment of microbial diversity the 27F (5󸀠 -AGAGTTTGATCMTGGCTCAG-3󸀠 ) and 519R
within the complex gut ecosystem by allowing examination of (5󸀠 -GWATTACCGCGGCKGCTG-3󸀠 ) sites was amplified
organisms not easily cultured in laboratory [8, 9]. [10]. MA-27F, to which multiplex identifiers (MIDs), MID-
Although microorganisms in the digestive tract of Jungle 151 (5󸀠 -CCATCTCATCCCTGCGTGTCTCCGACTCAG-
Crow have been studied by using conventional culture-based TGCTAGTCAG-3󸀠 ), MID-152 (5󸀠 -CCATCTCATCCCTGC-
assays [3], sufficient coverage of nonabundant and uncul- GTGTCTCCGACTCAGTGTATCACAG-3󸀠 ), and MID-153
turable microbial groups in the gut have not been obtained. (5󸀠 -CCATCTCATCCCTGCGTGTCTCCGACTCAGTGT-
Interestingly, in spite of its importance in the public health GCGCGTG-3󸀠 ), were attached, and MB-519R (5󸀠 -CCTATC-
relevance, no information on crow gut microbiota is available. CCCTGTGTGCCTTGGCAGTCTCAGGWATTACCG-
The present study aims to characterize the microbial commu- CGGCKGCTG-3󸀠 ) were provided from Hokkaido System
nity within the intestine of Jungle Crow based on information Science (Sapporo, Hokkaido, Japan). MIDs distinguished the
of thousands of reads targeted to microbial 16S rRNA genes amplified DNA fragments derived from three Jungle Crows.
obtained by the next-generation DNA sequencing. The thermal cycling conditions were composed of a denatur-
ing step at 94∘ C for 1 min, 30 thermal cycles of 94∘ C for 30 s,
2. Materials and Methods 60∘ C for 30 s, and 68∘ C for 1 min, and an additional extension
step at 72∘ C for 5 min. The PCR products were analyzed by the
2.1. Preparation of Crow Intestinal Contents. All procedures agarose (1.2% w/v) gel electrophoresis in TBE buffer (89 mM
involving crows were performed in accordance with the Tris-borate pH 8.3, 2 mM EDTA). The PCR fragments were
guidelines of Utsunomiya University (animal experiment per- stained with ethidium bromide and visualized with a UV
mission A11-0012). Adult three Jungle Crows (Corvus macro- transilluminator after electrophoresis. The PCR products for
rhynchos) were caught by traps set at the Experimental pyrosequencing were purified using QIAquick PCR purifica-
Farm of Utsunomiya University located in Moka city during tion kit (Qiagen, Tokyo, Japan).
their nonbreeding season (October 2011, December 2011, and
January 2012). Permits to trap crows were obtained from the
2.4. Pyrosequencing and Population Analysis. The PCR prod-
Tochigi prefecture (number 0010). All the materials that came
ucts with MIDs, whose quality was analyzed with a bioanaly-
in contact at processing and preparation of samples were
zer (Agilent 2100, Agilent Technologies, Palo Alto, CA, USA),
cleaned with disinfectant or sterilized. The crows were euth-
were equally mixed and subjected to commercially available
anized by an overdose of pentobarbital sodium (50 mg/kg
body weight). The crows were killed using sharp knife and pyrosequencing using the Roche Genome Sequencer GS
allowed to drain all blood. Then, the crows were dissected FLX+ system (Hokkaido System Science, Sapporo, Hokkaido,
immediately with scissors. The intestine was occluded by the Japan). Determined nucleotide sequences not containing
occluding clamps, aseptically removed from the abdominal unspecified nucleotide and with read length of longer than
cavity of crow, and soaked in distilled water in a Petri dish. All 250 bp were extracted as passed-filter (PF) reads and cate-
the contents were gently squeezed out of the intestines to the gorized into 3 groups according to MIDs, subjected to the
distilled water and transferred into polypropylene centrifuge BLAST search against the DNA Data Bank of Japan (DDBJ)
tubes. The intestinal contents were harvested with centrifu- bacterial 16S rRNA gene database, and converted to the data
gation for 5 min at 4∘ C with 10,000 rpm. sets composed of taxonomy ID and genus, based on informa-
tion of a top hit where alignment length was more than 30%
2.2. DNA Extraction and Purification from Intestinal Content. of submitted nucleotide sequence. Then, genera showing the
Genomic DNA was isolated and purified from fresh intestinal BitScores higher than 680 were chosen for further analysis.
content by using QIAamp DNA Stool Mini kit (Qiagen, Finally, the genera identified were confirmed by the Ribo-
Tokyo, Japan). For this purpose, 200 𝜇g fresh weight of sedi- somal Database Project (RDP) [11] Classifier with the con-
ment of crow intestinal content was thoroughly homogenized fidence threshold of 80%.
according to the manufacturer’s protocol. Finally, genomic
DNA was dissolved with 200 𝜇L buffer AE (10 mM Tris-Cl, 2.5. Phylogenetics and Statistics. Using RDP, key tag and
0.5 mM EDTA, pH 9.0). DNA purity and concentration were primer sequences were trimmed off and sequences with low
Table 1: Summary of numbers of reads, numbers of OTU, and diversity measurements.
Crow Sampling month, year Number of PF read Number of analyzed read Distance label Number of OTU 𝛼 diversity 𝛽 diversity
1 October, 2011 17635 15050 0.05 70 1.03 Crows 1-2, 0.270;
2 December, 2011 14757 12142 0.05 65 1.90 Crows 1–3, 0.087;
3 January, 2012 17712 14163 0.05 59 1.48 Crows 2-3, 0.165
quality were removed from the PF reads, and then the quali- M A B C D
fied sequences were aligned and compared by RDP to create
column formatted distance matrix data. Finally, operational
taxonomic units (OTUs) were created based on the distance
matrix data by mothur [12]. By counting the number of
sequencing in each OTU, the reciprocal of the Simpson index
was calculated as 𝛼 diversity, which represents observed rich-
ness within a group, and the 𝜃YC distance of Yue and Clayton
was calculated as 𝛽 diversity, which represents compositional
heterogeneity between two habitats [13]. A heatmap was
created based on the OTU data by R 3.0.1 for Windows.
3. Results and Discussion

3.1. Evaluation of Extracted DNA and Pyrosequencing from
Crow Intestinal Contents. For analyzing the quality of total
DNA extracted from crow intestinal microbiota, PCR was
attempted using LA Taq DNA polymerase in combination
with UNIV341F and UNIV907R primers. The expected prod- Figure 1: Amplified partial fragment of 16S rRNA gene from crow 1
ucts size of 550 bp with sufficient DNA concentrations was (A), crow 2 (B), and crow 3 (C) for evaluation of intestinal microbio-
obtained from intestinal DNA extracted from three speci- mes. Control PCR was performed under the same conditions except
mens of crow intestinal content (Figure 1). Without template for without template DNA (D). Molecular size marker (M) shows 1.5,
DNA for PCR, DNA fragments were not detected, indicating 1.0, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, and 0.1 kb from top to bottom.
that the amplified DNA fragments were derived from intesti-
nal microbial DNA.
For pyrosequencing of 16S rRNA genes in crow intestinal within OTU2-130. Sixteen OTUs for bacteria and one OTU
microbiome, PCR was attempted using LA Taq DNA poly- (OTU1) for Eimeria sp. were shared among the three crows
merase in combination with MA-27F and MB-519R primers. (Figure 3). However, uniquely found OTUs in each crow were
The amplified DNA fragments had an almost uniform size composed of small numbers of reads (<10), except for OTU6
distribution between 480 bp and 490 bp. As a result of pyrose- (126 reads) and OTU14 (22 reads) in crow 3.
quencing, more than 10,000 PF reads were obtained for each
crow (Table 1). A total of 303 reads for crow 1, 4,563 reads for 3.3. Taxonomy of Potentially Pathogenic Bacteria. Among the
crow 2, and 3,143 reads for crow 3 matched the DDBJ bacterial 8009 PF reads in the three crows matched with sequences in
16S rRNA gene sequences with the confident level. the bacterial 16S rRNA gene databases, the sequences mostly
clustered into OTU2 were abundantly found (12% in crow
3.2. Microbial Community Analysis. Using RDP, tag and 1, 54% in crow 2, and 4.0% in crow 3) and identified as the
primer sequences were removed from the PF reads, and gene of the genus Helicobacter. The highest hits were obtained
short sequences that did not contain the 519R sequence were against H. suncus (99% identity) and H. pametensis (99%
excluded. A total of 41,355 reads from the three crows were identity). Within the crow 1 and crow 3 intestinal microbiotas,
analyzed, of which 130 OTUs were clustered at the distance Lactococcus sp. and Campylobacter sp. (OTU3) were the
label of 0.05 (Figure 2). OTU1 occupied 98% of the analyzed most dominant bacteria (16% in crow 1 and 76% in crow 3),
reads obtained from crow 1, 70% of those from crow 2, and respectively. The highest hit among the genus Campylobacter
81% of those from crow 3. Among the analyzed reads, 97% was obtained against C. jejuni (100% identity). Among the
and 96% of the total numbers were clustered into OTU1, PF reads obtained with the crow 2 microbiota, the sequences
OTU2, and OTU4 in crow 2 and OTU1 and OTU3 in crow 3. mostly clustered into OTU4 were secondly abundant (28%)
The sequences clustered within OTU1 were highly homolo- and showed 100% identity with the gene of Brachyspira pilos-
gous to those of 16S rRNA genes in the protozoan phylum icoli.
Apicomplexa. The highest identity (96% in 469 bp, 1% gap)
was obtained with that of Eimeria praecox. 3.4. Possibility of the Jungle Crow as a Vector of Pathogen.
Alpha and beta diversities were dependent on bacterial Although the genus Helicobacter consists of over 20 recog-
abundance (Table 1), which was shown as the numbers of read nized species, H. pylori, whose infection represents a key
Crow 1
Crow 3
31
12 10
17
Crow 2
28 8 24
Crow 2 Crow 3
Figure 3: Venn diagram showing numbers of shared and unique

Crow 1
taxa among the three crows.
1 130
Number of OTU agents of some diseases and virulent strains to domestic fowls
and humans.
100 101 102 103 104 4. Conclusions

Number of read
Herein, the crow intestinal microbiotas were characterized in
Figure 2: Heatmap representation of the clustered distant matrix terms of bacterial and protozoan genera that potentially cause
data based on the pyrosequencing for 16S rRNA genes of crow some diseases to domestic fowls and in some cases to humans
intestinal microbiome. Number of OTU: numbers of operational
taxonomic unit; number of read: numbers of read clustered.
using pyrosequencing of 16S rRNA genes. Diversity of micro-
bial populations by each crow clarified that Eimeria sp. in the
protozoan phylum Apicomplexa was highly abundant in all
three microbiotas of crow intestine and bacterial genera such
factor in the etiology of various gastrointestinal diseases [14],
as Campylobacter and Brachyspira were dominantly included
was not found in the three crows. The gene sequences of H.
depending on the individual crow bodies. The obtained infor-
suncus and H. pametensis, which were isolated from the house
mation suggests roles and importance of Jungle Crows on
musk shrews [15], wild birds, and a domestic swine [16], were
transferring pathogenic microorganisms within the human
commonly found among the intestinal microbiotas of three
crows. However, it has not been well investigated whether habitat.
these bacteria possess pathogenicity as observed in H. pylori.
The fact that the sequences highly homologous to the gene Acknowledgments
of B. pilosicoli, which is a zoonotic intestinal spirochaete and
causes avian intestinal spirochaetosis [17], and to the gene This work was supported by a Grant-in-Aid for Scientific
of foodborne pathogen Campylobacter jejuni [18] exist in the Research (A) (23248053) to Isamu Maeda, Tsutomu Nozawa-
crow intestinal microbiotas suggests that the crow’s feces is Takeda, and Shoei Sugita and by a Grant-in-Aid from the
a potential source that spreads bacterial infectious diseases Utsunomiya University Center for Optical Research and Edu-
harmful for domestic fowls and humans. cation (CORE) to Isamu Maeda, Mohammad Shohel Rana
Siddiki, Naoki Tsukahara, and Shoei Sugita.
This is the first study in which a next-generation sequenc-
ing approach to characterize the crow intestinal microbiotas
was carried out. Based on bioinformatics, which can handle References
a huge amount of data coupled with the next-generation
[1] H. Higuchi, “Habitat segregation between the Jungle crow and
sequencing of PCR-amplified DNA, the existence and vestige
the carrion crow, Corpus macrorhynchos, C. corone, in Japan,”
of pathogenic organisms could be clarified. Jungle Crows Japanese Journal of Ecology, vol. 29, no. 2, pp. 353–358, 1979.
do not only acquire pathogenic microorganisms from their
[2] A. M. McKnite, M. E. Perez-Munoz, L. Lu et al., “Murine gut
native habitats, but also return them via excretion. Garbage is microbiota is defined by host genetics and modulates variation
preferred by Jungle Crows (Corvus macrorhynchos) than Car- of metabolic traits,” PLoS One, vol. 7, no. 6, Article ID e319191,
rion Crows (Corvus corone) [1]. These crows are highly adapt- 2012.
able in urban area, and their droppings are everywhere; thus [3] Y. Aruji, K. Tamura, S. Sugita, and Y. Adachi, “Intestinal micro-
contamination is inevitable. The discovery by this study dem- flora in 45 crows in Ueno Zoo and the in vitro susceptibilities of
onstrating inhabitation of large population of Eimeria sp., 29 Escherichia coli isolates to 14 antimicrobial agents,” Journal of
whose pathogenesis to chickens has been shown [19], in Veterinary Medical Science, vol. 66, no. 10, pp. 1283–1286, 2004.
crow intestine also supports the hypothesis that Jungle Crows [4] H. Y. Lee, A. Stephen, D. Sushela, and M. Mala, “Detection of
might be potential vectors and natural reservoirs of the causal protozoan and bacterial pathogens of public health importance
in faeces of corvus spp. (large-billed crow),” Tropical Biomedi-

cine, vol. 25, no. 2, pp. 134–139, 2008.
[5] V. Gomez-Alvarez, R. P. Revetta, and J. W. Santo Domingo,
“Metagenomic analyses of drinking water receiving different
disinfection treatments,” Applied and Environmental Microbiol-
ogy, vol. 78, no. 17, pp. 6095–6102, 2012.
[6] G. E. Liu, “Applications and case studies of the next-generation
sequencing technologies in food, nutrition and agriculture,”
Recent Patents on Food, Nutrition & Agriculture, vol. 1, no. 1, pp.
75–79, 2009.
[7] J. Shendure and H. Ji, “Next-generation DNA sequencing,”
Nature Biotechnology, vol. 26, no. 10, pp. 1135–1145, 2008.
[8] P. Kovatcheva-Datchary, E. G. Zoetendal, K. Venema, W. M. de
Vos, and H. Smidt, “Tools for the tract: understanding the func-
tionality of the gastrointestinal tract,” Therapeutic Advances in
Gastroenterology, vol. 2, no. 4, pp. 9–22, 2009.
[9] T. J. Lavery, B. Roudnew, J. Seymour, J. G. Mitchell, and T.
Jeffries, “High nutrient transport and cycling potential revealed
in the microbial metagenome of Australian sea lion (Neophoca
cinerea) faeces,” PLoS One, vol. 7, no. 5, Article ID e36478, 2012.
[10] D. M. Akob, H. J. Mills, and J. E. Kostka, “Metabolically active
microbial communities in uranium-contaminated subsurface
sediments,” FEMS Microbiology Ecology, vol. 59, no. 1, pp. 95–
107, 2007.
[11] J. R. Cole, Q. Wang, E. Cardenas et al., “The Ribosomal Database
Project: improved alignments and new tools for rRNA analysis,”
Nucleic Acids Research, vol. 37, supplement 1, pp. D141–D145,
2009.
[12] P. D. Schloss, S. L. Westcott, T. Ryabin et al., “Introduc-
ing mothur: open-source, platform-independent, community-
supported software for describing and comparing microbial
communities,” Applied and Environmental Microbiology, vol. 75,
no. 23, pp. 7537–7541, 2009.
[13] P. D. Schloss, D. Gevers, and S. L. Westcott, “Reducing the effects
of PCR amplification and sequencing artifacts on 16s rRNA-
based studies,” PLoS ONE, vol. 6, no. 12, Article ID e27310, 2011.
[14] J. G. Kusters, A. H. van Vliet, and E. J. Kuipers, “Pathogenesis
of Helicobacter pylori infection,” Clinical Microbiology Reviews,
vol. 19, no. 3, pp. 449–490, 2006.
[15] K. Goto, H. Ohashi, S. Ebukuro et al., “Isolation and character-
ization of Helicobacter species from the stomach of the house
musk shrew (Suncus murinus) with chronic gastritis,” Current
[16] F. E. Dewhirst, C. Seymour, G. J. Fraser, B. J. Paster, and J. G. Fox,
“Phylogeny of Helicobacter isolates from bird and swine feces
and description of Helicobacter pametensis sp. nov,” Interna-
tional Journal of Systematic Bacteriology, vol. 44, no. 3, pp. 553–
560, 1994.
[17] L. J. Mappley, M. A. Tchorzewska, A. Nunez, M. J. Woodward,
and R. M. La Ragione, “Evidence for systemic spread of the
potentially zoonotic intestinal spirochaete Brachyspira pilosicoli
in experimentally challenged laying chickens,” Journal of Medi-
cal Microbiology, vol. 62, part 2, pp. 297–302, 2013.
[18] F. Jorgensen, J. Ellis-Iversen, S. Rushton et al., “Influence of sea-
son and geography on Campylobacter jejuni and C. coli subtypes
in housed broiler flocks reared in Great Britain,” Applied and
[19] R. B. Williams, R. N. Marshall, M. Pagès, M. Dardi, and E. del
Cacho, “Pathogenesis of Eimeria praecox in chickens: virulence
of field strains compared with laboratory strains of E. praecox
and Eimeria acervulina,” Avian Pathology, vol. 38, no. 5, pp. 359–
366, 2009.
http://dx.doi.org/10.1155/2013/470867
Research Article
Microbial Adhesion and Biofilm Formation on
Microfiltration Membranes: A Detailed Characterization Using
Model Organisms with Increasing Complexity
L. Vanysacker,1 C. Denis,2 P. Declerck,2 A. Piasecka,1 and I. F. J. Vankelecom1

1
Centre for Surface Chemistry and Catalysis, KU Leuven, Kasteelpark Arenberg 23, P.O. Box 2461, 3001 Heverlee, Belgium
2
Laboratory of Aquatic Ecology and Evolutionary Biology, KU Leuven, Charles Deberiotstraat 32, 3000 Leuven, Belgium
Correspondence should be addressed to L. Vanysacker; louise.vanysacker@bio.kuleuven.be
Received 2 April 2013; Accepted 24 June 2013
Copyright © 2013 L. Vanysacker et al. This is an open access article distributed under the Creative Commons Attribution License,
Since many years, membrane biofouling has been described as the Achilles heel of membrane fouling. In the present study,
an ecological assay was performed using model systems with increasing complexity: a monospecies assay using Pseudomonas
aeruginosa or Escherichia coli separately, a duospecies assay using both microorganisms, and a multispecies assay using activated
sludge with or without spiked P. aeruginosa. The microbial adhesion and biofilm formation were evaluated in terms of bacterial
cell densities, species richness, and bacterial community composition on polyvinyldifluoride, polyethylene, and polysulfone
membranes. The data show that biofouling formation was strongly influenced by the kind of microorganism, the interactions
between the organisms, and the changes in environmental conditions whereas the membrane effect was less important. The findings
obtained in this study suggest that more knowledge in species composition and microbial interactions is needed in order to
understand the complex biofouling process. This is the first report describing the microbial interactions with a membrane during
the biofouling development.
1. Introduction microorganisms, since bacterial attachment is a necessary

first step for membrane biofilm formation [5]. Bacterial
Bacterial biofilms are ubiquitous in the environment and can adhesion by one species on the membrane surface has
be found on almost any hydrated surface. Whilst many of been extensively studied. However, membrane biofouling as
these communities can be used for the synthesis of valuable it occurs in purification systems is much more complex
products and the treatment of waste, others can cause as species may interact additionally with each other. For
problems in industrial and/or medical applications [1]. Very example, it is known that when species are occurring in
often in man-made systems, unwanted biofilm formation is a biofilm, they behave differently by changing their gene
referred to as biofouling [2]. Biofouling communities present expression and growth rate [6]. They undergo a transition
in wastewater treatment systems, like for example, membrane from a planktonic (“loner”) to a community-based existence
applications, are extremely complex and contain noncellular in which they interact with various bacterial species in close
material such as minerals, corrosion particles, clay, or silt par- proximity [7]. Some species can coexist and cooperate by
ticles but also, most importantly, numerous bacterial species exchanging, for example, metabolic products, while they may
and other microorganisms [3]. Ranging from membrane outcompete each other when nutrients become scarce [8, 9].
bioreactor (MBR) to reverse osmosis systems, the biofouling From a membrane biofouling point of view, it is important
processes are the same: cells attach to the membrane, start to to understand these bacterial interactions. This will not only
multiply, produce exopolymeric substances (EPS), and finally lead to a complete comprehension of what exactly happens
block the membrane pores [4]. One of the most important on the membrane surface, but it will also allow to improve
issues of membrane biofouling is the initial attachment of filtration processes and fouling control strategies. The present
paper reports on the adhesion and biofilm formation of three protein (rfp) labelled (kindly given by Prof. N. Boon, UGent,
types of biofoulants on membranes that are ubiquitously used Belgium) were cultivated in a shaking incubator at 28∘ C in
in membrane bioreactors. The objective of this study is to Lysogeny Broth (LB) (10 g tryptone, 5 g yeast extract, and
gain insight in to the mutual bacterial interactions and their 5 g NaCl per liter). P. aeruginosa cultures were supplemented
interactions with the membrane surface. The biofoulants with 100 mg L−1 ampicillin, 10 mg L−1 gentamycin, 25 mg L−1
complexity increases stepwise, starting from monospecies kanamycin, and 25 mg L−1 streptomycin [19]. E. coli was
Pseudomonas aeruginosa over duospecies P. aeruginosa and cultured using 50 mg L−1 kanamycin and 100 mg L−1 nalidixic
Escherichia coli to diluted activated sludge spiked with P. acid. At the start of each experiment, cells from a 24 h culture
aeruginosa and finally activated sludge. were harvested and suspended in sterile Ringer’s solution
(Oxoid). The initial cell concentration was spectrophotomet-
2. Material and Methods rically (𝜆 650 nm ) determined, and countable 10-fold dilutions
were additionally plated as a control [20].
2.1. Experimental Design. In order to analyze the adhesion Activated sludge originating from a lab-scale MBR fed
and biofilm formation of the microorganisms, four types with molasses based wastewater was used. The operational
of static experiments using an increasing biofoulant com- parameters of the MBR and the sludge characteristics are
plexity were performed (Figure 1). At first, a static assay in detail described by Piasecka et al. [21]. During the third
was performed in test tubes using two model monospecies experiment (Figure 1(c)), activated sludge was diluted 1000
suspensions of P. aeruginosa and E. coli (Figure 1(a)). Sec- times and spiked with 104 P. aeruginosa cells mL−1 .
ondly, the two model organisms were mixed in an equal
concentration and used as biofoulant in a similar test tube 2.3. Membrane Preparation. Two lab-made and one com-
set-up (Figure 1(b)). In a third experiment, the biofouling mercial membrane were used. PVDF and PSF microfiltra-
capacity of diluted activated sludge spiked with P. aeruginosa, tion membrane sheets were produced via the phase inver-
both in approximately equal concentration,was investigated sion procedure [22–24]. Briefly, a 12 wt% PVDF (Sigma-
in a culture vessel (Figure 1(c)). Finally, a fourth exper- Aldrich) and 10 wt% PSF (BASF-Ultrason) solution was
iment was performed in a lab-scale MBR reactor where prepared in dimethylacetamide (Sigma-Aldrich) and N-
activated sludge was used as biofoulant (Figure 1(d)). For Methyl-2-pyrrolidone (ACROS) solutions, respectively. Sub-
every experiment, the initial adhesion and the subsequent sequently, the membranes were cast (250 𝜇m thickness) on
biofilm formation were evaluated by an incubation time of a polypropylene/polyethylene support (Viledon nonwoven
1 h and 24 h, respectively. In addition, a third condition was FO 2471, kindly supplied by Freudenberg, Germany). The
created in the lab-scale MBR assay where the membranes solvents were allowed to evaporate during 60 s, followed by
were incubated in Ringer’s solution to mimic oligotrophic coagulation of the polymer film in deionized water. The
conditions (Figure 1(d)). PE membrane was purchased from Kubota (cartridge type
During every experiment three microfiltration mem- 203). These flat sheet membranes were made of chlorinated
branes, commonly used in MBR systems [10], were tested: PE with a nonwoven cloth base. Out of every membrane
polyvinyldifluoride (PVDF), polyethylene (PE), and polysul-
sheet, replicate coupons of 4.5 cm2 were cut. Each coupon
fone (PSF) membranes. To allow accurate statistical analysis,
was sterilized during three hours in 70% EtOH, followed by
at least three replicate membranes for every experiment
two rinsing steps using sterile deionized water to remove
were used. For the mono- and duospecies experiments, P.
the remaining EtOH. Each sterile membrane coupon was
aeruginosa and E. coli were quantified using standard plate
stored in a 50 mL falcon tube filled with 40 mL sterile dH2 O
count. In the third experiment, P. aeruginosa was quantified
water at 4∘ C until further use. Each membrane coupon
using P. aeruginosa gene specific quantitative real-time PCR
was used only once. The nominal pore size and porosity of
(qPCR) [11]. In experiment three and four, the activated
the membranes were determined using scanning electron
sludge bacterial community composition was determined by
microscopy (SEM, Philips SEM XL30 FEG with Adax dx-4i
bacterial fingerprints using Automated Ribosomal Intergenic
system) and image processing software (ImageJ) and were
Spacer Analysis (ARISA) [12], and the total amount of cells
compared with data from the literature (Figure 2) [25, 26].
were quantified by 16S rRNA qPCR [13].
The membrane surface hydrophobicity was determined using
contact angle goniometry (VCA Optima video camera sys-
2.2. Model Microorganisms and Culture Conditions. As men-
tem, AST Products, Billerica) with the sessile drop method.
tioned in the experimental design, P. aeruginosa, E. coli, and
The contact angle was measured at least at five different
activated sludge were used as biofoulants. Reasons why P.
positions and at different time points: immediately after the
aeruginosa and E. coli have been chosen as model organisms
drop reached the membrane and 10 min later (Table 1).
are: (1) they are both Gram-negative bacteria, (2) they have
approximately the same growth rate and hydrophobicity, (3)
they are commonly present throughout aquatic environments 2.4. Experimental Procedure
[14] and activated sludge [15], and (4) they are the best 2.4.1. Bacterial Adhesion and Biofilm Formation of P. aerug-
studied model organisms for biofilm formation since years inosa and E. coli under Monospecies and Duospecies Con-
[16–18]. P. aeruginosa (PA14) green fluorescent protein (gfp) ditions. Membrane pieces of 4.5 cm2 were immersed in
labelled (kindly given by Prof. B. Koch, Technical Univer- LB medium containing kanamycin (25 mg L−1 ). After five
sity of Denmark) and E. coli (LMG 2092T ) red fluorescent minutes conditioning, the specific bacterial species was/were
1h 1h
Adhesion
Monospecies Monospecies Adhesion
P. aeruginosa E. coli Duospecies
24 h P. aeruginosa + E. coli 24 h
Biofilm Biofilm
Membrane Membrane
(a) (b)
Adhesion Biofilm Ringer
1h 24 h 1 h → 24 h
1h
Adhesion
Membrane
24 h
Biofilm
Membrane
Culture vessel
P. aeruginosa + diluted activated sludge Lab-scale MBR
(c) (d)
Figure 1: Graphical illustration of the experimental design used in this study. In the first experiment, the adhesion and biofilm formation of
monospecies P. aeruginosa and E. coli were measured on PE, PSF, and PVDF membranes. (a) In a second experiment, the two microorganisms
were mixed, and the adhesion and biofilm formation were enumerated. (b) The third experiment was performed in a culture vessel. Diluted
activated sludge spiked with P. aeruginosa was used as model biofoulants. (c) The fourth experiment used real activated sludge as biofoulant
and was done in a lab-scale MBR. In this set up, the microbial adhesion, the biofilm formation in the reactor, and the biofilm formation in
Ringer’s solution after 1 h adhesion in the reactor were measured (d).
Table 1: Characteristics of the membranes used in this study.
Contact angle (∘ )
Type Pore size (𝜇m) Surface porosity (%)
5 sec 10 min
PSF Labmade a a 74 ± 1 71 ± 2
PVDF Labmade 0.01 19 73 ± 7 49 ± 2
PE Kubota 0.4 25 103 ± 2 73 ± 3
a
Not possible to determine with ImageJ.
added. A cell concentration of 108 cells mL−1 of each microor- a culture vessel with a working volume of 12 L (Nalgene,
ganism was used in the mono- and duospecies experiments Thermo Scientific) using activated sludge harvested from
[27–29]. To avoid sedimentation of the cells on the mem- a lab-scale MBR [21]. Prior to inoculation, the bacterial
brane, the membranes were placed in a vertical position in density in the biomass was determined in order to have an
small screw capped tubes during 1 h at 20∘ C on a rotary shaker idea of the number of cells present in the activated sludge.
(4 g). Hereafter, the membranes were removed, washed in This was done by 16S rRNA qPCR (see section “Biofouling
sterile Ringer’s solution to remove the loosely bound cells, community characterization and cell quantification”). The
and placed in a new sterile tube [30]. The adhered cells total cell density was approximately 109 cells mL−1 . To
were harvested and quantified, as described in the section gradually increase the biofoulant complexity and see the
“Biofouling community characterization and cell quantifica- effect of spiked P. aeruginosa cells, the sludge was diluted
tion.” In order to observe further biofilm development and 1000 times using synthetic wastewater (detailed composition
since bacterial cells are more likely to attach to surfaces in is described by Piasecka et al. [21]). The cells were allowed to
a low nutrient environment [29, 31], the washed membranes adapt and acclimatize to their new environment by mixing
containing adhered cells were incubated for a second time the suspension during 24 h. Thereafter, the 105 cells mL−1
in Ringer’s solution to create oligotrophic circumstances. sludge cells were spiked with approximately 104 P. aeruginosa
During this step, the membranes were incubated for 24 h cells mL−1 and mixed during 10 min, and finally the mem-
at 20∘ C on a rotary shaker (4 g). After a washing step, the brane pieces were placed in the culture vessel. After 1 h and
cells were removed from the membrane and quantified as 24 h incubation, the adhesion and biofilm formation were,
described in section “Biofouling community characterization respectively, enumerated as described in section “Biofouling
and cell quantification.” community characterization and cell quantification.”
2.4.2. Adhesion and Biofilm Formation by P. aeruginosa Spiked 2.4.3. Adhesion and Biofilm Formation by Activated Sludge in
Activated Sludge. The third experiment was performed in a Lab-Scale MBR. In the fourth experiment, the attachment
(a) (b) (c)
Figure 2: SEM pictures of pristine PE (a), PSF (b), and PVDF (c) membranes. The scale bar represents 10 𝜇m.
and biofilm formation of microorganisms present in acti- (expressed as bacterial cell density) and bacterial species rich-
vated sludge were investigated (Figure 1(d)). A description ness (total number of Operational Taxonomic Units (OTU)
of the lab-scale MBR can be found in Bilad et al. [32] detected by ARISA) were further analyzed with experimen-
and on http://www.biw.kuleuven.be/cok/membrane2/. The tal condition (mono- versus duospecies (Figures 1(a) and
operational parameters and wastewater characteristics are 1(b)); adhesion versus biofilm versus Ringer (Figure 1(d)))
mentioned by Piasecka et al. [21]. Three replicates of each and membrane type as explanatory variables. Bacterial cell
membrane type were randomly immersed in the reactor densities were logarithmically transformed prior to analysis.
tank. The membranes were incubated during 1 h in order to Being mainly interested in the interaction between species
measure the cell adhesion. After a washing step in Ringer’s and membranes, the significant main effects were further
solution, the cells on the membranes were removed (see explored by comparing all combinations using post hoc Tukey
below), the total amount of cells was quantified, and bacterial honest significant difference tests. All univariate analyses
community fingerprints were generated, as described in the were performed with Statistica 11.0 software.
following section. In addition, two biofilm formation assays Concerning the bacterial community data analysis, the
with different environmental conditions were performed: (1) associations among the community composition (height data
after 1 h incubation in the reactor, the washed membranes of OTUs) were explored firstly using Principal Component
were further incubated in sterile Ringer’s solution (olig- Analyses (PCA). Secondly, the relationship between the
otrophic biofilm condition) since bacterial cells are more observed bacterial community composition (height data
likely to attach to surfaces in a low nutrient environment of OTUs) and the experimental factors (membrane type
[29, 31] and (2) the membranes were further incubated during and experimental condition (adhesion versus biofilm
24 h in the reactor itself to compare the biofilm formation (Figure 1(c))); adhesion versus biofilm versus Ringer
with the oligotrophic biofilm condition (Figure 1(d)). (Figure 1(d))) was explored by means of Redundancy
Analysis (RDA). RDA is a direct gradient ordination
2.4.4. Biofouling Community Characterization and Cell Quan- technique that allows to formally test for the significance
tification. The biofouling was harvested in the same way of experimental factors on community data, as well as the
for the four experiments and for the adhesion and biofilm interaction effects among factors [35]. Significance tests
formation (Figure 1). Briefly, cells were removed from the were performed with random Monte Carlo permutation tests
membrane by scraping using a cell scraper, sonication (9999 unrestricted permutations per test). Both analyses were
(10 min, 42 kHz ± 6%, Bransonic 2510), and vortex (30 s, max performed using the software package Canoco for Windows,
speed) in 10 mL Ringer’s solution [33]. Biofouling associated version 4.5 (Biometris Plant Research International).
with P. aeruginosa and E. coli were quantified by the standard
plating method using LB medium with the required antibi- 3. Results
otics. Activated sludge biofouling was characterized using 3.1. P. aeruginosa and E. coli: Monospecies Condition. Using a
DNA-based assays. For this end, DNA of the acquired cell static assay in test tubes, the adhesion rate and biofilm forma-
suspension was extracted using the Mobio Ultraclean soil tion of P. aeruginosa and E. coli were measured (Figure 1(a)).
DNA kit (Cambio) [34]. Afterwards, the total amount of The bacterial cell densities are shown in Figure 3(a). No
cells was quantified by 16S rRNA qPCR [13], and bacterial membrane effect was found for E. coli during the adhesion
community fingerprints were generated by ARISA [12]. The (one-way ANOVA, 𝑃 = 0.099) and biofilm formation (one-
spiked P. aeruginosa in activated sludge was measured using way ANOVA, 𝑃 = 0.636). On the other hand, a small
a species specific qPCR, as described by Shannon et al. [11]. but significant membrane effect was found during the P.
aeruginosa adhesion (one-way ANOVA, 𝑃 = 0.024). A
2.4.5. Data Analysis. Depending on the experimental set- better attachment was detected on the PVDF membrane in
up, different statistical analyses were performed. By one-way comparison with the PE membrane (post hoc Tukey test, 𝑃 =
analyses of variance (ANOVA), membrane effects were tested 0.020, Figure 3(a)).
on the bacterial cell densities for each species and condition The cell increase during the biofilm formation was rather
separately. By two-way ANOVA, the amount of extracted cells low for both microorganisms. P. aeruginosa showed a small
8 8
Bacterial cell density (log cell cm−2 ) ∗
Bacterial cell density (log cell cm−2 )

∗
6 6
4 4
2 2
0 0
Adh E. coli Adh P. aer Bio E. coli Bio P. aer Adh E. coli Adh P. aer Bio E. coli Bio P. aer
PE PE
PSF PSF
PVDF PVDF
(a) Monospecies condition (b) Duospecies condition
−2
Figure 3: Bacterial cell density expressed in log cells cm membrane after adhesion and biofilm formation of P. aeruginosa and E. coli in
monospecies (a) and duospecies (b) conditions on PE, PSF, and PVDF membranes. Cells were enumerated using standard plating and log
transformed before analysis. Error bars represent standard deviation (𝑛 = 3). P. aer = P. aeruginosa, Adh = adhesion, Bio = Biofilm, and ∗ =
significantly different.
increase in cells, especially on the PE membrane. On the Table 2: Two-way factorial ANOVA results of the duospecies
other hand, E. coli was not able to produce a biofilm as lower experiment using P. aeruginosa and E. coli. Bacterial cell density
cell densities were found in comparison with the adhesion. (log cells cm−2 membrane) was selected as dependent variable and
When the two model organisms are mutually compared, P. membrane type (PSF, PVDF, and PS) and bacterial coexistence
aeruginosa adhered approximately 100 times more than E. coli (mono versus duospecies) as predictor variables.
on each membrane type (Figure 3(a)). 𝐹 𝑃
Duospecies experiment—adhesion
3.2. P. aeruginosa and E. coli: Duospecies Condition. In the E. coli
second experiment (Figure 1(b)), the adhesion and biofilm
Membrane 1.004 0.391
development was examined in a duospecies condition. The
Coexistence (mono- and duospecies) 26.63 <0.001
“coexistence effect” was analyzed by two-way ANOVA with
coexistence and membrane type as explanatory variables. Membrane ∗ coexistence 2.083 0.162
For the coexistence variables, the values of the cell densities P. aeruginosa
measured during the mono- and duospecies experiments Membrane 8.50 0.005
were used. Coexistence (mono- and duospecies) 0.11 0.744
A significant coexistence effect was found during the E. Membrane ∗ coexistence 2.11 0.164
coli attachment (Table 2, 𝑃 < 0.001), meaning that E. coli Duospecies experiment—biofilm
was less able to bind to the membranes when P. aeruginosa E. coli
was present. On the other hand, the adhesion of P. aeruginosa Membrane 10.03 0.967
was not limited when E. coli was present as no significant Coexistence (mono- and duospecies) 11.41 0.005
coexistence effect was found (𝑃 = 0.74). During the adhesion
Membrane ∗ coexistence 1.513 0.263
step, a membrane effect was found for P. aeruginosa (Table 2,
P. aeruginosa
𝑃 = 0.005). A post hoc Tukey test showed that this effect
was mainly attributed to the differences between PE and Membrane 1.274 0.321
PVDF in the monospecies experiment (𝑃 = 0.02), which has Coexistence (mono- and duospecies) 5.639 0.032
been mentioned in the previous section. Different significant Membrane ∗ coexistence 1.937 0.198
combinations were found for E. coli (statistical data not Significant values are in bold. Bacterial cell density data were log transformed
shown), but as no main membrane effect was found (Table 2, before analysis.
𝑃 = 0.39), these differences are attributed to the strong
coexistence effect (mono versus duospecies). E. coli. Also the other way around, the biofilm formation of
When looking at the biofilm formation (Figure 3(b)), a P. E. coli was still hampered when P. aeruginosa was present.
aeruginosa coexistence effect showed up (Table 2, 𝑃 = 0.03), For both species, no pairwise combination effects were
meaning that, in contrast with the attachment, the biofilm found (statistical data not shown). This underlines the poor
formation of P. aeruginosa is affected by the presence of membrane effect.
6 ∗ 40
Richness (species diversity membrane−1 )

PE
Feed
5 ∗
30
4
PE
3 20 PE
PSF PE
PE PE PSF
PCA 2 21.8%
2 Feed
PSF PVDF PVDF PE
10
PE
1 PVDF
PE PVDF
PVDF
0 0
PE
Adh 16S
Bio 16S
Adh
Bio
Bio P. aer
Adh P. aer
Feed PE
Feed
PE
PSF
PE
PVDF
Figure 4: Bacterial cell density (left axis) and species richness (right PCA 1 32.6%
axis) expressed, respectively, as log cells and species diversity cm−2
Figure 5: PCA ordination diagram illustrating the relationships
membrane when diluted activated sludge spiked with P. aeruginosa
among the bacterial communities found on the PE, PSF, and PVDF
was used as biofoulant. The bacterial cell density was enumerated
membranes. The diagram is based in the height data of the ARISA
with 16S and P. aeruginosa specific qPCR. The species richness was
data. Samples corresponding to the adhesion data are represented
calculated as the sum of the number of peaks (OTU’s) within each
by squares and samples from the biofilm data by diamonds. The
ARISA electropherogram. Error bars represent standard deviation
bacterial community in the culture vessel (feed) is plotted as
(𝑛 = 3). P. aer = P. aeruginosa, Adh = adhesion, Bio = Biofilm, and ∗
triangles. The diagram shows the first and second PCA axes, that,
= significantly different.
respectively, account for 32.6% and 21.8% of the explained variability.
3.3. Activated Sludge Spiked with P. aeruginosa . To gradually culture vessel was measured as 35 ± 3.9 OTUs mL−1 . The
increase the biofoulant complexity, activated sludge was PVDF membrane had for the biofilm condition a high species
diluted to approximately 105 cells mL−1 and spiked with richness and was significantly different from the PSF mem-
104 P. aeruginosa cells mL−1 (Figure 1(c)). To have an idea brane (statistical data not shown). No clear differences were
how many bacterial cells (16S rRNA) were attached in total found between the adhesion and biofilm species richness data
and what the proportion of P. aeruginosa was within the on the same membrane.
activated sludge community on each membrane, both cell In addition to the species richness, bacterial community
types were quantified after 1 and 24 h using qPCR (Figure 4). profiles were generated using the ARISA dataset and were
The bacterial cell density in the culture vessel after 24 h analyzed and visualized by RDA and PCA, respectively. In
filtration was 7.70 ± 2.29× 103 P. aeruginosa cells mL−1 and contrast with the results found using the species richness, no
4.29 ± 0.78× 107 total bacteria cells mL−1 . membrane effect was found (𝑃 = 0.08). Also the adhesion
Two-way ANOVA models were performed on the adhe- and biofilm communities were similar (𝑃 = 0.294). These
sion and biofilm data separately using membrane and organ- results are also visible on the PCA plot (Figure 5), where the
ism (P. aeruginosa and 16S cell) type as variables. All the adhesion and biofilm samples are randomly spread. Although
variables were significantly different (Table 3). During the the membrane effect was not significant, most of the PVDF
adhesion and the biofilm formation, high cell densities were membrane samples have the tendency to group together.
found on the PVDF membranes especially when the universal These profiles might explain the significant effects found
16S primers were used (Figure 4). using the bacterial density and species richness data.
P. aeruginosa was well represented on the membrane
surface in comparison to the other community members. 3.4. Activate Sludge in a Lab-Scale MBR. The results of the
Twenty-one and 38% of the microbial population were P. total amount of cells, quantified by 16S rRNA qPCR, are
aeruginosa during the adhesion process on the PE and PSF shown in Figure 6(a). The data was further analyzed with
membrane, respectively. Also during the biofilm formation, a two-way ANOVA. A significant interaction (𝑃 = 0.047)
P. aeruginosa was able to attach to the PE and PSF membrane was observed (Table 4), which could be assigned to the high
but in smaller quantities: 16 and 13% of the population were cell densities found in the PSF-biofilm condition (Figure 6).
assigned as P. aeruginosa, respectively. The 16S cell densities These results are highlighted by the post hoc Tukey test in
were much higher on the PVDF. This led to a lower P. Table 5. No analogue results were found during the adhesion
aeruginosa proportion: 2.3 and 5.3% during the adhesion and or biofilm formation in Ringers’ solution on the other PE
biofilm formation, respectively (Figure 4). membranes (Figure 6).
Similar, but less pronounced, results were found using The ARISA dataset was used to compare the over-
the species richness (Figure 4). The species richness in the all community structure among samples. The generated
4 400
∗
Richness (species diversity membrane−1 )

3 300
2 200
1 100
0 0 †
Adhesion Ringer Biofilm Adhesion Ringer Biofilm
PE PE
PSF PSF
PVDF PVDF
(a) (b)
Figure 6: Bacterial cell density (a) and species richness (b) expressed, respectively, as log cells and species diversity cm−2 membrane after
adhesion and biofilm formation of activated sludge on PE, PSF, and PVDF membranes. Bacterial cell densities were enumerated using 16S
rRNA qPCR and log transformed before analysis. The richness was calculated by the sum of the number of peaks (OTU’s) within each ARISA
electropherogram. Results for the adhesion on the PE membrane are not available due to repeated unsuccessful PCR amplification († ). Error
bars represent standard deviation (𝑛 = 3). Adhesion = 1 h incubation in the lab-scale MBR, Ringer = 1 h incubation in the lab-scale MBR
followed by 24 h biofilm in Ringer’s solution, biofilm = 24 h incubation in the lab-scale MBR, and ∗ = significantly different.
Table 3: Two-way ANOVA results of the activated sludge exper- Table 4: Two-way factorial ANOVA and post hoc Tukey results
iment spiked with P. aeruginosa in the culture vessel. Bacterial of the activated sludge experiment in the lab-scale MBR. Bacterial
cell density (log cells cm−2 membrane) was selected as dependent cell density (log cells cm−2 membrane) was selected as dependent
variable and membrane (PSF, PVDF, and PS) and organism (P. variable and membrane type (PSF, PVDF, and PS) and treatment
aeruginosa and 16S) as predictor variables. (adhesion, biofilm formation in Ringer’s solution, and biofilm
formation in lab-scale MBR) as predictor variables. The pairwise
𝐹 𝑃 comparison was performed between the bacterial cell density (log
Spiked activated sludge experiment—adhesion cells cm−2 membrane) among the different membrane types and
Membrane 38.77 <0.001 treatment. Bacterial cell density data were log transformed before
Organism (P. aeruginosa and 16S) 39.26 <0.001 analysis.
Membrane ∗ organism 6.413 0.007 Activated sludge experiment in lab-scale MBR 𝐹 𝑃
Spiked activated sludge experiment—biofilm Membrane 3.030 0.073
Membrane 31.67 <0.001 Treatment (adhesion-Ringer-biofilm) 6.419 0.008
Organism (P. aeruginosa and 16S) 6.532 0.016 Membrane ∗ treatment 2.972 0.047
Membrane ∗ organism 11.96 <0.001 Post hoc Tukey test
Significant values are in bold. Bacterial cell density data were log transformed PSF biofilm-PE biofilm 0.016
before analysis.
PSF biofilm-PE Ringer 0.006
PSF biofilm-PSF Ringer 0.034
electropherograms delivered OTUs ranging from 224 to
PSF biofilm-PVDF Ringer 0.011
1148 bp. The majority of them had a length between 243 and
1086 bp. The species richness, which is the sum of the OTUs
in each sample, is shown in Figure 5 (right). Due to repeated
unsuccessful PCR amplification, PE-adhesion data were not densities, RDA showed a significant treatment and inter-
available, and no statistical tests using the species richness action effects (membrane∗treatment) (Table 5). The treat-
were performed. A high variation inside each treatment ment effect explained 36% of the total bacterial community
(adhesion, Ringer, biofilm) and each membrane was found variation compared with only 8% for the, non significant,
(Figure 5). Also the samples processed after 1 h adhesion and membrane effect. Moreover, the interaction effect was more
24 h biofilm formation in the MBR showed a much higher pronounced than the main effects of the experimental factors,
species diversity compared to the Ringer treatment. as it explained 43% of the community variability. These results
Using RDA, the effects of membrane and treatment are also visible in Figure 7. The different conditions have the
(adhesion, Ringer and biofilm) type on the bacterial tendency to group together whereas the membrane types are
community profiles were explored. Similar as with the cell more randomly dispersed in the community profile.
Table 5: Results of RDA permutation analyses on bacterial community data derived from ARISA, testing for the effects of membrane type
(PE, PSF, and PVDF), treatment (adhesion, Ringer, biofilm), and the interaction between both factors. 𝜆 1 and 𝜆 2 represent the eigenvalues of
the first and second axes. 𝜆 tot represents the percentage of total community variation explained by the experimental factors.
𝜆1 𝜆2 𝜆 tot 𝐹 𝑃
Static assay in lab-scale MBR
Membrane 0.066 0.014 8% 1.062 0.382
Treatment (adhesion-Ringer-biofilm) 0.247 0.113 36% 4.809 <0.001
Membrane ∗ treatments 0.263 0.114 43% 2.161 0.007
Significant values are in bold.
hardly any membrane effect could be observed. Only during

Ringer P. aeruginosa adhesion, a slightly higher affinity was found
PSF PE PE
for the PVDF membrane in comparison to the PE membrane
PVDF
PE ∗ Ringer
(Figure 3(a)), but this membrane effect disappeared after 24 h
PSF ∗ Ringer PVDF ∗ adhesion
PSF PVDF Adhesion
of further incubation, and a higher P. aeruginosa number
PVDF ∗ Ringer PSF
PVDF
PSF
PSF
PSF ∗ adhesion were then detected on the PE membrane.
𝜆 2 11%
PSF
PVDF
PSF 4.2. P. aeruginosa and E. coli: Duospecies Condition. Once
PSF
microorganisms are strongly attached to the membrane, and
PVDF PE
PE after the EPS production is activated, the microorganisms
PVDF ∗ biofilm PE PE ∗ biofilm
PSF ∗ biofilm
eventually become embedded in the hydrated matrix and
PVDF are immobilized. As a result, the cells are influenced by
the exchange of nutrients with neighboring cells in the
Biofilm
biofilm. An important feature is that the microorganisms
PSF
are immobilized in relatively close proximity to one another,
becoming continuously confronted with dynamic changes.
Therefore, the success rate of any given microorganism in
𝜆 1 26%
a multispecies biofilm strongly depends on the behavior
of the other participants [39]. A first factor determining
Figure 7: RDA ordination diagram based on the bacterial ARISA the fitness of a bacterial strain with respect to a given set
data, representing the interaction between membrane type and of nutrients is its growth rate [40]. Although the growth
treatment (B). 𝜆 1 represents the eigenvalue axis 1; 𝜆 2 represents rates of our two model organisms, at least in monoculture
the eigenvalue axis 2. Arrows represent significant environmental (approximately 26 and 31 min for P. aeruginosa and E. coli,
variables superimposed onto the ordination; the length of the arrow resp.), were similar, a higher attachment for P. aeruginosa was
indicates the correlation between the environmental variable and found. Approximately 100 and 1000 times more P. aerugi-
the ordination axis. Adhesion data (dots): 1 h incubation in the lab- nosa cells were quantified in the adhesion (Figure 3(a)) and
scale MBR, Ringer data (squares): 1 h incubation in the lab-scale
biofilm (Figure 3(b)) assay, respectively. A second possible
MBR followed by 24 h biofilm in Ringer’s solution, and biofilm data
(diamonds): 24 h incubation in the lab-scale MBR.
reason is the difference in motility. Indeed, the ability to
attach or to detach from a surface and the possibility to
migrate over the substrate are features affecting the biofoul-
ing potential of species. P. aeruginosa and E. coli possess
4. Discussion polar monotrichous and peritrichous flagellation, respec-
tively, allowing swarming/swimming motility [41]. However,
4.1. P. aeruginosa and E. coli: Monospecies Condition. Initial compared to the peritrichous flagella observed in E. coli,
bacterial adhesion to abiotic surfaces is likely to be highly the polar flagellum in P. aeruginosa is driven by sodium
dependent on physicochemical and electrostatic interactions motive force and may then propel the bacterium at relatively
between the bacterial envelope itself and the substrate, which high speeds towards the membrane. In contrast, E. coli cells,
is often conditioned by the fluids to which it is exposed propelled by numerous proton-powered lateral flagella, seem
[36]. Generally, for aqueous applications, fouling is less to be effective for attachment and biofilm formation [42]
pronounced on hydrophilic membranes than on hydrophobic but possibly not powerful enough in comparison with the
membranes [29, 37, 38]. The PE membrane had initially a motility of P. aeruginosa. Another distinct type of motility,
high hydrophobicity, but the contact angle changes rapidly named twitching, is mediated by type IV pili located at one
in time as water penetrated inside the membrane pores or both poles of the cell [41]. Although some pathogenic E.
during the measurement. This phenomenon occurred also on coli strains contain a related type IV motility system [43],
the PVDF membrane. The PSF membrane had the highest twitching motility did not occur in the LMG strain used
stability (Table 1). Normally, a higher affinity was expected here. The differences in motility between the two species
for the PE and PSF membranes. This was not the case, since may explain the adhesion differences found in Figure 3. Also
the fact that no condition effect (Table 2) was found for P. As this phenomenon was observed during the adhesion
aeruginosa during the adhesion suggests that P. aeruginosa and the biofilm formation, this may suggest that some
can outcompete the attachment of E. coli. species present in the culture vessel had a high affinity for
Interesting are the differences found between adhesion this polymer. Although no significant membrane effect was
and biofilm formation. Here, a decrease in E. coli cells after found using RDA, the PVDF samples had the tendency to
24 h incubation in oligotrophic circumstances was found, cluster together in the PCA plot (Figure 5). This suggests
which is in contrast with the increase of P. aeruginosa that a slightly different community was present on the
in monospecies condition (Figure 3(a)). This suggests that, PVDF membrane. Quite interesting is that at the time this
although E. coli could attach to the membranes, the bacterium experiment was performed, the lab-scale MBR, where the
was not able to stay on it and produce a biofilm. This is sludge was harvested from, was running using only PVDF
quite surprising as biofilm formation is normally enhanced membranes. Perhaps some species developed the ability to
by environmental stress in many bacterial species and seems bind on this polymer and formed a biofilm on it, and maybe
to be promoted by nonoptimal growth conditions or even this can explain why a higher species richness was found in
by cellular stresses [44], here mimicked by the oligotrophic comparison with PSF and PE. During the fourth experiment,
Ringer’s solution. The decrease in E. coli cells was even more which was performed in the lab-scale MBR itself (Figure 1(d),
pronounced when both species were mixed (Figure 3(b), discussed in the following section), the reactor was running
Table 2 coexistence effect), suggesting that rather antagonistic without membrane, and this might be a reason why no clear
effects occur between the two species. These findings are differences in cell densities and high variation in species
in contrast with the findings of Liu and Li [45], where an richness were found among the membranes (Figure 6). Or
enhanced attachment of E. coli on a P. aeruginosa biofilm maybe cells only need to adhere when they are in a stressful
layer was shown on porous media in LB broth. Also it is environment (the culture vessel in comparison with the
suggested that during the biofilm maturation, interbacterial MBR reactor), like in most biofilm formation. Adaptation
adhesion, rather than direct contact with the substrate, leads to environmental stress is indeed well documented for the
to progressive buildup of the mature biofilm. On the other used model organisms. Boles et al. demonstrated that P.
hand, it is also known that some species produce inhibitor aeruginosa undergoes genetic diversification during short-
substances against others [46]. Because the production of term biofilm growth, which might allow increased resistance
antimicrobial compounds can be a metabolic burden on of the population to environmental stresses [60].
the cell, some species evolved to start the production of Compared to the mono- and duospecies experiments,
such components when their cell numbers are high enough it seems that attachment of P. aeruginosa was not really
to ensure that the compound will reach an effective con- influenced by other community members (Figure 3 versus
centration, a phenomenon known as quorum sensing [39]. Figure 4). In brief, if one analyses Figure 3 in more detail, one
Different quorum sensing dependent systems have been can see that for both the mono- and duospecies experiments,
described between E. coli and P. aeruginosa [47, 48]. As a with an inoculation of 108 cells mL−1 , the enumerations of
different pattern was found between the adhesion and biofilm P. aeruginosa during the adhesion and biofilm formation
assay, it is conceivable that quorum sensing is one of the are quite similar in terms of order of magnitude (adhesion
occurring mechanisms, but further experiments are required 106 cells cm−1 ; biofilm 107 cells cm−1 ). Next, if one analyses
to confirm these hypotheses. However, the more effective Figure 4, one can see that for activated sludge spiked with
colonization behavior of P. aeruginosa on every membrane P. aeruginosa, with an inoculation of 104 P. aeruginosa cells
type was clearly highlighted in these experiments.
mL−1 , the adhesion and biofilm formation are 102 and 103
cells mL−1 , respectively. This means that, with respect to
4.3. Activated Sludge Spiked with P. aeruginosa . To gradually the original number of inoculated cells, the cell attachment
increase the complexity of the biofoulants, activated sludge on the membrane is independent from the biofoulant com-
harvested from a lab-scale MBR was diluted and spiked with plexity, as the difference of the cell attachment is identical
P. aeruginosa. It is well known that activated sludge contains to the difference between the number of inoculated cells,
a very complex community, and many different factors affect namely, 104 . This means that P. aeruginosa, even though
the exact community composition [49]. Different species are the bacterium is in competition with other community
present, and each group of species has its own habitat and members, still can preserve its place on the membrane. This
niche. Examples are floc-forming, filamentous, nitrifying,
is very clear for the PSF and PE membranes and highlights
and organotrophic bacteria. By introducing a new habitat,
again the good attachment and biofilm properties of the
like a membrane, some species can bind to it and exhibit a
bacterium (Figure 4). The reason why P. aeruginosa is such
preferred membrane associated life form. This is probably
the reason why the microbial communities on membrane an excellent biofilm pioneer is probably due to its high
surfaces are very different from the ones in the suspended adaptive response to environmental stress. It was recently
biomass [21, 50–59]. reviewed that 303 genes are differentially expressed under
In the presented experiment, a quite important mem- different conditions [61]. Functional categorization of the
brane effect was found (Table 3). The PVDF membrane, genes revealed that the most highly differentially regulated
which did not have a specifically high biofouling tendency genes encode membrane proteins, followed by genes involved
in the mono-, duospecies, or lab-scale MBR experiments, in the transport of small molecules (such as QS molecules),
showed high cell densities and species richness (Figure 4). adaptation and protection, and transcriptional regulators,
indicating an adaptive response of P. aeruginosa to the slower (experiments 1 and 2). However, when other species
changing environment [61]. are present together with P. aeruginosa like members, but they
are still in minority, changes in biofouling pattern may occur.
4.4. Activate Sludge in a Lab-Scale MBR. In order to mimic a In this study, this was highlighted by the high enumeration
more representative biofouling formation, membrane pieces on the PVDF membrane in experiment 3. Moreover it was
were submerged in a lab-scale MBR and processed in three suggested that not only the species diversity, but also the
different ways (1 h adhesion, 1 h adhesion followed by 24 h origin of the community had an important effect on the
further incubation in Ringer, or in the MBR). When both the biofouling tendencies. Similar ecological assays may lead to
bacterial density (Table 4) as well as the community composi- the identification of microbial key players and provide many
tion (Table 5) was used, a strong main treatment and interac- exciting insights into biofouling related microorganisms.
tion (membrane ∗ treatment) effect was found. Although the
membranes were incubated 24 h longer in the reactor or in
Ringer’s solution, the differences with the cell densities after Acknowledgments
1 h adhesion were rather small. This suggests that in 1 h time, This study was funded by the IDO/06/008 and OT 11/061
most of the membrane surface was covered (Figure 6). Only Projects of the KU Leuven and an FWO Project (G.0808.10N)
PSF-biofilm showed a higher enumeration during the MBR of the Flemish government. Louise Vanysacker acknowledges
biofilm formation and strongly influenced the post hoc Tukey the Fund for Scientific Research Flanders (FWO) for a
test (Table 4). The reason why is not clear as the effect is not PhD scholarship. The authors thank Prof. B. Koch from the
present in the other assays. Notwithstanding the fact that the Technical University of Denmark and Prof. N. Boon from
membrane replicates were randomly located in the reactor, the University of Ghent (Belgium) for the supply of the flu-
small differences in aeration and therefore a lower membrane orescent labeled microorganisms. Finally the authors want to
scouring cannot be neglected at lab-scale. Concerning the thank Xavier Vanysacker and Bart Boerjan for the graphical
species richness, high variations were found between the illustrations and proofreading of the paper, respectively. The
membranes. Also the species diversity in the 24 h Ringer authors have no conflict of interests to declare.
treatment was much lower compared to the 1 h adhesion and
24 h biofilm formation in the reactor. This suggests that some
species could not stay in the freshly formed biofilm (decrease References
in diversity (Figure 6(b)), while other species use the free [1] J. A. Garnett and S. Matthews, “Interactions in bacterial biofilm
space on the membrane and start to proliferate (unchanged development: a structural perspective,” Current Protein and
density (Figure 6(a)). Also this confirms why significantly Peptide Science, vol. 13, no. 8, pp. 739–755, 2012.
different community patterns were found using RDA for [2] H.-C. Flemming, “Biofouling in water systems—cases, causes
the three treatments (Table 5). The RDA plot illustrates that and countermeasures,” Applied Microbiology and Biotechnology,
the samples had the tendency to cluster together along the vol. 59, no. 6, pp. 629–640, 2002.
treatments (adhesion, Ringer, and biofilm) rather than along [3] R. M. Donlan, “Biofilms: microbial life on surfaces,” Emerging
the different membrane types (Figure 7). Infectious Diseases, vol. 8, no. 9, pp. 881–890, 2002.
[4] H.-C. Flemming, G. Schaule, T. Griebe, J. Schmitt, and A.
5. Conclusion Tamachkiarowa, “Biofouling—the Achilles heel of membrane
processes,” Desalination, vol. 113, no. 2-3, pp. 215–225, 1997.
The bacterial attachment and biofilm formation on three [5] P. Blanpain-Avet, C. Faille, G. Delaplace, and T. Bénézech, “Cell
types of polymeric microfiltration membranes were evaluated adhesion and related fouling mechanism on a tubular ceramic
in this study using different static assays. Notwithstanding microfiltration membrane using Bacillus cereus spores,” Journal
the fact that membranes with quite different characteristics of Membrane Science, vol. 385-386, no. 1, pp. 200–216, 2011.
(PE, PSF, and PVDF) were chosen, the membrane effects [6] C. J. Seneviratne, Y. Wang, L. Jin, S. S. W. Wong, T. D. K. Herath,
and L. P. Samaranayake, “Unraveling the resistance of microbial
in the four experiments were either nonexistent, rather low,
biofilms: has proteomics been helpful?” Proteomics, vol. 12, no.
disappearing after longer incubation times, or nonrepetitive 4-5, pp. 651–665, 2012.
when different biofoulants were used. On the other hand,
[7] B. Prakash, B. M. Veeregowda, and G. Krishnappa, “Biofilms: a
coexistence or treatment effects were significant in most of survival strategy of bacteria,” Current Science, vol. 85, no. 9, pp.
the cases. This means that the membrane biofouling was 1299–1307, 2003.
most influenced by the species composition, how species [8] S. Elias and E. Banin, “Multi-species biofilms: living with friend-
interact with each other, and the environmental situations ly neighbors,” FEMS Microbiology Reviews, vol. 36, pp. 990–
in which the membranes were situated. As expected, when 1004, 2012.
more complex but rather realistic biofoulants were used, [9] O. Rendueles and J.-M. Ghigo, “Multi-species biofilms: how to
the biofouling behavior gets more difficult to predict. For avoid unfriendly neighbors,” FEMS Microbiology Reviews, vol.
example, it is clear that when one dominant species, such as 36, pp. 972–989, 2012.
P. aeruginosa, with good attachment capabilities is present, [10] S. Judd, The MBR Book: Principles and Applications of Membrane
high fouling tendencies are expected on every membrane Bioreactors For Water and wasteWater Treatment, Elsevier
by this particular species. On opposite sides, when a feed Science, Oxford, UK, 2006.
is mainly composed by E. coli like species, which are less [11] K. E. Shannon, D.-Y. Lee, J. T. Trevors, and L. A. Beaudette,
able to attach to a substrate, the biofouling formation will be “Application of real-time quantitative PCR for the detection
of selected bacterial pathogens during municipal wastewater [27] S. B. S. Ghayeni, P. J. Beatson, R. P. Schneider, and A. G. Fane,
treatment,” Science of the Total Environment, vol. 382, no. 1, pp. “Adhesion of waste water bacteria to reverse osmosis mem-
121–129, 2007. branes,” Journal of Membrane Science, vol. 138, no. 1, pp. 29–42,
[12] M. Cardinale, L. Brusetti, P. Quatrini et al., “Comparison of 1998.
different primer sets for use in automated ribosomal intergenic [28] T. Knoell, J. Safarik, T. Cormack, R. Riley, S. W. Lin, and
spacer analysis of complex bacterial communities,” Applied and H. Ridgway, “Biofouling potentials of microporous pol-
Environmental Microbiology, vol. 70, no. 10, pp. 6147–6156, 2004. ysulfone membranes containing a sulfonated polyether-
[13] M. T. Suzuki, L. T. Taylor, and E. F. DeLong, “Quantitative ethersulfone/polyethersulfone block copolymer: correlation
analysis of small-subunit rRNA genes in mixed microbial of membrane surface properties with bacterial attachment,”
populations via 5’-nuclease assays,” Applied and Environmental Journal of Membrane Science, vol. 157, no. 1, pp. 117–138, 1999.
Microbiology, vol. 66, no. 11, pp. 4605–4614, 2000. [29] M. Pasmore, P. Todd, S. Smith et al., “Effects of ultrafiltra-
[14] K. N. Timmis, “Pseudomonas putida: a cosmopolitan oppor- tion membrane surface properties on Pseudomonas aeruginosa
tunist par excellence,” Environmental Microbiology, vol. 4, no. biofilm initiation for the purpose of reducing biofouling,”
12, pp. 779–781, 2002. Journal of Membrane Science, vol. 194, no. 1, pp. 15–32, 2001.
[15] Q. Lin and W. Jianlong, “Biodegradation characteristics of [30] L. Eshed, S. Yaron, and C. G. Dosoretz, “Effect of permeate
quinoline by Pseudomonas putida,” Bioresource Technology, vol. drag force on the development of a biofouling layer in a
101, no. 19, pp. 7683–7686, 2010. pressure-driven membrane separation system,” Applied and
Environmental Microbiology, vol. 74, no. 23, pp. 7338–7347,
[16] C. Beloin, A. Roux, and J.-M. Ghigo, “Escherichia coli biofilms,”
2008.
Current Topics in Microbiology and Immunology, vol. 322, pp.
249–289, 2008. [31] D. C. Ellwood, C. W. Keevil, P. D. Marsh, C. M. Brown, and
J. N. Wardell, “Surface-associated growth,” Philosophical Trans-
[17] E. L. Golovlev, “The mechanism of formation of Pseudomonas
actions of the Royal Society B, vol. 297, no. 1088, pp. 517–532, 1982.
aeruginosa biofilm, a type of structured population,” Microbiol-
ogy, vol. 71, no. 3, pp. 249–254, 2002. [32] M. R. Bilad, P. Declerck, A. Piasecka, L. Vanysacker, X. Yan,
and I. F. J. Vankelecom, “Development and validation of a
[18] J. Kato, H.-E. Kim, N. Takiguchi, A. Kuroda, and H. Ohtake,
high-throughput membrane bioreactor (HT-MBR),” Journal of
“Pseudomonas aeruginosa as a model microorganism for inves-
Membrane Science, vol. 379, no. 1-2, pp. 146–153, 2011.
tigation of chemotactic behaviors in ecosystem,” Journal of
Bioscience and Bioengineering, vol. 106, no. 1, pp. 1–7, 2008. [33] M. Hamilton, The Biofilm Laboratory: Step-By-Step Protocols
for Experimental Design, Analysis, and Data Interpretation,
[19] B. Koch, L. E. Jensen, and O. Nybroe, “A panel of Tn7-
Cytergy, Bozeman, Mont, USA, 2003.
based vectors for insertion of the gfp marker gene or for
delivery of cloned DNA into Gram-negative bacteria at a neutral [34] L. Vanysacker, S. A. J. Declerck, B. Hellemans, L. De Meester,
chromosomal site,” Journal of Microbiological Methods, vol. 45, I. Vankelecom, and P. Declerck, “Bacterial community analysis
no. 3, pp. 187–195, 2001. of activated sludge: an evaluation of four commonly used DNA
extraction methods,” Applied Microbiology and Biotechnology,
[20] P. Declerck, J. Behets, B. De Keersmaecker, and F. Ollevier,
vol. 88, no. 1, pp. 299–307, 2010.
“Receptor-mediated uptake of Legionella pneumophila by
Acanthamoeba castellanii and Naegleria lovaniensis,” Journal of [35] J. Lepš and P. Šmilauer, Multivariate Analysis of Ecological
Applied Microbiology, vol. 103, no. 6, pp. 2697–2703, 2007. Data Using CANOCO, Cambridge University Press, Cambridge,
Mass, USA, 2003.
[21] A. Piasecka, C. Souffreau, K. Vandepitte et al., “Analysis of
the microbial community structure in a membrane bioreactor [36] W. M. Dunne Jr., “Bacterial adhesion: seen any good biofilms
during initial stages of filtration,” Biofouling, vol. 28, no. 2, pp. lately?” Clinical Microbiology Reviews, vol. 15, no. 2, pp. 155–166,
225–238, 2012. 2002.
[22] P. Vandezande, L. E. M. Gevers, and I. F. J. Vankelecom, “Solv- [37] S. Kang, E. M. V. Hoek, H. Choi, and H. Shin, “Effect of
ent resistant nanofiltration: separating on a molecular level,” membrane surface properties during the fast evaluation of cell
Chemical Society Reviews, vol. 37, no. 2, pp. 365–405, 2008. attachment,” Separation Science and Technology, vol. 41, no. 7,
pp. 1475–1487, 2006.
[23] M. Mulder, “Preparation of synthetic membranes,” in Basic
Principles of Membrane Technology, M. Mulder, Ed., pp. 71– [38] Y.-F. Yang, Y. Li, Q.-L. Li, L.-S. Wan, and Z.-K. Xu, “Surface
154, Kluwer Academic Publishers, Dordrecht, The Netherlands, hydrophilization of microporous polypropylene membrane by
1996. grafting zwitterionic polymer for anti-biofouling,” Journal of
Membrane Science, vol. 362, no. 1-2, pp. 255–264, 2010.
[24] M. R. Bilad, P. Declerck, A. Piasecka, L. Vanysacker, X. Yan,
and I. F. J. Vankelecom, “Treatment of molasses wastewater [39] P. Moons, C. W. Michiels, and A. Aertsen, “Bacterial inter-
in a membrane bioreactor: Influence of membrane pore size,” actions in biofilms bacterial interactions in biofilms,” Critical
Separation and Purification Technology, vol. 78, no. 2, pp. 105– Reviews in Microbiology, vol. 35, no. 3, pp. 157–168, 2009.
112, 2011. [40] M. K. Banks and J. D. Bryers, “Bacterial species dominance
[25] P. van der Marel, A. Zwijnenburg, A. Kemperman, M. Wessling, within a binary culture biofilm,” Applied and Environmental
H. Temmink, and W. van der Meer, “Influence of membrane Microbiology, vol. 57, no. 7, pp. 1974–1979, 1991.
properties on fouling in submerged membrane bioreactors,” [41] M. T. Madigan, M. T. Madigan, and T. D. Brock, Brock Biology
Journal of Membrane Science, vol. 348, no. 1-2, pp. 66–74, 2010. of Microorganisms, Pearson, San Francisco, Calif, USA, 2009.
[26] M. R. Bilad, D. Vandamme, I. Foubert, K. Muylaert, and I. F. J. [42] T. K. Wood, A. F. González Barrios, M. Herzberg, and J. Lee,
Vankelecom, “Harvesting microalgal biomass using submerged “Motility influences biofilm architecture in Escherichia coli,”
microfiltration membranes,” Bioresource Technology, vol. 111, pp. Applied Microbiology and Biotechnology, vol. 72, no. 2, pp. 361–
343–352, 2012. 367, 2006.
[43] J. A. Girón, O. G. Gómez-Duarte, K. G. Jarvis, and J. B. [58] J. Zhang, H. C. Chua, J. Zhou, and A. G. Fane, “Factors
Kaper, “Longus pilus of enterotoxigenic Escherichia coli and its affecting the membrane performance in submerged membrane
relatedness to other type-4 pili—a minireview,” Gene, vol. 192, bioreactors,” Journal of Membrane Science, vol. 284, no. 1-2, pp.
no. 1, pp. 39–43, 1997. 54–66, 2006.
[44] M. A. Schembri, K. Kjærgaard, and P. Klemm, “Global gene [59] H. Choi, K. Zhang, D. D. Dionysiou, D. B. Oerther, and G.
expression in Escherichia coli biofilms,” Molecular Microbiology, A. Sorial, “Effect of activated sludge properties and membrane
vol. 48, no. 1, pp. 253–267, 2003. operation conditions on fouling characteristics in membrane
[45] Y. Liu and J. Li, “Role of Pseudomonas aeruginosa biofilm in the bioreactors,” Chemosphere, vol. 63, no. 10, pp. 1699–1708, 2006.
initial adhesion, growth and detachment of Escherichia coli in [60] B. R. Boles, M. Thoendel, and P. K. Singh, “Self-generated
porous media,” Environmental Science and Technology, vol. 42, diversity produces “insurance effects” in biofilm communities,”
no. 2, pp. 443–449, 2008. Proceedings of the National Academy of Sciences of the United
[46] A. G. Al-Bakri, P. Gilbert, and D. G. Allison, “Immigration States of America, vol. 101, no. 47, pp. 16630–16635, 2004.
and emigration of Burkholderia cepacia and Pseudomonas [61] D. Balasubramanian and K. Mathee, “Comparative transcrip-
aeruginosa between and within mixed biofilm communities,” tome analyses of Pseudomonas aeruginosa,” Human Genomics,
Journal of Applied Microbiology, vol. 96, no. 3, pp. 455–463, vol. 3, no. 4, pp. 349–361, 2009.
2004.
[47] P. Moons, R. Van Houdt, A. Aertsen, K. Vanoirbeek, Y.
Engelborghs, and C. W. Michiels, “Role of quorum sensing
and antimicrobial component production by Serratia plymuth-
ica in formation of biofilms, including mixed biofilms with
Escherichia coli,” Applied and Environmental Microbiology, vol.
72, no. 11, pp. 7294–7300, 2006.
[48] W. Chu, T. R. Zere, M. M. Weber et al., “Indole production
promotes Escherichia coli mixed-culture growth with Pseu-
domonas aeruginosa by inhibiting quorum signaling,” Applied
and Environmental Microbiology, vol. 78, no. 2, pp. 411–419, 2012.
[49] R. J. Seviour, Microbial Ecology of Activated Sludge, IWA,
London, UK, 2010.
[50] D.-W. Gao, Y. Fu, Y. Tao et al., “Linking microbial community
structure to membrane biofouling associated with varying
dissolved oxygen concentrations,” Bioresource Technology, vol.
102, no. 10, pp. 5626–5633, 2011.
[51] L.-N. Huang, H. De Wever, and L. Diels, “Diverse and distinct
bacterial communities induced biofilm fouling in membrane
bioreactors operated under different conditions,” Environmen-
tal Science and Technology, vol. 42, no. 22, pp. 8360–8366, 2008.
[52] P. Jinhua, K. Fukushi, and K. Yamamoto, “Bacterial community
structure on membrane surface and characteristics of strains
isolated from membrane surface in submerged membrane
bioreactor,” Separation Science and Technology, vol. 41, no. 7, pp.
1527–1549, 2006.
[53] S. Lim, S. Kim, K.-M. Yeon, B.-I. Sang, J. Chun, and C.-H.
Lee, “Correlation between microbial community structure and
biofouling in a laboratory scale membrane bioreactor with
synthetic wastewater,” Desalination, vol. 287, pp. 209–215, 2012.
[54] K. Zhang, H. Choi, D. D. Dionysiou, G. A. Sorial, and D. B.
Oerther, “Identifying pioneer bacterial species responsible for
biofouling membrane bioreactors,” Environmental Microbiol-
ogy, vol. 8, no. 3, pp. 433–440, 2006.
[55] Y. Miura, Y. Watanabe, and S. Okabe, “Membrane biofouling
in pilot-scale membrane bioreactors (MBRs) treating munic-
ipal wastewater: impact of biofilm formation,” Environmental
Science and Technology, vol. 41, no. 2, pp. 632–638, 2007.
[56] B. Wu, S. Yi, and A. G. Fane, “Microbial behaviors involved
in cake fouling in membrane bioreactors under different solids
retention times,” Bioresource Technology, vol. 102, no. 3, pp. 2511–
2516, 2011.
[57] S. Xia, J. Li, S. He et al., “The effect of organic loading on
bacterial community composition of membrane biofilms in
a submerged polyvinyl chloride membrane bioreactor,” Biore-
source Technology, vol. 101, no. 17, pp. 6601–6609, 2010.
http://dx.doi.org/10.1155/2013/863240
Review Article
Microbial Inoculants and Their Impact on Soil Microbial
Communities: A Review
Darine Trabelsi and Ridha Mhamdi

Laboratory of Legumes, Centre of Biotechnology of Borj-Cédria, P.O. Box 901, 2050 Hammam-Lif, Tunisia
Correspondence should be addressed to Ridha Mhamdi; ridha.mhamdi@cbbc.rnrt.tn
Received 10 April 2013; Revised 7 June 2013; Accepted 25 June 2013
Copyright © 2013 D. Trabelsi and R. Mhamdi. This is an open access article distributed under the Creative Commons Attribution
cited.
The knowledge of the survival of inoculated fungal and bacterial strains in field and the effects of their release on the indigenous
microbial communities has been of great interest since the practical use of selected natural or genetically modified microorganisms
has been developed. Soil inoculation or seed bacterization may lead to changes in the structure of the indigenous microbial
communities, which is important with regard to the safety of introduction of microbes into the environment. Many reports
indicate that application of microbial inoculants can influence, at least temporarily, the resident microbial communities. However,
the major concern remains regarding how the impact on taxonomic groups can be related to effects on functional capabilities
of the soil microbial communities. These changes could be the result of direct effects resulting from trophic competitions and
antagonistic/synergic interactions with the resident microbial populations, or indirect effects mediated by enhanced root growth
and exudation. Combination of inoculants will not necessarily produce an additive or synergic effect, but rather a competitive
process. The extent of the inoculation impact on the subsequent crops in relation to the buffering capacity of the plant-soil-biota is
still not well documented and should be the focus of future research.
1. Introduction inoculation were always neglected. Since inoculation consists

in supplying high densities of viable and efficient microbes for
The application of inoculants is seen as being very attractive a rapid colonization of the host rhizosphere, it would induce
since it would substantially reduce the use of chemical fertil- at least a transient perturbation of the equilibrium of soil
izers and pesticides, and there are now an increasing number microbial communities. Changes in microbial composition
of inoculants being commercialized for various crops [1]. may be undesirable if important native species are lost, thus
Microorganisms play an important role in agricultural sys- affecting subsequent crops. However, a modification in the
tems, particularly plant growth-promoting microorganisms bacterial community structure caused by inoculation could
(PGPMs). Plant growth benefits may be attributed mainly to be buffered by ecosystem resilience, which is driven by the
three mechanisms as follow. (i) PGPMs acting as biofertilizers level of diversity and interactions of the plant-soil-biota
(such as nitrogen-fixing bacteria and phosphate-solubilizing [11]. The loss of certain bacterial species may however not
bacteria) assist plant nutrient uptake by providing fixed nitro- change the functioning of the system because of the bacterial
gen or other nutrients [2]. (ii) Phytostimulators (microbes redundancy, since different bacterial species may carry out
expressing phytohormones such as Azospirillum) can directly the same functions [11, 12].
promote the growth of plants, usually by producing plant Soil microbial community is complex and dynamic and
hormones [3, 4]. (iii) Biological control agents (such as Tri- varies in composition between different compartments and
choderma, Pseudomonas, and Bacillus) protect plants against levels, which represents a real challenge in soil ecology. One
phytopathogenic organisms [5–7]. Several reviews have dis- of the crucial problems to face in this type of studies is
cussed various aspects of growth promotion by PGPMs [8– the representativeness of the sampling. Number of replicates,
10]. However, the potential environmental impacts related to sample size, whether sampling is randomized or at regular
intervals, spatial scaling, and microsite variation remain beneficial or nonpathogenic microbes biosynthesize N-AHLs
major concerns. Most researchers used rhizospheric soil, but degrading lactonases to disrupt QS by quorum-quenching
even in this case it is practically very difficult to define it [39]. The presence of biostimulating molecules or QS-mimics
precisely. However, the side-distance effect on bulk soil would may also establish microbes that degrade N-AHLs in the rhi-
be more reliable to address a more generalized response. zosphere, thus reducing the virulence of pathogens. Another
Time-course studies would be also necessary to monitor example shows that root volatiles may serve as foraging
inoculation effect in relation to the buffering capacity of the cues for parasitic entomopathogenic nematodes [40], and,
ecosystem. Nevertheless, the techniques used to investigate then, poorly available organic phosphorus could be made
soil microbial communities at taxonomic and functional available through the grazing by nematodes of phytase-
levels are laborious and limit the use of exhaustive samplings. producing bacteria [41]. There may be also a synergistic effect
In the culture-dependent methods, the analysis is usually of root volatiles and CO2 as attractants for nematodes [42].
confined to restricted samples, and a biased image is drawn. Kawasaki et al. [43] showed that some legumes respond to
However, the culture-independent methods do not usually certain contaminants in a systemic manner by utilizing root-
permit unambiguous identification of taxonomic groups colonizing microorganisms to protect themselves, and they
[13]. Besides the bias induced by DNA extraction and PCR may actively stimulate such microorganisms in the rhizo-
amplification, the culture-independent methods represent sphere by manipulation of exudates flux and composition.
also some inherent limitations (Table 1). Surfactant-active compounds in the root exudates increase
The high-throughput sequencing techniques, even being the solubility of the contaminant and make it more bioavail-
more informative [31, 32], are still not economically afford- able to root-colonizing microorganisms [44]. The production
able for inoculation impact analysis, because of the high of plant growth regulators such as auxin, cytokinin, and
number of samples and replicates involved. There is a growing gibberellin by PGPMs may also interfere on soil microbial
interest in the recent years in genes and transcripts coding communities through an enhanced root growth and an
for metabolic enzymes. Besides questions addressing redun- increased exudation rate [45], without excluding the possi-
dancy and diversity, more and more attention is given to bility of a direct effect on the microbial equilibrium leading
the abundance of specific DNA and mRNA in the different to the selection of beneficial populations. These few examples
habitats [33]. illustrate the complexity of the rhizospheric interactions and
prove that our knowledge about the plant-soil-biota is just
2. The Rhizosphere: The Unrevealed World beginning.
The rhizosphere represents one of the most diverse habitats

on the planet and is central to ecosystem functioning. Infinite 3. Inoculation Impact Analysis
dynamic interactions between root exudates, microbial activ-
Different soil microorganisms had been extensively used
ity, genetic exchange, nutrient transformation, and gradient
as inoculants, including rhizobia, Azospirillum, mycorrhizal
diffusion are most likely the factors shaping this below-
fungi, and biocontrol agents. The most significant studies
ground world (for review, please see [34]). Consequently,
reporting the effect of these inoculants on soil microbial
there is an increasing need to understand its functioning
communities are summarized in Table 2.
to effectively manage ecosystem and harness its potential
benefits. In particular, manipulation of the rhizosphere is
now considered as a key mechanism for solving critical 3.1. Rhizobia Inoculants. Rhizobia are reported to influence
issues facing the planet, including agricultural and forest crop growth, yield, and nutrient uptake by different mech-
sustainability, improving water quality, mitigation of climate anisms. They fix nitrogen, help in promoting free-living
change, and preservation of biodiversity. To face the range nitrogen-fixing bacteria, increase supply of other nutrients,
of biotic and abiotic stresses, plants interact with different such as phosphorus and iron, produce plant hormones,
members of the soil microbial community in many ways and enhance other beneficial bacteria or fungi, control bacterial
in a complex range of trophic cascades [35]. These relations and fungal diseases, and help in controlling insect pests [8,
involve positive and negative feedbacks between soil organ- 10].
isms and plants and their chemical environment. Rhizobacte- Field release of a Rhizobium etli strain containing
ria communicate within species using diffused chemicals. N- genes encoding trifolitoxin (an antibiotic peptide active
acylhomoserine lactones (N-AHLs) are common signals for against members of a specific group of 𝛼-proteobacteria
bacterial communications in various Gram-negative bacteria that enhances the ability to compete trifolitoxin-sensitive
[36]. Plant-Microbe interactions are generally regulated by strains) strongly reduced the diversity of trifolitoxin-sensitive
quorum sensing (QS) in a population-dependent manner members of 𝛼-proteobacteria in bean rhizosphere as shown
[37]. Crépin et al. [38] showed that the rhizosphere bacterium by ribosomal intergenic spacer analysis (RISA), with little
Rhodococcus erythropolis catabolizes the N-AHLs produced apparent effect on most microbes [48]. Using a cultivation-
by the pathogenic bacterium Pectobacterium atrosepticum, dependent approach and a cultivation-independent
thus attenuating its virulence, which suggests a tritrophic PCR-single-strand conformation polymorphism (SSCP),
belowground chemical interaction. Pathogens exploit QS Schwieger and Tebbe [14] showed that field release of
using N-AHLs to form microcolonies (and biofilm) in the Sinorhizobium meliloti strain L33 affected bacterial diversity
rhizosphere to inflict pathogenicity in the host. By contrast, in the rhizosphere of alfalfa by reducing the number
Table 1: Advantages and limitations of the culture-independent methods usually used to investigate soil microbial communities.
Method Advantages Limitations
Formation of heteroduplexes; limited phylogenetic information; only short
SSCP (single-strand Separates amplified 16S ssDNA by sequence-dependent higher-order
fragments (between 150–400 nucleotides) can be optimally separated. High
conformation structure; great simplicity and speed; automation possible. No GC-clamp is
rate of reannealing during electrophoresis; biases introduced by conformation
polymorphism) [14] necessary and no gradient gels.
variations.
Separates amplified molecules by %GC content on a denaturing gradient; Limited to dominant communities; samples with high levels of diversity are
DGGE excellent and effective to follow changes of microbial communities in time difficult to resolve; phylogenetic information is limited to bands that are able
(denaturing-gradient gel and space; well suited for monitoring complex communities dominated by to be removed and sequenced. A single band does not always mean a single
electrophoresis) [15–17] a few members; allows phylogenetic identification through excision and strain; needs careful calibration; limited to DNA fragments typically below
sequencing of individual bands. 500 bp in size.
TGGE Separates amplified molecules by %GC content on a temperature gradient;

Provides approximately the same degree of specificity as DGGE and possesses
(temperature-gradient gel operates on the same principle as DGGE; taxonomic information could be
the same advantages and limitations.
electrophoresis) [18] obtained from isolated bands.
RISA (ribosomal RNA Amplifies the prokaryotic ribosomal intergenic region creating a The preferential amplification of shorter sequences is a particular concern;
intergenic spacer analysis) community profile based on the species-specific length polymorphisms in biases imposed by secondary structures in the rDNA genes flanking the
[19–21] this region; great simplicity and speed; automation is possible. amplified region may also pose a problem; small database.
Separates amplified and digested molecules according to their length; only
T-RFLP (terminal terminal restriction fragments (TRFs) are detected and used for qualitative Technical problems arise from inherent pitfalls in the databases used for
restriction fragment length and quantitative analysis; powerful tool for assessing diversity and structure phylogenetic analysis; overestimation of diversity created by incomplete
polymorphism) [22–25] of complex microbial communities; Enables high-throughput and low-cost digestion of environmental DNA and the formation of pseudo (TRFs).
analysis.
Assesses total microbial communities using probes targeting functional
genes; the proportions of specific phylotypes can address metabolic
potential of the microbial biomass; has superior sensitivity and is more Requires careful calibration; requires extremely accurate controls for inferring
Real-time PCR [26] convenient and less expensive for the quantification of selected bacterial cell mass or gene copy; biases introduced by contamination and primer
populations; quantification of rRNA directly isolated from ribosomes may dimers.
be used to reveal the metabolically most active members of a bacterial
community.
MIDI-FAME (Microbial Creates a complex profile of fatty acids unique to each community sample;
Quite limited in the taxonomic information; many fatty acids are common to
ID, Inc.-fatty acid methyl useful in comparing one sample with another and in tracking changes in
different microorganisms.
ester) [27] community structure over time or at different sampling locations.
Determine the profile of substrates metabolized by the microbial
CLPP (community-level community; give an estimate of growth and catabolic potential of
physiological profiles of culturable microorganisms in the original community; relatively Long procedure; growth dependent, limited in taxonomic information.
carbon sources) [28, 29] inexpensive and commercially are available means of gathering large
amounts of information about whole communities of microorganisms.
Detect, identify, and potentially quantify thousands of distinct DNA
Limited in exploring entire bacterial diversity. Prior knowledge of the
Hybridization arrays molecules in a single experiment relatively rapidly and cost-effectively;
microbial composition is necessary for designing meaningful probes; suffer
(microarrays, beadarrays arrays rely on oligonucleotide probes of 16S rRNA from specific groups of
from cross hybridization between closely related species; genetic variations
and Phylochips) [30] organisms to discern phylogeny, community composition, or function; well
between strains within species; biases introduced by hairpin structures.
suited for identification via multiple-gene functional groups.
The longer reads yield more information about the 16S rRNA and thus give Intrinsic sequencing errors; overestimation of taxon abundance; primer pairs
a more accurate identification. Allows accurate comparison between greatly influence estimates of microbial community richness and evenness;
NGS (next-generation
environments; large communities can be studied based on phylogeny amplicon products are still subjected to the biases inherent to any PCR-based
sequencing) [31, 32]
and/or function; an average sequence depth of 5000 sequences/sample, up experiment; the reproducibility of amplicon sequencing across a large number
3
to 200 samples, could be sequenced in parallel. of biological replicates is still questionable.

Table 2: The most significant studies addressing the impact of inoculation on soil microbial communities.
Inoculant type Species Techniques Major results

The bacterial diversity in the rhizosphere of Medicago sativa was qualitatively
S. meliloti L33 SSCP and quantitatively affected. While the number of members of 𝛾-proteobacteria [14]
decreased, the number of members of 𝛼-proteobacteria increased.
Field inoculation showed a significant increase of total bacterial diversity due
Cocktail of Ensifer
DGGE to seasonal changes, but no effect of rhizobial inoculation was observed. [17]
strains
DGGE offered little information about bacterial communities.
The persistence of certain 𝛾-proteobacterial populations in the rhizosphere of
S. meliloti
Rhizobia TGGE alfalfa could be affected. TGGE proved to be better for identifying specific [18]
M401/M403
M403-dependent changes.
Field inoculation showed significant effects on bacterial structure and
diversity in the bulk soil of common bean. Both 𝛼- and 𝛾-proteobacteria
R. gallicum 8a3
T-RFLP together with Firmicutes and Actinobacteria were enhanced, including [22, 23]
E. meliloti 4H41
beneficial bacterial communities with PGPM potentialities. Dual inoculation
was less effective than simple inoculation and induced distinct effects.
The bacterial genes involved in nitrogen turnover were affected by
inoculation. The effectiveness of inoculation was related to the abundance of
S. meliloti S26/OS6 qPCR [26]
nif H genes in the late flowering phase. A higher number of amoA copies were
observed during flowering.
DGGE Field inoculation showed no prominent effects on bacterial communities of
A. brasilense Cd/Sp245 [21]
RISA maize in two different soils and in different growth systems.
Field inoculation of maize increased the intersample variability of the
Azospirillum A. lipoferum CRT1 RISA bacterial community between individual plants and sampling times without [19, 20]
modifying the total number of root bacteria.
A. brasilense Inoculation changed the community-level physiological profiles of the
CLPPs [29]
40M/42M cultivable microbial communities associated with rice roots.
Inoculation affected the composition of the rhizosphere bacterial community
of pea. Four to five specific bands were suppressed. Before flowering, the AMF
decreased rhizosphere respiration and number of protozoa, but it did not
G. intraradices DGGE [46]
affect bacterial number. During flowering and pod formation, the AMF
AMF stimulated rhizosphere respiration and the negative effect on protozoa
decreased.
Inoculation significantly modified the rhizosphere bacterial composition of
G. mosseae
DGGE tomato. The two AMFs had had similar bacterial communities; however, [47]
G. intraradices
specific species-dependent effects were observed.
Inoculation induced a transient effect on fungal community in the
T-RFLP
P. fluorescens 2P24 rhizosphere of cucumber suggesting that this biocontrol agent has a limited [25]
DGGE
validity. DGGE and T-RFLP showed similar results.
Biocontrol
agents Inoculation induced shifts in fatty acid methyl ester profiles of cultivable
P. chlororaphis IDV1 bacteria fractions, as well as total microbial communities in the rhizosphere of
FAME [27]
P. putida RA2 maize. The rhizosphere composition shifted from a Gram-positive-dominated
community to more Gram-negative populations.
Inoculation induced a significant modification in the bacterial community
Pseudomonas spp. structure. The type of PGPM consortium had more impact on the bacterial
DGGE [15]
AMF community structure than the presence of AMF. A synergistic effect of
coinoculation was observed.
Inoculation with the biocontrol agent did not show significant effects on
Coinoculation B. subtilis 101 fungal (18S rRNA) and bacterial (16S rRNA) communities in the rhizosphere
DGGE [16]
A. brasilense Sp245 of tomato. Combination of the two rhizobacteria had no synergistic or
comparable effects on plant biomass, with respect to their single applications.
Inoculation did not show significant impact on cultivable communities and
A. brasilense T-RFLP
nif H T-RFLP-patterns of diazotrophic bacteria associated with rice roots. A [24]
P. fluorescens CLPPs
synergistic effect of coinoculation was found.
of 𝛾-proteobacteria and increasing the number of 𝛼- In our lab, the effect of on-field inoculation of Phaseolus
proteobacteria. The shift was interpreted as a replacement vulgaris with two local rhizobial strains was addressed by
of more general bacteria (Acinetobacter calcoaceticus exploring community structure and diversity using similarity
and Pseudomonas sp.) by specialists (rhizobia). The analysis of T-RFLP profiles. Effects on bacterial community
indigenous microbial populations in the rhizosphere of structure and diversity were clearly observed in the bulk soil
alfalfa inoculated with S. meliloti strain M403 (a strain in the neighborhood of 25 cm of the roots [22]. Both 𝛼- and 𝛾-
with enhanced competitiveness for nodule occupancy proteobacteria together with Firmicutes and Actinobacteria
constructed by introducing an extra copy of a modified were enhanced by inoculation. The mono- and dual inocula-
proline dehydrogenase gene) or strain M401 (a control tion with Rhizobium gallicum strain 8a3 and Ensifer meliloti
strain carrying the same plasmid without the modified strain 4H41 induced the proliferation of bacterial communi-
gene) were evaluated by comparing restriction fragment ties that had been frequently reported as PGPMs, like Rah-
length polymorphism (RFLP) and temperature-gradient nella, Bacillus, Azospirillum, Mesorhizobium, Pseudomonas,
gel electrophoresis fingerprints (TGGE) of 16S rDNA genes Streptomyces, and Sinorhizobium, among others [23]. The
directly amplified from soil DNA [18]. Seasonal changes extent of these changes was also seen in the next rotation crop
were observed in RFLP patterns only when primers specific as indicated by the 32% increase observed in potato yield, and
for 𝛽- and/or 𝛾-proteobacteria were used. RFLP analysis also by the 56% decrease in potato wireworm infestation [23].
showed that inoculation permitted certain 𝛾-proteobacterial The number of TRFs is significantly higher in the inoculated
populations to be maintained longer and persist in the treatments than that in the nitrogen-fertilized treatment
rhizosphere of alfalfa. Seasonal changes were also detected during flowering and harvesting stages. Similarly, there was
by TGGE performed with 𝛼-proteobacterial primers but a clear trend of increase in intersample heterogeneity of
not with 𝛽-proteobacterial primers [18]. Therefore, in the bacterial communities in the inoculated treatment up to the
TGGE patterns appeared two bands specifically correlated harvesting stage. These data suggest that the perturbation of
to inoculation with M403, showing sequence similarity, the community due to inoculation with a rhizobial strain
respectively, to Rahnella aquatilis and Kluyvera georgiana. is higher than that due to chemical fertilization and that
The seasonal fluctuations induced by environment and alfalfa the evolution of the community in response to inoculants is
plants showed much stronger influence on the microbial someway more stochastic, suggesting that the introduction of
community than the inoculation-dependent effects. exogenous bacteria in a community is likely to produce more
The quantification of bacterial genes encoding the nitro- long-term unpredictable effects than organic nitrogen supply.
genase reductase (nif H), ammonia monooxygenase (amoA),
and nitrite reductase (nirK and nirS), as well as archaeal 3.2. Azospirillum Inoculants. The agronomic benefits of
amoA genes within the nitrogen cycle, was performed in Azospirillum are well documented [52, 53]. Several modes
alfalfa rhizosphere inoculated with S. meliloti strains S26 and of action are implicated [45], especially the synthesis of
OS6 [26]. At the late flowering phase, a clear correlation was phytohormones such as indole-3-acetic acid [3, 54]. Since
demonstrated between effectiveness of inoculation, as evalu- Azospirillum inoculation can have a great effect on root
ated by nitrogen and carbon contents, and the abundance of development and exudation [53, 55], it is likely that the use of
nif H genes. Moreover, the number of archaeal amoA copies these phytostimulatory PGPMs will also modify the resident
increased upon the more effective strain inoculation. microbial community structure of the rhizosphere. Using
The effects of growth stage, intercropping, and rhizo- the automated ribosomal intergenic spacer analysis (ARISA),
bial inoculation on diversity of soybean endophytic bacte- inoculation of maize with Azospirillum lipoferum CRT1 was
ria were demonstrated by the cluster analysis of terminal seen to increase the variability in the genetic diversity of the
restriction fragments and the redundancy analysis [49]. The rhizobacterial communities between individual field-grown
relative abundance in roots was enhanced for Bradyrhizobium maize plants and between sampling times without modifying
liaoningense and decreased for Sinorhizobium americanum, the total number of root bacteria [19, 20]. In other studies, the
demonstrating that the endophytic Sinorhizobium might plant growth development was improved with Azospirillum
be suppressed by competition from the introduced strain. brasilense Sp245 by production of auxins, cytokinins, and gib-
The difference in abundance of endophytic Sinorhizobium, berellins [56]. Root physiology and architecture changed, and
Bradyrhizobium, and Rhizobium in the soybean roots could root surface area increased through production of more root
be correlated to the nodulation autoregulation system of hairs, leading to an increase in mineral uptake. However, no
legumes, which is capable of sensing and responding to prominent effects were observed on bacterial communities of
the presence of inefficient rhizobia by applying sanctions to maize grown in two different soils and in different growth
inactive strains via active control of the permeability of root systems as indicated by DGGE and ARISA [21]. Naiman et
cortical cells to oxygen, thus limiting the growth of inefficient al. [28] showed also that inoculation with Azospirillum and
nodules [20]. A 517 bp TRF assigned to Comamonadaceae Pseudomonas may exert different effects on the number of
and Burkholderiales showed a higher abundance in inocu- specific subgroups of cultivable bacteria in the rhizosphere
lated soybeans [49]. The abundance of rhizosphere Coma- of wheat. Inoculation changed the profiles of carbon source
monadaceae bacteria has been previously reported to be (CS) utilization of the soil microflora at tillering and grain-
associated with mycorrhization [50], and, on the other hand, filling stages. The CS utilization is related on one hand to the
the latter could be stimulated by the rhizobial inoculation of number of microorganisms which are able to use each CS
soybean [51]. as a sole carbon supply. on the other hand, the importance
of growth is a reflection of the functional potential of the two months. It was reported that the changes observed in
community. the mesquite rhizosphere microbial community structure
Inoculation with two A. brasilense strains (40M and 42M) may be either a direct or an indirect effect of the AMF
isolated from maize roots changed also the community-level used. One type of direct interaction between the AMF and
physiological profiles (CLPPs) of the cultivable microbial bacterial populations in the rhizosphere is the stimulation or
communities associated with rice [29]. Although the average suppression of one or more susceptible populations [72]. It
values of absorbance for arginine showed that the microbial was shown that the exudates of the extraradical mycelium
communities associated with inoculated and control plants of G. intraradices could significantly inhibit the conidial
had significant differences in ability to use arginine, a mod- germination of the plant pathogen Fusarium oxysporum in
erate effect on the physiological and genetic characteristics of transformed carrot roots [73]. By contrast, the PGPM Pseu-
microbial communities associated with certain rice cultivars domonas chlororaphis was strongly stimulated. One type of
was found under field conditions [24]. The nif H T-RFLP- indirect effects of G. intraradices is the induction of systemic
patterns of diazotrophic communities associated with rice resistance against the parasitic nematodes Radopholus similis
roots showed that some of the restriction fragments per- and Pratylenchus coffeae in banana plants [74]. Lioussanne
mitted differentiation of inoculation treatments, such as the et al. [47] showed that direct root colonization with either G.
66 bp fragment that could correspond to Methylogaea oryzae mosseae or G. intraradices significantly modified the DGGE
which is a methanotrophic bacterium associated with rice bacterial community structure of tomato rhizosphere. The
[24]. These data suggest that the hormonal effect exercised two AMF species had similar bacterial communities after
by Azospirillum improves the efficiency of the nitrogen four weeks. The bacterial taxa associated with the rhizosphere
absorption leading to superior yields of biomass. The overall of tomato plants inoculated with G. mosseae were identified
rhizosphere diversity, mainly of specific functional groups, as Pseudomonas, Herbaspirillum, and Acidobacterium, while
is most likely influenced by the change in residual nitrogen a Bacillus simplex (clone TR03) was found to be affiliated
rather than the direct effect of inoculation. However, the only with G. intraradices. One clone (TR2) that was ambigu-
specific mechanisms causing these changes are still not clear ously identified as Pseudomonas entomophila, Pseudomonas
and need to be extensively studied. plecoglossicida, or Pseudomonas putida has been associated
with both G. intraradices and G. mosseae. Taken together,
3.3. Mycorrhizal Fungi Inoculants. Arbuscular mycorrhizal all these works showed that AMF can support modification
fungi (AMF), grouped into the phylum Glomeromycota [57], in microbial community structure within mycorrhizosphere.
have the ability to form mutualistic symbiosis with most land The identification of key microbial populations that are
plants and colonize a wider soil volume. They receive carbon correlated with improved biomass production will help to
from their host, benefiting plant growth through their ability understand the role of the microbial community in support-
to exploit resources and delivering minerals and water back ing plant growth, suppressing plant pathogen invasion, and
[58]. AMF affect the soil microorganisms associated with other AMF functional abilities.
their extraradical mycelium, leading to the formation of a
specific zone of soil called the mycorrhizosphere [59, 60]. 3.4. Biocontrol Inoculants. Most of the commercial rhizobac-
In the mycorrhizosphere, the AMF might affect negatively terial products have been marketed for biological control
[61], positively [62], or may have no effect [63] on microbial of plant diseases rather than augmenting plant nutrition
biomass and growth of specific microbial taxa [64]. Many or minimizing abiotic stress impacts [1]. A number of
studies have shown that some bacterial species respond to the microorganisms such as Trichoderma harzianum [5, 75],
presence of certain AMF [59, 65], suggesting a high degree of Pseudomonas fluorescens [6], and Bacillus subtilis [7] have
specificity between bacteria associated with AMF. Thus, the demonstrated antagonism against diseases caused by Fusar-
specific bacteria together with AMF may create more indirect ium spp., Pythium spp., Rhizoctonia spp., Sclerotium spp., and
synergism for plant growth [66] including nutrient acqui- so forth, leading to enhancement in plant growth or yield. It
sition [67] and enhancement of root branching [68]. AMF has been established also that application of B. subtilis [75],
may inhibit pathogen proliferation through the formation of Pochonia chlamydosporia [76, 77], and P. fluorescens [78–80]
a bacterial community that limits the pathogen invasion [69, can effectively control the diseases caused by nematodes.
70]. Glomus intraradices has been shown to affect positively Two species of Pseudomonas showing antagonistic
the bacterial and saprotrophic fungal biomass in a root-free activity against the tomato pathogen Ralstonia solanacearum
sand environment [62]. Using DGGE of 16S rRNA amplicons [27] were assessed for their impacts on the rhizosphere
from total DNA extracts of pea rhizosphere, Wamberg microbial community structure of maize. Shifts in fatty acid
et al. [46] showed that DGGE profiles were relatively methyl ester (FAME) profiles of cultivable bacteria fractions,
similar between AMF-inoculated and AMF-uninoculated as well as total rhizosphere microbial communities, were
treatments. However, G. intraradices-inoculated treatments determined in answer to inoculation. The introduced P.
showed suppression of four to five specific bright bands. chlororaphis IDV1 and P. putida RA2 showed good survival
These types of changes were also studied in the extreme in the maize rhizosphere. However, both inoculants revealed
conditions characteristic of mine tailings [71]. Canonical a small growth-reducing effect towards maize, probably
correspondence analysis of DGGE profiles showed that AMF due to changes in the indigenous rhizosphere communities.
inoculation significantly influenced the development of both The community structure transitorily shifted following
fungal and bacterial rhizosphere community structures after domination of specific slow-growing bacterial classes.
The cultivable fraction of bacteria was represented by high fungal community, indicating that the period of validity of the
biomass of Gram-positives (i.e., Bacillus and Arthrobacter) biocontrol agent may be limited. All of these results suggest
as indicated by the contents of branched FAMEs (especially that inoculation with biocontrol agents breaks the original
iso/anteiso 15:0/17:0). During the later growth stage of ecological balance of soil microbial community; however,
maize, this group was disturbed by the introduced strains. a progressive recovery is commonly observed following
Actinomycetes, as indicated by fatty acids with a methyl fading of the released strain. The effects on soil microbial
group at carbon 10 of the chain [81], could be supported communities seem more pronounced and more maintained
by young roots of maize, but this effect was inhibited by in case of biofertilizers and phytostimulators having direct
inoculation [27]. Both P. chlororaphis IDV1 and P. putida RA2 effects on plants. It is likely that plants amplify and contribute
survived at relatively high density after release, suggesting in maintaining the observed effect. Root colonization is
their direct impact on community structure. In addition, the certainly a key step in this process.
inoculation shift from Gram-positive-dominated community
to more Gram-negative populations might be indicative of a 4. Coinoculation versus Monoinoculation
progressive change from oligotrophic to more copiotrophic
conditions [82], which could be due to the treatment effect Most inoculants often rely on application of a single strain
(i.e., nutrients released from dead cells) [27]. which might partially account for the recorded inconsis-
To assess the impact of intentionally applied microor- tencies in field. A way to overcome this problem is to
ganisms on several key ecological aspects of the microbial include different species or strains of beneficial microbes
decomposer community in plant litter, wheat straw on pans in the same microbial formulation. Coinoculation would
of soil was inoculated with either the fungus Limonomyces combine different mechanisms without the need for genetic
roseipellis or the bacterium P. fluorescens strain Pf-5, biocon- engineering [89], increase plant performance, and enhance
trol agents for Pyrenophora tritici-repentis [83]. Pseudomonas the efficacy and reliability of healthy effects on crops [90].
fluorescens had insignificant effect, but L. roseipellis had Roesti et al. [15] analyzed the effects of PGPR/AMF
measurable effects on some aspects of microbial community inoculations on bacterial community structure of wheat
structure and function. Limonomyces roseipellis altered the rhizosphere. DGGE analysis showed that inoculants induced
frequency distribution of fungal taxa on straw by increasing a significant modification in the bacterial community struc-
yeasts and decreasing saprophytic fungi such as Alternaria ture. However, the type of PGPR consortium had more
and Cladosporium. Compared with nontreated straw, res- impact on the bacterial community structure (28.3% of the
piration under conditions of adequate moisture (−0.1 MPa) variance) than the presence of AMF (10.6% of the variance).
was increased by L. roseipellis, but it was unaffected at Even though the PGPR strains used produce the antibiotic
water potential of −7 MPa. The spectrum of sole carbon 2.4-diacetylphloroglucinol, which is known for its antifungal
sources utilized by the straw microflora was slightly altered properties, the AMF growth was not affected, and a synergis-
in Limonomyces-treated straw [83]. This change in metabolic tic effect of PGPR/AMF coinoculation was observed.
capabilities is attributed to changes in microbial community. The effect of inoculation of Pinus pinea with Bacillus
The effects of Douglas fir (Pseudotsuga menziesii) co- licheniformis CECT 5106 and Bacillus pumilus CECT 5105 was
inoculation with the mycorrhiza helper bacterial strain P. evaluated [91]. Both Bacillus strains promoted the growth of P.
fluorescens BBc6R8 and/or the fungal strain Laccaria bicolor pinea seedlings (probably by gibberellin production), but this
S238N on the indigenous bacterial and ectomycorrhizal biological effect was absent when both strains were coinocu-
communities were assessed using quantitative and qualitative lated, indicating a competition effect. The phospholipid fatty
approaches [84]. The inoculated bacterial strain BBc6R8 was acids composition was also altered, despite the low levels of
not detected after 4 years in any of the treatments where it inoculants found at the end of the assay [91].
was speculated that the lack of bacterial effect on seedling Also, the combination of B. subtilis and A. brasilense
growth was due to the nonpersistence of the inoculated did not show synergistic or comparable effects on tomato
bacteria. When P. fluorescens BBc6R8 was introduced at growth comparing with their single applications. Rather, the
sowing, Frey-Klett et al. [85] showed a positive effect of mutual presence of these microorganisms, other than reduc-
bacterial inoculation on the Douglas fir-L. bicolor symbiosis ing plant growth, could cause root-architectural alterations
despite the bacteria only surviving for up to 19 weeks in the [16]. Our results showed also [23] a reduced efficacy of
glasshouse and field soil. Heinonsalo et al. [84] suggested that dual inoculation comparing with single inoculations with
the absence of positive bacterial effect could be due to the too two rhizobial strains. TRF composition was more diverse in
late inoculation of bacteria on already mycorrhized seedlings the case of single inoculations. Some of these differentiating
and that the mycorrhiza helper bacteria promoted the pre- TRFs could be affiliated to the genera known as anaerobic
symbiotic survival of the fungus in the soil. Transient effects nitrogen-fixing consortia and phytohormone producers (i.e.,
on soil microbial communities were found following the Clostridium, Bacillus, Stenotrophomonas, and Xanthomonas),
inoculation with biocontrol agents, such as P. fluorescens [86], which may explain the difference in plant growth. Therefore,
Streptomyces melanosporofaciens [87], and Corynebacterium combination of inoculants will not necessarily produce an
glutamicumin [88]. In a recent work, the effect of P. fluorescens additive or synergic effect, but rather a competitive process,
2P24 on soil fungal community in cucumber rhizosphere was and, hence, growth enhancement could be reduced or even-
detected with T-RFLP and DGGE [25]. The results showed tually disappear. Consequently, the effects on soil microbial
that inoculation had a transient significant effect on soil communities will be also unpredictable.
5. Conclusion [6] S. Peighami-Ashnaei, A. Sharifi-Tehrani, M. Ahmadzadeh, and

K. Behboudi, “Interaction of different media on production
Microbial inoculation may cause tremendous changes in the and biocontrol efficacy of Pseudomonas fluorescens P-35 and
number and composition of the taxonomic groups. However, Bacillus subtilis B-3 against grey mould of apple,” Journal of
the observed impacts depend largely on the techniques used Plant Pathology, vol. 91, no. 1, pp. 65–70, 2009.
to address the dynamics of soil microbial communities. Some [7] S. Dawar, S. Wahab, M. Tariq, and M. J. Zaki, “Application of
works showed no effect or a transient effect; however, others Bacillus species in the control of root rot diseases of crop plants,”
evidenced a long-term effect. These changes may influence Archives of Phytopathology and Plant Protection, vol. 43, no. 4,
plant and soil and thereby induce unpredictable feedback pp. 412–418, 2010.
reactions. Effects on plant growth and protection are not [8] H. Antoun and D. Prevost, “Ecology of plant growth promoting
necessarily resulting from a direct effect of the inoculated rhizobacteria,” in PGPR: Biocontrol and Biofertilization, Z. A.
strain and may be related to induction or repression of Siddiqui, Ed., pp. 1–38, 2005.
resident microbial populations. There may be also synergis- [9] X. Zhuang, J. Chen, H. Shim, and Z. Bai, “New advances
tic/antagonistic interactions between target and nontarget in plant growth-promoting rhizobacteria for bioremediation,”
effects. These changes could well lead to changes in beneficial Environment International, vol. 33, no. 3, pp. 406–413, 2007.
soil functions such as nitrogen fixation or N-cycling bacteria. [10] B. S. Saharan and V. Nehra, “Plant growth promoting rhizobac-
The extent of these changes on soil biology is still not well teria: a critical review,” Life Sciences and Medicine Research, vol.
documented and needs to be further assessed. The major 21, pp. 1–30, 2011.
concern remains regarding how the impact on taxonomic [11] A. C. Kennedy, “Bacterial diversity in agroecosystems,” Agricul-
groups can be related to effects on functional capabilities of ture, Ecosystems and Environment, vol. 74, no. 1–3, pp. 65–76,
the soil microbial communities. The dynamics of these effects 1999.
in relation to the host crop, the side-distance effect, the mid- [12] P. Nannipieri, J. Ascher, M. T. Ceccherini, L. Landi, G.
term and long-term effects, the crop-rotation effect, and site Pietramellara, and G. Renella, “Microbial diversity and soil
variation are still not understood and need to be further functions,” European Journal of Soil Science, vol. 54, no. 4, pp.
investigated. Undesirable growing conditions, such as biotic 655–670, 2003.
and abiotic stresses, most likely contribute to inconstancy of [13] P. R. Hirsch, T. H. Mauchline, and I. M. Clark, “Culture-
results and further complicate the problem, but they should independent molecular techniques for soil microbial ecology,”
be expected as a normal functioning of agriculture. With Soil Biology and Biochemistry, vol. 42, no. 6, pp. 878–887, 2010.
the evolution of the DNA-sequencing techniques and their [14] F. Schwieger and C. C. Tebbe, “Effect of field inoculation with
accessibility for many working groups, more light will be shed Sinorhizobium meliloti L33 on the composition of bacterial
on the complexity of the metabolic potentials of soil microbial communities in rhizospheres of a target plant (Medicago sativa)
communities and their importance to soil ecosystem. and a non-target plant (Chenopodium album)—linking of 16S
rRNA gene-based single-strand conformation polymorphism
community profiles to the diversity of cultivated bacteria,”
Acknowledgments Applied and Environmental Microbiology, vol. 66, no. 8, pp.
3556–3565, 2000.
The authors would like to thank the anonymous reviewers
for their valuable comments and suggestions to improve this [15] D. Roesti, R. Gaur, B. N. Johri et al., “Plant growth stage, fer-
tiliser management and bio-inoculation of arbuscular mycor-
paper.
rhizal fungi and plant growth promoting rhizobacteria affect the
rhizobacterial community structure in rain-fed wheat fields,”
References Soil Biology and Biochemistry, vol. 38, no. 5, pp. 1111–1120, 2006.
[16] C. Felici, L. Vettori, E. Giraldi et al., “Single and co-inoculation
[1] G. Berg, “Plant-microbe interactions promoting plant growth of Bacillus subtilis and Azospirillum brasilense on Lycopersicon
and health: perspectives for controlled use of microorganisms esculentum: effects on plant growth and rhizosphere microbial
in agriculture,” Applied Microbiology and Biotechnology, vol. 84, community,” Applied Soil Ecology, vol. 40, no. 2, pp. 260–270,
no. 1, pp. 11–18, 2009. 2008.
[2] I. R. Kennedy and N. Islam, “The current and potential contri- [17] L. Herrmann, K. Sanon, A. M. Zoubeirou et al., “Seasonal
bution of asymbiotic nitrogen fixation to nitrogen requirements changes of bacterial communities in the rhizosphere of Acacia
on farms: a review,” Australian Journal of Experimental Agricul- senegal mature trees inoculated with Ensifer strains in Burkina
ture, vol. 41, no. 3, pp. 447–457, 2001. Faso and Niger,” Agriculture, Ecosystems and Environment, vol.
[3] S. Spaepen, J. Vanderleyden, and R. Remans, “Indole-3-acetic 157, pp. 47–53, 2012.
acid in microbial and microorganism-plant signaling,” FEMS [18] P. Van Dillewijn, P. J. Villadas, and N. Toro, “Effect of a
Microbiology Reviews, vol. 31, no. 4, pp. 425–448, 2007. Sinorhizobium meliloti strain with a modified putA gene on
[4] B. R. Glick, B. Todorovic, J. Czarny, Z. Cheng, J. Duan, and the rhizosphere microbial community of alfalfa,” Applied and
B. McConkey, “Promotion of plant growth by bacterial ACC Environmental Microbiology, vol. 68, no. 9, pp. 4201–4208, 2002.
deaminase,” Critical Reviews in Plant Sciences, vol. 26, no. 5-6, [19] E. Baudoin, S. Nazaret, C. Mougel, L. Ranjard, and Y. Moënne-
pp. 227–242, 2007. Loccoz, “Impact of inoculationwith the phytostimulatory PGPR
[5] F. A. Mohiddin, M. R. Khan, and S. M. Khan, “Why Tricho- Azospirillum lipoferum CRT1 on the genetic structure of the
derma is considered super hero (super fungus) against the evil rhizobacterial community of field-grown maize,” Soil Biology
parasites?” Plant Pathology Journal, vol. 9, pp. 1–11, 2010. and Biochemistry, vol. 41, no. 2, pp. 409–413, 2009.
[20] O. Schumpp and W. J. Deakin, “How inefficient rhizobia [33] G. Rastogi and R. K. Sani, “Molecular techniques to assess
prolong their existence within nodules,” Trends in Plant Science, microbial community structure, function, and dynamics in the
vol. 15, no. 4, pp. 189–195, 2010. environment,” in Microbes and Microbial Technology: Agricul-
[21] A. Lerner, Y. Herschkovitz, E. Baudoin et al., “Effect of Azospir- tural and Environmental Applications, I. Ahmad, F. Ahmad, and
illum brasilense inoculation on rhizobacterial communities J. Pichtel, Eds., pp. 29–57, 2011.
analyzed by denaturing gradient gel electrophoresis and auto- [34] G. Benckiser and S. S. Bamforth, “Role of pathogens, signal
mated ribosomal intergenic spacer analysis,” Soil Biology and recalcitrance, and organisms shifting for ecosystem recupera-
Biochemistry, vol. 38, no. 6, pp. 1212–1218, 2006. tion: a review,” Agronomy for Sustainable Development, vol. 31,
[22] D. Trabelsi, H. Ben Ammar, A. Mengoni, and R. Mhamdi, no. 1, pp. 205–215, 2011.
“Appraisal of the crop-rotation effect of rhizobial inoculation on [35] D. L. Jones and P. Hinsinger, “The rhizosphere: complex by
potato cropping systems in relation to soil bacterial communi- design,” Plant and Soil, vol. 312, no. 1-2, pp. 1–6, 2008.
ties,” Soil Biology and Biochemistry, vol. 54, pp. 1–6, 2012. [36] S. B. von Bodman, W. D. Bauer, and D. L. Coplin, “Quorum
[23] D. Trabelsi, A. Mengoni, H. Ben Ammar, and R. Mhamdi, sensing in plant-pathogenic bacteria,” Annual Review of Phy-
“Effect of on-field inoculation of Phaseolus vulgaris with rhizo- topathology, vol. 41, pp. 455–482, 2003.
bia on soil bacterial communities,” FEMS Microbiology Ecology, [37] N. A. Whitehead, A. M. L. Barnard, H. Slater, N. J. L. Simpson,
vol. 77, no. 1, pp. 211–222, 2011. and G. P. C. Salmond, “Quorum-sensing in Gram-negative
bacteria,” FEMS Microbiology Reviews, vol. 25, no. 4, pp. 365–
[24] I. E. Garcı́a de Salamone, J. M. Funes, L. P. Di Salvo et
404, 2001.
al., “Inoculation of paddy rice with Azospirillum brasilense
and Pseudomonas fluorescens: impact of plant genotypes on [38] A. Crépin, C. Barbey, A. Cirou et al., “Biological control of
rhizosphere microbial communities and field crop production,” pathogen communication in the rhizosphere: a novel approach
Applied Soil Ecology, vol. 61, pp. 196–204, 2012. applied to potato soft rot due to Pectobacterium atrosepticum,”
Plant and Soil, vol. 358, no. 1-2, pp. 27–37, 2012.
[25] G. Gao, D. Yin, S. Chen et al., “Effect of biocontrol agent
[39] H. P. Bais, “Shoot the messages not the messengers,” Plant and
Pseudomonas fluorescens 2P24 on soil fungal community in
Soil, vol. 358, pp. 7–10, 2012.
cucumber rhizosphere using T-RFLP and DGGE,” PLoS ONE,
vol. 7, no. 2, Article ID e31806, 2012. [40] T. C. J. Turlings, I. Hiltpold, and S. Rasmann, “The impor-
tance of root-produced volatiles as foraging cues for ento-
[26] K. H. Babić, K. Schauss, B. Hai et al., “Influence of different
mopathogenic nematodes,” Plant and Soil, vol. 358, pp. 51–60,
Sinorhizobium meliloti inocula on abundance of genes involved
2012.
in nitrogen transformations in the rhizosphere of alfalfa (Med-
[41] U. Irshad, A. Brauman, C. Villenave, and C. Plassard, “Phos-
icago sativa L.),” Environmental Microbiology, vol. 10, no. 11, pp.
phorus acquisition from phytate depends on efficient bacterial
2922–2930, 2008.
grazing, irrespective of the mycorrhizal status of Pinus pinaster,”
[27] J. Kozdrój, J. T. Trevors, and J. D. Van Elsas, “Influence Plant and Soil, vol. 358, no. 1-2, pp. 155–168, 2012.
of introduced potential biocontrol agents on maize seedling
[42] J. Jin, C. Tang, R. Armstrong, and P. Sale, “Phosphorus supply
growth and bacterial community structure in the rhizosphere,”
enhances the response of legumes to elevated CO2 (FACE) in a
Soil Biology and Biochemistry, vol. 36, no. 11, pp. 1775–1784,
phosphorus-deficient vertisol,” Plant and Soil, vol. 358, no. 1-2,
2004.
pp. 91–104, 2012.
[28] A. D. Naiman, A. Latrónico, and I. E. Garcı́a de Salamone, [43] A. Kawasaki, E. R. Watson, and M. A. Kertesz, “Indirect
“Inoculation of wheat with Azospirillum brasilense and Pseu- effects of polycyclic aromatic hydrocarbon contamination on
domonas fluorescens: impact on the production and culturable microbial communities in legume and grass rhizospheres,”
rhizosphere microflora,” European Journal of Soil Biology, vol. Plant and Soil, pp. 1–14, 2011.
45, no. 1, pp. 44–51, 2009.
[44] R. Child, C. D. Miller, Y. Liang et al., “Polycyclic aromatic
[29] I. E. G. de Salamone, L. P. Di Salvo, J. S. E. Ortega, P. M. F. hydrocarbon-degrading Mycobacterium isolates: their associa-
B. Sorte, S. Urquiaga, and K. R. S. Teixeira, “Field response tion with plant roots,” Applied Microbiology and Biotechnology,
of rice paddy crop to Azospirillum inoculation: physiology of vol. 75, no. 3, pp. 655–663, 2007.
rhizosphere bacterial communities and the genetic diversity of [45] Y. Bashan, G. Holguin, and L. E. De-Bashan, “Azospirillum-
endophytic bacteria in different parts of the plants,” Plant and plant relationships: physiological, molecular, agricultural, and
Soil, vol. 336, no. 1, pp. 351–362, 2010. environmental advances (1997–2003),” Canadian Journal of
[30] K. Schreiner, A. Hagn, M. Kyselková et al., “Comparison of Microbiology, vol. 50, no. 8, pp. 521–577, 2004.
barley succession and take-all disease as environmental factors [46] C. Wamberg, S. Christensen, I. Jakobsen, A. K. Müller, and
shaping the rhizobacterial community during take-all decline,” S. J. Sørensen, “The mycorrhizal fungus (Glomus intraradices)
Applied and Environmental Microbiology, vol. 76, no. 14, pp. affects microbial activity in the rhizosphere of pea plants (Pisum
4703–4712, 2010. sativum),” Soil Biology and Biochemistry, vol. 35, no. 10, pp.
[31] J. G. Caporaso, C. L. Lauber, W. A. Walters et al., “Global 1349–1357, 2003.
patterns of 16S rRNA diversity at a depth of millions of [47] L. Lioussanne, F. Perreault, M. Jolicoeur, and M. St-Arnaud,
sequences per sample,” Proceedings of the National Academy of “The bacterial community of tomato rhizosphere is modified
Sciences of the United States of America, vol. 108, no. 1, pp. 4516– by inoculation with arbuscular mycorrhizal fungi but unaf-
4522, 2011. fected by soil enrichment with mycorrhizal root exudates or
[32] I. Han, T. K. Lee, J. Han, T. V. Doan, S. B. Kim, and J. Park, inoculation with Phytophthora nicotianae,” Soil Biology and
“Improved detection of microbial risk of releasing genetically Biochemistry, vol. 42, no. 3, pp. 473–483, 2010.
modified bacteria in soil by using massive sequencing and [48] E. A. Robleto, J. Borneman, and E. W. Triplett, “Effects of
antibiotic resistance selection,” Journal of Hazardous Material, bacterial antibiotic production on rhizosphere microbial com-
vol. 227-228, pp. 172–178, 2012. munities from a culture-independent perspective,” Applied and
Environmental Microbiology, vol. 64, no. 12, pp. 5020–5022, [64] P. Marschner and S. Timonen, “Timonen, Bacterial community
1998. composition and activity in rhizosphere of roots colonized
[49] Y. Z. Zhang, E. T. Wang, M. Li et al., “Effects of rhizobial by arbuscular mycorrhizal fungi,” in Microbial Activity in the
inoculation, cropping systems and growth stages on endophytic Rhizosphere, K. G. Mukerji, C. Manoharachary, and J. Singh,
bacterial community of soybean roots,” Plant and Soil, vol. 347, Eds., pp. 139–154, Springer, Berlin, Germany, 2006.
no. 1, pp. 147–161, 2011. [65] G. Andrade, K. L. Mihara, R. G. Linderman, and G. J. Beth-
[50] P. Offre, B. Pivato, S. Siblot et al., “Identification of bacterial lenfalvay, “Bacteria from rhizosphere and hyphosphere soils of
groups preferentially associated with mycorrhizal roots of different arbuscular-mycorrhizal fungi,” Plant and Soil, vol. 192,
Medicago truncatula,” Applied and Environmental Microbiology, no. 1, pp. 71–79, 1997.
vol. 73, no. 3, pp. 913–921, 2007. [66] J. M. Barea, “Mycorrhiza-bacteria interactions on plant growth
[51] Z. P. Xie, C. Staehelin, H. Vierheilig et al., “Rhizobial nodulation promotion,” in Plant Growth Promoting Rhizobacteria, A.
factors stimulate mycorrhizal colonization of nodulating and Ogoshi, L. Kobayashi, Y. Homma, F. Kodama, N. Kondon, and
nonnodulating soybeans,” Plant Physiology, vol. 108, no. 4, pp. S. Akino, Eds., pp. 150–158, OECD Press, Paris, France, 1997.
1519–1525, 1995. [67] J. M. Barea, R. Azcón, and C. Azcón-Aguilar, “Mycorrhizo-
[52] Y. Okon and C. A. Labandera-Gonzalez, “Agronomic applica- sphere interactions to improve plant fitness and soil quality,”
tions of Azospirillum: an evaluation of 20 years worldwide field Antonie van Leeuwenhoek, International Journal of General and
inoculation,” Soil Biology and Biochemistry, vol. 26, no. 12, pp. Molecular Microbiology, vol. 81, no. 1–4, pp. 343–351, 2002.
1591–1601, 1994. [68] E. Gamalero, M. G. Martinotti, A. Trotta, P. Lemanceau, and G.
[53] S. Dobbelaere, A. Croonenborghs, A. Thys et al., “Responses of Berta, “Morphogenetic modifications induced by Pseudomonas
agronomically important crops to inoculation with Azospiril- fluorescens A6RI and Glomus mosseae BEG12 in the root system
lum,” Australian Journal of Plant Physiology, vol. 28, no. 9, pp. of tomato differ according to plant growth conditions,” New
871–879, 2001. Phytologist, vol. 155, no. 2, pp. 293–300, 2002.
[54] M. Malhotra and S. Srivastava, “An ipdC gene knock-out [69] M. St-Arnaud and V. Vujanovic, “Effect of the arbuscular myc-
of Azospirillum brasilense strain SM and its implications on orrhizal symbiosis on plant diseases and pests,” in Arbuscular
indole-3-acetic acid biosynthesis and plant growth promotion,” Mycorrhizae in Crop Production, C. Hamel and C. Plenchette,
International Journal of General and Molecular Microbiology, Eds., pp. 67–122, Haworth’s Food Products Press, New York, NY,
vol. 93, no. 4, pp. 425–433, 2008. USA, 2007.
[55] C. Jacoud, D. Faure, P. Wadoux, and R. Bally, “Development [70] B. Li, S. Ravnskov, G. Xie, and J. Larsen, “Biocontrol of Pythium
of a strain-specific probe to follow inoculated Azospirillum damping-off in cucumber by arbuscular mycorrhiza-associated
lipoferum CRT1 under field conditions and enhancement of bacteria from the genus Paenibacillus,” BioControl, vol. 52, no.
maize root development by inoculation,” FEMS Microbiology 6, pp. 863–875, 2007.
Ecology, vol. 27, no. 1, pp. 43–51, 1998. [71] F. A. Solı́s-Domı́nguez, A. Valentı́n-Vargas, J. Chorover, and
[56] O. Steenhoudt and J. Vanderleyden, “Azospirillum, a free- R. M. Maier, “Effect of arbuscular mycorrhizal fungi on plant
living nitrogen-fixing bacterium closely associated with grasses: biomass and the rhizosphere microbial community structure of
genetic, biochemical and ecological aspects,” FEMS Microbiol- mesquite grown in acidic lead/zinc mine tailings,” Science of the
ogy Reviews, vol. 24, no. 4, pp. 487–506, 2000. Total Environment, vol. 409, no. 6, pp. 1009–1016, 2011.
[57] A. Schüßler, D. Schwarzott, and C. Walker, “A new fungal [72] M. Mar Vázquez, S. César, R. Azcón, and J. M. Barea, “Interac-
phylum, the Glomeromycota: phylogeny and evolution,” Myco- tions between arbuscular mycorrhizal fungi and other micro-
logical Research, vol. 105, no. 12, pp. 1413–1421, 2001. bial inoculants (Azospirillum, Pseudomonas, Trichoderma) and
[58] S. E. Smith and D. J. Read, Mycorrhizal Symbiosis, Academic their effects on microbial population and enzyme activities in
Press, London, UK, 3rd edition, 2008. the rhizosphere of maize plants,” Applied Soil Ecology, vol. 15,
[59] V. Artursson, R. D. Finlay, and J. K. Jansson, “Interactions no. 3, pp. 261–272, 2000.
between arbuscular mycorrhizal fungi and bacteria and their [73] M. Filion, M. St-Arnaud, and J. A. Fortin, “Direct inter-
potential for stimulating plant growth,” Environmental Micro- action between the arbuscular mycorrhizal fungus Glomus
biology, vol. 8, no. 1, pp. 1–10, 2006. intraradices and different rhizosphere microorganisms,” New
[60] R. G. Linderman, “Mycorrhizal interactions with the rhizo- Phytologist, vol. 141, no. 3, pp. 525–533, 1999.
sphere microflora: the mycorrhizosphere effect,” Phytopathol- [74] A. Elsen, D. Gervacio, R. Swennen, and D. De Waele, “AMF-
ogy, vol. 78, pp. 366–371, 1988. induced biocontrol against plant parasitic nematodes in Musa
[61] T. R. Cavagnaro, L. E. Jackson, J. Six et al., “Arbuscular sp.: a systemic effect,” Mycorrhiza, vol. 18, no. 5, pp. 251–256,
mycorrhizas, microbial communities, nutrient availability, and 2008.
soil aggregates in organic tomato production,” Plant and Soil, [75] M. R. Khan, N. Khan, and S. M. Khan, “Evaluation of agricul-
vol. 282, no. 1-2, pp. 209–225, 2006. tural materials as substrate for mass culture of fungal biocontrol
[62] A. Albertsen, S. Ravnskov, H. Green, D. F. Jensen, and J. Larsen, agents of fusarial wilt and root knot nematode diseases,” Annals
“Interactions between the external mycelium of the mycorrhizal of Applied Biology, vol. 22, pp. 50–51, 2001.
fungus Glomus intraradices and other soil microorganisms as [76] B. R. Kerry, “Rhizosphere interactions and the exploitation of
affected by organic matter,” Soil Biology and Biochemistry, vol. microbial agents for the biological control of plant-parasitic
38, no. 5, pp. 1008–1014, 2006. nematodes,” Annual Review of Phytopathology, vol. 38, pp. 423–
[63] P. A. Olsson, E. Bååth, I. Jakobsen, and B. Söderström, “Soil 441, 2000.
bacteria respond to presence of roots but not to mycelium of [77] M. R. Khan, Plant Nematodes-Methodology, Morphology, Sys-
arbuscular mycorrhizal fungi,” Soil Biology and Biochemistry, tematics, Biology and Ecology, Science Publishers, Enfield, NH,
vol. 28, no. 4-5, pp. 463–470, 1996. USA, 2008.
[78] K. K. Pal, K. V. B. R. Tilak, A. K. Saxena, R. Dey, and C. S. and bacterial rhizosphere structure after inoculation with PGPR
Singh, “Antifungal characteristics of a fluorescent Pseudomonas Bacillus (B. licheniformis CECT 5106 and B. pumilus CECT
strain involved in the biological control of Rhizoctonia solani,” 5105),” Applied Soil Ecology, vol. 20, no. 2, pp. 75–84, 2002.
Microbiological Research, vol. 155, no. 3, pp. 233–242, 2000.
[79] M. R. Khan, “Prospects of microbial control of root-knot
nematodes infecting vegetable crops,” in Biotechnology: Plant
Health Management, N. Sharma and H. B. Singh, Eds., pp.
643–665, International Book Distributing Company, Lucknow,
India, 2007.
[80] R. Singh, T. N. Parameswaran, E. V. S. P. Rao et al., “Effect
of arbuscular mycorrhizal fungi and Pseudomonas fluorescens
on root-rot and wilt, growth and yield of Coleus forskohlii,”
Biocontrol Science and Technology, vol. 19, no. 8, pp. 835–841,
2009.
[81] L. Zelles, “Fatty acid patterns of phospholipids and lipopolysac-
charides in the characterisation of microbial communities in
soil: a review,” Biology and Fertility of Soils, vol. 29, no. 2, pp.
111–129, 1999.
[82] H. Yao, Z. He, M. J. Wilson, and C. D. Campbell, “Microbial
biomass and community structure in a sequence of soils with
increasing fertility and changing land use,” Microbial Ecology,
vol. 40, no. 3, pp. 223–237, 2000.
[83] W. F. Pfender, V. P. Fieland, L. M. Ganio, and R. J. Seidler,
“Microbial community structure and activity in wheat straw
after inoculation with biological control organisms,” Applied
Soil Ecology, vol. 3, no. 1, pp. 69–78, 1996.
[84] J. Heinonsalo, P. Frey-Klett, J. C. Pierrat, J. L. Churin, D.
Vairelles, and J. Garbaye, “Fate, tree growth effect and potential
impact on soil microbial communities of mycorrhizal and
bacterial inoculation in a forest plantation,” Soil Biology and
Biochemistry, vol. 36, no. 2, pp. 211–216, 2004.
[85] P. Frey-Klett, J. C. Pierrat, and J. Garbaye, “Location and survival
of mycorrhiza helper Pseudomonas fluorescens during establish-
ment of ectomycorrhizal symbiosis between Laccaria bicolor
and Douglas fir,” Applied and Environmental Microbiology, vol.
63, no. 1, pp. 139–144, 1997.
[86] A. Natsch, C. Keel, N. Hebecker, E. Laasik, and G. Défago,
“Impact of Pseudomonas fluorescens strain CHA0 and a deriva-
tive with improved biocontrol activity on the culturable resident
bacterial community on cucumber roots,” FEMS Microbiology
Ecology, vol. 27, no. 4, pp. 365–380, 1998.
[87] K. Prévost, G. Couture, B. Shipley, R. Brzezinski, and C.
Beaulieu, “Effect of chitosan and a biocontrol streptomycete on
field and potato tuber bacterial communities,” BioControl, vol.
51, no. 4, pp. 533–546, 2006.
[88] W. Vahjen, J. C. Munch, and C. C. Tebbe, “Carbon source
utilization of soil extracted microorganisms as a tool to detect
the effects of soil supplemented with genetically engineered and
non-engineered Corynebacterium glutamicum and a recom-
binant peptide at the community level,” FEMS Microbiology
Ecology, vol. 18, no. 4, pp. 317–328, 1995.
[89] W. Janisiewicz, “Ecological diversity, niche overlap, and coexis-
tence of antagonists used in developing mixtures for biocontrol
of postharvest diseases of apples,” Phytopathology, vol. 86, no. 5,
pp. 473–479, 1996.
[90] S. Marimuthu, P. Subbian, V. Ramamoorthy, and R. Samiyap-
pan, “Synergistic effect of combined application of Azospirillum
and Pseudomonas fluorescens with inorganic fertilizers on root
rot incidence and yield of cotton,” Journal of Plant Diseases and
Protection, vol. 109, no. 6, pp. 569–577, 2002.
[91] A. Probanza, J. A. Lucas Garcı́a, M. Ruiz Palomino, B. Ramos,
and F. J. Gutiérrez Mañero, “Pinus pinea L. seedling growth
http://dx.doi.org/10.1155/2013/491091
Research Article
Plant Growth Promotion Potential Is Equally Represented in
Diverse Grapevine Root-Associated Bacterial Communities from
Different Biopedoclimatic Environments
Ramona Marasco,1 Eleonora Rolli,1 Marco Fusi,1 Ameur Cherif,2 Ayman Abou-Hadid,3
Usama El-Bahairy,3 Sara Borin,1 Claudia Sorlini,1 and Daniele Daffonchio1
1
Department of Food, Environment, and Nutritional Sciences, University of Milan, Via Celoria 2, 20133 Milan, Italy
2
Laboratory of Microorganisms and Active Biomolecules, University of Tunis El Manar, Campus Universitaire, Rommana 1068,
Tunis BP 94, Tunisia and Laboratory BVBGR, ISBST, University of Manouba, La Manouba 2010, Tunisia
3
Department of Horticulture, Faculty of Agriculture, Ain Shams University, Shubra Elkheima, Cairo, Egypt
Correspondence should be addressed to Daniele Daffonchio; daniele.daffonchio@unimi.it
Received 13 March 2013; Revised 14 May 2013; Accepted 21 May 2013
Copyright © 2013 Ramona Marasco et al. This is an open access article distributed under the Creative Commons Attribution
cited.
Plant-associated bacteria provide important services to host plants. Environmental factors such as cultivar type and pedoclimatic
conditions contribute to shape their diversity. However, whether these environmental factors may influence the plant growth
promoting (PGP) potential of the root-associated bacteria is not widely understood. To address this issue, the diversity and PGP
potential of the bacterial assemblage associated with the grapevine root system of different cultivars in three Mediterranean
environments along a macrotransect identifying an aridity gradient were assessed by culture-dependent and independent
approaches. According to 16S rRNA gene PCR-DGGE, the structure of endosphere and rhizosphere bacterial communities was
highly diverse (𝑃 = 0.03) and was associated with a cultivar/latitudinal/climatic effect. Despite being diverse, the bacterial
communities associated with Egyptian grapevines shared a higher similarity with the Tunisian grapevines than those cultivated
in North Italy. A similar distribution, according to the cultivar/latitude/aridity gradients, was observed for the cultivable bacteria.
Many isolates (23%) presented in vitro multiple stress resistance capabilities and PGP activities, the most frequent being auxin
synthesis (82%), insoluble phosphate solubilisation (61%), and ammonia production (70%). The comparable numbers and types of
potential PGP traits among the three different environmental settings indicate a strong functional homeostasis of beneficial bacteria
associated with grape root.
1. Introduction (PGP) microorganisms that are naturally associated with

plants [3].
Grapevine is among the most ancient crops grown in the Little attention has been dedicated to bacteria associated
Mediterranean basin. Historically, grape and derived prod- with grapevine, and, subsequently, relatively few studies are
ucts held an important economic role in the area, and available. It was shown that grapevine tissues, including
this importance persists to the present day. Vineyards are flowers, berries pulp, and seeds, host an endophytic com-
extremely sensitive to phytopathogen attacks, and a huge munity that encompasses, among others, bacteria affiliated
interest has been devoted to understanding the mecha- to the Bacillus, Burkholderia, and Pseudomonas genera [4–9].
nisms of virulence [1] and environmental-friendly biocontrol Despite these studies, little information is currently available
approaches [2]. Certain biocontrol strategies rely on the on the taxonomic and functional diversity of the bacterial
exploitation of beneficial traits of plant growth-promoting communities associated with the endosphere and root system
of grapevine. Recently, we observed that Enterobacteriaceae 2. Materials and Methods

were dominant in the root system of five different Barbera
cv. rootstocks cultivated in Oltrepò Pavese, with a differential 2.1. Study Area and Sample Collection. Three different agro-
genera distribution and colonization of the rhizosphere and climatic regions of the Mediterranean basin were chosen for
endosphere [10]. sampling. During July 2008, grapevine root samples were
Even fewer studies have investigated the potential of collected from vineyards located in Mornag, North Tunisia,
grapevine-associated bacteria in promoting plant resistance (TN) and in an arid farm 30 km north-west of Cairo, Egypt
to abiotic stresses [11, 12]. Water deprivation reduces fruit (ET). In July 2009, grapevine roots were collected from a
yield remarkably and causes quality losses, and most of the vineyard of the “Le Frecce” farm, North Italy (IT). The sites
grapevine-growing regions worldwide suffer seasonal periods were located along a latitude gradient, from 44∘ to 30∘ N
of drought [13]. Although generally referred to as a drought- (IT: 44∘ 57󸀠 N, TN: 36∘ 84󸀠 N, ET: 30∘ 2󸀠 N). The roots (root
resistant plant, grapevine is severely affected in terms of fruit tissues and rhizosphere) of healthy grapevine plants were
yield and quality by the cooccurrence of elevated tempera- collected at 50–60 cm depth. After removing the roots, the
tures and high evaporation rates that reduce carbohydrate root-surrounding soil was collected and the bulk soil was
content in berries and cause wilting of leaves. Even at sampled at a distance of 4 m from the grapevine plants. All
temperate latitudes and cool climates, extensive periods of soil and root samples were collected under sterile conditions
drought coupled with certain soil types and topography in the using sterile tools. Recovered samples were stored at −20∘ C
vineyard may determine anticipated harvests and influence for molecular analysis or at 4∘ C for isolation and processed
phenology [14]. In both managed and natural ecosystems, the following day in the laboratory.
the interaction of plants with the associated native-drought
resistant microbiome is an important factor in supporting 2.2. Total DNA Extraction and PCR from All Root System
plant health and physiology under water stress [15, 16]. As Fractions. Grapevine roots with attached soil particles were
sessile organisms, plants adopt different strategies to persist placed in a 50 mL screw-cap tube containing 9 mL of phys-
in the face of unfavourable environmental settings, including iological solution (9 g/L NaCl) to separate the rhizosphere
phenotypic plasticity or “escape and migrate” mechanisms. soil by vortexing. The root tissues removed from the 50 mL
Rapid adaptation of plants to low soil moisture is achieved screw-cap tube were sterilized as described by Sun et al. [21]
through changes in the structure and functionality of the by immersion in 70% ethanol for 3 min, sodium hypochlorite
belowground microbiome [17]. The selective force exerted by solution (2.5% available Cl− ) for 5 min, and 70% ethanol for
water scarcity combined with unfavourable harsh environ- 30 seconds. After these treatments, the tissues were washed
mental conditions in arid lands enriches the rhizosphere of five times with sterile distilled water. The efficacy of the
beneficial bacteria that exhibit antagonistic activity against sterilization method was verified by plating the water of the
phytopathogens and promote plant resistance to drought last washing step on PAF medium (10 g/L proteose peptone,
[16, 18]. Recently, we observed that the treatment of grape 10 g/L hydrolyzed casein, 3 g/L MgSO4 , 1.5 g/L K2 HPO4 ,
plantlets with selected bacteria determined an increase in 10 mL/L glycerol, and 15 g/L agar for solid medium). Total
epigeal biomass and the formation of a larger root system DNA was extracted from the soil fractions (rhizosphere,
during drought stress [10]. root-surrounding soil, and bulk soil) and root tissues using
It is well known that bacteria play key roles in promot- a Power Soil kit (MoBio) and DNeasy Plant kit (Qiagen),
ing plant growth in conventional and extreme ecosystems respectively, according to the manufacturer’s procedure. The
[16, 19], and that a plethora of environmental factors such DNA was quantified and stored at −20∘ C until use. PCR
as the cultivar type or pedoclimatic conditions can affect amplification of the 16S rRNA gene was performed using the
and modulate the structure of bacterial microbiomes [20]. 907R and 357F primers, adding a GC-clamp to the forward
However, the influence of such environmental factors on primer [22]. PCR reaction was performed in 0.2 mL tubes
the PGP potential of root-associated bacteria is poorly in a final volume of 50 𝜇L containing the 1x diluted buffer,
understood. 1.5 mM MgCl2 , 5% DMSO, 0.12 mM of a mixture of dNTPs,
The present study aims to assess the range of bac- 0.3 𝜇M of each primer, 1 U Taq polymerase, and 10 ng of
terial diversity and functional PGP potential of cultur- template. When necessary, DNA was properly diluted. The
able bacteria associated with grapevine roots growing in amplification program consisted of an initial denaturing step
three different agrosystems in the Mediterranean basin, at 94∘ C for 4 min, followed by 10 cycles of 94∘ C for 0.5 min,
in order to evaluate whether such PGP potential is inde- 61∘ C for 1 min, and 72∘ C for 1 min, followed by further 20
pendent from the specific environmental conditions. The cycles at 94∘ C for 0.5 min, 56∘ C for 1 min, 72∘ C for 1 min, and a
sampling sites were located in North Italy, North Tunisia, final extension at 72∘ C for 7 min. Two 𝜇L of the PCR products
and North Egypt, drawing a latitudinal/aridity macro- was analyzed by electrophoresis in 1% agarose gels.
transect in the Mediterranean basin. The structure of the
bacterial communities associated with the root endosphere
and soil of grapevines in the three sites was dissected 2.3. PCR-DGGE and Profile Analysis. DGGE was per-
by 16S rRNA gene-based PCR-DGGE (denaturing gradient formed using polyacrylamide gel (8% of a 37 : 1 acrylamide-
gel electrophoresis) analysis. The results were compared bisacrylamide mixture in a Tris acetate EDTA (TAE) 1x
with the diversity of the culturable bacteria and their PGP buffer, 0.75 mm thick, 16 × 10 cm) with a 45–60% denaturant
potential. gradient. Gels were run overnight at 90 V in TAE 1x buffer
at 60∘ C in DCode apparatus (Bio-Rad, Italy). The gels were Isolates were dereplicated using the ITS-PCR fingerprinting
stained with 1x Sybr Green (Life Technologies) and scanned protocol [30–32]. Two 𝜇L of the PCR products was checked
with gel photo GS-800 system. The DGGE bands were excised by electrophoresis in 1.5% agarose gel and stained with
from the gels using a sterile scalpel and eluted in 50 𝜇L water ethidium bromide. Gel images were captured using Gel Doc
at 37∘ C for 3 hours. The DNA eluted from DGGE bands 2000 system (Bio-Rad, Milan, Italy), and bacteria redundancy
was amplified using 907R and 357F primers (without the was reduced by evaluating the different ITS profiles. One
GC-clamp) [22]. The PCR was performed in a final volume strain per each ITS haplotype was used in the phylogenetic
of 50 𝜇L with the same conditions as above and using the analysis and for further experiments. A total of 331 strains
following protocol: 95∘ C for 5 min, 30 cycles of 95∘ C for isolated on ACCd [28], R2A (Oxoid), and KB [29] media were
1 min, 61∘ C for 1 min, 72∘ C for 1 min, and a final extension at characterized by 16S rRNA gene sequencing. The reaction
72∘ C for 7 min. The PCR products obtained were sequenced mixture contained the diluted buffer, 1.5 mM of MgCl2 ,
by Macrogen Inc. (Korea). The DGGE band patterns were 0.12 mM of a mixture of dNTPs, 0.3 𝜇M of each primer,
converted to a binary dataset by using ImageJ software 1 U of Taq polymerase, and 10 ng of template. The universal
[23]. Principal component analysis (PCA) was carried out primers were 27F (3󸀠 -AGAGTTTGATCMTGGCTCAG-5󸀠 )
using XLSTAT (version 7.5.2 Addinsoft, France). Analysis and 1492R (3󸀠 -CTACGGCTACCTTGTTACGA-5󸀠 ). Condi-
of variance along the PCA axis was evaluated using the tions for amplification consisted of an initial denaturation at
statistical test ANOVA and Student’s t test with significance 94∘ C for 4 min, followed by 35 cycles of 94∘ C for 0.5 min, 55∘ C
at 𝑃 ≤ 0.05. for 1 min, 72∘ C for 2 min, and a final extension at 72∘ C for
The nonparametric statistical test PERMANOVA [24] 10 min. The PCR products were checked by electrophoresis
was used to test the null hypothesis, in which there in 1% agarose gel, and sequencing service was performed by
were no differences between microbial assemblages of sites; Macrogen Inc., South Korea.
PERMANOVA was conducted with the factors of site (fixed,
orthogonal, and three levels IT, TN, and ET) and microbial
community (fixed, orthogonal, two levels, endophyte, and 2.6. Characterization of Plant Growth Promoting Activity and
rhizosphere). Abiotic Stress Resistance. The 331 bacterial strains identified
To test for significant relationships among the micro- were screened for production of indole acetic acid (IAA),
biological assemblage and climate traits [25], a distance- siderophores, exopolysaccharide (EPS) and ammonia pro-
based multivariate analysis for a linear model (DistLM) and ductions, mineral phosphate solubilization, protease activity
distance-based redundancy analysis (dbRDA) [26] were used. and tolerance to drought, salt, and osmotic stress. The ability
PERMANOVA and DistLM were performed with software of isolated strains to produce IAA was evaluated in the
PERMANOVA + for PRIMER 6 [27]. original liquid medium supplemented with L-tryptophan
(100 mg/L) as described by Bric et al. [33]. Strains were
considered as IAA-producers for concentrations higher than
2.4. Isolation of Cultivable Bacteria. One gram of each soil
2 𝜇g/mL. Pure IAA (Sigma-Aldrich Co., Italy) was used to
fraction (rhizosphere, root-surrounding soil, and bulk soil)
prepare the standard curve and to quantify the amount of IAA
and one gram of triturated root were used as inoculum
produced. Siderophore release was detected in a modified
for ACC-deaminase enrichment culture as described by
PAF medium (without Fe) using the Chrome Azurol S (CAS)
Penrose and Glick [28]. The medium was supplemented with
method described by Schwyn and Neilands [34]. Bacterial
100 𝜇g/mL of the fungicide cycloheximide. CFUs per gram of
culture was streaked on a half plate containing growth
sample were calculated, and 50 colonies from each fraction
media. Plates were incubated at 30∘ C for 7 days, and the
were randomly selected and propagated on PAF medium
formation of orange or pink halos indicated the presence of
plates. One gram of each fraction sample, suspended in 9 mL
siderophore. The mineral P-solubilizing ability of the strains
of sterile physiological solution (9 g/L NaCl), was diluted in
was determined on Pikovskaya’s liquid medium amended
10-fold series and plated in triplicate onto KB medium (20 g/L
with 0.5% tricalcium phosphate [Ca3 (PO4 )2 ] as inorganic P
peptone, 1.5 g/L dipotassium sulphate, 3.2 g/L magnesium
[35]. Exopolysaccharides (EPSs) production was estimated as
dichloride, 10 mL/L glycerol, and pH = 7.2) [29] and on R2A
described by Santarella et al. [36], using the modified Weaver
medium (Oxoid). After three days of incubation at 30∘ C, a
mineral media enriched with 20 g/L sucrose. Bacterial isolates
total count was performed, and twelve colonies per medium
were tested for the production of ammonia in peptone water
per fraction were randomly selected. For colony purification,
(peptone 5 g/L). Freshly grown cultures were inoculated
all the isolated colony types were spread three times on the
in 5 mL peptone water in each tube and incubated for
original medium. A total of 769 purified isolates were frozen
72 h at 30∘ C. Nessler’s reagent (0.5 mL) was added in each
in 25% glycerol at −80∘ C until use.
tube. Development of yellow-brownish color indicated NH3
production [37]. Proteolytic activity (casein degradation) was
2.5. DNA Extraction of Isolates, Dereplication and PCR of 16S determined from clearing zones in skim milk agar after 4 days
rRNA Gene. Purified bacterial colonies were resuspended in of incubation at 30∘ C as described by Nielsen and Sørensen
50 𝜇L of sterile TE (10 mM Tris/HCl, pH 8, and 1 mM EDTA) [38]. Resistance to salt was assessed by adding 5, 8, and
in 1.5 mL tubes. Tubes were incubated at 95∘ C for 8 min 10% sodium chloride to culture media and incubating the
and centrifuged at 13000 rpm for 10 min. The supernatant plate at 30∘ C for 7 days. The ability to grow at 4, 42, and
containing the DNA was stored at −20∘ C and used for PCR. 50∘ C was verified in solid media placed in incubators set at
the indicated temperatures, and the growth was qualitatively in nutrient-poor soils and under severe abiotic stresses,
scored after 7 days. Tolerance to osmotic stress was evaluated as previously observed for herbaceous and arboreal plants
by adding 10–20% polyethylene glycol (PEG) to the original grown in arid lands [16, 41].
liquid media. We focused on the comparison of the structure of endo-
sphere and rhizosphere-associated microbial communities
of grapevines along the investigated transect (Figure 2(a)).
2.7. Nucleotide Sequence Identification and Accession Numbers.
A differentiation of the bacterial diversity among the two
Analysis of sequences was performed with the basic sequenc-
different fractions was observed by PCA of the PCR-DGGE
ing alignment BLAST program run against the database
profiles along axis 1 (𝑃 = 0.0032) and axis 2 (𝑃 =
(http://blast.ncbi.nlm.nih.gov/Blast.cgi). The sequences of
0.0006) (Figures 2(b) and 2(c)). The separation between the
the partial 16S rRNA genes for isolates were deposited in
rhizospheric and endophytic communities along the transect
the GeneBank database under the accession numbers from
was confirmed by statistical analysis (PERMANOVA, 𝐹 =
HF584760 to HF585082, from HE610893 to HE610899, and
30.36; df = 5; 𝑃 = 0.0001). A latitudinal gradient effect was
from HF562892 to HF562897. The DGGE sequences were
observed in the distribution of the bacterial community in E
submitted under the accession numbers from HF678228 to
and R (Figure 2(a)). Bacterial communities associated with
HF678357.
grapevines cultivated in Egypt clustered with those of plants
cultivated in Tunisia, while the rhizosphere community of
3. Results and Discussion grapevine cultivated in Italy was separated in a different
cluster, showing a significant difference from the Egyptian
3.1. Cultivation-Independent Analysis of Grapevine Root- and Tunisian communities along axis 2 (𝑃 < 0.05). Further-
Associated Bacterial Communities. The diversity of bacterial more, the PERMANOVA analysis performed considering the
communities associated with the grapevine root system from geographical origin showed a significant statistical difference
each of the three studied regions was investigated through in the distribution of the bacterial communities, indicating
the analysis of the diversity of the 16S rRNA gene in the an influence of the site of origin in shaping the microbial
root tissues (E), rhizosphere (R), and root-surrounding soil community of both fractions (PERMANOVA, 𝐹 = 30.45;
(S) fractions of three replicate plants including bulk soil (B) df = 5; 𝑃 = 0.0078). As previously reported, aridity seems to
as a comparison (Figure 1(a)). Multiple-band PCR-DGGE be the driving force influencing the structure of both Archea
profiles were observed in all fractions (E, R, and S) of the and bacterial communities in bare soils [42].
three grapevine root systems and in the respective bulk soils DISTLM multivariate analysis was performed in order
(Figure 1(a)). The endophyte fractions from the soil samples to correlate the differences in the structure of microbial
were different, showing the simplest profile composed of communities in the different agroclimatic sites with abiotic
a limited number of bands compared to the other root environmental parameters (Table 1). The selection of soil
system samples (Figure 1(a)). This observation was confirmed microorganisms by plants is a complex process controlled
by PCA of the PCR-DGGE band patterns that showed a by several factors, often not easily correlated to geochemical
sharp separation between E-associated bacterial commu- settings [43]. Nevertheless, we observed that among abi-
nities and those of soil fractions (Figures 1(b) and 1(c)). otic factors, annual rainfall exerted a statistical significant
The reduced microbial diversity in root tissues may reflect influence in determining the structure of the root system-
specific physiological requirements to enter the interior of associated bacterial community (Table 1). It is possible that
the roots and establish as endophytic populations [39]. The a cooccurrence of harsh environmental factors, including a
soil fractions showed a multiple-band profile in the DGGE hot and dry climate and reduced rainfall, may reflect the
gels, indicating the hosting of a high number of different differences among the rhizosphere fractions in the three
bacterial taxa. Several shared bands were observed among analysed sites (Table 1).
the R, S, and B fractions, suggesting that similar bacteria may
have colonized different soil portions (Figure 1(a)). This was
observed in the case of grapevines cultivated in North Italy 3.2. Phylogenetic Affiliation of DGGE Bands Representative of
and Tunisia where R, S, and B fractions tended to cluster Grapevine Root-Associated Microbial Diversity. One hundred
together according to the PCA (Figure 1(b)). On the contrary, and twenty-nine bands, representative of all fractions,
a clear rhizosphere effect was observed in the root system of were excised from the DGGE gel and affiliated to five
grapevines growing in Egypt; the structure of the rhizosphere bacterial phyla by sequencing: Acidobacteria, Actinobacteria,
bacterial communities diverged from the S and B samples Firmicutes, Proteobacteria, and Bacteroidetes (Figures 1(a)
(Figure 1(c)), according to the scores of axis 1 of PCA, whereas and 3, and Table 2). Fifty-one percent of the sequences from
a statistically significant difference was detected among the E fractions were affiliated to plant chloroplasts, indicating
bacterial communities associated with the endosphere and a contamination in DNA extracts with the plastid DNA,
soil fractions in Egypt, according to the scores of axis 2 of as observed in other studies characterizing the endophytic
PCA (Figure 1(d)). Generally, the rhizosphere is defined as population of grapevine [8, 9] and other plants [21, 44].
a transition zone between the root surface and soil, where Despite plant material contamination, the root tissues from
the released exudates favour microbial proliferation, inducing the grapevines grown in Egypt presented a higher bacterial
changes in the structure and chemical-physical features of diversity with the Gammaproteobacteria, Sphingobacteria,
the soil [40]. The rhizosphere effect is especially pronounced and Flavobacteria being specific classes of this site.
Grapevine cultivated in Italy Grapevine cultivated in Tunisia Grapevine cultivated in Egypt
R1
R2
R3
E1
E2
E3
R1
R2
R3
R1
R2
R3
E1
E2
E3
E1
E2
E3
S1
S2
S3
S1
S2
S3
S1
S2
S3
B
B
B
10
34 52 9
3
33 8
46 7
8
19
20
31 11 24
18 32 12
6 23 22 33
7
45 5
13 51 2
6 20 51 41
16 48 49 1 32 40
18 42
5 24 35 47 19 27
1
24 17 28 19 23 36 25 2
12 29 25 13 26
26 33 7 28 43 43
11 15 27 31 34 4 18 22 35 3
10 14 17 26 30
16
32 1 2 8 21
39 42 40 27 31
4
20 3 4 37 6
22 25 23 11 44
9 10 9 30 41 29
21 35 37 16 17 21 38 39
12 15 14 13
30 29 28
38
15
5 14
Chloroplast Flavobacteria Sphingobacteria Gammaproteobacteria Betaproteobacteria

Alphaproteobacteria Bacilli Actinobacteria Acidobacteria
(a)
30 20 30
20 10 R1 20
E1 S2
R3
Axis 2 (12.05%)
Axis 2 (22.15%)
S3
Axis 2 (13.82%)
B E3 R2 R3 S1 R2
10 0 10
E2 B R1
S2
S3 R3 R2 S3
0 −10 B S1 0
S1 E2
E3 S2 R1 E3
−10 E1 −20 −10
E2 E1
−20 −30 −20
−30 −20 −10 0 10 20 30 40 −30 −20 −10 0 10 20 30 −40 −30 −20 −10 0 10 20 30 40
Axis 1 (71.02%) Axis 1 (64.32%) Axis 1 (55.54%)
(b)
40 b 40 40 b
b
b ab
Score axis 1 of PCA
Score axis 1 of PCA
Score axis 1 of PCA
20 20 20
ab
0 0 0
a a
a
−20 −20 −20
−40 −40 −40

E R S B E R S B E R S B
(c)
20 20 20
a a b b
Score axis 2 of PCA
Score axis 2 of PCA
Score axis 2 of PCA
5 a a 5 5
a a
−10 −10 −10 a
−25 −25 −25

(d)
Figure 1: DGGE analysis of grapevine root-associated microbial communities. (a) Representative DGGE gels of the separation of 16S rRNA
gene fragments along a denaturing gradient. The analyzed fractions were root tissues (E), rhizosphere (R), root-surrounding soil (S), and bulk
soil (B) of three replicate plants of grapevine from a vineyard in Italy (left panel), Tunisia (central panel), and Egypt (right panel). (b) Principal
component analysis (PCA) of the plot line profiles that were obtained from DGGE fingerprinting of the bacterial community. E replicates are
represented by a circle, R samples by a triangle, S samples by a rhombus, and B by a square. (c)-(d) ANOVA analysis was performed on the
average values of the line plot score along axis 1 and 2, respectively, of PCA analysis in order to assess the degree of similarities among plant
and soil-associated bacterial communities in the three study sites. Different letters (a and b), shown at the top of the scatter plots in the graph,
indicate a statistical significance at 𝑃 ≤ 0.05 according to ANOVA analysis.
Table 1: Relationships between endophytic/rhizosphere bacterial assemblages and climate features of different regions using nonparametric
multivariate multiple regression analysis (DISTLM). Partial (conditional) test where the amount explained by each variable added to the
model is conditional on variables already in the model.
Variable IT TN ET AIC 𝐹 𝑃 % Var % Cumul Res. df

Annual rainfall (mm) 809.1 561.1 39.7 110.09 27.749 0.0048 28.388 28.388 7
Summer rainfall (mm) 101.4 7.4 0.1 109.66 18.639 >0.05 16.973 45.362 6
Average rainfall (mm) 9 1 0 109.66 0 >0.05 <0.0001 45.362 6
Max temperature (∘ C) 31 30 35 109.66 0 >0.05 <0.0001 45.362 6
Min temperature (∘ C) 19 21 23 109.66 0 >0.05 <0.0001 45.362 6
Average temperature (∘ C) 25 25.5 29 109.66 0 >0.05 <0.0001 45.362 6
Latitude 44.6 36.8 30.2 109.66 0 >0.05 <0.0001 45.362 6
Longitude 9.1 10.1 31.2 109.66 0 >0.05 <0.0001 45.362 6
Irradiation (w/m2 ) 720 870 880 109.66 0 >0.05 <0.0001 45.362 6
AIC: coefficient of regression; 𝐹: value of pseudo 𝐹; 𝑃: significance of 𝐹; % Var: percentage of variance explained by each single variable; % Cumul: cumulative
percentage of variance explained; Res. df: residual degrees of freedom.
30 20
Score axis 1 (47.42%)

10
ET-E1
0
20
−10
ET-E3
Axis 2 (23.20%)
−20
ET-E2
10 P = 0.0032
TN-E1 −30
TN-E2 Root Rhizosphere
IT-E1 TN-R1
TN-E3 (b)
0 20
ET-R3
IT-E3 IT-E2 ET-R1 P = 0.0006
Score axis 2 (23.20%)
TN-R3 15
TN-R2 ET-R2 10
−10 5
IT-R1
IT-R3
0
IT-R2
−5
−20 −10
−40 −30 −20 −10 0 10 20 30
−15
Axis 1 (47.42%) Root Rhizosphere
(c)
(a)
Figure 2: Comparison of the structure of endosphere and rhizosphere bacterial communities associated with grapevines along an aridity
transect in the Mediterranean basin. (a) Principal component analysis (PCA) of the PCR-DGGE profiles of endophyte (E) and rhizosphere
(R) fractions associated with grapevine cultivated in Italy (IT), Tunisia (TN), and Egypt (ET). Samples were run and analyzed in triplicate.
(b)-(c) Statistical analysis was applied to the average values of endophyte and rhizosphere samples along axis 1 and axis 2, respectively, of the
PCA analysis. Statistical significance (𝑃) was evaluated according to the Student’s t-test and is indicated in every graph.
The remaining bands from the endophytic fraction were on the cultivar and age of the host plant [48], on geochemical
attributed to Actinobacteria, Alphaproteobacteria, and and physiological conditions of soil [49] and on climatic
Betaproteobacteria (Figures 1(a) and 3), which were also variables, such as soil moisture and chemical features [50].
dominant in the rhizosphere fractions, confirming that Alphaproteobacteria were the best represented class in
endophytes could represent a subgroup of rhizobacteria all rhizospheric soils collected, while Betaproteobacteria,
which have the ability to enter and establish in the root which were not detected in Egypt, were equally represented
interior [4, 5, 39]. Although other portals for endophyte in the other soil fractions in Italy and Tunisia (Figures
entrance into plant tissues cannot be excluded, considering 1(a) and 3). Despite their diffusion in both reproductive
both stomata on the phylloplane of grape flowers and and vegetative plant organs [51], no sequences affiliated
xylem sap [4, 7] or endophyte transmission through seeds to Burkholderia spp. were retrieved in our study. Among
[4, 45], root cracks are considered the “hot spot” for bacteria Betaproteobacteria, sequences affiliated to Massilia spp. were
endophytic colonization [37, 46, 47]. Several studies have detected (Table 2). These bacteria were already detected in
suggested that the diversity of endophytic bacteria depends grape endora during flowering [4] and display a copiotrophic
Grapevine cultivated in Italy Grapevine cultivated in Tunisia Grapevine cultivated in Egypt

25 25 25
Number of bands
Number of bands
20 20 20
Number of bands
15 15 15
10 10 10
5 5 5
0 0 0
Chloroplast Bacilli Gammaproteobacteria

Alphaproteobacteria Sphingobacteria Acidobacteria
Flavobacteria Actinobacteria Betaproteobacteria
(a) (b) (c)
Figure 3: Phylogenetic identification of DNA fragments that were excised from the DGGE gel and successfully amplified and sequenced.
lifestyle at the root niche [52]. Some members affiliated to in grapevine tissues using a diverse array of methods includ-
this genus have been shown to efficiently colonize cucumber ing 16S rRNA gene libraries, length heterogeneity PCR,
seed coats and radicles and are found to be associated with and FISH hybridization [4, 9, 57]. Among other bacteria,
the hyphae of mycorrhizae infecting germinating seeds FISH analysis detected cells affiliated to Firmicutes and
[52]. Sequences related to Chryseobacterium spp. were to Gammaproteobacteria adhering on epidermal cells, associ-
be found associated with grape plants in Tunisia and Egypt ated with xylem elements and colonizing different organs of
(Table 2). A C. balustinum strain was characterized by its flowers, including ovaries [4]. ARDRA profiles demonstrated
stimulation of the release of flavonoids in Phaseolus vulgaris, that flowers, fruits, and seeds host a rather low diversity
specifically produced during plant-bacteria interaction, of endophytes compared to rhizosphere and root tissues,
that can presumably be metabolized by the bacterium suggesting that specific metabolic skills are required to
[53]. Gammaproteobacteria sequences were retrieved in translocate from the root interior to other plant tissues [4].
rhizosphere samples from the vineyard in Italy, while its Among the genetic determinants affecting the endophyte
relative abundance was lower in grapevine rhizosphere lifestyle, a gene codifying for 𝛽-galactosidase and two genes
samples from Tunisia (Figure 3). Interestingly, Rhizobiales for the expression of acyl homoserine-lactones (AHL) were
spp. was observed only in R fractions, in agreement with recently found in the genome of Methylobacterium sp. strain
previous findings regarding the isolation of Rhizobium spp. GFX4 [58]. This strain, associated with the xylem of Riesling
from grape rhizosphere during flowering [4]. Their role in grapevines, could communicate with other xylem-associated
supporting plant growth or nutrient assimilation remains endophytes, potentially influencing their behaviours; by par-
to be elucidated, considering that under low soil moisture tially digesting galactan, this strain could also improve its
plants suffer from a disturbed bioavailability of nutrients colonization in the xylem through modification of the plant
[17]. Indeed, as a consequence of its ability to release EPS, a cell wall [58]. Endophyte communication through the release
Rhizobium sp. strain contributed to improve root-adhering of AHL has been documented for strains isolated from both
soil (RAS) aggregation in sunflower plantlets grown both grapevine and sugarcane and could play a, still unknown, role
in drained and dry soils [54]. Acidobacteria were revealed for bacteria establishment in plant tissues [59].
only in soil fractions loosely or nonassociated with roots (S
and B) in grapevines cultivated in Italy. No bacteria affiliated
to the Firmicutes phylum were observed in the root system 3.3. Diversity of Culturable Bacteria Associated with Grapevine
(Figure 3). Members of the Flavobacteriaceae class, that were Root Systems. The isolation of native bacterial species asso-
retrieved in Egypt (endosphere) and Tunisia (rhizosphere), ciated with grapevine cultivated in North Italy, Tunisia,
were classified among the most abundant bacteria in root and Egypt was performed using different media in order
and phyllosphere of Arabidopsis thaliana plants grown in to select for oligotrophic bacteria, Pseudomonadaceae, and
the wild [55]. A larger genome size has been advocated as ACC deaminating bacteria, already well documented as plant
a possible reason for the success of Flavobacterium spp. in growth promoters [60–62]. The highest number of cultivable
plant organ colonization, supporting a higher metabolic bacteria expressed as colony-forming units (CFUs) per gram
flexibility for the use of complex sugar compounds secreted of sample was in the rhizosphere (108 -109 ) and progressively
by plants [56]. decreased passing from the S (107 –109 ) to the B fractions (105 –
Our findings on the grapevine-associated bacterial com- 107 ). The root tissues presented the lowest values (105 –107 ),
munity structure are in agreement with other studies car- supporting previous data obtained on grapevine-associated
ried out on grape roots. Few bacterial groups were found bacterial communities [4, 7, 63]. It is noteworthy that bulk
Table 2: Identification of the dominant bands in the PCR-DGGE fingerprinting profiles (marked in Figure 1). The codes of the different
fractions of the grapevine root systems are as follows: E, Endosphere; R, rhizosphere; S, root-surrounding soil; B, bulk soil.
DGGE Closest relative Closest describe relative
Frac. Acc. N∘ % Acc. N∘ % Class
band (NCBI database) (Ez Taxon database)
B IT-01 Ramlibacter sp. AY429716 88 Curvibacter fontanus AB120963 87 Betaproteobacteria
B IT-02 Unc. Comamonadaceae AY360707 86 Curvibacter fontanus AB120963 85 Betaproteobacteria
B IT-03 Unc. Variovorax sp. JN590646 90 Xylophilus ampelinus AF078758 89 Betaproteobacteria
B IT-04 Unc. Acidobacteria EF651005 95 Desulfomonile limimaris AF230531 81 Acidobacteria
B IT-05 Unc. bacterium HQ393158 98 Massilia plicata AY966000 96 Betaproteobacteria
B IT-06 Massilia sp. JF279920 99 Massilia niabensis EU808006 99 Betaproteobacteria
B IT-07 Unc. bacterium AY274152 97 Solibacter usitatus CP000473 92 Acidobacteria
B IT-08 Unc. Acidobacteria GU257870 94 Chloracidobacterium thermophilum CP002514 80 Acidobacteria
E IT-09 Micromonospora sp. FR692086 92 Micromonospora peucetia X92603 90 Actinobacteria
E IT-10 Chloroplast U189132 99 Chloroplast AM711640 99 Eukarya
E IT-11 Chloroplast HQ325745 100 Chloroplast DQ386163 99 Eukarya
E IT-12 Chloroplast HQ325745 100 Chloroplast DQ386163 100 Eukarya
R IT-13 Unc. Rhodocyclaceae EF643420 97 Thiobacter subterraneus AB180657 91 Betaproteobacteria
R IT-14 Unc. bacterium FR853277 91 Steroidobacter denitrificans EF605262 89 Gammaproteobacteria
R IT-15 Unc. bacterium JN855310 98 Bradyrhizobium pachyrhizi AY624135 97 Alphaproteobacteria
R IT-16 Unc. Pseudomonadales FJ889292 97 Steroidobacter denitrificans EF605262 96 Gammaproteobacteria
R IT-17 Unc. bacterium HM445266 92 Rubrivivax gelatinosus D16213 90 Betaproteobacteria
R IT-18 Unc. bacterium EU881322 99 Thiobacter subterraneus AB180657 90 Betaproteobacteria
R IT-19 Unc. Rhizobium sp. EF074979 92 Rhizobium giardinii U86344 91 Alphaproteobacteria
R IT-20 Sphingobacterium sp. EU580525 97 Dyadobacter hamtensis AJ619978 93 Sphingobacteria
R IT-21 Unc. bacterium EF019453 86 Methylogaea oryzae EU672873 79 Gammaproteobacteria
R IT-22 Unc. bacterium EF392989 96 Steroidobacter denitrificans EF605262 87 Gammaproteobacteria
R IT-23 Unc. bacterium FR853277 95 Steroidobacter denitrificans EF605262 93 Gammaproteobacteria
R IT-24 Unc. bacterium GU568879 87 Novosphingobium resinovorum EF029110 82 Alphaproteobacteria
R IT-25 Unc. bacterium FJ479326 93 Steroidobacter denitrificans EF605262 91 Gammaproteobacteria
R IT-26 Unc. bacterium JN855310 99 Bradyrhizobium pachyrhizi AY624135 98 Alphaproteobacteria
R IT-27 Unc. bacterium JN855310 95 Nitrobacter hamburgensis CP000319 94 Alphaproteobacteria
R IT-28 Unc. bacterium DQ643675 93 Rubrivivax gelatinosus D16213 91 Betaproteobacteria
R IT-29 Unc. bacterium GU291531 94 Novosphingobium resinovorum EF029110 94 Alphaproteobacteria
R IT-30 Unc. Alphaproteobacteria JN371328 81 Blastochloris sulfoviridis D86514 78 Alphaproteobacteria
R IT-31 Unc. Steroidobacter FN297970 100 Steroidobacter denitrificans EF605262 98 Gammaproteobacteria
S IT-32 Unc. Acidobacteria HQ597613 97 Chloracidobacterium thermophilum CP002514 81 Acidobacteria
S IT-33 Unc. bacterium JN855310 91 Nitrobacter winogradskyi CP000115 90 Alphaproteobacteria
S IT-34 Unc. bacterium FJ479326 99 Steroidobacter denitrificans EF605262 97 Gammaproteobacteria
S IT-35 Unc. Alphaproteobacteria FJ568851 95 Donghia mobilis FJ455532 91 Alphaproteobacteria
B TN-01 Unc. bacterium KC541101 100 Massilia aurea AM231588 99 Betaproteobacteria
B TN-02 Unc. bacterium HM186197 99 Ohtaekwangia koreensis GU117702 93 Sphingobacteria
B TN-03 Chryseobacterium indoltheticum AY468448 99 Chryseobacterium indoltheticum AY468448 98 Flavobacteria
B TN-04 Unc. bacterium HF546519 83 Xenophilus azovorans AF285414 77 Betaproteobacteria
E TN-05 Unc. bacterium FN667504 97 Streptomyces sodiiphilus AY236339 96 Actinobacteria
E TN-06 Chloroplast HQ336404 99 Chloroplast DQ386163 99 Cyanobacteria
E TN-09 Chloroplast EU189132 99 Chloroplast AM711640 99 Cyanobacteria
E TN-13 Chloroplast EU118126 100 Chloroplast EU118126 99 Cyanobacteria
E TN-16 Chloroplast EU118126 99 Chloroplast EU118126 98 Eukarya
E TN-17 Rhizobium radiobacter AJ389904 100 Rhizobium radiobacter AJ389904 100 Alphaproteobacteria
E TN-18 Hydrogenophilus thermoluteolus AB680730 98 Hydrogenophilus hirschii FR749905 98 Betaproteobacteria
E TN-19 Chloroplast HQ325745 99 Chloroplast L37580 96 Eukarya
E TN-20 Chloroplast HQ325745 99 Chloroplast DQ386163 99 Eukarya
R TN-21 Unc. bacterium JF198689 89 Steroidobacter denitrificans EF605262 84 Gammaproteobacteria
R TN-22 Rhizobium sullae NR 029330 89 Rhizobium sullae Y10170 82 Alphaproteobacteria
R TN-23 Rhizobium giardinii JX869993 91 Rhizobium selenitireducens EF440185 74 Alphaproteobacteria
R TN-24 Rhizobium huautlense HQ538618 99 Rhizobium huautlense AF025852 99 Alphaproteobacteria
R TN-25 Unc. bacterium HM328693 99 Novosphingobium resinovorum EF029110 95 Alphaproteobacteria
Table 2: Continued.
DGGE Closest relative ∘ Closest describe relative
Frac. band (NCBI database) Acc. N % (Ez Taxon database) Acc. N∘ % Class
R TN-26 Rhizobium etli NR 029184 100 Rhizobium vallis FJ839677 99 Alphaproteobacteria
R TN-27 Unc. Betaproteobacteria EU266802 92 Hydrogenophaga palleronii AF019073 78 Betaproteobacteria
R TN-28 Rhizobium giardinii EU410948 99 Rhizobium endophyticum EU867317 98 Alphaproteobacteria
R TN-29 Unc. bacterium HF546519 85 Piscinibacter aquaticus DQ664244 81 Betaproteobacteria
R TN-30 Unc. bacterium HE586741 85 Hydrogenophaga palleronii AF019073 78 Betaproteobacteria
R TN-31 Unc. bacterium JF175892 93 Niastella populi EU877262 81 Sphingobacteria
R TN-32 Unc. bacterium JF175892 93 Chitinophaga arvensicola D12657 84 Sphingobacteria
R TN-33 Unc. Flavobacterium sp. EU017400 99 Flavobacterium reichenbachii AM177616 84 Flavobacteria
R TN-34 Flavobacterium sp. EF601822 99 Flavobacterium pectinovorum AM230490 98 Flavobacteria
R TN-35 Unc. Betaproteobacteria EF662768 97 Azoarcus buckelii AJ315676 88 Betaproteobacteria
R TN-36 Sphingobacterium sp. EU580525 96 Dyadobacter hamtensis AJ619978 94 Sphingobacteria
R TN-37 Reichenowi ornate AY316684 82 Aurantimonas altamirensis DQ372921 80 Alphaproteobacteria
R TN-38 Unc. Gammaproteobacteria JN648252 97 Steroidobacter denitrificans EF605262 88 Gammaproteobacteria
R TN-39 Rhizobium sp. FN546874 98 Rhizobium endophyticum EU867317 97 Alphaproteobacteria
R TN-40 Rhizobium sp. AM922181 99 Rhizobium endophyticum EU867317 98 Alphaproteobacteria
R TN-41 Unc. bacterium FJ719038 97 Dyadobacter hamtensis AJ619978 95 Sphingobacteria
R TN-42 Rhizobium leguminosarum HQ853453 99 Rhizobium vallis FJ839677 98 Alphaproteobacteria
S TN-43 Unc. bacterium AF423222 99 Thiobacter subterraneus AB180657 89 Betaproteobacteria
S TN-44 Unc. Comamonadaceae AY360707 99 Acidovorax konjaci AF078760 97 Betaproteobacteria
S TN-45 Unc. Sphingobacteriales AM934931 99 Ohtaekwangia koreensis GU117702 93 Sphingobacteria
S TN-46 Unc. bacterium GQ023702 91 Hydrogenophilus hirschii FR749905 91 Betaproteobacteria
S TN-47 Unc. Burkholderiaceae AM935484 99 Thiobacter subterraneus AB180657 90 Betaproteobacteria
S TN-48 Pedobacter sp. AY599662 85 Pedobacter metabolipauper AM491370 84 Sphingobacteria
S TN-49 Pedobacter africanus NR 028977 99 Pedobacter africanus AJ438171 98 Sphingobacteria
S TN-50 Unc. bacterium HM049699 88 Pedobacter africanus AJ438171 87 Sphingobacteria
S TN-51 Unc. bacterium HM049699 99 Pedobacter steynii AM491372 98 Sphingobacteria
S TN-52 Flavobacterium sp. HM149210 99 Flavobacterium pectinovorum AM230490 98 Flavobacteria
B ET-01 Unc. Betaproteobacteria JF806989 99 Thiobacter subterraneus AB180657 90 Betaproteobacteria
B ET-02 Bacillus sp. FM992819 99 Bacillus pocheonensis AB245377 98 Bacilli
B ET-03 Bacillus sp. FM992819 87 Bacillus alcalophilus X60603 85 Bacilli
B ET-04 Unc. bacterium DQ398884 86 Hydrogenophaga flava AF078771 84 Betaproteobacteria
B ET-05 Unc. Sphingobacteriales AM934931 99 Ohtaekwangia koreensis GU117702 92 Sphingobacteria
B ET-06 Unc. bacteria JF681924 90 Sterolibacterium denitrificans AJ306683 84 Gammaproteobacteria
E ET-07 Unc. bacterium JQ357881 99 Chitinophaga pinensis CP001699 98 Sphingobacteria
E ET-08 Unc. bacterium JQ358300 99 Chitinophaga sancti AB078066 98 Sphingobacteria
E ET-09 Chryseobacterium wanjuense AB682410 100 Chryseobacterium wanjuense DQ256729 100 Flavobacteria
E ET-10 Flavobacterium sp. JF915324 99 Flavobacterium subsaxonicum AM934666 96 Flavobacteria
E ET-11 Chitinophaga sp. JQ659659 98 Chitinophaga ginsengisoli AB245374 96 Sphingobacteria
E ET-12 Chitinophaga sp. JQ659659 98 Chitinophaga ginsengisoli AB245374 96 Sphingobacteria
E ET-13 Novosphingobium sp. EU984513 99 Novosphingobium resinovorum EF029110 99 Alphaproteobacteria
E ET-14 Unc. bacterium FN667504 97 Streptomyces sodiiphilus AY236339 96 Actinobacteria
E ET-15 Glycomyces scopariae JQ342894 99 Glycomyces scopariae EU200682 99 Actinobacteria
E ET-16 Variovorax paradoxus AB680784 99 Variovorax paradoxus D88006 99 Betaproteobacteria
E ET-17 Pseudoxanthomonas sp. JF703645 99 Pseudoxanthomonas AB008507 98 Gammaproteobacteria
E ET-18 Rhizobium sp. AM922181 99 Rhizobium radiobacter AJ389904 99 Alphaproteobacteria
E ET-19 Chloroplast HQ325745 99 Chloroplast DQ386163 99 Eukarya
E ET-20 Rhizobium sp. AM922181 85 Rhizobium radiobacter AJ389904 83 Alphaproteobacteria
E ET-21 Chloroplast HQ325745 100 Chloroplast DQ386163 99 Eukarya
R ET-22 Chitinophaga terrae AB267724 98 Chitinophaga niabensis EU714259 95 Sphingobacteria
R ET-23 Chitinophaga sp. GQ281772 99 Chitinophaga niabensis EU714259 97 Sphingobacteria
R ET-24 Chitinophaga sp. JQ659659 97 Chitinophaga sancti AB078066 96 Sphingobacteria
R ET-25 Novosphingobium sp. AB453877 90 Novosphingobium resinovorum EF029110 88 Alphaproteobacteria
R ET-26 Unc. bacterium GQ074926 99 Novosphingobium resinovorum EF029110 97 Alphaproteobacteria
R ET-27 Unc. bacterium GQ169020 90 Blastochloris sulfoviridis D86514 88 Alphaproteobacteria
R ET-28 Unc. bacterium DQ814032 99 Mesorhizobium thiogangeticum AJ864462 99 Alphaproteobacteria
R ET-29 Pseudoxanthomonas mexicana JQ660737 100 Pseudoxanthomonas japonensis AB008507 99 Gammaproteobacteria
R ET-30 Mesorhizobium sp. JN688938 94 Sinorhizobium americanum AF506513 93 Alphaproteobacteria
R ET-31 Unc. bacterium AB540382 82 Steroidobacter denitrificans EF605262 79 Gammaproteobacteria
R ET-32 Sphingopyxis chilensis JF459975 98 Sphingopyxis panaciterrae AB245353 98 Alphaproteobacteria
R ET-33 Chitinophaga sp. JN680879 89 Chitinophaga niabensis EU714259 79 Sphingobacteria
R ET-34 Unc. bacterium GU291531 100 Novosphingobium resinovorum EF029110 99 Alphaproteobacteria
Table 2: Continued.
DGGE Closest relative ∘ Closest describe relative
Frac. band (NCBI database) Acc. N % (Ez Taxon database) Acc. N∘ % Class
R ET-35 Devosia chinhatensis EF433462 99 Devosia chinhatensis EF433462 99 Alphaproteobacteria
R ET-36 Ensifer sp. JF450188 99 Ensifer adhaerens AM181733 98 Alphaproteobacteria
R ET-37 Sinorhizobium fredii HQ836172 98 Sinorhizobium americanum AF506513 97 Alphaproteobacteria
R ET-38 Mesorhizobium loti HQ424911 93 Mesorhizobium plurifarium Y14158 92 Alphaproteobacteria
R ET-39 Mesorhizobium loti HQ424911 88 Mesorhizobium plurifarium Y14158 87 Alphaproteobacteria
S ET-40 Cupriavidus sp. AB266608 99 Cupriavidus oxalaticus AF155567 98 Betaproteobacteria
S ET-41 Unc. Burkholderiaceae JF681924 95 Thiobacter subterraneus AB180657 79 Betaproteobacteria
S ET-42 Unc. Bacteroidetes FM877553 87 Ohtaekwangia koreensis GU117702 85 Sphingobacteria
S ET-43 Unc. Bacteroidetes FM877553 99 Ohtaekwangia kribbensis GU117703 92 Sphingobacteria
Frac.: Fraction analysed; Unc.: Uncultured.
soil and endosphere host similar bacterial communities, in major bacterial genera (Figure 4(b)). According to the cluster
terms of size, suggesting that the ability to thrive in plant analysis at the genus level performed on the entire strain
tissues requires specific genetic requirements [39]. A total collection, the composition of the cultivable communities
of 769 isolates were obtained from the selected agar media associated with grapes cultivated in Egypt and Tunisia shared
as representative of the different populations/morphologies. a higher similarity (82%) than those in Italy (68%). A similar
To reduce genotypic redundancy, the ACCd bacterial collec- profile of bacteria distribution was observed in the root
tion was dereplicated using ITS-PCR fingerprinting, and a systems of all grapevines studied, with the dominance of
representative bacterium from each haplotype was selected Gammaproteobacteria followed by Firmicutes. Even at the
for further identification and characterization. In total, 331 genus level, differences are highlighted in the bacteria distri-
strains were identified by partial sequencing of the 16S bution in plant and soil fractions and within the studied sites,
rRNA gene. The phylogenetic identification of culturable particularly among Tunisia and Egypt. The rhizosphere of
bacteria highlighted the diversity in terms of composition of grapes cultivated in these countries differed for the different
the different fractions, revealing a predominance of Gram- percentages of bacteria affiliated to Pseudomonas (40% in
negative bacteria (66%), belonging to the Gammaproteobac- Tunisia and 8% in Egypt) and Enterobacter (10% in Tunisia
teria (63%), Alphaproteobacteria (2%), and Betaproteobacte- and 65% in Egypt). Endophytes from the Italian plants were
ria (1%) subclasses. The remaining Gram-positive isolates dominated by Pantoea (40%), a genus that was not detected
were affiliated to the Firmicutes (31%), Actinobacteria (2%), in grapes grown at lower latitudes, followed by Pseudomonas
and Bacteroidetes (1%) classes (Figure 4(a)). Members of (34%), Bacillus (14%), Enterobacter (8%), Arthrobacter (3%),
these taxa have been found to be associated with other grape and Rhodococcus (1%). The genus Pantoea has been fre-
plants cultivated in Italy [9, 57], France [4, 63], Turkey [64], quently associated with grape tissues and may contribute
Nova Scotia [7], and Australia [8]. A high Shannon-Weaver to prime plants for accelerated phytoalexin production after
diversity index, calculated from the number of individuals B. cinerea challenge [65]; its plant growth potential has
per genus, was found within the rhizosphere of grapes already been documented for several model plants [66,
cultivated in Italy (𝐻󸀠 = 1.52) and Tunisia (𝐻󸀠 = 1.36) and 67]. Egypt and Tunisia grape root tissues hosted a higher
in the endosphere of ungrafted Barbera in Italy (𝐻󸀠 = 1.37). percentage of isolates affiliated to Pseudomonas and Bacillus
On the contrary, the bacterial communities associated with genera, confirming previous findings on bacteria community
the root system of grapes cultivated in Egypt presented lower composition in grape tissues as assessed through isolation
diversity values (𝐻󸀠 = 1.037 in E and 𝐻󸀠 = 1.07 in R). and culture-independent methods [4, 8]. Pseudomonas spp.
The high genetic diversity of grape root systems presumably in particular was abundant in the soil fractions from Tunisia
resulted from the combined effects of root exudates and (55%), although it was also observed in the other two sites
agricultural management practices, particularly at the lower (25% in Egypt and 14% in Italy), confirming the widespread
latitude site where the arid pedoclimatic condition may have diffusion of this genus in root-influenced soils [68]. Despite
influenced the bacterial community composition [43]. The the presence of sequences affiliated to Rhizobiales in all three
microbial community in the bulk soil, not directly influenced vineyards, as observed by DGGE analysis, isolates belonging
by the root system or agricultural practices, showed the lowest to the Rhizobium genus were retrieved only in the root
diversity indexes, particularly in the samples collected from tissues of grapes cultivated in Egypt (13%). The role of these
the Southern Mediterranean sites (𝐻󸀠 = 0.69 in Tunisia bacteria remains to be elucidated, although their association
and 𝐻󸀠 = 0.88 in Egypt). On the contrary, the bulk soil with other crops has been proposed for field applications
collected in vineyards in Italy recorded a higher Shannon [69]. The most abundant genera that were associated with
index (𝐻󸀠 = 1.158), probably because of a more structured the endosphere were also retrieved from the rhizosphere,
soil texture that is able to host a richer microbial community although at lower percentages. This observation strongly
[43]. Significant differences were observed in the structure supports the theory regarding endophytic bacteria entry from
of the bacterial communities in the analyzed vineyards, in the root system to spread in plant tissues through xylem
particular for the differential distribution pattern of the translocation, as previously documented for both beneficial
100 application of crop management techniques based on the use

of natural fertilizers such as manure and plant residues [16,
80 18]. Nevertheless, representative species of Enterobacteriaceae
were widespread in several plant systems, suggesting a role in
Isolates (%)
60 plant colonization and plant promotion also during stressful

conditions [73]. In the S fractions, less influenced by root
40 exudates, the isolation frequency of Bacillus increased. This
shift in the bacterial community composition was evident in
20 the bulk soil, not subjected to amendment and irrigation pro-
cesses. In both S and B fractions, the Enterobacteriaceae dis-
0
appeared, while the spore-forming bacteria (Firmicutes and
IT TN ET IT TN ET IT TN ET IT TN ET
Root RHIZ SSR Bulk Actinobacteria), typical of poorly structured soil, increased
in incidence reaching 98% in the bulk soil in Egypt, 91% in
Bacteroidetes Gammaproteobacteria Italy, and 31% in Tunisia (Figure 3(b)). The prevailing genera
Bacilli Betaproteobacteria Arthrobacter, Bacillus, and Paenibacillus can survive as rest-
Actinobacteria Alphaproteobacteria
ing cells or spores under adverse environmental conditions,
(a) hence, making them typical taxa of uncultivated and arid soils
100
[43].
The observed diversity of the culturable fraction is in
agreement with the findings provided by PCR-DGGE anal-
80 ysis. Indeed, the same microbial taxa were retrieved with the
exception of Acidobacteria that were detected only through
Isolates (%)
60 molecular analysis. Despite the fact that the cultivation

approach generally favours some taxa [74] and that the
40 PCR-based techniques of metagenomes are biased by the
preferential amplification of certain bacterial groups [75],
the combination of cultivation independent and dependent
20
techniques revealed sharp differences among the structure
of microbial communities associated with root systems of
0 grapes cultivated in three Mediterranean regions.
IT TN ET IT TN ET IT TN ET IT TN ET
Root RHIZ SSR Bulk
Stenotrophomonas Buttiauxella Rhizobium 3.4. Determining the PGP Potential of Grapevine-Associated

Pseudomonas Variovorax Streptomyces Bacteria. One hundred and seventy-five isolates of the de-
Acinetobacter Achromobacter Microbacterium
replicated collection, 93 from the rhizosphere and 82 endo-
Serratia Sphingobacterium Arthrobacter
Paenibacillus Rhodococcus phytes, were further screened in vitro for the presence
Pantoea
Enterobacter Lysinibacillus of PGP traits. To characterize their PGP potential, auxin
Citrobacter Bacillus (IAA) production, phosphate solubilization, ammonia and
siderophores productions, protease activity, and exopolysac-
(b)
charide (EPS) release were evaluated. The majority (95%)
Figure 4: Phylogenetic identification of culturable bacteria asso- of isolates showed multiple PGP activities, which may pro-
ciated with grapevine plant and soil fractions. (a)-(b) Bacteria mote plant growth directly, indirectly, or synergistically. In
repartition in endosphere (ROOT), rhizosphere (RHIZ), root- particular, none of the rhizobacteria and only 4% of the
surrounding soil (SSR), and bulk soil (BULK) of grapevine grown endophytes showed only one or no activity, while about 80%
in Italy (IT), Tunisia (TN), and Egypt (ET) according to the class of isolates from both fractions displayed more than three
and genus level, respectively. PGP activities. Interestingly, the distribution of the number
of PGP activities displayed by the strains in the three study
and pathogenic strains such as Agrobacterium tumefaciens sites revealed a similar distribution profile (96% in Italy, 97%
[70], Burkholderia sp. strain PsJN [5], Yersinia enterocolitica in Tunisia, and 94% in Egypt), supporting the hypothesis that
strain [71], and Xylella fastidiosa [72]. a huge functional PGP potential is maintained in grapevine
Isolates from the R fraction were mainly affiliated to root systems (Figures 5(a), 5(b), and 5(c)). Among the most
the Enterobacteriaceae family. Enterobacter that was shared common PGP abilities was the production of auxin-like
among the rhizosphere soils of the three sites was found at a compounds (82%), followed by the synthesis of ammonia
higher percentage in the soils in Egypt (65%) and Italy (39%). (70%) and the solubilization of insoluble phosphates (61%).
In the rhizosphere of grapes from Tunisia, Buttiauxiella In terms of auxins, IAA role in the stimulation and elongation
(33%) was the main genus of the Enterobacteriaceae, while of the root apparatus is well documented, extending the
in Italy Citrobacter accounted for 24% of the local collection root surface involved in nutrient and water uptake [76]. In
(Figure 3(b)). The predominance of bacteria affiliated to our bacterial collection, the IAA production was equally
the Enterobacteriaceae family could be attributed to the distributed among endophytic (84%) and rhizospheric (80%)
60 100
50
80
40
Isolates (%)
Isolates (%)
60
30
40
20
10 20
0 0
0 1 2 3 4 5 6
Prot
5% NaCl
8% NaCl
10% NaCl
42∘ C
IAA
Sid
P Sol
EPS
20% PEG
NH3
4∘ C
50∘ C
Number of PGP activity
(a)
PGP activity Abiotic stress tolerance

(d)
60 100
50 80
Isolates (%)
40
Isolates (%)
60
30
40
20
10 20
0 0
0 1 2 3 4 5 6
Prot
IAA
Sid
EPS
10% NaCl
P Sol
NH3
5% NaCl
8% NaCl
4∘ C
42∘ C
50∘ C
20% PEG
(b)

(e)
60 100
50 80
40
Isolates (%)
Isolates (%)
60
30
40
20
20
10
0 0
Prot
5% NaCl
8% NaCl
10% NaCl
42∘ C
IAA
Sid
P Sol
EPS
NH3
4∘ C
50∘ C
20% PEG
0 1 2 3 4 5 6
Endophytic bacteria
Rhizobacteria
(c)
Endophytic bacteria
Rhizobacteria
(f)
Figure 5: PGP potential of grapevine-associated bacteria. (a)–(c) Percentage of isolates showing an increasing number of PGP abilities in the
strain collection isolated from grapevine grown in vineyards located in Italy, Tunisia, and Egypt, respectively. (d)–(f) Percentage of isolates
displaying the assayed PGP traits and abiotic stress tolerance in the bacterial collection of strains associated with grapevine cultivated in Italy,
Tunisia, and Egypt, respectively.
bacteria, and a similar trend was observed in all three sites (5%, 8%, and 10%). While 67% of the isolates were able
along the latitude transect (79% in Italy, 82% in Tunisia, to grow on media containing 5% NaCl, this percentage
and 85% in Egypt) (Figures 5(d), 5(e), and 5(f)), in agree- decreased to 22% and 16% at 8% and 10% NaCl, respectively.
ment with previous observations that IAA synthesis is a As shown in Figure 4, even at lower NaCl concentrations,
widespread PGP trait [77]. The IAA production ranged from bacteria from the endophytic fraction showed sensitivity
2.11 to 36.2 𝜇gmL−1 , with the highest amount produced by to salt, particularly among bacteria isolated from Egyptian
the isolates from the rhizosphere and endosphere of grape grape root, where 55% of isolates could not grow in the
cultivated in Italy (14.4 and 18.6 𝜇gmL−1 in E and R fractions, presence of salt. On the contrary, rhizobacteria isolated from
resp.). The production of ammonia can indirectly influence Egypt included the highest proportion of isolates resistant
plant growth through the supply of nitrogen [78]. In the to salinity (79%), followed by those isolated from Tunisia
present investigation, 70% of isolates displayed ammonia (64%) and Italy (37%). At increasing concentrations of salt,
production. A similar distribution among the endophytic and the percentage of resistant isolates decreased, with only
rhizospheric bacteria was observed (Figures 5(d), 5(e), and 17% of isolates able to grow at 10% NaCl. In particular,
5(f)), with the exception of strains isolated from Italian root the capacity to tolerate high salt concentration followed
tissues (48%). Similarly, phosphate solubilisation ability was the latitudinal/aridity transect from the south to the north
exhibited by 61% of the isolates collected. Phosphorous is a (Figures 5(a), 5(b), and 5(c)), with percentages ranging
key nutrient for plant growth, representing one of the main from 31% in the rhizosphere of grapes from Egypt to 4%
in the rhizosphere from Italy. Although halotolerance has
factors limiting plant development and productivity [79]. The
been studied in bacteria affiliated to Halomonas spp. [86],
ability of rhizobacteria to solubilize phosphate (79% in Egypt,
this study highlights that even under soil dryness bacteria
78% in Tunisia, and 56% in Italy) through the production
with moderate halotolerance can be observed. Interestingly,
of organic acids or phytases can support plant growth in
almost all of the isolates were able to grow under low water
nutrient-poor soils in drought-prone ecosystems, such as
availability induced by PEG [87]. All the strains isolated in
those studied in this work [80–82]. Moreover, the isolates
Tunisia and 98% of those associated with Italian grapevine
showed protease (46%), siderophore production (47%), and
root systems (100% in E and 96% in R) were able to grow
EPS release (41%). The synthesis of protease presented a
at 20% of PEG, while slightly lower percentages of tolerance
similar pattern of distribution (about 50%) along all the
were observed for bacteria isolated from grapes cultivated
fractions of grape root system analyzed, except for the endo-
in Egypt, that is, 80% in the endosphere and 93% in the
phytes associated with grapevine from Italy (12%). Several
rhizosphere (Figures 5(d), 5(e), and 5(f)). During drought,
siderophore-producing bacteria were observed mainly in
belowground microbiomes survive by using water reserves in
the rhizosphere (78% in Italy, 65% in Tunisia, and 35% in
soils that, in turn, are rapidly depleted by earlier development
Egypt), probably because this PGP trait confers competitive of plants [88]. Thus, osmotic tolerance is a key feature for
colonization ability in iron-limiting soil and exerts a bio- microbial survival.
control role, reducing iron-dependent spore germination of Finally, we observed that 63% and 61% of the isolates
fungi. A high percentage of siderophore-releasing bacteria could grow at 4∘ C and 42∘ C, respectively. The majority (52%
was recorded only among the endophytic bacteria isolated in E and 100% in R) of the strains isolated from Italian
from grapes cultivated in Italy (64%). Finally, EPS production vineyards were able to grow at 4∘ C (Figures 5(d), 5(e), and
was qualitatively evaluated. Only 49% of endophyte and 32% 5(f)), presumably adapted to the cold temperatures in autumn
of rhizobacteria were able to produce EPS, with the highest and winter when the average air temperatures can be as
percentages observed for the isolates associated with grape low as −1∘ C [25]. On the contrary, only bacteria associated
roots from Egypt (65% in R and 52% in E). Bacteria adapted to with grapevines grown in Egypt presented resistance to high
arid environments are well known to protect themselves from temperatures, with 39% and 38% of strains isolated from root
extreme climate conditions, producing EPS-rich biofilms that and rhizosphere, respectively, capable of growing at 50∘ C,
entrap water molecules and thus retaining moisture [83]. probably being adapted to hot summer temperatures [89].
Bacterial EPS production in clay-rich soils, such as those in Global warming is predicted to affect microbial communities,
Italy, may presumably play an additional role for favoring root hampering their physiology and growth [90]. The 29% of
penetration in hard soils such as dry clay soils [84]. the collected strains were able to grow both at low and high
Further analyses were performed to evaluate bacteria temperatures, confirming that these isolates were adapted to
resistance to abiotic stresses that are often associated with the peculiar temperature fluctuations of the studied envi-
drought, such as increased salinization of soils and air ronments. The 23% of isolates presented the potential to
temperatures that rise up to 50∘ C in daytime and drop down express their PGP ability in unfavourable environmental
during the night. Thus, we analyzed the ability of bacteria to conditions influenced by drought, simultaneously showing
survive in the presence of increasing concentrations of salt, to halotolerance, resistance to a variable temperature range, and
grow despite temperature fluctuations (4, 42 and 50∘ C), and low water availability.
to thrive in conditions of low water availability (Figures 5(d),
5(e), and 5(f)). Salinization of dry soils, together with drought 4. Conclusions
and temperature variations, deeply hamper plant physiology
and development [85]. As expected, the number of strains The Mediterranean is a closed basin, encompassing subtrop-
resistant to salt decreased with increasing NaCl concentration ical, arid, and continental climates that, to date, are rapidly
changing through the increase in length and extent of dry European Social Fund (FSE), and Regione Lombardia (con-
periods [91]. Among the most cultivated and economically tract “Dote Ricerca”). The authors thank Erika Corretto for
relevant crops, Vitis vinifera is widespread at all the basin the excellent technical assistance.
latitudes, and grape quality and yields are affected by pro-
longed drought events. The findings reported in the present References
study contribute to expand the knowledge on the diversity
and PGP potential of grapevine-associated bacteria under [1] X. Y. Shi, J. L. Bi, J. G. Morse, N. C. Toscano, and D. A. Cooksey,
three different agroclimatic conditions. Culture-dependent “Effect of xylem fluid from susceptible and resistant grapevines
and independent techniques highlighted that, according to a on developmental biology of Xylella fastidiosa,” European Jour-
specific and yet undefined selection mediated by the different nal of Plant Pathology, vol. 135, pp. 127–135, 2013.
cultivars and rainfall and temperature regimes, the rhizo- [2] E. Aballay, A. Mårtensson, and P. Persson, “Screening of rhi-
sphere and endosphere microbial communities are different zosphere bacteria from grapevine for their suppressive effect on
among the three different sites. Indeed, many environmental Xiphinema index Thorne & Allen on in vitro grape plants,” Plant
and Soil, vol. 347, no. 1, pp. 313–325, 2011.
factors may explain such diversity, such as soil moisture
and temperature [79]. While summer temperatures in Italy, [3] A. Balloi, E. Rolli, R. Marasco et al., “The role of microorgan-
isms in bioremediation and phytoremediation of polluted and
Tunisia, and Egypt are quite high and of a similar magnitude,
stressed soils,” Agrochimica, vol. 54, no. 6, pp. 353–369, 2010.
in winter, grapevines in the different study sites experience
[4] S. Compant, B. Mitter, J. G. Colli-Mull, H. Gangl, and A.
different levels of low and freezing temperatures that pre- Sessitsch, “Endophytes of grapevine flowers, berries, and eeds:
sumably contribute to the modelling of the grape-associated identification of cultivable bacteria, comparison with other
microbial communities. Despite the different biopedocli- plant parts, and visualization of niches of colonization,” Micro-
matic conditions of the three studied sites (in terms of dif- bial Ecology, vol. 62, no. 1, pp. 188–197, 2011.
ferent cultivars, soil types, and climate conditions), a large set [5] S. Compant, B. Reiter, A. Sessitsch, J. Nowak, C. Clément,
of PGP abilities is still displayed by the respective collections and E. A. Barka, “Endophytic colonization of Vitis vinifera L.
of isolates, independently from the site of origin, and with by plant growth-promoting bacterium Burkholderia sp. strain
a similar profile in terms of the number of PGP traits and PsJN,” Applied and Environmental Microbiology, vol. 71, no. 4,
activities. A redundant functional capability of the isolates pp. 1685–1693, 2005.
from the root systems has been recorded in all the three sites, [6] S. Compant, H. Kaplan, A. Sessitsch, J. Nowak, E. Ait Barka,
indicating that in these three environments, the root bacterial and C. Clément, “Endophytic colonization of Vitis vinifera L. by
communities are adapted to the respective conditions. For Burkholderia phytofirmans strain PsJN: from the rhizosphere to
instance, insoluble phosphate solubilisation, auxin synthesis, inflorescence tissues,” FEMS Microbiology Ecology, vol. 63, no.
1, pp. 84–93, 2008.
and ammonia production were exhibited by multiple strains
[7] C. R. Bell, G. A. Dickie, W. L. G. Harvey, and J. W. Y. F. Chan,
in all three sites, highlighting that despite site-specific chemi-
“Endophytic bacteria in grapevine,” Canadian Journal of Micro-
cal settings, grapevines in the three investigated agroclimatic biology, vol. 41, no. 1, pp. 46–53, 1995.
regions share similar physiological requirements that are,
[8] E. R. West, E. J. Cother, C. C. Steel, and G. J. Ash, “The char-
at least partially, provided by PGP-associated bacteria. Such acterization and diversity of bacterial endophytes of grapevine,”
results suggest that the great diversity of the bacterial world Canadian Journal of Microbiology, vol. 56, no. 3, pp. 209–216,
has enough resources to provide functional redundancy in 2010.
rather different environments, such as those examined, and [9] D. Bulgari, P. Casati, L. Brusetti et al., “Endophytic bacterial
that a functional homeostasis of the root system bacterial diversity in grapevine (Vitis vinifera L.) leaves described by 16S
communities may sustain grapevine life in rather different rRNA gene sequence analysis and length heterogeneity-PCR,”
environmental settings. Journal of Microbiology, vol. 47, no. 4, pp. 393–401, 2009.
[10] E. Rolli, R. Marasco, B. Ettoumi et al., “The plant growth pro-
moting microbiome increases grapevine resistance to drought
Authors’ Contribution stress: a collaborative study between Fondazione Bussolera
Branca, Le Fracce farm and Milan universities,” BOIV, vol. 85,
Ramona Marasco and Eleonora Rolli contributed equally to pp. 189–199, 2012.
the work. [11] O. Fernandez, A. Theocharis, S. Bordiec et al., “Burkholderia
phytofirmans PsJN acclimates grapevine to cold by modulating
carbohydrate metabolism,” Molecular Plant-Microbe Interac-
tions, vol. 25, no. 4, pp. 496–504, 2012.
Acknowledgments [12] A. Theocharis, S. Bordiec, O. Fernandez et al., “Burkholderia
phytofirmans PsJN primes Vitis vinifera L. and confers a better
This work was supported by the project BIODESERT GA- tolerance to low nonfreezing temperatures,” Molecular Plant-
245746 “Biotechnology from Desert Microbial Extremophiles Microbe Interactions, vol. 25, no. 2, pp. 241–249, 2012.
for Supporting Agriculture Research Potential in Tunisia and [13] M. M. Chaves, O. Zarrouk, R. Francisco et al., “Grapevine under
Southern Europe” (European Union), Fondazione Bussolera deficit irrigation: hints from physiological and molecular data,”
Branca, and the project BIOGESTECA n∘ 15083/RCC “Fondo Annals of Botany, vol. 105, no. 5, pp. 661–676, 2010.
per la Promozione di Accordi Istituzionali” (Regione Lom- [14] B. Seguin and I. G. de Cortazar, “Climate warming: conse-
bardia, Italy) through a fellowship to R. Marasco. E. Rolli quences for viticulture and the notion of ‘terroirs’ in Europe,”
was supported by Università degli Studi di Milano, DeFENS, Acta Horticulturae, vol. 689, pp. 61–69, 2005.
[15] R. Hayat, S. Ali, U. Amara, R. Khalid, and I. Ahmed, “Soil [32] D. Daffonchio, A. Cherif, and S. Borin, “Homoduplex and
beneficial bacteria and their role in plant growth promotion: a heteroduplex polymorphisms of the amplified ribosomal 16S-
review,” Annals of Microbiology, vol. 60, no. 4, pp. 579–598, 2010. 23S internal transcribed spacers describe genetic relationships
[16] R. Marasco, E. Rolli, B. Ettoumi et al., “A drought resistance- in the “Bacillus cereus group”,” Applied and Environmental
promoting microbiome is selected by root system under desert Microbiology, vol. 66, no. 12, pp. 5460–5468, 2000.
farming,” PLoS One, vol. 7, Article ID e48479, 2012. [33] J. M. Bric, R. M. Bostock, and S. E. Silverstone, “Rapid in situ
[17] J. A. Lau and J. T. Lennon, “Rapid responses of soil microorgan- assay for indoleacetic acid production by bacteria immobilized
isms improve plant fitness in novel environments,” Proceedings on a nitrocellulose membrane,” Applied and Environmental
of the National Academy of Sciences, vol. 109, pp. 14058–14062, Microbiology, vol. 57, no. 2, pp. 535–538, 1991.
2012. [34] B. Schwyn and J. B. Neilands, “Universal chemical assay for
[18] M. Köberl, H. Müller, E. M. Ramadan, and G. Berg, “Desert the detection and determination of siderophores,” Analytical
farming benefits from microbial potential in arid soils and Biochemistry, vol. 160, no. 1, pp. 47–56, 1987.
promotes diversity and plant health,” PLoS ONE, vol. 6, no. 9, [35] S. Mehta and C. S. Nautiyal, “An efficient method for qualitative
Article ID e24452, 2011. screening of phosphate-solubilizing bacteria,” Current Microbi-
[19] A. Ambrosini, A. Beneduzi, T. Stefanski, F. G. Pinheiro, L. K. ology, vol. 43, no. 1, pp. 51–56, 2001.
Vargas, and L. M. P. Passaglia, “Screening of plant growth pro-
[36] C. Santaella, M. Schue, O. Berge, T. Heulin, and W. Achouak,
moting rhizobacteria isolated from sunflower (Helianthus
“The exopolysaccharide of Rhizobium sp. YAS34 is not neces-
annuus L.),” Plant Soil, vol. 356, pp. 245–264, 2012.
sary for biofilm formation on Arabidopsis thaliana and Brassica
[20] A. Bachar, M. I. M. Soares, and O. Gillor, “The effect of resource
napus roots but contributes to root colonization,” Environmen-
islands on abundance and diversity of bacteria in arid soils,”
Microbial Ecology, vol. 63, no. 3, pp. 694–700, 2012.
[21] L. Sun, F. Qiu, X. Zhang, X. Dai, X. Dong, and W. Song, [37] J. C. Cappuccino and N. Sherman, Microbiology: A Labora-
“Endophytic bacterial diversity in rice (Oryza sativa L.) roots tory Manual, Benjamin/Cummings Publishing Company, New
estimated by 16S rDNA sequence analysis,” Microbial Ecology, York, NY, USA, 3rd edition, 1992.
vol. 55, no. 3, pp. 415–424, 2008. [38] P. Nielsen and J. Sørensen, “Multi-target and medium-
[22] G. Muyzer, E. C. De Waal, and A. G. Uitterlinden, “Profiling independent fungal antagonism by hydrolytic enzymes in
of complex microbial populations by denaturing gradient gel Paenibacillus polymyxa and Bacillus pumilus strains from barley
electrophoresis analysis of polymerase chain reaction-amplified rhizosphere,” FEMS Microbiology Ecology, vol. 22, no. 3, pp. 183–
genes coding for 16S rRNA,” Applied and Environmental Micro- 192, 1997.
biology, vol. 59, no. 3, pp. 695–700, 1993. [39] P. R. Hardoim, L. S. van Overbeek, and J. D. V. Elsas, “Prop-
[23] W. S. Rasband, US National Institutes of Health, ImageJ, erties of bacterial endophytes and their proposed role in plant
Bethesda, Md, USA, 1997–2012. growth,” Trends in Microbiology, vol. 16, no. 10, pp. 463–471,
[24] M. J. Anderson, “A new method for non-parametric multivari- 2008.
ate analysis of variance,” Austral Ecology, vol. 26, no. 1, pp. 32– [40] T. S. Walker, H. P. Bais, E. Grotewold, and J. M. Vivanco, “Root
46, 2001. exudation and rhizosphere biology,” Plant Physiology, vol. 132,
[25] http://www.worldweatheronline.com . no. 1, pp. 44–51, 2003.
[26] M. J. Anderson, R. N. Gorley, and K. R. Clarke, PERMANOVA [41] A. Bachar, M. I. M. Soares, and O. Gillor, “The effect of resource
for PRIMER, Guide to Software and Statistical Methods, islands on abundance and diversity of bacteria in arid soils,”
PRIMER E, Plymouth, UK, 2008. Microbial Ecology, vol. 63, no. 3, pp. 694–700, 2012.
[27] M. J. Anderson, DISTLM V.5: A FORTRAN Computer ProgrAm [42] R. Angel, M. I. M. Soares, E. D. Ungar, and O. Gillor, “Biogeog-
to Calculate a Distance-Based Multivariate Analysis for a Linear raphy of soil archaea and bacteria along a steep precipitation
Model, Department of Statistics, University of Auckland, Auck- gradient,” ISME Journal, vol. 4, no. 4, pp. 553–563, 2010.
land , New Zealand, 2005.
[43] A. Bachar, A. Al-Ashhab, M. I. M. Soares et al., “Soil microbial
[28] D. M. Penrose and B. R. Glick, “Methods for isolating and char- abundance and diversity along a low precipitation gradient,”
acterizing ACC deaminase-containing plant growth-promoting Microbial Ecology, vol. 60, no. 2, pp. 453–461, 2010.
rhizobacteria,” Physiologia Plantarum, vol. 118, no. 1, pp. 10–15,
2003. [44] N. Garcias-Bonet, J. M. Arrieta, C. N. de Santana, C. M.
Duarte, and N. Marbà, “Endophytic bacterial community of
[29] E. O. King, M. K. Ward, and D. E. Raney, “Two simple media
a Mediterranean marine angiosperm (Posidonia oceanica),”
for the demonstration of pyocyanin and fluorescin,” The Journal
Front Microbiol, vol. 3, article 342, 2012.
of Laboratory and Clinical Medicine, vol. 44, no. 2, pp. 301–307,
1954. [45] P. R. Hardoim, C. C. P. Hardoim, L. S. van Overbeek, and J. D.
[30] D. Daffonchio, A. De Biase, A. Rizzi, and C. Sorlini, “Interspe- van Elsas, “Dynamics of seed-borne rice endophytes on early
cific, intraspecific and interoperonic variability in the 16S rRNA plant growth stages,” PLoS ONE, vol. 7, no. 2, Article ID e30438,
gene of methanogens revealed by length and single-strand con- 2012.
formation polymorphism analysis,” FEMS Microbiology Letters, [46] J. Sørensen and A. Sessitsch, “Plant-associated bacteria lifestyle
vol. 164, no. 2, pp. 403–410, 1998. and molecular interactions,” in Modern Soil Microbiology, J. D.
[31] D. Daffonchio, S. Borin, G. Frova, P. L. Manachini, and C. van Elsas, J. K. Jansson, and J. T. Trevors, Eds., pp. 211–236, CRC
Sorlini, “PCR fingerprinting of whole genomes: the spacers Press, 2nd edition, 2006.
between the 16s and 23S rRNA genes and of intergenic tRNA [47] R. P. Ryan, K. Germaine, A. Franks, D. J. Ryan, and D.
gene regions reveal a different intraspecific genomic variability N. Dowling, “Bacterial endophytes: recent developments and
of Bacillus cereus and Bacillus licheniformis,” International Jour- applications,” FEMS Microbiology Letters, vol. 278, no. 1, pp. 1–9,
nal of Systematic Bacteriology, vol. 48, no. 1, pp. 107–116, 1998. 2008.
[48] S. D. Siciliano and J. J. Germida, “Taxonomic diversity of Microbiology and Biotechnology, vol. 20, no. 11, pp. 1577–1584,
bacteria associated with the roots of field-grown transgenic 2010.
Brassica napus cv. Quest, compared to the non-transgenic B. [62] B. R. Glick, D. M. Penrose, and J. Li, “A model for the lowering
napus cv. Excel and B. rapa cv. Parkland,” FEMS Microbiology of plant ethylene concentrations by plant growth-promoting
Ecology, vol. 29, no. 3, pp. 263–272, 1999. bacteria,” Journal of Theoretical Biology, vol. 190, no. 1, pp. 63–68,
[49] H. Izumi, I. C. Anderson, K. Killham, and E. R. B. Moore, 1998.
“Diversity of predominant endophytic bacteria in European [63] P. Trotel-Aziz, M. Couderchet, S. Biagianti, and A. Aziz, “Char-
deciduous and coniferous trees,” Canadian Journal of Microbi- acterization of new bacterial biocontrol agents Acinetobacter,
ology, vol. 54, no. 3, pp. 173–179, 2008. Bacillus, Pantoea and Pseudomonas spp. mediating grapevine
[50] H. H. Long, D. G. Sonntag, D. D. Schmidt, and I. T. Baldwin, resistance against Botrytis cinerea,” Environmental and Exper-
“The structure of the culturable root bacterial endophyte com- imental Botany, vol. 64, no. 1, pp. 21–32, 2008.
munity of Nicotiana attenuata is organized by soil composition [64] K. Karagöz, F. Ateş, H. Karagöz, R. Kotan, and R. Çakmakçı,
and host plant ethylene production and perception,” New “Characterization of plant growth-promoting traits of bacteria
Phytologist, vol. 185, no. 2, pp. 554–567, 2010. isolated from the rhizosphere of grapevine grown in alkaline
[51] S. Compant, J. Nowak, T. Coenye, C. Clément, and E. Ait Barka, and acidic soils,” European Journal of Soil Biology, vol. 50, pp.
“Diversity and occurrence of Burkholderia spp. in the natural 144–150, 2012.
environment,” FEMS Microbiology Reviews, vol. 32, no. 4, pp. [65] B. Verhagen, P. Trotel-Aziz, P. Jeandet, F. Baillieul, and A. Aziz,
607–626, 2008. “Improved resistance against Botrytis cinerea by grapevine-
[52] M. Ofek, Y. Hadar, and D. Minz, “Ecology of root colonizing associated bacteria that induce a prime oxidative burst and
Massilia (Oxalobacteraceae),” PLoS One, vol. 7, Article ID e40117, phytoalexin production,” Phytopathology, vol. 101, no. 7, pp.
2012. 768–777, 2011.
[53] M. S. Dardanellia, F. J. Fernández de Córdobab, J. Estéveza [66] M. C. Quecine, W. L. Araujo, P. B. Rossetto et al., “Sugar-
et al., “Changes in flavonoids secreted by Phaseolus vulgaris cane growth promotion by the endophytic bacterium Pantoea
roots in the presence of salt and the plant growth-promoting agglomerans 33.1,” Applied and Environmental Microbiology, vol.
rhizobacterium Chryseobacterium balustinum,” Applied Soil 78, no. 21, pp. 7511–7518, 2012.
Ecology, vol. 57, pp. 31–38, 2012. [67] T. Taurian, M. S. Anzuay, J. G. Angelini et al., “Phosphate-
[54] Y. Alami, W. Achouak, C. Marol, and T. Heulin, “Rhizosphere solubilizing peanut associated bacteria: screening for plant
soil aggregation and plant growth promotion of sunflowers by growth-promoting activities,” Plant and Soil, vol. 329, no. 1, pp.
an exopolysaccharide-producing Rhizobium sp. strain isolated 421–431, 2010.
from sunflower roots,” Applied and Environmental Microbiol- [68] J. M. Chaparro, D. V. Badri, M. G. Bakker, A. Sugiyama, D. K.
ogy, vol. 66, no. 8, pp. 3393–3398, 2000. Manter, and J. M. Vivanco, “Root exudation of phytochemicals
[55] N. Bodenhausen, M. W. Horton, and J. Bergelson, “Bacterial in Arabidopsis follows specific patterns that are developmen-
communities associated with the leaves and the roots of Ara- tally programmed and correlate with soil microbial functions,”
bidopsis thaliana,” PLoS One, vol. 8, Article ID e56329, 2012. PLoS One, vol. 2, Article ID e55731, 2013.
[56] M. Kolton, S. J. Green, Y. M. Harel, N. Sela, Y. Elad, and E. [69] R. B. Bhattacharjee, A. Singh, and S. N. Mukhopadhyay, “Use
Cytryna, “Draft genome sequence of Flavobacterium sp. strain of nitrogen-fixing bacteria as biofertiliser for non-legumes:
F52, isolated from the rhizosphere of bell pepper (Capsicum prospects and challenges,” Applied Microbiology and Biotechnol-
annuum L. cv. Maccabi),” Journal of Bacteriology, vol. 194, pp. ogy, vol. 80, no. 2, pp. 199–209, 2008.
5462–5463, 2012. [70] F. A. Tarbah and R. N. Goodman, “Rapid detection of Agrobac-
[57] D. Bulgari, P. Casati, P. Crepaldi et al., “Restructuring of endo- terium tumefaciens in grapevine propagation material and basis
phytic bacterial communities in grapevine yellows-diseased and for an efficient indexing system,” Plant Disease, vol. 70, pp. 566–
recovered Vitis vinifera L. plants,” Applied and Environmental 568, 1986.
Microbiology, vol. 77, no. 14, pp. 5018–5022, 2011. [71] E. T. Thorne, B. M. Young, G. M. Young et al., “The structure
[58] H. M. Gan, T. H. Chew, A. O. Hudson, and M. A. Savkac, of xylem vessels in grapevine (Vitaceae) and a possible passive
“Genome sequence of Methylobacterium sp. strain GXF4, a mechanism for the systemic spread of bacterial disease,” Amer-
xylem-associated bacterium isolated from Vitis vinifera L. ican Journal of Botany, vol. 93, no. 4, pp. 497–504, 2006.
grapevine,” Journal of Bacteriology, vol. 194, pp. 5157–5158, 2012. [72] D. S. Chatelet, M. A. Matthews, and T. L. Rost, “Xylem structure
[59] A. O. Hudson, N. H. Ahmad, R. Van Buren, and M. A. and connectivity in grapevine (Vitis vinifera) shoots provides a
Savka, “Sugarcane and grapevine endophytic bacteria: isolation, passive mechanism for the spread of bacteria in grape plants,”
detection of quorum sensing signals and identification by 16S Annals of Botany, vol. 98, no. 3, pp. 483–494, 2006.
v3 rDNA sequence analysis,” in Current Research, Technology [73] S. M. Nadeem, Z. A. Zahir, M. Naveed, and S. Nawaz, “Mitiga-
and Education Topics in Applied Microbiology and Microbial tion of salinity-induced negative impact on the growth and yield
bioTechnology, A. M. Vilas, Ed., pp. 801–806, 2010. of wheat by plant growth-promoting rhizobacteria in naturally
[60] H. Bertrand, R. Nalin, R. Bally, and J.-C. Cleyet-Marel, “Iso- saline conditions,” Annals of Microbiology, vol. 63, pp. 225–232,
lation and identification of the most efficient plant growth- 2013.
promoting bacteria associated with canola (Brassica napus),” [74] T. Hamaki, M. Suzuki, R. Fudou et al., “Isolation of novel
Biology and Fertility of Soils, vol. 33, no. 2, pp. 152–156, 2001. bacteria and actinomycetes using soil-extract agar medium,”
[61] M. A. Siddikee, P. S. Chauhan, R. Anandham, G.-H. Han, and Journal of Bioscience and Bioengineering, vol. 99, no. 5, pp. 485–
T. Sa, “Isolation, characterization, and use for plant growth 492, 2005.
promotion under salt stress, of ACC deaminase-producing [75] M. T. Suzuki and S. J. Giovannoni, “Bias caused by template
halotolerant bacteria derived from coastal soil,” Journal of annealing in the amplification of mixtures of 16S rRNA genes
by PCR,” Applied and Environmental Microbiology, vol. 62, no.

2, pp. 625–630, 1996.
[76] M. Kochar and S. Srivastava, “Surface colonization by Azospiril-
lum brasilense SM in the indole-3-acetic acid dependent growth
improvement of sorghum,” Journal of Basic Microbiology, vol.
52, no. 2, pp. 123–131, 2012.
[77] O. Barazani and J. Friedman, “Is IAA the major root growth
factor secreted from plant-growth-mediating bacteria?” Journal
of Chemical Ecology, vol. 25, no. 10, pp. 2397–2406, 1999.
[78] P. A. Wani, M. S. Khan, and A. Zaidi, “Synergistic effects of the
inoculation with nitrogen-fixing and phosphate-solubilizing
rhizobacteria on the performance of field-grown chickpea,”
Journal of Plant Nutrition and Soil Science, vol. 170, no. 2, pp.
283–287, 2007.
[79] D. P. Schachtman, R. J. Reid, and S. M. Ayling, “Phosphorus
uptake by plants: from soil to cell,” Plant Physiology, vol. 116, no.
2, pp. 447–453, 1998.
[80] P. Vyas, P. Rahi, and A. Gulati, “Stress tolerance and genetic
variability of phosphate-solubilizing fluorescent Pseudomonas
from the cold deserts of the trans-himalayas,” Microbial Ecology,
vol. 58, no. 2, pp. 425–434, 2009.
[81] K. W. Ndung’u-Magiroi, L. Herrmann, J. R. Okalebo, C. O.
Othieno, P. Pypers, and D. Lesueur, “Occurrence and genetic
diversity of phosphate-solubilizing bacteria in soils of differing
chemical characteristics in Kenya,” Annals of Microbiology, vol.
62, no. 3, pp. 897–904, 2011.
[82] F. Mapelli, R. Marasco, A. Balloi et al., “Mineral-microbe inter-
actions: biotechnological potential of bioweathering,” Journal of
[83] S. S. Branda, Å. Vik, L. Friedman, and R. Kolter, “Biofilms: the
matrix revisited,” Trends in Microbiology, vol. 13, no. 1, pp. 20–
26, 2005.
[84] S. Timmusk, N. Grantcharova, and E. G. H. Wagner, “Paeni-
bacillus polymyxa invades plant roots and forms biofilms,”
7292–7300, 2005.
[85] D. Bartels and R. Sunkar, “Drought and salt tolerance in plants,”
Critical Reviews in Plant Sciences, vol. 24, no. 1, pp. 23–58, 2005.
[86] B. Jha, I. Gontia, and A. Hartmann, “The roots of the halo-
phyte Salicornia brachiata are a source of new halotolerant
diazotrophic bacteria with plant growth-promoting potential,”
Plant and Soil, vol. 356, no. 1-2, pp. 265–277, 2011.
[87] V. Sandhya, A. S. Z., M. Grover, G. Reddy, and B. Venkateswarlu,
“Alleviation of drought stress effects in sunflower seedlings by
the exopolysaccharides producing Pseudomonas putida strain
GAP-p45,” Biology and Fertility of Soils, vol. 46, no. 1, pp. 17–26,
2009.
[88] C. Parmesan, “Ecological and evolutionary responses to recent
climate change,” Annual Review of Ecology, Evolution, and
Systematics, vol. 37, pp. 637–669, 2006.
[89] V. Thiel, “Extreme environments,” Encyclopedia of Geo-
biology, 2012, http://www.springerreference.com/docs/html/
chapterdbid/ 187271.html.
[90] C. S. Sheik, W. H. Beasley, M. S. Elshahed, X. Zhou, Y. Luo, and
L. R. Krumholz, “Effect of warming and drought on grassland
microbial communities,” ISME Journal, vol. 5, no. 10, pp. 1692–
1700, 2011.
[91] S. M. Vicente-Serrano, A. Zouber, T. Lasanta, and Y. Pueyo,
“Dryness is accelerating degradation of vulnerable shrublands
in semiarid Mediterranean environments,” Ecological Mono-
graphs, vol. 4, pp. 407–428, 2012.
http://dx.doi.org/10.1155/2013/526362
Research Article
Seasonal and Spatial Variability of Virioplanktonic
Abundance in Haihe River, China
Lili Ma, Rui Sun, Guannan Mao, Hui Yu, and Yingying Wang
Key Laboratory of Pollution Processes and Environmental Criteria (Ministry of Education), Tianjin Key Laboratory of
Environmental Remediation and Pollution Control, College of Environmental Science and Engineering, Nankai University,
Tianjin 300071, China
Correspondence should be addressed to Yingying Wang; wangyy@nankai.edu.cn
Copyright © 2013 Lili Ma et al. This is an open access article distributed under the Creative Commons Attribution License, which
permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
In order to understand the composition and dynamics of planktonic viruses and their relationship with environmental parameters
in natural freshwater, flow cytometry was optimized with filtration/fixation/staining/dilution and then applied to the analysis
of samples collected from 9 stations (covering urban, rural, and estuarial areas) along the Haihe River, China, over a one-year
period of study. The total viral abundance exhibited an apparent peak in the spring. Spatially, the highest viral abundance was
recorded in estuarial areas. The correlation analysis indicated that the bacteria in the Haihe River significantly influenced viral
abundance. The relationship between abiotic variables and viral abundance remained the same as with bacterial abundance,
indicating that environmental parameters could possibly influence viral abundance in virtue of their bacterial host cells. The
influence of environmental factors on viral abundance differed in the three sampling areas, suggesting different drivers of viral
abundance in different stretches of the river associated with their utilization and surroundings.
1. Introduction Easy and precise assays for the rapid counting of viruses
are crucial for studies of viral ecology. Viruses in natural
Over the last two decades, it has been realized that viruses are samples are traditionally enumerated by culturing methods
the most abundant biological entities in aquatic ecosystems and microscopic techniques such as transmission electron
(105 –108 mL−1 ) and an important component in the aquatic microscopy (TEM) [1, 19–22] and epifluorescence micro-
microbial food web [1–4]. Viruses play critical roles in scopy (EFM) [22–24]. Since the first report by Marie and
shaping aquatic communities and determining ecosystem coworkers [25], viruses in natural samples have been counted
dynamics and have been shown to affect nutrient cycling by flow cytometry (FCM) in numerous studies [15, 26–31].
[5], microbial and dimethylsulfide release [6], and genetic While Brussaard and colleagues [32] provided a detailed
material transfer [7]. Early studies of the roles of planktonic discussion of FCM procedures for virus detection, this area
viruses were mainly focused on the marine pelagic environ- remains partly unexplored as previous studies were predom-
ment, and more recently on other aquatic habitats such as inantly focused on marine ecosystems. Because of the notably
lakes and rivers [8–10]. As a consequence, the roles of fresh- different properties of seawater and freshwater and their
water virioplankton remain much less studied. Heterotrophic different microbial communities, it is worth of discussing the
bacteria and/or chlorophyll-a have often been reported to effects of the protocol in freshwater ecosystems.
correlate with viral abundance [11–14]. Moreover, environ- The aims of the current study were (1) to detect and
mental parameters have also been observed to influence viral quantify free viruses in natural freshwater using FCM com-
abundance [15–18]. The viral abundance and dynamics in bined with optimization of the filtration/fixation/staining/
aquatic environments can be influenced not only by biotic dilution parameters, (2) to analyze the seasonal and spatial
factors, but also by the combined effects of biotic and abiotic changes in viral abundance along the Haihe River, China,
factors. using the optimized FCM method, and (3) to compare the
Beijing
Tianjin
China S1 Dongli district
Urban areas
S2
S3 S4 S7
S5 S9
S6 H
aih S8
e Riv
er
Bohai Sea
Tianjin Jinnan district
(km)
0 20
Sampling sites Coastline

Haihe River Fish pond
Tributary Dairy or Feedlot
Districts boundaries
Figure 1: Location of sampling sites along the Haihe River, China.
influence of biotic and abiotic factors on viral abundance in nitrogen during transport and storage at −80∘ C after arrival
freshwater. in the laboratory. Na2 EDTA (5 mM) was added to samples
immediately prior to staining, followed by incubation with
2. Materials and Methods 5 𝜇L mL−1 SYBR Green I (100X in DMSO; Molecular Probes)
at 80∘ C for 10 min in the dark. Samples were diluted in Milli-Q
2.1. Sampling Location. Natural freshwater samples for dif- water prior to FCM analysis. To optimize the protocol, indi-
ferent tests were collected from the Haihe River, which is vidual parameters (i.e., filtration, fixation, EDTA addition,
the largest water system in Northern China. Water samples staining temperature, and dilution solution) were optimized
were collected at 9 sites along the Haihe River during the while keeping the other parameter fixed.
autumn (September, 2011), winter (December, 2011), spring For enumeration of bacterial abundance, water samples
(April, 2012), and summer (July, 2012) (Figure 1). Sites M1 and were put on ice during transport and storage at 4∘ C after
M2 were located in urban areas. Sites from M3 to M7 were arrival in the laboratory and analysed within 6 hours. Staining
located in agriculturally influenced areas. Sites M8 and M9 and FCM analysis of bacteria were performed as described
were located near the mouth of the Haihe River at distances by Berney and colleagues [33]. In short, samples (1 mL) were
of 8 and 1 km, respectively, from the Bohai Sea. Water samples stained with 10 𝜇L mL−1 SYBR Green I/Propidium Iodide
were collected in clean, 2 L sterile bottles from a depth of staining solution and incubated for 25 min in the dark at
0.5 m at each sampling sites. room temperature. Prior to flow cytometric analysis, water
samples were diluted with Milli-Q water.
2.2. Treatments and FCM Analysis. Working solutions of Flow-cytometric measurements were made using a Partec
the dyes were prepared as follows: SYBR Green I (10,000X CyFlow Space flow cytometer (Partec GmbH, Münster, Ger-
in dimethyl sulfoxide (DMSO), Invitrogen) was diluted 100 many) with 488 nm excitation from a blue solid-state laser at
times in 0.20 𝜇m-filtered DMSO. Propidium Iodide (30 mM 50 mW. All samples were collected as logarithmic signals and
in DMSO, Invitrogen) was mixed with SYBR Green I stock at were triggered on the green fluorescence. Data were acquired
a ratio of 1 : 50. on two-parameter dot plots of green fluorescence (520 ±
The method for virus detection in freshwater was first 20 nm) versus sideward scatter (SSC) for virus and green
optimized with aquatic samples from the Haihe River, and fluorescence (520 nm, FL1) versus red fluorescence (630 nm,
then the optimal protocol was applied to the investigation of FL3) for bacteria. The output data were analysed with the
viral abundance variation. The procedures used as a reference Flomax software.
protocol were filtration of 10 mL water samples through
sterile syringe filters with 0.22 𝜇m pore size (PES, Millipore, 2.3. Environmental Parameters. A YSI EC300 Water Quality
USA) followed by fixation with glutaraldehyde (0.25% final Sonde was used to measure water temperature and salinity
concentration) for 15 min at 4∘ C and freezing in liquid on location. Total suspended solids (TSS), total nitrogen
1.2 slightly decreased compared to the unfiltered sample, but

none of the virus groups were missing. Based on these results,
(standardized to reference sample)
1 a good compromise choice would be to filter the sample

through membrane filters with a pore size of 0.22 𝜇m, not
only to reduce the bacteria and the influence of their cell
Total virus counts
0.8
lysates, but also to maintain reliable virus counts.
0.6
3.1.2. Fixation. The loss of counts from fixed samples has
0.4 been well documented. Similar to other reports [31], it was
observed that the number of viruses in samples significantly
0.2 decreases with the increased amount of glutaraldehyde (𝑃 <
0.05). This could result from glutaraldehyde being a potent
0 virucidal agent. For bacteriophages, glutaraldehyde possibly
1 2 3 4 5 6 7 forms protein-DNA cross-links, inhibiting DNA synthesis
Treatments [35]. Instead of the optimized fixation concentration (0.5%)
Figure 2: Comparison of total virus counts obtained by different reported for marine viruses, we found that 0.25% glutaralde-
treatments (1: reference; 2: filters with pore sizes of 0.1 𝜇m; 3: no hyde gave a better fixation for freshwater viruses.
filtration; 4: fixation with 0.5% glutaraldehyde; 5: staining at room
temperature; 6: without EDTA; 7: diluted in TE). The total virus 3.1.3. Staining. EDTA was added to samples immediately
count of each treatment was normalized to samples tested under
before staining. Adding EDTA to samples not only positively
reference conditions. Error bars represent the standard deviation
(𝑛 = 9).
influenced the total virus count, but also enhanced the FL1
and SSC signals compared to those of samples without EDTA
(Figures 2 and 3(c)). In the present study there was a 39%
reduction (𝑃 < 0.05) in the total virus count of samples
(TN), nitrate (NO3 –N), total phosphorus (TP), total dis-
stained at room temperature compared to samples heated at
solved phosphorus (TDP), total organic carbon (TOC), and
80∘ C, which suggests that the high staining temperature is
chlorophyll a (chl a) were measured according to standard
crucial to avoid underestimating the natural virioplanktonic
methods [34]. The rainfall data were acquired from the
abundance. This is consistent with a previous report that
Tianjin water resources bulletin.
two of the tested marine heterotrophic phages showed a
significant reduction in total virus counts at 60∘ C [31].
2.4. Statistical Analysis. Seasonal and spatial distribution
differences in viral and bacterial abundance were tested with
one-way analysis of variance (ANOVA). Potential relation- 3.1.4. Dilution. To avoid coincidence, the virus samples are
ships between all microbial and environmental data sets normally diluted before loading onto a flow cytometer.
were tested by Spearman’s rank correlation analysis. Statistical However, the choice of diluting agent influences the measured
analysis was performed with the Statistical Product and virus concentration. To examine the effects of diluting agent,
Service Solutions (SPSS) software (version 13.0). we compared autoclaved TE buffer through 0.1 𝜇m pore
size membrane filters with dilution in Milli-Q water. It
was observed that dilution in Milli-Q water provides a low
3. Results and Discussion instrument background and good results for discrimination
of virus groups (Figure 3(a)). In addition, the quality of Tris
3.1. Optimization of Procedures for Enumeration of Freshwater may differ depending on the supplier, so it is likely to affect the
Viruses by Flow Cytometry quality of TE buffer and small batches of the TE buffer should
be prefiltered immediately prior to use. It appears, therefore,
3.1.1. Filtration. The total virus counts of samples passed that instead of TE buffer, dilution in Milli-Q water is more
through membrane filters with pore sizes of 0.22 𝜇m and convenient and consistent.
0.1 𝜇m were 7% and 24% lower than the counts of samples
without filtration, respectively (Figure 2). However, this does
not mean that filtration is negligible. Not every fluorescent 3.1.5. Detection Limit and Comparisons with EFM. The pre-
dot might be a virus but could instead be DNA bound cision and detection limit of FCM for measuring total virus
to colloids [11]. DNA fragments could be generated from counts was determined with a serial dilution of freshwater
bacterial cell lysate as a result of the treatment steps (e.g., samples. Freshwater samples from the same sampling site
freezing step). A filtration step prevents overestimation of in the Haihe River were serially diluted with Milli-Q water.
the virus counts by bacterial lysate. When filtering through Figure 4(a) shows the detection limit of the instrument used
a 0.1 𝜇m pore size filter, it was evident that the natural virus in this study; 𝑟 of the trend line is 0.99 (𝑛 = 9) and the
communities lost a group of viruses compared to the sample lowest viral concentration detected is 4.04 × 104 counts mL−1
filtered through a 0.22 𝜇m pore size filter and the unfiltered (Figure 4(a)). Comparisons between FCM and EFM (using
sample (Figures 3(a) and 3(b)). The total virus counts of the protocol of Patel and coworkers [36]) for the enumeration
the sample passed through the 0.22 𝜇m pore size filter were of viruses were performed with 10 samples from the Haihe
1000 1000
100 100
FL1
FL1
10 Virus group 3 10
Virus group 2 Virus group 2
1 Virus group 1 1 Virus group 1
Background Background
0.1 0.1
1 10 100 1000 1 10 100 1000
SSC SSC
(a) (b)
1000 1000
100 100
FL1
FL1
10 10
Virus group 3
Virus group 2
1 Virus group 1 1
Background
Background
0.1 0.1
1 10 100 1000 1 10 100 1000
SSC SSC
(c) (d)
Figure 3: Flow cytometric analysis of viral populations by reference method (a), the same sample filtered through membrane filters with pore
sizes 0.1 𝜇m (b), without EDTA (c), blank sample (using autoclaved 0.1 𝜇m pore-size prefiltered natural freshwater) analysed by reference
method (d).
River. A good linear relationship was observed between FCM 3.2. Seasonal and Spatial Distribution of Bacterioplankton and
and EFM (𝑟 = 0.84, 𝑛 = 9, Figure 4(b)). Virioplankton. Viable and dead bacteria were discriminated
The results obtained in this study suggest that when between on the basis of their membrane integrity using a
studying the abundance of viruses in freshwater, samples combination of SYBR Green I and PI [33]. The total cell count
should be filtered with a 0.22 𝜇m pore size membrane, fixed is considered to be the sum of viable and dead cells. The
with glutaraldehyde (0.25% final concentration) for 15 min abundance of viable bacteria showed a dramatic fluctuation
at 4∘ C, frozen in liquid nitrogen, and stored at −80∘ C. For ranging from 3.64 × 106 to 3.93 × 107 counts mL−1 , with a
FCM analysis, samples should be added with EDTA (5 mM) clear seasonal pattern as evidenced by low abundance during
immediately before staining and incubated with 5 𝜇L mL−1 the winter (Figure 5(a)). The optimized treatment method
SYBR Green I (100X in DMSO) at 80∘ C for 10 min in the dark, was applied to virus enumeration of samples collected at 9
then diluted in Milli-Q water prior to analysis. sites along the Haihe River over a one-year period. Total viral
Total virus concentration (counts mL −1 × 104 )
100000 800
700
FCM viral abundance (1 counts mL−1 × 106 )

600
500 400
80000 400
300
200 350
60000 100
0
0 0.2 0.4 0.6 0.8 1 300
40000
250
20000 200
150
0
0 20 40 60 80 100 140 160 180 200 220 240 260 280 300 320 340 360
Hai River sample (%) EFM viral abundance (mL−1 × 106 )
(a) (b)
Figure 4: FCM detection limit and comparisons with EFM. Precision and detection limit of FCM for measuring total virus count (𝑛 = 9)
(a); linear regression of viral abundance determined by FCM and EFM (FCM = 1.11EFM + 1.12, 𝑟 = 0.84, 𝑛 = 9) (b).
abundance over the course of the investigation ranged from and 5(f)). The VBR fluctuated from 5.12 to 46.94 with an
7.35×107 counts mL−1 to 8.88×108 counts mL−1 (Figure 5(c)), overall mean of 21.44, which demonstrated the numerical
which is consistent with the range reported by previously predominance of viruses over bacteria, consistent with previ-
published studies on viral abundance values in freshwater ous reports [2]. The VBR values were significantly higher in
environments such as rivers, ponds, and lakes [8, 10, 26]. the spring and lower in the summer (𝑃 < 0.01) although the
Viral abundance in freshwater is reported to undergo bacterial abundance in the two seasons did not significantly
stronger seasonal changes than those in marine environ- change. This indicates that there are possible factors other
ments, especially in lakes [9]. In this study, the viral abun- than the host influencing the dynamics of viral abundance
dance peaked in spring and reached its lowest level in winter in the study area. The VBR values were significantly different
(𝑃 < 0.01). The lowest values of viral concentrations observed among urban, rural, and estuarial areas (𝑃 < 0.01), indicating
in winter may be explained by the fact that the host bacteria the inconsistency of virus-host interactions between the three
were least abundant in the winter, and thus fewer viruses areas.
would be released into the water. In contrast to previous
reports [37, 38], the highest viral abundance was recorded 3.3. Correlation Analysis. The nutrient concentrations (TN,
in the spring rather than in the summer (Figure 5(c)) in the NO3 –N, TP, and TDP), physical parameters (temperature,
present study. The reason may be due to the high intensities of salinity, and TSS), TOC, and chl a of each site are shown
solar radiation in the hot weather in northern China, which in Table 1. Rainfall displayed a significant seasonal variability
could accelerate viral degradation and viral decay [16, 17]. with 17.02 mm in the spring, 97.29 mm in the summer,
As shown in the boxplot (Figures 5(a) and 5(c)), several 60.53 mm in the autumn, and 5.81 mm in the winter. In
outliers were identified in sampling site 1, site 2, site 8, and this study, correlation analysis was used to identify physical,
site 9, which indicated that the spatial location potentially chemical, and biological variables that are associated with
influenced the microbial abundance. Thus, according to the changes in viral abundance.
different environmental types of the sampling sites, microbial
abundance was categorized into urban areas, rural areas, and 3.3.1. Biotic Influence on Viruses in the Haihe River. Previous
estuarial areas (Figures 5(b) and 5(d)). It is notable that the reports have demonstrated that high viral abundance is
average concentrations of viable and dead bacteria were both typically associated with high bacterial abundance and/or
higher in estuarial areas, followed by urban areas, and lower chlorophyll-a concentration [10, 13, 15, 39, 40]. Maranger
in rural areas. The results from one-way ANOVA showed and Bird reported that a strong correlation was found
that the viable, dead bacterial abundance was significantly between viral abundance and chlorophyll-a concentration,
different between the three areas (𝑃 < 0.01). Although no but not with bacterial abundance in lakes, whereas viral
significant spatial changes in viral abundance were observed, abundance in marine systems was strongly correlated with
the average abundance of virus increased from 3.01 × 108 bacterial abundance [12]. However, in our study, significant
counts mL−1 in urban areas to 3.49×108 counts mL−1 in rural relationships between viral abundance and chlorophyll-a
areas and continuously went up to 3.96 × 108 counts mL−1 in concentration (0.440, 𝑃 < 0.01) and between viral and
estuarial areas. bacterial abundance (0.575, 𝑃 < 0.01) were found from
The virus-to-bacteria ratio (VBR), indicating the relation- analysis of data for the 9 sites. As shown in Table 2, when
ship between viral and bacterial communities, exhibited a the data were analyzed in separate regions, the relationships
clear seasonal and spatial variation in the study (Figures 5(e) between viral abundance and chlorophyll-a concentration
4E7 4E7
Bacterial abundance (counts mL−1 )

Bacterial abundance (counts mL−1 )
1
3E7 3E7
18
2E7 ∗ 2E7
17
∗ ∗71
∗
63 ∗72
1E7 46 1E7 46
∗ ∗
44
0E0 65 0E0
Autumn Winter Spring Summer Urban areas Rural areas Estuarial areas
VB VB
DB DB
(a) (b)
5E8 1E9
Viral abundance (counts mL −1 )
Viral abundance (counts mL −1 )

4E8 8E8 21
3E8 6E8
2E8 4E8
1E8 2E8
29
0E0 0E0
(c) (d)
50 50
40 40
30 30
VBR
VBR
20 20
10 10
0 0
(e) (f)
Figure 5: Boxplots of microbial abundance and VBR from samples collected along the Haihe River. (a) Seasonal distribution of bacterial
abundance (VB, viable bacteria, 𝑛 = 36; DB, dead bacteria, 𝑛 = 36); (b) spatial distribution of bacterial abundance (VB, viable bacteria,
𝑛 = 36; DB, dead bacteria, 𝑛 = 36); (c) seasonal distribution of viral abundance (𝑛 = 36); (d) spatial distribution of viral abundance (𝑛 = 36);
(e) seasonal distribution of VBR (𝑛 = 36); (f) spatial distribution of VBR (𝑛 = 36); outliers numbers 1 and 46 refer to site S1; 29 and 65 refer
to site S2; 21 refers to site S3; 17, 44, and 71 refer to site S8; and 18, 63, and 72 refer to site S9.
became weak, but the bacterial abundance still strongly several factors (rainfall, TN, temperature, salinity) in addition
influenced the viral abundance. to the host that directly or indirectly possibly influenced
the dynamics of viral abundance. We recognized that viral
3.3.2. Effects of Environmental Parameters on Viral Abun- abundance was significantly negatively correlated with salin-
dance. Overall, no significant correlation was found between ity in rural areas (𝑟 = −0.602, 𝑃 < 0.01), supporting a
virioplanktonic abundance and abiotic variables over the previous hypothesis [38], which could be due to the changed
whole river. However, when the data were analysed separately ionic strength affecting the viral replication ability and even
by different sampling regions, it was observed that there were absorption to particles [2, 12]. However, viral abundance
Table 1: Environmental data recorded for the Haihe River from September 2011 to June 2012.
Autumn Winter Spring Summer
Mean (min–max) Median (Q1, Q3) Mean (min–max) Median (Q1, Q3) Mean (min–max) Median (Q1, Q3) Mean (min–max) Median (Q1, Q3)
Temp (∘ C) 28.2 (27.6–29.4) 28.2 (27.8, 28.5) 12.6 (11.0–15.6) 12 (11.7, 13.6) 13.4 (12.5–14.2) 13.3 (24.9, 25.3) 25.0 (24.0–25.7) 24.9 (11.7, 13.6)
TSS (mg L−1 ) 5.59 (3.12–9.68) 5.00 (4.06, 6.25) 8.19 (0.31–15.62) 7.50 (5.00, 13.43) 8.73 (0.31–21.77) 5.62 (10.62, 31.33) 31.85 (4.06–100.00) 20.66 (5.00, 13.43)
TN (mg L−1 ) 4.68 (3.71–5.41) 4.80 (4.49, 4.98) 4.50 (3.04–5.20) 4.65 (4.27, 5.10) 7.79 (4.33–10.14) 8.76 (6.23, 6.33) 6.25 (6.00–6.43) 6.23 (4.27, 5.10)
Nitrate (mg L−1 ) 2.14 (1.63–2.41) 2.16 (2.03, 2.35) 2.44 (2.06–2.85) 2.39 (2.36, 2.48) 2.08 (1.26–2.87) 2.09 (0.88, 2.02) 1.53 (0.82–2.90) 1.50 (2.36, 2.48)
Chla (𝜇g L−1 ) 107.61 (84.40–156.00) 107.00 (86.2, 116) 15.97 (4.00–47.00) 13.00 (9.30, 17.80) 62.67 (6.63–236.35) 25.72 (34.35, 132.41) 91.62 (3.76–262.31) 69.95 (9.30, 17.80)
TP (mg L−1 ) 0.76 (0.60–0.99) 0.74 (0.74, 0.77) 0.42 (0.34–0.54) 0.39 (0.36, 0.48) 1.96 (0.27–8.08) 0.55 (0.58, 1.09) 0.82 (0.14–1.26) 0.96 (0.36, 0.48)
TDP (mg L−1 ) 0.47 (0.37–0.56) 0.47 (0.45, 0.51) 0.26 (0.17–0.44) 0.24 (0.18, 0.31) 0.34 (0.12–0.69) 0.24 (0.21, 0.58) 0.45 (0.10–0.71) 0.55 (0.18, 0.31)
TOC (mg L−1 ) 17.14 (0.10–53.93) 10.77 (9.55, 13.9) 30.80 (16.71–41.13) 31.26 (25.39, 36.80) 22.05 (6.31–68.78) 16.97 (18.30, 26.76) 20.54 (5.84–28.86) 21.85 (25.39, 36.80)
Salinity (‰) 2.6 (1–6) 2 (2, 2) 3.9 (1–9) 3 (2, 3) 5.2 (1–18) 2 (2, 2) 4.7 (1–20) 2 (3, 3)
Q1: 1st quartile; Q3: 3rd quartile; Temp: temperature; TSS: total suspended solid; TN: total nitrogen; chl a: chlorophyll a; TP: Total phosphorus; TDP: total dissolved phosphorus.
7
Table 2: Correlation matrix (𝑟) of biotic and abiotic parameters in the Haihe River.
Temp TSS TN Nitrate Chl a TP TDP Rainfall TOC Salinity DB VB TV TB

Whole river
DB −0.304 0.262 −0.156 0.191 0.025 0.005 −0.148 −0.175 0.362 0.530 1.000 0.146 0.105 0.387∗
VB 0.323 0.106 0.164 −0.214 0.514∗∗ 0.171 −0.004 0.409∗ −0.117 0.126 0.146 1.000 0.575∗∗ 0.946∗∗
∗∗ ∗∗
TV 0.169 −0.085 0.177 0.025 0.440 0.189 0.064 0.220 −0.227 −0.177 0.105 0.575 1.000 0.568∗∗
TB 0.285 0.116 0.033 −0.186 0.489∗∗ 0.178 0.032 0.380∗ −0.049 0.232 0.387∗ 0.946∗∗ 0.568∗∗ 1.000
Urban areas
DB −0.595 −0.671 −0.476 0.619 −0.119 0.429 0.381 −0.927∗∗ 0.381 0.591 1.000 −0.190 0.524 0.286
VB 0.167 0.299 0.119 0.024 0.619 0.167 −0.048 0.293 −0.214 −0.206 −0.190 1.000 0.619 0.833∗
TV −0.500 −0.455 −0.381 0.619 0.286 0.286 0.071 −0.488 −0.190 −0.027 0.524 0.619 1.000 0.881∗∗
TB −0.095 −0.096 −0.333 0.262 0.643 0.429 0.190 −0.146 −0.095 −0.014 0.286 0.833∗ 0.881∗∗ 1.000
Rural areas
DB −0.294 0.322 −0.423 0.391 0.019 −0.362 −0.524∗ −0.124 0.444 0.131 1.000 −0.090 −0.086 0.180
∗∗ ∗ ∗∗ ∗∗ ∗ ∗∗
VB 0.680 0.151 0.127 −0.494 0.587 0.202 0.229 0.760 −0.203 −0.477 −0.090 1.000 0.651 0.937∗∗
∗ ∗ ∗∗ ∗∗
TV 0.461 −0.201 0.191 −0.234 0.432 0.088 0.103 0.543 −0.126 −0.602 −0.086 0.651 1.000 0.633∗∗
∗∗ ∗∗ ∗∗ ∗ ∗∗ ∗∗
TB 0.618 0.137 −0.104 −0.357 0.594 0.043 0.068 0.721 −0.144 −0.498 0.180 0.937 0.633 1.000
Estuarial areas
DB −0.071 0.786∗ 0.667 −0.238 0.048 0.333 0.228 0.439 0.119 0.667 1.000 0.286 0.548 0.714
VB 0.024 0.286 0.571 0.262 0.690 0.143 −0.575 −0.049 0.000 0.357 0.286 1.000 0.690 0.833∗
∗
TV 0.190 0.286 0.762 0.024 0.667 0.619 −0.204 0.390 −0.167 0.690 0.548 0.690 1.000 0.833∗
∗ ∗ ∗ ∗
TB 0.048 0.643 0.810 −0.095 0.619 0.357 −0.228 0.293 0.000 0.690 0.714 0.833 0.833 1.000
∗ ∗∗
𝑃 < 0.05; 𝑃 < 0.01.
DB: dead bacteria; VB: viable bacteria; TV: total viral abundance; TB: total bacteria abundance.
was positively influenced by salinity in estuarial areas (both soil flowing into water bodies. Meanwhile, the salinity could
𝑟 = 0.69 and 𝑃 = 0.05), indicating the different roles of also be diluted, thus promoting growth of host bacteria in
salinity in the two areas. The correlation between bacterial this region and leading to increased virus counts. During
abundance and salinity was also negative in the rural areas our sampling period, bacterial and viral abundance was
but positive in estuarial areas (Table 2), which indicates that significantly and positively correlated with temperature in
salinity may indirectly affect viral abundance via their hosts. rural areas, supporting the idea that temperature is an impor-
Otherwise, the opposite effects of salinity in the two areas tant environmental factor controlling microbial growth [9].
suggest that the predominant bacterial and viral communities Temperature may play an indirect role in viral abundance,
were possibly changed between rural and estuarial areas and as high temperatures benefit bacteria growth resulting in a
that the upstream production of native microbes could not higher number of hosts for bacteriophages. Moreover, higher
contribute to the abrupt growth of microbes in estuarial areas. numbers of viruses were associated with elevated nutrient
The rainfall pattern over the Haihe River is strongly seasonal concentrations (TN and TP) in estuarial areas (Table 2). Such
with wet summers and low winter rainfall. We found that the a positive correlation has also been documented in other
amount of rainfall positively influenced the viral (𝑃 < 0.05) water bodies [9, 42–44]. Some of these patterns probably
and bacterial abundances (𝑃 < 0.01) in rural areas (Table 2). result from the promotion of bacterial productivity under
Such correlations have been reported before, with seasonal eutrophic conditions, as well as the encouraged growth of
differences in viral and bacterial abundance and virus-host bacterial populations that might serve as viral hosts.
interactions being greatly influenced by rainfall [41]. During These results show that viral and bacterial abundance
this study, we found that the amount of rainfall in rural areas was both influenced by environmental factors. The level of
greatly affected the measured environmental parameters. correlation between abiotic variables and viral abundance
For instance, we observed a positive correlation with water was always similar to that of the relationship between abiotic
temperature (𝑟 = 0.753, 𝑃 < 0.001), TP (𝑟 = 0.617, 𝑃 < 0.01; variables and bacterial abundance (Table 2), indicating the
often regarded as nonpoint pollutions in agricultural areas), importance of indirect influence on viruses via their bacte-
TDP (𝑟 = 0.586, 𝑃 < 0.01) and chl a (𝑟 = 0.574, 𝑃 < rial host cells. The environmental factors influencing viral
0.01), and a negative correlation with salinity (𝑟 = −0.523, abundance varied in the three sampling areas, suggesting that
𝑃 < 0.05). Thus, the positive impact of rainfall on microbial the viral abundance in different stretches of the river with
abundance could be explained by the significant input of their specific surroundings was possibly driven directly and
nutrients from the surface runoff, with phosphorus from the indirectly by different abiotic factors.
4. Conclusions [10] N. Jiao, Y. Zhao, T. Luo, and X. Wang, “Natural and anthro-
pogenic forcing on the dynamics of virioplankton in the Yangtze
(1) The present study has provided an adequate and fast river estuary,” Journal of the Marine Biological Association of the
method using flow cytometry to investigate the vari- United Kingdom, vol. 86, no. 3, pp. 543–550, 2006.
ation of viral abundance in freshwater environments. [11] M. G. Weinbauer, “Ecology of prokaryotic viruses,” FEMS Micr-
(2) Viral distribution could be influenced both by biotic obiology Reviews, vol. 28, no. 2, pp. 127–181, 2004.
and abiotic variables. [12] R. Maranger and D. F. Bird, “Viral abundance in aquatic sys-
tems: a comparison between marine and fresh waters,” Marine
(3) The effects of abiotic variables on viral abundance
Ecology Progress Series, vol. 121, no. 1–3, pp. 217–226, 1995.
may also lie in the indirect influence via their bacterial
host cells. [13] J. L. Clasen, S. M. Brigden, J. P. Payet, and C. A. Suttle, “Evidence
that viral abundance across oceans and lakes is driven by
(4) There were clear differences in microbial abundance different biological factors,” Freshwater Biology, vol. 53, no. 6,
among the three sampling environments (urban, pp. 1090–1100, 2008.
rural, and estuarial areas), and the abiotic influence [14] C. Corinaldesi, E. Crevatin, P. del Negro et al., “Large-scale
on viral abundance was different in all three environ- spatial distribution of virioplankton in the adriatic sea: testing
ments. the trophic state control hypothesis,” Applied and Environmental
Acknowledgments [15] M. Wang, Y. Liang, X. Bai, X. Jiang, F. Wang, and Q. Qiao,
“Distribution of microbial populations and their relationship
The authors are grateful to the financial support from the with environmental parameters in the coastal waters of Qing-
National Basic Research Program of China (2010CB530304), dao, China,” Environmental Microbiology, vol. 12, no. 7, pp. 1926–
National Science Foundation of China (31000247), New Cen- 1939, 2010.
tury Talent Program, Ministry of Education, China (NCET- [16] K. E. Wommack, R. T. Hill, T. A. Muller, and R. R. Colwell,
09-0487), and EPP program of Eawag, Switzerland. They “Effects of sunlight on bacteriophage viability and structure,”
would like to thank Professor Thomas Egli, Dr. Frederik Applied and Environmental Microbiology, vol. 62, no. 4, pp.
1336–1341, 1996.
Hammes, and Professor Mark Bartlam for their consistent
guidance, invaluable comments, and suggestion during the [17] S. Jacquet and G. Bratbak, “Effects of ultraviolet radiation on
whole experiment. marine virus-phytoplankton interactions,” FEMS Microbiology
Ecology, vol. 44, no. 3, pp. 279–289, 2003.
[18] R. T. Noble and J. A. Fuhrman, “Virus decay and its causes in
References coastal waters,” Applied and Environmental Microbiology, vol.
63, no. 1, pp. 77–83, 1997.
[1] O. Bergh, K. Y. Børsheim, G. Bratbak, and M. Heldal, “High
abundance of viruses found in aquatic environments,” Nature, [19] G. Bratbak, J. K. Egge, and M. Heldal, “Viral mortality of the
vol. 340, no. 6233, pp. 467–468, 1989. marine alga Emiliania huxleyi (Haptophyceae) and termination
[2] K. E. Wommack and R. R. Colwell, “Virioplankton: viruses of algal blooms,” Marine Ecology Progress Series, vol. 93, no. 1-2,
in aquatic ecosystems,” Microbiology and Molecular Biology pp. 39–48, 1993.
Reviews, vol. 64, no. 1, pp. 69–114, 2000. [20] R. Thomas, L. Berdjeb, T. Sime-Ngando, and S. Jacquet, “Viral
[3] R. A. Edwards and F. Rohwer, “Viral metagenomics,” Nature abundance, production, decay rates and life strategies (lysogeny
Reviews Microbiology, vol. 3, no. 6, pp. 504–510, 2005. versus lysis) in Lake Bourget (France),” Environmental Microbi-
ology, vol. 13, no. 3, pp. 616–630, 2011.
[4] C. A. Suttle, “Viruses in the sea,” Nature, vol. 437, no. 7057, pp.
356–361, 2005. [21] K. Y. Børsheim, G. Bratbak, and M. Heldal, “Enumeration
[5] S. W. Wilhelm and C. A. Suttle, “Viruses and nutrient cycles in and biomass estimation of planktonic bacteria and viruses by
the sea—viruses play critical roles in the structure and function transmission electron microscopy,” Applied and Environmental
of aquatic food webs,” BioScience, vol. 49, no. 10, pp. 781–788, Microbiology, vol. 56, no. 2, pp. 352–356, 1990.
1999. [22] K. E. Wommack, R. T. Hill, M. Kessel, E. Russek-Cohen, and
[6] G. Malin, W. H. Wilson, G. Bratbak, P. S. Liss, and N. H. Mann, R. R. Colwell, “Distribution of viruses in the Chesapeake Bay,”
“Elevated production of dimethylsulfide resulting from viral Applied and Environmental Microbiology, vol. 58, no. 9, pp.
infection of cultures of Phaeocystis pouchetii,” Limnology and 2965–2970, 1992.
Oceanography, vol. 43, no. 6, pp. 1389–1393, 1998. [23] S. Hara, K. Terauchi, and I. Koike, “Abundance of viruses in
[7] R. V. Miller, S. Ripp, J. Relicon, O. A. Ogunseitan, and T. A. marine waters: assessment by epifluorescence and transmission
Kokjohn, “Virus-mediated gene transfer in freshwater environ- electron microscopy,” Applied and Environmental Microbiology,
ment,” in Gene Transfer and Environment, M. J. Gauthier, Ed., vol. 57, no. 9, pp. 2731–2734, 1991.
pp. 51–62, Springer, Berlin, Germany, 1992. [24] K. P. Hennes and C. A. Suttle, “Direct counts of viruses in nat-
[8] D. Lymer, J. B. Logue, C. P. D. Brussaard, A.-C. Baudoux, K. ural waters and laboratory cultures by epifluorescence micro-
Vrede, and E. S. Lindstrom, “Temporal variation in freshwater scopy,” Limnology and Oceanography, vol. 40, no. 6, pp. 1050–
viral and bacterial community composition,” Freshwater Biol- 1055, 1995.
ogy, vol. 53, no. 6, pp. 1163–1175, 2008. [25] D. Marie, C. P. D. Brussaard, R. Thyrhaug, G. Bratbak, and D.
[9] S. W. Wilhelm and A. R. Matteson, “Freshwater and marine viri- Vaulot, “Enumeration of marine viruses in culture and natural
oplankton: a brief overview of commonalities and differences,” samples by flow cytometry,” Applied and Environmental Micro-
Freshwater Biology, vol. 53, no. 6, pp. 1076–1089, 2008. biology, vol. 65, no. 1, pp. 45–52, 1999.
[26] S. Personnic, I. Domaizon, U. Dorigo, L. Berdjeb, and S. Jacquet, [42] R. Maranger and D. F. Bird, “High concentrations of viruses in
“Seasonal and spatial variability of virio-, bacterio-, and pico- the sediments of Lac Gilbert, Québec,” Microbial Ecology, vol.
phytoplanktonic abundances in three peri-alpine lakes,” Hydro- 31, no. 2, pp. 141–151, 1996.
biologia, vol. 627, no. 1, pp. 99–116, 2009. [43] I. Hewson, J. M. O’Neil, J. A. Fuhrman, and W. C. Dennison,
[27] C. P. D. Brussaard, D. Marie, and G. Bratbak, “Flow cytometric “Virus-like particle distribution and abundance in sediments
detection of viruses,” Journal of Virological Methods, vol. 85, no. and overlying waters along eutrophication gradients in two
1-2, pp. 175–182, 2000. subtropical estuaries,” Limnology and Oceanography, vol. 46, no.
[28] W. K. W. Li and P. M. Dickie, “Monitoring phytoplankton, 7, pp. 1734–1746, 2001.
bacterioplankton, and virioplankton in a coastal inlet (Bedford [44] P. Peduzzi and F. Schiemer, “Bacteria and viruses in the water
Basin) by flow cytometry,” Cytometry, vol. 44, pp. 236–246, column of tropical freshwater reservoirs,” Environmental Micro-
2001. biology, vol. 6, no. 7, pp. 707–715, 2004.
[29] C. P. D. Brussaard, K. R. Timmermans, J. Uitz, and M. J. W.
Veldhuis, “Virioplankton dynamics and virally induced phyto-
plankton lysis versus microzooplankton grazing southeast of
the Kerguelen (Southern Ocean),” Deep-Sea Research Part II,
vol. 55, no. 5–7, pp. 752–765, 2008.
[30] J. P. Payet and C. A. Suttle, “Physical and biological correlates
of virus dynamics in the southern Beaufort Sea and Amundsen
Gulf,” Journal of Marine Systems, vol. 74, no. 3-4, pp. 933–945,
2008.
[31] C. P. D. Brussaard, “Optimization of procedures for counting
viruses by flow cytometry,” Applied and Environmental Micro-
biology, vol. 70, no. 3, pp. 1506–1513, 2004.
[32] C. P. D. Brussaard, J. P. Payet, C. Winter, and M. G. Weinbauer,
“Quantification of aquatic viruses by flow cytometry,” in Manual
of Aquatic Viral Ecology, pp. 102–109, ASLO, 2010.
[33] M. Berney, M. Vital, I. Hülshoff, H.-U. Weilenmann, T. Egli,
and F. Hammes, “Rapid, cultivation-independent assessment of
microbial viability in drinking water,” Water Research, vol. 42,
no. 14, pp. 4010–4018, 2008.
[34] X. C. Jin and Q. Y. Tu, Eds., The Standard Methods for Obser-
vation and Analysis of Lake Eutrophicationed, Environmental
Science Press, Beijing, China, 2nd edition, 1990.
[35] G. McDonnell and A. D. Russell, “Antiseptics and disinfectants:
activity, action, and resistance,” Clinical Microbiology Reviews,
vol. 12, pp. 147–179, 1999.
[36] A. Patel, R. T. Noble, J. A. Steele, M. S. Schwalbach, I. Hew-
son, and J. A. Fuhrman, “Virus and prokaryote enumera-
tion from planktonic aquatic environments by epifluorescence
microscopy with SYBR Green I,” Nature Protocols, vol. 2, no. 2,
pp. 269–276, 2007.
[37] S. J. Williamson, L. A. Houchin, L. McDaniel, and J. H. Paul,
“Seasonal variation in lysogeny as depicted by prophage induc-
tion in Tampa Bay, Florida,” Applied and Environmental Micro-
biology, vol. 68, no. 9, pp. 4307–4314, 2002.
[38] J. C. Auguet, H. Montanié, D. Delmas, H. J. Hartmann, and
V. Huet, “Dynamic of virioplankton abundance and its envi-
ronmental control in the Charente estuary (France),” Microbial
Ecology, vol. 50, no. 3, pp. 337–349, 2005.
[39] M. G. Weinbauer and C. A. Suttle, “Comparison of epifluo-
rescence and transmission electron microscopy for counting
viruses in natural marine waters,” Aquatic Microbial Ecology,
vol. 13, no. 3, pp. 225–232, 1997.
[40] J. H. Paul, J. B. Rose, S. C. Jiang, C. A. Kellogg, and L. Dickson,
“Distribution of viral abundance in the reef environment of Key
Largo, Florida,” Applied and Environmental Microbiology, vol.
59, no. 3, pp. 718–724, 1993.
[41] C. Säwström and P. Pollard, “Environmental influences on
virus-host interactions in an Australian subtropical reservoir,”
Environmental Microbiology Reports, vol. 4, no. 1, pp. 72–81,
2012.
http://dx.doi.org/10.1155/2013/838145
Research Article
The Occurrence of Two Species of Entomophthorales
(Entomophthoromycota), Pathogens of Sitobion avenae and
Myzus persicae (Hemiptera: Aphididae), in Tunisia
Ibtissem Ben Fekih,1,2 Sonia Boukhris-Bouhachem,1 Jørgen Eilenberg,3

Mohamed Bechir Allagui,1 and Annette Bruun Jensen3
1
Plant Protection Laboratory of National Institute of Agricultural Research of Tunisia, Rue Hédi Karray, 2049 Ariana, Tunisia
2
National Institute of Agronomy of Tunisia, University of Carthage, 43, Avenue Charles Nicolle, 1082 Cité Mahrajène, Tunisia
3
Department of Plant and Environmental Sciences, University of Copenhagen, Thorvaldsensvej 40, 1871 Frederiksberg, Denmark
Correspondence should be addressed to Ibtissem Ben Fekih; fekih.ibtissem@gmail.com
Received 1 March 2013; Revised 22 April 2013; Accepted 28 May 2013
Copyright © 2013 Ibtissem Ben Fekih et al. This is an open access article distributed under the Creative Commons Attribution
cited.
The natural occurrence of entomophthoralean fungi pathogenic towards aphids on cereal and potato crops was investigated in
the years 2009, 2010, and 2011. Infected aphids were sampled in three bioclimatic zones in Tunisia (Beja, Cap bon, and Kairouan)
and fungal species were determined based on morphological characters such as shape, size, and number of nuclei in the primary
conidia. Polymerase Chain Reaction (PCR) on the internal transcribed spacer 1 region (ITS1) was used to verify morphological
determination. Both methods gave consistent results and we documented for the first time the natural occurrence of two fungal
species from the order Entomophthorales (phylum Entomophthoromycota), Pandora neoaphidis and Entomophthora planchoniana.
Both fungi were recorded on the aphid species Sitobion avenae and Myzus persicae on barley ears and potato leaves, respectively.
Moreover, natural mixed infections by both species (P. neoaphidis and E. planchoniana) were documented on the target aphids. This
investigation provides basic information of entomopathogenic fungi infecting economically important aphids in Tunisia.
1. Introduction sustainable control strategy to manage pest insects like aphids

[6]. Field observations have shown that aphid populations
Aphids (Hemiptera: Aphididae) are one of the most impor- are commonly regulated by a range of natural enemies, such
tant groups of insect pests in agriculture. They weaken their as predators, parasitoids, and also fungal pathogens [7]. In
host plants in diverse ways by causing direct damage as temperate regions, fungal species from the phylum Ento-
phloem feeders and also by indirect damage as plant virus mophthoromycota are important pathogens of aphids [8–
vectors [1]. A total of approximately 4000 aphid species have 10]. Their ability to cause epizootics among their host insects
been described, of which 157 species have been reported within a short time makes them potentially valuable for pest
in Tunisia [2]. Most of the identified species in Tunisia are control and by that an element in future IPM systems [11–15].
considered to be pests and cause significant yield losses to A new taxonomical revision assigned entomophthoralean
important crops such as cereals [3]. Furthermore, dissemi- fungi under the phylum Entomophthoromycota with three
nation of the Potato Virus Y (PVY) by aphids is considered new orders, Entomophthorales, Neozygitales, and Basidiobo-
as big problem in Tunisian potato fields [4]. The develop- lales [9]. The most common species worldwide infecting
ment of insecticide resistance among aphids has stimulated aphids belong to the Entomophthorales, particularly to the
an interest in developing alternative methods of control families Entomophthoraceae and Ancylistaceae, and the
[5]. Integrated pest management (IPM) can be seen as a Neozygitales represented by the family Neozygitaceae [9].
The aphid pathogenic species have been documented almost 2.2.3. Scanning Electron Microscopy. Infected aphids were
worldwide with most records from temperate climatic zones examined with an environmental scanning electron micro-
[8, 16–18]. Little is known, however, about the natural occur- scope (ESEM) aiming to obtain information about detailed
rence of these fungal pathogens in North Africa. A previous structures of the fungi. Samples were carbon coated using a
study in Egypt recorded twelve entomopathogenic fungal conductive carbon disk and observation was done through
species of which seven belonged to Entomophthoromycota the ESEM model QUANTA 200I-D7827 with tungsten (W)
[19]. The approach using entomopathogenic fungi in biologi- filament electron source.
cal control is a new field in Tunisia. So far, only few studies
on two Fusarium (Ascomycota) species on the artichoke
2.3. Molecular Characterization: DNA Extraction and PCR
aphid species Capitophorus elaeagni have been performed in
Amplification. Genomic DNA from 14 infected aphids
Tunisia [20, 21]. Thus, basic knowledge about occurrence,
(Myzus persicae and Sitobion avenae) and one healthy Sitobion
distribution, and prevalence over time of entomophthoralean
avenae taken from rearing chamber and considered as a
fungi in Tunisia is completely lacking. In this study we wanted
negative control was extracted using a Chelex extraction
to explore the natural occurrence of entomophthoralean
protocol [25]. DNA extraction was done by adding 20 𝜇L
fungi in relation to Sitobion avenae [22], among important
phosphate buffered saline (PBS PH7.2) and 5 𝜇L proteinase
aphid species infesting barley ears and Myzus persicae [22], a
K (10 mg/mL) to each 1.5 mL Eppendorf tube and the aphids
common pest aphids infesting potato fields in Tunisia.
were homogenized with a DNA-free pestle. After a quick
spin at 10,500 ×g for 30 s, 100–200 𝜇L (depending on aphid
2. Materials and Methods size) of a 10% Chelex solution was added and the samples
were incubated overnight at 56∘ C. Next day the samples were
2.1. Sampling. The investigations of mycoses in aphid pop- incubated at 94∘ C for 15 min and after a spin at 10,500 ×g
ulation (apterae and alate specimens) were done in three for 30 s, the supernatants were transferred to new Eppendorf
regions of Tunisia: in the north west, Beja: site of the Regional tubes and stored at −20∘ C.
Field Crop Research Center (36∘ 44󸀠 00󸀠󸀠 N, 9∘ 11󸀠 00󸀠󸀠 E), a sub- PCR was performed on the internal transcribed
humid area; in the north east, Cap bon: Site of Soliman spacer 1 (ITS 1) using two genus specific forward primers
(36∘ 40󸀠 40󸀠󸀠 N, 10∘ 28󸀠 20󸀠󸀠 E), situated in the semiarid area of for Entomophthora and Pandora, respectively: Ml2:
the region; in the center, Kairouan: Site of Sidi Mahmoud 5󸀠 -GCAACGGATCATCATGTAA-3󸀠 and PnCNf: 5󸀠 -
(35∘ 37󸀠 07󸀠󸀠 N, 9∘ 55󸀠 34󸀠󸀠 E), a continental zone with arid cold TTTGGGTTTAAATAGAAGGTTGA-3󸀠 and reverse prim-
winter. Barley and potato fields infested with aphids were ers Nu-5.8S-3󸀠 : 5󸀠 -ACTACGTTCTTCATCGATGA-3󸀠 [10]
used for random sampling of fungal infected aphids between and PnCNr: 5󸀠 -AGGCAAAGCCTAGAGCACTT-3󸀠 (unpu-
March and June of the years 2009, 2010, and 2011. Aphid bished). The primers were chosen to detect and confirm the
cadavers with symptoms of fungal infection were placed into identity of the fungi infecting the field collected aphids.
ventilated plastic boxes and carried to the laboratory.
Positive DNA controls provided from ARSEF Collection
of Entomopathogenic Fungal Cultures: ARSEF 2583: Pandora
2.2. Morphological Characterization. Fungus identification neoaphidis, isolated from the aphid species Acyrthosiphon
was based on the shape, size, and nuclei numbers in the pisum (USA, 1988) and ARSEF 6918: Entomophthora muscae,
primary conidia [23, 24]. The number of aphids subject to isolated from the Dipteria Coenosia tigrina (Denmark, 1999)
morphological identification of fungal infection was 730 for and negative water controls were included in each set of PCR
M. persicae and 980 for S. avenae. reactions.
PCR amplifications were performed in 50 𝜇L reaction
volumes containing 2 𝜇L of chelex-extracted DNA 1 : 1 or
2.2.1. Sample Preparation. The aphid cadavers with fresh
diluted 1 : 10, 10 𝜇L Phusion HF Buffer (5 × 7.5 mM MgCl2 ),
conidiophores were inverted over a glass slide in moist boxes
10 mM dNTPs, 0.5 𝜇M of each primer, 0.5 U Phusion High-
at 20∘ C to allow conidia ejection for first 5 hours and then
Fidelity DNA Polymerase (Finnzymes, Espoo, FI). For both
8 hours. Subsequently, some of the cadavers were stored
primers, the PCR conditions were denaturation at 98∘ C for
individually in 96% ethanol to be used in later molecular
30 s, followed by 38 cycles of denaturation at 98∘ C for 10 s,
examination. Light and electron microscopic studies were
annealing at 55∘ C for ML2 or 60∘ C for PnCNf for 20 s, and
used for the morphological determination of the fungal
extension at 72∘ C for 1 min, with a final extension at 72∘ C for
species.
10 min. The size of the PCR amplifications was estimated by
electrophoresis on a 1.5% agarose gel in 0.5 × TBE, and the
2.2.2. Light Microscopy. Conidia projected on the slide were products visualized with EZ-Vision (AMRESCO LLC, USA).
mounted in either lactic acid or aceto-orcein for measuring Sequencing was used to verify the species identity based
size or counting nuclei numbers, respectively. The shape and on the length of the PCR product, in particular for the P.
size (length and width) and the nuclei number per conidia neoaphidis primers PnCNf/PnCNr which were tested for the
of twenty randomly chosen conidia per aphid were measured first time in this study. Prior to sequencing the PCR products
and counted on a computer screen coupled to an Olympus were purified with Qiagen kit. The different PCR products
microscope at 400x magnification. 20 aphids per crop were were sequenced by Eurofins MWG and a sequences similarity
used. search using NCBI BLAST was performed in Genbank.
(a) (b)
c1
(c) (d)
Figure 1: Pandora neoaphidis. (a) Myzus persicae killed and with fungus outgrowth, (b) primary conidia (stained by lactic acid), (c) primary
conidia (stained by aceto-orceine) documenting one nucleus per conidium. (c1) Secondary conidium being produced by primary conidium.
(d) SEM image of a germinating primary conidium on host cuticle.
3. Results and Discussion 11.7–18.2, respectively (Table 1). Nuclei numbers and conidia
dimensions of this species are thus within the description of
Our study documented for the first time in Tunisia fungal E. planchoniana [24].
species within the phylum Entomophthoromycota, family In this study, primary conidia were the main fungal
Entomophthoraceae [9]. We identified two species of the sub- structure used for species identification. However, another
families Erynioideae and Entomophthoroideae, respectively: fungal structure emerging from the host aphid cadavers
Pandora neoaphidis, (Remaudière and Hennebert) [26] and named rhizoids ensuring the attachment of the aphids to
Entomophthora planchoniana [27] found on M. persicae and the plants was also documented in this study. The presence
S. avenae, respectively. or absence of rhizoids and their characters (monohyphal
or compound) with or without specialized holdfasts should
be taken into consideration during the fungus identification
3.1. Morphological Characterization. The species P. neoap-
[15]. E. planchoniana have particular monohyphal rhizoids
hidis is characterized by visible ellipsoid mononucleate
with disc-like ending which is a characteristic of the species
primary conidia [23, 24] (Figure 1). The Pandora conidia
(Figures 3(a) and 3(b)). However, P. neoaphidis rhizoids
from the Tunisian material contained one nucleus each
are monohyphal ending with irregularly terminal branches
and measured in length and width 22.1–30.9 × 15.6–18.8,
(Figure 3(c)).
respectively (Table 1). This is well within the known range of
P. neoaphidis conidia; however, the conidia obtained during
this investigation tended to be a bit larger when compared to 3.2. Molecular Analysis. We showed that two genus specific
data in [23]. ITS 1 primer sets ML2/Nu-5.8S-3󸀠 (Entomophthora) and
The species E. planchoniana is characterized by bell PnCNf/PnCNr (Pandora) worked well on the aphid cadavers
shaped plurinucleate primary conidia with sharp apical point collected in Tunisia. The amplification of the ITS1 region
and a broad flattened papilla [24] (Figure 2). The Entomoph- from 14 infected aphids using the two primers sets showed
thora conidia from the Tunisian material contained between different profiles. The Entomophthora primers amplified a
4 and 7 nuclei and measured in length and width 16.5–29.9 × single amplicons with expected size estimated to 350 bp for
(a) (b)
d1
(c) (d)
Figure 2: Entomophthora planchoniana. (a) Sitobion avenae killed and with fungus outgrowth. (b) Primary conidia (stained by lactic acid).
(c) Primary conidia (stained by aceto-orceine). (d) SEM image of hyphae, conidiophores, and one visible conidium (d1).
Table 1: Number of nuclei and dimensions (in 𝜇m) of primary conidia of Pandora neoaphidis and Entomophthora planchoniana.
Fungal species Number of nuclei Length × width of the primary conidia Reference
1 15–40 × 9–16 Humber [23]
P. neoaphidis
1 22.1–30.9 × 15.6–18.8 This study
4–8 15–20 × 12–16 Keller [24]
E. planchoniana
4–7 16.5–29.9 × 11.7–18.2 This study
13 samples which was the same size for the positive control product was obtained from uninfected aphid and the water
ARSEF 6918 (E) (Figure 4(a)). Sample numbers 11 and 14 (W) both considered as negative control.
show less intense amplicons which could be related to the Similarity search of nucleotide sequences in GenBank
depletion of fungal material in the hosts due to the full shows that all E. planchoniana sequences were 100 per-
discharge of Entomophthora conidia. This was congruent with cent similar to E. planchoniana isolate. For P. neoaphidis
the morphological analysis of conidia from the same infected all the sequence were 99 percent similar to many P.
aphid cadavers that were all identified as E. planchoniana neoaphidis isolates in Genbank. This is in favor of future
(Table 2). DNA extractions from the same 14 aphids were also use of PnCNf/PnCNr for molecular characterization of P.
screened with the Pandora primers. One clear single band neoaphidis isolates and probably also other Pandora species.
was produced from 4 infected cadavers with same size for PCR profiles indicated possible mixed fungal infections
the positive control ARSEF 2583 (P). The amplification profile in samples 10, 12, and 14 (Figures 4(a) and 4(b)). The mor-
gave strong amplicons for the samples number 7, 10, and 14 phological examination of the conidia projected from those
whereas it was less intense for the aphid sample number 12 aphid samples showed the presence of both P. neoaphidis
(Figure 4(b)). This was congruent with the morphological and E. planchoniana. This “concomitant” or “mixed infection”
analysis of conidia from the same infected aphid cadavers that seems to be relatively common in nature in our pest-pathogen
were all identified as P. neoaphidis (Table 2). No amplification systems and the frequency of such mixed infections may
(a) (b)
(c)
Figure 3: Monohyphal rhizoids. (a) Entomophthora planchoniana rhizoids with disc-like ending emerging from the ventral abdominal
region of infected Rhopalosiphum padi. (b) SEM Image of E. planchoniana rhizoids. (c) Pandora neoaphidis rhizoids with irregularly terminal
branches.
M T 1 2 3 4 5 6 7 8 9 10 11 12 13 14 P E W M T 1 2 3 4 5 6 7 8 9 10 11 12 13 14 P E W
800 bp
350 bp
(a) (b)
Figure 4: Amplification of ITS1 region using (a) the Entomophthora specific primer set ML2/Nu-5.8S-3󸀠 and (b) Pandora neoaphidis primers
PnCNf / PnCNr, using DNA extracted from a healthy aphid (T), 14 entomophthoralean killed cadavers (1–14), Pandora neoaphidis DNA (P),
and Entomophthora muscae DNA (E). Size marker (MW) = 100 pb.
actually be influenced by both environmental conditions and Both species were found in all the bioclimatic zones
other factors [28, 29]. included in this study: an arid region (Kairouan), a sub-
The simultaneous usage of both morphological and humid region (Beja), and a semi-arid region (Soliman, Cap
molecular methods gave a very strong background both with bon). Cereal and potato areas are mainly situated in the
respect to the correct identification of P. neoaphidis and E. north and center of Tunisia, where the climate switch from
planchoniana and with respect to future studies determining subhumid to arid which is considered suitable environment
the full spectrum of entomophthoralean fungi infecting for the development of entomophthoralean infection. The
aphids and their wider distribution of these fungi in Tunisia. occurrence of P. neoaphidis and E. planchoniana in the
Table 2: Fungal identification using morphological and molecular analysis.
Morphological analysis Molecular analysis

Aphids code Aphid species Host plant Sampling areas
P. neoaphidis E. planchoniana P. neoaphidis E. planchoniana
1 S. avenae Barely Beja − + − +
4 S. avenae Barely Soliman − + − +
5 M. persicae Potato Soliman − + − +
6 M. persicae Potato Kairouan − + − +
7 S. avenae Barely Beja + − + −
8 S. avenae Barely Soliman − + − +
10 M. 𝑝𝑒𝑟𝑠𝑖𝑐𝑎𝑒∗ Potato Soliman + + + +
11 M. persicae Potato Soliman − + − +
12 M. 𝑝𝑒𝑟𝑠𝑖𝑐𝑎𝑒∗ Potato Soliman + + + +
14 S. 𝑎V𝑒𝑛𝑎𝑒∗ Barely Soliman + + + +
(+) Documented fungi species; (−) undocumented fungi species.
∗
Aphids with mixed infection.
investigated regions seems to be in concordance with their economically important aphids in Tunisia. Further surveys
distribution in different bioclimatic zones. They have both a will be considered to explore other pathogen species and
worldwide distribution (Europe, Australia, North and South prevalence studies will be adopted to explore the poten-
America, North and South Africa, and Asia), and while P. tial of this order. Such information could be used in the
neoaphidis is common when the temperature is moderate and establishment of a framework for a national program of
the humidity high, E. planchoniana can be common in dry integrated pest management. In addition, it will improve
and moderately humid environment [8, 16]. our understanding of the worldwide distribution of aphid
Aphid pathogenic fungal species belong either to the pathogenic Entomophthoromycota in particular from North
phylum Ascomycota within the order Hypocreales (gen- Africa.
era Beauveria, Fusarium, Paecilomyces, Lecanicillium, and
others) or to the phylum Entomophthoromycota [9, 23].
Previous investigation of pathogens on the artichoke aphid Conflict of Interests
species Capitophorus elaeagni in Tunisia reported two Fusar- The authors declare no financial gain of conflicts or interests
ium species (F. sacchari and F. semitectum), both show- with regard to Fusion, EZvision, that they were referring to
ing potential against different aphid species [20, 21]. The within the paper. This just happens to be the polymerase and
investigation was done in two region of Tunisia: Bizerte the dye that were applicable in the current study. The same
(humid zone) and Sousse (semiarid zone). Interestingly, no goes for Eurofins MWG, with which there is no conflict of
entomophthoralean species was recorded by these authors, interests.
despite their widespread occurrence in such bioclimatic
zones documented in our studies. Which might be due to the
sampling strategy. References
Species from the order Entomophthorales are host-
specific with high potential to regulate aphid populations [1] C.-A. Dedryver, A. Le Ralec, and F. Fabre, “The conflicting
in field crops. Both identified species are considered to be relationships between aphids and men: a review of aphid
damage and control strategies,” Comptes Rendus Biologies, vol.
dominant among 30 species of Entomophthorales infecting
333, no. 6-7, pp. 539–553, 2010.
a wide range of hosts from Aphidoidea (Hemiptera) [8,
16]. However, obstacles related to the mass production and [2] S. Boukhris-Bouhachem, R. Souissi, H. Sellami, and M. Hullé,
inoculums formulation of both fungi are still not solved “Diversité des espèces de pucerons et évolution des populations
[8]. To date, biological control guidelines of aphids with des espèces vectrices,” in Proceedings of the 23nd International
Forum for Biological Sciences, pp. 21–24, Tunisian Association
Entomophthorales highlight often the conservation biologi-
of Biological Sciences, 2012.
cal control approach which enhances the natural occurring
of those pathogens [14]. [3] H. Harbaoui, K. M. Mezghani, M. Marrakchi, M. Makni, and
H. Makni, “Etude morphologique et toxicologique des pucerons
(homoptères: aphididae) des céréales en Tunisie,” Bulletin de la
4. Conclusion Société Zoologique de France, vol. 133, no. 1–3, pp. 203–213, 2008.
The present research provides fundamental information [4] S. Boukhris-Bouhachem, R. Souissi, E. Turpeau et al., “Aphid
on the occurrence of fungal entomopathogens infecting (hemiptera: Aphidoidea) diversity in Tunisia in relation to seed
potato production,” Annales de la Societe Entomologique de [20] J. Guesmi-Jouini, N. Boughalleb-M’hamdi, and M. Ben Halima-
France, vol. 43, no. 3, pp. 311–318, 2007. Kamel, “Etudes préliminaires sur les champignons ento-
[5] H. Hesketh, P. G. Alderson, B. J. Pye, and J. K. Pell, “The mopathogènes des pucerons de l’artichaut en Tunisie,” Faunistic
development and multiple uses of a standardised bioassay Entomology, vol. 63, no. 3, pp. 171–181, 2011.
method to select hypocrealean fungi for biological control of [21] G. Jouda, B. H. K. Monia, and B. Naima, “First report
aphids,” Biological Control, vol. 46, no. 2, pp. 242–255, 2008. of aphidopathogenic fungi fusarium semitectum (Berkeley
and Ravenel, 1875) and Fusarium sacchari (Butler and Hafiz
[6] A. Jarraya, Principaux nuisibles des plantes cultivées et des
Khan) Gams (1971) on Capitophorus elaeagni (Del Guer-
denrées stockées en Afrique du Nord. Leur biologie, leurs ennemis
cio) (Hemiptera: Aphididae),” African Journal of Agricultural
naturels, leurs dégâts et leur contrôle, Climat Pub, Tunis, Tunisie,
Research, vol. 5, no. 4, pp. 290–293, 2010.
2003.
[22] R. L. Blackman and V. F. Eastop, Aphids on the World’s crops:
[7] A. B. Jensen, L. M. Hansen, and J. Eilenberg, “Grain aphid An Identification and Information Guide, John Wiley & Sons,
population structure: no effect of fungal infections in a 2-year Chichester, UK, 1984.
field study in Denmark,” Agricultural and Forest Entomology,
[23] R. A. Humber, “Fungi: identification,” in Manual of Techniques
vol. 10, no. 3, pp. 279–290, 2008.
in Insect Pathology, L. A. Lacey, Ed., pp. 153–186, Academic
[8] M. Barta and L. Cagan, “Aphids-pathogenic entomophthorales Press, San Diego, Calif, USA, 1997.
(their taxonomy, biology and ecology),” Biologia, vol. 61, no. 21, [24] S. Keller, “Entomophthorales attacking aphids with a descrip-
pp. 543–616, 2006. tion of two new species,” Sydowia, vol. 58, no. 1, pp. 38–74, 2006.
[9] R. A. Humber, “Entomophthoromycota: a new phylum and [25] M. Traugott, J. R. Bell, G. R. Broad et al., “Endoparasitism
reclassification for entomophthoroid fungi,” Mycotaxon, vol. in cereal aphids: molecular analysis of a whole parasitoid
120, pp. 477–492, 2012. community,” Molecular Ecology, vol. 17, no. 17, pp. 3928–3938,
[10] A. B. Jensen, J. Eilenberg, and C. López Lastra, “Differential 2008.
divergences of obligately insect-pathogenic Entomophthora [26] R. A. Humber, “Synopsis of a revised classification for the
species from fly and aphid hosts,” FEMS Microbiology Letters, Entomophthorales (Zygomycotina),” Mycotaxon, vol. 34, pp.
vol. 300, no. 2, pp. 180–187, 2009. 441–460, 1989.
[11] J. Eilenberg, Biology of fungi from the order Entomophthorales [27] M. Cornu, “Note sur une nouvelle espèce d’ Entomophthora (E.
[D.Sc. dissertation], Department of Ecology, The Royal Veteri- planchoniana),” Bulletin de la Société botanique de France, vol.
nary and Agriculture University, Copenhagen, Denmark, 2002. 20, pp. 189–190, 1873.
[12] M. S. Goettel, J. Eilenberg, and T. Glare, “Entomopathogenic [28] M. B. Thomas, E. L. Watson, and P. Valverde-Garcia, “Mixed
fungi and their role in regulation of insect populations,” in infections and insect-pathogen interactions,” Ecology Letters,
Comprehensive Molecular Insect Science, L. I. Gilbert, K. Iatrou, vol. 6, no. 3, pp. 183–188, 2003.
and S. Gill, Eds., vol. 6, pp. 361–406, Elsevier, London, UK, 2004. [29] S. Vojvodic, J. J. Boomsma, J. Eilenberg, and A. B. Jensen,
[13] A. E. Hajek and I. Delalibera Jr., “Fungal pathogens as classical “Virulence of mixed fungal infections in honey bee brood,”
biological control agents against arthropods,” BioControl, vol. Frontiers in Zoology, vol. 9, p. 5, 2012.
55, no. 1, pp. 147–158, 2010.
[14] H. Hesketh, H. E. Roy, J. Eilenberg, J. K. Pell, and R. S.
Hails, “Challenges in modelling complexity of fungal ento-
mopathogens in semi-natural populations of insects,” BioCon-
trol, vol. 55, no. 1, pp. 55–73, 2010.
[15] S. Keller, Arthropod-Pathogenic Entomophthorales: Biology,
Ecology, Identification, Office for Official Publications of the
European Communities, Brussels, Luxembourg, 2007.
[16] M. Barta and L. Cagáň, “Observations on the occurrence of
Entomophthorales infecting aphids (Aphidoidea) in Slovakia,”
BioControl, vol. 51, no. 6, pp. 795–808, 2006.
[17] J. L. Hatting, T. J. Poprawski, and R. M. Miller, “Prevalences of
fungal pathogens and other natural enemies of cereal aphids
(Homoptera: Aphididae) in wheat under dryland and irrigated
conditions in South Africa,” BioControl, vol. 45, no. 2, pp. 179–
199, 2000.
[18] A. C. Scorsetti, R. A. Humber, J. J. Garcı́a, and C. C. L. Lastra,
“Natural occurrence of entomopathogenic fungi (Zygomycetes:
Entomophthorales) of aphid (Hemiptera: Aphididae) pests of
horticultural crops in Argentina,” BioControl, vol. 52, no. 5, pp.
641–655, 2007.
[19] A. Y. Abdel-Mallek, M. A. A. Abdel-Rahman, S. A. Omar,
and G. H. A. Hamam, “A Comparative abundance of ento-
mopathogenic fungi of cereal Aphids in Assiut, Egypt,” in
Proceedings of the 2nd Saudi Science Conference, pp. 167–174,
Faculty of Science, King Abdulaziz University, 2004.
http://dx.doi.org/10.1155/2013/969750
Research Article
Effects of Systemic Pesticides Imidacloprid and Metalaxyl on
the Phyllosphere of Pepper Plants
Constantinos Moulas, Christos Petsoulas, Konstantina Rousidou, Chiara Perruchon,

Panagiotis Karas, and Dimitrios G. Karpouzas
University of Thessaly, Department of Biochemistry and Biotechnology, Ploutonos 26 and Aiolou Street, 41221 Larisa, Greece
Correspondence should be addressed to Dimitrios G. Karpouzas; dkarpouzas@bio.uth.gr
Received 18 April 2013; Accepted 22 May 2013
Copyright © 2013 Constantinos Moulas et al. This is an open access article distributed under the Creative Commons Attribution
cited.
Microbes inhabiting the phyllosphere of crops are exposed to pesticides applied either directly onto plant foliage or indirectly
through soil. Although, phyllosphere microbiology has been rapidly evolving, little is still known regarding the impact of pesticides
on the epiphytic microbial community and especially on fungi. We determined the impact of two systemic pesticides (metalaxyl and
imidacloprid), applied either on foliage or through soil, on the epiphytic fungal and bacterial communities via DGGE and cloning.
Both pesticides induced mild effects on the fungal and the bacterial communities. The only exception was the foliage application
of imidacloprid which showed a more prominent effect on the fungal community. Cloning showed that the fungal community
was dominated by putative plant pathogenic ascomycetes (Erysiphaceae and Cladosporium), while a few basidiomycetes were also
present. The former ribotypes were not affected by pesticides application, while selected yeasts (Cryptococcus) were stimulated by
the application of imidacloprid suggesting a potential role in its degradation. A less diverse bacterial community was identified
in pepper plants. Metalaxyl stimulated an Enterobacteriaceae clone which is an indication of the involvement of members of this
family in fungicide degradation. Further studies will focus on the isolation of epiphytic microbes which appear to be stimulated by
pesticides application.
1. Introduction In a pioneering study, Yang et al. [7] demonstrated that

the microbial diversity on plant phyllosphere is much higher
Phyllosphere is the habitat of a diverse microbial community than what had been estimated before based on culture-
dominated by bacteria, fungi, and yeasts while archaea are not dependent methods but it is still lower than the microbial
particularly abundant [1]. Until recently most studies on the diversity in rhizosphere or even bulk soil [8]. Phyllosphere
microbiology of the phyllosphere had focused on the ecology is an oligotrophic environment with patchy distribution of
and interactions of microbial plant pathogens with the plant, C sources where microorganisms are exposed to stress con-
whereas little was known regarding the role and ecology of ditions including extreme exposure to UV radiation, violent
nonplant pathogenic microorganisms on plant phyllosphere. fluctuations of temperature, and limited water availability
It is now well documented that epiphytic microorganisms [9]. In order to survive under these conditions, phyllo-
could serve significant functional roles including (a) suppres- sphere microorganisms have developed various mechanisms
sion of plant pathogens in the phyllosphere of agricultural including pigmentation [10], DNA repair mechanisms [11],
crops [2], (b) nitrogen fixation [3], (c) methanol utilization production of biosurfactants [12], and extracellular polymeric
[4], and (d) degradation of organic pollutants [5]. In addi- substances [13].
tion, microbial interactions on plant phyllosphere have been Apart from the abiotic and biotic stress conditions
found to determine colonization of edible parts of plants by described, microorganisms on the phyllosphere of cultivated
human pathogens [6]. This is particularly important for the plants are exposed to pesticides. There is a wealth of literature
consumption of fresh salad, fruits, and vegetables. regarding the impact of pesticides on soil microorganisms
[14]. This is not surprising considering that soil constitutes of a natural phyllosphere microbial community as much as
largely the final deposit of both foliar and soil-applied possible. During the acclimation period the pepper plants
pesticides. However, only a few studies so far have addressed were watered and fertilized as needed.
the impact of pesticides onto nontarget microorganisms on At the end of the acclimation period the 18 pots with
plant phyllosphere. A series of studies by Zhang et al. [15–17] the pepper plants were divided into 6 groups of three. The
showed that the insecticides cypermethrin and abamectin first three pots received a foliage treatment with an aqueous
induced changes in the structure of the bacterial commu- suspension of the insecticide imidachloprid (CONFIDOR,
nity in pepper, cucumber, and broccoli phyllosphere. In a 200SL), while the next three pots received a soil drenching
similar study the application of the fungicide enostroburin with an aqueous suspension of the same insecticide. Similarly,
induced substantial changes in bacterial community in wheat the next two groups of pots received a foliar or a soil
phyllosphere [18]. All those studies have focused on poten- drenching application of the fungicide metalaxyl (RIDOMIL
tial effects on the bacterial community after foliar applica- GOLD, 46.52SL). The application rates in both foliar and soil
tion of pesticides. However, only limited data are available applications were as suggested for the control of the target
regarding pesticides effects on nontarget fungi inhabiting pests and diseases. Finally, the remaining two groups of pots
plant phyllosphere. Apart from foliar application, systemic received the same volume of water applied through foliar or
pesticides are commonly applied via soil drenching and soil application of the two pesticides to serve as untreated
they are translocated through the phloem to the aerial parts controls. Five days after pesticide applications 10 leaves per
of the plants offering protection from pest and pathogens. plant were collected and placed into sterile plastic bags and
However, nothing is known regarding the impact of such soil transported on ice to the laboratory where they were stored
applications on the epiphytic microbial community. at −20∘ C until further used.
Imidacloprid is a systemic insecticide which has gained
registration for 140 uses in 120 countries [19]. It is applied
2.2. DNA Extraction. The microbial DNA of the phyllosphere
either directly on the foliage or via soil drenching for
was extracted as described by Yang et al. [7] with slight modi-
the control of aphids (Myzus persicae, Myzus nicotianae),
fications. Briefly, leaf samples were transferred aseptically into
white fly (Trialeurodes vaporariorum), and Colorado beetle
polypropylene tubes containing 0.1 M potassium phosphate
(Leptinotarsa decemlineata) in fruits crops, vegetables, and
buffer (pH 7.0) and sonicated for 10 min in an ultrasonic
potatoes. Metalaxyl-M is a systemic fungicide which is
bath to dislodge microorganisms from the leaf surface. The
applied either on foliage or via soil drenching for the control
leaf remains were removed by a mild centrifugation step
of oomycetes such as Phytophthora parasitica (tobacco),
(3 min 500 × g) and the clear suspension was subjected to
Phytopthora infestans (potato), and Pythium sp. [20]. Nothing
centrifugation at 7000 × g for 15 min. The supernatant was
is known regarding the impact of those pesticides on the
removed and the microbial pellet obtained was used for DNA
microbial community of plant phyllosphere.
extraction using the NucleoSpin Tissue kit (Macherey-Nagel,
We aimed to investigate the impact of the systemic
Germany) according to manufacturers’ extraction.
pesticides imidacloprid and metalaxyl on the fungal and
bacterial communities on the phyllosphere of pepper plants.
The influence of the mode of application, foliar versus 2.3. PCR-DGGE Analysis. For studying the bacterial com-
soil application, on the magnitude and type of effects was munity, a nested PCR amplification of the 16S rRNA gene
also determined via denaturing gradient gel electrophoresis was used. In the first PCR round, DNA was amplified
(DGGE) and cloning. with universal bacterial primers 63f-1087r (ca. 1000 bp). The
product obtained (1 mL) was nested with primers 357f+GC
and 534r which amplify a 194 bp fragment of the 16S rRNA
2. Materials and Methods gene, including the variable V3 region. A 40 bp GC clamp
at the 5 end of primer 357f was used [22]. For studying the
2.1. Experimental Setup. Three-week-old pepper plants (Cap- fungal community, DNA was amplified with primers ITS1F-
sicum annum L. cv Ozho) (kindly supplied by AgriPlant ITS4 (ca. 600 bp) [23]. The products obtained were used as
A.S.) were initially transplanted into 3 L plastic pots which templates (1 mL) for a second semi nested PCR with the
had been filled with appropriate amounts of a 1 : 1 mixture primers ITS1F + GC and ITS2 (ca. 300 bp). Thermocycling
of sand and soil (sandy, pH 7.81, electrical conductivity conditions and the concentrations of the reagents used were
0.017 mmhos cm−1 , organic matter content 8.8 g kg−1 , P- as described elsewhere [24].
Olsen 2 mg g−1 , K 215 mg g−1 , and Mg 265 mg g−1 ). In total DGGE analyses were carried out on an INGENYphorU-
18 pots were prepared and placed randomly in the growth 2x2 system (Ingeny International BV, The Netherlands).
chamber at 22∘ C using a 16 h light/8 h night period. The plants Polyacrylamide gels (8%) in 1 × TAE buffer (40 mM Tris base,
were watered as needed and 30 mL of Hoagland solution 20 mM acetic acid, and 1 mM disodium EDTA, pH 8.2) were
[21] was applied twice weekly. Pepper plants were grown prepared. The polyacrylamide gels were made with denatu-
under these conditions for a week and then transferred to rating gradient of 30–55% and 50–60% for DGGE profiling
a commercial greenhouse situated in the area of Velestino, of the fungal and bacterial communities, respectively (where
Magnesia, Greece, where the experiment was contacted. The 100% denaturant contains 7 M urea and 40% formamide). The
plants were left in the greenhouse for a period of three electrophoresis was run for 16 h at 60∘ C and 75 V and gels
weeks to become acclimatized and allow the development were silver stained. The image was captured using a digital
camera and subsequent analysis was performed with Cross cluster was composed only by the samples which received a
Checker 2.9 v (Wageningen University, The Netherlands). soil application of imidacloprid and showed >70% similarity
Binary data for the presence/absence of bands in all samples with the samples of the first cluster. In the first cluster,
were derived and used for statistical analysis. samples were further separated according to the pesticide
applied with soil-treated samples of imidacloprid separated
2.4. Clone Libraries. Clone libraries for both communities from the metalaxyl-treated samples and the controls which
were constructed based on the fragments generated by the grouped together. Within the latter subcluster the foliage-
first PCR step. Since the results showed that replicate samples treated metalaxyl samples and the control samples clustered
of the same treatment showed minimum variability, the together (>90% similarity) while the soil-treated metalaxyl
triplicate PCR products from the same treatment were pooled samples were separated.
and purified/concentrated to a final volume of 30 𝜇L using Clone libraries were developed to identify the main
the NucleoSpin II PCR clean-up kit (Macherey-Nagel GmbH, members of the fungal community and fungi which were
Germany). Cloning into the pGEM-T vector (Promega, responsive to pesticide applications. Overall the phyllosphere
Madison, USA) was performed as described by Sambrook et was dominated by ascomycetes while a few basidiomycetous
al. [25]. Subsequent screening of the clone libraries by PCR yeasts were also present and were represented by bands
and DGGE was carried out as described by Liang et al. [26]. appearing mostly in the upper part of the gel belonging to the
Thirty-five white colonies were selected for each treatment orders Sporidiobolales (bands 17 and 18), Cystofilobasidiales
and were subjected to colony PCR using primers 357f+GC- (band 19), and Filobasidiales (bands 13 and 1) (Table 1).
534r and ITS1F+GC-ITS2 for bacterial and fungal libraries, Generally the lower part of the DGGE patterns in all treat-
respectively. Positive clones were screened on a DGGE gel to ments, which constitutes the high GC content region, was
determine their electrophoretic mobility compared with the dominated by ascomycetes of the order Erysiphales (bands
band pattern of the original environmental sample. Repre- 2, 3, 4, 5, 10, 11, and 12). Band 15 was the most dominant
sentative clones for each band type matching the migration band in all treatments and sequencing analysis of the clones
pattern of bands in the original samples were sequenced. showed highest homology (99.6%) to a Cladosporium allii
In cases where several clones showed identical migration strain (Table 1).
pattern with a single DGGE band, three clones or more A few members of the fungal community were responsive
were sequenced in order to check for possible comigration to pesticide applications. Therefore, bands 1 and 13 were
of diverse sequences. For sequencing, plasmid DNA was present only in the samples treated with imidacloprid, espe-
extracted and purified from selected colonies using the cially foliage-treated samples (Figure 1). Clones associated
NucleoSpin Plasmid kit (Macherey-Nagel GmbH, Germany) with those bands showed highest sequence homology to a
and sent for sequencing. Sequences were deposited in the Cryptococcus adeliensis (97.1%) and an uncultured Crypto-
European Molecular Biology Laboratory (EMBL) database coccus clone (100%), respectively. On the contrary, band 3
and their accession numbers are HF947094-HF947095 and disappeared from the samples which received a foliage appli-
HF947030-HF947093 for bacterial and fungal clones, respec- cation of imidacloprid. The single clone associated with this
tively. band showed highest sequence homology to an uncultured
unclassified fungal clone (99.4%).
Regarding metalaxyl-treated samples, band 21 was only
2.5. Statistical Analysis. The binary data matrices obtained present in the samples treated with the fungicide regardless
for each DGGE gel were used for multivariate statistical of the application mode. Clones associated with this band
analysis to compare the effect of pesticides and their mode of showed highest sequence homology to an uncultured Saccha-
application on the structure of the microbial communities on romyceta clone (99.6%). In addition, band 9 was stimulated
phyllosphere. Dendrograms from Jaccard distance matrices in the metalaxyl-treated samples. Clones associated with
using the group average algorithm were prepared using this band showed highest sequence homology to a Periconia
the MultiVariate Statistical Package (MVSP) 3.13v software macrospinosa strain (99.8%).
(http://www.kovcomp.com/).
3. Results 3.2. Effects of Pesticides on the Bacterial Community. DGGE

analysis of the bacterial community showed a less complex
3.1. Effects of Pesticides on the Fungal Community. The banding pattern, compared to the fungal community finger-
fingerprints produced by the three replicates of the same prints, with band numbers not exceeding 15 in any of the
treatment were highly similar (Figure 1). Overall, DGGE treatments (Figure 3). Cluster analysis separated samples into
analysis of the fungal community provided rather complex two main clusters. The first cluster contained all samples
banding patterns with band numbers exceeding 20 in all except one of the replicates of the metalaxyl soil application
treatments. Cluster analysis of the DGGE banding patterns (Figure 4). Within the first cluster samples shared more than
resulted in the development of two main clusters (Figure 2). 90% similarity and were further separated into two sub-
The first main cluster comprised all control and metalaxyl- clusters: the first one comprising all soil controls and the
treated samples along with the samples which received a two replicates of the metalaxyl soil application, while the
soil application of imidacloprid. Samples contained within second subcluster contained all other control, imidacloprid-
the first cluster shared a similarity of >84%. The second and metalaxyl-treated samples.
Control Imidacloprid Metalaxyl

Foliage Soil Foliage Soil Foliage Soil
M M M M
19
18
17
16
15 20
21 14
13
6
7
9 8
12
10 11
5
4
3
2
Figure 1: DGGE analysis of the fungal community in the phyllosphere of pepper plants subjected to foliage or soil applications of
imidachloprid, metalaxyl, or water (untreated control). Lanes designated with M correspond to a marker which contained 20 ng 𝜇L−1 of
the ITS-PCR products of the following fungi with the sequence they appear on the gel from top to bottom: Pleurotus djamor, Fusarium
oxysporum f.sp. radicis-lycopersici, F. solani, P. eryngii, P. ostreatus, and P. cystidiosus. Bands identified through screening with clone libraries
are designated with arrows accompanied with a code number as shown in Table 1.
Control F (100%). Similarly, band 1 was present only in the control

Control F samples and the clone associated with this band showed
Control S the highest sequence homology with a Propionibacterium sp.
Control S
strain (100%).
Control F
Control S
Metalaxyl F 4. Discussion
Metalaxyl F
Metalaxyl F Microorganisms inhabiting the aerial parts of crops are
Metalaxyl S
commonly exposed to pesticides either directly through
Metalaxyl S
Metalaxyl S foliage applications or indirectly through soil application. The
Imidacloprid S latter is valid only for systemic pesticides. Recent studies
Imidacloprid S provided clear evidence for the existence of a previously
Imidacloprid S unknown microbial diversity on the plant phyllosphere and
Imidacloprid F the involvement of epiphytic microbes on a wide array of
Imidacloprid F important services [1, 27]. Little is known regarding the
Imidacloprid F
interactions of pesticides and epiphytic microorganisms. We
1 0.95 0.9 0.85 0.8 0.75 investigated the impact of two systematic pesticides and their
Jaccard similarity index mode of application (soil versus foliage) on the community
structure of fungi and bacteria on the phyllosphere of pepper
Figure 2: Cluster analysis (group average, Jaccard similarity index) plants.
of the banding patterns generated by DGGE fingerprinting analysis Overall, DGGE analysis revealed a well-established fun-
of the fungal community. gal community and a less complex bacterial community.
On the one hand, this is against the general perception
that bacteria constitute the main microbial group on plant
Clone libraries developed aimed to identify the main phyllosphere [8]. On the other hand, this finding is not
members of the bacterial community and bacteria responsive surprising considering that pepper plants were grown under
to pesticide applications. Fingerprints of all samples were greenhouse conditions which favour the rapid establishment
dominated by band 2. Thus, 15 clones showing identical elec- of fungal pathogens on aerial plant parts. In line with this
trophoretic mobility with this band were sequenced. Results are the results of the clone libraries which verified that the
showed that this band was of plant origin since all clones fungal community on the phyllosphere of pepper plants was
showed highest sequence homology to chloroplast sequence dominated by ascomycetes with the majority of them being
of Capsicum annum and were excluded from further analysis. putative plant pathogens. Recent reports by Jumpponen and
Band B5 was unique in the samples which received a foliage Jones [27] also found a massive dominance of ascomycetes
application of metalaxyl and the clone sequenced showed the over other fungal phyla on the phyllosphere of temperate
highest sequence homology to an Enterobacteriaceae clone Quercus macrocarpa. Several fungal ribotypes belonged to the
Control Imidacloprid Metalaxyl

Foliage Soil Foliage Soil Foliage Soil
M M M M
B5
B1
B2 B2
B2
Figure 3: DGGE analysis of the bacterial community in the phyllosphere of pepper plants subjected to foliage or soil applications of
imidachloprid, metalaxyl, or water (untreated control). Lanes designated with M correspond to a marker which contained 20 ng 𝜇L−1 of the
16S rRNA-PCR products of each of the following bacteria appearing on the gel from top to bottom: Pseudomonas aeruginosa, Pseudomonas
sp., P. putida, Flavobacterium sp., Stenotrophomonas maltophilia, Xanthomonas sp., and Agrobacterium sp.
Table 1: Identity of selected DGGE bands from clones obtained for the fungal community on the pepper phyllosphere. The numbers following
% sequence similarity provide the numbers of clones showing highest similarity to this specific fungal ribotype.
Band no. Clones sequenced Closest match from GenBank (% sequence similarity by BLAST) GenBank Acc. no.
1 1 Cryptococcus adeliensis strain ZIM600 (97.1%) FN400760
Golovinomyces cichoracearum isolate (99.6%)—5 AB077656
2 6
Podosphaera fusca isolate (100%)—1 AB525915
3 1 Uncultured unclassified fungal clone (99.4%) AY843157
Golovinomyces cichoracearum isolate UMSG1 (99.8%)—2 HM449077
4 4
Uncultured Sordariales clone 9A6S46N (94%)—2 HQ389517
Erysiphe cichoracearum (99.6%)—2 AF031282
5 7
Lewia infectoria (99.8%)—5 AY154692
6 2 Alternaria alternata voucher TC0811057 (100%) HM013816
7 1 Stemphylium sp. FA-8J (100%) JX164072
Pyrenophora avenae isolate 94-1b (99.8%)—2 EF452453
8 3
Podospora communis strain NZ206 (99.5%)—1 EU621831
9 2 Periconia macrospinosa strain KS00113 (99.8%) FJ536208
Neoerysiphe galeopsidis (99.8%)—3 AB498946
10 4
Uncultured fungus clone (99.8%)—1 JF289165
11 2 Erysiphe betae isolate EB1-1 (99.4%) DQ164432
12 4 Erysiphe cruciferarum (99.7%) EU140958
13 2 Uncultured Cryptococcus clone (100%) JF432980
14 3 Cladosporium cladosporioides strain M61 (99.6%) JQ936096
15 9 Cladosporium allii strain CBS 101.81 (99.6%) JN906977
16 2 Unclassified Pleosporales isolate (100%) FN548155
17 4 Sporidiobolus sp. isolate FA-8H (99.8%) JX164071
18 1 Sporobolomyces roseus strain IWBT-Y851 (99.8%) JQ993392
Itersonilia perplexans strain JCM 10245 (99.8%)—1 AB072233
19 2
Uncultured fungus clone (99.8%)—1 AB520396
20 2 Dioszegia hungarica strain JCM 9046 (99.8%) AB049614
21 2 Uncultured Saccharomyceta clone (99.6%) HQ211757
Control F the stimulation of Cryptococcus yeasts on the phyllosphere

Control F of pepper plants could be attributed to their involvement
Control F in the degradation of imidacloprid. Further studies will
Metalaxyl S
Metalaxyl S
aim to isolate such strains and evaluate their degrading
Control S capacity against imidacloprid. Regarding metalaxyl, its appli-
Control S cation, either through soil or foliage, stimulated a Periconia
Control S macrospinosa and an uncultured ribotype belonging to the
Imidacloprid F unranked taxon of Saccharomyceta. Periconia macrospinosa
Imidacloprid F has been identified as a common plant endophyte in several
Imidacloprid F studies [38, 39].
Imidacloprid S
Overall, pesticide application induced only minor
Imidacloprid S
Imidacloprid S
changes in the structure of the bacterial community with
Metalaxyl F the exception of one of the metalaxyl soil-treated samples
Metalaxyl F which diverged from all the other samples. Previous studies
Metalaxyl F by Gu et al. [18] showed that the application of the fungicide
Metalaxyl S enostroburin induced substantial changes in the bacterial
1 0.95 0.9 0.85 0.8 0.75 community. A few members of the bacterial community were
Jaccard similarity index responsive to pesticide application. Thus foliage application
of metalaxyl stimulated the appearance of an Enterobac-
Figure 4: Cluster analysis (group average, Jaccard similarity index) teriaceae ribotype. This family includes several pathogens
of the banding patterns generated by DGGE fingerprinting analysis found in fresh salads destined for human consumption such
of the bacterial community. as Salmonella, Escherichia, Yersinia [6]; plant pathogens like
Erwinia [40]; and human pathogens like Klebsiella [41]. The
stimulation observed could be related to the involvement
of this Enterobacteriaceae ribotype in the degradation of
family of Erysiphaceae whose members are the causal agents metalaxyl; however further studies are required to confirm
of powdery mildew in different crops like sugar beet, beetroot this. In a similar study Gu et al. [18] suggested a putative
(Erysiphe betae) [28], and brassicas (Erysiphe cruciferarum) role of another Enterobacteriaceae ribotype (Pantoea sp.)
[29]. Among Erysiphaceae the most common ribotype was in the degradation of the fungicide enostroburin in the
identified as Golovinomyces cichoracearum which is the causal phyllosphere of wheat. Members of Enterobacteriaceae
agent of powdery mildew in pumpkin, cucumber, and melon belonging to the genus Enterobacter or Klebsiella have shown
crops, while it has been found as an epiphytic fungus in enhanced degrading capacities against the organophosphate
several other plants [30]. The dominant ribotype in the chlorpyrifos [42, 43]. In contrast, pesticide application
phyllosphere of pepper plants was identified as Cladosporium appeared to suppress a Propionibacterium ribotype which
allii, which is the causal agent of leaf blotch in leek [31]. was present in the untreated plants. Bacteria of this genus
Cladosporium species have been identified as very common were previously identified as members of the phyllosphere
inhabitants of plant phyllosphere [32]. All the above putative community in Magnolia grandiflora plants [44] and as
plant pathogenic fungi were not affected by the application of endophytes in potatoes [45].
the two pesticides. Generally, the DGGE screening of clone libraries followed
Generally, pesticides induced rather subtle changes in the in our study revealed the detection of two or more different
structure of the fungal community, with foliage application of ribotypes represented by the same DGGE band in one-third
imidacloprid inducing the most prominent changes. Little is of the bands analyzed via cloning. This has been identified
known regarding the impact of insecticides on the microbial as an inherent problem of the DGGE fingerprinting method
phyllosphere. A series of studies by Zhang et al. [15–17] and it is commonly observed when complex microbial
showed limited effects of cypermethrin and abamectin on the communities are studied [46, 47]. Further analysis of the
fungal biomass, whereas potential effects on the structure of clones obtained by the bacterial community showed that the
the fungal community were not provided. The limited impact major band of the fingerprint was related to the chloroplasts
of metalaxyl on the fungal community could be attributed to of pepper plants. Irrespective of the method employed to
its selectivity against oomycetes [33], which were absent in the isolate DNA from epiphytic microbial communities, the
phyllosphere of pepper plants in our study. Previous in vitro extracted microbial DNA is generally contaminated with
and leaf assays showed that broad-spectrum fungicides like chloroplasts [48, 49]. Since the chloroplast 16S shares high
metiram and captan had a detrimental effect on the growth sequence similarity with bacterial 16S rRNA sequences [50],
of epiphytic bacteria, fungi, and yeasts with the latter being contamination with plant DNA poses a serious challenge for
the most sensitive group of microbes [34]. the application of PCR-based methods to profile and quantify
Foliage application of imidachloprid stimulated a Cryp- bacterial populations in plant environments. In a similar
tococcus adeliensis and an uncultured Cryptococcus ribotype. study Hunter et al. [51], using the same primer pair as in our
Cryptococcus yeasts are common epiphytes [32, 35] and study, also detected several sequences of chloroplast origin in
members of this genus are known to effectively transform their clone libraries of the phyllosphere bacterial community
phenolics [36] and benzene [37]. Thus it is probable that in lettuce.
5. Conclusions [12] L. Bunster, H. J. Fokkema, and B. Schippers, “Effect of surface

activity of Pseudomonas sp. On leaf wettability,” Applied and
Our study provides first evidence that the application of Environmental Microbiology, vol. 55, pp. 1340–1345, 1989.
pesticides, either directly on foliage or through soil, induced [13] L. E. B. Baldotto and F. L. Olivares, “Phylloepiphytic interaction
rather mild changes on the structure of the fungal and between bacteria and different plant species in a tropical
bacterial communities. Further studies will aim to isolate agricultural system,” Canadian Journal of Microbiology, vol. 54,
and characterize fungal strains which were stimulated upon no. 11, pp. 918–931, 2008.
pesticide application and identify their true ecological role [14] G. Imfeld and S. Vuilleumier, “Measuring the effects of pes-
and interactions with the pesticides studied. ticides on bacterial communities in soil: a critical review,”
European Journal of Soil Biology, vol. 49, pp. 22–30, 2012.
Acknowledgment [15] B. Zhang, H. Zhang, B. Jin et al., “Effect of cypermethrin insec-
ticide on the microbial community in cucumber phyllosphere,”
This study was funded by resources of the M.S. course Journal of Environmental Sciences, vol. 20, no. 11, pp. 1356–1362,
“Biotechnology—Quality of Nutrition and Environment” of 2008.
the Department of Biochemistry and Biotechnology, Univer- [16] B. Zhang, Z. Bai, D. Hoefel et al., “The impacts of cypermethrin
sity of Thessaly. pesticide application on the non-target microbial community of
the pepper plant phyllosphere,” Science of the Total Environment,
vol. 407, no. 6, pp. 1915–1922, 2009.
References [17] B.-G. Zhang, L. Tang, Z.-M. Li et al., “Effect of abamectin insec-
[1] C. Knief, N. Delmotte, S. Chaffron et al., “Metaproteogenomic ticide on the microbial community in broccoli phyllosphere,”
analysis of microbial communities in the phyllosphere and Environmental Science, vol. 30, no. 5, pp. 1292–1297, 2009.
rhizosphere of rice,” ISME Journal, vol. 6, pp. 1378–1390, 2012. [18] L. Gu, Z. Bai, B. Jin et al., “Assessing the impact of fungicide
[2] J. M. Whipps, P. Hand, D. Pink, and G. D. Bending, “Phyl- enostroburin application on bacterial community in wheat
losphere microbiology with special reference to diversity and phyllosphere,” Journal of Environmental Sciences, vol. 22, no. 1,
plant genotype,” Journal of Applied Microbiology, vol. 105, no. 6, pp. 134–141, 2010.
pp. 1744–1755, 2008. [19] A. Elbert, M. Haas, B. Springer, W. Thielert, and R. Nauen,
[3] T. Miyamoto, M. Kawahara, and K. Minamisawa, “Novel endo- “Applied aspects of neonicotinoid uses in crop protection,” Pest
phytic nitrogen-fixing clostridia from the grass Miscanthus Management Science, vol. 64, no. 11, pp. 1099–1105, 2008.
sinensis as revealed by terminal restriction fragment length [20] A. Evenhuis, H. G. Spits, and H. T. A. M. Schepers, “Efficacy of
polymorphism analysis,” Applied and Environmental Microbiol- fungicidal protection of newly developing potato leaves against
ogy, vol. 70, no. 11, pp. 6580–6586, 2004. Phytophthora infestans,” Crop Protection, vol. 25, no. 6, pp. 562–
[4] N. Delmotte, C. Knief, S. Chaffron et al., “Community proteoge- 568, 2006.
nomics reveals insights into the physiology of phyllosphere [21] D. R. Hoagland and D. I. Arnon, “The water-culture method for
bacteria,” Proceedings of the National Academy of Sciences of the growing plants without soil,” California Agricultural Experimen-
United States of America, vol. 106, no. 38, pp. 16428–16433, 2009. tal Station, vol. 347, pp. 1–39, 1938.
[5] J. Ning, Z. Bai, G. Gang et al., “Functional assembly of
[22] G. Muyzer, E. C. de Waal, and A. G. Uitterlinden, “Profiling
bacterial communities with activity for the biodegradation of
of complex microbial populations by denaturing gradient gel
an organophosphorus pesticide in the rape phyllosphere,” FEMS
electrophoresis analysis of polymerase chain reaction-amplified
Microbiology Letters, vol. 306, no. 2, pp. 135–143, 2010.
genes coding for 16S rRNA,” Applied and Environmental Micro-
[6] J. M. Whipps, P. Hand, D. A. C. Pink, and G. D. Bending, biology, vol. 59, no. 3, pp. 695–700, 1993.
“Human pathogens and the phyllosphere,” Advances in Applied
Microbiology, vol. 64, pp. 183–221, 2008. [23] M. Gardes and T. D. Bruns, “ITS primers with enhanced
specificity for basidiomycetes–application to the identification
[7] C.-H. Yang, D. E. Crowley, J. Borneman, and N. T. Keen,
of mycorrhizae and rusts,” Molecular Ecology, vol. 2, no. 2, pp.
“Microbial phyllosphere populations are more complex than
113–118, 1993.
previously realized,” Proceedings of the National Academy of
Sciences of the United States of America, vol. 98, no. 7, pp. 3889– [24] C. Rousidou, K. K. Papadopoulou, G. Zervakis, B. K. Singh, C.
3894, 2001. Ehaliotis, and D. G. Karpouzas, “Fractioned land application of
[8] J. A. Vorholt, “Microbial life in the phyllosphere,” Nature diluted olive mill wastewater induces changes in the structure of
Reviews Microbiology, vol. 10, pp. 828–840, 2012. the soil microbial community,” European Journal of Soil Biology,
vol. 46, no. 1, pp. 34–40, 2010.
[9] S. E. Lindow and M. T. Brandl, “Microbiology of the phyllo-
sphere,” Applied and Environmental Microbiology, vol. 69, no. 4, [25] J. Sambrook, E. F. Fritsch, and T. Maniatis, Molecular Cloning:
pp. 1875–1883, 2003. a Laboratory Manual, Cold Spring Harbor Laboratory, Cold
[10] N. J. Fokkema and B. Schippers, “Phyllosphere vs rhizosphere as Spring Harbor, NY, USA, 2nd edition, 1989.
environments for saprotrophic colonization,” in Microbiology of [26] Z. Liang, R. A. Drijber, D. J. Lee, I. M. Dwiekat, S. D. Harris, and
the Phyllosphere, N. J. Fokkema and J. Van den Heuvel, Eds., pp. D. A. Wedin, “A DGGE-cloning method to characterize arbus-
137–151, Cambridge University Press, London, UK, 1986. cular mycorrhizal community structure in soil,” Soil Biology and
[11] J. J. Kim and G. W. Sundin, “Regulation of the rulAB mutagenic Biochemistry, vol. 40, no. 4, pp. 956–966, 2008.
DNA repair operon of Pseudomonas syringae by UV-B (290 [27] A. Jumpponen and K. L. Jones, “Massively parallel 454 sequenc-
to 320 nanometers) radiation and analysis of rulAB-mediated ing indicates hyperdiverse fungal communities in temperate
mutability in vitro and in planta,” Journal of Bacteriology, vol. Quercus macrocarpa phyllosphere,” New Phytologist, vol. 184,
182, no. 21, pp. 6137–6144, 2000. no. 2, pp. 438–448, 2009.
[28] M. Fernández-Aparicio, E. Prats, A. A. Emeran, and D. Rubiales, sample of waste water treatment plant in damascus,” Folia
“Characterization of resistance mechanisms to powdery mildew Microbiologica, vol. 52, no. 4, pp. 423–427, 2007.
(Erysiphe betae) in beet (Beta vulgaris),” Phytopathology, vol. 99, [44] C. R. Jackson and W. C. Denney, “Annual and seasonal variation
no. 4, pp. 385–389, 2009. in the phyllosphere bacterial community associated with leaves
[29] F. Mert-Türk, M. K. Gül, and C. Ö. Egesel, “Nitrogen and fungi- of the southern Magnolia (Magnolia grandiflora),” Microbial
cide applications against Erysiphe cruciferarum affect quality Ecology, vol. 61, no. 1, pp. 113–122, 2011.
components of oilseed rape,” Mycopathologia, vol. 165, no. 1, pp. [45] F. Rasche, T. Lueders, M. Schloter et al., “DNA-based stable
27–35, 2008. isotope probing enables the identification of active bacterial
[30] Á. M. R. Almeida, E. Binneck, F. F. Piuga, S. R. R. Marin, P. endophytes in potatoes,” New Phytologist, vol. 181, no. 4, pp.
R. Z. Ribeiro Do Valle, and C. A. Silveira, “Characterization 802–807, 2009.
of powdery mildews strains from soybean, bean, sunflower, [46] T. Vallaeys, E. Topp, G. Muyzer et al., “Evaluation of denaturing
and weeds in Brazil using rDNA-ITS sequences,” Tropical Plant gradient gel electrophoresis in the detection of 16S rDNA
Pathology, vol. 33, no. 1, pp. 20–26, 2008. sequence variation in rhizobia and methanotrophs,” FEMS
[31] M. M. Jordan, R. T. Burchill, and R. B. Maude, “Epidemiology Microbiology Ecology, vol. 24, no. 3, pp. 279–285, 1997.
of Cladosporium allii and Cladosporium allii-cepae, leaf blotch [47] C. R. Jackson, E. E. Roden, and E. E. Churchill, “Denaturing
pathogens of leek and onion. II. Infection of host plants,” Annals gradient gel electrophoresis can fail to separate 16S rDNA
of Applied Biology, vol. 117, no. 2, pp. 327–336, 1990. fragments with multiple base differences,” Molecular Biology
[32] Y.-H. He, S. Isono, M. Shibuya et al., “Oligo-DNA cus- Today, vol. 1, pp. 49–51, 2000.
tom macroarray for monitoring major pathogenic and non- [48] L. Sun, F. Qiu, X. Zhang, X. Dai, X. Dong, and W. Song,
pathogenic fungi and bacteria in the phyllosphere of apple “Endophytic bacterial diversity in rice (Oryza sativa L.) roots
trees,” PLoS ONE, vol. 7, no. 3, Article ID e34249, 2012. estimated by 16S rDNA sequence analysis,” Microbial Ecology,
[33] A. Kerkenaar and A. K. Sijpesteijn, “Antifungal activity of vol. 55, no. 3, pp. 415–424, 2008.
metalaxyl and furalaxyl,” Pesticide Biochemistry and Physiology,
[49] G. Rastogi, J. J. Tech, G. L. Coaker, and J. H. J. Leveau, “A
vol. 15, no. 1, pp. 71–78, 1981.
PCR-based toolbox for the culture-independent quantification
[34] M. Walter, C. M. Frampton, K. S. H. Boyd-Wilson, P. Harris- of total bacterial abundances in plant environments,” Journal of
Virgin, and N. W. Waipara, “Agrichemical impact on growth Microbiological Methods, vol. 83, no. 2, pp. 127–132, 2010.
and survival of non-target apple phyllosphere microorganisms,”
[50] M. Sakai, A. Matsuka, T. Komura, and S. Kanazawa, “Appli-
Canadian Journal of Microbiology, vol. 53, no. 1, pp. 45–55, 2007.
cation of a new PCR primer for terminal restriction fragment
[35] N. Čadež, J. Zupan, and P. Raspor, “The effect of fungicides on length polymorphism analysis of the bacterial communities in
yeast communities associated with grape berries,” FEMS Yeast plant roots,” Journal of Microbiological Methods, vol. 59, no. 1,
Research, vol. 10, no. 5, pp. 619–630, 2010. pp. 81–89, 2004.
[36] À. Fonseca, G. Scorzetti, and J. W. Fell, “Diversity in the
[51] P. J. Hunter, P. Hand, D. Pink, J. M. Whipps, and G. D. Bending,
yeast Cryptococcus albidus and related species as revealed
“Both leaf properties and microbe-microbe interactions influ-
by ribosomal DNA sequence analysis,” Canadian Journal of
ence within-species variation in bacterial population diversity
and structure in the lettuce (Lactuca species) phyllosphere,”
[37] W. J. Middelhoven, “Catabolism of benzene compounds by Applied and Environmental Microbiology, vol. 76, no. 24, pp.
ascomycetous and basidiomycetous yeasts and yeastlike fungi. 8117–8125, 2010.
A literature review and an experimental approach,” Antonie van
Leeuwenhoek, vol. 63, no. 2, pp. 125–144, 1993.
[38] K. Mandyam, T. Loughin, and A. Jumpponen, “Isolation and
morphological and metabolic characterization of common
endophytes in annually burned tallgrass prairie,” Mycologia, vol.
102, no. 4, pp. 813–821, 2010.
[39] Z.-L. Yuan, C.-L. Zhang, F.-C. Lin, and C. P. Kubicek, “Identity,
diversity, and molecular phylogeny of the endophytic myco-
biota in the roots of rare wild rice (Oryza granulate) from a
nature reserve in Yunnan, China,” Applied and Environmental
[40] H. H. El-Hendawy, M. E. Osman, and H. A. Ramadan, “Pectic
enzymes produced in vitro and in vivo by Erwinia spp. isolated
from carrot and pepper in Egypt,” Journal of Phytopathology,
vol. 150, no. 8-9, pp. 431–438, 2002.
[41] R. Podschun and U. Ullmann, “Klebsiella spp. as nosoco-
mial pathogens: epidemiology, taxonomy, typing methods, and
pathogenicity factors,” Clinical Microbiology Reviews, vol. 11, no.
4, pp. 589–603, 1998.
[42] B. K. Singh, A. Walker, J. A. W. Morgan, and D. J. Wright,
“Biodegradation of chlorpyrifos by Enterobacter strain B-14 and
its use in bioremediation of contaminated soils,” Applied and
[43] I. Ghanem, M. Orfi, and M. Shamma, “Biodegradation of
chlorpyrifos by klebsiella sp. isolated from an activated sludge
http://dx.doi.org/10.1155/2013/926985
Research Article
Combinatorial Mutagenesis and Selection to Understand and
Improve Yeast Promoters
Laila Berg,1 Trine Aakvik Strand,2 Svein Valla,1 and Trygve Brautaset2
1
Department of Biotechnology, Norwegian University of Science and Technology, Sem Sælands vei 6, 7491 Trondheim, Norway
2
Department of Molecular Biology, SINTEF Materials and Chemistry, Sem Sælands vei 2, 7465 Trondheim, Norway
Correspondence should be addressed to Trygve Brautaset; trygve.brautaset@sintef.no
Copyright © 2013 Laila Berg et al. This is an open access article distributed under the Creative Commons Attribution License,
Microbial promoters are important targets both for understanding the global gene expression and developing genetic tools for
heterologous expression of proteins and complex biosynthetic pathways. Previously, we have developed and used combinatorial
mutagenesis methods to analyse and improve bacterial expression systems. Here, we present for the first time an analogous strategy
for yeast. Our model promoter is the strong and inducible 𝑃𝐴𝑂𝑋1 promoter in methylotrophic Pichia pastoris. The Zeocin resistance
gene was applied as a valuable reporter for mutant 𝑃𝐴𝑂𝑋1 promoter activity, and we used an episomal plasmid vector to ensure a
constant reporter gene dosage in the yeast host cells. This novel design enabled direct selection for colonies of recombinant cells
with altered Zeocin tolerance levels originating solely from randomly introduced point mutations in the 𝑃𝐴𝑂𝑋1 promoter DNA
sequence. We demonstrate that this approach can be used to select for 𝑃𝐴𝑂𝑋1 promoter variants with abolished glucose repression
in large mutant libraries. We also selected 𝑃𝐴𝑂𝑋1 promoter variants with elevated expression level under induced conditions. The
properties of the selected 𝑃𝐴𝑂𝑋1 promoter variants were confirmed by expressing luciferase as an alternative reporter gene. The tools
developed here should be useful for effective screening, characterization, and improvement of any yeast promoters.
1. Introduction pWWO of Pseudomonas putida [1]. XylS belongs to the AraC-

XylS family of transcription factors, and XylS can upon bind-
Microorganisms are widely used in biotechnology as cell ing of effector molecules (alkylbenzoates) bind upstream of
factories for sustainable production of proteins, biopoly- Pm and activate transcription (for a review, see [2]). We have
mers, and chemicals for medical, pharmaceutical and indus- constructed expression vectors by combining xylS/Pm with
trial applications. Today, bioprospecting programs exploring the minimal replicon trfA/oriV of the natural broad host
microbial diversity around the world lead to the identification range plasmid RK2 and demonstrated industrial level pro-
of new species, new genes, and new products. As the majority duction of several medical recombinant proteins in Escher-
of natural microorganisms are not cultivable under labora- ichia coli under high-cell-density cultivations [3, 4]. By using
tory conditions, the application of robust microbial hosts for combinatorial mutagenesis and selection methods we have
heterologous expression of genes and gene clusters is crucial selected variants of the Pm promoter with improved expres-
both to understand microbial genetic diversity as well as to sion properties in E. coli, and we have also shown that such
develop new value-added products. Access to well-defined methods are highly useful to generate new basic knowledge
and specialized promoter systems is important to enable of bacterial gene expression [5–7]. The major principle
functional expression and concomitant verification of the of this approach was on one hand to generate large and
collected genes and pathways. We have previously developed mutant promoter sequence libraries and on the other hand to
combinatorial mutagenesis and selection methods to improve establish a direct selection method enabling efficient screen-
bacterial promoter systems for diverse purposes. In partic- ing of the libraries for promoter variants with the desired
ular, we have worked with the strong and flexible xylS/Pm improved properties. The selection system was based on
regulator/promoter system originating from TOL plasmid using the bla gene (encoding the ampicillin-inactivating
enzyme 𝛽-lactamase) as reporter placed under the transcrip- methanol to maintain the induction, for agar media this
tional control of the Pm promoter on a plasmid, enabling was obtained by adding 150 𝜇L (9 cm plates) or 500 𝜇L
direct selection of recombinant E. coli host cells with altered (14 cm plates) methanol to the lid of inverted plates. G418
ampicillin tolerance levels. (500 𝜇g/mL) was added as appropriate.
Besides bacteria, yeast is an important and sometimes
a preferred host for recombinant expression of genes and 2.2. Standard DNA Manipulations. A RbCl protocol (http://
pathways. The methylotrophic yeast Pichia pastoris has www.promega.com/) was used for transformation of E. coli.
become an extensively used expression host, and the strong Plasmids were isolated from E. coli by using WizardPlus SV
and inducible 𝑃𝐴𝑂𝑋1 promoter is one of the most frequently minipreps DNA purification kit (Promega) or Qiafilter maxi
applied promoters for gene expression in this organism [8– plasmid purification kit (Qiagen), and enzymatic manipu-
11]. Expression from the 𝑃𝐴𝑂𝑋1 promoter is tightly regulated lations were performed as described by the manufacturers.
by the present carbon source, involving strong repression in PCR reactions were performed using the Expand high-fidel-
the presence of glucose or glycerol and full induction by ity PCR system kit (Boehringer Mannheim) for cloning
methanol when the repressive carbon sources are absent [12, purposes and the PfuUltra II Fusion HS DNA polymer-
13]. Yeast promoters are in general longer and more complex ase (Strategene) for site-directed mutagenesis applications.
than bacterial promoters, and the understanding of the about Sequencing analyses were purchased from Eurofins MWG
950 base pairs long P. pastoris 𝑃𝐴𝑂𝑋1 promoter DNA region is operon (http://www.eurofinsdna.com/).
limited, both with regard to its cis-acting elements and the For transformation of yeast, in general 10 𝜇L of circular
molecular mechanisms underlying its regulation. In recent plasmid preparation was transformed to P. pastoris in accor-
years, several attempts have been made to understand and dance with the electroporation protocol from Invitrogen
improve the 𝑃𝐴𝑂𝑋1 promoter by using rational design and (http://www.Invitrogen.com/). Plasmids were isolated from
site-directed mutagenesis [14–16]. In spite of considerable P. pastoris by using the Easy Yeast Plasmid Isolation kit
efforts, the details about the cis-acting elements involved in (Clontech).
the glucose regulation of the 𝑃𝐴𝑂𝑋1 promoter and the under-
lying molecular mechanisms are still not known. Also, the
improvements of the induced expression level in these reports 2.3. Agar Plate Assay for Determination of the Maximal Zeocin
were moderate. This emphasizes the need for effective meth- Tolerance Levels of the P. pastoris Host Cells. An agar plate
ods to analyse and improve yeast promoter systems. assay was used to indirectly determine the yeast promoter
In the present study, we introduce a novel combinatorial activity as the maximal Zeocin tolerance levels of the P.
mutagenesis and selection strategy in yeast by using P. pastoris pastoris host cells. This is a modified version of the previously
and the 𝑃𝐴𝑂𝑋1 promoter as model system. We demonstrate for used agar plate assay developed for the indirect determi-
the first time that this approach also works in yeast and that it nation of the Pm promoter activity in E. coli as maximal
can be used both to alter the regulation properties as well as ampicillin resistance levels of the host cells [17]. In detail,
to improve the maximized expression level of the 𝑃𝐴𝑂𝑋1 pro- freshly transformed P. pastoris GS115 strains were inoculated
moter. The methodology presented here should be valuable into 100 𝜇L YPD broth in 96 micro-well plates and incubated
both to manipulate and improve yeast promoters, for pro- at 30∘ C with shaking overnight. Next, the cells were diluted
moter probe approaches, and for applied perspectives as well with a 96-pin replicator in 0.9% NaCl (100 𝜇L) and plated on
as to generate basic understanding of microbial promoters. YPD agar or minimal agar media with methanol containing
various Zeocin concentrations. Usually, the cells were also
plated on YPD agar containing G418. The plates were incu-
bated at 30∘ C for three days. The maximal Zeocin tolerance
2. Materials and Methods level determinations were carried out with minimum four
2.1. Microbial Strains and Growth Media. Strains and vectors technical recurrences.
used in this study are listed in Table 1, and primers used for
the vector constructions are listed in Table 2. E. coli cells were 2.4. Construction of Promoter Mutant Libraries. The mutant
generally grown at 37∘ C in L broth (1% tryptone, 0.5% yeast libraries were established and stored as plasmid libraries in E.
extract, and 0.5% NaCl) or on L agar (L broth with 2% agar). coli DH5𝛼, and circular plasmids were transformed to P. pas-
When Zeocin was used for selection, the pH in the L broth toris GS115 (see below). In total, two different mutant libraries
and agar was adjusted to 7.5. Ampicillin (100 𝜇g/mL), kana- were made in the 𝑃𝐴𝑂𝑋1 promoter with the 𝑍𝑐𝑅 gene as the
mycin (50 𝜇g/mL), and Zeocin (25 𝜇g/mL) were added as promoter reporter. The presence of both a kanamycin- and an
appropriate. ampicillin resistance gene allows for selection of the plasmid
P. pastoris cells were generally grown at 30∘ C in rich media libraries in E. coli, and the kanamycin resistance gene also
with glucose (YPD; 2% peptone, 1% yeast extract, and 2% glu- confers resistance against the antibiotic G418 in P. pastoris,
cose) or on YPD agar (YPD broth with 2% agar). YPDS agar avoiding that 𝑃𝐴𝑂𝑋1 expression levels (Zeocin resistance)
(YPD agar with 1 M sorbitol) was used for transformation could affect the composition of the libraries in any way.
purposes. The minimal media contained 1.34% yeast nitrogen Plasmid pPPE20 was used when constructing the mutant
base, 4 ⋅ 10−5 % biotin, 0.004% histidine, and 0.5% methanol. library in the promoter core region of the 𝑃𝐴𝑂𝑋1 promoter
2% agar was included for agar media preparation. Methanol (plasmid library LC). The construction of the LC plasmid
(0.5%) was added once a day to minimal media containing library is similar to the procedure previously described for
Table 1: Strains and vectors used in this study.
Reference or
Strain or plasmid Descriptiona,b,c
source
E.coli strains
DH5𝛼 General cloning host. BRL
ER2925 Host used for cloning experiments involving methylation-sensitive restriction enzymes. NEB
P. pastoris strain
GS115 Host used for all of the P. pastoris screening and growth experiments. Invitrogen
Vectors
Bacterial RK2-based expression vector harboring Pm/xylS regulatory promoter system with the
pIB11 [6]
bla gene as reporter, KmR .
pGL3-control
Firefly luciferase reporter vector, ApR . Promega
vector
P. pastoris episomal expression vector harbouring the 𝑃𝐺𝐴𝑃 promoter. E. coli-P. pastoris shuttle
pBGP1 [19]
vector, ZcR , ApR .
P. pastoris expressional vector harbouring the 𝑃𝐴𝑂𝑋1 promoter. E. coli-P. pastoris shuttle vector,
pPICZ A Invitrogen
ZcR .
pBGP1 derivative in which the three BspHI restriction sites are deleted including one situated in
pPPE3 This study
the 𝑃𝐺𝐴𝑃 promoter by using site-directed mutagenesis with primer pair 1, 2, and 3, ZcR , ApR .
pPPE3 derivative in which a BspHI restriction enzyme site was inserted in the start of the
pPPE8 This study
zeocin-resistance gene by using site-directed mutagenesis with primer pair 4, ApR .
pPPE8 derivative in which the zeocin-resistance gene was substituted with the
kanamycin-resistance gene from plasmid pIB11. The 𝐾𝑚𝑅 gene was amplified from pIB11 with
pPPE10 This study
primer pair 5 and end digested with BspHI and EcoRV prior to ligation into the corresponding
sites of pPPE8, ApR , KmR , G418R .
pPPE10 derivative in which a NcoI restriction enzyme site was inserted in the start of the Alpha
pPPE11 This study
signal by using site-directed mutagenesis with primer pair 6, ApR , KmR , G418R .
pPPE11 derivative in which the zeocin-resistance gene was inserted as reporter for the modified
𝑃𝐺𝐴𝑃 promoter. The zeocin-resistance gene was amplified from plasmid pBGP1 with primer pair 7
pPPE12 This study
and end digested with NcoI and NotI prior to ligation into the corresponding sites of pPPE11,
ApR , KmR , G418R .
pPPE12 derivative in which the modified 𝑃𝐺𝐴𝑃 promoter was substituted with the 𝑃𝐴𝑂𝑋1 . The
pPPE17 𝑃𝐴𝑂𝑋1 promoter was amplified from pPICZ A with primer pair 8 and end digested with SpeI and This study
NcoI prior to ligation into the corresponding sites of pPPE12, ApR , KmR , G418R .
pPICZ A derivative in which the BspHI restriction enzyme site in the 𝑃𝐴𝑂𝑋1 promoter is deleted
pIPA1 This study
by using site directed mutagenesis with primer pair 9, ZcR .
pIPA1 derivative in which an EcoRI restriction enzyme site is inserted directly upstream of the
pIPA4 putative TATA-box in the 𝑃𝐴𝑂𝑋1 promoter by using site-directed mutagenesis with primer pair 10, This study
ZcR .
pIPA4 derivative in which a NheI restriction enzyme site is inserted downstream of the
pIPA5 transcriptional start site in the 𝑃𝐴𝑂𝑋1 promoter by using site-directed mutagenesis with primer This study
pair 11, ZcR .
Similar to pPPE17 except for that the 𝑃𝐴𝑂𝑋1 promoter was substituted with the modified 𝑃𝐴𝑂𝑋1
promoter from plasmid pIPA5. The 𝑃𝐴𝑂𝑋1 promoter was amplified from plasmid pIPA5 with
pPPE20 This study
primer pair 12 and end digested with SpeI and NcoI prior to ligation into the corresponding sites
of pPPE12, ApR , KmR , G418R .
Similar to pPPE17 except for that the 𝑃𝐴𝑂𝑋1 promoter was substituted with the wild-type 𝑃𝐺𝐴𝑃
pPPE22 promoter from plasmid pBGP1 by subcloning the SpeI/MfeI fragment from plasmid pBGP1 into This study
the corresponding sites of plasmid pPPE12, ApR , KmR , G418R .
Similar to pPPE17 except for that the 𝑍𝑐𝑅 gene is substituted with the luc+ gene (encoding firefly
luciferase). The luc+ gene was amplified from plasmid pGL3-control vector with primer pair 13
pPPE35 This study
and end digested with NcoI and NotI prior to ligation into the corresponding sites of pPPE17,
ApR , KmR , G418R .
Similar to pPPE20 except for that a KpnI restriction enzyme site is inserted in the 𝑃𝐴𝑂𝑋1 promoter
pPPE50 about 90 base pairs upstream of the EcoRI restriction enzyme site by using site-directed This study
mutagenesis with primer pair 14, ApR , KmR , G418R .
a
Ap: ampicillin; Km: kanamycin; Zc: Zeocin, G418: aminoglycoside antibiotic.
b
Sequences of the primer pairs are listed in Table 2.
c
The target region was subcloned and verified by sequencing for all of the site-directed mutagenesis applications. Also, all cloned regions amplified by PCR
were verified by sequencing.
Table 2: Primers used in genetic modifications. region with KpnI- and EcoRI-compatible ends when
annealed for subsequent easy cloning into the pPPE50 vector.
Primer
pair Sequencea As before, one of the oligonucleotides corresponded to the
number wild-type sequence (5󸀠 -AATTCGAACAGTTAAATT-
5󸀠 -TGTATCCGCTAATGAGACAAT-3󸀠 and TTGATCATTAACGTTAGGCTATCAGCAGTATTC-
1
5󸀠 -TTGTCTCATTAGCGGATACA-3󸀠 CCACCAGAATCTTGGAAGCATACAATGTGGAGA-
2
5󸀠 -GATTTTGGTAATGAGATTAT-3󸀠 and CAATGCATAAGGTAC-3󸀠 ), and the oligonucleotide
5󸀠 -ATAATCTCATTACCAAAATC-3󸀠
corresponding to the complementary strand was randomly
5󸀠 -GGACGCATGTAATGAGATTATT-3󸀠 and
3
5󸀠 -AATAATCTCATTACATGCGTCC-3󸀠
mutagenized by the use of a doped oligonucleotide solution
5󸀠 -AGGAACTAAATCATGACCAAGTTGAC-3󸀠 and
(5󸀠 -C22324132242121131322423241221133432212442444332-
4
5󸀠 -GTCAACTTGGTCATGATTTAGTTCCT-3󸀠 312412432341123314223324321333322233124221G-3󸀠 , where
5󸀠 -TAATAATCATGAGCCATATTCAAC-3󸀠 and the numbers in the sequence indicate the doping percentages
5
5󸀠 -TAGATATCATTAGAAAAACTCATC-3󸀠 of the nucleotides: 1 = 88% C, 4% A, 4% T, and 4% G; 2 = 88%
5󸀠 -ATTTCGAAACCATGGGATTTCCTTC-3󸀠 and T, 4% A, 4% C, and 4% G; 3 = 88% A, 4% C, 4% T, and 4% G;
6
5󸀠 -GAAGGAAATCCCATGGTTTCGAAAT-3󸀠 4 = 88% G, 4% A, 4% T, and 4% C). Approximately 1 million
5󸀠 -TAATAACCATGGCCAAGTTGACC-3󸀠 and E. coli DH5𝛼 transformants were obtained, constituting the
7
5󸀠 -TAATAAGCGGCCGCATCAGTCCTGCTCCTC-3󸀠 LU plasmid library.
5󸀠 -TAATAAACTAGTAGATCTAACATCCAA-3󸀠 and
8
5󸀠 -TAATAACCATGGTTTCGAATAATTAG-3󸀠
5󸀠 -AGCCTAACGTTAATGATCAAAATT-3󸀠 and 2.5. Screening for 𝑃𝐴𝑂𝑋1 Promoter Variants. The Qiafilter
9
5󸀠 -AATTTTGATCATTAACGTTAGGCT-3󸀠
maxi plasmid isolation kit (Qiagen) was used for high-
5󸀠 -AAAATTTAACTGTTCGAATTCCTACTTGACAGCAA-3󸀠
10 and
yield purification of the plasmid libraries from E. coli DH5𝛼
5󸀠 -TTGCTGTCAAGTAGGAATTCGAACAGTTAAATTTT-3󸀠 according to the procedure described by the manufacturers in
5󸀠 -TCATAATTGCGACTGCTAGCAATTGACAAGCTTT-3󸀠 order to obtain sufficient transformation efficiencies of these
11 and libraries into P. pastoris. Circular plasmids were transformed
5󸀠 -AAAGCTTGTCAATTGCTAGCAGTCGCAATTATGA-3󸀠
to P. pastoris GS115 and the transformation suspensions were
5󸀠 -TAATAAACTAGTAGATCTAACATCCAA-3󸀠 and
12
5󸀠 -TAATAACCATGGTTTCGAATAATTAG-3󸀠
plated on rich agar media with glucose or minimal agar
5󸀠 -TAATAACCATGGAAGACGCCAAAAACA-3󸀠 and
media with methanol containing increasing concentrations
13 of Zeocin to directly select for the transformants containing
5󸀠 -TAATAAGCGGCCGCATTACACGGCGATCTTTC-3󸀠
5󸀠 -ACCCGCTTTTTGGTACCTTATGCATTGT-3󸀠 and the 𝑃𝐴𝑂𝑋1 promoter variants leading to stimulated expression
14
5󸀠 -ACAATGCATAAGGTACCAAAAAGCGGGT-3󸀠 compared to that of the wild-type. Normally, 100 𝜇L of the
a
The restriction enzyme site is underlined. transformation suspension was plated on 9 cm agar plates,
except for the big-scale screening experiment of the LC plas-
mid library on rich agar media with glucose (plating above
100000 transformants) in which 2.1 mL of the transformation
the Pm promoter libraries [5, 6]. The oligonucleotides were suspension was plated on square Corning bioassay plates
designed to constitute a double-stranded DNA fragment
(24.5 × 24.5 cm). To determine an approximate number of
containing the 𝑃𝐴𝑂𝑋1 promoter core region with EcoRI- and
the transformants containing the plasmid that were plated,
NheI-compatible ends when annealed for subsequent easy
the transformation suspension was also plated on YPDS agar
cloning into the pPPE20 vector. One of the oligonucleotides
containing G418.
corresponded to the wild-type sequence (5󸀠 -CTAGCA-
GTCGCAATTATGAAAGTAAGCTAATAATGATGA-
TAAAAAAAAAGGTTTAAGACAGGGCAGCTTCCT-
2.6. Luciferase Activity Assay. Freshly transformed P. pastoris
TCTGTTTATATATTGCTGTCAAGTAGG-3󸀠 ), and the
GS115 strains were grown according to the protocols for
oligonucleotide corresponding to the complementary strand
induced expression of the 𝑃𝐴𝑂𝑋1 promoter by Invitrogen
was randomly mutagenized by the use of a doped oligonu-
cleotide solution (5󸀠 -AATTC1231224313413323232333134- (http://www.invitrogen.com/). For repressed expression,
334433412411124212233311222222222321321322322341223- freshly prepared P. pastoris GS115 transformants were
12221323322414312G-3󸀠 , where the numbers in the sequence inoculated to 10 mL YPD in 125 mL bottles and incubated at
indicate the doping percentages of the nucleotides: 1 = 79% 225 rpm and 30∘ C overnight. Next, overnight cultures were
C, 7% A, 7% T, and 7% G; 2 = 79% T, 7% A, 7% C, and 7% G; 3 inoculated to 50 mL YPD in 250 mL baffled bottles to a final
= 79% A, 7% C, 7% T, and 7% G; 4 = 79% G, 7% A, 7% T, and optical density at 600 nm of 0.008 and incubated at 225 rpm
7% C). Approximately 410000 E. coli DH5𝛼 transformants and 30∘ C for 3 days. Samples were collected as indicated in
were obtained, constituting the LC plasmid library. the results. Lysis of the cell samples was carried out according
Plasmid pPPE50 was used when constructing the plas- to the protocols by Invitrogen and, the luciferase activity
mid library LU, and the construction of this library is identical assay was performed as described by the manufacturers
to that of the LC plasmid library, except that the oligonucleot- (http://www.promega.com/). The luciferase activity analyses
ides were designed to constitute a double-stranded DNA were repeated at least twice, and measurements were carried
fragment containing the 𝑃𝐴𝑂𝑋1 promoter core upstream out with minimum three technical recurrences.
· · · GGATGATTATGCATTGTCTCCACATTGTATGCTTCCAAGATTCTGGTGGGAATACTGCTG
ATAGCCTAACGTTCATGATCAAAATTTAACTGTTCTAACCCCTACTTGACAGCAATATATAAACAGAAGGAAG
Putative TATA-box
CTGCCCTGTCTTAAACCTTTTTTTTTATCATCATTATTAGCTTACTTTCATAATTGCGACTCCAATTAATTGAC
Transcriptional start site
AAGCTTTTGATTTTAACGACTTTTAACGACAACTTGAGAAGATCAAAAAACAACTAATTATTCGAAACCATGG
NcoI
NcoI
NotI
PAOX1 ZcR
Spel
AOX1 TT
PTEF1
pPPE17 PEM7
ApR (5498 bps)
KmR
PARS1
pUC origin
Figure 1: Physical map of the E. coli-P. pastoris shuttle plasmid pPPE17 and its derivatives. The restriction enzyme sites shown are unique.
Abbreviations: 𝑃𝐴𝑂𝑋1 : P. pastoris AOX1 promoter; 𝑍𝑐𝑅 : Zeocin resistance gene; AOX1 TT: P. pastoris AOX1 transcription termination region;
𝑃𝑇𝐸𝐹1 : yeast promoter TEF1; 𝑃𝐸𝑀7 : bacterial promoter EM7; 𝐾𝑚𝑅 : kanamycin resistance gene (confers resistance against kanamycin in bacteria
and G418 in yeast); PARS1: P. pastoris autonomous replication sequence; pUC origin: bacterial replication sequence; 𝐴𝑝𝑅 : ampicillin resistance
gene (bla). DNA sequence of the 3󸀠 -part of the wild-type 𝑃𝐴𝑂𝑋1 promoter (base pairs 670 to 950) is displayed above the plasmid map.
The promoter core region is shown in bold (see also text). The transcriptional start site (A) and the putative TATA-box (TATATAAA) are
highlighted in grey. The BspHI (TCATGA) restriction enzyme site is written in bold and underlined and is deleted by the introduction of a
point mutation (TAATGA) in plasmids pPPE20 and pPPE50. GGATGA is mutated to the KpnI (GGTACC) restriction enzyme site in plasmids
pPPE50. TAACCC is mutated to the EcoRI (GAATTC) restriction enzyme site in plasmids pPPE20 and pPPE50. CCAATT is mutated to the
NheI (GCTAGC) restriction enzyme site in plasmid pPPE20 and pPPE50.
3. Results and Discussion levels, it is critical to ensure a constant 𝑍𝑐𝑅 gene dosage in
the yeast host cells. Therefore, we chose to use the P. pastoris
3.1. Design and Construction of the Cassette Expression Plas- episomal (i. e. autonomously replicating) plasmid vector
mid pPPE17 to Enable Direct Selection of Yeast Promoter pBGP1 [19] as the carrier for the expression cassette, in favour
Variants with Altered Expression Properties. It has been well of using chromosomally integrated vectors which may lead to
documented that xylS/Pm is a strong and flexible promoter different expression levels depending on the location of the
system useful for high-level, as well as for controlled, recom-
integration and possible multicopy insertions of the expres-
binant expression of genes and pathways in Gram-negative
sion cassette. Plasmid pBGP1 was used as basis to generate the
bacteria (see Section 1). Still, Pm promoter variants with
new and improved properties have been generated by using pPPE17 plasmid, which contains the 𝑍𝑐𝑅 gene under control
combinatorial mutagenesis and in such approaches we have of the 𝑃𝐴𝑂𝑋1 promoter. This design enables the substitution
taken advantage of the correlation between the expression of the yeast promoter and/or gene of interest in a one-step
level of the bla gene (encoding 𝛽-lactamase) and the corre- cloning procedure (Figure 1). In addition, plasmid pPPE22
sponding ampicillin tolerance level of the host cells [5, 6, 18]. was constructed by substituting the 𝑃𝐴𝑂𝑋1 promoter in
We here chose to apply the Zeocin resistance gene 𝑍𝑐𝑅 as pPPE17 with the constitutive P. pastoris 𝑃𝐺𝐴𝑃 promoter (see
reporter due to its documented usefulness in identifying Figure 1), and it was included as a valuable control system.
yeast transformants with multicopy chromosomal insertions The latter promoter is reported to give strong expression
based on the Zeocin tolerance levels of the host cells (http:// under both the induced (presence of methanol and absence of
www.invitrogen.com/). When using this strategy to directly repressive carbon sources) and repressed (presence of glu-
select recombinant yeast strains with altered Zeocin tolerance cose) growth conditions for the 𝑃𝐴𝑂𝑋1 promoter [20].
Both plasmids pPPE17 and pPPE22 were transformed P. pastoris GS115 (pPPE17) under glucose-repressed condi-
into host strain P. pastoris GS115, and the maximal Zeocin tions. Over 100000 transformants representing library LC
tolerance levels were determined under 𝑃𝐴𝑂𝑋1 -promoter- and over 5000 transformants representing library LU were
repressed conditions for strains GS115, GS115 (pPPE17), and screened, and five totally different 𝑃𝐴𝑂𝑋1 promoter variants
GS115 (pPPE22). We chose to use rich agar media with (named LC-1, LC-2, LU-1, LU-2, and LU-3) with strongly ele-
glucose (YPD) for the repressed growth condition purpose. vated Zeocin tolerance levels were isolated. The selected pro-
Recombinant strains GS115 (pPPE17) and GS115 (pPE22) moter variants provided significantly higher maximal Zeocin
displayed about 5-fold and 400-fold higher Zeocin tolerance tolerance levels of the GS115 cells under repressed conditions
levels (25 𝜇g/mL and 2000 𝜇g/mL, resp.) compared to that of (up to 2000 𝜇g/mL) compared to that of the wild-type 𝑃𝐴𝑂𝑋1
host GS115 (up to 5 𝜇g/mL). The results show that the expres- promoter (25 𝜇g/mL). The five selected promoter variants had
sion level from the 𝑃𝐴𝑂𝑋1 promoter is low under the glucose- between 2 (LU-2) and 18 (LC-1) point mutations (Figure 2).
repressed conditions tested, as expected. Further, the results The LU-2 and the LC-2 promoter variants have only two
demonstrate a correlation between the expected expression and three point mutations, respectively, suggesting that single
levels of the 𝑍𝑐𝑅 gene from the two promoters and the toler- nucleotides directly upstream of the TATA box are important
ance levels of the corresponding host cells. for glucose repression of this promoter. The LC-1, LU-1, and
Plasmid pPPE17 was further modified into a promoter- LU-3 promoter variants, on the other hand, have multiple
cloning cassette system by deletion of a BspHI restriction (between 11 and 19) point mutations distributed over the
enzyme recognition site and insertion of two unique restric- entire randomized promoter regions and thus both upstream
tion enzyme recognition sites (EcoRI and NheI) in the 𝑃𝐴𝑂𝑋1 and downstream of the TATA-box. Together, these data may
promoter, generating plasmid pPPE20 (Figure 1). These mod- suggest the existence of several cis-acting promoter elements
ifications did not affect the expression properties of the involved in the glucose repression of the 𝑃𝐴𝑂𝑋1 promoter.
plasmid compared to pPPE17 (data not shown), and pPPE20 These results demonstrated that the combinatorial mutagen-
was used for screening of 𝑃𝐴𝑂𝑋1 promoter mutant libraries for esis and selection strategy can be used to efficiently select for
variants with altered expression properties as described next. altered promoter variants, and they also show that the 𝑃𝐴𝑂𝑋1
promoter core region and the 90 base pairs upstream region
can have high impact on the glucose repression under the
3.2. Selection of 𝑃𝐴𝑂𝑋1 Promoter Variants with Abolished conditions tested.
Glucose Repression. The 𝑃𝐴𝑂𝑋1 cis-acting promoter elements
involved in the glucose repression are not yet identified (see
Section 1). We here initially selected for 𝑃𝐴𝑂𝑋1 promoter vari- 3.3. Selected 𝑃𝐴𝑂𝑋1 Promoter Variants Retain Their Charac-
ants with altered glucose repression to establish and assess the teristics When Expressing an Alternative Reporter Protein, the
combinatorial mutagenesis approach. Previous mutagenesis Firefly Luciferase. Previous combinatorial mutagenesis work
strategies aiming at identifying 𝑃𝐴𝑂𝑋1 promoter variants with with the bacterial Pm promoter has shown that there can be
altered glucose repression have focused on the promoter a strong context dependency between the promoter sequence
region upstream of the putative TATA-box [10, 12]. We here and the coding region of the reporter gene which often affects
decided to explore the impact on the glucose regulation of the expression level in an unpredictable manner [5, 18]. Two
two distinct regions of the 𝑃𝐴𝑂𝑋1 promoter. First, the 𝑃𝐴𝑂𝑋1 of the selected 𝑃𝐴𝑂𝑋1 promoter variants (LC-2 and LU-2) were
promoter core region was here defined as the region from therefore chosen for expression studies in liquid P. pastoris
about 15 base pairs upstream of the putative TATA-box to cultures with the alternative reporter gene luc+, encoding
about 35 base pairs downstream of the transcriptional start firefly luciferase. For this purpose, the 𝑍𝑐𝑅 gene in the pPPE17
site (see Figure 1). A mutant library was constructed in this plasmids containing the wild-type 𝑃𝐴𝑂𝑋1 promoter and each
𝑃𝐴𝑂𝑋1 promoter core region by the use of randomly mutated of the LC-2 and LU-2 promoter variants was substituted
synthetic oligonucleotides (designated plasmid library LC), with the luc+ gene. The luciferase activities of GS115 cells
cloned into the NheI/EcoRI sites of pPPE20 (Figure 1) and harbouring these plasmids were quantified under repressed
transformed into P. pastoris GS115. Second, we constructed a conditions and during the glucose-depletion phase in shake
promoter mutant library (designated plasmid library LU) by flasks (Figure 3). Both promoter variants LC-2 and LU-2 gave
randomizing the 90 base pairs region directly upstream of the strongly enhanced luciferase activities at the first time point
𝑃𝐴𝑂𝑋1 promoter core region, corresponding to the region that compared to that of the wild-type (about 300-fold and 600-
was extensively studied by Hartner et al. [14] by using rational fold, resp.). These differences decrease to around 10-fold for
design. To accomplish this, plasmid pPPE20 was modified by the remaining time points for both of the promoter variants
the insertion of a unique restriction enzyme site (KpnI) in the (the second time point represents the transition from expo-
𝑃𝐴𝑂𝑋1 promoter, generating plasmid pPPE50. In the latter nential to stationary growth phase). Thus, the increase in
plasmid, the promoter upstream region can be easily sub- activity after the first time point is much larger for the wild-
stituted by a one-step cloning procedure using the unique type, which shows that the LC-2 and LU-2 promoter variants
EcoRI/KpnI restriction sites (Figure 1). As for pPPE20 (see are far less affected by the depletion of glucose compared
above), this modification did not affect the expression prop- to that of the wild-type promoter (Figure 3). Taken together,
erties of pPPE50 compared to pPPE17 (data not shown). these results confirmed that LC-2 and LU-2 are 𝑃𝐴𝑂𝑋1 pro-
The LC and LU plasmid libraries were then screened for moter variants with abolished glucose repression irrespective
increased Zeocin tolerance levels relative to the control strain of the context of the gene to be expressed.
Wild-type -159 ggatgattatgcattgtctccacattgtatgcttccaagattctggtgggaatactgctgatagcctaacgttcatgatcaaaatttaactgttctaacccctacttgacagcaa

LC-1 -159 ggatgattatgcattgtctccacattgtatgcttccaagattctggtgggaatactgctgatagcctaacgttaatgatcaaaatttaactgttcGAATTCctgctcagtcgcga
LC-2 -159 ggatgattatgcattgtctccacattgtatgcttccaagattctggtgggaatactgctgatagcctaacgttaatgatcaaaatttaactgttcGAATTCcttcgcgacagcaa
LU-1 -158 GGTACC-tatgcattgtcgcgacattgcatgcctccacgattctggtgggaattctgcttatggaccaacgttcatgatcaaaatttaactgttcGAATTCctacttgacagcaa
LU-2 -158 GGTACC-tatgcattgtctccacattgtatgcttccaagattctggtgggaatactgctgatagcctaacgttaatgatctaaatttaactgctcGAATTCctacttgacagcaa
LU-3 -158 GGTACC-tatgcattgtctccacattgtaaactcccacgactcttgtgggaatacggcttacagccgaccgttaatgctcaaaatttaaccgttcGAATTCctacttgacagcaa
LU-4 -158 GGTACC-taaccattgtgtccacattgtatgctttcaagcttctggtgggaatactgctaataccctaacgttaatgatcaaaattttactgctcGAATTCctacttgacagcaa
LU-5 -158 GGTACC-tctgcattgtctccacattgtatgcttccaagattctggtgggaatttcgctgatagcataacgttaatgatcaaaatttaactgttcGAATTCctacttgacagcaa
LU-6 -158 GGTACC-tatgcgttgacttcacattccatgcttaccagattctggtgggcatactgctgatagcataacgttaatggtcaaaatttcactgctcGAATTCctacttgacagcaa
+1 Promoter variant phenotype

Wild-type -44 TATATAAAcagaaggaagctgccctgtcttaaacctttttttttAtcatcattattagcttactttcataattgcgactggttcc
LC-1 -44 TATATAACcagaaggaagctgctctcccttaaatcttgttttttCtcatcactatttacctactttcataattgccactGCTAGC 40-fold wild-type R
LC-2 -44 TATATAAAcagaaggaagctgccctgtcttaaacctttttttttAtcatcattattagcttactttcataattgcgactGCTAGC 80-fold wild-typeR
LU-1 -44 TATATAAAcagaaggaagctgccctgtcttaaacctttttttttAtcatcattattagcttactttcataattgcgactGCTAGC 40-fold wild-type R
LU-4 -44 TATATAAAcagaaggaagctgccctgtcttaaacctttttttttAtcatcattattagcttactttcataattgcgactGCTAGC >3-fold wild-typeI
Figure 2: DNA sequence of region between base pairs-159 to +41 (promoter core region and the about 90 base pairs region directly upstream
of the promoter core region, the transcriptional start site is set as +1) of the wild-type 𝑃𝐴𝑂𝑋1 promoter and variants LC-1, LC-2, and LU-1 to
LU-6. Mutations are shown in bold and underlined. Deletion mutations are indicated by short horizontal lines. The transcriptional start site
(A) and the putative TATA-box (TATATAAA) are written in upper case and highlighted in grey. Mutations leading to restriction-enzyme-
site insertions or deletions without affecting the promoter activity are written in grey. The KpnI (GGTACC), EcoRI (GAATTC), and NheI
(GCTAGC) restriction enzyme sites are written in upper case and underlined. The relative Zeocin tolerance levels of the promoter variants
compared to the wild-type promoter under repressed (wild-typeR ) and induced (wild-typeI ) growth conditions are shown.
10000 variants with increased expression levels under induced con-

ditions. As an initial test, the maximum Zeocin tolerance
Relative luciferase activities
levels under induced conditions (minimal agar media with

1000
methanol) of strains GS115 and GS115 (pPPE17) were deter-
mined and found to be 25 𝜇g/mL and 2000 𝜇g/mL, respec-
100 tively. Next, over 5000 transformants representing each of the
LC and LU promoter libraries (see above) were screened with
respect to increased Zeocin tolerance levels relative to GS115
10
cells harbouring original plasmid pPPE17 under 𝑃𝐴𝑂𝑋1 -
induced conditions. In total, three promoter variants (LU-4,
1 LU-5, and LU-6) were identified that caused increased Zeocin
0 20 40 60 80 100 120 tolerance levels of the GS115 cells under induced conditions
Time (hours) (7000 𝜇g/mL or more) compared to that of the wild-type
Figure 3: Relative luciferase activities under repressed growth promoter (2000 𝜇g/mL). The three selected promoter vari-
conditions and during the glucose-depletion phase for the wild- ants contained between 5 and 12 point mutations (Figure 2).
type P. pastoris 𝑃𝐴𝑂𝑋1 promoter (black squares) and variants LC- These promoter variants contain multiple (between 5 and
2 (grey triangles) and LU-2 (black circles, dashed line) in plasmid 12) point mutations distributed over the randomized region,
pPPE35 expressed in P. pastoris GS115. Enzyme activities are shown analogous to the LU-1 and LU-3 variants that were screened
in logarithmic scale. The values are the average of two biological for increased activity under glucose-repressed conditions (see
replicas, and the error bars show the standard deviations. The above). The biological reason for these results is unclear. One
activities are relative to the wild-type, for which the value at the first potential explanation might be that there are cis-acting pro-
time point is arbitrarily set to 1. moter elements important for both methanol induction and
glucose repression located in the same region of the 𝑃𝐴𝑂𝑋1
promoter.
3.4. Selection of 𝑃𝐴𝑂𝑋1 Promoter Variants with Increased For practical reasons we found that using Zeocin concen-
Expression Level under Induced Conditions. Finally, we trations higher than 7000 𝜇g/mL was undesirable, and due to
wanted to test if it was possible to select 𝑃𝐴𝑂𝑋1 promoter this the identified promoter variants were not further ranked
at this stage. Instead, luciferase expression studies with GS115 References

cells under induced growth conditions were performed in
shake flasks for two of these 𝑃𝐴𝑂𝑋1 promoter variants (LU- [1] J. L. Ramos, S. Marqués, and K. N. Timmis, “Transcriptional
control of the Pseudomonas TOL plasmid catabolic operons
4 and LU-5). Analyses of samples withdrawn at the time
is achieved through an interplay of host factors and plasmid-
points for transition from exponential to stationary growth encoded regulators,” Annual Review of Microbiology, vol. 51, pp.
phase (about 72 hours) showed a similar activity for the 341–373, 1997.
LU-5 promoter variant and the wild-type promoter, whereas [2] T. Brautaset, R. Lale, and S. Valla, “Positively regulated bacterial
variant LU-4 displayed a 2-fold increase in luciferase activity expression systems,” Microbial Biotechnology, vol. 2, pp. 15–30,
compared to that of the wild-type (data not shown). This 2009.
demonstrates that the combinatorial mutagenesis and selec- [3] H. Sletta, A. Nedal, T. E. V. Aune et al., “Broad-host-range
tion strategy can be applied to also generate improved 𝑃𝐴𝑂𝑋1 plasmid pJB658 can be used for industrial-level production of a
promoter variants under induced conditions. The potential is secreted host-toxic single-chain antibody fragment in Escher-
(further) likely to be greater than indicated here, considering ichia coli,” Applied and Environmental Microbiology, vol. 70, no.
that only a small fraction of the mutant libraries was screened 12, pp. 7033–7039, 2004.
and a narrow selection window was used (a concentration of [4] H. Sletta, A. Tøndervik, S. Hakvåg et al., “The presence of N-
Zeocin ranging from 2000 to 7000 𝜇g/mL). terminal secretion signal sequences leads to strong stimulation
of the total expression levels of three tested medically important
proteins during high-cell-density cultivations of Escherichia
4. Conclusions coli,” Applied and Environmental Microbiology, vol. 73, no. 3, pp.
906–912, 2007.
Mutational analysis is an efficient approach to map specific [5] I. Bakke, L. Berg, T. E. V. Aune et al., “Random mutagenesis
promoter functions and properties to defined sequence of the Pm promoter as a powerful strategy for improvement of
motifs and nucleotides. Rational methods including site- recombinant-gene expression,” Applied and Environmental
directed mutagenesis are preferred but such approaches rely Microbiology, vol. 75, no. 7, pp. 2002–2011, 2009.
on a detailed preinsight into the relation between the se- [6] L. Berg, R. Lale, I. Bakke, N. Burroughs, and S. Valla, “The
quence and function, which is typically not the case for most expression of recombinant genes in Escherichia coli can be
promoters. Accordingly, the need for creating libraries with strongly stimulated at the transcript production level by mutat-
random mutations distributed over defined regions might be ing the DNA-region corresponding to the 5󸀠 -untranslated part
preferred; however, such strategies are dependent on efficient of mRNA,” Microbial Biotechnology, vol. 2, no. 3, pp. 379–389,
selection and screening methods to identify mutations lead- 2009.
ing to the desired phenotypes. The current study demon- [7] R. Lale, L. Berg, F. Stüttgen et al., “Continuous control of the
strates for the first time a combinatorial mutagenesis and flow in biochemical pathways through 5󸀠 untranslated region
selection approach for generating new understanding of the sequence modifications in mRNA expressed from the broad-
regulation of promoter systems and to develop improved host-range promoter Pm,” Applied and Environmental Microbi-
ology, vol. 77, pp. 2648–2655, 2011.
expression systems in yeast. The tools and technology devel-
oped here should be valuable for investigations and improve- [8] C. P. Hollenberg and G. Gellissen, “Production of recombinant
proteins by methylotrophic yeasts,” Current Opinion in Biotech-
ment of any yeast promoter system, leading to new basic
nology, vol. 8, no. 5, pp. 554–560, 1997.
knowledge as well as to useful tools for heterologous expres-
[9] L. M. Damasceno, C. Huang, and C. A. Batt, “Protein secretion
sion of single genes and complex biosynthetic pathways in
in Pichia pastoris and advances in protein production,” Applied
yeast. Understanding promoters is crucial to generate new Microbiology and Biotechnology, vol. 93, pp. 31–39, 2012.
basic knowledge of microbial gene expression and regula-
[10] J. M. Cregg, K. R. Madden, K. J. Barringer, G. P. Thill, and
tion at all levels, and development of new and specialized C. A. Stillman, “Functional characterization of the two alcohol
expression systems is important to experimentally analyze oxidase genes from the yeast Pichia pastoris,” Molecular and Cel-
microbial diversity including metagenome libraries for new lular Biology, vol. 9, no. 3, pp. 1316–1323, 1989.
functions and products. Finally, the tools and technology [11] S. B. Ellis, P. F. Brust, and P. J. Koutz, “Isolation of alcohol oxidase
developed here should also be useful as a promoter probe and two other methanol regulatable genes from the yeast Pichia
system for efficient screening and characterization of natural pastoris,” Molecular and Cellular Biology, vol. 5, no. 5, pp. 1111–
promoter libraries in yeast. 1121, 1985.
[12] F. S. Hartner and A. Glieder, “Regulation of methanol utilisation
pathway genes in yeasts,” Microbial Cell Factories, vol. 5, pp. 39–
Conflict of Interests 59, 2006.
The authors of this paper have no conflict of interests and they [13] J. F. Tschopp, P. F. Brust, J. M. Cregg, C. A. Stillman, and T.
R. Gingeras, “Expression of the lacZ gene from two methanol-
have no direct financial relation with the commercial identi-
regulated promoters in Pichia pastoris,” Nucleic Acids Research,
ties mentioned in this paper.
vol. 15, no. 9, pp. 3859–3876, 1987.
[14] S. F. S. Hartner, C. Ruth, D. Langenegger et al., “Promoter library
Acknowledgment designed for fine-tuned gene expression in Pichia pastoris,”
Nucleic Acids Research, vol. 36, no. 12, article e76, 2008.
This work was supported by a grant from the Research Coun- [15] C. A. Staley, A. Huang, M. Nattestad et al., “Analysis of the 5󸀠 -
cil of Norway. untranslated region (5󸀠 UTR) of the alcohol oxidase 1 (AOX1)
gene in recombinant protein expression in Pichia pastoris,”

Gene, vol. 496, pp. 118–227, 2012.
[16] Y. Xuan, X. Zhou, W. Zhang, X. Zhang, Z. Song, and Y. Zhang,
“An upstream activation sequence controls the expression of
AOX1 gene in Pichia pastoris,” FEMS Yeast Research, vol. 9, no.
8, pp. 1271–1282, 2009.
[17] H. C. Winther-Larsen, J. M. Blatny, B. Valand, T. Brautaset, and
S. Valla, “Pm promoter expression mutants and their use in
broad-host-range RK2 plasmid vectors,” Metabolic Engineering,
vol. 2, no. 2, pp. 92–103, 2000.
[18] L. Berg, V. Kucharova, I. Bakke, S. Valla, and T. Brautaset,
“Exploring the 5󸀠 -UTR DNA region as a target for optimizing
recombinant gene expression from the strong and inducible Pm
promoter in Escherichia coli,” Journal of Biotechnology, vol. 158,
pp. 224–230, 2012.
[19] C. C. Lee, T. G. Williams, D. W. S. Wong, and G. H. Robertson,
“An episomal expression vector for screening mutant gene
libraries in Pichia pastoris,” Plasmid, vol. 54, no. 1, pp. 80–85,
2005.
[20] H. R. Waterham, M. E. Digan, P. J. Koutz, S. V. Lair, and J.
M. Cregg, “Isolation of the Pichia pastoris glyceraldehyde-3-
phosphate dehydrogenase gene and regulation and use of its
promoter,” Gene, vol. 186, no. 1, pp. 37–44, 1997.
http://dx.doi.org/10.1155/2013/463894
Review Article
Cyanobacterial Toxin Degrading Bacteria: Who Are They?
Konstantinos Ar. Kormas and Despoina S. Lymperopoulou

Department of Ichthyology & Aquatic Environment, School of Agricultural Sciences, University of Thessaly, 384 46 Volos, Greece
Correspondence should be addressed to Konstantinos Ar. Kormas; kkormas@uth.gr
Received 21 March 2013; Accepted 21 May 2013
Copyright © 2013 K. Ar. Kormas and D. S. Lymperopoulou. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.
Cyanobacteria are ubiquitous in nature and are both beneficial and detrimental to humans. Benefits include being food supplements
and producing bioactive compounds, like antimicrobial and anticancer substances, while their detrimental effects are evident
by toxin production, causing major ecological problems at the ecosystem level. To date, there are several ways to degrade or
transform these toxins by chemical methods, while the biodegradation of these compounds is understudied. In this paper, we
present a meta-analysis of the currently available 16S rRNA and mlrA (microcystinase) genes diversity of isolates known to degrade
cyanobacterial toxins. The available data revealed that these bacteria belong primarily to the Proteobacteria, with several strains
from the sphingomonads, and one from each of the Methylobacillus and Paucibacter genera. Other strains belonged to the genera
Arthrobacter, Bacillus, and Lactobacillus. By combining the ecological knowledge on the distribution, abundance, and ecophysiology
of the bacteria that cooccur with toxic cyanobacterial blooms and newly developed molecular approaches, it is possible not only to
discover more strains with cyanobacterial toxin degradation abilities, but also to reveal the genes associated with the degradation
of these toxins.
1. Introduction well. However, advances in molecular microbial ecology have

elucidated that strains of the bacterial genera Sphingomonas,
Cyanobacteria are some of the most “charismatic” microor- Sphingosinicella, Arthrobacter, Brevibacterium, Rhodococcus,
ganisms in the tree of life. Their ecological importance is and Burkholderia can degrade microcystins (MC) and nodu-
widely recognised in the scientific world [1]. Among their larins in time scales from hours to days [6, 7]. Moreover,
remarkable features, though, there is a contradiction, at we are now aware of eukaryotic mechanisms of cyanotoxin
least for human interests. On one hand, they are considered elimination in animal tissues [8–10]. This slowly growing
of pronounced biotechnological interest as they produce body of literature presents many opportunities for deeper
several bioactive compounds through their metabolism, from investigations into cyanotoxins and their biodegradation.
biofuels and biopolymers to drugs [2–4]. On the other hand, There is good evidence that toxic cyanobacterial water
they are capable of producing a wide range of nuisance blooms favour the occurrence of specific members of
secondary metabolites, that is, toxins (hereafter cyanotoxins). the bacterioplankton [11]. These prokaryotes, bacteria and
The toxicity of these compounds, which has been proved Archaea, are excellent candidates for having cyanotoxin-
for a great variety of organisms [1], dictates for solutions to degrading properties. The satellite prokaryotic communities
the problem caused by the accumulation of cyanotoxins in, of cyanobacterial water blooms (e.g., [12, 13]) could either
mostly, freshwater water bodies all over the world. Although degrade and/or assimilate the toxin and their degradation
this issue of toxicity has been known for several decades now, products or could be inhibited by the toxins during the
there has been little effort towards biotechnological remedies, bloom (e.g., [14, 15]). Although there are several well-studied
especially via degradation/transformation by microorgan- water bodies harbouring toxin-producing Cyanobacteria,
isms. This could be partially due to the notion that cyan- their accompanying prokaryotic communities are consider-
otoxins are relatively recalcitrant to chemical degradation ably understudied (e.g., [12]). Another deficit in our current
[5] and were thought as non labile for biodegradation as knowledge in cyanotoxin-degrading prokaryotes in such
studies is that despite the fact that natural freshwaters can har- Cyclo-MCLR
mlrD
bour diverse archaeal assemblages [16], practically nothing is
known about the role of Archaea in cyanotoxin degradation. Outside the cell Transporter protein-
However, it has been shown recently that hydrogenotrophic MlrD
methanogens are associated with Microcystis scum degra-
dation [17], opening, thus, a new line of research for the
Cyclo-MCLR
investigation of the role of Archaea in MC and/or other
(M.W. = 994) mlrA
cyanotoxin degradation.
Microcystinase MlrA-
putative metallopeptidase
1.1. Microcystins. Microcystins are the most widely distrib-
uted and known cyanotoxins. They consist of a group of cyclic a-MCLR
heptapeptide hepatotoxins [cyclo-(D-Ala1 -X2 -D-MeAsp3 -
Inside the cell

(M.W. = 1012)
mlrB
Z4 -Adda5 -D-Glu6 -Mdha7 )] produced by Cyanobacteria that
belong to the genera Microcystis, Anabaena, Nostoc, and Putative serine peptidase
Oscillatoria (Planktothrix) [18, 19]. Two of these peptides MlrB
(𝑋 and 𝑍) are variable, resulting in more than 70 variants NH2 -Adda-Glu-
[20]. MC-LR, in which leucine (L) at position 2 and arginine Mdha-Ala-OH
(R) at position 4 are found, is the most toxic among MCs (M.W. = 614) mlrC
[21]. Conventional water treatment processes (chlorine, per-
manganate, chloride dioxide, ozone, and advanced oxidation Putative metallopeptidase
MlrC
methods) have been found inadequate for the removal of
Small
MCs [22]. Recently, more environmental friendly approaches peptides and
have been applied to MC removal directly from natural amino acids
eutrophic freshwater systems, involving plant material and
minerals [23]. Whilst more advanced and efficient chemical
techniques (granular activated carbon, powdered activated Figure 1: The degradative pathway of microcystin LR and the
carbon, and membrane filtration) exist, they are considered formation of intermediate (less toxic) products by Sphingomonas sp.
too expensive to employ for the elimination of a contaminant strain ACM-3962. MW: molecular weight [25].
whose presence in the water bodies is occasional and hard to
predict [24]. Moreover, MCs’ stable cyclic structure against
physicochemical factors renders biodegradation inevitable,
as the most sustainable, efficient, and realistic method for Biochemical characterization has shown that the mlrA
their removal [20]. Microbial degradation is the most impor- gene encodes a metalloprotease [25], and its activation site
tant mechanism for the removal of MCs in the natural envi- consists of a typical to zinc metalloproteases motif, the
ronment and is considered as an alternative water treatment HXXMECX [27]. It has been speculated that it might as
strategy versus the physicochemical one [11]. well represent a new protease family with the function of
To date, only one biodegradation pathway for microcystin cleaving smaller cyclic peptides [27]. The replacement of the
by Sphingomonas sp. strain ACM-3962 has been character- R group by the A (alanine) group at MC-LA does not seem to
ized [25]. The mlr gene cluster plays a crucial role in the affect the degradation process, suggesting that the presence
sequential enzymatic hydrolyses of peptide bonds [21]. This of arginine may not be a critical site of cleavage [28]. In a
cluster consists of four genes, that is, mlrA, mlrB, mlrC, recent study [29] the translation of the mlrA gene sequence
and mlrD, and codes for at least three intracellular enzymes and the alignment of the resulting amino acid sequence with
(Figure 1) [25]. other putative MlrA in the database revealed a zinc-bonding
site and a transmembrane region, and the authors introduced
The mlrA is the most important gene of this cluster
the concept of a potential new protein family with unique
because it encodes the enzyme that cleaves the Adda-Arg pep- functions. mlrB encodes a putative serine peptidase which
tide bond in MC-LR and opens the cyclic structure [25]. The degrades the linear (or acyclo) MC-LR to the tetrapeptide
cyclic structure of MC is responsible for its stability against H-Adda-Glu-Mdha-Ala-OH, while MlrC exerts the action
physicochemical and biological factors that contribute to of a putative metallopeptidase and degrades the tetrapeptide
inactivation or degradation, such as pH, temperature, sun- to Adda or small amino acids [25]. The mlrD gene does
light, and other common proteases [26]. The initial hydrolysis not express hydrolytic activity and probably encodes the
mediated by the mlrA gene results in a substantial reduction transporter protein that allows the uptake of MCs into the cell
in molecular toxicity, which is displayed by the 160-fold since it exhibits high sequence similarity to the PTR2 family
reduction of activity of the linear MC-LR towards protein of oligopeptide symporters [25].
phosphatase compared to that of the cyclic MC-LR. The mlrA The MC degradation ability is not commonly present
gene sequence is very rare with no homologues found in the in the Sphingomonas genus but only in specific bacterial
public databases, and therefore its functional characteristics strains (Table 1) [27]. It has been suggested that the MC-
are difficult to be assigned. degradation trait was acquired by gene transfer during the
Table 1: Isolated microcystin degrading Bacteria.
16S rRNA gene MC-degrading Degradation
Bacteria Source Degradable analogues References
GenBank accession number genes of other toxins
Actinobacteria
FN392690, FN392691, FN392693,
Arthrobacter sp. Surface water MCLR, MCRR, MCLF — NOD [30]
FN392694, FN392695, FN392696
Brevibacterium sp. Surface water FN392692 MCLR, MCRR, MCLF — NOD [30]
Rhodococcus sp. Surface water FN392688, FN392689 MCLR, MCRR, MCLF — NOD [30]
Microbacterium sp. Lake water — MCLR Unknown [31]
MC-LR, MC-RR, MC-YR,
Bifidobacterium lactis Bb12 and 420 Probiotic strains — Unknown CYN [32, 33]
MC-LF, MC-LY, MC-LW
Bifidobacterium longum 46 Probiotic strains — Unknown CYN [32, 33]
MC-LF, MC-LY, MC-LW
Firmicutes
Bacillus sp. EMB/JSEM1 Artificial media FJ526332 MCLR, MCRR mlrA [34]
Lactobacillus rhamnosus GG and LC-705 Probiotic strains AY370682 Unknown CYN [32, 33]
MC-LF, MC-LY, MC-LW
𝛼-Proteobacteria
mlrA
mlrB
Sphingomonas sp. ACM-3962 or MJ-PV Surface water AF411072 MCLR, MCRR [35]
mlrC
mlrD
Sphingomonas sp. 7CY Lake water AB076083 Unknown [36]
Sphingomonas sp. B9 Lake water AB159609 MCLR, MCRR, NOD Unknown NOD
Sphingomonas sp. CBA4 AY920497 Unknown [37]
Sphingomonas sp. MD-1 Lake water AB110635 MCLR, MCRR, MCYR mlrA [27, 38]
Sphingomonas sp. MDB2 (Sphingosinicella sp.) Lake water AB219940 MCLR, MCRR, MCYR Unknown [39]
Sphingomonas sp. MDB3 (Sphingosinicella sp.) Lake water AB219941 MCLR, MCRR, MCYR Unknown [39]
Sphingomonas sp. Y2
Lake water NR 040927/AB084247 MCLR, MCRR, MCYR mlrA [40]
(Sphingosinicella microcystinivorans)
Sphingomonas stygia Lake water — MC-LR, MC-RR, MC-YR Unknown [38]
mlrA
Biological sand mlrB
Sphingopyxis sp. LH21 DQ112242 MCLR, MCLA [41–43]
filter mlrC
mlrD
MCLR, MCRR, MCYR,
Sphingopyxis sp. USTB-05 — mlrA [44–46]
MCLA
MCLR, MCRR, MCYR,
Sphingopyxis sp. TT25 Reservoir water JQ398614 mlrA [28]
MCLA
Sphingomonas sp. NV3 Lake water JN256930 Unknown mlrA Unpublished
mlrA
mlrB
Novosphingobium sp. THN1 Lake water HQ664117 MCLR [47]
mlrC
mlrD
3
4
Table 1: Continued.
16S rRNA gene MC-degrading Degradation
Bacteria Source Degradable analogues References
GenBank accession number genes of other toxins
Sphingopyxis sp. C-1 Water bloom AB161684 MCLR mlrA [48]
Rhizobium gallicum Lake water — MCLR Unknown [49]
𝛽-Proteobacteria
Burkholderia sp. Surface water DQ459360 MCLR Unknown [50]
Sludge from
Methylobacillus sp. J10 Cyanobacteria FJ418599 MCLR, MCRR — [51]
salvaged yard
Paucibacter toxinivorans Lake sediment NR 042941 MCLR, MCYR, NOD — [52]
Ralstonia solanacearum — — [53]
𝛾-Proteobacteria — —
Lake water active
Morganella morganii — MCLR — NOD [24]
anthracite biofilter
Pseudomonas aeruginosa Lake water — MCLR — [54]
Stenotrophomonas sp. EMS Lake water — MCLR, MCRR mlrA [29]
CYN: cylindrospermopsin, NOD: nodularin, MC: microcystin, A: alanine, F: phenylalanine, L: leucine, R: arginine, Y: tyrosine, W: tryptophan, mlr: mycrocystinase gene.
evolution of Sphingomonas, based on 16S rRNA gene analy- saxitoxin-degrading bacteria that had been isolated from the
ses that relate phylogenetically non-MC-degrading bacteria digestive tracts of blue mussels.
closer to the MC-degrader strain Y2 [27]. Moreover, Manage
et al. [30] managed to isolate bacteria that belonged to the 1.5. Anatoxins. Anatoxin-a is a low molecular weight neu-
Actinobacteria phylum (Arthrobacter sp., Brevibacterium sp., rotoxic alkaloid that is produced by cyanobacteria belong-
and Rhodococcus sp.) capable of degrading MC-LR. The mlr ing to the genera Anabaena, Aphanizomenon, Microcystis,
gene cluster was not detected, but this is the first report on Planktothrix, Raphidiopsis, Arthrospira, Cylindrospermum,
bacteria other than Sphingomonadaceae with the ability to Phormidium, Nostoc, and Oscillatoria [61]. There is limited
degrade MCs. Hu et al. [51] also isolated a Methylobacillus information on the biodegradation of anatoxin-a. A Pseu-
sp. strain from a cyanobacterial sludge with the ability to domonas sp. isolate was reportedly able to degrade anatoxin-a
completely degrade both MC-LR and MC-RR. To date, 20 [62], while Rapala et al. [63] also reported a significant
different MC-degrading bacteria have been isolated from (22–48%) reduction of anatoxin-a in sediments, but no
rivers, lakes, and biological filters, but without simultaneous further information on bacteria able to degrade or enzymatic
detection of the mlr gene cluster, which is considered to pathways and corresponding genes is available.
encode the hydrolytic proteins involved in the initial steps of
MC-degradation [41]. It seems that the presence of microbial
populations capable of degrading MCs and other peptides is 1.6. 𝛽-N-Methylamino-L-alanine (BMAA). BMAA is a highly
promoted by the prevalence of MC-containing blooms [20]. reactive nonessential amino acid which can be found either
Therefore, a greater diversity of bacterial genera might be able free or protein-bound. It is associated with amyotrophic lat-
to degrade MCs with as yet to be characterized degradation eral sclerosis and Parkinson dementia complex (ALS/PDC)
mechanisms. [64]. Although BMAA is the most recent cyanotoxin dis-
covered to date [65], it is produced from members from
all major cyanobacterial groups [66] and for this reason it
1.2. Nodularins. Nodularins are produced by Nodularia is believed to be widespread in freshwater systems. It has
spumigena and have been detected mostly in brackish habi- been demonstrated that it accumulates in higher trophic
tats [28, and references therein]. Their structure is simi- levels, including species that are consumed by humans [64].
lar to that of MCs with the difference that they consist As airborne algae and Cyanobacteria [67] have started to
of a pentapeptide [cyclo-(D-MeAsp1 -L-Arg2 -Adda3 -D-Glu4 - attract the scientific interest in a more focused way [68],
Mdhb5 )] [20]. Consequently, some bacteria which are capable it has been proposed recently that it can be dispersed via
of degrading MCs are also able to degrade NODs (Table 1), aerosolization of Cyanobacteria containing this toxin [69].
possibly due to the similar mode of action of MlrA that To date, no published data exist on the biodegradation
cleaves hydrolytically their cyclic structure at the Adda-Arg on BMAA.
peptide bond [28].
1.7. Nontoxic Nuisance Cyanobacterial Compounds. Apart
1.3. Cylindrospermopsin. Cylindrospermopsin (CYN) is a from the cyanotoxins, the water quality problems in freshwa-
group of alkaloid cytotoxins which are produced by Cylin- ter bodies are also related to the organoleptic traits of water
drospermopsis raciborskii, Umezakia natans, Anabaena bergii, which are easily perceived by consumers due to the altered
Anabaena lapponica, Aphanizomenon ovalisporum, Aphani- and usually unpleasant taste and odor. 2-Methylisoborneol
zomenon flosaquae, Raphidiopsis curvata, and Lyngbya wollei. (MIB) and geosmin, two cyclic aliphatic tertiary alcohols,
Three studies until now have demonstrated that CYN can be are the main responsible compounds that impart such
biodegraded in water habitats [55, 56]. Smith et al. [55] found problematic characteristics in waters (earthy/camphorous
that a linear relation between the biodegradation rate and odour), even though they both do not pose any known
the initial concentration of CYN could be established, while hazard for the human health [28]. MIB and geosmin
Mohamed and Alamri [56] reported the biodegradation can be produced by a range of Cyanobacteria including
of CYN by a Bacillus strain (AMRI-03) isolated from a Anabaena, Aphanizomenon, Geitlerinema, Symploca, Plank-
cyanobacterial bloom, with rapidly occurring degradation tothrix (Oscillatoria), Phormidium, Nostoc, Pseudanabaena,
rates, highly dependent on the initial CYN concentration. and Lyngbya [70]. There are several studies which elucidate
However, no defined metabolic pathway has been elucidated the ability of some bacteria to biodegrade both compounds
to date. [71–74], but no definite metabolic pathway is known. The
biodegradation reactions of both compounds seem to be
1.4. Saxitoxins. Saxitoxins form a group of alkaloid neurotox- mediated by monooxygenase enzymes in a way similar to the
ins that can be produced by dinoflagellates and many different biological Baeyer-Villiger reaction that takes place during the
cyanobacterial genera, including Anabaena, Raphidiopsis, biodegradation of camphor, a bicyclic ketone [75]. This is due
Lyngbya, Planktothrix (Oscillatoria), and Cylindrospermopsis to the similarity of their structure to alicyclic alcohols and
[18, 57]. While there are many studies that report the biotrans- ketones [75, 76]. The isolation and cloning of the cam operon
formation of saxitoxin variants to more toxic ones [58, 59], from a camphor-degrading strain of Pseudomonas putida
there is scarcity of literature regarding its biodegradation. confirmed this hypothesis [77], while camphor enrichment of
Donovan et al. [60] demonstrated an overall reduction of a camphor-degrading bacterial consortium has led to the iso-
saxitoxin mixture to 90% by seven unidentified but potential lation of all three available secondary metabolites of MIB [74].
Similarly, the biodegradation of geosmin has been reported The amino-acid sequences of the mlrA gene-carrying
by cyclohexanol-degrading strains of Nocardia and Acineto- bacterial strains were retrieved from the GenBank
bacter, equivalently to the Baeyer-Villiger reaction [75], while (http://www.ncbi.nlm.nih.gov/nuccore), and their phyloge-
Eaton and Sandusky [74] demonstrated geosmin biodegra- netic relationship was inferred using the neighbor-joining
dation by two terpene-degrading bacteria, Rhodococcus method. The evolutionary distances were computed using
wratislaviensis DLC-cam and Pseudomonas sp. SBR3-tpnb, the Poisson correction method. Bootstrapping was per-
but only after their induction with either camphor or terpene. formed with 1,000 replicates. Sphingopyxis sp. C-1󸀠 s amino-
In this paper, we review the available literature on the acid sequence was not available in the database and was
molecular diversity of bacterial strains which possess some translated by the nucleotide sequence with the EMBOSS
kind of cyanotoxin-degrading feature or, at least, are naturally (Sixpack) program (http://www.ebi.ac.uk/Tools/st/). There
associated with toxin-producing Cyanobacteria based on 16S were a total of 157 informative positions in the final
rRNA and mlrA gene sequence diversity. We aimed to depict dataset.
the major taxa that include such strains as a first step in
a focused approach for the isolation and biotechnological
development of microorganisms that can degrade cyanotox- 3. Results and Discussion
ins in natural and engineered aquatic systems.
Our intention was to identify ways to enhance the devel-
opment of biotechnological tools for biodegradation of
cyanotoxins by reviewing the current literature and mining
2. Materials and Methods databases for possible new prokaryotes of interest. The bulk
of the existing data deal with microcystins. Our analysis
16S rRNA gene sequences of bacterial isolates that either
revealed that the ubiquitous 𝛼- and 𝛽-Proteobacteria and
carry the mlrA gene or grow on/degrade MCs or both
Actinobacteria include the majority of potential cyanotoxin-
were retrieved from GenBank, and phylogenetic trees were
degrading bacteria (Figure 2). Based on their phylogenetic
constructed using the neighbor-joining algorithm based
taxonomy, the few recently found strains capable of degrading
on distances calculated by the Jukes-Cantor model and
microcystin or carrying the mlrA gene were closely affiliated
implemented in the MEGA5 software [78]. Bootstrap-
to known species or genera (Figure 2). The analysis of the
ping was performed with 1,000 replicates to assign con-
MlrA protein diversity (Figure 3) showed that all the available
fidence levels to the tree topology. The sequences were
sequences also belong to closely related members of the
screened for chimeras using the Bellerophon program
Sphingomonadaceae (𝛼-Proteobacteria), but there are also
at http://comp-bio.anu.edu.au/bellerophon/bellerophon.pl/.
several strains that cannot be assigned to known bacterial
No putative chimerical sequences were detected.
taxa.
The search of mlrA sequences in the GeneBank database
The majority of the taxa with cyanotoxin degradation
proved a not so straight forward procedure, as the search by
capability are known to have multiple other biodegrada-
using “mlrA” gives ambiguous results, with some of them
tion traits. Strains of the genus Rhodococcus are known
being totally irrelevant to the degradation gene. An initial
for degrading xenobiotics in various aquatic and terrestrial
search of the “mlrA” gene turned back several hundreds
habitats and also for producing surfactants [79]. The strains
of sequences. Most of these sequences are associated with that belonged to the Sphingomonadaceae family were affil-
Escherichia coli and representatives of the family of Enter- iated with the genera Sphingosinicella, Sphingopyxis, Sph-
obacteriaceae that correspond to genes unrelated to the ingomonas, and Novosphingobium. The sphingomonads are
degradation of microcystins. The majority of these genes are not new to biotechnological applications. They are involved
related to transcriptional regulators such as the HTH-type with novel catalyses, bioremediation, fossil fuel desulfuriza-
transcriptional regulator, the ABC-transporter, acetyl-xylan tion, novel enzymes, biotin, and polysaccharide production.
esterases, the McrR-like regulator A, and the merR family. Fil- Their ability to degrade hazardous organic compounds,
tering of these results by excluding all E. coli and Enterobac- for example, PCBs, creosote, pentachlorophenol, herbicides,
teriaceae bacteria (e.g., mlrA [All Fields] NOT “Escherichia” and the conversion of commonly occurring organics to
NOT “Shigella” NOT “Klebsiella” NOT “Enterobacter” NOT novel or specialty chemicals, is a key feature for the group,
“Salmonella” NOT “Cronobacter” NOT “Erwinia” NOT “Cit- especially in contaminated soils and sediments [80, and
robacter” NOT “Pantoea”) shrinks the results to a few tenths references therein]. Despite their biotechnological signifi-
of sequences that still include some irrelevant sequences cance and their widespread distribution, their ecology is
similar to those described above. After this stage, manual insufficiently studied. This could be due to their ability to
fishing of the mlrA sequences was necessary. Two sequences utilize a wide array of organic-often refractory substances,
of uncultured clones (FJ438526 and FJ438527) were excluded their metabolic diversity, and their ability to grow in olig-
due to their short length (120 bp). Four more sequences otrophic conditions, with the latter being more obvious in
of uncultured clones were excluded (AB600656, AB600658, the marine environment. These features also hinder their
AB600663, and AB600668) due to the high number of successful cultivation. The metabolic diversity of this group
potential but short amino acid (1–157 aa) ORFs into which is indicated by the lack of a culture medium specific for
they could be translated. The MlrA sequences we used sphingomonads, leaving the molecular approaches more
consisted of 184 or more amino acids. appropriate for their study. They are more frequently found in
Bifidobacteriaceae
Bifidobacteriales
100 Bifidobacterium longum subsp. longum [KC429787]
Bifidobacterium animalis subsp. lactis [NR 074884]
59 Rhodococcus sp. C1 [FN392688]
Nocardiaceae
92 Rhodococcus sp. C3 [FN392689]
81
99
Rhodococcus fascians [KC494315]
Rhodococcus corynebacterioides [NR 041873]
ae
Actinobacteria
Actinobacteria
ace
92 Brevibacterium sp. F3 [FN392692]
Actinomycetales
i
ter
98 Brevibacterium aureum [AY299092]
bac
63
vi
Brevibacterium avium [NR 026485]
Bre
Arthrobacter sp. C6 [FN392690]
65
98 Arthrobacter sp. R1 [FN392694]
Micrococcaceae
Arthrobacter sp. F10 [FN392691]
98 Arthrobacter sp. R4 [FN392695]

50
Arthrobacter sp. F7 [FN392693]
66
Arthrobacter sp. R6 [FN392696]
Bacillales Lactobacillales
Lactobacillaceae
83 Lactobacillus rhamnosus GG (ATCC 53103) [AY370682]
98 Lactobacillus casei [KC222511]
Firmicutes
Lactobacillus plantarum [KC429782]
Bacilli
90
Bacillaceae
Bacillus licheniformis [KC429647]
Bacillus subtilis JSEM1/EMB [FJ526332]
98
91 Bacillus amyloliquefaciens [JX157134]
56 Sphingomonas 7CY [AB076083]
Sphingomonas sp. B9 [AB159609]
79 Sphingosinicella microcystinivorans MDB2 [AB219940]
Sphingosinicella microcystinivorans Y2 [AB084247]
Sphingomonadaceae
100
Sphingomonadales
𝛼-Proteobacteria
Sphingosinicella microcystinivorans MDB3 [AB219941]
Proteobacteria
Sphingosinicella microcystinivorans Y2 [NR 040927]
Sphingomonas sp. MD-1 [AB110635]
100
84 Novosphingobium sp. THN1 [HQ664117]
99 Sphingomonas sp. ACM-3962/MJ-PV [AF411072]
Sphingopyxis sp. LH21 [DQ112242]
Sphingopyxis sp. TT25 [JQ398614]
Burkholderiaceae
80
65 Sphingopyxis sp. C-1 [AB161684]
Unclassified
Methylobacillus sp. J10 [FJ418599] Methylophilaceae Methylophilales
𝛽-Proteobacteria
Paucibacter toxinivorans 2C20 (DSM 16998) [NR 042941]
Burkholderiales
99
Burkholderia sp. [AB622656]
59
Ralstonia solanacearum [JX294073]
93 Burkholderiaceae
99 Ralstonia solanacearum [JX826635]
Anabaena cylindrica [HE975014] Cyanobacteria; Nostocales; Nostocaceae
Thermotoga maritima [NC 000853]
0.02
Figure 2: Phylogenetic tree of the 16S rRNA gene sequences of isolates that either carry the mlrA gene (in green) or degrade MC (in red) or
both (in blue), based on the neighbour-joining method and a Jukes-Cantor distance matrix. One thousand bootstrap analyses were conducted,
and percentages greater than 50% are indicated at nodes. The numbers in brackets are GenBank accession numbers. Thermotoga maritima
was used as an outgroup. Scale bar represents 2% estimated distance.
freshwater habitats like rivers, ponds, lakes, and groundwater, enzymes (along with their products) are evolved rapidly
but the majority of the available strains have been isolated through multiple mechanisms [82]. On the other hand, the
from contaminated to heavily contaminated sites [80, and ability of the cooccurring bacterioplankton in cyanobacterial
references therein]. Since these strains have been found water blooms to degrade/assimilate toxins (e.g., [14, 15])
to carry the mlrA gene and/or grow on microcystin, it is along with the metabolic diversity, but insufficiently studied
plausible to consider them as some of the most active—and ecology, of bacteria such as the sphingomonads [80, and
promising for future applications—players in the degradation references therein] is well documented. These two adversary
of microcystin. forces might have an important ecological function and
Microcystins and nodularins’ biosynthesis is also carried might be able to drive the coevolution of the NRPS/PKS-I
out by nonribosomal peptide synthetases (NRPS) and type enzymes and the cyanotoxin biodegradation pathways in a
I polyketide synthases (PKS-I) [81]. These extraordinary water bloom, leading to a “race of arms” against cyanotoxins.
Stenotrophomonas sp. EMS [ADB03118]

53
Sphingopyxis sp. C-1 [BAI47770]
80 Sphingopyxis sp. C-1 MlrA [AB161685]-EMBOSS 001 1 ORF2
Sphingomonadales; Sphingomonadaceae
Sphingopyxis sp. LH21 [AAZ16519]
Proteobacteria; 𝛼-Proteobacteria;
Sphingopyxis sp. TT25 [AFC37495]
81
Sphingomonas sp. USTB-05 [ADK25053]
Sphingosinicella microcystinivorans Y2 [BAC82713]
Uncultured bacterium [BAL15379]
Sphingomonas sp. ACM-3962 [AAL10286]
Novosphingobium sp. THN1 [AEC46646]
93
Sphingomonas sp. MD-1 [BAC82712]
74
Sphingomonas sp. NV3 [AER41696]
57
65
57 Uncultured bacterium [BAL15377]

Anabaena cylindrica PCC 7122 [YP 007157074]
0.02
Figure 3: Phylogenetic tree MlrA amino acid sequences retrieved from the GenBank database, based on the neighbor-joining method.
The evolutionary distances were computed using the Poisson correction method. One thousand bootstrap analyses were conducted, and
percentages greater than 50% are indicated at nodes. The numbers in brackets are GenBank accession numbers. Scale bar represents 2%
estimated distance.
Moreover, the retrieval of bacteria such as Actinobacteria that riboflavin [79, 83]. The ability of Arthrobacter to degrade
lack the known mlr gene cluster [30] implies the existence microcystins renders this genus an important one in future
of more genes, and thus pathways, associated with the biotechnological quests. Similar features are also found in
degradation of these toxins. Brevibacterium spp. (Figure 2), a genus that is taxonomically
The genus Arthrobacter (Figure 2), originating mostly and physiologically similar to Arthrobacter, and until recently,
from soil, is known in biotechnology mostly because of its frequently confused with it [81].
nutritional versatility of growing on a wide array of organic Bacillus spp. have been one of the pillars of biotech-
compounds, including herbicides, pesticides, n-alkanes, aro- nology for several decades now. Although they possess
matic compounds, and lower alcohols [83, and references numerous biodegradation competences, to date only one
therein], its potential for bioremediating their heavy metals isolate (Figure 2) seems to be associated with the microcystin
(Cd, Co, Cu, Ni, and Pb) [79], and also for the produc- degradation. This by no means diminishes the biotechnologi-
tion of surfactants, glutamic acid, 𝛼-ketoglutaric acid, and cal value of this genus; the reason for the underrepresentation
of Bacillus spp. is possibly due to the fact that the Firmicutes is Acknowledgments
a minor phylum in freshwater habitats [84] where most toxic
cyanobacterial blooms occur. The authors thank the two anonymous reviewers for their
Apart from the above-mentioned strains with already comments on the originally submitted version of this paper.
well-recognised biodegradation potential, the Methylobacil-
lus sp. and Paucibacter sp. (Figure 2) can be considered References
important emerging genera of microcystin degraders. This
suggests that with ongoing research and the application [1] B. A. Whitton, Ed., Ecology of Caynobacteria II. Their Diversity
of appropriate methodology new taxa of degraders can be in Space and Time, Springer, Dordrecht, The Netherlands, 2012.
revealed. [2] R. M. M. Abed, S. Dobretsov, and K. Sudesh, “Applications of
It has been shown that freshwater harbors a vast and cyanobacteria in biotechnology,” Journal of Applied Microbiol-
distinguishable bacterial diversity from other aquatic habitats ogy, vol. 106, no. 1, pp. 1–12, 2009.
[84, 85]. However, our meta-analysis showed that, to date, [3] D. C. Ducat, J. C. Way, and P. A. Silver, “Engineering cyanobac-
only a few bacterial species bear the mlrA gene, a proxy for teria to generate high-value products,” Trends in Biotechnology,
vol. 29, no. 2, pp. 95–103, 2011.
microcystin degradation potential. Moreover, several of these
bacteria are not abundant in the ecosystems where micro- [4] T. L. Simmons, E. Andrianasolo, K. McPhail, P. Flatt, and W.
H. Gerwick, “Marine natural products as anticancer drugs,”
cystins and other cyanotoxins are found. This could partially
Molecular Cancer Therapeutics, vol. 4, no. 2, pp. 333–342, 2005.
explain our perceived concept of the nonlabile nature of
[5] M. Pelaez, M. G. Antoniou, X. He et al., “Sources and occurrence
microcystins and possibly of other cyanotoxins. However, of cyanotoxins worldwide,” in Xenobiotics in the Urban Water
it could also be related to the methodological approaches Cycle, D. Fatta-Kassinos, K. Bester, and K. Kümmerer, Eds., pp.
used so far to isolate such degraders. The most common 101–127, Springer, New York, NY, USA, 2010.
practice involves culture-dependent approaches with the [6] H. Mazur-Marzec and M. Pliński, “Do toxic cyanobacteria
toxin being the sole carbon and/or energy source (e.g., [39, blooms pose a threat to the Baltic ecosystem?” Oceanologia, vol.
48]). We believe that novel/recent molecular methodologies 51, no. 3, pp. 293–319, 2009.
provide valuable complementary information because they [7] H. Kato, K. Tsuji, and K. Harada, “Microbial degradation of
overcome culturing limitations, and in silico approaches are cyclic peptides produced by bacteria,” Journal of Antibiotics, vol.
not dependent on metabolic traits of the taxa containing 62, no. 4, pp. 181–190, 2009.
genes for microcystin degradation. [8] I. R. Falconer, T. Buckley, and M. T. Runnegar, “Biological
Nowadays there are several omics methodologies that half-life, organ distribution and excretion of 125 I-labelled toxic
peptide from the blue-green alga Microcystis aeruginosa,” Aus-
can be tailored to specific scientific quests. Future inves-
tralian Journal of Biological Sciences, vol. 39, no. 1, pp. 17–21,
tigations targeting microorganisms with cyanotoxin degra- 1986.
dation could include (a) genomic analysis of the available [9] W. P. Brooks and G. A. Codd, “Distribution of Microcystic aerug-
strains, like the case of Sphingosinicella microcystinivorans inosa peptide toxin and interactions with hepatic microsomes in
[39]; gene mining of such genomes could depict the relevant mice,” Pharmacology and Toxicology, vol. 60, no. 3, pp. 187–191,
degradation pathways; (b) metagenomic libraries of habitats 1987.
where the sphingomonads are either abundant or with bac- [10] N. A. Robinson, J. G. Pace, C. F. Matson, G. A. Miura, and W. B.
terial communities occurring in close association with toxic Lawrence, “Tissue distribution, excretion and hepatic biotrans-
cyanobacterial blooms and followed by metatranscriptomics formation of microcystin-LR in mice,” Journal of Pharmacology
libraries of the same samples where the MlrA and other and Experimental Therapeutics, vol. 256, no. 1, pp. 176–182, 1991.
related enzymes have been proved to be transcribed (c) [11] H. Li, P. Xing, and Q. L. Wu, “The high resilience of the
single-cell genomics (SCG). SCG is of particular interest bacterioplankton community in the face of a catastrophic
to biotechnology and bioprospecting due to its ability to disturbance by a heavy Microcystis bloom,” FEMS Microbiology
attribute specific—often novel—biochemical pathways to Ecology, vol. 82, pp. 192–201, 2012.
specific cells/species [31]. Recently, a SSG study revealed for [12] K. A. Kormas, E. Vardaka, M. Moustaka-Gouni et al., “Molec-
the first time a great extent of the metabolic potential of a ular detection of potentially toxic cyanobacteria and their
ubiquitous freshwater actinobacterial species [86]. In the case associated bacteria in lake water column and sediment,” World
of cyanotoxin degraders, for example, it would be feasible to Journal of Microbiology and Biotechnology, vol. 26, no. 8, pp.
1473–1482, 2010.
assign the biodegradation pathway of cyanotoxins to specific
[13] L. Shi, Y. Cai, F. Kong, and Y. Yu, “Specific association between
bacteria (i.e., single-cells) from any toxic cyanobacterial water
bacteria and buoyant Microcystis colonies compared with other
bloom, regardless of their cultivability. The application of bulk bacterial communities in the eutrophic Lake Taihu, China,”
such approaches, along with the development of standardized Environmental Microbiology Reports, vol. 4, pp. 669–678, 2012.
and easy-to-use analytical methods for the measurement of [14] K. A. Berg, C. Lyra, K. Sivonen et al., “High diversity of
multiple cyanotoxins from the same sample, could speed cultivable heterotrophic bacteria in association with cyanobac-
progress towards the standardized usage of specific cyan- terial water blooms,” ISME Journal, vol. 3, no. 3, pp. 314–325,
otoxin degraders. Finally, special attention should be paid to 2009.
other toxins than the microcystins, as some of these are, at [15] C. Dziallas and H. Grossart, “Microbial interactions with the
least, of equal potency, distribution, and persistence in the cyanobacterium Microcystis aeruginosa and their dependence
environment. on temperature,” Marine Biology, vol. 159, pp. 2389–2398, 2012.
[16] E. O. Casamayor and C. Borrego, “Archaea,” in Encyclopedia of bacteria,” FEMS Microbiology Letters, vol. 270, no. 1, pp. 27–33,
Inland Waters, G. E. Likens, Ed., vol. 3, pp. 167–181, Elsevier, 2007.
Oxford, UK, 2009. [33] S. M. K. Nybom, S. J. Salminen, and J. A. O. Meriluoto, “Specific
[17] P. Xing, H. Li, Q. Liu, and L. Zheng, “Composition of the strains of probiotic bacteria are efficient in removal of several
archaeal community involved in methane production during different cyanobacterial toxins from solution,” Toxicon, vol. 52,
the decomposition of Microcystis blooms in the laboratory,” no. 2, pp. 214–220, 2008.
Canadian Journal of Microbiology, vol. 58, pp. 1153–1158, 2012. [34] L. Hu, F. Zhang, C. Liu, and M. Wang, “Biodegradation of
[18] K. Sivonen and J. Jones, “Cyanobacterial toxins,” in Toxic microcystins by Bacillus sp. strain EMB,” Energy Procedia, vol.
Cyanobacteria in Water. A Guide to Their Public Health Conse- 16, pp. 2054–2059, 2012.
quences, Monitoring and Management, I. Chorus and J. Bartram, [35] D. G. Bourne, G. J. Jones, R. L. Blakeley, A. Jones, A. P. Negri, and
Eds., E & FN Spon on behalf of the World Health Organization, P. Riddles, “Enzymatic pathway for the bacterial degradation of
1999. the cyanobacterial cyclic peptide toxin microcystin LR,” Applied
[19] I. R. Falconer, Cyanobacterial Toxins in Drinking Water Supplies: and Environmental Microbiology, vol. 62, no. 11, pp. 4086–4094,
Cylindrospermopsins and Microcystins, CRC Press, Boca Raton, 1996.
Fla, USA, 2005. [36] H. Ishii, M. Nishijima, and T. Abe, “Characterization of
[20] C. Edwards and L. A. Lawton, “Bioremediation of cyanotoxins,” degradation process of cyanobacterial hepatotoxins by a gram-
Advances in Applied Microbiology, vol. 67, pp. 109–129, 2009. negative aerobic bacterium,” Water Research, vol. 38, no. 11, pp.
[21] S. Imanishi, H. Kato, M. Mizuno, K. Tsuji, and K. Harada, “Bac- 2667–2676, 2004.
terial degradation of microcystins and nodularin,” Chemical [37] A. M. Valeria, E. J. Ricardo, P. Stephan, and D. A. Wunderlin,
Research in Toxicology, vol. 18, no. 3, pp. 591–598, 2005. “Degradation of microcystin-RR by Sphingomonas sp. CBA4
[22] A. A. de la Cruz, M. G. Antoniou, A. Hiskia et al., “Can isolated from San Roque reservoir (Córdoba-Argentina),”
we effectively degrade microcystins?—implications on human Biodegradation, vol. 17, no. 5, pp. 447–455, 2006.
health,” Anti-Cancer Agents in Medicinal Chemistry, vol. 11, no. [38] T. Saitou, N. Sugiura, T. Itayama, Y. Inamori, and M. Mat-
1, pp. 19–37, 2011. sumura, “Degradation characteristics of microcystins by iso-
[23] D. K. Kim, K. E. O’Shea, and W. J. Cooper, “Degradation of lated bacteria from Lake Kasumigaura,” Journal of Water Supply,
MTBE and related gasoline oxygenates in aqueous media by vol. 52, no. 1, pp. 13–18, 2003.
ultrasound irradiation,” Journal of Environmental Engineering, [39] T. Maruyama, H. Park, K. Ozawa et al., “Sphingosinicella
vol. 128, no. 9, pp. 806–812, 2002. microcystinivorans gen. nov., sp. nov., a microcystin-degrading
[24] L. Eleuterio and J. R. Batista, “Biodegradation studies and bacterium,” International Journal of Systematic and Evolutionary
sequencing of microcystin-LR degrading bacteria isolated from Microbiology, vol. 56, no. 1, Article ID 63789, pp. 85–89, 2006.
a drinking water biofilter and a fresh water lake,” Toxicon, vol. [40] H. Park, Y. Sasaki, T. Maruyama, E. Yanagisawa, A. Hiraishi,
55, no. 8, pp. 1434–1442, 2010. and K. Kato, “Degradation of the cyanobacterial hepatotoxin
[25] D. G. Bourne, P. Riddles, G. J. Jones, W. Smith, and R. microcystin by a new bacterium isolated from a hypertrophic
L. Blakeley, “Characterisation of a gene cluster involved in lake,” Environmental Toxicology, vol. 16, no. 4, pp. 337–343, 2001.
bacterial degradation of the cyanobacterial toxin microcystin [41] D. Hoefel, L. Ho, P. T. Monis, G. Newcombe, and C. P.
LR,” Environmental Toxicology, vol. 16, no. 6, pp. 523–534, 2001. Saint, “Biodegradation of geosmin by a novel Gram-negative
[26] G. J. Jones and P. T. Orr, “Release and degradation of microbacterium; isolation, phylogenetic characterisation and degra-
cystin following algicide treatment of a Microcystis aeruginosa dation rate determination,” Water Research, vol. 43, no. 11, pp.
bloom in a recreational lake, as determined by HPLC and 2927–2935, 2009.
protein phosphatase inhibition assay,” Water Research, vol. 28, [42] L. Ho, A. Gaudieux, S. Fanok, G. Newcombe, and A. R.
no. 4, pp. 871–876, 1994. Humpage, “Bacterial degradation of microcystin toxins in
[27] T. Saito, K. Okano, H. Park et al., “Detection and sequencing of drinking water eliminates their toxicity,” Toxicon, vol. 50, no. 3,
the microcystin LR-degrading gene, mlrA, from new bacteria pp. 438–441, 2007.
isolated from Japanese lakes,” FEMS Microbiology Letters, vol. [43] L. Ho, D. Hoefel, C. P. Saint, and G. Newcombe, “Isolation and
229, no. 2, pp. 271–276, 2003. identification of a novel microcystin-degrading bacterium from
[28] L. Ho, E. Sawade, and G. Newcombe, “Biological treatment a biological sand filter,” Water Research, vol. 41, no. 20, pp. 4685–
options for cyanobacteria metabolite removal—a review,” Water 4695, 2007.
Research, vol. 46, no. 5, pp. 1536–1548, 2012. [44] J. Wang, P. Wu, J. Chen, and H. Yan, “Biodegradation of
[29] J. Chen, L. B. Hu, W. Zhou et al., “Degradation of microcystin- microcystin-RR by a new isolated Sphingopyxis sp. USTB-05,”
LR and RR by a Stenotrophomonas sp. strain EMS isolated from Chinese Journal of Chemical Engineering, vol. 18, no. 1, pp. 108–
Lake Taihu, China,” International Journal of Molecular Sciences, 112, 2010.
vol. 11, no. 3, pp. 896–911, 2010. [45] M. Zhang, G. Pan, and H. Yan, “Microbial biodegradation
[30] P. M. Manage, C. Edwards, B. K. Singh, and L. A. Lawton, of microcystin-RR by bacterium Sphingopyxis sp. USTB-05,”
“Isolation and identification of novel microcystin-degrading Journal of Environmental Sciences, vol. 22, no. 2, pp. 168–175,
bacteria,” Applied and Environmental Microbiology, vol. 75, no. 2010.
21, pp. 6924–6928, 2009. [46] H. Yan, J. Wang, J. Chen, W. Wei, H. Wang, and H. Wang, “Char-
[31] R. Stepanauskas, “Single cell genomics: an individual look at acterization of the first step involved in enzymatic pathway for
microbes,” Current Opinion in Microbiology, vol. 15, pp. 613–620, microcystin-RR biodegraded by Sphingopyxis sp. USTB-05,”
2012. Chemosphere, vol. 87, no. 1, pp. 12–18, 2012.
[32] S. M. K. Nybom, S. J. Salminen, and J. A. O. Meriluoto, “Removal [47] Y. Jiang, J. Shao, X. Wu, Y. Xu, and R. Li, “Active and silent
of microcystin-LR by strains of metabolically active probiotic members in the mlr gene cluster of a microcystin-degrading
bacterium isolated from Lake Taihu, China,” FEMS Microbiol- [64] S. Jonasson, J. Eriksson, L. Berntzon et al., “Transfer of a
ogy Letters, vol. 322, no. 2, pp. 108–114, 2011. cyanobacterial neurotoxin within a temperate aquatic ecosys-
[48] K. Okano, K. Shimizu, Y. Kawauchi et al., “Characteristics of tem suggests pathways for human exposure,” Proceedings of the
a microcystin-degrading bacterium under alkaline environ- National Academy of Sciences of the United States of America,
mental conditions,” Journal of Toxicology, vol. 2009, Article ID vol. 107, no. 20, pp. 9252–9257, 2010.
954291, 8 pages, 2009. [65] S. J. Murch, P. A. Cox, and S. A. Banack, “A mechanism
[49] A. Ramani, K. Rein, K. G. Shetty, and K. Jayachandran, for slow release of biomagnified cyanobacterial neurotoxins
“Microbial degradation of microcystin in Florida’s freshwaters,” and neurodegenerative disease in Guam,” Proceedings of the
Biodegradation, vol. 23, no. 1, pp. 35–45, 2012. National Academy of Sciences of the United States of America,
[50] G. A. F. Lemes, R. Kersanach, L. da S. Pinto, O. A. Dellagostin, J. vol. 101, no. 33, pp. 12228–12231, 2004.
S. Yunes, and A. Matthiensen, “Biodegradation of microcystins [66] P. A. Cox, S. A. Banack, S. J. Murch et al., “Diverse taxa
by aquatic Burkholderia sp. from a South Brazilian coastal of cyanobacteria produce 𝛽-N-methylamino-L-alanine, a neu-
lagoon,” Ecotoxicology and Environmental Safety, vol. 69, no. 3, rotoxic amino acid,” Proceedings of the National Academy of
pp. 358–365, 2008. Sciences of the United States of America, vol. 102, no. 14, pp. 5074–
5078, 2005.
[51] L. B. Hu, J. D. Yang, W. Zhou, Y. F. Yin, J. Chen, and Z. Q.
Shi, “Isolation of a Methylobacillus sp. that degrades microcystin [67] S. Genitsaris, K. A. Kormas, and M. Moustaka-Gouni, “Air-
toxins associated with cyanobacteria,” New Biotechnology, vol. borne algae and cyanobacteria: occurrence and related health
26, no. 3-4, pp. 205–211, 2009. effects,” Frontiers in Bioscience, vol. 3, pp. 772–787, 2011.
[52] J. Rapala, K. A. Berg, C. Lyra et al., “Paucibacter toxinivorans [68] S. Genitsaris, M. Moustaka-Gouni, and K. A. Kormas, “Air-
gen. nov., sp. nov., a bacterium that degrades cyclic cyanobac- borne microeukaryote colonists in experimental water con-
terial hepatotoxins microcystins and nodularin,” International tainers: diversity, succession, life histories and established food
Journal of Systematic and Evolutionary Microbiology, vol. 55, no. webs,” Aquatic Microbial Ecology, vol. 62, no. 2, pp. 139–152, 2011.
4, pp. 1563–1568, 2005. [69] E. W. Stommel, N. C. Field, and T. A. Caller, “Aerosolization of
[53] H. Yan, G. Pan, H. Zou, X. Li, and H. Chen, “Effective removal of cyanobacteria as a risk factor for amyotrophic lateral sclerosis,”
microcystins using carbon nanotubes embedded with bacteria,” Medical Hypotheses, vol. 80, pp. 142–145, 2012.
Chinese Science Bulletin, vol. 49, no. 16, pp. 1694–1698, 2004. [70] F. Jüttner and S. B. Watson, “Biochemical and ecological control
of geosmin and 2-methylisoborneol in source waters,” Applied
[54] S. Takenaka and M. F. Watanabe, “Microcystin LR degradation
and Environmental Microbiology, vol. 73, no. 14, pp. 4395–4406,
by Pseudomonas aeruginosa alkaline protease,” Chemosphere,
2007.
vol. 34, no. 4, pp. 749–757, 1997.
[71] A. Saito, T. Tokuyama, A. Tanaka, T. Oritani, and K. Fuchigami,
[55] M. J. Smith, G. R. Shaw, G. K. Eaglesham, L. Ho, and J. D.
“Microbiological degradation of (-)-geosmin,” Water Research,
Brookes, “Elucidating the factors influencing the biodegrada-
vol. 33, no. 13, pp. 3033–3036, 1999.
tion of cylindrospermopsin in drinking water sources,” Environ-
mental Toxicology, vol. 23, no. 3, pp. 413–421, 2008. [72] D. Hoefel, L. Ho, W. Aunkofer et al., “Cooperative biodegra-
dation of geosmin by a consortium comprising three gram-
[56] Z. A. Mohamed and S. A. Alamri, “Biodegradation of cylindros-
negative bacteria isolated from the biofilm of a sand filter
permopsin toxin by microcystin-degrading bacteria isolated
column,” Letters in Applied Microbiology, vol. 43, no. 4, pp. 417–
from cyanobacterial blooms,” Toxicon, vol. 60, pp. 1390–1395,
423, 2006.
2012.
[73] R. W. Eaton and P. Sandusky, “Biotransformations of 2-
[57] S. A. Murray, T. K. Mihali, and B. A. Neilan, “Extraordinary methylisoborneol by camphor-degrading bacteria,” Applied and
conservation, gene loss, and positive selection in the evolution Environmental Microbiology, vol. 75, no. 3, pp. 583–588, 2009.
of an ancient neurotoxin,” Molecular Biology and Evolution, vol.
[74] R. W. Eaton and P. Sandusky, “Biotransformations of
28, no. 3, pp. 1173–1182, 2011.
(+/−)−geosmin by terpene-degrading bacteria,” Biodegrada-
[58] Y. Kotaki, “Screening of bacteria which convert gonyautoxin 2,3 tion, vol. 21, no. 1, pp. 71–79, 2009.
to saxitoxin,” Nippon Suisan Gakkaishi, vol. 55, p. 1239, 1989.
[75] P. W. Trudgill, “Microbial degradation of the alicyclic ring:
[59] N. Kayal, G. Newcombe, and L. Ho, “Investigating the fate of structural relationships and metabolic pathways,” in Microbial
saxitoxins in biologically active water treatment plant filters,” Degradation of Organic Compounds, D. T. Gibson, Ed., pp. 131–
Environmental Toxicology, vol. 23, no. 6, pp. 751–755, 2008. 180, Marcel Dekker, New York, NY, USA, 1984.
[60] C. J. Donovan, J. C. Ku, M. A. Quilliam, and T. A. Gill, “Bacterial [76] B. E. Rittmann, C. J. Gantzer, and A. Montiel, “Biological
degradation of paralytic shellfish toxins,” Toxicon, vol. 52, no. 1, treatment to control taste-and-odor compounds in drinking
pp. 91–100, 2008. water treatment,” in Advances in Taste-and-Odor Treatment
[61] J. Osswald, S. Rellán, A. Gago, and V. Vasconcelos, “Toxicology and Control, I. H. Suffet, J. Mallevialle, and E. Kawczynski,
and detection methods of the alkaloid neurotoxin produced by Eds., American Water Works Association Research Foundation,
cyanobacteria, anatoxin-a,” Environment International, vol. 33, Denver, Colo, USA, 1995.
no. 8, pp. 1070–1089, 2007. [77] E. Oikawa, A. Shimizu, and Y. Ishibashi, “2-methylisoborneol
[62] J. Kiviranta, K. Sivonen, K. Lahti, R. Luukkainen, and S. I. degradation by the cam operon from Pseudomonas putida
Niemelae, “Production and biodegradation of cyanobacterial PpG1,” Water Science and Technology, vol. 31, no. 11, pp. 79–86,
toxins—a laboratory study,” Archiv Für Hydrobiologie, vol. 121, 1995.
pp. 281–294, 1991. [78] K. Tamura, D. Peterson, N. Peterson, G. Stecher, M. Nei, and
[63] J. Rapala, K. Lahti, K. Sivonen, and S. I. Niemela, “Biodegrad- S. Kumar, “MEGA5: molecular evolutionary genetics analysis
ability and adsorption on lake sediments of cyanobacterial using maximum likelihood, evolutionary distance, and max-
hepatotoxins and anatoxin-a,” Letters in Applied Microbiology, imum parsimony methods,” Molecular Biology and Evolution,
vol. 19, no. 6, pp. 423–428, 1994. vol. 28, no. 10, pp. 2731–2739, 2011.
[79] M. Fingerman and R. Nagabhushanam, Bioremediation of

Aquatic and Terrestrial Ecosystem, Science Publishers, Enfield,
NH, USA, 2005.
[80] D. L. Balkwill, J. K. Fredrickson, and M. F. Romine, “Sphin-
gomonas and related genera,” in The Prokaryotes. Third Edition.
A Handbook on the Biology of Bacteria, M. Dworkin, S. Falkow,
E. Rosenberg, K.-H. Schleifer, and E. Stackenbrandt, Eds., vol.
7 of Proteobacteria: Delta and Epsilon subclasses. Deeply rooting
Bacteria, pp. 605–629, Springer, Berlin, Germany, 2006.
[81] K. Nikolouli and D. Mossialos, “Bioactive compounds syn-
thesized by non-ribosomal peptide synthetases and type-I
polyketide synthases discovered through genome-mining and
metagenomics,” Biotechnology Letters, vol. 34, no. 8, pp. 1393–
1403, 2012.
[82] G. D. Amoutzias, Y. Van de Peer, and D. Mossialos, “Evolu-
tion and taxonomic distribution of nonribosomal peptide and
polyketide synthases,” Future Microbiology, vol. 3, no. 3, pp. 361–
370, 2008.
[83] D. Jones and R. M. Keddie, “The genus Arthrobacter,” in
The Prokaryotes. Third Edition. A Handbook on the Biology of
Bacteria, M. Dworki, S. Falkow, E. Rosenberg, K.-H. Schleifer,
and E. Stackenbrandt, Eds., vol. 3 of Proteobacteria: Delta
and Epsilon subclasses. Deeply rooting Bacteria, pp. 945–960,
Springer, Berlin, Germany, 2006.
[84] R. J. Newton, S. E. Jones, A. Eiler, K. D. McMahon, and S.
Bertilsson, “A guide to the natural history of freshwater lake
bacteria,” Microbiology and Molecular Biology Reviews, vol. 75,
no. 1, pp. 14–49, 2011.
[85] A. Barberán and E. O. Casamayor, “Global phylogenetic com-
munity structure and 𝛽-diversity patterns in surface bacterio-
plankton metacommunities,” Aquatic Microbial Ecology, vol. 59,
no. 1, pp. 1–10, 2010.
[86] S. L. Garcia, K. D. McMahon, M. Martinez-Garcia et al.,
“Metabolic potential of a single cell belonging to one of the most
abundant lineages in freshwater bacterioplankton,” The ISME
Journal, vol. 7, pp. 137–147, 2013.
http://dx.doi.org/10.1155/2013/248078
Research Article
Potential for Plant Growth Promotion of
Rhizobacteria Associated with Salicornia Growing in
Tunisian Hypersaline Soils
Francesca Mapelli,1 Ramona Marasco,1 Eleonora Rolli,1 Marta Barbato,1 Hanene Cherif,2
Amel Guesmi,2 Imen Ouzari,2 Daniele Daffonchio,1 and Sara Borin1
1
DeFENS, Department of Food, Environment and Nutritional Sciences (DeFENS), University of Milan, Via Celoria 2,
20133 Milan, Italy
2
Laboratory of Microorganisms and Active Biomolecules, University of Tunis El Manar, Campus Universitaire, 2092 Tunis, Tunisia
Correspondence should be addressed to Sara Borin; sara.borin@unimi.it
Received 15 March 2013; Revised 30 April 2013; Accepted 3 May 2013
Copyright © 2013 Francesca Mapelli et al. This is an open access article distributed under the Creative Commons Attribution
cited.
Soil salinity and drought are among the environmental stresses that most severely affect plant growth and production around
the world. In this study the rhizospheres of Salicornia plants and bulk soils were collected from Sebkhet and Chott hypersaline
ecosystems in Tunisia. Depiction of bacterial microbiome composition by Denaturing Gradient Gel Electrophoresis unveiled the
occurrence of a high bacterial diversity associated with Salicornia root system. A large collection of 475 halophilic and halotolerant
bacteria was established from Salicornia rhizosphere and the surrounding bulk soil, and the bacteria were characterized for the
resistance to temperature, osmotic and saline stresses, and plant growth promotion (PGP) features. Twenty Halomonas strains
showed resistance to a wide set of abiotic stresses and were able to perform different PGP activities in vitro at 5% NaCl, including
ammonia and indole-3-acetic acid production, phosphate solubilisation, and potential nitrogen fixation. By using a gfp-labelled
strain it was possible to demonstrate that Halomonas is capable of successfully colonising Salicornia roots in the laboratory
conditions. Our results indicated that the culturable halophilic/halotolerant bacteria inhabiting salty and arid ecosystems have
a potential to contribute to promoting plant growth under the harsh salinity and drought conditions. These halophilic/halotolerant
strains could be exploited in biofertilizer formulates to sustain crop production in degraded and arid lands.
1. Introduction Among abiotic stresses soil salinity is one of the strongest

factors affecting plant growth and yield [7]. Conditions of
The influence of microbes on plant fitness has been rec- high salt concentrations in the soil are very frequent in arid
ognized both in conventional and extreme habitats, where and semiarid regions on Earth, where different halophytic
the ability of rhizobacteria to facilitate plant adaptation and species can be found. Halophytes have been proposed as
promote growth and productivity has been reported [1–6]. key players for saline soils reclamation [8], phytoremediation
Root-associated bacteria can promote plant growth by direct of hydrocarbon and heavy metals polluted saline soils [9,
and indirect mechanisms, the former including nutrient 10], and forage and oil seed production [11, 12]. Salicornia
fixation and solubilisation and phytohormones synthesis. (Chenopodiaceae) is a subcosmopolitan plant genus com-
Indirect activities include biocontrol, the ability to reduce or prising annual species strictly occurring in salty environ-
avoid the harmful effects of phytopathogens. Both the host ments and widespread in several countries, including those
plant and its associated microbiome gain an evolutionary of the Mediterranean basin. Salicornia densely colonises
advantage to survive under harsh conditions by establishing different areas of southern Tunisia, including Sebkhet and
tight interplays. Chott ecosystems, dominated by extreme values of aridity
and soil salinity. Intense evaporation rates render Sebkhet and plot where salt crusts were absent (BDV4-S4, BDV4-S5, and
Chott as dry salt lakes which are inhospitable for most of the BDV4-S6). Replicates of bulk soils were also sampled from
organisms. the site BDV4 (presence of salt crust: BDV4-B1; absence
The manipulation of natural resources to increase plant of salt crust: BDV4-B4). Rhizospheric soil was defined as
productivity in lands traditionally considered unsuitable for soil particles tightly adhering to roots (1–3 mm) after gently
agriculture is a challenging but necessary task, in the light shaking. Bulk soil was collected as control about 2 m far
of the increasing world population and the need for food from any vegetation. Rhizosphere and bulk soils will be,
production [13]. The efforts which aimed to the production respectively, indicated in the text with the codes S and B.
of salt-resistant crops include conventional breeding, marker- Soil samples were collected using sterile spoons and stored
assisted selection, and the creation of transgenic plants and in sterile bags at −20∘ C for molecular analyses and at 4∘ C for
are nowadays focusing also on the halophyte potential to microbiological isolation. Soil salinity was measured with a
guarantee a suitable food production in a salinized planet hand refractometer (Atago, Tokyo, Japan) after the extraction
[14]. Different works in the last years highlighted the impor- of pore water from approximately 2 g of soil.
tance of plant growth promoting bacteria in facilitating salt
tolerance in plants devoted to food production [3, 7, 15,
16], and few reports emphasized the role of PGP bacteria 2.2. Metagenome Extraction and 16S rRNA Amplification.
associated with Salicornia spp. [17–22]. The investigation of DNA was extracted from 0.5 g of soil using the protocol
the rhizobacterial community associated to plants naturally established by Schbereiter-Gurtner et al. [25].
adapted to cope with extreme saline conditions might lead to DNA was quantified using NanoDrop 1000 spectropho-
several knowledge outputs: (i) the understanding of the plant- tometer (Thermo Scientific, Waltham, MA, USA).
microbe interaction under saline conditions, (ii) definition Bacterial 16S rRNA gene fragments (∼550 bp) were PCR
of the mechanisms underlying plant growth with promotion amplified using primers 907R (3󸀠 -CCGTCAATTCCTTTG-
under the salinity stress, and (iii) identification of bacterial AGTTT-5󸀠 ) and GC-357F (3󸀠 -CCTACGGGAGGCAGCAG-
strains to design biological fertilizers exploitable for agricul- 5󸀠 with a 5󸀠 -end GC-clamp) targeting a portion of the 16S
ture in arid and saline lands. To achieve the best results in rRNA gene that include the hypervariable V3 regions [26].
terms of plant growth promotion under salinity and drought PCR reactions were performed in a 50 𝜇L final volume
stress it is essential to focus on the fraction of the culturable containing 1X buffer, 2.5 mM MgCl2 , 5% of DMSO, 0.12 mM
bacteria that is able to thrive under these specific conditions. of dNTPs mixture, 0.3 𝜇M of each primer, 1.5 U Taq poly-
Therefore, the aims of this work were (i) the isolation of merase, and 10 ng of template, applying the following thermic
halophilic/halotolerant bacteria from Salicornia rhizosphere protocol: 94∘ C for 4 min, followed by 10 cycles of 94∘ C for
and bulk soils collected in hypersaline ecosystems in southern 0.5 min, 61∘ C for 1 min, and 72∘ C for 1 min; followed by
Tunisia, (ii) the characterization of their resistance to abiotic further 20 cycles of 94∘ C for 0.5 min, 56∘ C for 1 min, and 72∘ C
stresses and their plant growth promoting (PGP) potential, for 1 min; and a final extension at 72∘ C for 7 min. Presence
and (iii) the description of taxonomic diversity of both the and length of PCR products were verified by electrophoresis
halophilic/halotolerant culturable fraction and the whole in 1% w/v agarose gel prior to Denaturing Gradient Gel
bacterial microbiome inhabiting Salicornia rhizosphere and Electrophoresis (DGGE) analysis.
bulk soils.
2. Materials and Methods 2.3. Denaturing Gradient Gel Electrophoresis. PCR products
(∼150 ng) were loaded in a 0.5 mm polyacrylamide gel
2.1. Site Description, Soil Sampling, and Soil Characterization. (7% (w/v) acrylamide-bisacrylamide, 37.5 : 1) containing 40
The studied sites, named BDV4 (N 34∘ 26󸀠 951󸀠󸀠 ; E 09∘ 54󸀠 102󸀠󸀠 ), to 60% urea-formamide denaturing gradient (100% corre-
BDV11 (N 34∘ 08󸀠 735󸀠󸀠 ; E 08∘ 04󸀠 417󸀠󸀠 ), and BDV20 (N sponds to 7 M urea and 40% (vol/vol) formamide) according
33∘ 57󸀠 252󸀠󸀠 ; E 08∘ 24󸀠 508󸀠󸀠 ) corresponded, respectively, to to the method described by Muyzer et al. [26]. The gels were
Sebkhet El Naouel, Chott El Gharsa, and Chott El Jerid and run for 15 h at 60∘ C by applying a constant voltage of 90 V
were located in southern Tunisia. in 1X Tris-acetate-EDTA (TAE) buffer. After electrophoresis,
Visual inspection of the sites identified Salicornia as the the gels were stained for 30 min in 1X TAE buffer containing
only present plant. The plants were identified according to the 1X SYBR Green (Molecular Probes, Leiden, the Netherlands)
plant morphology as S. strobilacea [23, 24], a widespread plant according to manufacturer’s instructions and rinsed twice for
in southern Tunisia. 10 min with distilled water. Gels images were captured using
Between the sites different conditions in respect of super- a Gel Doc 2000 apparatus (Bio-Rad, Milan, Italy). The band
ficial salt crust presence (BDV11) or absence (BDV20) were patterns of DGGE gels were analysed using Image J software
observed (Table 1). Rhizospheric and bulk soils were sampled (available for free download at http://rsb.info.nih.gov/ij/) and
from triplicate specimens of Salicornia from sites BDV11 Microsoft Excel XLSTAT software (Addinsoft Inc., New York,
and BDV20. In the site BDV4 different microenvironments NY, USA) as previously described [5]. DGGE bands were
were identified, and a total of six Salicornia specimens were excided from the gels with a sterile scalpel and eluted in 50 𝜇L
collected. In this site three sampled specimens were growing of sterile Milli-Q water at 37∘ C for 4 hours. Subsequently, 8 𝜇L
on salt crust covered soil (BDV4-S1, BDV4-S2, and BDV4- of eluted DNA was reamplified by PCR using primers 357F
S3), and three sampled specimens were growing on a soil and 907R as described in the previous paragraph. Positive
Table 1: Sample code, location and characteristics of the Salicornia rhizospheres and bulk soils collected in Tunisia and analysed in the present
study.
Sample code Soil fraction Site Coordinates Feature

BDV4-S1,2,3 Rhizosphere Sebkhet en Naouel N 34∘ 26󸀠 951󸀠󸀠 , E 09∘ 54󸀠 102󸀠󸀠 soil covered by salt crust
BDV4-S4,5,6 Rhizosphere Sebkhet en Naouel N 34∘ 26󸀠 951󸀠󸀠 , E 09∘ 54󸀠 102󸀠󸀠 soil
BDV11-S1,2,3 Rhizosphere Chott El Gharsa N 34∘ 08󸀠 735󸀠󸀠 , E 08∘ 04󸀠 417󸀠󸀠 soil covered by salt crust
(17.3 ± 1.3% of salinity)
BDV20-S1,2,3 Rhizosphere Chott El Jerid N 33∘ 57󸀠 252󸀠󸀠 , E 08∘ 24󸀠 508󸀠󸀠 soil
BDV4-B1-1,2,3 Bulk soil Sebkhet en Naouel N 34∘ 26󸀠 951󸀠󸀠 , E 09∘ 54󸀠 102󸀠󸀠 soil covered by salt crust
BDV4-B4-1,2,3 Bulk soil Sebkhet en Naouel N 34∘ 26󸀠 951󸀠󸀠 , E 09∘ 54󸀠 102󸀠󸀠 soil
BDV11-B1,2,3 Bulk soil Chott El Gharsa N 34∘ 08󸀠 735󸀠󸀠 , E 08∘ 04󸀠 417󸀠󸀠 soil covered by salt crust
(19.1 ± 0.4% of salinity)
BDV20-B1,2,3 Bulk soil Chott El Jerid N 33∘ 57󸀠 252󸀠󸀠 , E 08∘ 24󸀠 508󸀠󸀠 soil
amplifications were partially sequenced by Macrogen Inc., genotypic identification through 16S rRNA gene sequenc-
Korea (http://www.macrogen.com/) using the primer 357F. ing. 16S rRNA amplification was performed by using the
universal primers 27F (3󸀠 -AGAGTTTGATCMTGGCTCAG-
2.4. Bacteria Isolation. Rhizospheric and bulk soil triplicates 5󸀠 ) and 1492R (3󸀠 -CTACGGCTACCTTGTTACGA-5󸀠 ) [28]
were pooled prior to bacteria isolation, except in the case of and applying the same protocol of ITS-PCR. Partial 16S
the specimens BDV4-S4, BDV4-S5, and BDV4-S6, collected rRNA sequences were obtained from Macrogen Inc., Korea
at site BDV4 in absence of surface salt crust, and processed (http://www.macrogen.com/).
separately. Thus, we compared the structure of culturable
halotolerant/halophilic bacteria among the different study 2.6. Nucleotide Sequence Analyses and Accession Numbers.
sites, and additionally the taxonomic diversity within the Nucleotide sequences were edited in Chromas Lite 2.01
same site was evaluated by comparing the replicates. To (http://www.technelysium.com.au) and subjected to BLAST
determine the bacterial cell number, 1 g of rhizospheric and search (http://blast.ncbi.nlm.nih.gov/Blast.cgi). The partial
bulk soils collected at the different sites was shaken with 16S rRNA gene sequences obtained from the bacterial isolates
9 mL of sterile saline solution (0.9% NaCl). The suspensions and the excised DGGE bands have been deposited in the
were serially diluted and plated in triplicate on solidified EMBL, respectively, under accession numbers HF678717–
R2A medium (Oxoid) enriched with 10 and 15% NaCl. After HF678862 and HF678127–HF678194.
1-week incubation at 30∘ C, the colony-forming unit (cfu)
per gram was determined, and, for each sample, a number
of colonies comprised between 8 and 42 per medium were 2.7. Resistance to Abiotic Stresses. Resistance to salt stress was
randomly selected. The bacterial isolates were stored as 25% assessed by growing the isolates at 30∘ C in R2A supplemented
glycerol stocks at −80∘ C. by different sodium chloride concentrations, ranging from
0 to 20% w/v. The ability to grow under osmotic stress was
tested at 30∘ C by adding 5–20% of polyethylene glycol (PEG)
2.5. Genotypic Characterization and Identification. DNA
to R2A broth medium. Finally, the capability to growth in a
extraction was performed on each isolated strain. The bacte-
wide range of temperatures was verified by incubating the
ria collection has been dereplicated through the application
R2A plates at 4∘ , 42∘ , and 50∘ C. A control, consisting in a
of 16S-23S rRNA Intergenic Transcribed Spacer-PCR (ITS-
sterile plate or tube, was also run parallel to each experiment.
PCR) using ITS-F (3󸀠 -GTCGTAACAAGGTAGCCGTA-5󸀠 )
and ITS-R (3󸀠 -CTACGGCTACCTTGTTACGA-5󸀠 ) primers
as previously described [5, 27]. PCR amplification was 2.8. In Vitro Screening of Plant Growth Promoting Activities.
performed in 25 𝜇L reaction containing 1X buffer, 1.5 mM Each isolate was grown as pure culture to evaluate its PGP
MgCl2 , 0.12 mM of dNTPs mixture, 0.3 𝜇M of each primer, features in suitable media enriched with 5% w/v NaCl. Only
1 U Taq polymerase, and 10 ng of template, applying the two isolates were not able to grow at 5% NaCl, and their
following thermic protocol: 94∘ C for 4 min, followed by 35 PGP activities were tested at 10% NaCl. The production of
cycles of 94∘ C for 0.5 min, 55∘ C for 1 min, and 72∘ C for indole-3-acetic acid was detected by the method described
2 min and a final extension at 72∘ C for 10 min. The ITS- by Bric et al. [29]. The ability to solubilise insoluble phos-
PCR products were run on 2% agarose gels and stained phate compounds was estimated according to Ahmad et
with ethidium bromide. Gel images were captured using al. [30]. The ammonia synthesis assay was performed as
a Gel Doc 2000 apparatus (Bio-Rad, Milan, Italy) and recommended by Cappuccino and Sherman [31]. Protease
ITS-fingerprinting profiles were visually analysed to clus- activity was determined from clearing zones in skimmed milk
ter together the bacterial isolates showing the same band agar according to Nielsen and Sørensen [32]. Atmospheric
pattern. From each cluster, at least a representative strain nitrogen fixation ability and ACC-deaminase activity were
has been selected for subsequent PGP characterization and determined by the method of Penrose and Glick [33]. nifH
gene detection has been performed by PCR test using failing of PCR-DGGE amplification. Although the bulk soils
the primer sets PolF (3󸀠 -TGCGAYCCSAARGCBGACTC-5󸀠 ) collected from Tunisian Sebkhet and Chott were characterized
and PolR (3󸀠 -ATSGCCATCATYTCRCCGGA-5󸀠 ) [34]. PCR by extreme dryness and salinity values, DGGE band profiles
amplification was performed in 25 𝜇L reaction containing 1X highlighted that a rich and diverse bacterial microbiome was
buffer, 1.5 mM MgCl2 , 0.12 mM of dNTPs mixture, 0.3 𝜇M present in all the samples (Figure 1(a), left panel). Principal
of each primer, 1 U Taq polymerase, and 10 ng of template, Component Analysis (PCA) performed on the line plots
applying the following thermic protocol: 94∘ C for 4 min, derived from DGGE band profiles (Figure 1(b), left panel)
followed by 35 cycles of 94∘ C for 0.5 min, 55∘ C for 1 min, and indicated that bulk soils clustered according to the site of
72∘ C for 2 min and a final extension at 72∘ C for 10 min. provenience. On axis 1, describing the 44% of the samples
similarity, bulk soil samples were distributed according to
2.9. Chromosomal gfp-Tagging of Halotolerant Halomonas the presence (stations BDV4-B1 and BDV11-B) or absence
Strains by Conjugation Procedure. To stably transform strains (stations BDV20-B and BDV4-B4) of salt crusts covering
affiliated to the genus Halomonas, we adopted the method the soil surface. Salinity is known as one of the strongest
based on mini-Tn7 transposon system [35]. Briefly, the abiotic factors influencing the assemblages of a huge variety
mobilisation of the gfp-harbouring fragment was achieved by of bacterial populations sheltered by the soil [38]. The factors
a four-parental conjugation formed by a cellular suspension shaping the composition of the bacterial community include
of 1010 cells of the strain to be transformed and 109 cells also biotic interactions, and the role of root exudates in
for E. coli strains carrying helper, delivery, and mobilization the selection of a peculiar microbiome is well known [39].
plasmids [35]. To select for gfp-transformed cells, after the The DGGE pattern obtained from Salicornia rhizosphere
mating, the cellular suspension was plated in R2A medium samples was different from those observed in bulk soils
supplemented with 10% NaCl and the required antibiotics. (Figure 1(a)). According to DGGE fingerprints the bacterial
The gfp-labelling procedure was successful for a strain of H. communities of the rhizospheric soil triplicates collected at
elongata, as visualized by fluorescence microscopy. both sites BDV20 and BDV11 clustered together (Figure 1(a),
right panel). Similarly, the rhizospheres collected from the
salt crust covered soil at site BDV4 (BDV4-S1, 2, and 3)
2.10. In Vitro Bacterial Rhizocompetence Test. To evaluate showed a high level of homogeneity. A high number of
gfp-labelled strain ability to adhere and potentially colonise DGGE bands were observed in all the rhizospheres, with the
plant root system, an in vitro assay was performed on two exclusion of BDV4-S6 sample, probably affected by biases
model plants: Arabidopsis thaliana and Salicornia plantlets in PCR amplification. Overall, the rhizosphere of Salicornia
collected in marine dune ecosystems in south Italy. After was proved to be a habitat characterized by a highly rich
an overnight growth in liquid selective medium, bacterial bacterial community. Principal Component Analysis of the
cell concentration was microscopically evaluated, and a 108 DGGE patterns (Figure 1(b), right panel) indicated, except
cell/mL suspension was prepared. Salicornia plant roots were for rhizosphere samples BDV4-S5 and BDV4-S1, a higher
dipped in MS salt half strength medium (SIGMA, Italy) similarity among the rhizospheres collected from different
supplemented with 2% NaCl and the prepared bacterial stations than among the bulk soils suggesting that the
suspension. For Arabidopsis rhizocompetence test, NaCl rhizosphere acts as a selection factor that tend to uniform
addition was avoided since this plant is extremely salt stress- bacterial diversity independent from the soil type. Principal
sensitive. After an overnight incubation (∼16 h), plant roots Component Analysis (Figure 1(b), right panel) pointed out
were gently washed to remove no- or weakly-bound bacterial the overall similarity of the bacterial communities hosted
cells and observed under a confocal laser scanning micro- by the Salicornia rhizospheres collected in different sites.
scope (Leica TCSNT). Images were acquired using Leica The even structure of the rhizosphere bacterial community
Confocal Software and analysed by using the MBF ImageJ confirmed the importance of plant inputs in the selection
software. of specific bacterial taxa associated with the roots, a well-
known phenomenon generally reported as “rhizosphere
3. Results and Discussion effect.” Despite that, the occurrence of a degree of variability
within the bacterial microbiome associated with Salicornia
3.1. DGGE Analysis of the Bacterial Microbiome Inhabiting specimens collected in different microenvironments within
Salicornia Rhizosphere and Surrounding Bulk Soil. The intro- site BDV4 (related to presence and absence of salt crusts on
duction of fingerprint-based analyses [26, 36, 37] turned soil surface) was perceived by DGGE analysis. This result
into the application of cultivation-independent techniques as confirmed that, besides the selection driven by the plant,
routine tools to depict the overall microbiome composition in also the environmental parameters play a role in shaping the
environmental samples and to infer which factors influence rhizosphere bacterial microbiome.
the abundance and distribution of specific microbial taxa. According to DGGE band sequencing analyses, the
Here, DGGE analysis of 16S rRNA gene was applied to pro- prevalent taxonomic groups associated with Salicornia roots
vide a snapshot of both culturable and unculturable bacterial and bulk soils were Alpha-, Beta- and Gammaproteobac-
assemblages in Salicornia rhizosphere and the surrounding teria, Bacilli, and Actinobacteria (Figure 1(c)), as previously
bulk soil not affected by the plant. Bulk soil samples from observed in the bacterial communities associated to dif-
all sites were analysed in triplicate, except in BDV4-B4 ferent plant species (Capsicum annuum) growing in desert
sites where only duplicate samples were analysed due to the areas under water stress condition [5]. In addition, two
BDV4-B1-2
BDV4-B4-2
BDV4-B4-1
BDV4-B1-3
BDV4-B1-1
BDV20-B2
BDV11-B3
BDV11-B2
BDV11-B1
BDV20-B3
BDV20-B1
BDV20-S3
BDV20-S2
BDV20-S1
BDV11-S3
BDV11-S2
BDV11-S1
BDV4-S6
BDV4-S5
BDV4-S4
BDV4-S3
BDV4-S2
BDV4-S1
56
27
169
54
70
60
107
174
71
115 66
154 83
88 84
85 34
35
163 89 33
78 36 22
130 110 37 3
162
44 13
113 99 79 49 45 25 9
68 50 40
31
142 51
105 63
98 106 75 58 16
95 67 42 62 15
97
138 66
96 74
8
5
178
179 171
177
(a)
20 20
BDV4-B4-2
15
Number of bands
BDV4-B1-2
10 BDV4-B4-1
BDV4-B1-3
F2 (22.76%)
10
BDV4-B1-1
0 BDV20-B3
BDV20-B2 5
BDV20-B1
−10 BDV11-B3
BDV11-B1
0
BDV11-B2
BDV4-B1 BDV4-B4 BDV11-B BDV20-B
−20 12
−30 −20 −10 0 10 20
F1 (43.61%) 10
Number of bands
1000 8
BDV4-S5
6
500
4
BDV4-S4
F2 (13.4%)
BDV20-S1 2
0 BDV4-S5
BDV20-S2 BDV20-S3
BDV11-S3 BDV4-S3 0
BDV11-S2 BDV11-S1 BDV4-S4-5-6 BDV4-S1-2-3 BDV11-S1-2-3 BDV20-S1-2-3
BDV4-S2
BDV4-S1
−500 Rhizobiales Burkholderiales

Flavobacteriales Unc. Actinobacteria
Rhodobacterales Legionellales
−1000 Sphingobacteriales Unc. Bacteria
−1000 −500 0 500 1000 1500 Sphingomonadales Bacillales
F1 (52.85%) Actinomycetales
(b) (c)
Figure 1: DGGE analysis performed on the bulk (left panel) and rhizospheric (right panel) soil bacterial community. (a) In the left panel,
DGGE patterns of the bulk soils collected at sites BDV20, BDV11 and BDV4. In the right panel, DGGE patterns of the rhizospheric soils
collected at sites BDV20, BDV11, and BDV4. The numbers represented the three analysed replicates. (b) Principal Component Analysis based
on the DGGE profiles of the bacterial community inhabiting bulk (left) and rhizospheric (right) soil associated with Salicornia specimens.
(c) Taxonomic identification of bacterial 16S rRNA sequences excised from DGGE bands cut from rhizospheric and bulk soil profiles.
1E+11 work is nevertheless the first report about the quantifica-

1E+10 tion of halotolerant microbes in root-associated extremely
1E+09
1E+08 saline soils, enlarging the concept of the rhizosphere effect
1E+07
Soil (cfu/g)
to specific bacterial groups highly adapted to the hostile

1E+06
1E+05 environmental conditions.
1E+04 A large collection of 475 isolates, representing the halotol-
1E+03 erant culturable fraction of the bacterial diversity associated
1E+02
1E+01 with Salicornia specimens and bulk soil of Sebkhet and
1E+00 Chott ecosystems, was established. In the case of BDV4-S6
BDV4-S1
BDV4-S4
BDV4-S5
BDV4-S6
BDV11-B
BDV11-S1
BDV20-S1
only eight colonies growing in a medium containing 15%
NaCl were isolated, whilst for the rest of the samples a
number of colonies comprising between 22 and 42 were ran-
R2A 10% domly picked for both the salt enrichment conditions. ITS-
R2A 15% PCR fingerprinting was applied to dereplicate the bacteria
collection allowing the identification of 136 clusters corre-
Figure 2: Evaluation of the halophilic/halotolerant culturable bac- sponding to different ITS profiles and representing different
teria number of Salicornia rhizosphere and bulk soils. Microbial cell species/subspecies. From each haplotype at least one strain
number is reported as colony-forming unit (cfu) per gram of fresh was arbitrarily selected for 16S rRNA partial gene sequencing
soil. Dark grey bars represent the cfu per gram of fresh soil detected
and for the phenotypic tests. The taxonomic identification of
on R2A medium enriched with 10% NaCl. Light grey bars represent
the cfu per gram of fresh soil detected on R2A medium enriched
the bacteria showed the prevalence of the Halomonas genus
with 15% NaCl. among the strains isolated at 15% NaCl (Figure 3). Excluding
samples BDV4-S5 and BDV4-S6, where bacteria belonging to
the genus Nesterenkonia were also retrieved, the strains iso-
lated from both bulk and rhizosphere soils on medium con-
taining 15% NaCl were represented exclusively by Halomonas.
different orders belonging to the Bacteroidetes phylum Hence, the subcollection obtained on 15% sodium chloride
were retrieved, namely, Sphingobacteriales and Flavobac- supplemented medium was characterized by high dominance
teriales, the latter exclusively present in the bulk soil and low values of the Shannon diversity index (Table 2).
(Figure 1(c)). Both in rhizospheric and bulk soils Beta- and Similarly, bacteria isolated on 10% NaCl containing medium
Gammaproteobacteria and Bacilli were represented by only from BDV11-B and BDV11-S1 belonged exclusively to the
one taxonomic order, while Alphaproteobacteria composi- Halomonas genus (Figure 3). This genus, together with the
tion differed in the two soil fractions. Alphaproteobacteria genus Chromohalobacter, represented the totality of BDV20-
in rhizospheric soils were represented exclusively by Rhi- S1 halophilic culturable community (Figure 3). The preva-
zobiales spp. while bulk soils were colonised by a more lence of the Halomonas genus was demonstrated within
diverse bacterial community including, in addition to Rhi- the halotolerant root-associated bacteria collection, where
zobiales, the orders Sphingomonadales and Rhodobacterales. Halomonas elongata, along with the species H. eurihalina, H.
sinaiensis, H. halmophila, H. ilicicola, H. indalina, H. vari-
abilis, H. xinjiangensis, and H. taeheungii, where retrieved.
3.2. Bacteria Isolation and Identification. Viable halotolerant The isolation of Halomonas sp. from the roots of Salicornia
bacteria were cultured on oligotrophic medium from all the brachiata was recently reported [19]. Besides harbouring
collected rhizospheres and from the bulk soil sampled at higher salt tolerant bacterial counts, Salicornia rhizosphere
BDV11 site. Culturable bacteria abundance (Figure 2) was was richer than bulk soils also in terms of biodiversity,
considerably variable between the different sites, and the since all the bacteria isolated from BDV11-B belonged to
number of colony-forming units (cfu) ranged between 9 × the specie Halomonas elongata. Overall, the strains isolated
104 and 1.6 × 1010 per gram of fresh soil. Similar counts were at 10% NaCl from rhizosphere soil of site BDV4 displayed
observed within the sites on the same medium supplemented a higher biodiversity at the genus level (Figure 3) that is
with 10% and 15% of sodium chloride. This result indicates reflected by low dominance values and high values of the
that most of the halotolerant culturable bacteria are able Shannon diversity index (Table 2). BDV4 rhizospheric soils,
to cope with the higher value of salt content, close to the characterized by lower abundance of culturable halotoler-
salinity measured in the soil pore water of BDV11 station. ant/halophilic bacteria compared to those collected at sites
Halotolerant bacteria abundance in the BDV11-B bulk soil BDV11 and BDV20 (Figure 2), hosted a more diverse halo-
was in accordance with values previously reported in similar tolerant bacterial community comprising different genera
environments [40]. The interaction with the plant and the that were previously reported in other hypersaline environ-
presence of root exudates could be responsible for the higher ments [41]. Strains belonging to the Chromohalobacter genus
abundance of halotolerant/halophilic bacteria detected in were isolated at 10% NaCl from BDV4-S6 and BDV20-S1.
the rhizosphere (Figure 2), as compared to the bulk soils in This genus, along with Halomonas, represents an important
BDV11 site. These results are in agreement with the “rhizo- member of the family Halomonadaceae, a taxonomic group
sphere effect” described by several authors in conventional within the Gammaproteobacteria typical of hypersaline envi-
and extreme environments [5, 39]. To our knowledge, this ronments, whose taxonomy is still under revision [42]. A
Table 2: Diversity indices of halotolerant/halophilic bacteria collection. The calculation based on the genera distribution in the different analysed rhizospheric and root-free soils. The
percentage indicates in brackets refers to the NaCl concentration used in the isolation medium.
BDV4-S1 BDV4-S1 BDV4-S4 BDV4-S4 BDV4-S5 BDV4-S5 BDV4-S6 BDV4-S6 BDV20-S1 BDV20-S1 BDV11-B BDV11-B BDV11-S1 BDV11-S1
(10%) (15%) (10%) (15%) (10%) (15%) (10%) (15%) (10%) (15%) (10%) (15%) (10%) (15%)
Genera 4 1 4 1 4 2 5 2 2 1 1 1 1 1
Individuals 28 38 32 32 24 40 22 8 42 42 42 42 42 42
Dominance 0.4668 1 0.3203 1 0.5382 0.8613 0.3471 0.7813 0.8673 1 1 1 1 1
Shannon 0.9887 0 1.245 0 0.8824 0.2664 1.254 0.3768 0.2573 0 0 0 0 0
Evenness 0.6719 1 0.8682 1 0.6042 0.6526 0.7006 0.7288 0.6467 1 1 1 1 1
7
45 100
40 90
80
Number of isolates
Isolates (%)
35 70
30 60
50
25 40
20 30
15 20
10
10 0
5 0% 5% 10% 15% 20% 4∘ C 42∘ C 50∘ C 5% 10% 20%
0 NaCl NaCl NaCl NaCl NaCl PEG PEG PEG
BDV4-S1-10
BDV4-S4-10
BDV4-S5-10
BDV4-S6-10
BDV20-S1-10
BDV11-B-10
BDV11-S1-10
BDV4-S1-15
BDV4-S4-15
BDV4-S5-15
BDV4-S6-15
BDV20-S1-15
BDV11-B-15
BDV11-S1-15
(a)
100
90
80
Isolates (%)
70
60
Chromohalobacter Halomonas 50
Marinococcus Oceanobacillus 40
30
Halobacillus Kushneria 20
Nesterenkonia Virgibacillus 10
0
IAA P-sol N-fix NH3 Prot. ACC
Figure 3: Taxonomic composition of the halophilic/halotolerant
fraction of culturable bacteria associated with Salicornia rhizosphere (b)
and bulk soils shown as genera distribution. The numbers 10 and 15
in the sample name indicate the percentage of NaCl supplemented Figure 4: Spread of abiotic resistance and plant growth promoting
to the medium during isolation procedures. traits among the halophilic/halotolerant bacteria isolated from
Salicornia rhizosphere and bulk soils. (a) Abiotic stresses resistance.
PEG: polyethylene glycol. (b) Plant growth promotion features. IAA:
third genus, Kushneria, of the family Halomonadaceae was indole-3-acetic acid production; P-sol: phosphate solubilization; N-
isolated from the BDV4-S4 rhizosphere. The additional root- fix: putative nitrogen fixation ability; NH3 : ammonia production;
associated halotolerant bacteria belonged to the classes Bacilli Prot.: Protease activity; ACC: 1-aminocyclopropane-1-carboxylate
(Halobacillus trueperi, Marinococcus halophilus, Oceanobacil- deaminase.
lus picturae and Virgibacillus olivae) and Actinobacteria, the
latter being represented exclusively by the species Nesterenko- the 136 ITS groups identified by collection dereplication.
nia halobia. Both Oceanobacillus picturae, and Nesterenkonia In particular the bacteria collection was screened for the
halobia were previously isolated from a different saline capability (i) to grow at extreme temperature values, (ii)
ecosystem, namely, mangrove sediments [43, 44], where O. to thrive in presence of different salt concentrations and
picturea was described for the first time as a phosphate- (iii) in conditions of low water availability. The strains able
solubilising bacterium able to promote mangrove seedling. to grow at 42∘ C represented 93% of the bacteria collection
Taxonomic analyses on the bacteria isolated from the (Figure 4(a)) whereas only 13% of the isolates survived at
rhizosphere of plants growing in different niches of the higher temperature (50∘ C). The ability to flourish at low
same site (BDV4-S4, BDV4-S5, and BDV4-S6) permitted temperature (4∘ C) was observed for 71% of the total isolates,
to assess the occurrence of intrasite environmental selective and twenty strains were able to grow in a large temperature
forces shaping the composition of the culturable halophilic interval, between 4 and 50∘ C. In arid and saline soils the
community, as it was already shown for the total bacterial vegetation is generally sparse, a factor that contributes to the
community by DGGE-fingerprinting results. Even though strong temperature fluctuations affecting the soil. Hence, the
the enrichment on 15% NaCl resulted in an even taxonomic ability of the plant associated microbes to grow in a large
distribution of the isolates, a certain degree of variability temperature range and to survive at temperature fluctuations
could be observed in the bacteria isolated at 10% NaCl is useful to efficiently colonise barren and extreme desert
(Figure 3), suggesting that microvariations within the site habitats.
may influence the prevalence of different bacterial popu- All the isolates were isolated in the presence of 10 and 15%
lations in the culturable halotolerant fraction. This result of sodium chloride in the growing medium. The majority of
strengthens the diversity pattern described by DGGE on the the isolates, corresponding, respectively, to 99% and 92% of
total bacterial microbiome inhabiting the Salicornia rhizo- the bacteria collection, grew at lower (5%) or higher (20%)
spheric soils collected at site BDV4 (Figure 1, right panel), salinity (Figure 4(a)). A significant fraction of the collection
indicating a partial overcoming of environmental factors on (74%) was constituted by halophiles, unable to grow in the
the rhizosphere effect imposed by the host plant. absence of NaCl in the medium (Figure 4(a)).
A high percentage (90%) of the isolated bacteria was
3.3. Resistance to Abiotic Stresses. Aiming to identify the most able to grow in presence of 5% polyethylene Glycol (PEG)
suitable rhizobacteria to design a biofertilizer for sustaining (Figure 4(a)), a molecule which induces a decrease of the
plant growth in saline and arid soils, the ability of the isolated water potential when added to the cultivation media [45].
bacteria to cope with different abiotic stresses typical of A decline in the number of isolates that positively grew at
arid lands was tested on 164 bacterial strains belonging to increasing PEG concentration was observed; nevertheless the
percentage of bacteria able to grow at 10 and 20% of PEG were and Halomonas xinjiangensis, displayed ACC-deaminase
noteworthy and corresponded, respectively, to 87 and 81% of activity in presence of 5% NaCl (Figure 4(b)). ACC-
the bacteria collection. deaminase activity in the genus Halomonas was recently
Tolerance to abiotic stresses was widespread within the reported in the ambit of the investigation of PGP features
bacteria collection, and represented a common trait even of halophilic bacteria isolated from halophytes, including
in phylogenetic unrelated strains, as expected since many the species Salicornia brachiata [22, 55]. Nonetheless, the
of the retrieved species were previously isolated from saline low percentage (2%) of ACC-deaminase activity among
and hypersaline habitats as in the case of Virgibacillus spp. the collection established in this work is in agreement with
[46, 47], Halomonas sinaiensis [48], and Kushneria and previous studies reporting the detection of ACC-deaminase
Halomonas spp. [49, 50]. activity only for a minor fraction of bacteria isolated from
In vitro tests showed that twenty Halomonas strains the rhizosphere of wheat growing in salinized soil [6, 16].
were particularly resistant to extreme values of different Direct mechanisms of plant growth promotion include
abiotic factors. These isolates, belonging to the species H. those metabolisms that, by supplying nutrients to the plant,
elongata and H. sinaiensis, were able to actively grow (i) on enhance its fitness. The established halotolerant/halophile
R2A medium containing a percentage of sodium chloride bacteria collection was analysed for the capability to solu-
comprised between 5 and 20%, (ii) in the presence of 5, 10, bilise phosphate, fix nitrogen, and produce ammonia. The
and 20% of PEG in the medium, and (iii) when incubated in phosphate solubilisation activity was present in 65% of the
a wide range of temperature (from 4 to 50∘ C). whole collection (Figure 4(b)), including all the genera except
The resistance of the isolates to the extreme physical- for Kushneria. The potential activity of nitrogen fixation has
chemical parameters of Tunisian Sebkhet and Chott ecosys- been phenotypically tested by the strain capability to grow
tems is a prerequisite to select efficient PGP bacteria able to in nitrogen-free medium and confirmed by molecular inves-
sustain plant growth since the effectiveness of a microbial tigation by PCR amplification of the nifH gene, codifying
consortium strictly depends on its competitive root coloni- for a subunit of the nitrogenase enzyme. Six percent of the
sation [51, 52], a reason that explains why the use of PGP analysed bacterial strains were positive to both the tests
bacteria isolated from different soil and climate conditions showing the putative ability to fix nitrogen (Figure 4(b)).
can be a largely unsuccessfully strategy in arid and saline Putative nitrogen fixation activity was detected in only a
lands [53, 54]. minor fraction of the ITS clusters, belonging to the species
Halomonas elongata, H. eurihalina, H. indalina, Kushneria
3.4. Plant Growth Promotion Test. The PGP activities of marisflavi, and Chromohalobacter canadensis. Ammonia pro-
164 isolates, belonging to the 136 ITS-PCR clusters and duction was also a common PGP trait shown by 93% of
representing the whole taxonomic diversity of the established the isolates (Figure 4(b)). All the bacteria genera present in
bacteria collection, were tested in vitro by using specific strain collection were positive to the ammonia production
media supplemented by 5% sodium chloride. assay, thus potentially contributing to plant nitrogen nutri-
One of the strategies adopted by PGP bacteria to induce tion. The inability to produce ammonia did not show any
plant growth is the influence on the plant hormonal balance. species-related pattern. The widespread ability to increase the
96% of the isolates showed the ability to produce indole- concentration of bioavailable nutrients in the isolate collec-
3-acetic acid (IAA) (Figure 4(b)), one of the main plant tion from Salicornia rhizosphere suggested the contribution
hormones of the auxin family. This trait was shared by all of these halotolerant and halophilic bacteria to the plant
the genera retrieved from the analysed rhizospheric and bulk nutrient balance. These direct PGP features were generally
soils while it was not detected in the isolated Chromohalobac- simultaneously present in the same strain, possibly acting
ter marismortui and C. salexigens strains. Few strains unable in a synergic manner to directly promote plant growth, as
to produce IAA belong to the species Oceanobacillus picturae previously reported [30].
and Halomonas halophila, characterized by an uneven dis- Besides direct PGP activity, several representatives of all
tribution of this PGP feature within their ITS clusters. The the taxonomic classes retrieved in our collection (11%) also
capability to modulate the plant stress level by providing displayed in vitro protease activity, a result that indicated
indole-3-acetic acid (IAA), a molecule involved in lateral their possible role as biocontrol agents. The bacterial iso-
roots development, was previously reported for halotolerant lates displaying protease activity comprised bacterial strains
bacteria isolated from coastal soils [55], halophyte roots in of the genera Chromohalobacter, Halomonas, Kushneria,
Argentina [56], and rhizosphere of C. annum growing in Marinococcus, Nesterenkonia, and Virgibacillus.
desert areas [5]. In addition, the recent study by Tiwari et al. The results about the investigation of the PGP traits
[16] demonstrated that inoculation of wheat with Halomonas occurrence among the bacteria collection established from
sp., the most abundant genus in our strains collection, the Salicornia rhizospheric and bulk soils are in general
resulted in higher content of IAA in the rhizosphere of the agreement with the observations reported by other studies
treated plants than control experiment. realized on halophyte [22, 55, 56] and crop plant growing
Rhizobacteria can also positively influence the health under saline conditions [5, 6, 15, 16].
status of the host plant by reducing the concentration of
stress signaling molecule such as 1-aminocyclopropane-1-
carboxylate, a precursor of ethylene. Only three strains out of 3.5. In Vitro Colonisation of Salicornia Root System. Besides
the collection, belonging to the species Halomonas taeheungii performing in vitro activities involved in biostimulation,
∗ ∗
∗ ∗
(a) (b) (c)
Figure 5: Representative images of gfp-tagged Halomonas elongata strain on Salicornia root acquired through BP530/30 GFP filter (excitation
at 488 nm). (a) Fluorescence image showing gfp-H. elongata cells and microcolonies. (b) Bright field image of (a) showing Salicornia root
surface (open arrow) and root hairs (arrow). (c) Overlapping of images (a) and (b) showing the colonisation of Salicornia root surface (open
arrow in the upper right of the panel) and root hairs (arrow on the right side of the panel) by the gfp-tagged H. elongata strain. Asterisks
indicated in the bright field images (a and b) show the biofilm matrix associated with the root surface. The scale bars of the images in the
figure correspond to 10 𝜇m.
biocontrol, or biofertilization, to play an effective role in of salt crusts on the soil surface was a driving force involved
plant growth promotion, a bacterial strain should be able to in shaping the structure of the hypersaline soil dwelling
colonise the plant root system. The potential ability of PGP bacterial community. The same approach demonstrated that
isolates to efficiently colonise plant root system was tested Salicornia selected similar bacterial communities in the rhi-
by performing an adhesion assay exploiting a gfp-tagged zosphere, independently from the site of sampling. Notably,
PGP bacterium [57]. The adhesion test was performed on rhizosphere associated bacterial communities differed from
a nonhalophyte model plant, Arabidopsis thaliana, already that colonizing the root-free soil. Overall DGGE fingerprint-
used to study plant-microbe interactions [5, 58, 59], and ing indicated that a peculiar bacterial microbiome is stably
on a wild Salicornia collected in southern Italy. Different associated with Salicornia roots, possibly having a role in
rhizospheric bacterial strains belonging to the Halomonas promoting plant growth and stress tolerance.
genus were selected, based on their promising multiple PGP The establishment of a large collection of halophilic
activities in vitro and the ability to cope with several abiotic and halotolerant bacterial strains and their identification
stresses, as candidate for the chromosomal gfp-tagging. H.
widened the knowledge on the rhizocompetent bacterial
elongata strain BDV11S17A was successfully transformed with
community associated with halophytes in saline and arid
the gene encoding the Green Fluorescent Protein (gfp) that
soils. Furthermore, the isolation from the halophilic plant
was stably inserted in the bacterium genome. The gfp-labelled
Salicornia of a large collection of bacteria which tolerate
H. elongata BDV11S17A was used to track bacterial adhesion
temperature, saline, and osmotic stresses and also showed in
on Arabidopsis and Salicornia roots in vitro by exposing the
vitro the ability at medium-high salinity value (5% NaCl) to
roots to a gfp-labelled bacterial suspension for 16 hours.
(i) positively influence the nutrients and hormonal balance
The gfp-tagged strain was unable to colonise Arabidopsis
and (ii) putatively express biocontrol activity as indicated
root system, and despite several attempts and the analysis
by the protease activity test is a novelty presented in this
of different root specimens, only few cells were observed
at the fluorescence microscope. On the contrary, confocal study. Furthermore, the gfp-labelled PGP Halomonas elon-
analysis of Salicornia roots showed an extensive colonisation gata strain isolated from rhizospheric soil showed the ability
by gfp-labelled strain (Figure 5). H. elongata BDV11S17A-gfp, to massively adhere on Salicornia roots in vitro, demon-
previously shown to be able to grow under different stresses strating the suitability of halophilic plants rhizobacteria to
(high and low temperatures, high saline concentrations, and set up effective PGP inocula. The great potential of PGP
water stress) and to perform PGP activities in vitro, showed a halophilic and halotolerant bacteria should be carefully taken
good rhizocompetence efficiently colonizing Salicornia root into account to satisfy the increased need of food production
surface and root hairs. Such features make the strain a in the frame of a raising world population and ongoing
potential candidate for in vivo PGP experiments. climate changes. The present work contributes to expand
the current knowledge on PGP bacteria, presenting a wide
4. Conclusions bacterial strain collection that could be exploited to set up
specifically designed microbial consortia able to enhance
DGGE fingerprinting on the total bacterial microbiome plant growth and productivity in soils impacted by salt and
colonising the bulk soils showed that the presence/absence drought stresses.
Acknowledgments [11] E. P. Glenn, J. W. O’Leary, M. C. Watson, T. L. Thompson, and

R. O. Kuehl, “Salicornia bigelovii Torr.: an oilseed halophyte for
The authors would like to thank Dr. Umberto Fascio ((CIMA) seawater irrigation,” Science, vol. 251, no. 4997, pp. 1065–1067,
Centro Interdipartimentale di Microscopia avanzata of the 1991.
University of Milan) for technical support at the confocal [12] F. M. Attia, A. A. Alsobayel, M. S. Kriadees, M. Y. Al-Saiady,
microscope and Raffaella Tassoni for technical support in and M. S. Bayoumi, “Nutrient composition and feeding value
performing the PGP tests. The authors are grateful to Dr. of Salicornia bigelovii torr meal in broiler diets,” Animal Feed
Lotte Lambertsen for the kind gifts of the E. coli strains Science and Technology, vol. 65, no. 1–4, pp. 257–263, 1997.
for gfp-labelling of the bacteria. This work was financially [13] United Nations Secretary-General’s High-level Panel on Global
supported by the European Union in the ambit of Project Sustainability, Resilient People, Resilient Planet: A Future Worth
BIODESERT (European Community’s Seventh Framework Choosing, United Nations, New York, NY, USA, 2012.
Programme CSA-SA REGPOT-2008-2 under Grant agree- [14] C. J. Ruan, J. A. T. da Silva, S. Mopper, Q. Pei, and S. Lutts,
ment no. 245746). F. Mapelli and E. Rolli were supported “Halophyte improvement for a salinized world,” Critical Reviews
in Plant Sciences, vol. 29, no. 6, pp. 329–359, 2010.
by Università degli Studi di Milano, DeFENS, European
Social Found (FSE) and Regione Lombardia (Contract “Dote [15] D. Ramadoss, V. K. Lakkineni, P. Bose, S. Ali, and K. Anna-
purna, “Mitigation of salt stress in wheat seedlings by halotol-
Ricerca”). R. Marasco was supported by a fellowship in the
erant bacteria isolated from saline habitats,” SpringerPlus, vol. 2,
ambit of BIOGESTECA Project (n∘ 15083/RCC “Fondo per no. 1, article 6, 2013.
la promozione di accordi istituzionali”).
[16] S. Tiwari, P. Singh, R. Tiwari et al., “Salt-tolerant rhizobacteria-
mediated induced tolerance in wheat (Triticum aestivum) and
References chemical diversity in rhizosphere enhance plant growth,” Biol-
ogy and Fertility of Soils, vol. 47, no. 8, pp. 907–916, 2011.
[1] A. Balloi, E. Rolli, R. Marasco et al., “The role of microorgan- [17] M. Argandoña, R. Fernández-Carazo, I. Llamas et al., “The
isms in bioremediation and phytoremediation of polluted and moderately halophilic bacterium Halomonas maura is a free-
stressed soils,” Agrochimica, vol. 54, no. 6, pp. 353–369, 2010. living diazotroph,” FEMS Microbiology Letters, vol. 244, no. 1,
[2] L. E. de-Bashan, J. P. Hernandez, Y. Bashan, and R. M. Maier, pp. 69–74, 2005.
“Bacillus pumilus ES4: candidate plant growth-promoting bac- [18] Y. Bashan, M. Moreno, and E. Troyo, “Growth promotion of the
terium to enhance establishment of plants in mine tailings,” seawater-irrigated oilseed halophyte Salicornia bigelovii inoc-
Environmental and Experimental Botany, vol. 69, no. 3, pp. 343– ulated with mangrove rhizosphere bacteria and halotolerant
352, 2010. Azospirillum spp.,” Biology and Fertility of Soils, vol. 32, no. 4,
[3] D. Egamberdieva, F. Kamilova, S. Validov, L. Gafurova, Z. pp. 265–272, 2000.
Kucharova, and B. Lugtenberg, “High incidence of plant [19] I. Gontia, K. Kavita, M. Schmid, A. Hartmann, and B. Jha,
growth-stimulating bacteria associated with the rhizosphere of “Brachybacterium saurashtrense sp. nov., a halotolerant root-
wheat grown on salinated soil in Uzbekistan,” Environmental associated bacterium with plant growth-promoting potential,”
Microbiology, vol. 10, no. 1, pp. 1–9, 2008. International Journal of Systematic and Evolutionary Microbiol-
[4] R. Hayat, S. Ali, U. Amara, R. Khalid, and I. Ahmed, “Soil ogy, vol. 61, part 12, pp. 2799–2804, 2011.
beneficial bacteria and their role in plant growth promotion: a [20] T. Ozawa, J. Wu, and S. Fujii, “Effect of inoculation with
review,” Annals of Microbiology, vol. 60, no. 4, pp. 579–598, 2010. a strain of Pseudomonas pseudoalcaligenes isolated from the
endorhizosphere of Salicornia europea on salt tolerance of the
[5] R. Marasco, E. Rolli, B. Ettoumi et al., “A drought resistance-
glasswort,” Soil Science and Plant Nutrition, vol. 53, no. 1, pp.
promoting microbiome is selected by root system under desert
12–16, 2007.
farming,” PLoS ONE, vol. 7, no. 10, Article ID e48479, 2012.
[21] E. Rueda-Puente, T. Castellanos, E. Troyo-Diéguez, J. L. D. de
[6] S. K. Upadhyay, D. P. Singh, and R. Saikia, “Genetic diver- León-Alvarez, and B. Murillo-Amador, “Effects of a nitrogen-
sity of plant growth promoting rhizobacteria isolated from fixing indigenous bacterium (Klebsiella pneumoniae) on the
rhizospheric soil of wheat under saline condition,” Current growth and development of the halophyte Salicornia bigelovii as
Microbiology, vol. 59, no. 5, pp. 489–496, 2009. a new crop for saline environments,” Journal of Agronomy and
[7] S. Mayak, T. Tirosh, and B. R. Glick, “Plant growth-promoting Crop Science, vol. 189, no. 5, pp. 323–332, 2003.
bacteria confer resistance in tomato plants to salt stress,” Plant [22] B. Jha, I. Gontia, and A. Hartmann, “The roots of the halo-
Physiology and Biochemistry, vol. 42, no. 6, pp. 565–572, 2004. phyte Salicornia brachiata are a source of new halotolerant
[8] K. C. Ravindran, K. Venkatesana, V. Balakrishnana, K. P. Chel- diazotrophic bacteria with plant growth-promoting potential,”
lappana, and T. Balasubramanian, “Restoration of saline land by Plant and Soil, vol. 356, no. 1-2, pp. 265–277, 2012.
halophytes for Indian soils,” Soil Biology and Biochemistry, vol. [23] Forssk, Fl. Aeg. Arab. 2. 1775.
39, pp. 2661–2664, 2007. [24] E. Le Floc’h, L. Boulos, and E. Vela, Catalogue synonymique
[9] D. M. Al-Mailem, N. A. Sorkhoh, M. Marafie, H. Al-Awadhi, commenté de la flore de Tunisie, République tunisienne min-
M. Eliyas, and S. S. Radwan, “Oil phytoremediation potential istère de l’environnement et du développement durable banque
of hypersaline coasts of the Arabian Gulf using rhizosphere nationale de genes, 2010.
technology,” Bioresource Technology, vol. 101, no. 15, pp. 5786– [25] C. Schbereiter-Gurtner, G. Piñar, W. Lubitz, and S. Rölleke, “An
5792, 2010. advanced molecular strategy to identify bacterial communities
[10] E. Manousaki and N. Kalogerakis, “Halophytes present new on art objects,” Journal of Microbiological Methods, vol. 45, no.
opportunities in phytoremediation of heavy metals and saline 2, pp. 77–87, 2001.
soils,” Industrial and Engineering Chemistry Research, vol. 50, [26] G. Muyzer, E. C. de Waal, and A. G. Uitterlinden, “Profiling
no. 2, pp. 656–660, 2011. of complex microbial populations by denaturing gradient gel
electrophoresis analysis of polymerase chain reaction-amplified heterotrophic bacteria from the Great Salt Plains of Oklahoma,”
genes coding for 16S rRNA,” Applied and Environmental Micro- Microbial Ecology, vol. 48, no. 4, pp. 449–462, 2004.
biology, vol. 59, no. 3, pp. 695–700, 1993. [41] J. Tang, A. P. Zheng, E. S. P. Bromfield et al., “16S rRNA
[27] M. Cardinale, L. Brusetti, P. Quatrini et al., “Comparison of gene sequence analysis of halophilic and halotolerant bacteria
different primer sets for use in automated ribosomal intergenic isolated from a hypersaline pond in Sichuan, China,” Annals of
spacer analysis of complex bacterial communities,” Applied and Microbiology, vol. 61, no. 2, pp. 375–381, 2011.
Environmental Microbiology, vol. 70, no. 10, pp. 6147–6156, 2004. [42] R. R. de la Haba, D. R. Arahal, M. C. Márquez, and A. Ventosa,
[28] D. Daffonchio, S. Borin, G. Frova, P. L. Manachini, and C. “Phylogenetic relationships within the family Halomonadaceae
Sorlini, “PCR fingerprinting of whole genomes, the spacers based on comparative 23S and 16S rRNA gene sequence
between the 16S and 23S rRNA genes and of intergenic tRNA analysis,” International Journal of Systematic and Evolutionary
gene regions reveal a different intraspecific genomic variability Microbiology, vol. 60, no. 4, pp. 737–748, 2010.
of Bacillus cereus and Bacillus licheniformis,” International Jour-
[43] A. C. F. Dias, F. D. Andreote, F. Dini-Andreote et al., “Diversity
nal of Systematic Bacteriology, vol. 48, pp. 107–116, 1998.
and biotechnological potential of culturable bacteria from
[29] J. M. Bric, R. M. Bostock, and S. E. Silverstone, “Rapid in situ Brazilian mangrove sediment,” World Journal of Microbiology
assay for indoleacetic acid production by bacteria immobilized and Biotechnology, vol. 25, no. 7, pp. 1305–1311, 2009.
on a nitrocellulose membrane,” Applied and Environmental
Microbiology, vol. 57, no. 2, pp. 535–538, 1991. [44] K. A. El-Tarabily and T. Youssef, “Enhancement of morpholog-
ical, anatomical and physiological characteristics of seedlings
[30] F. Ahmad, I. Ahmad, and M. S. Khan, “Screening of free-living
of the mangrove Avicennia marina inoculated with a native
rhizospheric bacteria for their multiple plant growth promoting
phosphate-solubilizing isolate of Oceanobacillus picturae under
activities,” Microbiological Research, vol. 163, no. 2, pp. 173–181,
greenhouse conditions,” Plant and Soil, vol. 332, no. 1, pp. 147–
2008.
162, 2010.
[31] J. C. Cappuccino and N. Sherman, “Negative staining,” in
[45] B. E. Michel and M. R. Kaufmann, “The osmotic potential of
Microbiology: A Laboratory Manual, J. C. Cappuccino and N.
polyethylene glycol 6000,” Plant Physiology, vol. 51, no. 5, pp.
Sherman, Eds., pp. 125–179, Benjamin/Cummings, Redwood
914–917, 1973.
City, Calif, USA, 3rd edition, 1992.
[32] P. Nielsen and J. Sørensen, “Multi-target and medium- [46] N. P. Hua, A. Hamza-Chaffai, R. H. Vreeland, H. Isoda, and
independent fungal antagonism by hydrolytic enzymes in T. Naganuma, “Virgibacillus salarius sp. nov., a halophilic
Paenibacillus polymyxa and Bacillus pumilus strains from barley bacterium isolated from a Saharan salt lake,” International
rhizosphere,” FEMS Microbiology Ecology, vol. 22, no. 3, pp. 183– Journal of Systematic and Evolutionary Microbiology, vol. 58, no.
192, 1997. 10, pp. 2409–2414, 2008.
[33] D. M. Penrose and B. R. Glick, “Methods for isolating and char- [47] T. Quesada, M. Aguilera, J. A. Morillo, A. Ramos-Cormenzana,
acterizing ACC deaminase-containing plant growth-promoting and M. Monteoliva-Sánchez, “Virgibacillus olivae sp. nov.,
rhizobacteria,” Physiologia Plantarum, vol. 118, no. 1, pp. 10–15, isolated from waste wash-water from processing of Spanish-
2003. style green olives,” International Journal of Systematic and
[34] F. Poly, L. J. Monrozier, and R. Bally, “Improvement in the RFLP Evolutionary Microbiology, vol. 57, no. 5, pp. 906–910, 2007.
procedure for studying the diversity of nifH genes in communi- [48] I. Romano, L. Lama, P. Orlando, B. Nicolaus, A. Giordano,
ties of nitrogen fixers in soil,” Research in Microbiology, vol. 152, and A. Gambacorta, “Halomonas sinaiensis sp. nov., a novel
no. 1, pp. 95–103, 2001. halophilic bacterium isolated from a salt lake inside Ras
[35] L. Lambertsen, C. Sternberg, and S. Molin, “Mini-Tn7 trans- Muhammad Park, Egypt,” Extremophiles, vol. 11, no. 6, pp. 789–
posons for site-specific tagging of bacteria with fluorescent 796, 2007.
proteins,” Environmental Microbiology, vol. 6, no. 7, pp. 726–732, [49] C. Sánchez-Porro, R. R. de la Haba, N. Soto-Ramı́rez, M. C.
2004. Márquez, R. Montalvo-Rodrı́guez, and A. Ventosa, “Descrip-
[36] W. T. Liu, T. L. Marsh, H. Cheng, and L. J. Forney, “Characteriza- tion of Kushneria aurantia gen. nov., sp. nov., a novel member of
tion of microbial diversity by determining terminal restriction the family Halomonadaceae, and a proposal for reclassification
fragment length polymorphisms of genes encoding 16S rRNA,” of Halomonas marisflavi as Kushneria marisflavi comb. nov., of
Applied and Environmental Microbiology, vol. 63, no. 11, pp. Halomonas indalinina as Kushneria indalinina comb. nov. and
4516–4522, 1997. of Halomonas avicenniae as Kushneria avicenniae comb. nov,”
[37] M. M. Fisher and E. W. Triplett, “Automated approach for International Journal of Systematic and Evolutionary Microbiol-
ribosomal intergenic spacer analysis of microbial diversity and ogy, vol. 59, no. 2, pp. 397–405, 2009.
its application to freshwater bacterial communities,” Applied [50] N. Soto-Ramı́rez, C. Sánchez-Porro, S. Rosas et al., “Halomonas
and Environmental Microbiology, vol. 65, no. 10, pp. 4630–4636, avicenniae sp. nov., isolated from the salty leaves of the black
1999. mangrove Avicennia germinans in Puerto Rico,” International
[38] C. A. Lozupone and R. Knight, “Global patterns in bacterial Journal of Systematic and Evolutionary Microbiology, vol. 57, no.
diversity,” Proceedings of the National Academy of Sciences of the 5, pp. 900–905, 2007.
United States of America, vol. 104, no. 27, pp. 11436–11440, 2007. [51] S. Compant, B. Duffy, J. Nowak, C. Clément, and E. A. Barka,
[39] K. Smalla, G. Wieland, A. Buchner et al., “Bulk and rhizosphere “Use of plant growth-promoting bacteria for biocontrol of
soil bacterial communities studied by denaturing gradient plant diseases: principles, mechanisms of action, and future
gel electrophoresis: plant-dependent enrichment and seasonal prospects,” Applied and Environmental Microbiology, vol. 71, no.
shifts revealed,” Applied and Environmental Microbiology, vol. 9, pp. 4951–4959, 2005.
67, no. 10, pp. 4742–4751, 2001. [52] S. Compant, C. Clément, and A. Sessitsch, “Plant growth-
[40] T. M. Caton, L. R. Witte, H. D. Ngyuen, J. A. Buchheim, M. promoting bacteria in the rhizo- and endosphere of plants:
A. Buchheim, and M. A. Schneegurt, “Halotolerant aerobic their role, colonization, mechanisms involved and prospects for
utilization,” Soil Biology and Biochemistry, vol. 42, no. 5, pp.

669–678, 2010.
[53] R. S. Redman, Y. O. Kim, C. J. D. A. Woodward et al.,
“Increased fitness of rice plants to abiotic stress via habitat
adapted symbiosis: a strategy for mitigating impacts of climate
change,” PLoS ONE, vol. 6, no. 7, Article ID e14823, 2011.
[54] S. Timmusk, V. Paalme, T. Pavlicek et al., “Bacterial distribution
in the rhizosphere of wild barley under contrasting microcli-
mates,” PLoS ONE, vol. 6, no. 3, Article ID e17968, 2011.
[55] M. A. Siddikee, P. S. Chauhan, R. Anandham, G. H. Han, and
T. Sa, “Isolation, characterization, and use for plant growth
promotion under salt stress, of ACC deaminase-producing
halotolerant bacteria derived from coastal soil,” Journal of
Microbiology and Biotechnology, vol. 20, no. 11, pp. 1577–1584,
2010.
[56] V. Sgroy, F. Cassán, O. Masciarelli, M. F. del Papa, A. Lagares,
and V. Luna, “Isolation and characterization of endophytic plant
growth-promoting (PGPB) or stress homeostasis-regulating
(PSHB) bacteria associated to the halophyte Prosopis strombu-
lifera,” Applied Microbiology and Biotechnology, vol. 85, no. 2, pp.
371–381, 2009.
[57] M. Bacilio, H. Rodriguez, M. Moreno, J. P. Hernandez, and Y.
Bashan, “Mitigation of salt stress in wheat seedlings by a gfp-
tagged Azospirillum lipoferum,” Biology and Fertility of Soils, vol.
40, no. 3, pp. 188–193, 2004.
[58] B. Fan, X. H. Chen, A. Budiharjo, W. Bleiss, J. Vater, and
R. Borriss, “Efficient colonization of plant roots by the plant
growth promoting bacterium Bacillus amyloliquefaciens FZB42,
engineered to express green fluorescent protein,” Journal of
[59] B. Fan, R. Borriss, W. Bleiss, and X. Wu, “Gram-positive rhi-
zobacterium Bacillus amyloliquefaciens FZB42 colonizes three
types of plants in different patterns,” Journal of Microbiology, vol.
50, no. 1, pp. 38–44, 2012.
http://dx.doi.org/10.1155/2013/387230
Research Article
Streptomyces rochei ACTA1551, an Indigenous Greek
Isolate Studied as a Potential Biocontrol Agent against
Fusarium oxysporum f.sp. lycopersici
Grammatiki S. Kanini, Efstathios A. Katsifas,

Alexandros L. Savvides, and Amalia D. Karagouni
National and Kapodistrian University of Athens, Faculty of Biology, Department of Botany, Microbiology Group,
Panepistimioupolis, Zografou, 15781 Athens, Greece
Correspondence should be addressed to Amalia D. Karagouni; akar@biol.uoa.gr
Received 15 March 2013; Accepted 24 April 2013
Copyright © 2013 Grammatiki S. Kanini et al. This is an open access article distributed under the Creative Commons Attribution
cited.
Many studies have shown that several Greek ecosystems inhabit very interesting bacteria with biotechnological properties. Therefore
Streptomyces isolates from diverse Greek habitats were selected for their antifungal activity against the common phytopathogenic
fungus Fusarium oxysporum. The isolate encoded ACTA1551, member of Streptomyces genus, could strongly suppress the fungal
growth when examined in antagonistic bioassays in vitro. The isolate was found phylogenetically relative to Streptomyces rochei after
analyzing its 16S rDNA sequence. The influence of different environmental conditions, such as medium composition, temperature,
and pH on the expression of the antifungal activity was thoroughly examined. Streptomyces rochei ACTA1551 was able to protect
tomato seeds from F. oxysporum infection in vivo while it was shown to promote the growth of tomato plants when the pathogen
was absent. In an initial effort towards the elucidation of the biochemical and physiological nature of ACTA1551 antifungal
activity, extracts from solid streptomycete cultures under antagonistic or/and not antagonistic conditions were concentrated
and fractionated. The metabolites involved in the antagonistic action of the isolate showed to be more than one and produced
independently of the presence of the pathogen. The above observations could support the application of Streptomyces rochei
ACTA1551 as biocontrol agent against F. oxysporum.
1. Introduction Mexico, Spain, Sweden, The Netherlands, UK, and USA, had
reported the disease [4, 5] and results in severe losses in the
Fusarium wilt diseases, including symptoms like wilting, greenhouses, open field crops, and hydroponic cultures.
chlorosis, necrosis, premature leaf drop, browning of the Like all formae speciales of F. oxysporum, this fungus
vascular system, stunting, and damping-off, were caused by is a soil inhabitant and extremely difficult to control [6, 7].
pathogenic forma speciales of Fusarium oxysporum. These However, several control methods have been used so far.
symptoms are the results of fungal spores and mycelium that Chemical treatments using benomyl, carbendazim, prochlo-
block the xylem, toxin production and host defense responses raz, fludioxonil, bromuconazole, or azoxystrobin [8] not only
such as tyloses, gums, and gels [1]. Fusarium wilting leads to result in fungicide resistant pathogens [9] but also cannot
severe losses in tomato and a variety of other crop plants [2]. be characterized from a wide range of action while they are
Fusarium oxysporum f.sp. lycopersici, a soilborne plant involved directly in the atmospheric pollution, undesirable
pathogen in the class Hyphomycetes, causes Fusarium wilt effects on nontarget organisms, and possible carcinogenicity
specifically in tomato. This disease was first described by [10–14]. Additionally, soil amendments have been used in
Massee in England in 1985 [3]. It is of worldwide economic order to reduce the severity of the pathogen in crops by
importance as at least 32 countries, including Australia, Bel- increasing the soil pH using fertilizers containing nitrate
gium, Canada, France, Germany, Greece, Israel, Italy, Japan, nitrogen [15] or by crop rotation [16, 17], but they have
rarely provided long-term control in any production area. and the optimal conditions of its antagonistic behavior,
Currently, resistant cultivars have been used for the control examination of culture extracts for antifungal activity, and
of the disease [18, 19], but research was hampered by the lack studies to control the Fusarium oxysporum in vivo, using the
of knowledge of the fungus genetic variability. plant Lycopersicon esculentum (Family: Solanaceae, Common
Thus, the above difficulties in controlling Fusarium wilt name: Tomato) as a model target aiming to complete the
diseases have stimulated renewed interest in biological con- profile of Streptomyces rochei ACTA1551 as biocontrol agent.
trol as a disease management alternative [6]. For this purpose,
root colonizing plant beneficial microorganisms have been
used including species of Pseudomonas (Pseudomonas fluo-
2. Materials and Methods
rescens, P. putida, P. aureofaciens) [20–24], Bacillus (Bacillus 2.1. Microbial Strains. Streptomyces strains isolated from the
subtilis, B. Polymyxa, and B. cereus) [25–27], non-patho- rhizosphere of indigenous plants (Table 1(A)) and nonrhizo-
genic Fusarium [28, 29], and Actinobacteria [30–33]. These sphere samples of special Greek habitats (Table 1(B)) were
microbes possess many traits that make them well suited as examined for their antifungal activity against Fusarium
biocontrol and growth-promoting agents. These include the oxysporum. The streptomycetes were maintained as spore
ability to (i) grow rapidly in vitro and to be mass produced; (ii) suspensions in 30% (w/v) glycerol at −20∘ C [49].
rapidly utilize seed and root exudates; (iii) colonize and mul- Fusarium oxysporum DSM62059 (Fusarium oxysporum
tiply in the rhizosphere and spermosphere environments and Schlechtendahl: Fries f.sp. lycopersici, a laboratory strain
in the interior of the plant; (iv) produce a wide spectrum of provided by the DSMZ culture collection) was used as fungal
bioactive metabolites (i.e., antibiotics, siderophores, volatiles, target. The pathogenic fungal strain was grown on Potato
and growth-promoting substances); (v) compete aggressively Dextrose Agar (PDA) or in Potato Dextrose Broth (PDB) at
with other microorganisms and (vi) adapt to environmental 25∘ C and was maintained on PDA at 4∘ C.
stresses. The major weakness of pseudomonads as biocontrol
agents is their inability to produce resting spores, which
2.2. Biocontrol Assay In Vitro. Streptomycete aliquots (30 𝜇L)
complicates formulation of the bacteria for commercial use
from a spore suspension in 30% (w/v) glycerol were used as
[20]. Bacillus strains even if they are well equipped genetically
inoculum for all in vitro antagonism bioassays. For the same
to produce a vast array of antibiotics, only a limited part of
test we used two full-loops of F. oxysporum mycelium from a
this antibiotic-devoted genetic background may be readily
5-day-old culture on PDA.
expressed in soil and thus, only a part of this arsenal may
The ability of the Streptomyces isolates to inhibit the fun-
be actually produced under natural conditions [34]. The
gal growth was determined using an agar plate antagonism
nonpathogenic Fusarium strains need to be applied in large
bioassay [33]. F. oxysporum inoculums was placed on either
concentrations as the basic mode of action of these biocontrol
side of a two-days-old colony of each streptomycete. After
agents is the competition of nutrients that leads to fungistasis
incubation at 28∘ C for 5 days the inhibition zone was formed
(inhibition of chlamydospores germination).
and its presence characterized the Streptomyces isolate as
The species of the genus Streptomyces belong to the
positive.
phylum of Actinobacteria characterized by a wide range of
For the quantification of the antifungal activity the
modes of action like antibiotic production, lysis of fungal cell
average (triplicate plates per strain × three independent
walls; competition and hyperparasitism have been proved to
experimental sets) of the quotient of the area of the inhibition
be effective biocontrol agents either alone or in combination
zone, which was formed around and over the area of the strep-
with other biocontrol agents. Several Streptomyces species
tomycete colony itself, was calculated (Antifungal activity =
such as S. lydicus, S. lividans, S. olivaceoviridis, S. scabies, S.
plicatus, S. hydroscopicus, S. violaceusniger, S. humidus, S. 𝜋𝑅𝑧2 /𝜋𝑅str
2
(𝑅𝑧 = radius of inhibition zone and 𝑅str = radius of
avermitilis, S. aureofaciens, and S. roseoflavus are well-known streptomycete colony)) [33].
producers of important compounds that are active against a
wide variety of fungal pathogens [35]. These include a 2.3. Taxonomy of Streptomycetes. The selected isolate was
wide range of antibiotics as well as a variety of enzymes further characterized through the amplification of its 16S
(i.e., chitinases), which degrade the fungal cell wall [35– rRNA gene. The 16S rDNA fragment was amplified by PCR
40]. Metabolites from streptomycetes have been used in using two universal primers [50, 51]: pA (5󸀠 -AGA GTT
agriculture as growth promoters [41–44] and selected strains TGA TCC TGG CTC AG-3󸀠 ) and R1492 (5󸀠 -TAC GGY
of the genus also have been used as direct biocontrol agents TAC CTT GTT ACG ACT T-3󸀠 ). Amplification reactions
for other plant diseases [30, 45–48]. were performed in volumes of 50 𝜇L containing 40 ng
In this study, a total of 528 Streptomyces isolates from template DNA, 0.4 𝜇M of each primer, 1X buffer with Mg2+ ,
the Athens University Microbiology Laboratory Culture Col- 1 unit of Taq DNA Polymerase (Biotools England) and
lection were examined for their antifungal activity against 0.2 mM dNTPs. Nucleases free water was used to bring
Fusarium oxysporum. The measurement of their antifungal the reaction volume to 50 𝜇L. After initial denaturation at
strength in vitro resulted in the selection of Streptomyces 95∘ C for 2 min, samples were cycled for 30 PCR cycles
rochei ACTA1551 as the isolate that expressed the higher using the following cycle profile: 95∘ C denaturation for
antagonistic activity. Streptomyces rochei ACTA1551 was used 30 sec, primer annealing at 53∘ C for 30 sec, and primer
for further studies evaluating the potential of its use as extension at 72∘ C for 2 min, plus a final 2 min elongation
biocontrol agent. We employed determination of the range step at 72∘ C. Amplified PCR products were separated by
Table 1: Streptomyces strains with antifungal activity originated from twelve studied Greek habitats.
Number (percentage) of isolates

Number of with antifungal activity against
Sampling area
isolates tested F. oxysporum DSM62059
{highest; lowest; average}1
(A) Rhizosphere samples
(1) Rhizosphere of Ebenus sibthorpii 28 3 (10.7%) {2.36; 1.3; 1.5}
(2) Rhizosphere of Ceratonia siliqua 42 1 (2.4%) {1.96; 1.96; 1.96}
(3) Rhizosphere of Olea europea 73 6 (8.2%) {2.57; 2.13; 1.34}
(4) Rhizosphere of Pinus brutia from Crete 20 3 (15.0%) {4.84; 3.26; 1.56}
(5) Rhizosphere of evergreen woody shrubs from an island of the Aegean Sea 21 5 (23.8%) {2.69; 1.87; 1.47}
(6) Rhizosphere of evergreen woody shrubs from an island of the Ionian Sea 28 0 (0.0%)
(7) Rhizosphere of coniferous trees (Arcadian forest) 84 6 (7.1%) {2.91; 2.05; 1.56}
Rhizosphere subtotals 296 21 (7.0%)
(B) Nonrhizosphere samples
(8) Hot spring water of Thermopiles thermal springs (Viotia district) 22 2 (9.1%) {1.63; 1.46; 1.28}
(9) Sediment from a volcanic area (Santorini island—Aegean sea) 5 1 (20%) {1.3; 1.3; 1.3}
(10) Soil derived from cultivated area (Marathon, Attica district) 179 10 (5.6%) {2.25; 1.6; 1.25}
(11) Soil from protected natural forest area (Kessariani, Attica District) 26 2 (7.7%) {1.39; 1.37; 1.34}
Nonrhizosphere subtotals 232 15 (6.0%)
Total 528 39 (7.0%)
1
Antagonistic activity levels as expressed by the quotient of the inhibition zone area over streptomycete colony area (See Section 2.2).
gel electrophoresis on 1.2% (w/v) agarose gel and then and Wellington (1990) [49] and used for the tomato seed
purified using Nucleospin Extract PCR kit (Macherey-Nagel, treatments. F. oxysporum was cultured in Nutrient Broth (NB,
Germany). The 16S rDNA fragment (>1400 bp) was fully Biokar Diagnostics, Beauvais, France) for 5 days at 28∘ C and
sequenced (Macrogen, Republic of Korea) and the results 180 rpm. The mycelium was aseptically collected on filter
were used for strain identification. For the detection of closest paper and washed with three culture volumes of deionized-
relatives, all sequences were compared with the BLAST sterile water. Three grams of wet mycelium (dry weight 15
function (http://www.ncbi.nlm.nih.gov/BLAST/). Sequence to 18% w/w) were resuspended in 1000 mL deionized sterile
data were compiled using the MEGA 5.10 software [52] and water and were thoroughly mixed with 1 kg of sterile soil [55].
aligned with sequences obtained from the GenBank (http:// Streptomyces isolate encoded ACTA1551 was selected for
www.ncbi.nlm.nih.gov/) databases using the ClustalW align- in vivo experiments due to the strong suppression it caused to
ing utility. Phylogenetic analysis was performed using F. oxysporum growth in vitro. A sandy silt loam soil (ASTM
neighbour joining method implemented in MEGA 5.10. One classification) with a pH of 7.9, taken from an area under
thousand bootstrap analyses (distance) were conducted. intense agricultural exploitation in the Marathon area (42 km
NE from the centre of Athens), was used. Prior to its use, the
2.4. Influence of Culture Conditions on Antagonistic Activity. soil was air-dried in the dark at 22∘ C for at least 3 months,
The agar plate antagonism bioassay was repeated for the passed through a 2 mm sieve, and autoclaved twice (121∘ C,
selected Streptomyces isolate using five different solid media; 60 min) on two separate days.
PDA (Potato Dextrose Agar, DSMZ media no. 129), AGS Tomato seeds were sterilised for 30 minutes in a 20%
(Arginine Glycerol Salts, [49]), NA (Nutriend Agar, Biokar chlorine suspension and then dried under sterile conditions.
Diagnostics, Beauvais, France), CzA (Czapek Agar, [53]) and A number of sterile seeds were immersed into a suspension
SAA (Streptomyces Antibiotic Agar, [53]), and four different of streptomycetes spores in Ringer 1/4 salt solution [54]
pH values (pH 6, pH 7, pH 8, and pH 9). All media—pH (109 spores per mL) for 30 minutes and then dried under
combination plates were incubated at 26∘ C, 28∘ C, and 30∘ C sterile conditions. Untreated sterile tomato seeds and sterile
for 2 days prior to fungal inoculation and five days after that. tomato seeds treated with the selected streptomycetes were
planted in pots containing sterile soil either amended with
2.5. Biocontrol Assay In Vivo. For in vivo antagonism tests, a F. oxysporum (3 g of wet washed mycelium per kg of soil) or
suspension of streptomycetes spores in Ringer 1/4 salt solu- not [55]. For every treatment, 20 seeds were planted in each
tion (NaCl 2.15 g/L, KCl 0.15 g/L, CaCl2 0.075 g/L, K2 HPO4 pot (three replicates for each pot were prepared). Each full
0.5 g/L according to Wellington et al., 1990 [54]) (109 spores experiment was conducted in four different occasions, over a
per mL) was prepared from a 5-day-old culture on Argi- time period of eight months. The pots were incubated at 28∘ C
nine Glycerol Salts Agar (AGS), as described by Herron under fluorescent light, and moisture was controlled daily
at the level of 40% (w/w) for 25 days. The number of seeds antagonistic isolates was 7% (39 isolates out of the 528 tested
that survived and/or germinated was evaluated in order to could suppress the growth of F. oxysporum) (Table 1).
estimate the ability of the examined streptomycete to control The sampling area that gave the higher percentage of
the fungi in vivo. In addition, the height and weight of the isolates that could suppress the growth of F. oxysporum was
emerged plants were measured for the estimation of the in the rhizosphere of evergreen woody shrubs from an island
vivo antagonism strength. of the Aegean Sea but the strongest antifungal activity was
observed by streptomycetes derived from the rhizosphere of
Pinus brutia (Crete). The streptomycetes that were isolated
2.6. Extraction and Fractionation of Streptomycete Metabolites from the rhizosphere of evergreen woody shrubs from an
from Solid Cultures. In parallel, the selected Streptomyces iso- island of the Ionian Sea lacked any antifungal activity, whilst
late was grown on SAA (Streptomyces Antibiotic Agar, [53]) very low numbers of potential biocontrol agents rose from
since it was selected as optimum medium for high antifungal the rhizosphere of Ceratonia siliqua or the soil derived from
activity expression. The cultures were incubated at 28∘ C for 7 cultivated area (Marathon, Attica District).
days. The medium around the formed streptomycete colony Although the percentage of the antagonistic isolates
in a radius of 2 cm was cut off and blended for three minutes. screened was low, these results in comparison with previous
The slurry was centrifuged at 4.000 g for 60 min and the observations [33, 56–58] support the hypothesis that the
cell free supernatant was collected. For the determination Greek Streptomyces isolates are multiactive and promising
of antifungal activity, 400 𝜇L from the concentrated culture for their application in several biotechnological processes,
supernatant were placed into wells on SAA plates (formed including biocontrol of soil borne fungal pathogens.
using a cork borer—diameter 1 cm, depth 1 cm) that were Focusing on the ecophysiological characteristics of the
inoculated with the fungus. sampling areas that gave the most active isolates, we can
Additionally, the inhibition zones of cocultures of the conclude the following correlation: the Aegean Sea climate
selected streptomycete ACTA1551 and F. oxysporum on SAA provides environmental stress on the indigenous Strepto-
plates were removed and blended for three minutes. The myces populations as long as periods of high temperatures
slurry was collected by centrifugation as described above. and drought are reflected in a soil poor in nutrients. Thus, the
After filtration, 400 𝜇L were placed in a similar manner into mentioned soil microorganisms with antagonistic properties
wells on SAA agar plates inoculated with the fungus for the seemed to be dominant. Furthermore the participation of the
determination of the antifungal activity. anthropogenic impact on the Aegean Sea islands, could prob-
All extracts from the solid cultures were fractionated into ably explain the high percentages of isolates collected from
a high molecular weight (protein) and a low molecular weight the rhizosphere soil of the plant Pinus brutia (indigenous
(nonprotein) component on a Pharmacia PD-10 gel filtration plant of Crete) and of the evergreen shrubs that were antag-
column using de-ionized water for elution, according to the onistic to F. oxysporum in vitro. On the contrary, the very
manufacturer’s recommendation. Each fraction was concen- different climatic conditions of the soil of the rhizosphere
trated by lyophilisation and examined for antifungal activity. of evergreen shrubs in the Ionian Sea islands (areas of high
All the experiments performed were in three independent humidity and low anthropogenic impact) resulted in isolates
sets using triplicate plates. which were not able to suppress the growth of the pathogenic
target.
2.7. Antimicrobial Activity of the Bioactive Extractions against The analysis of the above data resulted in the selection
Common Microbial Indicators. 400 𝜇L from the agar extracts of the isolate encoded ACTA1551, derived from the rhizo-
of the five streptomycetes were placed into wells (diameter sphere soil of the plant Pinus brutia (indigenous plant of
1 cm) of agar plates (triplicate plates, three individual experi- Crete) and expressed the highest antifungal activity (4.84
mental tests) inoculated with Escherichia coli, Bacillus subtilis, ± 0.05) of all the 39 isolates that were antagonistic against
Pseudomonas fluorescens, and Saccharomyces cerevisiae. The F. oxysporum. Analysis of 16S rRNA gene sequence data
inhibition zones were observed and measured. grouped the selected isolate to the species of Streptomyces
rochei (99% identity with the 16S rDNA sequence of the
closest phylogenetic relative, GenBank accession Number
2.8. Data Analysis. Statistical analysis for the various data JN167525). The position of the selected streptomycete among
sets was conducted through paired/unpaired 𝑡-tests, using the phylum of Actinobacteria is shown in Figure 1.
SifmaStat/Plot software program (ver. 12.0; Systat Software
Inc., Chicago, IL, USA). In all runs, a significance level of 3.2. The effect of Nutrients, Temperature, and pH on Antag-
<0.05 was used. onistic Activity. Incubation of the selected streptomycete on
SAA, containing glucose as carbon source and peptone from
3. Results and Discussion soya as nitrogen source, at 28∘ C and pH 8, was determined as
the optimum culture conditions for maximum antagonistic
3.1. Biocontrol Assay In Vitro. A total of 528 streptomycetes activity against F. oxysporum (Figure 2). The temperature of
were isolated using selective media, assigned to the genus 28∘ C is very close to the average of Mediterranean climatic
Streptomyces on the basis of their phenotypic characteristics. temperature, while pH 8 is the value that characterizes the
Isolates with inhibitory characteristics were selected and Greek cultivar soil. Therefore, one might suggest that the
screened by the agar plate assay. The total percentage of antifungal activity of the selected streptomycete ACTA1551
gi|75992798|gb|DQ192118.1|Streptomyces exfoliatus isolate GRE23 16S ribosomal RNA gene partial sequence
gi|75992792|gb|DQ192112.1|Streptomyces cyaneus isolate GRE28 16S ribosomal RNA gene partial sequence
88
gi|75992791|gb|DQ192111.1|Streptomyces cyaneus isolate GRE8 16S ribosomal RNA gene partial sequence
gi|219846073|ref|NR 025663.1|Streptomyces aureus strain B7319 16S ribosomal RNA partial sequence
84 gi|343204837|ref|NR 043822.1|Streptomyces kanamyceticus strain NRRL B-2535 16S ribosomal RNA partial sequence
gi|3550677|emb|Y15504.1|Streptomyces aureofaciens 16S rRNA gene strain NRRL 2209
Streptomyces
58 gi|444439472|ref|NR 074787.1|Streptomyces griseus subsp. griseus NBRC 13350 strain NBRC 13350 16S ribosomal RNA complete sequence
83 gi|444304135|ref|NR 074559.1|Streptomyces flavogriseus ATCC 33331 strain ATCC 33331 16S ribosomal RNA complete sequence
89 gi|75992801|gb|DQ192121.1|Streptomyces exfoliatus isolate GRE18 16S ribosomal RNA gene partial sequence
92 gi|90960097|dbj|AB184281.1|Streptomyces lydicus gene for 16S rRNA partial sequence strain: NBRC 13058
gi|340537755|gb|JN167524.1|Streptomyces fulvissimus strain GRE24 16S ribosomal RNA gene partial sequence
gi|343200523|ref|NR 041210.1|Streptomyces fulvissimus strain NBRC 3717 16S ribosomal RNA partial sequence
gi|343200438|ref|NR 041125.1|Streptomyces flavofungini strain NBRC 13371 16S ribosomal RNA partial sequence
98
gi|343202890|ref|NR 043383.1|Streptomyces vinaceusdrappus strain NRRL 2363 16S ribosomal RNA partial sequence
74
61 gi|343200404|ref|NR 041091.1|Streptomyces rochei strain NBRC 12908 16S ribosomal RNA partial sequence
gi|340537756|gb|JN167525.1|Streptomyces rochei strain GRE25 16S ribosomal RNA gene partial sequence
gi|2052220|emb|Z76678.1|S.coelicolor (DSM 40233T) 16S rRNA gene
99 gi|3859859|gb|AF064702.1|AF064702 Cellulomonas iranensis strain O 16S ribosomal RNA gene partial sequence
gi|27657416|emb|AJ536198.1| Micrococcus luteus 16S rRNA gene type strain DSM 20030T
gi|1747341|emb|X93187.1| A.philippinensis 16S rRNA gene
actinobacteria
98
68 gi|535086|emb|X72779.1| D.aurantiacum gene for 16S rRNA (DSM 43157)
Other
gi|1550757|emb|X92355.1| G.obscurus 16S ribosomal RNA (strain DSM43161)
32 96 gi|860736|emb|X84447.1| Actinosynnema mirum 16S rRNA gene strain DSM 43827T
33 gi|5915972|gb|AF114812.1| Saccharothrix syringae 16S ribosomal RNA gene partial sequence
gi|28933391|gb|AF430059.1| Nocardia veterana strain DSM 40777 16S ribosomal RNA gene complete sequence
29
100 gi|640002|emb|X80625.1| Rhodococcus ruber 16S rRNA gene strain DSM43338T
gi|5139793|dbj|AB028654.1| Planobispora rosea gene for 16S ribosomal RNA
gi|125487485|gb|EF217305.1| Agrobacterium tumefaciens strain 204 16S ribosomal RNA gene partial sequence
0.02
Figure 1: Phylogenetic tree of the 16S rDNA based on the neighbour-joining method, showing the position of the Streptomyces rochei
ACTA1551. One thousand bootstrap analyses (distance) were conducted. The tree was rooted with Agrobacterium tumefaciens. Scale bar
represents 2% estimated distance.
10,00 10,00
9,00 9,00
8,00 8,00
Antifungal activity
Antifungal activity
7,00 7,00
6,00 6,00
5,00 5,00
4,00 4,00
3,00 3,00
2,00 2,00
1,00 1,00
0,00 0,00
26 28 30 6 7 8 9
Temperature (∘ C) pH
(a) (b)
10,00
9,00
8,00
Antifungal activity
7,00
6,00
5,00
4,00
3,00
2,00
1,00
0,00
PDA NA SAA AGS CzA
(c)
Figure 2: Influence of temperature, pH value, and media on the strength of antifungal activity expressed from ACTA1551 (Streptomyces rochei)
on Fusarium oxysporum; (a) temperature effect, (b) pH effect, (c) carbon and nitrogen source effect.
could be enhanced under the above conditions which are activity expressed in vitro (Table 2) and the observation
similar to real farming processes. from previous work [33, 56–60] that it is multiproducer of
bioactive substances. Thus, it was characterized as promising
biocontrol agent in vivo.
3.3. Biocontrol Assay In Vivo. Using the results from the Analysis of particle size of the soil that was used for this
in vitro antagonistic assay we were led to the selection of purpose indicated the presence (%, dry weight) of sand, 50;
ACTA1551 (Streptomyces rochei) so as to use it for in vivo silt, 36; clay, 14. Mineralogy analysis showed the presence of
studies. This selection was based on its very high antifungal (%, dry weight) illite, 65; chlorite, 7; kaolinite, 10; smectite,
Table 2: Germination data of tomato seeds during in vivo antagonism bioassays. The total number of planted tomato seeds was (20 per pot)
× (three replicates for each pot) × (four independent experiments) = 240. All experimental data were combined for the analysis.
Soil infected with Fusarium oxysporum Soil not infected with Fusarium oxysporum
Number of Number of Number of Number of
Germination Germination
planted germinated planted germinated
percentage1 percentage1
tomato seeds tomato seeds tomato seeds tomato seeds
Seeds treated with
240 181 ± 2 75% 240 218 ± 2.2 90%
ACTA1551
Untreated tomato
240 108 ± 1 45% 240 228 ± 1.9 95%
seeds
1
Unpaired 𝑡-test on the number of germinated seeds: seeds treated with ACTA1551 in infected soil versus untreated tomato seeds in infected soil were
significantly different (𝑃 < 0.001); seeds treated with ACTA1551 in noninfected soil versus untreated tomato seeds in noninfected soil were significantly
different (𝑃 < 0.001).
Table 3: Mean height and weight data of tomato plants derived from emerged tomato seeds during in vivo antagonism bioassays. The total
number of planted tomato seeds was (24 per pot) × (three replicates for each pot) × (four independent experiments) = 288. All experimental
data were combined for the analysis.
Soil infected with Fusarium oxysporum Soil not infected with Fusarium oxysporum
Mean height (cm)1 Mean weight (g)2 Mean height (cm)1 Mean weight (g)2
Seeds treated with ACTA1551 13.42 ± 0.1 0.58 ± 0.01 19 ± 0.2 0.79 ± 0.01
Untreated tomato seeds 11.16 ± 0.1 0.35 ± 0.01 15.01 ± 0.1 0.51 ± 0.01
1
Unpaired 𝑡-test on plant heights: seeds treated with ACTA1551 in infected soil versus untreated tomato seeds in infected soil were significantly different (𝑃 <
0.001); seeds treated with ACTA1551 in noninfected soil versus untreated tomato seeds in noninfected soil were significantly different (𝑃 < 0.001).
2
Unpaired 𝑡-test on plant weights: seeds treated with ACTA1551 in infected soil versus untreated tomato seeds in infected soil significantly different (𝑃 < 0.001);
seeds treated with ACTA1551 in noninfected soil versus untreated tomato seeds in noninfected soil were significantly different (𝑃 < 0.001).
12; talc, 6 and calcite <1. Phosphorus content was 124 mg/L the ability of ACTA1551 to protect tomato seed from the
dry soil and organic carbon was 1.23% (dry weight). pathogenic effects of F. oxysporum (Table 3, Figure 3).
The first evidence of the effective usage of ACTA1551 as Moreover, the in vivo results for noninfected soil showed
biocontrol agent in vivo came from macroscopic observa- that the presence of ACTA1551 produced plants of statisti-
tions. The F. oxysporum infected soil, where untreated tomato cally significantly higher mean values of weight and height
seeds were planted, was fully cover by the white mycelium of (𝑃 ≤ 0.001) compared to the positive controls (untreated
the fungi, while only limited white spots of fungal mycelium tomato seeds planted into noninfected soil). The presence of
were observed onto infected soil where ACTA1551 treated ACTA1551 on the seeds promoted the growth of the emerged
tomato seeds were planted. plants as the height of the treated plants in noninfected soil
Considering the emergence of the tomato plants, the in was greater than the height of untreated plants in noninfected
vivo experiments showed that the tomato seeds treated with soil and the treated plants were heavier than the untreated,
ACTA1551 germinated in statistically significant (𝑃 < 0.001) in both cases, in a statistically significant level (𝑃 ≤ 0.001)
higher rates into F. oxysporum infected soil compared to (Table 3, Figure 3).
untreated seeds. Even if the germination of ACTA1551 treated The above results provided strong evidence that
tomato seeds was lower than the positive control treatment ACTA1551 could protect tomato plants from wilt symptoms
(untreated tomato seeds planted into noninfected soil), in a caused from F. oxysporum under the studied conditions and
statistically significant level (𝑃 < 0.001), it is clearly higher are in agreement with those from previews studies on the
compared to the results when ACTA1551 was absent. These potential of Streptomyces isolates to be used as biocontrol
results provide evidence for an effective antifungal behavior agents. For instance, Reddi and Rao [61] reported that isolates
of the selected streptomycete in vivo. However, the presence of Streptomyces ambofaciens were able to control Pythium
of ACTA1551 on the tomato seeds couldnot enhance the damping-off in tomato plants and Fusarium wilt in cotton
germination rate when planted in noninfected soil (Table 2). plants, in an artificially infested soil while Mahadevan and
The analysis of the estimated values of mean weight and Crawford [62] found that Streptomyces lydicus was identified
height of the emerged tomato plants showed that although as a broad spectrum biocontrol agent and the work of Farrag
the number of untreated plants that emerged in infected [63] enhanced that finding.
soil was low, the plants themselves were statistically signif- Sterilization of soils by pasteurisation, fumigation, or
icantly higher and stronger than the plants emerged from autoclaving usually allows the pathogen to proliferate giv-
untreated seed in infected soil (𝑃 < 0.001). The treatment of ing the opportunity to the biocontrol agent to express its
the seeds with ACTA1551 enhanced the weight and the height antagonistic activity under optimal conditions for the target’s
of the plants compared to the negative control (untreated growth. Thus, the results from these experiments clearly
tomato seeds planted into infected soil). These results showed suggested the antagonistic relationship between the pathogen
(a) (b) (c) (d)
Figure 3: In vivo antifungal ability of ACTA1551 (Streptomyces rochei). Growth of the tomato plants (from left to right). (a) Untreated seed
planted in untreated sterile soil (positive control), (b) seed treated with ACTA1551 (Streptomyces rochei) and planted in F. oxysporum infected
soil, (c) seed treated with ACTA1551 (Streptomyces rochei) and planted in F. oxysporum noninfected soil, (d) untreated seed planted in R. solani
DSM843 infected soil (negative control).
and the biocontrol agent. Other observations have noted that 3.4. Extraction and Fractionation of Streptomycetes Metabo-
in nonsterile soils the native microflora slowed up the growth lites from Solid Cultures. The concentrated extracts from
of the pathogen giving the opportunity to the biocontrol agent the agar inhibition zones of the selected Streptomyces strain
to express its antimicrobial activity more effectively [64]. ACTA1551 (Streptomyces rochei) could strongly suppress the
Of additional importance, is that through our experimental growth of F. oxysporum (Figure 4(a)). Additionally, the same
approach, the streptomycete can protect the plant just by antifungal activity was observed from the extracts of the solid
simple coating of the seeds with spore suspension prior culture of ACTA1551 without the presence of F. oxysporum.
to sowing. This inoculation method has been proved more This observation may suggest that the streptomycete does not
effective as the biocontrol agent can rapidly and extensively require the target fungus in its area to induce the production
cover the surface of the seeds [33, 55]. Early colonization by a of the antifungal substances. The antifungal activity is fungus
biocontrol agent often is required to fill the critical niches and independent and the antifungal components may belong to
to effectively compete against pathogenic fungi [65]. Thus, the main metabolism of Streptomyces rochei ACTA1551.
seed coating with bacterial and fungal biocontrol agents often Following fractionation of the bioactive extracts in a gel
is utilized or required to control aggressive, rapidly growing filtration column, the antifungal activity was observed in both
soil-borne pathogens [65, 66]. Additionally, this procedure the low molecular weight and the high molecular weight
is much easier to implement and more applicable for large fractions (Figure 4(a)). These results provide a combined
scale cultivation, compared to the classic one which includes action of different antifungal compounds. Each of them
enrichment of the soil with the biocontrol agents which is can independently suppress the growth of the pathogen
both time consuming and difficult to apply in real farming in vitro, but their combination (crude culture extract) can
conditions [48, 67]. However, experiments of a larger scale enhance the antifungal action. Although, further biochemical
under greenhouse conditions or in the field should support analysis, using High Performance Liquid Chromatography
this potential in a more applicable way. The undeniable fact fractionation coupled with Mass Spectrometry and Nuclear
is that ACTA1551 strain meets the conditions of a promising Magnetic Resonance techniques will allow us to refine its
biocontrol agent. structure.
Additionally, the same results give some information Moreover, examining the antimicrobial activity of the
about the mechanism of action of the selected streptomycete bioactive extractions against common microbial indicators
in soil. As it was shown, ACTA1551 could promote the growth (Escherichia coli, Bacillus subtilis, Pseudomonas fluorescens,
of the tomato plants in statistically significant levels, when it and Saccharomyces cerevisiae) it was shown that the agar
was the only microorganism in the soil, leading the tomato extract from S. rochei (ACTA1551) was able to inhibit growth
plants to overcome the growth levels they reached with no of all the examined indicator microorganisms (Figure 4(b)).
pathogen in their environment. Therefore, the possibility of Therefore the antifungal activity of the selected streptomycete
acting as a growth promoter could raise further examination. has a wider antimicrobial activity.
(a i) (a ii) (a iii)
(a iv) (a v) (b)
Figure 4: (a) Antifungal activity of the concentrated medium extracts derived from the solid culture of Streptomyces rochei ACTA1551
with or without F. oxysporum. Inhibition zone caused from (i) Streptomyces rochei ACTA1551 culture extract when F. oxysporum is present,
(ii) Streptomyces rochei ACTA1551 culture extract when F. oxysporum is absent, (iii) Streptomyces rochei ACTA1551 low molecular weight
fractions, (iv) Streptomyces rochei ACTA1551 high molecular weight fractions and negative control (no streptomycete extract added) (v). (b)
Antifungal activity of the concentrated medium extracts derived from the solid culture of Streptomyces rochei ACTA1551 against E. coli.
4. Conclusion [2] N. Ozbay and S. E. Newman, “Fusarium crown and root rot of
tomato and control methods,” Plant Pathology Journal, vol. 3,
This study focused on screening the special and biotechno- pp. 9–18, 2004.
logically interesting Greek Streptomyces microflora in order [3] G. E. Massee, The British Fungus-Flora, a Classified Text Book of
to detect strains able to control the phytopathogenic fungi Mycology, George Bell & Sons, London, UK, 1985.
Fusarium oxysporum. The best antifungal producer, Strep- [4] D. Brayford, “Fusarium oxysporum f. sp. radicis-lycopersici, IMI
tomyces rochei ACTA1551 from the rhizosphere soil of the descriptions of fungi and bacteria no. 1270,” Mycopathologia,
plant Pinus brutia (indigenous plant of Crete), proved to be vol. 133, pp. 61–63, 1996.
capable of protecting tomato plants from Fusarium wilting [5] W. R. Jarvis, “Fusarium crown and root rot of tomatoes,” Phyto-
under laboratory conditions while evidence that it promoted protection, vol. 69, pp. 49–64, 1988.
the plants growth was derived. The same streptomycete strain [6] C. Borrero, J. Ordovàsa, M. I. Trillasb, and M. Avilés, “Tomato
showed to be a promising producer of metabolites of a wide Fusarium wilt suppressiveness. The relationship between the
range of antimicrobial activity. Therefore, the present study organic plant growth media and their microbial communities
demonstrated that Streptomyces rochei ACTA1551 is a new as characterised by biology,” Soil Biology & Biochemistry, vol.
potential biocontrol agent at least against the pathogenic 38, pp. 1631–1637, 2006.
fungus Fusarium oxysporum. [7] W. H. Elmer, D. A. Johnson, and G. I. Mink, “Epidemiology and
management of the diseases causal to asparagus decline,” Plant
Disease, vol. 80, pp. 117–125, 1996.
References [8] Y. Uesugi, “Fungicide classes: chemistry, uses and mode of
action,” in Fungicidal Activity: Chemical and Biological Appro-
[1] C. H. Beckman, The Nature of Wilt Diseases of Plants, APS Press, aches To Plant Protection, D. Hutson and J. Miyamoto, Eds., pp.
St. Paul, Minn, USA, 1987. 23–56, John Wiley & Sons, New York, NY, USA, 1998.
[9] B. C. Baldwin and W. G. Rathmell, “Evolution of concepts for [24] A. Heydari and M. Pessarakli, “A review on biological control of
chemical control of plant disease,” Annual Review in Phytopa- fungal plant pathogens using microbial antagonists,” Journal of
thology, vol. 26, pp. 265–283, 1988. Biological Sciences, vol. 10, no. 4, pp. 273–290, 2010.
[10] G. N. Agrios, Plant Pathology, Academic Press, New York, NY, [25] Y. J. Zhang, P. S. Fan, X. Zhang, C. J. Chen, and M. G. Zhou,
USA, 3rd edition, 1988. “Quantification of Fusarium graminearum in harvested grain
[11] R. J. Cook, “Making greater use of microorgainsms for biocon- by real-time polymerase chain reaction to assess efficacies of
trol of plant pathogens,” Annual Reviews in Phytopathology, vol. fungicides on fusarium head blight, deoxynivalenol contami-
31, pp. 53–80, 1993. nation, and yield of winter wheat,” Phytopathology, vol. 99, no.
[12] A. Heydari, “Biological control of turfgrass fungal diseases,” in 1, pp. 95–100, 2009.
Turfgrass Management and Physiology, M. Pessarakli, Ed., CRC [26] V. K. Chebotar’, N. M. Makarova, A. I. Shaposhnikov, and L. V.
Press, Boca Raton, Fla, USA, 2007. Kravchenko, “Antifungal and phytostimulating characteristics
[13] A. Heydari, I. J. Misaghi, and G. M. Balestra, “Pre-emergence of Bacillus subtilis Ch-13 rhizospheric strain, producer of bio-
herbicides influence the efficacy of fungicides in controlling prepations,” Applied Biochemistry and Microbiology, vol. 45, no.
cotton seedling damping-off in the field,” International Journal 4, pp. 419–423, 2009.
of Agricultural Research, vol. 2, no. 12, pp. 1049–1053, 2007. [27] V. Nihorimbere, M. Ongena, H. Cawoy et al., “Beneficial effects
[14] M. Hernandez-Suarez, F. D. Hernandez-Castillo, G. Gallegos- of Bacillus subtilis on field-grown tomato in Burundi: reduction
Morales, R. H. Lira-Saldivar, and R. Rodriguez-Herrera, “Bio- of local Fusarium disease and growth promotion,” African
control of soil fungi in tomato with microencapsulates con- Journal of Microbiology Research, vol. 4, no. 11, pp. 1135–1142,
taining Bacillus subtilis,” American Journal of Agricultural and 2010.
Biological Sciences, vol. 6, pp. 189–195, 2011. [28] C. Olivain, C. Humbert, J. Nahalkova, J. Fatehi, F. L’Haridon,
[15] K. Handreck and N. Black, Growing Media For Ornamental and C. Alabouvette, “Colonization of tomato root by pathogenic
Plants and Turf, University of New South Wales Press, Sydney, and nonpathogenic Fusarium oxysporum strains inoculated
Australia, 3rd edition, 2002. together and separately into the soil,” Applied and Environmen-
[16] C. Y. Shen, “Integrated management of Fusarium and Verticil-
lum wilts of cotton in China,” Crop Protection, vol. 4, no. 3, pp. [29] K. He, E. B. Rimm, A. Merchant et al., “Fish consumption and
337–345, 1985. risk of stroke in men,” Journal of the American Medical Associa-
tion, vol. 288, no. 24, pp. 3130–3136, 2002.
[17] A. Miguel, J. V. Maroto, A. San Bautista et al., “The grafting of
triploid watermelon is an advantageous alternative to soil fumi- [30] M. G. Bakkera, J. D. Gloverb, J. G. Maib, and L. L. Kinkela, “Plant
gation by methyl bromide for control of Fusarium wilt,” Scientia community effects on the diversity and pathogen suppressive
Horticulturae, vol. 103, no. 1, pp. 9–17, 2004. activity of soil streptomycetes,” Applied Soil Ecology, vol. 46, pp.
35–42, 2010.
[18] S. C. Hwang, “Recent development in Fusarium research &
development of banana in Taiwan,” in Banana Fusarium Wilt [31] S. Compant, B. Reiter, A. Sessitsch, J. Nowak, C. Clément,
Management: Towards Sustainable Cultivation: Proceedings of and E. A. Barka, “Endophytic colonization of Vitis vinifera L.
the International Workshop on Banana Fusarium Wilt Disease, by plant growth-promoting bacterium Burkholderia sp. strain
A. B. Molina, N. H. N. Masdek, and K. W. Liew, Eds., pp. 39–49, PsJN,” Applied and Environmental Microbiology, vol. 71, no. 4,
INIBAP, Montpellier, France, 1999. pp. 1685–1693, 2005.
[19] M. Ulloa, J. M. Stewart, E. A. Garcia-Castañeda, S. GodoyAvila, [32] A. Minuto, D. Spadaro, A. Garibaldi, and M. L. Gullino, “Con-
A. Gaytan-Mosqueta, and S. Acosta-Nuñez, “Cotton genetic trol of soilborne pathogens of tomato using a commercial
resources in the western states of Mexico: In situ conservation formulation of Streptomyces griseoviridis and solarization,” Crop
status and germplasm collection for ex situ preservation,” Protection, vol. 25, no. 5, pp. 468–475, 2006.
Genetic Resources and Crops Evolution, vol. 53, pp. 653–668, [33] G. S. Kanini, E. A. Katsifas, A. L. Savvides, D. G. Hatzinikolaou,
2006. and A. D. Karagouni, “Greek indigenous streptomycetes as
[20] D. M. Weller, “Pseudomonas biocontrol agents of soilborne biocontrol agents against the soil-borne fungal plant pathogen
pathogens: looking back over 30 years,” Phytopathology, vol. 97, Rhizoctonia solani,” Journal of Applied Microbiology, vol. 114, no.
no. 2, pp. 250–256, 2007. 5, pp. 1468–1479, 2013.
[21] L. C. Dekkers, I. H. M. Mulders, C. C. Phoelich, T. F. C. Chin- [34] H. Cawoy, W. Bettiol, P. Fickers, and M. Ongena, Pesticides in
A-Woeng, A. H. M. Wijfjes, and B. J. J. Lugtenberg, “The sss the Modern World—Pesticides Use and Management, InTech,
colonization gene of the tomato-Fusarium oxysporum f. sp. Rijeka, Croatia, 2011, Edited by: M. Stoytcheva.
radicislycopersici biocontrol strain Pseudomonas fluorescens [35] T. Taechowisan, N. Chuaychot, S. Chanaphat, A. Wanbanjob,
WCS365 can improve root colonization of other wild-type and P. Tantiwachwutikul, “Antagonistic effects of Streptomyces
Pseudomonas spp. bacteria,” Molecular Plant-Microbe Interac- sp. SRM1 on colletotrichum musae,” Biotechnology, vol. 8, no. 1,
tions, vol. 13, no. 11, pp. 1177–1183, 2000. pp. 86–92, 2009.
[22] Z. Zhang and L. S. Pierson, “A second quorum-sensing system [36] K. Chamberlain and D. L. Crawford, “In vitro and in vivo antag-
regulates cell surface properties but not phenazine antibiotic onism of pathogenic turfgrass fungi by Streptomyces hygro-
production in Pseudomonas aureofaciens,” Applied and Envi- scopicus strains YCED9 and WYE53,” Journal of Industrial
ronmental Microbiology, vol. 67, no. 9, pp. 4305–4315, 2001. Microbiology and Biotechnology, vol. 23, no. 1, pp. 641–646, 1999.
[23] A. Bolwerk, A. L. Lagopodi, A. H. M. Wijfjes et al., “Interactions [37] R. F. de Souza, R. R. R. Coelho, A. Macrae et al., “Streptomyces
in the tomato rhizosphere of two Pseudomonas biocontrol lunalinharesii sp. nov., a chitinolytic streptomycete isolated
strains with the phytopathogenic fungus Fusarium oxysporum from cerrado soil in Brazil,” International Journal of Systematic
f. sp. radicis-lycopersici,” Molecular Plant-Microbe Interactions, and Evolutionary Microbiology, vol. 58, no. 12, pp. 2774–2778,
vol. 16, no. 11, pp. 983–993, 2003. 2008.
[38] K. Getha and S. Vikineswary, “Antagonistic effects of Strepto- [52] K. Tamura, D. Peterson, N. Peterson, G. Stecher, M. Nei, and
myces violaceusniger strain G10 on Fusarium oxysporum f.sp. S. Kumar, “MEGA5: molecular evolutionary genetics analysis
cubense race 4: indirect evidence for the role of antibiosis in using maximum likelihood, evolutionary distance and maxi-
the antagonistic process,” Journal of Industrial Microbiology and mum parsimony methods,” Molecular Biology and Evolution,
Biotechnology, vol. 28, no. 6, pp. 303–310, 2002. vol. 28, no. 10, pp. 2731–2739, 2011.
[39] R. C. Gomes, L. T. A. S. Semêdo, R. M. A. Soares, C. S. Alviano, [53] R. M. Atlas, Handbook of Microbiological Media, CRC Press,
L. F. Linhares, and R. R. R. Coelho, “Chitinolytic activity of acti- Boca Raton, Fla, USA, 1993, Edited by: L. C. Parks.
nomycetes from a cerrado soil and their potential in biocontrol,” [54] E. M. H. Wellington, N. Cresswell, and V. A. Saunders, “Growth
Letters in Applied Microbiology, vol. 30, no. 2, pp. 146–150, 2000. and survival of streptomycete inoculants and extent of plasmid
[40] B. K. Hwang, S. W. Lim, B. S. Kim, J. Y. Lee, and S. S. Moon, transfer in sterile and nonsterile soil,” Applied and Environmen-
“Isolation and in vivo and in vitro antifungal activity of tal Microbiology, vol. 56, no. 5, pp. 1413–1419, 1990.
phenylacetic acid and sodium phenylacetate from Streptomyces [55] Z. Lu, R. Tombolini, S. Woo, S. Zeilinger, M. Lorito, and J. K.
humidus,” Applied and Environmental Microbiology, vol. 67, no. Jansson, “In vivo study of Trichoderma-pathogen-plant interac-
8, pp. 3739–3745, 2001. tions, using constitutive and inducible green fluorescent protein
[41] K. A. El-Tarabily, “Promotion of tomato (Lycopersicon esculen- reporter systems,” Applied and Environmental Microbiology, vol.
tum Mill.) plant growth by rhizosphere competent 1-amino- 70, no. 5, pp. 3073–3081, 2004.
cyclopropane-1-carboxylic acid deaminase-producing strepto- [56] S. Baur, J. Niehaus, A. D. Karagouni et al., “Fluostatins C-E,
mycete actinomycetes,” Plant and Soil, vol. 308, no. 1-2, pp. 161– novel members of the fluostatin family produced by Strepto-
174, 2008. myces strain acta 1383,” Journal of Antibiotics, vol. 59, no. 5, pp.
[42] Y. Igarashi, M. Ogawa, Y. Sato et al., “Fistupyrone, a novel inhibi- 293–297, 2006.
tor of the infection of Chinese cabbage by Alternaria brassici- [57] T. Paululat, E. A. Katsifas, A. D. Karagouni, and H. P. Fiedler,
cola, from Streptomyces sp. TP-A-0569,” Journal of Antibiotics, “Grecoketides A and B: new naphthoquinones from Strepto-
vol. 53, no. 10, pp. 1117–1122, 2000. myces sp. Acta 1362,” European Journal of Organic Chemistry,
[43] M. Ichinose, M. Saito, N. Fujii et al., “Modulation of the control vol. 2008, no. 31, pp. 5283–5288, 2008.
of muscle sympathetic nerve activity during incremental leg [58] T. Paululat, A. Kulik, H. Hausmann et al., “Grecocyclines: new
cycling,” Journal of Physiology, vol. 586, no. 11, pp. 2753–2766, angucyclines from Streptomyces sp. Acta 1362,” European Jour-
2008. nal of Organic Chemistry, no. 12, pp. 2344–2350, 2010.
[44] S. D. Schrey and M. T. Tarkka, “Friends and foes: streptomycetes [59] E. A. Katsifas, E. P. Giannoutsou, and A. D. Karagouni,
as modulators of plant disease and symbiosis,” Antonie van “Diversity of streptomycetes among specific Greek terrestrial
Leeuwenhoek, vol. 94, no. 1, pp. 11–19, 2008. ecosystems,” Letters in Applied Microbiology, vol. 29, no. 1, pp.
[45] G. Godoy-Lutz, S. Kuninaga, J. R. Steadman, and K. Powers, 48–51, 1999.
“Phylogenetic analysis of Rhizoctonia solani subgroups associ- [60] E. A. Katsifas, T. G. Koraki, and A. D. Karagouni, “Determina-
ated with web blight symptoms on common bean based on ITS- tion of metabolic activity of streptomycetes in soil microcosms,”
5.8S rDNA,” Journal of General Plant Pathology, vol. 74, no. 1, pp. Journal of Applied Microbiology, vol. 89, no. 1, pp. 178–184, 2000.
32–40, 2008. [61] G. S. Reddi and A. S. Rao, “Antagonism of soil actinomycetes
[46] E. C. Neeno-Eckwall, L. L. Kinkel, and J. L. Schottel, “Compe- to some soil-borne plant pathogenic fungi,” Indian Phytopathol-
tition and antibiosis in the biological control of potato scab,” ogy, vol. 24, pp. 649–657, 1971.
Canadian Journal of Microbiology, vol. 47, no. 4, pp. 332–340, [62] B. Mahadevan and D. L. Crawford, “Properties of the chiti-
2001. nase of the antifungal biocontrol agent Streptomyces lidicus
[47] N. Shekhar, D. Bhattacharya, D. Kumar, and R. K. Gupta, “Bio- WYEC108,” Enzyme Microbiology and Technology, vol. 20, pp.
control of wood-rotting fungi with Streptomyces violaceusniger 293–489, 1997.
XL-2,” Canadian Journal of Microbiology, vol. 52, no. 9, pp. 805– [63] A. A. Farrag, “Efficiency of different biocontol agents on both
808, 2006. susceptible and resistant bean plants and their protein pattern
[48] W. M. Yuan and D. L. Crawford, “Characterization of Strep- consequences,” Journal of American Science, vol. 7, no. 4, pp. 7–
tomyces lydicus WYEC108 as a potential biocontrol agent 14, 2011.
against fungal root and seed rots,” Applied and Environmental [64] L. W. Burgess, C. M. Liddell, and B. A. Summerell, Laboratory
Microbiology, vol. 61, no. 8, pp. 3119–3128, 1995. Manual For Fusarium Research, University of Sydney, Sydney,
[49] P. R. Herron and E. M. H. Wellington, “New method for extrac- Australia, 2nd edition, 1988.
tion of streptomycete spores from soil and application to the [65] D. A. Mitchell, “Growth patterns, growth kinetics and the
study of lysogeny in sterile amended and nonsterile soil,” modeling of growth in solid-state cultivation,” in Solid Substrate
Applied and Environmental Microbiology, vol. 56, no. 5, pp. Cultivation, H. W. Doelle, D. A. Mitchell, and C. E. Rolz, Eds.,
1406–1412, 1990. pp. 87–114, Elsevier Applied Science, New York, NY, USA, 1992.
[50] U. Edwars, T. Rogall, H. Blöcker, M. Emde, and E. C. Böttger, [66] Y. Nerey, J. Pannecoucque, H. P. Hernandez et al., “Rhizoctonia
“Isolation and direct complete nucleotide determination of spp. causing root and hypocotyl rot in Phaseolus vulgaris in
entire genes. Characterization of a gene coding for 16S riboso- Cuba,” Journal of Phytopathology, vol. 158, no. 4, pp. 236–243,
mal RNA,” Nucleic Acids Research, vol. 17, no. 19, pp. 7843–7853, 2010.
1989. [67] J. M. Whipps, “Microbial interactions and biocontrol in the
[51] D. J. Lane, “16S/23S rRNA sequencing,” in Nucleic Acid Tech- rhizosphere,” Journal of Experimental Botany, vol. 52, pp. 487–
niques in Bacterial Systematic, E. Stackebrandt and M. Goodfel- 511, 2001.
low, Eds., pp. 115–147, John Wiley & Sons, New York, NY, USA,
1991.
http://dx.doi.org/10.1155/2013/260395
Research Article
A Phylogenetic Analysis of Greek Isolates of
Aspergillus Species Based on Morphology and Nuclear and
Mitochondrial Gene Sequences
Antonios Krimitzas,1 Ioanna Pyrri,2 Vassili N. Kouvelis,1

Evangelia Kapsanaki-Gotsi,2 and Milton A. Typas1
1
Department of Genetics and Biotechnology, Faculty of Biology, National and Kapodistrian University of Athens,
Panepistemiopolis, 15701 Athens, Greece
2
Department of Ecology and Systematics, Faculty of Biology, National and Kapodistrian University of Athens,
Panepistemiopolis, 15784 Athens, Greece
Correspondence should be addressed to Milton A. Typas; matypas@biol.uoa.gr
Received 13 March 2013; Accepted 9 April 2013
Copyright © 2013 Antonios Krimitzas et al. This is an open access article distributed under the Creative Commons Attribution
cited.
Aspergillus species originating from Greece were examined by morphological and molecular criteria to explore the diversity of
this genus. The phylogenetic relationships of these species were determined using sequences from the ITS and IGS region of
the nuclear rRNA gene complex, two nuclear genes (𝛽-tubulin (benA) and RNA polymerase II second largest subunit (rpb2))
and two mitochondrial genes (small rRNA subunit (rns) and cytochrome oxidase subunit I (cox1)) and, where available, related
sequences from databases. The morphological characters of the anamorphs and teleomorphs, and the single gene phylogenetic
trees, differentiated and placed the species examined in the well-supported sections of Aenei, Aspergillus, Bispori, Candidi,
Circumdati, Clavati, Cremei, Flavi, Flavipedes, Fumigati, Nidulantes, Nigri, Restricti, Terrei, Usti, and Zonati, with few uncertainties.
The combined use of the three commonly employed nuclear genes (benA, rpb2, and ITS), the IGS region, and two less often
used mitochondrial gene sequences (rns and cox1) as a single unit resolved several taxonomic ambiguities. A phylogenetic tree
was inferred using Neighbour-Joining, Maximum Parsimony, and Bayesian methods. The strains examined formed seven well-
supported clades within the genus Aspergillus. Altogether, the concatenated nuclear and mitochondrial sequences offer additional
tools for an improved understanding of phylogenetic relationships within this genus.
1. Introduction worldwide distribution necessitates the detailed taxonomic

and phylogenetic study of Aspergillus species in order to
The genus Aspergillus is one of the most common and provide more appropriate tools for their accurate and fast
important genera of microfungi. Several species of the genus identification.
proved very useful cell factories for biotechnology purposes The first most comprehensive taxonomic treatise of the
since they produce organic acids, extracellular enzymes, genus was based on morphological criteria and included
nutraceuticals (isoprenoids), and aroma compounds and 150 taxa that were separated in “groups” [7]. These clusters
are capable of fermenting various substrates under extreme of species (“groups”) were later revised to 18 “sections” in
cultivation conditions, usually intolerable to bacteria and 6 subgenera, in order to receive nomenclatural status [8].
yeasts [1–4]. Furthermore, the genus contains a number of In followup, a list of 182 species valid names was compiled
species that produce toxic metabolites, act as opportunistic [9], and expanded to ∼250 species [3], and it is expected to
pathogens to humans, or decompose a wide variety of increase further as new species are discovered and speciation
substrates [5, 6]. Their high diversity combined with the concepts are refined [10].
During the last decade, a revision of the genus Aspergillus The strains of Aspergillus were three-point inoculated
is in progress, as a result of molecular analyses which have in selective nutrient media and incubated on Czapek Yeast
offered a better insight into taxonomic and phylogenetic rela- Autolysate Agar (CYA) at 25∘ C and 37∘ C, on Malt Extract
tions [10–13]. The revision is based on polyphasic approaches Agar (MEA) at 25∘ C, and on G25N at 25∘ C [29]. The
that include molecular data, as well as morphological, phys- macromorphology of the colonies on each of the media, as
iological and ecological characteristics. Although no single well as the micromorphology on MEA, were studied after 7
method worked flawlessly in recognising species, molecular days of incubation. Microscopic examination was performed
data have largely supported previously inferred relationships by teasing apart the sample in a drop of 70% ethanol on a glass
that were based on the other characters [3]. Single locus DNA slide, a coverslip was placed on it, and it was observed under
sequence studies were a common practice in fungi in the past, a Zeiss microscope in plain light or Differential Interference
thus, large amounts of information have been accumulated Contrast. The photographs were taken with an Axiocam
in databases for several parts of the nuclear ribosomal rRNA digital camera (Zeiss, Germany). When the evaluation of the
repeat (in particular the ITS and 28S) [14–16] and genes like morphological data was completed, all strains were further
the 𝛽-tubulin [17]. The molecular identification of Aspergillus examined with molecular markers and the data was used
species is currently based on sequences from genes 𝛽-tubulin, in phylogenetic analyses. Additional taxa were included to
calmodulin, actin, and ITS [5, 12, 13, 18–22]. A phylogenetic represent most of the sections recognized in the genus
study of Aspergillus species, based on four single nuclear Aspergillus.
genes that were concatenated and used as one unit, attempted Mycelium preparation and fungal DNA extraction pro-
to resolve the existing taxonomic ambiguities and has resulted cedures used are previously described [30]. PCR primers
in a reevaluation of the genus Aspergillus [23, 24]. used in the amplification reactions for the ITS1-5.8S-ITS2
In other fungal genera, like Lecanicillium, Verticillium, region were 18S-ITS1 and 28S-ITS2 [27] and for the IGS
and Beauveria, the use of mitochondrial (mt) genes and region were CNL12 and CNS1 [31]. For the nuclear genes
mt intergenic regions proved an extremely useful tool to and specifically for the RNA polymerase II second largest
reveal species differences within a genus and even helped to subunit gene (rpb2), the primers used were RPB2-6F and
resolve taxonomic ambiguities (e.g., the small rRNA subunit RPB2-7R [32] and for the 𝛽-tubulin gene (benA) the primers
(rns), the NADH dehydrogenase subunits 1 (nad1) and 3 were bt2a and bt2b [33]. Primers for the amplification of mt
(nad3), the mt intergenic domains nad3–atp9 and the atp6- genes were NMS1 and NMS2 for rns [34] and cox1F and cox1R
rns) [25–27]. Similarly, other nuclear regions/genes like the for cox1 [35]. The PCR amplicons were sequenced using a
RNA polymerase II largest subunit gene (rpb1) and the IGS Thermo Sequenase Primer Cycle Sequencing kit (Amersham
region of the rRNA repeat also proved to be excellent tools Biosciences, Amersham, UK), with a LICOR 4200 IR2 (LI-
for discriminating species within Metarhizium, Verticillium COR, Lincoln NE, USA) automated sequencer. All fragments
and Aspergillus [23, 27, 28]. were read in both directions and nucleotide sequences were
In the present study the efficiency of combined nuclear submitted to GenBank database (Table 1).
and mitochondrial gene sequence data was evaluated for Maximum parsimony (MP), Neighbour-Joining (NJ),
the identification of species that are distributed in most of and Bayesian inference (BI) analyses were employed in order
Aspergillus sections. In detail, the commonly used (ITS1-5.8S- to create the phylogenetic trees. MP, and NJ analyses of
ITS2, RNA polymerase II second largest subunit gene (rpb2) nucleotide datasets were performed with PAUP∗ 4.0b10 [36]
and b-tubulin (benA)) and the less often exploited (IGS) using 1,000 and 10,000 replicates, respectively, with random
nuclear regions were assessed along with two mitochondrial addition of taxa and tree-bisection reconnection branch
genes, cytochrome oxidase subunit I (cox1) and small rRNA swapping. Reliability of nodes was assessed using 1,000 and
subunit (rns), for the identification of Aspergillus species and 100 bootstrap iterations (for NJ and MP analyses, resp.)
the evaluation of their phylogenetic affinities. The objective and for the NJ analyses the Kimura-2 parameter model was
of this work is to test hypotheses based on morphological employed. The BI was performed with MrBayes v3.1 [37].
criteria and data obtained from a multilocus DNA sequence A gamma distribution model of site variation was used,
analysis on the phylogeny within the genus Aspergillus. calculated with PAML [38]. For datasets ITS1-5.8S-ITS2,
IGS, benA, rpb2, rns, cox1 and the concatenated data, alpha
(𝛼) was 0.470, 0.630, 0.540, 0.400, 0.348, 0.527, and 0.403,
2. Materials and Methods respectively. Random starting trees were used and the burn-
in period was 1,000,000 cycles, as this was found to be clearly
Several strains of Aspergillus have been isolated from the sufficient for the likelihood and the model parameters to
indoor air of a food industry in Greece, as well as from reach equilibrium. After the burn-in, 1,000 trees were sam-
food samples. Additional strains of representative Aspergillus pled every 100 cycles during the sampling period (1,000,000
spp. were studied from the ATHUM Culture Collection cycles). Two independent MCMCMC searches were run for
of Fungi in the University of Athens and ex-type cultures each dataset using different random starting points. The
were obtained from the Agricultural Research Service ARS hypocrealean fungus Metarhizium anisopliae from the Class
(NRRL) at Peoria, USA (kindly provided by Dr. S. Peterson). Sordariomycetes, a close relative to Eurotiomycetes, was used
A total of thirty six strains belonging to thirty-one Aspergillus as outgroup in all datasets. When the species Coccidioides
species were studied (Table 1). All strains are maintained in immitis from Onygenales, an order within Eurotiomycetes,
the ATHUM Culture Collection of Fungi. was alternatively employed, the tree topologies from all
Table 1: The strains of Aspergillus spp used in this study.
GenBank accession number
Species Accession number Substrate Origin
ITS IGS benA rpb2 cox1 rns
A. aculeatus ATHUM 5028a Indoor air Greece EU982028 EU982062 EU982087 NT EU982142 EU982176
𝐴. 𝑎𝑒𝑛𝑒𝑢𝑠T NRRL 4769b Forest soil Somalia EF652474 EU982060 EF652298 EF652210 EU982140 EU982174
𝐴. 𝑎𝑟𝑒𝑛𝑎𝑟𝑖𝑢𝑠T NRRL 5012 Soil India EU021615 EU982048 EU021674 EU021653 NT NT
A. awamori FRR 4795 = ATHUM 5181 Unrecorded source Unknown EU982010 EU982035 EU982069 EU982094 EU982117 EU982150
𝐴. 𝑏𝑖𝑠𝑝𝑜𝑟𝑢𝑠T NRRL 3693 Soil stored 10 years at 30∘ C USA EF661208 EU982052 EF661121 EF661077 EU982133 EU982166
A. brunneouniseriatus NRRL 4273 soil under Dalbergia sissoo USA EF652141 EU982054 EF652123 EF652089 EU982135 EU982168
𝐴. 𝑐𝑎𝑚𝑝𝑒𝑠𝑡𝑟𝑖𝑠T NRRL 13001 Native mixed prairie soil USA EU014091 EU982049 EF669577 EF669619 EU982130 EU982163
𝐴. 𝑐𝑙𝑎V𝑎𝑡𝑜𝑓𝑙𝑎V𝑢𝑠T NRRL 5113 Forest soil Australia EF669713 EU982058 EF669686 EF669668 NT EU982172
ATHUM 5032 Dairy product Greece EU982014 EU982039 EU982073 EU982098 EU982121 EU982154
A. clavatus
𝐴. 𝑒𝑙𝑜𝑛𝑔𝑎𝑡𝑢𝑠T NRRL 5176 Alkaline soil India EF652502 EU982059 EF652326 EF652238 EU982139 EU982173
ATHUM 5015 Indoor air Greece EU982011 EU982036 EU982070 EU982095 EU982118 EU982151
A. flavus
A. fumigatus ATHUM 5013 Indoor air Greece EU982013 EU982038 EU982072 EU982097 EU982120 EU982153
𝐴. 𝑔𝑖𝑔𝑎𝑛𝑡𝑒𝑢𝑠T NRRL 10 Bat dung Mexico EF669928 NT EF669789 EF669716 EU982146 EU982181
𝐴. 𝑗𝑎𝑛𝑢𝑠T NRRL 1787 Soil Panama EU021598 EU982061 EU014076 EF669620 EU982141 EU982175
ATHUM 2539 Outdoor air Greece EU982009 EU982034 EU982068 EU982092 EU982116 EU982149
A. niger
A. niveus ATHUM 5029 Indoor air Greece EU982023 EU982050 EU982082 EU982105 EU982131 EU982164
A. ochraceus ATHUM 5014 Indoor air Greece EU982029 EU982063 EU982088 EU982111 EU982143 EU982177
A. oryzae ATHUM 4958 Outdoor air Greece EU982022 EU982047 EU982081 EU982104 EU982129 EU982162
ATHUM 5037 Dairy product Greece EU982020 EU982045 EU982079 NT EU982127 EU982160
A. parasiticus
ATHUM 5038 Dairy product Greece EU982021 EU982046 EU982080 NT EU982128 EU982161
A. puniceus ATHUM 5434 Dairy product Greece EU982019 EU982044 EU982078 EU982101 EU982126 EU982159
𝐴. 𝑟𝑒𝑠𝑡𝑟𝑖𝑐𝑡𝑢𝑠T NRRL 154 Clinical isolate from cloth England EF652042 EU982053 EF651880 EF651978 EU982134 EU982167
𝐴. 𝑠𝑐𝑙𝑒𝑟𝑜𝑡𝑖𝑜𝑟𝑢𝑚T NRRL 415 Food, fruit (apple) USA EF661400 EU982064 EF661337 EF661287 NT EU982178
A. sydowii ATHUM 5093 Dairy product Greece EU982025 EU982055 NT NT EU982136 EU982169
A. terreus ATHUM 4761 Outdoor air Greece EU982024 EU982051 EU982083 EU982106 EU982132 EU982165
ATHUM 5097 Indoor air Greece EU982026 EU982056 EU982085 EU982108 EU982137 EU982170
A. ustus
A. versicolor ATHUM 2541 Culture contaminant Greece EU982032 NT EU982091 EU982114 EU982147 EU982182
Emericella nidulans ATHUM 5022 Indoor air Greece EU982031 EU982066 EU982090 EU982113 EU982145 EU982180
E. variecolor ATHUM 5019 Indoor air Greece EU982030 EU982065 EU982089 EU982112 EU982144 EU982179
Eurotium amstelodami ATHUM 5082 Indoor air Greece EU982017 EU982042 EU982076 NT EU982124 EU982157
E. rubrum FRR 0326 = ATHUM 5183 Stored pack of prunes Australia EU982018 EU982043 EU982077 NT EU982125 EU982158
Neosartorya fischeri ATHUM 5030 Dairy product Greece EU982016 EU982041 EU982075 EU982100 EU982123 EU982156
Metarhizium anisopliae AF218207 AF218207 AY995134 DQ522453 AY884128 AY884128
a
ATHUM: Culture Collection of Fungi, National and Kapodistrian University of Athens.
b
NRRL: National Center For Agricultural Utilization Research.
NT: nontested.
Type strains designated by a superscript T.
3
Table 2: Phenotypic characters of Aspergillus species originating from Greece.
Conidia
Species ATHUM number Vesicle shape Vesicles (𝜇m) Metulae (𝜇m) Phialides (𝜇m)
Size (𝜇m) Ornamentation
Aspergillus aculeatus 5028 Globose 36–50 — 6–10 × 3.5–5 4-5 Finely rough
Aspergillus clavatus 5032, 5036 Clavate 15–150 × 15–55 — 7–10 × 2.5–4 4-5(-8) × 2.8–4 Smooth
Aspergillus flavus 5015 Globose 30–55 9–13 × 4-5 9–14 × 2-3 3–6 Smooth
Aspergillus fumigatus 5013 Spathulate (13-)17–22 — 6–9 × 2-3 2.2–3.5 Minutely echinulate
Aspergillus niger 2539, 5044 Globose 40–65 14–22 × 4–6 7–9 × 3-4 4–5.5 Rough
Aspergillus niveus 5029 Spathulate 9–16 6-7 × 2-3 6-7 × 2-3 2-3 Smooth
Aspergillus ochraceus 5014 Globose 22–40 10–15 × 4-5 10-11 × 2-3 2-3 Smooth
Aspergillus oryzae 4958 Subglobose 25–45 9–12 × 4-5 9–12 × 4–6 5–8 Rough
Aspergillus parasiticus 5037, 5038 Globose 25–40 — 7–11 × 3-4 3.5–6 Rough
Aspergillus sydowii 5093 Pyriform 8–17 4–6(-7) × 3-4 6–8 × 2-3 3-4 Echinulate
Aspergillus terreus 4761 Globose 10–17 5-6 × 2-3 7-8 × 1.5–2 2–2.5 Smooth
Aspergillus ustus 5097, 5103 Pyriform 11–17 5–7 × 3.5–4 4–7 × 2.5–3.5 3-4 Rough
Aspergillus versicolor 2541 Pyriform 10–15 × 3–5 5–7 × 3–5 5–8 × 3–5 3-4 Rough
Emericella nidulans 5022 Spathulate 10–14 7-8 × 3-4 6-7 × 2-3 3-4 Rough
Emericella variecolor 5019 Spathulate 16–19 6–9 × 2-3 6–9 × 2-3 2-3 Rough
Eurotium amstelodami 5082 Subglobose 15–35 — 6–9 × 3-4 4–6 Echinulate
Neosartorya fischeri 5030 Spathulate 12–16 — 6-7 × 2–2.5 2-3 Smooth
analyses were identical (data not shown). All sequences of and 805–847 for cox1), reflecting to the apparent genetic
Aspergillus species and their teleomorphs not amplified in variability between different Aspergillus species. Amplicons
this study were taken from GenBank and are provided with were sequenced in both directions, sequences were combined
their Accession Numbers and the species names provided and compared with all known related sequences in the data-
by the database. Aligned matrices are available in Treebase, banks, and appropriate phylogenetic trees were constructed.
http://purl.org/phylo/treebase/phylows/study/12165. Phylogenetic trees deduced from NJ analyses were drawn
(Figures 2, 3, 4, 5, and 6), and parsimony and Bayesian
methods were used to examine the sensitivity of the resulting
3. Results trees and tree topologies. Trees remained largely invariant
to these manipulations, and topologies were similar for
3.1. Morphological Analysis. The cultural characteristics of
each gene tested independently of the phylogenetic method
the Aspergillus spp. have been recorded for all the strains
applied.
and their evaluation was critical in cases of closely related
The phylogenetic analysis of ITS and benA sequences
species. The colony texture, colour and growth rate on the
included 227 and 257 taxa, respectively, belonging to 210
nutrient media, the formation of exudates, soluble pigments
and 235 species of Aspergillus and their teleomorphs (i.e.,
and sclerotia, and the thermotolerance were examined. The
Emericella, Eurotium, Fenellia, Neocarpenteles, Neosartorya,
most important diagnostic characters are the morphological
Neopetromyces, Petromyces, and Warcupiella), in order to
features of the conidial heads, including their growth pattern,
determine their phylogenetic positions (Figures 2 and 3). For
size and seriation, dimensions and surface texture of the
the datasets of IGS (Supplemental Figure S1 of Supplementary
conidiophore, the shape and size of the metulae, phialides
Material available online at http://dx.doi.org/10.1155/2013/
and conidia, and the colour and the wall ornamentation of
260395) and cox1 (Supplemental Figure S2) the sequences
the phialidospores (Figure 1). The range of variability found
analysed were significantly fewer than above, since there are
for the main morphological characteristics of the strains
no other entries in the databanks apart from those examined
of Aspergillus spp. originating from Greece is presented in
in our work (Table 1). Finally, the datasets of rpb2 (Figure 4)
Table 2. In addition, the morphology of the teleomorph
and rns (Figure 5) were enriched with the addition of 147 and
has been examined in detail when the holomorph was
10 more species, respectively, from databanks.
present. There is a great morphological variation in the type
The ITS dataset included 634 characters, 376 of which
of ascomata, from loosely interwoven hyphae to compact
were parsimony informative, resulting to 10 parsimonious
sclerotioid structures, as well as in the size, ornamentation,
trees (tree length: 2,606, consistency index (CI): 0.360 and
and sculpture of the ascospores.
retention index (RI): 0.824). Among the 790 polymorphic
sites of the benA sequences, 533 were phylogenetically infor-
3.2. DNA Analyses. Amplified parts of genes and regions mative. The topology of the NJ tree is the same as one of
had variable sizes (i.e., in bp, 478–532 for ITS, 650–850 for the 9,980 most parsimonious trees inferred by the PAUP
IGS, 411–587 for benA, 785–811 for rpb2, 536–555 for rns programme (length: 7,986 steps, CI: 0.183, and RI: 0.754). The
(a) (b) (c)
(d) (e) (f)
(g) (h) (i) (j)
(k) (l) (m) (n)
Figure 1: Anamorphic and teleomorphic characters of Aspergillus spp. (a) Aspergillus ochraceus ATHUM 5014. (b) Aspergillus flavus ATHUM
5033. (c) Aspergillus parasiticus ATHUM 5037. (d) Aspergillus fumigatus ATHUM 5013. (e) Aspergillus clavatus ATHUM 5036. (f) Aspergillus
giganteus NRRL 10. (g) Aspergillus janus NRRL 1787. (h) Aspergillus versicolor ATHUM 2541. (i) Aspergillus sydowii ATHUM 5093. ((j)–
(k)) Eurotium amstelodami ATHUM 5082. ((l)–(n)) Neosartorya fischeri ATHUM 5030. ((a)–(i), (k), (n)) Conidial heads with phialides and
conidia. (j) Part of cleistothecium with asci and ascospores. (l) Cleistothecia. (m) Ascospores. Bars = ((a)–(e), (g)–(n)) 10 𝜇m; (f) 20 𝜇m; (l)
50 𝜇m.
rpb2 and cox1 dataset included 1,140 and 908 characters, with CI: 0.354, and RI: 0.516). Finally, the rns dataset included 614
655 and 303 parsimony informative characters, and produced characters, with 130 parsimony informative characters (tree
20 and 1,185 equally parsimonious trees (tree length: 8,927 length: 512, CI: 0.534, and RI: 0.731).
and 1,185, CI: 0.152 and 0.594, and RI: 0.690 and 0.670), Thirty-six strains studied in this work belong to 31
respectively. The IGS dataset included 901 characters, with Aspergillus species (see Table 1). ITS sequences from these
735 parsimony informative characters (tree length: 5,829, strains, together with sequences from other species of the
(a)
(b)
Figure 2: Phylogenetic tree constructed from unambiguously aligned DNA sequences of the ITS domain as produced by NJ. Sequences
obtained during this study are presented in bold, whereas those retrieved from GenBank are shown in roman with their Accession Numbers.
Clade credibility using NJ calculated from 1,000 replicates (numbers in roman), parsimonial BS support calculated from 100 replicates
(numbers in italics) using PAUP, and PPs produced by 1,000,000 generations (numbers in bold) using MrBayes are shown.
(a)
(b)
Figure 3: Phylogenetic tree constructed from unambiguously aligned combined DNA sequences of the benA gene as produced by NJ.
Sequences obtained during this study are presented in bold, whereas those retrieved from GenBank are shown in roman with their Accession
Numbers. Clade credibility using NJ calculated from 1,000 replicates (numbers in roman), parsimonial BS support calculated from 100
replicates (numbers in italics) using PAUP, and PPs produced by 1,000,000 generations (numbers in bold) using MrBayes are shown.
(a)
(b)
Figure 4: Phylogenetic tree constructed from unambiguously aligned combined DNA sequences of the nuclear rpb2 gene as produced by
NJ. Sequences obtained during this study are presented in bold, whereas those retrieved from GenBank are shown in roman with their
Accession Numbers. Clade credibility using NJ calculated from 1000 replicates (numbers in roman), parsimonial BS support calculated from
100 replicates (numbers in italics) using PAUP, and PPs produced by 1,000,000 generations (numbers in bold) using MrBayes are shown.
genus that were retrieved from the GenBank (altogether ranged from 51–100%. Members of Candidi, Cervini, and
around 200), clustered well within their corresponding sec- Circumdati clustered with 100% and likewise Raperi, and
tions of genus Aspergillus (Figure 2). The NJ, MP bootstrap, Restricti showed higher than 90% support for all three meth-
and Posterior Probabilities (PPs) support of all sections ods. Species of sections Aspergillus (Eurotium xerophilum and
Figure 5: Phylogenetic tree constructed from unambiguously aligned combined DNA sequences of the mitochondrial rns gene as produced
by NJ. Sequences obtained during this study are presented in bold, whereas those retrieved from GenBank are shown in roman with their
Accession Numbers. Clade credibility using NJ calculated from 1,000 replicates (numbers in roman), parsimonial BS support calculated from
100 replicates (numbers in italics) using PAUP, and PPs produced by 1,000,000 generations (numbers in bold) using MrBayes are shown.
Eurotium halophilicum are excluded) and Flavi (A. alliaceus but sections Bispori, Ochraceorosei, and Sparsi are scattered.
and A. leporis are excluded) presented higher than 90% sup- Finally, strains from different sections, that is, A. bisporus
port for at least two different approaches of bootstrap anal- NRRL 3693 (Bispori), A. brunneouniseriatus NRRL 4273 (Cre-
yses, whereas sections Clavati, Cremei, Fumigati, and Sparsi mei), A. aeneus NRRL 4769 (Nidulantes) and A. paradoxus
clustered with excellent support (>90%) for one methodology AY373860 (Clavati) formed a mixed cluster (Figure 2). This
and at least good (51–89%) for the other two (Figure 2). cluster grouped as a sister clade to a group consisting of A.
Species belonging to sections Nigri, Nidulantes, and Usti crystallinus and A. malodoratus (see Figure 2) with excellent
were divided into two well discriminated, well supported NJ, MP and PP (100%) support.
groups in each case (Figure 2). Subgenus Nidulantes based All benA amplicons contained three introns with the
on this dataset seems to be polyphyletic; that is, sections exception of Emericella variecolor and A. versicolor which
Nidulantes, Raperi, Silvati, and Usti form a single cluster, lack the last intron. As shown in Figure 3, the single tree
Figure 6: Phylogenetic tree constructed from unambiguously aligned concatenated DNA sequences of the nuclear ITS, IGS, benA, and rpb2
genes and the mitochondrial cox1 and rns genes as produced by NJ. Sequences obtained during this study are presented in bold, whereas
those retrieved from GenBank are shown in roman with their Accession Numbers. Clade credibility using NJ calculated from 1,000 replicates
(numbers in roman), parsimonial BS support calculated from 100 replicates (numbers in italics) using PAUP, and PPs produced by 1,000,000
generations (numbers in bold) using MrBayes are shown. Solid lines with numbers in circles represent the seven clusters, while the dotted
lines show the sections of the species examined and their teleomorphs.
based on benA nucleotide sequences discriminated clearly A. bahamensis, A. homomorphus, and A. uvarum; Figure 3).
species that belong to sections Candidi, Cervini, Circumdati, Species within sections Aspergillus, Cremei, Nidulantes, Peni-
Clavati, Flavi, Fumigati, Nigri, Raperi and Terrei. Topology cillium, Usti, and Restricti although they showed good cluster-
support was extremely good for these sections with at least ing, they failed to produce high bootstrap support due to the
one methodology showing 100% bootstrap value in all cases odd discrepancies and must therefore be approached more
(except section Sparsi: 98, 92, and 62% for NJ, MP bootstrap cautiously.
and PP, resp.). As in the case of the ITS dataset section In accordance with ITS and benA, analyses of the rpb2
Nigri was divided into two groups with the second group dataset provided similar or better phylogeny (Figure 4) for
containing five members (i.e., A. aculeatus, A. aculeatinus, species of sections Aspergillus, Clavati, Cervini, Circumdati,
Fumigati, Penicillium, Raperi, and Restricti grouping with be related to the developmental patterns of the teleomorphs
excellent bootstrap support (93–100%, irrelevant to the boot- (Figure 6). In the subgenus Circumdati, clade 1 includes mem-
strap methodology applied). Here again, section Nigri species bers of section Circumdati with connection to Neopetromyces
were placed into two distinct groups with bootstrap support [40], a genus with light-coloured multilocular ascostromata.
of 100, 100, 99% and 77, 56, 98% (for the NJ and MP bootstrap It is distantly related to section Flavi (clade 6), with a recently
and PP of the BI analyses of groups 1 and 2, resp.), but the two confirmed connection to Petromyces [41, 42], characterized
groups were sister clades, supported with 69, 56, and 89% NJ by black multilocular ascostromata and section Nigri (clade
and MP bootstrap and PP values, with the exception of A. 7) with unknown teleomorphs, even though sclerotia are
japonicus, which was placed basally to rest of the Nigri species. formed that should be considered as ascomatal primordia
The phylogenies found after the analyses of IGS (Supple- [43]. In the subgenus Nidulantes, the sections Nidulantes,
mental Figure S1) and cox1 (Supplemental Figure S2) did not and Usti in clade 2 cluster with Emericella teleomorphs
improve the picture drawn from ITS, benA, and rpb2 trees but that produce ascomata surrounded by Hülle cells. Addi-
instead added some ambiguities which could be attributed tionally, few species belonging to subgenus Nidulantes are
mainly to the limited number of species examined since this in clade 3 which forms a cluster intermixed with subgenus
may inhibit the good resolution of the species phylogenetic Aspergillus sections Aspergillus and Restricti that contain
relationships. Nevertheless, phylogenies from both IGS and Eurotium teleomorphs with thin layered ascomata loosely
cox1 can be used for better distinguishing species within the suspended in the mycelium. In the subgenus Fumigati, in
same section (e.g., A. niger with A. awamori), as also found clade 5, the section Fumigati is connected to Neosartorya
in a recent paper with other genes by Perrone et al. [39], with ascomata composed of abundant sterile hyphae and the
but are not suitable to discriminate the sections themselves. section Clavati is connected to Neocarpentelles with uniloc-
Finally, the tree produced by the mitochondrial rns dataset ular stromatic ascomata. Clade 4 includes taxa in sections
showed very good clustering of species belonging to sections Terrei, Candidi, and Bispori without evident teleomorphic
Aspergillus, Clavati, Circumdati, Nigri, Nidulantes, and Usti connections.
(with the exception of A. elongatus which is basal to the other
Usti species) with excellent bootstrap support (values range
from 93 to 100%; Figure 5) and secondarily weaker bootstrap 4. Discussion
values of 55–90% for species that belong to Flavi, Terrei,
and Fumigati (Figure 5). Thus, a general conclusion from Fungal phylogeny based solely on morphological criteria or
single gene analyses is that subgenera Aspergillus, Candidi, on single genes like the rRNA gene sequences may not always
Fumigati, and Terrei evidently form well supported clusters, elucidate the taxonomic status of the organisms examined
while sections of the subgenera Circumdati, and Nidulantes [14]. Also, studies based on a single gene do not always faith-
are dispersed. fully represent the history of the entire genome containing it
To achieve full exploitation of the information obtained and comparisons may give the wrong conclusions about the
from nuclear and mitochondrial genes, a tree was constructed relationship of a fungus with members of the same species or
based on the combined gene datasets. The concatenated even the same genus [44]. Certainly, the nuclear rDNA repeat
dataset included 4,507 characters, with 2,217 parsimony is the most popular region in molecular phylogenetic studies
informative characters and parsimony analysis provided a because it is multicopied and contains the highly conserved
single tree. The tree length was based on 13,497 steps (CI: genes 18S, 5.8S, and 28S, as well as the variable domains
0.40, HI: 0.60, RI: 0.52, RC: 0.21). Analysis of the same of ITS1 and ITS2, and the nontranscribed IGS region. In
dataset with NJ and BI methods produced similar trees with addition, the 18S and 28S regions often harbour group I
identical topologies wherever there was a strong support introns, inserted at highly conserved positions, contributing
(Figure 6). The sections that contained the larger numbers further information in regard to this region. These properties
of strains examined, that is, Flavi, Nigri, Nidulantes, along have been exploited at length in studies of Aspergillus species
with sections Aspergillus, Circumdati, Clavati, Fumigati, and [11, 15, 45] or other fungal genera [28, 46, 47]. In a similar
Terrei, were well discriminated with better support compared way, the mtDNA of fungi has lately attracted considerable
to the results produced by single gene analyses (Figure 6). interest as an alternative or complementary molecule for
In addition, the tree produced by the concatenated dataset phylogenetic studies [26, 35, 48], because of its high copy
further supports the polyphyly of subgenus Circumdati, as numbers, richness in AT sequences, lack of methylation,
single gene phylogenies have also shown, since sections Nigri, universal gene functions, highly conserved regions, as well
Flavi, and Circumdati form separate clades at the base of as variable domains and introns [49]. Currently, the most
the tree. The section Circumdati, represented in this study common mt genes used in fungal phylogenetic studies are
by species A. ochraceus and A. sclerotiorum, forms a sister rns and cox1, because the primers designed for these genes
clade to section Zonati, represented by A. clavatoflavus. Both can be applied to a wide range of taxonomically different
sections can be a sibling clade to section Nidulantes based fungi [50, 51]. Our analyses of representative species from the
only on the NJ analysis. Both MP and BI analyses show that most common sections of Aspergillus showed for the first time
this relation is not supported and thus sections Circumdati here that mitochondrial based phylogenies indicate different
and Zonati collapse at the base of the tree. evolutionary pathways which may provide valuable data for
Most importantly, the tree of the concatenated dataset taxonomic purposes when combined with morphological
presented a backbone of seven major clades which could and nuclear based molecular datasets.
It has often been argued that the reconstruction of suggest that section Usti should be retained as an independent
phylogenies is better achieved by multiple datasets, not only section within the subgenus Nidulantes with further support,
of nuclear but also of mitochondrial origin [52]. In the present since both the ITS and concatenated data analyses show that
study, the ITS and IGS regions from the nu-rRNA gene species of Usti, Nidulantes, Silvati, and Raperi comprise a
complex and rns and cox1 from the mt genomes were chosen single clade supported (e.g., from ITS dataset) by NJ-MP
to study the phylogenetic relationships of Aspergillus species. bootstrap (51% and 71%, resp.), and BI Posterior Probability
To overcome possible ambiguities raised by the analyses of (95%), and thus, altogether may be considered as one group
these genes, data from the nuclear genes rpb2 and benA within subgenus Nidulantes (Figures 2 and 6).
that are generally accepted as phylogenetically informative Until now, several genes have been proposed for the
were also used [5, 45, 53]. Although benA sequences were identification of species within the section of black aspergilli.
excluded from the concatenated data used in the most recent However, the results obtained are very variable. IGS and
multigene approach, since the presence of two or three cox1 sequences were found unsuitable to discriminate species
introns in the corresponding benA amplicons was attributed within this section (Supplemental Figures S1 and S2), because
to amplification of paralogous genes [23], we have chosen to they either exhibited too high intraspecies variability or this
use this data in our work for three reasons: (a) because 34 variability was also extended to inter-species comparisons
out of the 36 species examined contained all three introns, [3, 20]. Similarly, ITS sequences could distinguish four
only two species were missing one intron, (b) the identity groups within the A. niger species complex but failed to
levels of the 34 amplicon sequences (i.e., containing all three do so for species like A. carbonarius and A. sclerotioniger,
introns) was >84%, and (c) exon identity between all 36 or A. japonicus, A. aculeatus and A. uvarum which had
amplicons ranged between 86 and 96%. As shown in results, identical ITS sequences [20]. Also, mt cytochrome b (cob)
this choice was justified because the benA based phylogenetic gene sequences were informative for many species within
tree provided additional information as, for example, in the Nigri section but they failed to distinguish between A.
the case of members of section Nigri that were clustered niger and A. tubingensis [54]. Thus, the best discriminatory
together but clearly differentiated the uniseriate from the results based on single gene sequences were those obtained
biseriate species. Thus, all data from nuclear and mt genes from 𝛽-tubulin and calmodulin phylogenetic analysis which
were combined as a single unit in order to blend information placed 26 taxa within subgenus Circumdati section Nigri,
from two independent heritage lines, to examine whether this further dividing them into 5 main clades, grouping all
approach provides useful information in resolving phyloge- uniseriate species (7) in only one clade (the A. aculeatus
netic ambiguities within the genus Aspergillus. clade) [13, 45]. In the present work, single gene datasets with
The single gene trees obtained from ITS, IGS, rpb2, and many individual sequences (ITS, benA, and rpb2) and the
rns and, mainly, the concatenated datasets, strengthen the concatenated dataset clearly distinguished members of Nigri
fraternisation of Clavati to subgenus Fumigati, as species of and divided them into two groups: the first containing A.
section Clavati form a sister group to species within section aculeatus, A. aculeatinus, A. bahamensis, A. uvarum—in all
Fumigati, with MP bootstrap and PP support that reached 89 datasets—and A. homomorphus, A. japonicus—wherever data
and 100%, respectively, at the combined dataset (Figure 6). were available—, and the second composed of all the rest
This is in full accordance with results obtained from a recent species (see e.g., ITS or benA trees, Figures 2 and 3). It is
multilocus analysis that was based solely on nuclear genes interesting to note that these two clades correspond to the
and the addition of section Cervini species in this subgenus uniseriate and biseriate conidial head formation, which may
[23, 24]. Interestingly enough, our analysis based on the be a possible explanation why previous studies that did not
ITS dataset showed that although sufficient bootstrap or take into account this characteristic, failed to cluster them
posterior probability support was lacking, uniseriate species accordingly [13, 20, 45]. An additional explanation for the
that belong to subgenera Fumigati and Aspergillus clustered splitting of section Nigri species into two subgroups may be
together, and apart from other species of Aspergillus that are attributed to the range of variability resulting from adaptation
biseriate (Figure 2). of the species to the Mediterranean type ecosystem. This is
Section Usti is of particular interest since its elimination also evident by the variability found in the morphological
and the transfer of species A. ustus, A. deflectus, A. puniceus, characteristics of the Greek Aspergillus isolates (Table 2).
A. granulosus, and A. pseudodeflectus to section Nidulantes, The biseriate species within section Nigri cluster with good
and species A. conjunctus, A. funiculosus, A. silvaticus, A. bootstrap support (67% bootstrap for NJ and 84% PP) with
panamensis and A. anthodesmis to section Sparsi was initially biseriate species belonging to sections Flavipedes, Terrei,
proposed [15], only to be rejected a few years later, by Ochraceorosei, and Sparsi, while the uniseriate species of
exploiting the results of polyphasic approaches in which the Nigri are basal to the above-mentioned biseriate clades of
sequences of the nuclear 𝛽-tubulin, calmodulin, actin, and sections Flavi with good support (70% PP, Figure 2). Thus,
ITS genes were used [5, 18]. These works not only preserved the differentiation of Nigri species studied here into uniseriate
section Usti but also added eight more species in it, namely, and biseriate not only helped to group them in combination
A. ustus, A. puniceus, A. granulosus, A. pseudodeflectus, A. with molecular markers into two clearly distinct clades, but
calidoustus, A. insuetus, A. kevei, and Emericella heterothal- in addition clustered together biseriate species of all sections.
lica. Finally, more recent works added nine more Aspergillus Species of the Flavi section were grouped well together,
species in it, seven and two, respectively, bringing the total with high bootstrap support (>80%), irrelevant to the dataset
number to 21 [11, 24]. In full agreement with these, our results used. While single gene analysis, like benA, failed to separate
A. flavus from A. oryzae, the concatenated dataset clearly In conclusion, our study clearly shows that all molecules
demonstrated its advantages. It could fully discriminate all used contributed to the resolution of phylogenetic relation-
species examined and could also confirm that taxa within ships in Aspergillus, irrespective to the different ways that
the section, like A. oryzae, A. flavus, and A. parasiticus the phylogenetic data were processed. It also confirms that
are phylogenetically distantly related to A. ochraceus, as single gene based analyses do not solve all ambiguities and
previously proposed [55] and also recently shown with the do not always represent the evolutionary history of the
polyphyly of A. flavus, and in extent of section Flavi [56]. species. Certainly, Aspergillus is a very diverse genus and it
Previous reports showed that sections Circumdati, Flavi is not always easy to determine the phylogenetic relationships
and Nigri do not cluster together [23] and this was further between species. Thus, the combination of sequence informa-
confirmed by our phylogenetic analyses of ITS, benA, and tion from nuclear and mitochondrial genes resolves several
rpb2 datasets. However, it must be pointed out that this is circumscriptions and relationships of Aspergillus species
in contradiction with our morphological observations which within different sections and demonstrates the advantages of
this multigenic approach. It is expected that the vast amount
demonstrated significant morphological similarities, that is,
of information derived from the released complete genomes
biseriate conidial heads, large globose vesicles, ascostromata
of Aspergillus species will be utilized in future studies by many
when teleomorph present, and suggest a possible common researchers to select new additional genes from both (nuclear
origin of these sections. The latter is further supported and mitochondrial) evolutionary lines. Such combined data is
by the analysis of the mt rns sequences, even though the expected to provide the appropriate levels of resolving power
bootstrap values do not seem significant enough. Thus, in future phylogenetic studies.
this intriguing controversy between molecular data and
morphological observations underlines the necessity to use
polyphasic approaches in one hand and increase the number Acknowledgments
of species examined on the other in order to solve difficult The authors are greatly indebted to Dr. Stephen Peterson,
taxonomy problems. Peoria, USA, for providing ex-type strains of Aspergillus spp.
The single tree produced by our concatenated dataset This research was supported by the Greek General Secretariat
showed that although A. niveus and A. terreus, which belong of Research and Technology (Project TR-5) in the frame of the
to section Terrei, are closely related and with excellent support “Competitiveness” Program of European Commission.
(100% irrelevant of the method applied), they are far apart
from A. janus (section Flavipedes) and A. arenarius, both
of which were previously placed closely to species of Terrei, References
Flavipedes, and Candidi [23]. This causes some concern [1] V. Meyer, B. Wu, and A. F. J. Ram, “Aspergillus as a multi-
because our dataset places A. arenarius as a sister clade to purpose cell factory: current status and perspectives,” Biotech-
section Flavi with good support (bootstrap of 90% for NJ nology Letters, vol. 33, pp. 469–476, 2011.
and 100% for MP analyses; 69% Posterior Probabilities for [2] D. Lubertozzi and J. D. Keasling, “Developing Aspergillus as a
the BI analyses; Figure 6), contrary to suggestions that it is host for heterologous expression,” Biotechnology Advances, vol.
unrelated to any section of Aspergillus [24]. Similarly, our 27, no. 1, pp. 53–75, 2009.
concatenated analysis contradicts the proposed merger of [3] D. M. Geiser, M. A. Klich, J. C. Frisvad, S. W. Peterson, J. Varga,
sections Terrei and Flavipedes [24, 57] because A. janus is and R. A. Samson, “The current status of species recognition
excluded from being a sister clade of section Terrei (Figure 6). and identification in Aspergillus,” Studies in Mycology, vol. 59,
Since the datasets from the other genetic loci, examined in pp. 1–10, 2007.
this work, show a random positioning of A. janus and A. [4] J. C. Frisvad, T. O. Larsen, R. de Vries et al., “Secondary
arenarius, it is obvious that a thorough molecular analysis metabolite profiling, growth profiles and other tools for species
of the relations between Terrei and Flavipedes is urgently recognition and important Aspergillus mycotoxins,” Studies in
needed, especially if previous members of Versicolores, like Mycology, vol. 59, pp. 31–37, 2007.
A. janus and A. arenarius, are to be included within these [5] S. A. Balajee, J. Houbraken, P. E. Verweij et al., “Aspergillus
sections. Recently, Jurjevic et al. [58] undertook the study of species identification in the clinical setting,” Studies in Mycol-
section Versicolores and provided useful phylogenetic data on ogy, vol. 59, pp. 39–46, 2007.
species of this section based on multilocus DNA phylogeny, [6] P. Krijgsheld, R. Bleichrodt, G. J. van Veluw et al., “Development
but still the inclusion of these species in this section has not in Aspergillus,” Studies in Mycology, vol. 74, pp. 1–29, 2013.
been resolved. [7] B. K. Raper and D. I. Fennell, The Genus Aspergillus, The
As for the subgenus Nidulantes, the tree based on con- Williams and Wilkins, Baltimore, Md, USA, 1965.
catenated data showed that it is a compact assemblage of [8] W. Gams, M. Christensen, A. H. S. Onions, J. I. Pitt, and R.
species from sections Nidulantes and Usti, as expected [23, A. Samson, “Infrageneric taxa of Aspergillus,” in Advances in
24]. Finally, the intermixed cluster of five species from several Penicillium and Aspergillus Systematics, R. A. Samson and J. I.
sections that was not resolved by the analysis of the six genetic Pitt, Eds., pp. 55–62, Plenum Press, New York, NY, USA, 1985.
loci used in our work is another interesting area of research [9] J. I. Pitt, R. A. Samson, and J. C. Frisvad, “List of accepted
that will certainly require the introduction of either more species and their synonyms in the family Trichocomaceae,”
genes or more taxa, as it has already been done in other in Integration of Modern Taxonomic Methods For Penicillium
studies facing similar problems [3, 52]. and Aspergillus Classification, R. A. Samson and J. I. Pitt, Eds.,
pp. 9–49, Harwood Academic Publishers, Amsterdam, The [26] V. N. Kouvelis, A. Sialakouma, and M. A. Typas, “Mitochondrial
Netherlands, 2000. gene sequences alone or combined with ITS region sequences
[10] R. A. Samson and J. Varga, “What is a species in Aspergillus?” provide firm molecular criteria for the classification of Lecani-
Medical Mycology, vol. 47, supplement 1, pp. S13–S20, 2009. cillium species,” Mycological Research, vol. 112, no. 7, pp. 829–
844, 2008.
[11] R. A. Samson, J. Varga, M. Meijer, and J. C. Frisvad, “New taxa in
Aspergillus section Usti,” Studies in Mycology, vol. 69, pp. 81–97, [27] M. P. Pantou, O. K. Strunnikova, V. Y. Shakhnazarova, N. A.
2011. Vishnevskaya, V. G. Papalouka, and M. A. Typas, “Molecular
and immunochemical phylogeny of Verticillium species,” Myco-
[12] R. A. Samson, S. W. Peterson, J. C. Frisvad, and J. Varga, “New
logical Research, vol. 109, no. 8, pp. 889–902, 2005.
species in Aspergillus section Terrei,” Studies in Mycology, vol.
69, pp. 39–55, 2011. [28] M. P. Pantou, A. Mavridou, and M. A. Typas, “IGS sequence
variation, group-I introns and the complete nuclear ribosomal
[13] J. Varga, J. C. Frisvad, S. Kocsubé et al., “New and revisited DNA of the entomopathogenic fungus Metarhizium: excellent
species in Aspergillus section Nigri,” Studies in Mycology, vol. 69, tools for isolate detection and phylogenetic analysis,” Fungal
pp. 1–17, 2011. Genetics and Biology, vol. 38, no. 2, pp. 159–174, 2003.
[14] M. L. Berbee, D. A. Carmean, and K. Winka, “Ribosomal DNA [29] M. A. Klich and J. I. Pitt, A LaboraTory Guide to Common
and resolution of branching order among the Ascomycota: how Aspergillus Species and Their Teleomorphs, CSIRO, North Ryde,
many nucleotides are enough?” Molecular Phylogenetics and Australia, 1988.
Evolution, vol. 17, no. 3, pp. 337–344, 2000.
[30] M. A. Typas, A. M. Griffen, B. W. Bainbridge, and J. B. Heale,
[15] S. W. Peterson, “Phylogenetic relationships in Aspergillus based “Restriction fragment length polymorphisms in mitochondrial
on rDNA sequence analysis,” in Integration of Modern Taxo- DNA and ribosomal RNA gene complexes as an aid to the
nomic Methods for Penicillium and Aspergillus Classification, R. characterization of species and sub-species populations in the
A. Samson and J. I. Pitt, Eds., pp. 323–355, Harwood Academic genus Verticillium,” FEMS Microbiology Letters, vol. 95, no. 2-3,
Publishers, Amsterdam, The Netherlands, 2000. pp. 157–162, 1992.
[16] M. L. Berbee, “The phylogeny of plant and animal pathogens in [31] B. Henrion, F. Le Tacon, and F. Martin, “Rapid identification of
the Ascomycota,” Physiological and Molecular Plant Pathology, genetic variation of ectomycorrhizal fungi by amplification of
vol. 59, no. 4, pp. 165–187, 2001. ribosomal RNA genes,” New Phytologist, vol. 122, pp. 289–298,
[17] P. J. Keeling, M. A. Luker, and J. D. Palmer, “Evidence from beta- 1992.
tubulin phylogeny that microsporidia evolved from within the [32] Y. J. Liu, S. Whelen, and B. D. Hall, “Phylogenetic relationships
fungi,” Molecular Biology and Evolution, vol. 17, no. 1, pp. 23–31, among ascomycetes: evidence from an RNA polymerase II
2000. subunit,” Molecular Biology and Evolution, vol. 16, pp. 1799–
[18] J. Houbraken, M. Due, J. Varga, M. Meijer, J. C. Frisvad, and R. 1808, 1999.
A. Samson, “Polyphasic taxonomy of Aspergillus section Usti,” [33] N. L. Glass and G. C. Donaldson, “Development of primer
Studies in Mycology, vol. 59, pp. 107–128, 2007. sets designed for use with the PCR to amplify conserved genes
[19] R. A. Samson, S. Hong, S. W. Peterson, J. C. Frisvad, and J. Varga, from filamentous ascomycetes,” Applied and Environmental
“Polyphasic taxonomy of Aspergillus section Fumigati and its Microbiology, vol. 61, no. 4, pp. 1323–1330, 1995.
teleomorph Neosartorya,” Studies in Mycology, vol. 59, pp. 147– [34] T. J. White, T. Bruns, S. Lee, and J. W. Taylor, “Amplification
203, 2007. and direct sequencing of fungal ribosomal RNA genes for
[20] R. A. Samson, P. Noonim, M. Meijer, J. Houbraken, J. C. Frisvad, phylogenetics,” in PCR ProTocols, M. A. Innis, D. H. Gelfand,
and J. Varga, “Diagnostic tools to identify black aspergilli,” J. J. Sninsky, and T. J. White, Eds., pp. 315–322, Academic Press,
Studies in Mycology, vol. 59, pp. 129–145, 2007. San Diego, Calif, USA, 1990.
[21] J. Varga, M. Due, J. C. Frisvad, and R. A. Samson, “Taxonomic [35] D. V. Ghikas, V. N. Kouvelis, and M. A. Typas, “The com-
revision of Aspergillus section Clavati based on molecular, plete mitochondrial genome of the entomopathogenic fungus
morphological and physiological data,” Studies in Mycology, vol. Metarhizium anisopliae var. anisopliae: gene order and trn gene
59, pp. 89–106, 2007. clusters reveal a common evolutionary course for all Sordari-
omycetes, while intergenic regions show variation,” Archives of
[22] J. Varga, J. C. Frisvad, and R. A. Samson, “Polyphasic taxonomy
of Aspergillus section Candidi based on molecular, morpholog-
ical and physiological data,” Studies in Mycology, vol. 59, pp. 75– [36] D. L. Swofford, PAUP∗ . Phylogenetic Analysis Using Parsimony
88, 2007. (∗ ) and Other Methods, Version 4, Sinauer Associates, Sunder-
land, Mass, USA, 1998.
[23] S. W. Peterson, “Phylogenetic analysis of Aspergillus species
using DNA sequences from four loci,” Mycologia, vol. 100, pp. [37] J. P. Huelsenbeck, MrBayes: Bayesian Inference of Phylogeny
205–226, 2008. (Software), University of Rochester, New York, NY, USA, 2000.
[24] S. W. Peterson, J. Varga, J. C. Frisvad, and R. A. Samson, “Phy- [38] Z. Yang, “PAML: a program package for phylogenetic analysis
logeny and subgeneric taxonomy of Asperillus,” in Aspergillus by maximum likelihood,” CABIOS, vol. 13, pp. 555–556, 1997.
in the Genomic Era, J. Varga and R. A. Samson, Eds., pp. [39] G. Perrone, G. Stea, F. Epifani, J. Varga, J. C. Frisvad, and R.
33–56, Wageningen Academic Publishers, Wageningen, The A. Samson, “Aspergillus niger contains the cryptic phylogenetic
Netherlands, 2008. species A. awamori,” Fungal Biology, vol. 115, pp. 1138–1150, 2011.
[25] V. N. Kouvelis, D. V. Ghikas, S. Edgington, M. A. Typas, and [40] J. C. Frisvad and R. A. Samson, “Neopetromyces gen. nov. and an
D. Moore, “Molecular characterization of isolates of Beauveria overview of teleomorphs of Aspergillus subgenus Circumdati,”
bassiana obtained from overwintering and summer populations Studies in Mycology, vol. 2000, no. 45, pp. 201–207, 2000.
of Sunn Pest (Eurygaster integriceps),” Letters in Applied Micro- [41] B. W. Horn, G. G. Moore, and I. Carbone, “Sexual reproduction
biology, vol. 46, no. 3, pp. 414–420, 2008. in Aspergillus flavus,” Mycologia, vol. 101, pp. 423–429, 2009.
[42] B. W. Horn, J. H. Ramirez-Prado, and I. Carbone, “The sexual

state in Aspergillus parasiticus,” Mycologia, vol. 101, pp. 275–280,
2009.
[43] D. M. Geiser, “Sexual structures in Aspergillus: morphology,
importance and genomics,” Medical Mycology, vol. 47, supple-
ment 1, pp. S21–S26, 2009.
[44] B. F. Lang, M. W. Gray, and G. Burger, “Mitochondrial genome
evolution and the origin of eukaryotes,” Annual Reviews Genet-
ics, vol. 33, pp. 351–397, 1999.
[45] J. Varga, S. Kocsubé, B. Tóth et al., “Aspergillus brasiliensis sp.
nov., a biseriate black Aspergillus species with world-wide dis-
tribution,” International Journal of Systematic and Evolutionary
[46] C. Neuveglise, Y. Brygoo, and G. Riba, “28s rDNA group-I
introns: a powerful tool for identifying strains of Beauveria
brongniartii,” Molecular Ecology, vol. 6, no. 4, pp. 373–381, 1997.
[47] C. Wang, L. I. Zengzhi, M. A. Typas, and T. M. Butt, “Nuclear
large subunit rDNA group I intron distribution in a population
of Beauveria bassiana strains: phylogenetic implications,” Myco-
logical Research, vol. 107, no. 10, pp. 1189–1200, 2003.
[48] A. M. Kretzer and T. D. Bruns, “Use of atp6 in fungal phyloge-
netics: an example from the boletales,” Molecular Phylogenetics
and Evolution, vol. 13, no. 3, pp. 483–492, 1999.
[49] A. Cambell, J. Mrazek, and s. Karlin, “Genome signature
comparisons among prokaryote, plasmid, and mitochondrial
DNA,” Proceedings of the National Academy of Sciences USA, vol.
96, pp. 9184–9189, 1999.
[50] K. N. Li, D. I. Rouse, and T. L. German, “PCR primers that allow
intergeneric differentiation of ascomycetes and their application
to Verticillium spp,” Applied and Environmental Microbiology,
vol. 60, no. 12, pp. 4324–4331, 1994.
[51] X. J. Min and D. A. Hickey, “Assessing the effect of varying
sequence length on DNA barcoding of fungi,” Molecular Ecology
Notes, vol. 7, pp. 365–373, 2007.
[52] A. Rokas and S. B. Carroll, “More genes or more taxa? The
relative contribution of gene number and taxon number to
phylogenetic accuracy,” Molecular Biology and Evolution, vol.
22, no. 5, pp. 1337–1344, 2005.
[53] Y. Tanabe, M. Saikawa, M. M. Watanabe, and J. Sugiyama,
“Molecular phylogeny of Zygomycota based on EF-1𝛼 and RPB1
sequences: limitations and utility of alternative markers to
rDNA,” Molecular Phylogenetics and Evolution, vol. 30, no. 2, pp.
438–449, 2004.
[54] K. Yokoyama, L. Wang, M. Miyaji, and K. Nishimura, “Identi-
fication, classification and phylogeny of the Aspergillus section
Nigri inferred from mitochondrial cytochrome b gene,” FEMS
Microbiology Letters, vol. 200, no. 2, pp. 241–246, 2001.
[55] S. Nikkuni, H. Nakajima, S. I. Hoshina et al., “Evolutionary rela-
tionships among Aspergillus oryzae and related species based
on the sequences of 18S rRNA genes and internal transcribed
spacers,” Journal of General and Applied Microbiology, vol. 44,
no. 3, pp. 225–230, 1998.
[56] S. S. Gonçalves, J. F. Cano, A. M. Stchigel et al., “Molecular phy-
logeny and phenotypic variability of clinical and environmental
strains of Aspergillus flavus,” Fungal Biology, vol. 116, pp. 1146–
1155, 2012.
[57] J. Varga, B. T. Tóth, S. Kocsubé et al., “Evolutionary relationships
among Aspergillus terreus isolates and their relatives,” Antonie
Van Leeuwenhoek, vol. 88, pp. 141–150, 2005.
[58] Z. Jurjevic, S. W. Peterson, and B. W. Horn, “Aspergillus section
Versicolores: nine new species and multilocus DNA sequence
based phylogeny,” IMA Fungus, vol. 3, pp. 59–79, 2012.
http://dx.doi.org/10.1155/2013/703130
Research Article
Unraveling the Lipolytic Activity of Thermophilic Bacteria
Isolated from a Volcanic Environment
Panagiota M. Stathopoulou, Alexander L. Savvides,

Amalia D. Karagouni, and Dimitris G. Hatzinikolaou
Microbiology Group, Sector of Botany, Department of Biology, National and Kapodistrian University of Athens,
Zografou Campus, Zografou 15784, Attica, Greece
Correspondence should be addressed to Dimitris G. Hatzinikolaou; dhatzini@biol.uoa.gr
Received 9 February 2013; Accepted 25 March 2013
Copyright © 2013 Panagiota M. Stathopoulou et al. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.
In a bioprospecting effort towards novel thermostable lipases, we assessed the lipolytic profile of 101 bacterial strains isolated from
the volcanic area of Santorini, Aegean Sea, Greece. Screening of lipase activity was performed both in agar plates and liquid cultures
using olive oil as carbon source. Significant differences were observed between the two screening methods with no clear correlation
between them. While the percentage of lipase producing strains identified in agar plates was only 17%, lipolytic activity in liquid
culture supernatants was detected for 74% of them. Nine strains exhibiting elevated extracellular lipase activities were selected
for lipase production and biochemical characterization. The majority of lipase producers revealed high phylogenetic similarity
with Geobacillus species and related genera, whilst one of them was identified as Aneurinibacillus sp. Lipase biosynthesis strongly
depended on the carbon source that supplemented the culture medium. Olive oil induced lipase production in all strains, but
maximum enzyme yields for some of the strains were also obtained with Tween-80, mineral oil, and glycerol. Partially purified
lipases revealed optimal activity at 70–80∘ C and pH 8-9. Extensive thermal stability studies revealed marked thermostability for the
majority of the lipases as well as a two-step thermal deactivation pattern.
1. Introduction enzymes have been purified either from the wild-type culture
supernatants [8–10] or following cloning and expression in
Thermophilic microorganisms unequivocally represent a mesophilic hosts [11–14].
valuable source of highly thermostable enzymes, with numer- Volcanic areas represent ecological niches of rich meta-
ous advantages towards biotechnological applications due bolic diversity especially among bacteria and archaea [15–17],
to their overall inherent stability and high reaction rates and the isolation of several thermophilic microorganisms has
at elevated temperatures [1, 2]. Among them, lipases (EC been described from such natural sources. The vast majority
3.1.1.3), the enzymes that catalyze both the synthesis and of these studies, though, is focused on the phylogenetic
hydrolysis of long chain fatty acid esters (depending on water diversity of the corresponding microbial communities, and
availability), constitute one of the most versatile and widely only few reports are available in the literature assessing the
used biocatalytical group [3]. They are used in numerous enzymatic potential of the microbial strains isolated either
diverse biotechnological applications ranging from biodiesel from volcanic [18] or other extreme environments [19] with
and biopolymers production to the synthesis of fine chem- culture dependent or metagenomic approaches.
icals for medical, agrochemical, and cosmetic applications Pursuant to the above mentioned statements, screening
[3–5]. Due to this fact, novel thermostable lipases are in of thermophilic microorganisms for lipolytic activities
continuous demand for commercial applications especially in could facilitate the discovery, for industrial purposes, of
detergent, food, and pulp and paper industries [6]. As a result, novel lipases that are stable and function optimally at high
several thermophilic microbial strains able to produce ther- temperatures. In this work, we examined the enzymatic
mostable lipases have been isolated [7] and the corresponding lipolytic potential of 101 bacterial strains which had
previously been isolated from the volcanic habitat of the 2.3. Molecular Characterization of Isolates. Bacterial identifi-
Santorini volcano, at South Aegean Sea [20]. A thorough cation of the nine selected isolates was based on 16S rDNA
biochemical screening for extracellular lipase activity was sequence analysis. Genomic DNA extraction from liquid
performed both in solid and liquid media in an attempt to cultures was performed according to Haught et al. [22], and
establish benchmark criteria for such an analysis. The most the quantity of the isolated DNA was determined photomet-
active strains were selected and characterized with respect to rically [23]. The 16S rDNA was amplified by polymerase chain
their 16S rDNA sequences and the effect of various carbon reaction (PCR) using two universal primers [24, 25]:
sources on the production of extracellular lipolytic activity.
Finally, the pH and temperature-activity optima as well as the pA (5󸀠 -AGA GTT TGA TCC TGG CTC AG-3󸀠 ) and
corresponding thermal stability kinetics were determined
for the partially purified and concentrated lipases from all R1492 (5󸀠 -TAC GGY TAC CTT GTT ACG ACT T-3󸀠 ).
selected isolates.
Amplification reactions were performed in volumes of
50 𝜇L containing 40 ng template DNA, 0.4 𝜇M of each primer,
2. Materials and Methods 1X buffer with Mg2+ , 1 U KAPA Taq DNA polymerase
(Kapa Biosystems, Woburn, MA, USA), and 0.2 mM dNTPs.
2.1. Microbial Strains. The 101 thermophilic bacterial strains Nucleases free water was used to bring the reaction vol-
used in this study have been previously isolated from the ume to 50 𝜇L. After initial denaturation at 95∘ C for 2 min,
seawater and sediment near the active volcano of Santorini samples were cycled for 30 PCR cycles using the following
island at Aegean Sea, Greece [20]. All isolates were preserved cycle profile: 95∘ C denaturation for 30 s, primer annealing
in 30% (w/v) sterile glycerol solution at −80∘ C. at 53∘ C for 30 s, and primer extension at 72∘ C for 2 min,
plus a final 2 min elongation step at 72∘ C. Amplified PCR
products were separated by gel electrophoresis on 1.2% (w/v)
2.2. Growth Media and Conditions. All isolates were prelim- agarose gel and then purified using NucleoSpin Extract
inary screened for their lipolytic activity both in solid and PCR kit (MACHEREY-NAGEL, Germany). The 16S rDNA
liquid cultures. A standard inoculum was prepared for each fragment (>1400 bp) was fully sequenced in both direc-
strain from glycerol stocks on Nutrient Agar (Nutrient broth tions (Macrogen, Republic of Korea). and the correspond-
plus 3% w/v agar) plates. Following incubation for 24 h at ing gene sequences have been submitted to the GenBank
60∘ C, bacterial biomass collected in sterile water and diluted databases under accession numbers: JQ808132 to JQ808140
to an OD600 of 0.3 was served as inoculum. (http://www.ncbi.nlm.nih.gov/). The 16S sequences were
Solid cultures were conducted in sterile 12-well plates used for the determination of the evolutionary history (phy-
filled with Rhodamine B—Olive Oil agar medium containing logenetic tree) of the nine isolates using the Neighbor-Joining
10 g⋅L−1 olive oil (ROA) [8]. Twenty 𝜇L of standard inoculum method [26] in MEGA5 software [27]. The percentage of
was placed in the middle of each well (3 wells per strain), replicate trees in which the associated taxa clustered together
and the plates were incubated at 60∘ C up to 48 hours. was calculated by the bootstrap test (1000 replicates) [28].
Lipase production was evaluated from the diameter of the Evolutionary distances were computed using the Maximum
orange fluorescent halo (reaction of liberated fatty acids with Composite Likelihood method [29].
Rhodamine B) observed under UV exposure [21].
Liquid cultures were conducted in 100 mL Erlenmeyer 2.4. Preparation of Partially Purified Lipase Samples. Liquid
flasks with 20 mL working volume. The basal medium used cultures of the nine selected strains were conducted in 2 L
consisted of (g⋅L−1 ): NaNO3 , 3; K2 HPO4 , 1; KCl, 0.5; CaCl2 , Erlenmeyer flasks of 400 mL working volume (3 flasks per
0.1; MgSO4 ⋅7H2 O, 0.5; yeast extract, 1; trace elements solu- strain) as described above. Biomass was removed by cen-
tion, 1 mL⋅L−1 , supplemented with 20 g⋅L−1 olive oil as sole trifugation (6000 rpm, 5 min), and the resulting supernatant
carbon and energy source. Triplicate flasks were inoculated was passed through Whatman number 1 filter paper in
for each strain, with 0.5 mL standard inoculum, and all flasks order to remove the residual olive oil. The filtrate was then
were incubated in a thermostated orbital shaker (SANYO subjected to ammonium sulphate precipitation. Optimum
Biomedical, UK) at 60∘ C and 180 rpm. Evaporation was daily precipitation range was determined by employing a gradual
checked gravimetrically and compensated with the addition ammonium sulphate procedure (5% saturation step) and
of sterile water. Samples were collected every 24 h for the measuring specific lipase activity on the resulting super-
determination of biomass concentration and extracellular natants and precipitates. The optimum precipitation condi-
lipolytic activity. The effect of other carbon sources on tions were subsequently applied to each culture supernatant
lipase production was determined in a similar manner, by and the collected precipitates were resuspended in 20 mM
substituting olive oil with either glucose, mineral oil, corn potassium phosphate buffer (pH 8) and desalted using PD-
cob, wheat bran, eicosane, Tween-80, or glycerol at an initial 10 gel filtration columns (GE Healthcare) using the above
concentration of 20 g⋅L−1 . Lipase production was followed by buffer for elution. The concentrated and desalted enzyme
assaying of the lipase activity in the supernatant as described solutions were used for the determination of the optimum pH
in enzyme assays. All lipolytic activities were determined in and temperature and thermal stability of the corresponding
triplicate and reported as averages. lipolytic activities.
Table 1: Comparison of the lipolytic activities among the 101 Table 1: Continued.
thermophilic bacterial isolates in Rhodamine olive oil agar (ROA)
and olive oil liquid (OLM) cultures. ROA OLM
Strain
24 h 48 h 48 h 72 h
Strain
ROA OLM SP48 − − 0.043 0.027
24 h 48 h 48 h 72 h SP49 − − 0.033 0.032
SP1 − − − − SP50 − − 0.042 0.037
SP2 − − 0.040 0.063 SP51 − − 0.058 0.032
SP3 − − − − SP52 − − − −
SP4 − + 0.062 0.102 SP53 − − − −
SP5 − − − − SP54 − − 0.037 0.045
SP6 + + 0.058 0.073 SP55 − − 0.035 0.062
SP7 − − 0.250 0.085 SP56 − − 0.052 0.000
SP8 − − 0.250 0.270 SP57 − − 0.038 0.183
SP9 − − 0.307 0.167 SP58 − − − −
SP10 − − 0.500 0.200 SP59 − − 0.000 0.085
SP11 − − 0.358 0.217 SP60 − − − −
SP12 − − 0.350 0.142 SP61 − − 0.040 0.037
SP13 − − 0.583 0.175 SP62 − − − −
SP14 − − 1.070 2.367 SP63 − − 0.038 0.053
SP15 − − 0.108 0.147 SP64 − − 0.057 0.000
SP16 − − 0.022 0.065 SP65 − − 0.045 0.043
SP17 − − − − SP66 − − 0.115 0.047
SP18 − − 0.200 0.067 SP67 − − 0.053 0.075
SP19 − − 0.217 0.248 SP68 − − − −
SP20 − − 0.153 0.055 SP69 − − − −
SP21 − − 0.550 0.427 SP70 − − 0.043 0.070
SP22 − − 1.618 0.450 SP71 + + 0.043 0.052
SP23 − − 0.567 0.192 SP72 − − 0.208 0.244
SP24 − − 0.217 0.133 SP73 + ++ 1.083 0.740
SP25 − + 0.044 0.092 SP74 − − 0.088 0.000
SP26 − − − − SP75 + ++ 1.540 1.667
SP27 + + 0.150 0.288 SP76 − − 1.297 1.317
SP28 − − − − SP77 − − 0.483 0.228
SP29 − − 0.967 0.620 SP78 − − 0.220 0.150
SP30 − + 0.422 1.027 SP79 − − 1.167 1.245
SP31 − − − − SP80 + + 0.137 0.110
SP32 − + 0.011 0.085 SP81 − − 0.543 0.293
SP33 − − − − SP82 − + 0.235 0.250
SP34 − − − − SP83 + + 0.625 1.433
SP35 − − 0.095 0.258 SP84 − + 0.167 0.133
SP36 − − − − SP85 − − 0.077 0.000
SP37 − − − − SP86 − − 0.063 0.070
SP38 − − 0.178 0.767 SP87 − + 0.058 0.000
SP39 − − − − SP88 − − − −
SP40 − − 0.063 0.115 SP89 − − − −
SP41 − − − − SP90 − + 0.593 0.233
SP42 − − 0.450 0.192 SP91 − − 0.517 0.767
SP43 − − 0.120 0.072 SP92 + + 0.517 0.522
SP44 − − 0.120 0.123 SP93 ++ ++ 0.610 0.633
SP45 − − 0.153 0.088 SP94 − − 0.550 0.257
SP46 − − − − SP95 − − 1.027 0.600
SP47 − − 0.038 0.045 SP96 − − 0.073 0.085
Table 1: Continued. significant lipolytic activity (>0.05 nkat/mL) was detected in

ROA OLM
the culture supernatant of 56 strains (55.4%) while for 19
Strain strains (18.9%) the corresponding average activities were very
24 h 48 h 48 h 72 h
low (<0.05 nkat/mL) but detectable. For 26 strains (25.7%) no
SP97 − − − −
activity was detected. No clear correlation could be identified
SP98 − − − − among the two screening methods used. As a result strains
SP99 − − − − such as SP6 and SP71 yielded clear fluorescent halos during
SP100 − − 0.400 0.642 growth on plates while the same strains produced only low
SP101 − − 0.000 0.122 levels of lipase in liquid cultures; in addition strains like SP14,
In ROA cultures, qualitative evaluation was performed based on fluorescence SP22, SP29, SP76, and SP79 revealed the opposite lipolytic
halo production (−: no halo; +: partial coverage of the plate; ++: full phenotype (no halo on ROA, high extracellular lipolytic
coverage of the plate). In OLM cultures, values represent the extracellular activity in OLM). Only four out of the 101 strains, namely,
lipolytic activity in nkat⋅mL−1 (average of triplicate flasks: max. SD ± 9.4%).
Bold/italicized lines represent those strains selected for further study.
SP73, SP75, SP83, and SP93, revealed an equivalent lipolytic
profile both on solid and liquid cultures. In addition, we
could not descry a single strain yielding fluorescence halos
2.5. Enzyme Assays. Lipase activity assays in liquid samples on Rhodamine B agar plates without displaying even a small
were routinely performed using p-nitro phenyl palmitate amount of extracellular lipolytic activity in liquid cultures.
(pNPP) as substrate in 50 mM Tris-HCl buffer, pH 8, The Rhodamine B plate assay has been widely used in
essentially as described by Vorderwülbecke et al. [30] with the past by many researchers as the sole screening method
minor modifications. pNPP was dissolved in propane-2-ol for the identification of lipase producing microorganisms
at a final concentration of 7.5 mg⋅mL−1 . The final substrate from various environmental samples [31], including extreme
solution (950 𝜇L) was prepared by dropwise addition, under environments such as hot springs and soda lakes [32–34].
continuous vortexing, of 95 𝜇L pNPP solution into 855 𝜇L Our results clearly demonstrate the fact that the solid-
of buffer solution (prepared by dissolving 0.4 g of Triton X- phase Rhodamine system has failed to detect all the lipase
100 and 0.1 g gum arabic in 90 mL of buffer). The reaction producing microorganisms even though olive oil (a true
was initiated by adding 50 𝜇L sample into the above substrate lipase inducer) was employed as sole carbon and energy
solution followed by incubation in a water bath at 60∘ C source. Liquid cultures on olive oil combined with an extra-
for 15 min. The reaction was terminated by placing the cellular enzyme assay proved more reliable in identifying the
reaction mixture on ice for 10 min. p-nitro phenol (pNP) lipolytic potential of the strains under consideration. The
concentration was determined by measuring the absorbance lower sensitivity of Rhodamine assay that is sometimes used
at 410 nm using a calibration curve constructed at the assay to explain the discrepancy between the two approaches can
buffer (pH 8). For blank, we used heat sterilized (15 min, be easily overruled in our case, by the fact that we were able
121∘ C) samples undergone the same procedure. Activity was to identify strains that yielded very clear halos with only
expressed as nkat⋅mL−1 . traces of lipase activity in liquid cultures. These results clearly
show that liquid culture screening on various lipase inducers
yields more reliable results and should be preferred, especially
2.6. Determination of Lipase Thermal Stability and Activity today, where several high throughput liquid enzymatic assay
Optima. Optimum temperature for lipase activity was deter- screening methods are being available [35, 36].
mined under the above described assay conditions at pH 8,
using an incubation temperature range from 50 to 100∘ C.
Optimum pH values were determined in the same manner 3.2. Phylogeny of Lipase Producers. Following the results
at 60∘ C using an assay buffer range from 6 to 10. A different of this primary screening, we decided to proceed with
calibration curve was constructed and used for pNP at each our work by selecting those strains that either excreted
pH value used. For temperature stability, enzyme samples high average extracellular lipolytic activity in liquid cultures
were incubated in sealed screw-cap vials, at pH 7 (phosphate (>0.75 nkat⋅mL−1 ) or produced intense fluorescent halos
buffer, 100 mM) at various temperatures, sample aliquots in Rhodamine media. The nine strains that fulfill these
were withdrawn periodically, and remaining activity was criteria are in bold italics in Table 1. The selected strains
determined at regular assay conditions. All enzyme activities were subjected to 16S rDNA sequencing (NCBI Accession
were determined in triplicate and reported as averages. numbers: JQ808132 to JQ808140) and phylogenetic analysis.
A phylogenetic tree constructed using the corresponding
3. Results and Discussion sequences along with those from annotated Geobacillus and
other related genera is given in Figure 1. All nine lipolytic
3.1. Detection of Lipolytic Activity. The lipolytic activities of all strains revealed great phylogenetic similarity (>99%) with
isolates were assessed both qualitatively (agar plate cultures) known Geobacillus and other closely related species. This high
and quantitatively (liquid cultures) using olive oil as sole car- 16S rDNA similarity is characteristic of this relatively recently
bon and energy source. The corresponding results are sum- characterized genus [37] that includes species with promising
marized in Table 1. Only 17 out of 101 (16.8%) strains produced biotechnological potential [38]. Members of geobacilli have
clear fluorescent halos, designating free fatty acid release, dur- been reported to represent the majority of the species that
ing growth on ROA plates. On the contrary, in liquid cultures inhabit a variety of diverse “hot” environments such as
SP73 (JQ808135)
Geobacillus kaustophilus HTA426 (NC 006510)
42 SP93 (JQ808140)
Geobacillus kaustophilus NBRC 102445 (AB681787)
52 SP75 (JQ808136)
Geobacillus kaustophilus L17 (FJ823106)
SP22 (JQ808133)
88 SP79 (JQ808138)
Geobacillus vulcani BCRC 17563 (EU484349)
41 SP76 (JQ808137)
45 Geobacillus thermoleovorans NP33 (JQ343209)

54 Geobacillus lituanicus N-3 (AY044055)
Geobacillus zalihae NBRC 101842 (AB681570)
75
85
Geobacillus stearothermophilus NBRC 100862 (AB681268)
100 Geobacillus jurassicus DS1 (AY312404)
Geobacillus subterraneus 34 (T) (AF276306)

99
83 SP29 (JQ808134)
91 Geobacillus kaue BGSC W9A78 (AY608975)
87
Geobacillus thermoglucosidasius ATCC43742 (AB021197)
74
Anoxybacillus amylolyticus MR3C (AJ618979)
Aeribacillus pallidus DSM 3670 (Z26930)

100 Aeribacillus pallidus NCCP-131 (AB543491)
93 SP14 (JQ808132)
Bacillus subtilis DSM10 (AJ276351)
Aneurinibacillus danicus NBRC 102444 (AB681786)
100 SP83 (JQ808139)

100 Aneurinibacillus thermoaerophilus DSM 10154T (AB112726)
0.01
Figure 1: Evolutionary relationships of the selected strains using the Neighbor-Joining method. The bootstrap consensus tree was inferred
from 1000 replicates. Branches corresponding to partitions reproduced in less than 50% bootstrap replicates are collapsed. The percentage
of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) is shown next to the branches. The
tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree.
The evolutionary distances were computed using the Maximum Composite Likelihood method and are in the units of the number of base
substitutions per site. All positions containing gaps and missing data were eliminated.
marine thermal vents [39, 40], high temperature oil fields although it was possible to identify a provisional putative
[37], hot springs [41], geothermal regions [42], and sugar lipase gene in the G. kaustophilus HTA426 complete genome
refinery wastewaters [43]. sequence (NCBI Accession number NC 006510, Gene ID:
In our case, three of the strains, namely, SP73, SP75, and 3185865). Isolates SP22 and SP79 that revealed identical 16S
SP93 have been clustered together with known Geobacillus rDNA partial sequences clustered together with close phylo-
kaustophilus strains most probably representing strains of this genetic similarity to the G. kaustophilus group, while isolates
species. To our knowledge there are no published reports SP76 and SP29 can clearly be characterized as members of
on lipolytic activities from G. kaustophilus related strains the Geobacillus group closely related to G. thermoleovorans
Glycerol 1.18 1.1 1.43 3.4
Eicosane 2.2 1.38 1.41 2.4 1.87 2.86
Wheat bran 4.88 1.13
Tween-80 1.85 1.28 1.21 5.6 6.21 1.15
Mineral oil 1.65 1.59 1.03 3.48 1.23 2.8
Olive oil 2.64 2.05 1.61 1.29 4.61 1.48 1.16 1.58
Glucose 1.78 1.92 4.12
SP14 SP22 SP29 SP73 SP75 SP76 SP79 SP83 SP93

−1
Figure 2: Effect of carbon source type on maximum extracellular lipase production (nkat⋅mL ) in liquid cultures, achieved between 48 and
72 h of growth, depending on the source used. Circle areas are proportional to enzyme activities, while only activities above 1 nkat⋅mL−1 are
shown with numerical values. Black circles denote no detection of lipolytic activity. Values represent the mean of triplicate flasks. Maximum
SD ± 18.2%.
and G. kaue, respectively. G. thermoleovorans is a known isolates. Olive oil, a well-known lipase inducer, was able to
lipase producer among the geobacilli, and the corresponding induce significant lipase production levels in all nine strains,
enzymes from various strains of this species have been although it proved to be the optimal carbon source only
purified and characterized either from wild-type cultures for strain SP75. These results justify its use as the preferred
[9, 44] or following heterologous expression in E. coli [12, carbon source in various screening studies for lipolytic
45, 46]. On the contrary, there is no information in the activity in thermophilic bacteria [7] or in studies concerning
literature concerning lipolytic activity by G. kaue strains. the optimization of lipase production using wild type strains
Isolate SP14 clustered together with the recently characterized [53]. Tween-80, an oleic acid ester (Polyoxyethylene sorbitan
Aeribacillus pallidus species, the only characterized member monooleate), yielded maximum lipase titers in strains SP79
in the Aeribacillus taxonomic group [47]. Concerning lipase and SP83, but it completely inhibited lipase production in
production, only an A. pallidus strain isolated from thermal strains SP22 and SP75. Its addition in culture media for lipase
sites in Mexico, has been reported to express lipolytic activity production by members of the Bacillaceae family has often
in ROA plates [48], although this was not the case for strain been evaluated, exhibiting either inducing [54–56] or inhibit-
SP14 that showed high extracellular lipase activity only in ing effect [8, 53, 57, 58]. Optimum lipase production with
liquid cultures. Finally, strain SP83 revealed an identical Tween-80 as sole carbon source has been also achieved with a
16S rDNA sequence with Aneurinibacillus thermoaerophilus G. stearothermophilus strain following medium optimization
DSM 10154, representing the most phylogenetically distant [59]. All but one strains were able to produce at least some
member among the lipase producers in our study. The lipase activity in the presence of pure alkanes (eicosane)
Aneurinibacillus genus does not belong to the Bacillaceae or alkane mixtures (mineral oil) as sole carbon sources.
family (like Geobacillus and Aeribacillus), since is a member Stimulation of lipase production by aliphatic hydrocarbons
of the Paenibacillaceae family [49]. Strains related to this in bacteria, although documented several decades ago [60],
group have been isolated from geothermal soils throughout has been reported only once for members of the Bacillaceae
the globe [50, 51], while lipase production has been optimized family [44]. It is also noteworthy that eicosane and mineral
in submerged cultures of an A. thermoaerophilus strain oil revealed very similar lipase production patterns among
isolated from a Malaysian hot spring [52]. the nine strains (Figure 2), suggesting that lipase biosynthesis
by alkanes in thermophilic bacteria is induced in a manner
independent of their chain length.
3.3. Lipase Production Pattern. The effect of carbon source on The two low molecular weight compounds examined,
lipase production by the nine selected isolates was studied glycerol and glucose, showed limited lipase induction ability
using the minimal basal medium supplemented with an with two notable exceptions of an apparent inducing effect;
initial concentration of 20 g⋅L−1 of various carbon sources. that of glycerol for strain SP93 and that of glucose for strain
Samples were withdrawn twice a day, and the lipase activity SP79. A repressing effect of glucose [55, 61] or glycerol [62] on
was determined in the cell free supernatant. Growth of lipase production by members of the Bacillaceae family has
the microorganisms was observed in all cases as verified been reported in several studies, but this does not seem to be a
by either OD600 or cfu measurements (data not shown). universal phenomenon, since both glucose and glycerol were
Figure 2 summarizes the corresponding lipase production found to be excellent carbon sources for lipase production
pattern which appeared to be quite diverse among the nine by the G. stearothermophilus strain-5 [59, 63]. Wheat bran,
a complex agricultural byproduct, was also examined on the and G. thermoleovorans CCR11 [12], 65∘ C for Geobacillus sp.
basis of its use in several reports as an inducer of lipase strain ARM [68], G. stearothermophilus strain-5 [63], and G.
activity by various fungi [64, 65], despite the fact that there are thermoleovorans [45, 71] and 70∘ C for the lipase of Geobacillus
no references concerning its use as carbon source for lipase sp. T1 [70], and G. zalihae [7]. These results indicate that the
production in bacteria. Wheat bran was indeed a poor lipase Santorini volcanic habitat is most probably an environment
inducer for the majority of the bacterial strains, but quite that hosts bacterial species that harbor enzymatic activities
surprisingly it was proved to be the optimal carbon source of exceptional thermophilicity.
for the Aeribacillus sp. strain SP14. The latter isolate was also The above promising results concerning the temperature
among the three strains that were able to produce lipase in all optima of the nine partially purified lipases led us to pursue
carbon sources examined, the other two being strains SP76 a thorough study on their thermal stability properties in 10∘ C
and SP79. increments. In agreement with their high optimum tempera-
Lipase production among the nine isolates did not reveal tures, the lipases from the Santorini geobacilli revealed very
any specific correlations with their evolutionary relation- significant thermal stability properties. With the exception of
ships. Phylogenetically close isolates such as SP22 and SP79 the lipases from isolates SP75 and SP76 that maintained 82%
showed quite diverse lipolytic activity patterns, while the and 75% of their activity, respectively, after 1 h incubation at
opposite was observed with several other pairs of isolates such 60∘ C all other lipases maintained over 90% of their activity at
as SP14 and SP79. The only weak relation involved the three G. this temperature and incubation time (data not shown). For
kaustophilus strains of our study (SP73, SP75, and SP93) that the temperature range from 70 to 100∘ C, thermal deactivation
proved to be the poorest lipase producers, both in substrate curves are given in Figure 4. Analysis of the time point data
diversity and maximum observed lipolytic activities. It is for all enzymes using SigmaPlot software revealed that first
noteworthy though to point out that the maximum lipase order deactivation was not able to adequately describe the
activities obtained in this experimental set are comparable deactivation kinetics of the lipases at various temperatures.
to those reported in studies concerning the optimization of In an effort to find the best fit model we applied the approach
medium/culture parameters for lipase production by various of Henley and Sadana [72] where lipase thermal inactivation
thermophilic Bacillus and Geobacillus strains [53], despite the is sought to take place following a two-step deactivation
fact that we did not perform any optimization efforts. process, through an intermediate state, as follows:
𝑘1 𝑘2
𝐸0 󳨀→ 𝐸1 󳨀→ 𝐸2 , (1)
3.4. Assessment of Lipolytic Activities. Following partial
purification from the corresponding culture supernatants, we where 𝐸0 , 𝐸1 , and 𝐸2 are the specific activities of the initial,
performed a biochemical characterization of the nine lipases intermediate, and final state, respectively, and 𝑘1 and 𝑘2 are
by determining their temperature and pH-activity optima first-order deactivation constants. By defining 𝛼1 = 𝐸1 /𝐸0
(Figure 3) as well as their thermostability kinetics (Figure 4). and 𝛼2 = 𝐸2 /𝐸0 and assuming that the 𝐸2 state is completely
All lipases had a relatively basic pH optimum between 8 and inactive (𝑎2 = 0), the observed residual enzyme activity as a
9. The functional pH range was relatively narrow between 7 function of time, 𝛼, is given by (2) following integration over
and 9.5 with most of the enzymes showing a sharp decrease time [72]:
in the activity at pH 10. The lipases from strains SP22 and
SP83 were the most alkalophilic, showing optimum activity 𝛼1 𝑘1
at pH 9 and maintaining more than 30% of their activity 𝛼 = [1 + ] ⋅ exp (−𝑘1 𝑡)
𝑘2 − 𝑘1
at pH 10. These pH optima are similar to those reported (2)
for several Bacillus species [8, 11, 66, 67], although the pH- 𝑎𝑘
− [ 1 1 ] ⋅ exp (−𝑘2 𝑡) .
activity profiles for the majority of mesophilic Bacilli are 𝑘2 − 𝑘1
wider, probably because pH effects are being determined at
moderate temperatures between 20 and 30∘ C. As far as the The fitting of the experimental data into (2) for each lipase
lipases from other geobacilli are concerned, the reported pH and temperature was performed using the nonlinear regres-
optima are also 8 [32, 63, 68] or 9 [12, 13, 69, 70] indicating a sion routines of Sigmaplot software, and the results are
universal lipolytic pH-activity pattern within the genus. presented in Figure 4 and Table 2. Regression was allowed
The temperature optima determined for the nine partially to converge freely into the least-squares value and the only
purified enzymes were relatively high and reached the level restriction that was applied concerned the 𝛼1 specific activity
of either 70 (4 isolates) or 80∘ C (5 isolates). The lipases for ratio that was confined between the values of 0 and 1.
strains SP14 (A. pallidus), SP22, SP73, SP79 (G. kaustophilus), All lipases seemed to follow the two-step deactiva-
and SP29 (G. kaue) had the highest temperature-activity tion scheme, since we observed an excellent fitting of the
optima maintaining (with the exception of isolate SP79) more experimental data into (2). This means that the enzymes
than 80% of their optimum activity even at the temperature upon the effect of temperature undergo a relatively rapid
of 90∘ C. These lipolytic activity optima are significantly inactivation into the 𝐸1 intermediate form, which has lower
higher than those reported for other geobacilli. More specif- activity compared to the native enzyme (𝐸0 ). The deactivation
ically, the optimum reported activity temperatures were rate, though, from the intermediate form to the completely
55∘ C for the G. stearothermophilus JC [13] and Geobacillus inactive enzyme (𝐸2 ) is significantly slower, advocating for a
sp. TW1 [32] lipases, 60∘ C for Geobacillus sp. SBS-4S [69] more compact and heat resistant character for the 𝐸1 enzyme.
Temperature (∘ C) Temperature (∘ C) Temperature (∘ C)

40 50 60 70 80 90 40 50 60 70 80 90 40 50 60 70 80 90
SP14 SP22 SP29
100 100 100
80 80 80
Relative activity (%)

60 60 60
40 40 40
20 20 20
0 0 0
5 6 7 8 9 10 5 6 7 8 9 10 5 6 7 8 9 10
pH pH pH
∘
Temperature ( C) Temperature (∘ C) Temperature (∘ C)
40 50 60 70 80 90 40 50 60 70 80 90 40 50 60 70 80 90
SP73 SP75 SP76
100 100 100
80 80 80
60 60 Relative activity (%) 60
40 40 40
20 20 20
0 0 0
5 6 7 8 9 10 5 6 7 8 9 10 5 6 7 8 9 10
pH pH pH
Temperature (∘ C) Temperature (∘ C) Temperature (∘ C)

40 50 60 70 80 90 40 50 60 70 80 90 40 50 60 70 80 90
SP79 SP83 SP93
100 100 100
80 80 80
60 60 60
40 40 40
20 20 20
0 0 0
5 6 7 8 9 10 5 6 7 8 9 10 5 6 7 8 9 10
pH pH pH
Figure 3: Effect of pH () and temperature (I) on the activity of the partially purified lipases from the nine selected strains. Error bars
represent the SD from triplicate determinations.
1 SP14 1 SP22
Residual activity 0.8 0.8
Relative activity
0.6 0.6
0.4 0.4
0.2 0.2
0 0
0 60 120 180 24012002400 0 60 120 180 24012002400
1 SP29 1 SP73
0.8 0.8
Relative activity
Relative activity
0.6 0.6
0.4 0.4
0.2 0.2
0 0
0 60 120 180 24012002400 0 60 120 180 24012002400
1 SP75 1 SP76
0.8 0.8
Relative activity
Relative activity
0.6 0.6
0.4 0.4
0.2 0.2
0 0
0 60 120 180 24012002400 0 60 120 180 2401200 2400
1 SP79 1 SP83
0.8 0.8
Relative activity
Relative activity
0.6 0.6
0.4 0.4
0.2 0.2
0 0
0 60 120 180 24012002400 0 60 120 180 2401200 2400
Time (min)
1 SP93
0.8
Relative activity
0.6
0.4
0.2
0
0 60 120 180 24012002400
Time (min)
Figure 4: Thermal stability profiles of the nine partially purified lipases. 󳵻: 70∘ C; 󳵳: 80∘ C; ◻: 90∘ C; ◼: 100∘ C. Straight lines represent the
non-linear fitting of (2) in the experimental data.
Table 2: Values of the kinetic constants for the thermal deactivation kinetics as determined through nonlinear regression fitting of the
experimental data in (2). Activation energies for k1 and k2 were determined from the Arrhenius equation through linear regression.
SP14 SP22 SP29 SP73 SP75

70∘ C 0.647 ± 0.021 0.589 ± 0.060 0.662 ± 0.055 0.740 ± 0.021 0.267 ± 0.030
80∘ C 0.759 ± 0.007 0.879 ± 0.035 0.673 ± 0.040 0.726 ± 0.028 0.459 ± 0.044
𝛼1
90∘ C 0.690 ± 0.026 0.687 ± 0.026 0.566 ± 0.031 0.562 ± 0.059 0.335 ± 0.005
100∘ C 0.328 ± 0.027 0.432 ± 0.006 0.500 ± 0.011 0.399 ± 0.034 0.302 ± 0.009
70∘ C 2.24 ± 0.37 × 10−2 5.06 ± 1.14 × 10−3 1.43 ± 0.44 × 10−2 3.21 ± 0.74 × 10−2 2.54 ± 0.30 × 10−1
80∘ C 1.54 ± 0.12 × 10−1 3.45 ± 0.36 × 10−1 1.57 ± 0.54 × 10−1 1.66 ± 0.47 × 10−1 2.07 ± 0.45 × 10−1
𝑘1 (min−1 ) 90∘ C 2.30 ± 0.65 × 10−1 5.33 ± 0.36 × 10−1 2.04 ± 0.45 × 10−1 1.45 ± 0.52 × 10−1 3.96 ± 0.13 × 10−1
100∘ C 3.59 ± 0.40 × 10−1 8.81 ± 0.37 × 10−1 6.47 ± 0.54 × 10−1 6.27 ± 1.27 × 10−1 1.42 ± 0.08 × 100
𝐸𝑎 (kJ/mole) 93.72 171.58 125.24 93.56 60.96
70∘ C 1.82 ± 0.27 × 10−4 9.42 × 1.49 × 10−5 8.26 ± 1.13 × 10−4 7.06 ± 0.62 × 10−4 3.59 ± 1.33 × 10−4
80∘ C 1.83 ± 0.10 × 10−3 4.85 ± 0.31 × 10−4 1.76 ± 0.64 × 10−3 3.21 ± 0.35 × 10−3 9.20 ± 1.48 × 10−3
𝑘2 (min−1 ) 90∘ C 2.81 ± 0.45 × 10−3 4.96 ± 0.56 × 10−3 3.05 ± 0.66 × 10−3 3.27 ± 1.22 × 10−3 9.09 ± 0.34 × 10−3
100∘ C 7.84 ± 2.32 × 10−3 1.45 ± 0.07 × 10−2 7.43 ± 0.68 × 10−3 9.42 ± 2.68 × 10−3 6.19 ± 0.32 × 10−2
𝐸𝑎 (kJ/mole) 125.29 185.85 80.47 83.19 165.09
SP76 SP79 SP83 SP93
70∘ C 0.493 ± 0.055 0.639 ± 0.013 0.627 ± 0.026 0.662 ± 0.006
80∘ C 0.377 ± 0.008 0.384 ± 0.047 0.458 ± 0.029 0.380 ± 0.011
𝛼1
90∘ C 0.270 ± 0.048 0.330 ± 0.022 0.254 ± 0.053 0.366 ± 0.083
100∘ C 0.176 ± 0.054 0.313 ± 0.013 0.491 ± 0.052 0.271 ± 0.065
70∘ C 5.36 ± 1.21 × 10−2 7.23 ± 1.63 × 10−2 4.12 ± 0.53 × 10−2 2.57 ± 0.12 × 10−2
80∘ C 2.02 ± 0.08 × 10−1 8.01 ± 1.17 × 10−2 1.64 ± 0.23 × 10−1 7.22 ± 0.23 × 10−2
𝑘1 (min−1 ) 90∘ C 2.90 ± 0.68 × 10−1 1.48 ± 0.10 × 10−1 2.77 ± 0.60 × 10−1 2.02 ± 0.05 × 10−1
100∘ C 3.68 ± 0.58 × 10−1 5.05 ± 0.27 × 10−1 1.19 ± 0.47 × 101 1.87 ± 0.46 × 100
𝐸𝑎 (kJ/mole) 66.00 67.92 185.06 147.11
70∘ C 3.77 ± 0.82 × 10−3 1.39 ± 0.16 × 10−3 3.30 ± 0.24 × 10−3 5.63 ± 0.13 × 10−4
80∘ C 4.21 ± 0.25 × 10−3 3.04 ± 0.64 × 10−3 4.88 ± 0.80 × 10−3 1.09 ± 0.03 × 10−2
𝑘2 (min−1 ) 90∘ C 2.06 ± 0.35 × 10−2 7.81 ± 1.43 × 10−3 1.12 ± 0.49 × 10−2 2.62 ± 0.61 × 10−2
100∘ C 3.23 ± 1.50 × 10−2 1.77 ± 0.27 × 10−2 9.49 ± 0.88 × 10−2 7.25 ± 1.50 × 10−2
𝐸𝑎 (kJ/mole) 85.26 91.24 115.04 165.59
This was verified by the fact that for all enzymes and at all following 30 min incubation at 80∘ C [7], the Geobacillus sp.
temperatures examined the values of the deactivation con- SBS-4S lipase when exposed for 100 min at 60∘ C [69], and the
stant 𝑘2 were on average two orders of magnitude lower than lipase from G. stearothermophilus JC after 30 min at 70∘ C [13],
the corresponding 𝑘1 levels (Table 2). With the exceptions of while the Geobacillus sp. strain ARM lipase had half lives of
the lipases from isolates SP75 and SP76 above mentioned, only few minutes at 60 and 70∘ C [68].
all other enzymes showed exceptional thermal stabilities that
allowed clear detection of activity even after 1 h incubation at
100∘ C. The most thermostable enzymes were those of strains 4. Conclusions
SP14, SP22, SP29, and SP73. The most significant differenti-
ation of these four lipases that seemed to contribute in their An extensive assessment of the lipolytic potential in a set of
observed high thermal stability was the fact that the specific thermophilic bacteria isolated from the Santorini volcanic
activity ratio 𝛼1 for these enzymes was markedly higher habitat was performed. By combining both solid and liquid
than the other five lipases, and more importantly, its value culture techniques, a widespread lipolytic phenotype of vari-
was practically stable in the temperature range from 70 to able intensity was revealed, that involved almost 75% of the
90∘ C. As was the case with their temperature-activity optima, isolates. The genus Geobacillus was the dominant one within
the lipases from the Santorini thermophilic bacilli were also the subset of nine strains selected for their high extracellular
more thermostable than the majority of other Geobacillus lipase production. Enzyme levels were strongly dependent on
sp. lipases reported in the literature. Although this is the the type of carbon source, but we were not able to clearly
first extensive study on the thermal stabilities for Geobacillus identify a universal phenotype for lipase induction among the
lipases, literature data fully support our claim, since complete strains. The biochemical characterization of the nine lipases
loss of activity has been reported for the G. zalihae lipase marked a significant differentiation to most other Geobacillus
lipases with respect to the effect of temperature on the activity [12] R. Quintana-Castro, P. Dı́az, G. Valerio-Alfaro, H. S. Garcı́a, and
and stability of the corresponding enzymes. The majority R. Oliart-Ros, “Gene cloning, expression, and characterization
of lipases from the Santorini thermophilic bacteria revealed of the Geobacillus thermoleovorans CCR11 thermoalkaliphilic
exceptional thermostability with high optimum activity lipase,” Molecular Biotechnology, vol. 42, no. 1, pp. 75–83, 2009.
temperatures, thus representing very promising candidate [13] Y. Jiang, X. Zhou, and Z. Chen, “Cloning, expression, and
enzymes for a variety of high temperature industrial lipolytic biochemical characterization of a thermostable lipase from
applications. Such an endeavor would probably require their Geobacillus stearothermophilus JC,” World Journal of Microbiol-
efficient cloning and overexpression in mesophilic hosts, ogy and Biotechnology, vol. 26, no. 4, pp. 747–751, 2010.
even though that for some of the strains lipase production [14] T. C. Leow, R. N. Abd Rahman, M. Basri, and A. B. Salleh, “High
was at relatively high levels compared to other wild-type level expression of thermostable lipase from Geobacillus sp.
strain T1,” Bioscience, Biotechnology and Biochemistry, vol. 68,
thermophilic bacterial strains.
no. 1, pp. 96–103, 2004.
[15] R. J. Martinez, H. J. Mills, S. Story, and P. A. Sobecky, “Prokary-
Conflict of Interests otic diversity and metabolically active microbial populations in
sediments from an active mud volcano in the Gulf of Mexico,”
All the authors of the paper do not have a direct financial Environmental Microbiology, vol. 8, no. 10, pp. 1783–1796, 2006.
relation with any identity mentioned in the paper that might [16] M. G. Pachiadaki, V. Lykousis, E. G. Stefanou, and K. A.
lead to a conflict of interests. Kormas, “Prokaryotic community structure and diversity in the
sediments of an active submarine mud volcano (Kazan mud
References volcano, East Mediterranean Sea),” FEMS Microbiology Ecology,
vol. 72, no. 3, pp. 429–444, 2010.
[1] C. Vieille and G. J. Zeikus, “Hyperthermophilic enzymes: [17] M. I. Prokofeva, I. V. Kublanov, O. Nercessian et al., “Cultivated
sources, uses, and molecular mechanisms for thermostability,” anaerobic acidophilic/acidotolerant thermophiles from terres-
Microbiology and Molecular Biology Reviews, vol. 65, no. 1, pp. trial and deep-sea hydrothermal habitats,” Extremophiles, vol. 9,
1–43, 2001. no. 6, pp. 437–448, 2005.
[2] P. Turner, G. Mamo, and E. N. Karlsson, “Potential and utiliza- [18] I. V. Kublanov, A. A. Perevalova, G. B. Slobodkina et al., “Biodi-
tion of thermophiles and thermostable enzymes in biorefining,” versity of thermophilic prokaryotes with hydrolytic activities in
Microbial Cell Factories, vol. 6, p. 9, 2007. hot springs of Uzon caldera, Kamchatka (Russia),” Applied and
[3] R. Gupta, N. Gupta, and P. Rathi, “Bacterial lipases: an overview Environmental Microbiology, vol. 75, no. 1, pp. 286–291, 2009.
of production, purification and biochemical properties,” [19] G. F. da Cruz, C. F. F. Angolini, L. G. De Oliveira et al., “Search-
Applied Microbiology and Biotechnology, vol. 64, no. 6, pp. ing for monooxygenases and hydrolases in bacteria from an
763–781, 2004. extreme environment,” Applied Microbiology and Biotechnology,
[4] K.-E. Jaeger and T. Eggert, “Lipases for biotechnology,” Current vol. 87, no. 1, pp. 319–329, 2010.
Opinion in Biotechnology, vol. 13, no. 4, pp. 390–397, 2002. [20] C. Meintanis, K. I. Chalkou, K. A. Kormas, and A. D. Karagouni,
[5] H. Treichel, D. de Oliveira, M. A. Mazutti, M. Di Luccio, and “Biodegradation of crude oil by thermophilic bacteria isolated
J. V. Oliveira, “A review on microbial lipases production,” Food from a volcano island,” Biodegradation, vol. 17, no. 2, pp. 105–111,
and Bioprocess Technology, vol. 3, no. 2, pp. 182–196, 2010. 2006.
[6] M. Salameh and J. Wiegel, “Lipases from extremophiles and [21] G. Kouker and K. E. Jaeger, “Specific and sensitive plate assay
potential for industrial applications,” Advances in Applied for bacterial lipases,” Applied and Environmental Microbiology,
Microbiology, vol. 61, pp. 253–283, 2007. vol. 53, no. 1, pp. 211–213, 1987.
[7] R. N. Abd Rahman, T. C. Leow, A. B. Salleh, and M. Basri,
[22] C. Haught, D. L. Wilkinson, K. Zgafas, and R. G. Harrison,
“Geobacillus zalihae sp. nov., a thermophilic lipolytic bacterium
“A method to insert a DNA fragment into a double-stranded
isolated from palm oil mill effluent in Malaysia,” BMC Microbi-
plasmid,” BioTechniques, vol. 16, no. 1, pp. 46–48, 1994.
ology, vol. 7, p. 77, 2007.
[23] J. Sambrook, E. F. Fritsch, and T. Maniatis, Molecular Cloning:
[8] C. Schmidt-Dannert, H. Sztajer, W. Stocklein, U. Menge, and
A Laboratory Manual, Cold Spring Harbor Laboratory, Cold
R. D. Schmid, “Screening, purification and properties of a ther-
Spring Harbor, NY, USA, 1989.
mophilic lipase from Bacillus thermocatenulatus,” Biochimica et
Biophysica Acta, vol. 1214, no. 1, pp. 43–53, 1994. [24] U. Edwars, T. Rogall, H. Blocker, M. Emde, and E. C. Bottger,
[9] L. D. Castro-Ochoa, C. Rodrı́guez-Gómez, G. Valerio-Alfaro, “Isolation and direct complete nucleotide determination of
and R. Oliart Ros, “Screening, purification and characterization entire genes. Characterization of a gene coding for 16S riboso-
of the thermoalkalophilic lipase produced by Bacillus thermal RNA,” Nucleic Acids Research, vol. 17, no. 19, pp. 7843–7853,
moleovorans CCR11,” Enzyme and Microbial Technology, vol. 37, 1989.
no. 6, pp. 648–654, 2005. [25] D. J. Lane, “16S/23S rRNA Sequencing,” in Nucleic Acid
[10] M. A. Salameh and J. Wiegel, “Purification and characterization Techniques in Bacterial Systematics, E. Stackebrandt and M.
of two highly thermophilic alkaline lipases from Thermosyn- Goodfellow, Eds., pp. 115–147, John Wiley & Sons, New York,
tropha lipolytica,” Applied and Environmental Microbiology, vol. NY, USA, 1991.
73, no. 23, pp. 7725–7731, 2007. [26] N. Saitou and M. Nei, “The neighbor-joining method: a new
[11] S. Sinchaikul, B. Sookkheo, S. Phutrakul, F. M. Pan, and S. method for reconstructing phylogenetic trees,” Molecular Biol-
T. Chen, “Optimization of a thermostable lipase from Bacillus ogy and Evolution, vol. 4, no. 4, pp. 406–425, 1987.
stearothermophilus p1: overexpression, purification, and charac- [27] K. Tamura, D. Peterson, N. Peterson et al., “MEGA5: molecu-
terization,” Protein Expression and Purification, vol. 22, no. 3, pp. lar evolutionary genetics analysis using maximum likelihood,
388–398, 2001. evolutionary distance, and maximum parsimony methods,”
Molecular Biology and Evolution, vol. 28, no. 10, pp. 2731–2739, Journal of Microbiology and Biotechnology, vol. 27, no. 11, pp.
2011. 2683–2696, 2011.
[28] J. Felsenstein, “Confidence limits on phylogenies: an approach [43] S.-K. Tai, H.-P. Lin, J. Kuo, and J. K. Liu, “Isolation and charac-
using the bootstrap,” Evolution, vol. 39, no. 4, pp. 783–791, 1985. terization of a cellulolytic Geobacillus thermoleovorans T4 strain
[29] K. Tamura, M. Nei, and S. Kumar, “Prospects for inferring from sugar refinery wastewater,” Extremophiles, vol. 8, no. 5, pp.
very large phylogenies by using the neighbor-joining method,” 345–349, 2004.
Proceedings of the National Academy of Sciences of the United [44] D.-W. Lee, Y.-S. Koh, K.-J. Kim et al., “Isolation and character-
States of America, vol. 101, no. 30, pp. 11030–11035, 2004. ization of a thermophilic lipase from Bacillus thermoleovorans
[30] T. Vorderwülbecke, K. Kieslich, and H. Erdmann, “Comparison ID-1,” FEMS Microbiology Letters, vol. 179, no. 2, pp. 393–400,
of lipases by different assays,” Enzyme and Microbial Technology, 1999.
vol. 14, no. 8, pp. 631–639, 1992. [45] Y. R. Abdel-Fattah and A. A. Gaballa, “Identification and
[31] Z. Shu, R. Lin, H. Jiang, Y. Zhang, M. Wang, and J. Huang, over-expression of a thermostable lipase from Geobacillus
“A rapid and efficient method for directed screening of lipase- thermoleovorans Toshki in Escherichia coli,” Microbiological
producing Burkholderia cepacia complex strains with organic Research, vol. 163, no. 1, pp. 13–20, 2008.
solvent tolerance from rhizosphere,” Journal of Bioscience and [46] A. R. Cho, S. K. Yoo, and E. J. Kim, “Cloning, sequencing and
Bioengineering, vol. 107, no. 6, pp. 658–661, 2009. expression in Escherichia coli of a thermophilic lipase from
[32] H. Li and X. Zhang, “Characterization of thermostable lipase Bacillus thermoleovorans ID-1,” FEMS Microbiology Letters, vol.
from thermophilic Geobacillus sp. TW1,” Protein Expression and 186, no. 2, pp. 235–238, 2000.
Purification, vol. 42, no. 1, pp. 153–159, 2005. [47] D. Miñana-Galbis, D. L. Pinzón, J. G. Lorén, A. Manresa, and R.
[33] N. S. Abdul Hamid, H. B. Zen, O. B. Tein, Y. M. Halifah, N. Saari, M. Oliart-Ros, “Reclassification of Geobacillus pallidus (Scholz
and F. Abu Bakar, “Screening and identification of extracellular et al. 1988) Banat et al. 2004 as Aeribacillus pallidus gen. nov.,
lipase-producing thermophilic bacteria from a Malaysian hot comb. nov,” International Journal of Systematic and Evolutionary
spring,” World Journal of Microbiology and Biotechnology, vol. Microbiology, vol. 60, no. 7, pp. 1600–1604, 2010.
19, no. 9, pp. 961–968, 2003. [48] D. L. Pinzón-Martı́nez, C. Rodrı́guez-Gómez, D. Miñana-
[34] V. A. Vargas, O. D. Delgado, R. Hatti-Kaul, and B. Mattiasson, Galbis et al., “Thermophilic bacteria from Mexican thermal
“Lipase-producing microorganisms from a Kenyan alkaline environments: isolation and potential applications,” Environ-
soda lake,” Biotechnology Letters, vol. 26, no. 2, pp. 81–86, 2004. mental Technology, vol. 31, no. 8-9, pp. 957–966, 2010.
[35] J.-L. Reymond and P. Babiak, “Screening systems,” in White [49] M. Heyndrickx, L. Lebbe, M. Vancanneyt et al., “A polyphasic
Biotechnology, R. Ulber and D. Sell, Eds., pp. 31–58, Springer, reassessment of the genus Aneurinibacillus, reclassification
Berlin, Germany, 2007. of Bacillus thermoaerophilus (Meier-Stauffer et al. 1996) as
[36] D. Navarro, M. Couturier, G. G. D. da Silva et al., “Automated Aneurinibacillus thermoaerophilus comb. nov., and emended
assay for screening the enzymatic release of reducing sugars descriptions of A. aneurinilyticus corrig., A. migulanus, and A.
from micronized biomass,” Microbial Cell Factories, vol. 9, p. 58, thermoaerophilus,” International Journal of Systematic Bacteriol-
2010. ogy, vol. 47, no. 3, pp. 808–817, 1997.
[37] T. N. Nazina, T. P. Tourova, A. B. Poltaraus et al., “Taxo- [50] R. N. Allan, L. Lebbe, J. Heyrman, P. De Vos, C. J. Buchanan, and
nomic study of aerobic thermophilic bacilli: descriptions of N. A. Logan, “Brevibacillus levickii sp. nov. and Aneurinibacillus
Geobacillus subterraneus gen. nov., sp. nov. and Geobacillus terranovensis sp. nov., two novel thermoacidophiles isolated
uzenensis sp. nov. from petroleum reservoirs and transfer of from geothermal soils of northern Victoria Land, Antarctica,”
Bacillus stearothermophilus, Bacillus thermocatenulatus, Bacil- International Journal of Systematic and Evolutionary Microbiol-
lus thermoleovorans,” International Journal of Systematic and ogy, vol. 55, no. 3, pp. 1039–1050, 2005.
Evolutionary Microbiology, vol. 51, no. 2, pp. 433–446, 2001. [51] M. B. Stott, M. A. Crowe, B. W. Mountain et al., “Isolation of
[38] G. McMullan, J. M. Christie, T. J. Rahman, I. M. Banat, N. G. novel bacteria, including a candidate division, from geothermal
Ternan, and R. Marchant, “Habitat, applications and genomics soils in New Zealand,” Environmental Microbiology, vol. 10, no.
of the aerobic, thermophilic genus Geobacillus,” Biochemical 8, pp. 2030–2041, 2008.
Society Transactions, vol. 32, no. 2, pp. 214–217, 2004. [52] M. Masomian, R. N. Abd Rahman, A. B. Salleh, and M. Basri,
[39] T. L. Maugeri, C. Gugliandolo, D. Caccamo, and E. Stacke- “A unique thermostable and organic solvent tolerant lipase
brandt, “A polyphasic taxonomic study of thermophilic bacilli from newly isolated Aneurinibacillus thermoaerophilus strain
from shallow, marine vents,” Systematic and Applied Microbiol- HZ: physical factor studies,” World Journal of Microbiology and
ogy, vol. 24, no. 4, pp. 572–587, 2001. Biotechnology, vol. 26, no. 9, pp. 1693–1701, 2010.
[40] T. L. Maugeri, C. Gugliandolo, D. Caccamo, and E. Stacke- [53] Y. R. Abdel-Fattah, “Optimization of thermostable lipase
brandt, “Three novel halotolerant and thermophilic Geobacillus production from a thermophilic Geobacillus sp. using Box-
strains from shallow marine vents,” Systematic and Applied Behnken experimental design,” Biotechnology Letters, vol. 24,
Microbiology, vol. 25, no. 3, pp. 450–455, 2002. no. 14, pp. 1217–1222, 2002.
[41] S. Canakci, K. Inan, M. Kacagan, and A. O. Belduz, “Evaluation [54] S. Kumar, K. Kikon, A. Upadhyay, S. S. Kanwar, and R. Gupta,
of arabinofuranosidase and xylanase activities of Geobacillus “Production, purification, and characterization of lipase from
spp. isolated from some hot springs in Turkey,” Journal of thermophilic and alkaliphilic Bacillus coagulans BTS-3,” Protein
Microbiology and Biotechnology, vol. 17, no. 8, pp. 1262–1270, Expression and Purification, vol. 41, no. 1, pp. 38–44, 2005.
2007. [55] R. Kumar, S. Mahajan, A. Kumar, and D. Singh, “Identification
[42] A. Cihan, B. Ozcan, N. Tekin et al., “Phylogenetic diversity of variables and value optimization for optimum lipase produc-
of isolates belonging to genera Geobacillus and Aeribacillus tion by Bacillus pumilus RK31 using statistical methodology,”
isolated from different geothermal regions of Turkey,” World New Biotechnology, vol. 28, no. 1, pp. 65–71, 2011.
[56] S. Ertuǧrul, G. Dönmez, and S. Takaç, “Isolation of lipase pro- [70] T. C. Leow, R. N. Abd Rahman, M. Basri, and A. B. Salleh, “A
ducing Bacillus sp. from olive mill wastewater and improving its thermoalkaliphilic lipase of Geobacillus sp. T1,” Extremophiles,
enzyme activity,” Journal of Hazardous Materials, vol. 149, no. 3, vol. 11, no. 3, pp. 527–535, 2007.
pp. 720–724, 2007. [71] N. A. Soliman, M. Knoll, Y. R. Abdel-Fattah, R. D. Schmid,
[57] F. M. Shariff, T. C. Leow, A. D. Mukred, A. B. Salleh, M. Basri, and S. Lange, “Molecular cloning and characterization of ther-
and R. N. Abd Rahman, “Production of L2 lipase by Bacillus mostable esterase and lipase from Geobacillus thermoleovorans
sp. strain L2: nutritional and physical factors,” Journal of Basic YN isolated from desert soil in Egypt,” Process Biochemistry, vol.
Microbiology, vol. 47, no. 5, pp. 406–412, 2007. 42, no. 7, pp. 1090–1100, 2007.
[58] A. Ebrahimpour, R. N. Abd Rahman, D. H. Ean Ch’ng, M. [72] J. P. Henley and A. Sadana, “Categorization of enzyme deacti-
Basri, and A. B. Salleh, “A modeling study by response surface vations using a series-type mechanism,” Enzyme and Microbial
methodology and artificial neural network on culture param- Technology, vol. 7, no. 2, pp. 50–60, 1985.
eters optimization for thermostable lipase production from a
newly isolated thermophilic Geobacillus sp. strain ARM,” BMC
Biotechnology, vol. 8, p. 96, 2008.
[59] M. Sifour, T. I. Zaghloul, H. M. Saeed, M. M. Berekaa, and Y.
R. Abdel-fattah, “Enhanced production of lipase by the ther-
mophilic Geobacillus stearothermophilus strain-5 using statisti-
cal experimental designs,” New Biotechnology, vol. 27, no. 4, pp.
330–336, 2010.
[60] C. Breuil, D. B. Shindler, J. S. Sijher, and D. J. Kushner, “Stimu-
lation of lipase production during bacterial growth on alkanes,”
Journal of Bacteriology, vol. 133, no. 2, pp. 601–606, 1978.
[61] S. Takaç and B. Marul, “Effects of lipidic carbon sources on
the extracellular lipolytic activity of a newly isolated strain
of Bacillus subtilis,” Journal of Industrial Microbiology and
[62] H. A. El-Shafei and L. A. Rezkallah, “Production, purifica-
tion and characterization of Bacillus lipase,” Microbiological
Research, vol. 152, no. 2, pp. 199–208, 1997.
[63] M. M. Berekaa, T. I. Zaghloul, Y. R. Abdel-Fattah, H. M. Saeed,
and M. Sifour, “Production of a novel glycerol-inducible lipase
from thermophilic Geobacillus stearothermophilus strain-5,”
World Journal of Microbiology and Biotechnology, vol. 25, no. 2,
pp. 287–294, 2009.
[64] N. G. Edwinoliver, K. Thirunavukarasu, R. B. Naidu, M. K.
Gowthaman, T. N. Kambe, and N. R. Kamini, “Scale up of
a novel tri-substrate fermentation for enhanced production
of Aspergillus niger lipase for tallow hydrolysis,” Bioresource
Technology, vol. 101, no. 17, pp. 6791–6796, 2010.
[65] J. G. S. Mala, N. G. Edwinoliver, N. R. Kamini, and
R.Puvanakrishnan, “Mixed substrate solid state fermentation
for production and extraction of lipase from Aspergillus niger
MTCC 2594,” Journal of General and Applied Microbiology, vol.
53, no. 4, pp. 247–253, 2007.
[66] M. L. Rúa, C. Schmidt-Dannert, S. Wahl, A. Sprauer, and
R. D. Schmid, “Thermoalkalophilic lipase of Bacillus thermo-
catenulatus: large-scale production, purification and properties:
aggregation behaviour and its effect on activity,” Journal of
[67] S. Imamura and S. Kitaura, “Purification and characterization
of a monoacylglycerol lipase from the moderately thermophilic
Bacillus sp. H-257,” Journal of Biochemistry, vol. 127, no. 3, pp.
419–425, 2000.
[68] A. Ebrahimpour, R. N. Abd Rahman, M. Basri, and A. B.
Salleh, “High level expression and characterization of a novel
thermostable, organic solvent tolerant, 1,3-regioselective lipase
from Geobacillus sp. strain ARM,” Bioresource Technology, vol.
102, no. 13, pp. 6972–6981, 2011.
[69] M. Tayyab, N. Rashid, and M. Akhtar, “Isolation and identifi-
cation of lipase producing thermophilic Geobacillus sp. SBS-4S:
cloning and characterization of the lipase,” Journal of Bioscience
and Bioengineering, vol. 111, no. 3, pp. 272–278, 2011.
http://dx.doi.org/10.1155/2013/540465
Research Article
Characterization and Dynamic Behavior of Wild Yeast during
Spontaneous Wine Fermentation in Steel Tanks and Amphorae
Cecilia Díaz,1 Ana María Molina,2 Jörg Nähring,1 and Rainer Fischer1
1
Molecular Biology Division, Fraunhofer Institute for Molecular Biology and Applied Ecology, 57392 Schmallenberg, Germany
2
Facultad de Ingenierı́a y Tecnologı́a, Universidad San Sebastián, 4030000 Concepcion, Chile
Correspondence should be addressed to Cecilia Dı́az; cecilia.diaz@ime.fraunhofer.de
Received 22 January 2013; Revised 28 March 2013; Accepted 9 April 2013
Copyright © 2013 Cecilia Dı́az et al. This is an open access article distributed under the Creative Commons Attribution License,
We studied the dynamic behavior of wild yeasts during spontaneous wine fermentation at a winery in the Valais region of
Switzerland. Wild yeasts in the winery environment were characterized using a PCR-RFLP method. Up to 11 different yeast
species were isolated from the vineyard air, whereas only seven were recovered from the grapes surface. We initially investigated
a cultureindependent method in pilot-scale steel fermentation tanks and found a greater diversity of yeasts in the musts from two
red grape varieties compared to three white grape varieties. We found that the yeasts Metschnikowia pulcherrima, Rhodotorula
mucilaginosa, Pichia kluyveri, P. membranifaciens and Saccharomyces cerevisiae remained active at the end of the fermentation. We
also studied the dynamic behavior of yeasts in Qvevris for the first time using a novel, highlysensitive quantitative real-time PCR
method. We found that non-Saccharomyces yeasts were present during the entire fermentation process, with R. mucilaginosa and P.
anomala the most prominent species. We studied the relationship between the predominance of different species and the output of
the fermentation process. We identified so-called spoilage yeasts in all the fermentations, but high levels of acetic acid accumulated
only in those fermentations with an extended lag phase.
1. Introduction because no additional wine yeasts are introduced into the

process.
Low-intervention winemaking methods based on sponta- Wine flavor is influenced by the large number of
neous fermentation are becoming more popular among yeast species present during spontaneous fermentation [3,
wine producers and consumers [1, 2]. Some wine producers 8–11], including those from the genera Hanseniaspora,
and viticulturists have readopted traditional winemaking Metschnikowia and Candida, and more occasionally Toru-
methods to generate unique attributes that differentiate their laspora and Pichia. Most of these non-Saccharomyces yeasts
products, improve wine quality, and increase the variety of grow during the early fermentation stages, whereas the
complex flavors that characterize regional vineyards. process is eventually completed by Saccharomyces cerevisiae
Spontaneous fermentation is a complex process influ- because it can tolerate higher levels of alcohol and lower levels
enced by many factors, including the endogenous microbial of oxygen [9–15].
flora, the grape variety, climatic conditions, and the winemak- Previous studies have shown that non-Saccharomyces
ing process [3–7]. The outcome of the fermentation process yeasts can be detected throughout the fermentation process
can therefore be difficult to predict and can differ from [15]. They influence the course of fermentation and the char-
year to year. The natural yeast flora, found on grapes and acteristics of the resulting wine by producing extracellular
in wineries, play a significant role during fermentation and enzymes and metabolites of oenological significance that
are particularly important during spontaneous fermentation modify the sensory and organoleptic properties of wine,
introducing a broader spectrum of aromas and flavors [16– Table 1: Grape varieties used for the analysis of dynamic wild
18]. yeast populations during spontaneous fermentation in stainless steel
Microbiology techniques are often used to isolate and tanks and Qvevris.
identify wild yeasts, but this requires different types of Harvest
media and different culture protocols that influence species
which are recovered. The metabolic status of the cells also 2008-2009 (steel tank) 2010-2011 (Qvevris)
results in the presence of viable but nonculturable (VBNC) Pinot Noir (R) Chardonnay (W) Resi (W)
microbes whose influence on fermentation can be underesti- Grape
Cornalin (R) Petit Arvine (W) Ermitage (W)
mated because the population dynamics cannot be evaluated variety
accurately. Quantitative real-time PCR (qPCR) is a faster Gutedel (W)
and more reliable alternative to identify and quantify yeasts (R): red variety; (W): white variety.
during fermentation [19] and is particularly advantageous
for VBNC yeasts because of its sensitivity [20]. Although
the technique cannot distinguish living cells from intact from the inner surface of clean fermentation tanks before
dead cells, it remains the most widely used method for filling them with grape juice, and 1000 L of air inside the
the evaluation of wild yeast dynamics during fermentation cellar was also filtered and the collected residues were plated
because VBNC cells continue to influence wine flavor and as above. All the environmental samples were collected in
palatability regardless of their actual status [21]. triplicate.
We compared microbiology methods (viable counts) Five different grape varieties from the 2008-2009 harvests
and novel molecular biology techniques: polymerase chain (Table 1) were processed by spontaneous fermentation to
reaction/restriction fragment length polymorphism (PCR- determine the predominant yeast species at the different
RFLP) and qPCR for the identification of yeast species, and fermentation stages. Prefermentation steps such as harvesting
we characterized their dynamic behavior during spontaneous and pressing were carried out according to routine winery
wine fermentation in the Valais region of Switzerland in procedures. Pressed berries were fermented with the skin
the 2008–2011 harvests. We used these new methods to to make red wine, or were clarified before fermentation to
identify the predominant species present during spontaneous produce white wine. Duplicate fermentations were carried
fermentation, establishing a standard for the semiquantitative out in the winery cellar using new 110-L stainless steel tanks,
detection of yeasts with antibodies in a biochip format. Such a without starting yeast cultures. Liquid samples (50 mL, in
device would allow winemakers to make early decisions about duplicate) from the fermenting musts were collected daily,
the suitability of grapes and the likely success of spontaneous frozen immediately at −20∘ C, and stored in the dark prior
fermentation. Also for the first time, we studied the dynamic to analysis. Immediately after defrosting, liquid samples were
behavior of wild yeasts during spontaneous fermentation in centrifuged at 4000 rpm for 5 min. The supernatant was
Qvevris (amphora-like clay vessels), the use of which is an tested for chemical parameters, and the pellet was resus-
emerging trend among European winemakers. pended in 100 𝜇L distilled water, plated on RBCA medium
and incubated at 30∘ C for 3–7 days, and then stored at 4∘ C
2. Materials and Methods prior to analysis.
2.1. Samples. Grapes and must samples were collected from 2.3. Fermentation Parameters. The fermentations were mon-
the vineyards of the winery Albert Mathier et Fils S.A., in itored by measuring glucose/fructose consumption and
Salgesch, Valais, Switzerland, during the 2008–2011 harvest ethanol formation during fermentation, and the acetic acid
seasons. The grape varieties we studied are listed in Table 1. content at the end of fermentation. The parameters were
Samples were collected in situ and frozen at −20∘ C during determined by spectrophotometry at 20 ± 1∘ C using the D-
transport prior to analysis. Glucose/D-Fructose, Ethanol, and Acetic Acid enzymatic kits
provided by R-Biopharm (Germany), according to the manu-
2.2. Isolation of Yeasts from the Winery Environment and facturer’s instructions. Standards and controls were provided
Fermentation Samples. During 2008 and 2009, we screened in the kits. All measurements (duplicate fermentations) were
yeasts present in the winery environment (i.e., the vineyard, taken in triplicate.
winery facilities and cellar) and in the fermenting wine musts.
In the vineyard, grape berries were placed in direct contact 2.4. Identification of the Predominant Yeast Species. A ter-
with plates containing Rose Bengal Chloramphenicol Agar minal restriction fragment length polymorphism (T-RFLP)
(RBCA), a selective medium for yeasts and molds (15 g L−1 method was developed and optimized for yeast identification,
agar, 10 g L−1 glucose, 5 g L−1 papain-digested soybean meal, based on restriction patterns generated from the genomic
1 g L−1 KH2 PO4 , 0.5 g L−1 MgSO4 × 7H2 O, 0.05 g L−1 Rose region spanning the internal transcribed spacers (ITS1 and
Bengal, and 10 g L−1 chloramphenicol). We pumped 1000 L of ITS2) and the 5.8S rRNA gene. These regions show low
vineyard air surrounding the grapes through a Millipore M intraspecific polymorphism and high interspecific variability
Air Tester T (Millipore, USA) and plated the collected residue and have previously been shown to distinguish 26 yeast
on RBCA as above. We also sampled environmental yeast species found on grapes, in cellars and/or in wine musts
flora from the winery facilities. Contact samples were taken [22, 23].
Total DNA from the isolated colonies was extracted agar (10 g L−1 yeast extract, 20 g L−1 peptone, 20 g L−1 glucose,
using the First-Beer Magnetic DNA kit (GEN-IAL GmbH, and 0.1 mg L−1 chloramphenicol) at 27∘ C for 24 h. The cells
Troisdorf, Germany) and amplified using primers ITS-5 (5󸀠 - were counted using a Neubauer chamber. DNA was extracted
GGA AGT AAA AGT CGT AAC AAG G-3󸀠 ) and ITS-4 (5󸀠 - using the First-Beer Magnetic DNA kit and serially diluted
TCC TCC GCT TAT TGA TAT GC-3󸀠 ) followed by a second (1 : 10) from 107 -108 down to 1 cell mL−1 . Each point on the
round of amplification with the nested primers ITS-1 (5󸀠 -TCC calibration curve was measured in duplicate. Conventional
GTA GGT GAA CCT GCG G-3󸀠 ) and ITS-2 (5󸀠 -GCT GCG and real-time PCR was carried out using a range of yeast
TTC TTC ATC GAT GC-3󸀠 ) [24]. The first reaction mixture species to verify the specificity of each primer set.
comprised 200 𝜇M of each dNTP, 10x PCR buffer, 0.5 𝜇M of
each primer, 1 𝜇L of extracted yeast DNA and 1.25 U Hot Start
2.6. Wild Yeast Dynamics during Spontaneous Fermentation in
Polymerase (5 Prime, Hamburg, Germany) in a total volume
Qvevris. Spontaneous fermentation in Qvevris was studied
of 25 𝜇L. The samples were amplified in a thermocycler
during the 2010 and 2011 harvest seasons. The white grape
(VRW, Pennsylvania, USA) by denaturing at 95∘ C for 3 min
varieties Resi and Ermitage (Table 1) were harvested, crushed,
followed by 15 cycles of denaturing at 95∘ C for 30 s, annealing
and fermented in 1500-L Qvevris without clarification. We
at 57∘ C for 30 s and extension at 72∘ C for 1 min, and a final
took 50-mL samples in triplicate at 2-3-day intervals through-
extension step at 72∘ C for 5 min. The nested amplification
out fermentation; that is, every time the Qvevris were opened
mixture comprised 200 𝜇M of each dNTP, 10x PCR buffer,
to stir the must. The samples were frozen immediately at
0.5 𝜇M of each primer (labeled if necessary for product size
−20∘ C and stored in the dark prior to analysis. DNA was
determination, see below), 2.5 U Hot Start Polymerase (5
extracted from the must using a modified CTAB method
Prime, Hamburg, Germany), and 0.5 𝜇L template DNA (from
[25] in which 10 mL samples were centrifuged for 1 min at
the first-round PCR) in a total volume of 50 𝜇L. The mixture
3000 rpm to sediment the skin and seeds before the standard
was denatured at 95∘ C for 3 min then amplified by 20 cycles
protocol was applied. The extracted DNA was then tested
of denaturing at 95∘ C for 30 s, annealing at 62∘ C for 30 s and
by qPCR to identify the wild yeast species present during
extension at 72∘ C for 1 min, followed by a final extension
spontaneous fermentation as discussed above.
at 72∘ C for 5 min. The products were digested with BstYI
(New England BioLabs, Ipswich) at 60∘ C for 1 h. The length of
the terminal fragment was determined using a 3130 Genetic 3. Results
Analyzer (Applied Biosystems, Darmstadt, Germany) prior
to the purification of the samples using the Cycle Pure Kit 3.1. Establishing a PCR-RFLP Method for Yeast Identification.
(Omega Bio-tek, USA). Yeast genomic DNA was amplified using primers ITS4
and ITS5 (first round), and the products were amplified
with the nested primers ITS1 and ITS2. The sizes of both
2.5. Primer Design and Real-Time PCR. Primers specific the digested and undigested PCR products are unique to
for the 10 predominant yeasts found in the winery and particular yeast genera and also allow the differentiation of
in fermentation samples during the 2008-2009 harvest certain species, resulting in the unambiguous identification
season were designed to anneal within the 26S rDNA of up to 28 species (Table 3). There were only three cases in
region and amplify products 150–200 bp in length (Table 2). which we were unable to distinguish two different species:
Each primer pair was designed by processing available (1) Hanseniaspora guilliermondii and H. uvarum; (2) Saccha-
sequences using CLC Combined Workbench 3 Software romyces bayanus and S. pastorianus; (3) Dekkera bruxellensis
(CLC-Bio, Denmark), and the properties of each primer and Cryptococcus flavus. The method was optimized using
were verified using Primer Tool (Sigma-Aldrich, USA; http:// species obtained from the Deutsche Sammlung von Mikroor-
www.sigmagenosys.com/calc/DNACalc.asp). The specificity ganismen und Zellkulturen GmbH (DSMZ), Braunschweig,
of each primer pair was controlled by searching GenBank Germany. Even so, wild yeast species in wineries are often
using BLAST (http://www.ncbi.nlm.nih.gov /BLAST/). local subspecies that are subject to different environmental
Real-time PCR was carried out using an ABI 7300 Real- selection conditions and their sequences and PCR product
Time PCR System (Applied Biosystems, Hitachi, Japan). sizes can differ slightly from purchased strains. Therefore,
Each reaction comprised 7.5 𝜇L Platinum SYBR Green qPCR and in order to validate the method, we selected 4 isolated
SuperMix-UDG (Bio-Rad, Hercules, CA, USA), 200 nM of yeasts and sequenced the first-round PCR products (NCBI
each primer (Metabion, Germany), and 0.3 𝜇L template DNA accession numbers KC869927, KC869928, KC869929, and
extracted from must, in a total volume of 15 𝜇L. The mixture KC869930) to compare these empirical sequences to those in
was heated to 50∘ C for 2 min and then 95∘ C for 2 min, GenBank by using the empirical sequences as BLAST queries.
followed by 40 cycles of denaturation at 95∘ C for 15 s, and
annealing/extension at 60–63∘ C (depending on the primers) 3.2. Natural Flora in Vineyard and Cellar Environments.
for 45 s. The cycling temperature was then increased by 0.3∘ C Yeasts naturally present in the vineyard environment were
every 10 s from 63 to 95∘ C to obtain the melting curve. The isolated from the grape surface and from the air around the
DNA concentration in the samples was limited to 50 ng per grapes using culture-dependent methods (see Section 2.2).
analysis, except for standard curves prepared from samples During 2008 and 2009, up to 11 different yeast species could
containing a known number of yeast cells. All yeast species be isolated from the vineyard air although Bulleromyces albus
were cultivated in Yeast Extract Peptone Dextrose (YPD) and Sporidiobolus pararoseus were the only species recovered
Table 2: Specific primers used for qPCR analysis.
Yeast species Primer name Primer sequence Product size Annealing temperature ∘ C
CZ-5fw 5 -CGATGAGATGCCCAATTCCA-3󸀠
󸀠
Candida zeylanoides 191 bp 58
CZ-3bw 5 -GAAGGGAACGCAAAATACCAA-3󸀠
󸀠
ZF-5fw 5󸀠 -CTTGAGCTCCTTGTAAAGC-3󸀠
Zygosaccharomyces florentinus 256 bp 55
ZF-3bw 5󸀠 -CTAGGTTTTCTGCTGCCG-3󸀠
MP-5fw 5󸀠 -CAACGCCCTCATCCCAGA-3󸀠
Metschnikowia pulcherrima 253 bp 60
MP-3bw 5󸀠 -AGTGTCTGCTTGCAAGCC-3󸀠
WS-5fw 5󸀠 -GGGTGTCCAGTGCTTTG-3󸀠
Williopsis saturnus 199 bp 56
WS-3bw 5 -CCCAAGAAGGGAAGATAATCAC-3󸀠
󸀠
PK-5fw 5󸀠 -AGTCTCGGGTTAGACGT-3󸀠
Pichia kluyveri 169 bp 55
PK-3bw 5󸀠 -GCTTTTCATCTTTCCTTCACA-3󸀠
RM-5fw 5󸀠 -GCGCTTTGTGATACATTTTC-3󸀠
Rhodotorula mucilaginosa 169 bp 54
RM-3bw 5󸀠 -CCATTATCCATCCCGGAAAA-3󸀠
PANG-5fw 5󸀠 -GTGTCCATTTCCGTGTAAGA-3󸀠
Pichia angusta 175 bp 56
PANG-3bw 5󸀠 -AGCCCACCCACAAG-3󸀠
PA-5fw 5 -ACGTCATAGAGGGTGAGAAT-3󸀠
󸀠
Pichia anomala 197 bp 57
PA-3bw 5 -AAACACCAAGTCTGATCTAATG-3󸀠
󸀠
CG-5fw 5󸀠 -GAGGGTGTCAGTTCTTTGT-3󸀠
Candida glabrata 224 bp 56
GC-3bw 5󸀠 -GTGAGCTGCGAGAGTC-3󸀠
HU-5fw 5󸀠 -GGCGAGGGATACCTTTTCTCTG-3󸀠
Hanseniaspora uvarum 172 bp 59
HU-3bw 5󸀠 -GAGGCGAGTGCATGCAA-3󸀠
PF-5fw 5 -TTGCCTATGCTCTGAGGCC-3󸀠
󸀠
Pichia fermentans 170 bp 61
PF-3bw 5󸀠 -TCCATGTCGGGCGCAAT-3󸀠
SC-5fw 5 -AGGAGTGCGGTTCTTTCTAAAG-3󸀠
󸀠
Saccharomyces cerevisiae 215 bp 59
SC-3bw 5󸀠 -TGAAATGCGAGATTCCCCCA-3󸀠
TD-5fw 5󸀠 -GTGGCGAGGATCCCAG-3󸀠
Torulaspora delbrueckii 186 bp 58
TD-3bw 5󸀠 -CTATCGGTCTCTCGCAA-3󸀠
in both years (Table 4). We recovered seven yeast species from varieties Pinot Noir and Cornalin, which contained 12 and
the grape surface, and species appear to be dependent on 9 different yeast species, respectively, and the white varieties
the variety and the year of harvest (Table 4). Aurebasidium Gutedel, Chardonnay, and Petite Arvine, which contained 12,
pullulans, Cryptococcus magnus, Rhodotorula mucilaginosa, 12, and 7 yeast species, respectively (Table 5).
and Zygosaccharomyces florentinus were the only species Most of the yeast species we identified were present
isolated from both the air and the grape surface. Most of in more than one of the musts (Table 5). M. pulcherrima,
the yeasts isolated from the vineyard air were also present S. cerevisiae, S. bayanus, and T. delbrueckii were found
in the grape juice at the beginning of fermentation. In the in all five musts at some point during fermentation. Six
cellar environment, yeasts were isolated from the surface of yeast species were only found in one type of must, and
clean and empty barrels (i.e., before filling the fermentation only at the beginning of fermentation. Four yeast species
tanks with the grape must) and from the air inside the cellar found in Gutedel musts did not grow in any of the other
room. In 2008, four different species were isolated from musts: Bulleromyces albus, Candida zeylanoides, Cryptococcus
the cellar air and three from the clean fermentation tank flavus/Dekkera bruxellensis (the latter could not be distin-
(Table 4), whereas in 2009 only one species (R. mucilaginosa) guished on the basis of their PCR-RFLP patterns), and Filoba-
was isolated from the clean fermentation tank. All the species sidium floriforme. Similarly, Pichia burtonii and P. holstii only
in the cellar environment were also found in the vineyard, and found in Cornalin musts (Table 5). There were no species
all species identified in the cellar environment were also later associated exclusively with red or white grape varieties.
found in the fermenting must. The composition of the yeast populations also changed
significantly during fermentation. Initially, 7–12 different
species were found in the musts (depending on the variety),
3.3. Yeast Flora in Steel-Tank Fermentations. Changes in the but this declined to 1–5 species by the midfermentation, when
composition of the yeast population during spontaneous nitrogen becomes limiting and the ethanol concentration
fermentation in steel tanks were measured using culture- begins to increase rapidly (Table 5). By the end of fermenta-
dependent methods. We investigated the musts of five grape tion, only six different yeast species could be recovered from
varieties during the 2008 and 2009 harvest seasons: the red the musts. The ethanol-resistant strain S. bayanus made up
Table 3: Sizes of digested and undigested nested PCR products 100

representing different yeast species derived using the T-RFLP 90
method.
80
Yeast contribution (%)

Size (bp) 70
Yeast species
Nested PCR Digested 60
Aureobasidium pullulans∗ 254 — 50
Botryotinia fuckeliana∗ — 223 40
Bulleromyces albus 195 164 30
Candida glabrata 478 224 20
Candida zemplinina 200 10
Candida zeylanoides 269 238 0
Pinot Noir Cornalin Chardonnay Petit Arvine Gutedel
Cryptococcus flavus 169 134
Cryptococcus magnus 211 — Metschnikowia pulcherrima Saccharomyces bayanus
Dekkera bruxellensis 168 134 Saccharomyces cerevisiae Pichia klyveri
Epicoccum nigrum∗ 216 — Rhodotorula mucilaginosa Pichia membranifaciens
Filobasidium floriforme 237 204 Figure 1: Yeast population recovered at the end of spontaneous
Hanseniaspora guilliermondii 364 66 fermentation in steel tanks during the 2008 and 2009 harvest
Hanseniaspora uvarum 365 66 seasons.
Hyphopichia burtonii 159 124
Issatchenkia orientalis 178 148 as well as acetic acid production (Table 6). The red varieties
Kluyveromyces lactis 305 273 Pinot Noir and Cornalin reached dryness (less than 4 g L−1 of
Metschnikowia pulcherrima 140 107 total sugar) 6–11 days after pressing (Figure 2). The lag phase
Metschnikowia sp. 348 — of the Pinot Noir fermentations in 2008 (i.e., the period before
Pichia angusta 379 351 glucose consumption increases rapidly) was relatively long
compared to the Cornalin fermentations in the same year (2-
Pichia anomala 258 228
3 days) (Figure 2). In contrast, the white grape varieties failed
Pichia fermentans 156 122 to reach dryness in fermentations during 2008 and 2009, and
Pichia holstii — 248 the Petite Arvine and Chardonnay vessels contained high
Pichia kluyveri 162 128 levels of residual sugar at the end of fermentation (Figure 2).
Pichia membranifaciens 165 130 In 2008, the fermentation of Gutedel grapes was delayed at
Rhodotorula mucilaginosa 229 198 the midexponential phase (days 6–13) whereas Petite Arvine
was characterized by sluggish fermentation from the late
Saccharomyces bayanus 435 405
exponential phase (day 11) onwards (Figure 2).
Saccharomyces cerevisiae 441 410
Saccharomyces pastorianus 435 405
3.4. Real-Time PCR. Thirteen pairs of specific primers were
Sporidiobolus pararoseus 222 197 designed for the rapid identification and quantification of
Torulaspora delbrueckii (wild yeast the yeast species we detected. The primers designed for Z.
370 340
isolate) florentinus, C. glabrata, and P. fermentans showed evidence
Williopsis saturnus 252 134 of nonspecific annealing and were therefore eliminated from
Zygosaccharomyces florentinus 244 213 the study. The sequences and annealing temperatures of the
∗ remaining primers are summarized in Table 2. The melt curve
Molds.
analysis for each PCR showed a single peak (data not shown).
Standard curves were established for each pair of primers. The
a substantial proportion of the yeasts in all musts (Figure 1) reaction efficiencies ranged between 72.54% (P. anomala) and
and was the only strain detected in Gutedel musts during the 98.68% (S. cerevisiae) with high reproducibility. The lowest
mid- and late fermentation stages. In contrast, S. cerevisiae detection limit was 102 cells L−1 .
was found in the Chardonnay, Pinot Noir, and Petite Arvine
musts at the end of fermentation, and M. pulcherima was 3.5. Yeast Flora in Qvevri Fermentations. The dynamic behav-
present in the Chardonnay, Pinot Noir, and Cornalin musts ior of the yeast populations in Qvevri spontaneous fermen-
at the end of fermentation. The other species retrieved at tations was monitored by qPCR during the 2010 and 2011
the end of the fermentation were P. klyveri (Chardonnay and harvest seasons. There was a slight tendency towards higher
Cornalin musts), P. membranifaciens (Chardonnay must), yeast diversity in the Resi variety compared to Ermitage, with
and R. mucilaginosa (Pinot Noir must) (Figure 1). 10 and 8 different yeast species, respectively (Table 7). Most
The progress of fermentation was monitored by measur- of the species were present in varieties, and M. pulcherima,
ing sugar consumption and ethanol production (Figure 2), R. mucilaginosa, P. anomala, H. uvarum, S. cerevisiae, and
300 300
250 250
Concentration (g L−1 )
200 200
150 150
100 100
50 50
0 0
0 5 10 0 5 10 15
Day Day
Total sugar 09 Ethanol 09 Total sugar 09 Ethanol 09

(a) (b)
300 300
250 250
200 200
150 150
100 100
50 50
0 0
0 5 10 15 20 0 5 10 15 20
Day Day
(c) (d)
250
200
150
100
50
0
0 5 10 15 20
Day
Total sugar 09 Ethanol 09
Total sugar 08 Ethanol 08
(e)
Figure 2: Spontaneous fermentation profile, expressed in g L−1 , for the white grape varieties in stainless steel tanks during the 2008 and 2009
harvest seasons. (a) Pinot Noir; (b) Cornalin; (c) Chardonnay; (d) Petit Arvine; (e) Gutedel.
Table 4: Yeasts isolated from the winery environment during the 2008 and 2009 harvest seasons.
Air vineyard Air cellar Tank surface Grape surface∗

Yeast species
2008 2009 2008 2009 2008 2009 2008 2009
Aureobasidium pullulans + Gu Gu
Bulleromyces albus + +
Candida zemplinina +
Cryptococcus magnus + Gu
Filobasidium floriforme Ch Ch, Pa, Gu
Hanseniaspora uvarum + +
Metschnikowia pulcherrima + +
Pichia angusta
Pichia anomala + + Co
Pichia kluyveri + +
Rhodotorula mucilaginosa + + + Pn Pa, Gu
Sporidiobolus pararoseus + +
Williopsis saturnus Pa
Zygosaccharomyces florentinus + + + Ch Co
∗
Varieties of grapes and musts: Pinot Noir (Pn); Cornalin (Co); Chardonnay (Ch); Petite Arvine (Pa); Gutedel (Gu).
Table 5: Yeasts found during spontaneous fermentation in stainless steel tanks, during the 2008 and 2009 harvest seasons.
Grape variety
Yeast species
Pinot Noir Cornalin Chardonnay Petit Arvine Gutedel
Bulleromyces albus a
Candida zemplinina a a; b a a
Candida zeylanoides a
Candida Krusei or Issatchenkia orientalis a a
Cryptococcus flavus or Dekkera bruxellensis a
Filobasidium floriforme a
Hanseniaspora uvarum a; b a a
Metschnikowia pulcherrima a; b; c a; b; c a; b; c a a
Pichia anomala a a a
Pichia burtonii a
Pichia holstii a
Pichia kluyveri a a; c a a
Pichia membranifaciens a; b; c a
Rhodotorula mucilaginosa c a a a
Saccharomyces bayanus a; b; c a, b; c a; b; c a, b; c a, b; c
Saccharomyces cerevisiae a; b; c a a; b; c a; c a
Sporidiobolus pararoseus b a
Torulaspora delbrueckii a a a a a
Zygosaccharomyces florentinus a a
a: detected at the beginning of the fermentation; b: detected during log phase; c: detected during stationary phase.
T. delbrueckii were also found at every fermentation stage fermentations during 2010 and 2011 (Table 7), and in both
(Figures 3 and 4). In contrast, C. zemplinina and P. angusta cases the dominant species in the must before fermentation
were found only in the Resi variety, and although P. kluyveri were R. mucilaginosa, and P. anomala although they were
was found in both varieties, it was present only at certain more abundant in 2011 (Figure 3). The less-abundant species
fermentation stages during the 2010 harvest and was not were H. uvarum, S. cerevisiae, T. delbrueckii, and C. zem-
detected in 2011 (Table 7). plinina, although all of them were present throughout the
R. mucilaginosa was the dominant species in the 2011 fermentation. These species were 10 times more abundant
Ermitage fermentations whereas P. anomala was the domi- in 2010 than in 2011, except R. mucilaginosa, which was
nant species in the Resi fermentations in both harvest years. more abundant in 2011. S. cerevisiae was the most abundant
Up to eight yeast species were detected in the Ermitage species at the beginning of the 2010 fermentations and it
Beginning End Beginning End

1 × 107 1 × 107
1 × 106 1 × 106
1 × 105 1 × 105
Cells (mL−1 )
Cells (mL−1 )
1 × 104 1 × 104
1 × 103 1 × 103
1 × 102 1 × 102
1 × 101 1 × 101
1 3 5 7 9 11 13 15 1 3 5 7 9 11 13 15
Day Day
M. pulcherrima S. cerevisiae M. pulcherrima S. cerevisiae

H. uvarum P. anomala H. uvarum P. anomala
R. mucilaginosa R. mucilaginosa
(a) (b)
Figure 3: Dynamic behavior of wild yeast populations during the spontaneous fermentation of Ermitage grapes in Qvevris, measured by
qPCR: (a) 2010 harvest; (b) 2011 harvest.
Beginning End Beginning End

1 × 107 1 × 107
1 × 105 1 × 105
Cells (mL−1 )
Cells (mL−1 )
1 × 103 1 × 103
1 × 101 1 × 101
0 3 6 9 12 15 0 1 2 3 4 5 6
Day Day
M. pulcherrima S. cerevisiae M. pulcherrima S. cerevisiae
H. uvarum P. anomala H. uvarum P. anomala
R. mucilaginosa R. mucilaginosa
(a) (b)
Figure 4: Dynamic behavior of wild yeast populations during the spontaneous fermentation of Resi grapes in Qvevris, measured by qPCR:
(a) 2010 harvest; (b) 2011 harvest.
proliferated rapidly, reaching its maximum concentration 2010 than 2011 although the onset of fermentation in 2011
(1 × 106 cells mL−1 ) by day 6. In contrast, R. mucilaginosa was more rapid, beginning after 1 day (Figure 4). In 2010,
was the most abundant species in the 2011 fermentations the highest concentration of S. cerevisiae (1 × 106 cells mL−1 )
(1 × 106 cells mL−1 ) and S. cerevisiae proliferated more slowly, was achieved 2 days after fermentation began, whereas in
reaching its maximum concentration after 14 days. In 2010, 2011 the concentration increased rapidly during the first day
the yeast population declined slowly during fermentation and remained high until the end of the fermentation (1 ×
whereas in 2011 the P. anomala, S. cerevisiae, and R. mucilagi- 106 cells mL−1 ).
nosa populations remained high (Figure 3). The progress of the Ermitage and Resi fermentations was
The Resi fermentations during 2010 and 2011 began monitored and compared. The onset of the Ermitage fer-
rapidly (before 5 days in both cases) with P. anomala mentation took longer in 2010 but was nevertheless complete
dominating throughout fermentation and H. uvarum and T. after 14 days in both 2010 and 2011 (Figure 5). The Ermitage
delbrueckii present at lower levels (Figure 4). The concentra- fermentations did not reach dryness by day 14 in 2011, and
tion of yeast, including S. cerevisiae, was slightly higher in the ethanol content was lower than in the 2010 fermentation,
150 150
125 125
Concentration (gL−1 )
Concentration (gL−1 )
100 100
75 75
50 50
25 25
0 0
0 5 10 15 20 0 5 10 15
Day Day
Ethanol 10 Ethanol 11 Glucose 10 Glucose 11
Glucose 10 Glucose 11 Ethanol 10 Ethanol 11
(a) (b)
Figure 5: Spontaneous fermentation profile in Qvevris during the 2010 and 2011 harvests: (a) Ermitage; (b) Resi.
Table 6: Acetic acid production in the spontaneous fermentations.
Fermentation vessel Variety Harvest season Acetic acid (mg L−1 )

2008 65.61 ± 2.5
Pinot Noir
2009 154.69 ± 23.48
2008 150.71 ± 18.62
Cornalin
2009 145.0 ± 7.07
2008 154.64 ± 91.43
Steel tanks Chardonnay
2009 379.15 ± 0
2008 131.64 ± 76.88
Petite Arvine
2009 280.0 ± 0
2008 24.93 ± 0
Gutedel
2009 94.0 ± 39.0
2010 180.0 ± 0
Resi
2011 270.0 ± 10.0
Qvevri
2010 160.0 ± 30.0
Ermitage
2011 650 ± 20.0
concomitant with the production of significant amounts of yeasts that were present in the winery environment and in
acetic acid (Table 6). No measurements were taken beyond wine musts undergoing spontaneous fermentation in steel
day 14 because the Ermitage and Resi wines were blended tanks, thus favoring the detection and proliferation of some
at this stage. The Resi fermentations become more rapidly in yeast species over others. Rose Bengal Chloramphenicol Agar
2011 than in 2010, beginning on the same day (or shortly after) (RBCA) medium was chosen instead of Sabouraud medium
the Qvevris were filled. However the fermentation process because the latter favored mold growth over yeasts (data not
reached dryness in both years. The Ermitage must took longer shown). Freezing the samples prior to analysis may have
to begin fermentation than Resi, starting 2 and 10 days later reduced the viability of the yeast although it is thought that
in 2010 and 2011, respectively. this is a minor effect [26, 27]. Therefore, we acknowledge
that yeast species present in low numbers are unlikely to be
4. Discussion detected using this method, whereas abundant species are
more likely to be recognized. Thus, only seven yeast species
4.1. Isolation and Identification of Predominant Yeast Species. were isolated from the grape surface (A. pullulans, C. magnus,
The isolation media we used enabled us to select different F. floriforme, R. mucilaginosa, W. saturnus, and Z. florentinus),
Table 7: Yeasts identified by qPCR during spontaneous fermentations in Qvevris during the 2010 and 2011 harvest seasons.
Resi Ermitage
Yeast species
2010 2011 2010 2011
Candida zemplinina a; b; c b
Metschnikowia pulcherima a; b; c a; b; c a; b; c a; b; c
Williopsis saturnus a; c a b; c
Pichia kluyveri a b
Rhodotorula mucilaginosa a; b; c a; b; c a; b; c a; b; c
Pichia angusta b c
Pichia anomala a; b; c a; b; c a; b; c a; b; c
Hanseniaspora uvarum a; b; c a; b; c a; b; c a; b; c
Saccharomyces cerevisiae a; b; c a; b; c a; b; c a; b; c
Torulaspora delbrueckii a; b; c a; b; c a; b; c a; b; c
a: detected at the beginning of the fermentation; b: detected during log phase; c: detected during stationary phase.
with A. pullulans and R. mucilaginosa previously reported the yeast populations through the different stages of fermen-
as colonizers of the grape surface [28]. A further 11 yeast tation in steel tanks also differed among grape varieties. The
species were isolated from vineyard air samples, all of detection of some yeast species only during the later stages
which had previously been detected in winery environmental of fermentation probably reflects their proliferation to cell
samples [28, 29]. The anamorphic yeast Kloeckera apiculata, numbers above the detection threshold of our assay rather
previously reported as the predominant yeast species on the than their genuine absence at the beginning of fermentation.
grape surface and in air samples [28], was not found in our The relative greater diversity of yeast species in red compared
investigation but instead the teleomorphic species H. uvarum to white wines is consistent with the higher pH of red wines,
was found in our environmental samples. providing favorable conditions for yeast growth [34]. In white
We found that differences in yeast diversity were often wines, yeasts isolated from the grape skin were not found in
dependent on the grape variety. This phenomenon can be the must, probably because they remained in the skin fraction
attributed to several factors, including the different stages during clarification, and this may also have contributed to the
of berry ripening at harvest, physical damage to the grape lower species diversity we observed.
surface, and pest management practices [29]. Although we The higher yeast diversity during the early stages of
studied different grape varieties grown in the same area and fermentation predominantly reflects the low ethanol toler-
processed at the same winery, microclimatic conditions and ance of non-Saccharomyces species [3, 9, 10, 17, 35, 36].
viticultural practices may have influenced the yeast diversity Nevertheless, we found that non-Saccharomyces yeasts such
we detected. as P. klyveri, P. membranifaciens, R. mucilaginosa, and M.
Most of the yeasts isolated from the vineyard air were also pulcherima were active in the late fermentation stages in
present in the grape juice at the beginning of fermentation. some must varieties. This is consistent with previous reports
All the yeasts identified in the cellar were also found later of ethanol tolerance in M. pulcherima [10, 35, 37], but R.
in the fermenting must. R. mucilaginosa was found in air mucilaginosa is usually found during the early stages of
samples from both the vineyard and the cellar, and on the fermentation, and its presence along with the Pichia species
grape surface, but not on the tank surface. During 2008, Z. later in fermentation could add complexity but also reduce
florentinus was the only species found in all environmental the wine quality [34, 38].
samples (air and contact samples, from both the vineyard and Considering the results from the 2008 and 2009 harvests
the cellar). together, we observed that the generally higher yeast diver-
The viable counts of the environmental samples showed sity in the must at the beginning of the fermentation was
the presence of only non-Saccharomyces species. Although coincident with the rapid onset of the exponential phase.
S. cerevisiae and related species such as S. bayanus are We evaluated the interrelation between the yeast species
predominantly responsible for fermentation, they represent and the success of fermentation. We found that despite the
only a small fraction of the diversity we identified, which diversity of yeasts in red and white varieties, white musts
is consistent with other reports showing that S. cerevisiae is generally contained higher residual sugar levels than red
rarely isolated from natural sources such as berry and leaf musts and that sluggish fermentation was more likely. Such
surfaces when using viable count methods [30–33]. The small fermentations were characterized by the initial predominance
number of species isolated from the cellar environment (air of C. zemplinina and S. bayanus, as well as lower levels of
and tank surface) during 2009 compared to 2008 may have M. pulcherima and S. cerevisiae, contrasting with the red
been caused by the sanitary conditions adopted by the winery wine musts. The impact of these properties on fermentation
after the sampling results in 2008. The dynamic behavior of reflects the better performance of S. cerevisiae compared with
the lower fructose uptake capacity of S. bayanus [39], which further experiments to determine the influence of Qvevris on
is consistent with our results. spontaneous fermentation. Comparative studies with steel-
tank fermentations, using the same raw materials (grape
4.2. Dynamic Behavior of Wild Yeasts during Spontaneous variety and harvest year), should be carried out to investigate
Fermentation in Qvevris. We developed a novel qPCR the impact of Qvevris in more detail.
method for the rapid, sensitive, and culture-independent
detection of yeast species throughout fermentation, revealing 5. Conclusions
that the non-Saccharomyces yeast R. mucilaginosa and P.
anomala dominated the final stages of spontaneous fermen- The predominant yeasts found in the winery (i.e.,
tation in Qvevris. These results are important because non- Metschnikowia pulcherima, Rhodotorula mucilaginosa,
Saccharomyces yeasts can influence the flavor and quality of and some Pichia species) were used as a basis for the
wine in both positive and negative ways [40–42] despite their development of an antibody chip for the identification and
metabolic activity and abundance [19, 20, 43]. semiquantitative detection of wild yeast. The effect of the
The diversity of the yeast species was variety dependent initial yeast concentration and the berry/must temperature
and vintage dependent, with C. zemplinina and P. angusta on the length of the fermentation lag phase and thus the
present only in the variety of Resi, and P. kluyveri present quality of spontaneous fermentation will be investigated
in both wines but only during the 2010 harvest. H. uvarum in more detail to improve the performance of this device.
has previously been identified as the predominant species We have also provided the first quantitative evidence
during the early stages of fermentation [9–11, 35] but we describing the dynamic behavior of yeast populations during
found no evidence for this species on the grape surface spontaneous fermentation in amphora vessels.
(viable cell count method) and found it was less prevalent
during amphora fermentations (qPCR method). In con- Conflict of Interests
trast, R. mucilaginosa was found to be abundant in both
the amphora and steel-tank fermentations using qPCR and The authors declare that there is no conflict of interests.
culture-dependent methods, respectively.
The Resi fermentations commenced almost immediately Acknowledgments
in 2011, even though similar numbers of yeast cells were
present at the beginning of fermentation in both years, The authors thank Albert Mathier et Fils winery, especially
and S. cerevisiae was less abundant in 2011 than 2010. The the owner (Amédée Mathier) and oenologist (Fadri Kuonen)
minimal lag phase and rapid fermentation (completed in 3 for kindly providing the samples for testing. This work was
days) could be explained by the climatic conditions in the funded by the Fraunhofer Institute.
weeks prior to harvest, which increased the temperature of
the berries and the must after crushing (data not shown),
favoring the rapid proliferation of S. cerevisiae. This sug- References
gests that berry temperature before pressing could play a [1] M. R. Provenzano, H. El Bilali, V. Simeone, N. Baser, D.
key role in the success of spontaneous fermentation. We Mondelli, and G. Cesari, “Copper contents in grapes and wines
tested this hypothesis by studying parallel fermentations. from a Mediterranean organic vineyard,” Food Chemistry, vol.
Ermitage fermentations underwent a longer lag phase in 122, no. 4, pp. 1338–1343, 2010.
2011 (11 days) than 2010 (5 days, reaching dryness by day [2] W. Kaltzin, “Natural wines, als Trend,” 2012, http://www.der-
14). During 2011, R. mucilaginosa and P. anomala were the winzer.at/?id=2500%2C5109787%2C%2C%2CY2Q9NDg%3D.
predominant species throughout fermentation, and these
[3] G. H. Fleet, “Yeast interactions and wine flavour,” International
are considered spoilage yeasts [44]. Several previous studies Journal of Food Microbiology, vol. 86, no. 1-2, pp. 11–22, 2003.
have shown that longer lag phases provide an opportunity
[4] A. Cuadros-Inostroza, P. Giavalisco, J. Hummel, A. Eckardt,
for non-Saccharomyces yeasts and other microorganisms to
L. Willmitzer, and H. Peña-Cortés, “Discrimination of wine
outcompete beneficial microbes and produce toxic and/or attributes by metabolome analysis,” Analytical Chemistry, vol.
noxious compounds, causing spoilage [45–49]. Accordingly, 82, no. 9, pp. 3573–3580, 2010.
we found that 0.65 g L−1 acetic acid was produced in this
[5] J. Marais, “Effect of different wine-making techniques on the
fermentation, which is above the upper range in normal wines composition and quality of pinotage wine. I. Low-temperature
and is considered undesirable [8, 50, 51]. skin contact prior to fermentation,” South African Journal of
Despite the presence of spoilage yeast, the success of Enology and Viticulture, vol. 24, no. 2, pp. 70–75, 2003.
spontaneous fermentation seems to correlate with the length
[6] P. Hernandez-Orte, M. Cersosimo, N. Loscos, J. Cacho, E.
of the lag phase, since fermentations with a longer lag Garcia-Moruno, and V. Ferreira, “Aroma development from
phase were more likely to fail. The onset of fermentation non-floral grape precursors by wine lactic acid bacteria,” Food
also depended on the temperature of the must, so this is Research International, vol. 42, no. 7, pp. 773–781, 2009.
a key factor to consider when predicting the outcome of a [7] P. Romano, C. Fiore, M. Paraggio, M. Caruso, and A. Capece,
spontaneous fermentation. “Function of yeast species and strains in wine flavour,” Interna-
Our integration of novel analytical methods with tra- tional Journal of Food Microbiology, vol. 86, no. 1-2, pp. 169–180,
ditional winemaking using Qvevris provides the basis for 2003.
[8] M. G. Lambrechts and I. S. Pretorius, “Yeast and its importance medically important yeasts,” Journal of Clinical Microbiology,
to wine aroma—a review,” South African Journal of Enology and vol. 39, no. 11, pp. 4042–4051, 2001.
Viticulture, vol. 21, special issue, pp. 97–129, 2000. [24] T. White, T. Bruns, S. Lee, and J. Taylor, “Amplification and
[9] M. Combina, A. Elı́a, L. Mercado, C. Catania, A. Ganga, and C. direct sequencing of fungal ribosomal RNA genes for phyloge-
Martinez, “Dynamics of indigenous yeast populations during netics,” in PCR Protocols: A Guide to Methods and Applications,
spontaneous fermentation of wines from Mendoza, Argentina,” M. A. Innis, D. H. Gelfand, J. J. Sninsky, and T. J. White, Eds.,
International Journal of Food Microbiology, vol. 99, no. 3, pp. pp. 315–322, Academic Press, 1990.
237–243, 2005.
[25] M. A. Lodhi, G. N. Ye, N. F. Weeden, and B. I. Reisch, “A
[10] E. Di Maro, D. Ercolini, and S. Coppola, “Yeast dynamics during simple and efficient method for DNA extraction from grapevine
spontaneous wine fermentation of the Catalanesca grape,” cultivars and Vitis species,” Plant Molecular Biology Reporter,
International Journal of Food Microbiology, vol. 117, no. 2, pp. vol. 12, no. 1, pp. 6–13, 1994.
201–210, 2007.
[26] S. Pardo, M. A. Galvagno, and P. Cerrutti, “Studies of via-
[11] A. Rementeria, J. A. Rodriguez, A. Cadaval et al., “Yeast bility and vitality after freezing of the probiotic yeast Sac-
associated with spontaneous fermentations of white wines from charomyces boulardii: physiological preconditioning effect,”
the ”Txakoli de Bizkaia” region (Basque Country, North Spain),” Revista Iberoamericana de Micologia, vol. 26, no. 2, pp. 155–160,
International Journal of Food Microbiology, vol. 86, no. 1-2, pp. 2009.
201–207, 2003.
[27] C. Alves-Araújo, M. J. Almeida, M. J. Sousa, and C. Leão,
[12] G. H. Fleet, Wine Microbiology and Biotechnology, Taylor &
“Freeze tolerance of the yeast Torulaspora delbrueckii: cellular
Francis, 1993.
and biochemical basis,” FEMS Microbiology Letters, vol. 240, no.
[13] E. H. Hansen, P. Nissen, P. Sommer, J. C. Nielsen, and N. 1, pp. 7–14, 2004.
Arneborg, “The effect of oxygen on the survival of non-
Saccharomyces yeasts during mixed culture fermentations of [28] A. Martini, “Origin and domestication of the wine yeast
grape juice with Saccharomyces cerevisiae,” Journal of Applied Saccharomyces cerevisiae,” Journal of Wine Research, vol. 4, no.
Microbiology, vol. 91, no. 3, pp. 541–547, 2001. 3, pp. 165–176, 1993.
[14] R. B. Boulton, V. L. Singleton, L. F. Bisson, and R. E. Kunkee, [29] P. Raspor, D. M. Milek, J. Polanc, S. Smole Možina, and N.
Principles and Practices of Winemaking, Springer, 1998. Čadež, “Yeasts isolated from three varieties of grapes cultivated
[15] N. P. Jolly, O. P. H. Augustyn, and I. S. Pretorius, “The role and in different locations of the Dolenjska vine-growing region,
use of non-saccharomyces yeasts in wine production,” South Slovenia,” International Journal of Food Microbiology, vol. 109,
African Journal of Enology and Viticulture, vol. 27, no. 1, pp. 15– no. 1-2, pp. 97–102, 2006.
38, 2006. [30] A. Martini, “Origin and domestication of the wine yeast
[16] M. Ciani and F. Maccarelli, “Oenological properties of non- Saccharomyces cerevisiae,” Journal of Wine Research, vol. 4, no.
Saccharomyces yeasts associated with wine-making,” World 3, pp. 165–176, 1993.
Journal of Microbiology and Biotechnology, vol. 14, no. 2, pp. 199– [31] A. Vaughan-Martini and A. Martini, “Facts, myths and legends
203, 1998. on the prime industrial microorganism,” Journal of Industrial
[17] C. M. Egli, W. D. Edinger, C. M. Mitrakul, and T. Henick-Kling, Microbiology, vol. 14, no. 6, pp. 514–522, 1995.
“Dynamics of indigenous and inoculated yeast populations and [32] M. J. De La Torre, M. C. Millan, P. Perez-Juan, J. Morales, and J.
their effect on the sensory character of Riesling and Chardonnay M. Ortega, “Indigenous yeasts associated with two Vitis vinifera
wines,” Journal of Applied Microbiology, vol. 85, no. 5, pp. 779– grape varieties cultured in southern Spain,” Microbios, vol. 100,
789, 1998. no. 395, pp. 27–40, 1999.
[18] A. Soden, I. L. Francis, H. Oakey, and P. A. Henschke, “Effects [33] I. S. Pretorius, “Tailoring wine yeast for the new millennium:
of co-fermentation with Candida stellata and Saccharomyces novel approaches to the ancient art of winemaking,” Yeast, vol.
cerevisiae on the aroma and composition of Chardonnay wine,” 16, no. 8, pp. 675–729, 2000.
Australian Journal of Grape and Wine Research, vol. 6, no. 1, pp.
21–30, 2000. [34] T. Deak and L. R. Beuchat, “Yeasts associated with fruit juice
concentrates,” Journal of Food Protection, vol. 56, no. 9, pp. 777–
[19] N. Hierro, B. Esteve-Zarzoso, A. González, A. Mas, and J. M.
782, 1993.
Guillamón, “Real-time quantitative PCR (QPCR) and reverse
transcription-QPCR for detection and enumeration of total [35] M. J. Torija, N. Rozès, M. Poblet, J. M. Guillamón, and A. Mas,
yeasts in wine,” Applied and Environmental Microbiology, vol. “Yeast population dynamics in spontaneous fermentations:
72, no. 11, pp. 7148–7155, 2006. comparison between two different wine-producing areas over
[20] K. Zott, O. Claisse, P. Lucas, J. Coulon, A. Lonvaud-Funel, and I. a period of three years,” International Journal of General and
Masneuf-Pomarede, “Characterization of the yeast ecosystem in Molecular Microbiology, vol. 79, no. 3-4, pp. 345–352, 2001.
grape must and wine using real-time PCR,” Food Microbiology, [36] P. Satora and T. Tuszynski, “Biodiversity of yeasts during plum
vol. 27, no. 5, pp. 559–567, 2010. Wegierka Zwykla spontaneous fermentation,” Food Technology
[21] L. Cocolin and D. Ercolini, Molecular Techniques in the Micro- and Biotechnology, vol. 43, no. 3, pp. 277–282, 2005.
bial Ecology of Fermented Foods, Springer, 2007. [37] A. Querol, M. Jimenez, and T. Huerta, “Microbiological and
[22] B. G. Baldwin, “Phylogenetic utility of the internal transcribed enological parameters during fermentation of musts from poor
spacers of nuclear ribosomal DNA in plants: an example from and normal grape-harvests in the region of Alicante (Spain),”
the compositae,” Molecular Phylogenetics and Evolution, vol. 1, Journal of Food Science, vol. 55, no. 6, pp. 1603–1606, 1990.
no. 1, pp. 3–16, 1992. [38] V. Loureiro and M. Malfeito-Ferreira, “Spoilage yeasts in the
[23] Y. C. Chen, J. D. Eisner, M. M. Kattar et al., “Polymorphic wine industry,” International Journal of Food Microbiology, vol.
internal transcribed spacer region 1 DNA sequences identify 86, no. 1-2, pp. 23–50, 2003.
[39] I. Magyar and T. Tóth, “Comparative evaluation of some

oenological properties in wine strains of Candida stellata, Can-
dida zemplinina, Saccharomyces uvarum and Saccharomyces
cerevisiae,” Food Microbiology, vol. 28, no. 1, pp. 94–100, 2011.
[40] J. Mora, J. I. Barbas, and A. Mulet, “Growth of yeast species dur-
ing the fermentation of musts inoculated with kluyveromyces
thermotolerans and saccharomyces cerevisiae,” American Jour-
nal of Enology and Viticulture, vol. 41, no. 2, pp. 156–159, 1990.
[41] E. Longo, J. B. Velázquez, C. Sieiro, J. Cansado, P. Calo, and T.
G. Villa, “Production of higher alcohols, ethyl acetate, acetalde-
hyde and other compounds by 14 Saccharomyces cerevisiae
wine strains isolated from the same region (Salnés, N.W. Spain),”
World Journal of Microbiology & Biotechnology, vol. 8, no. 5, pp.
539–541, 1992.
[42] C. Lema, C. Garcia-Jares, I. Orriols, and L. Angulo, “Contribu-
tion of Saccharomyces and non-Saccharomyces populations to
the production of some components of Albariño wine aroma,”
American Journal of Enology and Viticulture, vol. 47, no. 2, pp.
206–216, 1996.
[43] I. Andorrà, S. Landi, A. Mas, J. M. Guillamón, and B. Esteve-
Zarzoso, “Effect of oenological practices on microbial popula-
tions using culture-independent techniques,” Food Microbiol-
ogy, vol. 25, no. 7, pp. 849–856, 2008.
[44] J. I. Pitt and A. D. Hocking, Fungi and Food Spoilage, Springer,
3rd edition, 2009.
[45] G. S. Drysdale and G. H. Fleet, “The growth and survival
of acetic acid bacteria in wines at different concentrations of
oxygen,” American Journal of Enology and Viticulture, vol. 40,
no. 2, pp. 99–105, 1989.
[46] L. F. Bisson, “Stuck and sluggish fermentations,” American
Journal of Enology and Viticulture, vol. 50, no. 1, pp. 107–119,
1999.
[47] C. G. Edwards, R. B. Beelman, C. E. Bartley, and A. L.
Mcconnell, “Production of decanoic acid and other volatile
compounds and the growth of yeast and malolactic bacteria
during vinification,” American Journal of Enology and Viticul-
ture, vol. 41, no. 1, pp. 48–56, 1990.
[48] C. G. Edwards, K. M. Haag, and M. D. Collins, “Identification
and characterization of two lactic acid bacteria associated with
sluggish/stuck fermentations,” American Journal of Enology and
Viticulture, vol. 49, no. 4, pp. 445–448, 1998.
[49] M. J. Torija, G. Beltran, M. Novo et al., “Effects of fermentation
temperature and Saccharomyces species on the cell fatty acid
composition and presence of volatile compounds in wine,”
International Journal of Food Microbiology, vol. 85, no. 1-2, pp.
127–136, 2003.
[50] M. Vilanova, Z. Genisheva, L. Bescansa, A. Masa, and J. M.
Oliveira, “Volatile composition of wines from cvs. Blanco
lexı́timo, Agudelo and Serradelo (Vitis vinifera) grown in
Betanzos (NW Spain),” Journal of the Institute of Brewing, vol.
115, no. 1, pp. 35–40, 2009.
[51] I. Mato, S. Suárez-Luque, and J. F. Huidobro, “Simple determi-
nation of main organic acids in grape juice and wine by using
capillary zone electrophoresis with direct UV detection,” Food
Chemistry, vol. 102, no. 1, pp. 104–112, 2007.
http://dx.doi.org/10.1155/2013/780145
Review Article
Engineering Microbial Cells for the Biosynthesis of
Natural Compounds of Pharmaceutical Significance
Philippe Jeandet,1 Yann Vasserot,1 Thomas Chastang,2 and Eric Courot2

1
Laboratory of Enology and Applied Chemistry, Research Unit on Vines and Wines in Champagne, UPRES EA 4707,
Faculty of Sciences, University of Reims, P.O. Box 1039, 51687 Reims Cedex 02, France
2
Laboratory of Stress, Defenses, and Plant Reproduction, Research Unit on Vines and Wines in Champagne,
UPRES EA 4707, Faculty of Sciences, University of Reims, P.O. Box 1039, 51687 Reims Cedex 02, France
Correspondence should be addressed to Philippe Jeandet; philippe.jeandet@univ-reims.fr
Received 13 February 2013; Accepted 30 March 2013
Copyright © 2013 Philippe Jeandet et al. This is an open access article distributed under the Creative Commons Attribution License,
Microbes constitute important platforms for the biosynthesis of numerous molecules of pharmaceutical interest such as antitumor,
anticancer, antiviral, antihypertensive, antiparasitic, antioxidant, immunological agents, and antibiotics as well as hormones,
belonging to various chemical families, for instance, terpenoids, alkaloids, polyphenols, polyketides, amines, and proteins.
Engineering microbial factories offers rich opportunities for the production of natural products that are too complex for cost-
effective chemical synthesis and whose extraction from their originating plants needs the use of many solvents. Recent progresses
that have been made since the millennium beginning with metabolic engineering of microorganisms for the biosynthesis of natural
products of pharmaceutical significance will be reviewed.
1. Introduction sacrifice of two to four fully grown trees of Taxus brevifolia.

Finally, microbial fermentations are very easy to transpose at
Genetic engineering of cells, particularly microorganisms, the industrial level, as previously reported for the production
can be successfully applied for the development of strains of resveratrol [8].
dedicated to the overproduction of natural products. In fact, Currently, Escherichia coli, on one hand, and Saccha-
microbes are widely used for the biosynthesis of numerous romyces cerevisiae, on the other hand, are employed for the
valuable molecules such as antitumor, anticancer, antiviral, microbial synthesis of almost all natural products of interest,
antiparasitic, antioxidant, immunological, agents, antibiotics although new platform microorganisms are emerging [6].
and hormones as well as biofuels [1–7]. Biotechnology might This review summarizes, without being exhaustive, efforts
represent a powerful means for large-scale synthesis of that have been made since the millennium beginning with
natural compounds as the use of microorganisms allows for the metabolic engineering of microbes for the biosynthesis
low cost and rapid production of biologically active molecules of natural products with a particular emphasis on com-
through an environmentally benign route. Such approaches pounds showing antibiotic, antitumoral, antiparasitic, antihy-
avoid organic solvents, heavy metal catalyzers, and strong pertensive, immunological, hormonal, and antioxidant activ-
acids and bases that are currently employed upon synthetic ities and belonging to various chemical families, including
chemistry-based routes. Some natural compounds including terpenoids, polyphenols, polyketides, alkaloids, stilbenoids,
nicotianamine for instance can reach unrealistic prices ($ amines, and proteins.
350,000 per gram). On the other hand, extraction of valuable
natural products from native plant sources is difficult as these
2. Antibiotics
compounds generally accumulate in very small amounts, also
needing the use of many solvents. As an example, the doses of Among the alkaloid family which are nitrogen-containing
taxol needed for the treatment of a single patient requires the molecules of low molecular weight synthesized by numerous
plants, fungi, bacteria, and animals, benzylisoquinoline MAO

alkaloids constitute a remarkable group of pharmaceutical Dopamine 3,4-dihydroxyphenyl
molecules including, for example, the narcotic analgesic acetaldehyde
morphine and the antibacterial agents berberine, scoulerine, Escherichia coli
NCS
palmatine, and magnoflorine. Besides, these alkaloids have
been reported such as anti-human immunodeficiency virus (𝑆)-norlaudanosoline (𝑆)-3󳰀 -hydroxycoclaurine
(HIV) and antioxidant and anticancer agents (see [3, 4] 60 MT CNMT
and the references therein). Because these compounds are (𝑆)-3󳰀 -hydroxy-𝑁-
methylcoclaurine
too complex for cost-effective chemical total synthesis,
microbial systems for benzylisoquinoline-type alkaloid 4󳰀 OMT
production were used [9, 10], but attempts to reconstruct BBE (𝑆)-
the entire metabolic pathway leading to (S)-reticuline and, (𝑆)-scoulerine reticuline (𝑆)-reticuline
beyond this branch-point biochemical intermediate, to
the synthesis of many other related alkaloids have been Corytuberine CYP80G2
Berberine
achieved recently by the group of Minami and coworkers
[11]. They established a microbial system, combining two Corytuberine-𝑁-
different microorganisms together with microbial and methyltransferase
plant-derived genes. Reticuline production was achieved (CNMT?)
Magnoflorine
first in engineered Escherichia coli and the biosynthesis
of specific alkaloids was terminated using a coculture of Saccharomyces cerevisiae
engineered E. coli and Saccharomyces cerevisiae [11]. This
Figure 1: Benzylisoquinoline alkaloid biosynthetic pathway recon-
study resulted in (i) the biosynthesis of (S)-reticuline, the
structed in Escherichia coli and Saccharomyces cerevisiae. For in vivo
key intermediate in the formation of benzylisoquinoline production of (S)-reticuline, transgenic E. coli expresses biosynthetic
alkaloids, only when using a methyl group donor and genes MAO, NCS, 6OMT, CNMT, and 4󸀠 OMT. For in vivo produc-
(ii), the production of specific alkaloids, magnoflorine, and tion of (S)-scoulerine and magnoflorine from reticuline originating
corytuberine on one hand, and, on the other hand, scoulerine from engineered E. coli, transgenic S. cerevisiae expresses, respec-
depending on the targeted genes. Reticuline in engineered tively, the genes BBE, CYP80G2 and CNMT. All genes are from Cop-
E. coli was obtained as follows (Figure 1): dopamine tis japonica except the gene MAO which originates from Micrococcus
was converted first to 3,4-dihydroxyphenylacetaldehyde luteus. MAO: monoamine oxidase; NCS: norcoclaurine synthase;
by a monoamine oxidase from Micrococcus luteus; (S)- 6OMT: norcoclaurine 6-O-methyltransferase; CNMT: coclaurine-
norlaudanosoline was synthesized by the condensation N-methyltransferase; 4󸀠 OMT: 3󸀠 -hydroxy-N-methylcoclaurine-4󸀠 -
of dopamine and 3,4-dihydroxyphenylacetaldehyde by O-methyltransferase; BBE: berberine bridge enzyme; CYP80G2:
P450 corytuberine synthase.
a norcoclaurine synthase from Coptis japonica, then
leading to (S)-3󸀠 -hydroxycoclaurine via a norcoclaurine
6-O-methyltransferase from C. japonica. (S)-3󸀠 -hydroxy-
N-methylcoclaurine was obtained through the action of the production of many other alkaloids of medical interest
coclaurine-N-methyltransferase from C. japonica, leading [11].
to (S)-reticuline via the 3󸀠 -hydroxy-N-methylcoclaurine-4󸀠 - Recently, an important modification of the pathway
O-methyltransferase from C. Japonica. When S-adenosyl- leading to (S)-reticuline was reported by the same group [12].
L-methionine (SAM) was used as a methyl group donor, An E. coli strain that over produces L-tyrosine by amino-acid
the engineered biosynthesis pathway leads predominantly fermentation was first generated. This was accomplished by
to the stereospecific (S)-reticuline with a yield of 55 mg/L three steps of genetic engineering. Production of dopamine
within 1 h. S. cerevisiae expressing heterologous genes was then obtained via the introduction in E. coli of a tyrosi-
were added to E. coli cells engineered with the reticuline nase and its adaptor protein from Ralstonia solanacearum,
biosynthetic genes with 5 mM dopamine in the medium after able to convert L-tyrosine to L-dihydroxyphenylalanine (L-
a certain period of culture. S. cerevisiae was engineered to DOPA) along with a L-DOPA decarboxylase from Pseu-
harbor the genes encoding the P450 enzyme, corytuberine domonas putida, leading to L-dopamine. This short pathway
synthase (CYP80G2), and a CNMT from C. japonica closed the gap between L-tyrosine and the synthetic pathway
(possibly a CNMT-like methyltransferase), leading to to reticuline via the monoamine oxidase from M. luteus
magnoflorine at a yield of 7.2 mg/L within 72 h without any and leading to the 3,4-dihydroxyphenylacetaldehyde (see the
addition of SAM. Otherwise, the second alkaloid produced, previous section). In the final step of the fermentative pro-
(S)-scoulerine, was obtained with a S. cerevisiae strain duction of (S)-reticuline, the dopamine-producing pathway
expressing the berberine bridge enzyme (BBE) from C. described above was combined with the synthetic pathway
japonica at a yield of 8.3 mg/L within 48 h. The conversion from dopamine to (S)-reticuline.
efficiencies of reticuline to magnoflorine or scoulerine in Engineering microbes also offers great opportunities
S. cerevisiae cells were 65.5% and 75.5%, respectively. This for the biosynthesis of plant polyketides of medical interest
underlines that such a working system results in the efficient including antibiotics (erythromycin, rifamycin), anticancer
production of benzylisoquinoline alkaloids in engineered drugs (tetracyclines, anthracyclines, and epothilones),
microbes. The system could be generalized and applied to antiparasitic agents (avermectin), cholesterol-lowering
SCoA SCoA
H O
O 6x methylmalonyl-
CO2 H
CoA extender units
Propionyl-
CoA primer unit
DBES 1 DBES 2 DBES 3
Loading Module 1 Module 4 Module 6

Module 2 Module 3 Module 5 Release
AT ACP KS AT KR ACP KS AT KR ACP KS AT ACP KS AT DH ER KR ACP KS AT KR ACP KS AT KR ACP TE
S S S S S S S
O O O O O O O
OH OH O OH OH
OH OH O OH
OH OH O
OH O
OH
OH
OH
OH
O
O
Erythromycin A
HO OH
OH N(CH3 )2
HO OH
O
O O
O
O OH
O OCH3
O
OH
O OH
6-deoxyerythronolide B
= 4󳰀 -phosphopantetheine cofactor
Figure 2: Biosynthesis of deoxyerythronolide B in the route to erythromycin A (adapted from [19]). Deoxyerythronolide B synthases (DEBS)
catalyze iterative claisen-type condensations by progressively associating one propionyl-CoA primer unit with six methylmalonyl-CoA
extender units. Each DEBS molecule contains two modules, these modules representing a physical location for a claisen-type condensation.
Once loaded onto the ketosynthase (KS) of module 1 by the loading acyl transferase (AT) and the acyl carrier protein (ACP) domains,
propionyl-CoA condenses with a methylmalonyl unit loaded onto module 1 ACP by the module 1 AT domain. After that, ketoreductases
(KR), specific for each module, reduce the resulting ketone group. The chain then passes in a progressive manner from module 1 ACP (via
a phosphopantetheine cofactor) to the KS domain of the next module (module 2). Following these iterative condensations, the polyketide
chain grows before being released and cyclized by a thioesterase/claisen cyclase (TE). Other abbreviations: ER are enoyl reductase and DH:
dehydratase.
agents (lovastatin), and immunosuppressants (rapamycin) The macrocyclic core of erythromycin, 6-

[13]. They are synthesized from acyl-CoA precursors by deoxyerythronolide B (6-dEB), a broad spectrum antibiotic
polyketide synthases (PKSs) [14], enzymes that are grouped synthesized by the soil bacterium Saccharopolyspora
in three different classes: (i) type I refers to large and erythraea, is a prototype of polyketides. As mentioned
multifunctional enzymes; (ii) type II refers dissociable above, polyketides are obtained from single building blocks,
complexes usually composed of monofunctional enzymes acetyl-CoA, propionyl-CoA, and methylmalonyl-CoA
found in bacteria [15]; (iii) type III consists of homodimeric (Figure 2) [19], requiring the production of the last two
enzymes of relatively small size found in plants [16] as well compounds reengineering of E. coli metabolism due to
as in bacteria [17] and fungi [18]. Type III PKSs catalyze limited capabilities for molecular biological manipulation
iterative condensations of malonyl-CoA units with a great of microorganisms especially actinomycetes [20]. For an
variety of CoA-linked starter substrates (see below). efficient 6-dEB production, the intracellular activity of the
deoxyerythronolide synthase (DEBS) catalyzing formation without or with heterologous metK expression within 5 days
of the macrocyclic core of erythromycin has to be well [21].
synchronized with precursor biosynthesis.
First attempts for engineering this polyketide pathway in
E. coli addressed expression of the genes encoding DEBS. 3. Antitumor Antibiotics
Each of the three genes encoding for DEBS 1, DEBS 2, and
Important human antibiotics and antitumor antibiotics such
DEBS 3 from S. erythraea was cloned into E. coli BL21 (DE
as tetracyclines and anthracyclines are of bacterial origin.
3) along with the sfp phosphopantetheinyl transferase gene
These compounds are synthesized by bacterial type II polyke-
from Bacillus subtilis, the plasmid-based coexpression of this
tide synthases (PKSs). The main limitation for reconstituting
gene facilitating posttranslational modification of the DEBS
the biosynthesis of aromatic polyketides in E. coli consists in
protein in E. coli [20]. A single copy of the sfp gene under
the inability to generate the elongated poly-𝛽-ketone back-
the control of the T7 RNA polymerase promoter was thus
bone from malonyl-CoA, the starting block of the synthesis
integrated in the prp operon of strain BL21 (DE 3), yielding E.
pathway [24]. This requires a minimal PKS comprising a
coli strain BAP1. This site was chosen because it suppresses the
ketosynthase- (KS-) chain length factor (CLF) and an acyl-
possibility that the bacterium cell uses propionate, a starting
carrier protein (ACP). In fungi, PKSs are megasynthases in
block for 6-dEB biosynthesis, as a source for catabolism
which, unlike bacterial aromatic PKSs, the enzymatic com-
and anabolism. Secondly, along with the sfp gene, the prpE
ponents are not dissociated. Megasynthases include a starter-
gene of the strain BAP1 was placed under control of an
unit acyltransferase (SAT), a KS, a malonyl-CoA:acyl carrier
IPTG inducible T7 promoter. PrpE is thought to convert
protein (ACP) transacylase (MAT), a product template (PT),
propionate into propionyl-CoA, thus allowing accumulation
an acyl-carrier protein (ACP), and a thioesterase/claisen-
of this precursor inside the cell that is well synchronized
cyclase (TE/CLC) (Figure 2) (see [24] and the references
with DEBS expression. Finally, a suitable pathway for the
therein). A fungal megasynthase, the PKS4 from Gibberella
biosynthesis of (2S)-methylmalonyl-CoA was engineered in
fujikuroi, can be expressed by E.coli and the biosynthesis
E. coli. The strain BAP1 was thus transformed with plasmids
of aromatic polyketides can aslo be reconstituted in E.
harboring the propionyl-CoA carboxylase genes pccA and
coli [25]. In order to produce polyketides with cyclization
pccB from Streptomyces coelicolor. Coexpression of the birA
regioselectivities not observed among fungal polyketides,
biotin ligase from the bacterium was shown to enhance the
the PKS4 KS MAT didomain and the ACP domain were
activity of the biotinylated subunit pccA [20].
extracted as stand-alone proteins, leading to a dissociated
To resume, engineering of E. coli to produce 6-dEB
PKS4 minimal PKS. Such an approach mimics the dissoci-
included introduction of the DEBS genes from S. erythraea,
ation of the components that are observed in bacterial type
introduction of a propionyl-CoA carboxylase from B. subtilis,
II minimal polyketide synthases [24]. In a second approach,
introduction of a propionyl-CoA carboxylase from S. coeli-
Tang and coworkers [24] used a compact synthetic polyketide
color for (2S)-methylmalonyl-CoA biosynthesis and deletion
megasynthase PKS WJ consisting of KS, MAT, and ACP on a
of the endogenous prpRBCD genes to avoid utilization of
single polypeptide. Two plasmids encoding the PKS WJ were
propionate as a carbon source for catabolism and anabolism
transformed into the E. coli strain BAP1. Under fed-batch
of the cell along with overexpression of the endogenous
fermentation, high cell density cultures of engineered E. coli
prpE and birA genes (to ensure conversion of propionate
lead to the production of 3 mg/L of the aromatic polyketide
to propionyl-CoA and to enhance conversion of propionyl-
anthraquinone within 60 h after addition of isopropyl 𝛽-D-1-
CoA to (2S)-methylmalonyl-CoA, resp.). A major challenge
thiogalactopyranoside. In contrast, no polyketide production
in metabolic engineering is indeed to achieve an optimal flow
was observed in the dissociated PKS4 minimal PKS. Using
through a given heterologous metabolic pathway for high
engineered fungal PKSs in bacteria thus paves the way for
yield and productivity. The best performing system led to a
a more general approach towards the production of various
calculated specific productivity of 0.1 mmol, that is, 38.6 mg
aromatic polyketides.
of 6-dEB/g cellular protein/day.
Production of 6-dEB in E. coli was improved further by
the same group [21] by cloning an S-adenosylmethionine 4. Anticancer Drugs
(SAM) synthase gene (metK) from Streptomyces spectabilis
into an expression plasmid under the control of an inducible Taxol (paclitaxel) and its structural analogs are potent
T7 promoter according to the assumption that the pro- and commercially successful anticancer drugs [26, 27].
duction of several antibiotics (for instance, streptomycin Taxol is a diterpene originating from the isoprenoid (ter-
and erythromycin) is improved by increasing the intra- penoid) pathway. Opportunities for biosynthesis of natu-
cellular concentration of SAM. The same observation was ral terpenoid drugs using engineered microorganisms have
made with the production of nicotianamine by Wada been reviewed elsewhere [28–30]. Terpenoids are synthe-
and coworkers [22], indicating that increased SAM levels sized from the condensation of two C5 units, respec-
may be a generally useful tool for improving the produc- tively, isopentenyl-pyrophosphate (IPP) and its isomer
tion of various natural products. This plasmid was intro- dimethylallyl-pyrophosphate (DMAPP), as starting blocks,
duced into the engineered BAP1 engineered E. coli strain both originating from either the mevalonate pathway or
described above [20], improving the specific productiv- the 2C-methyl-D-erythritol-4-phosphate pathway (Figure 3)
ity of 6-dEB from 10.86 mg/L/OD600 to 20.08 mg/L/OD600 [31–35]. Condensation of these two C5 units and larger
expression of the Sulfolobus acidocaldarius geranylgeranyl-

Sugars
diphosphate synthase along with the codon-optimized taxa-
diene synthase from Taxus chinensis, a truncated version of
G3P Pyruvate
yeast 3-hydroxyl-3-methylglutaryl-CoA reductase (tHMG-
Methylerythritol-phosphate pathway
CoA reductase) (which was also shown to increase produc-
(𝑑𝑥𝑠, 𝑖𝑑𝑖, 𝑖𝑠𝑝𝐷, and 𝑖𝑠𝑝𝐹 genes) tion of artemisinic acid in recombinant yeast, see the next
FPP GPP section and [36]) and the UPC2-1 transcription factor gene,
synthase synthase leads to taxadiene levels of 8.7 mg/L [38].
FPP GPP IPP
To go farther in the production of taxol, Stephanopoulos
DMAPP
and co-workers designed a multivariate modular approach
GGPP
synthase
and succeeded in increasing the titer of taxadiene up to
Geranyl- 1 g/L [23]. The multivariate modular pathway engineering
geranyl PP was composed of two modules separated at the level of
Taxadiene isopentenyl pyrophosphate (IPP)/dimethylallyl pyrophos-
synthase phate (DMAPP). The first module comprised an eight-
Taxa- Baccatin Taxol gene, upstream, methylerythritol-phosphate (MEP) pathway
Taxadiene III
4(5),11(12) 5𝛼-OL native to E. coli, the pathway in which the expression of
-diene only four genes (dxs, idi, ispD, and ispF) that deemed to be
Taxadiene
5𝛼- rate-limiting was modulated. To channel the overflow flux
hydroxylase
from the starting blocks, IPP and DMAPP, towards taxol
Figure 3: Pathway for in vivo production of taxadiene and 5- production, a second module comprised a two-gene (encod-
𝛼-hydroxytaxadiene in the route to taxol via the methylerythri- ing for the geranyl geranyldiphosphate synthase GGPPS and
tol phosphate pathway in E. coli (adapted from [23]). Genes the taxadiene synthase, TS, resp.) downstream, heterologous
dxs, idi, ispD, and ispF, respectively, encoding for 1-deoxy-D- pathway to taxadiene. Using this modular approach, taxa-
xylulose-5-phosphate synthase, isopentenyl diphosphate isomerase, diene production exhibited a 15,000-fold increase over the
isoprenoid synthase domain, and 2C-methyl-D-erythritol-2,4- control, yielding 1.02 g/L in defined media with controlled
cyclodiphosphate synthase. IPP: isopentenyl diphosphate; DMAPP: glycerol feeding. This has to be compared with the milligrams
dimethylallyl diphosphate; GPP: geranyl diphosphate; FPP: farnesyl of taxadiene obtained in previous studies [38].
diphosphate; GGPP: geranylgeranyl diphosphate. As the cyclic olefin taxadiene undergoes multiple steps of
stereospecific oxidations (including oxygenation at seven dif-
ferent positions mediated by CYP450-dependent monooxy-
genases) [39], acylations, and benzoylation to form baccatin
IPP- and DMAPP-derived building blocks such as the C10 III, the next step in taxol biosynthesis leading to taxadien-5𝛼-
unit, geranyl pyrophosphate (GPP), the C15 unit, farnesyl ol was engineered. The first oxygenation step is the hydroxy-
pyrophosphate (FPP), and the C20 unit, geranylgeranylpy- lation of the C5 position catalyzed by a CYP450, taxadien-
rophosphate (GGPP), yields monoterpenes, sesquiterpenes 5𝛼-hydroxylase (Figure 3) [40]. To succeed in the functional
(for example, artemisin, see below), triterpenes (steroids), expression of plant CYP450 in E. coli, engineering taxol
tetraterpenes (carotenoids), and diterpenes such as taxol. P450 oxidation chemistry in the bacterium was performed
Taxol has been isolated from Taxus brevifolia (Pacific yew by the construction of a chimera protein from taxadien-
tree) (see [23] and the references therein). Unfortunately, 5𝛼-hydroxylase with its redox partner, Taxus cytochrome
sufficient taxol dosage for one patient requires sacrificing P450 reductase. One of the chimera enzymes generated,
two to four fully grown trees of this species. Moreover, the At24T5𝛼OH-tTCPR, carried out the first hydroxylation step
chemical synthesis of this compound is extremely complex with a yield higher than 98%, leading to taxadien-5𝛼-ol along
(requiring 35 to 51 steps), with a highest total yield of with a byproduct.
the best synthesis of 0.4% (see [23] and the references Owing to the fact that there are still six more hydroxy-
therein). Although a semisynthetic route using baccatin III lation steps (and probably some other modifications of the
(an intermediate in the taxol biosynthetic pathway isolated taxane core structure that have not yet been identified) until
from plant sources) (Figure 3) was described, limitations reaching the suitable taxol precursor, baccatin III, a lot of
were encountered in terms of productivity and scalability (see work remains for engineering the complete pathway to taxol
[23] and the references therein). Metabolic engineering of in microbes.
microbes thus appears along with plant cell cultures, as a valu- The epothilones are potential anticancer agents useful
able alternative for the production of such compounds, which particularly in case of paclitaxel- (taxol-) resistant tumor cell
eliminates reliance on yew tree plantations. As the biosynthe- lines [41]. Epothilones A and B are synthesized by the action
sis of taxol comprises many enzyme-catalyzed steps, research of a hybrid nonribosomal peptide synthetase (NRPS)/type
was limited to attempts made towards the production of I polyketide synthase (PKS) together with a P450 epoxi-
taxadiene, the first committed taxol intermediate. Metabolic dase that converts desoxyepothilone into epothilone, by the
engineering of taxadiene production as an essential step myxobacterium, Sorangium cellulosum [42]. Inactivation of
towards taxol production was first described in yeast S. the epoxidase or deletion of the gene encoding for this
cerevisiae [38]. In the best producing system, heterologous enzyme yields epothilones C and D as direct products of the
Acetyl-CoA
Simple sugar
Acetoacetyl-CoA
Mevalonate pathway
IPP DMAPP
Ergosterol GPP
synthase
ERG20
Squalene GPP
FPP
Successive synthase
Artemisinin hydroxylations Gossypol
ERG20
through a P450 FPP
monooxygenase
Semisynthesis CDS
(CYP71AV1 gene) ADS CAH
8-hydroxy-
Artemisinic Amorpha- Cadinene
cadinene
acid 4,11-diene
Figure 4: Engineering the pathways for the biosynthesis of artemisinic acid and 8-hdroxycadinene via the mevalonate pathway (adapted
from [36, 37]). ADS: amorphadiene synthase; CDS: cadinene synthase; CAH: cadinene hydroxylase. For other abbreviations, see legend of
Figure 3.
NRPS/PKS. Heterologous expression of an entire epothilone most successful examples of terpenoid biosynthetic path-
gene cluster composed of six open reading frames (ORFs), way engineering is the production of artemisinic acid, the
epoA, epoB, epoC, epoD, epoE, and epoF encoding a sin- precursor for the antimalarial agent artemisinin, in yeast
gle NRPS module, eight PKS modules, and a C-terminal [36]. Engineering of the pathway leading to artemisinic acid
thioesterase, has already been described in Streptomyces was achieved in yeast S. cerevisiae in three steps (Figure 4):
coelicolor but with a low yield of epothilones [43] and (1) increasing farnesyl pyrophosphate (FPP) production, (2)
Myxococcus xanthus, the growth of which is slow compared conversion of FPP to amorpha-4,11-diene, and (3) three-step
to E. coli [44]. In a more recent work [42], the genes oxidation of amorphadiene to artemisinic acid. Combina-
encoding the entire cluster epoA, epoB, epoC, epoD, epoE, tion, on one hand, of the overexpression of a truncated,
and epoF were redesigned and synthesized to allow for soluble form of 3-hydroxy-3-methylglutaryl-CoA reductase
expression into two polypeptides with compatible module (tHMGR gene) along with the down regulation of ERG9
linkers. Splitting this large protein into two polypeptides was which encodes squalene synthase (leading to the production
necessary to obtain its expression in a soluble form along of sterols) increased amorphadiene production fivefold and
with the lowering of the incubation temperature (15∘ C), co twofold, respectively, in yeast. Downregulating ERG9 and
expression of molecular chaperones, and switching from a overexpressing upc2-1 which enhances the activity of UPC2
T7 promoter to an arabinose-inducible PBAD promoter. The (a global transcription factor regulating the biosynthesis
entire cluster was expressed in the modified E. coli K207-3, of sterols in yeast) were combined with integration of an
which is a derivative of BL21 (DE3), the so-called strain BAP1 additional copy of tHMGR into the chromosome. Intro-
already mentioned in the production of the macrocyclic core ducing the amorphadiene synthase gene from Artemisia
of erythromycin, 6-deoxyerythronolide B [20]. This, in turn, annua to this high FPP engineered producer resulted in
resulted in the complete biosynthesis of epothilones C and D an increased production of amorphadiene up to 149 mg/L.
in E. coli as clearly evidenced by the LC/MS/MS analysis of Finally, overexpression of the gene encoding FPP synthase
E. coli extracts. The design of the synthetic epothilone genes (ERG20) enhanced amorphadiene production by ca 10%, that
together with E. coli expression provides an excellent platform is, 153 mg/L in the resulting engineered yeast strain EPY224
for the production of epothilone analogues, according to (about 500-fold the production previously reported [45]).
protein engineering strategies, to redesign a large gene cluster. This transgenic yeast strain was transformed with a vector
harboring a cytochrome P450 monooxygenase and its redox
5. Antiparasitic Agents partner CYP71AV1/CPR from A. annua that performs a three-
step oxidation of amorphadiene to artemisinic acid under
Terpenoids and their derivatives together with their roles in the control of galactose-inducible promoters. The produced
respiration, electron transport, photosynthesis, and hormone acid was efficiently transported out of the cell, making its
signaling may serve as antiparasitic agents. One of the purification relatively easy. In a one-litre aerated bioreactor,
115 mg of artemisinic acid was produced, of which 76 mg was production in the engineered SCY4-pINNAS strain reached
recovered following a one-step silica gel column chromato- up to 328 𝜇g/g wet weight within 48 h when the growth
graphic separation of ether extracts from the washed buffers. medium was supplemented with glycine and formate (leading
Artemisinic acid is produced by the engineered S. cerevisiae to a 100-fold increased accumulation of the usual level of
strain EPY224 at a biomass fraction comparable to that the precursor SAM). The intracellular concentration of NA
produced by the plant but within few days for the yeast versus reached 766 𝜇g/g wet weight within 60 h when the transgene
several months in A. annua. Artemisinin is then obtained was introduced with the high-copy episomal vector without
from artemisinic acid by hemisynthesis [46]. Further experi- any noticeable effect on yeast growth (strain SCY4-pEPNAS).
ments achieved by the same group have improved production As the chemical synthesis of NA for commercial use as an
of artemisinic acid in a transgenic yeast transformed with antihypertensive agent is very expensive ($ 350/mg), this
an amorphadiene synthase, an amorphadiene oxidase, and a makes its production through a relatively simple fermentative
cytochrome P450 reductase expressed from a single plasmid. process very efficient [22].
Modulating the selection markers and using appropriate
medium formulation, production of artemisinic acid in the 7. Hormones and Immunological Agents
engineered yeast reached 250 𝜇g/mL in shake-flask cultures
and 1 g/L in bioreactors [47]. Engineering microbes for the production of hormones
More recently, a novel semibiosynthetic route towards and immunological agents has also been described. Mazor
artemisinin has been reported using the long-chain fatty acid and coworkers have reported facile isolation of full-length
P450BM3 hydroxylase from Bacillus megaterium capable of immunoglobulin G antibodies from combinatorial libraries
forming artemisinic −11S, 12-epoxide from amorphadiene, expressed in E. coli [51]. Full-length heavy and light chains are
this epoxide being an interesting intermediate to form dihy- secreted into the periplasmic space, where they assemble into
droartemisinic alcohol [48]. An engineered E. coli expressing aglycosylated IgGs that are captured by an Fc-binding protein
this particular P450 hydroxylase yielded high titers of the anchored in the inner membrane of the bacterium.
epoxide (250 mg/L). Finally, engineering of the mevalonate The use of Bacillus subtilis for recombinant protein
pathway in Seo E. coli with equivalent genes from Staphy- expression has been well established. Advantageously, this
lococcus aureus yielded an average amount of 22.7 g/L of bacterium is able to secrete functional extracellular proteins
amorphadiene in the culture medium [49]. directly into the culture medium, making their purification
Difficulties in efficient translations and functional expres- easy; it is not pathogenic for humans and lacks endotoxins
sion of key enzymes including plant P450 enzymes are in the cell wall. B. subtilis was used for establishing a system
encountered in E. coli. To overcome this limitation, the for the suitable production of human interleukin-3 (hIL-
production of functionalized terpenoids in E. coli, such as the 3), which is a cytokine that regulates blood-cell production
conversion of cadinene to 8-hydroxycadinene, a precursor by controlling the production, differentiation, and function
for the phytoalexin gossypol, was achieved by Chang and of granulocytes and macrophages [52]. To facilitate optimal
co-workers by cloning a cadinene hydroxylase along with a secretion of hIL-3 by B. subtilis, three signal peptides were
cytochrome P450 reductase (Figure 4) [37]. tested: the modified AmyL, Pel, and SacB signal peptides.
Use of the modified AmyL signal peptide resulted in a
reproducibly high secretion from the bacterial cell. The other
6. Antihypertensive Agents signal peptides used (SacB and Pel, resp.) did not result in
productive secretion of hIL-3 or led to plasmid instability,
Nicotianamine (NA) is an ubiquitous compound in plants such that it is impossible to predict which signal is optimal
thought to play a role, as a metal chelator, in the internal for the secretion of a particular heterologous protein such
transport of metal nutrients as well as to be a precursor of as hIL-3. The hIL-3 purified showed full biological activity,
mugineic acids secreted from the roots and facilitating iron that is, it was able to specifically induce proliferation of
solubilization in the soil. Interestingly, NA possesses anti- the hIL-3-dependent leukemia cell line MO7e. The best
hypertensive effects via the inhibition of the angiotensin I- production system was reported for the B. subtilis WB800
converting enzyme which catalyzes hydrolysis of angiotensin strain in combination with the pP43LatIL3 vector containing
I to the potent vasoconstrictor angiotensin II [50]. NA the AmyL signal peptide, yielding 100 mg/L hIL-3 within 24 h
biosynthesis is quite simple as it is obtained from the conden- of culture.
sation of three S-adenosylmethionine (SAM) units by nico- As B. subtilis, E. coli is a common host used for over-
tianamine synthase (NAS). Large amounts of NA (766 𝜇g/g expression of proteins that do not require glycosylation. This
wet weight) were obtained following the introduction of is due to the fact that recombinant proteins can easily be
the Arabidopsis AtNAS2 gene into the yeast Saccharomyces transported in and easily released from the periplasmic space
cerevisiae strain SCY4 [22]. Two vectors for NAS expression by osmotic shock and that there are fewer proteases in the
in S. cerevisiae were constructed, the first one containing an periplasm compared to the cytoplasm. According to this,
integrative vector, pINNAS (strain SCY4-pINNAS) and the periplasmic production of human interferon-𝛾 (hINF-𝛾) but
second one a high-copy episomal vector, pEPNAS (strain not of human interleukin-2 (hIL-2) by the Tat (twin arginine
SCY4-pEPNAS). Both contained the AtNAS2 ORF from translocation) pathway in E. coli strain BL21-SI was described
Arabidopsis under the control of the GAPDH promoter to [53]. The Tat system was identified as a transport mechanism
induce constitutive and strong expression of AtNAS2. NA for a specific group of periplasmic proteins [54]. Expression
of recombinant proteins was obtained using the pEMR vector

Shikimate
containing the Tat-dependent modified penicillin acylase pathway
signal peptide (mSPpac) driven by the T7 promoter.
Phenylalanine
E. coli was also engineered for the expression of a
recombinant gonadotropin-releasing hormone (GnRH) vac- PAL
Tyrosine
cine that could be used as a vaccine for the control of
Cinnamic acid
fertility and hormone-dependent diseases [55]. To this aim,
a highly effective expression plasmid pED-GNRH3 was TAL
4CH
constructed to express GnRH3-hinge-MVP in the form of
a fusion protein. The recombinant gene encoding GnRH3- 𝑝-coumaric acid Sugar
hinge-MVP which contains three repeated GnRH units, a metabolism
fragment of hinge region, and a T-cell epitope of measles virus 4CL
protein was cloned into E. coli. A crucial point was to insure
plasmid stability during the fermentation process affecting 𝑝-coumaroyl- + Malonyl-coenzyme A
the production of the recombinant protein. This was achieved coenzyme A
by the optimization of many culture conditions including CHS
STS
fermentation temperature and the culture medium. Both the
stability of the plasmid and the expression of GnRH3-hinge- Resveratrol Naringenin
MVP are of real concern for the industrial production of the chalcone
engineered strain. Figure 5: In vivo biosynthesis of resveratrol from phenylala-
nine or tyrosine via the phenylpropanoid/polymalonate pathway.
PAL/TAL: phenylalanine/tyrosine ammonia lyase; C4H: cinnamate-
8. Antioxidant and Anticancer Agents 4-hydroxylase; C4L: coumarate:coenzyme A ligase; STS: stilbene
(resveratrol) synthase; CHS: chalcone synthase.
Hydroxystilbenes including resveratrol and its derivatives
possess a large spectrum of biological activities in human
health as cardioprotective, antitumor, and antioxidant agents.
Engineering bacteria or yeast for resveratrol might represent to p-coumarate (4CH), 4CL, and STS activities. The biosyn-
a valuable means for its production in large quantities [8, 56– thetic pathway has been introduced successfully into the
59]. At the same time, the potential role of resveratrol as oleaginous yeast Yarrowia lipolytica completed by constitu-
an antimicrobial compound acting as an allelochemical or a tive expression of malonyl-CoA, and resulting in a resveratrol
phytoalexin and protecting plants from fungal attacks [60– production of 1.46 mg/L [69]. The entire resveratrol pathway
62] or increasing their antioxidant capabilities has led to has been introduced in other microorganisms such as the
numerous studies concerning molecular engineering of this baker yeast S. cerevisiae, molds such as Aspergillus niger and
compound in plants [59, 63–66]. A. oryzae, and bacteria such as Lactococcus lactis and E.
Resveratrol is obtained through the universal phenyl- coli. In all cases, pathway expression started with PAL except
propanoic acid pathway beginning with phenylalanine (via for E. coli in which TAL was the first enzyme introduced
the phenylalanine ammonia lyase, PAL) or with tyrosine (via [70]. Resveratrol synthesis was only reported in E. coli that
the tyrosine ammonia lyase, TAL) (Figure 5). The obtained expressed the TAL gene upon feeding with p-coumarate. The
phenylpropanoic acids are then activated by ligation to necessary addition of p-coumarate to the culture medium has
coenzyme A by coumarate:CoA ligase (CL), often acting the consequence that the TAL or PAL activities remained low
only in the presence of a 4-hydroxyphenyl group and thus in the bacterium. As a result, much of the ongoing efforts
termed 4-coumaroyl:CoA ligase (4CL). Hydroxystilbenes are are dedicated to increasing these activities in bacteria, for
then synthesized by type III polyketide synthases, stilbene example, by aiming to develop novel TAL or PAL enzymes
synthases (STS) which compete with chalcone synthases, [71, 72]. Two more recent works constructed the complete
implied in the biosynthesis of the C6 -C3 -C6 flavonoid ring resveratrol biosynthetic pathway in S. cerevisiae to pro-
[67, 68]. PKS act to condense successive units of malonyl- duce resveratrol from phenylalanine. Trantas and coworkers
CoA with p-coumaroyl-CoA, forming a linear polyketide reported a resveratrol production of 0.29 mg/L within 120 h
molecule before cyclization. of cultivation after feeding with 10 mM of phenylalanine
Two main strategies have already been used in the [73]. Finally, Shin and coworkers engineered the S. cerevisiae
engineering of resveratrol in bacteria or yeast: (i) introduc- strain (W303-1A) with four heterologous genes, the PAL
ing the entire biosynthetic pathway using aromatic amino gene from Rhodosporidium toruloides, the C4H and the 4CL1
acids as substrates (L-phenylalanine or L-tyrosine), and (ii) genes from A. thaliana, and the STS gene from Arachis
introducing specific genes, such as 4CL and STS, starting with hypogea [74]. Overexpression of the acetyl-CoA carboxylase
para-coumaric acid as a substrate. (ACC1) gene for increasing the pool of malonyl-CoA resulted
Engineering the entire pathway consists in the production in an engineered strain termed W303-1A/ACC1 harboring
of resveratrol from its precursor phenylalanine or tyrosine the p425PAL, p423C4H, and pESCTRP4CL1-STS plasmids
through the introduction of the genes encoding for PAL or capable of synthesizing 5.8 mg/L resveratrol upon feeding
TAL, the cinnamate-4-hydroxylase, leading from cinnamate with 12 mM tyrosine in YP medium containing 2% galactose.
Introducing selective genes is an alternative strategy 14.4 mol% yield [82]. When the 4CL and the STS-encoding
directed towards transforming microorganisms with only genes added to the yeast genome were submitted to protein
two genes (4CL and STS) under utilization of p-coumaric fusion, expression of the 4CL::STS fusion protein increased
acid. In fact, p-coumaric acid can be used advantageously as a resveratrol production in yeast by up to 10-fold (5.25 mg/L
precursor for the bioproduction of resveratrol in engineered resveratrol). This can be compared to the coexpression of
bacteria rather than the basic amino-acid phenylalanine. This 4CL and STS without any protein fusion (0.65 mg/L) [81].
amino-acid has indeed to be converted to p-coumaric acid Works also clearly demonstrated that synthetic scaffolds
from cinnamic acid via the P450 enzyme, C4H. However, can optimize engineered metabolic pathways. Use of a syn-
P450 monooxygenases failed to be effectively expressed in E. thetic scaffold to recruit the 4CL1 and STS enzymes of the
coli. Additionally, p-coumarate is a relatively cheap precursor resveratrol pathway improved resveratrol production in S.
useful for bioproduction systems [75]. cerevisiae, resulting in a 5-fold improvement of the resveratrol
Recombinant E. coli strains able to produce resveratrol production over the nonscaffolded control (14.4 mg/L) [83].
were obtained [76–78]. An accumulation of 3.6 mg/L resver- Finally, engineered S. cerevisiae yeast cells expressing 4CL and
atrol was reported in the E. coli BL21 (DE3) engineered strain STS genes along with the araE gene encoding for a high-
through an increase in malonate transport and malonyl-CoA capacity E. coli transporter exhibited a resveratrol production
synthase activity [78]. Transformation of the BL21 E. coli of 3.1 mg/L [84].
strain with the 4CL gene from tobacco and the grapevine The efficacy of recombinant microorganisms for resver-
STS gene resulted in a resveratrol production of 16 mg/L [76]. atrol production depends on various factors, such as the
Similarly, a metabolically engineered E. coli strain, which was species and the strain, the origin of the transferred genes,
transformed with the 4CL gene from A. thaliana and a STS and culture conditions as well as other parameters such as
gene from A. hypogea, resulted in a final high titer >100 mg/L plasmids or precursors used. A recent study has underlined
of resveratrol [77]. Replacement of the 4CL gene from A. the crucial role played by the culture medium [85]. Resver-
thaliana by the 4CL gene from Lithospermum erythrorhizon atrol production by an industrial Brazilian sugar cane yeast
in the same JM109 E. coli strain, leads to a resveratrol expressing the 4CL1 gene from A. thaliana and the grapevine
production of 171 mg/L after feeding with p-coumaroyl-CoA STS gene, in a rich medium, reached up to 391 mg/L.
[79]. When modified cinnamoyl-CoA was used, unnatural
stilbenes were formed, suggesting that engineering microbes 9. Conclusions
paves the way for the biosynthesis of novel compounds with
potent biological activities [79]. There is an increasing number of examples demonstrating
An original approach was developed more recently by the that engineering microbial cells for the production of natural
group of Lim and coworkers in which two different E. coli compounds of pharmaceutical interest is a successful method
strains (BL21 Star and BW27784), two different promoters to gain access to some bioactive substances only accumu-
(the strong T7 promoter and the constitutive promoter of the lating in low amounts in plants and whose total chemical
glyceraldehyde-3-phosphate dehydrogenase), and a library of synthesis is too complex. Moreover, recovery of unnatural
two 4CLs from Petroselinum crispum and A. thaliana and compounds with significant biological properties has also
nine STS (from pine, Polygonum, Psilotum, Vitis, and Arachis) been reported through metabolic engineering of microbes
in different combinations were systematically examined to [79].
optimize resveratrol production from p-coumaric acid in E. Some particularly interesting results have been obtained
coli [75]. Both STS from Vitis vinifera (VvSTS) gene and from for engineered systems where relatively simple genetic con-
A. hypogea (AhSTS) gene were more efficient in producing structs are needed. Extraordinary progresses have also been
resveratrol when coupled with the 4-CL from A. thaliana done in the engineering of very complex biosynthetic path-
(At4CL) gene than with the 4CL from Petroselinum crispum ways such as benzylisoquinoline alkaloids, artemisinic acids
(Pc4CL) gene. The best performing system was observed or taxol pathways. Stephanopoulos and co-workers recently
with the BW27784 pUCo-VvSTS-At4CL strain, at which achieved the production of taxadiene, a key intermediate in
0.05 mM cerulenin was added to the culture medium for the route for taxol, at low cost in E. coli [23]. However, still
improving the malonyl-CoA pool. Resveratrol production work has to be done before optimizing taxol production by
from p-coumaric acid reached the impressive value of 2.3 g/L. microorganisms as many steps remain to be engineered until
Due to the prohibitive cost of cerulenin, the simultaneous reaching taxol.
overexpression of both the acetyl-CoA carboxylase (ACC) Moreover, successful endeavors in the future for optimiz-
gene and the biotin ligase (BirA) gene from Photorhabdus ing microbial production of natural compounds will require,
luminescens was obtained allowing acetyl-CoA to be more besides the functional integration of a complete or a partial
efficiently converted to malonyl-CoA. heterologous pathway in a given microbe, adaptation of
There are many reports of selective genes being intro- the host organism to its environment to improve product
duced in S. cerevisiae, resulting in various productions of titers. For example, resveratrol production by a recombinant
resveratrol ranging from 0.65 to 5.8 mg/L [76, 80, 81]. An industrial yeast strain has been shown to be considerably
engineered yeast strain W303-1A containing a 4CL gene enhanced according to the culture medium used [85].
(4CL1 gene) from A. thaliana and as STS gene from A. New techniques for metabolic engineering will be
hypogea produced unglycosylated resveratrol at a titer of exploited in the next years to come. Namely, methodologies
3.1 mg/L after feeding with 15.3 mg/L p-coumaric acid, that is, for generating high-quality libraries of enzyme variants [86]
and novel high-throughput screening (HTS) technologies biology,” Current Opinion in Biotechnology, vol. 24, pp. 291–299,
[87] will open the way for the engineering of enzymes for 2013.
the biosynthesis of various compounds with potent biological [8] D. Donnez, P. Jeandet, C. Clément, and E. Courot, “Bioproduc-
activities. Specifically, HTS technologies can rapidly lead tion of resveratrol and stilbene derivatives by plant cells and
to the identification of genes which modulate a particular microorganisms,” Trends in Biotechnology, vol. 27, no. 12, pp.
biosynthesis pathway. Gathering all genes encoding for a 706–713, 2009.
biomolecular pathway will allow the assembly of genetic [9] D. A. Rathbone and N. C. Bruce, “Microbial transformation of
constructs for the synthesis of a given product. New methods alkaloids,” Current Opinion in Microbiology, vol. 5, no. 3, pp.
for the rapid cloning of single genes together with the 274–281, 2002.
availability of synthetic operons, such as bacterial operons [10] J. Keasling, “From yeast to alkaloids,” Nature Chemical Biology,
(commonly used in the biosynthesis of many medically and vol. 4, no. 9, pp. 524–525, 2008.
pharmaceutically important compounds), have accelerated [11] H. Minami, J. S. Kim, N. Ikezawa et al., “Microbial production of
the construction of synthetic multigene pathways [88, 89]. plant benzylisoquinoline alkaloids,” Proceedings of the National
A very recent application was the coordinated induction Academy of Sciences of the United States of America, vol. 105, no.
of a five-gene pathway leading to zeaxanthin production in 21, pp. 7393–7398, 2008.
tailored S. cerevisiae [90]. [12] A. Nakagawa, H. Minami, J. S. Kim et al., “A bacterial platform
A major challenge in metabolic engineering is to achieve for fermentative production of plant alkaloids,” Nature Commu-
an optimal flow through a given heterologous metabolic nications, vol. 2, no. 1, article 326, 2011.
pathway for high yield and productivity. A simple method to [13] S. Horinouchi, “Combinatorial biosynthesis of plant medicinal
fine tune metabolic pathways, named customized optimiza- polyketides by microorganisms,” Current Opinion in Chemical
tion of metabolic pathways by combinatorial transcriptional Biology, vol. 13, no. 2, pp. 197–204, 2009.
engineering (COMPACTER), was described for rapid tuning [14] C. Hertweck, “The biosynthetic logic of polyketide diversity,”
of gene expression in a heterologous pathway under different Angewandte Chemie, vol. 48, no. 26, pp. 4688–4716, 2009.
metabolic backgrounds. This was achieved by the creation of [15] B. Shen, “Polyketide biosynthesis beyond the type I, II and III
a library of mutant pathways using de novo assembly of pro- polyketide synthase paradigms,” Current Opinion in Chemical
moter mutants for each pathway gene in a target microorgan- Biology, vol. 7, no. 2, pp. 285–295, 2003.
ism followed by high-throughput screening/selection [91]. [16] I. Abe and H. Morita, “Structure and function of the chalcone
Finally, systems biology, metabolic engineering, and synthase superfamily of plant type III polyketide synthases,”
“omics” technologies (genomics, functional genomics, and Natural Product Reports, vol. 27, no. 6, pp. 809–838, 2010.
metagenomics) have opened the way for protein and [17] N. Funa, Y. Ohnishi, I. Fujli, M. Shibuya, Y. Ebizuka, and
biomolecular pathway engineering. Production of novel S. Horinouchi, “A new pathway for polyketide synthesis in
small molecules of pharmaceutical interest by designing microorganisms,” Nature, vol. 400, no. 6747, pp. 897–899, 1999.
novel proteins and pathways is also under progress [92]. [18] Y. Seshime, P. R. Juvvadi, I. Fujii, and K. Kitamoto, “Discov-
These new methodologies will thus pave the way for very ery of a novel superfamily of type III polyketide synthases
important advances in metabolic engineering of microbial in Aspergillus oryzae,” Biochemical and Biophysical Research
cell factories. Communications, vol. 331, no. 1, pp. 253–260, 2005.
[19] Y. A. Chan, A. M. Podevels, B. M. Kevany, and M. G. Thomas,
“Biosynthesis of polyketide synthase extender units,” Natural
References Product Reports, vol. 26, no. 1, pp. 90–114, 2009.
[1] J. A. Chemler and M. A. Koffas, “Metabolic engineering for [20] B. A. Pfeifer, S. J. Admiraal, H. Gramajo, D. E. Cane, and C.
plant natural product biosynthesis in microbes,” Current Opin- Khosla, “Biosynthesis of complex polyketides in a metabolically
ion in Biotechnology, vol. 19, no. 6, pp. 597–605, 2008. engineered strain of E. coli,” Science, vol. 291, no. 5509, pp. 1790–
[2] N. Ferrer-Miralles, J. Domingo-Espı́n, J. Corchero, E. Vázquez, 1792, 2001.
and A. Villaverde, “Microbial factories for recombinant phar- [21] Y. Wang, B. A. Boghigian, and B. A. Pfeifer, “Improving het-
maceuticals,” Microbial Cell Factories, vol. 8, article 17, 2009. erologous polyketide production in Escherichia coli by overex-
[3] S. Y. Lee, H. U. Kim, J. H. Park, J. M. Park, and T. Y. Kim, pression of an S-adenosylmethionine synthetase gene,” Applied
“Metabolic engineering of microorganisms: general strategies Microbiology and Biotechnology, vol. 77, no. 2, pp. 367–373, 2007.
and drug production,” Drug Discovery Today, vol. 14, no. 1-2, pp. [22] Y. Wada, T. Kobayashi, M. Takahashi, H. Nakanishi, S. Mori,
78–88, 2009. and N. K. Nishizawa, “Metabolic engineering of Saccharomyces
[4] J. Du, Z. Shao, and H. Zhao, “Engineering microbial factories cerevisiae producing nicotianamine: potential for industrial
for synthesis of value-added products,” Journal of Industrial biosynthesis of a novel antihypertensive substrate,” Bioscience,
Microbiology and Biotechnology, vol. 38, pp. 873–890, 2011. Biotechnology and Biochemistry, vol. 70, no. 6, pp. 1408–1415,
[5] M. Papagianni, “Recent advances in engineering the central car- 2006.
bon metabolism of industrially important bacteria,” Microbial [23] P. K. Ajikumar, W. H. Xiao, K. E. J. Tyo et al., “Isoprenoid
Cell Factories, vol. 11, article 50, 2012. pathway optimization for Taxol precursor overproduction in
[6] J. Marienhagen and M. Bott, “Metabolic engineering of Escherichia coli,” Science, vol. 330, no. 6000, pp. 70–74, 2010.
microorganisms for the synthesis of plant natural products,” [24] W. Zhang, Y. Li, and Y. Tang, “Engineered biosynthesis of
Journal of Biotechnology, vol. 163, pp. 166–178, 2013. bacterial aromatic polyketides in Escherichia coli,” Proceedings of
[7] P. Xu, N. Bhan, and M. A. G. Koffas, “Engineering plant the National Academy of Sciences of the United States of America,
metabolism into microbes: from systems biology to synthetic vol. 105, no. 52, pp. 20683–20688, 2008.
[25] S. M. Ma, J. Zhan, K. Watanabe et al., “Enzymatic synthesis [41] R. J. Kowalski, P. Giannakakou, and E. Hamel, “Activities of
of aromatic polyketides using PKS4 from Gibberella fujikuroi,” the microtubule-stabilizing agents epothilones A and B with
Journal of the American Chemical Society, vol. 129, no. 35, pp. purified tubulin and in cells resistant to paclitaxel (Taxol),”
10642–10643, 2007. Journal of Biological Chemistry, vol. 272, no. 4, pp. 2534–2541,
[26] M. C. Wani, H. L. Taylor, M. E. Wall, P. Coggon, and A. T. 1997.
McPhail, “Plant antitumor agents. VI. The isolation and struc- [42] S. C. Mutka, J. R. Carney, Y. Liu, and J. Kennedy, “Heterologous
ture of taxol, a novel antileukemic and antitumor agent from production of epothilone C and D in Escherichia coli,” Biochem-
Taxus brevifolia,” Journal of the American Chemical Society, vol. istry, vol. 45, no. 4, pp. 1321–1330, 2006.
93, no. 9, pp. 2325–2327, 1971. [43] L. Tang, S. Shah, L. Chung et al., “Cloning and heterologous
[27] D. G. I. Kingston, “The shape of things to come: structural and expression of the epothilone gene cluster,” Science, vol. 287, no.
synthetic studies of taxol and related compounds,” Phytochem- 5453, pp. 640–642, 2000.
istry, vol. 68, no. 14, pp. 1844–1854, 2007. [44] L. Tang, L. Chung, J. R. Carney, C. M. Starks, P. Licari, and L.
[28] D. Klein-Marcuschamer, P. K. Ajikumar, and G. Stephanopou- Katz, “Generation of new epothilones by genetic engineering
los, “Engineering microbial cell factories for biosynthesis of iso- of a polyketide synthase in Myxococcus xanthus,” Journal of
prenoid molecules: beyond lycopene,” Trends in Biotechnology, Antibiotics, vol. 58, no. 3, pp. 178–184, 2005.
vol. 25, no. 9, pp. 417–424, 2007. [45] B. E. Jackson, E. A. Hart-Wells, and S. P. T. Matsuda, “Metabolic
[29] P. K. Ajikumar, K. Tyo, S. Carlsen, O. Mucha, T. H. Phon, and engineering to produce sesquiterpenes in yeast,” Organic Let-
G. Stephanopoulos, “Terpenoids: opportunities for biosynthesis ters, vol. 5, no. 10, pp. 1629–1632, 2003.
of natural product drugs using engineered microorganisms,” [46] N. Acton and R. J. Roth, “On the conversion of dihy-
Molecular Pharmaceutics, vol. 5, no. 2, pp. 167–190, 2008. droartemisinic acid into artemisinin,” Journal of Organic Chem-
[30] J. Kirby and J. D. Keasling, “Biosynthesis of plant isoprenoids: istry, vol. 57, no. 13, pp. 3610–3614, 1992.
perspectives for microbial engineering,” Annual Review of Plant [47] D. K. Ro, M. Ouellet, E. M. Paradise et al., “Induction of multiple
Biology, vol. 60, pp. 335–355, 2009. pleiotropic drug resistance genes in yeast engineered to produce
[31] M. Rohmer, M. Knani, P. Simonin, B. Sutter, and H. Sahm, an increased level of anti-malarial drug precursor, artemisinic
“Isoprenoid biosynthesis in bacteria: a novel pathway for the acid,” BMC Biotechnology, vol. 8, article 83, 2008.
early steps leading to isopentenyl diphosphate,” Biochemical [48] J. A. Dietrich, Y. Yoshikuni, K. J. Fisher et al., “A novel semi-
Journal, vol. 295, no. 2, pp. 517–524, 1993. biosynthetic route for artemisinin production using engineered
[32] T. Kuzuyama and H. Seto, “Diversity of the biosynthesis of the substrate-promiscuous P450BM3,” ACS Chemical Biology, vol.
isoprene units,” Natural Product Reports, vol. 20, no. 2, pp. 171– 4, no. 4, pp. 261–267, 2009.
183, 2003. [49] H. Tsuruta, C. J. Paddon, D. Eng et al., “High-level production
[33] V. J. J. Martin, D. J. Piteral, S. T. Withers, J. D. Newman, and J. of amorpha-4, 11-diene, a precursor of the antimalarial agent
D. Keasling, “Engineering a mevalonate pathway in Escherichia artemisinin, in Escherichia coli,” PLoS ONE, vol. 4, no. 2, Article
coli for production of terpenoids,” Nature Biotechnology, vol. 21, ID e4489, 2009.
no. 7, pp. 796–802, 2003. [50] A. Hayashi and K. Kimoto, “Studies on the mechanism of anti-
[34] D. J. Pitera, C. J. Paddon, J. D. Newman, and J. D. Keasling, hypertensive action by nicotianamine,” Journal of Nutritional
“Balancing a heterologous mevalonate pathway for improved Science and Vitaminology, vol. 56, no. 4, pp. 242–246, 2010.
isoprenoid production in Escherichia coli,” Metabolic Engineer- [51] Y. Mazor, T. V. Blarcom, R. Mabry, B. L. Iverson, and G.
ing, vol. 9, no. 2, pp. 193–207, 2007. Georgiou, “Isolation of engineered, full-length antibodies from
[35] Y. Zhao, J. Yang, B. Qin et al., “Biosynthesis of isoprene in libraries expressed in Escherichia coli,” Nature Biotechnology,
Escherichia coli via methylerythritol phosphate (MEP) path- vol. 25, no. 5, pp. 563–565, 2007.
way,” Applied Microbiology and Biotechnology, vol. 90, no. 6, pp. [52] L. Westers, D. S. Dijkstra, H. Westers, J. M. van Dijl, and W.
1915–1922, 2011. J. Quax, “Secretion of functional human interleukin-3 from
[36] D. K. Ro, E. M. Paradise, M. Quellet et al., “Production of Bacillus subtilis,” Journal of Biotechnology, vol. 123, no. 2, pp. 211–
the antimalarial drug precursor artemisinic acid in engineered 224, 2006.
yeast,” Nature, vol. 440, no. 7086, pp. 940–943, 2006. [53] E. Medina-Rivero, V. E. Balderas-Hernández, L. G. Ordoñez-
[37] M. C. Y. Chang, R. A. Eachus, W. Trieu, D. K. Ro, and J. Acevedo, L. M. T. Paz-Maldonado, A. P. Barba-De La Rosa,
D. Keasling, “Engineering Escherichia coli for production of and A. De León-Rodrı́guez, “Modified penicillin acylase signal
functionalized terpenoids using plant P450s,” Nature Chemical peptide allows the periplasmic production of soluble human
Biology, vol. 3, no. 5, pp. 274–277, 2007. interferon-𝛾 but not of soluble human interleukin-2 by the Tat
[38] B. Engels, P. Dahm, and S. Jennewein, “Metabolic engineering pathway in Escherichia coli,” Biotechnology Letters, vol. 29, no. 9,
of taxadiene biosynthesis in yeast as a first step towards Taxol pp. 1369–1374, 2007.
(Paclitaxel) production,” Metabolic Engineering, vol. 10, no. 3-4, [54] T. Palmer, F. Sargent, and B. C. Berks, “Export of complex
pp. 201–206, 2008. cofactor-containing proteins by the bacterial Tat pathway,”
[39] R. Kaspera and R. Croteau, “Cytochrome P450 oxygenases of Trends in Microbiology, vol. 13, no. 4, pp. 175–180, 2005.
taxol biosynthesis,” Phytochemistry Reviews, vol. 5, no. 2-3, pp. [55] J. Xu, W. Li, J. Wu et al., “Stability of plasmid and expression
433–444, 2006. of a recombinant gonadotropin-releasing hormone (GnRH)
[40] S. Jennewein, R. M. Long, R. M. Williams, and R. Croteau, vaccine in Escherichia coli,” Applied Microbiology and Biotech-
“Cytochrome P450 taxadiene 5𝛼-hydroxylase, a mechanisti- nology, vol. 73, no. 4, pp. 780–788, 2006.
cally unusual monooxygenase catalyzing the first oxygenation [56] Y. Wang, H. Chen, and O. Yu, “Metabolic engineering of resver-
step of taxol biosynthesis,” Chemistry and Biology, vol. 11, no. 3, atrol and other longevity boosting compounds,” BioFactors, vol.
pp. 379–387, 2004. 36, no. 5, pp. 394–400, 2010.
[57] K. V. Kiselev, “Perspectives for production and application of [73] E. Trantas, N. Panopoulos, and F. Ververidis, “Metabolic
resveratrol,” Applied Microbiology and Biotechnology, vol. 90, no. engineering of the complete pathway leading to heterologous
2, pp. 417–425, 2011. biosynthesis of various flavonoids and stilbenoids in Saccha-
[58] Y. Wang, S. Chen, and O. Yu, “Metabolic engineering of romyces cerevisiae,” Metabolic Engineering, vol. 11, no. 6, pp. 355–
flavonoids in plants and microorganisms,” Applied Microbiology 366, 2009.
and Biotechnology, vol. 91, pp. 949–956, 2011. [74] S. Y. Shin, S. M. Jung, M. D. Kim, N. S. Han, and J. H.
[59] P. Jeandet, B. Delaunois, A. Aziz et al., “Metabolic engineering Seo, “Production of resveratrol from tyrosine in metabolically-
of yeast and plants for the production of the biologically engineered Saccharomyces cerevisiae,” Enzyme and Microbial
active hydroxystilbene, resveratrol,” Journal of Biomedicine and Technology, vol. 51, pp. 211–216, 2012.
Biotechnology, vol. 2012, Article ID 579089, 14 pages, 2012. [75] C. G. Lim, Z. L. Fowler, T. Hueller, S. Schaffer, and M. A.
[60] P. Langcake, “Disease resistance of Vitis spp. and the production G. Koffas, “High-yield resveratrol production in engineered
of the stress metabolites resveratrol, 𝜀-viniferin, 𝛼-viniferin and Escherichia coli,” Applied and Environmental Microbiology, vol.
pterostilbene,” Physiological Plant Pathology, vol. 18, pp. 213– 77, no. 10, pp. 3451–3460, 2011.
226, 1981. [76] J. Beekwilder, R. Wolswinkel, H. Jonker, R. Hall, C. H. De Rie
[61] M. Adrian, P. Jeandet, J. Veneau, L. A. Weston, and R. Bessis, Vos, and A. Bovy, “Production of resveratrol in recombinant
“Biological activity of resveratrol, a stilbenic compound from microorganisms,” Applied and Environmental Microbiology, vol.
grapevines, against Botrytis cinerea, the causal agent for gray 72, no. 8, pp. 5670–5672, 2006.
mold,” Journal of Chemical Ecology, vol. 23, no. 7, pp. 1689–1702, [77] K. T. Watts, P. C. Lee, and C. Schmidt-Dannert, “Biosynthesis of
1997. plant-specific stilbene polyketides in metabolically engineered
[62] M. Adrian and P. Jeandet, “Effects of resveratrol on the ultra- Escherichia coli,” BMC Biotechnology, vol. 6, article 22, 2006.
structure of Botrytis cinerea conidia and biological significance [78] L. L. Huang, Z. Xue, and Q. Q. Zhu, “Method for the production
in plant/pathogen interactions,” Fitoterapia, vol. 83, pp. 1345– of resveratrol in a recombinant bacterial host cell,” US Patent
1350, 2012. 2007/0031951A1, 2007.
[63] P. Jeandet, A. C. Douillet-Breuil, S. Debord, M. Sbaghi, R. Bessis, [79] Y. Katsuyama, N. Funa, I. Miyahisa, and S. Horinouchi, “Syn-
and M. Adrian, “Phytoalexins from the vitaceae: biosynthesis, thesis of unnatural flavonoids and stilbenes by exploiting the
phytoalexin gene expression in transgenic plants, antifungal plant biosynthetic pathway in Escherichia coli,” Chemistry and
activity, and metabolism,” Journal of Agricultural and Food Biology, vol. 14, no. 6, pp. 613–621, 2007.
Chemistry, vol. 50, no. 10, pp. 2731–2741, 2002. [80] J. V. W. Becker, G. O. Armstrong, M. J. Van Der Merwe, M.
[64] C. Halls and O. Yu, “Potential for metabolic engineering of G. Lambrechts, M. A. Vivier, and I. S. Pretorius, “Metabolic
resveratrol biosynthesis,” Trends in Biotechnology, vol. 26, no. 2, engineering of Saccharomyces cerevisiae for the synthesis of the
pp. 77–81, 2008. wine-related antioxidant resveratrol,” FEMS Yeast Research, vol.
[65] B. Delaunois, S. Cordelier, A. Conreux, C. Clément, and P. 4, no. 1, pp. 79–85, 2003.
Jeandet, “Molecular engineering of resveratrol in plants,” Plant [81] Y. Zhang, S. Z. Li, J. Li et al., “Using unnatural protein fusions
Biotechnology Journal, vol. 7, no. 1, pp. 2–12, 2009. to engineer resveratrol biosynthesis in yeast and mammalian
[66] P. Jeandet, B. Delaunois, A. Conreux et al., “Biosynthesis, cells,” Journal of the American Chemical Society, vol. 128, no. 40,
metabolism, molecular engineering, and biological functions of pp. 13030–13031, 2006.
stilbene phytoalexins in plants,” BioFactors, vol. 36, no. 5, pp. [82] S. Y. Shin, N. S. Han, Y. C. Park, M. D. Kim, and J. H. Seo, “Pro-
331–341, 2010. duction of resveratrol from p-coumaric acid in recombinant
[67] M. B. Austin and J. P. Noel, “The chalcone synthase superfamily Saccharomyces cerevisiae expressing 4-coumarate:coenzyme A
of type III polyketide synthases,” Natural Product Reports, vol. ligase and stilbene synthase genes,” Enzyme and Microbial
20, no. 1, pp. 79–110, 2003. Technology, vol. 48, no. 1, pp. 48–53, 2011.
[68] O. Yu and J. M. Jez, “Nature’s assembly line: biosynthesis of [83] Y. Wang and O. Yu, “Synthetic scaffolds increased resveratrol
simple phenylpropanoids and polyketides,” Plant Journal, vol. biosynthesis in engineered yeast cells,” Journal of Biotechnology,
54, no. 4, pp. 750–762, 2008. vol. 157, pp. 258–260, 2012.
[69] L. L. Huang, Z. Xue, and Q. Q. Zhu, “Method for the production [84] Y. Wang, C. Halls, J. Zhang, M. Matsuno, Y. Zhang, and
of resveratrol in a recombinant oleaginous microorganism,” US O. Yu, “Stepwise increase of resveratrol biosynthesis in yeast
patent 2009/0082286A1, 2009. Saccharomyces cerevisiae by metabolic engineering,” Metabolic
[70] M. Katz, J. Förster, H. David et al., “Fluxome Sciences A/S, Engineering, vol. 13, pp. 455–463, 2011.
Metabolically engineered cells for the production of resveratrol [85] T. Sydor, S. Schaffer, and E. Boles, “Considerable increase in
or an oligomeric or glycosidically-bound derivative thereof,” US resveratrol production by recombinant industrial yeast strains
Patent 2008/0286844A1, 2008. with use of rich medium,” Applied and Environmental Microbi-
[71] T. Vannelli, Z. Xue, S. Breinig, W. W. Qi, and F. S. Sariaslani, ology, vol. 76, no. 10, pp. 3361–3363, 2010.
“Functional expression in Escherichia coli of the tyrosine- [86] A. V. Shivange, J. Marienhagen, H. Mundhada, A. Schenk, and
inducible tyrosine ammonia-lyase enzyme from yeast Tri- U. Schwaneberg, “Advances in generating functional diversity
chosporon cutaneum for production of p-hydroxycinnamic for directed protein evolution,” Current Opinion in Chemical
acid,” Enzyme and Microbial Technology, vol. 41, no. 4, pp. 413– Biology, vol. 13, no. 1, pp. 19–25, 2009.
422, 2007. [87] G. Yang and S. G. Withers, “Ultrahigh-throughput FACS-based
[72] Z. Xue, M. McCluskey, K. Cantera, A. Ben-Bassat, F. S. screening for directed enzyme evolution,” ChemBioChem, vol.
Sariaslani, and L. Huang, “Improved production of p- 10, no. 17, pp. 2704–2715, 2009.
hydroxycinnamic acid from tyrosine using a novel thermostable [88] Z. Shao, H. Zhao, and H. Zhao, “DNA assembler, an in
phenylalanine/tyrosine ammonia lyase enzyme,” Enzyme and vivo genetic method for rapid construction of biochemical
Microbial Technology, vol. 42, no. 1, pp. 58–64, 2007. pathways,” Nucleic Acids Research, vol. 37, no. 2, article e16, 2009.
[89] M. Blanusa, A. Schenk, H. Sadeghi, J. Marienhagen, and U.

Schwaneberg, “Phosphorothioate-based ligase-independent
gene cloning (PLICing): an enzyme-free and sequence-
independent cloning method,” Analytical Biochemistry, vol.
406, no. 2, pp. 141–146, 2010.
[90] J. Liang, J. C. Ning, and H. Zhao, “Coordinated induction of
multi-gene pathways in Saccharomyces cerevisiae,” Nucleic Acids
Research, vol. 41, article e54, 2013.
[91] J. Du, Y. Yuan, T. Si, J. Lian, and H. Zhao, “Customized optimiza-
tion of metabolic pathways by combinatorial transcriptional
engineering,” Nucleic Acids Research, vol. 40, article e142, 2012.
[92] P. Xu, N. Bhan, and M. A. G. Koffas, “Pathway and protein
engineering approaches to produce novel and commodity small
molecules,” Current Opinion in Biotechnology. In press.
http://dx.doi.org/10.1155/2013/640163
Research Article
An Internalin A Probe-Based Genosensor for Listeria
monocytogenes Detection and Differentiation
Laura Bifulco, Angela Ingianni, and Raffaello Pompei

Department of Biomedical Sciences, Section of Microbiology, University of Cagliari, via Porcell 4, 09124 Cagliari, Italy
Correspondence should be addressed to Raffaello Pompei; rpompei@unica.it
Received 19 December 2012; Revised 13 February 2013; Accepted 22 February 2013
Copyright © 2013 Laura Bifulco et al. This is an open access article distributed under the Creative Commons Attribution License,
Internalin A (InlA), a protein required for Listeria monocytogenes virulence, is encoded by the inlA gene, which is only found in
pathogenic strains of this genus. One of the best ways to detect and confirm the pathogenicity of the strain is the detection of
one of the virulence factors produced by the microorganism. This paper focuses on the design of an electrochemical genosensor
used to detect the inlA gene in Listeria strains without labelling the target DNA. The electrochemical sensor was obtained by
immobilising an inlA gene probe (single-stranded oligonucleotide) on the surfaces of screen-printed gold electrodes (Au-SPEs)
by means of a mercaptan-activated self-assembled monolayer (SAM). The hybridisation reaction occurring on the electrode
surface was electrochemically transduced by differential pulse voltammetry (DPV) using methylene blue (MB) as an indicator. The
covalently immobilised single-stranded DNA was able to selectively hybridise to its complementary DNA sequences in solution
to form double-stranded DNA on the gold surface. A significant decrease of the peak current of the voltammogram (DPV) upon
hybridisation of immobilised ssDNA was recorded. Whole DNA samples of L. monocytogenes strains could be discriminated from
other nonpathogenic Listeria species DNA with the inlA gene DNA probe genosensor.
1. Introduction specific interaction with its host cell receptor E-cadherin [16–
19].
Listeria monocytogenes is a Gram-positive, aerobic, rod- Conventionally, the detection and identification of bacte-
shaped, foodborne pathogenic bacterium inducing listeriosis, ria mainly rely on specific microbiological and biochemical
an illness characterized by encephalitis, septicaemia, and identification methods, which require at least 3 and as many
meningitis [1–4]. It is the only pathogenic species of Listeria in as 7 days to yield results. Genetic characterisation methods
humans and has been the cause of several well-documented are more rapid than the classical identification methods and
food poisoning outbreaks [1, 5–9]. It can also cause gas- lead to unequivocal species identification [20, 21]. Among
troenteritis in otherwise healthy individuals and more severe these, polymerase chain reaction (PCR), followed by hybridi-
invasive diseases in immunocompromised patients, pregnant sation of the PCR amplified target with a labelled single-
women, newborns, and elderly people [10–15]. stranded oligonucleotide probe is an effective method of
L. monocytogenes enters mammalian cells by inducing its sequence-specific DNA detection [16, 17].
own phagocytosis. Internalin A (InlA) is an 80 kDa surface Rapid and reliable detection methods of this pathogenic,
protein which allows Listeria to enter the cells. It is a complex toxin-producing bacterium are required since it is able to
key virulence factor protein encoded by the inlA gene and is survive and grow at low temperatures [22] and because the
specific only for L. monocytogenes and not for other listerial mortality rate for infected individuals is much higher than
species or for other genera. It mediates the attachment of for other common foodborne pathogens [23–27].
Listeria to, and the invasion of, hepatocytes, epithelial, and Although there are many DNA hybridisation assays
endothelial cells. The bacterial adhesion and invasion of currently suitable for diagnosis, faster, cheaper, miniaturised,
human intestinal epithelial cells is also mediated through multianalyte, easier to use, and more sensitive approaches
are highly desirable, especially in the case of decentralised H2 SO4 solution was used for electrochemical cleaning of the
analysis. In this context, electrochemical detection of DNA electrodes.
hybridisation events offer innovative routes [28–37].
An effective and sensitive biosensor requires a probe that 2.1.2. Microbial Strains and Conditions. 6 listerial strains
can be immobilized on a sensing platform. An ideal probe from foods (Lys 1: L. innocua, Lys 2: L. monocytogenes,
should be able to achieve sensitive and specific detection of Lys 3: L. monocytogenes, Lys 4: L. monocytogenes, Lys 5:
the target analyte. It must also be easy to produce and with- L. monocytogenes, and Lys 6: L. ivanovii) and 4 listerial
stand environmental stresses, such as changes in temperature collection strains (Lys 7: C 315 L. innocua, Lys 8: C 276 L.
and pH. The proposed methodology aims at the detection innocua, Lys 9: C 383 L. monocytogenes, and Lys 10: C 483
of L. monocytogenes incidence in either environmental or L. monocytogenes) taken from our Institute’s collection and
clinical samples, based on the detection of the inlA gene in representing important species of the genus Listeria were
DNA extracts from isolated strains. For achieving our goal, used in this work. DNA samples were prepared as described.
we have embedded an inlA-specific probe on the surface of a The strains were grown on BHI plates and reidentified by
SPE using several signal enhancing protocols and measured metabolic tests according to Ingianni et al. [16] and use of the
hybridization events in DNA extracts and control templates, API Listeria galleries (bioMèrieux Italia, Milan, Italy).
based on generated electrochemical signals [29–31, 38–41].
At the best of our knowledge, this is the only genosensor
2.1.3. DNA Extracts. DNA samples were prepared as follows:
so far described that allows the discrimination of L. mono-
the bacteria strains were incubated overnight in BHI broth
cytogenes from different nonpathogenic Listeria strains based
and washed twice in PBS before DNA extraction using an
on the detection of a specific listerial pathogenic factor such
Easy-DNA kit (Invitrogen, Carlsbad, Ca, USA) following
as internalin A. It proved to allow a definite and significant
the manufacturer’s protocol. DNA concentration and purity
differentiation of pathogenic from nonpathogenic listerial
were determined by UV light absorbance measured by an
species.
Ultrospec III spectrophotometer (Pharmacia LKB).
2. Materials and Methods 2.1.4. PCR Performance. PCR was performed as described
2.1. Apparatus, Chemicals, and Probe. Differential pulse by Ingianni et al. [16]. DNA oligonucleotide stock solutions
voltammetry (DPV), for measurements, and cyclic voltam- (100 mg/L) and Listeria DNA extracts (dsDNA 100 mg/L)
metry (CV), for electrode cleaning, were carried out using an were prepared with TE solution (10 mM Tris-HCl, 1 mM
AUTOLAB PGSTAT 30 electrochemical analysis system and EDTA, and pH 8.00) and kept frozen. Working DNA solu-
a GPES 4.8 software package (Eco Chemie, The Netherlands). tions were prepared with either 500 mM acetate buffer (pH
Electrodes: screen-printed gold electrodes (Au-SPEs) were 4.80) or 20 mM Tris-HCl buffer (pH 7.00), according to the
obtained from Ecobioservices & Researches s.r.l. (Florence, hybridisation protocol [38].
Italy). The two 59-base oligonucleotide sequences, the L.
monocytogenes internalin inlA gene probe (A) (Gen Bank 2.2. SAM Preparation and Electrode Modification. The SAM
M67471.1) originally designed by Ingianni et al. [16], and modification of Au-SPEs was performed following a protocol
its complementary sequence target (B) were obtained from described by Gooding et al. [38] and Kerman et al. [39] for
Invitrogen. gold rod electrodes. This protocol was adjusted for screen-
The sequences were as follows [16]: printed electrode modification. The gold surfaces of the work-
ing electrodes were prepared by electrochemical cleaning
DNA probe (59-base sequence A): before modification. The electrodes were cleaned by cycling
5 -CCATTAGCTAATTTAACAACACTAGAA- between the 0 V and +1.5 V potentials in a 50 mM H2 SO4
CGACTAGATATTTCAAGTAATAAGGTGTCA- solution at a scan rate of 100 mV/s for approximately 15 min.
GA-3 ; until reproducible scans were recorded. The electrodes were
DNA target (59-base sequence B): rinsed with sterile distilled water before SAM modification.
SAMs were prepared by covering the surfaces of the clean
5 -TCTGACACCTTATTACTTGAAATATCT- Au-SPEs with a freshly prepared 75 : 25 (v/v) ethanol:water
AGTCGTTCTAGTGTTGTTAAATTAGCTAAT- solution containing 20 mM MPA. Au-SPEs were incubated in
GG-3 . this ethanolic solution overnight for approximately 15 h. The
Au-SPEs/SAM were rinsed with 75 : 25 (v/v) ethanol:water
2.1.1. Reagents, Buffers, and Solutions. Methylene blue (MB) and then with water, prior to covalent activation by immer-
was purchased from Difco. The Methylene blue solution was sion in the 50 mM phosphate buffer solution (pH 7.40)
prepared with 20 𝜇M MB and 20 mM NaCl in 20 mM Tris- containing 2 mM EDC and 5 mM NHS for 1 h. Then, the
HCl buffer (pH 7.00). 3-mercaptopropionic acid (MPA), N- Au-SPEs/SAM/Linker surfaces were rinsed with the 50 mM
hydroxysulfosuccinimide (NHS), and 𝑁󸀠 -ethylcarbodiimide phosphate buffer solution (pH 7.40).
hydrochloride (EDC) were obtained from Sigma-Aldrich Next, DNA immobilisation was performed on the work-
(Steinheim, Germany). All chemicals were of an analytical ing electrode surfaces. 20 𝜇L of 500 mM acetate buffer solu-
reagent grade. In-house distilled and sterilised water was tion (pH 4.80) containing 100 ppm probe were pipetted onto
used for the preparation of all buffers and solutions. 50 mM the surface of each Au-SPEs/SAM/Linker. The probe droplets
were left to air-dry overnight. Sensors were then soaked ×10−6

in water for 2 h and rinsed again with water to remove 1
unbound DNA. Thus, inlA probe-modified Au-SPEs were 0.9
obtained. Three inlA probe-modified electrodes were utilised 0.8
as a control (inlA probe) for each experiment. The cost of each
0.7
electrode was about 1.5 euros, and it was found to be stable for
Current (A)
at least 1 week, when kept in the refrigerator. 0.6
0.5
2.3. Hybridisation. 20 𝜇l of 20 mM Tris buffer solution (pH 0.4
7.00) containing 100 ppm target (complementary sequences 0.3
or whole Listeria DNAs) were pipetted onto the inlA probe- 0.2
modified Au-SPE surfaces. Whole Listeria DNA samples
0.1
were prepared immediately before hybridisation by high tem-
perature denaturation (94∘ C) for 10 min. to obtain ssDNA. 0
The target droplets were air-dried for 30 min. This allowed
Bare electrode Activated SAM
hybrid-modified Au-SPEs to be obtained. Each test required InlA probe
MPA SAM
about 50–60 min of work by a technician.
Figure 1: Comparison of the MB reduction peaks. MB reduction
2.4. MB Binding. MB was accumulated on the surface of at the bare electrode (first gray column), after SAM modification
(second green column) and activation (third pale yellow column)
either the modified or the hybridised electrodes, by pipetting
steps and after inlA probe covalent binding (last blue column)
20 𝜇L of 20 mM Tris-HCl buffer (pH 7.00) containing 20 mM (average of 10 electrodes; inlA probe versus MPA, activated SAM,
MB with 20 mM NaCl, which was then left for 5 min. and bare electrode: 𝑃 < 0.05).
without applying any potential. After MB accumulation, the
electrodes were rinsed with 20 mM Tris-HCl buffer (pH 7.00)
for a few seconds.
Measurements of MB reduction were carried out at
the bare electrodes (Figure 1), at the MPA-SAM modified
2.5. Voltammetric Transduction. The reduction signal of the electrodes, at the EDC/NHS-activated SAM electrodes and
accumulated MB was measured by using differential pulse at the SAM/ssDNA inlA probe-modified electrodes. The
voltammetry (DPV) with an amplitude of 10 mV and scan voltammetric signal of MB reduction at the bare elec-
rate of 20 mV/s. Experiments were carried out in 20 mM Tris- trodes decreased after SAM modification and activation and
HCl buffer (pH 7.00). Each experiment was carried out in increased again after inlA probe linking. The MPA-SAM
triplicate. restricted MB access to the electrode but still allowed signif-
icant electrochemistry to occur at the underlying electrode
3. Results and Discussion (MPA-SAM in Figure 1). The activation of this carboxylic
acid terminated SAM with EDC/NHS further restricted MB
3.1. Genosensors. The genosensors relied on the electro- access to the electrode without completely passivising it
chemical transduction of the hybridisation between the (activated SAM in Figure 1). Immobilisation of the probe on
immobilised ssDNA probe and its unlabelled complementary the SAM-modified electrodes resulted in an increase in MB
sequences. By following the modified protocol, we could peak currents (inlA probe in Figure 1) due to the affinity
form the SAMs on the surfaces of the screen-printed gold of MB for the free guanine bases of the DNA as previous
electrodes and activate them. Then, the original inlA probe reported [30, 42]. The values of inlA probe against bare, SAM,
was covalently linked onto the gold-working electrodes. The and activated SAM were found to be significant: 𝑃 < 0.05.
sensors were optimised for use with the complementary
oligonucleotide and then tested on samples of Listeria culture 3.3. Detection of the InlA Complementary DNA Sequence.
DNA extracts. Hybridisation detection was accomplished by The sensors were studied for hybridisation detection using
measuring the MB reduction signal. Electroactivity of this the complementary sequence of the immobilised probe.
label could discriminate the hybrid from the probe. The The genosensors were usable for one shot only. Therefore,
decrease in the magnitude of the MB voltammetric reduction we compared the data obtained from series of 3 to 5
signals, thus, reflected the extent of hybrid formation. Probe genosensors produced during each experiment. Data shown
specificity and probe-method sensitivity were further tested are the average of each experiment. Figure 2 shows the DP
using PCR products of inlA gene targets as templates. voltammograms for the MB reduction signal at the inlA
probe-immobilised Au-SPEs (blue) and after hybridisation
3.2. Probe Immobilisation. To understand probe coverage with the target (red). The shown voltammetric curves are
and surface organization at the Au-SPEs/SAM, the Au- the average of 5 electrodes. The highest MB reduction
SPEs/SAM/activated, and the Au-SPEs/SAM/probe-modifi- signal was observed with the ssDNA probe on the electrode
ed electrodes, we recorded peak current magnitudes at the alone (Figure 2, blue), because MB has a strong affinity
respective electrodes after incubation in MB solutions. for the free guanine bases; hence, the greatest amount
9𝐸−07 ×10−6
8𝐸−07 0.9
7𝐸−07
6𝐸−07 0.8
Current (A)
5𝐸−07 0.7
4𝐸−07
0.6
3𝐸−07
Current (A)
2𝐸−07 0.5
1𝐸−07 0.4
0𝐸+0
−1𝐸−07 0.3
0
−0.03418
−0.06836
−0.1025
−0.1367
−0.1709
−0.2051
−0.2393
−0.2734
−0.3076
−0.3418
−0.376
−0.4102
−0.4443
−0.4785
0.2
0.1
Potential (V)
0
InlA probe
Complementary
SAM L. monocytogenes 10
Figure 2: DP voltammograms of inlA probe-modified electrodes InlA probe L. innocua 1
and hybrid-modified electrodes with the complementary sequence. Complementary L. innocua 7
L. monocytogenes 2 L. ivanovii 6
Figure 4: Mb reduction test of pathogenic and nonpathogenic

listerial species. Comparison of two samples of L. monocytogenes,
1.2𝐸−06
two samples of L. innocua, and one sample of L. ivanovii versus probe
1𝐸−06 and complementary oligonucleotide (the media of L. monocytogenes
against other Listeria species was statistically highly significant: 𝑃 =
8𝐸−07 0.0016). Probe, complementary, and DNA samples concentrations
Current (A)
were 100 ppm.

6𝐸−07
4𝐸−07
2𝐸−07
The differences recorded in the reduction signals indicate a
0𝐸+00 different grade of hybridization between L. monocytogenes
Probe 25 ppm 50 ppm 75 ppm 100 ppm 150 ppm strain DNAs. It could be due to the presence of a different
Complementary DNA concentration number of copies of the inlA gene; thus, the voltammogram
Figure 3: Detection of the inlA complementary DNA sequence.
peaks of L. monocytogenes DNAs showed a minimum high
Different concentrations of complementary DNA were tested versus when a low number of copies are present, while L. monocyto-
100 ppm probe. Each point is the average of 5 genosensors. The MB genes peaks were higher for a larger number of copies of the
reduction signal decreases to a plateau as the hybridisation increases. gene.
Genosensors were then tested on whole DNA samples
of different Listeria strains. Figure 4 shows the comparison
of MB reduction peaks after hybridisation with two DNA
samples of L. monocytogenes strains, one from L. ivanovii
of MB accumulation occurs on this surface. An obvious and two DNA samples of L. innocua strains. The inlA gene
decrease in the voltammetric peak was observed for the sequences are only present in the DNA of L. monocytogenes.
indicator after double-strand formation (Figure 2, red), since However, the L. monocytogenes voltammogram current
the interaction between MB and the guanine residues of the was always lower than Listeria non-monocytogenes voltam-
probe was prevented by hybrid formation on the electrode mograms (the difference between the media of L. monocyto-
surface. genes and the media of other Listeria species was found to be
The sensors were tested with different concentrations of highly statistically significant: 𝑃 = 0.0016).
complementary oligonucleotide. A voltammetric signal was MB reduction peak data of the six L. monocytogenes
still observed (Figures 3 and 4), even when all the DNA probe strains were compared with the voltammograms of the other
was completely hybridised to a duplex, because MB can also strains. Figure 5 shows that the L. innocua voltammograms
act as an intercalator. However, the rapid decrease of the MB are comparable to the inlA probe signals, while L. monocyto-
signal after hybridisation, as shown in the calibration curve, genes voltammograms are comparable to the complementary
indicates that the voltammetric signal due to intercalation is oligonucleotide signals.
small compared to the signal from direct interaction with the Experiments were carried out to investigate genosensor
guanine bases. stability. They were stored at 4∘ C after preparation, and
measurements were performed after 24, 48, 72, and 96 hours.
3.4. Detection of L. monocytogenes Strains and Discrimination When kept in a freezer at −20∘ C the genosensors presented
from Different Listeria Strains. Genosensors were tested on the same responses for at least 6 weeks, and at −80∘ C, they
whole DNA samples of different L. monocytogenes strains. were still efficient after 3 months.
0.0000009 [2] R. J. Meldrum, P. W. Ellis, P. T. Mannion, D. Halstead, and J.

0.0000008 Garside, “Prevalence of Listeria monocytogenes in ready-to-eat
0.0000007 foods sampled from the point of sale in wales, United Kingdom,”
0.0000006 Journal of Food Protection, vol. 73, no. 8, pp. 1515–1518, 2010.
Current (A)
0.0000005 [3] D. Nucera, S. Lomonaco, D. M. Bianchi et al., “A five year

0.0000004 surveillance report on PFGE types of Listeria monocytogenes
0.0000003 isolated in Italy from food and food related environments,”
0.0000002 International Journal of Food Microbiology, vol. 140, no. 2-3, pp.
0.0000001 271–276, 2010.
0 [4] L. O’Connor, M. O’Leary, N. Leonard et al., “The characteriza-
−1𝐸−07 tion of Listeria spp. isolated from food products and the food-
0
−0.04395
−0.08789
−0.1318
−0.1758
−0.2197
−0.2637
−0.3076
−0.3516
−0.3955
−0.4395
−0.4834
processing environment,” Letters in Applied Microbiology, vol.
51, no. 5, pp. 490–498, 2010.
Potential (V) [5] Centers for Disease Control and Prevention (CDC), “Surveil-
InlA probe Complementary lance for foodborne disease outbreaks: United States, 2007,”
L. monocytogenes L. innocua Morbidity and Mortality Weekly Report, vol. 59, no. 31, pp. 973–
979, 2010.
Figure 5: Voltammograms of listerial strains. Two hybridisation [6] Centers for Disease Control and Prevention (CDC), “Surveil-
voltammograms of L. monocytogenes and L. innocua are shown in lance for foodborne disease outbreaks—United States, 2006,”
comparison with an inlA probe voltammogram and a hybridisation Morbidity and Mortality Weekly Report, vol. 58, pp. 609–615,
test with a complementary sequence voltammogram. 2006.
[7] A. di Pinto, L. Novelleo, F. Montemurro, E. Bonerba, and G.
Tantillo, “Occurrence of Listeria monocytogenes in ready-to-eat
4. Conclusions foods from supermarkets in Southern Italy,” New Microbiolog-
ica, vol. 33, no. 3, pp. 249–252, 2010.
In this work, we investigated the possibility of an Internalin A [8] C. W. Donnelly, “Listeria monocytogenes: a continuing chal-
(inlA) probe application for the construction of a genosensor lenge,” Nutrition Reviews, vol. 59, no. 6, pp. 183–194, 2001.
for the identification of the pathogenic bacterium L. mono- [9] J. Koch, R. Dworak, R. Prager et al., “Large listeriosis outbreak
cytogenes. The inlA probe utilised was previously designed linked to cheese made from pasteurized milk, Germany, 2006-
in our laboratory. Due to the formation of the alkanethiol 2007,” Foodborne Pathogens and Disease, vol. 7, no. 12, pp. 1581–
SAMs adsorbed on Au-SPE surfaces, the inlA probe could be 1584, 2010.
attached using covalent linkers such as EDC and NHS. These [10] S. B. Barbuddhe and T. Chakraborty, “Listeria as an enteroinva-
genosensors were used for the detection of hybridisation on sive gastrointestinal pathogen,” Current Topics in Microbiology
the Au-SPE surfaces by means of MB as the electroactive and Immunology, vol. 337, no. 1, pp. 173–195, 2009.
reporter. The inlA probe-modified Au-SPEs were shown [11] I. A. Gillespie, P. Mook, C. L. Little, K. Grant, and G. K. Adak,
to transduce hybridisation with complementary and, more “Listeria monocytogenes infection in the over-60s in England
interestingly, with whole DNA samples of L. monocytogenes between 2005 and 2008: a retrospective case-control study
that contained the inlA gene. Furthermore, discrimination utilizing market research panel data,” Foodborne Pathogens and
between different pathogenic and nonpathogenic Listeria Disease, vol. 7, no. 11, pp. 1373–1379, 2010.
species was recorded. The test is quite inexpensive, requires [12] H. Hof, “History and epidemiology of listeriosis,” FEMS
less than 60 min of technical work, when DNA extracts are Immunology and Medical Microbiology, vol. 35, no. 3, pp. 199–
available, and can be useful for assaying DNA extracted from 202, 2003.
clinical isolates, as well as from environmental and food [13] K. A. Jackson, M. Iwamoto, and D. Swerdlow, “Pregnancy-
strains, especially in the case of decentralised analysis. associated listeriosis,” Epidemiology and Infection, vol. 138, no.
10, pp. 1503–1509, 2010.
[14] J. McLauchlin, “Human listeriosis in Britain, 1967–85, a sum-
Acknowledgments mary of 722 cases: 1. Listeriosis during pregnancy and in the
newborn,” Epidemiology and Infection, vol. 104, no. 2, pp. 181–
The authors wish to thank Dr. Ilaria Palchetti from the “Ugo 189, 1990.
Schiff ” Dipartimento di Chimica, Florence, Italy, for the stim- [15] P. Mook, K. A. Grant, C. L. Little, G. Kafatos, and I. A. Gillespie,
ulating discussions and critical suggestions. The authors also “Emergence of pregnancy-related listeriosis amongst ethnic
acknowledge the financial support from the Consorzio per le minorities in England and Wales,” Eurosurveillance, vol. 15, no.
Ricerche e lo Sviluppo delle Biotecnologie Biotecne, Cagliari, 27, pp. 17–23, 2010.
Italy, and the Fondazione Banco di Sardegna, Sassari, Italy [16] A. Ingianni, M. Floris, P. Palomba, M. A. Madeddu, M. Quartuc-
(Project 2010-11). cio, and R. Pompei, “Rapid detection of Listeria monocytogenes
in foods, by a combination of PCR and DNA probe,” Molecular
References and Cellular Probes, vol. 15, no. 5, pp. 275–280, 2001.
[17] R. H. Orsi, H. C. D. Bakker, and M. Wiedmann, “Listeria
[1] C. G. Clark, J. Farber, F. Pagotto et al., “Surveillance for Listeria monocytogenes lineages: genomics, evolution, ecology, and
monocytogenes and listeriosis, 1995–2004,” Epidemiology and phenotypic characteristics,” International Journal of Medical
Infection, vol. 138, no. 4, pp. 559–572, 2010. Microbiology, vol. 301, no. 2, pp. 79–96, 2011.
[18] A. J. Roberts, S. K. Williams, M. Wiedmann, and K. K. Night- [34] F. Lucarelli, S. Tombelli, M. Minunni, G. Marrazza, and M.
ingale, “Some Listeria monocytogenes outbreak strains demon- Mascini, “Electrochemical and piezoelectric DNA biosensors
strate significantly reduced invasion, inlA transcript levels, for hybridisation detection,” Analytica Chimica Acta, vol. 609,
and swarming motility in vitro,” Applied and Environmental no. 2, pp. 139–159, 2008.
Microbiology, vol. 75, no. 17, pp. 5647–5658, 2009. [35] B. Meric, K. Kerman, G. Marrazza, I. Palchetti, M. Mascini, and
[19] W. D. Schubert, C. Urbanke, T. Ziehm et al., “Structure of M. Ozsoz, “Disposable genosensor, a new tool for the detection
internalin, a major invasion protein of Listeria monocytogenes, of NOS-terminator, a genetic element present in GMOs,” Food
in complex with its human receptor E-cadherin,” Cell, vol. 111, Control, vol. 15, no. 8, pp. 621–626, 2004.
no. 6, pp. 825–836, 2002. [36] O. Pänke, A. Kirbs, and F. Lisdat, “Voltammetric detection of
[20] D. Elhanafi, V. Utta, and S. Kathariou, “Genetic characterization single base-pair mismatches and quantification of label-free
of plasmid-associated benzalkonium chloride resistance deter- target ssDNA using a competitive binding assay,” Biosensors and
minants in a Listeria monocytogenes train from the 1998-1999 Bioelectronics, vol. 22, no. 11, pp. 2656–2662, 2007.
outbreak,” Applied and Environmental Microbiology, vol. 76, no. [37] J. P. Tosar, K. Keel, and J. Laı́z, “Two independent label-
24, pp. 8231–8238, 2010. free detection methods in one electrochemical DNA sensor,”
[21] C. Mammina, A. Aleo, C. Romani et al., “Characterization of Biosensors and Bioelectronics, vol. 24, no. 10, pp. 3036–3042,
Listeria monocytogenes isolates from human listeriosis cases in 2009.
Italy,” Journal of Clinical Microbiology, vol. 47, no. 9, pp. 2925– [38] J. J. Gooding, P. Erokhin, and D. B. Hibbert, “Parameters impor-
2930, 2009. tant in tuning the response of monolayer enzyme electrodes
[22] R. O. Azizoglu, J. Osborne, S. Wilson, and S. Kathariou, “Role fabricated using self-assembled monolayers of alkanethiols,”
of growth temperature in freeze-thaw tolerance of Listeria spp.,” Biosensors and Bioelectronics, vol. 15, no. 5-6, pp. 229–239, 2000.
Applied and Environmental Microbiology, vol. 75, no. 16, pp. [39] K. Kerman, D. Ozkan, P. Kara, B. Meric, J. J. Gooding, and
5315–5320, 2009. M. Ozsoz, “Voltammetric determination of DNA hybridization
[23] P. P. Banada, S. Guo, B. Bayraktar et al., “Optical forward- using methylene blue and self-assembled alkanethiol mono-
scattering for detection of Listeria monocytogenes and other layer on gold electrodes,” Analytica Chimica Acta, vol. 462, no.
Listeria species,” Biosensors and Bioelectronics, vol. 22, no. 8, pp. 1, pp. 39–47, 2002.
1664–1671, 2007.
[40] B. Meric, K. Kerman, D. Ozkan et al., “Electrochemical DNA
[24] P. Banerjee and A. K. Bhunia, “Cell-based biosensor for rapid biosensor for the detection of TT and hepatitis B virus from
screening of pathogens and toxins,” Biosensors and Bioelectron- PCR amplified real samples by using,” Talanta, vol. 56, no. 5,
ics, vol. 26, pp. 99–106, 2010. pp. 837–846, 2002.
[25] T. Curtis, R. M. Z. G. Naal, C. Batt, J. Tabb, and D. Holowka, [41] D. Ozkan, P. Kara, K. Kerman et al., “DNA and PNA sensing
“Development of a mast cell-based biosensor,” Biosensors and on mercury and carbon electrodes by using methylene blue as
Bioelectronics, vol. 23, no. 7, pp. 1024–1031, 2008. an electrochemical label,” Bioelectrochemistry, vol. 58, no. 1, pp.
[26] V. Nanduri, A. K. Bhunia, S. I. Tu, G. C. Paoli, and J. D. Brewster, 119–126, 2002.
“SPR biosensor for the detection of L. monocytogenes using
[42] W. Yang, J. J. Gooding, and D. B. Hibbert, “Characterisation of
phage-displayed antibody,” Biosensors and Bioelectronics, vol.
gold electrodes modified with self-assembled monolayers of L-
23, no. 2, pp. 248–252, 2007.
cysteine for the adsorptive stripping analysis of copper,” Journal
[27] E. Tully, S. P. Higson, and R. O’Kennedy, “The development of Electroanalytical Chemistry, vol. 516, no. 1-2, pp. 10–16, 2001.
of a “labeless” immunosensor for the detection of Listeria
monocytogenes cell surface protein, Internalin B,” Biosensors and
Bioelectronics, vol. 23, no. 6, pp. 906–912, 2008.
[28] F. Farabullini, F. Lucarelli, I. Palchetti, G. Marrazza, and
M. Mascini, “Disposable electrochemical genosensor for the
simultaneous analysis of different bacterial food contaminants,”
Biosensors and Bioelectronics, vol. 22, no. 7, pp. 1544–1549, 2007.
[29] Y. Jin, X. Yao, Q. Liu, and J. Li, “Hairpin DNA probe based elec-
trochemical biosensor using methylene blue as hybridization
indicator,” Biosensors and Bioelectronics, vol. 22, no. 6, pp. 1126–
1130, 2007.
[30] P. Kara, K. Kerman, D. Ozkan et al., “Electrochemical genosen-
sor for the detection of interaction between methylene blue and
DNA,” Electrochemistry Communications, vol. 4, no. 9, pp. 705–
709, 2002.
[31] K. Kerman, M. Kobayashi, and E. Tamiya, “Recent trends
in electrochemical DNA biosensor technology,” Measurement
Science and Technology, vol. 15, pp. R1–R11, 2004.
[32] F. Lucarelli, A. Kicela, I. Palchetti, G. Marrazza, and M. Mascini,
“Electrochemical DNA biosensor for analysis of wastewater
samples,” Bioelectrochemistry, vol. 58, no. 1, pp. 113–118, 2002.
[33] F. Lucarelli, G. Marrazza, A. P. F. Turner, and M. Mascini,
“Carbon and gold electrodes as electrochemical transducers for
DNA hybridisation sensors,” Biosensors and Bioelectronics, vol.
19, no. 6, pp. 515–530, 2004.
http://dx.doi.org/10.1155/2013/503805
Research Article
Brazilian Cerrado Soil Actinobacteria Ecology
Monique Suela Silva,1 Alenir Naves Sales,1 Karina Teixeira Magalhães-Guedes,1

Disney Ribeiro Dias,2 and Rosane Freitas Schwan1
1
Department of Biology, Federal University of Lavras (UFLA), Campus Universitário, 37.200-000 Lavras, MG, Brazil
2
Department of Food Science, Federal University of Lavras (UFLA), Campus Universitário, 37.200-000 Lavras, MG, Brazil
Correspondence should be addressed to Disney Ribeiro Dias; diasdr@dca.ufla.br
Received 15 October 2012; Revised 4 December 2012; Accepted 19 December 2012
Copyright © 2013 Monique Suela Silva et al. This is an open access article distributed under the Creative Commons Attribution
cited.
A total of 2152 Actinobacteria strains were isolated from native Cerrado (Brazilian Savannah) soils located in Passos, Luminárias,
and Arcos municipalities (Minas Gerais State, Brazil). The soils were characterised for chemical and microbiological analysis. The
microbial analysis led to the identification of nine genera (Streptomyces, Arthrobacter, Rhodococcus, Amycolatopsis, Microbacterium,
Frankia, Leifsonia, Nakamurella, and Kitasatospora) and 92 distinct species in both seasons studied (rainy and dry). The rainy
season produced a high microbial population of all the aforementioned genera. The pH values of the soil samples from the Passos,
Luminárias, and Arcos regions varied from 4.1 to 5.5. There were no significant differences in the concentrations of phosphorus,
magnesium, and organic matter in the soils among the studied areas. Samples from the Arcos area contained large amounts of
aluminium in the rainy season and both hydrogen and aluminium in the rainy and dry seasons. The Actinobacteria population
seemed to be unaffected by the high levels of aluminium in the soil. Studies are being conducted to produce bioactive compounds
from Actinobacteria fermentations on different substrates. The present data suggest that the number and diversity of Actinobacteria
spp. in tropical soils represent a vast unexplored resource for the biotechnology of bioactives production.
1. Introduction of the isolates are able to produce novel secondary metabo-

lites [1]. Since the first antibiotic from an Actinobacterium,
Actinobacteria are a distinct group of bacteria that are widely reported more than 50 years ago [7], more than 4000 new
distributed in nature [1]. Currently [2], Actinobacteria com- bioactive compounds have been discovered. The search for
prise eight groups with 48 genera. Special attention has been new species or strains of the Actinobacterium group is still of
given to this bacterial group in biotechnological applica- primary interest to the biotechnology field.
tions, which are a natural result of their great metabolic Actinobacteria taxonomy is extremely complex, and clas-
diversity [3]. Actinobacteria are the most common source of sification using only the traditional methods, which are
antibiotics [4] and are a promising source of a wide range based on morphological and physiological characteristics,
of enzymes, enzyme inhibitors, immunomodifiers, and vita- has led to very heterogeneous suprageneric groups. Recently,
mins [5]. In nature, Actinobacteria play an important role in three main approaches have been suggested to identify
the cycling of organic compounds and have been associated species of Actinobacteria: chemotaxonomy (differentiation
with soil organic matter production, including production of of species by chemical composition), numerical taxonomy
the black pigments called melanin, which are related to soil (differentiation of species by phenotypic similarity number),
humic acid [6]. and molecular systematics (use of DNA study to the species
Streptomyces is the most common Actinobacteria genus differentiation). A combination of three techniques becomes
found in soils and constitutes up to 90% of the isolates. How- more complete [1].
ever, new approaches for the isolation of soil Actinobacteria Brazilian Cerrado soils have an enormous biodiversity
have revealed that other genera are also present in significant potential. Some of these soils have been described as habitats
numbers. Many new species have been isolated, and most with high biological activity but have not been extensively
(b)
Sampling point (SP) georeferenced

Extension of the Cerrado Subsample points around SP
( km)
0 360 3m
MAPA L.F.MARTW 6m
(a)
(c)
Figure 1: (a) Location of Cerrado soil in Minas Gerais, Brazil. (b) Cities where samples were collected. (c) Distribution of sampling point.
Sampling point scheme: one composed soil sample (12 subsamples) was collected around each sampling point.
explored for the search and discovery of novel Actinobacteria Table 1: Location and description of the Brazilian Cerrado soil
spp. In this study, isolation of Actinobacteria species from collection sites.
Brazilian Cerrado soils and comparisons of the Actinobacte- Site name Location
ria communities of the Cerrado soil with the physicochemical
Region of Passos
characteristics of these soils were performed. Some of the
Point 1 20∘ 49󸀠 57.7󸀠󸀠 S; 046∘ 30󸀠 29.3󸀠󸀠 W
isolates that showed promise for use in biotechnology were
identified and tested for the production of bioactives (e.g., Point 2 20∘ 49󸀠 56.8󸀠󸀠 S; 046∘ 30󸀠 30.1󸀠󸀠 W
enzymes and antibiotics). Point 3 20∘ 49󸀠 48.0󸀠󸀠 S; 046∘ 30󸀠 54.9󸀠󸀠 W
Point 4 20∘ 49󸀠 47.1󸀠󸀠 S; 046∘ 30󸀠 54.5󸀠󸀠 W
Point 5 20∘ 49󸀠 47.8󸀠󸀠 S; 046∘ 30󸀠 51.5󸀠󸀠 W
2. Materials and Methods
Region of Luminárias
2.1. Soil Sampling. Thirty composite soil samples were col- Point 6 21∘ 37󸀠 51.0󸀠󸀠 S; 044∘ 58󸀠 22.7󸀠󸀠 W
lected during the months of January (rainy season) and Point 7 21∘ 37󸀠 50.6󸀠󸀠 S; 044∘ 58󸀠 22.7󸀠󸀠 W
August (dry season) from the Passos, Arcos, and Luminárias Point 8 21∘ 37󸀠 51.5󸀠󸀠 S; 044∘ 59󸀠 11.0󸀠󸀠 W
municipalities. These soils are highly conserved (native), and Point 9 21∘ 37󸀠 55.3󸀠󸀠 S; 044∘ 59󸀠 29.3󸀠󸀠 W
the locations are georeferenced in Table 1 and Figure 1. Each Point 10 21∘ 37󸀠 54.6󸀠󸀠 S; 044∘ 59󸀠 54.0󸀠󸀠 W
sample was obtained according to the procedure of Lima et al. Region of Arcos
[8]. Twelve subsamples of 75 to 100 g were collected from each
Point 11 20∘ 16󸀠 27.7󸀠󸀠 S; 045∘ 29󸀠 14.6󸀠󸀠 W
point in two concentric circles with radii of 3 and 6 m from
Point 12 20∘ 14󸀠 47.9󸀠󸀠 S; 045∘ 25󸀠 35.9󸀠󸀠 W
the centre and a depth of 0 to 20 cm using a flamed auger. The
collected material from each point was mixed in a sterile bag Point 13 20∘ 14󸀠 51󸀠󸀠 S; 045∘ 31󸀠 40.8󸀠󸀠 W
and stored at 4∘ C until analysis. Point 14 20∘ 14󸀠 48.6󸀠󸀠 S; 045∘ 31󸀠 33.4󸀠󸀠 W
Point 15 20∘ 14󸀠 58.0󸀠󸀠 S; 045∘ 31󸀠 54.0󸀠󸀠 W
2.2. Physicochemical Analysis of Soils. Approximately 200 g of
each soil sample was subjected to physicochemical analysis
using the procedure of the Embrapa [9]. The concentrations regions and the physicochemical soil variables were subjected
of potassium (K), phosphorus (P), aluminium (Al), magne- to statistical analysis (principal component analysis or PCA)
sium (Mg), organic matter (OM), hydrogen and aluminium using the Unscrambler 9.7 software (CAMO, Oslo, Norway).
(Al + H) and exchangeable bases (SB) as well as the pH and
soil texture were evaluated. The Sisvar 5.1 program of the 2.3. Bacterial Isolation and Culture Purification. Ten grams of
SAS System 9.1 software (SAS Institute Inc., Cary, NC, USA) soil from each composite sample was added to 90 mL of sterile
was used for statistical analysis of the differences between peptone water (bacto peptone, 1 g/L) and homogenised by
the means. Correlations between the Brazilian Cerrado soils’ stirring at 130 rpm for 10 min (dilutions of 10−1 to 10−8 ). These
Table 2: Actinobacterial-specific primer used in bacterial communities in the Brazilian Cerrado soils, according to Schäfer et al. [12].
Amplified
Primer Sequence (5󸀠 –3󸀠 ) Primers PCR conditions
fragment (bp)
27f GAG TTT GAT CMT GGC TCA G Denatured for 5 min at
Bacterial universal primer ∼1500 95∘ C. 30 cycles: denaturing
1492r ACG GYT ACC TTG TTA CGA CTT
at 92∘ C for 60 s, annealing
Com2xf AAA CTC AAA GGA ATT GAC GG Actinobacterial-specific at 55∘ C for 60 s, and
∼270
Ac1186r CTT CCT CCG AGT TGA CCC primer extension at 72∘ C for 60 s;
SC-Act-235aS20 CGC GGC CTA TCA GCT TGT TG Actinobacterial-specific final extension for 10 min at
∼640 72∘ C
SC-Act-878aA19 CCG TAC TCC CCA GGC GGG G primer
sample dilutions were used for inoculations by spreading cluster analysis using the Bionumerics 2.50 software (Applied
100 𝜇L on the surface of Aaronson’s medium according to Maths, Sint-Martens-Latem, Belgium).
Silva et al. [10] and humic acid vitamin medium according
to Hayakawa and Nonomura [11]. The plates were incubated
2.6. PCR Amplification and DNA Sequencing of the 16S
for 72 h to 120 h at 28∘ C.
rRNA Gene. Representative isolates of each Rep-PCR
From the plates containing 30 to 300 CFU, a number of
profile were selected for amplification of the 16S rRNA
colonies equal to the square root of the number of different
gene as described by Pereira et al. [13]. DNA (2 𝜇L) was
quantified of each colonial morphotypes were isolated [13].
added to 30 𝜇L of Taq PCR Master Mix (Qiagen,
These morphotype strains were cultured for 72 to 120 h at
São Paulo, Brazil), 26 𝜇L of H2 O, 1 𝜇L of primer 27f
28∘ C in 500 𝜇L of nutrient broth supplemented with glycerol
(5󸀠 -AGAGTTTGATCCTGGCTCAG-3󸀠 ), and 1 𝜇L of primer
to a final concentration of 20%. The isolates were purified
1512r (5󸀠 -ACGGCTACCTTGTTACGACT-3󸀠 ). The PCR
by successive restreak and were preserved by freezing at
reaction was performed as follows: initial denaturation at
−20∘ C. The strains were reactivated on nutrient agar by
95∘ C for 10 min; 25 cycles at 93∘ C for 1 min, 50∘ C for 1 min,
incubating for 72 to 120 h at 28∘ C and were then characterised
and 72∘ C for 1 min 30 s; and a final elongation at 72∘ C for
for bacterial colony morphology (i.e., size, shape, eleva-
5 min. The presence of PCR products was confirmed by
tion, brightness, texture and colour) by making comparison
electrophoresis on a 1% agarose gel in 1x TAE buffer at 70 V
between the colonies that were originally isolated from the
for 30 min, stained with SYBR Green (Invitrogen, Foster
culture media. The pure cultures were preserved under the
City, CA, USA), and visualised under a transilluminator.
conditions described above.
The sequencing of amplicons was performed at Macrogen
Inc. (Seoul, South Korea), and the sequences were compared
2.4. PCR Primer System for Selective Amplification of Acti- with the GenBank database using the BLAST algorithm
nobacteria. Pure cultures of the various colonial morpho- (http://www.ncbi.nlm.nih.gov/BLAST/).
types were characterised by actinobacterial-specific primers
according to Schäfer et al. [12] and described in Table 2.
The 27f and 1492r universal bacterial primers were used 3. Results
as controls. The actinobacterial strains were subjected to
3.1. Physicochemical Characteristics of the Brazilian Cerrado
molecular characterisation by REP-PCR as described below.
Soil Samples. The chemical and biochemical properties of the
Cerrado soil from the Passos, Luminárias, and Arcos regions
2.5. Molecular Characterisation Based on Repetitive Extra- during the rainy and dry seasons are shown in Table 3. The
genic Palindromic-PCR (Rep-PCR). Total genomic DNA was pH values of the soils ranged from 4.7 to 5.5, 5.0 to 5.4,
extracted as described by Pereira et al. [13]. The molecular and 4.1 to 5.0 for the Passos, Luminárias, and Arcos regions,
characterisation of selected isolates was performed by poly- respectively. These soils had high acidity.
merase chain reaction sequencing by REP-PCR as described Differences in the soil textures from the Arcos, Passos,
by Gevers et al. [14]. Two microliters of DNA were added to and Luminárias regions were observed. The three analysed
12.5 𝜇L of Taq PCR Master Mix (Qiagen, São Paulo, Brazil), areas revealed no significant differences in organic matter
8 𝜇L H2 O, 0.25 𝜇L bovine serum albumin (BSA), 0.25 𝜇L of values. In general, the physical and chemical characteristics
formamide, and 2 𝜇L of primer GTG5 (5󸀠 -GTG GTG GTG for all the soils analysed were similar.
GTG GTG-3󸀠 ) [13]. PCR was performed under the following A multivariate analysis using frequency values for the
cycling conditions: 5 min initial denaturation at 94∘ C; 30 chemical characteristics of the Brazilian Cerrado soils was
cycles of 95∘ C for 30 s, 45∘ C for 60 s and 60∘ C for 5 min; performed (Figure 2). Samples obtained from the Arcos
and a final elongation at 60∘ C for 16 min. The PCR products region were significantly different from the Luminárias and
were separated by electrophoresis on a 2% agarose gel in Passos regional samples because of the high concentration of
1x TAE buffer at 60 V for 4 h, stained with SYBR Green phosphorus during the rainy season and high concentrations
(Invitrogen, Foster City, CA, USA), and visualised under of aluminium and potassium during the dry season in the
a transilluminator. The Rep-PCR profiles were subjected to Arcos region.
4
Table 3: Chemical and physical characteristics of the Brazilian Cerrado soil samples.
Sample and Season pH P (mg/dm3 ) K (mg/dm3 ) Mg (mg/dm3 ) Al (mg/dm3 ) H + Al (Cmol/dm3 ) OM (dag/Kg) SB (mg/dm3 ) Texture
Point 1 5.3 ± 0.1a 1.5 ± 0.1a 25 ± 1a 0.1 ± 0.0a 0.6 ± 0.1a 3.6 ± 0.1a 1.4 ± 0.1a 0.3 ± 0.1a Sandy loam
Point 2 5.4 ± 0.1a 1.5 ± 0.1a 56 ± 2a 0.1 ± 0.0a 0.6 ± 0.1a 4.5 ± 0.1a 2.0 ± 0.1a 0.4 ± 0.1a Medium loam
PA Rainy Point 3 5.5 ± 0.1a 1.2 ± 0.1a 33 ± 1a 0.2 ± 0.0a 0.4 ± 0.1a 2.6 ± 0.1a 1.1 ± 0.1a 0.3 ± 0.1a Medium loam
Point 4 5.5 ± 0.1a 1.0 ± 0.1a 70 ± 1b 0.1 ± 0.0a 0.5 ± 0.1a 3.6 ± 0.1a 1.5 ± 0.1a 0.5 ± 0.1a Medium loam
Point 5 5.4 ± 0.1a 0.7 ± 0.1a 9 ± 1b 0.1 ± 0.0a 0.1 ± 0.1a 1.7 ± 0.1a 0.4 ± 0.1b 0.2 ± 0.1a Medium loam
Point 6 5.4 ± 0.1a 1.2 ± 0.1a 28 ± 1a 0.2 ± 0.0a 0.5 ± 0.1a 7.9 ± 0.1a 3.4 ± 0.1a 0.3 ± 0.1a Clay loam
LU Rainy Point 8 5.1 ± 0.1a 1.2 ± 0.1a 11 ± 1b 0.2 ± 0.0a 0.3 ± 0.1a 2.6 ± 0.1a 1.1 ± 0.1a 0.2 ± 0.1a Sandy loam
Point 9 5.2 ± 0.1a 2.0 ± 0.1a 20 ± 1a 0.1 ± 0.0a 0.9 ± 0.2a 7.0 ± 0.1a 2.4 ± 0.1a 0.3 ± 0.1a Medium loam
Point 10 5.1 ± 0.1a 1.5 ± 0.1a 34 ± 1a 0.1 ± 0.0a 0.8 ± 0.1a 8.8 ± 0.3b 2.7 ± 0.1a 0.3 ± 0.1a Clay loam
AC Rainy Point 13 4.1 ± 0.1a 1.8 ± 0.1a 27 ± 1a 0.3 ± 0.0a 2.1 ± 0.1b 15.3 ± 1b 3.4 ± 0.1a 0.3 ± 0.1a Clay loam
Point 14 4.1 ± 0.1a 1.8 ± 0.1a 33 ± 1a 0.1 ± 0.0a 2.4 ± 0.1b 17.1 ± 2b 4.0 ± 0.1a 0.3 ± 0.1a Clay loam
Point 15 5.0 ± 0.1a 1.8 ± 0.1a 69 ± 2b 0.1 ± 0.0a 1.8 ± 0.1b 12.3 ± 1b 2.7 ± 0.1a 0.4 ± 0.1a Clay loam
Point 1 4.7 ± 0.1a 1.7 ± 0.1a 113.8 ± 1b 0.1 ± 0.0a 0.2 ± 0.1a 13.7 ± 0.1b 3.9 ± 0.1a 0.5 ± 0.1a Sandy loam
Point 2 5.1 ± 0.1a 1.7 ± 0.1a 88.9 ± 1b 0.1 ± 0.0a 0.4 ± 0.1a 5.6 ± 0.1a 2.4 ± 0.1a 0.7 ± 0.1a Medium loam
PA Dry Point 3 5.1 ± 0.1a 1.4 ± 0.1a 137.28 ± 1b 0.1 ± 0.0a 0.4 ± 0.1a 4.5 ± 0.1a 2.2 ± 0.1a 0.8 ± 0.1a Medium loam
Point 6 5.1 ± 0.1a 2.5 ± 0.1a 37.4 ± 1a 0.1 ± 0.0a 0.6 ± 0.1a 6.3 ± 0.1a 2.2 ± 0.1a 0.1 ± 0.1a Clay loam
Point 7 5.1 ± 0.1a 0.9 ± 0.1a 37.4 ± 1a 0.1 ± 0.0a 1.5 ± 0.1b 7.0 ± 0.1a 2.8 ± 0.1a 0.2 ± 0.1a Clay loam
LU Dry Point 8 5.2 ± 0.1a 0.9 ± 0.1a 39 ± 1a 0.1 ± 0.0a 0.7 ± 0.1a 7.8 ± 0.1a 2.8 ± 0.1a 0.2 ± 0.1a Sandy loam
Point 9 5 ± 0.1a 1.2 ± 0.1a 46 ± 1a 0.1 ± 0.0a 1.5 ± 0.1b 10.9 ± 0.1b 3.0 ± 0.1a 0.3 ± 0.1a Medium loam
Point 10 5 ± 0.1a 1.2 ± 0.1a 67 ± 1b 0.1 ± 0.0a 1.4 ± 0.1b 9.88 ± 0.1b 3.1 ± 0.1a 0.3 ± 0.1a Clay loam
Point 11 4.7 ± 0.1a 2.0 ± 0.1a 149.7 ± 1b 0.1 ± 0.0a 0.4 ± 0.1a 8.7 ± 0.1a 2.4 ± 0.1a 1.3 ± 0.1a Clay loam
Point 12 4.8 ± 0.1a 1.4 ± 0.1a 48.3 ± 1a 0.6 ± 0.0a 0.1 ± 0.1a 7.0 ± 0.1a 1.8 ± 0.1a 0.2 ± 0.1a Clay loam
AC Dry Point 13 4.3 ± 0.1a 1.4 ± 0.1a 54.6 ± 1a 0.1 ± 0.0a 0.1 ± 0.1a 15.3 ± 1b 2.8 ± 0.1a 0.3 ± 0.1a Clay loam
Point 14 4.2 ± 0.1a 1.7 ± 0.1a 39.0 ± 1a 0.1 ± 0.0a 0.1 ± 0.1a 17.1 ± 1b 3.0 ± 0.1a 0.2 ± 0.1a Clay loam
Point 15 4.8 ± 0.1a 1.2 ± 0.1a 84.2 ± 2b 0.1 ± 0.0a 0.1 ± 0.1a 10.9 ± 1b 1.9 ± 0.1a 0.4 ± 0.1a Clay loam
Data are average values of duplicate ± standard deviation. Different letters indicate significant differences (𝑃 < 0.05). Soil classification in sandy (content clay <15), Medium (content clay between 15 and 35), and
clay (content clay ≥35). Abbreviations: PA: Passos; LU: Luminárias; AC: Arcos. K: potassium; P: phosphorus; Al: aluminum; Ca: calcium; Mg: magnesium; H + Al: hydrogen + aluminum; OM: organic matter; SB:
(exchangeable bases) the sum of Ca, Mg, Na, and K.
Principal components 1 and 2 (79.08%)

8
6
Principal component 2 (25.28%)
4 PAS
K Mg
2
ARS
Al
0 Ca Phi PAC
MO LUMS
−2 P LUMC
ARC
−4
−6
−10 −8 −6 −4 −2 0 2 4 6 8 10
Principal component 1 (53.8%)
Figure 2: Principal component analysis (PCA) of chemical charac-

teristics of the Brazilian Cerrado soil of Minas Gerais. Abbreviations:
K = potassium; P = phosphorus; Al = aluminum, Ca = calcium, Mg
= magnesium; PAC = Passos (rainy season); PAS = Passos (dry
season); ARC = Arcos (rainy season); ARS = Arcos (dry season);
LUMC = Luminárias (rainy season); LUMS = Luminárias (dry
season).
Table 4: Actinobacteria count of the population in log CFU/g of soil

in differents medium during the rainy and dry season.
Region
Aaronsons’s medium Humic acid vitamin medium
Dry season Figure 3: Similarity analysis between the bands’ profiles (Rep-PCR)
Arcos 7.9 ± 0.1a 6.6 ± 0.1b of the Actinobacteria isolates of the Brazilian Cerrado soils of three
Luminárias 7.8 ± 0.1a 6.8 ± 0.1b regions: Arcos, Passos, and Luminárias. ( ) OTUs quantification.
Passos 7.8 ± 0.1a 6.7 ± 0.1b
Rainy season
Arcos 9.1 ± 0.2c 7.2 ± 0.1d 3.3. Identification and Distribution of Isolates. The analysis of
Luminárias 8.9 ± 0.2c 7.1 ± 0.1d
the 16S rRNA gene sequence amplification products led to the
identification of nine Actinobacteria genera (Streptomyces,
Passos 8.8 ± 0.1c 7.2 ± 0.2d
Artrobacter, Rhodococcus, Amycolatopsis, Microbacterium,
Data are mean values of duplicate ± standard deviation.
Frankia, Leifsonia, Nakamurella, and Kitasatospora) and 92
Different letters indicate significant differences (𝑃 < 0.05).
distinct species within the genera (Figure 3 and Table 5).
The genera Streptomyces, Artrobacter, Rhodococcus, Amy-
colatopsis, and Microbacterium were found in all three regions
3.2. Microbial Isolation and Characterisation. Of the culture analysed. Leifsonia, Nakamurella, and Kitasatospora were
media tested (Aaronsons’s medium and humic acid vitamin found only in the Arcos region, and the genus Frankia was
medium), all were able to recover colonies from all of the soil not found in the Passos region (Table 5).
samples (Table 4). The distributions of Actinobacteria genera were different
Total bacterial counts were compared between the rainy in the Cerrado soils during the rainy and dry seasons
and dry seasons. A statistically significant difference (𝑃 < (776 isolates were from the dry season, and 1376 isolates
0.05) was observed for all the analysed areas (Table 4). The were from the rainy season) (Table 5). The rainy season
rainy season exhibited higher microbial counts (∼9.1 log produced a higher microbial population for all described
CFU/g) compared to the dry season (∼7.9 log CFU/g). genera (Figure 4).
A total of 2152 isolates were characterised. The isolates
were selected from the groups and subjected to group analysis 4. Discussion
corresponding to each region. The selected isolates from each
region (188 isolates) were characterised by Rep-PCR, and 78 The Cerrado biome has two distinct seasons: the dry season
different band profiles were obtained (Figure 3). (May to September) and the rainy season (November to
6
Table 5: Actinobacteria diversity in different Cerrado regions of Minas Gerais, Brazil. OTUs and abundance quantification.
Region Season Total abundance Actinobacteria
Arthrobacter nitroguajacolicus HQ202810.1 (12), Arthrobacter methylotrophus NR025083.1 (9), Arthrobacter aurescens JN662517.1 (12),
Arthrobacter oxydans DQ122301.1 (12), Streptomyces yokosukanensis NR043496.1 (7), Streptomyces griseochromogenes subsp. suitaensis
Rainy 153 NR043851.1 (10), Streptomyces globifer AB184472.2 (10), Streptomyces viridochromogenes AB045858.1 (14), Arthrobacter boritolerans
AB288059.1 (12), Arthrobacter ilicis FR87442.1 (12), Arthrobacter methylotrophus NR025083.1 (12), Arthrobacter alkaliphilus NR041401.1
Passos (12), Streptomyces globosus HM230830.1 (19)
Arthrobacter nitroguajacolicus HQ202810.1 (10), Arthrobacter methylotrophus NR025083.1 (11), Arthrobacter aurescens JN662517.1 (11),
Arthrobacter oxydans DQ122301.1 (9), Streptomyces yokosukanensis NR043496.1 (12), Streptomyces griseochromogenes subsp. suitaensis
Dry 105 NR043851.1 (10), Streptomyces globifer AB184472.2 (12), Streptomyces viridochromogenes AB045858.1 (5), Arthrobacter boritolerans
AB288059.1 (4), Arthrobacter ilicis FR87442.1 (4), Arthrobacter methylotrophus NR025083.1 (1), Arthrobacter alkaliphilus NR041401.1
(9), Streptomyces globosus HM230830.1 (7)
Streptomyces gelaticus EU741111.1 (12), Amycolatopsis hippodromi HQ021203.1 (14), Amycolatopsis circi HQ021202.1 (12), Amycolatopsis
equine HQ021204.1 (12), Arthrobacter nicotinovorans GQ284335.1 (16), Arthrobacter alkaliphilus NR041401.1 (12), Arthrobacter
nitroguajacolicus HQ202810.1 (10), Arthrobacter aurescens JN662517.1 (10), Arthrobacter methylotrophus NR025083.1 (11), Streptomyces
sp. SIIA 2050 EF657884.1 (10), Arthrobacter sp. ACT2G HM063855.1 (3), Streptomyces sp. BK14 EF657884.1 (19), Streptomyces
roseogriseus AB184461.2 (17), Streptomyces hygroscopicus EU367978.1 (18), Streptomyces corchorusii FJ461617.1 (15), Streptomyces
olivaceoviridis NR041108.1 (12), Arthrobacter alkaliphilus NR041401.1 (12), Arthrobacter ramosus AM039435.1 (13), Microbacterium sp.
Rainy 414
PSSUR4 JF768719.1 (9), Frankia alni ACN14a CT573213.2 (11), Streptomyces showdoensis AY999741.1 (12), Streptomyces termitum
HQ680451.1 (14), Streptomyces violaceorectus EU570697.1 (14), Streptomyces lipmanii AB045861.1 (15), Streptomyces laurentii FJ972687.1
(13), Streptomyces carpinensis DQ026664.1 (15), Microbacterium phyllosphaerae JF700409.1 (10), Microbacterium azadirachtae
EU912487.1 (12), Arthrobacter boritolerans AB288059.1 (11), Streptomyces castaneoglobisporus AB184452.2 (9), Streptomyces
malachitofusus NR041105.1 (6), Rhodococcus erythropolis EF608231.1 (5), Rhodococcus qingshengii HQ202829.1 (6), Rhodococcus
Luminárias globerulus GU332596.1 (12), Streptomyces aureus EU841581.1 (12),
Streptomyces gelaticus EU741111.1 (10), Amycolatopsis hippodromi HQ021203.1 (9), Amycolatopsis circi HQ021202.1 (7), Amycolatopsis
equine HQ021204.1 (5), Arthrobacter nicotinovorans GQ284335.1 (4), Arthrobacter alkaliphilus NR041401.1 (10), Arthrobacter
nitroguajacolicus HQ202810.1 (11), Arthrobacter aurescens JN662517.1 (9), Arthrobacter methylotrophus NR025083.1 (12), Streptomyces
sp. SIIA 2050 EF657884.1 (9), Arthrobacter sp. ACT2G HM063855.1 (1), Streptomyces sp. BK14 EF657884.1 (7), Streptomyces
roseogriseus AB184461.2 (8), Streptomyces hygroscopicus EU367978.1 (6), Streptomyces corchorusii FJ461617.1 (5), Streptomyces
olivaceoviridis NR041108.1 (8), Arthrobacter alkaliphilus NR041401.1 (4), Arthrobacter ramosus AM039435.1 (12), Microbacterium sp.
Dry 245
PSSUR4 JF768719.1 (7), Frankia alni ACN14a CT573213.2 (9), Streptomyces showdoensis AY999741.1 (10), Streptomyces termitum
HQ680451.1 (10), Streptomyces violaceorectus EU570697.1 (11), Streptomyces lipmanii AB045861.1 (9), Streptomyces laurentii FJ972687.1
(12), Streptomyces carpinensis DQ026664.1 (12), Microbacterium phyllosphaerae JF700409.1 (4), Microbacterium azadirachtae
EU912487.1 (7), Arthrobacter boritolerans AB288059.1 (2), Streptomyces castaneoglobisporus AB184452.2 (2) ,Streptomyces
malachitofusus NR041105.1 (2), Rhodococcus erythropolis EF608231.1 (2), Rhodococcus qingshengii HQ202829.1 (2), Rhodococcus
globerulus GU332596.1 (3), Streptomyces aureus EU841581.1 (4),
Table 5: Continued.
Region Season Total abundance Actinobacteria
Arthrobacter ramosus AM039435.1 (12), Amycolatopsis albidoflavus AB327251.1 (12), Arthrobacter methylotrophus NR025083.1 (17),
Amycolatopsis hippodromi HQ021203.1 (12), Streptomyces castaneoglobisporus HQ607438.1 (12), Streptomyces cavourensis subsp.
cavourensis NR043851.1 (17), Microbacterium sp. 4N2-2 HQ833039.1 (12), Arthrobacter sp. LC9 AB248530.1 (10), Streptomyces
michiganensis NR041071.1 (11), Streptomyces xanthochromogenes NR043847.1 (12), Microbacterium azadirachtae EU912487.1 (12),
Microbacterium phyllosphaerae JF700409.1 (10), Arthrobacter alkaliphilus NR041401.1 (12), Streptomyces xanthochromogenes
NR043847.1 (19), Nakamurella multipartita DSM 44233 CP001737.1 (12), Arthrobacter sp. S2.TSA.004 HM063855.1 (10), Leifsonia
shinshuensis DQ334864.1 (18), Streptomyces olivochromogenes EU841608.1 (10), Streptomyces crystallinus NR041177.1 (9), Streptomyces
phaeochromogenes EU594477.1 (12), Streptomyces microflavus AB045861.1 (15), Streptomyces termitum HQ680451.1 (12), Streptomyces
showdoensis AY999741.1 (18), Arthrobacter nitroguajacolicus GQ181046.1 (17), Streptomyces sp. 4728 DQ487015.1 (11), Streptomyces
anulatus HQ995503.1 (12), Streptomyces flavogriseus CP002475.1 (15), Streptomyces chromofuscus FJ486284.1 (12), Streptomyces
Rainy 865 gilvosporeus FJ196597.1 (12), Streptomyces natalensis AB184356.2 (12), Streptomyces albulus JN566022.1 (12), Arthrobacter sp. 21S1
AB248530.1 (12), Arthrobacter sp. ACT2K AB248530.1 (12), Streptomyces melanogenes NR041089.1 (12), Arthrobacter sp. S5.TSA.012
AB248530.1 (12), Arthrobacter sp. J9 AB248530.1 (16), Streptomyces globosus HM230830.1 (12), Rhodococcus qingshengii HQ202829.1
(12), Rhodococcus erythropolis GU726138.1 (12), Rhodococcus globerulus HM217119.1 (12), Streptomyces cavourensis HQ610450.1 (12),
Streptomyces bacillaris FJ792550.1 (12), Streptomyces mauvecolor JN180187.1 (17), Streptomyces rameus AB184798.1 (22), Streptomyces
spinichromogenes AB184534.1 (22), Streptomyces gelaticus EU741111.1 (12), Streptomyces pulveraceus EU240417.1 (12), Streptomyces
nodosus subsp. asukaensis AB184497.1 (12), Streptomyces sp. HB200 EF657884.1 (12), Arthrobacter aurescens GU171380.1 (20),
bacterium Ellin 5075 AY234492.1 (12), Amycolatopsis rubida NR025072.1 (12), Amycolatopsis circi HQ021202.1 (19), Amycolatopsis
equine HQ021204.1 (12), Streptomyces laurentii FJ972687.1 (12), Streptomyces showdoensis EU124781.1 (12), Streptomyces sp. RB72
EF657884.1 (17), Streptomyces xanthocidicus NR043370.1 (12), Streptomyces tsukiyonensis AB184594.1 (14), Kitasatospora kifunensis
AJ781341.1 (12), Streptomyces recifensis HM062992.1 (19)
Arcos
Arthrobacter ramosus AM039435.1 (10), Amycolatopsis albidoflavus AB327251.1 (10), Arthrobacter methylotrophus NR025083.1 (12),
Amycolatopsis hippodromi HQ021203.1 (12), Streptomyces castaneoglobisporus HQ607438.1 (9), Streptomyces cavourensis subsp.
cavourensis NR043851.1 (8), Microbacterium sp. 4N2-2 HQ833039.1 (12), Arthrobacter sp. LC9 AB248530.1 (12), Streptomyces
michiganensis NR041071.1 (12), Streptomyces xanthochromogenes NR043847.1 (12), Microbacterium azadirachtae EU912487.1 (12),
Microbacterium phyllosphaerae JF700409.1 (3), Arthrobacter alkaliphilus NR041401.1 (4), Streptomyces xanthochromogenes NR043847.1
(3), Nakamurella multipartita DSM 44233 CP001737.1 (2), Arthrobacter sp. S2.TSA.004 HM063855.1 (12), Leifsonia shinshuensis
DQ334864.1 (10), Streptomyces olivochromogenes EU841608.1 (12), Streptomyces crystallinus NR041177.1 (12), Streptomyces
phaeochromogenes EU594477.1 (12), Streptomyces microflavus AB045861.1 (12), Streptomyces termitum HQ680451.1 (12), Streptomyces
showdoensis AY999741.1 (2), Arthrobacter nitroguajacolicus HQ202810.1 (2), Streptomyces sp. 4728 DQ487015.1 (12), Streptomyces
anulatus HQ995503.1 (3), Streptomyces flavogriseus CP002475.1 (19), Streptomyces chromofuscus FJ486284.1 (13), Streptomyces
Dry 370 gilvosporeus FJ196597.1 (14), Streptomyces natalensis AB184356.2 (7), Streptomyces albulus JN566022.1 (5), Arthrobacter sp. 21S1
AB248530.1 (3), Arthrobacter sp. ACT2K AB248530.1 (9), Streptomyces melanogenes NR041089.1 (12), Arthrobacter sp. S5.TSA.012
AB248530.1 (12), Arthrobacter sp. J9 AB248530.1 (12), Streptomyces globosus HM230830.1 (12), Rhodococcus qingshengii HQ202829.1
(7), Rhodococcus erythropolis GU726138.1 (2), Rhodococcus globerulus HM217119.1 (1), Streptomyces cavourensis HQ610450.1 (2),
Streptomyces bacillaris FJ792550.1 (1), Streptomyces mauvecolor JN180187.1 (1), Streptomyces rameus AB184798.1 (1), Streptomyces
spinichromogenes AB184534.1 (2), Streptomyces gelaticus EU741111.1 (2), Streptomyces pulveraceus EU240417.1 (2), Streptomyces nodosus
subsp. asukaensis AB184497.1 (2), Streptomyces sp. HB200 EF657884.1 (2), Arthrobacter aurescens GU171380.1 (12), bacterium Ellin
5075 AY234492.1 (1), Amycolatopsis rubida NR025072.1 (3), Amycolatopsis circi HQ021202.1 (4), Amycolatopsis equine HQ021204.1 (3),
Streptomyces laurentii FJ972687.1 (3), Streptomyces showdoensis EU124781.1 (5), Streptomyces sp. RB72 EF657884.1 (6), Streptomyces
xanthocidicus NR043370.1 (4), Streptomyces tsukiyonensis AB184594.1 (3), Kitasatospora kifunensis AJ781341.1 (2), Streptomyces
recifensis HM062992.1 (1)
( ) Distribution of isolates OTUs. Access Number—99% of similarity.
7
Rainy season Dry season
Streptomyces
400 Streptomyces
150
Kitasatospora 300 Arthrobacter Arthrobacter
Kitasatospora
100
200
100 50
Nakamurella Rhodococcus Nakamurella Rhodococcus
0 0
Leifsonia Amycolatopsis Leifsonia Amycolatopsis
Frankia Microbacterium Frankia Microbacterium

Arcos Arcos
Passos Passos
Figure 4: Actinobacteria genera abundance distribution in Cerrado soil in the rainy and dry seasons.
April). An important factor relevant to this study is that the The Arcos regional soil contained large amounts of alu-
soil samples were collected at the peak of each season; thus, minium (∼2.4 mg/dm3 ) during the rainy season and large
the distinction between the samples was maximised because amounts of hydrogen and aluminium during the rainy and
water was either limited or abundant [15]. The samples dry seasons (∼15 Cmol/dm3 ) (Table 3). High quantities of
were collected in January (high rainfall, 1000–2100 mm) and soluble aluminium in the soil can cause toxicity in plants,
August (low rainfall, 20–200 mm) of 2010. An important as aluminium competes with other elements (e.g., essential
caveat is the heterogeneity of the distribution of microor- nutrients) for the same chemical sites and promotes soil
ganisms in the soil because microbial growth is usually impoverishment [22]. However, in this study, the Microbiota
observed in patches rather than homogeneously [16–18]. We was not affected by high aluminium levels in the soil based
minimised this potential bias by collecting samples from on the similarity of this population to those at the other
different spots in each area studied and mixing the individual sampled sites. The three analysed areas (Luminárias, Arcos,
samples to obtain a composite sample. and Passos) displayed no significant differences in organic
Microorganisms are the key drivers of biogeochemical matter contents, which may be due to the similarity in the
processes in the soil. Thus, it is important to evaluate the vegetation and riverbank forest profiles of the three areas [15].
physicochemical properties of the soil and how these prop- Brazilian Cerrado soils cover a vast area, representing up
erties could be related to microbial profiles in different soils to 25% of the country [1]. Despite having a low pH (∼5.0),
[19]. These changes in the soil affect the native microbial large amounts of aluminium and iron, and a low nutrient
populations. Seasonal variations in the moisture and pH of content, these soils are extensively used in agriculture. One
the soil can lead to changes in the distribution patterns of of the few studies examining the microbial population of
the microbial species. For example, bacteria prefer neutral these soils reported on high numbers of Actinobacteria
to alkaline conditions, whereas yeasts and filamentous fungi [21]. Huddleston and collaborators [23] found a culturable
prefer acidic conditions. Some microbial species also have streptomycete population of approximately 8.0 log CFU/g
preferences for soils with high or low moisture contents [20]. in soil. In the previous study, the number of Actinobac-
These Cerrado soils had high acidity, consistent with the val- teria isolated from those soils was on the same order of
ues found in Cerrado soils that have been reported by studies magnitude (∼8.8 log CFU/g of soil) as the number isolated
of others [15, 21]. The Cerrado soils are commonly acids. from soils of the Brazilian Cerrado regions of the Minas
This may be due to the vegetative and microbial population Gerais State (Luminárias, Passos, and Arcos). These findings
present [15, 21]. Despite having low pH, large amounts of suggest that the Cerrado soils represent a large, unexplored
aluminium and iron, and low nutrient content, these soils are environment for the potential isolation of Actinobacteria.
extensively used in agriculture. One of the few studies dealing Morphological and physiological data led to either a partial
with the Microbiota of these soils reported high numbers of or complete identification of 2152 isolates from the Cerrado
Actinobacteria [21]. soil. Dendrograms of the identified strains indicated that
The pH values of the Cerrado soils were similar; however, they were phenotypically diverse from the Actinobacteria
moisture influenced the total microbial population counts. species already described with the majority clustering in a
Samples collected in the rainy season are contained in a separate and isolated group. Tropical soils present a myriad of
higher microbial population (∼9.1 log CFU/g) than those microhabitats scarcely explored microbiologically. According
collected in the dry season (∼7.9 log CFU/g) (Table 4). to Zucchi et al. [1], of the 16,013 fungal species described
as new to science over a ten-year period (from 1981 to industry as intermediates in the synthesis of insecticide,
1990), 49% of the species were from tropical countries. This fungicide, and herbicide and cause serious environmental
observation may be extended to other microbial groups, damage [28].
including Actinobacteria, for which there are no statistics The strains identified in this study have been previously
available concerning Brazilian tropical soils. characterised as important for biotechnology applications.
Actinobacteria are Gram-positive, morphologically and Studies are currently being conducted to produce bioactive
physiologically very diverse bacteria with a high GC content compounds from Actinobacteria fermentations on different
in their DNA, and they are one of the main phyla within substrates. The present data suggest that the number and
the domain Bacteria. The class Actinobacteria contains six diversity of Actinobacteria in tropical soils represent a vast
orders—Acidimicrobiales, Rubrobacterales, Coriobacterales, unexplored resource for the biotechnology of bioactives
Bifidobacteriales, Actinomycetales, and Nitriliruptorales. production.
Actinobacteria are dominant colonizers in soils. Many species
produce extracellular enzymes for degradation of macro- 5. Future Perspectives
molecules such as lignin, cellulose, chitin, and, in part, starch.
Therefore, Actinobacteria often occur in materials where We commented on the introduction on the global develop-
organic materials are degraded [12]. In particular, investiga- ment of soil microbiology and the renaissance taking place
tions in the indoor environment demonstrated their presence in natural product research. Furthermore, we reiterated our
in water-damaged building materials and soils beside fungi belief that natural product search and discovery with soil
[12]. This may explain the high presence of Actinobacteria Actinobacteria shows exceptional promise. Our optimism is
in these soils of Brazilian Cerrado. In nature, Actinobacteria based on the spectacular technological armamentarium that
play an important role in the cycling of organic compounds is now available as well as the relatively complete but slowly
and have also been associated with soil organic matter developing understanding of soil biology. This optimism is
production, owing to their black pigments called melanins, also encouraged by the wide range of natural products that
which are related, in some respects, to soil humic acid [12, 21]. may exist with a diversity of applications (e.g., enzymes
In the present work, the soils studied were characterised [5], antibiotics [4], fertilizer, pesticide [16], etc.). However,
as being especially rich in the Streptomyces genus, as are other in this study, the focus was on the microbial diversity of
soils throughout the world. The Streptomyces genus has been Actinobacteria. Soil actinobacterial research and discovery
the focus of research because of the commercial applicability is an important component of natural product research but
of substances produced as well as the systematics of this the development of discoveries to yield products must also
group, which have been modified with advances in molecular be addressed. Although there are encouraging signs that the
biology [24]. Among the species isolated, S. cavourensis newer biotechnology companies are focusing on soil organ-
is a producer of the antibiotic chromomycin [25] and S. isms, medical necessity as much as business opportunity
michiganensis is involved in the synthesis of anthelmintic and should ultimately be the driver behind investment in natural
antiprotozoal substances [26]. product drugs.
Members of the Arthrobacter genus are widely distributed
in ecosystems and can be isolated from diverse environments,
such as air, water (fresh and salt), soil, oil, airborne infections, Acknowledgments
tobacco leaves, human skin, and activated sludge. Arthrobac- The authors would like to thank the Coordenação de Aper-
ter spp. exhibit great metabolic versatility and are able to feiçoamento de Pessoal de Nı́vel Superior (CAPES), the
degrade pollutants and xenobiotics, such as heavy metals Conselho Nacional de Desenvolvimento Cientı́fico e Tec-
(As, Cd, Cr, Cu, Hg, Ni, Pb, Se, V, and Zn) [27]. One of nológico (CNPq), and the Fundação de Amparo à pesquisa
the species identified in this work, A. ramosus, is involved in do Estado de Minas Gerais (FAPEMIG) for financial support
the synthesis of protease, an enzyme important for the food, and scholarships.
pharmaceutical, leather, and detergent industries. A. ramosus
is also highly resistant to a variety of heavy metals and may
be useful for bioremediation processes [27]. References
Species of the Microbacterium genus can be isolated from
[1] T. D. Zucchi, A. S. Guidolin, and F. L. Cônsoli, “Isolation and
air, soil, water, fungi, plants, and humans. Many Microbac- characterization of actinobacteria ectosymbionts from Acro-
terium spp. play a significant role in human health, industry, myrmex subterraneus brunneus (Hymenoptera, Formicidae),”
agriculture, environment, bioengineering, and biotechnology Microbiological Research, vol. 166, no. 1, pp. 68–76, 2011.
and have applicability in the production of exopolysaccha- [2] J. G. Holt, N. R. Krieg, P. H. A. Sneath, J. T. Stanley, and S.
ride, degradation of oil, degradation of xylan, metal tolerance, T. Williams, Bergey’s Manual of Determinative Bacteriology,
production of biosurfactants, degradation of dimethylsul- Williams & Wilkins, Baltimore, Md, USA, 9th edition, 1994.
phide, degradation of lactone, and as a growth promoter in [3] J. M. Piret and A. L. Demain, “Actinomycetes in biotechnology:
plants [28]. Microbacterium phyllosphaerae was found in all an overview,” in Actinomycetes in Biotechnology, M. Good-
three regions examined in this study. In combination with fellow, S. T. Williams, and M. Mordarski, Eds., pp. 461–482,
Burkholderia sp. and Candida tropicallis, M. phyllosphaerae Academic Press, London, UK, 1988.
performs the processes responsible for the biodegradation of [4] Y. Okami and K. Hotta, “Search and discovery of new antibi-
chlorophenols, which are commonly used in the chemical otics,” in Actinomycetes in Biotechnology, M. Goodfellow, S. T.
Williams, and M. Mordarski, Eds., pp. 37–67, Academic Press, in the chemistry, biochemistry and bacterial community struc-
London, UK, 1988. tures of soils in the Cerrado,” Antonie van Leeuwenhoek, vol. 98,
[5] W. Peczynska-Czoch and M. Mordarski, “Actinomycetes en- no. 3, pp. 403–413, 2010.
zymes,” in Actinomycetes in Biotechnology, M. Goodfellow, S. T. [20] J. D. Bresolin, M. M. C. Bustamante, R. H. Krüger, M. R. S. S.
Williams, and M. Mordarski, Eds., pp. 219–284, Academic Press, Silva, and K. S. Perez, “Structure and composition of bacterial
London, UK, 1988. and fungal community in soil under soybean monoculture in
[6] R. C. Gomes, A. S. Mangrich, R. R. R. Coelho, and L. F. Lin- the Brazilian Cerrado,” Brazilian Journal of Microbiology, vol.
hares, “Elemental, functional group and infrared spectroscopic 41, no. 2, pp. 391–403, 2010.
analysis of actinomycete melanins from Brazilian soils,” Biology [21] R. R. R. Coelho and A. Drozdowicz, “The occurrence of
and Fertility of Soils, vol. 21, no. 1-2, pp. 84–88, 1996. actinomycetes in a Cerrado soil in Brazil,” Revue d’Ecologie et
[7] A. Schatz, E. Bugle, and S. A. Waksman, “Streptomycin, a sub- de Biologie du Sol, vol. 15, pp. 459–473, 1978.
stance exhibiting antibiotic activity against Gram-positive and [22] P. G. C. Ruggiero, M. A. Batalha, V. R. Pivello, and S. T. Meirelles,
Gram-negative bacteria,” Experimental Biology and Medicine, “Soil-vegetation relationships in cerrado (Brazilian savanna)
vol. 55, no. 1, pp. 66–69, 1944. and semideciduous forest, Southeastern Brazil,” Plant Ecology,
vol. 160, no. 1, pp. 1–16, 2002.
[8] A. S. Lima, R. S. A. Nóbrega, A. Barberi, K. Da Silva, D. F.
Ferreira, and F. M. D. S. Moreira, “Nitrogen-fixing bacteria [23] A. S. Huddleston, N. Cresswell, M. C. P. Neves et al., “Molecular
communities occurring in soils under different uses in the detection of streptomycin-producing streptomycetes in Brazil-
Western Amazon Region as indicated by nodulation of siratro ian soils,” Applied and Environmental Microbiology, vol. 63, no.
(Macroptilium atropurpureum),” Plant and Soil, vol. 319, no. 1-2, 4, pp. 1288–1297, 1997.
pp. 127–145, 2009. [24] R. F. Souza, R. R. R. Coelho, A. Macrae et al., “Streptomyces
[9] Embrapa: Empresa Brasileira de Pesquisa Agropecuária, Man- lunalinharesii sp. nov., a chitinolytic streptomycete isolated
ual de Métodos de Análise de Solo, Rio de Janeiro, Brazil, 2nd from cerrado soil in Brazil,” International Journal of Systematic
edition, 1997. and Evolutionary Microbiology, vol. 58, no. 12, pp. 2774–2778,
2008.
[10] C. F. Silva, R. S. Azevedo, C. Braga, R. Silva, E. S. Dias, and R.
[25] J. D. Skarbek and L. R. Brady, “Streptomyces cavourensis sp. nov.
F. Schwan, “Microbial diversity in a bagasse-based compost
(nom. rev.) and Streptomyces cavourensis subsp. washingtonensis
prepared for the production of Agaricus brasiliensis,” Brazilian
subsp. nov., a Chromomycin-Producing Subspecies,” Interna-
Journal of Microbiology, vol. 40, no. 3, pp. 590–600, 2009.
tional Journal of Systematic Bacteriology, vol. 28, no. 1, pp. 45–53,
[11] M. Hayakawa and H. Nonomura, “Humic acid-vitamin agar, a 1978.
new medium for the selective isolation of soil actinomycetes,”
[26] G. Cassinelli, E. Cotta, G. D’Amico et al., “Thaimycins, new
Journal of Fermentation Technology, vol. 65, no. 5, p. 501, 1987.
anthelmintic and antiprotozoal antibiotics produced by Strep-
[12] J. Schäfer, U. Jäckel, and P. Kämpfer, “Development of a new tomyces michiganensis var. amylolyticus var. nova,” Archiv für
PCR primer system for selective amplification of Actinobacte- Mikrobiologie, vol. 70, no. 3, pp. 197–210, 1970.
ria,” FEMS Microbiology Letters, vol. 311, no. 2, pp. 103–112, 2010. [27] A. Bafana, K. Krishnamurthi, M. Patil, and T. Chakrabarti,
[13] G. V. M. Pereira, K. T. Magalhães, E. R. Lorenzetii, T. P. Souza, “Heavy metal resistance in Arthrobacter ramosus strain G2
and R. F. Schwan, “A multiphasic approach for the identification isolated from mercuric salt-contaminated soil,” Journal of Haz-
of endophytic bacterial in strawberry fruit and their potential ardous Materials, vol. 177, no. 1–3, pp. 481–486, 2010.
for plant growth promotion,” Microbial Ecology, vol. 63, no. 2, [28] L. Lin, W. Guo, Y. Xing et al., “The actinobacterium Microbac-
pp. 405–417, 2012. terium sp. 16SH accepts pBBR1-based pPROBE vectors, forms
[14] D. Gevers, G. Huys, and J. Swings, “Applicability of rep-PCR biofilms, invades roots, and fixes N2 associated with microprop-
fingerprinting for identification of Lactobacillus species,” FEMS agated sugarcane plants,” Applied Microbiology and Biotechnol-
Microbiology Letters, vol. 205, no. 1, pp. 31–36, 2001. ogy, vol. 93, no. 3, pp. 1185–1195, 2012.
[15] A. P. Castro, B. F. Quirino, G. Pappas Jr., A. S. Kurokawa, E. L.
Neto, and R. H. Krüger, “Diversity of soil fungal communities of
Cerrado and its closely surrounding agriculture fields,” Archives
of Microbiology, vol. 190, no. 2, pp. 129–139, 2008.
[16] M. S. Girvan, J. Bullimore, A. S. Ball, J. N. Pretty, and A. M.
Osborn, “Responses of active bacterial and fungal communities
in soils under winter wheat to different fertilizer and pesticide
regimens,” Applied and Environmental Microbiology, vol. 70, no.
5, pp. 2692–2701, 2004.
[17] K. Schwarzenbach, J. Enkerli, and F. Widmer, “Objective criteria
to assess representativity of soil fungal community profiles,”
Journal of Microbiological Methods, vol. 68, no. 2, pp. 358–366,
2007.
[18] B. E. Wolfe, D. L. Mummey, M. C. Rillig, and J. N. Klironomos,
“Small-scale spatial heterogeneity of arbuscular mycorrhizal
fungal abundance and community composition in a wetland
plant community,” Mycorrhiza, vol. 17, no. 3, pp. 175–183, 2007.
[19] R. S. Peixoto, G. M. Chaer, N. Franco, F. B. Reis Jr., I. C. Mendes,
and A. S. Rosado, “A decade of land use contributes to changes

Microbial Diversity For Biotechnology

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Microbial Diversity For Biotechnology

Uploaded by

Copyright:

Available Formats

BioMed Research International

Microbial Diversity for Biotechnology

Guest Editors: George Tsiamis, Dimitrios Karpouzas, Ameur Cherif,

Microbial Diversity for Biotechnology

Guest Editors: George Tsiamis, Dimitrios Karpouzas,

Cultivation-Dependant Assessment, Diversity, and Ecology of Haloalkaliphilic Bacteria in Arid Saline

Population Abundance of Potentially Pathogenic Organisms in Intestinal Microbiome of Jungle Crow

Microbial Adhesion and Biofilm Formation on Microfiltration Membranes: A Detailed

Plant Growth Promotion Potential Is Equally Represented in Diverse Grapevine Root-Associated

The Occurrence of Two Species of Entomophthorales (Entomophthoromycota), Pathogens of Sitobion

Engineering Microbial Cells for the Biosynthesis of Natural Compounds of Pharmaceutical

An Internalin A Probe-Based Genosensor for Listeria monocytogenes Detection and Differentiation,

George Tsiamis,1 Dimitrios Karpouzas,2 Ameur Cherif,3 and Konstantinos Mavrommatis4

Correspondence should be addressed to George Tsiamis; gtsiamis@upatras.gr

Received 24 December 2013; Accepted 24 December 2013; Published 23 January 2014

Interestingly, the metabolites involved in the antagonistic

Correspondence should be addressed to Ameur Cherif; cherif.ameur@gmail.com

Received 30 April 2013; Revised 31 August 2013; Accepted 14 September 2013

Academic Editor: George Tsiamis

1. Introduction To adapt in such conditions, haloalkaliphilic microorganisms

W E Ksar Ghilane sites

Accession Phylogenetic Closet described species Accession Sampling Sample Type of

Representative strains NaCl tolerance pH tolerance Production of extracellular enzymes

Isolates (number of strains/16S rRNA(%) similarity, [salt tolerance]) Assignation ITS-haplotypes

100 BMG ED25 (1/99%, [0–10%]) H 21

Halomonas taeanensis (NR 043087.1)

Halomonas Salinicoccus Halomonas

Chott Douz Chott Djerid

Acknowledgments [15] H. Sahay, S. Singh, R. Kaushik, A. K. Saxena, and D. K. Arora,

Sofia Nikolaki and George Tsiamis

Correspondence should be addressed to George Tsiamis; gtsiamis@upatras.gr

Received 30 April 2013; Revised 26 August 2013; Accepted 26 August 2013

Academic Editor: Dimitrios Karpouzas

1. Introduction contain a vast array of biochemical transformations, and the

Table 2: Technical specifications of Next Generation Sequencing platforms.

Table 3: Oligonucleotide primers that can be used for 16S rRNA

(Figure 1). Hypervariable regions of the 16S rRNA gene are

Complex environmental sample

Isolation of single cells

Multiple displacement PCR and 16S rRNA

Figure 2: General overview of the single cell genomics approach.

Superphylum Candidate phylum Proposed name Etymology

Latescibacter a hiding small rod

Cloacimonas a unit from a sewer

Paceibacter Pace’s bacterium

Gracilibacteria -gracilis (lat.), slim, slender, slight, meager, simple

Ca.lesc.a.man’tes. L. v. calesco, to become warm, grow hot; L. v. amo, to love,

Hydrogenedens a hydrogen consumer te.re.phtha’li.cus. N.L. n. (acidum)

Acetothermus indicates a vinegar organism living in hot places and

Fer.vi.do.bac’ter. L. adj. fervidus, hot, steaming; -i-connecting vowel; N.L. n.

Parv.ar.chae’a. A higher taxonomic unit comprising the genera Parvarchaeum

[138] H. Heuer, K. Hartung, G. Wieland, I. Kramer, and K. Smalla,

Okba Selama,1 Phillip James,2 Farida Nateche,1

Correspondence should be addressed to Hocine Hacène; h hacene@yahoo.fr

Received 14 March 2013; Revised 11 July 2013; Accepted 13 August 2013

Academic Editor: Konstantinos Mavrommatis

1. Introduction environments on scales ranging from 0.002 km to 20,000 km

Algorithm 1: Biodiversity and biogeography—GBIF Filter.

If there is not taxonomy for the sequence (no

Algorithm 2: Biodiversity and Biogeography—NCBI Nucleotide Tracker.

Table 1: Occurrences overview of records with coordinates from GBIF Database.

data from NCBI Nucleotide Database 50% than data from

3.2.3. Occurrences of Records in Different Geographical Areas.