Appl Microbiol Biotechnol (2011) 89:429439 DOI 10.

1007/s00253-010-2884-9

ENVIRONMENTAL BIOTECHNOLOGY

Enhancement of the antimicrobial performance of biocidal formulations used for the preservation of white mineral dispersions
Nicola Di Maiuta & Patrick Schwarzentruber & Crawford S. Dow

Received: 30 June 2010 / Revised: 3 September 2010 / Accepted: 7 September 2010 / Published online: 28 September 2010 # Springer-Verlag 2010

Abstract Biocides play an important role in the preservation of white mineral dispersions (WMD). Due to the occurrence of biocide-resistant bacteria and technical limitations in the use of biocides, new preservation strategies are requiredlike the enhancement of biocides by non-biocidal compounds. The aim of this study was to evaluate the biocide enhancement performance of lithium against various biocide-resistant bacteria in WMD. Subsequently, the minimal enhancing concentration (MEC) of lithium and the bioavailability of lithium in respect to the mode of introduction into WMD were investigated. The antimicrobial performance of biocidal formulations comprising isothiazolinones and formaldehyde releasers or isothiazolinones and glutaraldehyde has been evaluated against the related resistant bacterial spectrum in the presence of lithium. The MEC of lithium ranged from 1,350 to 1,500 ppm (based on the liquid phase weight of a WMD with 75% solids) for formaldehyde releasers and glutaraldehyde-based biocidal formulations, respectively. The biocide enhancing property of lithium was independent of whether lithium was introduced into WMD via a lithiumneutralised dispersant, added during the calcium carbonate grinding step, or dosed into the final product. Lithium is a non-biocidal compound which has been discovered to be a potent and universal biocide enhancer. Lithium boosts the biocidal activity of various biocides and provides a novel technique to overcome biocide resistance in WMD. Such a
N. Di Maiuta (*) : P. Schwarzentruber Omya Development AG, R&D Microbiology, 4665 Oftringen, Switzerland e-mail: nicola.dimaiuta@omya.com N. Di Maiuta : C. S. Dow Department of Biological Sciences, University of Warwick, Coventry CV4 7AL, UK

biocide enhancer represents a breakthrough that offers a potential tool to revolutionise the consumption of biocidal agents in the WMD producing industry. Keywords Biocide . Enhancement . Resistance . Bacteria . Lithium . Calcium carbonate slurry . White mineral dispersion

Introduction White mineral dispersions (WMD; ground and precipitated calcium carbonate, modified calcium carbonate, clay and talc) are water-based mineral dispersions used as filler and pigment in the paper and paint industries (Rohleder and Huwald 2001). Although the alkaline pH of WMD ranges from 8 to 10, and the water content from 25% to 80% (w/w), the presence of various salts and the available oxygen are sufficient to promote an adequate microbial propagation environment. Bacterial contamination of WMD often leads to acidification of the product. A decrease in pH is usually associated with an increase in viscosity and consequently involves alteration of the rheological parameters such as fluidity and pumpability, which are required to ensure trouble-free handling to the final user. Discolouration and development of unpleasant odours can occur under certain conditions when oxygenation of the WMD product stops and aerobic growth is replaced by anaerobic growth. Therefore, biocides play an important role in the preservation of WMD in order to maintain high hygiene and quality standards, such as brightness, rheologic parameters as well as odour neutrality. Due to the occurrence of biocide-resistant bacteria in WMD, as well as technical limitations on the use of biocides (e.g. acidic biocide formulations), there is increas-

430

Appl Microbiol Biotechnol (2011) 89:429439

ing demand for new preservation strategies to maintain the high quality required of WMD in respect of brightness, rheological parameters and odour neutrality (Di Maiuta et al. 2009). A lack of biocidal activity having been recognised, the only available options are to increase the biocide dosage or to substitute the antimicrobial compound currently in-use with another one (Schwarzentruber 2003; Schwarzentruber and Gane 2005). Apart from the economic and ecological impact caused by increased application of biocide, in this context, it is essential to highlight the fact that there exist only a limited number of antimicrobials that are suited to preserving WMD. This is due to the more rigorous regulatory situation created by the BPD (Biocidal Products Directive 98/8/EC), as well as the incompatibility of certain antimicrobials with the chemical properties of WMD. These circumstances have led to the shelving of development and registration of new biocidal substances, owing to the laborious and expensive research required. (Khknen and Nordstrm 2008) in addition, some biocidal compounds used for the preservation of calcium carbonate slurries, such as formaldehyde releasers, are under pressure due to their possible classification as carcinogenics, while some EU member states have reduced their occupational exposure limits. The few remaining actives such as glutaraldehyde, isothiazolinones (CMIT/MIT/BIT), 2-bromo-2nitropropane-1,3-diol (BNPD) or 2,2-dibromo-3-nitrilopropionamide (DBNPA) are less stable, while the acidic character of some compounds constitutes a technical hurdle for calcium carbonate slurries which are rather alkaline. Of the strategies for overcoming biocide resistance and to keep within technical limitations on the application of biocides, one approach is to use combinations of antimicrobial compounds. Combinations of biocides have been described to exhibit a synergistic effect in that the effectiveness of the combined antimicrobials is greater than the sum of the individual compounds (Lambert et al. 2003; Botelho 2000; Denyer et al. 1985, 1986; Acar 2000). However, the synergy is an effect brought about by combining reduced concentrations of different antimicrobials, rather than combining the in-use biocides with nonantimicrobial compounds. The term enhancement is therefore more appropriate to describe an increase of the antimicrobial activity of a biocide by means of a nonantimicrobial compound (Hodges and Hanlon 1991). Previous studies have reported the enhancement of biocides by non-antimicrobial compounds such as EDTA (Vaara 1992; Ayres et al. 1999; Alakomi et al. 2006) and polyethylenimine (Helander et al. 1997; Alakomi et al. 2006; Khalil et al. 2008). Generally, the enhancement activity has been suggested as originating from destabilisation of the bacterial cell membrane of Gram-negative bacteria, thus increasing its permeability to the antimicrobials. This effect has been reported in connection with a

vast assortment of other compounds such as polycations, lactoferrin, transferrin and citric acid (Hancock and Wong 1984). Additionally, interference with the quorum sensing system (Vestby et al. 2009) or the substrate as a promoter of the uptake of antimicrobials has already been described (Maillard and Russell 2001; Denyer and Maillard 2002). Metals such as silver (Silvestry-Rodriguez et al. 2007), mercury (Morrier et al. 1998), copper (Sondossi 1990) and caesium (Avery 1995) have also been noted to exhibit biocide potential or even toxicity towards bacteria. In general, heavy metal ions are more toxic than the other classes of metal ions (Harrison et al. 2005). Lithium has been determined as exhibiting antiviral activity (Amsterdam et al. 1990), as well as antimicrobial activity against Gram-negative bacteria (Eisenberg et al. 1991), Gram-positive bacteria and synergistic activity when combined with antibiotics (Lieb 2002, 2004). In addition, even though the molecular mechanism of electropermeabilisation is not fully understood, lithium is routinely used to perform lithium-cation induced electrotransformation of bacteria. This effect is believed to be related to a transient increase in permeability of the cell membrane (Papagianni et al. 2007; Ramon and Fonzi 2009). Since lithium is easy to introduce into calcium carbonate slurry via dispersant neutralisation, the main objective of this study was to evaluate the enhancement performance of lithium to the antimicrobial activity of the biocidal formulations in-use in calcium carbonate slurries against inherent biocide-resistant bacteria. The effects of concentration and the mode of introduction of lithium into calcium carbonate slurry were determined. In addition, the inherent activity and the mechanism of enhancement of lithium were investigated.

Materials and methods Chemicals and biocide dosage A commercial biocidal formulation of (ethylenedioxy) dimethanol (EDDM, CAS No. 3586-55-8), 5-chlor-2methyl-2H-isothiazolin-3-one (CMIT, CAS No. 26172-55-4) and N-methyl-isothiazolin-3-one (MIT, CAS No. 2682-20-4) was obtained from Dow Microbial Control (Horgen, Switzerland). A commercial biocidal formulation of glutaraldehyde (GDA, CAS No. 111-30-8), CMIT and MIT was obtained from Lanxess Deutschland GmbH (Leverkusen, Germany). Unless otherwise stated, all other chemicals and the culture media were purchased from SigmaAldrich (Buchs, Switzerland) or Becton Dickinson (Allschwil, Switzerland), respectively. In liquid cultures, commercial biocide formulations were dosed in parts per million (w/w) of the commodity based on

Appl Microbiol Biotechnol (2011) 89:429439

431

the total weight of the solution. In contrast, the biocide amount dosed in calcium carbonate slurry is reported in parts per million (w/w) of the commodity relative to the weight of the aqueous phase of a WMD with a solid content of 75% (w/w). Manufacture of calcium carbonate slurry Calcium carbonate slurry is typically prepared by wet grinding marble in the presence of 0.2% to 1.3% (w/w) of a radically polymerised polyacrylic acid wherein the carboxylic acid groups are neutralised by alkali metals. In this study, Hydrocarb 9075% (HC90, 75% solids content, pH 9 to 9.5, and particle size 90%<2 m) was used. Sterile slurry was prepared by sterilisation at 121 C for 15 min in an autoclave. Lithium was introduced into the calcium carbonate slurries via the dispersant (fully or partially neutralised with lithium) or during the wet grinding of marble in the form of lithium carbonate, as well as added to the final slurry product by means of a lithium carbonate solution. The nomenclature and details of the calcium carbonate slurries investigated in this study are summarised in Table 1. Identification of microorganisms in WMD and culture conditions The biocide-resistant cultures used in this study have been deposited in the Culture Collection of Switzerland (CCOS) repository. The biocide-resistant calcium carbonate slurry cultures, GDA/CMIT/MIT-resistant (rGCM; CCOS 29) and EDDM/CMIT/MIT-resistant (rECM; CCOS 30), were obtained from product storage tanks located at WMD manufacturing plants that used the related biocide to preserve the WMD. The in-use concentrations of GDA/ CMIT/MIT formulation and EDDM/CMIT/MIT formulation were 1,350 and 750 ppm, respectively. The formaldehyderesistant WMD cultures of Pseudomonas putida (CCOS 31) and facultative methylotrophic Methylobacterium extorquens (CCOS 32) were described previously (Di Maiuta et al.

2009). Biocide-resistant colonies isolated from biocidesupplemented calcium carbonate slurry were phylogenetically identified by partially sequencing the 16S rRNA gene (Di Maiuta et al. 2009). WMD bacteria cultures utilised for the determination of the antimicrobial activity of biocides are batch culture which have been standardised by incubation time (statically for 48 h at 30 C and TVC >105 cfuml1). Fifty grams of fresh sterile Hydrocarb 9075%, containing the corresponding concentration of biocide if required, was inoculated with 1 ml of the preceding slurry culture. After 48 h of incubation, bacterial growth was confirmed by plating on PCA plates a 1:10 dilution of the slurry in phosphate buffered saline. The slurry cultures provided the basis for inoculation of the challenge testing assays in calcium carbonate slurries. Challenge testing Fifty-gram aliquots of sterilised calcium carbonate slurry were each supplemented with different concentrations of biocide formulation. An aliquot containing no biocide was included to ensure that the used WMD cultures grow in the unpreserved product. The samples were then incubated for 3 days at 30 C. Subsequently, the samples were inoculated with 1 ml of contaminated WMD culture as described above. After 24 h to 3 days of incubation at 30 C, the total viable count was determined according to the plate count method. Samples that did not show any counts (<100 cfu ml1) were further inoculated with 1 ml of the same WMD culture as above. No more than three inoculations were performed. Cell counts greater than 104 cfuml1 were evaluated qualitatively by estimating the order of magnitude (105, 106 or higher). All analyses have been repeated two times with high reproducibility. Lithium ion analysis Slurry liquid phase was extracted by pressure filtration (Fann Instruments filter press series 300, special hardened filter paper 3.500, retention 25 m, 6 bar) and passed

Table 1 Details of the calcium carbonate slurry products used in this study Product ID HC90-Li HC90-MgNa5050 HC90-MgLi5050 HC90-MgLi7525 HC90-gLi ND not determined
a b

Dispersant neutralisation (%) 100 Li 50:50 Mg:Na 50:50 Mg:Li 75:25 Mg:Li 50:50 Mg:Na

Lithium contenta (ppm) 1,580 0 920 660 1,500

Lithium measuredb (ppm) 407 0 ND ND 711

pH

Remarks Lithium via dispersant Standard HC 90 slurry Lithium via dispersant Lithium via dispersant Ground with Li2CO3

9.2 9.4 9.4 9.6

Theoretical value Measured by IC in the water phase

432

Appl Microbiol Biotechnol (2011) 89:429439

through a Sartorius (Dietikon, Switzerland) 0.2-m pore size syringe filter (Minisart RC) prior to measurement of lithium. Lithium concentration was determined by ion chromatography using a Metrohm (Zofingen, Switzerland) 882 Compact IC plus system with autosampler. Twenty millilitres of filtrate was injected into a Metrosep A Supp 5 (1504.0 mm) column. The eluent (flow rate 0.7 mlmin1) composition was NaHCO3 (1 mmoll1)Na2CO3 (3.2 mmol l1) 1:1v/v. The analysis was performed at room temperature. The total lithium concentration was determined by means of inductively coupled plasma optical emission spectrometry (ICPOES). Firstly, the water of the calcium carbonate slurry samples was removed by drying the sample at 120 C. Eight millilitres of 69% HNO3 was then added to 2 g of dried sample and boiled for 5 min to carry out digestion. Finally, the boiled sample was cooled to 20 C and adjusted to 100 ml with ultrapure water (Milli-Q integral water purification system). The OPTIMA 3200 XL ICPOES system from Perkin Elmer (Schwerzenbach, Switzerland) was used. Lithium ions were detected in the emission line at wavelengths of 610.35 and 670.76 nm. Leakage of cellular constituents caused by lithium Pseudomonas mendocina cells isolated from the EDDM/ CMIT/MIT-resistant culture were used to investigate the effect of lithium on bacterial cells. Bacteria were grown in tryptic soy broth at 30 C and 160 rpm on an orbital shaker overnight (151 h). Cells were harvested by centrifugation (5,000 rcf for 10 min), washed twice with 10 mmoll1 HEPES buffer pH 7.4 (Denyer et al. 1986) and resuspended in HEPES buffer to achieve a cell density of 107 to 108 cellsml1. The HEPES buffercell suspension was then supplemented with biocide, lithium or a combination thereof and analysed at 0, 30 and 60 min after the respective treatments. Two biological replicates and two technical replicates of all samples were analysed. The absorbance measurements at 600 and at 260 nm were performed using a Vaudaux-Eppendorf (Basel, Switzerland) Bio-Photometer. Prior to A260 analysis and determination of the potassium ion concentration, the suspension was passed trough a 0.2-m pore size syringe filter to remove the cells. The potassium ion concentration was measured by means of ICPOES. Potassium ions were detected in the emission line at a wavelength 766.47 nm as described above. Membrane permeabilisation and depolarisation assay The NPN (1-N-phenylnaphthylamine) uptake assay was performed to assess membrane permeabilisation as described by Helander and Mattila-Sandholm (2000). Fluorescence was detected using a Synergy HT microplate reader

(BioTek Instruments Inc., Winooski, VT, USA). Wells were read from the bottom with a sensitivity value of 70, and filter settings were 360/40 nm for the excitation and 420/ 50 nm for the emission. P. mendocina cells were prepared as described for the leakage experiments and resuspended in 5 mmoll1 HEPES pH 7.2. The NPN stock solution (0.5 mmoll1) was prepared in acetone. For each substance, 16 replicates (wells) in a 96-microtitre plate were analysed (two biological replicates and eight technical replicates were analysed). Fifty microlitres NPN (40 moll1, diluted in 5 mmoll1 HEPES pH 7.2) and 25 l of each substance (in HEPES buffer; 8 higher in order to achieve the desired final concentration in the well) were adjusted to a total volume of 100 l with HEPES buffer if required. Just before measurement, 100 l of bacterial suspension was added to the wells, the plate briefly shaken for 15 s and fluorescence values recorded within 5 min. Controls included buffer alone (200 l), buffer (100 l) and bacterial suspension, buffer (150 l) and NPN (50 l) as well as buffer (50 l), NPN (50 l) and bacterial suspension (100 l). The relative fluorescence unit value (rfu) was calculated as follows: rfu= (cells+test substance+NPN)(cells+NPN). The NPN factor was calculated as the ratio of the rfu value of the substance(s) to the rfu value of the NPN supplemented buffer background [(buffer+NPN)(buffer)]. Control experiments included treatment of the cells with saline solution (0.9% NaCl) or ETDA. Statistically significant changes of the NPN uptake factor were determined by the Student's t test (unpaired, equal variance) based on the untreated sample (buffer only). The CellFacts II analyzer from CellFacts Instruments (Coventry, UK) was used to assess membrane depolarisation. Two hundred fifty grammes sterilised calcium carbonate slurry was inoculated with 5 ml of rECM culture, and after incubation for 48 h at 30 C, 50 g aliquots were each supplemented with different amounts of lithium. Lithium was supplemented in the form of lithium carbonate. Membrane potential of the cells was determined 30 min after the addition of lithium to the slurry cultures as described by Di Maiuta et al. (2009) with minor modifications. To stain the cells, 5 l of 100 M 3,3-dipropylthiadicarbocyanine iodide (DiSC3(5), Invitrogen) was added into the tube, mixed well and incubated in darkness for 15 min. The CellFacts II instrument was operated according to the manufacturer's instructions. Two biological replicates and two technical replicates were analysed.

Results Characterisation of biocide-resistant bacteria in WMD Biocide-resistant colonies isolated from biocide-supplemented calcium carbonate slurry were identified by sequencing

Appl Microbiol Biotechnol (2011) 89:429439

433

approximately 800 bp of the 16S rRNA gene. Sequence analysis of the isolated strains revealed that both resistant cultures rGCM and rECM contained bacteria belonging to the genus Pseudomonas ( Table 2). The closest relatives based on the similarity of the 16S rRNA gene sequence were the species Pseudomonas pseudoalcaligenes, P. mendocina and Pseudomonas alcaliphila. However, by means of the 16S rRNA gene, the resolution between the Pseudomonas spp. is limited (Mulet et al. 2010). Resistance was analysed by means of the challenge test. The MIC of the biocides against the corresponding resistant bacteria were determined as the concentration preventing bacterial growth detection over three cycles of inoculation. The resistance and the susceptibility level of the biocideresistant bacteria were compared to the MIC of the biocidesusceptible WMD bacterial population H90-i found in slurry produced under laboratory conditions in small scale (Pseudomonas sp.). In comparison to the H90-i culture, the increase in biocide concentrations used to preserve calcium carbonate slurry over three bacteria challenges with biocide-resistant bacteria was 20- and ninefold higher for the biocidal formulations GDA/CMIT/MIT and EDDM/ CMIT/MIT, respectively. These MIC values clearly showed that the WMD samples obtained from the biocide-using production plants were contaminated with bacteria that were adapted or even resistant to the applied biocide formulation. Lithium content of the investigated calcium carbonate slurries The lithium concentrations summarised in Table 1 show that the determined lithium concentration of slurry deviated significantly from the theoretical calculated value based on the dosed lithium-neutralised dispersant or lithium carbonate amount, respectively. The lithium concentration was initially determined by means of ion chromatography in the aqueous phase of the calcium carbonate. These data suggested that the lithium introduced into the calcium carbonate slurry via dispersant was not completely dis-

solved in the water phase and not completely dissociated from the polyacrylate dispersant. Similarly, lithium introduced into the calcium carbonate slurry in the form of lithium carbonate reacted with the polyacrylates and was partially sequestered to neutralise free carboxyl groups of the polyacrylate polymer. These observations were corroborated by the fact that using ICPOES to determine the total lithium content revealed a concentration of 1,499 ppm lithium for the H90-Li slurry manufactured with fully lithium-neutralised dispersant. Addition of lithium in WMD via neutralised dispersant The enhancement efficacy of lithium to EDDM biocide was investigated in various calcium carbonate slurries where lithium was introduced into the product via lithiumneutralised dispersant ( Table 3). After a single challenge with the formaldehyde-resistant WMD cultures of P. putida or facultative methylotrophic M. extorquens, significant growth of both bacterial species was observed in all calcium carbonate slurry products without biocide. However, product HC90-Li dispersed with lithium-neutralised dispersant showed inhibition of both formaldehyde-resistant microorganisms when 1,500 ppm EDDM biocide had been added to the slurry. On the other hand, the HC90 slurries manufactured by means of dispersant neutralised with only 25% or 50% all showed considerable bacterial contamination. The biocide-enhanced performance to EDDM of a HC90-Li dilution series in conventional lithium-free HC90MgNa5050 was also examined ( Table 4). These results indicated that a 1:2 dilution of HC90-Li with conventional HC90 reduced the lithium content significantly. In the diluted slurries, a lack of lithium-derived biocide-enhanced efficacy against both M. extorquens and Pseudomonas sp. formaldehyde-resistant bacteria was determined. Addition of lithium into WMD via lithium carbonate The biocide-enhancing performance of lithium introduced into WMD in the form of lithium carbonate was determined

Table 2 Calcium carbonate slurry isolated species from the biocide-resistant culture identified on the basis of 16S rRNA gene sequence alignment, with use of BlastN and RDPII databases Taxonomic affiliation Pseudomonas Closest relativea Pseudomonas pseudoalcaligenes Pseudomonas mendocina Pseudomonas alcaliphila
a b c

Accession No.b AB109888 AF232713 AB030583

Maximum identityc 99 99 99

Length (bp) 756 778 925

Closest relative from comparison with BlastN and RDPII databases Accession number of the closest relative entry Maximum identity for the covered sequence length

434 Table 3 Enhancement of the biocidal activity of EDDM by various concentrations of lithium introduced into WMD via dispersant Product ID

Appl Microbiol Biotechnol (2011) 89:429439 Total viable count (cfuml1) as mean (standard deviation) M. extorquens EDDM (ppm) HC90-MgNa5050 HC90-Li HC90-MgLi5050 HC90-MgLi7525 0 >106 >106 >106 >106 1,500 3.2104 (4103) <102 4.4104 (9.6103) 2.4104 (3.7103) Pseudomonas sp. 0 >106 >106 >106 >106 1,500 1.6104 (3.3103) <102 8.0104 (7.1103) 1.6104 (6.8103)

by comparing the addition of lithium carbonate during the mineral grinding step with the dosing of lithium carbonate directly into the final product. The calcium carbonate slurry HC90-gLi ground in the presence of lithium carbonate (1,500 ppm lithium) showed excellent enhancing properties ( Fig. 1). Independent of which biocide formulation was used, a concentration of 1,500 ppm lithium was sufficient to preserve calcium carbonate slurry over three bacterial challenges with the related biocide-resistant culture. The minimal enhancing concentration (MEC) of lithium was elucidated for both biocidal formulations against their respective biocide-resistant microorganisms by dosing various concentrations of lithium carbonate into regular HC90-MgNa5050 calcium carbonate slurry ( Figs. 2 and 3). The MEC of lithium for the biocide formulation EDDM/ CMIT/MIT and GDA/CMIT/MIT over three challenges with biocide-resistant bacteria were determined to be 1,050 and 1,500 ppm, respectively. No significant differences in total viable counts were observed in either biocide-free/ lithium-containing calcium carbonate slurry samples or vice versa inoculated with the biocide-resistant bacteria culture. The challenge testing results clearly show that the biocide enhancing property of lithium is dose-dependent, and that the slurry culture rECM was more susceptible to the enhancing effect of lithium. A closer investigation of the enhancing concentration of lithium for a single bacterial challenge revealed that complete eradication of the resistant bacteria was achieved at a dose of between 750 and 1,050 ppm. As a consequence of lithium showing an antimicrobial enhancing performance to different biocides against the
Table 4 Enhancement performance of lithium in serial dilutions of HC90-Li to 1,500 ppm EDDM biocide against formaldehyde-resistant WMD bacteria Product

relative resistant slurry bacteria, it was assumed that other monovalent cations such as potassium or sodium might also be potential candidates to enhance the performance of the biocide. Sodium did not come into consideration since it is being used for neutralisation of the dispersant whereas potassium in the equivalent concentration to lithium (1,500 ppm) did not show any enhancement of the antimicrobial performance of the biocides EDDM/CMIT/ MIT and GDA/CMIT/MIT against the relative resistant slurry cultures (data not shown). Effect of lithium on P. mendocina cells isolated from WMD P. mendocina cells isolated from the EDDM/CMIT/MITresistant culture were used to investigate the influence of lithium on the bacterial cells on its own and in the presence of biocide formulation EDDM/CMIT/MIT. The leakage of cytoplasmic constituents absorbing at 260 nm (such as nucleotides, nucleosides and aromatic amino acids) along with potassium was recorded over a time period of 60 min after exposing the cells to lithium, biocide or combinations thereof ( Fig. 4a, b). Changes in cell morphology (lysis or swelling) were followed by monitoring the absorbance at 600 nm ( Fig. 4c). Firstly, in bacterial cells exposed to 750 ppm of the biocide mixture EDDM/CMIT/MIT only, or in combination with 1,500 ppm lithium, a significant leakage of cellular constituents absorbing at 260 nm was observed. Thirty minutes after the treatment, the absorbance (260 nm) increased to an average level of 0.28 units, which is equivalent to a secretion of 13.5 gml1 of dsDNA
Total viable count (cfuml1) mean (standard deviation) M. extorquens Pseudomonas sp. >106 <102 3.2104 (1.4104) 5.0104 (7.4103) 5.0104 (1.4104)

HC90-MgNa5050 HC90-Li Dilution ratiosa 1:2 1:4 1:8

>106 <102 >106 >106 >106

HC90-Li:HC90-MgNa5050

Appl Microbiol Biotechnol (2011) 89:429439

435

with 750 ppm of the biocide EDDM/CMIT/MIT mixture did not show significant uptake of NPN. Similar results were obtained by replacing the biocide/lithium volume with saline solution (0.9% w/w NaCl), indicating that the permeabilisation effect is fully attributable to the effect of lithium on the bacterial membrane. Finally, using the fluorescent dye DiSC3(5), a significant depolarisation of the bacterial membrane in the presence of lithium was observed (Fig. 6).

Fig. 1 Enhancement of EDDM/CMIT/MIT and GDA/CMIT/MIT biocide formulations in HC90-gLi calcium carbonate slurry ground in the presence of lithium carbonate. Calcium carbonate samples were inoculated with the resistant bacteria rECM and rGCM, respectively

Discussion The development of biocide resistance and the incidence of multiple biocide-resistant bacteria in calcium carbonate slurries is a primary factor determining future preservation strategies for WMD. For this reason, WMD preservation research activities have been increased on novel and innovative preservation approaches. Synergistic combinations of biocidal actives such as aldehyde-based or aldehydereleasing compounds with isothiazolinones, BNPD or DBNPA have been used for many years to preserve WMD (Schwarzentruber 2003; Schwarzentruber and Gane 2005). Nevertheless, WMD-contaminating bacteria resistant to several combinations of biocidal actives have been discovered. The estimation of synergy between combinations of antimicrobials has been described in detail as the sum of the fractional inhibitory concentration assessed by means of the checkerboard titration principle (Berenbaum 1978). However, using fractional inhibitory concentrations to investigate synergistic effects of antimicrobials assumes a linear dose response, and it has been suggested that the synergism occasionally arises from the combination of antimicrobials with different dose-responses, rather than from their synergistic performance (Lambert and Lambert 2003; Lambert et al. 2004, 2003). In the case of WMD, a modified definition of the terms biocide synergy and enhancement have been introduced: (1) a synergistic biocidal formulation is considered to be a blend of two or more antimicrobial actives to preserve WMD, and (2) a

(1.0 A260 unit = 50 g ml 1). In contrast, the same concentration of lithium on its own did not show any effect on the leakage of cellular constituents determined at 260 nm. On the other hand, 1,500 ppm of lithium induced a rapid outflow of potassium ions either on its own or in combination with 750 ppm EDDM/CMIT/MIT biocide, ranging from 14 to 18 ppm. Morphological effects monitored at 600 nm were observed in bacterial cells exposed to lithium or lithium combined with the biocide. An increase of absorbance at 600 nm probably indicated that the treatment with lithium resulted in a cellular swelling or promoted cytokinesis. Secondly, the permeabilisation of P. mendocina treated with lithium, EDDM/ CMIT/MIT biocide blend or a mixture was assessed utilising the NPN uptake assay. The NPN uptake factors are shown in Fig. 5. Lithium on its own (1,500 ppm) was significantly (p<0.01) permeabilising P. mendocina cells as demonstrated by the enhanced uptake of NPN. Similarly, lithium in combination with the biocide mixture EDDM/ CMIT/MIT (750 ppm) showed a reduced but still significantly increased uptake of NPN compared to the untreated control (p < 0.01). Control experiments with EDTA (180 ppm) instead of biocide/lithium showed a significant increased uptake of NPN too. In contrast, cells treated only
Fig. 2 Biocide enhancement performance of lithium against rECM WMD culture in the presence of 750 ppm EDDM/ CMIT/MIT determined by means of the challenge test. For each sample, a three-cycle challenge test was performed. Total viable count (TVC) detection limit was 102 cells ml1

436 Fig. 3 Biocide enhancement performance of lithium against rGCM WMD culture in the presence of 1,350 ppm GDA/ CMIT/MIT determined by means of the challenge test. For each sample, a three-cycle challenge test was performed. Total viable count (TVC) detection limit was 102 cells ml1

Appl Microbiol Biotechnol (2011) 89:429439

biocide enhancer is a substance that is not noted to be a biocide and which at the applied concentration does not exhibit antimicrobial activity in WMD. The results of this study clearly showed that lithium is a universal enhancer of biocides used for the preservation of WMD. WMD bacteria with resistance to the in-use biocide concentration were susceptible to the biocide in the

presence of lithium ions. The major advantage is that lithium can be introduced into WMD simply via dispersant neutralisation. No difference in the biocide enhancer properties of lithium was observed whether it was introduced into the calcium carbonate slurry during the grinding procedure or afterwards into the final product. In this context, it is important to mention that for lithium, the

Fig. 4 Pseudomonas mendocina leakage of cytoplasmic components at 260 nm (a) and potassium (b) as well as absorbance at 600 nm (c) measured 30 and 60 min after exposure to the following combinations: no biocide/lithium control (black square), 750 ppm EDDM/CMIT/

MIT (white circle), 1,500 ppm lithium (white triangle) and 750 ppm EDDM/CMIT/MIT+1,500 ppm lithium (black circle). All concentrations are given in parts per million based on the weight of the solution. Values are presented as meanstandard deviation (n=4)

Appl Microbiol Biotechnol (2011) 89:429439

437

Fig. 5 NPN uptake factor obtained after exposing Pseudomonas mendocina cells to different treatments. The following concentrations were used: 750 ppm EDDM/CMIT/MIT, 1,500 ppm lithium, 750 ppm EDDM/CMIT/MIT + 1,500 ppm lithium and 1,500 ppm NaCl 180 ppm EDTA. Values are presented as meanstandard deviation (n=16); **p<0.01 for Student's t test (unpaired, equal variance) based on the untreated sample

fractional inhibitory concentration index (FICI) calculated according to Berenbaum (1978) revealed a FICI value of 0.8 for the EDDM/CMIT/MIT biocide and 0.7 for the GDA/CMIT/MIT biocide, respectively. Synergism of drugs or antimicrobials has been described as occurring when the FICI is <1 (Berenbaum 1978); nevertheless, that interpretation was revised by the American Society of Microbiology, which has defined a synergy at FICI<0.5. Controversially, other authors have defined a partial synergy as 0.5>FICI<1 (Botelho 2000). Therefore, in view of the fact that lithium is not an antimicrobial, yet enhances the antimicrobial performance of biocides in calcium carbonate slurries, it is more appropriate to use the term enhancement rather than synergism. Lithium has in fact been used since the 1950s to treat bipolar disorders in psychiatry (Cruceanu et al. 2009), and several clinical observations indicated that lithium might posses an antimicrobial and antiviral performance both in vitro and in vivo (Amsterdam et al. 1990; Lieb 2002, 2004, 2007). There are a few studies that draw differing conclusions about the antimicrobial activity of lithium ions against bacteria. A recent study found that the enzyme alcohol dehydrogenase of M. extorquens was inhibited to 28% by 5 mmoll1 lithium (34.5 ppm; Koutsompogeras et al. 2006). Others authors reported a toxicity of lithium toward Escherichia coli at a concentration of 4,830 ppm and the arrest of cell division and differentiation of protozoa of the group Trypanosomatidae by 200 mmoll1 LiCl (1,380 ppm Li+; Spiegel and Soares 1999). Some other authors have demonstrated that 100 mmoll1 LiCl (690 ppm Li+) in the presence of various sole carbon sources inhibited the growth of Salmonella thyphimurium and E. coli due to an accumulation of Li+ in the cell cytoplasm that might cause inhibition of the enzyme

pyruvate kinase (Niiya et al. 1980; Umeda et al. 1984), whereas others ruled out the enhancement of the antimicrobial performance of fluoride by lithium (Eisenberg et al. 1991). Within the scope of this investigation, attempts were made to hypothesise the enhancement mechanism of lithium in the presence of biocide. Experimental data showed that an outflow of cellular constituents absorbing at 260 nm only took place in the presence of the biocide blend (EDDM/CMIT/MIT), suggesting that the release is caused by interaction of the cell envelope with both formaldehyde and isothiazolinones. In fact, neither formaldehyde nor isothiazolinones are membrane-active biocidesthe kind that disrupt the cell membrane and thus cause an outflow of cytoplasmic constituents (McDonnell and Russell 1999). However, the release of cell wall components absorbing at 260 nm might be related to the initiation of cellular autolysis caused by formaldehyde (Denyer 1995; Musterman and Morand 1977) and the rapid association of CMIT/MIT with the cells (Diehl and Chapman 1999). On the other hand, an outflow of potassium ions was observed only when lithium was present alone or in combination with the biocide (EDDM/CMIT/MIT). The greatest efflux was observed with the combination of lithium and biocide. A leakage of potassium has been described as the first sign of an increase in membrane permeability and biocide-induced membrane damage (Lambert and Hammond 1973). Furthermore, the presence of lithium leads to an increase of absorbance at 600 nm, a swelling of the cells that is probably the result of water diffusing into the cell due to a rise in salinity. Previous

Fig. 6 Influence of increasing lithium concentrations on the membrane potential of rECM WMD culture cells in standard calcium carbonate slurry determined by means of the membrane potential sensitive dye Disc3(5) using the CellFacts II instrument. Lithium was added to the slurry culture, and fluorescence was determined after 30 min. Values are presented as meanstandard deviation of two independent replicates

438

Appl Microbiol Biotechnol (2011) 89:429439

studies described an increase of the absorbance at 600 nm to be related to an alteration of the refractive index, the release of light scattering material or to swelling of the cells due to intracellular accumulation of biocide (Sheppard et al. 1997; Chun and Hancock 2000). Another possible explanation for the slight increase in absorbance at 600 nm is the effect of lithium on bacterial cell division. This might result in the promotion of the cytokinesis within the bacterial population and thus to a shift of the average population size to smaller cells, as well as to a change in the refractive index. In a similar fashion, the data provided by the NPN uptake test demonstrated that lithium significantly increased the permeability of the cell membrane to the hydrophobic NPN probe. Based on the results, it can be assumed that the biocideenhancing action of lithium is the consequence of interference with the Na+(Li+)/H+-antiporter systems, and in turn, the cation balance of the bacterial cells is compromised. While the cytoplasmic membrane is not permeable to cations without the involvement of membrane transport systems, the Na+(Li+)/H+-antiporter exchanges a cytoplasmic Na+ or Li+ for an external H+ (Inaba et al. 1994, 1997). In an alkaline environment, like that found in calcium carbonate slurries, monovalent ions and especially Na+ ions are key players in pH homeostasis, regulation of the Na+ content and cell volume (Padan et al. 2005; Hunte et al. 2005). Moreover, these cationic antiporters are also involved in the cationic detoxification of the cells since the cytotoxicity of excess sodium and lithium increases as the pH rises, and depends on the cytoplasmic K + concentration (Padan et al. 2005; Hunte et al. 2005). An increased cationic load as a result of dosing lithium into the alkaline calcium carbonate slurry probably results in a toxic accumulation of Na+ ions due to the competitive saturation of the antiporter system with Li+. The elevated extrusion of Li+ ions consequently influences the transmembrane electrochemical gradient of protons or even the transmembrane pH gradient. Other authors postulated that the efflux of potassium ions probably results from the dissipation of the transmembrane electrochemical gradient of protons (Kroll and Patchett 1991). In calcium carbonate slurry supplemented with lithium, a depolarisation of the bacterial cell membrane and hence the loss of membrane integrity was observed by means of the fluorescent dye DiSC3(5). As a consequence, dissipation of the proton motive force leads to the leakage of essential molecules (Friedrich et al. 1999). Another possible explanation is that either the formaldehyde or the isothiazolinones deriving from the EDDM/ CMIT/MIT biocide inhibited the antiporter system thus causing an inhibitory growth effect because of the disabled extrusion of surplus cations. Finally, lithium and magnesium are alike in some chemical and physical characteristics such as their atomic and ionic radii; these similarities are known as a diagonal relationship. Chemical resemblances might lead to an

interaction between lithium and the magnesium ions and destabilise the lipopolysaccharide structure of the cell membrane in a fashion similar to the metal chelator EDTA. In calcium carbonate slurries, the interference of lithium with the cationic antiporter is probably also supported by the increased pH observed after lithium carbonate is introduced. However, an enhancement of the biocide due to the alkaline stress caused by the pH shift in the presence of lithium carbonate can be ruled out. Control experiments using potassium carbonate instead of lithium carbonate did not show any biocide enhancement activity. Summing up, experimental data provided evidence that lithium enhances the biocidal activity of various biocides and provides a novel technology to minimise biocide resistance in WMD. Additionally, lithium can be introduced into calcium carbonate slurries via the dispersant efficiently and without any additional dosage steps. This approach may solve the emerging problems of bacterial resistance or adaptation by using an enhancer compound to assure the effectiveness of the biocide at its action site. Such a biocide enhancer represents a breakthrough that offers a potential tool to revolutionise the consumption of biocidal agents in the WMD producing industry. Two final items of particular importance are the specifications of enhancer compounds, which must be chemically compatible with WMD, and the economic and ecological sustainability of their deployment.
Acknowledgments We thank Daniel Oschwald, Michael Jggi, Daniel Schild and Karin Fischer for their technical assistance with analytical measurements. We also thank Jan Sinstadt for his critical reading of the manuscript.

References
Acar JF (2000) Antibiotic synergy and antagonism. Med Clin North Am 84:13911406 Alakomi HL, Paananen A, Suihko ML, Helander IM, Saarela M (2006) Weakening effect of cell permeabilizers on gram-negative bacteria causing biodeterioration. Appl Environ Microbiol 72:46954703 Amsterdam JD, Maislin G, Rybakowski J (1990) A possible antiviral action of lithium carbonate in herpes simplex virus infections. Biol Psychiatry 27:447453 Avery SV (1995) Caesium accumulation by microorganisms: uptake mechanisms, cation competition, compartmentalization and toxicity. J Ind Microbiol 14:7684 Ayres HM, Furr JR, Russell AD (1999) Effect of permeabilizers on antibiotic sensitivity of Pseudomonas aeruginosa. Lett Appl Microbiol 28:1316 Berenbaum MC (1978) A method for testing for synergy with any number of agents. J Infect Dis 137:122130 Botelho MG (2000) Fractional inhibitory concentration index of combinations of antibacterial agents against cariogenic organisms. J Dent 28:565570 Chun W, Hancock RE (2000) Action of lysozyme and nisin mixtures against lactic acid bacteria. Int J Food Microbiol 60:2532 Cruceanu C, Alda M, Turecki G (2009) Lithium: a key to the genetics of bipolar disorder. Genome Med 1:79

Appl Microbiol Biotechnol (2011) 89:429439 Denyer SP (1995) Mechanisms of action of antibacterial biocides. Int Biodeterior Biodegrad 36:227245 Denyer SP, Maillard JY (2002) Cellular impermeability and uptake of biocides and antibiotics in Gram-negative bacteria. J Appl Microbiol 92(Suppl):3545 Denyer SP, Hugo WB, Harding VD (1985) Synergy in preservative combinations. Int J Pharm 25:245253 Denyer SP, Hugo WB, Harding VD (1986) The biochemical basis of synergy between the antibacterial agents, chlorocresol and 2phenylethanol. Int J Pharm 29:2936 Di Maiuta N, Hubschmid S, Giuliani N, Schwarzentruber P, Dow CS (2009) Microbial degradation of formaldehyde in white mineral dispersions preserved with formaldehyde-releasing biocides. Int Biodeterior Biodegrad 63:769777 Diehl MA, Chapman JS (1999) Association of the biocide 5-chloro-2methyl-isothiazol-3-one with Pseudomonas aeruginosa and Pseudomonas fluorescens. Int Biodeterior Biodegrad 44:191199 Eisenberg AD, Oldershaw MD, Curzon ME, Handelman SL (1991) Effects of fluoride, lithium, and strontium on growth and acid production of mutants streptococci and Actinomyces viscosus. Caries Res 25:179184 Friedrich C, Scott MG, Karunaratne N, Yan H, Hancock RE (1999) Salt-resistant alpha-helical cationic antimicrobial peptides. Antimicrob Agents Chemother 43:15421548 Hancock RE, Wong PG (1984) Compounds which increase the permeability of the Pseudomonas aeruginosa outer membrane. Antimicrob Agents Chemother 26:4852 Harrison JJ, Turner RJ, Ceri H (2005) High-throughput metal susceptibility testing of microbial biofilms. BMC Microbiol 5:53 Helander IM, Mattila-Sandholm T (2000) Fluorometric assessment of gram-negative bacterial permeabilization. J Appl Microbiol 88:213219 Helander IM, Alakomi H-L, Latva-Kala K, Koski P (1997) Polyethyleneimine is an effective permeabilizer of Gram-negative bacteria. Microbiology 143:31933199 Hodges NA, Hanlon GW (1991) Detection and measurement of combined biocide action. Soc Appl Bacteriol Tech Ser 27:297 310 Hunte C, Screpanti E, Venturi M, Rimon A, Padan E, Michel H (2005) Structure of a Na+/H+ antiporter and insights into mechanism of action and regulation by pH. Nature 435:11971202 Inaba K, Kuroda T, Shimamoto T, Kayahara T, Tsuda M, Tsuchiya T (1994) Lithium toxicity and Na+(Li+)/H+ antiporter in Escherichia coli. Biol Pharm Bull 17:395398 Inaba K, Utsugi J, Kuroda T, Tsuda M, Tsuchiya T (1997) Na+(Li+)/ H+ antiporter in Pseudomonas aeruginosa and effect of Li+ on cell growth. Biol Pharm Bull 20:621624 Khknen E, Nordstrm K (2008) Toward a nontoxic poison: current trends in (European Union) biocides regulation. Integrated Environ Assess Manag 4:471477 Khalil H, Chen T, Riffon R, Wang R, Wang Z (2008) Synergy between polyethylenimine and different families of antibiotics against a resistant clinical isolate of Pseudomonas aeruginosa. Antimicrob Agents Chemother 52:16351641 Koutsompogeras P, Kyriacou A, Zabetakis I (2006) Characterizing NAD-dependent alcohol dehydrogenase enzymes of Methylobacterium extorquens and strawberry (Fragaria x ananassa cv. Elsanta). J Agric Food Chem 54:235242 Kroll R, Patchett R (1991) Biocide-induced perturbations of aspects of cell homeostasis: intracellular pH, membrane potential and solute transport. Soc Appl Bacteriol Tech Ser 27:189202 Lambert PA, Hammond SM (1973) Potassium fluxes, first indications of membrane damage in micro-organisms. Biochem Biophys Res Commun 54:796799 Lambert RJ, Lambert R (2003) A model for the efficacy of combined inhibitors. J Appl Microbiol 95:734743

439 Lambert RJ, Johnston MD, Hanlon GW, Denyer SP (2003) Theory of antimicrobial combinations: biocide mixturessynergy or addition? J Appl Microbiol 94:747759 Lambert RJ, Hanlon GW, Denyer SP (2004) The synergistic effect of EDTA/antimicrobial combinations on Pseudomonas aeruginosa. J Appl Microbiol 96:244253 Lieb J (2002) Lithium and antidepressants: inhibiting eicosanoids, stimulating immunity, and defeating microorganisms. Med Hypotheses 59:429432 Lieb J (2004) The immunostimulating and antimicrobial properties of lithium and antidepressants. J Infect 49:8893 Lieb J (2007) Lithium and antidepressants: stimulating immune function and preventing and reversing infection. Med Hypotheses 69:811 Maillard JY, Russell AD (2001) Biocide activity: prospects for potentiation. Chim Oggi 12:3336 McDonnell G, Russell AD (1999) Antiseptics and disinfectants: activity, action, and resistance. Clin Microbiol Rev 12:147179 Morrier JJ, Suchett-Kaye G, Nguyen D, Rocca JP, Blanc-Benon J, Barsotti O (1998) Antimicrobial activity of amalgams, alloys and their elements and phases. Dent Mater 14:150157 Mulet M, Lalucat J, Garcia-Valdes E (2010) DNA sequence-based analysis of the Pseudomonas species. Environ Microbiol 12:15131530 Musterman JL, Morand JM (1977) Formaldehyde as a preservative of activated sludge. Res J Water Pollut Control Fed 49:4556 Niiya S, Moriyama Y, Futai M, Tsuchiya T (1980) Cation coupling to melibiose transport in Salmonella typhimurium. J Bacteriol 144:192199 Padan E, Bibi E, Ito M, Krulwich TA (2005) Alkaline pH homeostasis in bacteria: new insights. Biochim Biophys Acta 1717:6788 Papagianni M, Avramidis N, Filioussis G (2007) High efficiency electrotransformation of Lactococcus lactis spp. lactis cells pretreated with lithium acetate and dithiothreitol. BMC Biotechnol 7:16 Ramon AM, Fonzi WA (2009) Genetic transformation of Candida albicans. Methods Mol Biol 499:169174 Rohleder J and Huwald E. (2001) CalciumcarbonatA modern resource, In: Calcium Carbonate: From the Cretaceous Period into the 21st Century. (Tegethoff FW, Rohleder J and Kroker E, eds.), Birkhuser, Basel Schwarzentruber P (2003) Microbiological characterisation of white pigment slurriesa strategy for bacteria management. PhD Thesis, Department of Biological Sciences, University of Warwick, Coventry Schwarzentruber P and Gane PAC. (2005) Application of microbicides for the storage protection of mineral dispersions, In: Directory of Microbicides for the Protection of Materials. (Paulus,W. ed.), Springer Netherlands, pp. 251262. Sheppard FC, Mason DJ, Bloomfield SF, Gant VA (1997) Flow cytometric analysis of chlorhexidine action. FEMS Microbiol Lett 154:283288 Silvestry-Rodriguez N, Sicairos-Ruelas EE, Gerba CP, Bright KR (2007) Silver as a disinfectant. Rev Environ Contam Toxicol 191:2345 Sondossi M (1990) The potentiation of industrial biocide activity with Cu2+. I: synergistic effect of Cu2+ with formaldehyde. Int Biodeterior 26:5161 Spiegel CN, Soares MJ (1999) Biological effects of lithium chloride on Herpetomonas muscarum muscarum and Blastocrithidia culicis (Kinetoplastida: Trypanosomatidae). J Parasitol 85:729733 Umeda K, Shiota S, Futai M, Tsuchiya T (1984) Inhibitory effect of Li+ on cell growth and pyruvate kinase activity of Escherichia coli. J Bacteriol 160:812814 Vaara M (1992) Agents that increase the permeability of the outer membrane. Microbiol Rev 56:395411 Vestby LK, Lonn-Stensrud J, Moretro T, Langsrud S, Aamdal-Scheie A, Benneche T, Nesse LL (2009) A synthetic furanone potentiates the effect of disinfectants on Salmonella in biofilm. J Appl Microbiol 108:771778