This article appeared in a journal published by Elsevier.

The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elseviers archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/authorsrights

Author's personal copy

Food Chemistry 147 (2014) 2533

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Biochemical heterogeneity of malt is caused by both biological variation

and differences in processing: I. Individual grain analyses of biochemical
parameters in differently steeped barley (Hordeum vulgare L.) malts
Maik Kleinwchter a, Christian Mller b, Frank-Jrgen Methner b, Dirk Selmar a,
a
b

Institut fr Panzenbiologie, Technische Universitt Braunschweig, Mendelssohnstrae 4, D-38106 Braunschweig, Germany

Institut fr Biotechnologie, Technische Universitt Berlin, Seestrae 13, D-13353 Berlin, Germany

a r t i c l e

i n f o

Article history:
Received 30 April 2013
Received in revised form 28 August 2013
Accepted 16 September 2013
Available online 25 September 2013
Keywords:
Hordeum vulgare L.
Malting
Biochemical homogeneity
Steep aeration

a b s t r a c t
Using individual grain analyses, the degree of inherent biological variation in germinating barley seeds
has been established. Even under homogenous laboratory conditions, the activities of the germinationrelated enzymes a-amylase, b-amylase and b-glucanase varied by a factor of two to three. The comparison with single grain analyses of different industrially produced malts (steeping systems without
aeration, with air suction and pressurised aeration) revealed that the heterogeneity of these malts nearly
tripled. This increase may be due to the gradients in O2 and CO2 that arise in large industrial steeping
vessels. The most homogenous malting in the industrial systems was achieved without any aeration
during steeping. Therefore, to improve homogeneity, the common practise of steep aeration should be
omitted. Germination progression was quite different within the three exhaustively aerated attempts,
which indicated that gaseous composition was not the only factor affecting germination progression.
2013 Elsevier Ltd. All rights reserved.

1. Introduction
Malting is the process of converting cereal seeds into malt for
use in brewing or distilling. In the rst stage of this process, the
cereal seeds1, e.g., barley (Hordeum vulgare L.), are germinated. Germination is induced by the imbibition of the barley seeds in steeping
vessels for 12 d. This so-called steeping process generally includes
23 immersion phases which are split up with dry rest periods. After
steeping, the germinating grains are incubated for a further 34 d in
germination boxes for sprouting. Seedling development is then terminated by kiln-drying. The aim of the germination phase of malting
is to induce the synthesis of germination-related enzymes, i.e., cell
wall and protein-degrading enzymes (e.g., b-glucanases and different proteases), as well as starch mobilising enzymes (e.g., a- and
b-amylases). Owing to the concerted activities of these enzymes,
the starch stored in amyloplasts can be hydrolysed, leading to the re Corresponding author. Tel.: +49 (0) 531 391 5881; fax: +49 (0) 531 391 8180.
E-mail addresses: m.kleinwaechter@tu-bs.de (M. Kleinwchter), c.mueller@
tu-berlin.de (C. Mller), frank-juergen.methner@tu-berlin.de (F.-J. Methner),
d.selmar@tu-bs.de (D. Selmar).
1
Sensu stricto barley grains represent fruits, i.e., caryopses. Due to the specic
fusion of seed coat and pericarp, it is not possible to distinguish between fruit and
seed. With respect to most aspects, grains indeed should be denoted as fruits.
However, in all works that mainly deal with the characteristic physiological attributes
of seeds, e.g., germination, the term seed should be used, since the basic metabolic
reactions are related to seed physiology and not to fruit physiology. Accordingly, in
this treatise, barley grains are denoted as seeds.
0308-8146/$ - see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodchem.2013.09.090

lease of low molecular carbohydrates, such as glucose, maltose, maltotriose, etc. These provide nutrition to the developing embryo.
Furthermore, a suitable degree of cellular modication is required
for the subsequent brewing process, since the brewing performance
of the malt (e.g., the lterability of the mash) strongly depends on
the degree of breakdown in cellular structures. In the third stage
of malting, kilning, the seedlings are dried and slightly roasted in a
kiln-oven at temperatures ranging from 50 C to 90 C. This rapidly
aborts all metabolic processes by water deprivation and heat effects
and, due to the high temperatures, Maillard reactions take place,
which lead to the formation of several of the typical colour and avour compounds found in malt (for detailed information on malting,
refer to Kunze, 2010; Narziss, 1999).
Many studies have demonstrated that barley seed germination
and early seedling development are strongly inuenced by external factors, such as the temperature, the availability of O2 and
the prevailing CO2 concentration (Kleinwchter, Meyer, & Selmar,
2012; Narziss, 1999). It is well-known that a certain concentration
of O2 is required for regular barley seed germination and seedling
development: under low O2 partial pressure, the germination progression is retarded, while it is completely inhibited in the absence
of O2 (e.g., Kleinwchter et al., 2012; Narziss, 1999). After being
incubated for 78 d under anoxia, barley seeds even die (Perata,
Loreti, Guglielminetti, & Alpi, 1998). With respect to the underlying
mechanisms, Hanson and Jacobson (1984) reported that the transcription of germination-related enzymes (i.e., a-amylase) did
not occur under anoxia. For this reason, barley embryos are

Author's personal copy

26

M. Kleinwchter et al. / Food Chemistry 147 (2014) 2533

supposed to die due to carbohydrate starvation. However, further

details on the inhibition of barley seed germination under anoxious
conditions have not been elucidated until now. Interestingly, similar inhibitory effects on barley seed germination have recently
been shown for CO2 despite the presence of ambient O2 concentrations (Kleinwchter et al., 2012). Yet, the mechanisms behind this
phenomenon are unclear.
In line with these ndings, it is commonplace in industrial malting to aerate extensively during steeping and subsequent germination in order to avoid delays in germination caused by O2
deciency or enhanced CO2 concentrations (Albers, Drost, & Pesman, 1983; French & McRuer, 1990; Gibbons, 1983). Nonetheless,
the practise of steep aeration is considered controversial by some
authors; Wilhelmson et al. (2006) showed that aeration at the
beginning of the steeping process had only a limited effect on barley seed germination and malt quality. Therefore, aeration seems
not to be necessary at the beginning of steeping. Other researchers
generally question the common practise of steep aeration, since an
excessive supply of O2 could also strongly promote the growth of
the embryos, which would lead to enhanced substance losses during malting (Cantrell, 1987; Enari, Linnahalme, & Linko, 1961; Kelly
& Briggs, 1992).
It is also known that besides the overall levels of biochemical
parameters (e.g., enzyme activities and b-glucan contents), their
distribution, i.e., the homogeneity in the batch, also plays an
important role in the brewing performance of the malt (Palmer,
2000), but the impact of steep aeration on the biochemical homogeneity of malt has not been analysed until now. In the last decade,
de S & Palmer published three related articles (de S & Palmer,
2004, 2006; Palmer, 2000) which exhaustively dealt with the
determination of biochemical homogeneity (i.e., that of b-glucanase activity, b-glucan contents and nitrogen content), as well as
with the impact of biochemical homogeneity on brewing performance (i.e., the lterability of the mash). Unfortunately, the standard (recommended) methods for malt analysis are not capable
of detecting inhomogeneities, since these methods are based on
the analysis of pooled, homogenised samples, from which only
average values can be obtained. Thus, there is a need to perform
individual grain analysis and to arrange and analyse the data in frequency distributions instead of displaying them as average values
(Palmer, 2000). Furthermore, the authors found that the homogeneity of the protein and the b-glucan breakdown had a direct impact on the brewing performance of the malt. In turn, de S and
Palmer (2004) came to the conclusion that inhomogeneities in
endosperm modication, where a small, but signicant, number
of grains showed limited breakdown of cell wall (b-glucan) materials, could cause severe brewhouse problems with respect to the
lterability of the mash.
This study addresses how the processing, in particular steep
aeration, impacts on the biochemical homogeneity of malt. Initially, the corresponding enzyme assays for standard malt analysis
had to be scaled-down so that they were able to quantify the enzyme activities in individual barley grains (3050 mg). In order
to monitor the progression and homogeneity of the germination
processes, the activities of a- and b-amylase and b-glucanase were
determined. Additionally, c-aminobutyric acid (GABA) contents
were analysed as an indicator for plant stress metabolism (Kleinwchter et al., 2012).
In a second step, small-scale laboratory germinations to ensure
uniform germination conditions were performed in order to establish a reference for assessing the biochemical homogeneity of the
malts produced under different processing conditions. Subsequently, samples from three different steeping systems (without
aeration, air suction and pressurised aeration) were analysed by
individual grain analysis. In order to assess homogeneity, the data
were arranged in graphs that display the distribution of the values

in combination with the average value and the coefcient of variation (measure for the degree of homogeneity).
2. Materials and methods
2.1. Germination under laboratory conditions (continuous aeration)
The barley seeds (Hordeum vulgare L., var. Braemar) used for this
experiment derived from a single batch of malting barley (Cargill
Deutschland GmbH, Salzgitter, Germany). The barley was steeped
in a steeping vessel and then germinated for several days in special
germination boxes to mimic the industrial malting process in a
simplied manner. For this, exactly 150 g of the barley seeds were
soaked for 24 h in 1 L of tap water in an Erlenmeyer ask. Pressurised air was introduced through a gas frit (30 L/h). To ensure an
evenly distribution of oxygen, the steeping good was consistently
mixed using a magnetic stirrer. After steeping, the moisture content of the barley seeds was about 42%. Then the seeds were spread
out on a slightly elevated sieve placed on a plastic socket in a closable plastic container (26  10  9 cm). The plastic socket allowed
the seedlings to be uniformly exposed to a stream of air (30 L/h)
throughout the entire germination period (72 h). The air introduced into the germination system was saturated with water using
an impinger. The layer thickness of the germination good was
<2 cm and the temperature was held at 18 C during the entire germination process (steep and germination box).
After 0, 24, 48, 72 and 96 h, sample material (about 510 g) was
removed from the steeping vessel or the germination box. The
plant material was directly shock-frozen in liquid N2. Single, randomly chosen grains from these samples were homogenized using
a Retsch MM 200 ball mill and stored at 20 C before enzyme and
GABA assays were performed.
2.2. Industrial-scale germination with varying steep aeration
The same raw material, barley (H. vulgare, var. Braemar) was
transported from a storage silo to the different steeping vessels.
Three different steeping systems, as they are in use in commercial
malt production, were analysed: without aeration; with air suction
during dry rests, and with temporary pressurised aeration. The
steeping vessel for the pressurised aeration was rectangular and
the two steeping vessels for the other two processes were cylindroconical, having diameters of 34 m and depths of between
3.54.0 m, which gave capacities of 27, 15 and 17 t of barley,
respectively. The steeping programs were as follows:
Without aeration: 1st wet steep (3 h), 1st dry rest (25 h), 2nd
wet steep (4 h).
Air suction: 1st wet steep (3 h), 1st dry rest (22.5 h), 2nd wet
steep (4.5 h). Ambient air (T = app. 20 C) was exhaustively sucked
through the grain bed during dry rests from top to bottom.
Temporary pressurised aeration: 1st wet steep (3.5 h), 1st dry
rest (16 h), 2nd wet steep (0.6 h), 2nd dry rest (23 h), 3rd wet steep
(0.25 h). Ambient air (T = app. 20 C) was exhaustively forced
through the grain bed from bottom to top every 12 h during the
wet steeps and constantly for 14 h during each dry rest.
The temperature varied between 15 C and 19 C during steeping, i.e., due to the metabolic activity of the germinating seeds, the
temperature increased slightly in the course of the steeping process. The O2 content in the steeping water and in the air of the
grain bed during dry rests was measured. Except for the aerated
phases, where the conditions were almost O2-saturated, the O2
was depleted after about 3 h during the rst wet steep, after 7 h
during the rst dry rest and after about 3040 min during the second wet steep. After steeping, the differently steeped barley seeds
were separately transferred to germination boxes (two rectangular

Author's personal copy

M. Kleinwchter et al. / Food Chemistry 147 (2014) 2533

Saladin boxes and one circular box). In the germination boxes, the
barley seedlings were intensively aerated with humidied and
tempered air. The beds of barley were thoroughly stirred three
times a day by vertical screws. The layer thicknesses of the germination beds were 0.9 m (Saladin boxes) and 1.20 m (circular box).
The temperature in the different germination boxes varied between 15 C and 19 C. Samples were taken after 96 h total germination time (including the steeping period) at three randomly
chosen points in the barley beds using a sampling lance. After a
thorough mixing, the plant material was directly shock-frozen in
liquid N2. Single, randomly chosen grains from these samples were
then homogenized using a Retsch MM 200 ball mill and stored at
20 C before enzyme activity and GABA analysis were performed.
2.3. Determination of a- and b-amylase and b-glucanase activities
In order to determine a- and b-amylase and b-glucanase activities, standard assays from Megazyme (Ireland) were used (Ceralpha, Betamyl-3, b-amylase, and malt & Bacterial b-glucanase &
cellulase, respectively). The substrate, p-nitrophenyl maltoheptaoside, was used to determine a-amylase activity. Its hydrolysis
yields p-nitrophenyl maltosaccharide, which is further cleaved in
the presence of excess a-glucosidase to yield glucose and p-nitrophenol, the concentration of which is measured photometrically
(400 nm). The betamyl-3 method, used for the determination of
b-amylase activity, employs p-nitrophenyl-b-maltotrioside as the
substrate yielding p-nitrophenyl-b-glucoside. This is hydrolysed
by b-glucosidase and gives rise to glucose and p-nitrophenol,
which is quantied photometrically as in the a-amylase assay.
Analysis of b-glucanase activity is based on the cleavage of an
insoluble azo-barley glucan substrate. When the dyed substrate
is hydrolysed by glucanases, dyed fragments are liberated, which
are then separated from the insoluble substrate by centrifugation.
The absorbance (590 nm) of the supernatant corresponds to the
glucanase activity. The standard procedures recommended by the
manufacturer were used, but the sample weight and extraction
buffer volumes were reduced proportionally. In order to determine
the a- and b-amylase activities, the homogenised plant materials
of single barley grains (3050 mg, 15 per time point) were extracted with 500 lL of extraction buffer, corresponding to a reduction of 1/10 in the suggested values. In the case of b-glucanase,
single homogenised barley grains were extracted with 800 lL of
extraction buffer, which also corresponded to a reduction of 1/10.
Corresponding comparisons of the mean enzyme activities gained
from 15 individual grain analyses with those from the analyses
of pooled samples proved the applicability of the reduction of 1/
10 for both assays.
2.4. Determination of GABA contents
GABA was quantied by HPLC according to Bytof, Knopp, Schieberle, Teutsch, and Selmar (2005). Before extraction, 20 mg of
the homogenised barley grains was spiked with norvaline
(40 nmol) as internal standard. For extraction, 10 mL of sulfosalicylic acid (4% w/v) was added to the freeze dried powder and then
the extraction mixture was sonicated for 30 min and incubated
overnight at 4 C. After mixing, a 2 mL aliquot of the extraction
solution was centrifuged and ltered. The amino acids were derivatised with o-phthaldialdehyde (OPA) before HPLC analysis using
a Spark Holland Midas autosampler. The amino acid derivatives
were separated on a C18 column (Nucleosil 100, 5 lm, Macherey
Nagel, 250  4.0 mm) using a binary gradient (A, 5% MeOH, 5%
ACN, 2% THF, 88% 50 mM sodium acetate buffer, pH 6.2; B, 40%
MeOH, 40% ACN, 20% sodium acetate buffer) at a ow rate of
1.3 mL/min. The derivatives were detected using an RF-551 Shimadzu uorescence detector (kex = 334 nm; kem = 425 nm) and

27

quantied by external calibration. Duplicate analyses were carried

out on each sample.
2.5. Data analysis
Each series of analyses consisted of 15 single analyses of 15
individual barley grains. In order to display the biochemical homogeneity of the malts, the data were arranged in frequency distributions (establishment of a reference) or in bar charts (comparison
between different steeping systems). In the bar charts, the mean
value of the 15 single analyses, the maximal relative differences
from the mean value, and the coefcient of variation (relative standard deviation for the data set) are displayed. The GABA contents
were not statistically analysed to clearly distinguish between this
stress related low molecular metabolite and the enzymes, which
are used as markers for germination processes.
3. Results and discussion
3.1. Small-scale germination assays: estimation of the inherent
biological variation
Before the variability caused by the different industrial processing methods could be estimated, the inherent biological variation
had to be determined. Thus, small-scale germination assays, ensuring identical conditions for each seed, were conducted. The principal nding of this experiment was the large individual extent to
which the analysed biochemical parameters varied (a-amylase,
b-amylase and b-glucanase activity as well as GABA contents). This
can be seen by the wide spread of the values in the distribution
charts after 96 h of germination (Fig. 1). However, these variations
were not seen in the ungerminated barley seeds. The variability
increased over time during seedling development, which is shown
by the time-dependent broadening of the distribution pattern.
Fluctuations by a factor of two to three seem to be standard (e.g.,
a-amylase activity after 96 h of germination; Fig. 1A). A detailed
evaluation revealed that the parameters followed different
progression patterns.
The a-amylase and b-glucanase activity (Fig. 1A and C) was absent or very low in the ungerminated barley seeds (0 and 24 h), a
fact that is well-known from the analysis of pooled samples
(Narziss, 1999). With respect to the variations in these two parameters, the ungerminated seed material was nearly homogenous.
After being incubated for more than 24 h, the distribution patterns
of a-amylase and b-glucanase activity shifted to the right, which
reected an increase in enzyme activity. Intriguingly, the individual levels of enzyme activity in the older seedlings varied more
than at the beginning of germination, which is shown by the much
broader width of the distribution patterns. In particular, after 96 h,
the a-amylase activity (Fig. 1A) varied from approximately 1 to
3 nkat/mg DW, corresponding to a factor of three. Over the course
of germination, biosynthetic activities are induced; the corresponding enzymes are synthesised and their activities increase
steadily. The speed of the increase in a-amylase and b-glucanase
activity was different within individual seedlings because of differences in their individual germination-associated biosynthetic
activity. However, since particular attention was paid to ensuring
uniform germination conditions (e.g., temperature control, low
layer thickness, uniform ow of air through the bed etc.), differences in the metabolic reactions caused by varying external conditions can be excluded. The differences must therefore be due to the
inherent biological variation.
Differing patterns for the changing enzyme activity were found
for b-amylase activity (Fig. 1B). The ungerminated seeds (0 h) already revealed distinct enzyme activities, which varied from 0.15

Author's personal copy

M. Kleinwchter et al. / Food Chemistry 147 (2014) 2533

-amylase

frequency of a measured value

A 15

B
0h

0
15

24 h
0
15

48 h
0
15

72 h
0
15

96 h

frequency of a measured value

28

0
0.5

1.5

24 h
0
15

48 h
0
15

72 h
0
15

96 h

2.5

-amylase activity [ nkat /mg DW] (0.5 nkat / mg- intervals)

-glucanase

15

0h
0
15

24 h
0
15

48 h
0
15

72 h
0
15

96 h
0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

-amylase activity [ nkat /mg DW] (0.05 nkat / mg- intervals)

frequency of a measured value

frequency of a measured value

0h
0
15

0
0

-amylase
15

GABA
15

0h
0
15

24 h
0
15

48 h
0
15

72 h
0
15

96 h
0

10

12

14

16

18

-glucanase activity [ nkat /mg DW ] ( 0.5 nkat/g -intervals)

10

20

30

40

GABA concentration [ mg/100 g DW ] ( 2 mg/100 g -intervals)

Fig. 1. Distribution of biochemical parameters in barley seeds during germination and early seedling growth. Special emphasis was put on ensuring uniform germination
conditions (e.g., temperature control, low layer thickness and uniform ow of air). Germination processes were monitored in terms of (A) a-amylase, (B) b-amylase and (C) bglucanase activity. The level of stress metabolism (D) was determined using the concentration of c-aminobutyric acid (GABA).

to 0.4 nkat/mg DW, corresponding to a factor of two to three. Thus,

the variation range for b-amylase activity in the original, ungerminated barley seeds was quite similar to that found for a-amylase
and b-glucanase activity after the seed had been germinated for
96 h. At the beginning of the germination period (24 h), the mean
b-amylase activities increased slightly, which was shown by a minor shift in the distribution pattern to the right to higher values.
Subsequently, the b-amylase activities remained on a similar level
for the rest of the incubation period (4896 h). The variation range
remained relatively constant (by a factor of two to three). The nding that b-amylase, in contrast to a-amylase and b-glucanase, is already present in mature barley seeds, is in accordance with the
results of previous studies (Narziss, 1999; Sopanen & Laurire,
1989). The slight increase in b-amylase activity at the beginning
of germination (24 h) is not related to any additional de novo synthesis (Sopanen & Laurire, 1989), but due to the fact that some of
this enzyme occurs in a bound, inactive form, which is instantly released to its free, active form by proteases during germination.
The GABA distribution charts (Fig. 1D) show that the accumulation pattern of this typical stress metabolite was completely different from the distribution patterns of the enzymes. Most of the
barley seeds consistently had relatively low GABA concentrations,
ranging from 2 to 10 mg/100 g DW over the course of the entire
incubation period. However, with the progression of germination
and growth processes (7296 h), some grains accumulated extraordinarily high GABA concentrations (e.g., 24 and 44 mg/100 g DW;
96 h). The general germination-related increase in GABA concentration previously reported (Inatomi & Slaughter, 1971; Kleinwchter
et al., 2012), is not due to a slight uniform increase in the GABA contents in all seeds of a malt batch, but is caused by a low, but significant, number of deviators. One explanation of this unexpected

nding is that some grains may suffer from severe stress, whereas
others are not affected. In this context, it has to be noted that GABA
is accumulated in response to various stresses (e.g., high temperatures, water deciency, physical injuries, biotic stresses etc.; for review see Bown & Shelp, 1997; Satya Narayan & Nair, 1990).
Accordingly, there might be numerous particular causes, e.g., slight
injuries or spatial infections, which are responsible for the individual differences in GABA accumulation.
The data on the enzyme activity mentioned above were compared with the results of the individual grain analyses for b-glucanase activity reported by de S and Palmer (2004), which are the
only individual grain analyses published so far. This comparison
revealed that the variations by factors of two to three established
in this study have to be considered as very low. The variations in
b-glucanase activity reported by de S and Palmer were much
higher (between a factor of six to eight). One suggestion for this
may be that the micro-malting system used in their studies gave
rise to much higher heterogeneities, with respect to the germination conditions, compared to the small-scale apparatus used in this
study. Alternatively, it could be due to the fact that the barley used
by these authors was relatively inhomogeneous and had higher
inherent biological heterogeneity, which may be caused by uneven
ripening of the grains in the eld.
Since the differences in the germination state, and thus in the
activities of germination-related enzymes, are most pronounced
during the later stages of the malting process, the evaluation of
homogeneity of the malts derived from different processing approaches, should be performed using malts from the late phase
of germination (i.e., 96 h). Accordingly, this strategy was followed
in order to evaluate and to visualise the heterogeneity of malts derived from the different industrial-scale processing systems.

Author's personal copy

29

M. Kleinwchter et al. / Food Chemistry 147 (2014) 2533

3.2. Industrial-scale germinations: impact of steep aeration on the

biochemical homogeneity of malt
Malt samples produced via three different industrial-scale
steeping systems (without aeration, air suction during dry rests,
and temporary pressurised aeration) were investigated by single
grain analyses. The germination conditions were similar in all three
processes. The seeds were subjected to different conditions by the
three industrial approaches only in the rst 3040 h (for details see
Section 2.2.). The water content of the seeds after steeping was
about 4143% and this was more or less identical for all samples.
As mentioned above, the enzyme activities for establishing the heterogeneities were determined after 96 h. In the corresponding gures, the small-scale laboratory experiment data (germination
under almost homogeneous conditions) were added as a standard
and a reference (Figs. 25) for the data from the three different
industrial malting processes.
With respect to a-amylase activity (Fig. 2), the most obvious
nding was that the malt germinated under dened, uniform laboratory conditions (continuous aeration) was much more homogeneous than all the industrially processed malts. This is best shown
by a standard deviation (CV) of only 19.4% for the laboratory approach compared to 33.9% for the malt produced without aeration;
65.9% for air suction and 45.6% for pressurised aeration (Fig. 2). The
same situation is shown by the individual grain analyses for bamylase activity (Fig. 3). Again, the malt produced under dened
laboratory conditions had the lowest heterogeneity (CV = 26.1%)
followed by the industrial-scale process without steep aeration
(CV = 31.1%). Just like a-amylase, the b-amylase activity for the
two industrial-scale processes, where steep aeration was applied,
was considerably more heterogeneous (CV = 48.1% and 70.8%).
The data on b-glucanase activity conrmed the previous ndings
(Fig. 4) that the highest biochemical homogeneity was estimated

without steep aeration

4

CV = 33.9%

33.2%

= 2.2
2

92.5%

-amylase activity [ nkat / mg DW ]

-amylase activity [ nkat / mg DW ]

for the malt produced under dened laboratory conditions

(CV = 17.7%), whereas the malts derived from the industrial processes were less homogeneous, as shown by the CV-values, which
more than doubled. For further clarication, the CV-values for all
three enzymes (a- and b-amylase and b-glucanase) were averaged
for each processing method and the corresponding mean values
are displayed in Table 1. The malt germinated under dened and
uniform lab conditions had only small heterogeneities, shown by
a mean coefcient of variation of just 21.1%. In contrast, the mean
CV of all the industrially produced malts was more than double
(47.1%). Thus, the tremendous increase in heterogeneity must be
due to heterogeneities in the malting conditions at the industrial
facilities. Accordingly, it could be deduced that the most heterogeneous conditions are to be found in the steeping vessels with air
suction, followed by those with pressure aeration. The most
homogenous conditions seem to be found in industrial malting approaches where aeration during steeping is omitted.
The main differences between steeping systems with and without aeration should be due to the concentrations of O2 and CO2. In
this context, it has to be considered that the germinating seeds
consume O2 and liberate CO2. Whereas in unaerated approaches,
related equilibrium concentrations will result throughout the
batch, directed aeration will create massive differences. Substantial
gradients in the concentration of O2 and CO2 will be established
due to the large size of the industrial steeping vessels (Albers
et al., 1983). Moreover, the aeration may also affect the temperature of the malt bed (Narziss, 1999). The spatial heterogeneities
in gaseous composition and temperature create strongly inhomogeneous germination conditions, which will directly impact on
the progression of metabolic processes. As a consequence, the biochemical homogeneity of the complete malt batch could be impaired. In line with these arguments, in the lab scale trial, the
heterogeneities were much lower because the bed layer was only

air suction
4

2
156.5%

72.8%

individual seedlings

CV = 45.6%

2
98.3%

= 1.2
1
82.8%

individual seedlings

-amylase activity [ nkat / mg DW ]

-amylase activity [ nkat / mg DW ]

temporary steep aeration

4

= 0.7

individual seedlings

CV = 65.9%

laboratory
4

CV = 19.4%
3

= 2.1

30.6%

2
42.2%

individual seedlings

Fig. 2. The a-amylase activities of 15 individual barley seedlings of differently processed green malts after 96 h of germination: (A) without steep aeration, (B) with air
suction during dry rests and (C) with temporary pressurised steep aeration. As a reference, the data obtained by germination under dened laboratory conditions have been
added (D; see also Fig. 1A, 96 h). In addition, the average enzyme activity () of the 15 individual grain analyses is given. The heterogeneity of the malt is displayed by the
coefcient of variation (CV; relative standard deviation for the corresponding data set) and the maximum relative differences from the mean value.

Author's personal copy

M. Kleinwchter et al. / Food Chemistry 147 (2014) 2533

-amylase activity [ nkat / mg DW ]

without steep aeration

0.8

0.6

CV = 31.1%

77.7%

= 0.4

0.4

31.4%

0.2

-amylase activity [ nkat / mg DW ]

30

air suction

0.8

0.6

CV = 70.8%

0.4

= 0.3
0.2

75.3%

individual seedlings

individual seedlings

temporary steep aeration

0.8

CV = 48.1%

0.6

80.7%

0.4

= 0.3
0.2
97.0%

-amylase activity [ nkat / mg DW ]

-amylase activity [ nkat / mg DW ]

179.8%

laboratory

0.8

CV = 26.1%

0.6

38.7%

= 0.4

0.4

58.2%

0.2

individual seedlings

individual seedlings

Fig. 3. The b-amylase activities of 15 individual barley seedlings of differently processed green malts after 96 h of germination: (A) without steep aeration, (B) with air suction
during dry rests and (C) with temporary pressurised steep aeration. As a reference, the data obtained by germination under dened laboratory conditions have been added (D;
see also Fig. 1B, 96 h). In addition, the average enzyme activity () of the 15 individual grain analyses is also given. The heterogeneity of the malt is displayed by the coefcient
of variation (CV; relative standard deviation for the corresponding data set) and the maximum relative differences from the mean value.

16

without steep aeration

CV = 41.9%

B
50.5%

= 11.0

12

8
96.8%

-glucanase activity [ nkat / g DW ]

-glucanase activity [ nkat / g DW ]

CV = 48.0%

16

67.6%

12

= 9.1
8
91.6%

individual seedlings

individual seedlings

temporary steep aeration

16

CV = 39.0%

40.6%

12

= 9.4
8
77.3%

individual seedlings

-glucanase activity [ nkat / g DW ]

-glucanase activity [ nkat / g DW ]

air suction

laboratory

16

CV = 17.7%

35.0%

= 12.4

12
27.4%

individual seedlings

Fig. 4. The b-glucanase activities of 15 individual barley seedlings of differently processed green malts after 96 h of germination: (A) without steep aeration, (B) with air
suction during dry rests and (C) with temporary pressurised steep aeration. As a reference, the data obtained by germination under dened laboratory conditions have been
added (D; see also Fig. 1C, 96 h). In addition, the average enzyme activity () of the 15 individual grain analyses is also given. The heterogeneity of the malt is displayed by the
coefcient of variation (CV; relative standard deviation for the corresponding data set) and the maximum relative differences from the mean value.

about 12 cm, which provided identical gaseous concentrations

and temperatures for all seeds. Thus, with regards to malt homogeneity, the common practise of steep aeration should be omitted.

Yet, to substantiate these ndings further studies are required, in

which, for instance, the causes of biochemical heterogeneities
should be elucidated.

Author's personal copy

31

M. Kleinwchter et al. / Food Chemistry 147 (2014) 2533

GABA content [ mg / 100 g DW ]

GABA content [ mg / 100 g DW ]

120

80

40

= 20.0
0

120

80

D
GABA content [ mg / 100 g DW ]

GABA content [ mg / 100 g DW ]

individual seedlings

temporary steep aeration

120

80

= 36.3

40

= 36.2

40

individual seedlings

air suction

without steep aeration

laboratory

120

80

40

= 7.7
0

individual seedlings

individual seedlings

Fig. 5. The c-aminobutyric acid contents (GABA) of 15 individual barley seedlings of differently processed green malts after 96 h of germination: (A) without steep aeration,
(B) with air suction during dry rests and (C) with temporary pressurised steep aeration. As a reference, the data obtained by germination under dened laboratory conditions
have been added (D; see also Fig. 1D, 96 h). In addition, the average GABA content () of the 15 individual grain analyses is also given.

Table 1
Mean coefcients of variation ( CV) for all the a- and b-amylase and b-glucanase activities (n = 45) and mean GABA concentrations of the differently processed malts. The GABA
values were recalculated after the exclusion of statistical outliers (variance more than double the mean value).

CV (%)
GABA (mg/100 g DW)

Lab-scale

Industrial

Steep aeration

Without aeration

Air suction

Forced air

Mean

21.1
3.5

35.6
20.0

61.6
23.8

44.2
21.2

47.1
21.7

3.3. Industrial-scale germinations: impact of steep aeration on

germination progression
Apart from the estimation and evaluation of homogeneity, the
extent of the corresponding enzyme activities is also relevant for
malt quality. Accordingly, germination progression is of special
interest. Steep aeration is recommended for rapid germination
and thus for cellular modication and mobilization of starch. However, high germination rates are undesirable because of the increase in malting losses, and some researchers suggest that steep
aeration should be omitted (Kelly & Briggs, 1992). In contrast,
other researchers have reported delays in germination due to O2
deciency or enhanced CO2 concentrations, when steep aeration
was omitted (Albers et al., 1983; French & McRuer, 1990; Gibbons,
1983). Since the activities of a-amylase and b-glucanase rise dramatically as germination time increases, germination progression
can be evaluated by analyzing the activities of these enzymes.
The results from this study showed that the mean a-amylase activity of the aerated lab-scale approach and that of the industrial
steeping process without aeration were quite similar (2.1 and
2.2 nkat/mg DW, respectively; Fig. 2). The induction and progression of metabolic activity in the two batches of seeds were the
same, even though the O2 supply and the CO2 concentrations
should be quite different. Surprisingly, the mean activities of the
two aerated industrial-scale processes were drastically lower (1.2
and 0.7 nkat/mg DW, respectively; Fig. 2), which suggested that

germination had been strongly retarded. Similar effects were also

observed for the mean b-glucanase activities, although the differences observed were less pronounced. Again, the aerated lab approach and the industrial malting without aeration had very
similar mean activities (12.4 and 11.0 nkat/g DW, respectively;
Fig. 4), which were higher than the two aerated industrial maltings
(9.4 and 9.1 nkat/g DW, respectively). When changes in enzyme
activities are used to document germination progression, it has
to be considered that ungerminated seeds reveal already signicant b-glucanase activity, whereas a-amylase is totally absent in
ungerminated seeds (Fig. 1). When comparing the mean b-amylase
activity, all the values were very similar. This was expected as large
amounts of this enzyme are present in the ungerminated seeds and
only minor increases of its activity are likely to occur. Consequently, there are only minor differences in b-amylase activity,
even when germination progression differs strongly. A comparison
of the mean values conrms the suggestions made for a-amylase
and b-glucanase with regards to the differences in germination
rates. The mean values for b-amylase in the aerated lab approach
and industrial malting without aeration were identical (0.4 nkat/
mg DW; Fig. 3) and were slightly higher than the two aerated
industrial approaches (0.3 nkat/mg DW).
These ndings clearly indicate that germination progression
differs strongly, depending on the steeping conditions. As outlined
in the introduction, some researchers argue that the common practise of aeration in industrial steeping is not mandatory to achieve

Author's personal copy

32

M. Kleinwchter et al. / Food Chemistry 147 (2014) 2533

optimal germination progression (Wilhelmson et al., 2006),

although such an approach is generally applied in the daily malting
routine. According to these researchers, the omission of aeration
had no negative effect, unless the time of O2 deciency does not
exceed the initial phase of the process. Consequently, there should
be no difference in germination progression in both the aerated
and non-aerated approaches. In this context, the nding that the
germination rate in the two aerated industrial-scale approaches
was drastically lower than that in the non-aerated approach was
surprising. The identication of possible reasons for this becomes
even more complicated if one considers that the fastest germination occurred with the continuously aerated lab-scale approach,
where high O2 and low CO2 concentrations were present, and in
the non-aerated industrial approach, where the O2 levels were
low and the CO2 concentration high. These contrasting conditions
for the two processes may indicate that gaseous composition is
not solely responsible for the observed differences in the germination progression. For clarication, a further experimental lab-scale
approach was conducted, where during steeping (24 h wet steep)
the seeds had been either aerated, as outlined in Section 2.1., were
not aerated, or were purged with nitrogen. The subsequent germination conditions were identical in all cases and the seeds were
aerated. The activity of a-amylase and b-glucanase did not differ
signicantly between the three approaches and non-aeration
turned out to be slightly benecial (Figs. S1 and S2, supplemental
data). In contrast, coleoptile growth was fastest using the steep
aerated approach (Fig. S3, supplemental data), a fact that is in
accordance with the well known acceleration of germination related growth processes by aeration (e.g., Albers et al., 1983; French
& McRuer, 1990). Clearly, the growth processes and biosynthesis of
germination related enzymes were not progressing in parallel,
which indicates that the regulation of these two processes may
be different and is not impacted by exogenous factors in the same
way.
The question arises, which factors, apart from gaseous composition, may directly or indirectly affect the various germination processes. As purging with air causes a vaporization-related
temperature decline, differing temperature in the differentially
aerated assays might also contribute to the observed differences
in germination progression. Such effects can easily be substantiated using lab scale approaches. However, in industrial steeping
systems, these differences in temperature do not become apparent
due to the massive production of heat by the metabolic activities of
the germinating seeds.
3.4. Industrial-scale germinations: impact of steep aeration on stress
metabolism
The plant stress metabolite GABA (Fig. 5) generally shows a
completely different distribution pattern to the enzyme activities.
As already described in the context of Fig. 1D (lab germination;
Section 3.1.), the bulk of the barley seeds accumulated relatively
low amounts of GABA (2030 mg/100 g DW; Fig. 5AD), but some
outliers had extraordinary high GABA contents (e.g., >120 mg/
100 g DW; Fig. 5B and C). Thus, some of the barley grains seemed
to suffer from severe stress, whereas others were not affected.
However, the particular reasons for these tremendous GABA accumulations are not clear (e.g., injury or partial infection; see also
Section 3.1.). Yet, when the estimated outliers, i.e., those grains
containing more than twice the mean GABA content, were excluded from the calculation, the mean values for all the industrially produced malts were quite similar: 20.0, 23.8 and 21.1 mg/
100 g DW, respectively. However, the corresponding mean value
of 21.7 mg/100 g DW was about six times higher than that of
the grains processed under laboratory conditions (3.5 mg/100 g
DW), which suggested a markedly higher general stress level in

industrially produced malts. Yet, as the supply of O2 strongly differed due to the aeration regime applied, anoxia-related GABA
accumulation (e.g., Chung, Jang, Cho, & Lim, 2009; Kleinwchter
et al., 2012) could be excluded as a possible cause. Moreover,
due to the relatively similar temperatures during steeping and
the equal water content of the steeped grains, temperature as
well as water effects should also be ineligible as stress inducers.
Thus, the causes of the signicant differences in the stress levels
between industrially steeped barley seeds and those generated
under laboratory conditions remain unclear. Apart from the factors already mentioned, the major difference between industrially
processed malts and those processed under laboratory conditions
are the mechanical strains due to the higher layer thickness in
industrial steeping and germination systems. However, it is not
known how such mechanical strains could induce these corresponding stress reactions. GABA accumulation is a complex issue,
and has been reported to be triggered by a number of quite different stress reactions (Inatomi & Slaughter, 1971; for review
see Bown & Shelp, 1997; Satya Narayan & Nair, 1990). Accordingly, further studies are required, in order to draw unambiguous
conclusions on the feasibility of using individual grain analyses of
the GABA contents as a marker for process-related biochemical
inhomogeneities.

4. Conclusion
In conclusion, our ndings clearly indicate that steep aeration is
not a basic requirement for obtaining appropriate germination rate
and sufcient levels of the relevant enzyme activities, and that it
rather should be omitted in order to prevent the establishment
of biochemical inhomogeneities during malting in industrial
facilities.
Moreover, it is intriguing that the differences in the steeping
procedure still are manifested at the end of the germination period, although the conditions for all samples throughout the following three days had been identical. This is in accordance with
the old saying of German maltsters Das Malz wird in der Weiche
gemacht, which means The malt is made whilst steeping.
This article represents the rst part of a two-part study dealing
with the impact of steep aeration on the homogeneity of barley
malt. As such, it involved the preparatory work for the establishment of individual grain analysis for the assessment of biochemical
homogeneity and comparative analysis of differently steeped barley malts. The second part of this two-part study, entitled Impact
of aeration differences on the manifestation of biochemical parameters, will concentrate on the causes of biochemical heterogeneities in differently steeped malts.
Acknowledgements
This research project was supported by the German Ministry of
Economics and Technology (via AiF) and the FEI (Forschungskreis
der Ernhrungsindustrie e.V., Bonn, Germany). Project AiF16299N. We thank Winrich von Bierbrauer zu Brennstein (Oettinger Brauerei, Braunschweig, Germany), Batrice Conde-Petit, Urs
Keller and Eliana Zamprogna (all Buhler Group, Uzwil, Switzerland)
for their engagement and support of this research project. Their
contributions and ideas are greatly acknowledged.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.foodchem.2013.
09.090.

Author's personal copy

M. Kleinwchter et al. / Food Chemistry 147 (2014) 2533

References
Albers, V. J., Drost, B. W., & Pesman, L. (1983). Aerated steeping systems. MBAA
Technical Quarterly, 20, 7479.
Bown, A. W., & Shelp, B. J. (1997). The metabolism and functions of c-aminobutyric
acid. Plant Physiology, 115, 15.
Bytof, G., Knopp, S. E., Schieberle, P., Teutsch, I., & Selmar, D. (2005). Inuence of
processing on the generation of -aminobutyric acid in green coffee beans.
European Food Research and Technology, 220, 245250.
Cantrell, I. C. (1987). Cambridge prize lecture: Industrial scale research into malting
and brewing problems. Journal of the Institute of Brewing, 93, 487495.
Chung, H. J., Jang, S. H., Cho, H. Y., & Lim, S. T. (2009). Effects of steeping
and anaerobic treatment on GABA (c-aminobutyric acid) content in
germinated waxy hull-less barley. Food Science and Technology, 42,
17121716.
de S, R. M., & Palmer, G. H. (2004). Assessment of enzymatic endosperm
modication of malting barley using individual grain analysis. Journal of the
Institute of Brewing, 110, 4350.
de S, R. M., & Palmer, G. H. (2006). Analysis of b-glucan in single grains of
barley and malt using NIR-spectroscopy. Journal of the Institute of Brewing,
112, 916.
Enari, T. M., Linnahalme, T., & Linko, M. (1961). Effects of air and carbon dioxide in
the steeping of barley. Journal of the Institute of Brewing, 67, 358361.
French, B. J., & McRuer, G. R. (1990). Malt quality as affected by various steep
aeration regimes. MBAA Technical Quarterly, 27, 1014.
Gibbons, G. C. (1983). The effects of oxygen, nitrogen, and carbon dioxide steeping
regimes on endosperm modication and embryo growth. Carlsberg Research
Communications, 48, 3556.

33

Hanson, A. D., & Jacobson, J. V. (1984). Control of lactate dehydrogenase, lactate

glycolysis and a-amylase by O2 decit in barley aleurone layers. Plant
Physiology, 75, 566572.
Inatomi, K., & Slaughter, J. C. (1971). The role of glutamate decarboxylase and caminobutyric acid in germinating barley. Journal of Experimental Botany, 72,
561571.
Kelly, L., & Briggs, D. E. (1992). Barley maturity and the effects of steep aeration on
malting. Journal of the Institute of Brewing, 98, 329334.
Kleinwchter, M., Meyer, A.-K., & Selmar, D. (2012). Malting revisited: Germination
of barley (Hordeum vulgare L.) is inhibited by both oxygen deciency and high
carbon dioxide concentrations. Food Chemistry, 132, 476481.
Kunze, W. (2010). Technology Brewing and Malting (fourth ed.). Berlin: Versuchsund Lehranstalt fr Brauerei (VLB).
Narziss, L. (1999). Band 1. Die Technologie der Malzbereitung, 7. neu bearbeitete
Auage. In K. Schuster, F. Weinfurtner, & L. Narziss (Eds.), Die Bierbrauerei in drei
Bnden. Stuttgart: Ferdinand Enke Verlag.
Palmer, G. H. (2000). Malt performance is more related to inhomogeneity of protein
and b-glucan breakdown than to standard malt analysis. Malt performance, 106,
189192.
Perata, P., Loreti, E., Guglielminetti, L., & Alpi, A. (1998). Carbohydrate metabolism
and anoxia tolerance in cereal grains. Acta Botanica Neerlandica, 47, 269283.
Satya Narayan, V., & Nair, P. M. (1990). Metabolism, enzymology and possible roles
of 4-aminobutyrate in higher plants. Phytochemistry, 29, 367375.
Sopanen, T., & Laurire, C. (1989). Release and activity of bound b-amylase in a
germinating barley grain. Plant Physiology, 89, 244249.
Wilhelmson, A., Laitila, A., Vilpola, A., Olkku, J., Kotaviita, E., Fagerstedt, K., et al.
(2006). Oxygen deciency in barley (Hordeum vulgare) grain during malting.
Journal of Agricultural and Food Chemistry, 54, 409416.