Beruflich Dokumente
Kultur Dokumente
Food Microbiology
journal homepage: www.elsevier.com/locate/fm
Department of Biochemical Science and Technology, National Chiayi University, Chiayi City 60004, Taiwan
Institute of Biochemistry and Molecular Biology, National Yang-Ming University, Taipei City 11221, Taiwan
School of Medical Laboratory and Biotechnology, Chung Shan Medical University, Taichung City 40201, Taiwan
d
San Ying Foods Co., Ltd., Chiayi County 62151, Taiwan
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 20 September 2011
Received in revised form
14 July 2012
Accepted 2 October 2012
Available online 9 October 2012
The diversity of bacteria associated with the fermentation of inyu, also known as black soy sauce, was
studied through the nested PCR-denaturing gradient gel electrophoresis (DGGE) of samples collected
from the fermentation stages of the inyu production process. The DGGE proles targeted the bacterial 16S
rDNA and revealed the presence of Citrobacter farmeri, Enterobacter cloacae, Enterobacter hormaechei,
Enterococcus faecium, Klebsiella pneumoniae, Pantoea agglomerans, Salmonella enterica, Serratia marcescens, Staphylococcus sciuri and Weissella confusa. The bacterial compositions of 4 fermented samples were
further elucidated using the plate count method. The bacteria isolated from the koji-making stage
exhibited the highest diversity; Brachybacterium rhamnosum, E. hormaechei, K. pneumoniae, Kurthia
gibsonii, Pantoea dispersa, Staphylococcus gallinarum, Staphylococcus kloosii and S. sciuri were identied.
Koji collected during the preincubation stage presented the largest cell counts, and E. hormaechei,
K. pneumoniae, E. cloacae and Enterobacter pulveris were identied. In brine samples aged for 7 and 31
days, the majority of the bacteria isolated belonged to 4 Bacillus species, but 4 Staphylococcus species and
Delftia tsuruhatensis were also detected. This study demonstrates the benets of using a combined
approach to obtain a more complete picture of microbial populations and provides useful information for
the control or development of bacterial ora during inyu fermentation.
2012 Elsevier Ltd. All rights reserved.
Keywords:
inyu
Soy sauce
Bacterial diversity
Nested PCR-DGGE
16S rDNA
1. Introduction
Soy sauce is a traditional fermented seasoning in Asia. The
common method used to produce soy sauce is soaking soybeans in
water, steaming them, and then mixing them with baked wheat
our. The mixture is then inoculated with Aspergillus oryzae or
Aspergillus sojae and incubated at 25e30 C for 2e3 days, in
a process known as koji making. Subsequently, the koji is mixed
with a high concentration of salty water to obtain brine with a nal
concentration of 22e23% salt. The mixture is then aged at room
temperature for 6e12 months (Huang and Teng, 2004; Su et al.,
2005; Ko et al., 2009). Generally, soy sauces are divided into
* Corresponding author. Tel.: 886 2 2826 7125; fax: 886 2 2826 4843.
E-mail address: tsaiyc@ym.edu.tw (Y.-C. Tsai).
0740-0020/$ e see front matter 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.fm.2012.10.001
mixed with salt, loaded in a vat with salty water and covered with
a layer of salt. Finally, the resultant mixture is aged outdoors
without stirring for 2e4 months with exposure to sunlight.
Because of the nonsterile nature of koji making, there are
various types of microorganisms living in the fermentation
medium that are responsible for the unique avors and tastes of the
soy sauce. A. oryzae or A. sojae, the molds that the koji is inoculated
with during koji fermentation, produces extracellular enzymes that
hydrolyze the proteins and carbohydrates of the ingredients that
are involved in the brine fermentation (Lioe et al., 2010). Halophilic
lactic acid bacteria (LAB), such as Tetragenococcus halophilus, grow
during the early stage of brine fermentation and produce lactic acid
that acidify the brine (Huang and Teng, 2004). Salt-tolerant yeasts,
such as Zygosaccharomyces rouxii, grow rapidly, are associated with
decreases in the brine pH, and are capable of alcoholic fermentation
and the hydrolysis of various amino acids into their respective
alcohols during the middle stage of brine fermentation (Huang and
Teng, 2004). Salt-resistant yeasts, such as Torulopsis versatilis,
produce odor compounds during the maturation stage of brine
fermentation (Huang and Teng, 2004). Bacillus subtilis and Bacillus
mesentericus also contribute to the production of odor compounds
and create a more complex avor during brine aging (Huang and
Teng, 2004). However, the competition of the Bacillus species
with the molds decreases the production of amylases and proteases, rendering the bacteria as undesirable contaminants of koji
(Takazane et al., 1998). Because there are increasing demands from
the consumers and producers of fermented foods for high-quality
products that are free from bacterial contaminants, it is necessary
to understand and control the changes in bacterial ora that occur
during the fermentation process.
The traditional approach for studying the diversity of a microbial community consists of isolating and enumerating microbial
groups by growing them on various selective culture media and
identifying the predominant populations using phenotypic and
molecular techniques. However, in addition to being laborious and
time intensive, these culture-dependent approaches will not
identify strains or species that cannot grow on regular nutrient
media. As a result, culture-independent approaches that apply
molecular techniques based on the direct detection of DNA or RNA
in microbial ecosystems are particularly attractive for studying
microbial population dynamics both reliably and rapidly. Among
these methods, denaturing gradient gel electrophoresis (DGGE)
analysis is one of the most widely applied molecular techniques
(Muyzer et al., 1993). However, the GC-clamped primers used in
such analyses present sensitivity limitations when targeting lowabundance samples in natural environments (Vanbroekhoven
et al., 2004). Recently, this drawback has been addressed to
assess the microbial diversity of fermented soybean foods by using
nested PCR combined with DGGE (Kim et al., 2009; Lee et al.,
2010).
The aim of this study was to investigate bacterial communities
that were either contributively or undesirably associated with the
process of inyu fermentation by using a combination of culturedependent and culture-independent methods. Nested PCR-DGGE
of the V6eV8 regions of the bacterial 16S rDNA was used to
monitor the diversity of the bacterial populations that are associated with the 9 stages of inyu fermentation. Both koji samples and 2
brine samples, which appeared to contain all bands of all brine
samples in DGGE proles, were subjected to further evaluation of
their bacterial diversities using the plate count method. The isolates
were grouped and identied on the basis of their genotypic characteristics, as determined by randomly amplied polymorphic DNA
(RAPD) and 16S rDNA sequencing. To the best of our knowledge,
this is the rst report to reveal the bacterial diversity associated
with inyu production.
253
Black soybean
Soaking
4 hours
Steaming
Mixing
Room temperature
6 days
7 days
Koji
Washing
Incubating
4550 C
95% humidity
8 hours
6 hours
Sample PI
Mixing
Covering
Covering
Aging
1 day
1 week
Outdoor
Sunlight
4 months
2 weeks
Sample 1D
Sample 1W
Sample 2W
1 month
Sample 1M
2 months
Sample 2M
3 months
Sample 3M
Inyu brine
Sample 4M
Fig. 1. The production diagram of inyu used in this study with the sampling points
indicated.
254
Table 1
Primers used in this study.
Positiona
Reference
Genomic DNA
Genomic DNA
Genomic DNA
Bacterial 16S rDNA
Bacterial 16S rDNA
Bacterial 16S rDNA
Bacterial 16S rDNA
Bacterial 16S rDNA
Bacterial 16S rDNA
Bacterial 16S rDNA
Bacterial 16S rDNA
8e27
515e530
933e954
1541e1522
531e517
804e787
968e985
1514e1492
968e985
1401e1385
Primer name
Primer sequence
Target
A
8
OPL-05
8F
520F
930F
15R
520R
800R
F-968
Bac-1492R-II
CCGCAGCCAA
ACGCGCCCT
ACGCAGGCAC
AGAGTTTGATCMTGGCTCAG
CAGGAGTGCCAGCAGCCGCGG
GCACAAGCGGTGGAGCATGTGG
AAGGAGGTGATCCAACCGCA
ACCGCGGCTGCTGGC
CAGGACTACCAGGGTATCTAAT
AACGCGAAGAACCTTAC
ACGGYTACCTTGTTACGCGACTT
CGCCCGGGGCGCGCCCCGGGCG
GGGCGGGGGCACGGGGGGAACG
CGAAGAACCTTAC
CGGTGTGTACAAGACCC
F-968-GC
R-1401
a
8F
F-968
520F
930F
16S rDNA
520R
255
800R
R-1401
100 bp
Bac-1492R-II
15R
Fig. 2. The position of primers used in this study with regard to the bacterial 16S rDNA.
3. Results
3.1. Bacterial enumeration, pH and salt concentration
measurements
The counts for viable bacterial colonies observed on NB medium
during the fermentation of inyu are shown in Table 2. Bacteria were
quantied at a concentration of 6.2 107 cfu/g in the KM stage.
After washing and preincubation, the koji exhibited the largest
bacterial population (2.2 108 cfu/g) of all of the collected samples.
After placing the koji in a vat with salty water and salt, the
concentrations of the bacteria were slightly reduced to 3.7 107
and 1.8 107 cfu/mL in brines that had been exposed to sunlight for
1 day and 1 week, respectively. The bacterial concentrations of the
brines continued to decrease, reaching a range of 1.1 106 to
1.2 105 cfu/mL after aging from 2 weeks to 4 months.
The pH of brine was determined to be 5.7 after aging in sunlight
for 1 day, and then, the pH gradually decreased to 5.4 after aging for
1 month (Table 2). The pH subsequently decreased further to 4.9
after aging for 2 months and then remained stable.
The salt concentration of brine aged for 1 day was 30%; the salt
concentration then increased from 35% to 39% after aging for 1
week to 3 months (Table 2). At the end of the aging period, the salt
concentration had reached 42%.
Total DNA was directly extracted from the koji and brine
samples, and the V6eV8 regions of the bacterial 16S rDNA were
amplied to yield nested PCR products for the characterization of
bacterial diversity using DGGE on a 30e65% denaturing gradient
gel (Fig. 3a). Based on the premise that bands that migrated to the
same position were identical, bands AeX (highlighted in Fig. 3a)
were selected, excised, re-amplied and subjected to sequencing.
The sequencing results suggested that all of the bands were
mixtures of amplicons. In fact, the 16S rDNA fragments might have
identical melting properties even though they had different
sequences (Ercolini, 2004), and therefore, they cannot be separated
by this DGGE condition. Re-amplied products of these bands were
thus applied to further separation by a second round of DGGE
analysis with 3 narrow ranges of denaturing gradient gels,
including 30e50% (bands AeG; Fig. 3b), 40e60% (bands HeO;
Fig. 3d) and 50e70% (bands PeX; Fig. 3c). The initial brine
samples of 1W and 2M were included on the 30e50% and 50e70%
gels (Fig. 3b and c), respectively, to conrm the origin of these reamplied products. In addition to analysis on the 50e70% gel, the
re-amplied products of bands P1 and Q1 were also included in the
40e60% gel (Fig. 3d) to exclude the possibility that the re-amplied
products were artifacts. The bands highlighted in Fig. 3bed were
excised and re-amplied prior to sequencing. The identication
results are presented in Table 3.
Bands E, G, H, I and O shown in Fig. 3a had corresponding bands
(with similar migration positions) in all samples. In the subsequent
DGGE analysis (Fig. 3b and d), the bands developed from E1eE3,
G1e2, H1e2, I1e3 and O1e3 yielded sequences with close
homologies to Staphylococcus sciuri (band 1), Klebsiella pneumoniae
(bands 2, 4e10, 12e13, 19e21 and 23e26), Enterobacter cloacae
(bands 11, 15, 22 and 31), Salmonella enterica (band 14), Enterobacter
hormaechei (bands 17e18 and 32e33), Citrobacter farmeri (band
16), Enterococcus faecium (band 35) and Weissella confusa (band 34).
With the exception of bands 34 and 35 that were only present in
brine aged for 1 week (Fig. 3D), the other bands were present in
both the koji and brine samples (Fig. 3b and d). K. pneumoniae was
also identied from band 36 but was present only in brine aged for
2 months and not in brine aged for 1 week or 1 month (Fig. 3c).
Band C in Fig. 3a only exhibited corresponding bands in koji
samples from the KM and PI stages. The secondary DGGE ngerprint, as shown in Fig. 3b, also revealed that band 3 only had clear
corresponding bands in these koji samples. However, its sequence
corresponded to Pantoea agglomerans and was the same as that of
band 37 developed from Q2, for which corresponding bands were
found in all of the brine samples shown in Fig. 3c. These results
suggested that P. agglomerans was present in both the koji and
brine samples.
Bands with positions similar to bands JeN shown in Fig. 3a were
present in all of the brine samples but were not found in the koji
samples. In the subsequent DGGE analysis (Fig. 3d), bands 27 and
28e30, with sequences homologous to sequences of Serratia
Table 2
The bacterial counts, pH values and salt concentrations in the fermented samples of inyu.
Sample
Koji preparationa
KM
PI
1D
1W
2W
1M
2M
3M
4M
Cell counts
pH value
Salt concentration (%)
7.79
e
e
8.34
e
e
7.57
5.7
30
7.25
5.6
35
6.06
5.4
38
5.98
5.4
37
5.6
4.9
39
5.56
5.1
39
5.06
5.1
42
e: Not determined.
a
Bacteria counts are represented as log10 cfu/g.
b
Bacteria counts are represented as log10 cfu/ml.
256
KM PI
1D 1W
2W 1M 2M
3M
4M
1W
1M
P1 Q1 X
KM
PI
C1 E1 C2 E2 G1 A
KM
PI
I1 O1 H1 I2 O2 H2 I3 J
1W
D E3 F
G2
1W
1W
K L M N O3 P1 Q1
2M
P2 Q2 R
2M
Fig. 3. DGGE proles of the V6eV8 regions of bacterial 16S rDNA. (a) A 30e65% denaturing gradient gel was used to analyze nested PCR products amplied from DNA directly
extracted from inyu fermentation samples. Lanes KM and PI: koji samples from the koji-making and preincubation stages; lanes 1D, 1W, 1M, 2M, 3M and 4M: brine samples from
outdoor aging for 1 day, 1 week, 2 weeks, 1 month, 2 months, 3 months and 4 months. Bands with different migration distances (AeX) were selected from different samples. The
different numbers behind the same letter indicate bands obtained from different samples that migrated the same distance. The bands AeG, PeX and HeO were further excised, reamplied and subjected to a second round of DGGE analysis using (b) 30e50%, (c) 50e70% or (d) 40e60% denaturing gradient gels. Bands indicated by numbers alone were excised,
re-amplied and subjected to sequencing.
had been aged for 1 week and 1 month and appeared to contain all
of the bands in all of the brine samples, together with both koji
samples, were further subjected to an evaluation of their bacterial
diversities using the plate count method. A total of 212 isolates
were analyzed by RAPD-PCR using 3 sets of random primers (data
not shown).
RAPD analysis has been widely used for studying the genetic
diversity of bacteria (Atienzar and Jha, 2006). Isolates that showed
identical RAPD patterns were regarded as identical strains and,
thus, as members of the same group. To verify the RAPD proles,
a total of 120 groups were obtained from koji samples in the KM (21
groups) and PI (5 groups) stages and from brine samples that had
been aged for 1 week (46, 21 and 7 groups selected from 5%, 10%
Banda
Identityb
364/364 (100%)
16
28e30
11
15
22
31
17e18, 32e33
2, 4e10, 12e13,
23e26, 36
19
20e21
3
37
14
27
363/364
364/364
363/364
361/364
362/364
330/330
364/364
364/364
(99%)
(100%)
(99%)
(99%)
(99%)
(100%)
(100%)
(100%)
364/364
364/364
364/364
363/364
363/364
363/364
(100%)
(100%)
(100%)
(99%)
(99%)
(99%)
34
35
364/364 (100%)
364/364 (100%)
and 15% NaCl plates, respectively) and 1 month (12 and 8 groups
selected from 5% to 10% NaCl plates, respectively). One representative isolate from each group was chosen for full-length
sequencing of its 16S rDNA. A total of 8 genera and 17 species of
bacteria, including the Bacillus (4 species), Brachybacterium (1
species), Delftia (1 species), Enterobacter (3 species), Klebsiella (1
species), Kurthia (1 species), Pantoea (1 species) and Staphylococcus
(5 species) genera, were identied (Table 4). All of the isolates
yielded similarity scores of over 99% to their corresponding species,
with the exception of the isolate obtained from the koji during the
PI stage that had a 98.6% similarity score for Enterobacter pulveris. A
total of 32 isolates that were identied as different species that had
been selected from the various samples and plates with differing
NaCl concentrations were further subjected to molecular phylogenetic analysis on the basis of their 16S rDNA (Fig. 4).
The koji from the KM stage exhibited the highest diversity of
bacteria, with a total of 6 genera and 8 species. Among these,
Brachybacterium rhamnosum made up the bulk of the community
(34%) with a concentration of 2.0 107 cfu/g; E. hormaechei (28%)
was found at a concentration of 1.7 107 cfu/g; Staphylococcus spp.
(24%) presented concentrations of 7.0 106, 4.5 106 and
2.5 106 cfu/g for Staphylococcus gallinarum, Staphylococcus kloosii
and S. sciuri, respectively; and K. pneumoniae, Pantoea dispersa and
Kurthia gibsonii made up 14% of the community, with concentrations of 5.0 106, 2.5 106 and 5.0 105 cfu/g, respectively.
However, most of these species were absent from the subsequently
collected samples; only the levels of E. hormaechei (2.5 107 cfu/g)
257
and K. pneumoniae (1.6 108 cfu/g) increased in the koji from the PI
stage by 1.5 and 32.0 fold, respectively. Although the koji from the
PI stage yielded the largest population among all of the collected
samples, with the exception of E. hormaechei and K. pneumoniae,
only two additional species, E. cloacae (3.0 107 cfu/g) and E.
pulveris (5.0 106 cfu/g), were identied.
In brine samples that had been aged for either 1 week or 1
month, 4 species of Bacillus were identied as being dominant in
the isolates selected from the agar plates containing 5% NaCl.
Bacillus pumilus (2.5 102 and 3.0 104 cfu/mL in brines aged for 1
week and 1 month, respectively) was only isolated from 5% NaCl
agars, whereas B. subtilis and Bacillus licheniformis were detected in
isolates selected from both 5% NaCl agar (5 104 and 8.6 104 cfu/
mL of B. subtilis in brines aged for 1 week and 1 month, respectively; 5 102 and 6.9 103 cfu/mL of B. licheniformis in brines aged
for 1 week and 1 month, respectively) and 10% NaCl agar (2.1 103
and 1.1 103 cfu/mL of B. subtilis in brines aged for 1 week and 1
month, respectively; 1.6 103 and 3.9 103 cfu/mL of
B. licheniformis in brines aged for 1 week and 1 month, respectively). Bacillus amyloliquefaciens increased from 2.7 104 to
1.5 105 cfu/mL while aging from 1 week to 1 month in isolates
selected from 5% NaCl agar; this species was also detected in
isolates selected from 10% NaCl agar (5.7 102 cfu/mL) in brine
aged for 1 month. In brine aged for 1 week, Delftia tsuruhatensis and
4 species of Staphylococcus (including Staphylococcus cohnii,
Staphylococcus condimenti, S. gallinarum and S. kloosii), with
concentrations ranging from 1.1 102 to 1.2 103 cfu/mL, were also
identied in isolates selected from 5 to 15% NaCl agar plates
(Table 4).
4. Discussion
Recently, the microbial communities involved in the soy sauce
manufacturing process were analyzed using the PCR-DGGE method
(Tanaka et al., 2012). However, no such evaluations that focused on
inyu fermentation have been reported. The production of inyu,
a type of fermented soy sauce made in Taiwan, employs a longer
fermentation period for koji making and a higher temperature
for brine aging than traditional soy sauces; these characteristics led
to the hypothesis that a much higher level of bacterial diversity
would be associated with inyu production. In addition, an articial
method and nonsterile operations are used for koji making,
washing, preincubation, loading in vats and sunlight exposure
aging during inyu production. These manufacturing processes that
are characteristic of inyu production led us to elucidate the associated bacterial communities, providing valuable information
regarding the functions of the bacteria during fermentation. Here,
we revealed the bacterial compositions during the different
stages of the inyu fermentation using both culture-independent
and culture-dependent techniques. Four species (E. cloacae,
E. hormaechei, K. pneumoniae and S. sciuri, belonging to 3 genera)
were identied using both techniques, 6 additional species
(C. farmeri, P. agglomerans, S. enterica, S. marcescens, E. faecium and
W. confusa, belonging to 6 genera) were identied using nested PCRDGGE (Table 3), and 13 additional species (B. amyloliquefaciens,
B. licheniformis, B. pumilus, B. subtilis, B. rhamnosum, D. tsuruhatensis,
E. pulveris, K. gibsonii, P. dispersa, S. cohnii, S. condimenti, S. gallinarum
and S. kloosii, belonging to 7 genera) were identied using a traditional plating technique (Table 4).
Although DGGE provided a broad and rapid overview of the
microbial population dynamics, the DGGE proles of the bacterial
V6eV8 16S rDNA amplied directly from fermented samples were
a mixture of amplicons (Fig. 3aed). The excised bands were reamplied and separated by DGGE using 3 different ranges of
denaturing gradients (Fig. 3bed). These results suggested that
258
Table 4
Quantication of bacterial species in the fermented samples of inyu.
Species
Samples
KMa
PIa
1Wb
0%
5%
10%
15%
5%
10%
15%
e
7.22
e
6.70
6.40
7.48
7.40
6.70
8.20
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
7.29
5.70
e
e
6.85
6.65
6.40
e
e
e
e
e
e
3.07
2.33
e
1.70
2.32
e
2.06
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
2.70
e
e
e
e
e
e
e
e
4.43
2.70
2.40
4.70
e
3.19
e
3.33
e
e
e
e
5.17
3.84
4.48
4.93
2.75
3.59
e
3.05
e
e
e
e
0%
Enterobacteria
Enterobacter cloacae
Enterobacter hormaechei
Enterobacter pulveris
Klebsiella pneumoniae
Pantoea dispersa
Brachybacterium
Brachybacterium rhamnosum
Kurthia
Kurthia gibsonii
Staphylococcus
Staphylococcus cohnii
Staphylococcus condimenti
Staphylococcus gallinarum
Staphylococcus kloosii
Staphylococcus sciuri
Delftia
Delftia tsuruhatensis
Bacillus
Bacillus amyloliquefaciens
Bacillus licheniformis
Bacillus pumilus
Bacillus subtilis
1Mb
e: Not found.
a
Bacteria counts are represented as log10 cfu/g.
b
Bacteria counts are represented as log10 cfu/mL.
c
Salt concentration in the agar plates.
259
1W10C9
Delftia tsuruhatensis IFO 16741T (NR_024786)
PI2
1000
1000
1000
KM9B
1000
KM6
PI1
960
991
5%
1000
999
977
1000
1000
1000
1000
1000
1000
5%
KM1
Kurthia gibsonii NBRC15534T (AB271738)
1000 KM4A
Staphylococcus sciuri DSM20345T (NR_025520)
Staphylococcus condimenti DSM11674T (NR_029345)
1000
1W5D2
Staphylococcus carnosus GTC1232T (AB233329)
1W10A13
1000
KM5A
Staphylococcus kloosii ATCC43959T (AB009940)
1W10A11
991
1W15A2
Staphylococcus cohnii ATCC49330T (AB009936)
1W10A3
Staphylococcus arlettae ATCC43957T (NR_024664)
1W10A10
Staphylococcus
gallinarum ATCC35539T (D83366)
966
KM7
Staphylococcus succinus ATCC700337T (AF004220)
Bacillus aerophilus 28K (AJ831844)
1M5C7
1W5D1
Bacillus pumilus DSMZ27T (AY456263)
1W5B14
1M5B31
1W10C10
1000
1M10A10
Bacillus licheniformis BCRC11702T (EF433410)
1000
Bacillus aerius 24K (AJ831843)
Bacillus vallismortis DSM11031T (AB021198)
1M5B1
1W5B26
Bacillus nematotocita B-16T (AY820954)
Bacillus amyloliquefaciens NBRC15535T (AB325583)
965
1M10A32
Bacillus mojavensis NBRC15718T (AB021191)
1W5B29
1W10C4
1M5A1
1M10A33
Bacillus subtilis IAM12118T (AB042061)
Fig. 4. Phylogenetic tree based on the 16S rDNA sequence analysis showing the phylogenetic placement of species isolated from kojis at the koji-making (KM) and preincubation
(PI) stages and from brines aged for 1 week (1W) and 1 month (1M) during the fermentation of inyu. Panels a and b illustrate the results for species of gram-negative and grampositive bacteria, respectively. The tree was constructed by the neighbor-joining method (Escherichia coli was used as the outgroup). Bootstrap values (expressed as the percentages
of 1000 replications) greater than 90% are shown at branch points. The scale bar represents 5% sequence divergence.
260
for 1 week using the plate count method (Table 4). To our knowledge, this is the rst report describing the presence of these species
in fermented food. B. rhamnosum, belonging to the family Dermabacteraceae, class Actinobacteria, and the closely related species
(based on 16S rDNA) Brachybacterium squillarum have been isolated
from salt-fermented seafood in Korea (Park et al., 2011).
D. tsuruhatensis has been isolated from activated sludge collected
from a domestic wastewater treatment plant in Japan and was able
to degrade various aromatic hydrocarbon compounds (Shigematsu
et al., 2003). This species has been described as a plant growthpromoting bacterium and has been proposed as a potential
biocontrol agent against plant pathogens (Han et al., 2005; Validov
et al., 2007). Recently, this species has also been described as the
causative agent of a human catheter-related infection (Preiswerk
et al., 2011).
In conclusion, nested PCR-DGGE analysis combined with a plate
count method was used for the rst time to assess the bacterial
communities involved in inyu production. The combination of the 2
techniques overcame the limitations of each individual method and
provided a more global insight into the bacterial diversity that is
associated with inyu fermentation. This study might be a great
benet for the further modication of the fermentation process to
produce inyu with more attractive avors without safety concerns.
In brief, most of the bacterial species found in the KM koji,
including Brachybacterium, Kurthia and Staphylococcus species,
were absent after koji washing. The family Enterobacteriaceae, the
sole survivors in the PI koji, also disappeared in the subsequent
brine sample. In addition to the unusual high-salt and hightemperature environment used for the fermentation of inyu
brine, Staphylococcus and Bacillus species might also contribute to
the inhibition of pathogenic bacteria as well as the possible role of
S. gallinarum and S. kloosii in soy sauce fermentation (Tanaka et al.,
2012) and thus provide a natural safety control. Furthermore,
bacteria from both genera, especially the latter, which was also
found in 1M brine, might produce useful catalysts that give inyu
special avors during brine fermentation. Their possible thermotolerant and halotolerant properties might be valuable for both
other food fermentations and industrial applications. On the other
hand, the safety issues related to the Staphylococcus species could
be determined by analyzing the hemolytic activity of the bacteria as
well as inyu brine. Certainly, the desirable or undesirable effects of
bacteria during brine fermentation should be claried in further
studies. Once these bacteria are classied to be undesirable, they
might be prevented by introducing bacteriocin-producing LAB to
lower the pH and produce antimicrobial proteins (Chang and
Chang, 2011; Sarika et al., 2012) in the early brine fermentation
steps.
Acknowledgments
We thank Prof. H.-M. Ho and Prof. H.-C. Mei (National Taipei
University of Education, Taipei) for their technical assistance with
the nested PCR-DGGE analysis.
References
Alcantara-Hernandez, R.J., Rodriguez-Alvarez, J.A., Valenzuela-Encinas, C., Gutierrez-Miceli, F.A., Castanon-Gonzalez, H., Marsch, R., Ayora-Talavera, T.,
Dendooven, L., 2010. The bacterial community in taberna a traditional
beverage of southern Mexico. Letters in Applied Microbiology 51, 558e563.
Atienzar, F.A., Jha, A.N., 2006. The random amplied polymorphic DNA (RAPD) assay
and related techniques applied to genotoxicity and carcinogenesis studies:
a critical review. Mutation Research 613, 76e102.
Bevilacqua, A., Cannarsi, M., Gallo, M., Sinigaglia, M., Corbo, M.R., 2010. Characterization and implications of Enterobacter cloacae strains, isolated from Italian
table olives Bella di Cerignola. Journal of Food Science 75, M53eM60.
261
Nubel, U., Engelen, B., Felske, A., Snaidr, J., Wieshuber, A., Amann, R.I., Ludwig, W.,
Backhaus, H., 1996. Sequence heterogeneities of genes encoding 16S rRNAs in
Paenibacillus polymyxa detected by temperature gradient gel electrophoresis.
Journal of Bacteriology 178, 5636e5643.
Onda, T., Yanagida, F., Tsuji, M., Shinohara, T., Yokotsuka, K., 2003. Time series
analysis of aerobic bacterial ora during Miso fermentation. Letters in Applied
Microbiology 37, 162e168.
Ouoba, L.I.I., Diawara, B., Annan, N.T., Poll, L., Jakobsen, M., 2005. Volatile
compounds of Soumbala, a fermented African locust bean (Parkia biglobosa)
food condiment. Journal of Applied Microbiology 99, 1413e1421.
Ouoba, L.I.I., Diawara, B., Jespersen, L., Jakobsen, M., 2007. Antimicrobial activity of
Bacillus subtilis and Bacillus pumilus during the fermentation of African locust
bean (Parkia biglobosa) for Soumbala production. Journal of Applied Microbiology 102, 963e970.
Park, S.K., Kim, M.S., Jung, M.J., Nam, Y.D., Park, E.J., Roh, S.W., Bae, J.W., 2011. Brachybacterium squillarum sp. nov., isolated from salt-fermented seafood. International Journal of Systematic and Evolutionary Microbiology 61, 1118e1122.
Preiswerk, B., Ullrich, S., Speich, R., Bloemberg, G.V., Hombach, M., 2011. Human
infection with Delftia tsuruhatensis isolated from a central venous catheter.
Journal of Medical Microbiology 60, 246e248.
Sarika, A.R., Lipton, A.P., Aishwarya, M.S., Dhivya, R.S., 2012. Isolation of a bacteriocin-producing Lactococcus lactis and application of its bacteriocin to
manage spoilage bacteria in high-value marine sh under different storage
temperatures. Applied Biochemistry and Biotechnology 167, 1280e1289.
Shigematsu, T., Yumihara, K., Ueda, Y., Numaguchi, M., Morimura, S., Kida, K., 2003.
Delftia tsuruhatensis sp. nov., a terephthalate-assimilating bacterium isolated
from activated sludge. International Journal of Systematic and Evolutionary
Microbiology 53, 1479e1483.
Steinkraus, K.H., 1991. African alkaline fermented foods and their relation to similar
foods in other parts of the world. In: Westby, A., Reilly, P. (Eds.), Proceedings of
a Regional Workshop on Traditional African FoodsdQuality and Nutrition.
International Foundation for Science, pp. 87e92.
Su, N.W., Wang, M.L., Kwok, K.F., Lee, M.H., 2005. Effects of temperature and sodium
chloride concentration on the activities of proteases and amylases in soy sauce
koji. Journal of Agricultural and Food Chemistry 53, 1521e1525.
Takazane, S., Endo, T., Shida, O., Tagami, H., Takagi, H., Kadowaki, K., 1998. Inuences
of Bacillus species in shoyu koji on shoyu brewing (part 1) isolation and identication of antifungal antibiotics producing bacteria. Shoyu Kenkyujyo Kenkyu
Houkoku 24, 77e82.
Tanaka, Y., Watanabe, J., Mogi, Y., 2012. Monitoring of the microbial communities
involved in the soy sauce manufacturing process by PCR-denaturing gradient
gel electrophoresis. Food Microbiology 31, 100e106.
Teng, D.F., Lin, C.S., Hsieh, P.C., 2004. Fermented whole soybeans and soybean paste.
In: Hui, Y.H. (Ed.), Handbook of Food and Beverage Fermentation Technology.
CRC Press.
Thanh, V.N., Mai le, T., Tuan, D.A., 2008. Microbial diversity of traditional Vietnamese alcohol fermentation starters (banh men) as determined by PCRmediated DGGE. International Journal of Food Microbiology 128, 268e273.
Toyokawa, Y., Takahara, H., Reungsang, A., Fukuta, M., Hachimine, Y., Tachibana, S.,
Yasuda, M., 2010. Purication and characterization of a halotolerant serine
proteinase from thermotolerant Bacillus licheniformis RKK-04 isolated from Thai
sh sauce. Applied Microbiology and Biotechnology 86, 1867e1875.
Validov, S., Kamilova, F., Qi, S., Stephan, D., Wang, J.J., Makarova, N.,
Lugtenberg, B., 2007. Selection of bacteria able to control Fusarium oxysporum
f. sp. radicis-lycopersici in stonewool substrate. Journal of Applied Microbiology 102, 461e471.
Vanbroekhoven, K., Ryngaert, A., Wattiau, P., Mot, R., Springael, D., 2004. Acinetobacter diversity in environmental samples assessed by 16S rRNA gene PCRDGGE ngerprinting. FEMS Microbiology Ecology 50, 37e50.
Weisburg, W.G., Barns, S.M., Pelletier, D.A., Lane, D.J., 1991. 16S ribosomal DNA
amplication for phylogenetic study. Journal of Bacteriology 173, 697e703.
Wesley, E.K., Carol, G.G., Jennifer, S.O., Linda, V.P., Mary, L.M., Barbara, L.Z.,
Eugene, M., Melvin, P.W., Stanley, M., 1998. Staphylococcus hominis subsp.
Novobiosepticus subsp. nov., a novel trehalose- and N-acetyl-D-glucosaminenegative, novobiocin and multiple-antibiotic-resistant subspecies isolated from
human blood cultures. International Journal of Systematic Bacteriology 48,
799e812.
Wesley, E.K., Schleifer, K.H., Smith, R.F., 1976. Characterization of Staphylococcus
sciuri sp. nov. and its subspecies. International Journal of Systematic Bacteriology 26, 22e37.
Yoo, S.K., Cho, W.H., Kang, S.M., Lee, S.H., 1999. Isolation and identication of
microorganisms in Korean traditional soybean paste and soybean sauce. Korean
Journal of Applied Microbiology and Biotechnology 27, 113e117.
Zaman, M.Z., Abu Bakar, F., Jinap, S., Bakar, J., 2011. Novel starter cultures to inhibit
biogenic amines accumulation during sh sauce fermentation. International
Journal of Food Microbiology 145, 84e91.
Zell, C., Resch, M., Rosenstein, R., Albrecht, T., Hertel, C., Gotz, F., 2008. Characterization of toxin production of coagulase-negative staphylococci isolated
from food and starter cultures. International Journal of Food Microbiology
127, 246e251.