Changes in Microbial Community Composition during Production of Takanazuke

Abstract

Changes in microbial community composition of takana produced by a conventional fermenatation method with 6% (w/w) NaCl were analyzed. Harvested takana leaves were pickled with 6% (w/w) NaCl in Aso city, transported to our laboratory in Kumamoto city, and then incubated at the average monthly temperatures of the Aso area. The lactate concentration and D- to L-lactate ratio increased during the 180-day fermentation. Partial sequencing of the 16S rRNA gene was used to determine the microbial composition of takanazuke on days 0, 19, 40, 90, and 180. Only Pseudomonas was detected on day 0, whereas Lactobacillus curvatus was clearly the dominant species on day 19. After 40 days, L. curvatus (56.9%), L. (para)plantarum (39.2%), and L. sakei (3.9%) were detected, whereas on day 90, the bacterial population consisted of L. (para)plantarum (50.9%), L. sakei (45.6%), L. curvatus (1.7%), and members of the genus Weissella (1.7%). The final microbial composition on day 180 of the fermentation was L. (para) plantarum (59.6%), genus Weissella (32.7%), L. curvatus (3.8%), L. alimentarius (1.9%), and genus Clostridium (1.9%).

Introduction

Takana (Brassica juncea Coss.) is cultivated in the Aso area of Kumamoto prefecture. Takana is sown in the fall and is harvested in March and April for the production of takanazuke, a fermented food that is traditionally used to preserve takana and which is a regional specialty product of Kumamoto. For the production of takanazuke, salt (approximately 6% [w/w] NaCl) is added to harvested takana to prevent microbial contamination during the fermentation period. When preparing takanazuke to be preserved for the long-term and consumed after the fall season, 8% (w/w) NaCl is used for the fermentation.

Pickled vegetables are naturally fermented by various species of lactic acid-producing bacteria (LAB) (Hutkins, 2006), and there have been many reports of using LAB to ferment vegetables worldwide (Di Cagno et al., 2013). As the conditions for pickle production are neither constant nor aseptic, it is difficult to regulate product quality. The traditional method of production is superior for pickle storage and food safety, but often alters the taste of the fermented product. Starter bacterial cultures have been used for production to maintain the quality of the final products (Holzapfel, 2002). Utilization of starter LAB for takanazuke production also stabilizes product quality and reduces the requirement for added salt. We previously isolated LAB from takanazuke in 2010 for use as starter strains. The isolated LABs were identified by partial 16S rRNA gene sequencing, and their characteristics were analyzed (Sakai et al., 2013). Two isolated strains, B17–4 (Lactobacillus (para)plantarum) and C120–3 (Pediococcus parvulus), with potential as suitable starter strains were used for takanazuke production (Sakai, M., Nagano, M., Ohta, H., Kida, K., and Morimura, S., unpublished data). However, the microbial community composition found in takanazuke produced by the traditional method has yet to be characterized.

The isolation and microbial community analysis of LAB from pickled vegetables have been widely reported (Tamang et al., 2005; Plengvidhya et al., 2007; Chao et al., 2009; Tanganurat et al., 2009; Paramithiotis et al., 2010). However, as certain species of LAB are not readily isolated by common laboratory methods (Ampe et al., 1999), both culture-dependent and -independent methods are currently used for the analysis of microbial communities in fermented food (Giraffa, 2004, Temmerman et al., 2004). Recently, the results obtained by both methods have been extensively analyzed and compared (Endo et al., 2008; Nguyen et al., 2013; Wouters et al., 2013a; Wouters et al., 2013b). In the present study, we analyzed changes in the microbial community composition of takanazuke produced in our laboratory based on 16S rRNA gene sequence analysis. We also compared the microbial community composition to that isolated in a previous study.

Materials and Methods

Production of takanazuke Takana was harvested in April 2011 from Ichihara Farm in Aso city, Kumamoto. Specimens were mixed with 6% (w/w) NaCl (day 0) and transported to our laboratory in Kumamoto city. The samples were kept at room temperature for 1 day, and were then divided into 110 g portions in sealed containers. The specimens were further incubated at the average monthly temperatures of the previous year (2010) for the Aso region, as follows: days 1 – 28 (April) at 10.8°C; days 29 – 59 (May) at 15.9°C; days 60 – 89 (June) at 20.0°C; days 90 – 120 (July) at 23.6°C; days 121 – 151 (August) at 25.0°C; and days 152 – 181 (September) at 21.8°C. Approximately 4% (w/w) of water was removed from each 110 g portion of takanazuke on days 2, 4, 6, 17, and 31.

Determination of microbial community of takanazuke Takanazuke on days 0, 19, 40, 90, and 180 was placed in a triple layer of gauze and squeezed to release liquid. The obtained solution was centrifuged at 8000 × g at 4°C for 15 min, and the precipitate was suspended in sterilized water. The suspension was centrifuged again under the same conditions, and the precipitate was suspended in 200 μL sterilized water. The bead beating method was then used to extract DNA, which was used as a template for the polymerase chain reaction (PCR) targeting partial 16S rRNA gene sequences using the primer pair 27F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 518R (5′-GTATTACCGCGGCTGCTGG-3′). The PCR was performed with AmpliTaq Gold DNA polymerase (Applied Biosystems, Carlsbad, CA) and the following conditions: preheating at 95°C for 5 min, followed by 25 cycles of denaturation at 95°C for 1 min, annealing at 50°C for 1 min, and extension at 72°C for 2 min. PCR products were purified using an Ultra Clean PCR Clean-up Kit (MO BIO, Carlsbad, CA) according to the manufacturer's protocol. Purified PCR products were ligated with pT7Blue vector (Novagen, Darmstadt, Germany) using a Ligation Mix Kit (Takara, Kyoto, Japan) according to the manufacturer's instructions. The ligation mixture (10 μL) was then transformed into 100 μL chemically competent Escherichia coli DH5α cells (Takara) by a conventional method. White colonies were selected on LB-ampicillin plates, and plasmids were extracted using the Wizard SV Minipreps DNA Purification System (Promega, Madison, WI). Extracted plasmids were digested with EcoRI (Takara) and PstI (Takara) to determine the size of the inserted DNA. Sequence analysis of the inserted DNA was completed by Takara Co., Ltd. Data from the analysis were compared to the NCBI database by a BLAST search to identify obtained clones. Chimeric data were removed following comparison between known sequences and the obtained data.

Effect of initial pH on the growth of three LAB strains Three previously isolated LAB strains, Lactobacillus curvatus B17–2, L. sakei A165–1, and L. (para)plantarum B17–4 (Sakai et al., 2013) were grown from glycerol stocks in MRS medium (10 g/L peptone, 10 g/L meat extract, 5 g/L yeast extract, 20 g/L glucose, 80 1 g/L Tween, 2 g/L K₂HPO₄, 5 g/L sodium acetate, 2 g/L diammonium hydrogen citrate, 0.2 g/L MgSO₄·7H₂O, and 0.05 g/L MnSO₄·nH₂O) supplemented with 2% NaCl. The pH of the medium was adjusted to 6.0 – 6.5, and agar (15 g/L) was added for the preparation of solid medium. The three strains were cultivated on MRS plates containing 2% NaCl at 30°C for 48 h. A loopful of LAB was inoculated in 50 mL MRS medium containing salt and precultured at 24°C for 24 h. Five milliliters of the preculture was inoculated in 50 mL fresh MRS medium containing salt and with an initial pH of either 6.5, 6.0, 5.5, 5.0, 4.5, or 4.0, and the culture was then incubated statically at 30°C for 48 h. The absorbance at 660 nm of each culture was measured periodically with a spectrophotometer (UV-1700; Shimadzu, Kyoto, Japan).

Analysis Takanazuke on days 0, 4, 7, 19, 30, 40, 90, and 180 was wrung with a triple layer of gauze. The obtained solution was centrifuged at 8000 × g at 4°C for 15 min, and the supernatant was filtrated through a 0.45-μm cellulose acetate membrane filter (Advantec, Tokyo, Japan). The pH of the filtered solution was measured using a pH meter (HM-25G; DKK-TOA, Tokyo, Japan) and lactate concentrations were measured using an F-kit Lactate (Roche Diagnostics Corp., Basel, Switzerland). The concentration of extracted DNA was determined by the measurement of absorbance at 260 nm with a spectrophotometer (DU-530; Beckman, Brea, CA).

The sequences of the partial 16S rRNA gene obtained in this work have been deposited in the DNA Data Bank of Japan (DDBJ) under accession numbers AB852111 – AB852186.

Results and Discussion

Change in lactate concentration during fermentation The pH values of water collected from fermenting takanazuke were 6.62, 6.51, 6.11, 5.75, 5.19, 4.19, 4.22, and 3.83 on days 0, 4, 7, 19, 30, 40, 90, and 180, respectively, and lactate concentrations were 0.7, 8.7, 10.7, 18.5, 25.8, and 28.8 g/L on days 7, 19, 30, 40, 90, and 180 (Fig. 1). The ratios of L-lactate to total lactate were 63.4%, 97.0%, 84.1%, 47.3%, 43.8%, and 42.7% on days 7, 19, 30, 40, 90, and 180, respectively (Fig. 2). L-lactate was mainly produced during the early stage of fermentation, and the amount of D-lactate increased over time. Together, these results suggest that the LAB community changed over the course of the fermentation.

Fig. 1.

Changes in pH and lactate concentration during fermentation. Closed circles, lactate; closed squares, pH.

Fig. 2.

Change in D/L-lactate concentration during fermentation. Black bars, D-lactate; white bars, L-lactate.

Effect of initial pHon growth of LAB Prior to analyzing the changes in the LAB community associated with takanazuke, the effect of pH on three previously isolated LAB strains, L. curvatus, L. sakei, and L. (para)plantarum was investigated (Sakai et al., 2013). As shown in Fig. 3, both L. curvatus and L. sakei were unable to grow at initial pHs of 4.0 and 4.5, whereas L. (para) plantarum was comparatively acid tolerant.

Fig. 3.

Growth curves of L. curvatus B17–2 (a), L. sakei A165–1 (b), and L. (para)plantarum B17–4 (c) in MRS broth medium at various pHs. Closed squares, pH 6.5; open squares, pH 6.0; closed triangles, pH 5.5; open triangles, pH 5.0; closed circles, pH 4.5; open circles, pH 4.0.

Change in microbial community composition during fermentation The concentrations of DNA extracted from takanazuke were 19, 100, 70, 103, and 115 ng/μL on days 0, 19, 40, 90, and 180 of fermentation, respectively. Partial 16S rRNA gene sequences were amplified from the extracted DNA, and a total of 29, 46, 51, 57, and 52 clones were identified on days 0, 19, 40, 90, and 180, respectively. The genus of each clone was determined by sequencing analysis, and the profiles of each sample are presented in Fig. 4. Only the genus Pseudomonas was detected on day 0, and the Pseudomonads were divided into 7 groups: P. syringae, P. fluorescens, P. chlororaphis, P. putida, P. stutzeri, P. aeruginosa, and P. pertucinogena (Anzai et al., 2000; Garrity et al., 2005). Twenty-six of the 29 clones identified on day 0 belonged to P. syringae, as shown in Fig. 5. At the time of harvesting the takana specimens, several of the leaves of contained ulcers, which may have been possibly caused by P. syringae, as this species has been previously reported as a plant pathogen (Takikawa et al., 1989) and was a dominant clone in our analyses.

Fig. 4.

Temporal profiles of the bacterial communities during fermentation. Legend for bars: cross hatched, Pseudomonas sp.; dotted, L. curvatus; checkered, L. (para)plantarum; striped, L. sakei; gray, Weissella sp.; black, L. alimentarius; white, Clostridium sp.

Fig. 5.

Phylogenetic tree of clones based on similarities in partial 16S rRNA gene sequences measured on day 0.

On day 19, Pseudomonas was no longer detected in the takanazuke, and L. curvatus had instead become the dominant species, at which point lactate concentration was 8.7 g/L. It was reported that LAB increased during leek fermentation from log 2-log 3 cfu/mL at day 0 to log 8-log 9 cfu/mL after 1-week incubation on MRS plates, whereas Pseudomonadaceae decreased from log 5 cfu/mL to almost zero during the same period on Pseudomonas cetrimide plates (Wouters et al., 2013b). This result is similar to that of the microbial community profile observed during takana fermentation in the present study. Although organic acids such as lactate can inhibit the growth of bacteria and fungi (Holzapfel et al., 1995), the growth of Pseudomonas during takanazuke fermentation appeared to be inhibited by salt and acid. All 46 clones on day 19 were most closely related to L. curvatus, which was isolated in our previous study. L. curvatus B17-2 produced 54% L-lactate and 46% D-lactate in our previous fermentation experiment using takana juice as a substrate (Sakai et al., 2013). However, nearly all of the lactate produced by day 19 of the present takana fermentation was L-lactate (Fig. 2). L. curvatus produces lactate racemase following induction by L-lactate (Garvie, 1980; Schleifer, 2009); therefore, this enzyme may have been active in the previous fermentation with takana juice medium, but was possibly only expressed at low levels in the present takana fermentation due to the low L-lactate concentration on day 19.

On day 40, L. (para)plantarum (39.2%) and L. sakei (3.9%) were detected in addition to L. curvatus (56.9%). On day 90, L. (para) plantarum (50.9%) and L. sakei (45.6%) were the dominate species in the culture, whereas the abundance of L. curvatus (1.7%) and members of the genus Weissella (1.7%) were markedly reduced. Although L. sakei was acid-sensitive compared with L. (para)plantarum (Fig. 3), L. sakei accounted for 45.6% of clones. L. sakei can utilize arginine as a carbon source when the sugar content is low (van de Guchte et al., 2002; Rimaux et al., 2012), unlike L. curvatus (Schleifer, 2009). The synthesis of citrulline from arginine was repressed in synthetic medium containing more than 27.5 mM glucose (Montel and Champomier, 1987). L. sakei was first detected and isolated from fermented sausage (Kesmen et al., 2012), which contains high protein levels and low sugar content, as is typical of fermented meat. Under such conditions, L. sakei uses the arginine deiminase (ADI) pathway (Rimaux et al., 2012) to convert arginine to ornithine and ATP, and also producing ammonia as a byproduct. L. sakei CTC 494 has been reported to use the ADI pathway at pH 4.5, albeit not to a high degree (Rimaux et al., 2011). As the sugar content was low and L. sakei might use the ADI pathway, L. sakei was able grow on day 90 of fermentation. The apparent utilization of the ADI pathway may have also contributed to the pH increase from 4.19 to 4.22 between days 40 and 90 by the production of ammonia, although the lactate concentration increased during this period. On day 180, L. (para) plantarum (59.6%) was dominant, although members of the genus Weissella (32.7%), and L. curvatus (3.8%), L. alimentarius (1.9%), and genus Clostridium (1.9%) were also detected. L. (para) plantarum increased gradually after 40 days of fermentation. This increase may have been due to the relatively high acid tolerance of this species compared with L. curvatus (Fig. 3). The increase production of D-lactate during fermentation might be dependent on the growth of L. (para) plantarum.

In the production of sauerkraut, heterotypic Leu. mesenteroides initiates lactate fermentation, after which the acid-tolerant LAB strains L. plantarum and L. brevis continue the process, with homotypic L. plantarum remaining dominant at the final stage of fermentation (Hutkins, 2006, Plengvidhya et al., 2007). In the traditional fermentation of cauliflower (B. oleracea) in Greece, the Leu. mesenteroides group is dominant during the early stages, with the L. plantarum group becoming dominant at subsequent stages (Paramithiotis et al., 2010). In the fermentation of turnips (B. rapa) for the production of brovada in Italy, heterotypic L. hilgardii was isolated at the early stage and Pediococcus parvulus was isolated mainly during following stages, with L. plantarum being consistently isolated (Maifreni et al., 2004). During the fermentation of mustard (suan-tsai) in Taiwan, Lactobacillus, Leuconostoc, Weissella, and Pediococcus were isolated after 3 days of fermentation, and L. plantarum and L. brevis were predominantly isolated after 2 months (Chao et al., 2009). In our previous study, Leu. mesenteroides was mainly isolated on day 13 of takanazuke (NaCl 6% [w/w]) fermentation, and L. sakei, L. curvatus, L. (para)plantarum, and Leu. mesenteroides were isolated on day 17 after the pH had decreased (Sakai et al., 2013). L. (para)plantarum was predominantly isolated with continued fermentation, whereas L. brevis was isolated at the late stages (Sakai et al., 2013). In addition, we also isolated Enterococcus faecium during takana fermentation (NaCl 8% (w/w)) (Sakai et al., 2013). In contrast to the fermentation method used in that study, the traditional method of takana fermentation in the Aso region yielded L. (para)plantarum, L. parabrevis, and P. parvulus as the dominant species (Sakai et al., 2013). However, in the present study, E. faecium, L. parabrevis, and P. parvulus were not detected during the takanazuke fermentation. This finding suggests that these three species might not be associated with takana plants or the surrounding soil, but rather, are found in fermentation pots in the Aso region.

We previously selected L. plantarum B17–4 as the starter LAB for production of takanazuke, and in the present study, L. plantarum was detected between active lactate fermentation and the end stages of fermentation. Taken together, these past and present findings indicate that L. plantarum is a suitable starter strain for takanazuke production.

Conclusions

We analyzed changes in the microbial community composition during the production of takanazuke. Although Pseudomonas was dominant on day 0, L. curvatus was the dominant species on day 19 of fermentation, which corresponded to high l-lactate production. Notably, the microbial community of LAB changed continuously during the 180-day fermentation period. Although the microbial composition might differ between the specific conditions found in Aso and Kumamoto, L. plantarum isolated from takanazuke in our previous study was the major LAB and therefore may be a suitable starter strain for takanazuke production.

Acknowledgements

We thank Mr. Tsuyoshi Ichihara of Ichihara Farm, Aso city for his kind contribution of takana specimens and advice regarding takanazuke production.

References

•  Ampe,  F.,  ben Omar,  N.,  Moizan,  C.,  Wacher,  C., and  Guyot,  J.P. (1999). Polyphasic study of the spatial distribution of microorganisms in Mexican pozol, a fermented maize dough, demonstrates the need for cultivation-independent methods to investigate traditional fermentations. Appl. Environ. Microbiol., 65, 5464–5473.
•  Anzai,  Y.,  Kim,  H.,  Park,  J.Y.,  Wakabayashi,  H., and  Oyaizu,  H. (2000). Phylogenetic affiliation of the pseudomonads based on 16S rRNA sequence. Int. J. Syst. Evol. Microbiol., 50, 1563–1589.
•  Chao,  S.H.,  Wu,  R.J.,  Watanabe,  K., and  Tsai,  Y.C. (2009). Diversity of lactic acid bacteria in suan-tsai and fu-tsai, traditional fermented mustard products of Taiwan. Int. J. Food Microbiol., 135, 203–210.
•  Di Cagno,  R.,  Coda,  R.,  De Angelis,  M., and  Gobbetti,  M. (2013). Exploitation of vegetables and fruits through lactic acid fermentation. Food Microbiol., 33, 1–10.
•  Endo,  A.,  Mizuno,  H., and  Okada,  S. (2008). Monitoring the bacterial community during fermentation of sunki, an unsalted, fermented vegetable traditional to the Kiso area of Japan. Lett. Appl. Microbiol., 47, 221–226.
•  Garrity,  G.M.,  Bell,  J.A,, and  Lilburn,  T. (2005). Pseudomonadales, p. 323–379. In  Garrity,  G.,  Brenner,  D.J.,  Krieg,  N.R., and  Staley,  J.R. (ed.), Bergey's Manual of Systematic Bacteriology, Vol.2, The proteobacteria Part B. Springer.
•  Garvie,  E. I. (1980). Bacterial lactate dehydrogenases. Microbiol. Rev., 44, 106–139.
•  Giraffa,  G. (2004). Studying the dynamics of microbial populations during food fermentation. FEMS Microbiol. Rev., 28, 251–260.
•  Holzapfel,  W. H.,  Geisen,  R., and  Schillinger,  U. (1995). Biological preservation of foods with reference to protective cultures, bacteriocins and food-grade enzymes. Int. J. Food Microbiol., 24, 343–362.
•  Holzapfel,  W.H. (2002). Appropriate starter culture technologies for smallscale fermentation in developing countries. Int. J. Food Microbiol., 75, 197–212.
•  Hutkins,  R. W. (2006). Fermented vegetables. In: Microbiology and Technology of Fermented Foods. Blackwell Publishing, Oxford, pp. 233–259.
•  Kesmen,  Z.,  Yetiman,  A. E.,  Gulluce,  A.,  Kacmaz,  N.,  Sagdic,  O.,  Cetin,  B.,  Adiguzel,  A.,  Sahin,  F., and  Yetim,  H. (2012). Combination of culture-dependent and culture-independent molecular methods for the determination of lactic microbiota in sucuk. Int. J. Food Microbiol., 153, 428–435.
•  Maifreni,  M.,  Marino,  M., and  Conte,  L. (2004). Lactic acid fermentation of Brassica rapa: chemical and microbial evaluation of a typical Italian product (brovada). Eur. Food Res. Technol., 218, 469–473.
•  Montel,  M.C. and  Champomier,  M.C. (1987). Arginine catabolism in Lactobacillus sake isolated from meat. Appl. Environ. Microbiol., 53, 2683–2685.
•  Nguyen,  D.T.,  Van Hoorde,  K.,  Cnockaert,  M.,  De Brandt,  E.,  Aerts,  M.,  Binh,  Thanh, L., and  Vandamme,  P. (2013). A description of the lactic acid bacteria microbiota associated with the production of traditional fermented vegetables in Vietnam. Int. J. Food Microbiol., 163, 19–27.
•  Paramithiotis,  S.,  Hondrodimou,  O. L., and  Drosinos,  E. H. (2010). Development of the microbial community during spontaneous cauliflower fermentation. Food Res. Int., 43, 1098–1103.
•  Plengvidhya,  V.,  Breidt,  F.,  Lu,  Z., and  Fleming,  H. (2007). DNA fingerprinting of lactic acid bacteria in sauerkraut fermentations. Appl. Environ. Microbiol., 73, 7697–7702.
•  Rimaux,  T.,  Rivière,  A.,  Illeghems,  K.,  Weckx,  S.,  De Vuyst,  L., and  Leroy,  F. (2012). Expression of the arginine deiminase pathway genes in Lactobacillus sakei is strain dependent and is affected by the environmental pH. Appl. Environ. Microbiol., 78, 4874–4883.
•  Rimaux,  T.,  Vrancken,  G.,  Pothakos,  V.,  Maes,  D.,  De Vuyst,  L., and  Leroy,  F. (2011). The kinetics of the arginine deiminase pathway in the meat starter culture Lactobacillus sakei CTC 494 are pH-dependent. Food Microbiol., 28, 597–604.
•  Sakai,  M.,  Nagano,  M.,  Ohta,  H.,  Kida,  K., and  Morimura,  S. (2013). Isolation of lactic acid bacteria from takanazuke as a starter strain to reduce added salt and stabilize fermentation. Food Sci. Technol. Res., 19, 577–582.
•  Schleifer,  K.H. (2009). Lactobacillales, p.465–511. In  Vos,  P.,  Garrity,  G.,  Jones,  D.,  Krieg,  N.R.,  Ludwig,  W.,  Rainey,  F.A.,  Schleifer,  K.H., and  Whitman,  W.B. (ed.), Bergey's Manual of Systematic Bacteriology, Vol.3, The Firmicutes. Springer.
•  Tamang,  J. P.,  Tamang,  B.,  Schillinger,  U.,  Franz,  C. M.,  Gores,  M., and  Holzapfel,  W. H. (2005). Identification of predominant lactic acid bacteria isolated from traditionally fermented vegetable products of the Eastern Himalayas. Int. J. Food Microbiol., 105, 347–356.
•  Tanganurat,  W.,  Quinquis,  B.,  Leelawatcharamas,  V., and  Bolotin,  A., (2009). Genotypic and phenotypic characterization of Lactobacillus plantarum strains isolated from Thai fermented fruits and vegetables. J. Basic. Microbiol., 49, 377–385.
•  Takikawa,  Y.,  Serizawa,  S.,  Ichikawa,  T.,  Tsuyumu,  S., and  Goto,  M. (1989). Pseudomonas syringae pv. actinidiae pv. nov.: the causal bacterium of canker of kiwifruit in Japan. Ann. Phytopath. Soc. Japan, 55, 437–444.
•  Temmerman,  R.,  Huys,  G., and  Swings,  J. (2004). Identification of lactic acid bacteria: culture-dependent and culture-independent methods. Trends Food Sci. Technol., 15, 348–359.
•  van de Guchte,  M.,  Serror,  P.,  Chervaux,  C.,  Smokvina,  T.,  Ehrlich,  S.D., and  Maguin,  E. (2002). Stress responses in lactic acid bacteria. _Antonie van Leeuwenhoek, 82, 187–216.
•  Wouters,  D.,  Grosu-Tudor  S.,  Zamfir,  M., and  De Vuyst,  L. (2013a). Bacterial community dynamics, lactic acid bacteria species diversity and metabolite kinetics of traditional Romanian vegetable fermentations. J. Sci. Food Agric., 93, 749–760.
•  Wouters,  D.,  Bernaert,  N.,  Conjaerts,  W.,  Van Droogenbroeck,  B.,  De Loose,  M., and  De Vuyst,  L. (2013b). Species diversity, community dynamics, and metabolite kinetics of spontaneous leek fermentations. Food Microbiol., 33, 185–196.