Contribution of glutamate decarboxylase in Lactobacillus reuteri to acid resistance and persistence in sourdough fermentation
© Su et al; licensee BioMed Central Ltd. 2011
Published: 30 August 2011
Skip to main content
© Su et al; licensee BioMed Central Ltd. 2011
Published: 30 August 2011
Acid stress impacts the persistence of lactobacilli in industrial sourdough fermentations, and in intestinal ecosystems. However, the contribution of glutamate to acid resistance in lactobacilli has not been demonstrated experimentally, and evidence for the contribution of acid resistance to the competitiveness of lactobacilli in sourdough is lacking. It was therefore the aim of this study to investigate the ecological role of glutamate decarboxylase in L. reuteri.
A gene coding for a putative glutamate decarboxylase, gadB, was identified in the genome of L. reuteri 100-23. Different from the organization of genetic loci coding for glutamate decarboxylase in other lactic acid bacteria, gadB was located adjacent to a putative glutaminase gene, gls3. An isogenic deletion mutant, L. reuteri ∆gadB, was generated by a double crossover method. L. reuteri 100-23 but not L. reuteri ∆gadB converted glutamate to γ-aminobutyrate (GABA) in phosphate butter (pH 2.5). In sourdough, both strains converted glutamine to glutamate but only L. reuteri 100-23 accumulated GABA. Glutamate addition to phosphate buffer, pH 2.5, improved survival of L. reuteri 100-23 100-fold. However, survival of L. reuteri ∆gadB remained essentially unchanged. The disruption of gadB did not affect growth of L. reuteri in mMRS or in sourdough. However, the wild type strain L. reuteri 100-23 displaced L. reuteri ∆gadB after 5 cycles of fermentation in back-slopped sourdough fermentations.
The conversion of glutamate to GABA by L. reuteri 100-23 contributes to acid resistance and to competitiveness in industrial sourdough fermentations. The organization of the gene cluster for glutamate conversion, and the availability of amino acids in cereals imply that glutamine rather than glutamate functions as the substrate for GABA formation. The exceptional coupling of glutamine deamidation to glutamate decarboxylation in L.Â reuteri likely reflects adaptation to cereal substrates.
Sourdough is used in bread production as a leavening agent, or as a baking improver to improve the texture, flavour, and shelf life of bread [1, 2]. Sourdoughs used as the sole leavening agent (type I sourdoughs) are maintained by frequent back-slopping. These conditions select for fast growing microorganisms, and type I sourdough microbiota are typically dominated by Lactobacillus sanfranciscensis[3–5]. Industrial sourdoughs used for the production of baking improvers (type II sourdoughs) are fermented at 35 – 40°C for extended fermentation times (2 – 5 days) to achieve high levels of total titrable acidity [1, 3]. These sourdoughs are typically dominated by thermophilic, acid-tolerant lactobacilli, including Lactobacillus pontis, and Lactobacillus panis[3–5]. Lactobacillus reuteri is generally considered an intestinal organism  but also occurs in type II sourdoughs [3–5], and was shown to persist in an industrial sourdough fermentation over a period of more than one decade, corresponding to several 10,000 generations of bacterial growth . Model sourdough fermentations confirmed that long fermentation times at high temperature (42°C) select for L. reuteri[8, 9].
Owing to long fermentation times, acid stress impacts the persistence of lactobacilli in type II sourdough fermentation. Mechanisms of acid tolerance in Gram-positive bacteria generally include the expression of stress proteins, the overexpression of proton pumps, and modification of metabolic pathways to consume protons [10, 11]. Remarkably, metabolic pathways contributing to acid resistance of L. reuteri also improve bread quality. The formation of dextran, reuteran, or levan by lactic acid bacteria contributes to acid resistance [12, 13] and improves bread texture and volume . Conversion of arginine to ornithine improves the acid resistance of L. reuteri[10, 15, 16]. Ornithine is also a precursor of the character impact compound of wheat bread crust flavour, 2-acetyl-1-pyrroline . The deamidation of glutamine to glutamate by sourdough lactobacilli generates umami taste in sourdough bread, and improves growth of L. reuteri at low pH .
Glutamate decarboxylation to γ-aminobutyric acid (GABA) contributes to the acid resistance of Escherichia coli, Listeria monocytogenes, and Lactococcus lactis[19–21]. The antiport of glutamate and GABA generates a ΔpH and ΔΨ contributing to a proton motive force for ATP synthesis . However, although glutamate decarboxylases were biochemically characterized in L. brevis and L. paracasei[23, 24], the physiological function of glutamate decarboxylase in aciduric Lactobacillus is unclear. Moreover, experimental evidence for the contribution of any acid resistance mechanism to the competitiveness of lactobacilli in sourdough is lacking. It was therefore the aim of this study to investigate the ecological role of glutamate decarboxylase in cereal-associated L. reuteri. Experiments were conducted with L. reuteri 100-23 . This strain has a known genome sequence, harbours a glutamate decarboxylase, and decarboxylates glutamate during sourdough fermentation . An isogenic deletion mutant, L. reuteri ∆gadB, was generated by a double crossover method to elucidate the importance of GadB in acid resistance and the competitiveness in type II sourdoughs.
Bacterial strains and plasmids used in this study
Strain or plasmid
Source or reference
Escherichia coli JM109
Cloning host for pGEMTeasy- and pJRS233-derivative plasmids
Lactobacillus reuteri 100-23
Rodent isolate; wild type strain
Wild-type strain derivative with a deletion in gadB
Cloning vector used in E. coli; 3.0 kb; Ampr
pGEMTeasy containing 1 kb of the DNA sequence upstream of gadB; 4.0 kb; Ampr
pGEMTeasy containing 0.9 kb of the DNA sequence downstream of gadB; 3.9 kb; Ampr
pGEMTeasy containing the upstream and downstream sequences of gadB; 4.9 kb; Ampr
Shuttle vector used in the hosts E. coli and L. reuteri 100-23; 6.0 kb; Ermr
pJRS233 containing 1.9 kb of the flanking sequences of gadB; 7.9 kb; Ermr
DNA was isolated from overnight cultures using the Blood & Tissue Kit (Qiagen, Mississauga, Canada) according to instructions of the manufacturer. Oligonucleotides were purchased from Integrated DNA Technologies (San Diego, CA), restriction enzymes from New England Biolabs (Pickering, Canada), T4 DNA ligase from Epicentre (Markham, Canada) and Taq DNA polymerase from Invitrogen (Burlington, Canada). DNA sequencing was performed by Macrogen (Rockville, Maryland).
The glutamate decarboxylase (GadB) protein sequence of Lactobacillus paracasei (GI:169264609) was used to identify a homologous gene encoding gadB in the genome of L. reuteri 100-23 (http://www.jgi.doe.gov/) . Gene loci of putative gadB genes and flanking nucleotide sequences in Lactococcus lactis subsp. cremoris MG1363 , L. reuteri 100-23 and L. plantarum WCFS1  were analyzed with the BLASTx program against the National Center for Biotechnology Information databases (http://blast.ncbi.nlm.nih.gov/Blast.cgi). The protein sequences of GadB from various species were retrieved from Uniprot database and aligned to calculate the score of similarity using Geneious alignment (Geneious version 5.1.6, http://www.geneious.com).
The gene coding for GadB in L. reuteri 100-23 was truncated using pJRS233  according to a deletion strategy described earlier . A 5’ 1016-bp fragment of gadB was amplified from genomic DNA of L. reuteri 100-23 using primers gadB-KO1-PstI (5’-AACTGCAGGTTCAATTTTCAGCACATG-3’) and gadB-KO2-XbaI (5’- GCTCTAGATATCCTGCCATAGATAAAACCTC-3’). The amplicon was ligated into pGEMTeasy vector (Promega) to generate pGadB-A. The plasmid pGadB-B containing the 3’ flanking fragment of gadB was created using gadB-KO3-XbaI (5’- GCTCTAGATTCACTCATTAAACCTTAGAA-3’) and gadB-KO4-BamHI (5’- CGGGATCCAATGGCTGCAGGGATA-3’). The flanking fragments of gadB from these plasmids were digested with the respective restriction enzymes, purified, and ligated to create pGadB-AB. The DNA fragment in pGadB-AB was cut with PstI and BamHI and ligated into the PstI-BamHI sites in pJRS233. The resulting plasmid pKO-gadB-AB was electrotransformed in competent L. reuteri 100-23 cells resuspended in water with 30% (v/v) polyethylene glycol (MW 3350; J.T. Baker Chemical, Phillipsburg, NJ). Transformants were grown in mMRS-erythromycin broth at 42 – 44°C for 80 generations to select for single crossover mutants. L. reuteri with pKO-gadB-AB integrated into chromosome were cured by culturing in mMRS broth at 37°C for 100 generations. The culture was plated on mMRS agar and erythromycin-sensitive double crossover mutants were identified by replica plating mMRS and mMRS-erythromycin agar. The truncation of gadB in L. reuteri 100-23∆gadB was confirmed by PCR with the primers gadB-KO1-PstI and gadB-KO4-BamHI. An amplicon with the expected size of 3000 bp was obtained with DNA from the L. reuteri 100-23 as template. PCR with DNA from L. reuteri 100-23∆gadB yielded a 2000 bp amplicon. A second PCR with primers gadB-5F (5’-GGTCTTATTACCGTTCCTAAT-3’) and gadB-6R (5’-ACATTTCTTATGGGATTGCAT-3’) yielded 1700 bp and 500 bp amplicons from the L. reuteri 100-23 and L. reuteri ΔgadB, respectively. DNA sequencing was conducted to verify the deletion region using primers gadB-5F and gadB-6R.
Growth of L. reuteri 100-23 and L. reuteri ΔgadB was assessed in mMRS (pH 6.7) and mMRS acidified to pH values of 4.7 and 3.8. Media were inoculated with overnight cultures, incubated at 37°C, and growth was monitored by measuring the optical density (OD) at 600 nm. To evaluate acid resistance, overnight grown cultures were harvested by centrifugation, washed in 50 mM Na2HPO4 buffer (pH 7), and resuspended in 50 mM potassium phosphate buffer (pH 2.5) to an OD600nm of 1.0. To assess the contribution of amino acid metabolism to acid resistance, parallel experiments were conducted in potassium phosphate buffer (pH 2.5) supplemented with 10 mM glutamate or arginine. A pH of 2.5 adjusted with HCl was chosen to match conditions previously used to determine the effect of glutamate decarboxylase on acid resistance in E. coli and L. monocytogenes[10, 19, 20]. Samples were taken after 0, 1, 3, 5, 8 and 24 h of incubation at pH 2.5 for quantification of amino acids (see below), and to monitor bacterial survival. For determination of bacterial survival, samples were immediately mixed with phosphate buffered saline (PBS, 137 mmol L-1 NaCl, 2.7 mmol L-1 KCl, 10 mmol L–1 Na2HPO4, and 2 mmol L-1 KH2PO4, pH 8.0) and diluted in PBS buffer prior to plating on MRS agar. For the samples of 24 h acid treatment, 10 mL of culture was centrifuged, resuspended in PBS buffer and plated on MRS agar to determine the bacterial survival. Growth curves and survival at pH 2.5 were determined in triplicate independent experiments.
Sourdough fermentations were performed with L. reuteri 100-23 and L. reuteri ∆gadB to examine the effect of glutamate decarboxylase on growth, acidification, pH, and amino acid concentrations. Doughs were prepared from 40 g whole wheat flour and sterile tap water to achieve a dough yield of 200 [(dough mass/flour mass) × 100], and inoculated at 37°C with an initial cell count of 1 ± 0.5 × 107 CFU g-1. Initially, sourdoughs were inoculated with L. reuteri 100-23 or L. reuteri ∆gadB separately, and samples were collected after 0, 6, 12, 24, 48, 72, and 96 h of fermentation. In a second experiment, L. reuteri 100-23 and ΔgadB were co-cultured in the same sourdough. Doughs were fermented for 96 h (trial I), or for 240 h by backslopping every 48 h (trial II) to match conditions known to select for L. reuteri. At each refreshment step, 5% of the ripe sourdough was used as an inoculum for the subsequent fermentation. Samples were taken to measure pH and viable cell counts. The dough samples were stored at -20°C for subsequent DNA extraction and amino acid analysis (see below). Sourdough fermentations were carried out in two independent experiments, and each sample was analyzed in duplicate.
Cells from phosphate buffer (pH 2.5) were removed by centrifugation. The supernatant (1 vol) was mixed with deionized water (4 vol), saturated potassium borate (4 vol) and β-aminobutyric acid used as internal standard (1 vol). Sourdough samples were lyophilized, extracted with water at an extraction ratio of 1:6 (w/v) and diluted with water, potassium borate, and internal standard as described above. Amino acids were quantified by HPLC after derivatisation with o-phtaldialdehyde .
Two grams of sourdough were washed with 4 ml of PBS buffer (pH 7.8) and resuspended in PBS. Flour solids were removed by 1500 × g for 10 min and cells were harvested by centrifugation at 13,000 × g. Cells were resuspended in 660 µL lysis buffer (10 mmol L-1 Tris-Cl, 0.5 mmol L-1 EDTA, 10% SDS, pH 7.5), and incubated at 65°C for 10 min. Protein was precipitated by addition of 340 µL 5 mol L-1 potassium acetate and incubation at -20°C for 10 min. After centrifugation, 900 µL of the supernatant was mixed with an equal volume of isopropanol, DNA was pelleted by centrifugation and washed with 95% ice-cold ethanol. DNA was resuspended in 20 µL of 10 mmol L-1 Tris-Cl (pH 8.5), and the DNA concentration was determined by UV spectrometry using a NanoDrop spectrophotometer ND-1000 (Thermo Fisher Scientific Inc., Wilmington, DE, USA).
The cell counts of L. reuteri 100-23 relative to the L. reuteri ΔgadB was quantified using a probe-based qPCR assay targeting the native and disrupted regions of gadB, respectively. Two μg of chromosomal DNA of L. reuteri 100-23 and L. reuteri ΔgadB were digested separately with NotI, purified and diluted to a known DNA concentration corresponding to a known copy number gadB or ∆gadB for use as standards. DNAs extracted from sourdough were also digested with NotI and purified using the QIAquick PCR Purification kit (Qiagen). Strain-specific primers and probes were designed using Primer Express Software 3.0 (Applied Biosystems, Streetsville, Ontario, Canada). Primers and probe specific for L. reuteri 100-23 (WT-qPCR-F, 5’-TGGGATTTCCAACTAAAGAATGTTG-3’; WT-qPCR-R, 5’-CAACCAATACCAGGATAAACTAAACCA-3’; and WT-qPCR-probe, 5’-Hex/TCCATTAACGCTTCCGGCCACAAGT/IABkFQ-3’) target the DNA region of gene gadB, which is deleted in L. reuteri ΔgadB strain. Primers and probe specific for L. reuteri ΔgadB (ΔgadB-qPCR-F, 5’-CATATCATTGCAAATTCAGACGAAA-3’; ΔgadB-qPCR-R, 5’-ATCTAAGCAAGTTGTTATGCTTGTTTAGATC-3’; and ΔgadB-qPCR-probe (5’-TET/CCTAGGAGGTTTTATCTATGGCAGGATAATCTAGATTCAC/IABkFQ-3’) target the joint site of deleted gadB sequence which is absent in the wild type strain. PCR was carried out with Rox Brilliant II Master Mix (Stratagene, Mississauga, Ontario, Canada) in a 7500 Fast Real-time PCR instrument (Applied Biosystems). Standard curve, determination of the PCR efficiency, and calculation of the copy number were carried out with 7500 Fast System SDS v1.4 software.
DNA sequences of L. reuteri 100-23 were obtained from the National Center for Biotechnology Information databases (GenBank: AAPZ02000002.1). The gene regions of predicted glutamate/γ-aminobutyrate antiporter (gadC2), glutaminase (gls3), glutamate/γ-aminobutyrate antiporter (gadC1), glutamate decarboxylase (gadB) and xanthine/uracil/vitamin C permease are 549316..550869, complement (548219..549139), complement (546271..547809), complement (544847..546253) and 543444..544787, respectively. The sequence of the truncated gadB in L. reuteri ΔgadB was deposited with accession number JF339969.
The effect of glutamate on acid resistance of L. reuteri 100-23 and L. reuteri ∆gadB was compared to the effect of arginine. The effect of arginine on survival of L. reuteri 100-23 was comparable to the effect of glutamate (Fig. 3). Notably, L. reuteri ΔgadB was more resistant in arginine-containing medium than the wild-type strain, indicating that the loss of the function of acid resistance may be compensated by over-expression of alternative pathways to achieve pH homeostasis.
The evaluation of glutamine and glutamate metabolism by L. reuteri 100-23 and L. reuteri ∆gadB in sourdough revealed that disruption of gadB had no influence on the concentration of total amino acids or the concentration of (glutamine + glutamate + GABA). After 96 h of incubation, the concentration of total amino acids for the wild type and ∆gadB strains were 141 ± 7 and 129 ± 8 mmol kg-1 DM, respectively. The concentrations of (glutamine + glutamate + GABA) for the wild type and ∆gadB strains were 26 ± 4 and 21 ± 4 mmol kg-1 DM, respectively. In sourdoughs fermented with both strains, glutamine levels remained low through out fermentation, indicating a quantitative conversion by L. reuteri (Fig. 4B). L. reuteri 100-23 converted glutamine to GABA, which accumulated to 25 mmol kg-1 DM after 96 h of fermentation (Fig. 4B). In contrast, L. reuteri ∆gadB accumulated glutamate. Low levels of GABA at the beginning of the fermentation and in the sourdoughs fermented with L. reuteri ∆gadB are attributable to interference by tyrosine, whose retention time was close to that of GABA.
This study investigated the role of gadB coding for glutamate decarboxylase in the acid resistance of L. reuteri, and demonstrated that glutamate decarboxylation contributes to the competitiveness of L. reuteri in type II sourdoughs that are propagated by continuous backslopping.
The contribution of metabolic traits of L. reuteri to its competitiveness in the intestine was previously evaluated with insertional mutations generated by single-crossover mutagenesis [6, 33]. This study employed a double crossover method to generate an in-frame deletion of gadB in L. reuteri 100-23. This approach avoids interference by antibiotic-resistance genes or other plasmid-borne foreign genes that remain on the chromosome of mutant strains generated by single-crossover mutagenesis.
The identification of lactic acid bacteria in sourdoughs was previously determined by using species-specific, semi-quantitative denaturing gradient gel electrophoresis (DGGE) [8, 9, 34, 35] or culture-dependent enumeration, followed by strain identification through random amplified polymorphic DNA [34, 36]. However, culture-dependent enumeration or DGGE fails to account for species or strains that contribute less than 1% to the total cell counts. The qPCR method used in this study achieved an accurate and consistent relative quantification of L. reuteri 100-23 and L. reuteri ∆gadB even in sourdoughs in which L. reuteri ∆gadB contributed less than 0.5% of the total population.
The arginine deiminase (ADI) pathway generates ATP, consumes intracellular protons, and causes the alkalization of the fermentation substrate [Fig. 6; [10, 11, 15]]. In keeping with prior studies, arginine improved the survival of L. reuteri at pH 2.5 about 10-100 fold , this study]. Glutamate decarboxylase was characterized in L. brevis and L. paracasei[23, 24], and GadB-positive lactobacilli were employed to accumulate GABA as a functional food ingredient in food fermentations [26, 38]. The strain-dependent expression of glutamate decarboxylase in lactobacilli, particularly L. reuteri, indicates that GadB is not essential for growth . This was confirmed by comparison of the growth of L. reuteri 100-23 and the ∆gadB mutant strain in mMRS and sourdough (Figs. 2, 4 and 5). However, although glutamate decarboxylation generally contributes to bacterial resistance to acid [10, 15], its contribution to acid resistance in acid-tolerant lactobacilli has not been demonstrated experimentally . Glutamate decarboxylation is particularly effective at pH values below the pKA of GABA, 4.23, as protonation of GABA below the pKA results in alkalization of the external fermentation substrate (Fig. 6). Accordingly, this study demonstrated a protective effect of glutamate decarboxylation at pH 2.5 in buffer, and at pH 3.6 in sourdough. Sourdough is acidified by lactic and acetic acids rather than HCl. However, growth and survival of lactobacilli in sourdough is determined by the pH and not by the concentration of undissociated organic acids . The protective effect of glutamate on the survival of L. reuteri at pH 2.5, an about 100-fold increase of cell counts after acid challenge, is comparable to the effect in E. coli[19, 37]. In L. monocytogenes, glutamate addition improved survival after exposure pH 2.5 of a wild type strain more than 6 log compared to a gadAB deficient mutant .
A contribution of glutamine deamidation to bacterial acid resistance has not been demonstrated [10, 15, 40] but glutamine improved the growth of L. reuteri and L. sanfranciscensis at low pH . Glutamine accounts for about 30% of the amino acids wheat proteins but the amount of glutamate is low . Consequently, proteolysis in wheat dough liberates glutamine rather than glutamate [18, 26]. Glutamate-mediated acid resistance in sourdough thus depends on the conversion of glutamine to glutamate. L. reuteri 100-23 harbours three putative glutaminase genes and gls3 is located adjacent to gadB. The substrate availability in cereal fermentations and the arrangement of gene loci strongly indicates that the glutamate-based acid resistance of L. reuteri 100-23 uses glutamine as a substrate (Fig. 6). In contrast, L. lactis MG1363 has genes coding for glutamine synthase downstream of the gadBC operon, and in L. plantarum WCFS1, no genes related to glutamine metabolism are located close to the gadB (Fig. 1). The exceptional arrangement of glutamine / glutamate catabolic genes in L. reuteri may reflect the adaptation to cereal substrates. Glutamine:GABA antiport by gadC or analogues has not been shown experimentally . However, ATP-dependent glutamate transport enzymes in L. lactis and L. delbrueckii also transport glutamine with high affinity [42, 43].
The disruption of gadB did not influence the survival of L. reuteri during long-term sourdough fermentation unless the sourdough was backslopped. The loss of one metabolic pathway for pH homeostasis in L. reuteri is apparently partially compensated by alternative pathways (Fig. 6). However, monitoring of the microbiota in sourdoughs maintained by continuous backslopping is considered the most appropriate tool to identify competitive strains [8, 34–36]. The observation that the wild type strain displaced L. reuteri ∆gadB after only a few refreshments demonstrates that acid resistance in general and glutamate decarboxylase in particular contributes to the competitiveness of L. reuteri in type II sourdoughs.
In addition to its occurrence in sourdough fermentations, L. reuteri is recognized as a stable member of the intestinal microbiota of humans and animals, and strains in the species have evolved to colonize the gastrointestinal tracts of specific hosts . In pigs, poultry, and rodents, L. reuteri colonizes the pars oesophagus, the crop, and the forestomach, respectively . It is interesting to note that sourdough isolates but not human or poultry isolates of L. reuteri are able to colonize rodents [45, 46]. Colonization and biofilm formation on the stratified, squamous epithelia lining the proximal gastrointestinal tract of animals occurs upstream of the stomach [44, 45]. Therefore, intestinal L. reuteri are exposed to stomach acidity. The conversion of glutamate to GABA in L. reuteri is strain-specific but was reported in human, rodent, and sourdough isolates [This study, . Mechanisms of acid resistance that are relevant in sourdough thus likely contribute to the competitiveness of L. reuteri in intestinal ecosystems, and may improve gastro-intestinal survival of L. reuteri commercially used as probiotics.
This study demonstrated that the conversion of glutamate to GABA by L. reuteri 100-23 contributes to acid resistance and to competitiveness in type II sourdough fermentations. The organization of the gene cluster for glutamate conversion, and the availability of amino acids in cereals imply that glutamine rather than glutamate functions as the substrate for GABA formation. The exceptional coupling of glutamine deamidation to glutamate decarboxylation in L.Â reuteri may reflect ecological adaptation to cereals, or to the proximal intestinal tracts of animals that predominantly feed on cereal grains.
high performance liquid chromatography
phosphate buffered saline
quantitative real time polymerase chain reaction.
The authors would like to thank Dr. Shih-Tung Liu for the critical discussions, Dr. Jens Walter for the access to the JGI and IMG databases. NSERC is acknowledged for funding. M. Gänzle acknowledges the Canada Research Chairs Program for financial support.
This article has been published as part of Microbial Cell Factories Volume 10 Supplement 1, 2011: Proceedings of the 10th Symposium on Lactic Acid Bacterium. The full contents of the supplement are available online at http://www.microbialcellfactories.com/supplements/10/S1.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.