If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Intracellular pH homeostasis through the extrusion of a proton by F0F1-ATPase is one of the key mechanisms used by lactic acid bacteria in response to acid stress, and also influences their post-fermentation acidification. In this study, the genotypic and phenotypic stability of a low post-fermentation acidification (LPA) mutant (designated as DGCC12411m) of Lactobacillus plantarum DGCC12411 was assessed. Compared with its mother strain, the pH of DGCC12411m in De Man, Rogosa, and Sharpe (MRS) broth after 48-h cultivation was 0.35 pH units higher. Incorporation of DGCC12411m in yogurt stored at ambient temperature (ambient yogurt) showed a reduced post-fermentation acidification during storage at 25°C for 120 d. Whole-genome sequencing analysis showed a SNP mutation (GGT > GAT at positions 505 to 507) in DGCC12411m, which resulted in the substitution of a highly conserved glycine residue by aspartic acid at the Walker A motif of the F0F1-ATPase α-subunit. However, degeneration of the LPA phenotype was observed after 5 passages of DGCC12411m in MRS broth. Analysis of DNA sequencing on both the whole population and the isolates showed that a back mutation occurred at the SNP site (GAT changed back to GGT) over the passaging, and the reversion gradually increased from a ratio of 10.8% at P5 to 60.0% at P10. We also found that the LPA phenotype stability of DGCC12411m was improved by supplementing 0.1 M potassium phosphate buffer to the growth medium as well as by reducing the inoculation rate of DGCC12411m to 2% (vol/vol). Such LPA Lactobacillus strains have potential for use as starter cultures in fermented foods with less change in acidity during shelf-life storage.
Most of the fermented foods consumed around the world are produced by lactic acid bacteria (LAB), which utilize carbohydrates from different raw materials to produce lactic acid as the main final product of their metabolism (
). Owing to its resistance to low pH and ability to utilize a wide range of carbohydrate sources, Lactobacillus plantarum has been used as starter culture in various food fermentations such as a nondairy fermented beverage from a blend of cassava and rice (
Probiotic strain Lactobacillus plantarum NBIMCC 2415 with antioxidant activity as a starter culture in the production of dried fermented meat products.
). However, strong post-fermentation acidification by L. plantarum limits its broader industrial application in fermented foods, such as dairy products, for which consumers are increasingly showing a preference for a relatively mild taste. Studies focused on starter cultures' post-fermentation acidification improvement have been reported (
Streptococcus thermophilus strain exhibiting unusual acidification kinetics in milk. P. 106 in 12th International Symposium of Lactic Acid Bacteria Abstract Book.
). Therefore, to understand the science in controlling post-fermentation acidification and developing methods to generate low post-fermentation acidification (LPA) variants are valuable to its wider industrial application. Low post-fermentation acidification is generally defined as pH value declining less than or equal to 0.3 pH units from pH 4.5 in an extended fermentation period of 24 h, according to industrial practice (
Potential probiotic characterization of Lactobacillus reuteri from traditional Chinese highland barley wine and application for room-temperature-storage drinkable yogurt.
), selection of acid-sensitive mutants, which are unable to continue fermentation below a certain pH, is another promising approach to improve control of post-fermentation acidification. F0F1-ATPase is a key enzyme in maintaining intracellular pH through extrusion of protons out of the cells by hydrolysis of ATP (
Yoghurt fermented by Lactobacillus delbrueckii ssp. bulgaricus H+-ATPase-defective mutants exhibits enhanced viability of Bifidobacterium breve during storage.
Int. J. Food Microbiol.2007; 116 (17434219): 358-366
demonstrated a negative correlation between ATPase activity and neomycin resistance in Escherichia coli. Reduced F0F1-ATPase activity resulted in reduction of aminoglycoside antibiotics uptake, thus increasing neomycin resistance (
Yoghurt fermented by Lactobacillus delbrueckii ssp. bulgaricus H+-ATPase-defective mutants exhibits enhanced viability of Bifidobacterium breve during storage.
Int. J. Food Microbiol.2007; 116 (17434219): 358-366
Rheological and physical characteristics of non-fat set yogurt prepared with EPS-producing Streptococcus thermophilus and an H+-ATPase-defective mutant Lactobacillus delbrueckii ssp. bulgaricus..
), a mutation in this gene that reduces ATPase activity and increases acid sensitivity may cause competitive disadvantages in growth media or food matrices without neomycin pressure. Subsequently, this may influence the mutant's stability and performance in food fermentation. However, the stability of F0F1-ATPase mutation in LPA LAB mutants has never been assessed. In this study, an acid-sensitive L. plantarum variant with an SNP mutation on its F0F1-ATPase α-subunit was generated. To understand the performance of DGCC12411m in industrial application, we tested its post-fermentation acidification and cell count during shelf-life storage in a yogurt stored at ambient temperature (hereafter termed “ambient yogurt”). We selected this application model because ambient yogurt is the largest fermented dairy product category in China, with continued high sales growth. In addition, genotypic and phenotypic stability of this variant were evaluated. Methods to improve its stability during propagation were also developed.
MATERIALS AND METHODS
Strain and Growth Conditions
The strain L. plantarum DGCC12411 was obtained from the DuPont Global Culture Collection (Niebüll, Germany); this strain was originally isolated from a traditional fermented vegetable product. This strain was deposited in Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (Braunschweig, Germany) under the accession number DSM32493. The strain was stored in 15% (vol/vol) glycerol at −80°C until use. The working cultures of DGCC12411 and its mutant strains were prepared by inoculating 1% (vol/vol) of the glycerol stock of each culture into fresh De Man, Rogosa, and Sharpe (MRS) broth (Oxoid, Basingstoke, UK) and incubated aerobically, under ambient gas conditions, at 30°C for 24 h. Similar aerobic cultivation in MRS medium was used throughout the experiment, unless mentioned otherwise.
Generated LPA Mutants
The acid-sensitive mutants were generated as previously described (
Yoghurt fermented by Lactobacillus delbrueckii ssp. bulgaricus H+-ATPase-defective mutants exhibits enhanced viability of Bifidobacterium breve during storage.
Int. J. Food Microbiol.2007; 116 (17434219): 358-366
), with minor modification. The DGCC12411 was revived by dissolving 100 mg of freeze-dried culture in 10 mL of sterile phosphate buffer (3M Co., Maplewood, MN), serially diluted, and plated at the appropriate dilution on MRS agar. The MRS agar plates were incubated at 30°C for 48 h. A single colony was picked and transferred into 10 mL of MRS broth and incubated at 30°C overnight until mid-exponential phase [optical density (OD)595nm = 0.8 to 0.9]. The OD reading was measured using Tecan Infiniti PRO 200 spectrophotometer (Thermo Fisher Scientific, Waltham, MA). To derive acid-sensitive mutants from DGCC12411, 200 µL [2% (vol/vol)] of the overnight culture was transferred to 10 mL of half-strength (26 g/L) MRS broth containing 700 μg/mL neomycin sulfate (CAS 1405-10-3; Rebio Technologies Ltd., Cranleigh, UK; purity ≥ 98%) and incubated at 30°C for 48 to 72 h until mid-exponential phase (OD595nm approximately 0.5). Subsequently, the culture was diluted in a sterile phosphate buffer (3M Co.) and appropriate dilutions plated on the half-strength MRS agar containing 700 μg/mL neomycin sulfate. The agar plates were then incubated at 30°C for 3 to 5 d, until visible colonies were observed. Two thousand single-colony isolates were randomly selected and grown in 96-well microtiter plates (Nunc, Roskilde, Denmark) containing 200 μL of MRS broth (pH 6.3) per well and incubated at 30°C for 48 h. After incubation, 1% (vol/vol) culture from each well of the microtiter plates was inoculated into 2 new microtiter plates containing 100 μL/well of fresh medium: one containing MRS broth at pH 6.3, and the other containing MRS broth adjusted to pH 4.5. The microtiter plates were incubated at 30°C for 72 h. Variants that grew well, as indicated by turbidity change at pH 6.3 (OD595nm ≥ 1.0 at 72 h after subtracted with the blank) but not at pH 4.5 (OD595nm ≤ 0.3 at 72 h after subtraction of the blank) were recognized as potential acid-sensitive mutants of DGCC12411.
The post-fermentation acidification of 8 selected variants was further validated by assessing their acidification in MRS. In brief, both DGCC12411 and the potential variants were grown in 10 mL of MRS broth at 30°C for 24 h. The cultures were adjusted to OD595nm 0.5 and inoculated at 2% (vol/vol) to 100 mL of fresh MRS broth. The culture was then incubated at 30°C in a water bath, and the acidification activity of each strain was monitored by iCINAC (AMS Alliance, Rome, Italy) every 3 min for 60 h. One mutant (designated as DGCC12411m) showing the highest stable final pH was selected for further study.
Post-Fermentation Acidification and Cell Count Assessment in Ambient Yogurt
A commercial ambient yogurt product is a fermented milk product that has been subjected to pasteurization after the end of fermentation. The pasteurization step inactivates the starter cultures used for fermentation, and this product contains no living cells. In this study, DGCC12411 and its mutant strain were respectively supplemented to commercially available ambient yogurt as adjunct cultures. Approximately 1.0 × 107 cfu/mL of DGCC12411 or DGCC12411m were inoculated into a commercial pasteurized ambient yogurt (Momchilovtsi, Bright Dairy, Shanghai, China) and stored at 25°C for 120 d. The pH and cell count were measured at d 0, 30, 60, 90, and 120. For cell count enumeration, the yogurt samples inoculated with DGCC12411 or its mutant strain were mixed well using a vortexer, and 11 g of the homogeneous sample was diluted in 99 mL of sterile phosphate buffer (3M Co.). Subsequently, 10-fold serial dilution was performed by using sterile phosphate buffer (3M Co.), and 100 μL of appropriate dilutions were spread-plated onto the MRS agar. The MRS agar plates were then incubated at 30°C for 48 h, followed by colony enumeration. Three individual replicates were performed for each treatment for pH and cell count measurements.
ATPase Gene PCR and Sequence Analysis
Based on the aligned F0F1-ATPase gene sequences of DGCC12411 and DGCC12411m, primers flanking the mutation site (ATPase-F: 5′-TCAACCAGTTGACGGTTCAG-3′, and ATPase-R: 5′-TTTGGTTCAAAATGGCATCA-3′) were designed. The 40-μL PCR reaction mixture contained 0.5 μL of overnight cultures in 1 × buffer, 2.5 mM MgCl2, dNTPs and 0.04 U Taq DNA polymerase (each 0.5 mM; Promega, Madison, WI), and 0.25 μM primers (MajorBio, Shanghai, China). The PCR reaction was performed using a Veriti 96 Well Thermal Cycler (Applied Biosystems, Waltham, MA) using the following amplification protocol: initial denaturation at 95°C for 5 min; 30 cycles of 95°C for 45 s, 52°C for 45 s, and 72°C for 30 s; and a final extension at 72°C for 5 min. The PCR amplicons were loaded onto a 1.5% (wt/vol) agar gel pre-stained with SyBr Safe DNA dye (Invitrogen, Carlsbad, CA) for direct visualization after electrophoresis for 60 min at 100 V. Molecular ladders of 100 bp (New England Biolabs, Ipswich, MA) were used to estimate the size of the PCR amplicons (approximately 224 bp). Subsequently, PCR amplicons were subjected to gel purification, and Sanger sequencing analysis was performed by using ATPase-F primer (MajorBio).
) to confirm the identity of the gene sequences by comparison with type strains available at the National Center for Biotechnology Information (NCBI, Bethesda, MD). The ATPase gene sequences of DGCC12411 and DGCC12411m were aligned and compared by ClustalW (https://embnet.vital-it.ch/software/ClustalW.html), using L. plantarum WCFS1 (YP_004890068.1) as reference.
Stability Assessment of DGCC12411m Strain
Stability of the LPA phenotype and the F0F1-ATPase SNP mutation of DGCC12411m over 10 passages were assessed. With the 10% (vol/vol) inoculation rate, we estimated that 3.3 generations occurred in each passage, a total of 33 generations throughout 10 passages. The pH after 48-h cultivation (pH48h) was used as the post-fermentation acidification indicator. Three colonies of DGCC12411m were randomly chosen and inoculated from the MRS agar plate into 10 mL tubes of MRS broth, respectively, and incubated at 30°C for 24 h. Then, 1 mL of fresh culture from each tube was inoculated into a new tube containing 9 mL of fresh MRS medium and incubated at 30°C for 24 h. After 24-h incubation, these cultures were designated as passage 1 (P1) and subsequently used as the inoculum for passage 2 (P2). The initial culture used to inoculate P1 was further incubated at 30°C for another 24 h for the pH measurement (Mettler Toledo pH meter, Greifensee, Switzerland) after 48 h of incubation. The same process was repeated for another 9 consecutive passages (P2 to P10). The mother strain DGCC12411 was used as a control. The F0F1-ATPase α-subunit gene of each passage was sequenced, and 100 μL of the 24-h broth culture of each passage was serially diluted and plated on MRS plates. At each passage, 10 colonies per sample were selected for further post-fermentation acidification and ATPase gene analysis.
Effects of Inoculum Size and Potassium Phosphate Buffer on Phenotypic Stability
Effects of inoculum size on phenotypic stability were measured by inoculating 2 or 10% (vol/vol) DGCC12411m into MRS broth for 10 passages, using the method previously described. With the 2% (vol/vol) inoculation rate, we estimated that 5.6 generations occurred in each passage, hence a total of 56 generations throughout 10 passages. We used MRS supplemented with neomycin sulfate (600 µg/mL) as a positive control. The effect of potassium phosphate buffer was assessed by comparing phenotype stability between the untreated control and MRS medium supplemented with 0.05 or 0.10 M potassium phosphate buffer (7.214 g of sodium phosphate dibasic heptahydrate and 10.086 g of sodium phosphate monobasic monohydrate in 1 L of MRS medium; pH adjusted to 6.3). Three individual replicates were performed for each test.
Statistical Analysis
Data were statistically analyzed with Minitab software (version 18; Minitab Inc., State College, PA). One-way ANOVA (Tukey test) was used to study the significant differences between means, with significant difference at P < 0.05 or P < 0.01.
Sequence Deposition
The F0F1-ATPase gene nucleotide sequences of DGCC12411 and DGCC12411m were deposited in GenBank under accession numbers MN886309 and MN886310.
RESULTS AND DISCUSSION
LPA L. plantarum Variant and the Putative Causal Mutation
Lactobacillus plantarum DGCC12411 is of high industrial interest, as it is an exopolysaccharide-producing bacterium that contributes to high thickness and ropiness texture for fermented food. However, controlling its post-fermentation acidification during food fermentation is challenging. To reduce its post-fermentation acidification, a spontaneous mutation and a selective screening method were adopted to generate natural LPA variants. After screening of 2,000 isolates, 8 variants that could not grow in acidified MRS broth (pH 4.5) were obtained. By monitoring their acidification in normal MRS (pH 6.3), we found that 7 variants were slow in acidification but stabilized at the same pH as the mother strain after 60-h fermentation (data not shown). Only DGCC12411m with pH 4.21 at 24 h was selected as an LPA variant. At extended fermentation to 60 h, DGCC12411m stopped at pH 4.09, which is 0.35 pH unit higher than the mother strain DGCC12411 (Figure 1). The pH of both DGCC12411 and DGCC12411m remained stable for 36 to 60 h of culture; thus we selected the pH at 48 h, a central point, as an end pH to compare the post-fermentation acidification of DGCC12411 and its variants in further tests.
Figure 1Acidification curves of Lactobacillus plantarum DGCC12411 (dashed line) and its acid-sensitive mutant strain (DGCC12411m; solid line) in De Man, Rogosa, and Sharpe broth. Each treatment has duplicates.
The industry is looking for cultures for use in ambient yogurt products that possess LPA effects during ambient-temperature storage. However, we also noticed that DGCC12411m was slower than its mother strain in fermentation (Figure 1). This is a disadvantage for starter cultures in applications that require fast fermentation but an advantage for adjunct cultures in foods, especially those stored at ambient temperature, when it is desirable to minimize the strain's acidification activity during shelf-life for better product stability. In this study, after inoculation of 1 × 107 cfu/mL DGCC12411m or DGCC12411 into the ambient yogurt and storage at 25°C, a significant drop in pH mainly occurred during the first 2 mo; then the pH of the yogurt became stable for the yogurts inoculated with either strain. Consistent with the acidification results obtained using MRS, DGCC12411m also showed better (P < 0.01) post-fermentation acidification control in the ambient yogurt, with a ΔpH of 0.2 pH unit higher than the parent at d 30 and 0.12 at d 120 (Figure 2A). Changes in the taste of the yogurt inoculated with DGCC12411m appeared more acceptable than with DGCC12411 due to its lower acidity (data not shown). The cell count of DGCC12411m in ambient yogurt remained at or slightly higher than 1 × 107 cfu/mL for up to 60 d, and only 1 log reduction in cell count was observed on d 120. Our results ruled out the possibility that the improvement in post-acidification control was due to reduction in cell count during shelf-life.
Figure 2(A) pH and (B) cell count of Lactobacillus plantarum DGCC12411 (dashed line) and its acid-sensitive mutant DGCC12411m (solid line) in the model of yogurt at ambient temperature during shelf-life storage (25°C for 120 d). Error bars represent SD of 3 replicates.
To identify possible mutations responsible for the LPA phenotype, the whole genome of DGCC12411m was sequenced and compared with its mother strain (data not shown). An SNP mutation (GGT > GAT at positions 505 to 507) in the F0F1-ATPase α-subunit was detected in the acid-sensitive mutant. After BLAST analysis of the DGCC12411m ATPase gene against the whole-genome sequences of other L. plantarum strains in GenBank, we found that this region is extremely conserved (Figure 3). In a previous study, the mutations in 3 LPA L. plantarum mutants that were generated using a similar method occurred on the F0F1-ATPase β-subunit (
also reported that mutations that occurred in or around the subunit-subunit interaction interface resulted in a profound reduction of ATPase activity, but mutations that took place outside or far from the motif did not show this effect.
have also reported on the generation of acid-sensitive L. plantarum strains, but the mutations were not identified in their study. In DGCC12411m, the mutation from G to A at position 506 resulted in a change from a glycine residue at position 169 to an aspartic acid residue. Glycine-169 is located as the first glycine residue at the Walker A motif (also known as a phosphate-binding loop or a P-loop), a glycine-rich loop that commonly presents in the α- and β-subunits of F1-ATPase, myosin, and other ATP-utilizing enzymes (
Distantly related sequences in the alpha- and beta-subunits of ATP synthase, myosin, kinases and other ATP-requiring enzymes and a common nucleotide binding fold.
). The P-loop consists of a sequence of 8 amino acids [GXXXXGKT(S), G: glycine; K: lysine; T: threonine; S: serine; X: any amino acid], with the first glycine residue highly conserved in enzymes containing this motif across archaea, prokaryotes, and eukaryotes (
demonstrated that the substitution of the first glycine residue with alanine in the catalytic A-subunit of A-ATPase of Pyrococcus horikoshii OT3 resulted in flattening of the P-loop conformation, which inevitably closed the active site region and hindered nucleotide binding.
Figure 3Alignment of the complete gene sequences of F0F1-ATPase α-subunit of Lactobacillus plantarum DGCC12411 and its mutant strain, DGCC12411m, with other L. plantarum strains deposited in the National Center for Biotechnology Information (NCBI, Bethesda, MD) with 99.87% homology: L. plantarum WCFS1 (NC_004567.2), L. plantarum DR7 (CP031318.1), L. plantarum JDM1 (CP001617.1), L. plantarum P-8 (CP005942.2), and L. plantarum ST-III (CP002222.1). Site of SNP mutation (position 506) marked with an asterisk (*).
Due to the different growth environments and selective pressures between laboratory experiments and commercial-scale production, phenotype stability is one of the major concerns in the application of mutants in industrial production, especially with mutations in the enzymes that are core to bacterial activity. Despite F0F1-ATPase-based LPA LAB mutants being developed and reported in several previous studies (
), their stability has rarely been assessed. A previous study reported on the stability of an uncharacterized acid-sensitive L. delbrueckii ssp. bulgaricus mutant (
). During continuous passage in milk, they observed a continuous reduction in the final pH by the mutant, without characterizing the genetic changes that contributed to the LPA phenotype nor the cause of the phenotype instability.
In this study, we found that the back mutation from aspartic acid residue to glycine reverses the post-fermentation acidification phenotype. Therefore, we believe the single amino acid residue change at this conserved motif of F0F1-ATPase α-subunit caused the LPA phenotype of DGCC12411m. To verify our postulation, the stability of the SNP mutation as well as the LPA phenotype of DGCC12411m were assessed across 10 passages [estimated 33 generations, 10% (vol/vol) inoculation], which is about 50% more than standard industrial practice (about 22 generations). Figure 4A shows that the pH48h of DGCC12411m decreased gradually from pH 4.15 ± 0.00 at P1 to 3.94 ± 0.05 at P5 (about 17 generations) and drastically dropped to 3.77 ± 0.03 (SD) at P6 (about 20 generations), which was similar to results for the mother strain, DGCC12411. Then, the pH48h remained stable until P10. Such low stability in a phenotype could cause problems during strain preservation and production. To further understand the post-fermentation acidification evolution in this population, 30 single colonies were isolated from each passage, and their post-fermentation acidification phenotype was determined. Figure 5 shows that the pH48h of all colonies from P1 to P4 remained stable and was not significantly different than that of DGCC12411m. At P5 (about 17 generations), variants emerged with significantly (P < 0.01) lower end pH (pH48h < 4.0), which accounted for 43% (13 out of 30) of the colonies. These low-pH48h colonies became dominant from P6 to P10 (around 20 to 33 generations) in all 3 replicates. This result showed a clear evolution of the dominant population from LPA to a normal post-fermentation acidification phenotype.
Figure 4Stability of the low post-fermentation acidification phenotype and genotype during passaging. (A) pH of the 48-h culture of acid-sensitive mutant Lactobacillus plantarum DGCC12411m (•, solid line) and L. plantarum DGCC12411 (○, dashed line) in fresh De Man, Rogosa, and Sharpe medium for 10 passages. Asterisks (*) denote significant differences (P < 0.01) between pH of DGCC12411m and its mother strain at different passages. Inoculation rate was 10% (vol/vol). Error bars represent SD of 3 replicates. (B) Alignment of the partial gene sequences of F0F1-ATPase α-subunit of DGCC12411m broth culture (pooled from 3 replicates) from P1 to P10. Site of SNP mutation (position 506) marked with #; the 3 codons inside the frame encoded for glycine-aspartic acid-169.
Figure 5Stability of the low post-fermentation acidification phenotypes of the acid-sensitive mutant Lactobacillus plantarum DGCC12411m over 10 passages. A = replicate 1; B = replicate 2; C = replicate 3. • represents colonies with GAT at positions 505 to 507; ○ represents colonies with GGT at positions 505 to 507. Inoculation rate was 10% (vol/vol). pH48h = pH after 48 h of cultivation.
). These revertants might be due to the compensatory changes from one or more mutations in a different gene or genes, one or more regulatory signals within the original gene but at a site other than the causal mutation, or a true reversion that restored the original amino acid sequence (
). To understand whether the LPA phenotype degeneration was caused by back mutation of GAT to GGT, the F0F1-ATPase α-subunit genes of the population (pooled from 3 replicates) from each passage were analyzed by Sanger sequencing. We noticed that an SNP back mutation appeared at P5 (about 17 generations) and became dominant after P7 (about 23 generations; Figure 4B). To further understand whether the SNP back mutation is the only cause of the LPA phenotype degeneration, the F0F1-ATPase α-subunit of these 300 isolates were sequenced (Figure 5). All the colonies isolated from P1 to P4 (up to 13 generations) were GAT at positions 505 to 507. In P5 (about 17 generations), 3 GGT colonies were detected, which accounted for 10% of the isolates. These GGT colonies indicated that a true reversion occurred and restored the original amino acid sequence. The proportion of GGT colonies increased to 100% after P6 (about 20 generations) in replicate 3 (Figure 5C). However, we also noticed that not all low-pH48h colonies were GGT. The GGT colonies accounted for only 48.7% (79 out of 162) of the low-pH48h colonies. These GAT colonies with low pH48h suggested that, beyond true reversion, other mechanisms were involved that compensated for its acid tolerance.
have reported that a continuous decrease in the pH of variants isolated from each passage was caused by a phenotypic adaptation rather than by genetic reversion. However, other enzymes, such as glutamate decarboxylase and arginine deiminase, also play important roles in pH homeostasis (
The effect of inoculum size and incubation temperature on cell growth, acid production and curd formation during milk fermentation by Lactobacillus plantarum Dad 13.
). With a higher inoculation level, the probability of transferring the revertant to the next passage will increase. In addition, it will also shorten the time to stationary phase, resulting in a longer exposure time at lower pH for bacteria cultivated for 24 h. To assess the effects of inoculation ratio on end pH during passaging, we compared the pH48h of DGCC12411m with 10% and 2% inoculum from P1 to P10; these inocula are estimated to grow about 3.3 and 5.6 generations per passage, respectively. We used MRS supplemented with neomycin sulfate (600 µg/mL) as a reference. As expected, with the strong selective pressure of neomycin, the LPA phenotype of DGCC12411m was extended up to P9. Compared with the 10% inoculation rate, for which pH48h started to drop after P4 (about 13 generations), the pH48h of DGCC12411m inoculated at 2% was significantly (P < 0.05) higher from P5 to P7 (28 to 40 generations), which demonstrated an improvement of LPA stability (Figure 6A).
Figure 6Effects of (A) inoculation size and (B) potassium phosphate buffer on the stability of the low post-fermentation acidification phenotype (as measured by final pH at 48 h) of the acid-sensitive mutant of Lactobacillus plantarum DGCC12411, designated as DGCC12411m. • (solid line) = DGCC12411m in De Man, Rogosa, and Sharpe medium (MRS) supplemented with neomycin sulfate (600 µg/mL, positive control); • (dashed line) = DGCC12411 in MRS (negative control); ▴ = 10% (vol/vol) DGCC12411m in MRS; ○ = 2% (vol/vol) DGCC12411m in MRS (untreated control for panel B); ▪ = DGCC12411m in MRS supplemented with 0.1 M potassium phosphate buffer; □ = DGCC12411m in MRS supplemented with 0.05 M potassium phosphate buffer. Error bars represent SD of 3 replicates. Asterisks (*) denote significant differences (P < 0.05) in pH between 10% (vol/vol) and 2% (vol/vol) DGCC12411m in MRS. Hash sign (#) denotes significant difference (P < 0.05) in pH between 0.1 M potassium phosphate buffer and untreated control.
In addition, prolonged acid stress exposure is generally more favorable for the selection of variants that can grow to a lower pH, which may be adverse for the maintenance of the LPA phenotype. Therefore, we speculated that relief of H+ stress during pH homeostasis by addition of buffering salt in the growth medium may help extend the stability. Previous study has also reported on supplementing buffering salts (such as potassium phosphate) as a pH stabilizer during fermentation (
). Adding 0.05 M potassium phosphate buffer into the medium slightly improved the pH48h stability of DGCC12411m (Figure 6B). Greater improvement in the stability of DGCC12411m was observed when we supplemented 0.1 M potassium phosphate buffer to MRS, even though significant (P < 0.05) improvement compared with the untreated control was only observed at P10. The pH48h stability of DGCC12411m can be maintained at the same level up to P8 (about 45 generations), which is twice the number of generations for strain scale-up in industrial production practices.
CONCLUSIONS
In this study, an acid-sensitive L. plantarum mutant was developed, which showed improved post-fermentation acidification under laboratory conditions, as well as improved shelf-life pH stability in the ambient yogurt application model for up to 4 mo. We identified that a novel SNP mutation (which resulted in the replacement of a highly conserved glycine residue by aspartic acid) in the F0F1-ATPase α-subunit led to the LPA phenotype. We also demonstrated that the mutant developed in the laboratory using strong selective pressure could have issues in phenotype and genotype stability during industrial-scale production, as back mutation occurred after 5 passages (about 17 generations). Stabilizing the mutation is of utmost importance if the strain is to be used for future industry application, as degeneration of the LPA phenotype will negatively affect the benefits of the desired phenotype in commercial fermented dairy products, especially those stored at ambient temperature. In view of this, a thorough assessment of genotype and phenotype stability is of utmost importance during strain functionality improvement. Moreover, we also developed methods to stabilize the LPA phenotype during passaging, which may enable the industrial application of such mutants. This study provided a practical way for developing milder starter cultures for industrial application.
ACKNOWLEDGMENTS
This work was funded by DuPont Nutrition and Biosciences, Shanghai, China. The authors thank Julia Salabai (intern from Erhvervsakademi Aarhus, Aarhus, Denmark) and Meng Chen (technician, DuPont Nutrition and Biosciences) for their contribution to the experiments, and Dennis Romero (technical fellow, DuPont Nutrition and Biosciences) and Cindy Stewart (Global Technology and Innovation Leader, DuPont Nutrition and Biosciences) for their support and comments on the manuscript. The authors have not stated any conflicts of interest.
REFERENCES
Bernardeau M.
Guguen M.
Vernoux J.P.
Beneficial lactobacilli in food and feed: Long-term use, biodiversity and proposals for specific and realistic safety assessments.
Potential probiotic characterization of Lactobacillus reuteri from traditional Chinese highland barley wine and application for room-temperature-storage drinkable yogurt.
Streptococcus thermophilus strain exhibiting unusual acidification kinetics in milk. P. 106 in 12th International Symposium of Lactic Acid Bacteria Abstract Book.
Rheological and physical characteristics of non-fat set yogurt prepared with EPS-producing Streptococcus thermophilus and an H+-ATPase-defective mutant Lactobacillus delbrueckii ssp. bulgaricus..
Probiotic strain Lactobacillus plantarum NBIMCC 2415 with antioxidant activity as a starter culture in the production of dried fermented meat products.
Yoghurt fermented by Lactobacillus delbrueckii ssp. bulgaricus H+-ATPase-defective mutants exhibits enhanced viability of Bifidobacterium breve during storage.
Int. J. Food Microbiol.2007; 116 (17434219): 358-366
Distantly related sequences in the alpha- and beta-subunits of ATP synthase, myosin, kinases and other ATP-requiring enzymes and a common nucleotide binding fold.
The effect of inoculum size and incubation temperature on cell growth, acid production and curd formation during milk fermentation by Lactobacillus plantarum Dad 13.
30507-5&id=gr1.jpg){kind=link}
30507-5&id=gr2.jpg){kind=link}
30507-5&id=gr3.jpg){kind=link}
30507-5&id=gr4.jpg){kind=link}
30507-5&id=gr5.jpg){kind=link}
30507-5&id=gr6.jpg){kind=link}