This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental material
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Plumed-Ferrer, C.
Right arrow Articles by von Wright, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Plumed-Ferrer, C.
Right arrow Articles by von Wright, A.
Agricola
Right arrow Articles by Plumed-Ferrer, C.
Right arrow Articles by von Wright, A.

 Previous Article  |  Next Article 

Applied and Environmental Microbiology, September 2008, p. 5349-5358, Vol. 74, No. 17
0099-2240/08/$08.00+0     doi:10.1128/AEM.00324-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Comparative Study of Sugar Fermentation and Protein Expression Patterns of Two Lactobacillus plantarum Strains Grown in Three Different Media{triangledown} ,{dagger}

Carme Plumed-Ferrer,1* Kaisa M. Koistinen,1,{ddagger} Tiina L. Tolonen,2 Satu J. Lehesranta,1,§ Sirpa O. Kärenlampi,1 Elina Mäkimattila,3 Vesa Joutsjoki,3 Vesa Virtanen,2 and Atte von Wright1

Applied Biotechnology, Department of Biosciences, University of Kuopio, FIN-70211 Kuopio, Finland,1 Laboratory of Biotechnology, University of Oulu, FIN-88600 Sotkamo, Finland,2 Biotechnology and Food Research, MTT Agrifood Research Finland, FIN-31600 Jokioinen, Finland3

Received 7 February 2008/ Accepted 12 June 2008


arrow
ABSTRACT
 
A comparative study of two strains of Lactobacillus plantarum (REB1 and MLBPL1) grown in commercial medium (MRS broth), cucumber juice, and liquid pig feed was performed to explore changes to the metabolic pathways of these bacteria, using a proteomics approach (two-dimensional electrophoresis and liquid chromatography-tandem mass spectrometry) combined with analyses of fermentable sugars and fermentation end products. The protein expression showed that even with an excess of glucose in all media, both strains could metabolize different carbohydrates simultaneously and that hexoses could also be used via a phosphoketolase pathway with preferential expression in liquid feed. Sugar analyses showed that the fermentation of sugars was homolactic for all media, with some heterolactic activity in liquid feed, as shown by the production of acetate. Cucumber juice (the medium with the highest glucose content) showed the lowest hexose consumption (10%), followed by liquid feed (33%) and MRS broth (50%). However, bacterial growth was significantly higher in cucumber juice and liquid feed than in MRS broth. This discrepancy was due to the growth benefit obtained from the utilization of the malate present in cucumber juice and liquid feed. Despite different growth conditions, the synthesis of essential cellular components and the stress response of the bacteria were unaffected. This study has improved our understanding of the mechanisms involved in the growth performance of an appropriate lactic acid bacterium strain to be used for food and feed fermentation, information that is of crucial importance to obtain a high-quality fermented product.


arrow
INTRODUCTION
 
Lactic acid bacteria (LAB) represent a group of microorganisms that are functionally related by their ability to produce lactic acid during fermentation. Acidification due to lactic acid production and other enzymatic processes contribute to the flavor, texture, and keeping qualities of fermented food and feed products (2, 14). Understanding the mechanisms involved in the growth performance of an appropriate LAB strain to be used for food and feed fermentations is of crucial importance to obtaining high-quality fermented product.

Lactobacillus plantarum is a facultative heterofermentative LAB encountered in a large number of environmental niches, from dairy and meat products to a variety of vegetable and plant fermentations (8). It is also one of the Lactobacillus species present in the human intestinal tract (1). Due to its high acid tolerance, L. plantarum has been used as the starter culture of food and feedstuffs (16, 24), but it has also been used as a probiotic (6, 12) and as a delivery vehicle for therapeutic compounds (14, 18) due to its ability to survive and persist in the gastrointestinal tract. Therefore, studies of the adaptation of L. plantarum to different environmental niches are of great interest due to the extensive use of this bacterium in food and feed fermentations.

The two strains of L. plantarum used in this study (REB1 and MLBPL1 strains) are used in starter cultures for sauerkraut, cucumber juice, and liquid pig feed fermentations (16, 20). In our previous study of the growth phase-dependent proteomes of L. plantarum strains REB1 and MLBPL1 grown in MRS broth, the growth phase-dependent preferential expression of the different metabolic pathways was elucidated (9). A similar study has also been performed with L. plantarum WCFS1 grown in MRS broth (4). Summing up the general findings of these two studies, the bacteria created a pool of acetyl phosphate and nucleotides during the lag phase and activated a stress response; during the early exponential phase, the bacteria showed preferential expression of proteins for the simultaneous use of different carbohydrates; during the late exponential and early stationary phases, the bacteria showed preferential expression for enzymes involved in the biosynthesis of proteins and cell division pathways and for membrane proteins; and during the late stationary phase, preferential expression was shown for numerous stress proteins (4, 9). The indicated simultaneous use of different carbohydrates in the early exponential phase was very interesting because these results contradict the standard view of carbohydrate metabolism in LAB, according to which, glucose represses the proteins associated with the metabolism of other sugars (8, 22).

Previous proteomic studies of L. plantarum have been performed using MRS broth, the standard rich medium used for laboratory cultivations. However, MRS broth cannot be considered a representative medium from which conclusions can be made about gene expression under actual conditions of applications of L. plantarum to food and feed fermentations. In this study, the protein expression patterns of L. plantarum strains REB1 and MLBPL1 were investigated with media (cucumber juice and liquid pig feed) that are relevant to the practical fermentation processes in which these strains are actually used as starters, and the proteomes were compared to those seen after growth in MRS medium. The aim was to find out the application-specific proteomic patterns and whether strains with different backgrounds displayed any specific adaptation to the respective fermentation media.


arrow
MATERIALS AND METHODS
 
Bacteria and their cultivation.
The L. plantarum strain REB1 was isolated from a spontaneously fermented cereal-based feed (17), and the L. plantarum strain MLBPL1 was isolated from a spontaneously fermented white cabbage (20). A 2% overnight aerobic culture in MRS broth (Lab M, Lancashire, United Kingdom) was used to inoculate the experimental media (MRS broth, cucumber juice, and liquid pig feed). Cultures (25 ml) from the MRS broth and cucumber juice and 50-ml cultures from liquid pig feed were aerobically incubated at 30°C with slow agitation. Five independent cultures were made from each strain and medium. Samples were taken after 8 h of incubation, corresponding to the exponential growth phase of all cultures, as shown by the growth behavior of both strains (Fig. 1). The bacterial count of each sample at 8 h was determined by plating on MRS agar (Lab M, Lancashire, United Kingdom). Liquid pig feed and cucumber juice samples were subsequently filtered (Nylon filter; 11-µm pore size; Millipore, Billerica, MA). Bacteria were pelleted (3,000 x g for 5 min at 4°C) and frozen immediately in liquid N2. Samples were stored at –70°C and used later for proteomics and sugar analyses.


Figure 1
View larger version (13K):
[in this window]
[in a new window]

  		FIG. 1. Growth curves (log CFU ml–1) of L. plantarum strains REB1 and MLBPL1 in MRS broth, cucumber juice, and liquid feed.

Experimental media.
The experimental media used in this study were MRS broth, cucumber juice, and liquid pig feed. Approximately 400 ml of cucumber juice was obtained by homogenizing and centrifuging one Finnish cucumber (500 g). Cucumber juice was inoculated with 2% glucose and subsequently autoclaved at 121°C for 5 min. Liquid pig feed was prepared by mixing the dry feed with three parts water. The dry feed was formulated as described by Plumed-Ferrer et al. (16). Briefly, dry feed contained 69% barley, 12.8% soybean meal, 18% concentrated liquid whey, and 0.2% mineral vitamin premix.

Extraction of soluble proteins.
Extraction of soluble proteins was performed as described by Koistinen et al. (9). Briefly, the frozen bacterial pellets were washed and lysed with lysozyme at 37°C for 45 min. Subsequently, DNase treatment was performed, and after the lysate was centrifuged, the supernatants were collected, and the proteins were precipitated in ice-cold acetone with trichloroacetic acid. After the protein pellets were washed with ice-cold acetone, they were dried and stored at –70°C.

Two-dimensional electrophoresis.
One two-dimensional electrophoresis (2-DE) gel was run for each biological replicate. The 2-DE was performed as described by Koistinen et al. (9). Briefly, the proteins were resuspended and solubilized, and the protein concentrations were calculated using a protein assay dye reagent (Bio-Rad, Hercules, CA). For isoelectric focusing, 24-cm immobilized pH gradient (IPG) strips with a linear pH range of 4 to 7 (Amersham Biosciences, Uppsala, Sweden) were used. The focusing was done in an Ettan IPGPhor system (Amersham Biosciences), and the strips were subsequently stored at –70°C. The sodium dodecyl sulfate-polyacrylamide gel was run overnight with 19- by 23-cm homogeneous 12% gels in a Hoefer DALT system (Amersham Biosciences). The gels were stained with SYPRO Ruby (Bio-Rad), and gel images were acquired with an FLA-3000 fluorescent image analyzer (Fuji Photo Film Co. Ltd., Tokyo, Japan).

Gel image analysis and statistical analysis.
Image analyses were performed with PDQuest software (Bio-Rad). Spot intensities were normalized to the total intensity of valid spots (which, by default, expresses normalized values as parts per million) to minimize errors caused by differences among the amounts of proteins and staining intensities. In the case of strain REB1, the majority of spots with normalized quantity values of ≤7.1389 had a quality score of 0, as given by PDQuest, and spots were considered background noise, while spots with normalized quantity values of >7.1389 tended to have quality scores of >0. For data derived from the MLBPL1 strain, spots with normalized quantity values of ≤3.4053 were set to 0. Differences among spot intensities from the different growth media were analyzed statistically with the nonparametric Kruskal-Wallis test and principal component analysis (PCA), using SPSS software (SPSS, Chicago, IL). The criteria for differential expression were as follows: protein expression levels in different media were first compared using the Kruskal-Wallis test. When there was an indication of a statistically significant (P < 0.01) difference, the expression patterns were checked visually to observe in which medium the spot intensity differed from the others. Furthermore, PCA was used to assess whether the samples of bacteria grown in different media would separate from each other, based on their protein expression patterns, and to identify groups of proteins responsible for the differences observed between media. Spots with at least five non-0 values were included in PCA. For PCA, the quantity values were transformed to square roots to normalize the data. Moreover, one-way analysis of variance and Tukey's honestly significant difference post hoc test were used to analyze the differences between the bacterial growth in each medium, using SPSS software.

Mass spectrometry identification of proteins.
SYPRO Ruby-stained 2-DE gels were restained with silver, using a mass spectrometry-compatible method (19), with the modification that the gels were washed first in 40% ethanol and 5% acetic acid and then washed twice in water for 30 min. Matches with confidence intervals of 95% were selected, and each match was verified manually to increase the reliability of the identification. Tryptic digestion of the protein spots excised from the gels and sample preparation was performed according to the method described by Koistinen et al. (10).

The tryptic peptides were analyzed with one of two different liquid chromatography-tandem mass spectrometry systems, as described by Lehesranta et al. (11). The peptides were identified against the L. plantarum genome (8) or the complete proteome in UniProtKB or the NCBI nonredundant protein database or the known proteins from lactobacilli. Matches between tandem mass spectrometry spectra and sequences in the databases were verified manually.

Sugar, organic acids, and ethanol analyses.
Sugars and organic acids (three replicates) were determined with a high-performance liquid chromatography (1100 series machine; Agilent Technologies, Santa Clara, CA) system equipped with a diode array detector and a refractive index detector. For the analysis of disaccharides, thawed samples were centrifuged (13,000 x g; 10 min; 20°C), filtered through a 0.45-µm syringe filter (Palm Gelman Laboratory, New York, NY), and analyzed with a Zorbax carbohydrate column (250-mm by 4.6-mm inside diameter [ID]; 5 µm; Agilent Technologies), and eluted with 1.4 ml min–1 of 80% acetonitrile. We analyzed monosaccharides and organic acids by mixing aliquots of 500 µl of thawed samples with 20 µl of 7% sulfuric acid and then diluting the samples twofold with distilled water. The mixtures were subsequently centrifuged (13,000 x g; 10 min; 20°C), filtered through a 0.45-µm syringe filter (Palm Gelman Laboratory), and analyzed with a Rezex RHM-monosaccharide H+ column (8%) (300-mm by 7.8-mm ID; 8 µm; Phenomenex Inc., Torrance, CA), eluted with 0.4 ml min–1 of 0.5 mM sulfuric acid. Standard solutions of glucose, fructose, galactose, lactose, sucrose, lactic acid, malic acid, and acetic acid were used as external standards for quantification.

Gas chromatography (HP6890GC model; Agilent Technologies) coupled to a mass-selective detector (HP5973 machine; Agilent Technologies) was used to determine the amount of ethanol in the culture samples. Separation of ethanol from the other compounds was achieved with an HP-FFAP column (25-mm by 0.20-mm ID; 0.30 µm; Agilent Technologies) and helium as the carrier gas. Samples (1 µl) were injected using the split mode (200:1) at 250°C, and compounds were separated using the following temperature gradient program: 60°C for 2 min, 15°C/min to 200°C, and postrun at 220°C for 2 min. Quantification was accomplished using the selected ion-monitoring mode, with ethanol as the external standard.


arrow
RESULTS
 
Growth of L. plantarum REB1 and MLBPL1 in different media.
The colony counts for L. plantarum REB1 and MLBPL1 in MRS broth, cucumber juice, and liquid feed were calculated at the moment the bacteria were collected (Fig. 2). The results showed that the colony counts for both strains were significantly (P < 0.01 or P < 0.05, respectively) higher in cucumber juice and liquid feed than in MRS broth (Fig. 2). The cell counts for the REB1 strain were slightly but not significantly higher than that of the MLBPL1 strain (P > 0.05).


Figure 2
View larger version (21K):
[in this window]
[in a new window]

  		FIG. 2. Bacterial growth (log CFU ml–1) after 8 h of incubation. MRS, MRS broth; Cu, cucumber juice; Feed, liquid pig feed. Groups with different letters differ significantly: a, P < 0.01; b, P < 0.05.

Protein expression patterns of L. plantarum strains REB1 and MLBPL1.
Five biological replicates of each strain and medium were analyzed, and one 2-DE gel was run for each biological replicate. The representative 2-DE gels of the two L. plantarum strains are shown in Fig. SM1 and SM2 in the supplemental material; and the numbers shown correspond to the identified proteins listed in Tables 1 and 2, as well as in Tables SM3 to SM5 in the supplemental material. The number of protein spots quantified from the gels was 924 (REB1 strain) and 905 (MLBPL1). A total of 232 proteins were identified. Of these proteins, 80 and 55 protein spots from REB1 and MLBPL1, respectively, showed statistically significant differences between media (P < 0.01). Some additional spots, which did not show statistically significant differences (P > 0.01) between media, were identified because of their high intensities. In a few cases, different spots from the same L. plantarum strain were identified as the same protein, suggesting the presence of different isoforms. Several protein spots yielded more than one protein identification with high confidence, indicating the presence of comigrating proteins or proteins sharing the same sequence (although the latter was not the case in this experiment). These proteins were excluded because the quantitative expression of the two comigrating proteins could not be measured reliably. Classification of the proteins was done according to the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (7) and to LacplantCyc, a L. plantarum pathway/genome database used as a reference for LAB (21).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Identified proteins with the most statistically significant differences in their expression in L. plantarum REB1a


View this table:
[in this window]
[in a new window]

  		TABLE 2. Identified proteins with the most statistically significant differences in their expression in L. plantarum MLBPL1a

PCA was used to identify groups of proteins responsible for the differences observed between media. Samples of bacteria grown in the three media could be separated very clearly, with some differences shown between the two strains (Fig. 3). Several proteins that had a high positive or negative loading in the PCs, i.e., those that contributed to the differences seen for the components, were identified. For the REB1 strain, samples grown in liquid feed separated from those in the MRS broth and cucumber juice in the first component (PC1), whereas all three media separated from each other in the second component (PC2). The protein spots with high positive loading in the PC1 tended to be highly expressed (e.g., R049), and spots with high negative loadings tended to be expressed at low levels in liquid feed (e.g., R064) (Table 1; see also Table SM3 in the supplemental material). For the MLBPL1 strain, all three media separated from each other in the PC1, whereas samples grown in cucumber juice separated from those grown in MRS broth and liquid feed samples in the PC2. In this case, the protein spots with high positive loading in the PC2 tended to be expressed at low levels (e.g., M121), and spots with high negative loading tended to be highly expressed in cucumber juice (e.g., M089) (Table 2; see also Table SM4 in the supplemental material). The two components accounted for 53.4% and 56.2% of the variation between the growth media used for the REB1 and MLBPL1 strains, respectively.


Figure 3
View larger version (9K):
[in this window]
[in a new window]

  		FIG. 3. Principal component (PC) scores for the L. plantarum strains REB1 (a) and MLBPL1 (b) grown in MRS broth ({square}), cucumber juice (•), and liquid pig feed ({blacktriangleup}). Values in parentheses indicate the percentages of total variation accounted for by each principal component.

Proteins differentially expressed between the media and strain.
Eighty protein spots from REB1 and 55 spots from MLBPL1 showed statistically significant differences in protein expression between media (P < 0.01). Tables 1 and 2 show the most statistically significant differences, and the complete list can be found in Tables SM3 and SM4 in the supplemental material. Data showed that in MRS broth, the identified proteins involved in glycolysis and pyruvate metabolism generally showed the highest expression, except for the enzyme pyruvate kinase proteins, which had the lowest expression in this medium (Fig. 4). The proteins for the de novo synthesis of purines and pyrimidines showed high expression in MRS broth, as well as the enzyme adenine phosphoribosyltransferase, involved in the salvage pathway of purines (Fig. 5a).


Figure 4
View larger version (33K):
[in this window]
[in a new window]

  		FIG. 4. Carbohydrate pathways used by L. plantarum strains REB1 and MLBPL1. Expression of the identified enzymes that catalyze the different steps is indicated by white (MRS), gray (cucumber), and black (feed) bars. The strain and statistical significance of each protein pattern is indicated by R (REB1) or M (MLBPL1) and by black (P < 0.01) or gray (P > 0.01). Abbreviations: Ack1, acetate kinase; Als, acetolactate synthase; EnoA1, enolase 1; Fba, fructose bisphosphate aldolase; FruK, 1-phosphofructokinase; GalU, UTP-glucose-1-phosphate uridylyltransferase; GapB, glyceraldehyde-3-phosphate dehydrogenase; Glk, glucokinase; Gnd2, phosphogluconate dehydrogenase; Gpd, Glucose-6-phosphate 1-dehydrogenase; LdhD, D-lactate dehydrogenase; LdhL1, L-lactate dehydrogenase; Lox, lactate oxidase; LoxD, lactate oxidase (oxidoreductase); Mae, malic enzyme; Map, maltose phosphorylase; MleS, malolactic enzyme; MtlD, mannitol-1-phosphate 5-dehydrogenase; Pfk, 6-phosphofructokinase; Pgi, glucose-6-phosphate isomerase; Pgk, phosphoglycerate kinase; Pgm, phosphoglucomutase; Pmg9, phosphoglyceromutase 2; Pmi, mannose-6-phosphate isomerase; Pox2, pyruvate oxidase; Prs1, ribose-phosphate pyrophosphokinase; Pts9AB, mannose PTS, EIIAB; PtsI, phosphoenolpyruvate-protein phosphatase; Pyk, pyruvate kinase; RbsK2, ribokinase; Rpe, ribulose-phosphate 3-epimerase; RpiA, ribose 5-phosphate isomerase A; Sack, fructokinase; TpiA, triosephosphate isomerase; Xpk1, phosphoketolase.


Figure 5
View larger version (16K):
[in this window]
[in a new window]

  		FIG. 5. Expression profiles of some identified proteins. The quantities of the spots in each growth phase are shown as the means of spot quantity ± standard errors of the means. Abbreviations: (a) PurA, adenylosuccinate synthase; PurH, bifunctional purine biosynthesis protein purH; PyrD, dihydroorotate oxidase; Apt, adenine phosphoribosyltransferase; (b) AtpA, H+-transporting two-sector ATPase, alpha subunit; AtpG, H+-transporting two-sector ATPase, gamma subunit; (c) AspB, aspartate aminotransferase; MetC1, cystathionine beta-lyase.

Protein expression levels for carbohydrate metabolism in cucumber juice varied considerably (Fig. 4). The proteins for nucleotide synthesis showed expression levels that were comparable to those observed for MRS broth. However, the adenine phosphoribosyltransferase protein expression was found at low levels in cucumber juice. Interestingly, the alpha subunit of an ATPase showed the highest expression in cucumber juice, whereas the gamma subunit showed no expression in this medium (Fig. 5b).

Glycolytic enzymes and those involved in pyruvate metabolism generally showed the lowest expression in liquid feed. However, the phosphotransferases (PTS) protein identified, as well as the enzymes for the first steps of the degradation of hexoses (glucokinase, glucose-6-phosphate isomerase, phosphoglucomutase, and 1-phosphofructokinase), were highly expressed in liquid feed. The highest expression of the key enzyme for the oxidative branch of the phosphoketolase pathway (phosphogluconate dehydrogenase) was also detected in this medium (Fig. 4). The proteins for nucleotide metabolism showed the lowest expression in this medium compared to that in MRS broth and cucumber juice (Fig. 5a). The expression of enzymes for amino acid metabolism such as aspartate aminotransferase and cystathionine beta-lyase (Fig. 5c) and some proteins involved in DNA metabolism, transcription, and translation was found at low levels in liquid feed.

The expression patterns of the proteins identified from both L. plantarum strains were very similar. Generally, the REB1 strain showed more proteins with medium-dependent expression (i.e., more strongly up- and downregulated) than the MLBPL1 strain, which had a similar protein expression pattern but with smaller differences (not significantly different) in expression. Within the same strain, there were also protein isoforms identified at different spots that showed differences in their expression profiles, e.g., isoforms of fructose bisphosphate aldolase showed differential expression in cucumber juice, as well as isoforms of carbamoyl-phosphate synthase (pyrimidine-specific small chain) in MRS broth, and those of glutamyl-tRNA amidotransferase (subunit A) in liquid feed (Fig. 6).


Figure 6
View larger version (22K):
[in this window]
[in a new window]

  		FIG. 6. Expression profiles of some identified proteins for which two isoforms of the same protein showed different expression patterns. The quantities of the spots in each growth phase are shown as the means of spot quantity ± standard errors of the means. Abbreviations: (a) Fba, fructose bisphosphate aldolase; (b) PyrAA, carbamoyl-phosphate synthase, pyrimidine-specific small chain; (c) GatA, glutamyl-tRNA amidotransferase, subunit A.

Sugars, organic acids, and ethanol analyses.
The total consumption of hexoses (mainly glucose) in MRS broth was 48 mM for the REB1 strain and 55 mM for MLBPL1. The total production of lactic acid was 83 mM and 110 mM for REB1 and MLBPL1, respectively. These results show that approximately 1 mol of hexose was converted to 2 mol of lactic acid, as in a classic homofermentative pattern. Moreover, the production of acetic acid or ethanol was practically nonexistent (Tables 3 and 4).


View this table:
[in this window]
[in a new window]

  		TABLE 3. Effects of the growth of L. plantarum REB1 on the concentrations of sugars, organic acids, and ethanol in MRS broth, cucumber juice, and liquid feed


View this table:
[in this window]
[in a new window]

  		TABLE 4. Effects of the growth of L. plantarum MLBPL1 on the concentrations of sugars, organic acids, and ethanol in MRS broth, cucumber juice, and liquid feed

Cucumber juice is normally supplemented with 2% glucose because of the decrease of sugars when cucumbers are stored. For this reason, the final glucose concentration in cucumber juice was particularly high. However, the total consumption of hexoses, as well as the total lactic acid production, was lower in cucumber juice than in MRS broth. In cucumber juice, the fermentation of hexoses was also mainly homolactic. The production of acetic acid was very low, and even though the results showed that production of acetic acid for REB1 (2.2 mM) was higher than that for MLBPL1 (0.8 mM), the differences were not significant (P > 0.05). Cucumber juice contained approximately 14 mM of malic acid, which was completely consumed by both strains. Ethanol was practically undetectable (Tables 3 and 4).

The total consumption of hexoses in liquid feed was approximately 30 mM for both strains, showing consumption that was higher than that in cucumber juice. The production of lactic acid was also higher in liquid feed than in cucumber juice (76 and 87 mM by REB1 and MLBPL1 strains, respectively). The highest concentrations of acetic acid were detected in this medium (8 mM by REB1 and 9 mM by MLBPL1). Liquid feed also contained approximately 8.5 mM of malic acid, which was totally consumed in both strains. No ethanol production was observed (Tables 3 and 4). The fermentation of hexoses in liquid feed was, thus, both homo- and heterolactic.


arrow
DISCUSSION
 
The extensive adaptation capacity of L. plantarum to different environments and its wide range of applications make this bacterium especially interesting and challenging. The sequence analysis provided very valuable information, showing that this bacterium's most striking feature was the high potential to import and metabolize a large number of carbohydrates (8). Genomic analysis provides a static view of an organism, whereas proteomic analysis allowed a more dynamic observation and, thus, a more complete picture of the cellular events under conditions approaching the actual applications (e.g., in food and feed fermentations). Proteomic analysis was, therefore, chosen to check the changes on the metabolic routes of two L. plantarum strains of different origin, grown in two media of diverse composition or in a standard rich medium (MRS broth). The bacteria were collected at the exponential phase, during which the metabolism indicated by the soluble proteome is most versatile, as shown in previous studies (4, 9).

Sugar metabolism.
The main quality of LAB is the efficient fermentation of different carbohydrates and related compounds linked to the synthesis of ATP by substrate level phosphorylation. The ATP produced is subsequently used for biosynthetic purposes (2). L. plantarum is a facultative heterofermentative bacterium, indicating that hexoses are fermented via glycolysis and pentoses via the phosphoketolase pathway, leading to homo- or heterolactic fermentation, respectively (2). However, under stress conditions, the fermentation of hexoses via the phosphoketolase pathway by L. plantarum has been reported (15). In the study reported herein, two strains of L. plantarum were grown in three considerably different media of variable sugar composition. MRS broth contained mainly glucose, cucumber juice contained glucose and fructose, and liquid feed had a more complex carbohydrate composition, in which lactose was the predominant sugar (Tables 3 and 4). In all three media, the fermentation of sugars was primarily homolactic. However, some heterolactic activity was detected in liquid feed, indicated by the production of acetic acid after fermentation.

The expression of proteins for the sugar metabolism was generally rather variable. The highest expression of glycolytic proteins occurred in MRS broth and the lowest in liquid feed. Interestingly, the phosphotransferase systems identified showed the highest expression in MRS broth and liquid feed, as did the enzymes for the first steps of degradation of hexoses (Fig. 4). These expression levels correlate with the total hexose consumption, which was the lowest in cucumber juice (roughly 10%, compared with 50% in MRS broth and 33% in liquid feed) (Tables 3 and 4). Despite the low sugar consumption, the growth of both L. plantarum strains was significantly higher (P < 0.01 or P < 0.05) in cucumber juice than in MRS broth and liquid feed (Fig. 2). This discrepancy between sugar consumption and growth could be explained by the decarboxylation of malate. It is known that the cofermentation of malic acid with another carbohydrate such as glucose increases biomass yield and lactic acid production more than fermentation of glucose alone (2, 13). Thus, the consumption of malic acid in cucumber juice (and to a minor degree in liquid feed) gave the bacteria a growth benefit over that of the bacteria grown solely on carbohydrates. This is supported by the expression of the malolactic enzyme, which showed the highest expression in cucumber juice, followed by expression in liquid feed (Fig. 4).

The highest expression of 6-phosphogluconate dehydrogenase, the key enzyme from the oxidative branch of the phosphoketolase pathway, occurred in liquid feed, followed by that in cucumber juice. These results suggest that both of the L. plantarum strains examined here can oxidize hexoses via the phosphoketolase pathway, as reported previously for L. plantarum (15) and as concluded in our previous study of the growth phase-dependent proteome of these two strains (9). Moreover, degradation of hexoses via the phosphoketolase pathway appeared to be medium dependent, suggesting that the combination of different sugars in a medium may influence the pathway used for the degradation of hexoses. Expression of the enzyme ribokinase was highest in MRS broth and cucumber juice, whereas expression levels of the phosphoketolase enzyme were very similar in all three media. These results suggest that the riboses generated in the phosphoketolase pathway in MRS broth and cucumber juice were most likely used for nucleotide synthesis rather than for energy production. This is supported by the fact that the proteins for nucleotide synthesis (purines and pyrimidines) also showed preferential expression in MRS broth and cucumber juice (see Table SM5 in the supplemental material).

In all three media, the major monosaccharide was glucose, and it was still in excess at the time point at which bacteria were harvested. In liquid feed, the preferred sugars appeared to be sucrose and lactose, the breakdown of which contributed to the net increase of glucose and fructose during fermentation. No accumulation of galactose was seen, indicating that this breakdown product of lactose was also consumed rapidly. Moreover, there was some indication of heterolactic fermentation in liquid feed, which is contrary to the expectation that glucose is the preferred substrate as long as it is available (2, 5), and the consumption of malic acid had a significant effect on carbohydrate utilization. Thus, sugar metabolism seemed to have a different, or more complicated, regulatory system than that known for other LAB, at least in the strains of L. plantarum examined in this study.

Biosynthetic metabolism.
The expression of proteins involved in purine and pyrimidine metabolism reflected the accessibility of these compounds in the media. The proteins from the de novo synthesis of nucleotides in cucumber juice and MRS broth showed higher expression than in liquid feed (Fig. 5a). It seems that barley, the major ingredient in liquid feed, provided enough nucleotides, so no synthesis was necessary. There was also the indication that the bacteria used riboses (formed in the phosphoketolase pathway) to obtain the pentose ring for the nucleotide synthesis in MRS broth and cucumber juice, as stated above (Fig. 4). However, when there is no need for nucleotide synthesis, as in liquid feed, pentoses follow the heterofermentative pathway, leading to the formation of acetate.

Expression of the enzymes for amino acid metabolism was relatively varied (see Fig. SM5 in the supplemental material). However, expression was generally lower in liquid feed than in the other media. These results reflect the possibility that liquid feed also contains large amounts of amino acids and utilizable peptides. The expression levels of proteins for DNA metabolism were very similar in all three media, as was the expression level of enzymes for protein synthesis.

These results indicate that the L. plantarum strains REB1 and MLBPL1 have very similar biosynthetic behavior in all three media, beside the requirements for essential components such as nucleotides and amino acids. This suggests that environmental variability does not affect the biosynthetic pathways.

Stress response.
The genes involved in the general stress response are highly conserved among LAB. These genes expressed proteins involved in the maturation of newly synthesized proteins, in the refolding denatured proteins, and in DNA repair (24). However, there are different stress conditions that activate these responses, such as acid and basic stress, oxidative stress, heat and cold stress, osmotic stress, and starvation (3, 9, 15, 24). In our previous study using MRS broth, stress response proteins were prominent in the lag phase, suggesting that the bacteria were adapting to the new fresh batch, whereas in exponential and early stationary phases, the stress response showed a lower and more constant expression level (9). In this study, the identified stress proteins were present but did not show significant differences between the media.

Conclusions.
To improve the flavor, texture, or microbiological quality of fermented foods and feeds, it is important to know the metabolism of the responsible microorganisms. This study provides this type of information by addressing the protein expression of two starter strains with different histories under experimental conditions that mimic their actual applications (with fermented vegetables and liquid feed).

The results demonstrated the ability of L. plantarum to adapt to different environmental conditions as shown by its capacity to change its metabolic pathways according to the medium used. The carbohydrate fermentation appeared to be very flexible, allowing the bacterium the simultaneous use of different fermentable sugars. This is a very important feature of this species because it makes it very suitable as a potential starter culture for any kind of fermented food and feed products. The performances of the two strains studied here were very similar; REB1, however, showed more proteins that were strongly up-/downregulated than those of MLBPL1. This capacity to switch a specific protein expression is apparently an important characteristic of this bacterium.

The two strains used in this study demonstrate that despite the basic similarity between the overall metabolic profiles of REB1 and MSBPL1, there were also differences, probably reflecting their previous history of use. Further understanding of the specific adaptations will allow selection of starter strains for specific purposes, including strains originating from sources other than food or feed (i.e., intestinal strains).


arrow
ACKNOWLEDGMENTS
 
This work was supported by the National Technology Agency of Finland (TEKES).

We thank Jim McNicol (Biomathematics and Statistics Scotland, Dundee, Scotland, United Kingdom) for advice on multivariate analysis. T.L.T. thanks Mari Mannila and staff of the Laboratory of Biotechnology at the University of Oulu for help with the chemical analysis.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Applied Biotechnology, Department of Biosciences, University of Kuopio, P.O. Box 1627, FIN-70211 Kuopio, Finland. Phone: 358 16173573. Fax: 358 17163322. E-mail: Carme.Plumed{at}uku.fi Back

{triangledown} Published ahead of print on 20 June 2008. Back

{dagger} Supplemental material for this article may be found at http://aem.asm.org/. Back

{ddagger} Present address: School of Biological Sciences, Royal Holloway, University of London, Egham, Surrey TW20 0EX, United Kingdom. Back

§ Present address: Department of Biological and Environmental Sciences, University of Helsinki, FI-00014 Helsinki, Finland. Back


arrow
REFERENCES
 
    1
  1. Adlerberth, I., S. Ahrne, M. L. Johansson, G. Molin, L. A. Hanson, and A. E. Wold. 1996. A mannose-specific adherence mechanism in Lactobacillus plantarum conferring binding to the human colonic cell line HT-29. Appl. Environ. Microbiol. 62:2244-2251.[Abstract]
    2. 2
  2. Axelsson, L. 2004. Lactic acid bacteria: classification and physiology, p. 1-66. In S. Salminen, A. von Wright, and A. Ouwehand (ed.), Lactic acid bacteria: microbial and functional aspects, 3rd ed. Marcel Dekker, New York, NY.
    3. 3
  3. Castaldo, C., R. A. Siciliano, L. Muscariello, R. Marasco, and M. Sacco. 2006. CcpA affects expression of the groESL and dnaK operons in Lactobacillus plantarum. Microb. Cell Fact. 5:35.[CrossRef][Medline]
    4. 4
  4. Cohen, D. P., J. Renes, F. G. Bouwman, E. G. Zoetendal, E. Mariman, W. M. de Vos, and E. E. Vaughan. 2006. Proteomic analysis of log to stationary growth phase Lactobacillus plantarum cells and a 2-DE database. Proteomics 6:6485-6493.[CrossRef][Medline]
    5. 5
  5. Gänzle, M. G., N. Vermeulen, and R. F. Vogel. 2007. Carbohydrate, peptide and lipid metabolism of lactic acid bacteria in sourdough. Food Microbiol. 24:128-138.[CrossRef][Medline]
    6. 6
  6. Goossens, D., D. Jonkers, E. Stobberingh, and R. Stockbrügger. 2005. The effect of the probiotic L-plantarum 299v on the faecal and mucosal bacterial flora of patients with inactive ulcerative colitis. Eur. J. Gastroenterol. Hepatol. 17:A62.
    7. 7
  7. Kanehisa, M. 1997. A database for post-genome analysis. Trends Genet. 13:375-376.[CrossRef][Medline]
    8. 8
  8. Kleerebezem, M., J. Boekhorst, R. van Kranenburg, D. Molenaar, O. P. Kuipers, R. Leer, R. Tarchini, S. A. Peters, H. M. Sandbrink, M. W. E. J. Fiers, W. Stiekema, R. M. K. Lankhorst, P. A. Bron, S. M. Hoffer, M. N. N. Groot, R. Kerkhoven, M. de Vries, B. Ursing, W. M. de Vos, and R. J. Siezen. 2003. Complete genome sequence of Lactobacillus plantarum WCFS1. Proc. Natl. Acad. Sci. USA 100:1990-1995.[Abstract/Free Full Text]
    9. 9
  9. Koistinen, K. M., C. Plumed-Ferrer, S. J. Lehesranta, S. O. Kärenlampi, and A. von Wright. 2007. Comparison of growth-phase dependent cytosolic proteomes of two Lactobacillus plantarum strains used in food and feed fermentations. FEMS Microbiol. Lett. 273:12-21.[CrossRef][Medline]
    10. 10
  10. Koistinen, K. M., V. H. Hassinen, P. A. M. Gynther, S. J. Lehesranta, S. I. Keinänen, H. I. Kokko, E. J. Oksanen, A. I. Tervahauta, S. Auriola, and S. O. Kärenlampi. 2002. Birch PR-10c is induced by factors causing oxidative stress but appears not to confer tolerance to these agents. New Phytol. 155:381-391.[CrossRef]
    11. 11
  11. Lehesranta, S. J., H. V. Davies, L. V. T. Shepherd, N. Nunan, J. W. McNicol, S. Auriola, K. M. Koistinen, S. Suomalainen, H. I. Kokko, and S. O. Kärenlampi. 2005. Comparison of tuber proteomes of potato varieties, landraces, and genetically modified lines. Plant Physiol. 138:1690-1699.[Abstract/Free Full Text]
    12. 12
  12. McNaught, C. E., N. P. Woodcock, A. D. G. Anderson, and J. MacFie. 2005. A prospective randomised trial of probiotics in critically ill patients. Clin. Nutr. 24:211-219.[CrossRef][Medline]
    13. 13
  13. Passos, F. V., H. P. Fleming, H. M. Hassan, and R. F. McFeeters. 2003. Effect of malic acid on the growth kinetics of Lactobacillus plantarum. Appl. Microbiol. Biotechnol. 63:207-211.[CrossRef][Medline]
    14. 14
  14. Pavan, S., P. Hols, J. Delcour, M. C. Geoffroy, C. Grangette, M. Kleerebezem, and A. Mercenier. 2000. Adaptation of the nisin-controlled expression system in Lactobacillus plantarum: a tool to study in vivo biological effects. Appl. Environ. Microbiol. 66:4427-4432.[Abstract/Free Full Text]
    15. 15
  15. Pieterse, B., R. J. Leer, F. H. J. Schuren, and M. J. van der Werf. 2005. Unravelling the multiple effects of lactic acid stress on Lactobacillus plantarum by transcription profiling. Microbiology 151:3881-3894.[Abstract/Free Full Text]
    16. 16
  16. Plumed-Ferrer, C., I. Kivelä, P. Hyvönen, and A. von Wright. 2005. Survival, growth and persistence under farm conditions of a Lactobacillus plantarum strain inoculated into liquid pig feed. J. Appl. Microbiol. 99:851-858.[CrossRef][Medline]
    17. 17
  17. Plumed-Ferrer, C., M. Llopis, P. Hyvönen, and A. von Wright. 2004. Characterization of the microbial community and its changes in liquid piglet feed formulations. J. Sci. Food Agric. 84:1315-1318.[CrossRef]
    18. 18
  18. Pouwels, P. H., R. J. Leer, M. Shaw, M. J. H. den Bak-Glashouwer, F. D. Tielen, E. Smit, B. Martinez, J. Jore, and P. L. Conway. 1998. Lactic acid bacteria as antigen delivery vehicles for oral immunization purposes. Int. J. Food Microbiol. 41:155-167.[CrossRef][Medline]
    19. 19
  19. Shevchenko, A., M. Wilm, O. Vorm, and M. Mann. 1996. Mass spectrometric sequencing of proteins from silver stained polyacrylamide gels. Anal. Chem. 68:850-858.[Medline]
    20. 20
  20. Tamminen, M., T. Joutsjoki, M. Sjöblom, M. Joutsen, A. Palva, E. L. Ryhänen, and V. Joutsjoki. 2004. Screening of lactic acid bacteria from fermented vegetables by carbohydrate profiling and PCR-ELISA. Lett. Appl. Microbiol. 39:439-444.[CrossRef][Medline]
    21. 21
  21. Teusink, B., F. H. J. van Enckevort, C. Francke, A. Wiersma, A. Wegkamp, E. J. Smid, and R. J. Siezen. 2005. In silico reconstruction of the metabolic pathways of Lactobacillus plantarum: comparing predictions of nutrient requirements with those from growth experiments. Appl. Environ. Microbiol. 71:7253-7262.[Abstract/Free Full Text]
    22. 22
  22. Titgemeyer, F., and W. Hillen. 2002. Global control of sugar metabolism: a Gram-positive solution. Antonie van Leeuwenhoek 82:59-71.[CrossRef][Medline]
    23. 23
  23. van de Guchte, M., P. Serror, C. Chervaux, T. Smokvina, S. D. Ehrlich, and E. Maguin. 2002. Stress responses in lactic acid bacteria. Antonie van Leeuwenhoek 82:187-216.[CrossRef][Medline]
    24. 24
  24. Yoon, K. Y., E. E. Woodams, and Y. D. Hang. 2006. Production of probiotic cabbage juice by lactic acid bacteria. Bioresour. Technol. 97:1427-1430.[CrossRef][Medline]

Applied and Environmental Microbiology, September 2008, p. 5349-5358, Vol. 74, No. 17
0099-2240/08/$08.00+0     doi:10.1128/AEM.00324-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental material
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Plumed-Ferrer, C.
Right arrow Articles by von Wright, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Plumed-Ferrer, C.
Right arrow Articles by von Wright, A.
Agricola
Right arrow Articles by Plumed-Ferrer, C.
Right arrow Articles by von Wright, A.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»