Malic Enzyme and Malolactic Enzyme Pathways Are Functionally Linked but Independently Regulated in Lactobacillus casei BL23

Abstract

Lactobacillus casei is the only lactic acid bacterium in which two pathways for l-malate degradation have been described: the malolactic enzyme (MLE) and the malic enzyme (ME) pathways. Whereas the ME pathway enables L. casei to grow on l-malate, MLE does not support growth. The mle gene cluster consists of three genes encoding MLE (mleS), the putative l-malate transporter MleT, and the putative regulator MleR. The mae gene cluster consists of four genes encoding ME (maeE), the putative transporter MaeP, and the two-component system MaeKR. Since both pathways compete for the same substrate, we sought to determine whether they are coordinately regulated and their role in l-malate utilization as a carbon source. Transcriptional analyses revealed that the mle and mae genes are independently regulated and showed that MleR acts as an activator and requires internalization of l-malate to induce the expression of mle genes. Notwithstanding, both l-malate transporters were required for maximal l-malate uptake, although only an mleT mutation caused a growth defect on l-malate, indicating its crucial role in l-malate metabolism. However, inactivation of MLE resulted in higher growth rates and higher final optical densities on l-malate. The limited growth on l-malate of the wild-type strain was correlated to a rapid degradation of the available l-malate to l-lactate, which cannot be further metabolized. Taken together, our results indicate that L. casei l-malate metabolism is not optimized for utilization of l-malate as a carbon source but for deacidification of the medium by conversion of l-malate into l-lactate via MLE.

INTRODUCTION

Lactobacillus casei is a facultatively heterofermentative lactic acid bacterium (LAB) isolated from a wide variety of habitats, including raw and fermented milk, the gastrointestinal tracts of animals, and plant materials (1). Lb. casei strains are used as cheese starter cultures, but a major interest in this species has arisen from the probiotic properties of some strains (2). Lb. casei is also remarkable because is the only LAB in which both the malic enzyme and the malolactic enzyme l-malate dissimilation pathways have been demonstrated (3, 4). Most LAB decarboxylate l-malate to l-lactate by a NAD⁺ and Mn²⁺-dependent malolactic enzyme (MLE). A few of them, however, can convert l-malate into pyruvate by the action of a malic enzyme (ME). This pathway was first detected in Enterococcus faecalis (5) and later in Lb. casei (4, 6) and Streptococcus bovis (7). Although there is evidence showing that some LAB strains can utilize lactate as a carbon source (8–12), most LAB cannot channel lactate into the gluconeogenic pathway. For this reason, the utilization of l-malate through MLE cannot sustain their growth, whereas the utilization of the ME pathway enables these organisms to grow with l-malate as a carbon source (3, 13).

The metabolism of l-malic acid by LAB has led to considerable interest because of its relevance in winemaking (14), since the degradation of l-malate leads to a reduction in the acidity of wine, and it provides microbiological stability by preventing the secondary growth of LAB after bottling. However, whereas MLE has been the focus of an extensive research effort, the physiological role and the regulation of ME have received less attention.

In a previous study (3), we identified a gene cluster consisting of two diverging operons, maePE and maeKR, encoding a putative malate transporter (maeP), an ME (maeE), and a two-component system (TCS) belonging to the citrate family (maeK and maeR; Fig. 1). Our results showed that ME is required for growth with l-malate and that the TCS is essential for expression of maePE. Similar results have been obtained in E. faecalis JH2-2, which harbors an identical gene arrangement (15). Furthermore, transcriptional analyses showed that expression of maeE is induced by l-malic acid and repressed by glucose, whereas the TCS-encoding genes expression was induced by l-malic acid, and it was not repressed by glucose (3).

Fig 1.

Schematic representation of the mae and mle gene clusters in Lb. casei BL23 and derivative strains used in the present study.

A survey of the Lb. casei genome sequences available allows the identification of a second cluster of genes involved in l-malate metabolism (Fig. 1). This cluster is constituted by three genes encoding a putative malolactic enzyme (mleS), a l-malate transporter (mleT) and, oriented in the opposite direction, a LysR-type transcriptional regulator (mleR). Only the closely related species Lactobacillus rhamnosus also harbors an MLE-encoding gene cluster and an ME-encoding gene cluster. The presence of two pathways for l-malate utilization in Lb. casei prompted us to investigate their role in l-malate utilization as a carbon source, whether their expression is concertedly regulated at a transcriptional level and whether both pathways are functionally intertwined.

MATERIALS AND METHODS

Strains and growth conditions.

The strains and plasmids used in the present study are listed in Table 1. Lb. casei was routinely grown in MRS broth (Oxoid) at 37°C under static conditions. Lactococcus lactis MG1363 was grown in M17 medium (Oxoid) supplemented with 27.75 mM glucose. Escherichia coli DH5α strains were grown in LB medium (BD Difco) at 37°C with aeration. The antibiotics used were 100 μg ampicillin of ml⁻¹ for E. coli and 5 μg of erythromycin ml⁻¹ for Lb. casei and Lc. lactis.

Table 1.

Strains and plasmids used in this study

Strain or plasmid	Characteristics or relevant genotypea	Source or reference
Strains
Escherichia coli DH5α	F− endA1 hsdR17 gyrA96 thi-1 recA1 relA1 supE44 ΔlacU169 (ϕ80 lacZ ΔM15)	Stratagene
Lactobacillus casei
BL23	Wild-type strain	B. Chassy, University of Illinois
BL315	BL23 ΔmaeR	3
MRST	BL23 ΔmleRST	This study
MR	BL23 mleR::pRV300; Eryr	This study
MT	BL23 mleT::pRV300; Eryr	This study
MTc	MRST carrying plasmid pT1mleT	This study
MS	BL23 ΔmleS; in-frame deletion, aa 121–378	This study
MPs	BL23 maeP; stop codon after aa 16	This study
MPT	BL23 maeP mleT::pRV300; Eryr	This study
Lactococcus lactis MG1363	Plasmid-free derivative of NCDO712	16
Plasmids
pRV300	Insertional vector for Lactobacillus; Ampr Eryr	22
pRVmaePstop	pRV300 containing a 1-kb fragment with maeP carrying a stop codon	This study
pRVmle	pRV300 containing a 1.9-kb fragment with fused flanking regions from the mle operon	This study
pRVmleR	pRV300 containing a 0.6-kb internal fragment of mleR	This study
pRVmleS	pRV 300 containing a 0.7-kb fragment of mleS with a 774-bp in-frame deletion	This study
pRVmleT	pRV300 containing a 0.6-kb internal fragment of mleT	This study
pT1NX	Expression vector for Gram-positive bacteria harboring the constitutive P1 promoter; Eryr	17
pT1mleT	pT1NX carrying mleT under control of the P1 promoter	This study

^aAmp^r, ampicillin resistance; Ery^r, erythromycin resistance. aa, amino acid(s).

Growth assays and gene expression analyses were carried out at 30°C in malic enzyme induction (MEI) medium (tryptone, 5 g liter⁻¹; yeast extract, 5 g liter⁻¹; K₂HPO₄, 6 g liter⁻¹; KH₂PO₄, 4 g liter⁻¹; MgSO₄·7H₂O, 0.2 g liter⁻¹; MnSO₄, 0.05 g liter⁻¹; Tween 80, 1 ml liter⁻¹; cysteine, 0.5 g liter⁻¹) as previously described (3). Medium pH was adjusted to 6.8, 5.5, or 4.5 with HCl. At least three independent replicates of each growth curve were obtained. The results were expressed as averages ± the standard deviations.

DNA techniques.

Standard methods were used for cloning in E. coli (18). Restriction enzymes and T4 DNA ligase were purchased from New England BioLabs. Taq DNA polymerase for PCR screening was from Biotools (B&M Labs, Madrid, Spain). Platinum Pfx DNA polymerase (Life Technologies S.A., Madrid, Spain) was used for cloning purposes. Plasmids were isolated with a GFX Micro Plasmid Prep kit (GE Healthcare). DNA from Lb. casei was isolated with the DNA isolation kit for cells and tissues (Roche). E. coli strains were transformed by electroporation with a Gene Pulser apparatus (Bio-Rad), as recommended by the manufacturer, Lc. lactis was transformed as described by Holo and Nes (19), and Lb. casei strains were transformed by electroporation as described previously (20).

Real-time quantitative reverse transcription-PCR (RT-qPCR).

Isolation of total RNA from Lb. casei strains, synthesis of cDNA and, real-time quantitative PCR were carried out as described previously (3). Unless otherwise stated, samples of cultures growing in media containing either glucose or ribose were taken at mid-exponential phase (optical density at 595 nm [OD₅₉₅] of 0.6). Samples of cultures containing only l-malic acid were taken after 50 h of incubation_. Primers were designed by using the Primer-BLAST service (http://www.ncbi.nlm.nih.gov/tools/primer-blast) in order to generate amplicons ranging from 100 to 150 bp in size. The primers used were as follows (see Table S1 in the supplemental material): maeE-qF and maeE-qR (maeE), maeP-qR and maeP-qR (maeP), mleS-qF and mleS-qR (mleS), and mleT-qF and mleTqR (mleT). For each set of primers, the cycle threshold values (crossing points) were determined by the automated method implemented in the LightCycler software 4.0 (Roche). The genes fusA, leuS, pyrG, and recG were selected as reference genes (3). The relative expression based on the expression ratios between the target genes and reference genes was calculated using the software tool REST (relative expression software tool) (21). Linearity and amplification efficiency were determined for each primer pair. Every real-time PCR determination was performed at least six times.

Construction of strains.

The genetic organization of the mae and mle gene clusters in the strains used in the present study is shown in Fig. 1. To construct the mleR- and mleT-defective mutants, internal fragments of the mleR and mleT genes were amplified by PCR using Pfx polymerase and the oligonucleotide pairs MleR1/MleR2 and MleT1/MleT2, respectively (see Table S1 in the supplemental material). The PCR products were digested with XhoI/SacI, ligated to the integrative vector pRV300 (22) digested with the same enzymes, and transformed into E. coli DH5α. The resulting plasmids (pRVmleR and pRVmleT) were used to transform Lb. casei BL23, and single-crossover integrants were selected by their resistance to erythromycin and confirmed by PCR using a combination of an oligonucleotide annealing at the pRV300 polylinker and an external oligonucleotide (MleR3 and MleT3, respectively). One strain of each type of integration was selected and named MR (mleR::pRV300) and MT (mleT::pRV300).

In order to obtain a BL23 derivative strain harboring a deletion of the mle gene cluster, flanking fragments of the region to be deleted were amplified using the primer sets MledelF1/MledelR1 and MledelF2/MledelR2, respectively (see Table S1 in the supplemental material). Since mleS and mleT genes are probably translationally coupled, an in-frame deletion in mleS that removed the sequences coding for amino acids 121 to 378 was constructed using the primer sets MleSdelF1C/MleSdelR1C and MleSdelF2C/MleSdelR2A (see Table S1 in the supplemental material). The corresponding pairs of PCR fragments were combined into one fragment by PCR, digested with XhoI/SacI, and cloned in pRV300 to generate the plasmids pRVmle and pRVmleS, respectively, as described previously (3). Lb. casei was transformed with pRVmle or pRVmleS, and for each plasmid one erythromycin-resistant clone carrying the plasmid integrated by a single crossover was grown in MRS without erythromycin for ∼200 generations. The cells were plated on MRS and replica plated on MRS plus erythromycin. Antibiotic-sensitive clones were isolated and, among them, one was selected in which a second recombination event led to the deletion of the mle gene cluster or an internal in-frame deletion in gene mleS, as subsequently confirmed by sequencing of PCR-amplified fragments spanning the deleted regions. The resulting strains were named MRST (Δmle) and MS (mleS), respectively (Table 1). Sequence alignments of Lb. casei BL23 and each derivative strain around the deletion point are shown in Fig. S1 in the supplemental material.

Translational stop codons were introduced into maeP by a recombination strategy. The primer sets MaePsF1/MaePsR1 and MaePsF2/MaePsR2 (see Table S1 in the supplemental material) were used to amplify two internal and overlapping fragments of maeP. The primers MaePsR1 and MaePsF2 are complementary and contain two in-frame stop codons and a SpeI restriction site. The two DNA fragments were merged by PCR, and the resulting fragment was digested with XhoI/EcoRI and cloned in pRV300. The resulting plasmid, pRVmaePstop, was transformed in Lb. casei BL23, and single-crossover recombinants were selected for their resistance to erythromycin. As outlined above, one isolate was grown without selective pressure. Erythromycin-sensitive clones were selected and checked by PCR amplification with the primer pair MaePsF1/MaePsR2 and subsequent digestion of the amplified fragments with SpeI. Introduction of the mutation was subsequently confirmed by DNA sequencing. The resulting sequence of the derivative strain MPs is shown in Fig. S1 in the supplemental material.

A maeP mleT double mutant was obtained by transforming strain MPs (maeP) with the plasmid pRVmleT. Single-crossover integrants were selected by their resistance to erythromycin and confirmed by PCR as described above. One isolate was selected and named MPT (maeP mleT::pRV300).

Complementation of mleT mutation was achieved by cloning mleT in plasmid pT1NX under the control of the constitutive promoter P1. To this end, a PCR fragment encompassing gene mleT was amplified by using primers MleT-C1 and MleT-C2 (see Table S1 in the supplemental material), digested with BglII and SpeI, and ligated to pT1NX digested with the same enzymes, resulting in plasmid pT1mleT. The ligation mixture was transformed into Lc. lactis MG1363 by electroporation (19), and transformants were checked by restriction analysis and subsequent DNA sequencing. Plasmid from one isolate was used to transform Lb. casei MRST (Δmle), and the resulting strain was named MTc (Δmle [pT1mleT]).

Intracellular l-malate accumulation assay.

Bacteria were grown in MEI medium supplemented with 33.3 mM ribose and 37.3 mM l-malic acid at 30°C until the mid-exponential phase (OD₅₉₅ of 0.6 to 0.8). Cells were collected by centrifugation (12,000 × g, 5 min, 4°C) and washed twice in cold 25 mM sodium phosphate buffer (pH 6) with 1 mM MgCl₂. The cells were finally suspended in the same buffer at an OD₅₉₅ of 6.0 and kept on ice for immediate use. The reaction mixture consisted of 300 μl of cell suspension, to which 3 μl of 10% (wt/vol) peptone and 3 μl of 1.1 M glucose were added. The mixture was incubated at 30°C for 5 min prior to the addition of 400 nCi of l-[U-¹⁴C]malic acid (Hartmann Analytic GmbH). After 15 s, the reaction mixture was rapidly filtered through a 0.45-μm-pore-size nitrocellulose filter (Millipore) and washed twice with 5 ml of cold 0.1 M lithium chloride. The filters were dried, and the radioactivity retained in the cells was determined by liquid scintillation counting. Six independent replicates of each malate uptake assay were carried out. For statistical analyses, an F test was used to compare variances, and a two-tailed unpaired t test with Welch's correction was used to compare means.

Analysis of organic acids.

Samples of cultures grown in MEI medium supplemented with 33.5 mM l-malic acid (MEIM) were taken at different times during growth. The samples were centrifuged, and the supernatant was filtered through 0.22-μm-pore-size Millex-GV syringe-driven filter units (Millipore) and stored at −80°C until use. Samples were analyzed using high-pressure liquid chromatography equipment (Agilent, series 1200) with an isocratic pump (Agilent G1310A) according to the procedure described by Frayne (38) with minor modifications. The mobile phase consisted of a solution of 0.75 ml of 85% H₃PO₄ per liter of deionized water, with a flow rate of 0.7 ml min⁻¹. An Agilent G1322A degasser was used. Samples (5 μl) were injected automatically (Agilent G1367B). The separation of the components was carried out using an Aminex HPX-87H precolumn (Bio-Rad) coupled to two Aminex HPX-87H ion exclusion columns (300 by 7.8 mm; Bio-Rad) thermostatically controlled at 65°C (Agilent G1316A). The compounds were detected by a variable wavelength detector (Agilent G1314B) set to 210 nm and a refractive index detector (Agilent G1362A) in series. External calibration was performed.

RESULTS

Lb. casei harbors a gene cluster encoding the MLE pathway.

Biochemical evidence had shown that Lb. casei strains possess both ME and MLE activities (4). Inspection of available Lb. casei genome sequences allowed us to identify a gene cluster consisting of a putative transcriptional regulator of the LysR family (mleR) and two divergently transcribed genes which would code for a malolactic enzyme and a putative l-malate transporter (mleS and mleT), respectively (Fig. 1). This genetic organization is identical to that of mle clusters found in related LAB such as Oenococcus oeni (23). A BLASTP search (http://blast.ncbi.nlm.nih.gov) showed that Lb. casei MleS shares 377 (70%) identical and 452 (84%) conserved residues with the biochemically characterized MLE of O. oeni (24). MleT is a putative membrane transport protein (Pfam PF03547) and shares 207 (64%) identical and 253 (78%) conserved residues with its O. oeni counterpart. Finally, MleR shares 108 (39%) identical and 164 (60%) conserved residues with its O. oeni counterpart. On the basis of this evidence, we concluded that these three genes constitute the malolactic gene cluster of Lb. casei.

Expression of genes involved in l-malic acid metabolism in Lb. casei BL23.

The relative transcript levels of genes maeE, maeP, mleS, and mleT of Lb. casei BL23 were determined in cells grown in MEI medium (pH 6.8) supplemented with glucose, ribose, glucose/l-malic acid, ribose/l-malic acid, or l-malic acid. The transcript levels in cells grown with glucose were taken as the reference condition. The results obtained showed that maeE and maeP genes were induced in the presence of l-malic acid and in the absence of glucose (Fig. 2) in agreement with our previous results (3). A moderate increment in mae transcripts was observed in the presence of ribose possibly due to the relieving of carbon catabolite repression by CcpA in the presence of this sugar (25). Notwithstanding, maximal induction occurred in MEIM, indicating that induction by MaeR still operated in the presence of ribose. On the other hand, mle transcripts were more abundant in the presence of l-malic acid, and glucose had no significant effect on their expression (Fig. 2). This result strongly suggests that expression of mle genes is induced by l-malic acid and that it is not subjected to carbon catabolite repression.

Fig 2.

RT-qPCR analysis of the relative transcript levels of l-malic acid utilization genes in Lb. casei BL23 grown with different carbon sources compared to the same strain grown with glucose. GM, glucose plus l-malic acid; M, l-malic acid; R, ribose; RM ribose plus l-malic acid. RE, relative gene expression ratio; means ± the standard errors are represented.

The effect of external pH on mae and mle genes expression was also studied. To this end, strain BL23 (wild type) was grown in MEI medium, supplemented with glucose or l-malic acid, and adjusted to pH 5.5 or 4.5. Samples for RNA isolation were obtained as indicated above using RT-PCR (see Materials and Methods) except for cultures in MEI supplemented with l-malic acid which were taken after 10 h, due to the faster growth of strain BL23 under these culture conditions compared to its growth in MEIM adjusted to pH 6.8 (results not shown). Small differences in the expression of mle genes at the different pH values tested were observed (see Fig. S2 in the supplemental material), indicating that pH had a minor effect on the expression of the mle operon under these experimental conditions. In contrast, a significant increase in the expression of mae genes was observed (see Fig. S2 in the supplemental material), especially at pH 5.5. This result indicates that pH affects the expression of mae genes under these experimental conditions.

MleR is required for induction of the expression of mleS and mleT, whereas MaeR does not affect the expression of mle genes.

The role of the mleR gene was evaluated by comparing the transcript levels of genes involved in the l-malic acid metabolism of Lb. casei BL23 (wild type) and a mleR-defective mutant (MR strain) using cells grown under the conditions described above. Inactivation of mleR resulted in a loss of induction of mle genes (Fig. 3A). Therefore, MleR is a transcriptional activator required for induction of expression of mle genes in the presence of l-malic acid. Compared to the transcript levels present in the parental strain, modest differences were observed for mae genes (Fig. 3B). The basal level of mae transcripts was higher in the mleR mutant grown with ribose and lower when grown with ribose and l-malic acid or l-malic acid compared to the corresponding cultures of strain BL23. These results indicate that MleR does not regulate the transcription of mae genes, although the inactivation of MleR may have an indirect effect on the transcription of mae genes under certain growth conditions.

Fig 3.

Effect of a mleR mutation on the expression of mle and mae genes. (A) RT-qPCR analysis of the relative transcript levels of l-malic acid utilization genes in Lb. casei MR (mleR) strain grown with different carbon sources compared to the same strain grown with glucose. (B) RT-qPCR analysis of the relative transcript levels of l-malic acid utilization genes in Lb. casei MR (mleR) strain grown with different carbon sources compared to corresponding cultures of the wild-type strain Lb. casei BL23. G, glucose; GM, glucose plus l-malic acid; M, l-malic acid; R, ribose; RM ribose plus l-malic acid. RE, relative gene expression ratio; means ± the standard errors are represented.

On the other hand, inactivation of MaeR did not affect the expression of mle genes (see Table S2 in the supplemental material). These results indicate that MaeR does not regulate expression of mle genes and therefore each gene cluster is independently regulated.

A functional malate transporter is required for induction of mle genes but not for mae genes.

The expression of l-malic acid metabolic genes was also studied in mutant strains defective in one or both putative l-malic acid transporters present in Lb. casei BL23 (MaeP and MleT). Since gene maeP is located upstream of maeE (Fig. 1), a strain harboring a stop codon in maeP (MPs strain) was obtained in order to minimize polar effects on the expression of maeE (see Fig. S1 in the supplemental material).

Inactivation of mleT (MT strain) resulted in loss of induction of mleS in MEI supplemented with glucose and l-malic acid, whereas the expression of this gene was induced when cells were grown with ribose and l-malic acid (Fig. 4). Measurement of transcript levels from cells grown with l-malic acid could not be carried out because of the low quality of the RNA obtained from these cells, possibly due to the poor growth of this strain on this compound (see below). In contrast, the induction of either mle or mae genes was not affected by a mutation in maeP, whereas mleS was not induced in any growth condition tested in the double-mutant strain MPT (maeP mleT). However, maeE was still induced in cells grown with ribose and l-malic acid (Fig. 4). These results strongly suggest that internalization of l-malate is required for the induction of mle genes but not for the induction of mae genes.

Fig 4.

Effect of the inactivation of l-malic acid transporters on the expression of mle and mae genes. RT-qPCR analysis of the relative transcript levels of l-malic acid utilization genes in Lb. casei strains grown with different carbon sources compared to the corresponding strains grown with glucose. (A) Strain MT (mleT); (B) strain MPs (maeP); (C) strain MPT (maeP mleT). See Fig. 2 for additional details.

Inactivation of gene mleT leads to a major growth defect in MEIM.

In order to evaluate the relevance of the two putative malate transporters encoded by Lb. casei (MaeP and MleT) on the growth with l-malic acid, growth of Lb. casei BL23 and its derivative strains MT (mleT), MPs (maeP), and MPT (maeP mleT) in MEIM was monitored. All Lb. casei strains displayed a biphasic growth (Fig. 5). In the first stage of rapid growth, possibly due to consumption of residual sugars in the growth medium or reserve compounds (3), all strains grew at similar growth rates. This was followed by a lag phase and a second stage of slow growth in which the behavior of the different strains varied. The results obtained showed that the inactivation of mleT led to a major growth defect, whereas the inactivation of maeP resulted in a slight delay in growth, although both strains eventually reached similar values of maximal OD (Fig. 5). Inactivation of both transporters prevented growth on l-malic acid (Fig. 5).

Fig 5.

Growth of Lb. casei BL23 and derivative strains MPs (maeP), MT (melT), and MPT (maeP mleT) in MEIM. Error bars represent the standard deviations.

Both transporters MleT and MaeP contribute to malate accumulation in Lb. casei BL23.

In order to gain insight into the role of transporters MleT and MaeP in l-malic acid metabolism, the accumulation of malate by cells grown with ribose and l-malic acid was determined. This experiment could not be performed using cells grown with l-malic acid because reproducible results could not be obtained after repeated attempts. However, transcriptional analyses showed that both transporters are produced in cells growing with ribose and l-malic acid; therefore, the contribution of both transporters can be evaluated under this growth condition. Inactivation of any of the putative transporter encoding genes resulted in significant decreases in malate accumulation and inactivation of both mleT and maeP resulted in minimal accumulation of malate (Fig. 6). A significant difference in malate accumulation was observed between maeP and mleT strains (P = 0.042), indicating that MaeP was the main transporter under the assay conditions. In contrast, no significant difference was detected between the wild type and the mleR mutant (P = 0.612). Therefore, inactivation of MleR did not affect the malate accumulation ability of Lb. casei under the assay conditions.

Fig 6.

Accumulation of l-malate by Lb. casei BL23 and derivative strains grown in MEI supplemented with ribose and l-malic acid. Bars indicate the means of six independent determinations. Error bars represent the standard deviations.

Detrimental effect of MLE production on the growth of Lb. casei BL23 with l-malic acid.

The effect of a functional MLE pathway on the growth with l-malic acid was also addressed in the present study. The growth of strains MR (mleR), MRST (Δmle), and MS (mleS) (see Table 1) was monitored in MEIM. All three strains reached higher maximal OD values than the parental strain (Fig. 7). Again, a biphasic growth was observed, although interestingly, the behavior of the different strains varied. The mleR strain resumed growth as the parental strain Lb. casei BL23, although at a significantly higher growth rate. The mleS strain also grew at a higher growth rate than the parental strain, but it showed a longer lag phase, whereas the Δmle strain displayed an intermediate growth behavior (Fig. 7). In order to confirm that the difference observed between Δmle and mleS strains was only due to the absence of a functional mleT gene in the Δmle strain, plasmid pT1mleT constitutively producing MleT was introduced into MRST (Δmle) strain. The production of MleT resulted in a longer lag phase (see Fig. S3 in the supplemental material), indicating that the difference observed between strains MRST (Δmle) and MS (mleS) was due to the presence of MleT in the mleS mutant. In summary, strains lacking MLE (MS and MRST strains) or not inducing MLE production (MR strain) grew faster and reached higher maximal OD values than those producing MLE.

Fig 7.

Variation of OD, medium pH, and concentrations of l-malic acid, lactic acid, and acetic acid during the growth of Lb. casei BL23 and derivative strains in MEIM. (A) Strain BL23; (B) strain MT (mleT); (C) strain MS (ΔmleS); (D) strain MR (mleR); (E) strain MRST (ΔmleRST). Values represent the means of three independent experiments; error bars represent the standard deviations.

The variation of the pH of the growth medium and the concentrations of malic acid, lactic acid, and acetic acid were also monitored in this experiment. The results obtained showed that l-malic acid degradation resulted in an increase in the pH of the growth medium (Fig. 7). Again, remarkable differences were observed between strains. The parental strain BL23 and strain MT (mleT) raised the medium pH to ∼7.6, whereas strains MR (mleR), MS (mleS) and MRST (Δmle) only reached a pH value of ∼7.3 (Fig. 7). Furthermore, while growth and pH increase were coupled in strains with mutations in mleR, mleS, and in the Δmle strain, no correspondence between growth and pH variation was observed in the parental strain BL23 and strain MT (mleT). These results agreed with the degradation of l-malic acid and production of lactic acid and acetic acid. Strain BL23 degraded continuously the l-malic acid until its complete depletion (Fig. 7). In contrast, all other strains degraded l-malic acid rapidly during the first stage of growth, degradation rate diminished during the intermediate lag phase, and it increased again when growth was resumed (Fig. 7). The production of lactic acid and acetic acid markedly varied between strains. Strains expressing mleS (BL23 and MT) produced as much lactic acid as l-malic acid they consumed, whereas acetic acid was produced to a relatively low concentration (Fig. 7). On the other hand, strains with mutations in mleS, in mleR, or with all mle genes deleted produced less lactic acid and more acetic acid (Fig. 7). This result indicates that the main pathway of l-malic acid degradation in Lb. casei was the MLE even if this resulted in poor growth with l-malic acid as the only carbon source.

DISCUSSION

The unusual presence of two pathways for malate utilization in Lb. casei posed the question of what their respective roles are in l-malate utilization as a carbon source. The degradation of l-malic acid via MLE results in its direct decarboxylation to l-lactic acid. The free energy of the reaction is conserved by a chemiosmotic mechanism sustained by an electrogenic malate transport (26–28). However, although MLE can provide energy to the cell, it cannot sustain growth (3, 29, 30) and l-malic acid metabolism via this pathway is generally assumed as being a protective mechanism against acidification (9, 27, 31–33). On the other hand, ME converts l-malate into pyruvate, which can be subsequently directed to energy production, redox balance, or biosynthetic pathways (see Fig. S4 in the supplemental material). Therefore, the ability of Lb. casei to grow with l-malate as a carbon source depends on the activities of these two pathways.

A previous study by Schütz and Radler (4) has shown that Lb. casei strains produced MLE in the presence of glucose and l-malic acid, whereas ME was only detected in the presence of l-malic acid and the absence of glucose. Our results previously reported (3) and those reported here agree with this observation. RT-qPCR analysis of mae and mle transcripts showed that the expression of mae genes is induced by l-malic acid and repressed by glucose, whereas the expression of mle genes is induced by l-malic acid and is not subjected to glucose repression.

Furthermore, our results showed that mleR codes for a transcriptional activator required for induction of the expression of the mle metabolic genes (Fig. 3). The role of MleR as a transcriptional activator of mle genes was first reported for Lc. lactis (34) and subsequently confirmed for its homologous counterpart of Streptococcus mutans (35). Furthermore, these authors showed that the expression of mle genes in S. mutans was also affected by environmental pH independently of MleR and malate. Broadbent et al. (31) also observed the induction of mleS and mleT (mleP) genes in response to acid stress in Lb. casei ATCC 334 grown in MRS not supplemented with l-malic acid. However, no significant effect of external pH on the expression of mle genes was observed in the present study. Differences in growth medium and experimental design possibly account for this discrepancy. In contrast, a stimulating effect of acid pH on mae genes expression was observed (see Fig. S2 in the supplemental material). The response to low pH in Lb. casei has wide-ranging effects on gene expression (31); therefore, this observation does not necessarily reflect a direct regulatory effect of medium pH on mae gene expression.

The results reported here also demonstrated that cells must produce either MaeP or MleT for the induction of mle genes and that no other malate transporters are produced by Lb. casei, at least under the growth conditions assayed (Fig. 4). This is in agreement with the measurements of malate accumulation in cells grown with ribose and l-malic acid, which showed minimal accumulation of malate in strain MPT (maeP mleT; Fig. 6), and it would also agree with the inability of strain MPT to grow on l-malic acid (Fig. 5). MleR is a member of the LysR family of transcriptional regulators. LysR transcriptional regulators are cytoplasmic proteins constituted by an N-terminal DNA-binding domain and a C-terminal sensory domain. A common feature of these regulators is the need of a coinducer for transcriptional regulation (36). Our results strongly suggest that malate is the coinducer of MleR.

In contrast, inactivation of the transporters did not affect induction of maeE (Fig. 4). The expression of the genes maeE and maeP is under the control of the citrate family two-component system (TCS) constituted by MaeR and MaeK (3). For some other TCSs of the citrate family, the requirement of a cosensor, usually a cognate transporter or an ancillary solute-binding protein, has been described (37). Our results indicate that neither MaeP nor MleT is required for the induction of mae genes by MaeK/MaeR, and no gene encoding a putative solute-binding protein is located near the mae gene cluster. Therefore, these results suggest that the Mae TCS would not requires a cosensor. Furthermore, this TCS possibly senses the presence of extracellular l-malate, since uptake is not required to activate the transcription of maeP and maeE genes.

Inactivation of MleR did not affect expression of mae genes and, equally, inactivation of MaeR did not affect expression of mle genes. These results indicate that each pathway is independently regulated at the transcriptional level. However, at a functional level, both pathways are related since the data for the accumulation of malate showed that inactivation of any of the putative transporter encoding genes resulted in significant decreases in malate accumulation (Fig. 6). A schematic representation of this regulation and the contribution of both l-malate transporters to l-malate metabolism is presented in the Fig. S5 in the supplemental material. It is worth noting here that strain MR (mleR), which does not induce the production of MleT in the presence of l-malic acid, showed the best growth in MEIM (Fig. 7), and also no significant differences were observed in l-malate uptake with the parental strain (Fig. 6). This result indicates that a basal level of MleT is enough to provide optimal transport of l-malic acid under our assay conditions.

Analysis of the growth of Lb. casei BL23 and derivative strains in MEIM indicates that the interplay between both l-malic acid degradation pathways is more complex and has remarkable effects on Lb. casei physiology. The wild-type strain and a strain with an mleT mutation grew poorly in MEIM compared to strains with mleS, mleR, or Δmle mutations. The poor growth of the mleT strain cannot be attributed only to the absence of MleT, since this effect is not observed in the Δmle strain, which also lacks MleT. Monitoring of l-malic acid degradation and the production of lactic acid and acetic acid showed that l-malic acid consumption by wild-type and mleT strains was not coupled to growth and that most of the l-malic acid was degraded to lactic acid with a concomitant increase in pH (Fig. 7). These results strongly suggest that these strains consumed l-malic acid mainly via MLE and, as a consequence, little l-malic acid was available to enter biosynthetic pathways via ME. This hypothesis would also explain why strains lacking or producing at a low level MLE (i.e., MS [mleS] and MR [mleR] strains) grew faster and reached higher OD. In these strains, l-malic acid was metabolized via ME into pyruvate, which could subsequently be used to produce energy or channeled to biosynthetic pathways (see Fig. S4 in the supplemental material). The high production of acetate and the low production of lactic acid by these strains would agree with this hypothesis. Therefore, utilization of l-malic acid via MLE results in the wasteful degradation of this compound under this growth condition.

Notwithstanding, comparison of the growth of strains with mleS, mleR, or Δmle mutations also suggests that MLE doe not only have a detrimental effect on growth. All strains displayed a biphasic growth in MEIM (Fig. 7). The growth rate in the second growth stage varied between strains. Strain MS (mleS) had the lowest growth rate, and strain MR (mleR) had the highest growth rate. No significant differences were observed in the final concentrations of lactic acid and acetic acid. The mleR mutant does not induce the expression of genes encoding MLE and MleT; however, these proteins were produced at a basal level, as evidenced by the malate accumulation assay (Fig. 6). Therefore, these data suggest that a basal level of expression of mleS reduces the lag phase. The data available are insufficient to explain the lag phase observed, but the observation that the mleS mutant (which lacks MLE but produces the transporter MleT) displayed the longest lag phase suggests that an imbalance between l-malic acid uptake and metabolization may account for this observation.

In the present study we have shown that MleR and MaeR are two transcriptional activators that specifically regulate the mle and mae gene clusters, respectively, in Lb. casei. Although no cross talk exists in the regulation of these gene clusters, both l-malate transporters (MleT and MaeP) are required for maximal growth on l-malate. Utilization of l-malic acid as a carbon and energy source in Lb. casei BL23 is very inefficient, and it is subjected to strong catabolite repression. Thus, in this microorganism l-malate metabolism is optimized for raising external pH by conversion of l-malic acid to l-lactic acid via MLE. Rapid l-malate catabolism via MLE provides an advantage by preventing excessive acidification of the medium when cometabolized with substrates such as glucose, but it greatly diminishes the ability to grow on l-malic acid as a carbon source of Lb. casei.