Cross-talk between the L-sorbose and D-sorbitol (D-glucitol) metabolic pathways in Lactobacillus caseia

María J. Yebra1 and Gaspar Pérez-Martínez1

Departamento de Biotecnología, Instituto de Agroquímica y Tecnología de Alimentos, CSIC, Apdo. Correos 73, 46100 Burjassot, Spain1

Author for correspondence: Gaspar Pérez-Martínez. Tel: +34 96 390 0022. Fax: +34 96 363 6301. e-mail: gaspar.perez{at}iata.csic.es


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
A gene encoding sorbitol-6-phosphate dehydrogenase (SorF) belonging to the sorbose operon (sorFABCDG) has been characterized in Lactobacillus casei. Inactivation of this gene revealed the presence of another sorbitol-6-phosphate dehydrogenase that was induced by D-sorbitol (D-glucitol). The gene encoding this activity (gutF) has also been isolated, sequenced and disrupted. The sorbitol-6-phosphate dehydrogenase genes (sorF, gutF) were required for growth on L-sorbose and D-sorbitol, respectively. Biochemical and transcriptional analyses of the wild-type and mutant strains demonstrated that L-sorbose and D-sorbitol induced sorF and the gene encoding the sorbose operon activator (sorR), while the expression of gutF was only activated by D-sorbitol. Furthermore, these studies indirectly suggested that a common metabolite of the L-sorbose and D-sorbitol metabolic pathways (probably D-sorbitol 6-phosphate) would act as the effector of SorR. The same effector would also be the inducer of gutF, although the two pathways seem to be subject to distinct regulatory mechanisms.

Keywords: sorbitol-6-phosphate dehydrogenase, sugar catabolism, gene regulation

Abbreviations: PTS, phosphotransferase system; Sor-PR, L-sorbose-1-phosphate reductase; Stol-PDh, D-sorbitol-6-phosphate dehydrogenase

a The GenBank accession number for the sequence reported in this paper is AF396831.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Lactobacillus casei is a facultative heterofermentative lactic acid bacterium associated with fermented dairy products, which has received considerable attention in recent years due to its claimed properties as a probiotic (Aso & Akazan, 1992 ; Pedone et al., 2000 ; Forestier et al., 2001 ). Probiotic bacteria can efficiently use slowly absorbable and nonabsorbable carbohydrates as carbon sources, which helps to maintain a high population in the intestine. Hence, these carbohydrates are considered as prebiotics. Among them, the sugar alcohol D-sorbitol (D-glucitol) has been used as an alternative to oligosaccharides (Salminen et al., 1998 ). The genetics of D-sorbitol catabolism was first studied in Escherichia coli (Yamada & Saier, 1987 ). The sorbitol-specific enzyme II of the phosphoenolpyruvate-dependent phosphotransferase system (PTS) transports and phosphorylates D-sorbitol to D-sorbitol 6-phosphate, which is converted to D-fructose 6-phosphate by the D-sorbitol-6-phosphate dehydrogenase (Stol-PDh) (Fig. 1). Recently, other PTS-dependent catabolic pathways for D-sorbitol have been studied in the plant pathogen Erwinia amylovora (Aldridge et al., 1997 ) and in two Gram-positive bacteria, Clostridium beijerinckii (Tangney et al., 1998 ) and Streptococcus mutans (Boyd et al., 2000 ). As in Esc. coli, the metabolism of D-sorbitol in those bacteria involves a Stol-PDh which generates D-fructose 6-phosphate from D-sorbitol 6-phosphate. In lactobacilli, transport and metabolism of D-sorbitol had not yet been investigated. However, a L. casei mutant deficient in the general PTS component enzyme I was unable to ferment D-sorbitol (Viana et al., 2000 ), suggesting that in L. casei this polyol was also translocated by a PTS-like transporter.



View larger version (21K):
[in this window]
[in a new window]
 
Fig. 1. Schematic representation of the transport and metabolic pathways of L-sorbose and D-sorbitol. Sor-PR and Stol-PDh indicate L-sorbose-1-phosphate reductase and D-sorbitol-6-phosphate dehydrogenase, respectively.

 
Cloning, sequence analysis and characterization of the sorbose operon in L. casei has recently been reported (Yebra et al., 2000 ). During PTS transport, L-sorbose is phosphorylated to L-sorbose 1-phosphate, which is reduced to D-sorbitol 6-phosphate by an L-sorbose-1-phosphate reductase (Sor-PR), encoded by sorE. In a NAD+-dependent step, the Stol-PDh, encoded by sorF, catalyses the oxidation of D-sorbitol 6-phosphate to D-fructose 6-phosphate, which would be further catabolized by the glycolytic enzymes (Fig. 1). In addition, this operon included a distal sorG gene that possibly encodes a D-fructose-1,6-bisphosphate aldolase specific for L-sorbose catabolism. The sorbose genes are organized in two clusters, sorRE and sorFABCDG, which are induced by L-sorbose and repressed by D-glucose. sorR encodes a transcriptional activator that regulates both sor clusters. Analysis of the effects of mutations in sorBC, impaired in EIISor, and a previously characterized man mutant that is deficient in EIIMan (Veyrat et al., 1994 ), showed that L-sorbose is also transported by the D-mannose PTS in L. casei (Yebra et al., 2000 ). This article reports the cloning and sequence analysis of a gene encoding a second Stol-PDh (gutF) isolated from L. casei CECT 5275 (formerly ATCC 393 [pLZ15-]) (strain BL23). Sugar fermentation patterns of strains BL23F1 (sorF), BL23F2 (gutF::pRV300) and BL23F1-2 (sorF gutF::pRV300) showed that sorF and gutF are involved in L-sorbose and D-sorbitol metabolism in L. casei. Expression analysis of the two Stol-PDhs in strains BL23, BL23F1, BL23F2 and BL23R (sorR::pRV300) suggested that the metabolic pathways of the L-sorbose and D-sorbitol are interconnected at the regulation level.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains, culture conditions and plasmids.
The Lactobacillus casei and Escherichia coli strains used in this study are listed in Table 1. L. casei was grown at 37 °C under static conditions on MRS medium (Oxoid), MRS basal medium (Veyrat et al., 1994 ) or MRS fermentation medium (Adsa-Micro) supplemented with 0·5% sugar, as indicated. For growth pattern determination, 50 ml MRS basal medium was supplemented with 0·5% L-sorbose, 0·5% D-sorbitol, 0·5% D-ribose, 0·5% L-sorbose plus 0·5% D-ribose or 0·5% D-sorbitol plus 0·5% D-ribose. The tubes were inoculated to an initial OD550 of 0·1 and samples were taken at time intervals as indicated, to measure the growth. E. coli was used as the host in cloning experiments and it was grown in Luria–Bertani medium at 37 °C with vigorous shaking. Agar plates containing the same medium were prepared. E. coli and L. casei transformants were selected with ampicillin (50 µg ml-1) and erythromycin (5 µg ml-1), respectively.


View this table:
[in this window]
[in a new window]
 
Table 1. Strains and plasmids used in this study

 
The L. casei library (Gosalbes et al., 1997 ) was constructed in pJDC9 (Chen & Morrison, 1988 ). The vector pRV300 (Leloup et al., 1997 ) was used in E. coli for cloning purposes and as an integrative plasmid for insertional inactivation of genes in the L. casei chromosome. E. coli strains were transformed by electroporation with a Gene Pulser apparatus (Bio-Rad) as recommended by the manufacturer. L. casei was transformed as previously described by Posno et al. (1991) .

DNA manipulation and sequencing.
To amplify genes encoding Stol-PDh enzymes from L. casei by PCR, total DNA was used as the template with two synthetic primers, gut1 (5'-GGSGGHKCMTCDGGWATTGG-3') and gut2 (5'-ACKCCVKKRATRTARCTDGC-3'), designed from the conserved sequence regions of Stol-PDhs from E. coli (Yamada & Saier, 1987 ), Klebsiella pneumoniae (Wehmeier & Lengeler, 1994 ), C. beijerinckii (Tangney et al., 1998 ) and L. casei (Yebra et al., 2000 ). Amplified DNA fragments of the expected size were cloned into pRV300 and transformed into E. coli. Recombinant DNA techniques, plasmid DNA isolation and reverse PCR experiments were performed by standard procedures (Sambrook et al., 1989 ). DNA sequencing was carried out by using the ABI PRISM dRhodamine Terminator Cycle Sequencing Ready Reaction kit with AmpliTaq DNA polymerase and an automatic ABI 310 DNA sequencer (Applied Biosystems). M13 universal and reverse primers or primers annealing within the cloned DNA were used to sequence both strands of the DNA. Alignment of the sequences and analysis of the sequence for ORFs were carried out with version 4.0 of the DNAMAN for windows (Lynnon BioSoft). Sequence similarities were analysed with the BLAST program.

Chromosomal inactivation of the Stol-PDh genes.
The 490 bp PstI–HindIII sorF fragment was isolated from the original clone in pJDC9 (Chen & Morrison, 1988 ) and subcloned into the integrative vector pRV300. The resulting plasmid was digested with NheI, the generated 3' recessed ends were filled-in with the DNA Polymerase I Large (Klenow) Fragment and then it was self-ligated to render a frameshift in the sorF gene. The plasmid was then used to transform L. casei and the transformants were selected on agar plates with erythromycin. Chromosomal integration of this plasmid was confirmed by PCR analysis. Subsequently, one of the integrants was selected and used to inoculate fresh MRS medium without antibiotic. After growth for approximately 200 generations, appropriate dilutions were replicated onto agar plates with and without erythromycin. Strains that had undergone the second recombination and had therefore lost the integrated plasmid were selected as erythromycin sensitive. The frameshift mutation introduced in sorF was confirmed by PCR analysis and DNA sequencing, and one mutant strain named BL23F1 was selected.

The plasmid pRVgut3 (Table 1) containing a 720 bp PCR fragment corresponding to gutF was electroporated into L. casei and the transformants were selected on agar plates with erythromycin. Chromosomal integration of this plasmid was checked by PCR analysis and DNA sequencing. The sugar fermentation pattern of the integrants was analysed and one of them, designated BL23F2, was selected. The same procedure was used to inactivate the gutF gene in the sorF mutant (BL23F1) obtaining the double mutant gutF sorF (BL23F1-2).

Enzyme activity tests.
L. casei strains were grown in MRS fermentation medium supplemented with 0·5% of the appropriate sugar to OD550 0·8, then cells were harvested, washed and resuspended in 1 ml buffer (10 mM Tris/HCl pH 7·5, 1 mM DTT, 0·1 mM PMSF). Crude extracts were prepared by disruption of the cells with glass beads in a Fastprep apparatus (Savant Instruments) for four periods of 1 min with intervals of 1 min on ice. Sor-PR activity was determined by the method of Wöhrl & Lengeler (1990) , using 2·5 mM D-fructose 1-phosphate as substrate. Stol-PDh activity was determined as described by Yebra et al. (2000) using 2 mM D-sorbitol 6-phosphate (D-glucitol 6-phosphate) as substrate. The rates of NADH oxidation (for Sor-PR) and NAD+ reduction (for Stol-PDh) were determined by measuring the change of absorbance at 340 nm. NADH oxidase activity was determined in Tris/HCl buffer (pH 7·5) and 0·1 mM NADH. NADH oxidation was monitored in the reaction mixture (0·5 ml) by following the absorbance at 340 nm. Specific enzyme activities are given in nmol min-1 (mg protein)-1. Protein concentrations were measured by the method of Bradford (1976) .

Southern and Northern blot analysis.
DNA probes were synthesized by PCR using L. casei chromosomal DNA as template, deoxynucleoside triphosphate mix with digoxigenin-dUTP from the Roche digoxigenin-labelling kit and Taq (Expand High Fidelity PCR System) from Roche. Each probe was amplified with two internal primers: sorR probe with sorR1 (5'-AAGGCGTTGTTTCAATTGCC-3') and sorR3 (5'-ACACGTGTCCACAGGCTAAG-3'), sorF probe with sorF1 (5'-AGCAGAGAAGATTAATGGC-3') and manR21 (5'-TAGGATAATTGGTGCGCTAAAGG-3'), gutF with gut3 (5'-CGACTTTAAGGACCCTAA-3') and gut4 (5'-GACGACAACCCCATGTCC-3'). Chromosomal DNA from L. casei (6 µg) was digested with the restriction enzymes HindIII or SphI and the same amount of chromosomal DNA from E. coli was digested with SphI. The cleaved DNAs were separated on a 0·7% agarose gel, transferred to a nylon membrane and hybridized with the probes as described in Sambrook et al. (1989) . Total RNA was isolated from L. casei cells grown in MRS fermentation medium with 0·5 % appropriate sugars to OD550 0·8. Cells were collected by centrifugation, washed with 50 mM EDTA, pH 8·0, resuspended in 1 ml TRIZOL Reagent (Gibco-BRL) and then mechanically disrupted with glass beads in a Cell Disruptor (Savant Instruments). Total RNA was isolated according to the protocol of the TRIZOL manufacturer. Sample preparation, denaturing agarose gel electrophoresis and RNA transfer were performed by standard methods (Sambrook et al., 1989 ).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Sorbitol induction of Stol-PDh activities in L. casei
The metabolic relationship between D-sorbitol and L-sorbose was investigated in L. casei. The wild-type and different mutants of the sorbose operon (strains BL23, BL23S, BL23R, BL23D, BL23DS) could ferment sorbitol on agar plates (Table 2). Transcription of the sorbose genes is induced by L-sorbose through the action of the activator SorR and repressed by CcpA-mediated carbon catabolite repression (Yebra et al., 2000 ), hence, D-ribose was taken as the reference carbon source in the following experiments, for it is a non-PTS and non-repressing sugar (Monedero et al., 1997 ; Viana et al., 2000 ). L. casei wild-type and mutants were grown in liquid broth with D-ribose and D-ribose plus D-sorbitol, to test Stol-PDh activities (Table 2). Experimental data confirmed that Stol-PDh activity of ribose-grown cells was below the detection level. Interestingly, when D-sorbitol was added to the culture medium, Stol-PDh was induced in all strains tested, including BL23R where sorF-encoded Stol-PDh is not expressed (Yebra et al., 2000 ), indicating that L. casei may contain an additional Stol-PDh.


View this table:
[in this window]
[in a new window]
 
Table 2. Sorbitol and sorbose phenotypes and D-sorbitol-6-phosphate dehydrogenase activity in L. casei wild-type and man, sorBC and sorR mutant strains

 
Sequence analysis of a second Stol-PDh in L. casei
In view of the previous results, a screening of the L. casei genome was set up by PCR using degenerate oligonucleotides. Amino acid sequence alignment of the Stol-PDhs from E. coli (Yamada & Saier, 1987 ), K. pneumoniae (Wehmeier & Lengeler, 1994 ), C. beijerinckii (Tangney et al., 1998 ) and L. casei (Yebra et al., 2000 ) allowed us to design two primers from conserved sequence regions. PCR amplification of L. casei DNA with these primers yielded a DNA fragment with the expected size (720 bp), which was cloned in pRV300 and transformed into E. coli. Sequence analysis of the 720 bp insert in several clones revealed at least two types of clone. One type contained a DNA fragment identical to that of sorF from the sorbose operon (Yebra et al., 2000 ). The other type had some differences in the DNA sequence, its deduced amino acid sequence also showed high similarity to Stol-PDhs. New primers were used in reverse PCR amplifications to sequence a 110 bp fragment downstream of the 3' end and a 316 bp fragment upstream of the 5' end. Sequence analysis of the entire DNA fragment showed one ORF (288 aa) with a putative ribosome-binding site (Shine & Dalgarno, 1974 ) 6 nt upstream of the potential ATG start codon. The encoded protein shared 85% identity with the Stol-PDh from the sorbose operon of L. casei (Yebra et al., 2000 ) and 62% identity to the Stol-PDhs associated with sorbitol metabolism of C. beijerinckii (Tangney et al., 1998 ) and S. mutans (Boyd et al., 2000 ). This gene will hereafter be named gutF.

Southern blot analysis
To corroborate the presence of sorF and gutF in the L. casei genome, we used internal DNA fragments of each gene as probes in Southern hybridization experiments (Fig. 2). Chromosomal DNA of L. casei cleaved with HindIII showed bands of approximately 4 kb and 1·9 kb in size when sorF was used as a probe, and a single band of about 5 kb with gutF as a probe (Fig. 2, lanes 1a and 1b). Lanes containing SphI digests of L. casei DNA showed bands of 6 kb and 2·3 kb when sorF and gutF fragments were used as probes, respectively (Fig. 2, lanes 2a and 2b). The bands obtained with HindIII and SphI digests had fragments with expected sizes in the regions of the L. casei genome that contain the sorF gene (Yebra et al., 2000 ) or the gutF gene (genomic region known by reverse PCR analysis). The probes did not hybridize with E. coli chromosomal DNA (Fig. 2, lanes 3a and 3b). Although the alignment of the DNA sequences of sorF and gutF genes showed a high homology (74% identity), there was no cross-hybridization after overexposing the membranes to X-ray films (data not shown).



View larger version (101K):
[in this window]
[in a new window]
 
Fig. 2. Southern blot analysis using an internal DNA fragment of sorF (a) or gutF (b) as a probe. Six micrograms of genomic DNA from L. casei were digested with HindIII (lanes 1) or SphI (lanes 2), and from E. coli were digested with HindIII (lanes 3).

 
Chromosomal inactivation of sorF and gutF, and phenotype of the mutants
To ascertain the biological function of sorF in L-sorbose metabolism, a frameshift mutation was introduced at the first NheI restriction site of the sorF gene in the chromosome of L. casei by two sequential recombination steps. The BL23F1 mutant had lost its ability to ferment L-sorbose on MRS fermentation medium, but was still able to use D-sorbitol as the sole carbon source in the culture medium (Table 3). Then, the suicide vector pRVgut3 (Table 1) was also used to transform L. casei and the integrants obtained were studied for their ability to ferment L-sorbose or D-sorbitol on MRS fermentation medium. All integrants could grow on L-sorbose but not on D-sorbitol and one of them, named BL23F2, was selected for further analysis (Table 3). Additionally, a double mutant was obtained by disruption of the gutF gene in the BL23F1 mutant with the pRVgut3 construct as before. The resulting strain, named BL23F1-2, was unable to use L-sorbose or D-sorbitol as sole carbon source (Table 3).


View this table:
[in this window]
[in a new window]
 
Table 3. D-Sorbitol-6-phosphate dehydrogenase activity in L. casei mutant strains

 
Using the enzyme activity test described in Methods, Stol-PDh activity could not be detected in BL23 grown on D-sorbitol as the only carbon source (data not shown). This was due to a high NADH oxidase activity level [1573 nmol min-1 (mg protein)-1] measured in the cells cultured on D-sorbitol, but not on D-ribose [49 nmol min-1 (mg protein)-1]. It has been previously shown that S. mutans is able to ferment sugar alcohols D-mannitol and D-sorbitol with the concomitant production of NADH. In that strain a H2O-forming NADH oxidase has been identified, which is indispensable for the regeneration of NAD+ during aerobic D-mannitol metabolism (Higuchi et al., 1999 ). To avoid this effect, Stol-PDh activity was determined in strains BL23, BL23F1 and BL23F2 grown on MRS fermentation medium supplemented with D-ribose plus L-sorbose or D-ribose plus D-sorbitol (Table 3). All the strains showed a very high activity when grown on L-sorbose and surprisingly, this included also BL23F1. This could be due to a high induction of gutF expression, possibly as a result of accumulation of D-sorbitol 6-phosphate. The activity level of BL23F2 grown on D-ribose plus L-sorbose was similar to that of the wild-type. Furthermore, Stol-PDh activity in strains BL23F1, BL23F2 and BL23F1-2 was also determined in the presence of D-sorbitol. A remarkable activity was found in the double mutant, which could indicate the presence in L. casei of yet another enzyme with Stol-PDh activity.

Growth pattern of L. casei strains
The growth pattern of L. casei wild-type (BL23) and mutant strains on MRS basal medium supplemented with L-sorbose, D-sorbitol, D-ribose or a mixture of these sugars is shown in Fig. 3. BL23F1 (sorF), BL23F2 (gutF::pRV300) and BL23F1-2 (sorF gutF::pRV300) showed the same pattern of growth on D-ribose as the wild-type (data not shown). Both BL23 and BL23F2 could grow efficiently on L-sorbose, but BL23F1 and BL23F1-2 hardly proliferated (Fig. 3a). On MRS basal medium with D-sorbitol, BL23F1 displayed a slower growth rate than the wild-type, and BL23F2 and BL23F1-2 could barely grow (Fig. 3b). This indicated that sorF may also participate in D-sorbitol catabolism. When L-sorbose was supplemented with D-ribose, again BL23 and BL23F2 had a similar growth pattern; however, BL23F1 could grow to a high density, although with a lower growth rate than the wild-type on D-ribose (Fig. 3c). This could be due to a transient accumulation of sorbose 1-phosphate and sorbitol 6-phosphate in the cells, which could have a toxic effect. Growth of BL23F1-2 on L-sorbose plus D-ribose was very limited. In this strain, sorbose 1-phosphate may be accumulated with consequent toxicity for the cells. However, in the presence of D-sorbitol plus D-ribose, the double mutant showed a growth pattern similar to the wild-type with only D-ribose (Fig. 3d).



View larger version (25K):
[in this window]
[in a new window]
 
Fig. 3. Growth patterns of L. casei on MRS basal medium containing 0·5% L-sorbose (a), 0·5% D-sorbitol (D-glucitol) (b), 0·5% L-sorbose plus 0·5% D-ribose (c) or 0·5% D-sorbitol plus 0·5% D-ribose (d). {bullet}, Wild-type (BL23); {blacktriangledown}, sorF mutant (BL23F1); {blacksquare}, gutF mutant (BL23F2); {square}, sorF gutF double mutant (BL23F1-2). In (c) and (d) {circ} represents the growth of the wild-type on MRS basal medium supplemented with 0·5% D-ribose.

 
Effect of the different mutations on the transcription of sorR, sorF and gutF genes in L. casei
It has been previously mentioned that L-sorbose acts as an inducer of the sorbose metabolic operon (sorRE and sorFABCDG). The transcription patterns of sorR, sorF and gutF have been analysed by Northern blot in L. casei wild-type and mutant strains BL23R (sorR::pRV300) (Yebra et al., 2000 ), BL23F1 (sorF) and BL23F2 (gutF::pRV300) grown on D-sorbitol, D-ribose plus L-sorbose or D-ribose plus D-sorbitol. Samples of the mutants were not included in blots prepared with probes directed against their respective mutated genes. On D-sorbitol gutF, sorF and sorR were efficiently expressed, indicating that the enzymes encoded by the sor genes may also participate in the catabolism of D-sorbitol in L. casei (Fig. 4). When gutF was used as probe, a major transcript of 2·4 kb was detected (shown in Fig. 4), indicating that (an)other gene(s) may be transcribed with it. After overexposure of the X-ray film, several additional minor bands appeared (data not shown). The existence of a faint band of approximately 6·3 kb could suggest that gutF was part of a large operon, as described in other micro-organisms (Yamada & Saier, 1987 ; Aldridge et al., 1997 ; Boyd et al., 2000 ).



View larger version (73K):
[in this window]
[in a new window]
 
Fig. 4. Northern blot analysis using an internal DNA fragment of sorR (a), sorF (b) or gutF (c) as the probe. RNA (1 µg) was isolated from L. casei cells grown with 0·5% D-sorbitol (D-glucitol), L-sorbose plus 0·5% D-ribose and with 0·5% D-sorbitol plus 0·5% D-ribose. WT, sorF, gutF and sorR indicate L. casei wild-type, BL23F1 (sorF), BL23F2 (gutF::pRV300) and BL23R (sorR::pRV300) strains, respectively.

 
Transcription was also studied in the presence of D-ribose to compare these results with previous activity data. sorR mRNA was detected in the wild-type, BL23F1 (sorF) and BL23F2 (gutF) grown on D-ribose plus L-sorbose (Fig. 4a). On D-ribose plus D-sorbitol, no signal was observed in the wild-type; however, there were transcription products in BL23F1 and BL23F2, which is an indication that sorbitol 6-phosphate accumulated in both mutants could be the effector of the transcriptional activation of sorR (Fig. 4a). With the sorF probe, a strong signal was obtained in BL23 and BL23F2 grown on L-sorbose but no signal was found in BL23R (Fig. 4b). These last results corroborated that SorR is a positive regulator of the sorFABCDG operon. Weaker transcription bands were observed on D-ribose plus D-sorbitol, although the signal was stronger in BL23F2 (gutF) which again links the accumulation of sorbitol 6-phosphate to the activation of sorR and that of sorF (Fig. 4b). Fig. 4c shows the hybridization bands obtained with the gutF probe. On L-sorbose plus D-ribose only BL23F1 expressed gutF mRNA, which explains why BL23F1 could grow and displayed a high Stol-PDh activity on this sugar. In fact these data suggest that sorF encodes the Stol-PDh activity which is preferentially used by L. casei. Only in the sorF and sorR mutants, where sorF is not expressed, is gutF highly transcribed. When D-sorbitol and D-ribose were added to the culture medium, gutF was only noticeably transcribed in BL23F1. It is very evident that the presence of D-ribose in the culture medium reduced the expression of sorR, sorF and gutF. It was previously shown that Sor-PR and Stol-PDh activity levels decreased in the wild-type when the culture medium contained D-ribose in addition to L-sorbose (Yebra et al., 2000 ). This effect caused by the presence of D-ribose could be due to a yet unknown regulatory mechanism.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Close biochemical relationships are inherent to the L-sorbose and D-sorbitol metabolic pathways in Gram-negative bacteria (Sprenger & Lengeler, 1984 ; Yamada & Saier, 1987 ; Wehmeier et al., 1992 , 1995 ). They share a common enzymic step catalysed by a Stol-PDh that converts D-sorbitol 6-phosphate to D-fructose 6-phosphate with the concomitant reduction of NAD+ to NADH (Fig. 1). In fact, the gene encoding the sorbose-specific Stol-PDh could complement the lack of the sorbitol-specific Stol-PDh in E. coli (Sprenger & Lengeler, 1984 ; Wehmeier et al., 1992 ).

Certain Lactobacillus species can utilize L-sorbose and D-sorbitol, among them some strains of the L. casei–paracasei group (Acedo & Perez-Martínez, unpublished). During the study of the sorbose operon of L. casei CECT 5275 (formerly ATCC 393 [pLZ15-]), different mutants were obtained lacking the PTS transport elements (sorBC), the activator of the operon (sorR) and Stol-PDh (sorF) (Yebra et al., 2000 ). They constituted an excellent ground to start the study of the biochemical and possible regulatory links with the D-sorbitol metabolic pathway. In this work, it soon became evident that in L. casei BL23F1 (sorF) there was at least another Stol-PDh activity directly related to the consumption of D-sorbitol and induced by this sugar-alcohol. Then, a gene encoding a new Stol-PDh in L. casei was identified and sequenced. This gene, gutF, showed sequence differences with the previously characterized sorF, encoding the Stol-PDh of the sorbose operon (Yebra et al., 2000 ). Chromosomal inactivation of these two genes and analysis of the growth patterns on L-sorbose or D-sorbitol suggested that sorF could also be involved in the metabolism of D-sorbitol and gutF would only be involved in the catabolism of D-sorbitol. These data were confirmed by the transcription analysis in the wild-type, where sorF was transcribed at a very high rate on L-sorbose and on D-sorbitol (with and without D-ribose), while gutF was only transcribed on D-sorbitol. This may indicate that sorF provides the preferred Stol-PDh in L. casei. Also a double sorF gutF mutant was generated that showed a Sor-Gut- phenotype, but interestingly, this strain still had a significant Stol-PDh activity, suggesting that yet another enzyme could be present in this micro-organism with this activity profile. Then, the negative phenotype of gutF mutants on D-sorbitol might be explained by a likely polar effect of the plasmid insertion in gutF on the transcription of other D-sorbitol metabolic genes, such as PTS transporters (Yamada & Saier, 1987 ; Aldridge et al., 1997 ; Boyd et al., 2000 ).

The growth pattern of the different strains on liquid broth further illustrated the data obtained by growth on agar plates. BL23F1 (sorF) was unable to grow on L-sorbose; however, addition of D-ribose favoured a slow but steady growth, suggesting that L-sorbose may be transported by its own PTSSor elements or even through PTSMan (Yebra et al., 2000 ) and metabolized through the weak expression of sorE (M. J. Yebra, unpublished) and gutF (Fig. 4c). A poisoning effect has been attributed to sorbose 1-phosphate in E. coli (Wehmeier et al., 1992 ). Hence, transient accumulation of this sugar phosphate could prevent growth of this mutant on L-sorbose and may explain the growth delay when D-ribose was added. The difference in growth pattern between D-ribose plus L-sorbose and D-ribose plus D-sorbitol observed in the sorF gutF double mutant could also be explained by the toxic accumulation of sorbose 1-phosphate.

D-Ribose was chosen in this work as neutral carbon source (not inducing nor repressing sugar) to facilitate certain strains to grow on L-sorbose or D-sorbitol when they carried mutations in their respective pathways. It was also required to obtain cell batches for Stol-PDh quantification, where NADH oxidase activity was not induced. However, these experiments had an unexpected interference due to the inhibitory effect of D-ribose on the metabolism of D-sorbitol detected in the growth curves. This effect was taking place at the transcriptional level, since gutF mRNA was hardly expressed in the wild-type and sorR strains when D-ribose was present.

Additional difficulties were encountered in this work to determine Stol-PDh activity in D-sorbitol-grown cells by a standard NADH-consumption reaction, because the indigenous NADH-oxidase levels were 103-fold higher than in the control. The remarkable differences between the Stol-PDh activities of different strains grown on L-sorbose in Table 3 were possibly due to L-sorbose induction of sorbose-1-phosphate reductase in the sorbose operon. This enzyme would regenerate NAD+ and partially restore the redox balance in the cell. Full expression of sorF under these conditions, and especially of that of gutF, was also confirmed by the Northern blot assay, indicating an interesting relationship between the expression of Stol-PDh enzymes and the intracellular redox potential. This will be expanded in future studies.

The transcriptional activator of the sorbose operon in L. casei, SorR, showed significant amino acid sequence similarity to regulatory proteins such as SorC from K. pneumoniae, DeoR from Bacillus subtilis and YgaP from Bacillus megaterium (Yebra et al., 2000 ). These transcriptional regulators control the transcription of genes involved in phosphorylated sugar catabolism and, in the case of DeoR from B. subtilis, its binding to DNA is modulated by D-ribose 5-phosphate (Zeng et al., 2000 ). This work confirmed previous results (Yebra et al., 2000 ) indicating that SorR positively regulated expression of sorF at the transcriptional level during growth on L-sorbose. D-Sorbitol uptake in L casei occurs through PTS elements (Viana et al., 2000 ), possibly yielding sorbitol 6-phosphate, as in other bacteria (Yamada & Saier, 1987 ; Aldridge et al., 1997 ; Tangney et al., 1998 ; Boyd et al., 2000 ). Findings of this work are not just related to D-sorbitol metabolism, but have a significant impact on the mechanism of regulation of the sorbose operon. Three facts indicated that D-sorbitol 6-phosphate could be the actual effector of SorR: (i) in the wild-type, D-sorbitol induced expression of sorF as strongly as L-sorbose; (ii) inactivation of sorR inhibited expression of sorF on D-sorbitol and L-sorbose, indicating that both sugars induce SorR activation through the same effector; and (iii) in strain BL23F1 (sorF), D-sorbitol led to a higher expression of sorR (SorR is known to activate also its own transcription (Yebra et al., 2000 )) possibly due to accumulation of D-sorbitol 6-phosphate.

In relation to the likely regulatory mechanism of gutF, it has been stated that D-sorbitol induced expression of gutF. In addition, when grown on D-ribose plus L-sorbose, or D-ribose plus D-sorbitol, gutF is only induced in the sorF mutant, suggesting that sorbitol 6-phosphate could also be the actual inducer of this gene. However, inactivation of sorR had no effect on the expression of gutF during growth on D-sorbitol, which indicated that this gene may rely on a regulator different from SorR.

In this paper, the genetic and functional characterization of the genes encoding two Stol-PDh, sorF and gutF, from L. casei has been achieved. Results of these studies indicated that cross-talk occurs between these metabolic pathways and the redox state of the cell and also the genes involved in the L-sorbose and D-sorbitol catabolism interact through a common effector in their regulatory mechanisms.


   ACKNOWLEDGEMENTS
 
This work was supported by the Spanish Comisión Interministerial de Ciencia y Tecnología Grant (ALI98-0714) and the Biotech Programme of the European Community (contract BIO4-CT96-0380).


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Aldridge, P., Metzger, M. & Geider, K. (1997). Genetics of sorbitol metabolism in Erwinia amylovora and its influence on bacterial virulence. Mol Gen Genet 256, 611-619.[Medline]

Aso, Y. & Akazan, H. (1992). Prophylactic effect of a Lactobacillus casei preparation on the recurrence of superficial bladder cancer. Urol Int 49, 125-129.[Medline]

Boyd, D. A., Thevenot, T., Gumbmann, M., Honeyman, A. L. & Hamilton, I. R. (2000). Identification of the operon for the sorbitol (glucitol) phosphoenolpyruvate:sugar phosphotransferase system in Streptococcus mutans. Infect Immun 68, 925-930.[Abstract/Free Full Text]

Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72, 248-254.[Medline]

Chen, J.-D. & Morrison, D. A. (1988). Construction and properties of a new insertion vector, pJDC9, that is protected by transcriptional terminators and useful for cloning of DNA from Streptococcus pneumoniae. Gene 64, 155-164.[Medline]

Forestier, C., De Champs, C., Vatoux, C. & Joly, B. (2001). Probiotic activities of Lactobacillus casei rhamnosus: in vitro adherence to intestinal cells and antimicrobial properties. Res Microbiol 152, 167-173.[Medline]

Gosalbes, M. J., Monedero, V., Alpert, C.-A. & Pérez-Martínez, G. (1997). Establishing a model to study the regulation of the lactose operon in Lactobacillus casei. FEMS Microbiol Lett 148, 83-89.[Medline]

Higuchi, M., Yamamoto, Y., Poole, L. B., Shimada, M., Sato, Y., Takahashi, N. & Kamio, Y. (1999). Functions of two types of NADH oxidases in energy metabolism and oxidative stress of Streptococcus mutans. J Bacteriol 181, 5940-5947.[Abstract/Free Full Text]

Leloup, L., Ehrlich, S. D., Zagorec, M. & Morel-Deville, F. (1997). Single-crossover integration in the Lactobacillus sake chromosome and insertional inactivation of the ptsI and lacL genes. Appl Environ Microbiol 63, 2117-2123.[Abstract]

Monedero, V., Gosalbes, M. J. & Pérez-Martínez, G. (1997). Catabolite repression in Lactobacillus casei ATCC 393 is mediated by CcpA. J Bacteriol 179, 6657-6664.[Abstract]

Pedone, C. A., Arnaud, C. C., Postaire, E. R., Bouley, C. F. & Reinert, P. (2000). Multicentric study of the effect of milk fermented by Lactobacillus casei on the incidence of diarrhoea. Int J Clin Pract 54, 568-571.[Medline]

Posno, M., Leer, R. J., van Luijk, N., van Giezen, M. J. F., Heuvelmans, P. T. H. M., Lokman, B. C. & Pouwels, P. H. (1991). Incompatibility of Lactobacillus vectors with replicons derived from small cryptic Lactobacillus plasmids and segregational instability of the introduced vectors. Appl Environ Microbiol 57, 1822-1828.

Salminen, S., Bouley, C., Boutron-Ruault, M. C. & 7 other authors (1998). Functional food science and gastrointestinal physiology and function. Br J Nutr 80 (suppl 1), s147–s171.

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Shine, J. & Dalgarno, L. (1974). The 3' terminal sequence of Escherichia coli 16S ribosomal RNA: complementarity to nonsense triplets and ribosome binding sites. Proc Natl Acad Sci USA 71, 1342-1346.[Abstract]

Sprenger, G. A. & Lengeler, J. W. (1984). L-sorbose metabolism in Klebsiella pneumoniae and Sor+derivatives of Escherichia coli K-12 and chemotaxis towards sorbose. J Bacteriol 157, 39-45.[Medline]

Tangney, M., Brehm, J. K., Minton, N. P. & Mitchell, W. J. (1998). A gene system for glucitol transport and metabolism in Clostridium beijerinckii NCIMB 8052. Appl Environ Microbiol 64, 1612-1619.[Abstract/Free Full Text]

Veyrat, A., Monedero, V. & Pérez-Martínez, G. (1994). Glucose transport by the phosphoenolpyruvate:mannose phosphotransferase system in Lactobacillus casei ATCC 393 and its role in carbon catabolite repression. Microbiology 140, 1141-1149.[Abstract]

Viana, R., Monedero, V., Dossonnet, V., Vadeboncoeur, C., Pérez-Martínez, G. & Deutscher, J. (2000). Enzyme I and HPr from Lactobacillus casei: their role in sugar transport, carbon catabolite repression and inducer exclusion. Mol Microbiol 36, 570-584.[Medline]

Wehmeier, U. F. & Lengeler, J. W. (1994). Sequence of the sor-operon for L-sorbose utilization from Klebsiella pneumoniae KAY2026. Biochim Biophys Acta 1208, 348-351.[Medline]

Wehmeier, U. F., Nobelmann, B. & Lengeler, J. W. (1992). Cloning of the Escherichia coli sor genes for L-sorbose transport and metabolism and physical mapping of the genes near metH and iclR. J Bacteriol 174, 7784-7790.[Abstract]

Wehmeier, U. F., Wöhrl, B. M. & Lengeler, J. W. (1995). Molecular analysis of the phosphoenolpyruvate-dependent L-sorbose:phosphotransferase system from Klebsiella pneumoniae and of its multidomain structure. Mol Gen Genet 246, 610-618.[Medline]

Wöhrl, B. M. & Lengeler, J. W. (1990). Cloning and physical mapping of the sor genes for L-sorbose transport and metabolism from Klebsiella pneumoniae. Mol Microbiol 4, 1557-1565.[Medline]

Yamada, M. & Saier, M. H.Jr (1987). Physical and genetic characterization of the glucitol operon in Escherichia coli. J Bacteriol 169, 2990-2994.[Medline]

Yebra, M. J., Veyrat, A., Santos, M. A. & Pérez-Martínez, G. (2000). Genetics of L-sorbose transport and metabolism in Lactobacillus casei. J Bacteriol 182, 155-163.[Abstract/Free Full Text]

Zeng, X., Saxild, H. H. & Switzer, R. L. (2000). Purification and characterization of the DeoR repressor of Bacillus subtilis. J Bacteriol 182, 1916-1922.[Abstract/Free Full Text]

Received 16 January 2002; revised 10 April 2002; accepted 8 May 2002.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Yebra, M. J.
Articles by Pérez-Martínez, G.
Articles citing this Article
PubMed
PubMed Citation
Articles by Yebra, M. J.
Articles by Pérez-Martínez, G.
Agricola
Articles by Yebra, M. J.
Articles by Pérez-Martínez, G.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2002 Society for General Microbiology.