The Bacillus subtilis ywkA gene encodes a malic enzyme and its transcription is activated by the YufL/YufM two-component system in response to malate

Thierry Doan1, Pascale Servant1,{dagger}, Shigeo Tojo2, Hirotake Yamaguchi2, Guillaume Lerondel1, Ken-Ichi Yoshida2, Yasutaro Fujita2 and Stéphane Aymerich1

1 Génétique Moléculaire et Cellulaire, INRA (UMR216) CNRS (URA1925) and INAP-G, F-78850 Thiverval-Grignon, France
2 Department of Biotechnology, Fukuyama University, 985 Sanzo, Higashimura, Fukuyama, Japan

Correspondence
Stéphane Aymerich
stef{at}grignon.inra.fr


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
A transcriptome comparison of a wild-type Bacillus subtilis strain growing under glycolytic or gluconeogenic conditions was performed. In particular, it revealed that the ywkA gene, one of the four paralogues putatively encoding a malic enzyme, was more transcribed during gluconeogenesis. Using a lacZ reporter fusion to the ywkA promoter, it was shown that ywkA was specifically induced by external malate and not subject to glucose catabolite repression. Northern analysis confirmed this expression pattern and demonstrated that ywkA is cotranscribed with the downstream ywkB gene. The ywkA gene product was purified and biochemical studies demonstrated its malic enzyme activity, which was 10-fold higher with NAD than with NADP (kcat/Km 102 and 10 s-1 mM-1, respectively). However, physiological tests with single and multiple mutant strains affected in ywkA and/or in ywkA paralogues showed that ywkA does not contribute to efficient utilization of malate for growth. Transposon mutagenesis allowed the identification of the uncharacterized YufL/YufM two-component system as being responsible for the control of ywkA expression. Genetic analysis and in vitro studies with purified YufM protein showed that YufM binds just upstream of ywkA promoter and activates ywkA transcription in response to the presence of malate in the extracellular medium, transmitted by YufL. ywkA and yufL/yufM could thus be renamed maeA for malic enzyme and malK/malR for malate kinase sensor/malate response regulator, respectively.


Abbreviations: TCA, tricarboxylic acid; TCS, two-component system

The online version of this paper (at http://mic.sgmjournals.org) contains a table listing the oligonucleotides used in this study and a figure showing the overproduction and purification of His6-YufM.

{dagger}Present address: Institut de Génétique et Microbiologie, Université Paris-Sud, Bât. 409, F-91405 Orsay, France.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Malic enzymes catalyse the reversible oxidative decarboxylation of malate into pyruvate with reduction of NAD or NADP. In eukaryotes, several isoforms of this enzyme have been characterized. Three classes of malic enzymes have been defined on the basis of their coenzyme specificity and their capability to also decarboxylate oxaloacetate (Kobayashi et al., 1989). A first class of NADP-dependent malic enzymes (EC 1.1.1.40), decarboxylating malate and also oxaloacetate, are found in cytosol, chloroplasts and mitochondria; a second group of enzymes, preferentially utilizing NAD (EC 1.1.1.38), are also capable of decarboxylating oxaloacetate and are found in bacteria and insects; finally, a third class of NAD-dependent malic enzymes (EC 1.1.1.39), unable to use oxaloacetate, have been found only in mitochondrial matrix. In prokaryotes, malic enzymes are also widely distributed (Murai et al., 1971; Diesterhaft & Freese, 1973; Lamed & Zeikus, 1981; Knichel & Radler, 1982; Bartolucci et al., 1987; Kobayashi et al., 1989; Kawai et al., 1996; Voegele et al., 1999). However, few of them have been characterized so far, and in particular, few reports allowed distinction of whether the observed malic enzyme activity resulted from one or several isozymes. In Escherichia coli, both NAD- and NADP-dependent malic enzymes, named SfcA and B2463 respectively, have been characterized (Murai et al., 1971; Hansen & Juni, 1974, 1975; Stols & Donnelly, 1997) but their respective physiological functions have not been precisely documented. It has been suggested that SfcA is involved in gluconeogenesis by providing pyruvate, then converted into phosphoenolpyruvate by a phosphoenolpyruvate synthase (Hansen & Juni, 1975), and that B2463 is involved in the supply of NADPH and acetyl-CoA, necessary for anabolic reactions during growth on C4-dicarboxylates as sole carbon sources (Murai et al., 1971; Hansen & Juni, 1974, 1975). The synthesis of both enzymes is repressed in the presence of glucose while the expression of sfcA, encoding the NAD-dependent malic enzyme, is inducible by malate (Murai et al., 1971). The combined deletion of the two genes leads to a severe growth defect only on C4-dicarboxylates, thus indicating that malic enzyme activity is necessary under such growth conditions in E. coli (van der Rest et al., 2000). Rhizobium meliloti, a nitrogen-fixing Gram-negative symbiont of alfalfa, also synthesizes two malic enzymes named DME and TME, which are NAD- and NADP-dependent, respectively (Driscoll & Finan, 1997; Mitsch et al., 1998). The former has been shown to be involved in symbiotic nitrogen fixation as a result of its role for growth in the nodule, where C4-dicarboxylates are the major carbon and energy source provided by the host (Driscoll & Finan, 1993; Dunn, 1998), while the latter has been proposed to function as a generator of NADPH necessary for biosynthesis reactions (Driscoll & Finan, 1996).

Interestingly, the analysis of the Bacillus subtilis complete genome sequence revealed the presence of four paralogous genes encoding putative malic enzymes: ywkA, malS, ytsJ and mleA. None of these genes has been previously characterized. Indeed, little information has been reported about malic enzyme activity in this bacterial species. Diesterhaft & Freese (1973) demonstrated that the presence of malate stimulates the synthesis of an NADP/NAD-dependent malic enzyme activity.

Amino acid sequence comparisons indicate that the ywkA and malS products, on the one hand, and the ytsJ and mleA products, on the other hand, belong to two distinct subgroups (60 % identity between YwkA and MalS; 40 % identity between YtsJ and MleA; but 28 % identity between YtsJ and MalS). Furthermore, YwkA and MalS share 40 % identity with SfcA, an NAD-dependent malic enzyme which was found to be more similar to malolactic enzymes than to other malic enzymes (Groisillier & Lonvaud-Funel, 1999). However, malic and malolactic enzymes are difficult to distinguish on the basis of the amino acid sequence because they are highly similar and harbour common conserved sequence motifs (Denayrolles et al., 1994; Kawai et al., 1996; Groisillier & Lonvaud-Funel, 1999) and because only very few malolactic enzymes have been enzymically characterized (Ansanay et al., 1993; Arthurs & Lloyd, 1999). Recently, a function for mleA in B. subtilis has been proposed by Wei et al. (2000). mleA is the second gene of a bicistronic operon including the mleN gene. mleN has been demonstrated to encode an antiporter that achieves simultaneous Na+/H+ and malate/lactate antiport. Therefore, MleA is strongly predicted to be a malolactic enzyme.

The examination of the sequence of the B. subtilis chromosome indicates that ywkA is probably cotranscribed with ywkB, which encodes a protein of unknown function. malS is linked to ytnP, an unknown gene, and ytsJ is linked to dnaE, which encodes the second essential DNA polymerase (Dervyn et al., 2001). Thus, the genetic organization of these three genes did not give any precise clue about their possible physiological roles.

Bacteria have to adapt to frequent changes in nutritional availability and, more generally, in environmental conditions. This is particularly true for free-living soil bacteria like B. subtilis. To accommodate gene expression to environmental conditions, two-component sensor–regulator systems are frequently used by bacteria. The two-component systems (TCSs) are composed of sensory protein kinases and of response regulator proteins. The protein kinases are typically located in the cytoplasmic membrane and comprise an N-terminal sensory and a conserved C-terminal transmitter domain. Binding of the signal ligand to the sensory domain induces ATP hydrolysis and autophosphorylation of the transmitter domain. The phosphoryl group is subsequently transferred to the cognate response regulator, leading to the activation of its regulatory property, most commonly transcriptional activation. More than 30 TCSs can be predicted from the complete B. subtilis genome sequence. However, a precise adaptative role has been identified for only a few of them (Perego & Hoch, 2002). Among them, two, YdbFG (DctSR) (Asai et al., 2000) and CitST (Yamamoto et al., 2000), appeared to be able to sense the presence of tricarboxylic acid (TCA) cycle intermediates in the medium and to activate the expression of genes encoding transporters for these compounds. Furthermore, a comprehensive DNA microarray analysis of the B. subtilis TCSs has suggested that ywkA could be a target of YufL/YufM (Kobayashi et al., 2001).

The starting point of the study reported here is a global transcriptome comparison of a wild-type B. subtilis strain growing in minimal medium containing glucose or malate as sole carbon source. Among the genes which appeared to be more transcribed on malate than on glucose was ywkA, but not malS or ytsJ. The regulation of the expression of the ywkA gene has therefore been further investigated, and the uncharacterized YufL/YufM TCS has been demonstrated to control directly ywkA transcription in response to the presence of malate. The enzymic activity encoded by ywkA has been characterized. Finally, single and multiple mutant strains for ywkA and the other putative malic enzyme genes have been constructed and studied to investigate the physiological role of YwkA.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and growth conditions.
E. coli TG1 [K12 {Delta}(lac–pro) supE thi hsd-5/F' traD36 proA+B+ lacIq lacZ{Delta}M15] was used as a general cloning host and E. coli strain M15 pREP4 (Qiagen) was used for overproduction of His6-YufM and His6-YwkA. E. coli strains were grown in Luria–Bertani broth (LB) supplemented with antibiotics when necessary (ampicillin, 100 mg l-1; kanamycin, 25 mg l-1). Conventional calcium-shock or electroporation procedures were used for transformation (Sambrook et al., 1989).

The B. subtilis strains used in this work are listed in Table 1. Standard procedures were used to transform B. subtilis (Anagnostopoulos & Spizizen, 1961). B. subtilis strains were grown in LB, except when stated otherwise. Antibiotics for selection were added at 5 mg l-1 (chloramphenicol), 0·4 mg l-1 (erythromycin), 100 mg l-1 (spectinomycin), or 5 mg l-1 (kanamycin). All cultures were performed at 37 °C. Growth tests were performed in MM minimal medium [K2HPO4 85 mM, KH2PO4 40 mM, (NH4)2SO4 15 mM, trisodium citrate 6 mM, MgSO4 0·8 mM, pH 7] supplemented with tryptophan (0·005 %) and carbon sources (0·5 %). Cultures for {beta}-galactosidase assays were performed in CQTHC minimal medium [C minimal medium (Aymerich et al., 1986) supplemented with 0·005 % tryptophan, 0·05 % glutamine, 0·05 % casein hydrolysate] supplemented with 1 % glucose, 1 % glycerol, 1 % fumarate, 1 % malate, or 1 % succinate plus 1 % glutamate.


View this table:
[in this window]
[in a new window]
 
Table 1. B. subtilis strains used in this study

 
DNA manipulation.
PCR products were purified with the QIAquick PCR purification kit (Qiagen). Plasmid DNA was extracted and purified from E. coli with the QIAprep spin miniprep kit (Qiagen). Restriction enzymes and T4 DNA ligase were used as recommended by the manufacturers. Pyrobest DNA polymerase (Takara) was used for preparative polymerase chain reactions as recommended by the manufacturer. DNA fragments were purified from agarose gels using the Qiagen Gel Extraction Kit.

Oligonucleotides.
The oligonucleotides used in this study are listed in a table available as supplementary data with the online version of this paper (http://mic.sgm.journals.org).

Construction of strains.
A fusion of a ywkA promoter fragment to a promoter-less lacZ gene was constructed using the vector pDG1661 (Guérout-Fleury et al., 1996). The 277 bp promoter fragment to be tested was obtained by PCR using primers PS69 and PS70 (see supplementary table); this fragment was inserted between the EcoRI and BamHI sites of pDG1661 to generate plasmid pPS61, then used to insert the reporter construct into the B. subtilis amyE locus to generate strain PS1716 (Table 1).

In vivo replacement by homologous recombination of a kanamycin-resistance cassette for the lacZ ORF and erythromycin resistance gene of pMUTIN was obtained using pEC23, a plasmid specifically designed for this use, constructed by P. Stragier, Institut de Biologie Physico-Chimique, 75005 Paris, France.

Strain GTD110 was constructed by replacing the malS ORF by the spectinomycin-resistance cassette contained between the BamHI and XbaI sites of plasmid pIC333 (Steinmetz & Richter, 1994): two PCR fragments, one corresponding to the 500 bp upstream of malS ORF and one to the 500 bp downstream of the malS ORF, were generated with primer pairs EagI-ytnP/XbaI-malS and BamHI-ytzB/EcoRI-ytzB (see supplementary table), respectively. The two PCR fragments were then ligated together with the spectinomycin cassette flanked by the BamHI and XbaI restriction sites into vector pJH101 (Ferrari et al., 1982) to generate plasmid pTD101, which was used to transform B. subtilis.

RNA isolation and Northern blot analysis.
For RNA preparation, cell pellets were obtained from a mid-exponential-phase (OD600 0·8) culture in S6 medium (Fujita & Freese, 1981) containing 25 mM glucose or 37·5 mM malate. Total RNAs were isolated from cells essentially as previously described (Yoshida et al., 2001). For Northern blot analysis, RNA samples (2 µg each) were electrophoresed in glyoxal gels, transferred to a Hybond-N membrane (Amersham) and hybridized with a labelled probe as described previously (Yoshida et al., 1997). To prepare the probes for ywkA, ywkB and gapB, parts of their coding regions were amplified by PCR using chromosomal DNA of B. subtilis 168 as a template and primer pairs ywkA-N1 and ywkA-N2, ywkB-N1 and ywkB-N2, and gapB-N1 and gapB-N2, respectively (see supplementary table). The PCR products were labelled radioactively by using the Bca BEST labelling kit (Takara Shuzo) and [{alpha}-32P]dCTP (ICN Biomedicals).

Preparation of fluorescently labelled cDNA, hybridization and microarray analysis.
The fluorescently labelled cDNA probes used for hybridization to DNA microarrays were prepared by a two-step procedure, as described previously (Ogura et al., 2001).

The microarrays that we used in this study contained 4055 protein genes, 45 not being spotted due to a problem with DNA amplification by PCR, as well as 39 calf thymus DNA spots as negative controls. The hybridization and microarray analysis were performed as described previously (Ogura et al., 2001; Yoshida et al., 2001). The mean Cy3 and Cy5 fluorescence intensities for each spot were calculated, the background being taken as the mean of the intensities of the 39 calf thymus DNA spots. After subtracting the background from all the intensities of the B. subtilis gene spots, and their normalization using the total Cy3 and Cy5 intensities, we calculated the expression ratios. To get reliable ratios, we ignored the spots of the intensities used as numerators for this calculation which were less than the background, and replaced the intensities used as denominators with the standard deviation of the mean intensity of the negative controls if they were lower than them.

Primer extension.
To map the 5'-end of the ywkA transcript by primer extension, 50 µg of each RNA was annealed to primer rev (see supplementary table) that had been labelled at its 5'-end by a MEGALABEL kit (Takara Shuzo) and [{alpha}-32P]ATP (Amersham). Primer extension reactions were carried out as described previously (Yoshida et al., 1997).

Construction of the mini-Tn10 transposition library and genetic screening.
The mini-Tn10 delivery vector pIC333 (Steinmetz & Richter, 1994) was used for transposon mutagenesis. Outside the transposon, this plasmid carries a thermosensitive Gram-positive origin of replication, an erythromycin-resistance gene, and a gene encoding a modified transposase with relaxed target specificity. The mini-transposon itself confers spectinomycin resistance and contains the pUC origin of replication in E. coli. pIC333 was introduced into PS1716 at 25 °C using erythromycin selection. Single transformant colonies were used to inoculate 12 independent cultures in 2 ml LB containing spectinomycin. At the beginning of the exponential growth phase, the temperature was shifted from 25 to 37 °C and cells were grown for 4 h more at 37 °C. Dilutions of the cultures were spread on plates of MM plus casein hydrolysate (0·05 %) or LB medium, plus X-Gal supplemented with glucose (0·5 %) or malate (0·5 %) to screen for clones with deregulated ywkA' : : lacZ expression.

{beta}-Galactosidase assays.
{beta}-Galactosidase activities were measured using the method of Miller (1972) on cell extracts prepared by lysozyme treatment and centrifugation. Protein concentration was determined by the method of Bradford (1976) using the Bio-Rad protein assay solution. One unit of {beta}-galactosidase activity is defined as the amount of enzyme that produces 1 nmol o-nitrophenol min-1 at 20 °C.

Expression and purification of His6-YufM.
His6-YufM was overexpressed and purified using the QIAexpress kit (Qiagen). yufM coding sequence (705 bp) was integrated into E. coli expression vector pQE-30 using primers BamHI-YufM and PstI-YufM (see supplementary table) in order to produce a fusion protein with a hexahistidine tag at the N-terminus. The resulting plasmid, pTDp3, was transformed into E. coli M15 pREP4 groESL to generate the strain used for production, GTDp3. All the subsequent procedures were performed as described in the QIAexpress manual. His6-YufM was stored in a buffer containing 50 mM Tris pH 8, 150 mM NaCl, 1 mM DTT and 0·5 mM EDTA.

Expression and purification of His6-YwkA.
His6-YwkA was overexpressed and purified as described for His6-YufM. Thus, plasmid pTDp3 was built using primers BamHI-YwkA and SphI-YwkA (see supplementary table) and introduced into E. coli M15 pREP4 to generate strain GTDp4, used for production of His6-YwkA. Storage buffer was the same as used for His6-YufM.

Determination of kinetic parameters of YwkA.
YwkA malic enzyme activity was tested at 37 °C by monitoring spectrophotometrically at 340 nm the NAD(P)H formation during the reductive decarboxylation of malate [malate+NAD(P) -> pyruvate+NAD(P)H]. The reaction mixture was composed of 25 nmol YwkA in a Tris buffer containing divalent cations (50 mM Tris/HCl pH 8, 10 mM MgCl2, 10 mM MnCl2, 50 mM KCl, 10 mM {beta}-mercaptoethanol). For determination of the Km for malate, 5 mM of both cofactors were added to the reaction. For determination of Km and Vmax for each cofactor, the malate concentration was raised to 40 mM. For both cofactors, a molar absorption coefficient of 6·22x106 cm2 mol-1 was used for calculations. Demonstration of malic enzyme activity was assessed in a coupled reaction: at the end of the first reaction (end of NAD reduction), 40 units of strictly NAD-dependent lactate dehydrogenase from Bacillus stearothermophilus (Sigma) and 10 mM CaCl2 were added to the reaction mixture. Then, oxidation of NADH, if produced by the first conversion of malate to pyruvate, was followed.

DNA electrophoretic mobility shift assays.
The 277 bp DNA fragment of plasmid pPS61 containing the promoter region of ywkA was amplified by PCR with primers PS69 and PS70 (see supplementary table) and purified. Approximately 20 pmol of this fragment was end-labelled using T4 polynucleotide kinase (New England Biolabs) and 30 µCi (1·1 MBq) [{gamma}-32P]ATP (Isoblue, ICN). After precipitation, 10 000 c.p.m. of radiolabelled probe ({approx}0·05 pmol) was incubated for 5 min at room temperature in binding buffer (10 mM Tris/HCl pH 8·0, 50 mM NaCl, 1 mM EDTA, 1 mM DTT, 5 %, w/v, glycerol) supplemented with 1 µg herring sperm DNA. Varying amounts of His6-YufM were added for a total reaction volume of 20 µl. After incubating for 10 min at room temperature, samples were analysed by native 6 % acrylamide gel electrophoresis, 1x Tris/glycine, run at 4 °C and constant voltage (14 V cm-1). After migration, gels were dried and the radiolabelled bands revealed by the PhosphoImager system.

DNase I footprinting.
Primers PS69 and PS70 (see supplementary table), 30 pmol each, were both end-labelled using T4 polynucleotide kinase (Biolabs) and 30 µCi (1·1 MBq) [{gamma}-32P]ATP (Isoblue, ICN) and purified by precipitation. They were used to generate one-strand labelled PCR probes. The labelled DNA probes (60 000 c.p.m.; {approx}0·25 pmol) were incubated for 15 min at room temperature with varying amounts of His6-YufM (the amount of protein was increased fivefold compared to gel mobility shift experiments due to partial loss of YufM binding activity after freezing). DNase I (Amersham Pharmacia Biotech) reactions were then performed as previously described (Doan & Aymerich, 2003). All samples were analysed on a 6 % acrylamide sequencing gel, which was dried and revealed by autoradiography.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Transcriptome analysis reveals ywkA induction in the presence of malate
A transcriptome analysis originally designed to identify the B. subtilis genes whose expression varies depending on whether the central carbon flux is glycolytic or gluconeogenic was performed. Cells of strain 168CA grew at almost the same rate on either glucose or malate as the sole carbon source; their doubling times in S6 medium containing glucose and malate were 1·37 and 1·25 h, respectively. Thus, these growth conditions are suitable to detect by DNA microarray analysis specific changes in global gene expression during glycolytic growth versus gluconeonenic growth; changes of expression of numerous genes such as ribosome protein genes that are affected by growth rate would not interfere.

Fig. 1 shows the expression ratio of genes involved in the glycolytic pathway and the TCA cycle between the gluconeogenic and the glycolytic growth conditions, which were inferred from our DNA microarray analysis (the complete microarray data are available on the KEGG web site: http://www.genome.ad.jp/kegg/expression). Expression of genes boxed in the figure was changed at least twofold between glycolysis and gluconeogenesis. ywkA transcription appeared to be induced more than sevenfold in the presence of malate as compared to glucose. The transcription signal of ywkB, the gene located just downstream and most probably constituting a bicistronic operon with ywkA, was very weak. Therefore, the upregulation of ywkB transcription found under gluconeogenic growth conditions (ratio 1·96) cannot be considered as significant. The gapB, pckA and pdhABCD genes were also upregulated under gluconeogenic growth conditions. By contrast, the gapA, pgk, tpi, eno and yjmC genes were upregulated during glycolytic growth. The upregulation of gapB (Fillinger et al., 2000) and pckA (Yoshida et al., 2001) during gluconeogenesis was reported previously. It has also been reported that transcription of the gapA, pgk, tpi and eno genes, as well as that of the pgm gene, both belonging to the same gapA-pgk-tpi-pgm-eno operon, is stimulated during glycolysis (Fillinger et al., 2000; Ludwig et al., 2001, 2002; Doan & Aymerich, 2003).



View larger version (34K):
[in this window]
[in a new window]
 
Fig. 1. Expression change of the B. subtilis genes involved in glycolysis and gluconeogenesis. Genes involved in the glycolytic pathway and the TCA cycle are followed by their expression ratios (malate/glucose) obtained by the DNA microarray analysis. The expression ratio for each gene was taken as the mean of two ratios obtained from the different hybridization profiles. Genes whose expression was changed at least twofold between glycolytic and gluconeogenic growths, are boxed (glycolytic genes) or boxed and stippled (gluconeogenic genes). The underlined gapA, pgk, tpi, pgm and eno genes, on the one hand, and pdhABCD genes, on the other, are organized in two operons. The complete microarray data are available on the KEGG website (http://www.genome.ad.jp/kegg/expression).

 
ywkA expression is induced only when malate is present in the medium
To confirm that ywkA transcription is induced in the presence of malate and to define more precisely the physiological conditions in which it is turned on, a reporter strain of the activity of the ywkA promoter region, PS1716, was constructed (Table 1). This strain harbours a ywkA' : : lacZ fusion inserted into the chromosome at the ectopic amyE locus. The {beta}-galactosidase activity synthezised in strain PS716 grown in mineral medium supplemented with different carbon sources was measured (Table 2). The results indicated that ywkA promoter activity was very weak in the presence of succinate plus glutamate, or fumarate, and of glucose or glycerol, as sole gluconeogenic and glycolytic carbon source, respectively. Similarly, very weak basal {beta}-galactosidase activity was detected in the presence of other gluconeogenic carbon sources, pyruvate or aspartate (data not shown). By contrast, the expression of the reporter construct was very strong in the presence of malate (and rather constant throughout the exponential growth phase). Moreover, addition of glucose to malate minimal medium did not affect ywkA expression. These results showed that ywkA transcription is specifically induced in the presence of malate in the culture medium and that it is not subject to carbon catabolite repression. Furthermore, because no expression was detected in the presence of fumarate, a C4-dicarboxylic acid that enters the TCA cycle and is then converted in one step into intracellular malate, it could be assumed that only extracellular malate induces ywkA expression.


View this table:
[in this window]
[in a new window]
 
Table 2. Expression of a ywkA' : : lacZ reporter fusion in minimal medium supplemented with various carbon sources

 
ywkA does not contribute to the utilization of malate for growth
As ywkA is specifically expressed when malate is present in the culture medium, we wanted to find out whether this gene is involved in the utilization of malate as a carbon source for growth. A ywkA mutant strain, GTD102, was constructed and was tested for growth in minimal medium containing either malate, or succinate plus glutamate (entering the TCA cycle and subsequently providing malate), or a glycolytic carbon source, glucose. This growth test showed that strain GTD102 was able to use either malate (Fig. 2), or succinate plus glutamate or glucose (data not shown) as efficiently as the wild-type strain.



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 2. Comparative growth of strains 168CA ({circ}), GTD102 (ywkA, {blacksquare}), BFA91 (ytsJ, {square}), GTD110 (malS, {triangleup}), GTD121 (ytsJ ywkA, {bullet}) and GTD123 (ytsJ ywkA malS, {diamond}) in malate minimal medium (MM malate).

 
A ywkA inactivation might be functionally complemented by one or several other putative malic enzyme encoding genes. Therefore, we also constructed different strains harbouring different combinations of mutations of the ywkA, malS and ytsJ genes: single mutant strains BFA91 (ytsJ) and GTD110 (malS), double mutant strains GTD111 (ywkA, malS), GTD121 (ywkA, ytsJ) and GTD122 (ytsJ, malS), and a triple mutant strain GTD123 (ywkA, malS, ytsJ). All these strains were able to grow in glucose minimal medium as efficiently as the wild-type strain (data not shown). But when growing in the presence of malate as sole carbon source (Fig. 2), all the strains mutated for ytsJ, BFA91, GTD121, GTD122 (data not shown) and GTD123, exhibited the same lower growth rate than the wild-type strain. By contrast, all the other mutants, GTD102, GTD110 and GTD111 (data not shown) grew as well as the wild-type strain. The ywkA mutation is supposed to have a polar effect on ywkB, and the absence of phenotype could be due to the absence of expression of both ywkA and ywkB. Because the ywkA disruption was constructed using a pMUTIN derivative (Moriya et al., 1998; Vagner et al., 1998), the possible polar effect on ywkB can be suppressed by addition of IPTG. However, the growth pattern of the strains harbouring a ywkA mutation was identical in the presence and in the absence of IPTG (data not shown). These experiments confirmed that ywkA and malS are not necessary for growth under either of the conditions tested. By contrast, ytsJ appeared to play a major role in the utilization of malate for growth.

YwkA exhibits a malic enzyme activity and preferentially utilizes NAD
Does ywkA encode a functional malic enzyme? To answer this question, a characterization of the enzymic properties of the YwkA protein was performed. First, a vector designed to overexpress a YwkA tagged with His6 was constructed (pTDp3) and introduced into E. coli M15 pREP4. Overexpression was induced by cultivating the resulting strain in the presence of IPTG, and the His6-YwkA protein was purified to apparent homogeneity. Then, the capacity of this protein to reduce NAD or NADP in the presence of malate was assayed. The results showed that YwkA has a strong activity of reduction of either NAD+ or NADP+ that depends on the presence of malate. To demonstrate that this reducing activity is associated with the conversion of malate into pyruvate, i.e. that YwkA exhibits a malic enzyme activity and not a malate dehydrogenase or a malolactic enzyme activity, the assay of the activity of an enzyme, lactate dehydrogenase, using specifically pyruvate as substrate, was coupled to the assay of YwkA. Indeed, addition of lactate dehydrogenase after the end of the first reaction (no further increase of A430) led to rapid reoxidation of NADH (Fig. 3). This could result only from oxidative conversion of pyruvate into lactate and thus demonstrated that pyruvate is the product of the first reaction catalysed by YwkA. In a second series of experiments, we determined the kinetic parameters of YwkA activity. The Km value for malate was determined to be 6·6 mM. Km, kcat and kcat/Km values were 1·1 mM, 112 s-1 and 102 s-1 mM-1, respectively, for NAD, and 9·0 mM, 88 s-1, and 10 s-1 mM-1, respectively, for NADP. Thus, we concluded that YwkA is a malic enzyme with a dual specificity for NAD and NADP as a cofactor but with a strong preference for NAD [10-fold higher catalytic efficiency (kcat/Km) with NAD than with NADP].



View larger version (12K):
[in this window]
[in a new window]
 
Fig. 3. Characterization of the enzymic activity of YwkA. Addition of lactate dehydrogenase after the end of the first reaction (no further increase of A430) results in reoxidation of NADH, thus showing conversion of pyruvate into lactate.

 
Transcriptional characterization of ywkA
RNA samples prepared from cells of strains 168CA, 1A250 and 1A147 (ccpA1) grown in malate or glucose minimal medium were subjected to Northern analysis using specific probes for ywkA, ywkB, and gapB, a key gluconeogenic gene. As shown in Fig. 4, a specific transcript (2·8 kb) was detected with the ywkA probe only when cells were grown on malate (lanes 2, 4 and 6). This transcript is able to cover the ywkA gene as well as the downstream gene, ywkB. When the ywkB probe was used, this 2·8 kb transcript was only slightly visible due to the presence of the faint broad bands from rRNAs (3·0 and 1·5 kb), only when cells of the three strains were grown on malate (lanes 2, 4 and 6). These results strongly suggested that the ywkAB genes are cotranscribed as a single transcription unit. As a control, a transcript (1·7 kb) of gapB was observed, as expected, only when cells were grown on malate (Fig. 4, gapB lanes 2, 4 and 6) (Fillinger et al., 2000). Furthermore, this analysis also confirmed that repression of ywkA-ywkB transcription as well as gapB transcription (Fillinger et al., 2000; Yoshida et al., 2001) is independent of CcpA (compare lanes 4 and 6).



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 4. Northern analysis of the ywkA, ywkB and gapB transcripts. The RNA samples from strains 168CA (lanes 1 and 2), 1A250 (lanes 3 and 4) and 1A147 (lanes 5 and 6) grown on glucose (lanes 1, 3 and 5) or malate (lanes 2, 4 and 6) were subjected to Northern analyses using ywkA, ywkB and gapB probes. The specific bands for ywkA and ywkB were observed at 2·8 kb, while those for gapB were observed at 1·7 kb. Two faint bands indicated by arrowheads at about 1·5 and 3 kb are due to nonspecific hybridization to rRNAs.

 
Primer extension analysis mapped a 5'-end of the ywkAB transcript only with RNA prepared from cells of strain 168CA grown on malate, not from those grown on glucose (Fig. 5, lane 1), and allowed us to find a corresponding promoter sequence consisting of -10 (TACAAT) and -35 (TTGAAT) regions separated by a 17 bp spacer, which is probably recognized by {sigma}A RNA polymerase. No other clear 5'-end of the ywkAB transcript was detected in the region (nucleotides about +120 to -140) covering the translation initiation nucleotide of ywkA (+80) by our primer extension experiments, including ones using another more downstream primer (data not shown).



View larger version (49K):
[in this window]
[in a new window]
 
Fig. 5. Mapping of the 5' end of the ywkAB transcript. Reverse transcriptase reaction was performed using total RNA from strain 168CA grown in malate (lane 1) or glucose (lane 2). Lanes G, A, T and C correspond to the dideoxy sequencing ladder obtained with the same primer as in the reverse transcriptase reaction. The transcriptional start site is indicated by an arrow on the right side of the panel. On the left side, the 5' end position is marked by an arrow (+1) in the nucleotide sequence of the non-coding strand of the ywkA promoter region, where probable -10 and -35 regions are underlined.

 
In conclusion, the ywkAB genes are cotranscribed to produce a 2·8 kb transcript from the transcription initiation site (Fig. 5); this transcription is induced during growth on malate and this regulation is independent of CcpA.

Genetic screening for regulator(s) involved in the control of ywkA transcription
The ywkAB genes are not physically linked to any gene encoding a putative transcriptional regulator. So, we decided to search for a regulator of ywkAB transcription by a genetic screening. A mini-Tn10 transposant library was obtained from B. subtilis strain PS1716, which contains an ectopic ywkA' : : lacZ reporter fusion. This library was then screened on plates for deregulation of ywkA expression (see Methods). Several clones from independent pools of transposants were isolated in which ywkA was no longer induced in the presence of malate. The gene targeted by the transposon in 18 of these clones was identified. YufL was the gene most frequently found to be interrupted (eight occurrences). Back-crossing experiments from these eight transposants indicated that the loss of ywkA inducibility was associated with the insertion of the transposon. As predicted on the basis of its nucleotide sequence, yufL would encode the membrane-bound sensor-kinase of a TCS, associated with the transcriptional response-regulator encoded by the putatively co-transcribed gene yufM. However, no clone harbouring a transposon inserted into the yufM ORF was picked up in this genetic screening.

The TCS YufL/YufM is required for ywkA induction by malate
To confirm the role of the YufL/YufM TCS in the regulation of ywkA transcription, we introduced a yufM disruption into strain PS1716 and analysed the expression of the ywkA' : : lacZ reporter fusion in the resulting strain (GTD132). Strain GTD134, a PS1716-derived strain in which the yufL gene is disrupted by a plasmidic cassette that places the downstream gene yufM under the control of a Pspac promoter, and the parental strain PS1716 were also tested in parallel. The {beta}-galactosidase activity synthesized by these strains grown in the presence of glucose or malate as sole carbon source was measured (Table 3). The results showed that ywkA expression was no longer induced by malate in a yufL or yufM mutant strain. In the yufL mutant strain (GTD134), expression of ywkA was restored after the artificial overexpression of yufM by addition of IPTG, independently of the carbon source, glucose or malate, present in the medium. These findings thus demonstrated that the YufL/YufM TCS is required for ywkA induction in response to the presence of malate.


View this table:
[in this window]
[in a new window]
 
Table 3. Effect of yufL or yufM mutations on ywkA' : : lacZ reporter fusion expression

 
The growth rate and the biomass yield of the yufM and of the yufL mutant strains in the malate medium was significantly lower than that of the parental strain (doubling time 1·6 h versus 1·2 h; final OD600 0·7 versus 1·5, respectively). This observation suggests that the YufL/YufM TCS positively controls the expression of other genes that, in contrast to ywkA, are involved in the utilization of malate as carbon source for growth.

YufM is able to bind specifically to the regulatory region of ywkA
A His-tagged YufM protein was overexpressed in E. coli and purified to apparent homogeneity (a figure showing the purification is available as supplementary data with the online version of this paper at http://mic.sgmjournals.org). A gel mobility shift assay was then performed to test the capacity of YufM to bind to the ywkA regulatory region. The assay, which was carried out in the presence of a large excess of unspecific DNA competitor, showed that YufM can bind directly and specifically to the regulatory region of ywkA in a concentration-dependent manner (Fig. 6). The shift of the probe was complete only when the protein/DNA concentration ratio exceeded 150; this suggests that, as demonstrated for other TCS response regulators (Perego & Hoch, 2002), YufM requires to be phosphorylated to be fully active in DNA binding.



View larger version (19K):
[in this window]
[in a new window]
 
Fig. 6. Specific binding of His6-YufM to the promoter region of the ywkA operon. The 277 bp DNA fragment inserted into plasmid pPS61 (approx. 0·05 pmol) was labelled and then incubated with no or increasing amounts of His6-YufM in the presence of a large excess of unspecific DNA.

 
YufM protects a 50 bp DNA sequence upstream of the ywkA promoter sequence
To define more precisely the YufM binding region, DNase I footprinting experiments were performed. The 277 bp fragment present upstream of the ywkA coding sequence was labelled on one strand or the other, incubated with an increasing amount of YufM, and then hydrolysed by DNase I. As revealed after electrophoresis (Fig. 7a), YufM protected a 50 bp sequence shifted three bases from one strand to the opposite one (Fig. 7b). This protected area is located just upstream of the -35 sequence of the ywkA promoter, in agreement with the transcriptional activator activity of YufM. Sequence examination of that protected region revealed two distinct perfect direct repeats (Fig. 7b): ATTAAAAAATTN11ATTAAAAAATT and TAAGTAN11TAAGTA. Searches for these motifs (with one mismatch allowed in each sequence) in the B. subtilis complete genome sequence revealed no other occurrences. In particular, no sequence similar to these direct repeats can be detected in the region protected by YufM upstream of two other target genes of YufM (Tanaka et al., 2003). This suggests that none of these DNA sequences is strictly required for efficient binding of YufM (see also Discussion).



View larger version (79K):
[in this window]
[in a new window]
 
Fig. 7. (a) DNase I footprinting analysis of the interaction between YufM and the promoter region of ywkA examined on the coding and the non-coding strands. Approximately 0·25 pmol of labelled DNA fragment was incubated with no (lanes 2 and 7) or increasing amounts of His6-YufM (lanes 3, 74·58 pmol; lanes 4, 37·29 pmol; lanes 5, 9·32 pmol; lanes 6, 1·86 pmol). Lanes 1 and 8, G+A sequencing ladder obtained from the same DNA fragment. (b) Nucleotide sequence of the ywkA promoter region. The protected region is boxed: the protected nucleotides on both strands are in normal bold letters; italic letters indicate nucleotides that are protected only on the coding (5' side) or the non-coding (3' side) strand. Two direct repeats are indicated by continued or dashed arrows.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In this study, we have investigated the physiological function and the transcriptional regulation of ywkA, one of the four paralogous genes of B. subtilis that could be predicted to encode malic enzymes. On the basis of the results of a global transcriptomic analysis using DNA microarrays, we have further studied the expression pattern of the ywkA gene through lacZ reporter fusion, Northern and primer extension analyses. We have been able to demonstrate that ywkA expression is specifically induced by malate. A genetic study then identified the uncharacterized YufL/YufM TCS as the regulatory system necessary for the activation of ywkA transcription in response to the presence of malate in the culture medium. In vitro experiments demonstrated that YufM is able to bind directly to a region located just upstream of the ywkA promoter. Thus we propose to rename this TCS MalK/MalR for malate sensor kinase/malate regulator component. On the other hand, if our enzymic study clearly demonstrated that YwkA is a malic enzyme, our phenotypic analysis showed that ywkA is not necessary for utilization of malate as sole carbon source for growth.

The enzymic characterization reported here demonstrated that YwkA exhibits a malic enzyme activity and excludes the possibility that it is a malolactic enzyme or a malate dehydrogenase. Furthermore, it showed that YwkA has a 10-fold higher catalytic efficiency with NAD as a cofactor than with NADP. We therefore propose to class YwkA in the group of NAD-dependent malic enzymes, EC 1.1.1.38, and to rename its structural gene maeA for malic enzyme A.

Interestingly, our transcriptome analysis indicated that either malS or ytsJ, two paralogues of ywkA, were significantly and similarly transcribed in the presence of glucose or malate, suggesting that their expression is not regulated by malate nor by the glycolytic or gluconeogenic carbon flux orientation. By contrast, ywkA transcription was induced only in the presence of malate in the culture medium. Further addition of glucose or another glycolytic carbon source to the medium did not affect ywkA expression. Thus, ywkA is not subject to carbon catabolite repression. YwkA was not expressed during growth on fumarate or succinate plus glutamate, although these compounds enter the TCA cycle upstream of malate and thus very probably lead to the elevation of the intracellular malate concentration. It is thus very likely that extracellular malate is the signal for ywkA induction. Of course this does not exclude the possibility that the permeation of malate would be necessary for efficient transduction of the signal and thus full induction of ywkA transcription. The effects of the inactivation of the different putative malate transporters on the inducibility of ywkA need to be studied to test this possibility.

To try to get some clues concerning the biological role of ywkA, the regulatory mechanism of its expression has been investigated. Our analysis demonstrated that the YufL/YufM TCS is the positive regulatory system that mediates the induction of ywkA expression in response to the presence of malate in the culture medium, YufM being a direct transcriptional activator of ywkA transcription. This conclusion is in agreement with the results of a global transcriptome analysis performed by Kobayashi et al. (2001) indicating that ywkA expression is ninefold upregulated in a yufL mutant strain overexpressing YufM, compared to a wild-type strain. The YufL/YufM TCS belongs to the CitA/CitB family of TCSs. The members of this family that have already been characterized are CitA/CitB from E. coli (Kaspar & Bott, 2002) and Klebsiella pneumoniae (Bott et al., 1995; Kaspar et al., 1999), CitS/CitT from B. subtilis (Yamamoto et al., 2000) and Streptomyces coelicolor, YdbF/YdbG from B. subtilis (Asai et al., 2000) and DcuS/DcuR from E. coli (Zientz et al., 1998; Golby et al., 1999). All these TCSs are involved in the control of genes required for TCA cycle intermediates import and utilization. Sequence analyses strongly suggest that YufL is a transmembrane protein containing an extracellular sensor domain and a cytoplasmic kinase domain, as shown for the highly similar E. coli DcuS protein (Golby et al., 1999). Thus, YufL would be able to modulate YufM DNA-binding activity by phosphorylation/dephosphorylation depending on the presence of extracellular malate. The transcriptome analysis performed by Kobayashi et al. (2001) has revealed 97 other differentiallly expressed genes in the yufL mutant strain overexpressing yufM compared to the wild-type strain. Because this list included several competence genes, and especially comK and most of the members of the ComK regulon (Berka et al., 2002; Ogura et al., 2002), the authors suggested that the YufL/YufM TCS would be involved in competence development. However, we found no drastic transformability defect, if any, of yufL, yufM or ywkA mutant strains (T. Doan & S. Aymerich, unpublished data). The study of the genome-wide expression changes dependent on ComK, recently reported by Berka et al. (2002), led these authors to conclude that the ComK regulon defines a growth-arrested state, distinct from sporulation, of which competence for genetic transformation is but one notable feature. They suggest that this is a unique adaptation to stress and that it be termed the ‘K-state’. Thus the YufL/YufM TCS could participate in the signalling of the ‘K-state’.

Interestingly, simultaneously with our work, the expression of maeN (yufR), a gene located 5·3 kb downstream of yufL-yufM on the chromosome, has been found to be activated by the YufL/YufM TCS in the presence of malate (Tanaka et al., 2003). Wei et al. (2000) had previously reported that MaeN exhibits a malate–Na+ symporter activity. Ogasawara and colleagues indeed found that MaeN, and thus YufL/YufM, are essential for utilization of malate as sole carbon source for growth (Tanaka et al., 2003). This is in agreement with the growth defect on malate associated with the yufL and the yufM mutations that we observed during the test of their effects on ywkA expression; however, the defect was only partial in our experiments because the medium used contained not only malate but also glutamine and casein hydrolysate as carbon sources (Table 3). The coregulation of ywkA and maeN is of course physiologically relevant since MaeN could provide the YwkA malic enzyme with cytoplasmic malate. Furthermore, Krulwich and coworkers (Ito et al., 2000) have shown that, in some particular conditions, maeN expression was coupled to the expression of the downstream mrp gene cluster involved in alkali resistance and pH homeostasis. These authors suggested a possible requirement of coordinated Na+–malate symport and Na+ re-extrusion in exchange for H+. The protonmotive force could be generated by secondary NADH dehydrogenases putatively encoded by some mrp genes. Therefore, YwkA could be involved in the regeneration of NADH during this process.

Ogasawara and colleagues have also found that the expression of yflS, putatively encoding a 2-oxoglutarate/malate translocator, was under the control of the YufL/YufM TCS (Tanaka et al., 2003). Indeed, maeN and yflS have been detected as being up-regulated (ratios Mal/Glc 4·2 and 2·7, respectively) during growth on malate in our transcriptomic analysis. By contrast, they were not clearly identified (ratios 1·7 and 1·5) in the transcriptomic analysis of Kobayashi et al. (2001), which compared a yufL mutant strain overexpressing YufM with the wild-type strain. We have compared the sequence of the region protected by YufM upstream of the ywkA, the maeN and the yflS promoters and found no significant similarities. The most conspicuous feature of these three regions is their very high A+T content (88, 83 and 82 mol%, upstream of ywkA, maeN and yflS, respectively). On the other hand, no other occurrence of one or the other direct repeat included in the region protected by YufM upstream of ywkA (Fig. 6) can be detected in the complete B. subtilis genome. Thus one should conclude that the YufM DNA-binding activator does not recognize a unique primary sequence, a property similar to that of several DNA-binding components of other well-characterized TCSs (Perego & Hoch, 2002).

YwkA does have malic enzyme activity but it does not contribute to the very efficient utilization of malate as sole carbon source for growth by B. subtilis. This is in agreement with the fact that ywkA is not subject to carbon catabolite repression. Because neither a malS nor a double ywkA malS mutant strain showed any growth defect, it can also be concluded that, like ywkA, malS is not necessary for growth on malate. By contrast, the ytsJ gene is required for efficient growth on malate, as well as on other TCA cycle intermediates. ytsJ is expressed constitutively during the exponential phase of growth and encodes an NADP-dependent malic enzyme (T. Doan, G. Lerondel & S. Aymerich, unpublished data). YtsJ could thus be required for its capacity to catalyse an anaplerotic reaction that allows the TCA cycle to function as a real cycle when the only carbon source is one intermediate of this cycle, or for its capacity to regenerate NADPH necessary for numerous anabolic reactions, or for both of these activities. We would propose that ywkA is involved in a more specific biological process necessary only or principally when malate is present in the extracellular environment. The study of the ywkB gene, of which no clear orthologue can be detected in the genome of other bacilli, and the identification of the activity of more genes belonging to the yufL/YufM TCS regulon, should help to develop this hypothesis.


   ACKNOWLEDGEMENTS
 
We thank Patrick Stragier for the generous gift of plasmid pEC23 (unpublished). We are very grateful to Naotake Ogasawara and Uwe Sauer for communicating results before publication and for very helpful discussions. We thank Claude Gaillardin and Josef Deutscher for valuable discussions and for their continuous encouragement, and Dominique Le Coq and Ivan Mijakovic for their comments on the manuscript.

This work was supported by EU Quality of Life and Management of Living Resources, programme grant QL GZ-CT-1999-O1455, and by a Grand-in-Aid for Scientific Research on Priority Areas from the Ministry of Education, Science, Sports and Culture of Japan. Stéphane Aymerich, Guillaume Lerondel and Thierry Doan are on the staff at INRA (France).


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Anagnostopoulos, C. & Spizizen, J. (1961). Requirements for transformation in Bacillus subtilis. J Bacteriol 81, 741–746.

Ansanay, V., Dequin, S., Blondin, B. & Barre, P. (1993). Cloning, sequence and expression of the gene encoding the malolactic enzyme from Lactococcus lactis. FEBS Lett 332, 74–80.[CrossRef][Medline]

Arthurs, C. E. & Lloyd, D. (1999). Kinetics, stereospecificity, and expression of the malolactic enzyme. Appl Environ Microbiol 65, 3360–3363.[Abstract/Free Full Text]

Asai, K., Baik, S. H., Kasahara, Y., Moriya, S. & Ogasawara, N. (2000). Regulation of the transport system for C4-dicarboxylic acids in Bacillus subtilis. Microbiology 146, 263–271.[Abstract/Free Full Text]

Aymerich, S., Gonzy-Treboul, G. & Steinmetz, M. (1986). 5'-noncoding region sacR is the target of all identified regulation affecting the levansucrase gene in Bacillus subtilis. J Bacteriol 166, 993–998.[Medline]

Bartolucci, S., Rella, R., Guagliardi, A., Raia, C. A., Gambacorta, A., De Rosa, M. & Rossi, M. (1987). Malic enzyme from archaebacterium Sulfolobus solfataricus. Purification, structure, and kinetic properties. J Biol Chem 262, 7725–7731.[Abstract/Free Full Text]

Berka, R. M., Hahn, J., Albano, M., Draskovic, I., Persuh, M., Cui, X., Sloma, A., Widner, W. & Dubnau, D. (2002). Microarray analysis of the Bacillus subtilis K-state: genome-wide expression changes dependent on ComK. Mol Microbiol 43, 1331–1345.[CrossRef][Medline]

Bott, M., Meyer, M. & Dimroth, P. (1995). Regulation of anaerobic citrate metabolism in Klebsiella pneumoniae. Mol Microbiol 18, 533–546.[Medline]

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.[CrossRef][Medline]

Denayrolles, M., Aigle, M. & Lonvaud-Funel, A. (1994). Cloning and sequence analysis of the gene encoding Lactococcus lactis malolactic enzyme: relationships with malic enzymes. FEMS Microbiol Lett 116, 79–86.[CrossRef][Medline]

Dervyn, E., Suski, C., Daniel, R., Bruand, C., Chapuis, J., Errington, J., Janniere, L. & Ehrlich, S. D. (2001). Two essential DNA polymerases at the bacterial replication fork. Science 294, 1716–1719.[Abstract/Free Full Text]

Diesterhaft, M. D. & Freese, E. (1973). Role of pyruvate carboxylase, phosphoenolpyruvate carboxykinase, and malic enzyme during growth and sporulation of Bacillus subtilis. J Biol Chem 248, 6062–6070.[Abstract/Free Full Text]

Doan, T. & Aymerich, S. (2003). Regulation of the central glycolytic genes in Bacillus subtilis: binding of the repressor CggR to its single DNA target sequence is modulated by fructose-1, 6-bisphosphate. Mol Microbiol 47, 1709–1721.[CrossRef][Medline]

Driscoll, B. T. & Finan, T. M. (1993). NAD+-dependent malic enzyme of Rhizobium meliloti is required for symbiotic nitrogen fixation. Mol Microbiol 7, 865–873.[Medline]

Driscoll, B. T. & Finan, T. M. (1996). NADP+-dependent malic enzyme of Rhizobium meliloti. J Bacteriol 178, 2224–2231.[Abstract]

Driscoll, B. T. & Finan, T. M. (1997). Properties of NAD+- and NADP+-dependent malic enzymes of Rhizobium (Sinorhizobium) meliloti and differential expression of their genes in nitrogen-fixing bacteroids. Microbiology 143, 489–498.[Abstract]

Dunn, M. F. (1998). Tricarboxylic acid cycle and anaplerotic enzymes in rhizobia. FEMS Microbiol Rev 22, 105–123.[CrossRef][Medline]

Ferrari, F. A., Ferrari, E. & Hoch, J. A. (1982). Chromosomal location of a Bacillus subtilis DNA fragment uniquely transcribed by sigma-28-containing RNA polymerase. J Bacteriol 152, 780–785.[Medline]

Fillinger, S., Boschi-Muller, S., Azza, S., Dervyn, E., Branlant, G. & Aymerich, S. (2000). Two glyceraldehyde-3-phosphate dehydrogenases with opposite physiological roles in a nonphotosynthetic bacterium. J Biol Chem 275, 14031–14037.[Abstract/Free Full Text]

Fujita, Y. & Freese, E. (1981). Isolation and properties of a Bacillus subtilis mutant unable to produce fructose-bisphosphatase. J Bacteriol 145, 760–767.[Medline]

Golby, P., Davies, S., Kelly, D. J., Guest, J. R. & Andrews, S. C. (1999). Identification and characterization of a two-component sensor-kinase and response-regulator system (DcuS-DcuR) controlling gene expression in response to C4-dicarboxylates in Escherichia coli. J Bacteriol 181, 1238–1248.[Abstract/Free Full Text]

Groisillier, A. & Lonvaud-Funel, A. (1999). Comparison of partial malolactic enzyme gene sequences for phylogenetic analysis of some lactic acid bacteria species and relationships with the malic enzyme. Int J Syst Bacteriol 49, 1417–1428.[Abstract]

Guérout-Fleury, A. M., Frandsen, N. & Stragier, P. (1996). Plasmids for ectopic integration in Bacillus subtilis. Gene 180, 57–61.[CrossRef][Medline]

Hansen, E. J. & Juni, E. (1974). Two routes for synthesis of phosphoenolpyruvate from C4-dicarboxylic acids in Escherichia coli. Biochem Biophys Res Commun 59, 1204–1210.[Medline]

Hansen, E. J. & Juni, E. (1975). Isolation of mutants of Escherichia coli lacking NAD- and NADP-linked malic enzyme. Biochem Biophys Res Commun 65, 559–566.[Medline]

Ito, M., Guffanti, A. A., Wang, W. & Krulwich, T. A. (2000). Effects of nonpolar mutations in each of the seven Bacillus subtilis mrp genes suggest complex interactions among the gene products in support of Na+ and alkali but not cholate resistance. J Bacteriol 182, 5663–5670.[Abstract/Free Full Text]

Kaspar, S. & Bott, M. (2002). The sensor kinase CitA (DpiB) of Escherichia coli functions as a high-affinity citrate receptor. Arch Microbiol 177, 313–321.[CrossRef][Medline]

Kaspar, S., Perozzo, R., Reinelt, S., Meyer, M., Pfister, K., Scapozza, L. & Bott, M. (1999). The periplasmic domain of the histidine autokinase CitA functions as a highly specific citrate receptor. Mol Microbiol 33, 858–872.[CrossRef][Medline]

Kawai, S., Suzuki, H., Yamamoto, K., Inui, M., Yukawa, H. & Kumagai, H. (1996). Purification and characterization of a malic enzyme from the ruminal bacterium Streptococcus bovis ATCC 15352 and cloning and sequencing of its gene. Appl Environ Microbiol 62, 2692–2700.[Abstract]

Knichel, W. & Radler, F. (1982). D-Malic enzyme of Pseudomonas fluorescens. Eur J Biochem 123, 547–552.[Abstract]

Kobayashi, K., Doi, S., Negoro, S., Urabe, I. & Okada, H. (1989). Structure and properties of malic enzyme from Bacillus stearothermophilus. J Biol Chem 264, 3200–3205.[Abstract/Free Full Text]

Kobayashi, K., Ogura, M., Yamaguchi, H., Yoshida, K., Ogasawara, N., Tanaka, T. & Fujita, Y. (2001). Comprehensive DNA microarray analysis of Bacillus subtilis two-component regulatory systems. J Bacteriol 183, 7365–7370.[Abstract/Free Full Text]

Lamed, R. & Zeikus, J. G. (1981). Thermostable, ammonium-activated malic enzyme of Clostridium thermocellum. Biochim Biophys Acta 660, 251–255.[Medline]

Ludwig, H., Homuth, G., Schmalisch, M., Dyka, F. M., Hecker, M. & Stülke, J. (2001). Transcription of glycolytic genes and operons in Bacillus subtilis: evidence for the presence of multiple levels of control of the gapA operon. Mol Microbiol 41, 409–422.[CrossRef][Medline]

Ludwig, H., Rebhan, N., Blencke, H. M., Merzbacher, M. & Stülke, J. (2002). Control of the glycolytic gapA operon by the catabolite control protein A in Bacillus subtilis: a novel mechanism of CcpA-mediated regulation. Mol Microbiol 45, 543–553.[CrossRef][Medline]

Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Mitsch, M. J., Voegele, R. T., Cowie, A., Osteras, M. & Finan, T. M. (1998). Chimeric structure of the NAD(P)+- and NADP+-dependent malic enzymes of Rhizobium (Sinorhizobium) meliloti. J Biol Chem 273, 9330–9336.[Abstract/Free Full Text]

Moriya, S., Tsujikawa, E., Hassan, A. K., Asai, K., Kodama, T. & Ogasawara, N. (1998). A Bacillus subtilis gene-encoding protein homologous to eukaryotic SMC motor protein is necessary for chromosome partition. Mol Microbiol 29, 179–187.[CrossRef][Medline]

Murai, T., Tokushige, M., Nagai, J. & Katsuki, H. (1971). Physiological functions of NAD- and NADP-linked malic enzymes in Escherichia coli. Biochem Biophys Res Commun 43, 875–881.[Medline]

Ogura, M., Yamaguchi, H., Yoshida, K., Fujita, Y. & Tanaka, T. (2001). DNA microarray analysis of Bacillus subtilis DegU, ComA and PhoP regulons: an approach to comprehensive analysis of B. subtilis two-component regulatory systems. Nucleic Acids Res 29, 3804–3813.[Abstract/Free Full Text]

Ogura, M., Yamaguchi, H., Kobayashi, K., Ogasawara, N., Fujita, Y. & Tanaka, T. (2002). Whole-genome analysis of genes regulated by the Bacillus subtilis competence transcription factor ComK. J Bacteriol 184, 2344–2351.[Abstract/Free Full Text]

Perego, M. & Hoch, J. A. (2002). Two-component systems, phosphorelays, and regulation of their activities by phosphatases. In Bacillus subtilis and its Closest Relatives: from Genes to Cells, pp. 423–481. Edited by A. L. Sonenshein, J. A. Hoch & R. Losick. Washington, DC: American Society for Microbiology.

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

Steinmetz, M. & Richter, R. (1994). Easy cloning of mini-Tn10 insertions from the Bacillus subtilis chromosome. J Bacteriol 176, 1761–1763.[Abstract]

Stols, L. & Donnelly, M. I. (1997). Production of succinic acid through overexpression of NAD+-dependent malic enzyme in an Escherichia coli mutant. Appl Environ Microbiol 63, 2695–2701.[Abstract]

Tanaka, K., Kobayashi, K. & Ogasawara, N. (2003). The Bacillus subtilis YufLM two-component system regulates the expression of the malate transporters MaeN (YufR) and YflS, and is essential for utilization of malate in minimal medium. Microbiology 149, 2317–2329.[Abstract/Free Full Text]

Vagner, V., Dervyn, E. & Ehrlich, S. D. (1998). A vector for systematic gene inactivation in Bacillus subtilis. Microbiology 144, 3097–3104.[Abstract]

van der Rest, M. E., Frank, C. & Molenaar, D. (2000). Functions of the membrane-associated and cytoplasmic malate dehydrogenases in the citric acid cycle of Escherichia coli. J Bacteriol 182, 6892–6899.[Abstract/Free Full Text]

Voegele, R. T., Mitsch, M. J. & Finan, T. M. (1999). Characterization of two members of a novel malic enzyme class. Biochim Biophys Acta 1432, 275–285.[Medline]

Wei, Y., Guffanti, A. A., Ito, M. & Krulwich, T. A. (2000). Bacillus subtilis YqkI is a novel malic/Na+-lactate antiporter that enhances growth on malate at low protonmotive force. J Biol Chem 275, 30287–30292.[Abstract/Free Full Text]

Yamamoto, H., Murata, M. & Sekiguchi, J. (2000). The CitST two-component system regulates the expression of the Mg-citrate transporter in Bacillus subtilis. Mol Microbiol 37, 898–912.[CrossRef][Medline]

Yoshida, K., Aoyama, D., Ishio, I., Shibayama, T. & Fujita, Y. (1997). Organization and transcription of the myo-inositol operon, iol, of Bacillus subtilis. J Bacteriol 179, 4591–4598.[Abstract]

Yoshida, K., Kobayashi, K., Miwa, Y. & 9 other authors (2001). Combined transcriptome and proteome analysis as a powerful approach to study genes under glucose repression in Bacillus subtilis. Nucleic Acids Res 29, 683–692.[Abstract/Free Full Text]

Zientz, E., Bongaerts, J. & Unden, G. (1998). Fumarate regulation of gene expression in Escherichia coli by the DcuSR (dcuSR genes) two-component regulatory system. J Bacteriol 180, 5421–5425.[Abstract/Free Full Text]

Received 24 January 2003; revised 13 May 2003; accepted 1 July 2003.