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

Kousei Tanaka, Kazuo Kobayashi and Naotake Ogasawara

Department of Bioinformatics and Genomics, Graduate School of Information Science, Nara Institute of Science and Technology, 8916-5, Takayama, Ikoma, Nara 630-0101, Japan

Correspondence
Naotake Ogasawara
nogasawa{at}bs.aist-nara.ac.jp


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The Gram-positive bacterium Bacillus subtilis has a complete set of enzymes for the tricarboxylic acid (TCA) cycle and can grow aerobically using most of the TCA cycle intermediates (malate, fumarate, succinate and citrate) as a sole carbon source. The B. subtilis genome sequence contains three paralogous two-component regulatory systems, CitST, DctSR and YufLM. CitST and DctSR activate the expression of a transporter of the Mg2+–citrate complex (CitM) and a fumarate and succinate transporter (DctP), respectively. These findings prompted an investigation of whether the YufL sensor and its cognate regulator, YufM, play a role in malate uptake. This paper reports that the YufM regulator shows in vitro binding to the promoter region of two malate transporter genes, maeN and yflS, and is responsible for inducing their expression in vivo. It was also found that inactivation of the yufM or maeN genes resulted in bacteria that could not grow in a minimal salts medium containing malate as a sole carbon source, indicating that the induction of the MaeN transporter by the YufM regulator is essential for the utilization of malate as a carbon source. Inactivation of the yufL gene resulted in the constitutive expression of MaeN. This expression was suppressed by reintroduction of the kinase domain of YufL, indicating that the YufL sensor is required for proper signal detection and signalling specificity. The authors propose that a phosphatase activity of YufL plays an important role in the YufLM two-component regulatory system. The studies reported here have revealed that members of a set of paralogous two-component regulatory systems in B. subtilis, CitST, DctSR and YufLM, are involved in a related function – uptake (and metabolism) of the TCA cycle intermediates – but with distinct substrate specificities.


Abbreviations: TCA, tricarboxylic acid


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
C4-dicarboxylates such as malate, fumarate and succinate are metabolized aerobically or anaerobically by bacteria, which contain transporters for their uptake, exchange or efflux. Studies in Gram-negative bacteria have revealed that many contain more than one C4-dicarboxylate transporter, expression of which is induced only in the presence of the external substrate, usually through the action of a two-component regulatory system (Janausch et al., 2002). The two-component regulatory system, composed of a sensor kinase and a response regulator, is a ubiquitous mechanism that is able to sense and respond to various environmental stimuli (Stock et al., 2000). The sensor kinases monitor environmental signals and modulate functions of response regulators through phosphotransfer reactions. In many cases, the response regulators control gene expression directly as a transcriptional repressor or activator.

The Gram-positive bacterium Bacillus subtilis has a complete set of enzymes for the tricarboxylic acid (TCA) cycle and can grow aerobically on most of the TCA cycle intermediates (malate, fumarate, succinate and citrate; Fortnagel & Freese, 1968). The sequenced B. subtilis genome contains three paralogous sensor kinase genes, citS, dctS (ydbF) and yufL, which are highly similar to the Klebsiella pneumoniae citA and the Escherichia coli dcuS (CitA family; Fabret et al., 1999; Kasper et al., 1999). The CitA protein is a citrate sensor and controls the expression of genes for citrate fermentation in K. pneumoniae (Bott et al., 1995). The E. coli DcuS sensor is responsible for the activation of the dcuB and fumB genes encoding the anaerobic fumarate transporter and fumarase (Zientz et al., 1998; Golby et al., 1999). In B. subtilis, the CitS sensor kinase and its cognate response regulator, CitT, have been shown to activate the expression of the secondary transporter of the Mg2+–citrate complex, CitM, in the presence of citrate (Boorsma et al., 1996; Yamamoto et al., 2000). Although B. subtilis has a second citrate transporter, YxiQ (CitH), which transports the free citrate anion (Boorsma et al., 1996), the citSTM genes are essential for B. subtilis growth on minimal plates including citrate as a sole carbon source (Yamamoto et al., 2000). Furthermore, we have reported that the DctS sensor and the DctR (YdbG) regulator in B. subtilis regulate the expression of the DctP (YdbH) protein, a homologue of the C4-dicarboxylate transporters found in Gram-negative bacteria (Asai et al., 2000). Disruption of the dctS, dctR or dctP gene resulted in loss of the ability to utilize fumarate and succinate, but not malate, in the growth medium. These findings prompted us to study the role of the YufL sensor and its cognate regulator, YufM, in malate uptake.

Three B. subtilis secondary transporters having an ability to transport malate have been reported. The B. subtilis genome contains two genes, yufR and yxkJ, that code for membrane proteins belonging to the citrate–cation symporter family (Saier et al., 2002). The YufR protein (renamed MaeN) was expressed in E. coli and shown to catalyse Na+-coupled malate transport, although its physiological role in B. subtilis has not been studied (Wei et al., 2000). The YxkJ protein (renamed CimH) was also expressed in E. coli and found to translocate free citrate and malate anions (Krom et al., 2001). In addition, Wei et al. (2000) demonstrated that the YqkI protein (renamed MleN), which belongs to the Na+–H+ antiporter family (Saier et al., 2002), is an antiporter of H+-malate against Na+-lactate, and proposed that the transporter plays a role in malate uptake and Na+–H+ exchange in low protonmotive conditions. Furthermore, a search for possible malate transporter in B. subtilis revealed an additional candidate of the divalent anion–Na+ symporter family (Saier et al., 2002), YflS, which is highly homologous (54 % identity in 464 aa overlap) to the 2-oxoglutarate/malate translocator of spinach (Weber et al., 1995).

In this investigation, we examined whether expression of these malate transporter genes is induced by malate in the culture medium and is under the control of the YufLM two-component system. The genetic and biochemical studies described here demonstrated that YufLM induces the expression of the maeN and yflS genes in the presence of malate, and that the induction of the MaeN transporter by the YufM regulator is essential for the utilization of malate as a carbon source. The YufL sensor was required for proper signal detection and signalling specificity in this process.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and growth media.
The bacterial strains used in this study are listed in Table 1. E. coli strains were cultured in LB medium (5 g yeast extract l-1, 10 g polypeptone l-1 and 5 g NaCl l-1, pH 7·2) at 37 °C. B. subtilis strains were grown in Spizizen's minimal salts (SM) medium (2 g (NH4)2SO4 l-1, 14 g K2HPO4 l-1, 6 g KH2PO4 l-1, 1 g trisodium citrate.2H2O l-1; Anagnostopolos & Spizizen, 1961) containing 0·5 % (w/v) malate, a minimal salts medium (2 g (NH4)2SO4 l-1, 14 g K2HPO4 l-1, 6 g KH2PO4 l-1) containing 0·05 % (w/v) yeast extract (NYE medium; Willecke & Lange, 1974), and a minimal salts medium (2 g K2SO4 l-1, 10·8 g K2HPO4 l-1, 6 g KH2PO4 l-1, 1 g trisodium citrate.2H2O l-1) containing 0·5 % (w/v) glucose and 0·2 % (w/v) glutamine (GGM medium) with or without the addition of malate, fumarate or succinate at a final concentration of 2 mM. When required, ampicillin, erythromycin, kanamycin and chloramphenicol were added to a final concentration of 50, 0·5, 10 or 5 µg ml-1, respectively. IPTG was added at 1 mM where indicated. All B. subtilis growth media were supplemented with L-tryptophan (required for the growth of the parent strain, 168 trpC2; 50 µg ml-1).


View this table:
[in this window]
[in a new window]
 
Table 1. Bacterial strains

 
Plasmids and construction of B. subtilis mutants.
The plasmids and primers used in this study are listed in Tables 2 and 3, respectively. Transformation of B. subtilis cells was performed as previously described (Moriya et al., 1998).


View this table:
[in this window]
[in a new window]
 
Table 2. Plasmids constructed in this study

 

View this table:
[in this window]
[in a new window]
 
Table 3. Primers used in this study

 
To construct insertional inactivation mutants, internal segments of the yufL, yufM, maeN and yflS genes were amplified using the primers listed in Table 3, digested at the BamHI or HindIII sites within the primers, and inserted between the BamHI and HindIII sites of pMutinT3 (Vagner et al., 1998). The resulting plasmids were used to transform 168 trpC2 by a single crossover, with selection of the erythromycin-resistant transformants. pMutin-insertion mutants of cimH and mleN were constructed by members of the Japan and European Union consortia of B. subtilis functional genomics.

To create a deletion mutant of the yufL gene, the plasmid pIFDLC was constructed (Fig. 1). An 811 bp fragment encompassing the 7 N-terminal codons of yufL and the upstream region was amplified by PCR using the IFDL-BFF and IFDL-BFR primers, and a 1525 bp fragment encompassing the 5 C-terminal codons of yufL, including the termination codon, and the downstream region was obtained using the IFDL-BRF and IFDL-BRR primers. Then the two fragments were PCR ligated using an overlapping sequence contained within the primers. The resultant fragment (fragment A in Fig. 1) was restricted at the KpnI and BamHI sites in the primers and inserted into KpnI-BamHI-restricted pUC19. Next, a 2348 bp fragment covering the pbpD and yuxK region (fragment B in Fig. 1) was amplified with the IFDL-FF and IFDL-FR primers, digested at the KpnI and XhoI sites in the primers, and inserted between the KpnI and XhoI sites of pUC19 containing fragment A. Finally, a kanamycin-resistance gene of pDG780 (Anne-Marie et al., 1995) was excised by XhoI digestion and inserted between fragments A and B on pUC19. The resulting plasmid, pIFDLC, was linearized by BamHI digestion and used to transform 168 trpC2 cells by a double crossover with selection for kanamycin resistance.



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 1. Schematic of the structure of pIFDLC. Location and direction of transcription of genes are shown by arrows. Bars below the map indicate DNA fragments inserted into the plasmid. Small arrows indicate the primers used to amplify each DNA fragment (a, IFDL-FF; b, IFDL-FR; c, IFDL-BFF; d, IFDL-BFR; e, IFDL-BRF; f, IFDL-BRR). The black circle indicates the PCR ligation point. Above the gene map, a fragment from pDG780 containing a kanamycin-resistance gene is shown.

 
To place the yufL gene under the control of the IPTG-inducible spac promoter at the amyE locus on the B. subtilis chromosome, a 1618 bp fragment containing the Shine–Dalgarno (SD) sequence and the entire coding region of yufL was PCR amplified using the UFLF-F and UFLF-R primers, and cloned into the BamHI site of plasmid pDLT3 (Morimoto et al., 2002) to obtain pDLYufL. Plasmid pDLDSYL was constructed to express a fusion protein of the receiver domain of DctS and the kinase domain of YufL at the amyE locus. The C-terminus portion of yufL (from codon 295 to the termination codon) was amplified with the SLF-BF primer (which has a base substitution to create a KpnI restriction site without changing the coding amino acid) and the SLE-BR primer. The amplified fragment was BamHI restricted and inserted into a BamHI-restricted pDLT3. Next, the DctS receiver domain (codons 1–212) and the YufL kinase domain (codons 212–295) were amplified using primers SLF-MF (which contains the SD sequence of yufL) and SLF-MR, and SLF-BF and SLF-BR, respectively. The two fragments were then PCR ligated using overlapping sequences contained within the SLF-MR and SLF-BR primers. Finally, the resultant fragment was restricted at the XhoI and KpnI sites in the primers and inserted into XhoI-KpnI-restricted pDLT3 containing the C-terminal portion of yufL, to obtain pDLDSYL. Plasmids pDLYufL and pDLDSYL were linearized by ScaI digestion and used to transform 168 trpC2 cells by a double crossover with selection for chloramphenicol resistance.

To construct a series of strains containing transcriptional fusions of serially shortened maeN or yflS promoter regions and a promoterless Bacillus stearothermophilus bgaB gene encoding a thermostable {beta}-galactosidase (Hirata et al., 1986) at the amyE locus on the B. subtilis chromosome, plasmid pDLd (Nanamiya et al., 1998) was used. PCR products containing various lengths of the maeN or yflS promoter region were obtained with the primer sets listed in Table 2, digested at the EcoRI and BamHI sites in the primers, and cloned between the EcoRI and BamHI sites of pDLd. The resulting plasmids were transformed into B. subtilis cells by a double crossover with selection for chloramphenicol resistance.

To create B. subtilis cells having multiple mutations, chromosomal DNA of single mutants was used for transformation.

{beta}-Galactosidase assay.
B. subtilis cells (1 ml) were harvested by centrifugation when the OD600 of cells reached 0·35, and {beta}-galactosidase activity was measured as described by Youngman et al. (1985). Incubation for the assay was at 28 °C for LacZ and at 62 °C for BgaB. Units are expressed as pmol 4-methylumbelliferyl {beta}-D-galactoside hydrolysed min-1 (mg protein)-1. Protein concentration was determined using the Bio-Rad Protein Assay Kit.

Northern blotting and primer-extension analysis.
Total RNA was extracted from wild-type cells growing in GGM medium with and without malate at the exponential phase (OD600 0·35) by the method described by Igo & Losick (1986).

In the Northern analysis, aliquots containing 1 µg total RNA were electrophoresed and blotted onto a nylon membrane (positively charged; Roche). Hybridization and detection were performed using digoxigenin-labelled RNA probes (10 ng), according to the manufacturer's instructions (Roche). Inserts of the pMUFR and pMFLS plasmids containing an internal fragment of the yufR or yflS gene, respectively, were amplified with primers T3T7R and T3F, and used as templates for in vitro RNA synthesis using the DIG RNA Labelling Kit (Roche).

In the primer-extension analysis, 30 µg of the RNA sample was incubated with 32P end-labelled primers (20 fmol) specific for the maeN or yflS genes (EX-R and EX-S, respectively), 250 mM KCl, TE pH 7·4 (10 mM Tris/HCl pH 7·4 and 1 mM EDTA pH 8·0) and dNTPs (1·25 mM each) for 60 min at 65 °C, and gradually cooled to ambient temperature over a period of 60 min. After the addition of buffer (final concentration of 50 mM Tris/HCl pH 8·3, 75 mM KCl, 3 mM MgCl2, 10 mM DTT) and reverse transcriptase (M-MLV Reverse Transcriptase, 400 U; TaKaRa Biomedicals), the reaction mixture was incubated for 120 min at 42 °C. The cDNA products were then electrophoresed through a 6 % polyacrylamide/urea gel, and radioactive bands were detected with Imaging Plates and a BAS2500 scanner (Fuji Photo Film). The sequencing ladder used for size comparison was prepared with the same end-labelled primer and a cycle sequencing kit (TaKaRa Biomedicals) using B. subtilis chromosome DNA as template.

Purification of His6-YufM protein.
In order to express the YufM protein with a histidine tag (His6) at the N-terminus, the entire coding sequence including the initiation codon was PCR amplified with primers UFMF-F and UFMF-R and cloned into plasmid pET15b (Novagen), to obtain pETYufM. Cultivation of E. coli BL21(DE3) pLysS cells containing pETYufM and purification of the His6-YufM protein were performed as previously described (Morimoto et al., 2002).

DNase I footprinting experiments.
As probes for DNase I footprinting assays, a 199 bp fragment (-183 to +16 bp relative to the maeN initiation codon) and a 212 bp fragment (-162 to +50 bp relative to the yflS initiation codon) were amplified by PCR using the RFOOT-F and RFOOT-R primers and the EFLS-F2 and SFOOT-R primers, respectively. To prepare single-stranded labelled probes, one of the primers was end-labelled with [{gamma}-32P]ATP. A constant amount of the probes (100 fmol) and 0·5 µg poly(dI-dC)/poly(dI-dC) as a competitor DNA were incubated with increasing amounts of His6-YufM (8·8, 17, 35, 70 pmol) in 50 µl binding buffer (50 mM PIPES pH 6·25, 200 mM NaCl, 4 mM MgCl2, 4 mM DTT, 20 mM imidazole, 0·5 % Tween 20 and 10 % glycerol) at 25 °C for 30 min. Then 1 µl of a DNase I solution (0·1 U µl-1, TaKaRa Biomedicals) was added to the reaction mixture. After incubation for 1 min at 25 °C, the reaction was stopped by adding 100 µl 20 mM EDTA, and the mixture was subjected to phenol/chloroform/isoamyl alcohol (25 : 24 : 1) extraction and ethanol precipitation. The pellet was dissolved in a loading buffer (0·01 % bromophenol blue, 0·01 % xylene cyanol and 1 mM EDTA in 90 % formamide) and separated on a denaturing 6 % polyacrylamide gel. After electrophoresis, radioactive bands were detected with Imaging Plates and a BAS2500 scanner. DNA ladders used as size markers were created with the appropriate end-labelled primer and a cycle sequencing kit (TaKaRa Biomedicals).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The YufM regulator is essential for malate utilization in B. subtilis
The possible operonic structures of the genes investigated herein, together with positions of the pMutinT3 insertions, are shown schematically in Fig. 2 (see also http://bacillus.genome.ad.jp).



View larger version (23K):
[in this window]
[in a new window]
 
Fig. 2. Operon structure of genes investigated in this study. The location and direction of transcription of genes surrounding yufL, yufM, maeN, yflS, cimH and mleN on the B. subtilis chromosome are shown by arrows. The stem–loop structure indicates a possible transcriptional terminator. The insertion site of the pMutinT3 derivatives in each mutant is shown by inverted triangles. Small arrows indicate the direction of Pspac in integrated plasmids or that of the tet promoter.

 
In order to examine the function of yufL and yufM, we inserted the pMutinT3 plasmid (Vagner et al., 1998) into each gene and monitored the growth of the mutants in Spizizen's minimal salts medium (SM medium) containing malate as a sole carbon source (Fig. 3). Growth of wild-type B. subtilis cells in the malate-only medium was comparable to that in the medium containing glucose as carbon source. In contrast, the yufL and yufM mutants were unable to grow in the malate-only medium, indicating that the YufLM two-component system indeed regulates malate uptake and/or metabolism.



View larger version (17K):
[in this window]
[in a new window]
 
Fig. 3. Growth properties of the yufLM mutants in the SM medium containing malate as a carbon source. The OD600 of wild-type (168 trpC2, {square}), yufL : : pMutinT3 (NT101, {triangleup}) and yufM : : pMutinT3 (NT102, {lozenge}) cells was monitored. NT101 cells were also cultivated in the presence of 1 mM IPTG ({blacktriangleup}).

 
To test whether the growth defect in the yufL : : pMutinT3 mutant could be due to a polar effect of the plasmid insertion on the expression of the downstream yufM gene, we examined the growth of the yufL mutant in the presence of IPTG to guarantee yufM expression from the spac promoter in the integrated plasmid. Unexpectedly, IPTG restored the growth of the mutant to wild-type levels (Fig. 3). This phenomenon will be discussed later.

The expression of the MaeN and YflS transporters is induced by malate in the growth medium
Next, we tested whether possible B. subtilis malate transporter genes are induced by malate in the growth medium. As shown in Fig. 2, the maeN and yflS genes seem to be transcribed as a monocistronic operon. The cimH gene is cotranscribed with the downstream yxkI and yxzE genes of unknown function. The mleN gene constitutes an operon with the downstream mleA, which encodes the malolactate enzyme that converts malate directly to lactate (Wei et al., 2000). In order to examine the expression of the maeN, yflS, cimH and mleN genes, we inserted pMutinT3 into each gene and monitored the expression of the lacZ reporter gene on the integrated plasmid. We cultivated the mutants in a minimal salts medium containing yeast extract as carbon source (NYE medium; used to guarantee the growth of mutants) supplemented with glucose, malate, succinate or fumarate as additional carbon sources, as described previously (Asai et al., 2000). Measurements of the {beta}-galactosidase activity of pMutinT3-inserted mutants in the exponential growth phase revealed that expression of the maeN and yflS genes is specifically induced by malate (data not shown). The expression of yqkI was constitutive and no cimH expression was detected under these growth conditions (data not shown). An operon encoding a possible ABC transport system, YufNOPQ, is located between yufLM and maeN (Fig. 2A). However, these genes were not induced by malate (data not shown).

Activation of the maeN and yflS promoters in the maeN : : pMutinT3 and yflS : : pMutinT3 cells by malate but not by fumarate or succinate was observed even in a minimal medium containing glucose (GGM medium, Fig. 4A), indicating that these genes are not under the control of carbon catabolite repression. Northern blot experiments using RNA from wild-type cells confirmed that 1·35 and 1·45 kb transcripts encompassing the maeN and yflS genes, respectively, are induced by external malate, although the induction of the latter was very weak (Fig. 4B). We performed primer extension experiments using the same RNA preparation, and the transcription initiation site was detected at 27 bp upstream of the initiation codon of maeN, accompanied by a possible -10 sequence of the promoter recognized by E{sigma}A, TGTAGA (Fig. 4C, see also Fig. 8B). However, no clear -35 sequence could be identified positioned at an appropriate distance from the -10 sequence, suggesting that the maeN promoter is regulated by a transcriptional activator. However, we failed to determine the transcription initiation site of yflS, due to insufficient transcript levels and/or existence of a sequence that blocked reverse transcription. Growth of the maeN : : pMutinT3 and yflS : : pMutinT3 cells in SM medium containing malate as the carbon source (Fig. 4D) indicated that YflS apparently plays no role in this growth condition, although its expression is induced. By contrast, the maeN mutant was unable to grow in the malate-only medium, demonstrating that the MaeN transporter is essential for the uptake of malate in the medium.



View larger version (34K):
[in this window]
[in a new window]
 
Fig. 4. Induction of maeN and yflS transcription by malate, and the growth properties of their inactivation mutants. (A) {beta}-Galactosidase activity of maeN : : pMutinT3 (NT103, a) and yflS : : pMutinT3 (NT104, b) cells growing in GGM medium without (-) or with additional C4-dicarboxylates: 2 mM malate (M), 2 mM fumarate (F) or 2 mM succinate (S). (B) Northern analysis using RNA probes specific to the maeN (a) and yflS (b) genes. Total RNA was prepared from B. subtilis 168 trpC2 cells growing in GGM medium (OD600 0·35) with 2 mM malate (lane +) or without malate (lane -). The approximate positions of transcripts are indicated by triangles. Lane M contained an RNA molecular mass marker (Roche). (C) Determination of the maeN transcription-start site by primer extension experiment. RNA preparations from (B) were used for primer-extension analyses with the 32P-labelled EX-R primer complementary to a sequence from 63 to 86 bp downstream from the maeN translation start site (lanes - and +). Lanes T, A, C and G contained sequencing ladders generated by using the primer used for the reverse-transcriptase reaction. The sequence and the transcriptional-start site (indicated by an arrow) are shown. (D) Growth properties of the maeN and yflS mutants in SM medium containing malate as carbon source: wild-type (168 trpC2, {square}), maeN : : pMutinT3 (NT103, {triangleup}), and yflS : : pMutinT3 (NT104, {circ}) were monitored.

 


View larger version (83K):
[in this window]
[in a new window]
 
Fig. 8. Binding of the YufM protein to the cis-acting regulatory sequences for maeN and yflS expression. (A) DNase I footprinting assays of His6-YufM binding to the maeN and yflS promoter region. The YufM protein was fused to a histidine tag (His6) at the N-terminus, expressed in E. coli and purified to homogeneity. A 199 bp fragment encompassing from -183 to +16 bp relative to the maeN initiation codon was radioactively labelled at either end and used as coding and non-coding strand probes for the maeN promoter sequence (a). A 212 bp fragment encompassing from -162 to +50 bp relative to the yflS initiation codon was used as the probe for the yflS promoter sequence (b). A constant amount of the probes (100 fmol) was incubated with different amounts of His6-YufM. The amounts of proteins in lanes 1–6 were 0, 70, 35, 18, 8·8 and 0 pmol, respectively. Lanes A, C, G and T contained sequencing ladders. The maximum protected regions are shown by bars. The numbers are the distances with respect to the maeN and yflS initiation codon. (B) Nucleotide sequence of the maeN (a) and yflS (b) promoter regions. Lower-case letters indicate the His6-YufM protected site. Italic letters indicate the start codon. The experimentally determined transcription-start site is labelled with an arrow. Underlining indicates the putative -10 sequence of the promoter recognized by E{sigma}A. Triangles indicate the end-point of the promoter sequences used in the promoter deletion experiments (Fig. 7). Black, grey and white triangles indicate full, partial and no induction, respectively, following addition of malate to the medium.

 
The YufM regulator is essential for induction of maeN and yflS expression
Next we tested whether the YufM regulator modifies the induction of maeN and yflS. We transcriptionally fused the promoter sequence of each transporter gene, -381 to 230 bp (relative to the initiation codon) for maeN and -339 to 196 bp for yflS, respectively, to a promoterless B. stearothermophilus bgaB gene encoding a thermostable {beta}-galactosidase (Hirata et al., 1986), and integrated the constructs into the amyE locus of the B. subtilis genome. As shown in Fig. 5, the expression of PmaeN : : bgaB and PyflS : : bgaB in an otherwise wild-type background was induced by malate in a manner similar to that observed for the native promoters, while further introduction of the yufM : : pMutinT3 mutation completely abolished their induction. These results demonstrated that, as expected, the YufM regulator activates the maeN and yflS promoters in response to malate in the growth medium.



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 5. The YufM regulator regulates the induction of maeN and yflS transcription. Promoter sequences of each transporter gene, -381 to +230 bp (relative to the initiation codon) for maeN and -339 to +196 bp for yflS, respectively, were transcriptionally fused to the bgaB reporter gene and integrated into the amyE locus. {beta}-Galactosidase activities in otherwise wild-type background cells (NT401 and NT501, grey bars) and in yufM inactivated cells (NT402 and NT502, white bars) were compared. Cells were grown to OD600 0·35 in GGM medium without (-) or with additional C4-dicarboxylates: 2 mM malate (M), 2 mM fumarate (F) or 2 mM succinate (S).

 
Role of the YufL sensor in the regulation of maeN expression
The growth defect of the yufL : : pMutinT3 mutant in SM medium with malate as carbon source was restored by the addition of IPTG to the medium (Fig. 3). Consistent with this observation, we found that the expression of PmaeN : : bgaB at the amyE locus became constitutively active in yufL : : pMutinT3 mutant cells cultivated in the presence of IPTG (data not shown). We questioned whether transcription by the spac promoter resulted in an overproduction of the YufM regulator that led to constitutive activation of maeN expression, although such a phenomenon has not been observed in similar mutants of citS and dctS (Yamamoto et al., 2000; Asai et al., 2000). Alternatively, the ectopic placement of the maeN promoter might give unexpected results. To test these possibilities, we made a yufL deletion mutant that allowed expression of YufM from its native promoter in the absence of YufL. We introduced this into the maeN : : pMutinT3 cells to monitor the expression of the native maeN promoter (NT205). In addition, we removed the spac promoter in pMutinT3 in this strain, to avoid IPTG-induced side effects in subsequent experiments. The results shown in Fig. 6(a) indicate that maeN expression became constitutively active in the absence of the YufL sensor. Then we ectopically expressed the YufL protein at the amyE locus using the spac promoter (NT206). PmaeN expression was still constitutive in NT206 cells grown in the absence of IPTG (Fig. 6b), but the addition of IPTG to the culture medium restored the malate-dependent expression (Fig. 6c). Sensor kinases of the CitA family consist of the N-terminal sensor domain flanked by two transmembrane helices (TMH) and the C-terminal intracellular autokinase domain that is connected to the second TMH with a linker region of about 100 aa in length (Kaspar et al., 1999). The kinase domain of the sensor is expected to interact with the cognate regulator. Therefore, we also expressed the linker and kinase domains of YufL (residues 212–533) as a fusion with the extracellular sensor domain including two TMHs of the fumarate/succinate sensor, DcuS (residues 1–212), in NT205 cells. Interestingly, the expression of the kinase and linker domains was sufficient for the repression of the constitutive activation of PmaeN, although the expression of PmaeN was not activated by the addition of malate, fumarate or succinate. These results demonstrate that the YufL sensor is required for proper signal detection and signalling specificity, but in a manner different from that of typical two-component system kinases, including B. subtilis CitS and DctS.



View larger version (17K):
[in this window]
[in a new window]
 
Fig. 6. Role of the YufL sensor in the regulation of maeN expression. NT205 ({Delta}yufL, a), NT206 ({Delta}yufL amyE : : Pspac-yufL, b and c) and NT207 [{Delta}yufL Pspac-dctS'(sensor domain)-yufL' (linker and kinase domains), d] cells were grown to OD600 0·35 in GGM medium without (-) or with additional C4-dicarboxylates – 2 mM malate (M), 2 mM fumarate (F) or 2 mM succinate (S) – and {beta}-galactosidase activities were determined. IPTG was added at a final concentration of 1 mM to NT206 (c) and NT207 (d).

 
The sensor kinases usually activate the cognate response regulators by phosphorylating them in the presence of environmental signals. Besides directing the phosphorylation reaction, some sensor kinases have been reported to possess a phosphatase activity, enabling them to dephosphorylate their cognate response regulators (reviewed by Stock et al., 2000). Our observation would be explained if we assume that the YufM protein is nonspecifically phosphorylated independently of YufL, and that YufL specifically dephosphorylates the phospho-YufM in the absence of malate. When malate is present in the medium, the phosphatase activity of YufL would be suppressed and the specific phosphotransfer reaction to YufL activated.

The E. coli PhoB and the B. subtilis ComA regulators have been reported to be phosphorylated by acetyl phosphate in the absence of the cognate sensor kinase (Kim et al., 1996, 2001). Therefore we further disrupted the pta gene, encoding the phosphotransacetylase that is responsible for conversion of acetyl coenzyme A to acetyl phosphate, in NT205 cells. However, the constitutive maeN expression was not suppressed by the pta inactivation (data not shown).

Binding of the YufM regulator to cis-acting sequences required for the induction of maeN or yflS expression
In order to determine the cis-acting region necessary for the maeN and yflS induction by malate and YufM, we performed promoter deletion experiments using the previously described amyE : : PmaeN-bgaB and amyE : : PyflS-bgaB cells. We found that maeN promoter activity was induced by malate at the same level as in the initial construct if 98 bp of sequence upstream of the initiation codon was retained, but a further deletion of 6 bp resulted in about 80 % reduction in the induction and the additional deletion of 3 bp completely abolished the promoter activity (Fig. 7a; see also Fig. 8B). Similarly, a full induction of the yflS expression was observed in cells having at least 92 bp of sequence upstream of the initiation codon, while removal of an additional 7 bp abolished the promoter activity (Fig. 7b, see also Fig. 8B).



View larger version (22K):
[in this window]
[in a new window]
 
Fig. 7. Mapping of the cis-acting sequence that is required for the induction of maeN and yflS expression. Transcriptional fusions of the gradually shortened maeN (a) and yflS (b) promoter regions and the bgaB reporter gene at the amyE locus were constructed as shown schematically in the figure. The {beta}-galactosidase activities of these strains in GGM medium with or without malate are listed on the right. The lengths of the promoter regions (from the initiation codon) in each strain are listed on the left.

 
Then we confirmed the binding of YufM to the identified cis-acting regulatory sequences of maeN and yflS. The YufM protein was fused to an N-terminal histidine tag (His6), expressed in E. coli and purified to homogeneity by nickel-affinity chromatography (data not shown). Then, we performed DNase I footprinting experiments to precisely map the YufM binding sequence. Fig. 8(A) shows that YufM protected sequences of about 50 bp in both the coding and non-coding strands of the maeN and yflS promoters. These protected regions are consistent with the cis-regulatory region for the maeN and yflS induction identified by the promoter deletion experiments, as summarized in Fig. 8(B). However, the protection was only apparent at a high concentration of protein, 700-fold excess with respect to the DNA, suggesting that phosphorylation of YufM will increase the affinity to DNA and will be required for the in vivo action.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Gram-negative bacteria such as E. coli and K. pneumoniae have been shown to monitor external citrate or fumarate and regulate the expression of genes involved in their anaerobic metabolism including transporters for their uptake (reviewed by Janausch et al., 2002). The B. subtilis genome sequence contains three paralogous sensor kinase and response regulator pairs, CitST, DctSR and YufLM, consisting of the CitA and CitB families of sensor kinases and their cognate regulators (Fabret et al., 1999; Kasper et al., 1999). Our group and others have reported that the B. subtilis DctSR and CitST are essential for aerobic growth using citrate or fumarate/succinate as carbon sources, respectively, through the induction of expression of the transporter gene responsible for their uptake (Asai et al., 2000; Yamamoto et al., 2000). Here we report that the YufM regulator binds in vitro to the promoter region of two malate transporter genes, maeN and yflS, and is responsible for induction of maeN and yflS expression by malate in the growth medium in vivo. We also demonstrated that nonfunctional mutants of yufM and maeN completely lost the ability to grow in the minimal salts medium (SM medium) containing malate as a sole carbon source, indicating that the induction of the MaeN transporter through the action of the YufM regulator is essential for the utilization of malate.

We have reported DNA microarray analyses of 24 two-component systems, including yufLM, to identify the target genes of each system (Kobayashi et al., 2001). Each response regulator was overproduced in cells deficient for the cognate sensor kinase, and expression of 98 genes was found to be significantly induced or repressed by the YufM overproduction. However, maeN and yflS expression was not significantly affected. We found that expression of genes under the control of ComK was activated by YufM overproduction, for unknown reasons. This peculiar phenotype might be the reason why we did not detect maeN and yflS as target of YufLM.

It has been suggested that at least two transport systems for malate may be induced in B. subtilis (Willecke & Lange, 1974). The first system, required for the uptake of malate, fumarate and succinate, is induced in NYE medium. The second system is then induced by the addition of large amounts of malate to the growth medium. The Dct system (Asai et al., 2000) would correspond to the first C4-dicarboxylate transport system reported by Willecke & Lange (1974), although it has not yet been determined whether the DctP transporter uses malate as a substrate, and the second system will be the MaeN (and YflS) transporter characterized in this report.

Although its precise biochemical role is as yet unknown, the YflS protein seemed to act as a malate transporter in B. subtilis, in that it was induced by malate in a YufM-dependent manner. In addition to MaeN and YflS, CimH and YqkI have been reported to transport malate (Wei et al., 2000; Krom et al., 2001). However, we did not detect cimH expression in our cultivation conditions. We observed MleN expression independent of malate addition. Krom et al. (2000) reported that the growth of the mleN disruptant was indistinguishable from the wild-type cell in SM medium with either glucose or malate. Therefore, the physiological role of CimH and MleN is still unclear at this time.

The deletion analysis of maeN and yflS promoter regions and the DNase I footprinting experiments clearly demonstrated that the YufM regulator binds to the region essential for the activation of maeN and yflS transcription by external malate. Based on deletion analysis of the dcuP promoter, Asai et al. (2000) proposed that two tandem repeats of AGACCAAA compose the DcuR binding sequence. Yamamoto et al. (2000) identified the multiple repeats of (A/T)(A/T)CAAA in two CitT regulator binding regions in the citM promoter. However, we could not identify such clear repeat sequences in the YufM-binding regions of the maeN and yflS promoter sequences. Instead, we theorize that YufM might recognize some higher-order structure of the protected sequences.

The YufL sensor is required for proper signal detection and signalling specificity. Inactivation of the yufL gene resulted in the constitutive activation of YufM, which was suppressed by reintroduction of the kinase domain of YufL. Our observation suggests that YufM protein is phosphorylated independently of YufL and that YufL dephosphorylates the phospho-YufM in the absence of external signal, malate. Malate in the growth medium would suppress the phosphatase activity of YufL and activate the specific phosphotransfer reaction to YufL. It has been reported that the E. coli PhoB and the B. subtilis ComA regulators are phosphorylated by acetyl phosphate in the absence of the cognate sensor kinase, and that the inactivation of the pta gene abolishes the phosphorylation (Kim et al., 1996, 2001). However, the constitutive activation of YufM was not suppressed by the pta inactivation. YufM might be phosphorylated nonspecifically by two-component kinases other than YufL. It is also possible that the YufM protein is activated by a small phosphodonor other than acetyl phosphate.

In Gram-negative bacteria, two-component regulatory systems regulating the uptake of C4-dicarboxylates also regulate the expression of genes involved in their metabolism (Janausch et al., 2002). A search for orthologues of the yufLM and maeN genes in complete genome sequences of bacteria using the MBGD database (http://mbgd.genome.ad.jp) revealed that the yufNOPQ operon found between the yufLM and maeN genes in B. subtilis is missing in the Bacillus halodurans and Oceanobacillus iheyensis genomes and the ytsJ gene encoding a protein highly similar to malate oxidoreductase is located there in these organisms, suggesting that this additional malate metabolism gene is also under the control of YufLM. However, the expression of B. subtilis ytsJ was independent of malate in NYE medium (data not shown). Instead, another malate oxidoreductase gene in B. subtilis, ywkA, has recently been found to be regulated by YufM and induced by malate (Doan et al., 2003). This finding suggests that the B. subtilis CitST and DctSR might also regulate the expression of genes for metabolism of citrate or C4-dicarboxylates, although these have not yet been identified.

The studies reported here, together with previous reports (Asai et al., 2000; Yamamoto et al., 2000), have revealed that members of a set of paralogous two-component regulatory systems in B. subtilis, CitST, DctSR and YufLM, are involved in a related function, uptake (and metabolism) of the TCA cycle intermediates, but with distinct substrate specificities. In addition, different characteristics among them have been revealed. The DctS sensor requires an additional membrane protein, supposed to be C4-dicarboxylate binding protein, DctB (YdbE), for recognition of fumarate/succinate in the growth medium (Asai et al., 2000). Results presented in this report suggest that YufM protein would be nonspecifically phosphorylated by two-component kinases other than YufL, and the phosphatase activity of YufL would play an important role in proper signal detection and signalling specificity. The mode of recognition of and binding to DNA seem to be different between YufM and CitT/DctR. Further studies on the structure–function relationships of these homologous two-component sensors and kinases will be valuable in elucidating the evolution of the two-component regulatory systems.


   ACKNOWLEDGEMENTS
 
We are grateful to S. Aymerich and his colleagues for communicating unpublished results. This work was supported by a Grant-in-Aid for Scientific Research on Priority Areas, Genome Biology, from the Ministry of Education, Culture, Sports, Science and Technology of Japan.


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

Anne-Marie, G., Kamran, S., Niels, F. & Patrick, S. (1995). Antibiotic-resistance cassettes for Bacillus subtilis. Gene 167, 335–336.[CrossRef][Medline]

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]

Boorsma, A., van der Rest, M. E., Lolkema, J. S. & Konings, W. N. (1996). Secondary transporters for citrate and the Mg2+-citrate complex in Bacillus subtilis are homologous proteins. J Bacteriol 178, 6216–6222.[Abstract]

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

Doan, T., Servant, P., Tojo, S., Yamaguchi, H., Lerondel, G., Yoshida, K., Fujita, Y. & Aymerich, S. (2003). 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. Microbiology 149, 2331–2345.[Abstract/Free Full Text]

Fabret, C., Feher, V. A. & Hoch, J. A. (1999). Two-component signal transduction in Bacillus subtilis: how one organism sees its world. J Bacteriol 181, 1975–1983.[Free Full Text]

Fortnagel, P. & Freese, E. (1968). Analysis of sporulation mutants. II. Mutants blocked in the citric acid cycle. J Bacteriol 95, 1431–1438.[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]

Hirata, H., Fukazawa, T., Negoro, S. & Okada, H. (1986). Structure of a {beta}-galactosidase gene of Bacillus stearothermophilus. J Bacteriol 166, 722–727.[Medline]

Igo, M. & Losick, R. (1986). Regulation of a promoter that is utilized by minor forms of RNA polymerase holoenzyme in Bacillus subtilis. J Mol Biol 191, 615–624.[Medline]

Janausch, I. G., Zientz, E., Tran, Q. H., Kroger, A. & Unden, G. (2002). C4-dicarboxylate carriers and sensors in bacteria. Biochim Biophys Acta 1553, 39–56.[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]

Kim, S. B., Shin, B. S., Choi, S. K., Kim, C. K. & Park, S. H. (2001). Involvement of acetyl phosphate in the in vivo activation of the response regulator ComA in Bacillus subtilis. FEMS Microbiol Lett 195, 179–183.[CrossRef][Medline]

Kim, S. K., Wilmes-Riesenberg, M. R. & Wanner, B. L. (1996). Involvement of the sensor kinase EnvZ in the in vivo activation of the response-regulator PhoB by acetyl phosphate. Mol Microbiol 22, 135–147.[Medline]

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]

Krom, B. P., Warner, J. B., Konings, W. N. & Lolkema, J. S. (2000). Complementary metal ion specificity of the metal-citrate transporters CitM and CitH of Bacillus subtilis. J Bacteriol 182, 6374–6381.[Abstract/Free Full Text]

Krom, B. P., Aardema, R. & Lolkema, J. S. (2001). Bacillus subtilis YxkJ is a secondary transporter of the 2-hydroxycarboxylate transporter family that transports L-malate and citrate. J Bacteriol 183, 5862–5869.[Abstract/Free Full Text]

Morimoto, T., Loh, P. C., Hirai, T., Asai, K., Kobayashi, K., Moriya, S. & Ogasawara, N. (2002). Six GTP-binding proteins of the Era/Obg family are essential for cell growth in Bacillus subtilis. Microbiology 148, 3539–3552.[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]

Nanamiya, H., Ohashi, Y., Asai, K., Moriya, S., Ogasawara, N., Fujita, M., Sadaie, Y. & Kawamura, F. (1998). ClpC regulates the fate of a sporulation initiation sigma factor, {sigma}H protein, in Bacillus subtilis at elevated temperatures. Mol Microbiol 29, 505–513.[CrossRef][Medline]

Saier, M. H., Jr, Goldman, S. R., Maile, R. R., Moreno, M. S., Weyler, W., Yang, N. & Paulsen, I. T. (2002). Overview of transport in Bacillus subtilis. In Bacillus subtilis and Its Closest Relatives: from Genes to Cells, pp. 113–128. Edited by A. L. Sonenshein, J. A. Hoch & R. Losick, Washington, DC: American Society for Microbiology.

Stock, A. M., Robinson, V. L. & Goudreau, P. N. (2000). Two-component signal transduction. Annu Rev Biochem 69, 183–215.[CrossRef][Medline]

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

Weber, A., Menzlaff, E., Arbinger, B., Gutensohn, M., Eckerskorn, C. & Flugge, U. I. (1995). The 2-oxoglutarate/malate translocator of chloroplast envelope membranes: molecular cloning of a transporter containing a 12-helix motif and expression of the functional protein in yeast cells. Biochemistry 34, 2621–2627.[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]

Willecke, K. & Lange, R. (1974). C4-dicarboxylate transport in Bacillus subtilis studied with 3-fluoro-L-erythro-malate as a substrate. J Bacteriol 117, 373–378.[Medline]

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]

Youngman, P., Perkins, J. & Sandman, K. (1985). Use of Tn917-mediated transcriptional gene fusions to lacZ and cat-86 for the identification and study of regulated genes in the Bacillus subtilis chromosome. In Molecular Biology of Microbial Differentiation, pp. 47–54. Edited by J. A. Hoch & P. Setlow. Washington, DC: American Society for Microbiology.

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 27 January 2003; revised 6 May 2003; accepted 6 May 2003.