JB
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Joseph, P.
Right arrow Articles by Sonenshein, A. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Joseph, P.
Right arrow Articles by Sonenshein, A. L.
Journal of Bacteriology, June 2005, p. 4127-4139, Vol. 187, No. 12
0021-9193/05/$08.00+0     doi:10.1128/JB.187.12.4127-4139.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

A Region of Bacillus subtilis CodY Protein Required for Interaction with DNA

Pascale Joseph, Manoja Ratnayake-Lecamwasam,{dagger} and Abraham L. Sonenshein*

Department of Molecular Biology and Microbiology, Tufts University School of Medicine, Boston, Massachusetts 02111

Received 13 October 2004/ Accepted 10 March 2005


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Bacillus subtilis CodY protein is the best-studied member of a novel family of global transcriptional regulators found ubiquitously in low-G+C gram-positive bacteria. As for many DNA-binding proteins, CodY appears to have a helix-turn-helix (HTH) motif thought to be critical for interaction with DNA. This putative HTH motif was found to be highly conserved in the CodY homologs. Site-directed mutagenesis was used to identify amino acids within this motif that are important for DNA recognition and binding. The effects of each mutation on DNA binding in vitro and on the regulation of transcription in vivo from two target promoters were tested. Each of the mutations had similar effects on binding to the two promoters in vitro, but some mutations had differential effects in vivo.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The Bacillus subtilis CodY protein controls the expression of several hundred genes that are generally involved in the response to nutrient limitation, including genes that encode extracellular degradative enzymes, transport systems, intracellular catabolic systems, chemotaxis, motility, genetic competence, and sporulation (1, 5, 9, 19-21, 26, 27, 37). CodY is highly active as a repressor in rapidly growing cells in rich medium. Repression of CodY target genes is relieved as cells make the transition from rapid exponential growth to stationary phase.

Two types of small-molecule effectors are now known to modulate the activity of CodY. GTP binds to CodY and activates it as a repressor (33); the branched-chain amino acids (BCAAs) isoleucine and valine also interact with CodY and enhance its binding to target sites (39). Genes under CodY control are, therefore, repressed when intracellular pools of GTP and isoleucine-valine are high and derepressed when these pools are depleted by nutrient limitation.

In Lactococcus lactis, CodY has been shown to control expression of intracellular peptidase, peptide transport, and aminotransferase genes in response to BCAA availability (4, 12, 13). In fact, close homologs of CodY can be found encoded in the genomes of most of the low-G+C gram-positive bacteria (33). It is likely that CodY plays similar roles in many of these bacterial species.

CodY is an unusual transcription factor in that it shows no significant overall similarity to any other types of transcriptional regulators characterized to date. The mechanism by which CodY is able to recognize and bind to target promoters is unknown at this time. No consensus sequence common to the regulated promoters has yet been discerned. CodY-controlled promoters are very A+T-rich, and at least one (the dpp promoter) has an intrinsic DNA bend (43). Serror and Sonenshein (38) suggested that CodY might recognize a three-dimensional structure in the DNA rather than a linear nucleotide sequence. On the other hand, a putative helix-turn-helix (HTH) region (3, 7) was identified near the C terminus of CodY; an in-frame deletion of this region rendered the protein unable to bind to the dpp promoter in vitro (38). The same HTH deletion caused derepression of the dpp promoter in vivo. The apparent role of this HTH domain in DNA binding suggests that CodY may be a sequence-specific DNA-binding protein, even though no consensus sequence has been found.

The HTH motif is a DNA-binding domain frequently found in bacterial transcriptional regulators (3, 15, 31, 32). X-ray crystallographic and two-dimensional nuclear magnetic resonance studies of bacterial and phage transcription factors have revealed that the basic HTH motif spans approximately 20 amino acids and that the two {alpha}-helices are placed at an angle of about 120o. The helices are typically linked by a flexible region of 3 or 4 amino acids. The amino-terminal helix 1 (stabilizing helix) sits above the major groove, near the DNA backbone, and the flexible turn region allows the carboxy-terminal helix 2, called the recognition helix, to form sequence-specific interactions with DNA in the major groove (31). However, the interactions necessary for DNA binding are not limited to helix 2. In some cases, residues of helix 1 (e.g., in the Escherichia coli Trp repressor) and amino acids flanking the HTH motif (e.g., helix III of catabolite gene activator protein, the N-terminal arm of {lambda} repressor) participate in DNA binding (15). Thus, recognition and specificity of HTH-mediated protein-DNA interactions may depend on the context of the HTH motif in DNA-binding domains.

The amino acid sequence from residues 203 to 222 of CodY protein resembles a typical HTH motif (Fig. 1A). By comparison with the HTH regions of well-characterized transcription factors, such as phage {lambda} and 434 repressors and {lambda} Cro protein (32), residues arginine-214 (R214), serine-215 (S215), and valine-218 (V218) of CodY helix 2 might be expected to contact target DNA. In addition, alanine-207 (A207) of helix 1 might be implicated in a hypothetical hydrophobic interaction with isoleucine-217 (I217) of helix 2 (Fig. 1A). The putative CodY HTH motif is highly conserved among the CodY homologs (≥80% identity; Fig. 1B), suggesting that CodY homologs recognize and bind target promoters in a similar way. Although crystals of CodY protein have been obtained (2), the three-dimensional structure of CodY has not yet been determined. Despite the absence of such information, it should be possible to identify key interactions between CodY protein and target DNAs by in vitro binding studies and mutational analysis. Since B. subtilis CodY was the first member of a novel family of transcriptional regulators identified, potential DNA-interacting and helix-stabilizing residues of the putative CodY HTH region were subjected to site-directed mutagenesis. We report the effects of such mutations on DNA binding and oligomerization capabilities of CodY in vitro and on the regulation of target promoters in vivo. Interestingly, some mutations had differential effects on different target sites.



View larger version (40K):
[in this window]
[in a new window]
  		FIG. 1. The putative CodY helix-turn-helix motif. (A) The region encompassing residues 200 to 224 of the B. subtilis CodY protein is drawn as a double-barrel structure, showing the amino acid substitutions created in the various mutants used in this study. The positions of these and others residues in the full protein sequence are indicated in parentheses. A hypothetical interaction between residue 5 of helix 1 (A207) and residue 15 of helix 2 (I217) (31), is indicated by a dashed line. (B) Alignment of amino acid sequences of the putative HTH of CodY homologs using the Clustal W algorithm (44). Residues completely conserved in all CodY homologs are indicated with an asterisk at the bottom of the figure.

 

   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Bacterial strains and plasmids. The B. subtilis and E. coli strains used in this study are described in Tables 1 and 2, respectively.


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

 

View this table:
[in this window]
[in a new window]
  		TABLE 2. E. coli strains used in this study

 
Culture media and DNA manipulations. E. coli strains were grown in Luria-Bertani (LB) medium (25a) in the presence of ampicillin (50 µg ml–1) or chloramphenicol (25 µg ml–1) when appropriate. Plasmid DNA was isolated from E. coli by the method of Sambrook et al. (34) or by using a QIAprep spin miniprep kit (QIAGEN). Restriction enzyme digestions and ligations were carried out in accordance with instructions provided by the suppliers. DNA fragments were purified from agarose gels by using a GeneClean kit (Bio 101) or a QIAquick gel extraction kit (QIAGEN). Preparation and transformation of electrocompetent E. coli cells were as described previously by others (8). Electroporations were performed with a Gene Pulser apparatus (Bio-Rad Laboratories) or a BTX TransPorator Plus. B. subtilis chromosomal DNA isolation, competent-cell preparation, and transformation were done as described previously (10). B. subtilis cells were grown in DS medium (10). To monitor ß-galactosidase expression on plates, 5-bromo-4-chloro-3-indolyl-ß-D-galactoside (X-Gal) was added at 80 µg ml–1.

Site-directed mutagenesis of the CodY HTH region and overexpression of mutant CodY proteins in E. coli. Three related protocols were used to create point mutations in the CodY HTH region. In all cases, an overlap PCR procedure adapted from the method of Mikaelian and Sergeant (25) was used. In protocol 1 (used for strains with A207D, S215F, S215T, R214K, R214E, and V218D mutations), the template plasmid used for mutagenesis was pPS40 (38). To create each mutation, two separate PCRs were carried out. One was primed with a mutagenic oligonucleotide (listed in Table 3) and OMRL5 (5'-CGCCAAGCTCGAAATTAACCC-3'); the other was primed with OMRL4 (5'-CTTGTCCGAACCCTCAAAAAGGGTTC-3') and OMRL3 (5'-GGCCGCGGATCCTGAGGAATGTTTAGGAGG-3'). OMRL3 annealed upstream of the codY coding sequence and contained a BamHI restriction enzyme recognition site (indicated in boldface) at the 5' end. The products of these reactions were mixed, denatured, reannealed, and amplified using OMRL3 and OMRL5. Following the final PCR step, the amplified mutant codY gene products were digested with restriction endonucleases BamHI (at a site located in the OMRL3 sequence) and HindIII (at a site located immediately downstream of the codY gene) and ligated to the BamHI- and HindIII-digested pT7cat6 vector (16). The resultant recombinant plasmids were introduced into E. coli strain JM107 (48) by electroporation. The inserts were sequenced (35) to verify the presence of the expected mutation.


View this table:
[in this window]
[in a new window]
  		TABLE 3. Mutagenic oligonucleotides used in this study

 
To introduce the mutations into the codY gene in the B. subtilis chromosome, the mutant codY alleles were released from their pT7cat6 derivatives by digestion with BamHI and HindIII and ligated to pJPM1, a plasmid that replicates in E. coli but not in B. subtilis (28). Each pJPM1 derivative (pMRL7, pMRL8, pMRL9, pMRL10, pMRL11, and pMRL15) (Table 2) was then used to transform competent cells of B. subtilis strains carrying a dpp-lacZ fusion (PS59), generating merodiploid strains carrying both mutant and wild-type alleles of codY. The chromosomal DNA of strain PS56 was then used to transform competent cells of the merodiploid strains to introduce the abrB null mutation. The sequence of each mutant codY allele in the merodiploid strains was verified by amplifying the codY gene using OPJ1 (5'-GGGGGAGCTCGCTCGGAACGATAGCC-3') and OMRL5 primers and B. subtilis chromosomal DNA as the template. Primer OPJ1 anneals upstream of the codY gene and primer OMRL5 to the integrated pJPM1 sequence. The codY mutations in the merodiploid strains were transferred to strain PJB14 (ilvB{Delta}T-lacZ) by transformation.

To produce pure CodY for in vitro analysis, the mutant codY genes were specifically amplified by PCR in two steps. First, PCR was performed using OPJ1 and OMRL5 primers to amplify the promoter-proximal codY allele. The resulting PCR product was then used as a template for amplification with OKT1 and OKT2 primers (21). OKT1 anneals upstream of the codY gene and includes the codY Shine-Dalgarno sequence and a SacI restriction site. OKT2 appends six histidine codons to the 3' end of the codY gene, followed by a stop codon and an SphI restriction site. The PCR products were digested with SacI and SphI and ligated to SacI-and-SphI-digested pBAD30 (14) and introduced into E. coli strain JM107 by electroporation. As a result, the codY gene was placed under the control of the araBAD promoter (21). The integrity of each cloned codY gene was verified by sequencing using vector-specific primers OBB116 (5'-CTCCATACCCGTTTTTTTGG-3') and OBB117 (5'-CTCTCATCCGCCAAAACAG-3').

In protocols 2 and 3, the template plasmid used for mutagenesis was pKT1, a pBAD30 derivative harboring the wild-type codY gene with a six-histidine codon C-terminal extension (21). In protocol 2, used to create the mutation A207V, two separate PCRs were carried out, one with mutagenic primer OMRL14 (Table 3) and OBB117 and the other one with OKT1 and OKT2 primers. The two resulting PCR products were mixed, denatured, reannealed, and amplified using OKT1 and OBB117 primers. In protocol 3, used to create the mutations V218A and S215A, one PCR was carried out with the mutagenic primer OPJ2 or OPJ49 (Table 3) and OKT1 and the other one with mutagenic primer OMRL6 or OPJ48 and OKT2. The two PCR products were mixed, denatured, reannealed, and amplified using OKT1 and OKT2 primers. The final PCR products in both cases were digested with SacI and SphI and cloned in pBAD30, generating plasmids pJP2, pJP8, and pJP53. The sequence of each mutated codY gene was verified as indicated above.

To introduce mutations A207V, V218A, and S215A into the chromosomal codY gene, the mutated codY gene from plasmids pJP2, pJP8, and pJP53 was amplified using primers OPJ3 and OPJ4. OPJ3 (5'-GGGGAATTCACAAGAATTATTAACTCCATGCTGCAAGC-3') annealed to the codY coding sequence 18 bp after the initiating ATG codon and contained an EcoRI recognition site (indicated in boldface). OPJ4 (5'-GGGGCTGCAGTTAATGAGATTTTAGATTTTCTAATTC-3') contains the 3' end of the codY coding sequence and an PstI restriction site (indicated in boldface). The PCR products were digested with EcoRI and PstI and ligated to EcoRI-and-PstI-digested pJPM1 and introduced into E. coli strain JM107 by electroporation. The derivative plasmids, pJP9, pJP10, and pJP54, were used to transform B. subtilis strain FJS107. The chromosomal DNA of the resultant B. subtilis codY HTH mutant strains PJB1, PJB2, and PJB32 was then used to transform B. subtilis strain PS56 carrying the dpp-lacZ fusion and an abrB null mutation, creating strains PJB3, PJB4, and PJB34, respectively. The chromosomal DNA of strains PJB3 and PJB4 was then used to transform strain PJB14, carrying the ilvB{Delta}T-lacZ fusion, to introduce the HTH mutations into the codY gene. The plasmid pJP54 was used to transform competent cells of strain PJB14, generating strain PJB33. In each case, the B. subtilis chromosomal alleles of the codY mutants were verified by DNA sequencing after amplification by PCR with primers OMRL5 and OPJ1 as described above.

Purification of CodY and mobility shift DNA binding assays. E. coli strains carrying plasmid pBAD30 in which the different versions of the codY gene were cloned (Table 2) were grown until the optical density at 600 nm (OD600) reached {approx}0.6. Arabinose (0.2% final concentration) was added, and growth was continued for 4 to 5 h. Cells were then harvested by centrifugation and broken by sonication. CodY proteins were purified by cobalt affinity chromatography as previously described (21, 39) using 75 mM imidazole for elution. Elution fractions were free of detectable contaminating proteins as determined by Coomassie blue staining of sodium dodecyl sulfate (SDS)-polyacrylamide gels.

A 256-bp dpp promoter-containing DNA (from positions –172 to +84 relative to the transcription start site) was PCR amplified using vector-specific primers T3 and T7 (Stratagene) and the plasmid pFS48 (40) as the template. A 453-bp ilvB promoter-containing probe (from positions –306 to +147 relative to the transcription start site) was generated by PCR amplification using plasmid pRPS5 as the template and primers ORPS6 and ORPS7 (39). The purified PCR products were labeled using [{gamma}-32P]ATP and T4 polynucleotide kinase (Invitrogen) as described by the manufacturer. The labeled PCR products were purified on a 8% (wt/vol) nondenaturing polyacrylamide gel by using the Qiaex II gel extraction kit (QIAGEN). Mobility shift DNA binding assays were performed as described previously (27, 39) in the presence of 2 mM GTP and 10 mM each of isoleucine, leucine, and valine. The samples were loaded on a 10% nondenaturing polyacrylamide gel, and electrophoresis was carried out in 35 mM HEPES-43 mM imidazole electrophoresis buffer (pH 7.4) as described previously (39). Gels were dried under vacuum and exposed to a phosphorimager screen before analysis with a Molecular Dynamics Storm 860 Imager and ImageQuant version 1.2 Macintosh software.

DNase I footprinting assays. The ilvB DNA fragment was PCR amplified from plasmid pRPS5 using ORPS7 and 32P-labeled ORPS6 as primers. The dpp DNA fragment was PCR amplified from plasmid pFS48 using the T7 primer and 32P-labeled T3 primer. T3 and ORPS6 primers were labeled using T4 polynucleotide kinase (Invitrogen) and [{gamma}-32P]ATP as described previously (22). Conditions used for protein binding were the same as for gel shift assays but in a 20-µl reaction mixture volume. After incubation for 30 min at room temperature, 6 mM MgCl2, 6 mM CaCl2, and RQ1 DNase (Promega; 0.059 and 0.13 unit for dpp and ilvB fragments, respectively) were added to each mixture. After 1 min at room temperature, the reactions were stopped by the addition of EDTA to 25 mM and transferred to a dry ice-ethanol bath. Samples were extracted with phenol-chloroform, precipitated in ethanol, and resuspended in 5 µl of sequencing gel loading buffer (34). The samples were incubated for 5 min at 80°C and subsequently analyzed on a 6% urea-polyacrylamide DNA sequencing gel prepared in TBE (89 mM Tris, 89 mM borate, 2 mM EDTA). Sanger sequencing reactions for the dpp promoter were performed with the T3 primer on plasmid pFS48 using the Sequenase kit (United States Biochemical Corp.) and [{alpha}-35S]dATP. The dried gels were analyzed as described above.

Expression of mutant CodY proteins in B. subtilis. B. subtilis strains carrying a codY HTH mutation (PJB22, PJB23, PJB24, PJB25, PJB26, PJB27, PJB28, and PJB29) or a codY deletion mutation (PJB16) or the wild-type codY gene (PJB14) were grown in 20 ml of DS medium to an OD600 of approximately 1. A 10-ml sample of each culture was removed, and the cells were pelleted by centrifugation at 13,000 rpm for 3 min at 4°C. The cell pellets were then resuspended in 2 ml solution A (38) containing 1 mM phenylmethylsulfonyl fluoride. The cells were sonicated on ice in 30-second pulses with 30-second breaks between pulses. The lysates were then centrifuged at 13,000 rpm at 4°C. The quantity of total protein present in the supernatant fluid was determined using the Bio-Rad protein assay reagent. Ten micrograms of total soluble protein was loaded on a 12% denaturing polyacrylamide gel after boiling for 3 minutes in Laemmli loading buffer (62.5 mM Tris, pH 6.8, 2% SDS, 10% glycerol, 0.015% bromophenol blue) containing 2.5% ß-mercaptoethanol. Electrophoresis was conducted at a constant voltage of 110 V for 1 hour. Proteins were then electrotransferred to Immobilon-P membranes (Millipore) at 4°C under constant voltage of 100 V in electrotransfer buffer (192 mM glycine, 25 mM Tris, 15% methanol). The immunoblotting was performed as described previously (33) using rabbit polyclonal antibody to CodY prepared by Biodesign International (Kennebunkport, ME).

ß-Galactosidase assays. B. subtilis strains carrying fusions to the E. coli lacZ gene were assayed for ß-galactosidase activity after growth in DS medium as described previously (41).

In vitro cross-linking of CodY. Using an adaptation of the method of den Hengst et al. (6), we incubated wild-type or mutant CodY proteins (2.5 µM) for 15 min at room temperature in 1x cross-linking buffer (100 mM KCl, 15 mM Tris-HCl, pH 7.5) in a total volume of 12 µl. Formaldehyde was then added to the reaction mixture to a final concentration of 1% (vol/vol), followed by a 20-min incubation step at room temperature. Cross-linking with formaldehyde was stopped by the addition of 3 µl of 5x Laemmli loading buffer containing 2.5% ß-mercaptoethanol. The samples were heated at 37°C for 5 min and analyzed by SDS-polyacrylamide gel electrophoresis and Western blotting using rabbit antibody to CodY. As a control, parallel samples were heated at 96°C for 10 min to break the chemical cross-links (45).

Detection of codY mRNA by reverse transcription-PCR (RT-PCR). B. subtilis strain PJB23, expressing the mutant form of CodY, was grown in 20 ml of DS medium to an OD600 of approximately 0.9, and a 1-ml sample of culture was removed to extract total RNA using the QIAGEN RNeasy kit. Total RNA was subjected to vigorous DNase treatment using the TURBO DNA-free kit (Ambion) as described by the manufacturer. Reverse transcription was performed using SuperScript II (Invitrogen) with 1 µg of total RNA and 2 pmol of OPJ20 primer (5'-GGAGCTCAGGAGGAACTTTTGAAATGG-3') following the manufacturer's protocol. As a control, reactions were also carried out in the absence of SuperScript II to check for genomic contamination. Primers OPJ9 (5'-GATCATCGGAGGCGGGGAAAG-3') and OPJ20 were used to amplify a 427-bp segment of codY-coding sequence that would contain the relevant mutation. PCR was performed with Platinum Taq DNA polymerase (Invitrogen) and with 1 µl of cDNA as the template. In all cases, PCR products (500 ng) were subjected to DrdI restriction enzyme treatment and analyzed on a 2.5% agarose gel prepared in TBE.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Site-directed mutagenesis of potential CodY HTH residues. To test the functional significance of the putative HTH sequence of CodY and to determine whether specific residues in this motif are important in DNA binding, CodY HTH region residues that were expected to contact DNA or to form hydrophobic interactions between helix 1 and helix 2 were targeted (by comparison to well-characterized HTH-containing transcriptional factors such as phage {lambda} repressor and {lambda} Cro protein) for site-directed mutagenesis. Using overlap PCR primed by mutagenic oligonucleotides, each targeted residue (A207, R214, S215, and V218) of the putative CodY HTH region was separately substituted by either a similar or a contrasting (e.g., acidic versus basic, charged versus uncharged) amino acid (see Materials and Methods) (Fig. 1A). The targeted residues are 100% conserved in CodY homologs from 27 bacterial species (Fig. 1B). The mutated codY genes with a six-histidine C-terminal tag were cloned in an E. coli expression vector downstream of the araBAD promoter (14), enabling inducible expression of the various forms of CodY for purification (21).

Effects of CodY HTH mutations on DNA binding activity of CodY in vitro at the dpp and ilvB promoters. To allow analysis of DNA binding activity in vitro, wild-type and mutant versions of His6-tagged CodY were purified by cobalt affinity chromatography. Immunoblotting using antibodies raised to CodY protein verified that the introduction of the HTH mutations into the purified His6-tagged CodY proteins did not affect the stability of the corresponding proteins in E. coli (data not shown). The mutants were then tested for interaction with the dpp and ilvB target promoter regions by electrophoretic mobility shift assays (EMSAs). Since GTP and BCAAs are effectors of CodY that enhance its affinity to target DNAs (27, 39), the EMSAs were conducted in the presence of a mixture of GTP and BCAAs.

The wild-type CodY was found to bind to the ilvB promoter with high affinity (Fig. 2). The apparent Kd, based on the CodY concentration needed to shift 50% of the input DNA, was 4 to 8 nM, consistent with previous estimates obtained with an independent protein preparation (39). Multiple bands of different mobilities were observed with increased protein concentrations (Fig. 2a). Shivers and Sonenshein (39) showed, by footprinting experiments, that the appearance of these different shifted bands correlates with binding of CodY to multiple sites of various affinities in the ilvB promoter region.



View larger version (61K):
[in this window]
[in a new window]
  		FIG. 2. In vitro DNA binding activities of CodY proteins containing HTH mutations. The abilities of the purified wild-type and HTH mutant proteins to bind to the dpp and ilvB promoters were determined in EMSAs as described in Materials and Methods. The numbers above each lane indicate the CodY protein concentration expressed in nM. In all cases, the reaction mixtures contained the effectors GTP (2 mM) and isoleucine, valine, and leucine (10 mM each).

 
Wild-type CodY was found to bind to the dpp promoter with an affinity similar to that seen at the ilvB promoter (apparent Kd of ~2 to 4 nM). Interestingly, multiple shifted bands were also observed at increased protein concentrations (Fig. 2h). This result could be due to binding of CodY to several sites on the dpp promoter, as for the ilvB promoter, or to oligomerization of CodY at a single site on the DNA fragment or both. To distinguish between these possibilities, DNase I footprinting experiments were performed. As previously determined by Serror and Sonenshein (38), whose work predated the discovery that binding of CodY is enhanced by GTP and BCAAs, a 36-bp region on the template strand from positions –10 to +27 with respect to the transcription start site was protected by CodY in the presence of enhancing effectors (Fig. 3). In the presence of effectors, however, the concentration of CodY necessary to achieve 50% protection of this region was very low (<4 nM) or at least 18-fold lower than that seen in the absence of effectors (38). In addition, we found that at higher protein concentrations, the protected region was extended to position –63 relative to the transcription start site (Fig. 3). This additional region of protection represents binding to a lower-affinity site. Moreover, we found that, at the highest protein concentration used in the assay (256 nM), CodY protected two additional regions upstream of position –63 (Fig. 3). These two regions, representing the lowest-affinity binding sites, were fully protected at a CodY concentration of 1,024 nM (data not shown).



View larger version (119K):
[in this window]
[in a new window]
  		FIG. 3. DNase I footprint assay of wild-type CodY and CodY(A207V) with the dpp promoter region. A 256-bp DNA fragment (32P end labeled on the template strand) corresponding to positions –172 to +84 relative to the transcription start site was incubated with increasing concentrations (in nM) of purified wild-type CodY or CodY(A207V) in the presence of 2 mM GTP and 10 mM each of isoleucine, leucine, and valine. The footprinting assay was carried out as described in Materials and Methods. Protected regions observed only with wild-type CodY at 128 nM and 256 nM are marked by vertical lines. Sanger sequencing ladders are shown in the leftmost lanes (A, C, G, and T). The –35 and –10 boxes are indicated by vertical lines and the transcription start site by a bent arrow.

 
The V218A mutant form of CodY was able to bind to the ilvB and dpp promoters with an affinity similar to that of wild-type CodY (Fig. 2b and i), while the mutant forms with R214K, S215T, and S215A bound to the two promoters with slightly reduced affinities (four- to eightfold) compared to that of the wild-type CodY (Fig. 2c, d, e, j, k, and l). The A207V and S215F mutants showed a stronger defect in their ability to interact with the two promoters. The affinities of A207V and S215F mutants were reduced 32-fold for the dpp promoter (Fig. 2m and n), whereas the affinities were decreased 16-fold for the ilvB promoter (Fig. 2f and g). Interestingly, these two mutants also showed a defect in their capacity to form the highest-molecular-weight complexes, suggesting that the CodY A207V and S215F mutants were able to interact only with the high-affinity binding sites in the ilvB and dpp promoter regions, albeit with lower affinities than that of the wild-type protein.

To confirm this hypothesis, DNase I footprinting experiments were performed with the CodY protein with the A207V substitution [CodY(A207V)] and wild-type CodY protein on the 453-bp ilvB probe used in the gel shift assays. Analysis of the footprint showed the four protected regions for the wild-type protein, as previously seen (39) (Fig. 4). Regions I and II correspond to the high-affinity binding sites and regions III and IV to the low-affinity binding sites (Fig. 4). In addition, a fifth region was protected at higher CodY concentrations (Fig. 4). Regions I and II were also protected by CodY(A207V), but only at higher protein concentrations than those needed for the wild-type CodY (Fig. 4). In contrast, regions III to V were not protected by the mutant CodY even at the highest protein concentrations (Fig. 4). Similarly, only the high-affinity binding site in the dpp promoter region was found to be protected by CodY(A207V), and again only at higher protein concentrations than those needed for the wild-type protein to bind (Fig. 3). Therefore, the low-molecular-weight complexes seen for CodY(A207V) in gel shift assays (Fig. 2f and m) reflect binding to the high-affinity sites in the dpp and ilvB promoter regions.



View larger version (75K):
[in this window]
[in a new window]
  		FIG. 4. DNase I footprint assay of wild-type CodY and CodY(A207V) with the ilvB promoter region. A 453-bp PCR product, corresponding to positions –306 to +147 relative to the transcription start site and labeled at the 5' end of the nontemplate (coding) strand, was incubated with increasing concentrations (in nM) of purified wild-type CodY (WT-CodY) or CodY(A207V) in the presence of 2 mM GTP and 10 mM each of isoleucine, leucine, and valine. The footprinting assay was carried out as described in Material and Methods. Protected regions are marked by the vertical bars, as are the –35 and –10 boxes. The transcription start site is indicated by a bent arrow.

 
The CodY proteins carrying the nonconservative substitution A207D, R214E, or V218D showed the strongest defects in DNA binding. No binding to either promoter was observed even at very high protein concentrations (apparent Kd of >500 nM) (Fig. 2g and n). Overall, the orders of affinities of the mutant CodY proteins were very similar for the ilvB and dpp promoter regions. The mutants that retained DNA binding activity were able to respond to the presence of GTP and BCAAs, since their affinities for DNA were increased in the presence of either GTP or BCAAs (Table 4), but the extent of stimulation by effectors was not always as high as for the wild-type CodY protein (see Discussion).


View this table:
[in this window]
[in a new window]
  		TABLE 4. In vitro binding activities of wild-type and mutant versions of CodY and relative expression of the dpp-lacZ and ilvB-lacZ fusions

 
Introduction of the HTH mutations at the codY locus. To study the effects of the CodY HTH mutations in B. subtilis cells, each mutant codY gene was ligated to an integrative vector carrying a B. subtilis-compatible chloramphenicol resistance gene (28), and introduced by single-crossover homologous recombination into the chromosome of a B. subtilis strain carrying either a dpp-lacZ or an ilvB-lacZ fusion at the amyE locus. The resulting transformants were partial diploids, carrying both a mutant codY gene under the control of the cod operon promoter and a wild-type, promoterless copy of codY (Fig. 5A). Since all the HTH mutations were located within 100 bp of the 3' end of the codY gene, we were successful in introducing each mutation into the expressed copy of the codY gene on the B. subtilis chromosome at a high frequency. The presence and location of the expected mutations were verified by amplifying the expressed codY gene from the chromosome with primers OPJ1 and OMRL5 (Materials and Methods and Fig. 5A) and sequencing the PCR product. Only strains that had the expected mutation and no other mutations were used for ß-galactosidase assays.



View larger version (52K):
[in this window]
[in a new window]
  		FIG. 5. Transcription of codY in a merodiploid strain. (A) Introduction of an HTH mutation at the codY locus was performed as described in the text and in Materials and Methods. The resulting recombinant strains (e.g., PJB23) carried both a mutant codY gene (the mutation is indicated by an asterisk) under the control of the normal cod operon promoter (located upstream of codV and indicated by a broken arrow) and a wild-type, promoterless codY copy downstream of the integrated pJPM1 plasmid. The presence and location of the expected mutation were verified by amplifying the expressed codY gene from the chromosome with primers OPJ1 and OMRL5. codY mRNA of mutant strain PJB23 was reverse transcribed using primer OPJ20. The codY cDNA was then amplified using primers OPJ9 and OPJ20. The resulting RT-PCR product was subjected to digestion with DrdI restriction enzyme and loaded on a 2.5% agarose gel (lane 4). The corresponding PCR products obtained from amplification of the wild-type codY gene cloned in pBAD30 (lane 1) or the codY(A207D) allele cloned in pBAD30 (lane 2), from amplification of wild-type codY cDNA from strain PJB14 (lane 3), or from amplification of codY from genomic DNA of the PJB23 mutant strain (lane 5) were also subjected to DrdI digestion as controls. (B) Expression of CodY HTH mutant proteins in B. subtilis. Whole-cell lysates of strain PJB16 carrying a codY null mutation (lane 1), wild-type strain PJB14 (lane 2) or codY HTH mutant strains PJB22 (lane 3), PJB23 (lane 4), PJB24 (lane 5), PJB25 (lane 6), PJB26 (lane 7), PJB27 (lane 8), PJB28 (lane 9), and PJB29 (lane 10) were assayed by immunoblotting using anti-CodY antibodies as described in Materials and Methods. The CodY-specific band is indicated by an arrow on the right.

 
Even though the wild-type codY copy was promoterless, we still wanted to verify that only the mutant copy was expressed. To do so, we took advantage of the creation of new restriction enzyme recognition sites caused by some HTH mutations. Total RNA was extracted from cultures of recombinant strains and the codY mRNA was reverse transcribed using the primer OPJ20, which annealed to the 3' end of the codY-coding sequence (Materials and Methods and Fig. 5A). The codY cDNA was then amplified by PCR using primers OPJ9 and OPJ20 (Fig. 5A), and the resulting RT-PCR product was then incubated with the appropriate restriction enzyme. An example of the results obtained with strain PJB23 carrying the codY(A207D) mutation is shown in Fig. 5A. In this case, if the wild-type codY copy were expressed as well as the mutant codY was, we would find three bands after digestion with DrdI, an enzyme that can cut only the mutant codY allele. One band should correspond to the undigested wild-type copy (427 bp) and two others to the digested mutant copy (273 bp and 154 bp). As shown in Fig. 5A, lane 4, the RT-PCR product subjected to digestion generated only two bands (273 bp and 154 bp) identical in size to those generated by a control PCR obtained from the amplification of the cloned codY(A207D) mutant allele (Fig. 5A, lane 2). As expected, neither the wild-type PCR product obtained from amplification of codY cloned in pBAD30 (Fig. 5A, lane 1) nor the wild-type RT-PCR product obtained by amplification of codY cDNA from the PJB14 strain (Fig. 5A, lane 3) was cut by the enzyme. These results indicate that the wild-type codY copy was not expressed at a detectable level in strain PJB23. The results obtained for several other mutants were all consistent with those shown in Fig. 5A (data not shown).

To rule out the possibility that potential effects on gene regulation were due to instability of mutant CodY proteins, each B. subtilis strain was tested by immunoblotting using antibodies raised against CodY protein. The results shown in Fig. 5B indicate that all CodY mutants were produced as stable proteins in B. subtilis. Surprisingly, the protein having the A207D mutation migrated slightly more slowly than did either the wild-type protein or the other mutant proteins (Fig. 5B, lane 4). Single mutations that alter mobility in SDS-polyacrylamide gels have been described for other proteins (23, 24, 29).

Effects of CodY HTH mutations on dpp expression in B. subtilis. In wild-type cells, the dpp operon is induced in early stationary phase, coincident with relief of repression by two regulatory proteins, CodY and AbrB (41, 42). To analyze CodY activity in vivo in the absence of overlapping repression by AbrB, each codY HTH mutation was introduced into a B. subtilis strain (PS56) carrying a dpp-lacZ fusion at the amyE locus and an abrB mutation. B. subtilis strains carrying a codY HTH mutation or a codY deletion mutation (PS83) or the wild-type codY gene (PS56) were then assayed for ß-galactosidase activity at various times during growth in DS medium (Fig. 6). The codY HTH mutant strains fell into three classes based on their effects on dpp expression compared to the parent PS56 strain.



View larger version (28K):
[in this window]
[in a new window]
  		FIG. 6. Effects of CodY HTH mutations on transcription from the dpp promoter. B. subtilis strains carrying a dpp-lacZ fusion were grown in DS liquid medium and tested for ß-galactosidase specific activity at the indicated time point. The x axis shows the sampling time relative to the end of exponential growth phase (time zero). The experiment was performed in triplicate, and any variation in data points was less than 15% of the represented values. (A-D) PS56 (open squares, wild-type codY) and PS83 (filled squares, {Delta}codY). (A) MRLB36 [filled circles, codY(R214E)] and MRLB37 [open circles, codY(R214K)]. (B) PJB41 [filled circles, codY(S215F)], PJB40 [open circles, codY(S215T)], and PJB34 [open triangles, codY(S215A)]. (C) MRLB35 [filled circles, codY(V218D)] and PJB3 [open circles, codY(V218A)]. (D) PJB31 [filled circles, codY(A207D)] and PJB4 [open circles, codY(A207V)].

 
The first class comprises the strains containing the codY(V218A) and codY(A207V) mutations. These strains displayed a wild-type phenotype as shown in Fig. 6C and D. Thus, the A207V and V218A mutations had no effect on the ability of CodY to repress the dpp promoter in vivo. This result is in good agreement with the observation that the V218A mutation had little if any effect on the binding of CodY in vitro (Fig. 2i and Table 4). However, the result was surprising for the A207V mutation, since the CodY(A207V) mutant displayed a significant defect in binding to the dpp promoter compared to the wild-type protein (Fig. 2m and Table 4) (see Discussion).

The second class comprises the strains harboring the codY(R214K), codY(S215T), and codY(S215A) mutations. These mutations caused a moderate derepression of the dpp promoter early during the exponential phase, increasing ß-galactosidase activity to a level about 2.5- to 12-fold higher than in the wild-type strain, PS56 (Fig. 6A and B and Table 4). These strains showed an induction of dpp expression similar to (Fig. 6A) or higher than (Fig. 6B) that of the wild-type strain when they approached stationary phase. These results indicate that the R214K, S215T, and S215A substitutions prevent full repression of the dpp promoter during exponential phase and that this partial repression is relieved when stationary phase is reached. Interestingly, the substitution of Ser-215 by alanine had a greater effect on the ability of CodY to repress the dpp promoter than did the substitution by threonine (Fig. 6B) (see Discussion). The slightly decreased affinities of these CodY mutants for DNA, as measured by gel shift assays (Fig. 2j, k, and l), correlated with partial derepression of the dpp-lacZ fusion in vivo (Table 4). Therefore, the inability of the mutants CodY(R214K), CodY(S215T), and CodY(S215A) to repress the dpp promoter to the same extent as the wild-type protein is likely due to their reduced ability to bind efficiently to DNA in vivo.

The third class comprises the strains carrying the four nonconservative substitutions codY(A207D), codY(R214E), codY(S215F), or codY(V218D). The effects of these mutations on dpp expression were similar to the effect of a codY null mutation (Fig. 6A, B, C, and D). In these strains, the dpp promoter was expressed during exponential growth at levels 33- to 48-fold higher than in strain PS56 (Table 4). As shown above, these mutations abolished binding or dramatically reduced the affinity of CodY for the dpp promoter in vitro (Fig. 2n).

Effects of CodY HTH mutations on ilvB expression in B. subtilis. Among the many targets of CodY are the ilv genes that encode the enzymes of BCAA biosynthesis (27, 39). To test the effects of CodY mutations in vivo on the regulation of these genes, each mutation was integrated into the chromosome of a B. subtilis strain carrying an ilvB-lacZ fusion at the amyE locus. The ilvB{Delta}T-lacZ fusion used in our experiment was missing the T-box region that causes leucine-responsive transcription termination (11). B. subtilis strains carrying a codY HTH mutation, a codY deletion mutation (PJB16), or a wild-type codY gene (PJB14) were then assayed for ß-galactosidase activity at various times during growth in DS medium. The codY HTH mutant strains fell into two classes based on their effects on ilvB expression.

The first class included strains carrying the substitutions R214K, S215T, S215A, and V218A. In these strains, the ilvB{Delta}T-lacZ fusion was partially derepressed during exponential growth phase. The codY(R214K) and codY(S215T) mutations allowed about five- to sevenfold derepression. Partial derepression of the ilvB{Delta}T-lacZ fusion in strains carrying these mutations (R214K and S215T) fits well with the partial defect in DNA binding seen for the mutant proteins in vitro (Fig. 2c and d and Table 4). Interestingly, as for the dpp promoter, the codY(S215A) mutation had a much greater effect on the activity of CodY than did the codY(S215T) mutation, causing a 20-fold derepression of the ilvB{Delta}T-lacZ fusion (see Discussion). The codY(V218A) mutation caused an eightfold derepression of ilvB{Delta}T-lacZ expression, a surprising result, since CodY(V218A) bound to ilvB DNA in vitro with approximately the affinity of the wild-type protein (Fig. 2b and Table 4) (see Discussion). The strains also showed derepression of the ilvB promoter when the cells approached stationary phase and showed, in stationary phase, a slightly higher level of ß-galactosidase than in the wild-type strain (Fig. 7).



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 7. Effects of CodY HTH mutations on transcription from the ilvB promoter. B. subtilis strains carrying an ilvB-lacZ fusion were assayed for ß-galactosidase specific activity as described in the legend of Fig. 6. (A-D) PJB14 (open squares, wild-type codY) and PJB16 (filled squares, {Delta}codY). (A) PJB24 [filled circles, codY(R214E)] and PJB25 [open circles, codY(R214K)]. (B) PJB26 [filled circles, codY(S215F)], PJB27 [open circles, codY(S215T)], and PJB33 [open triangles, codY(S215A)]. (C) PJB28 [filled circles, codY(V218D)] and PJB29 [open circles, codY(V218A)]. (D) PJB23 [filled circles, codY(A207D)] and PJB22 [open circles, codY(A207V)].

 
By comparison, strains carrying mutations codY(R214E) (Fig. 7A), codY(S215F) (Fig. 7B), codY(V218D) (Fig. 7C), codY(A207D) and codY(A207V) (Fig. 7D) had ß-galactosidase activities similar to that of strain PJB16 carrying a codY null mutation. That is, introduction of the codY HTH point mutations led to approximately 30- to 48-fold derepression of the ilvB promoter. We noted again a good correlation between the severely reduced DNA binding affinities of these mutants in vitro and the total derepression of the ilvB promoter in vivo (Table 4).

Oligomerization of wild-type and HTH-mutated CodY proteins detected by in vitro cross-linking. Since the functional forms of many transcriptional regulators are dimers or oligomers, we sought to determine whether the HTH mutations affected potential CodY oligomerization. Native gel electrophoresis and light-scattering experiments indicated that purified CodY is primarily in the form of a dimer of approximately 60 kDa (2; K. Matsuno and A. L. Sonenshein, unpublished results). To study CodY oligomerization further, we performed in vitro cross-linking assays with wild-type and mutant CodY proteins.

First, the purified mutant and wild-type proteins were subjected to electrophoresis under nondenaturing conditions (Fig. 8A). Each protein sample migrated as a single species whose mobility varied slightly from sample to sample. Next, the purified proteins were subjected to formaldehyde cross-linking and SDS-polyacrylamide gel electrophoresis. An immunoblot showed that the monomer form of CodY predominated after denaturation; a small fraction of the protein was cross-linked with formaldehyde to generate forms with the mobilities expected for dimers, trimers, or larger oligomers (Fig. 8B). Although the efficiency of cross-linking was not high, the fractions of wild-type and mutant CodY proteins in the oligomeric forms were similar for the various proteins (Fig. 8B). Under the same experimental conditions, the molecular mass of ovalbumin, a monomeric protein, remained unchanged, indicating the specificity of cross-linking (data not shown). In addition, heating the samples to break the chemical cross-link reduced the oligomers to the monomer form in all cases (data not shown). Since the concentration of CodY in our cross-linking reaction mixtures (2.5 µM) was similar to that found inside the cell (2 to 4 µM [R. Shivers, personal communication]), formation of higher-order CodY complexes is probably favored under physiological conditions. We conclude that the HTH mutations tested do not affect significantly the global structure of the protein and that they have no discernible effect on the ability of CodY to oligomerize in solution.



View larger version (81K):
[in this window]
[in a new window]
  		FIG. 8. (A) Coomassie blue staining of purified wild-type and mutant CodY proteins subjected to nondenaturing polyacrylamide (10%) gel electrophoresis. (B) Cross-linking of purified wild-type and mutant CodY proteins. Formaldehyde was used to cross-link potential oligomers of CodY. After electrophoresis on SDS-polyacrylamide gels, the proteins were visualized by immunoblot analysis using a rabbit polyclonal serum antibody to CodY. The positions of monomers and putative dimers and trimers of CodY are indicated to the left. Lane M contains prestained molecular mass markers (in kDa; GIBCO-BRL). The positions of molecular mass markers (in kDa) are indicated to the right.

 

   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Prokaryotic and eukaryotic transcriptional regulators have distinct DNA-binding domains that contain helix-loop-helix, zinc finger, leucine zipper, ß-ribbon, or HTH motifs (32). Of these discrete motifs, the HTH motif is the best characterized to date. Although not homologous to any previously known family of HTH-containing protein, CodY has a predicted HTH-like motif (3, 7, 38) within which residues critical for repression correspond to residues of other HTH proteins implicated in interacting with DNA or stabilizing the helices.

Positions 12 and 13 of the HTH are commonly involved in sequence-specific interaction with DNA (32). Although we cannot conclude that residues at these positions in CodY make direct contact with DNA, replacement of arginine-214 (position 12) by glutamate (R214E) and serine-215 (position 13) by phenylalanine (S215F) caused a 30- to 48-fold derepression of expression from the dpp and ilvB promoters. When arginine-214 and serine-215 were replaced with structurally similar residues, lysine (R214K) and threonine (S215T), respectively, repression of the dpp and ilvB promoters was only partially affected in vivo, and the binding activity in vitro was slightly reduced compared to that of the wild-type protein. A study of the HTH region of the E. coli TyrR protein has shown that an alanine substitution at position 12 (histidine-494) results in a substantial derepression of the tyrP and aroF promoters during exponential phase (17). Histidine-494 in TyrR interacts with a cytosine and alteration of the base pair or the histidine-494 residue completely abolishes DNA binding activity (17, 18). Substitution by alanine at position 13 (threonine-495) in TyrR has a significant effect on repression, but a serine substitution does not (17, 47). Interestingly, alanine substitution of the corresponding residue in CodY protein, serine-215, has a greater effect on repressor activity at both the dpp and ilvB promoters than does the substitution by threonine. The hydroxyl group of serine-215 may therefore be involved in interaction with DNA, as has been proposed for the corresponding residue, threonine-495, in TyrR protein (17). In the Haemophilus influenzae TyrR nuclear magnetic resonance structure, H494 and T495 are part of a hydrophilic cluster that is accessible to solvent and therefore may be involved in DNA interaction as in E. coli TyrR protein (46). In the phage {lambda} and 434 repressors, residues at positions 12 (glutamine in both cases) and 13 (serine and glutamine, respectively) are also known to contact DNA (30, 32). Specifically, the conserved glutamine at position 12 contacts an adenine base in the operator site (30). In the {lambda} repressor, the serine at position 13 makes contact with a guanine, while in the 434 repressor, the glutamine at this position associates with thymine and guanine bases (30).

Position 16 in the phage repressor HTH domains, corresponding to valine-218 in CodY, also contacts DNA (32). Substitution of CodY residue Val218 by a negatively charged residue (V218D) caused total derepression of both the dpp and ilvB promoters and abolished binding in vitro. The substitution of residue Val218 by alanine (V218A), however, did not affect the ability of CodY to repress the dpp promoter in vivo or to bind to this target in vitro at all. The same V218A mutation also had no significant effect on the ability of CodY to bind to the ilvB promoter in vitro but caused a partial derepression of an ilvB{Delta}T-lacZ fusion in vivo. Since the affinity of CodY(V218A) for DNA is increased in the presence of effectors to the same extent as for wild-type CodY (Table 4), it is unlikely that the partial derepression of the ilvB{Delta}T-lacZ fusion is due to a defect in response to the effectors. The lack of effect of the alanine substitution on binding in vitro may indicate that Val218 does not contact DNA directly. However, the methyl groups of valine and isoleucine have been shown to interact with the methyl group of thymine in homeodomain proteins (32) and that type of interaction may be maintained, in the CodY case, by the methyl group of alanine. To explain the in vivo defect in repression of the ilvB promoter, we can imagine two possibilities. First, the V218A mutation may affect interaction of CodY with another protein. Alternatively, the mutation could alter the conformation of the HTH motif in such a way as to interfere with the ability of other residues to interact with target promoters.

In typical HTH motifs, hydrophobic interactions between the highly conserved alanine residue at position 5 and isoleucine or valine at position 15 are thought to stabilize the positioning of helices 1 and 2 (3, 31). In the E. coli Trp repressor, residue 5 is a lysine, and the HTH elbow is perturbed by its presence (36). Alanine-207 occupies position 5 in the putative helix 1 of CodY. The substitution of Ala207 by a charged residue, aspartate, rendered the protein unable to bind DNA even at very high protein concentrations. Thus, the potential Ala207-mediated structural interactions may indeed be required for target DNA recognition and binding. Substitution by another nonpolar residue, valine, decreased by 16- to 64-fold the affinities of CodY for the dpp and ilvB promoters. Interestingly, footprinting experiments showed that, even at high protein concentrations, the mutant CodY(A207V) was able to bind only to the high-affinity binding sites in the dpp and ilvB promoter regions. Although CodY(A207V) is able to oligomerize in solution to the same extent as wild-type CodY (Fig. 8), we cannot exclude the possibility that higher-order oligomerization of CodY upon binding to DNA, which might be required for binding to the low-affinity sites, is defective in this mutant. The expression of a dpp-lacZ fusion in a strain carrying the codY(A207V) mutation was similar to that in the wild-type strain, whereas the expression of an ilvB{Delta}T-lacZ fusion in a strain carrying the same mutation was similar to that in a strain carrying a codY null mutation. These diametrically opposed results are surprising, since the CodY protein carrying the A207V mutation had similarly decreased affinities for the two target promoters in vitro. The high-affinity binding site in the dpp promoter region encompasses the transcription start site as well as the –10 region. At the dpp promoter, even low-affinity versions of the proteins may be able to bind tightly enough in vivo to repress the promoter. By contrast, binding to the high-affinity sites in the ilvB promoter region may not be sufficient to repress the promoter. Also, we cannot exclude the possibility that another regulatory protein competes with CodY for the same binding site (R. P. Shivers and A. L. Sonenshein, Mol. Microbiol., in press). This regulatory protein could be a positive regulator that has a higher affinity for the ilvB promoter region than does CodY(A207V) mutant protein, destabilizing the interaction of CodY with DNA. It is interesting that the HTH mutations generally had a much greater effect on the repression of the ilvB{Delta}T-lacZ fusion than the dpp-lacZ fusion (Table 4).

Although the mutant forms of CodY that retained residual DNA binding activity all responded to the presence of GTP and BCAAs in vitro, the extent of the response was, in some cases, less than that for wild-type CodY. They also presumably responded to effectors in vivo, since their repression activity was reduced as cells made the transition from exponential phase to stationary phase. Whereas all the mutant proteins appear to have the same global structure in solution as that of wild-type CodY, they may be less susceptible to hypothetical conformational changes induced by effectors. The fact that G3, one of the motifs involved in GTP binding, is within the HTH domain (33) may help to explain why certain HTH mutations render CodY less responsive to effectors. The reduced response to effectors does not fully explain the phenotypes of these mutants, however. They also have a clear defect in intrinsic DNA binding, since their affinity for DNA in the absence of effectors is lower than that of the wild type.

In summary, our results show that, although CodY is not a member of any previously described family of transcriptional regulators, its DNA-binding domain is similar to those carried by proteins belonging to the HTH family of DNA-binding proteins. Interestingly, the HTH-like regions of all identified CodY homologs are extremely well-conserved (Fig. 1). However, since no CodY-binding consensus sequence has been discernible, it is conceivable that the CodY HTH motif is not the only part of the protein required for specific binding. Another domain of the protein may interact with a conserved DNA structural motif (e.g., bent DNA) in target promoters, whereupon the HTH domain may then be utilized to form specific interactions with appropriately located residues exposed on the DNA. If this is the case, CodY-regulated promoters may share a common structure but have variations in their primary nucleotide sequences. X-ray crystallographic studies currently in progress (2) should be very helpful in understanding this interaction in greater detail.


   ACKNOWLEDGMENTS
 
We are grateful to B. Belitsky, P. Serror, S. Dineen, A. Villapakkam, and R. Shivers for helpful discussions and detailed comments on the manuscript.

This work was supported by a research grant (GM042219) to A.L.S. from the U.S. Public Health Service.


   FOOTNOTES
 
* Corresponding author. Mailing address: Department of Molecular Biology and Microbiology, Tufts University School of Medicine, 136 Harrison Avenue, Boston, MA 02111. Phone: (617) 636-6761. Fax: (617) 636-0337. E-mail: linc.sonenshein{at}tufts.edu. Back

{dagger} Present address: Corporate-Sponsored Research and Licensing, Massachusetts General Hospital, Charlestown, MA 02129. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Bergara, F., C. Ibarra, J. Iwamasa, J. C. Patarroyo, R. Aguilera, and L. M. Marquez-Magana. 2003. CodY is a nutritional repressor of flagellar gene expression in Bacillus subtilis. J. Bacteriol. 185:3118-3126.[Abstract/Free Full Text]
  2. Blagova, E. V., V. M. Levdikov, K. Tachikawa, A. L. Sonenshein, and A. J. Wilkinson. 2003. Crystallization of the GTP-dependent transcriptional regulator CodY from Bacillus subtilis. Acta Crystallogr. Sect. D 59:155-157.[CrossRef][Medline]
  3. Brennan, R. G., and B. W. Matthews. 1989. The helix-turn-helix DNA binding motif. J. Biol. Chem. 264:1903-1906.[Free Full Text]
  4. Chambellon, E., and M. Yvon. 2003. CodY-regulated aminotransferases AraT and BcaT play a major role in the growth of Lactococcus lactis in milk by regulating the intracellular pool of amino acids. Appl. Environ. Microbiol. 69:3061-3068.[Abstract/Free Full Text]
  5. Debarbouille, M., R. Gardan, M. Arnaud, and G. Rapoport. 1999. Role of BkdR, a transcriptional activator of the SigL-dependent isoleucine and valine degradation pathway in Bacillus subtilis. J. Bacteriol. 181:2059-2066.[Abstract/Free Full Text]
  6. den Hengst, C. D., P. Curley, R. Larsen, G. Buist, A. Nauta, D. van Sinderen, O. P. Kuipers, and J. Kok. 2005. Probing direct interactions between CodY and the oppD promoter of Lactococcus lactis. J. Bacteriol. 187:512-521.[Abstract/Free Full Text]
  7. Dodd, I. B., and J. B. Egan. 1990. Improved detection of helix-turn-helix DNA-binding motifs in protein sequences. Nucleic Acids Res. 18:5019-5026.[Abstract/Free Full Text]
  8. Dower, W. J., J. F. Miller, and C. W. Ragsdale. 1988. High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Res. 16:6127-6145.[Abstract/Free Full Text]
  9. Fisher, S. H., K. Rohrer, and A. E. Ferson. 1996. Role of CodY in regulation of the Bacillus subtilis hut operon. J. Bacteriol. 178:3779-3784.[Abstract/Free Full Text]
  10. Fouet, A., and A. L. Sonenshein. 1990. A target for carbon source-dependent negative regulation of the citB promoter of Bacillus subtilis. J. Bacteriol. 172:835-844.[Abstract/Free Full Text]
  11. Grandoni, J. A., S. A. Zahler, and J. M. Calvo. 1992. Transcriptional regulation of the ilv-leu operon of Bacillus subtilis. J. Bacteriol. 174:3212-3219.[Abstract/Free Full Text]
  12. Guedon, E., P. Renault, S. D. Ehrlich, and C. Delorme. 2001. Transcriptional pattern of genes coding for the proteolytic system of Lactococcus lactis and evidence for coordinated regulation of key enzymes by peptide supply. J. Bacteriol. 183:3614-3622.[Abstract/Free Full Text]
  13. Guedon, E., P. Serror, S. D. Ehrlich, P. Renault, and C. Delorme. 2001. Pleiotropic transcriptional repressor CodY senses the intracellular pool of branched-chain amino acids in Lactococcus lactis. Mol. Microbiol. 40:1227-1239.[CrossRef][Medline]
  14. Guzman, L. M., D. Belin, M. J. Carson, and J. Beckwith. 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J. Bacteriol. 177:4121-4130.[Abstract/Free Full Text]
  15. Harrisson, S. C., and A. Aggarwal. 1990. DNA recognition by proteins with the helix-turn-helix motif. Annu. Rev. Biochem. 59:933-969.[CrossRef][Medline]
  16. Houman, F., M. R. Diaz-Torres, and A. Wright. 1990. Transcriptional antitermination in the bgl operon of E. coli is modulated by a specific RNA binding protein. Cell 62:1153-1163.[CrossRef][Medline]
  17. Hwang, J. S., J. Yang, and A. J. Pittard. 1997. Critical base pairs and amino acid residues for protein-DNA interaction between the TyrR protein and tyrP operator of Escherichia coli. J. Bacteriol. 179:1051-1058.[Abstract/Free Full Text]
  18. Hwang, J. S., J. Yang, and A. J. Pittard. 1999. Specific contacts between residues in the DNA-binding domain of the TyrR protein and bases in the operator of the tyrP gene of Escherichia coli. J. Bacteriol. 181:2338-2345.[Abstract/Free Full Text]
  19. Inaoka, T., and K. Ochi. 2002. RelA protein is involved in induction of genetic competence in certain Bacillus subtilis strains by moderating the level of intracellular GTP. J. Bacteriol. 184:3923-3930.[Abstract/Free Full Text]
  20. Inaoka, T., K. Takahashi, M. Ohnishi-Kameyama, M. Yoshida, and K. Ochi. 2003. Guanine nucleotides guanosine 5'-diphosphate 3'-diphosphate and GTP co-operatively regulate the production of an antibiotic bacilysin in Bacillus subtilis. J. Biol. Chem. 278:2169-2176.[Abstract/Free Full Text]
  21. Kim, H. J., S. I. Kim, M. Ratnayake-Lecamwasam, K. Tachikawa, A. L. Sonenshein, and M. Strauch. 2003. Complex regulation of the Bacillus subtilis aconitase gene. J. Bacteriol. 185:1672-1680.[Abstract/Free Full Text]
  22. Kim, H. J., A. Roux, and A. L. Sonenshein. 2002. Direct and indirect roles of CcpA in regulation of Bacillus subtilis Krebs cycle genes. Mol. Microbiol. 45:179-190.[CrossRef][Medline]
  23. McIver, K. S., and R. L. Myles. 2002. Two DNA-binding domains of Mga are required for virulence gene activation in the group A streptococcus. Mol. Microbiol. 43:1591-1601.[CrossRef][Medline]
  24. McMurry, L. M., M. Stephan, and S. B. Levy. 1992. Decreased function of the class B tetracycline efflux protein Tet with mutations at aspartate 15, a putative intramembrane residue. J. Bacteriol. 174:6294-6297.[Abstract/Free Full Text]
  25. Mikaelian, I., and A. Sergeant. 1992. A general and fast method to generate multiple site directed mutations. Nucleic Acids Res. 20:376.[Free Full Text]
  26. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
  27. Mirel, D. B., W. F. Estacio, M. Mathieu, E. Olmsted, J. Ramirez, and L. M. Marquez-Magana. 2000. Environmental regulation of Bacillus subtilis {sigma}D-dependent gene expression. J. Bacteriol. 182:3055-3062.[Abstract/Free Full Text]
  28. Molle, V., Y. Nakaura, R. P. Shivers, H. Yamaguchi, R. Losick, Y. Fujita, and A. L. Sonenshein. 2003. Additional targets of the Bacillus subtilis global regulator CodY identified by chromatin immunoprecipitation and genome-wide transcript analysis. J. Bacteriol. 185:1911-1922.[Abstract/Free Full Text]
  29. Mueller, J. P., G. Bukusoglu, and A. L. Sonenshein. 1992. Transcriptional regulation of Bacillus subtilis glucose starvation-inducible genes: control of gsiA by the ComP-ComA signal transduction system. J. Bacteriol. 174:4361-4373.[Abstract/Free Full Text]
  30. Noel, D., K. Nikaido, and G. F. Ames. 1979. A single amino acid substitution in a histidine-transport protein drastically alters its mobility in sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Biochemistry 18:4159-4165.[CrossRef][Medline]
  31. Pabo, C. O., A. K. Aggarwal, S. R. Jordan, L. J. Beamer, U. R. Obeysekare, and S. C. Harrison. 1990. Conserved residues make similar contacts in two repressor-operator complexes. Science 247:1210-1213.[Abstract/Free Full Text]
  32. Pabo, C. O., and R. T. Sauer. 1984. Protein-DNA recognition. Annu. Rev. Biochem. 53:293-321.[CrossRef][Medline]
  33. Pabo, C. O., and R. T. Sauer. 1992. Transcription factors: structural families and principles of DNA recognition. Annu. Rev. Biochem. 61:1053-1095.[CrossRef][Medline]
  34. Ratnayake-Lecamwasam, M., P. Serror, K. W. Wong, and A. L. Sonenshein. 2001. Bacillus subtilis CodY represses early-stationary-phase genes by sensing GTP levels. Genes Dev. 15:1093-1103.[Abstract/Free Full Text]
  35. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
  36. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467.[Abstract/Free Full Text]
  37. Schevitz, R. W., Z. Otwinowski, A. Joachimiak, C. L. Lawson, and P. B. Sigler. 1985. The three-dimensional structure of trp repressor. Nature 317:782-786.[CrossRef][Medline]
  38. Serror, P., and A. L. Sonenshein. 1996. CodY is required for nutritional repression of Bacillus subtilis genetic competence. J. Bacteriol. 178:5910-5915.[Abstract/Free Full Text]
  39. Serror, P., and A. L. Sonenshein. 1996. Interaction of CodY, a novel Bacillus subtilis DNA-binding protein, with the dpp promoter region. Mol. Microbiol. 20:843-852.[Medline]
  40. Shivers, R. P., and A. L. Sonenshein. 2004. Activation of the Bacillus subtilis global regulator CodY by direct interaction with branched-chain amino acids. Mol. Microbiol. 53:599-611.[CrossRef][Medline]
  41. Slack, F. J., J. P. Mueller, and A. L. Sonenshein. 1993. Mutations that relieve nutritional repression of the Bacillus subtilis dipeptide permease operon. J. Bacteriol. 175:4605-4614.[Abstract/Free Full Text]
  42. Slack, F. J., J. P. Mueller, M. A. Strauch, C. Mathiopoulos, and A. L. Sonenshein. 1991. Transcriptional regulation of a Bacillus subtilis dipeptide transport operon. Mol. Microbiol. 5:1915-1925.[CrossRef][Medline]
  43. Slack, F. J., P. Serror, E. Joyce, and A. L. Sonenshein. 1995. A gene required for nutritional repression of the Bacillus subtilis dipeptide permease operon. Mol. Microbiol. 15:689-702.[Medline]
  44. Strauch, M. A., and M. Ayazifar. 1995. Bent DNA is found in some, but not all, regions recognized by the Bacillus subtilis AbrB protein. Mol. Gen. Genet. 246:756-760.[CrossRef][Medline]
  45. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680.[Abstract/Free Full Text]
  46. Walburger, A., C. Lazdunski, and Y. Corda. 2002. The Tol/Pal system function requires an interaction between the C-terminal domain of TolA and the N-terminal domain of TolB. Mol. Microbiol. 44:695-708.[CrossRef][Medline]