Regulation of the expression of whiB1 in Mycobacterium tuberculosis: role of cAMP receptor protein

Mycobacterium tuberculosis, the aetiological agent of tuberculosis, accounts for nearly 2 million human deaths every year. Approximately one-third of the world's population is latently infected with M. tuberculosis, and 7 to 8 million new TB cases occur annually (Dye et al., 1999). M. tuberculosis can survive in diverse surroundings ranging from the human host to droplet nuclei in the atmosphere. Adaptation to such diverse conditions requires controlled regulation of the expression of key genes that allow the bacillus to alter its physiology in response to changes in environmental stimuli. Sequencing of the M. tuberculosis genome has revealed the presence of more than 100 regulatory proteins, 13 sigma factors and 11 two-component systems (Cole et al., 1998), which provide the bacterium with a high degree of adaptability.

The Wbl (WhiB-like) family of proteins is present throughout the actinomycetes but absent from all other organisms evaluated so far (Molle et al., 2000; Soliveri et al., 2000). Due to the presence of a conserved helixturnhelix motif, these proteins are believed to function as DNA-binding transcription regulators. The first of these proteins, WhiB, was identified in Streptomyces coelicolor, a Gram-positive sporulating bacterium closely related to M. tuberculosis. S. coelicolor whiB mutants produce abnormally long, tightly coiled aerial hyphae that are completely blocked in their ability to form sporulation septa (Chater, 1972; Davis & Chater, 1992; Flardh et al., 1999). Studies of whiB orthologues in mycobacteria have shown that the M. smegmatis whiB2 gene (also called whmD) is essential (Gomez & Bishai, 2000); M. tuberculosis whiB3 plays a role in virulence and its gene product may interact with a sigma factor of RNA polymerase (Steyn et al., 2002); whiB7 of M. tuberculosis is involved in multi-drug resistance (Morris et al., 2005). Each of the Wbl family of proteins contains four invariant cysteine residues, which are believed to be involved in binding a [4Fe4S] cluster (Jakimowicz et al., 2005). The functional importance of this cluster is emphasized by the observation that none of the four whiD alleles carrying mutations at these cysteine residues was able to complement the whiD mutant phenotype in S. coelicolor (Jakimowicz et al., 2005).

It has recently been reported that the M. tuberculosis gene Rv3676 encoding a CRP-FNR family protein (hereafter referred to as CRP^M) is defective in the vaccine strain Mycobacterium bovis BCG (Spreadbury et al., 2005). Moreover, an M. tuberculosis mutant lacking an intact Rv3676 gene is attenuated for virulence and shows reduced expression of several mycobacterial genes including resuscitation promoting factor, rpfA (Rickman et al., 2005). One of the genes that was dependent on the presence of Rv3676 for full expression was whiB1, suggesting the role of cAMP in the expression of whiB1 in M. tuberculosis.

In this study we show that whiB1 is transcribed by a class I-type CRP-dependent promoter. We provide evidence that whiB1 transcription is regulated by varying cAMP levels via a mechanism involving direct binding of cAMP-activated CRP^M to a consensus site adjacent to the whiB1 promoter.

Bacterial strains, plasmids and growth conditions.
In this study we used Escherichia coli strain DH5α (F'/endA1 hsdR17 [ ] glnV44 thi-1 recA1 gyrA [Nal^r] relA1 Δ[lacIZYAargF]U169 deoR [φ80dlacΔ(lacZ)M15]), from Stratagene, and BL21(DE3) (F ompT hsdS_B [ ] gal dcm [DE3]), from Novagen. M. tuberculosis CDC1551 was obtained from Colorado State University, CO, USA. Plasmid pSD5B (mycobacteriaE. coli shuttle vector carrying promoterless lacZ gene, Km^R; Jain et al., 1997) was kindly provided by Dr Anil Tyagi, University of Delhi South Campus, New Delhi, India, and pET-22a (expression vector with pBR322 origin of DNA replication and T7 promoter, Ap^R) was obtained from Novagen. LuriaBertani (LB) broth and LB agar media were used for culturing E. coli, and 7H9 broth and 7H10 agar, both supplemented with 1x OADC were used for culturing mycobacteria. Both E. coli and mycobacteria were grown at 37 °C with shaking at 200 r.p.m. For cAMP treatment, the M. tuberculosis cells were cultured in 7H9 medium supplemented with 1x OADC (oleic acid/albumin/dextrose/catalase) and harvested at the desired OD₆₀₀. The cell pellet was resuspended in 0.1 vol. 0.125 M potassium phosphate buffer, pH 7.4, and divided into two equivalent portions. One of these was incubated with a cell-diffusible form of cAMP, dibutyric cAMP (db-cAMP) (20 mM) and the other with an equivalent concentration of butyric acid (20 mM). After incubation for 2 h at 37 °C with shaking, the cell pellets were harvested and stored at 70 °C. Antibiotics were added when necessary: ampicillin (50 µg ml¹) and kanamycin (25 µg ml¹ for E. coli and 15 µg ml¹ for mycobacteria).

DNA techniques.
Restriction enzymes and T4 DNA ligase were purchased from New England Biolabs and Taq polymerase was purchased from Invitrogen. Protocols for DNA manipulations, including plasmid DNA preparation, restriction endonuclease digestion, agarose gel electrophoresis, and isolation and ligation of DNA fragments were as described by Sambrook et al. (1989). E. coli DH5α was transformed by the standard protocol (Sambrook et al., 1989) and M. tuberculosis by electroporation. PCR amplifications were carried out according to the manufacturer's specifications (Applied Biosystems). Each of the 30 cycles was carried out at 94 °C for 1 min, 60 °C for 1 min and 72 °C for 1 min, followed by final extension at 72 °C for 10 min. DNA fragments used for cloning and labelling reactions were purified by using the Qiagen gel extraction kit according to the manufacturer's specifications.

Mapping the 5' end of the whiB1 mRNA.
The 5'-RACE (rapid amplification of cDNA ends) technique (Frohman, 1993) was used to determine the transcription initiation site with a kit purchased from Invitrogen. Total RNA was isolated from M. tuberculosis (mid-exponential phase) by using the Trizol method according to the instructions provided by the supplier (Invitrogen). The abridged anchor primer (AAP) and abridged universal amplification primers (AUAP) were used in combination with the gene-specific primers. The gene-specific primers used for RT-PCR, nested PCR1 and nested PCR2 were whiB1/RACE1 and whiB1/RACE2, respectively (see Fig. 1b). The PCR products were subsequently sequenced by using the gene-specific primer whiB1/RACE2.

(36K): Fig. 1. Identification of the TSP of the M. tuberculosis whiB1 gene by 5'-RACE. (a) Mapping of the 5' end of the M. tuberculosis whiB1 transcript by the 5'-RACE method. Fragments were amplified by PCR using the abridged anchor primer (Invitrogen) in combination with the gene-specific primer whiB1/RACE1 (not shown). The resulting amplicon was used as the template for another PCR amplification by using the abridged universal amplification primer (Invitrogen) and gene-specific primer whiB1/RACE2, resulting in formation of an amplicon of ∼200 bp (lane 1). Lane 2, size markers. DNA sequencing of this fragment identified the TSP. (b) Sequence of the whiB1 5'UTR. The TSP (+1) identified by 5'-RACE is indicated by a right-angled arrow. Positions of bases upstream of the TSP are indicated by numbers above the respective bases. The predicted 10 and 35 sequences are indicated by boldface. The putative ribosome-binding site (RBS) and the translation start codon (ATG) are boldfaced and underlined. The putative CRP-binding site is shaded. Horizontal arrows indicate the positions and directions of the primers that were used for PCR amplification of DNA fragments.

Construction of P_whiB1lacZ transcriptional fusion construct and measurement of β-galactosidase activity.
A DNA fragment ∼250 bp in length containing the whiB1 promoter sequence from positions 214 to +33 with respect to the transcriptional start point was PCR-amplified from chromosomal DNA of M. tuberculosis using the primers PwhiB1(F) (carrying an XbaI site) and PwhiB1(R) (carrying an SphI site) (Fig. 1b). A transcriptional fusion construct was generated by cloning the PCR amplicon in pSD5B at the XbaI and SphI sites, and the recombinant vector was introduced into M. tuberculosis by electroporation. Promoter activity was determined by β-galactosidase assays using cell lysates of the cultures harvested at their exponential phase of growth. Assays were performed as described by Miller (1972); the values presented are means of three separate assays.

Generation of point mutations.
DNA fragments carrying the mutations in promoter sequences were constructed by two-step PCR amplification using overlapping modified primers, as described earlier (Good & Nazar, 1992). After cleavage with XbaI/SphI, these fragments were cloned in pSD5B, the recombinant vectors were used to transform the M. tuberculosis CDC1551 strain, and the transformants were grown on 7H10 agar plates containing kanamycin (25 µg ml¹) and X-Gal (60 µg ml¹).

Purification of CRP^M of M. tuberculosis.
The gene encoding the CRP^M of M. tuberculosis was PCR-amplified from M. tuberculosis chromosomal DNA using gene-specific primers, Pr3676(F) (5'-GGGGCATATGGACGAGATCCTGGCCAGGG-3') and Pr3676(R) (5'-GGGGCTCGAGCCTCGCTCGGCGGGCCAGTCT-3'). These primers were selected such that the amplicons carried the sequence of the complete ORF excluding the termination codon. The amplicons were cloned into the expression vector pET-22a at the NdeI and XhoI sites. The resulting construct pETcrp was used to transform E. coli BL21(DE3). CRP with histidine tags at the C-terminus was purified from the transformants by Ni-NTA chromatography according to the manufacturer's protocol (Qiagen). Purity of the protein was analysed by SDS-PAGE.

Electrophoretic mobility shift assay.
DNA probes used for electrophoretic mobility shift assay (EMSA) analysis were prepared by PCR amplification of the desired regions of the whiB1 promoter, using 5'-end biotin-labelled synthetic oligonucleotides as the primers. The amplicons were purified from agarose gels and used for gel-shift experiments. The EMSA reaction mixture (10 µl) contained 20 fmol biotin-labelled probe and CRP^M in binding buffer (40 mM HEPES/NaOH pH 7.5, 50 mM KCl, 2 mM MgCl₂, 0.01 mM EDTA, 1 mM DTT and 1 mg BSA ml¹) containing 1 µg nonspecific competitor DNA poly(dI-dC) (Amersham Pharmacia Biotech) and 1 mM cAMP. Following incubation at 37 °C for 15 min, the DNAprotein complexes were resolved in a native 5 % polyacrylamide gel (acrylamide : bisacrylamide, 30 : 1, w/w) in 0.5x TBE buffer (45 mM Tris base, 45 mM boric acid and 1 mM EDTA, pH 8.0). Samples were electrophoresed at 100 V, and transferred to positively charged nylon membranes (Amersham Pharmacia Biotech). After UV cross-linking (100 µJ cm² for 2 min), the biotinylated probes in the membrane were detected using the LightShift Chemiluminescent EMSA kit (Pierce). Each EMSA was performed three times to confirm CRP^M-DNA binding.

The upstream region of whiB1 possesses a CRP-binding site
Sequence analysis of the upstream region of the M. tuberculosis whiB1 gene indicated the presence of a putative CRP-binding site (AGTGANNNNNNCCACG) 169 bp upstream of the translation start site (Fig. 1b). In order to define the promoter region, we determined the transcription start point (TSP) of the whiB1 transcript by the 5'-RACE technique, as described in Methods. Total RNA was isolated from M. tuberculosis in mid-exponential phase (OD₆₀₀ 0.6), and final amplification of dCTP-tailed cDNA was performed using primers AUAP and whiB1/RACE2 (Fig. 1b). As shown in Fig. 1(a), PCR amplification of the RACE cDNA products revealed a major band (>200 bp in size) which upon DNA sequencing corresponded to a whiB1 TSP from a purine base, G, located 109 bp upstream of the translation start codon (Fig. 1b). Analysis of the sequence upstream of this TSP revealed the presence of promoter-like sequences at the 35 (TTGACA) and 10 (AACGAT) positions. The minor band (<100 bp in size) observed in Fig. 1(a) proved by sequence analysis to be an artefactual primer-dimer amplification product.

The CRP-binding consensus sequence in E. coli (5'-TGTGA-TCTAGA-TCACA-3') exhibits perfect twofold sequence symmetry, with the bold-faced bases representing the left and right arms for binding of the active CRP dimer (Berg & von Hippel, 1988; de Crombrugghe et al., 1984). The putative CRP-binding site in the whiB1 5'UTR (5'-AGTGAGATAGCCCACG-3') is located between positions 51 and 66 bp (centred at 58.5 bp) upstream of the TSP, and contains the critical residues, 5'-GTG-3' in the left arm and 5'-CAC-3' in the right arm (Fig. 1b). The left arm contains a near-perfect sequence (AGTGA) and the right arm contains 3/5 matching residues (CCACG).

Transcription of whiB1 is constitutive and positively regulated by CRP^M
In order to assess the expression of the whiB1 promoter in M. tuberculosis, a DNA fragment carrying the wild-type whiB1 5'UTR (spanning positions 214 to +33) was cloned upstream of a promoterless lacZ gene in pSD5B (a mycobacterial promoter probe vector), and the recombinant plasmid (pSD5B-wtwhiB1) was used to transform M. tuberculosis. Promoter activity was determined by β-galactosidase assays with cell lysates of bacilli harvested during different phases of growth (OD₆₀₀ 0.4, 0.8, 2.0 and 3.5, respectively). As shown in Fig. 2(b), the β-galactosidase specific activity resulting from whiB1 promoter expression at different growth phases remained constant, thus indicating that whiB1 is expressed constitutively in M. tuberculosis. Recently, a similar pattern of whiB1 expression was observed in M. tuberculosis by Geiman et al. (2006) by using the RT-PCR method.

(23K): Fig. 2. Base substitutions in the left arm of the putative M. tuberculosis whiB1 CRP-recognition site and their influence on whiB1 activity and CRPM binding. (a) Sequence analysis of the whiB1 promoter with wild-type and mutant CRP recognition sites. Boldfaced letters represent bases at the 35 and 10 positions. The underlined sequence represents the putative CRP-binding site. Positions of mutations in the CRP-binding site are indicated by asterisks. (b) Effects of base substitutions on whiB1 promoter specific activity. M. tuberculosis bacilli harbouring either pSD5B-wtwhiB1 (wild-type) or pSD5B-mtwhiB1 (mutant) promoter constructs were harvested at different phases of growth (OD600 0.4, 0.8, 2.0 and 3.5). Promoter activity was measured as β-galactosidase specific activity. The values are means±SD of three separate assays. (c) SDS-PAGE analysis of overproduced CRPM as a His6 fusion protein (left lane). Coomassie-blue-stained gel showing the isolated CRPM protein (400 pmol) following affinity chromatography. M, molecular mass markers. (d) Interaction of CRPM with the whiB1 5'UTR. Biotin-labelled whiB1 5'UTR (lanes 2 and 3) or biotin-labelled whiB1 5'UTR with mutations in the putative CRP-binding site (5'-AGTGAgatagcCCACG-3' to 5'-AGGTAgatagcCCACG-3') (lane 5) was incubated with CRPM in the presence of 1 mM cAMP before separation of CRPMwhiB1 complexes by gel electrophoresis. As a specificity control, unlabelled competitor DNA was added in 20-fold molar excess (lane 3). Lanes 1 and 4 contain whiB1 promoter fragments with wild-type and mutant CRP-recognition sequences, respectively, in the reaction mixture lacking CRPM.

To evaluate the role of the putative CRP^M-recognition site in the regulation of whiB1 expression, we constructed a mutant derivative of the whiB1 promoter. The nucleotides at the third and fourth positions of the CRP-binding site in the whiB1 5'UTR (Fig. 1b), known to be critical for CRP binding in E. coli (Jansen et al., 1987), were replaced, altering the wild-type left arm consensus sequence from AGTGA to AGGTA (Fig. 2a). The mutant promoter DNA fragment was cloned in pSD5B (resulting in the construct pSD5B-mtwhiB1), and the activities of both the wild-type and mutant whiB1 promoters were compared in M. tuberculosis at OD₆₀₀ 0.4, 0.8, 2.0 and 3.5, as described above. As shown in Fig. 2(b), the whiB1 promoter with a wild-type CRP-binding site exhibited β-galactosidase activities of ∼7500 nmol min¹ mg¹ in M. tuberculosis. Point mutations in the CRP-binding site resulted in a 34-fold reduction in β-galactosidase specific activity to ∼2000 nmol min¹ mg¹ at each OD₆₀₀ tested (Fig. 2b). These observations indicate that the putative CRP-binding site is required in its intact form for full-level whiB1 expression in M. tuberculosis.

CRP^M regulates expression of whiB1 by directly binding to the upstream CRP recognition site
Based on the whiB1 promoter activities in mycobacteria, we examined whether whiB1 expression was regulated by direct binding of CRP^M to the putative CRP-recognition sequence adjacent to the whiB1 promoter. We performed EMSA analyses using purified recombinant CRP^M and DNA comprising the whiB1 promoter region and putative CRP binding site. The target DNA region extended from base 214 to +33 relative to the TSP, and was PCR-amplified using forward primer PwhiB1(F) and biotin-labelled reverse primer PwhiB1(R) (Fig. 1b) as described in Methods. Recombinant CRP^M protein was expressed in E. coli as a fusion protein with a His₆ tag at its C-terminus and purified using metal affinity chromatography. Analysis of the purified protein by SDS-PAGE confirmed that the preparation was homogeneous, and CRP^M had an apparent molecular mass of approximately 26 kDa, which is in good agreement with its predicted molecular mass (Fig. 2c).

As shown in Fig. 2(d), CRP^M was able to bind the wild-type whiB1 5'UTR with a concentration of 1.5 µM per 2 nM template DNA fragment (Fig. 2d, lane 2). Addition of CRP^M at lower concentrations did not result in sufficient complex formation to be detected by EMSA under these conditions (data not shown). In order to confirm that the binding was specific, the binding reactions were carried out in the presence of a 20 molar excess of unlabelled target DNA. In the presence of competitor, binding was completely abolished (Fig. 2d, lane 3), indicating that CRP^M binds specifically to the whiB1 5'UTR. To verify that CRP^M binds to its predicted CRP recognition site in the whiB1 5'UTR, we performed the EMSA experiment with a whiB1 5'UTR containing a defective CRP-binding site (substitution of the left arm consensus sequence from 5'-AGTGA-3' to 5'-AGGTA-3', Fig. 2a) as the template. As shown in Fig. 2(d), lane 5, CRP^M is unable to bind to the mutant CRP-recognition site in whiB1 5'UTR, thus supporting the premise that CRP^M directly regulates the expression of whiB1 in M. tuberculosis by specifically binding to the CRP-recognition site adjacent to the whiB1 promoter.

Regulation of the expression of whiB1 by CRP^M requires cAMP
It is well established that both CRP and CRP^M specifically bind to DNA at CRP recognition sites in the presence of cAMP (Bai et al., 2005; Botsford & Harman, 1992; Busby & Buc, 1987; Crasnier, 1996; Kolb et al., 1993; Lawson et al., 2004). To characterize the role of cAMP in CRP^M binding to the whiB1 5'UTR, we carried out EMSA with CRP^M and the whiB1 5'UTR in the presence of a gradient of cAMP concentrations varying from 0.1 µM to 1 mM (Fig. 3a). Since no complex was observed when cAMP was omitted from the binding reaction, our results demonstrate that the mycobacterial CRP requires cAMP for significant binding to the whiB1 5'UTR (Fig. 3a, lane 2). Although binding of CRP^M to the whiB1 5'UTR was observed in the presence of cAMP at concentrations as low as 0.1 mM, maximal complex formation required 1 mM cAMP (Fig. 3a). In addition, CRP^M was unable to bind to the mutant CRP-recognition site in the whiB1 5'UTR at any concentration of cAMP used, thus indicating that CRP^M binds specifically to the CRP-recognition site in the whiB1 5'UTR (Fig. 3b). Our results therefore indicate that the ability of CRP^M to bind the whiB1 5'UTR varied directly with the cAMP concentration.

(33K): Fig. 3. Effects of cAMP concentration on CRPMwhiB1complex and expression of whiB1 in M. tuberculosis. (a) Effects of increasing cAMP concentrations on the binding of CRPM to the whiB1 5'UTR. Purified CRPM was incubated with the different concentrations of cAMP for 30 min on ice, before binding to the whiB1 5'UTR. Biotin-labelled whiB1 5'UTR (2 nM) was incubated with cAMP-CRPM (1.5 µM) for 15 min at 37 °C before separation of CRPMwhiB1 complexes by gel electrophoresis. The cAMP concentrations used are indicated above the respective lanes. Lane C, control reaction lacking CRPM. Complex, CRPMwhiB1 complex; Free, unbound DNA fragment. (b) Effects of increasing cAMP concentrations on the binding of CRPM to the whiB1 5'UTR with mutations in the CRP-binding site (5'-AGTGAgatagcCCACG-3' to 5'-AGGTAgatagcCCACG-3'). Biotin-labelled mutant whiB1 5'UTR (2 nM) was incubated with cAMP-CRPM (1.5 µM) for 15 min at 37 °C before separation of CRPMwhiB1 complexes by gel electrophoresis. The cAMP concentrations used are indicated above the respective lanes. Free, unbound DNA fragment. (c) Effects of db-cAMP on whiB1 promoter specific activity. M. tuberculosis bacilli harbouring either pSD5B-wtwhiB1 (wild-type) or pSD5B-mtwhiB1 (mutant) promoter constructs were harvested 2 h after exposure to either db-cAMP or butyric acid (BA), as described in Methods. Promoter activity was measured at different phases of growth (OD600 0.4, 0.8, 2.0 and 3.5) as β-galactosidase specific activity. The values are means±SD of three separate assays.

The requirement of cAMP for the binding of CRP^M to the whiB1 5'UTR in vitro implies that cAMP levels regulate the expression of whiB1 in bacteria in vivo. We therefore analysed the effects of the cAMP analogue db-cAMP on the activities of wild-type and mutant whiB1 promoters in M. tuberculosis. In order to assess early transcriptional effects of cAMP addition rather than late-stage potentially pleiotropic phenomena, M. tuberculosis bacilli harbouring either pSD5B-wtwhiB1 or pSD5B-mtwhiB1 were harvested 2 h after exposure to db-cAMP, as described in Methods. Promoter activity was determined by β-galactosidase assays with cell lysates of bacilli harvested after treatment with either db-cAMP or butyric acid at different phases of growth (OD₆₀₀ 0.4, 0.8, 2.0 and 3.5). As shown in Fig. 3(c), at all OD₆₀₀ levels, db-cAMP treatment resulted in a 22.5-fold increase in the activity of the wild-type whiB1 promoter, compared with its activity in bacilli grown in the presence of butyric acid. However, transcription by the mutant whiB1 promoter essentially remained constant under both these conditions (Fig. 3c). These results thus suggest that whiB1 expression in M. tuberculosis may be controlled by a molecular switch provided by the cellular cAMP levels. The sequenced genome of M. tuberculosis contains seven genes belonging to the whiB family. Recent studies of these mycobacterial whiB paralogues have underscored the importance of several of the WhiB proteins for M. tuberculosis physiology. Recently, S. coelicolor WhiD, the orthologue of M. tuberculosis WhiB3, was shown to bind a [4Fe4S] cluster (Jakimowicz et al., 2005). In all probability the four cysteine residues, conserved in all but one of the Wbl proteins identified so far, are involved in binding the ironsulfur cluster. As a consequence of this interaction, it has been postulated that the WhiD protein may play a role in maintaining the redox balance of the cell. The presence of a putative helixturnhelix motif in the WhiB protein family implies that these proteins may perform a regulatory role as DNA-binding proteins. Recently, Kim et al. (2005) proposed that the orthologue of WhiB1 in Corynebacterium glutamicum, WhcE, may be involved in the regulation of genes sensing redox changes at the different stages of cell growth.

Based on a recent report (Rickman et al., 2005), we have chosen to focus on CRP-dependent regulation of the expression of M. tuberculosis whiB1. In E. coli, the CRP-dependent promoters are grouped into three classes. Class I has a single CRP-binding site, centred at one of the positions 61.5, 71.5, 81.5 or 91.5; class II possesses a CRP-recognition site centred at 41.5, which thus overlaps with the 35 sequence; and class III requires two or more CRP-binding sites in the promoter for full activation (Busby & Ebright, 1999; Ebright, 1993; Ushida & Aiba, 1990). Based on the position of the CRP-recognition sequence 58.5 nt upstream of the TSP, the M. tuberculosis whiB1 promoter region most resembles the class I CRP-dependent promoters of E. coli.

Our study indicates that CRP^M binds to a CRP recognition site upstream of the M. tuberculosis whiB1 promoter in a cAMP-dependent manner in vitro, and that db-cAMP administered exogenously strongly induces M. tuberculosis whiB1 transcription in vivo. The higher activity of the wild-type whiB1 promoter in comparison to its mutated derivative, lacking a functional CRP-recognition sequence, indicates a requirement for CRP^M binding to drive full-level expression of whiB1 in M. tuberculosis. However, despite the presence of a functional CRP recognition site, we observed constitutive expression of the wild-type promoter through the growth cycle. This may reflect relatively constant intracellular cAMP levels during mycobacterial growth in the glucose-rich OADC-supplemented 7H9 medium. We also observed constitutive expression from the mutant whiB1 promoter lacking the CRP recognition sequence at levels approximately three- to fourfold lower than those with an intact CRP recognition site, and we postulate that this represents a basal expression level from the σ⁷⁰-like promoter consensus sequences at the 35 (TTGACA) and 10 positions (AACGAT) in the core whiB1 promoter.

An earlier study by Rickman et al. (2005) found that whiB1 transcription was reduced in an M. tuberculosis mutant lacking an intact Rv3676 gene, which encodes CRP^M. However, the CRP-recognition consensus sequence in the M. tuberculosis whiB1 5'UTR lacks perfect twofold symmetry and diverges from the canonical E. coli consensus sequence at several positions. Indeed, a recent report which identified 73 CRP binding motifs in M. tuberculosis using a Gibbs sampling computational search model did not detect the whiB1 promoter region as a significant match (Bai et al., 2005). Despite the lack of perfect twofold symmetry and several differences from the E. coli consensus sequence, our observations, together with those of Rickman et al. (2005), indicate that certain divergent CRP^M-recognition sites such as that of whiB1 are likely to retain biologically significant activity in M. tuberculosis.

cAMP is an important signalling molecule regulating various cellular functions, including virulence factors from a diverse range of pathogens (Alspaugh et al., 2002; Caler et al., 2000; D'Souza & Heitman, 2001; Gross et al., 2003; Lee et al., 2003; Li et al., 2002). There is little information about the role of cAMP in mycobacteria, despite its presence in both fast- and slow-growing, as well as pathogenic and nonpathogenic, species (Padh & Venkitasubramanian, 1977). The genome sequence of M. tuberculosis revealed 15 ORFs that contain a cyclase homology domain (Cole et al., 1998; McCue et al., 2000), which implies that cAMP-mediated signal transduction may be a central and versatile tool that this pathogen may employ to combat multiple environmental challenges. Indeed, in one novel role for cAMP in mycobacterial pathogenesis, Mycobacterium microti, a member of the M. tuberculosis complex, has been reported to secrete cAMP during infection of macrophages. It is postulated that secreted cAMP may prevent phagolysosome formation and thereby protect bacilli from host-mediated destruction (Lowrie et al., 1975, 1979). Recently, it has been observed that exogenous cAMP induced the expression of several biologically significant genes in M. bovis BCG, thus implicating cAMP-mediated regulation of gene expression as an important mechanism in the M. tuberculosis complex (Gazdik & McDonough, 2005).

We conclude that whiB1 is the first gene of the wbl family shown to be regulated by cAMP in M. tuberculosis. Future studies of the effects of cAMP on whiB1 expression and subsequent downstream effects mediated by whiB1 will better define the role of cAMP signalling and Wbl-mediated gene regulation in pathogenic mycobacteria.

The support of NIH grants AI51668, AI36973, AI37856 and AI43846 is gratefully acknowledged.