Báo cáo khoa học: CmtR, a cadmium-sensing ArsR–SmtB repressor, cooperatively interacts with multiple operator sites to autorepress its transcription in Mycobacterium tuberculosis pptx

CmtR, a cadmium-sensing ArsR–SmtB repressor, cooperatively interacts with multiple operator sites to autorepress its transcription in Mycobacterium tuberculosis Santosh Chauhan, Anil Kumar, Amit Singhal, Jaya Sivaswami Tyagi and H. Krishna Prasad Department of Biotechnology, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, India All organisms require metal ions as cofactors for several enzymatic reactions. The physiological concen- trations of metal ions are maintained by the coordi- nated action of a family of intracellular metal-sensing and transporter proteins [1,2]. One such family of metal-sensor proteins (SmtB ⁄ ArsR) functions exclu- sively as transcriptional repressors by regulating intra- cellular metal ion concentrations under conditions of surplus metal ions [2]. These repressors sense di- and multivalent heavy metal ions and regulate the expres- sion of gene(s) encoding protein(s) that specifically expel or chelate metal ion(s) [2]. These metalloregula- tory repressors bind the operator ⁄ promoter DNA regions in operons regulated by stress-inducing concen- trations of heavy metal ions. Protein binding to the operator ⁄ promoter DNA is strongly inhibited by metal binding to the sensing apoprotein. A number of these metal-responsive transcriptional regulatory proteins have been described in a variety of microbes, namely SmtB, a Zn 2+ -responsive transcriptional repressor in Keywords ArsR–SmtB repressor; autoregulation; CmtR; metalloregulatory repressors; Rv1994c Correspondence H. K. Prasad, TB Immunology Laboratory, Department of Biotechnology, All India Institute of Medical Sciences, New Delhi 110029, India Fax: +91 11 26589286 Tel: +91 11 26594994 E-mail: hk_prasad@hotmail.com J. S. Tyagi, TB Molecular Biology Laboratory, Department of Biotechnology, All India Institute of Medical Sciences, New Delhi 110029, India Fax: +91 11 26588663 Tel: +91 11 26588491 E-mail: jstyagi@aiims.ac.in (Received 27 January 2009, revised 4 April 2009, accepted 20 April 2009) doi:10.1111/j.1742-4658.2009.07066.x CmtR is a repressor of the ArsR–SmtB family from Mycobacterium tuber- culosis that has been shown to sense Cd(II) and Pb(II) in Mycobacte- rium smegmatis. We establish here that CmtR binds cooperatively to multiple sites in M. tuberculosis DNA and protects an unusually long 90 bp AT-rich sequence from )80 bp to +10 with respect to its own initiation codon. CmtR interacts with four hyphenated imperfect inverted repeats matching the consensus sequence TA ⁄ GTAA-N 4–5 -TT ⁄ GATA in the protected region. SDS–PAGE and formaldehyde crosslinking experi- ments showed that CmtR forms higher-order oligomers (up to an octamer). The oligomerization of CmtR is in agreement with the cooperative binding of CmtR to multiple sites on DNA. Two promoters transcribe cmtR, and the major promoter physically overlaps with CmtR binding sites. Autorepression of CmtR is mediated by cooperative interaction of CmtR with multiple sites on DNA that occlude the major operon promoter. The combined results of a GFP reporter assay, an electromobility shift assay and a DNase I footprinting experiment establish that Cd(II), not Pb(II), disrupts the interaction of CmtR with DNA to de-repress transcription of the cmtR–Rv1993c–cmtA operon in M. tuberculosis. Abbreviations EMSA, electromobility shift assay; TSP, transcription start point. 3428 FEBS Journal 276 (2009) 3428–3439 ª 2009 The Authors Journal compilation ª 2009 FEBS Synechococcus species [3], ArsR in Escherichia coli [4], ZiaR in Synechocystis species [5], MerR in Streptomy- ces lividans [6] and CadC in Staphylococcus aureus [7]. Mycobacterium tuberculosis, a versatile pathogen, sur- vives in a variety of harsh environmental conditions including the phagosome of the mammalian host cell. Inside the phagosome, M. tuberculosis must adapt to metal fluxes and modulate the expression of genes involved in metal detoxification and efflux. Ten putative SmtB ⁄ ArsR metal sensors have been identified in the M. tuberculosis genome [8], of which three – namely NmtR [9], CmtR [10–12] and KmtR [13] – have been partially characterized. CmtR from M. tuberculosis is a winged helical DNA-binding repressor that was shown to sense cadmium and lead in the surrogate host Myco- bacterium smegmatis and is proposed to bind a single 25 bp site in the cmtR operator ⁄ promoter region in the apo-form [10]. The cadmium–CmtR complex and apo- CmtR were both shown to exist as a homodimer [12]. CmtR was also shown to be upregulated (approximately threefold) upon entry into macrophages [14]. Cadmium is present as an air pollutant and in ciga- rette smoke [15]. Cadmium is known to accumulate in human aleovolar macrophages [15]. M. tuberculosis may be exposed to toxic concentrations of cadmium in macrophages. As CmtR senses cadmium, it may mod- ulate the expression of genes involved in detoxification and efflux of this toxic metal from the mycobacterium. Here we demonstrate through electromobility shift assay (EMSA) and DNase I footprinting that CmtR cooperatively interacts with multiple binding sites and protects a 90 bp sequence in the CmtR operator ⁄ pro- moter region. Consistent with these data, we show here that CmtR exists as multimers under non-reducing conditions. The combined results from EMSA, DNase I footprinting, transcription start point (TSP) mapping and a GFP reporter assay showed that CmtR represses its transcription by promoter occlusion, and that Cd(II) dislodges CmtR from the operator ⁄ promoter to de-repress transcription of the cmtR–Rv1993c–cmtA operon. Results Co-transcription of cmtR, Rv1993c and cmtA in M. tuberculosis cmtR is located in the proximity of Rv1993c and Rv1992c ⁄ cmtA ⁄ ctpG in the M. tuberculosis genome (Fig. 1A). cmtR was previously shown to be co-tran- scribed with Rv1993c and cmtA in M. bovis [10]. To establish that cmtR, Rv1993c and cmtA are co-tran- scribed and constitute an operon in M. tuberculosis, RT-PCR was performed with logarithmic-phase RNA using primers 94RTf and 94RTr (Fig. 1A). The antici- pated PCR product of 330 bp was detected (Fig. 1B, lane 2). Purification of recombinant CmtR protein M. tuberculosis CmtR protein with an N-terminal hexahistidine tag (His-CmtR) was overexpressed in cmtR Rv1995 156 bp Rv1993cRv1992c/cmtA/ctpG 94RTf 94RTr 123 330 bp 16 12.5 12 anti-CmtR Dimer Dimer 12 anti-His 16 Dimer Trimer A B C D Fig. 1. (A) Schematic representation of the cmtR–1993c–cmtA gene locus in M. tuberculosis. (B) Co-transcription of cmtR–1993c–cmtA genes in M. tuberculosis. RT-PCR products of RNA from cultures of M. tuberculosis H37Rv were amplified using primers 94RTf and 94RTr. Lane 1, negative control (without reverse transcriptase); lane 2, cDNA; lane 3, genomic DNA (amplification control). (C) Immunoblot analysis of purified CmtR. Nitrocellulose blots were probed using anti-CmtR (C) and anti-His (D) serum. Lane 1, His-CmtR; lane 2, CmtR. Dimer and trimer species are indicated by arrowheads. The molecular mass of bands (in kDa) as predicted based on the use of a protein molecular mass marker is indicated. S. Chauhan et al. Autoregulation by CmtR FEBS Journal 276 (2009) 3428–3439 ª 2009 The Authors Journal compilation ª 2009 FEBS 3429 E. coli. The fusion protein was purified by Ni-chelate affinity chromatography, and was determined to be approximately 16 kDa in western blots developed with anti-CmtR serum or anti-His monoclonal IgG (Fig. 1C,D, lane 1). The histidine tag of His-CmtR was removed using recombinant Tobacco etch virus (rTEV) protease to give a protein of approxi- mately 12.5 kDa, CmtR (Fig. 1C, lane 2). Further- more, proteins corresponding to dimers (32 and 25 kDa) and a trimer (48 kDa) were also detectable (Fig. 1C,D). Mapping of in vivo transcription initiation sites Two TSPs were identified upstream of cmtR by primer extension analysis (Fig. 2A). The major primer extension product (T1 cmtR ) was identified, starting with the ‘G’ nucleotide located 34 bp upstream of the CmtR translational start site (Fig. 2A,B). Another primer extension product (T2 cmtR ), which was rela- tively weak, was also identified, starting with the ‘G’ nucleotide located 111 bp upstream of CmtR transla- tional start site (Fig. 2A,B). The putative )10 promoter elements identified upstream of T1 cmtR (con- served at five of the six positions) and T2 cmtR (conserved at three of the six positions) showed a resemblance to the SigA )10 element (Fig. 2B), but )35 elements of both the promoters were modestly conserved (two of the six positions, SigA consensus sequence TTGACW-N 17 -TATAMT where W = A ⁄ T, M=A⁄ C [16]). CmtR interacts with a 90 bp sequence spanning the translational start site To map the CmtR binding site precisely, DNase I footprinting was performed using purified CmtR and 315 bp of DNA that includes the sequence from )191 bp to +124 bp with respect to the translational start site of CmtR. An unusually long 90 bp protected region was observed, which spans )80 bp to +10 bp with respect to translational start site of CmtR (Figs 3 and 4A). This result suggests that CmtR binds to mul- tiple sites and may interact as a functional multimer in the operator ⁄ promoter region. Two strong hypersensi- tive sites were observed at intervals of 18 bp from the 3¢ end of the footprint (Fig. 3), suggesting that this region of DNA is bent or distorted upon CmtR binding. The T1 cmtR TSP exactly overlaps the CmtR binding site, and T2 cmtR is present upstream of the CmtR binding site (Fig. 4A), which indicates that CmtR represses the cmtR–Rv1993c–cmtA operon by obstructing contact between the RNA polymerase and the promoter sequence. CmtR cooperatively interacts with multiple sites in the cmtR promoter region A close examination of the CmtR protected sequence revealed four hyphenated inverted repeats matching con- sensus sequence TA ⁄ GTAA-N 4-5 -TT ⁄ GATA (Fig. 4A,B), which could be the binding site of CmtR. EMSA assays were performed with various size fragments of CmtR Rv1995 –10 T2 cmtR (+1) –10 –35 –35 A T1 cmtR T2 cmtR T1 cmtR (+1) GC T 1 A B Fig. 2. TSP mapping. (A) Primer extension of M. tuberculosis RNA isolated from an aerobic culture (lane 1). Two TSPs were mapped. The experiment was repeated with the same results using two different sam- ples of RNA. The results of dideoxy sequencing reactions using the same primer are shown on the left. (B) Sequence encom- passing both TSPs and the putative )10 and )35 promoter sequences. Autoregulation by CmtR S. Chauhan et al. 3430 FEBS Journal 276 (2009) 3428–3439 ª 2009 The Authors Journal compilation ª 2009 FEBS the operator ⁄ promoter region to assess CmtR binding to these sites (Fig. 5A). Four differently migrating DNA–protein complexes were observed (C1, C2, C3 and C4) when the EMSA was performed with the larg- est F-1 fragment ()191 to +124 bp with respect to translational start site of CmtR, Fig. 5B), which may represent binding of CmtR to four different sites (labeled as sites 1, 2, 3 and 4, Fig. 4A). Interestingly, at least three of these complexes appeared at the low- est CmtR concentration (Fig. 5B), indicating that CmtR has high affinity for these sites. This was consis- tent with the nearly complete protection observed even at the lowest protein concentration in DNase I foot- printing. Cooperativity in binding can be assessed from the breadth of transition on a log plot. For non-coop- erative interactions, an increase of 1.81 log units in protein concentration is required to increase the bound fraction from 10% to 90% [17]. Positive cooperativity reduces this span; an increase of 0.5 log units of CmtR protein concentration (480 nm to 2.4 lm, Fig. 5B) increased the total bound fraction from 10% to 90% (Fig. 5C), which shows a cooperative interaction of CmtR with the cmtR operator ⁄ promoter region. The DNA–protein complexes were specific as they dissoci- ated in the presence of an excess of cold probe and were conserved in the presence of an excess of non- specific DNA [poly(dI-dC), Fig. 5F]. To map CmtR binding sites within the 90 bp pro- tected region, EMSA was performed with a smaller fragment F-2 ()67 to +124 bp relative to the CmtR translational start site) that lacks 13 bp from the 5¢ end of the 90 bp CmtR protected sequence and also the first half of site 1 (Figs 4A and 5A). Only three DNA–protein complexes (C2, C3 and C4) were observed (Fig. 5D), suggesting that deletion had indeed disrupted one of the binding sites. EMSA with the F-3 fragment ()33 to +124 bp), which include sites 3 and 4, showed only two retarded complexes (C3 and C4) (Fig. 5E). Previously, it was shown that CmtR binds to a single site in the region that spans )33 to )9bp [10], which includes site 3 and part of site 4 (Fig. 4A). Taken together, the results show that CmtR interacts with four binding sites in the 90 bp protected region, which corresponds fairly well to four predicted binding sites in this region. To establish that the predicted sites are genuine CmtR binding sites, we performed EMSA with oligonucleotides of approximately 20 bp carrying the predicted binding sites 1–4 (Fig. 5G). The EMSA results clearly show that CmtR protein interacts with each of these binding sites, but with low affinity. No interaction was observed with non-specific oligonucleo- tide (Fig. 5G). The decreased affinity of CmtR to DNA (carrying individual sites) could be attributed to a loss of cooperativity. CmtR forms dimer and higher-order oligomers Several repressors, e.g EthR [18] and RstR [19] that bind at tandem sites, interact as multimers. CmtR interacts with multiple binding sites in the opera- tor ⁄ promoter region. Two experiments were performed to examine the oligomeric status of CmtR. First, SDS–PAGE analysis of soluble His-CmtR under non- reducing condition shows that it forms dimers (approx- imately 32 kDa) and higher-order oligomers (Fig. 6A,B). Bands corresponding to positions up to an octamer (approximately 48, 64, 80, 96, 112 and 128 kDa) were detected in western blots probed with anti-His (lane 2, Fig. 6A) and anti-CmtR serum (data not shown). Similar results were obtained when CmtR from which the His tag had been removed was used (not shown). Under reducing conditions, only mono- mers and dimers of CmtR were observed. Interestingly, CmtR is partially dimeric even in the presence of * TA G C CmtR 2.4 0 4.8 12 7.2 9.6 T2 cmtR T1 cmtR ATG (µM) Fig. 3. CmtR interacts with a 90 bp sequence in the cmtR–Rv1995 intergenic region. DNase I footprinting was performed using cmtR– Rv1995 intergenic DNA (a 315 bp fragment amplified using primers P6 and 1994intR; the cmtR non-coding strand was labeled) and increasing concentrations of CmtR protein. Arrowheads (right side) indicate the hypersensitive sites and the protected region is indi- cated by a black box. The results of dideoxy sequencing reactions using the same primer and DNA template are also shown. Asterisk indicates the strand which is labelled. S. Chauhan et al. Autoregulation by CmtR FEBS Journal 276 (2009) 3428–3439 ª 2009 The Authors Journal compilation ª 2009 FEBS 3431 20 mm dithiothreitol (Fig. 6B) or 7.5% b-mercapto- ethanol (Fig. 6A). Second, a formaldehyde crosslinking experiment was performed to confirm the oligomeriza- tion of CmtR. Although not all oligomeric states were observed, bands corresponding to the position of a dimer (approximately 32 kDa) and an octamer (approximately 128 kDa, Fig. 6C) were apparent, showing that CmtR has an inherent ability to form oligomers. CmtR is a Cd sensing repressor in M. tuberculosis It has previously been shown that M. tuberculosis CmtR is a cadmium- and lead-sensing repressor in the surrogate host M. smegmatis [10]. In order to establish which metal ion(s) the cmtR promoter is responsive to, a 212 bp DNA fragment (P6–P3 region, Fig. 4) with 156 bp of the cmtR–Rv1995 intergenic region was fused to a promoterless GFP gene in plasmid pFPV27 to give pCmtR, and electroporated into M. tuberculosis H37Rv. The culture was grown to mid-logarithmic phase (attenuance at 595 nm of approximately 0.3), and subsequently diluted to an attenuance at 595 nm of 0.1. Metal ions CdCl 2 , NiCl 2 , CoCl 2 and Pb(NO 3 ) 2 were added to the culture at maximum permissive con- centrations as used previously [10]. No inhibition of growth was observed in the presence of any of the metal ions used (Fig. 7A, inset). Increased GFP flores- cence (approximately 2.5-fold) was observed on addi- tion of Cd(II) but not with any other metal ion tested (Fig. 7A). This shows that CmtR senses only Cd(II) but not Pb(II) in M. tuberculosis. CmtR (+1) Rv1995 P6 P4 1994 intR P3 (–43) –10 –35 –10 T2 cmtR T1 cmtR P8 Site 1 Site 2 Site 3 Site 4 –35 –10 –35 –10 –35 cmtR Rv1995 T1 (+1) T2 (+1) (–43) 1234 (+47) (+47) (+1) A B C Fig. 4. (A) Nucleotide sequence and salient features of the cmtR–Rv1995 intergenic region. The TSPs (T1 cmtR and T2 cmtR ) are indicated by angled arrows. The putative )10 and )35 promoter elements are indi- cated by dashed boxes. The CmtR DNase I- protected sequence ()43 to +47 bp) is boxed. Full arrows indicate the CmtR recog- nition sites 1, 2, 3 and 4. The positions of primers are indicated by half-headed arrows. The arrowheads indicate hypersensitive sites. (B) The sequences of four putative CmtR binding sites and the consensus sequence with which CmtR may interact. (C) Detailed map of the intergenic region. The four CmtR binding sites (1, 2, 3 and 4) are indicated by white boxes within the 90 bp CmtR recognition sequence (gray box). The TSPs mapped in this study are shown by angled arrows. The putative )10 and )35 promoter elements are indicated by small black boxes. Primers P3 and P6 were used to amplify the cmtR promoter DNA cloned in GFP reporter vector. DNase I foot- printing was performed using P6 and the 1994intR amplicon. Primers P6, P4, P8 and 1994intR were used for EMSA. Autoregulation by CmtR S. Chauhan et al. 3432 FEBS Journal 276 (2009) 3428–3439 ª 2009 The Authors Journal compilation ª 2009 FEBS Cadmium disrupts the CmtR–DNA sequence-specific interaction The effect of metal ions on in vitro binding of CmtR to the operator ⁄ promoter region was determined using DNase I footprinting and a gel shift assay. Ni(II), Co(II) and Pb(II) were not able to dissociate the com- plex even at high concentrations of 500, 200 and 50 lm, respectively (Fig. 7B), whereas Cd(II) disrupted the interaction of CmtR with DNA (Fig. 7B,C). % DNA bound [CmtR] µM 0.01 0.1 1 10 0 10 20 30 40 50 60 70 80 90 100 C1 C2 C3 F 0.12 0 0.24 0.36 0.48 0.60 0.72 0.84 0.96 1.08 1.2 1.8 2.4 3.0 3.6 4.8 5.4 6.0 (µM) CmtR C4 12345678 9101112131415161718 F-1 CmtR 123456 78910 C2 C3 F C4 0.1 0 0.2 0.3 0.4 0.6 0.7 0.8 1.0 1.2 F-2 C3 C4 F 0.12 0 0.24 0.48 0.84 1.2 2.4 3.6 4.8 5.4 6.0 CmtR 1 2 34567891011 F-3 F-1 (315 bp, –191 to +124) F-2 (191 bp, –67 to +124) F-3 (157 bp, –33 to +124) CmtR 12 34 234 34 P8 1994intR P4 P6 T1 12345 Site 2 Site 1 Site 3 Site 4 123 123 123 123 123 Control A BC D GF E (µM) (µ M) Fig. 5. CmtR binds cooperatively to multiple sites in the cmtR promoter region. (A) Various fragments used in EMSA. Angled arrows indi- cate the CmtR TSP and the ATG site. Primers used to amplify fragments are indicated by half-headed arrows. The putative CmtR binding sites (1, 2, 3, and 4) are indicated. (B) 32 P-labeled cmtR promoter DNA F-1 was incubated in the absence (lane 1) or presence (lanes 2–18) of increasing concentrations of CmtR protein. The arrowhead indicates the position of free DNA, and the arrows indicate the DNA–protein complexes. The arrow within the gel indicates the position of DNA–protein complex 2. (C) The fraction of bound F-1 DNA (estimated by subtracting free DNA from input DNA, Fig. 1A) versus CmtR concentration was plotted using SIGMAPLOT 2001 (www.sigmaplot.com). (D, E) 32 P-labeled F-2 (D) and F-3 (E) fragments were incubated in the absence (lane 1) or presence (lanes 2–10 ⁄ 11) of increasing concentrations of CmtR protein (l M). (F) A competition assay was performed with the F-1 fragment and 3 lM of CmtR with no competitor (lane 2), with 50 · non-specific competitor [poly(dI-dC), lane 3], or with 10· (lane 4) and 50· (lane 5) self-competitor; lane 1 contains labeled DNA only. (G) EMSA was performed with approximately 20 bp double-stranded DNA carrying or not carrying (control) a CmtR binding site (1, 2, 3 and 4) in the absence (lane 1) and presence of 1 l M (lane 2) or 2 lM (lane 3) of CmtR protein. Control, 38 bp double-stranded DNA known to bind to the DevR protein of M. tuberculosis (unpublished result). S. Chauhan et al. Autoregulation by CmtR FEBS Journal 276 (2009) 3428–3439 ª 2009 The Authors Journal compilation ª 2009 FEBS 3433 DNase I footprinting in the presence of metal ions confirmed that the site-specific interaction of CmtR was only dislodged by Cd(II). Together, the in vitro interaction studies and in vivo reporter assay shows that Cd(II) dissociates bound CmtR from DNA to de-repress transcription from the cmtR promoter. Discussion Intracellular concentrations of metals can be toxic to bacteria; therefore their uptake is tightly regulated by metal-dependent transcription regulators. CmtR is a Cd(II)-sensing repressor that may regulate genes involved in reducing the intracellular level of Cd(II). We demonstrate here by DNase I footprinting and EMSA that CmtR interacts cooperatively with multiple binding sites that span 90 bp of sequence up- and downstream of its own translational start site ()80 to +10 bp). Our results show that CmtR interacts with four imperfect hyphenated inverted repeats matching the consensus sequence TA ⁄ GTAA-N 4-5 -TT ⁄ GATA. Previously, on the basis of EMSA results, Cavet et al. (2003) proposed that CmtR may interact with a single degenerate 10-5-10 hyphenated inverted repeat [10]; however, no such instance of this predicted site was observed within the 90 bp CmtR protected sequence as shown in Fig. 4. However, this binding site does not resemble the sites described in the current study. In the EMSA assays performed by Cavet et al., the DNA–protein complexes were visualized by ethidium bromide staining [10], which is less sensitive compared to the radiolabeling and imaging techniques used in the present study. Hence they may not have visualized the multiple DNA–protein complexes observed in the pres- ent study. In addition, DNase I footprinting experi- ments show that CmtR binds to an extended 90 bp DNA sequence compared to the 25 bp sequence proposed by Cavet et al. (2003). To our knowledge, this is the first report of an ArsR–SmtB family repres- sor producing an exceptionally long footprint on DNA and interacting with multiple sites. Most of the SmtB ⁄ ArsR repressors have been proposed to recog- nize one or two degenerate AT-rich inverted repeats in their operator ⁄ promoter region [20–22]. ZntR, a SmtB ⁄ ArsR family repressor, has been proposed to interact with a hyphenated 9-2-9 inverted repeat (ATA TGAACA-AA-TATTCATAT) within the 49 bp of protected DNA [20]. SmtB has been proposed to inter- act with two hyphenated imperfect inverted repeats (6-2-6, TGAACA-GT-TATTCA and 7-2-7, CTGAA TC-AA-GATTCAG) in the smtB operator ⁄ promoter region [22]. Primer extension analysis identified two TSPs for cmtR. The major TSP (T1 cmtR ) and the putative )10 and )35 promoter elements completely overlap the CmtR binding sites, and the other TSP (T2 cmtR )is present upstream of the CmtR binding sites. This architecture of the promoter suggests that the interac- tion of the RNA polymerase is hindered by the 14 29 100 M 50 60 70 200 Monomer Dimer Trimer Tetramer Pentamer Hexamer 12 Heptamer Octamer 14 29 97 Monomer Dimer Higher oligomer (~octamer) 123M Monomer Dimer Higher oligomer Trimer Tetramer 12345 A B C Fig. 6. CmtR forms oligomers. Western blot analysis of His-CmtR probed with anti-His serum. CmtR was boiled in sample buffer in (A) the absence (lane 2) or presence of 7.5% b-mercaptoethanol (lane 1), or (B) the absence (lane 1) or presence of increasing con- centrations of dithiothreitol (1, 5, 10 and 20 m M, lanes 2–5, respec- tively) and resolved on 12% SDS–PAGE. (C) A crosslinking experiment with CmtR was performed in the absence (lane 1) or presence (lanes 2 and 3) of 0.1% formaldehyde. Lanes 1 and 3 contain 30 lg of CmtR, and lane 2 contains 20 lg of CmtR. ‘M’ represents molecular mass markers in kDa. Autoregulation by CmtR S. Chauhan et al. 3434 FEBS Journal 276 (2009) 3428–3439 ª 2009 The Authors Journal compilation ª 2009 FEBS binding of CmtR, resulting in repression of the cmtR–Rv1993c–cmtA operon. The CmtR binds cooperatively to multiple sites in a stepwise manner (in the EMSA). CmtR oligomers up to an octamer were observed. Hence, it may be possi- ble that CmtR oligomerizes in a stepwise manner on DNA as has been reported for other repressors [18,19]. All repressors of the SmtB ⁄ ArsR family, e.g. CadC, SmtB and ArsR [23–25], have been shown to exist as homodimers. The formation of higher oligomers has not been observed previously among SmtB ⁄ ArsR fam- ily repressors, but is not uncommon for repressors gen- erally. This property may allow the repressors to efficiently mask the promoter from RNA polymerase. CmtR was previously shown to exist as a dimer, but as dithiothreitol was used (1-5 mm) during protein purification and anaerobic conditions were used during further experimentation [11,12], it is possible that reducing conditions may have disrupted the higher- order oligomers to produce the dimers observed here (Fig. 6B). CmtR has six cysteine residues, Cys4, Cys24, Cys35, Cys57, Cys61 and Cys102, of which Cys57, Cys61 and Cys102 bind cadmium ions [11,12] and were shown to be important for cadmium-depen- dent activation [10]. Mutation of Cys24 of CmtR made the repressor non-functional, indicating that this resi- due may be involved in oligomerization of CmtR. Our in vivo GFP reporter assay suggests that CmtR binds to Cd(II) and induces cmtR expression in M. tuberculosis by derepression. This is in partial agreement with a previous report that Cd(II) and Pb(II) can both act as an inducer of cmtR in M. smegmatis [10]. Our in vitro experiments (EMSA and DNase I footprinting) also support the in vivo results, where the sequence-specific interaction was abolished by Cd(II) at physiological concentration (5 lm), but not by Pb(II) even at higher concentra- tions (50 lm). The dissociation of the DNA–protein 100 200 300 400 500 600 700 24 48 72 120 144 168 Time (h) RFU/Attenuance 595 Control A B C Cd Pb Ni Co 0.1 0.5 0.9 0 48 96 144 Time (h) Attenuance 595 CmtR – + + + + + Metal ion – – Ni Cd Co Pb CmtR – + Metal ion – – Cd Pb + + + + + + + + F 1 2 3 4 5 6 7 8 9 10 Fig. 7. (A) CmtR is a Cd-sensing repressor in M. tuberculosis. CmtR-directed GFP fluo- rescence in aerobic shaken M. tuberculosis cultures in the absence (control) or presence of metal ions (Ni, Cd, Co and Pb) at maximum permissive concentrations. GFP fluorescence is expressed as relative fluorescence unit (RFU) ⁄ attenuance after background subtraction. The mean values for two independent experiments are plot- ted. Growth curves for all the strains are shown in the inset. The metal ions were added to the culture when the attenuance at 595 nm was 0.1, and GFP fluorescence was measured at 24 h intervals. (B) DNase I footprinting was performed using fragment F-1 and 2.4 l M CmtR protein in the absence and presence of metal ions as indicated. Ni(II), Cd(II), Co(II) and Pb(II) were added at concentrations of 500, 5, 200 and 50 l M, respectively. (C) EMSA with fragment F-2 and CmtR protein (3 l M) in the absence (lane 2) or presence of increasing concentra- tions of Cd(II) (1, 2.5, 5 and 10 l M, lanes 3–6, respectively) and Pb(II) (5, 10, 20 and 40 l M, lanes 7–10, respectively). S. Chauhan et al. Autoregulation by CmtR FEBS Journal 276 (2009) 3428–3439 ª 2009 The Authors Journal compilation ª 2009 FEBS 3435 complex at physiological concentrations shows that Cd(II) has very high affinity for CmtR. The disruption of a repressor–promoter complex at physiological con- centration is rare, but was also observed in case of ZntR [20]. The difference between our results and the previous results [10] could be because of the different hosts used (M. tuberculosis versus M. smegmatis), which may differ in the regulation of genes [26–28]. M. smegmatis, which is a non-pathogenic saprophytic bacteria, may encounter varied metal toxicity com- pared to pathogenic M. tuberculosis, which may result in a different response to ions. Moreover, the homolog of M. tuberculosis CmtR in M. smegmatis (MSMEG_5603) has only 67% similarity with M. tuberculosis CmtR (data not shown). The cmtR, Rv1993c and cmtA genes constitute an operon. The role of Rv1993c and cmtA is not known, but the cmtA gene product has sequence similarity to the well-characterized metal transporting P 1 -ATPase pump, which pumps out metal ions that may otherwise be toxic to the bacterium. Our results show that CmtR binds to multiple sites to repress the operon when the Cd(II) ion is not present (Fig. 8). These sites, which overlap the major promoter (T1) and are located downstream of the T2 promoter, do not allow interac- tion of RNA polymerase with promoter DNA in the presence of bound CmtR. When present, the Cd(II) ion binds to CmtR and decreases its affinity for DNA, resulting in its release from DNA and hence transcrip- tion of the operon (Fig. 8). The increased concentra- tion of CmtA may actively pump the Cd(II) out. In the absence of Cd(II), the cmtR–Rv1993c–cmtA operon is repressed again by CmtR (Fig. 8). Experimental procedures Plasmids, bacterial strains and culture conditions M. tuberculosis H37Rv was cultured in Dubos medium con- taining 0.05% Tween-80 plus 0.5% albumin ⁄ 0.75% dex- trose ⁄ 0.085% NaCl at 37 °C under aerobic shaking conditions (220 r.p.m.). E. coli DH5a was grown in Luria– Bertani medium usually and in 2· YT medium [29] for pro- tein overexpression. Antibiotics, when required, were used at the concentrations indicated: ampicillin at 100 lgÆmL )1 and kanamycin at 25 lgÆmL )1 . All cloning steps were per- formed as described [29]. The plasmids and primers used in this study are listed in Tables 1 and 2, respectively. Cloning and purification of CmtR The cmtR coding sequence was amplified from M. tubercu- losis H37Rv DNA using primers P1 and P2 (Table 2) engi- neered to contain restriction sites for BamHI and HindIII, respectively. The amplified product was digested with the indicated restriction enzymes, and cloned into pPROEx- HTb (Invitrogen, Carlsbad, CA, USA) generating pPRO- CmtR. The construct was verified by DNA sequencing. Recombinant His-CmtR (an N-terminally histidine- tagged fusion protein of approximately 16.0 kDa) was over- expressed by growing recombinant E. coli DH5a at 37 ° C to an attenuance at 595 nm of 0.4–0.5, followed by induc- tion with 1 mm isopropyl thio-b-d-galactoside for 4 h at 37 °C. The induced cells were harvested, resuspended in buffer A (20 mm Tris pH 8.0, 500 mm NaCl, 20 mm imid- azole, 10% glycerol and 1 mm phenylmethanesulfonyl fluo- ride) and sonicated (Branson Ultrasonics, Danbury, CT, USA) on ice (duty cycle 60, four pulses of 2 min each). The –10 –35 cmtR Rv1995 T1 T2 Rv1993ccmtA Cd 2+ Cd 2+ - CmtR –10 –35 cmtR Rv1995 T1 T2 Rv1993ccmtA –10 –35 Fig. 8. Transcriptional regulation of the cmtR–1993c–cmtA operon by CmtR. CmtR (oval) represses transcription of the cmtR–1993c– cmtA operon by binding to multiple sites overlapping and down- stream of the T1 cmtR and T2 cmtR promoters, respectively. Cd(II) (black circles) acts as an inducer, binding to CmtR and releasing it from the DNA, resulting in the de-repression of operon trans cription. Table 1. Plasmids used in this study. Km R , kanamycin resistance. Plasmid Description Source or reference pPROEX-HTb E. coli expression vector (N-terminal histidine tag) Invitrogen pGEMT-Easy E. coli cloning vector Promega pFPV27 E. coli–mycobacteria shuttle plasmid containing a promoterless GFP gene, Km R [32] pPRO-CmtR cmtR coding region in pPROEX-HTb to overexpress CmtR with a N-terminal histidine tag This study pCmtR pFPV27 containing the 156 bp Rv1994c–Rv1995 intergenic region promoter (P6–P3 fragment) upstream of GFP This study Autoregulation by CmtR S. Chauhan et al. 3436 FEBS Journal 276 (2009) 3428–3439 ª 2009 The Authors Journal compilation ª 2009 FEBS sonicate was centrifuged at 12 000 g for 30 min and the supernatant was applied to a nickel-nitrilotriacetic acid column. Recombinant His-CmtR protein (approximately 16.0 kDa) was eluted in buffer A containing 250 mm imid- azole. The histidine tag was removed using rTEV protease (Invitrogen) according to the manufacturer’s protocol to yield a protein of approximately 12.5 kDa, referred to as CmtR. The purified His-CmtR and CmtR proteins were dialyzed against 50 mm Tris (pH 8.0), 50 mm NaCl and 50% glycerol, and stored at )20 °C. SDS–PAGE, western blotting and formaldehyde crosslinking Purified His-CmtR or CmtR was boiled for 5 min at 100 °C in SDS sample buffer (300 mm Tris ⁄ HCl pH 6.8, 12% SDS, 60% glycerol, 0.6% bromophenol blue) in the absence and presence of 7.5% b-mercaptoethanol or various concentrations of dithiothreitol. The samples were resolved by 12% SDS–PAGE. The bands were transferred to nitrocellulose membrane at 0.8 mAÆcm 2 )1 for 2 h using a semidry blotting apparatus (Bio-Rad, Hercules, CA, USA). The membrane was probed with horseradish peroxidase- conjugated anti-His serum (Qiagen, Valencia, CA, USA) or polyclonal antibody against purified recombinant M. tuber- culosis CmtR protein raised in rabbit and processed accord- ing to the manufacturer’s protocol (Qiagen) or as described previously [30]. A crosslinking experiment with purified His-CmtR was performed in the presence of 0.1% formal- dehyde in standard phosphate-buffered saline [29] for 30 min at 25 °C. Crosslinking was terminated by the addi- tion of SDS sample buffer (with 7.5% b-mercaptoethanol), and the products were resolved by 10% SDS–PAGE before transfer to nitrocellulose membrane as described above. The membrane was probed with horseradish peroxidase- conjugated anti-His serum (Qiagen) and developed using 3’,3’-diaminobenzidine. Electromobility shift assay and DNase I footprinting EMSA and DNase I footprinting were performed with purified CmtR (His tag removed). For EMSA, radiolabeled DNA fragments were generated by PCR using appropriate primers (Table 2 and Fig. 5A), one of which was end- labeled using c- 32 P ATP (approximately 3000 CiÆmmol )1 , Board of Radiation and Isotope Technology, Hyderabad, India). Binding of CmtR was performed in a 20 lL reac- tion where the protein was first incubated with metal ions for 10 min at room temperature (when required) and then with 32 P-labeled DNA (approximately 2 ng, approximately 15 000 cpm) or with double-stranded oligonucleotides for 30 min on ice in binding buffer [25 m m Tris ⁄ HCl pH 8.0, 6mm MgCl 2 , 5% glycerol, 0.02 mm dithiothreitol and 1 lg of poly(dI-dC)]. The reaction was electrophoresed on a 5% non-denaturing gel at 120 V (constant) in 0.5· Tris ⁄ borate buffer at 4 °C after pre-running the gel for 30 min under similar conditions. The gel was dried and analyzed by phosphor imaging using Quantity One software (Bio-Rad). To make double-stranded oligonucleotides, single-stranded oligonucleotides (Table 2) were annealed by incubating at 95 °C for 3 min in buffer containing 10 mm Tris ⁄ HCl and 100 mm NaCl, and allowed to cool slowly to 4 °C. The DNA was visualized by ethidium bromide staining. The DNase I footprinting assay was performed as described previously [26]. The binding and running buffers used were the same as in EMSA. DNA–protein interac- tion was performed as described above with approximately 150 000 cpm of labeled DNA, in a reaction volume of 50 lL. DNase I treatment with 0.2 units was performed for 3 min at 22 °C in the presence of 50 lL cofactor solu- tion (2.5 MgCl 2 and 5 mm CaCl 2 ), and the reaction was stopped by the addition of 90 lL stop solution (200 mm NaCl, 30 mm EDTA, 1% SDS and 66 lgÆ mL )1 yeast tRNA). The reaction products were phenol ⁄ chloroform- extracted, ethanol-precipitated, washed with 70% ethanol Table 2. List of primers ⁄ oligonucleotides used in the study. Primers Sequence (5¢-to3¢) Experiment P1 GTACTATTGGATCCATGCTGACG CmtR protein expression P2 GTCCTGTAAGCTTAAGTCGTGTC CmtR protein expression P3 TTCCCGCATCTCACACGTCA Reporter assay P4 CATATCTGCTATGGATGTAC EMSA P6 GTCACACCTTTCGTCGCAGC Reporter assay, EMSA, DNase I footprinting P8 TGTTATACCAGTATATGGTGTACTA EMSA 94RTf CTCGGCCTCAACTACAGTCGT Reverse transcription 94RTr ACAGGTAGCTGAGCAGCAGAC Reverse transcription 1994intR CAGCTAGCTGGCCGGGATAGC EMSA, DNase I footprinting, TSP mapping P1F GCCGATCATATCTGCTATGG EMSA, oligonucleotides for site 1 P1R CCATAGCAGATATGATCGGC P2F ATGTACAATTCAGCTCTTGCT EMSA, oligonucleotides for site 2 P2R AGCAAGAGCTGAATTGTACAT P3F GCTGTTATACCAGTATATGG EMSA, oligonucleotides for site 3 P3R CCATAT ACTGGTATAACAGC P4F TGGTGTACTAATTTGATCTATG EMSA, oligonucleotides for site 4 P4R CATAGATCAAATAGTACACCA H1 CGAGTCGACCGGAGGACCTTT GGCCCTGCGTCGACCGA EMSA, oligonucleotides used in control experiment H2 TCGGTCGACGCAGGGCCAAAG GTCCTCCGGTCGACTCG S. Chauhan et al. Autoregulation by CmtR FEBS Journal 276 (2009) 3428–3439 ª 2009 The Authors Journal compilation ª 2009 FEBS 3437 [...]...Autoregulation by CmtR S Chauhan et al and air-dried DNA was dissolved in formamide ⁄ urea loading dye, loaded on a 6% denaturing polyacrylamide gel alongside a DNA sequencing ladder generated using the primer used in the DNase I footprinting reaction, and run at 70 W The gel was dried and visualized by phosphor imaging (Bio-Rad) RNA isolation, RT-PCR and primer extension RNA was isolated from M tuberculosis. .. Schnappinger D, Ehrt S, Voskuil MI, Liu Y, Mangan JA, Monahan IM, Dolganov G, Efron B, Butcher PD, Nathan C et al (2003) Transcriptional adaptation of Mycobacterium tuberculosis within macrophages: insights into the phagosomal environment J Exp Med 198, 693–704 15 Grasseschi RM, Ramaswamy RB, Levine DJ, Klaassen CD & Wesselius LJ (2003) Cadmium accumulation and detoxification by alveolar macrophages... CamR family implicated in ethionamide resistance in mycobacteria, octamerizes cooperatively on its operator Mol Microbiol 51, 175–188 19 Kimsey HH & Waldor MK (2004) The CTXphi repressor RstR binds DNA cooperatively to form tetrameric repressor operator complexes J Biol Chem 279, 2640– 2647 20 Singh VK, Xiong A, Usgaard TR, Chakrabarti S, Deora R, Misra TK & Jayaswal RK (1999) ZntR is an autoregulatory... factors and global gene regulation in Mycobacterium tuberculosis J Bacteriol 186, 895–902 Sambrook J & Russell DW (2001) Molecular Cloning: A Laboratory Manual, 3rd edn Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY Saini DK, Malhotra V, Dey D, Pant N, Das TK & Tyagi JS (2004) DevR–DevS is a bona fide twocomponent system of Mycobacterium tuberculosis that is hypoxia-responsive in the absence... Commission, India A K was supported by grants from the Department of Biotechnology (DBT), under the Ministry of Science and Technology, India We thank Dr L Ramakrishnan (Department of Microbiology, University of Washington, Seattle, WA, USA) for the generous gift of plasmid pFPV27 M tuberculosis H37Rv DNA was obtained from the TB Research Materials and Vaccine Testing program of the US National Institute of Allergy... Authors Journal compilation ª 2009 FEBS S Chauhan et al 9 Cavet JS, Meng W, Pennella MA, Appelhoff RJ, Giedroc DP & Robinson NJ (2002) A nickel–cobalt-sensing ArsR–SmtB family repressor Contributions of cytosol and effector binding sites to metal selectivity J Biol Chem 277, 38441–38448 10 Cavet JS, Graham AI, Meng W & Robinson NJ (2003) A cadmium–lead-sensing ArsR–SmtB repressor with novel sensory sites. .. Complementary metal discrimination by NmtR and CmtR in a common cytosol J Biol Chem 278, 44560–44566 11 Wang Y, Hemmingsen L & Giedroc DP (2005) Structural and functional characterization of Mycobacterium tuberculosis CmtR, a PbII ⁄ CdII-sensing SmtB ⁄ ArsR metalloregulatory repressor Biochemistry 44, 8976–8988 12 Banci L, Bertini I, Cantini F, Ciofi-Baffoni S, Cavet JS, Dennison C, Graham AI, Harvie DR... Allergy and Infectious Diseases (Grant A1 -75320) The technical assistance of Mr Shailendra Kumar is acknowledged References 1 Brown NL, Stoyanov JV, Kidd SP & Hobman JL (2003) The MerR family of transcriptional regulators FEMS Microbiol Rev 27, 145–163 2 Busenlehner LS, Pennella MA & Giedroc DP (2003) The SmtB ⁄ ArsR family of metalloregulatory transcriptional repressors: structural insights into prokaryotic... absence of the DNA-binding domain of DevR Microbiology 150, 865–875 Bagchi G, Chauhan S, Sharma D & Tyagi JS (2005) Transcription and autoregulation of the Rv3134c– devR–devS operon of Mycobacterium tuberculosis Microbiology 151, 4045–4053 Valdivia RH, Hromockyj AE, Monack D, Ramakrishnan L & Falkow S (1996) Applications for green fluorescent protein (GFP) in the study of host–pathogen interactions Gene... log-phase cultures grown in Dubos medium under aerobic conditions as described previously [31] For RT-PCR, 0.5 lg M tuberculosis RNA was used as a template for synthesis of cDNA using a RevertAid firststrand cDNA synthesis kit and random primers (MBI Fermentas, St Leon-Rot, Germany) according to the manufacturer’s instructions The cDNA was used to amplify a product encompassing the entire Rv1993c gene and . CmtR, a cadmium-sensing ArsR–SmtB repressor, cooperatively interacts with multiple operator sites to autorepress its transcription in Mycobacterium tuberculosis Santosh. TTCCCGCATCTCACACGTCA Reporter assay P4 CATATCTGCTATGGATGTAC EMSA P6 GTCACACCTTTCGTCGCAGC Reporter assay, EMSA, DNase I footprinting P8 TGTTATACCAGTATATGGTGTACTA