The Mycobacterium tuberculosis phoPR Operon Is PositivelyAutoregulated in the Virulent Strain H37Rv�†
Jesus Gonzalo-Asensio,1,6 Carlos Y. Soto,2 Ainhoa Arbues,1,6 Javier Sancho,3 María del Carmen Menendez,4María J. García,4 Brigitte Gicquel,5 and Carlos Martín1,6*
Departamento de Microbiología, Medicina Preventiva y Salud Publica, Universidad de Zaragoza, Zaragoza, Spain1; Departamento deQuímica, Facultad de Ciencias, Universidad Nacional de Colombia, Bogota, Colombia2; Departamento de Bioquímica y
Biología Molecular y Celular, Facultad de Ciencias, and Institute for Biocomputation and Physics of Complex Systems (BIFI),Universidad de Zaragoza, Zaragoza, Spain3; Departamento de Medicina Preventiva, Universidad Autonoma de
The attenuated Mycobacterium tuberculosis H37Ra strain is an isogenic counterpart of the virulent paradigmstrain H37Rv. Recently, a link between a point mutation in the PhoP transcriptional regulator and avirulence ofH37Ra was established. Remarkably, a previous study demonstrated negative autoregulation of the phoP gene inH37Ra. These findings led us to study the transcriptional autoregulation of PhoP in the virulent H37Rv strain. Incontrast to the negative autoregulation of PhoP previously published for H37Ra, our experiments using a phoPpromoter-lacZ fusion showed that PhoP is positively autoregulated in both H37Rv and H37Ra compared with anH37Rv phoP deletion mutant constructed in this study. Using quantitative reverse transcription-PCR (RT-PCR)analysis, we showed that the phoP gene is transcribed at similar levels in H37Rv and H37Ra. Gel mobility shift andDNase I footprinting assays allowed us to identify the precise binding region of PhoP from H37Rv to the phoPpromoter. We also carried out RT-PCR studies to demonstrate that phoP is transcribed together with the adjacentgene phoR, which codes for the cognate histidine kinase of the phoPR two-component system. In addition, quanti-tative RT-PCR studies showed that phoR is independently transcribed from a promoter possibly regulated by PhoP.Finally, we discuss the possible role in virulence of a single point mutation found in the phoP gene from theattenuated H37Ra strain but not in virulent members of the M. tuberculosis complex.
The role of the phoP gene in Mycobacterium tuberculosisvirulence has been characterized extensively in various in vitrocultures and animal models. The inactivation of phoP in a fullyvirulent M. tuberculosis clinical isolate results in impairedgrowth in cultured macrophages and no bacillary multiplica-tion in mouse organs (32). Vaccination with the phoP mutanthas been shown to protect mice and guinea pigs against tuber-culosis (1, 27), suggesting that attenuated phoP mutants couldpotentially be used as live antituberculosis vaccine candidates(2).
Both the attenuated phenotype and protective efficacyagainst tuberculosis of the phoP mutant can be accounted forby the mechanism of action of PhoP, the response regulator(RR) of the two-component system (2CS) PhoPR. 2CSs aresignal transduction pathways enabling bacteria to detect andrespond to environmental stimuli, resulting in adaptation.Generally, 2CSs consist of two proteins, a membrane-associ-ated histidine kinase (HK) which senses an environmentalsignal(s) and an RR which is phosphorylated by its cognateHK. These changes in the phosphorylation state of the RR
result in transcriptional adaptation (14, 33). The M. tuberculo-sis genome encodes only 11 2CSs (11, 23), far fewer than thenumbers found in many other bacteria. The small number of2CSs present could be the result of M. tuberculosis adaptationto an intracellular lifestyle. Of the 11 2CSs that M. tuberculosispossesses, the phoPR 2CS has been demonstrated to be essen-tial for the virulence of the tubercle bacillus (32).
Biochemical analyses of M. tuberculosis phoP mutants havedemonstrated that PhoP positively regulates the biosynthesisof virulence-associated lipids, such as sulfolipid (SL), diacyl-trehaloses (DAT), and polyacyltrehaloses (PAT) (19). Consis-tent with these findings, the PhoP regulon identified in the M.tuberculosis H37Rv strain contains a number of genes involvedin the synthesis of these complex lipids along with severalgenes implicated in virulence (49).
The relationship between virulence and lipid compositionwas demonstrated in analyses comparing the virulent H37Rvstrain with the attenuated H37Ra strain, which, similar to M.tuberculosis phoP mutants, has been shown to lack SL, DAT,and PAT (8, 17). Conversely, a recent study identified a pointmutation in PhoP from H37Ra with respect to that fromH37Rv resulting in the amino acid substitution Ser219Leu,which probably affects the functionality of the DNA-bindingdomain. It was clearly demonstrated that this point mutation isresponsible for the absence of SL, DAT, and PAT in H37Ra(10). This finding may explain some striking similarities be-tween H37Ra and M. tuberculosis phoP mutants, as many of thephenotypes of the H37Ra strain, including virulence attenua-tion (9, 37), formation of smaller colonies on agar plates, loss
* Corresponding author. Mailing address: Departamento de Micro-biología, Medicina Preventiva y Salud Publica, Facultad de Medicina,Universidad de Zaragoza, C/Domingo Miral s/n, 50009 Zaragoza,Spain. Phone: 34 976 76 17 59. Fax: 34 976 76 16 64. E-mail: [email protected].
† Supplemental material for this article may be found at http://jb.asm.org/.
of acid fastness (44), lack of reactivity with neutral red (13, 29,41), and cording defects (17, 30), are correlated with the phe-notypes of phoP mutants constructed from virulent M. tuber-culosis strains (19, 32, 49).
Another remarkable example of the impact of the phoPmutation is the lack of secretion of the major T-cell antigenESAT-6 and its binding partner, the 10-kDa culture filtrateprotein CFP-10, in both H37Ra and an M. tuberculosis phoPmutant (16). The study established an interesting link betweenPhoP and ESAT-6/CFP-10 secretion, providing important in-formation about PhoP virulence regulation in M. tuberculosis.
In a previous work, Gupta et al. demonstrated transcrip-tional autorepression of PhoP due to a sequence-specific in-teraction with its promoter when they used the H37Ra versionof PhoP (21). Here we study many aspects of PhoP autoregu-lation in the virulent strain H37Rv, such as transcriptionalorganization, expression, and DNA-binding properties. Ourresults indicate that PhoP is positively autoregulated in theH37Rv and H37Ra strains of M. tuberculosis. We also describethe organization of the phoPR operon in H37Rv and addressthe importance of the PhoP mutation in the loss of virulence ofH37Ra.
MATERIALS AND METHODS
Bacterial strains, media, and growth conditions. M. tuberculosis H37Rv (11),H37Ra (ATCC 25177), and MT103 (32), Mycobacterium smegmatis mc2155 (40),and their derivatives were cultured at 37°C in Middlebrook 7H9 medium sup-
plemented with ADC (0.5% bovine serum albumin, 0.2% dextrose, 0.085%NaCl, 0.0003% beef catalase) (Difco) and 0.05% Tween 80. Escherichia coliBL21(DE3)/pLysS and its derivatives were grown in Luria-Bertani (LB) mediumat the appropriate temperature. When required, ampicillin (100 �g/ml), kana-mycin (20 �g/ml), hygromycin (20 �g/ml), 5-bromo-4-chloro-3-indolyl-�-D-galac-topyranoside (X-Gal; 50 �g/ml), and isopropyl-�-D-thiogalactopyranoside(IPTG; 1 mM) were used.
Construction of plasmids. The vectors and oligonucleotides used are shown inTable 1. For the overproduction of PhoP in E. coli, the phoP gene was amplifiedfrom M. tuberculosis genomic DNA by PCR, using the primers PhoPexp andPhoPR, and the PCR fragment was inserted into the pGEM-T Easy vector toobtain pMT1. The insert was then excised by digestion with NdeI and XhoI andligated into pET15b to give pTEX1. This recombinant plasmid contains thecodons for six histidine residues at the 5� end of phoP and therefore generates aprotein with a His6 motif at the N terminus (His6-PhoP). The insert in pTEX1was sequenced and confirmed to be identical to the phoP gene from the H37Rvstrain. To construct the phoP promoter-lacZ fusion plasmid, a 238-bp fragmentfrom positions �226 to �14 with respect to the phoP start codon was PCRamplified from M. tuberculosis MT103 genomic DNA by use of the primersBCG2B and PhoPro. The PCR product was cloned into pGEM-T Easy to obtainpTBLAC. The phoP promoter region was released from pTBLAC by digestionwith ApaI and KpnI and then inserted into the pJEM14 E. coli-mycobacterialshuttle plasmid to give pTBGAL. The sequence of the phoP promoter cloned inpTBGAL was identical to that in the H37Rv strain.
Quantification of �-galactosidase activity. Mycobacterial strains grown tostationary phase (optical density at 600 nm [OD600] � 2) were used to inoculate50 ml of fresh medium at a ratio of 1:50. Aliquots were taken from this cultureat the indicated times and centrifuged, and the pellet was washed with phos-phate-buffered saline and stored at �80°C. Cells were resuspended in 1 ml Zbuffer (0.06 M Na2HPO4, 0.04 M NaH2PO4, 0.01 M KCl, 1 mM MgSO4, and 0.05M �-mercaptoethanol [pH 7 at 25°C]) and sonicated to generate a cell extract.The chromogenic substrate o-nitrophenyl-D-galactoside was added to cell ex-
TABLE 1. Plasmids and primers used in this study
Plasmid or primer Description or sequence (5�–3� �position with respect to phoP start codon�) Reference
PlasmidspGEM-T Easy E. coli cloning vector for PCR fragments; Apr PromegapMT1 phoP gene in pGEM-T Easy vector This studypET15b E. coli expression vector; Apr NovagenpTEX1 phoP gene in pET15b This studypNBV1 E. coli-mycobacterial shuttle plasmid; Hygr 24pSO5 phoP gene in pNBV1 32pSO5K pSO5 derivative carrying a Kmr marker 19pJUZ1 phoPR genes in pNBV1 19pTBLAC phoP promoter in pGEM-T Easy vector This studypJEM14 E. coli-mycobacterial shuttle vector for promoter-lacZ fusions; Kmr 47pTBGAL phoP promoter in pJEM14 This study
PrimersPhoPexp CCCATATGCGGAAAGGGGTTGATCTCG (�1) This studySepho8 GTTCCTTGTTGCCCTTGCCC (�436) This studyPhoInt GTCGTATCTGCGCCGCAAG (�654) This studyPhoPF AATCTAGATCAAGCATCAGCC (�1000) 19PhoPR AATCTAGAGATCACCCGAACGTAGAACC (�1046) 19Sepho3 CGGCTTTCGTTGGCGCTCA (�1437) This studySepho6 ACGAGATTGCGCAGCACCT (�1900) This studyB1 GGCAACGGTCCAAGCTGA (�105) This studyPhoPro GGGGTACCCCTTTCCGCATTGGTTG (�14) This studyBCG2A GCCGTCCATCCCGGGCATC (�237) 42BCG2B CCATGTTCAAACCGGTGTC (�226) 42tsp2fw GGCAACGAGCTTTCAGGA (�52) This study
Primers for qRT-PCRRT phoP fw GCCTCAAGTTCCAGGGCTTT This studyRT phoP rv CCGGGCCCGATCCA This studyRT phoR fw AACGGAATGCTGGCACAAA This studyRT phoR rv GCCTTTTCGGCGGAAGAT This studyRT sigA fw CCGATGACGACGAGGAGATC This studyRT sigA rv CGGAGGCCTTGTCCTTTTC This study
VOL. 190, 2008 M. TUBERCULOSIS H37Rv PhoPR OPERON 7069
tracts at a final concentration of 0.66 mg/ml. The mixtures were incubated at28°C for 1 h, and the enzymatic reaction was stopped by adding 0.29 M Na2CO3.The A420 of the supernatant was determined, and �-galactosidase activity wascalculated in Miller units, using the following formula: �-galactosidase activity (1,000 A420)/(time [min] aliquot volume [ml] A600).
Overproduction and purification of PhoP. The PhoP protein from H37Rv wasoverproduced from pTEX1 in E. coli BL21(DE3)/pLysS and purified using aHisTrap affinity column (Amersham Pharmacia Biotech) as previously described(10).
EMSA. DNA probes for electrophoretic mobility shift assays (EMSAs) wereobtained by PCR amplification from M. tuberculosis MT103 genomic DNA.Amplification products were purified with a GFX PCR DNA purification kit(Amersham Biosciences). Purified DNA fragments were 3� end labeled with abiotin 3�-end-labeling kit (Pierce) for use in gel shift assays. In each 20-�l bindingreaction mix, 0.02 pmol of labeled fragment was incubated for 30 min at roomtemperature with 100 ng poly(dI-dC) and increasing amounts of PhoP in bindingbuffer (50 mM HEPES [pH 7.5], 50 mM KCl, 1 mM dithiothreitol, and 5%glycerol). Samples were loaded on a 6% nondenaturing polyacrylamide gelcontaining 1 Tris-borate-EDTA buffer (36). Electrophoresis was performed at100 V and 7 mA at 4°C until the bromophenol blue reached the bottom of thegel. The DNA-protein complexes were electroblotted onto a positively chargedHybond-N� nylon membrane (Amersham Pharmacia Biotech) and detected byenhanced chemiluminescence, using a Lightshift chemiluminescence EMSA kit(Pierce) and Hyperfilm ECL (Amersham Pharmacia Biotech).
DNase I footprinting. DNase I footprinting assays were performed by thefluorescence labeling procedure (34). The fluorescently labeled DNA was am-plified by PCR, using pJUZ1 (19) as a template, with the Cy5-labeled BCG2Bprimer (Sigma) and unlabeled BCG2A primer for the phoP coding strand andthe unlabeled BCG2B primer and Cy5-labeled BCG2A primer for the noncodingstrand. In each case, the same labeled primer was used to prime the sequencingreaction for the determination of molecular size. Labeled DNA fragment (100ng) and 1 �g of PhoP were included in each 20-�l binding reaction mix, asdescribed above for EMSA. The mixture was incubated for 30 min at roomtemperature, and DNase I (0.1 unit; Roche) was then added and allowed to actfor 3 min at 30°C. The nuclease digestion was stopped by adding 180 �l of stopsolution (10 mM Tris-HCl [pH 8], 40 mM EDTA). After chloroform extractionand ethanol precipitation, samples were loaded and analyzed on an ALF se-quencer, as previously described (7).
Isolation of RNA from mycobacteria. The M. tuberculosis wild-type, phoPmutant, and complemented strains were grown at 37°C until they reached thedesired OD600 under aerobic conditions. Cells were harvested and total RNAswere isolated using a Fast RNA Pro Blue kit (Qbiogene) according to themanufacturer’s recommendations. The extracted RNAs were treated withRNase-free DNase (Ambion), and the RNAs were then further purified using anRNeasy kit (Qiagen). The integrity of the RNAs was checked by gel electro-phoresis on a 1% agarose gel.
RT-PCR. Reverse transcription-PCR (RT-PCR) was carried out in two steps.RT was carried out with Expand reverse transcriptase (Roche), using 1 �g RNAas the template and the appropriate reverse primer. Reaction mixtures wereincubated at 43°C for 90 min. RT products were then subjected to PCR ampli-fication, using TaqGold DNA polymerase (Roche) and the appropriate primers.PCR amplification involved initial denaturation for 10 min at 94°C, followed by40 cycles of 94°C for 30 s, 59°C for 30 s, and 72°C for 3 min. Samples wereanalyzed by electrophoresis on a 1% agarose gel.
qRT-PCR. cDNA libraries from M. tuberculosis wild-type, mutant, and com-plemented strains were constructed as follows. One microgram of RNA wasmixed with 25 pmol of random hexanucleotide primers (Sigma) and 50 units ofExpand reverse transcriptase (Roche) in a final volume of 20 �l. Reactionmixtures were incubated at 30°C for 10 min and then at 43°C for 90 min.Expression of phoP and phoR mRNAs was measured and normalized withrespect to the level of sigA mRNA by quantitative real-time RT-PCR (qRT-PCR). qRT-PCR was carried out in a StepOne Plus (Applied Biosystems) in-strument, using the cDNA generated by RT from 25 ng of RNA as a template,1 Power SYBR green PCR master mix (Applied Biosystems), and the appro-priate primers (Table 1), each at a concentration of 250 nM. The PCR programinvolved an initial denaturation step for 10 min at 95°C, followed by 40 cycles of95°C for 15 s and 60°C for 1 min. The specificity of the PCR products wasconfirmed by the loss of fluorescence at a single temperature, when the double-stranded DNA melted to single-stranded DNA.
Bioinformatic analyses and homology modeling. The full genome sequences ofMycobacterium bovis strains BCG and AF2122/97 and M. tuberculosis strainsHaarlem, CDC1551, F11, C, H37Rv, and H37Ra were obtained from the NCBIwebsite (http://www.ncbi.nlm.nih.gov/sites/entrez). We also sequenced the cod-
ing and promoter regions of the phoP gene from M. tuberculosis MT103 andincluded them in the sequence alignment. The ExPASy proteomic server (http://au.expasy.org/) was used to predict the organization of the PhoP domain.
From the structure of the DNA-binding domain of PhoP from M. tuberculosis(50) (Protein Data Bank [PDB] code 2pmu), the Ser219Leu mutation was mod-eled using Deepview (20). On the other hand, superpositioning of the wild-typeand mutant PhoP structures was done with Deepview on the structure of thecomplex between the structurally homologous effector domain of E. coli PhoBand Pho box DNA (4) (PDB code 1gxp).
RESULTS
phoP promoter activity in the presence of PhoP from M.tuberculosis H37Rv. It has been reported that PhoP from theH37Ra strain represses its own transcription if M. smegmatis isused as a surrogate host (21). We investigated whether thisautoregulation was affected by the Ser219Leu mutation inPhoP by carrying out a similar experiment with PhoP from theH37Rv strain. A transcriptional fusion between the phoP pro-moter and a promoterless lacZ gene was constructed and in-serted into pJEM14 (47), an E. coli-mycobacterial shuttle plas-mid, yielding pTBGAL (Table 1). This plasmid was used totransform M. smegmatis mc2155. The resulting strain was co-transformed with pSO5 (32), which contains both the promoterregion and the entire phoP coding region in the multicopyplasmid pNBV1 (24). The cotransformed strain was grown at37°C in 7H9 medium supplemented with hygromycin and kana-mycin. Aliquots were collected at the indicated times, and�-galactosidase activities in the cell extract were evaluated todetermine levels of transcription from the phoP promoter inthe presence of PhoP. �-Galactosidase activity levels were twoto three times higher in M. smegmatis mc2155 cotransformedwith pTBGAL and pSO5 than in the control strain cotrans-formed with pTBGAL and pNBV1 (Fig. 1). Thus, phoP ex-pression seems to be positively autoregulated by the H37Rvversion of PhoP when M. smegmatis is used as a surrogate host.
Positive autoregulation of PhoP in M. tuberculosis H37Rvand H37Ra. In order to corroborate the previous finding andto rule out differences between the transcriptional machineriesof the fast-growing M. smegmatis strain and the slow-growingM. tuberculosis strain, we attempted to study the expression ofthe phoP promoter in H37Rv, H37Ra, and a defective phoPmutant. Firstly, we constructed and characterized an H37Rv�phoP::hyg mutant (see the supplemental material). M. tuber-culosis H37Rv, H37Ra, and H37Rv �phoP::hyg were trans-formed with either pJEM14 or pTBGAL and grown at 37°C in7H9 medium supplemented with kanamycin. Aliquots werecollected along the logarithmic growth phase (OD600 � 0.3,0.6, and 1), and �-galactosidase activities in the cell extractwere calculated. Our results indicate that the lacZ gene istranscribed from the phoP promoter at similar levels in H37Rvand H37Ra (Fig. 2). Moreover, we also observed that tran-scription of phoP is about twofold higher in H37Rv and H37Rathan in the H37Rv defective phoP mutant at early- and late-logarithmic growth phases, while no significant differences ex-ist at mid-log phase (Fig. 2). In addition, the H37Rv andH37Ra strains transformed with either pJEM14 or pTBGALwere plated on 7H10–oleic acid-albumin-dextrose-catalaseagar containing the �-galactosidase substrate X-Gal. It wasobserved that H37Rv and H37Ra transformed with pTBGALdeveloped blue coloration, while the pJEM14 transformants
remained uncolored in the presence of X-Gal (data notshown). Taken together, these results indicate that phoP couldbe positively autoregulated in both M. tuberculosis H37Rv andH37Ra, and this effect was more noticeable at the time pointsdescribed above.
EMSA with the phoP promoter and PhoP from H37Rv. Pre-vious works have demonstrated that PhoP from H37Ra bindsto its own promoter (21, 38). Here we studied whether PhoPfrom M. tuberculosis H37Rv specifically binds to its own pro-moter sequence, possibly mediating autoregulation of phoPexpression. The same 238-bp fragment from the phoP pro-moter previously used in �-galactosidase experiments was usedfor gel shift assays. This fragment was amplified by PCR usingthe BCG2B and PhoPro primers (Table 1) and then labeled foruse as a probe in mobility shift experiments. The labeled probedisplayed a clear shift in electrophoretic mobility in the pres-ence of PhoP, and larger amounts of protein resulted in agradual decrease in probe mobility (Fig. 3, lanes 2 to 6). Weconfirmed the specificity of binding by using an unlabeledfragment as a competitor. PhoP binding to the labeled phoPpromoter was reduced by adding a 100-fold excess of unlabeledprobe (Fig. 3, lane 7). Further support for the specificity of theinteraction with DNA was provided by experiments in which a
nonspecific fragment corresponding to the sigA coding se-quence was used, and as expected, no shift in DNA mobilitywas detected (data not shown). Similar experiments were per-formed in the presence of PhoP previously incubated with thephosphorylation reagent acetyl phosphate. Even if phosphory-lation of PhoP did not appear to affect binding specificity, wecannot rule out differences in binding affinity due to phosphor-ylation (data not shown).
DNase I protection of the phoP promoter with PhoP. Weattempted to compare the PhoP binding region in H37Rv withthat previously reported for H37Ra (21). The sequence towhich PhoP binds was mapped by DNase I footprinting assayswith a 463-bp fragment from positions �226 to �237 relativeto the phoP start codon. The regions protected by PhoP fromdegradation by DNase I were located between positions �66and �18 on the noncoding strand and �38 and �1 on thecoding strand, resulting in a 67-bp binding region, from posi-tions �66 to �1 (Fig. 4). The degree of protection was higherfor the coding strand than for the noncoding strand (Fig. 4).Similar results were obtained in other footprinting assays usingradiolabeled primers (data not shown). Our results indicatethat the PhoP binding site described here, although similar, isshorter to that previously described for the H37Ra strain,
FIG. 1. phoP promoter activity in M. smegmatis expressing the H37Rv variant of PhoP. The phoP promoter-lacZ fusion plasmid pTBGAL wasintroduced into M. smegmatis carrying either a PhoP expression vector (pSO5) or the expression plasmid lacking an insert (pNBV1) and culturedfor the indicated times. White bars indicate �-galactosidase activity of M. smegmatis carrying pTBGAL and pNBV1, and black bars represent theresults for the strain cotransformed with pTBGAL and pSO5. Results shown are the means and standard errors from two independentexperiments.
FIG. 2. phoP promoter activity in H37Rv and H37Ra. White, black, and hatched bars represent �-galactosidase activities from the H37Rv�phoP::hyg, H37Rv, and H37Ra strains transformed with pTBGAL and cultured until the indicated OD600. Miller units for the aforementionedstrains transformed with the empty vector (pJEM14) were below 10 and have been omitted for clarity. Results shown are the means and standarderrors from two independent experiments.
VOL. 190, 2008 M. TUBERCULOSIS H37Rv PhoPR OPERON 7071
which is located between positions �79 and �9 relative to thephoP start codon (21).
Transcriptional organization of the M. tuberculosis phoPRoperon. In the M. tuberculosis H37Rv chromosome, the phoPgene maps upstream of the phoR gene, coding for the HK ofthe 2CS PhoPR, and the stop codon of phoP is separated fromthe start codon of phoR by a 47-bp intergenic region (Fig. 5),suggesting that both genes may be transcribed in an operon.RT-PCR was used to determine whether phoP and phoR weretranscribed to give a single mRNA. We obtained amplificationproducts for the 5� end of phoP and the 3� end of phoR and a393-bp amplification product for the phoPR intergenic region(Fig. 5), indicating that these two genes are cotranscribed. Forconfirmation of the cotranscription of phoPR, a larger region,from positions �52 to �1900 with respect to the phoP startcodon, was amplified and the expected cotranscribed fragmentobtained (Fig. 5). We checked that the RNA samples were notcontaminated with genomic DNA by carrying out RT-PCRswithout reverse transcriptase. The lack of amplification prod-ucts for all of the fragments described above demonstratesclearly that the RT-PCR products obtained were amplifiedfrom RNA that had been reverse transcribed into cDNA (Fig.5). Our results indicate that the two genes are cotranscribed inan operon, as described for other 2CSs (22), but these resultsdo not exclude the possibility of independent promoters forphoP and phoR.
Quantification of phoP and phoR expression by qRT-PCR.M. tuberculosis H37Rv and H37Ra were cultured to mid-expo-nential growth phase. RNAs were extracted and reverse tran-scribed to generate cDNAs, which were then amplified in qRT-PCR experiments. The expression of the phoP gene wasmeasured by qRT-PCR and normalized to sigA expressionlevels. Transcription rates for the phoP gene were similar inH37Rv and H37Ra (Fig. 6A), in concordance with our previ-ous results from the �-galactosidase activity assays. Once wedemonstrated that phoP is similarly transcribed in H37Rv andH37Ra, we sought to compare phoR expression levels in bothstrains. qRT-PCR experiments using phoR-specific primers
(Table 1) indicated that phoR is transcribed about 10-foldmore in H37Rv than in H37Ra (Fig. 6A), suggesting that thephoP mutation could affect phoR transcription. To assess thispossibility, we compared phoR expression between an M. tu-berculosis clinical isolate MT103 wild-type strain and anMT103 defective phoP mutant (19). Similar to that observed inH37Ra, phoR expression was highly reduced in the phoP mu-tant of M. tuberculosis (Fig. 6B). To rule out the possibility thatphoR transcription was reduced as a consequence of a polareffect of the phoP mutation on the transcription of the entirephoPR operon, we studied phoR expression in the phoP mutantcomplemented with pSO5K (19), a multicopy plasmid carryingthe phoP gene from H37Rv. We observed that phoR transcrip-tion was restored in the complemented strain (Fig. 6B). More-over, the levels of phoR transcripts seemed to be correlateddirectly with PhoP production, since overexpression of PhoPfrom the multicopy plasmid resulted in a significant increase inthe level of phoR with respect to that in the wild-type strain(Fig. 6B). Altogether, these results indicate that the phoR genecould be transcribed from a PhoP-regulated promoter.
Characterization of the Ser219Leu mutation in PhoP. TheM. tuberculosis PhoPR system is highly conserved among slow-growing mycobacteria. Sequence alignment indicated full con-servation of the phoP gene in all eight M. tuberculosis complexgenomes annotated to date, with the remarkable exception ofthe H37Ra strain (Fig. 7A). The phoP gene from M. tubercu-losis H37Ra has acquired a point mutation in codon 219(TCG3TTG), resulting in the replacement of a serine by aleucine residue (Fig. 7A). Bioinformatic approaches based onsequence alignment led to the identification of two putativedistinct domains in PhoP. The PhoP N-terminal domain isinvolved in the phosphotransfer reaction through the con-served residue Asp71, whereas the C-terminal domain is in-volved in DNA binding (data not shown). The recent structureof the PhoP DNA-binding domain (50) situates the above-mentioned missense mutation in the DNA recognition helix(Fig. 7B), potentially affecting DNA interactions and conse-quently the role of this protein in transcriptional regulation.
A model of the three-dimensional structure of the PhoPSer219Leu mutant was constructed, and both the wild-type andmutant structures were superimposed on the structure of thePhoB-DNA complex from E. coli (Fig. 7B). The mutation islocated in the central part of the C-terminal �-helix (�3) of thethree-helix bundle of the effector domain (Fig. 7B). The helixis amphipathic, with the wild-type serine residue facing thesolvent. Replacement of a solvent-exposed serine residue by aleucine residue is unlikely to significantly reduce the confor-mational stability of the protein. On the other hand, the mu-tant leucine residue cannot easily favor any aggregation of theprotein because the region is highly charged. Apparently, theSer219Leu mutation should give rise to a protein with stabilityand solubility similar to those of the wild-type protein. Incontrast, the mutation is clearly expected to reduce the affinityand/or specificity of PhoP for its DNA-binding region. As thesuperposition of the model structure of PhoP with the PhoB-DNA complex indicates, the helix bearing the mutation (�3) isthe DNA recognition helix of a modified helix-turn-helixDNA-binding motif where loop �2-�3 replaces the turn (4).The mutation therefore takes places right at the DNA-bindingsite (Fig. 7B). Given the capability of serine residues to estab-
FIG. 3. Binding of PhoP to the phoP promoter. Electrophoreticmobilities of a 238-bp fragment containing the phoP promoter in theabsence (lane 1) and presence of increasing amounts of recombinantPhoP (180 �M [lane 2], 360 �M [lane 3], 720 �M [lane 4], 1.08 mM[lane 5], and 1.44 mM [lane 6]). Lane 7 represents the electrophoreticmobility of the phoP promoter in the presence of 360 �M PhoP and a100-fold excess of unlabeled probe.
FIG. 4. DNase I footprinting assay of the phoP promoter. (A) Fluorograms indicate fluorescence intensities after DNase I digestion offragments containing Cy5-labeled coding and noncoding strands in the absence (�PhoP) and presence (�PhoP) of recombinant PhoP. Sequencingreaction mixtures with Cy5-labeled primers were included in the gel (data not shown). Sites protected by PhoP are indicated by boxes. The ATGtranslation initiation triplet is indicated by asterisks. (B) PhoP binding region. The nucleotides to which PhoP binds are indicated in a box. TheATG translation initiation triplet is indicated by asterisks.
VOL. 190, 2008 M. TUBERCULOSIS H37Rv PhoPR OPERON 7073
lish hydrogen bonds, Ser219 could be important for the spec-ificity of the binding to DNA. But even if the contribution ofSer219 to specificity and/or binding affinity is small, its replace-ment by a bulky leucine residue is expected to interfere with atight protein-DNA interaction and therefore to reduce theaffinity of the complex.
DISCUSSION
In 1934, Steenken et al. observed that subculture of the H37strain by serial passages resulted in two different colony mor-phologies, with one of them being highly attenuated for guineapigs. The passages continued until two stable strains wereobtained, the virulent H37Rv strain and the attenuated H37Rastrain (28, 43–45). Since then, both strains have been usedextensively worldwide; indeed, H37Rv is the most widely usedvirulent M. tuberculosis laboratory reference strain today.
The H37Rv genome was the first mycobacterial genome tobe sequenced (11) and provided general insight into the biol-ogy of M. tuberculosis. However, the genetic basis for H37Raattenuation has remained unclear. H37Ra is likely to haveacquired multiple point mutations, deletions, and/or genomicrearrangements during in vitro passage. Somewhat analogouslyto the attenuation process for BCG, the current antitubercu-losis vaccine was derived by serial passage of M. bovis in thelaboratory over a period of 13 years (18), resulting in the lossof more than 100 genes (3).
Even if the presence of genomic variations between H37Rvand H37Ra has been confirmed, their role in H37Ra attenu-ation remains unclear (5). Pioneering studies in mycobacterialgenetics sought to restore virulence to the H37Ra strain by invivo complementation with an H37Rv cosmid library in anattempt to identify genomic fragments associated with viru-lence. Despite allowing the identification of a genomic frag-
ment which conferred an enhanced in vivo growth rate, thisstudy failed to completely restore virulence to the H37Rastrain (31).
The recent release of the H37Ra genome sequence availablefrom the NCBI website (NC_009525) should increase our un-derstanding of the mechanism of attenuation of H37Ra. Re-cent studies have identified a number of nucleotide polymor-phisms between H37Rv and H37Ra (16, 25), making it possibleto evaluate the role of discrete regions of the genome in M.tuberculosis virulence.
In this study, we focused on the phoP gene, which has beenshown to play an important role in virulence (19, 27, 32). PhoPis fully conserved in all of the annotated genomes of the M.tuberculosis complex except that of H37Ra. The phoP genefrom H37Ra contains a point mutation that results in thereplacement of the polar residue Ser219 by the nonpolar res-idue Leu in the DNA-binding domain of PhoP.
Apparently, this mutation should not compromise either thestability or the aggregation state of the protein. Indeed, circu-lar dichroism studies indicate that the point mutation does notaffect the global secondary structure of PhoP (10). However,homology modeling of both variants of PhoP clearly indicatesthat the replacement of the polar residue Ser by the nonpolarresidue Leu lies within the DNA recognition helix (50), andthis may result in a lower affinity for DNA binding as a con-sequence of both the impaired ability of the nonpolar residueLeu to establish hydrogen bonds with the bases of the DNAhelix and the steric difficulties imposed by the bulky Leu resi-due in establishing a tight PhoP-DNA interaction (Fig. 7B).
We found that PhoP from H37Rv binds to its own promoter(Fig. 3), like PhoP from H37Ra (21, 38). However, a recentwork demonstrated that PhoP from H37Rv, but not PhoP fromH37Ra, binds a 40-mer promoter region (10). In this study, we
FIG. 5. RT-PCR analysis of phoPR from M. tuberculosis. (A) Schematic diagram of M. tuberculosis phoPR and the gene-specific primers usedfor RT-PCR (Table 1). The direction of transcription for phoP and phoR is indicated by arrows. Primers used for RT-PCR and the sizes of thefragments obtained with each pair of primers are indicated. (B) RT-PCR of phoPR. The combination of primers is indicated above each set ofreactions. Amplification products for the 5� end of phoP (436 bp), the intergenic region (393 bp), and the 5� end of phoR (464 bp) are shown. Alarger fragment, from the phoP promoter region to the 3� end of phoR (1,952 bp), was also amplified. The positions of the standard DNA sizemarkers are indicated on the right. Each set of three reactions consists of a positive control PCR assay with genomic DNA as the template (�),an RT-PCR (�), and a negative control assay without reverse transcriptase (�).
used a 238-bp fragment containing the promoter region ofphoP to demonstrate that the point mutation in H37Ra ap-pears to diminish the DNA-binding affinity of PhoP (data notshown). The results presented in this work are in agreementwith a recent study which demonstrated that the H37Ra ver-sion of PhoP displays a decreased ability for DNA binding tothe PhoP binding motif (25).
The PhoP binding site in the phoP promoter extends fromnucleotides �79 to �9 relative to the phoP start codon inH37Ra (21) and from nucleotides �66 to �1 in the H37Rvstrain (Fig. 4). Even if the Ser219Leu substitution may beresponsible for these subtle variations, the differences in themethods used should also be taken into account.
PhoP negative autoregulation has been characterized usingthe protein from M. tuberculosis H37Ra (21). Here we carriedout similar experiments with PhoP from H37Rv and found thatunlike that reported for H37Ra, the phoP gene from H37Rv ispositively autoregulated with M. smegmatis as the expressionhost (Fig. 1). Furthermore, given that in M. tuberculosis phoP is
transcribed at similar levels in H37Rv and H37Ra, with tran-scription being reduced in the H37Rv phoP mutant at early-and late-logarithmic phases (Fig. 2A), we consider that phoP ispositively autoregulated in both H37Rv and H37Ra. We andothers (25) have demonstrated that the Ser219Leu mutation inH37Ra decreases the DNA-binding affinity of PhoP. However,the remaining DNA-binding activity may be sufficient to prop-erly autoregulate phoP expression. Consequently, this pointmutation might not affect the autoregulatory capacity of PhoP.
Transcription of the phoP promoter in the H37Rv phoP nullmutant indicates that in M. tuberculosis other transcriptionalregulators influencing phoP transcription, apart from PhoP,may exist. This is particularly observed at the mid-logarithmicphase of growth, wherein phoP is transcribed at similar levelsin the H37Rv phoP mutant, H37Rv, and H37Ra strains. It isalso possible that PhoP could be tightly autoregulated only inthe presence of an as yet unknown PhoR-stimulating signalthat would control PhoP phosphorylation and consequently theexpression of PhoP-regulated genes, including phoP itself.
Even though the Ser219Leu substitution does not appear toaffect the autoregulatory mechanism of PhoP, it seems to haveimportant implications in the virulence regulation of M. tuber-culosis. A recent study demonstrated that even if H37Ra andan M. tuberculosis phoP mutant produce the major antigenESAT-6/CFP-10, neither of these strains secretes the antigeninto the supernatant, resulting in decreased T-cell responsesagainst both ESAT-6 and CFP-10 (16). A plausible explanationcomes from the observation that the Rv3614c-to-Rv3616c(espA) gene cluster essential for ESAT-6/CFP-10 secretion (15,26) appears to be downregulated in both H37Ra and anH37Rv phoP mutant (16, 17, 49).
Results for the PhoP regulon in M. tuberculosis H37Rv andthe transcriptome comparison between H37Rv and H37Rafurther support the implications of the PhoP mutation inH37Ra avirulence. The Rv1184c, fadD21, and papA1 genes,encoding proteins involved in the biosynthesis of the virulence-associated lipids SL, DAT, and PAT (12, 39), are much lessexpressed in H37Ra than in H37Rv (17), which probably ac-counts for the absence of these lipids in H37Ra. Other genesdownregulated in H37Ra (Rv1639c, cdh, narK1, Rv2376c, nirA,fadD9, Rv3312A, Rv3479, lipF, Rv3686c, and Rv3822) (17)have also been shown to belong to the PhoP regulon identifiedin H37Rv (49). Altogether, these observations indicate that 14of the 22 genes differentially expressed between H37Rv andH37Ra are under the control of PhoP. This strongly suggeststhat the mutated version of PhoP contributes to M. tuberculosisH37Ra attenuation by leading to major global changes in thegene expression profile of this avirulent strain. Downregulationof PhoP-regulated genes in H37Ra could be a consequence ofthe decreased ability of PhoP for DNA binding and/or theinadequate phosphorylation of the protein as a result of de-creased expression of PhoR in H37Ra.
Despite these important contributions of the PhoP mutationto the attenuation process of H37Ra, complementation of thisstrain with the H37Rv version of PhoP only partially restoredthe virulence of H37Ra (16, 25). This is probably due to theexistence of other polymorphisms affecting the expression ofvirulence factors. Supporting this hypothesis, it has been foundthat the production of phthiocerol dimycocerosates—a familyof lipids implicated in virulence (6, 35)—is abrogated in the
FIG. 6. Quantification of phoP and phoR expression by qRT-PCR.(A) Quantification of phoP and phoR expression in H37Ra relative tothat in H37Rv. (B) Relative quantification of phoR expression inMT103 �phoP::hyg and the complemented strain MT103 �phoP::hyg-pSO5K with respect to phoR expression in the MT103 wild-type strain.Results are the means for two independent experiments; error barsindicate the standard deviations of the means.
VOL. 190, 2008 M. TUBERCULOSIS H37Rv PhoPR OPERON 7075
H37Ra strain and that their synthesis is independent of thePhoP mutation. Alternatively, given that phoR is expressedmuch less in H37Ra than in H37Rv, which may be a conse-quence of the phoP mutation in the former strain (Fig. 6),another plausible explanation for the partial virulence comple-mentation of H37Ra is that decreased expression of PhoR inthis strain could result in inadequate phosphorylation-medi-ated activation of PhoP. Accordingly, complementation ofH37Ra with the whole H37Rv phoPR operon could restorevirulence more proficiently than complementation with onlythe phoP gene.
Naturally occurring mutations causing attenuation of bacte-rial pathogens have already been described. Sequencing oflow-virulence field strains of Listeria monocytogenes recentlyshowed multiple point mutations affecting the virulence of thisintracellular pathogen (46). Remarkably, like the Ser219Leumutation in PhoP, the naturally occurring mutation Lys220Thrin the transcriptional regulator PrfA from L. monocytogenesresults in abrogated DNA binding, no expression of PrfA-regulated proteins, and attenuated virulence (48).
Our study also suggests that differences between M. tuber-culosis strains should be taken into account in genetic studies.
Continuous passages in synthetic laboratory media may wellresult in genetic polymorphisms and, consequently, in sub-strain variability. H37Rv is a highly pathogenic strain and canbe manipulated only in biosafety level 3 containment labora-tories. For this reason, many genetic studies have made use ofthe attenuated counterpart, H37Ra, based on the assumptionthat the results obtained could be extrapolated to all M. tuber-culosis complex strains. However, different works, includingthis one, clearly demonstrate the important implications of asingle mutation in virulence regulation.
ACKNOWLEDGMENTS
We thank Jordi Barbe for scientific advice about the nonradioactivefootprinting assay, Stewart Cole for kindly providing the H37Rv strain,Juan Carlos Ramírez for bioinformatic studies, and Alberto Cebolladafor statistical analysis.
This work was supported by the Spanish MEC (BIO2005-07949-C02-01), EU FP6 TB-VAC (LSHP-CT203-503367), INCO (ICA4-CT-2002-10063), BFU2007-61476/BMC, and PM076/2006 projects.
REFERENCES
1. Aguilar, D., E. Infante, C. Martin, E. Gormley, B. Gicquel, and R. HernandezPando. 2007. Immunological responses and protective immunity against tuber-
FIG. 7. Characterization of the point mutation in PhoP from H37Ra. (A) Domain organization of PhoP. The N-terminal domain of PhoP (blue)contains the Asp71 residue involved in the phosphotransfer reaction. The C-terminal domain (red) interacts with target DNA molecules,modulating gene expression. Alignment of the phoP gene sequences from the annotated genomes of the M. tuberculosis complex shows a pointmutation responsible for the Ser219Leu substitution in the H37Ra strain, indicated by an asterisk. (B) DNA-binding domain of PhoP (blue ribbon)(see text for details) superimposed on the structure of a PhoB-DNA complex (gray ribbon, protein; sticks in CPK color, DNA). Solid spheres showthe wild-type serine 219 residue (left) or the leucine residue that appears in the mutant protein (right). The mutant leucine residue is expectedto interfere with DNA binding and/or recognition.
culosis conferred by vaccination of Balb/C mice with the attenuated Mycobac-terium tuberculosis (phoP) SO2 strain. Clin. Exp. Immunol. 147:330–338.
2. Asensio, J. A., A. Arbues, E. Perez, B. Gicquel, and C. Martin. 2008. Livetuberculosis vaccines based on phoP mutants: a step towards clinical trials.Expert Opin. Biol. Ther. 8:201–211.
3. Behr, M. A., M. A. Wilson, W. P. Gill, H. Salamon, G. K. Schoolnik, S. Rane,and P. M. Small. 1999. Comparative genomics of BCG vaccines by whole-genome DNA microarray. Science 284:1520–1523.
4. Blanco, A. G., M. Sola, F. X. Gomis-Ruth, and M. Coll. 2002. Tandem DNArecognition by PhoB, a two-component signal transduction transcriptionalactivator. Structure 10:701–713.
5. Brosch, R., W. J. Philipp, E. Stavropoulos, M. J. Colston, S. T. Cole, andS. V. Gordon. 1999. Genomic analysis reveals variation between Mycobacte-rium tuberculosis H37Rv and the attenuated M. tuberculosis H37Ra strain.Infect. Immun. 67:5768–5774.
6. Camacho, L. R., D. Ensergueix, E. Perez, B. Gicquel, and C. Guilhot. 1999.Identification of a virulence gene cluster of Mycobacterium tuberculosis bysignature-tagged transposon mutagenesis. Mol. Microbiol. 34:257–267.
7. Campoy, S., M. Fontes, S. Padmanabhan, P. Cortes, M. Llagostera, and J.Barbe. 2003. LexA-independent DNA damage-mediated induction of geneexpression in Myxococcus xanthus. Mol. Microbiol. 49:769–781.
8. Cardona, P. J., C. Y. Soto, C. Martin, B. Giquel, G. Agusti, E. Guirado, T.Sirakova, P. Kolattukudy, E. Julian, and M. Luquin. 2006. Neutral-redreaction is related to virulence and cell wall methyl-branched lipids in My-cobacterium tuberculosis. Microbes Infect. 8:183–190.
9. Chen, M., H. Gan, and H. G. Remold. 2006. A mechanism of virulence:virulent Mycobacterium tuberculosis strain H37Rv, but not attenuatedH37Ra, causes significant mitochondrial inner membrane disruption in mac-rophages leading to necrosis. J. Immunol. 176:3707–3716.
10. Chesne-Seck, M. L., N. Barilone, F. Boudou, J. Gonzalo Asensio, P. E.Kolattukudy, C. Martin, S. T. Cole, B. Gicquel, D. N. Gopaul, and M.Jackson. 2008. A point mutation in the two-component regulator PhoP-PhoR accounts for the absence of polyketide-derived acyltrehaloses but notthat of phthiocerol dimycocerosates in Mycobacterium tuberculosis H37Ra. J.Bacteriol. 190:1329–1334.
11. Cole, S. T., R. Brosch, J. Parkhill, T. Garnier, C. Churcher, D. Harris, S. V.Gordon, K. Eiglmeier, S. Gas, C. E. Barry III, F. Tekaia, K. Badcock, D.Basham, D. Brown, T. Chillingworth, R. Connor, R. Davies, K. Devlin, T.Feltwell, S. Gentles, N. Hamlin, S. Holroyd, T. Hornsby, K. Jagels, A. Krogh,J. McLean, S. Moule, L. Murphy, K. Oliver, J. Osborne, M. A. Quail, M. A.Rajandream, J. Rogers, S. Rutter, K. Seeger, J. Skelton, R. Squares, S.Squares, J. E. Sulston, K. Taylor, S. Whitehead, and B. G. Barrell. 1998.Deciphering the biology of Mycobacterium tuberculosis from the completegenome sequence. Nature 393:537–544.
12. Dubey, V. S., T. D. Sirakova, and P. E. Kolattukudy. 2002. Disruption of msl3abolishes the synthesis of mycolipanoic and mycolipenic acids required forpolyacyltrehalose synthesis in Mycobacterium tuberculosis H37Rv and causescell aggregation. Mol. Microbiol. 45:1451–1459.
13. Dubos, R. J., and G. Middlebrook. 1948. Cytochemical reaction of virulenttubercle bacilli. Am. Rev. Tuberc. 58:698–699.
14. Fontan, P. A., S. Walters, and I. Smith. 2004. Cellular signaling pathways andtranscriptional regulation in Mycobacterium tuberculosis. Curr. Sci. 86:122–134.
15. Fortune, S. M., A. Jaeger, D. A. Sarracino, M. R. Chase, C. M. Sassetti, D. R.Sherman, B. R. Bloom, and E. J. Rubin. 2005. Mutually dependent secretionof proteins required for mycobacterial virulence. Proc. Natl. Acad. Sci. USA102:10676–10681.
16. Frigui, W., D. Bottai, L. Majlessi, M. Monot, E. Josselin, P. Brodin, T.Garnier, B. Gicquel, C. Martin, C. Leclerc, S. T. Cole, and R. Brosch. 2008.Control of M. tuberculosis ESAT-6 secretion and specific T cell recognitionby PhoP. PLoS Pathog. 4:e33.
17. Gao, Q., K. Kripke, Z. Arinc, M. Voskuil, and P. Small. 2004. Comparativeexpression studies of a complex phenotype: cord formation in Mycobacteriumtuberculosis. Tuberculosis (Edinburgh) 84:188–196.
18. Garnier, T., K. Eiglmeier, J. C. Camus, N. Medina, H. Mansoor, M. Pryor,S. Duthoy, S. Grondin, C. Lacroix, C. Monsempe, S. Simon, B. Harris, R.Atkin, J. Doggett, R. Mayes, L. Keating, P. R. Wheeler, J. Parkhill, B. G.Barrell, S. T. Cole, S. V. Gordon, and R. G. Hewinson. 2003. The completegenome sequence of Mycobacterium bovis. Proc. Natl. Acad. Sci. USA 100:7877–7882.
19. Gonzalo Asensio, J., C. Maia, N. L. Ferrer, N. Barilone, F. Laval, C. Y. Soto,N. Winter, M. Daffe, B. Gicquel, C. Martin, and M. Jackson. 2006. Thevirulence-associated two-component PhoP-PhoR system controls the biosyn-thesis of polyketide-derived lipids in Mycobacterium tuberculosis. J. Biol.Chem. 281:1313–1316.
20. Guex, N., and M. C. Peitsch. 1997. SWISS-MODEL and the Swiss-Pdb-Viewer: an environment for comparative protein modeling. Electrophoresis18:2714–2723.
21. Gupta, S., A. Sinha, and D. Sarkar. 2006. Transcriptional autoregulation byMycobacterium tuberculosis PhoP involves recognition of novel direct repeatsequences in the regulatory region of the promoter. FEBS Lett. 580:5328–5338.
22. Haydel, S. E., W. H. Benjamin, Jr., N. E. Dunlap, and J. E. Clark-Curtiss.2002. Expression, autoregulation, and DNA binding properties of the My-cobacterium tuberculosis TrcR response regulator. J. Bacteriol. 184:2192–2203.
23. Haydel, S. E., and J. E. Clark-Curtiss. 2004. Global expression analysis oftwo-component system regulator genes during Mycobacterium tuberculosisgrowth in human macrophages. FEMS Microbiol. Lett. 236:341–347.
24. Howard, N. S., J. E. Gomez, C. Ko, and W. R. Bishai. 1995. Color selectionwith a hygromycin-resistance-based Escherichia coli-mycobacterial shuttlevector. Gene 166:181–182.
25. Lee, J. S., R. Krause, J. Schreiber, H. J. Mollenkopf, J. Kowall, R. Stein,B. Y. Jeon, J. Y. Kwak, M. K. Song, J. P. Patron, S. Jorg, K. Roh, S. N. Cho,and S. H. Kaufmann. 2008. Mutation in the transcriptional regulator PhoPcontributes to avirulence of Mycobacterium tuberculosis H37Ra strain. CellHost Microbe 3:97–103.
26. MacGurn, J. A., S. Raghavan, S. A. Stanley, and J. S. Cox. 2005. A non-RD1gene cluster is required for Snm secretion in Mycobacterium tuberculosis.Mol. Microbiol. 57:1653–1663.
27. Martin, C., A. Williams, R. Hernandez-Pando, P. J. Cardona, E. Gormley, Y.Bordat, C. Y. Soto, S. O. Clark, G. J. Hatch, D. Aguilar, V. Ausina, and B.Gicquel. 2006. The live Mycobacterium tuberculosis phoP mutant strain ismore attenuated than BCG and confers protective immunity against tuber-culosis in mice and guinea pigs. Vaccine 24:3408–3419.
28. Middlebrook, G. 1950. The experimental analysis of virulence of tuberclebacilli. Bull. N. Y. Acad. Med. 26:498–506.
29. Middlebrook, G., C. M. Coleman, and W. B. Schaefer. 1959. Sulfolipid fromvirulent tubercle bacilli. Proc. Natl. Acad. Sci. USA 45:1801–1804.
30. Middlebrook, G., R. J. Dubos, and C. Pierce. 1947. Virulence and morpho-logical characteristics of mammalian tubercle bacilli. J. Exp. Med. 86:175–187.
31. Pascopella, L., F. M. Collins, J. M. Martin, M. H. Lee, G. F. Hatfull, C. K.Stover, B. R. Bloom, and W. R. Jacobs, Jr. 1994. Use of in vivo complemen-tation in Mycobacterium tuberculosis to identify a genomic fragment associ-ated with virulence. Infect. Immun. 62:1313–1319.
32. Perez, E., S. Samper, Y. Bordas, C. Guilhot, B. Gicquel, and C. Martin. 2001.An essential role for phoP in Mycobacterium tuberculosis virulence. Mol.Microbiol. 41:179–187.
33. Rison, S., S. Kendall, F. Movahedzadeh, and N. Stoker. 2005. The myco-bacterial two-component regulatory systems, p. 29–69. In T. Parish (ed.),Mycobacterium molecular microbiology. Horizon Bioscience, Wymondham,United Kingdom.
34. Rodriguez-Garcia, A., M. Ludovice, J. F. Martin, and P. Liras. 1997. Argi-nine boxes and the argR gene in Streptomyces clavuligerus: evidence for aclear regulation of the arginine pathway. Mol. Microbiol. 25:219–228.
35. Rousseau, C., N. Winter, E. Pivert, Y. Bordat, O. Neyrolles, P. Ave, M.Huerre, B. Gicquel, and M. Jackson. 2004. Production of phthiocerol dimy-cocerosates protects Mycobacterium tuberculosis from the cidal activity ofreactive nitrogen intermediates produced by macrophages and modulatesthe early immune response to infection. Cell. Microbiol. 6:277–287.
36. Sambrook, J., and D. W. Russell. 2001. Molecular cloning: a laboratorymanual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
37. Silver, R. F., Q. Li, and J. J. Ellner. 1998. Expression of virulence ofMycobacterium tuberculosis within human monocytes: virulence correlateswith intracellular growth and induction of tumor necrosis factor alpha butnot with evasion of lymphocyte-dependent monocyte effector functions. In-fect. Immun. 66:1190–1199.
38. Sinha, A., S. Gupta, S. Bhutani, A. Pathak, and D. Sarkar. 2008. PhoP-PhoPinteraction at adjacent PhoP binding sites is influenced by protein phosphor-ylation. J. Bacteriol. 190:1317–1328.
39. Sirakova, T. D., A. K. Thirumala, V. S. Dubey, H. Sprecher, and P. E.Kolattukudy. 2001. The Mycobacterium tuberculosis pks2 gene encodes thesynthase for the hepta- and octamethyl-branched fatty acids required forsulfolipid synthesis. J. Biol. Chem. 276:16833–16839.
40. Snapper, S. B., R. E. Melton, S. Mustafa, T. Kieser, and W. R. Jacobs, Jr.1990. Isolation and characterization of efficient plasmid transformation mu-tants of Mycobacterium smegmatis. Mol. Microbiol. 4:1911–1919.
41. Soto, C. Y., N. Andreu, I. Gibert, and M. Luquin. 2002. Simple and rapiddifferentiation of Mycobacterium tuberculosis H37Ra from M. tuberculosisclinical isolates through two cytochemical tests using neutral red and Nileblue stains. J. Clin. Microbiol. 40:3021–3024.
42. Soto, C. Y., M. C. Menendez, E. Perez, S. Samper, A. B. Gomez, M. J. Garcia,and C. Martin. 2004. IS6110 mediates increased transcription of the phoPvirulence gene in a multidrug-resistant clinical isolate responsible for tuber-culosis outbreaks. J. Clin. Microbiol. 42:212–219.
43. Steenken, W., Jr. 1950. Dissociation of the tubercle bacillus; a review. Am.Rev. Tuberc. 62:22–33.
44. Steenken, W. J., W. H. Oatway, and S. A. Petroff. 1934. Biological studies ofthe tubercle bacillus. III. Dissociation and pathogenicity of the R and Svariants of the human tubercle bacillus. J. Exp. Med. 60:515–544.
45. Steenken, W., and L. U. Gardner. 1946. History of H37 strain of tuberclebacillus. Am. Rev. Tuberc. 54:62–66.
VOL. 190, 2008 M. TUBERCULOSIS H37Rv PhoPR OPERON 7077
46. Temoin, S., S. M. Roche, O. Grepinet, Y. Fardini, and P. Velge. 2008.Multiple point mutations in virulence genes explain the low virulence ofListeria monocytogenes field strains. Microbiology 154:939–948.
47. Timm, J., E. M. Lim, and B. Gicquel. 1994. Escherichia coli-mycobacterialshuttle vectors for operon and gene fusions to lacZ: the pJEM series. J.Bacteriol. 176:6749–6753.
48. Velge, P., M. Herler, J. Johansson, S. M. Roche, S. Temoin, A. A. Fedorov,P. Gracieux, S. C. Almo, W. Goebel, and P. Cossart. 2007. A naturallyoccurring mutation K220T in the pleiotropic activator PrfA of Listeria mono-
cytogenes results in a loss of virulence due to decreasing DNA-binding af-finity. Microbiology 153:995–1005.
49. Walters, S. B., E. Dubnau, I. Kolesnikova, F. Laval, M. Daffe, and I. Smith.2006. The Mycobacterium tuberculosis PhoPR two-component system regu-lates genes essential for virulence and complex lipid biosynthesis. Mol.Microbiol. 60:312–330.
50. Wang, S., J. Engohang-Ndong, and I. Smith. 2007. Structure of the DNA-binding domain of the response regulator PhoP from Mycobacterium tuber-culosis. Biochemistry 46:14751–14761.
