• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Mol Microbiol. Author manuscript; available in PMC Sep 24, 2007.
Published in final edited form as:
PMCID: PMC1992516

Rv3133c/dosR is a transcription factor that mediates the hypoxic response of Mycobacterium tuberculosis


Unlike many pathogens that are overtly harmful to their hosts, Mycobacterium tuberculosis can persist for years within humans in a clinically latent state. Latency is often linked to hypoxic conditions within the host. Among M. tuberculosis genes induced by hypoxia is a putative transcription factor, Rv3133c/DosR. We performed targeted disruption of this locus followed by transcriptome analysis of wild-type and mutant bacilli. Nearly all the genes powerfully regulated by hypoxia require Rv3133c/DosR for their induction. Computer analysis identified a consensus motif, a variant of which is located upstream of nearly all M. tuberculosis genes rapidly induced by hypoxia. Further, Rv3133c/DosR binds to the two copies of this motif upstream of the hypoxic response gene alpha-crystallin. Mutations within the binding sites abolish both Rv3133c/DosR binding as well as hypoxic induction of a downstream reporter gene. Also, mutation experiments with Rv3133c/DosR confirmed sequence-based predictions that the C-terminus is responsible for DNA binding and that the aspartate at position 54 is essential for function. Together, these results demonstrate that Rv3133c/DosR is a transcription factor of the two-component response regulator class, and that it is the primary mediator of a hypoxic signal within M. tuberculosis.


Tuberculosis (TB) is a global health emergency of staggering proportions. Worldwide, Mycobacterium tuberculosis (MTB) causes about 8 million new infections and two million deaths each year (Bloom and Small, 1998; Dye et al., 1999). The remarkable success of MTB as a pathogen is closely associated with its ability to persist in humans for extended periods without causing disease. It is estimated that one-third of the world population, or about 1.9 billion people, harbours latent MTB infections (Enarson and Murray, 1996; Dye et al., 1999), which can last for years or decades (Manabe and Bishai, 2000). This enormous reservoir of latent disease greatly complicates efforts at tuberculosis control.

Despite significant effort in recent years, progress has been slow in understanding the natural history of latent tuberculosis and reactivation (Parrish et al., 1998). Important unresolved questions include the metabolic state of bacilli during latency, the role that metabolically dampened MTB may play in lengthening the time necessary for effective chemotherapy, and the nature of the bacterial genetic programme and host responses that underlie long-term persistence. The failure of bacteria to increase in numbers during latency, the lack of clinical sequelae and the enhanced resistance of latent TB to chemotherapy argue that the bacilli may be metabolically dormant (Mitchison, 1992; Wayne and Sramek, 1994; Gangadharam, 1995; Gupta and Katoch, 1997; Hu et al., 1998; Michele et al., 1999). However, there is no direct evidence from the genome (Cole et al., 1998) or the laboratory (Robertson, 1933; McCune et al., 1966; Parrish et al., 1998) that MTB is capable of a truly dormant, spore-like state. In addition, chemotherapy can reduce the rate of reactivation in persons with latent TB (Comstock and Woolpert, 1972; Comstock et al., 1979), and immunotherapy can protect against reactivation in mice (Lowrie et al., 1999). It is hard to see how these therapies would have any effect in the complete absence of mycobacterial metabolism.

Oxygen tension is one factor frequently associated with the establishment and maintenance of latent TB (Wayne and Sohaskey, 2001). In vivo, the number of bacilli in a lesion generally correlates well with the degree of oxygenation (Canetti, 1955), suggesting that O2 supply may limit MTB growth during infections. Also, inhibition of MTB growth in vivo is associated with the formation of hard, fibrous, hypoxic granulomas (Dannenberg, 1993; Yeager et al., 1996). Replication of MTB requires oxygen, but bacilli show a remarkable ability to survive for years without oxygen in vitro (Corper and Cohn, 1933; Canetti, 1955). MTB maintained under anaerobic conditions in vitro lose their acid-fast character (Gillespie et al., 1986), and some human studies (Parrish et al., 1998) have associated latent TB with tubercle bacilli that were no longer acid fast.

Based on these observations, Wayne has pioneered the use of hypoxic culture conditions to generate non-replicating persistent bacilli in vitro as a model for latency (Wayne and Diaz, 1967; Wayne and Sramek, 1994; Wayne and Hayes, 1996). Variants of this model have been used to identify MTB genes potentially important for the development or maintenance of the latent state (Imboden and Schoolnik, 1998; Yuan et al., 1998; Hu et al., 1999; Lim et al., 1999). One such gene is acr (also known as hspX, Rv2031), which encodes alpha-crystallin. MTB Acr is a dominant antigen in vivo, recognized by most TB patient sera (Lee et al., 1992; Verbon et al., 1992). Acr is a member of the small heat shock protein family that forms high-molecular-weight aggregates and has chaperone activity in vitro (Yuan et al., 1996). Under hypoxic conditions, Acr expression is dramatically and rapidly increased (Yuan et al., 1996; 1998; Manabe et al., 1999; Florczyk et al., 2001; Sherman et al., 2001).

We have exploited the powerful regulation of Acr under reduced O2 tension to provide insight into the nature of the genetic programme by which MTB adapts to potentially hypoxic microenvironments within the host. Previously, we described the subset of MTB genes (including acr) that respond rapidly to hypoxia and demonstrated that the MTB protein Rv3133c is required for hypoxic induction of acr (Sherman et al., 2001). Recently, it was demonstrated that long-term hypoxic survival of Mycobacterium bovis BCG required Rv3133c, which was named DosR for dormancy survival regulator (Boon and Dick, 2002). The DosR sequence shows homology to transcription factors of the two-component response regulator class, allowing us to make and test several hypotheses about the behaviour of this protein. Here, we define a sequence motif to which DosR binds. Transcriptome analysis of wild-type and DosR mutant MTB strains indicates that this protein is needed to induce nearly all the MTB genes that respond powerfully to a hypoxic signal. We show that DosR binds to motif sequences upstream of the acr coding region, and that binding is necessary, but not sufficient, for hypoxic gene induction. We conclude that Rv3133c/DosR is the primary transcription factor to mediate the genetic response to reduced oxygen tension in MTB.


Microarray analysis defines the DosR regulon

We have demonstrated previously that the initial hypoxic response of MTB involves 47 induced genes and results in growth arrest (Sherman et al., 2001). Recently, we showed that exposure to low levels of nitric oxide (NO) also produces growth arrest and that the same set of MTB genes is induced by both NO and hypoxia (M. I. Voskuil, submitted). Co-ordinate expression following multiple stimuli suggests that these genes may comprise a regulon under the control of a common transcription factor. Our previous work demonstrated that induction of the hypoxic response gene acr depends upon expression of the putative two-component response regulator DosR (Sherman et al., 2001). To investigate its role in the expression of other genes, we targeted DosR for deletion from the mycobacterial chromosome. Using a two-step method and sacB/sucrose counterselection (Pelicic et al., 1996; Parish and Stoker, 2000), we generated a strain of H37Rv in which DosR has been replaced by a kanamycin resistance determinant. This deletion (bp 3 499 282 to bp 3 499 933 of the H37Rv genome) extends from 18 bp upstream of the putative translation start site to 17 bp upstream of the putative stop codon. Successful gene replacement was confirmed by both site-specific polymerase chain reaction (PCR) and Southern blot (data not shown).

To test the role of DosR in the hypoxic response, we performed whole-genome expression profiling on wild-type H37Rv and mutant H37Rv:ΔdosR::kandosR::kan) under both normoxic (normal oxygen) and hypoxic conditions in vitro. Whole-genome microarray technology provides a robust tool to assess the expression of many genes simultaneously (DeRisi et al., 1997). The MTB microarray has provided a powerful method of monitoring genome-wide changes in MTB gene expression in response to the drug isoniazid (Wilson et al., 1999), of studying regulatory mutants (Manganelli et al., 2001; 2002; Kaushal et al., 2002; Rodriguez et al., 2002) and of assessing the effects of environmental perturbations (Sherman et al., 2001; Betts et al., 2002; Fisher et al., 2002). To determine whether genes in addition to acr are under the control of DosR, MTB strains H37Rv and ΔdosR::kan were maintained for 2 h at 0.2% O2 in N2 as described previously (Yuan et al., 1998; Sherman et al., 2001). RNA was isolated, labelled and applied to the array surface. The resulting hybridized array was scanned, and the fluorescent intensities identified regulated open reading frames (ORFs). Data for all ORFs induced or repressed at least 1.8-fold are available as Supplementary material, with highlights of the data described below. Consistent with previous results (Sherman et al., 2001), incubation of MTB for 2 h under hypoxic conditions resulted in the induction of numerous genes relative to expression under normoxic conditions (Supplementary material, Table S1). However, this response was dramatically muted in the dosR mutant (Supplementary material, Table S2). Genes whose induction depends on DosR are listed in Table 1. Notably, of the 27 genes most powerfully induced by hypoxia (≥5.7-fold), 26 require the presence of DosR. Genetic complementation demonstrated that the effects on hypoxic gene induction result from the mutation of DosR and not some unrelated MTB gene. Introduction of DosR to the dosR mutant restored hypoxic gene induction, as measured by Western blot for the Acr protein (data not shown). Also, in the closely related mutant H37Rv:Δ3134::kan that does not express dosR (Sherman et al., 2001) or respond genetically to reduced O2, complementation with dosR restored hypoxic induction of all DosR-dependent genes (data not shown). We conclude that DosR is a crucial component of the MTB hypoxic response.

Table 1
MTB genes whose induction by hypoxia requires DosR.

A consensus motif upstream of hypoxic response genes

Consistent with its role in gene expression, the DosR protein sequence predicts that it is a transcription factor of the two-component response regulator class (Dasgupta et al., 2000). To begin testing this prediction, we searched by computer analysis with the motif discovery program YMF (Sinha and Tompa, 2002) for shared sequence motifs upstream of hypoxic response genes. As transcription units in bacteria are often multigenic, a single binding site could be responsible for the activation of contiguous genes. Our analysis revealed a 20 bp palindromic consensus sequence 5′-TTSGGGACTWWAGTCCCSAA-3′ (S = C/G; W = A/T; Table 2), a variant of which is upstream of 42/50 (84%) MTB genes that are rapidly induced at least twofold by hypoxia. Using a scoring matrix with entries that are the log likelihood ratios of each base at each position, we determined the best matches to this motif in the H37Rv genome (Supplementary material, Table S4). The top 13 scoring sites are all upstream of genes induced by hypoxia and regulated by DosR (Table 3). The genes most powerfully induced by hypoxia are most likely to have this motif upstream: 94% of 16 genes induced at least 10-fold were preceded by this motif (Table 1). In addition, powerfully induced genes sometimes had multiple copies in their upstream regions.

Table 2
Scoring matrix from which the hypoxia consensus motif is derived.
Table 3
Best matches to the hypoxia consensus in H37Rv genome.

DosR binds to DNA upstream of the hypoxic response gene acr

To see whether DosR could alter gene expression by binding to the hypoxia motif sites, we assessed the ability of this protein to bind specific DNA sequences by electrophoretic mobility shift assay (EMSA). We expressed the DosR protein in Escherichia coli from an IPTG-inducible promoter (Fig. 1A, lanes 1–3) and mixed protein extracts with radiolabelled DNA corresponding to the 146 bp upstream of the acr coding sequence (Fig. 1B). Previously, we have shown that this region contains all the information necessary to direct the hypoxic induction of acr (Yuan et al., 1998; Sherman et al., 2001). A band of retarded mobility, indicating protein bound to the DNA, was evident when acr promoter DNA was mixed with E. coli extracts induced to express DosR (Fig. 1B, lane 2). In addition, DosR is predicted to be a two-component response regulator of the ROIII subclass (Parkinson and Kofoid, 1992; Dasgupta et al., 2000). Typical of such proteins, the DosR sequence divides into two domains: a C-terminal half with a helix–turn–helix (HTH) DNA-binding motif, and an N-terminal receiver half predicted to participate in a histidine-to-aspartate phosphorelay (Dasgupta et al., 2000). To begin testing these predictions, we separately expressed the N-terminal and C-terminal portions of DosR in E. coli (Fig. 1A, lanes 4–7) and assessed the binding of these portions to the acr promoter by mobility shift assay (Fig. 1B). As expected, a shift was evident upon incubation with extracts induced to express the C-terminus (amino acids 144–217; Fig. 1B, lane 4) but not the N-terminus (amino acids 1–134; Fig. 1B, lane 3). All mobility shift assays reported here included a large molar excess of unlabelled poly-(dI–dC) and calf thymus DNA, indicating that binding was specific for the acr promoter. Further, we conclude that the protein binding to the acr promoter is DosR and not some unrelated E. coli protein induced by IPTG, as E. coli extracts containing the N-terminus of DosR failed to produce a shift.

Fig. 1
DosR binds to DNA upstream of the acr coding sequence.
  1. SDS-PAGE analysis of DosR expression in E. coli. The dosR coding sequence was cloned into a pET expression vector, and expression was induced by IPTG. Lanes (1–3) full length: 1, uninduced;

To facilitate analysis of the interaction between DosR and the acr promoter, we mapped the acr transcription start site by primer extension. A synthetic oligonucleotide of 16 bases that terminates at bp 7 of the acr coding region was used in extension reactions with RNA from oxygenated (rolling) and hypoxic (static) log-phase cultures of MTB. A single start site was visualized 29 bp upstream of the initiator ATG (Fig. 2). Reactions with additional aliquots of MTB RNA and a different primer confirmed this result (data not shown), which is 4 bp downstream of the putative acr transcription start site reported by Hu and Coates (1999). This discrepancy may result from differences in technique, as Hu and Coates (1999) sized their extension products on a heterologous sequencing ladder. All base numberings below are relative to the acr transcription start site defined here.

Fig. 2
Localization of the acr transcription start site. Primer extension was performed with RNA from oxygenated (rolling) and hypoxic (static) log-phase cultures of MTB. Products were sized on a sequencing ladder of the acr promoter region.

To localize binding of DosR upstream of acr, we performed additional mobility shift assays with double-stranded oligonucleotides corresponding to overlapping regions of the acr promoter. We detected protein binding in two distinct regions, a more distal 33 bp stretch from −111 to −79 and a more proximal 33 bp stretch from −66 to −34 (Fig. 3A). Significantly, these regions each contain a copy of the motif defined above (Figs (Figs3B3B and and4A).4A). To assess the specificity of DosR binding to these regions, we repeated the mobility shift assays after adding a 100-fold molar excess of various unlabelled competitor DNAs. In each case, binding was abolished by the addition of the corresponding unlabelled DNA (Fig. 4B, lanes 3 and 9). In addition, the unlabelled DNA competed for binding to the other acr promoter region (Fig. 4B, lanes 5 and 11). To test specifically the role of the upstream motifs in DosR binding, we mutated four bases within each motif (−94 to −91 in the distal region and −53 to −50 in the proximal region; underlined in Fig. 4A). The mutated DNAs were no longer able to compete with the wild-type sequence for binding to DosR (Fig. 4B, lanes 4, 6, 10 and 12). Furthermore, when mutated versions of −111 to −79 and −66 to −34 were used in mobility shift assays, shifted bands were abolished, indicating that the bases mutated are necessary for DosR binding (data not shown).

Fig. 3
Localization of DosR binding within the acr promoter.
  1. EMSA of E. coli extracts and radiolabelled overlapping regions of acr promoter DNA. Labelled probe A (−149 to −96); B (−111 to −79); C (−89 to −49); D
Fig. 4
Effects of mutations in the hypoxia motifs upstream of the acr coding sequence.
  1. 20-mer hypoxia motifs in the native acr promoter. Positions are relative to the transcription start site. Underlined bases were mutated as follows: ‘ccaa’→‘ggtt’

DosR binding is required for hypoxic gene induction

These results indicate that DosR binds to two distinct regions of the acr promoter, each of which contains an example of the hypoxia motif. To test the functional significance of this binding, we introduced the mutations described above into the acr upstream sequence and assessed the hypoxic responsiveness of the wild-type and mutant promoters by luciferase reporter gene assay (Yuan et al., 1998; Sherman et al., 2001) (Fig. 4C). None of the mutations had any discernible effect under normal oxygenation. However, mutating from −94 to −91, which eliminated DosR binding to the distal motif sequence, reduced hypoxic responsiveness roughly by half. Mutating from −53 to −50 to eliminate DosR binding to the proximal region reduced hypoxic responsiveness nearly sixfold. Introducing both mutations simultaneously virtually abolished the induction, indicating that DosR binding to its motif is necessary for a proper hypoxic response in MTB.

Mutating DosR affects hypoxic gene induction

Based on alignment to known two-component response regulators, Asp-54 of DosR is predicted to be a site of phosphorylation that is essential for gene induction. To test the importance of this amino acid to DosR function, we mutated Asp-54 to Glu. This change should conserve the charge and structure of the protein but block the possibility of phosphorylation. Mutated DosR (Asp-54Glu) could still bind to the acr promoter (Fig. 5A). However, in functional assays, the mutated DosR was no longer able to mediate hypoxic induction of a luciferase reporter gene under the control of the acr promoter (Fig. 5B). These results indicate that Asp-54 is essential to DosR activity, possibly because it is a site of phosphorylation.

Fig. 5
DosR Asp-54 is required for gene induction but not for binding.
  1. EMSA of mutant DosR (Asp-54Glu) with labelled acr promoter DNA. Lanes: 1, no extract; 2, induced extract of DosR (Asp-54Glu); 3, induced extract of DosR (WT).
  2. Hypoxic induction of acr by mutant


We have shown that DosR of MTB is necessary for the induction of a robust genetic response to reduced oxygen tension. Further, we have defined a consensus sequence, a variant of which is upstream of nearly all the hypoxic response genes. We have shown that the DosR protein binds to this sequence upstream of the acr gene, and that binding is necessary, but not sufficient, for gene induction. Thus, we conclude that DosR is a transcription factor that mediates the induction of MTB genes in response to hypoxia. Further, consistent with the sequence-based prediction that DosR belongs to the two-component response regulator family of transcriptional activators, we find that the C-terminal 73 amino acids are sufficient for DNA binding and that the aspartate at position 54 is essential for activity. Previously, this protein was named DevR, because it is differentially expressed in virulent MTB strain H37Rv compared with avirulent H37Ra (Dasgupta et al., 2000; Mayuri et al., 2002; Saini et al., 2002). Given the functional data presented elsewhere (Boon and Dick, 2002) and here, we believe that the name DosR is more appropriate.

The presence of an upstream motif sequence defined here correlates strongly with genes induced by hypoxia and regulated by DosR. Further, the number of motif sequences and their degree of fit to the consensus, as defined by matrix score, roughly predict the extent of hypoxic gene induction. Experiments are under way to determine whether the matrix score predicts the strength of DNA binding. It will be interesting to see whether a perfect match, which exists nowhere in the H37Rv genome, elicits stronger binding and/or more robust gene induction. Still, the link between motif sequence, regulation by hypoxia and by DosR is not exact. For example, a reasonable match to the motif is present upstream of Rv1976c and Rv1977 (score = 11.7), yet these genes are not induced by hypoxia. Conversely, a few genes (e.g. Rv3129 or Rv3841; see Table 1) appear to be regulated by DosR without an obvious upstream match to the consensus. These inconsistencies suggest that aspects of sequence or context may affect DosR-mediated gene expression in ways that we have not yet discerned. Also, although nearly all the genes powerfully induced by hypoxia are controlled by DosR, there are numerous modestly induced genes for which DosR does not appear to play a role (Supplementary material, Table S1). We are now investigating whether another transcription factor(s) also affect(s) the MTB hypoxic response.

We are interested in the signal transduction pathway by which a drop in oxygen tension results in altered MTB gene expression. For instance, it seems reasonable that some haem protein may play a role in monitoring available O2, but the identity of this sensor is unknown. The identity of the kinase that activates DosR is also not yet clear. Directly adjacent to dosR on the MTB chromosome is a putative sensor kinase gene Rv3132c, but disruption of that gene had little effect, as measured by the accumulation of Acr protein (Sherman et al., 2001), and little impact on survival during long-term hypoxic dormancy (Boon and Dick, 2002). Strong sequence homology between Rv3132c and Rv2027c has been noted previously (Dasgupta et al., 2000; Sherman et al., 2001). We are currently investigating whether either or both these proteins participate in the phosphorelay that activates DosR.

Another outstanding question is the role of DosR-mediated gene expression and the hypoxic response during MTB infection of a mammalian host. In vitro, both hypoxia and low levels of NO cause MTB to enter a state of growth arrest (Wayne and Hayes, 1996; M. I. Voskuil et al., submitted). In vivo, MTB may be exposed to hypoxia and NO as granulomas form during a normal host response (Flynn and Chan, 2001; Wayne and Sohaskey, 2001). It is tempting to speculate that host-induced hypoxia/NO curtails bacterial replication, but the ensuing bacterial response results in long-term persistence rather than clearance. Indeed, we have shown recently that DosR response genes are powerfully induced during MTB infection of mice (M. I. Voskuil et al., submitted). However, despite its attractiveness, the link between hypoxia, NO and mycobacterial latency in vivo remains circumstantial and will require further testing. Experiments to define the phenotype of the dosR deletion mutant under hypoxic conditions in vitro and in vivo are under way.

Experimental procedures

Mycobacteria and culture conditions

H37Rv (ATCC 27294), H37Rv:ΔdosR::kan and M. bovis BCG-Montreal (ATCC 35735) were grown to mid-log phase in 7H9 media with 0.05% Tween 80 and ADC supplement (Becton Dickinson) and stored as 1 ml aliquots in 15% glycerol (final concentration) at −80°C. For individual experiments, bacilli were grown in roller bottles in Middlebrook 7H9 medium (Becton Dickinson) with 0.05% Tween 80 and ADC supplement or on Middlebrook 7H10 plates at 37°C as described previously (Sherman et al., 1995). When needed, kanamycin was used at 30 μg ml−1 (50 μg ml−1 for E. coli) and hygromycin at 50 μg ml−1 (200 μg ml−1 for E. coli).

Targeted MTB gene disruption

Targeted disruptions were performed as described previously (Sherman et al., 2001). Genomic regions (≈ 900 bp each) flanking dosR were amplified by PCR and cloned into the plasmid pKO in order to flank the kanR determinant. The fidelity of the cloned flanking sequences was confirmed by sequence analysis. Constructs were electroporated into mycobacteria as described previously (Wards and Collins, 1996) and selected on 7H10 plates with hygromycin. In the first screening step, each colony was tested by PCR with two primer pairs, one specific for integration upstream of the gene of interest and the other specific for integration downstream. Colonies positive at either end by PCR were grown to an OD600 = ≈ 1.0 and plated onto 7H10 plates containing 10% sucrose. Bacilli that grow on sucrose generally either have mutated copies of sacB or have lost the integrated plasmid. A portion of those in which the plasmid is lost will also lose the gene of interest. Colonies appearing on sucrose plates were picked into media and patched separately onto 7H10 plates with kanamycin and hygromycin. Sucrose-resistant, hygromycin-sensitive, kanamycin-resistant colonies (indicating loss of the integrated plasmid) were screened for loss of the gene of interest by PCR and confirmed by Southern blot.

Microarray analysis

RNA isolation was performed as described previously (Sherman et al., 2001). Steps in M. tuberculosis DNA microarray gene expression analysis were performed as described previously (Schoolnik et al., 2001; Ehrt et al., 2002). A 70-mer oligonucleotide-basezd microarray (tuberculosis oligonucleotide set; Qiagen) was used. Labelled cDNA was prepared as follows: 2 μg of total RNA and 4.4 μg of random oligonucleotide hexamers were incubated for 2 min at 98°C, cooled on ice, combined with Stratascript RTase buffer, 0.5 mM dA-, G-, CTP, 0.02 mM dTTP, 1.5 nmol of Cy3- or Cy5-dUTP (Amersham) and 1.8 μl of Stratascript RTase (Stratagene) in a total volume of 25 μl and incubated for 10 min at 25°C and for 90 min at 42°C. cDNA was purified by microcon-10 (Amicon) filtration. Hybridization solution (10 μl; labelled cDNA, 5 μg of tRNA, 2× SSC, 25% formamide, 0.1% SDS) was hybridized at 52°C. Microarray-determined ratios were calculated from three biological replicates with two microarrays for each biological replicate. Microarrays were scanned using a GenePix 4000 A (Axon Instruments). The intensities of the two dyes at each spot were quantified using SCANALYZE (M. Eisen; http://rana.lbl.gov/EisenSoftware.htm). All gene-specific spots on the microarray were used to normalize the intensities of Cy3 and Cy5 from each spot. Additional methods and analysis details are available at the following website: http://schoolniklab.stanford.edu/projects/tb.html.

Computer analysis

Of the genes whose induction by hypoxia requires DosR (Table 1), 27 have non-coding upstream regions of at least 25 bp. For each of these, 250 bp of upstream sequences was collected. The sequences for Rv 1737c and Rv 2032 were discarded, as they share ≈ 250 bp upstream sequence with Rv 1738 and acr, respectively, which are already in this set. The motif discovery program YMF (http://bio.cs.washington.edu/software.html; Sinha and Tompa, 2002) was applied to the remaining 25 upstream sequences. The four to six motifs with greatest z-scores produced for each of the lengths 6, 7, 8 and 9 were aligned, resulting in 19 high-scoring YMF motifs (Supplementary material, Table S3). All instances of these motifs in the 25 upstream sequences, together with the reverse complements of such instances, were collected and aligned similarly, resulting in 100 aligned motif instances from the upstream regions, each of length 20. These induced the 4 × 20 log likelihood ratio matrix M (first four columns of Table 2), where Mrj=log2(prjqr) for r [set membership] {A,T,G,C} and 1 ≤ j ≤ 20; prj = (nrj + qr)/(n + 1) is the likelihood of residue r in position j; nrj is the number of occurrences of residue r in position j among the 100 aligned motif instances; n = 100 is the number of aligned motif instances; and qr is the background frequency of residue r in the non-coding portion of MTB (qA = qT = 0.19, qG = qC = 0.31). The log likelihood ratio matrix M was used to score every 20-mer in the non-coding portion of the MTB genome, where the score of x[1]x[2]…x[20] is given by ΣjMx[j],j. The highest scoring 20-mers are shown in Table 3. Additional high-scoring 20-mers are shown in Supplementary material, Table S4.

Expression of DosR in E. coli

To express DosR in E. coli, a DNA fragment of full-length, N-terminal or C-terminal portions of the DosR ORF was amplified by PCR and cloned into pET-22b(+) (Novagen). The resultant plasmid was introduced to E. coli BL21 (DE3) (Novagen). To induce the expression of DosR, cultures grown to OD600 = 0.6 were treated with 1 mM IPTG at room temperature for 5 h. To prepare bacterial extracts, cells were harvested and washed with 5× sonication buffer (20 mM Tris-Cl, pH 8.0, 100 mM KCl, 10 mM MgCl2, 25% glycerol). After washing, cells were resuspended in the same buffer and disrupted by three 45 s pulses in a bead-beater (Qbiogene Fast prep FP120). Cell extract was obtained by centrifugation at 14 000 g at 4°C for 30 min.

Electrophoretic mobility shift assay

To assess protein binding, E. coli extracts (prepared as described above) were used as the protein source. Oligonucleotides were used as DNA probes, designed so that, when annealed, both ends had 5′ ‘a’ extensions that could be labelled with [32P]-dTTP by a Klenow fill-in reaction as described previously (Winterling et al., 1998). DNA probes were labelled in a 20 μl reaction mixture containing 2 μg of DNA. Gel mobility shift assay was carried out as described previously (Dhandayuthapani et al., 1997) with some modifications. Binding of protein to DNA fragments was carried out by incubation at room temperature for 10 min in a 20 μl reaction mixture. The mixture was composed of labelled probe (10 000 c.p.m.), 5–10 μg of protein, 25 mM Tris-HCl, pH 8.0, 0.5 mM EDTA, 20 mM KCl, 6 mM MgCl2, 1 mM dithiothreitol (DTT), 50 ng μl−1 poly-(dI–dC), 50 ng μl−1 calf thymus DNA and 5% glycerol. After incubation, the reaction was resolved in a 5% non-denaturing TBE polyacrylamide gel (Bio-Rad).

Primer extension

Primer extension was performed as described previously (Sambrook et al., 1989), using a synthetic oligonucleotide, sequence 5′-gaacgggaagggtggt-3′, and RNA (5–10 μg/rxn) from oxygenated (rolling) and hypoxic (static) log-phase cultures of MTB, isolated as described previously (Sherman et al., 2001).

Luciferase reporter gene assay

This assay was performed as described previously (Yuan et al., 1998; Sherman et al., 2001). Briefly, acr promoter (with described mutations) was amplified by PCR and cloned into the integrating mycobacterial shuttle plasmid pMH66 upstream of the firefly luciferase gene. The resultant construct was electroporated into BCG, where it integrated into the phage L5 attachment site. Luciferase activity was assayed in triplicate for 15 s by TD-20/20 luminometer (Turner Designs).

Other procedures

Mutated versions of dosR and the acr promoter sequence were generated by mutagenic gene fusion PCR (Yon and Fried, 1989). Mycobacterial transformations were carried out by electroporation as described previously (Wards and Collins, 1996; Sherman et al., 2001). E. coli strain DH5α (Invitrogen) was used for routine DNA manipulations.

Supplementary Material

Supplementary Tables


We thank Mark Hickey for mapping the acr transcription start site by primer extension, and Carleen Collins, Wim Hol, Goragot Wisedchaisri, David Roberts, Mark Hickey and members of the Sherman laboratory for helpful discussions. We gratefully acknowledge funding from NIH (grants AI47744 to D.R.S., AI44826 to G.K.S. and HG02602-01 to M.T.), NSF (grant DBI-9974498 to M.T.) and the Sequella Global Tuberculosis Foundation, of which D.R.S. is a Core Scientist.


Supplementary material

The following material is available from http://www.blackwellpublishing.com/products/journals/suppmat/mole/mole3474/mmi3474sm.htm

Table S1. Data for all ORFs of strain H37Rv induced or repressed at least 1.8-fold under hypoxic conditions.

Table S2. Data for all ORFs of strain H37Rv:DdosR induced or repressed at least 1.8-fold under hypoxic conditions.

Table S3. Alignment of 19 top-scoring YMF patterns used to generate the matrix in Table 2.

Table S4. Best 200 matches to the hypoxia motif in the H37Rv genome.


  • Betts JC, Lukey PT, Robb LC, McAdam RA, Duncan K. Evaluation of a nutrient starvation model of Mycobacterium tuberculosis persistence by gene and protein expression profiling. Mol Microbiol. 2002;43:717–731. [PubMed]
  • Bloom BR, Small PM. The evolving relation between humans and Mycobacterium tuberculosis. N Engl J Med. 1998;338:677–678. [PubMed]
  • Boon C, Dick T. Mycobacterium bovis response regulator essential for hypoxic dormancy. J Bacteriol. 2002;184:6760–6767. [PMC free article] [PubMed]
  • Canetti G. The Tubercle Bacillus in the Pulmonary Lesion of Man. Springer Publishing; New York: 1955. Growth of the tubercle bacillus in the tuberculosis lesion; pp. 111–126.
  • Cole ST, Brosch R, Parkhill J, Garnier T, Churcher C, Harris D, et al. Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature. 1998;393:537–544. [PubMed]
  • Comstock GW, Woolpert SF. Preventive treatment of untreated, non-active tuberculosis in an Eskimo population. Arch Environ Health. 1972;25:333–337. [PubMed]
  • Comstock GW, Baum C, Snider DE. Isoniazid prophylaxis among Alaskan Eskimos: a final report of the Bethel isoniazid studies. Am Rev Resp Dis. 1979;119:827–830. [PubMed]
  • Corper HJ, Cohn ML. The viability and virulence of old cultures of tubercle bacilli. Annu Rev Tuberc. 1933;28:856–874.
  • Dannenberg AM., Jr Immunopathogenesis of pulmonary tuberculosis. Hosp Pract. 1993;28:51–58. [PubMed]
  • Dasgupta N, Kapur V, Singh KK, Das TK, Sachdeva S, Jyothisri K, Tyagi JS. Characterization of a two-component system, devR-devS, of Mycobacterium tuberculosis. Tuber Lung Dis. 2000;80:141–159. [PubMed]
  • DeRisi JL, Iyer VR, Brown PO. Exploring the metabolic and genetic control of gene expression on a genomic scale. Science. 1997;278:680–686. [PubMed]
  • Dhandayuthapani S, Mudd M, Deretic V. Interactions of OxyR with the promoter region of the oxyR and ahpC genes from Mycobacterium leprae and Mycobacterium tuberculosis. J Bacteriol. 1997;179:2401–2409. [PMC free article] [PubMed]
  • Dye C, Scheele S, Dolin P, Pathania V, Raviglione MC. Consensus statement. Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country. WHO Global Surveillance and Monitoring Project. J Am Med Assoc. 1999;282:677–686. [PubMed]
  • Ehrt S, Voskuil MI, Schoolnik GK, Schnappinger D. Genome-wide expression profiling of intracellular bacteria: the interaction of Mycobacterium tuberculosis and macrophages. In: Kabelitz K, editor. Methods in Microbiology. Vol. 32. Academic Press; SanDiego, CA: 2002. pp. 169–180.
  • Enarson DA, Murray JF. Global epidemiology of tuberculosis. In: Rom WN, Garay S, editors. Tuberculosis. Little, Brown; Boston: 1996. pp. 55–75.
  • Fisher MA, Plikaytis BB, Shinnick TM. Microarray analysis of the Mycobacterium tuberculosis transcriptional response to the acidic conditions found in phagosomes. J Bacteriol. 2002;184:4025–4032. [PMC free article] [PubMed]
  • Florczyk MA, McCue LA, Stack RF, Hauer CR, McDonough KA. Identification and characterization of mycobacterial proteins differentially expressed under standing and shaking culture conditions, including Rv2623 from a novel class of putative ATP-binding proteins. Infect Immun. 2001;69:5777–5785. [PMC free article] [PubMed]
  • Flynn JL, Chan J. Immunology of tuberculosis. Annu Rev Immunol. 2001;19:93–129. [PubMed]
  • Gangadharam PR. Mycobacterial dormancy. Tuber Lung Dis. 1995;76:477–479. [PubMed]
  • Gillespie J, Barton LL, Rypka EW. Phenotypic changes in mycobacteria grown in oxygen-limited conditions. J Med Microbiol. 1986;21:251–255. [PubMed]
  • Gupta UD, Katoch VM. Understanding the phenomenon of persistence in mycobacterial infections. Indian J Lepr. 1997;69:385–393. [PubMed]
  • Hu Y, Coates AR. Transcription of the stationary-phase-associated hspX gene of Mycobacterium tuberculosis is inversely related to synthesis of the 16-kilodalton protein. J Bacteriol. 1999;181:1380–1387. [PMC free article] [PubMed]
  • Hu YM, Butcher PD, Sole K, Mitchison DA, Coates AR. Protein synthesis is shutdown in dormant Mycobacterium tuberculosis and is reversed by oxygen or heat shock. FEMS Microbiol Lett. 1998;158:139–145. [PubMed]
  • Hu YM, Butcher PD, Mangan JA, Rajandream MA, Coates ARM. Regulation of hmp gene transcription in Mycobacterium tuberculosis: Effects of oxygen limitation and nitrosative and oxidative stress. J Bacteriol. 1999;181:3486–3493. [PMC free article] [PubMed]
  • Imboden P, Schoolnik GK. Construction and characterization of a partial Mycobacterium tuberculosis cDNA library of genes expressed at reduced oxygen tension. Gene. 1998;213:107–117. [PubMed]
  • Kaushal D, Schroeder BG, Tyagi S, Yoshimatsu T, Scott C, Ko C, et al. Reduced immunopathology and mortality despite tissue persistence in a Mycobacterium tuberculosis mutant lacking alternative sigma factor, SigH. Proc Natl Acad Sci USA. 2002;99:8330–8335. [PMC free article] [PubMed]
  • Lee BY, Hefta SA, Brennan PJ. Characterization of the major membrane protein of virulent Mycobacterium tuberculosis. Infect Immun. 1992;60:2066–2074. [PMC free article] [PubMed]
  • Lim A, Eleuterio M, Hutter B, Murugasu-Oei B, Dick T. Oxygen depletion-induced dormancy in Mycobacterium bovis BCG. J Bacteriol. 1999;181:2252–2256. [PMC free article] [PubMed]
  • Lowrie DB, Tascon RE, Bonato VL, Lima VM, Faccioli LH, Stavropoulos E, et al. Therapy of tuberculosis in mice by DNA vaccination. Nature. 1999;400:269–271. [PubMed]
  • McCune RM, Feldmann FM, McDermott W. Microbial persistence. II. Characteristics of the sterile state of tubercle bacilli. J Exp Med. 1966;123:469–486. [PMC free article] [PubMed]
  • Manabe YC, Bishai WR. Latent mycobacterium tuberculosis-persistence, patience, and winning by waiting. Nature Med. 2000;6:1327–1329. [PubMed]
  • Manabe YC, Chen JM, Ko CG, Chen P, Bishai WR. Conditional sigma factor expression, using the inducible acetamidase promoter, reveals that the Mycobacterium tuberculosis sigF gene modulates expression of the 16-kilodalton alpha-crystallin homologue. J Bacteriol. 1999;181:7629–7633. [PMC free article] [PubMed]
  • Manganelli R, Voskuil MI, Schoolnik GK, Smith I. The Mycobacterium tuberculosis ECF sigma factor sigmaE: role in global gene expression and survival in macrophages. Mol Microbiol. 2001;41:423–437. [PubMed]
  • Manganelli R, Voskuil MI, Schoolnik GK, Dubnau E, Gomez M, Smith I. Role of the extracytoplasmic-function sigma factor sigma (H) in Mycobacterium tuberculosis global gene expression. Mol Microbiol. 2002;45:365–374. [PubMed]
  • Mayuri, Bagchi G, Das TK, Tyagi JS. Molecular analysis of the dormancy response in Mycobacterium smegmatis: expression analysis of genes encoding the DevR-DevS two-component system, Rv3134c and chaperone alpha-crystallin homologues. FEMS Microbiol Lett. 2002;211:231–237. [PubMed]
  • Michele TM, Ko C, Bishai WR. Exposure to antibiotics induces expression of the Mycobacterium tuberculosis sigF gene: implications for chemotherapy against mycobacterial persistors. Antimicrob Agents Chemother. 1999;43:218–225. [PMC free article] [PubMed]
  • Mitchison DA. The Garrod Lecture: understanding the chemotherapy of tuberculosis – current problems. J Antimicrob Chemother. 1992;29:477–493. [PubMed]
  • Parish T, Stoker NG. Use of a flexible cassette method to generate a double unmarked Mycobacterium tuberculosis tlyA plcABC mutant by gene replacement. Microbiology. 2000;146:1969–1975. [PubMed]
  • Parkinson JS, Kofoid EC. Communication modules in bacterial signaling proteins. Annu Rev Genet. 1992;26:71–112. [PubMed]
  • Parrish NM, Dick JD, Bishai WR. Mechanisms of latency in Mycobacterium tuberculosis. Trends Microbiol. 1998;6:107–112. [PubMed]
  • Pelicic V, Reyrat JM, Gicquel B. Generation of unmarked directed mutations in mycobacteria, using sucrose counter-selectable suicide vectors. Mol Microbiol. 1996;20:919–925. [PubMed]
  • Robertson HE. The persistence of tuberculosis infections. Am J Pathol. 1933;9:711–718. [PMC free article] [PubMed]
  • Rodriguez GM, Voskuil MI, Gold B, Schoolnik GK, Smith I. ideR, an essential gene in mycobacterium tuberculosis: role of IdeR in iron-dependent gene expression, iron metabolism, and oxidative stress response. Infect Immun. 2002;70:3371–3381. [PMC free article] [PubMed]
  • Saini DK, Pant N, Das TK, Tyagi JS. Cloning, overexpression, purification, and matrix-assisted refolding of DevS (Rv 3132c) histidine protein kinase of Mycobacterium tuberculosis. Protein Expr Purif. 2002;25:203–208. [PubMed]
  • Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: a Laboratory Manual. Cold Spring Harbor Laboratory Press; Cold Spring Harbor, NY: 1989.
  • Schoolnik GK, Voskuil MI, Schnappinger D, Yildiz FH, Meibom K, Dolganov NA, et al. Whole genome DNA microarray expression analysis of biofilm development by Vibrio cholerae O1 E1 Tor. In: Doyle R, editor. Methods in Enzymology. Vol. 336. Academic Press; SanDiego, CA: 2001. pp. 3–18. [PubMed]
  • Sherman DR, Sabo PJ, Hickey MJ, Arain TM, Mahairas GG, Yuan Y, et al. Disparate responses to oxidative stress in saprophytic and pathogenic mycobacteria. Proc Nat Acad Sci USA. 1995;92:6625–6629. [PMC free article] [PubMed]
  • Sherman DR, Voskuil M, Schnappinger D, Liao R, Harrell MI, Schoolnik GK. Regulation of the Mycobacterium tuberculosis hypoxic response gene encoding alpha-crystallin. Proc Natl Acad Sci USA. 2001;98:7534–7539. [PMC free article] [PubMed]
  • Sinha S, Tompa M. Discovery of novel transcription factor binding sites by statistical overrepresentation. Nucleic Acids Res. 2002;30:5549–5560. [PMC free article] [PubMed]
  • Verbon A, Hartskeerl RA, Schuitema A, Kolk AH, Young DB, Lathigra R. The 14,000-molecular-weight antigen of Mycobacterium tuberculosis is related to the alpha-crystallin family of low-molecular-weight heat shock proteins. J Bacteriol. 1992;174:1352–1359. [PMC free article] [PubMed]
  • Wards BJ, Collins DM. Electroporation at elevated temperatures substantially improves transformation efficiency of slow-growing mycobacteria. FEMS Microbiol Lett. 1996;145:101–105. [PubMed]
  • Wayne LG, Diaz GA. Autolysis and secondary growth of Mycobacterium tuberculosis in submerged culture. J Bacteriol. 1967;93:1374–1381. [PMC free article] [PubMed]
  • Wayne LG, Hayes LG. An in vitro model for sequential study of shiftdown of Mycobacterium tuberculosis through two stages of nonreplicating persistence. Infect Immun. 1996;64:2062–2069. [PMC free article] [PubMed]
  • Wayne LG, Sohaskey CD. Nonreplicating persistence of Mycobacterium tuberculosis. Annu Rev Microbiol. 2001;55:139–163. [PubMed]
  • Wayne LG, Sramek HA. Metronidazole is bactericidal to dormant cells of Mycobacterium tuberculosis. Antimicrob Agents Chemother. 1994;38:2054–2058. [PMC free article] [PubMed]
  • Wilson M, DeRisi J, Kristensen HH, Imboden P, Rane S, Brown PO, Schoolnik GK. Exploring drug-induced alterations in gene expression in Mycobacterium tuberculosis by microarray hybridization. Proc Natl Acad Sci USA. 1999;96:12833–12838. [PMC free article] [PubMed]
  • Winterling KW, Chafin D, Hayes JJ, Sun J, Levine AS, Yasbin RE, Woodgate R. The Bacillus subtilis DinR binding site: redefinition of the consensus sequence. J Bacteriol. 1998;180:2201–2211. [PMC free article] [PubMed]
  • Yeager HJ, Azumi N, Underhill CB. Fibrosis: the formation of the granuloma matrix. In: Rom WN, Garay SM, editors. Tuberculosis. Little, Brown; Boston: 1996. pp. 363–372.
  • Yon J, Fried M. Precise gene fusion by PCR. Nucleic Acids Res. 1989;17:4895. [PMC free article] [PubMed]
  • Yuan Y, Crane DD, Barry CE., III Stationary phase-associated protein expression in Mycobacterium tuberculosis: function of the mycobacterial alpha-crystallin homolog. J Bacteriol. 1996;178:4484–4492. [PMC free article] [PubMed]
  • Yuan Y, Crane DD, Simpson RM, Zhu Y, Hickey MJ, Sherman DR, Barry CE., III The 16-kDa α-crystallin (Acr) protein of Mycobacterium tuberculosis is required for growth in macrophages. Proc Nat Acad Sci USA. 1998;95:9578–9583. [PMC free article] [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem Compound links
  • Gene
    Gene links
  • GEO DataSets
    GEO DataSets
    GEO DataSet links
  • MedGen
    Related information in MedGen
  • Pathways + GO
    Pathways + GO
    Pathways, annotations and biological systems (BioSystems) that cite the current article.
  • Protein
    Published protein sequences
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links
  • Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...