SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Mycobacterium tuberculosis has two members of the α-crystallin (Acr) family of molecular chaperones. Expression of Acr1 is induced by exposure to hypoxia or nitric oxide and is associated with bacterial persistence in a non-replicating state. Expression of Acr2 is induced by heat shock, oxidative stress, and uptake by macrophages. We have shown that Acr2 continues to be expressed at a high level during both acute and chronic infection in the mouse model, with an increased ratio of acr2:acr1 mRNA in the persistent phase. Deletion of the acr2 gene resulted in a decrease in the resistance of M. tuberculosis to oxidative stress but did not impair growth in mouse bone marrow macrophages. There was no difference in bacterial load in mice infected with an acr2 deletion mutant, but a marked alteration in disease progression was evident from reduced weight loss over a prolonged infection. This correlated with reduced recruitment of T-cells and macrophages to the lungs of mice infected with the acr2 mutant and reduced immune-related pathology. These findings demonstrate that both α-crystallins contribute to persistent infection with M. tuberculosis and suggest that manipulation of acr expression can influence the host response to infection.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Mycobacterium tuberculosis remains one of the world's most prevalent and serious pathogens. It is estimated that every year 2 million people die as a direct result of tuberculosis and that there is a reservoir of 2 billion cryptically infected people (Dye et al., 1999). Of these asymptomatic carriers, around 10% will at some stage in their lives develop active disease and in doing so will contribute to the ongoing transmission of infection (Bloom and Murray, 1992). During this lifelong infection the bacterium has to survive in a range of different environments, including the phagocytic compartment of macrophages and the potentially hypoxic environment of granulomas. M. tuberculosis has developed a series of strategies both to reduce overall exposure to the antimicrobial defences of host phagocytes, and to neutralize their effectiveness. These include the inhibition of normal phagosomal processing (Russell, 2001), and the subversion of immune activation by interference with antigen presentation (Hmama et al., 1998; Noss et al., 2001) and cytokine signalling (Ting et al., 1999). In addition, the bacteria survive exposure to unfavourable conditions encountered during infection by taking advantage of the protection conferred by their resilient cell wall, by expressing a range of antioxidant enzymes (Li et al., 1998; Dussurget et al., 2001; Piddington et al., 2001; Master et al., 2002), by expression of mechanisms of resistance against nitrosative damage (Darwin et al., 2003) and by mounting a vigorous stress protein response following phagocytosis (Lee and Horwitz, 1995; Monahan et al., 2001; Schnappinger et al., 2003). There is growing evidence that the long-term survival of M. tuberculosis is associated with adaptation to a phenotypically resistant non-replicating form in response to exposure to nitric oxide (NO) within the activated macrophage or to low oxygen tension in the granuloma (Wayne and Sohaskey, 2001; Voskuil et al., 2003).

To study the general stress response of M. tuberculosis we have extensively characterized the bacterial response to heat shock in terms of protein expression (Young and Garbe, 1991), gene transcription and regulatory mechanisms (Stewart et al., 2001; 2002). In a recent whole genome expression study, we identified acr2/Rv0251c, encoding a member of the α-crystallin or small heat shock protein chaperone family, as the most strongly upregulated gene following heat shock at 45°C (Stewart et al., 2002). Expression of acr2 is subject to negative control by the heat shock regulator HspR (Stewart et al., 2002) and to positive control by the alternative sigma factors σΗ and σE, which are responsive to both heat and oxidative stress (Manganelli et al., 2001). Acr2 is also the most strongly upregulated gene following uptake of M. tuberculosis by macrophages (Schnappinger et al., 2003). Interestingly, acr2 is a paralogue of the DosR-controlled acr or hspX (Rv 2031c) (Sherman et al., 2001; Park et al., 2003) which is a prominent member of the hypoxia/NO regulon that is specifically induced in activated macrophages and has been implicated in mycobacterial persistence (Sherman et al., 2001; Voskuil et al., 2003).

The present study was designed to explore the role of α-crystallin chaperones in the biology of mycobacterial infection. We report on the distribution of α-crystallin genes in mycobacteria, the level of transcription of acr1 and acr2 genes during different stages of M. tuberculosis infection in mice, and the effect of acr2 deletion on the virulence of M. tuberculosis.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Three classes of α-crystallin genes in mycobacteria

The α-crystallin chaperones are structurally characterized by three domains, with variable N- and C-terminal domains flanking a conserved α-crystallin core domain of about 80 amino acids. The sequences of these proteins are generally less well conserved than those of other heat shock chaperones. A recent comparison of bacterial and archaeal α-crystallins revealed that as few as five amino acids were conserved in 80% of 60 sequences compared (Narberhaus, 2002). We searched five mycobacterial genomes for genes encoding α-crystallin proteins and compared these sequences with other bacterial orthologues and with the archetypal lens crystallins of the human eye. The alignments were made using clustal w and are displayed as an unrooted neighbour joining tree (Fig. 1). Consistent with the low sequence conservation characteristic of this protein family, the mycobacterial α-crystallins showed only 15–25% identity with orthologues from other bacterial genera or from humans. Similarity between paralogues tends to be much greater; for example, the two Escherichia coliα-crystallins share approximately 50% sequence identity and the three lactobacillus chaperones are 30–50% identical in pairwise comparisons. An exception to this is Bacillus subtilis in which the similarity of the three α-crystallins with themselves and with other α-crystallins is low, possibly because of the diversification of these proteins to functions such as spore formation (Narberhaus, 2002). The relationship between mycobacterial α-crystallins is more complex, with homologues falling into three distinct classes –acr1, acr2 and acr3– each sharing approximately 25–30% identity with the other two classes. The relationship between proteins within each class is closer; for example, M. tuberculosis acr2 shares 75%, 75% and 61% identity with its M. marinum, M. avium and M. smegmatis counterparts. Similarly M. tuberculosis acr1 shares 60% and 75% identity with respective orthologues in M. marinum and M. smegmatis.

image

Figure 1. Phylogenetic relationship of mycobacterial α-crystallins. Sequences were compared using clustal w and displayed as an unrooted neighbour-joining dendrogram. *Gene disrupted by insertion sequence.

Download figure to PowerPoint

Not all acr classes are represented in every mycobacterial species. M. tuberculosis possesses only an acr1 and acr2 whereas M. leprae has just a single acr3 (hsp18). M. marinum and M. smegmatis have one member of each acr class and M. avium possesses genes encoding one acr2 and four acr3 chaperones, although one of these is probably non-functional as a result of insertion of an IS1245 element.

Expression of acr2 and acr1 during murine infection

Expression of acr1 is upregulated by exposure to low oxygen conditions or to NO as part of the mycobacterial dormancy program (Voskuil et al., 2003). In contrast, acr2 expression is upregulated by oxidative stress (Schnappinger et al., 2003). This dichotomous response of the α-crystallins of M. tuberculosis stimulated us to investigate the expression pattern of the two genes during infection of C57Bl/6 mice. The mouse model of tuberculosis is characterized by a 2 week acute phase of infection during which the bacterial population expands, followed by an immune-mediated containment of infection and a chronic or persistent phase with a stable bacterial load (Stewart et al., 2003). RNA was extracted from infected lungs and spleens during the acute and chronic stages of infection, and real-time reverse-transcription PCR was used to assess the concentrations of mycobacterial acr1 and acr2 messenger RNA. Levels of message were expressed relative to 16S rRNA (Fig. 2) which has been shown to be proportional to bacterial load during murine infection (Shi et al., 2003). Acr1 expression in this model has been shown to increase concomitant with the production of interferon-γ (IFNγ) at the end of the acute phase of bacterial growth, with a subsequent decline during the chronic phase (Shi et al., 2003). Consistent with this finding, we observed a lower level of acr1 expression in infected lungs on day 63 in comparison to day 14. This reduction was not observed in the case of acr2 expression, and there was a 2.5-fold increase in the overall ratio of acr2 : acr1 in the lungs of mice during the chronic as compared to the acute phase of infection (P = 0.01) (Fig. 2A–C). Similarly, there was a threefold increase in acr2 : acr1 ratio in spleens taken at the two timepoints (P = 0.06) (Fig. 2D–F). These results demonstrate that the high level of expression of acr2 triggered by initial phagocytosis of M. tuberculosis is maintained throughout the different stages of infection in the murine model.

image

Figure 2. mRNA levels of acr1 and acr2 relative to 16S rRNA in lungs (A, B) and spleens (D, E) of acute (2 weeks) and chronic (9 weeks) M. tuberculosis-infected mice. Ratios of acr2 : acr1 are shown for lungs (C) and spleens (F). Error bars show the standard error of mean values from four mice; P-values were calculated by T-test. The experiment was repeated three times with similar results.

Download figure to PowerPoint

Effect of acr2 deletion on the virulence of M. tuberculosis

To generate a deletion mutant, a plasmid construct with the acr2 gene replaced by a hygromycin resistance determinant (pSMT183) was introduced into M. tuberculosis H37Rv by electroporation. Subsequent counter-selection in the presence of sucrose generated approximately 1000 transformants. Six colonies were grown in broth and in each case Southern blot analysis of genomic DNA gave hybridization patterns consistent with replacement of acr2 by the hyg gene (data not shown). Deletion of acr2 was further confirmed by Western blot using a rabbit polyclonal antiserum generated using recombinant Acr2 expressed as a His-tagged fusion protein in E. coli(Fig. 3). The antiserum bound strongly to Acr2, but showed no reactivity against Acr1. Western blotting against a whole cell homogenate of M. tuberculosis revealed a single band of approximately 17 kDa; this band was completely absent from the deletion mutant (Fig. 3).

image

Figure 3. Western blot demonstrating expression of Acr2 in M. tuberculosis H37Rv and absence of Acr2 protein in M. tuberculosis Δacr2. Reintroduction of acr2 on the integrating vector pSMT184 to complement the Δacr2 strain gives lower level of expression of Acr2 than in the wild type (Δacr2comp). The antiserum is not reactive against recombinant Acr1 (rAcr1). Equal loading concentrations were checked by running a duplicate gel for Coomassie staining, and even transfer of protein was verified by ponceau red staining of membranes after transfer. This figure is representative of several experiments.

Download figure to PowerPoint

To complement the Δacr2 strain we reintroduced a single copy of the acr2 gene together with its own promoter into the attB phage attachment site in the chromosome using the integrating construct pSMT184. Western blot analysis confirmed that Acr2 was expressed in the complemented strain, but at a level that was considerably lower than that seen in the wild-type strain (Fig. 3). This observation may indicate that we have not incorporated the entire acr2 transcriptional apparatus. The region that was cloned into the complementing construct includes the HspR binding site (−61 from the initiation codon) and the putative sE and σH joint recognition sequence (−94 from the initiation codon), but there may be other control elements involved. An alternative explanation is that the promoter is intact but that there is some feature of the natural chromosomal location of the acr2 gene that has an important influence on the level of expression and that is not reproduced by insertion in the attB site.

Expression of acr2 in M. tuberculosis is strongly upregulated following exposure to H2O2 or uptake by macrophages (Schnappinger et al., 2003). We have verified, by quantitative RT-PCR, the increase in acr2 transcription in both these environments (data not shown). To test whether upregulation is associated with an essential function, we compared survival of the mutant and wild-type strains in media containing H2O2 and over 4 days of culture in murine bone marrow-derived macrophages. In comparison to the wild type, the mutant survived significantly less well in both 1 mM (2.5-fold difference, P < 0.01) and 10 mM H2O2 (13-fold difference, P < 0.01) over 8 h (Fig. 4A). Reintroduction of acr2 in the complemented strain restored resistance to 1 mM H2O2 but not to 10 mM H2O2. Partial restoration of the resistant phenotype is consistent with the failure to restore the wild-type level of Acr2 expression in the complemented strain. No significant difference was observed between survival of the mutant and wild-type strains in the macrophage culture model (Fig. 4B).

image

Figure 4. A. Susceptibility to H2O2. The acr2-deleted M. tuberculosis strain (white bars) was more susceptible to killing by 1 mM and 10 mM H2O2 over 8 h than wild-type M. tuberculosis H37Rv (black bars) (both P < 0.01). The complemented Δacr2 mutant strain (grey bars) demonstrated an intermediate phenotype with a similar resistance to H37Rv at 1 mM H2O2 but decreased resistance at 10 mM (P < 0.01). Graph shows survival as a percentage of the original inocula. Error bars show standard error of the mean values from replicate wells. B. Survival in murine bone marrow-derived macrophages of M. tuberculosis H37Rv (black bars), M. tuberculosis Δacr2 (white bars) and the complemented mutant, Δacr2pSMT184 (grey bars). Values are mean cfu/well from four replicate wells and error bars show standard error of the mean. The experiment was repeated with similar results.

Download figure to PowerPoint

We next compared growth of the Δacr2 mutant and wild-type M. tuberculosis during the course of a 9 week infection in C57Bl/6 mice. No difference was found between the wild-type, the Δacr2 mutant, and the Δacr2pSMT184-complemented strain in terms of initial colonization, growth rate during the acute phase of infection, or bacterial load during the persistent phase in either lungs or spleens (Fig. 5A and B). Despite the similarity in bacterial load, visual inspection of mice infected with the Δacr2 mutant suggested that the disease pathology was less advanced than in those infected with the wild type. This was confirmed by observation of animals over a longer time-course, and is reflected in the weights of the different groups (Fig. 5C). After 20 weeks of infection, the Δacr2-infected mice were approximately 25% heavier than those infected with the wild-type M. tuberculosis; the complemented strain showed an intermediate phenotype.

image

Figure 5. In vivo phenotype of the acr2-deleted M. tuberculosis. C57Bl/6 mice were infected with the wild-type M. tuberculosis H37Rv (open circles), mutant Δacr2 (filled squares) and complemented mutant Δacr2pSMT184 (triangles). Bacterial load was measured in lungs (A) and spleens (B) over 9 weeks, and mouse body weight monitored over 20 weeks (C). There were no significant differences in bacterial load between any of the three strains at weeks 2, 4 or 9 (P > 0.05). There was a significant difference in the weight of mice infected with wild-type and Δacr2 strains at 15 weeks (P = 0.01) and at 20 weeks (P < 0.01) when the experiment had to be terminated because of severe weight loss in the group infected with wild-type M. tuberculosis. Mice infected with the complemented mutant were intermediate of the mutant and wild-type strains and were statistically distinct from both (each comparison P < 0.05). The experiment was repeated with similar results.

Download figure to PowerPoint

To investigate further the basis of the bacterial load-independent difference in pathology, we quantified the immune cell infiltration into the lungs elicited by the mutant and wild-type strains at 4 weeks post infection. We observed significantly more CD4+ T cells (Fig 6A, P = 0.013), CD8+ T cells (Fig 6B, P = 0.030) and CD11b+ cells (indicative of macrophages and granulocytes) (Fig 6C, P < 0.004) in the lungs of wild-type compared to Δacr2 mutant-infected mice. Further analysis revealed higher numbers of activated (CD45RBlo) CD4+ and CD8+ T cells in the lungs of wild-type infected mice, although the relative proportions of activation were similar between the groups (data not shown). Similarly there was no difference in the proportion of major histocompatibility complex (MHC) class II expression on CD11b+ cells, suggesting there is no qualitative reduction in antigen presenting cell activation during infection with the Δacr2 mutant, but simply that their overall numbers in the lung are reduced. Furthermore, examination of whole lungs and tissue sections revealed that the granulomas of Δacr2 mutant-infected mice did not enlarge at the same rate as wild-type granulomas through the course of infection. At 15 weeks post infection when the weight difference between mice was well established, the bacterial load was again measured to be equivalent between strains but the pathology of lungs was markedly different. Lungs from mice infected with the Δacr2 strain had relatively discrete lesions in contrast to the advanced and almost completely coalesced lesions in the wild-type infection, typical of the breakdown of the granulomatous response in the late stages of murine tuberculosis (Fig. 7).

image

Figure 6. Infection with wild-type M. tuberculosis H37RV elicits greater cellular infiltration of the lungs than the Δacr2 mutant strain. At 4 weeks post infection, flow cytometry revealed that the lungs of mice infected with the wild-type strain contain larger numbers of CD4+ T cells (A, P = 0.013), CD8+ T cells (B, P = 0.030) and CD11b+ cells (C, P = 0.004) than the lungs of Δacr2-infected mice. Mice infected with the complemented mutant Δacr2pSMT184 gave results intermediate between the wild type and Δacr2.

Download figure to PowerPoint

image

Figure 7. Gross pathology of lungs in mice infected with M. tuberculosis H37Rv wild-type and mutant Δacr2 strains. Two representative left lung lobes from each group at 15 weeks post infection are shown. Δacr2-infected lungs have discrete lesions whereas wild-type lungs have more diffuse lesions which have coalesced to fill most of the lung.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Members of the α-crystallin family of molecular chaperones are widespread through prokaryotic and eukaryotic organisms and function primarily to prevent the irreversible aggregation of proteins during stress conditions such as elevated temperature, low or high pH and oxidative conditions (Horwitz, 1992; Jofre et al., 2003; Takeuchi et al., 2003). In most bacteria, the expression of α-crystallins is negligible under normal growth conditions but when stressed their expression can be increased greatly. This holds true in M. tuberculosis with microarray gene expression experiments, indicating that acr2 is the most upregulated gene during heat shock (Stewart et al., 2002) and acr1 one of the most upregulated genes under hypoxic conditions (Sherman et al., 2001). In addition, it has recently been demonstrated that acr2 is also the most highly induced gene following phagocytosis of M. tuberculosis by resting or activated macrophages; acr1 is upregulated specifically in activated macrophages (Schnappinger et al., 2003). We reasoned that such a prominent part of the mycobacterial stress response warranted a more detailed investigation.

To begin to understand the biology of these molecules we made a comparison of predicted α-crystallin proteins in five mycobacterial species. On the basis of sequence similarity, the mycobacterial chaperones fell into three classes. To rationalize the terminology of mycobacterial α-crystallins we have called these classes Acr1, Acr2 and Acr3. It does not appear that pathogenicity depends on a particular class of α-crystallin as M. tuberculosis contains an acr1 (previously referred to as acr, or hspX) and an acr2 gene, M. leprae a single acr3, and M. avium one acr2 and three functional acr3 genes. The genetic complement of M. leprae has been extensively depleted as a result of reductive evolution (Cole et al., 2001) and retention of acr3 suggests that mycobacteria have a fundamental requirement for at least one α-crystallin gene. The mycobacterial species which reside outside of temperature-controlled environments are replete with α-crystallin genes. M. avium which is a facultative pathogen found in the environment has four functional acr genes, while M. smegmatis (a saprophyte) and M. marinum (a pathogen of cold blooded marine animals) contain representatives of each of the three acr classes. Perhaps this diversity allows these species to respond to a greater range of environmental conditions.

The Acr1 α-crystallin of M. tuberculosis has received considerable attention in the context of the dramatic increase in its expression associated with bacterial adaptation to a non-replicating, ‘dormant’ phenotype in a hypoxic environment (Yuan et al., 1996; Boon et al., 2001). Expression of acr1 is controlled by a two-component regulator, DosR, which is triggered by exposure to low oxygen tension or to nitric oxide within activated macrophages (Sherman et al., 2001; Wayne and Sohaskey, 2001; Park et al., 2003; Voskuil et al., 2003). Consistent with this, expression of acr1 has been shown to correlate with expression of IFNγ and inducible nitric oxide synthase (NOS2) in lungs at the transition from the acute to the chronic phase of murine tuberculosis (Shi et al., 2003). As acr2 is induced in quiescent macrophages and acr1 is specifically induced in activated macrophages, we anticipated that there may be a sequential induction of the two α-crystallins in the mouse model, with acr2 in the early acute phase replaced by acr1 following onset of the adaptive immune response. In fact, acr2 expression was maintained throughout the infection, with a 2.5- to threefold increase in the ratio of acr2 : acr1 in the chronic phase. The contribution of Acr2 is not restricted to the early stages of infection therefore, but may be important during persistent infection. As α-crystallins can form hetero-oligomeric complexes (Zantema et al., 1992; Studer and Narberhaus, 2000), it is possible that Acr1 and Acr2 are at least partially interchangeable in terms of their chaperone function.

To test whether the prominent induction of acr2 expression was associated with a functional contribution to the process of infection, we generated a mutant strain of M. tuberculosis from which the acr2 gene had been deleted. The Δacr2 mutant was significantly impaired in its resistance to H2O2, but we were unable to detect any change in its ability to survive in murine bone marrow macrophages. These results illustrate the risk of inferring a direct link between the induction of expression of a gene and its functional essentiality under particular set of conditions. Consistent with results in the macrophage culture, growth of the Δacr2 mutant was unimpaired during the initial acute phase of infection in the mouse model. At later stages of infection, there was no difference between mutant and wild type in terms of bacterial load in tissues, but monitoring of weight loss over 20 weeks demonstrated a significant reduction in the overall severity of the disease caused by the Δacr2 mutant. Complementation of the mutant by reintroduction of a single copy of the acr2 gene in the attB site on the chromosome resulted in partial restoration of Acr2 expression along with partial reversal of the H2O2 and mouse virulence phenotypes.

Reduced pathology in the absence of any detectable change in bacterial load has previously been reported for M. tuberculosis mutants lacking the sigH (Kaushal et al., 2002) and whiB3 (Steyn et al., 2002) genes, and differences in the rate of progression of pathology have been observed between clinical isolates of M. tuberculosis that differ in their ability to elicit a proinflammatory immune response (Manca et al., 2001). One possible explanation for the reduced pathology phenotype of the Δacr2 mutant is that the Acr2 protein itself elicits some detrimental form of immune response that promotes tissue destruction without affecting the bacterial population. We cannot discount this mechanism, but analysis of the human T cell response to Acr2 suggests an association with protection rather than pathology (K. A. Wilkinson et al., in preparation). A related hypothesis would be that the loss of Acr2 chaperone function causes a significant change in the repertoire of mycobacterial peptides that are made available for immune activation. An alternative explanation is that the difference in pathology is an indirect consequence of an effect of the mutation on bacterial physiology. If the loss of Acr2 shifts the balance within the persistent population from actively replicating bacteria controlled by immune killing towards a higher proportion of the bacteria in a non-replicating state, for example, this might result in a reduction in the progressive accumulation of immune mediated damage. Consistent with this hypothesis we demonstrate that the Δacr2 mutant elicits less recruitment of T cells and CD11b+ cells to the lung during the early stages of infection. As infection progresses this reduced immune response is manifest in smaller granulomatous lesions and a delayed progression to the end-point of infection where lung function is severely compromised and the mice rapidly lose weight. Further engineering of the α-crystallin content – for example, by deletion of both acr genes, or by switching the regulatory control of the two acr classes – may be useful in exploring strategies for manipulation of the virulence and immunogenicity of M. tuberculosis.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Bacterial strains and growth conditions

Escherichia coli DH5α and HMS174 (DE3) were grown at 37°C in Luria–Bertani broth and agar containing 150 µg ml−1 hygromycin or 50 µg ml−1 carbenicillin as appropriate. M. tuberculosis H37Rv was grown at 37°C in Middlebrook 7H9 broth (Difco) or on Middlebrook 7H11 medium (Difco) containing 10% albumin/dextrose/catalase (ADC) enrichment and 10% oleic acid/albumin/dextrose/catalase (OADC) enrichment respectively, and 50 µg ml−1 hygromycin or 15 µg ml−1 kanamycin as appropriate.

Production of recombinant Acr2 and generation of antiserum

The open reading frame encoding acr2 (Rv0251c) was amplified by PCR from M. tuberculosis genomic DNA using HF Expand polymerase (Roche) and the primers acr2For (5′-cccatatgaacaatctcgcattgtggtcgc-3′) and acr2Rev (5′-ccctc gagctacttcgtgatggcgatgcgc-3′). The product was subcloned into TOPO (Invitrogen) then excised with Xho1 and Nde1 and cloned into the His-Tag fusion expression vector pET15B (Novagen) to make pETACR2. pETACR2 was transformed into E. coli HMS174 (DE3). One litre cultures containing carbenicillin were grown to an OD of 0.5 and expression induced by addition of 0.1 mM IPTG for 2 h at 37°C. The cells were pelleted by centrifugation and lysed using Bugbuster (Novagen) as per manufacturer's instructions. The lysate was centrifuged and soluble His-tagged Acr2 was isolated on a Talon (Clontech) metal affinity column and assessed by SDS-PAGE.

New Zealand white rabbits were inoculated with 200 µg recombinant Acr2 in Freund's incomplete adjuvant and boosted after 3, 6 and 9 weeks. Specificity of the antiserum was tested by Western blot against 10 ng recombinant Acr2, 10 ng recombinant Acr1 (LIONEX, Braunschweig, Germany) and 20 µg M. tuberculosis homogenate separated by SDS-PAGE.

Genetic deletion and subsequent complementation of acr2 in M. tuberculosis

Regions of DNA flanking acr2 (Rv0251c) were PCR-amplified from genomic DNA using HF Expand polymerase and the primer pairs ACR1 (5′-cgggatccttgagtttggtcgacaggtc-3′)/ACR2 (5′-gctctagaaccggacgcgaccacaatgc-3′) and ACR3 (5′-gaagatctcagcgcatcgccatcacgaa-3′)/ACR4 (5′-gctctagacggca cgaatggcacgagat-3′) for up and downstream regions respectively. The regions were cloned around the hygromycin resistance gene (hyg) in the suicide delivery vector pSMT100. The resulting plasmid, pSMT183, was electroporated into M. tuberculosis H37Rv and gene replacement transformants (Δacr2) selected as previously described (Dussurget et al., 2001). Gene replacement of acr2 with hyg was confirmed by Southern blot of Kpn1-digested genomic DNA probed with pSMT184 vector (see below), and Western blot of SDS-PAGE separated whole cell lysates probed with the rabbit polyclonal Acr2 antiserum.

To complement the deletion of acr2 we chose to reintroduce a single copy of the acr2 gene with its own promoter. The acr2 open reading frame and 208 bp of upstream sequence were amplified from genomic DNA by PCR using the primers ACRcomp1 (cgcgagctcgagggtatgaggggcaaatt) and ACRcomp2 (cgcgagctcaactacttcgtgatggcgat). The product was digested with the Sac1 restriction enzyme and cloned into the integrating vector pKinta (Stewart et al., 2001). The resulting vector, pSMT184, was electroporated into M. tuberculosis Δacr2 and transformants selected by plating on 7H11 containing kanamycin.

Infection models

Mycobaterium tuberculosis strains were grown to mid-logarithmic phase, pelleted by centrifugation, washed and resuspended in saline at 1 × 106 cfu ml−1 6–8 week old female C57Bl/6 mice were infected by the intravenous route with 2 × 105 cfu M. tuberculosis H37Rv, Δacr2 or Δacr2pSMT184. Groups of four mice per strain were culled 1 day post infection and after 2, 4 and 9 weeks. Bacterial load was assessed by homogenization of spleens and lungs and plating of serial dilutions on Middlebrook 7H11. To assess disease progression, groups of nine mice infected with each strain were maintained for 20 weeks and weighed at regular intervals.

For immunological analysis groups of four infected mice were culled at 4 weeks post infection. Single-cell lung suspensions were prepared by homogenizing the tissues through 100 mm cell strainers. Cells were stained with FITC-conjugated anti CD45RB, peridinin-chlorophyl-protein (PerCP)-conjugated anti-CD4 and CD11b, allophycocyanin (APC)-conjugated CD8, and phycoerythrin (PE)-conjugated anti MHC II (I-A/I-E) (all from Pharmingen) for 30 min on ice. The samples were washed in PBS containing 1% BSA and fixed for 16 h in 4% paraformaldehyde in PBS. The samples were rinsed in PBS and then analysed on a FACSCalibur flow cytometer collecting data on at least 40 000 cells.

Murine bone marrow-derived macrophages were cultivated and infected with the three M. tuberculosis strains as previously described (Dussurget et al., 2001) but using macrophage SFM medium (Gibco, Invitrogen) supplemented with 40 ng ml−1 recombinant murine M-CSF (R and D Systems). Macrophages were seeded at 5 × 105 per well in 24-well plates and infected at an moi of 1.

Resistance to hydrogen peroxide

Approximately 1 × 106 log phase M. tuberculosis H37Rv, Δacr2 or Δacr2pSMT184 were added to 7H9 medium without catalase containing 1 mM or 10 mM H2O2 and incubated at 37°C for 8 h. The survival of bacteria was assessed by plating out on 7H11 agar samples at 0 and 8 h incubation.

Quantitative RT-PCR of acr1 and acr2 transcripts during acute and chronic murine infection

Mice were culled and approximately 100 mg sections of spleen and lung were immediately removed to BiopulverizerTM matrix B tubes (BIO 101) containing 1 ml of Trisol (Invitrogen) and pulsed for 45 s at maximum speed in a ribolyser (Hybaid). RNA was extracted according to Trizol manufacturer's guidelines (Invitrogen) and resuspended in a final volume of 40 µl. Contaminating genomic DNA was removed by addition of 2 µl of DNase1 and 4.7 µl 10× buffer (Invitrogen). Finally the RNA was cleaned up using the RNeasy kit (Qiagen).

To assess acr1 and acr2 transcript levels in infected tissues we used quantitative real-time RT-PCR with internal fluorescent hybridization probes in an ABI Prism 7700 Sequence Detection System (Applied Biosystems, Foster City, CA). Specific RT and PCR primers and a probe for acr2 were designed according to specifications recommended by ABI/PE Biosystems using Primer Express® software. These sequences were as follows: RT primer (−266RT) 5′-gcgtgtgctcgtcg-3′; Forward primer (160F) 5′-gcggtggtccgtttgga-3′; Reverse primer (227R) 5′-ccagggtcaagctcgacgtt-3′; Probe (181T) 5′-FAM-cccggcattgacgtcgacaagg-3′TAMRA. Sequences used to detect M. tuberculosis acr1 and 16S were as described previously (Desjardin et al., 1999; 2001). All probes were dually labelled with 5 carbofluorescein (FAM) at the 5′ end and N,N,N′,N′-tetramethyl-6-carborhodamine (TAMRA) at the 3′ end. The proximity of the dye (FAM) and the quencher (TAMRA) on the intact probe prevents detection of any fluorescence. However, degradation of the probe during the course of PCR allows the release and detection of FAM. Multiplex reverse transcription was performed for 1 h at 37°C using 500 n M of each specific RT primer, 10 U µl−1 of SuperscriptII reverse transcriptase (Invitrogen) in its own reaction buffer, in the presence of 500 µM of each dNTP. For PCR, 5 µl of each cDNA was assayed in a total reaction volume of 25 µl containing Taqman® Universal PCR Master Mix (Applied Biosystems). Primers and probes were used at the following concentrations: 16S forward and reverse primers at 250 n M and the probe at 100 n M; acr1 probe at 200 n M, forward primer at 150 n M and the reverse at 50 n M; and the acr2 probe 100 n M, forward primer at 50 n M and the reverse at 300 n M. Reaction conditions consisted of 1 cycle of 50°C for 2 min and 1 cycle of 95°C for 10 min and then 40 cycles of 95°C for 15 s followed by annealing and elongation at 65°C (16S), 67°C (acr1), or 60°C (acr2) for 1 min. The cycle threshold (CT) for each sample was compared with CT values of known amounts of a standard DNA from M. tuberculosis H37Rv. To assure lack of DNA contamination of the RNA samples, a duplicate tube of sample with no RT enzyme was included as control. Results are expressed as values normalized to the 16S rRNA content.

Computer analysis of α-crystallin genes

Genes encoding α-crystallin family proteins were retrieved from annotated genome sequences held at The Institute for Genome Research and at the NCBI. α-Crystallin family genes were identified in the unfinished genomes of M. smegmatis and M. avium 104 using tblastn (Altschul et al., 1997). Predicted protein sequences were aligned using clustal w and are displayed as an unrooted neighbour-joining dendrogram.

Statistical analysis

All statistical comparisons were performed using Student's t-test and significance inferred at P < 0.05.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

This work was supported by a Wellcome Trust Programme Grant to D.B.Y. and B.D.R., by Personal Fellowship support to R.J.W. (064261) and by a Wellcome Trust centre award (060079).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D.J. (1997) Gapped blast and psi-blast: a new generation of protein database search programs. Nucleic Acids Res 25: 33893402.
  • Bloom, B.R., and Murray, C.J. (1992) Tuberculosis: commentary on a reemergent killer. Science 257: 10551064.
  • Boon, C., Li, R., Qi, R., and Dick, T. (2001) Proteins of Mycobacterium bovis BCG induced in the Wayne dormancy model. J Bacteriol 183: 26722676.
  • Cole, S.T., Eiglmeier, K., Parkhill, J., James, K.D., Thomson, N.R., Wheeler, P.R., et al. (2001) Massive gene decay in the leprosy bacillus. Nature 409: 10071011.
  • Darwin, K.H., Ehrt, S., Gutierrez-Ramos, J.C., Weich, N., and Nathan, C.F. (2003) The proteasome of Mycobacterium tuberculosis is required for resistance to nitric oxide. Science 302: 19631966.
  • Desjardin, L.E., Perkins, M.D., Wolski, K., Haun, S., Teixeira, L., Chen, Y., et al. (1999) Measurement of sputum Mycobacterium tuberculosis messenger RNA as a surrogate for response to chemotherapy. Am J Respir Crit Care Med 160: 203210.
  • Desjardin, L.E., Hayes, L.G., Sohaskey, C.D., Wayne, L.G., and Eisenach, K.D. (2001) Microaerophilic induction of the alpha-crystallin chaperone protein homologue (hspX) mRNA of Mycobacterium tuberculosis. J Bacteriol 183: 53115316.
  • Dussurget, O., Stewart, G., Neyrolles, O., Pescher, P., Young, D., and Marchal, G. (2001) Role of Mycobacterium tuberculosis copper-zinc superoxide dismutase. Infect Immun 69: 529533.
  • Dye, C., Scheele, S., Dolin, P., Pathania, V., and Raviglione, M.C. (1999) Consensus statement. Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country. WHO Global Surveillance and Monitoring Project. JAMA 282: 677686.
  • Hmama, Z., Gabathuler, R., Jefferies, W.A., De Jong, G., and Reiner, N.E. (1998) Attenuation of HLA-DR expression by mononuclear phagocytes infected with Mycobacterium tuberculosis is related to intracellular sequestration of immature class II heterodimers. J Immunol 161: 48824893.
  • Horwitz, J. (1992) Alpha-crystallin can function as a molecular chaperone. Proc Natl Acad Sci USA 89: 1044910453.
  • Jofre, A., Molinas, M., and Pla, M. (2003) A 10-kDa class-CI sHsp protects E. coli from oxidative and high-temperature stress. Planta 217: 813819.
  • Kaushal, D., Schroeder, B.G., Tyagi, S., Yoshimatsu, T., Scott, C., Ko, C., et al. (2002) Reduced immunopathology and mortality despite tissue persistence in a Mycobacterium tuberculosis mutant lacking alternative sigma factor, SigH. Proc Natl Acad Sci USA 99: 83308335.
  • Lee, B.Y., and Horwitz, M.A. (1995) Identification of macrophage and stress-induced proteins of Mycobacterium tuberculosis. J Clin Invest 96: 245249.
  • Li, Z., Kelley, C., Collins, F., Rouse, D., and Morris, S. (1998) Expression of katG. Mycobacterium tuberculosis is associated with its growth and persistence in mice and guinea pigs. J Infect Dis 177: 10301035.
  • Manca, C., Tsenova, L., Bergtold, A., Freeman, S., Tovey, M., Musser, J.M., et al. (2001) Virulence of a Mycobacterium tuberculosis clinical isolate in mice is determined by failure to induce Th1 type immunity and is associated with induction of IFN-alpha/beta. Proc Natl Acad Sci USA 98: 57525757.
  • Manganelli, R., Voskuil, M.I., Schoolnik, G.K., and Smith, I. (2001) The Mycobacterium tuberculosis ECF sigma factor σE: role in global gene expression and survival in macrophages. Mol Microbiol 41: 423437.
  • Master, S.S., Springer, B., Sander, P., Boettger, E.C., Deretic, V., and Timmins, G.S. (2002) Oxidative stress response genes in Mycobacterium tuberculosis: role of ahpC in resistance to peroxynitrite and stage-specific survival in macrophages. Microbiology 148: 31393144.
  • Monahan, I.M., Betts, J., Banerjee, D.K., and Butcher, P.D. (2001) Differential expression of mycobacterial proteins following phagocytosis by macrophages. Microbiology 147: 459471.
  • Narberhaus, F. (2002) Alpha-crystallin-type heat shock proteins: socializing minichaperones in the context of a multichaperone network. Microbiol Mol Biol Rev 66: 6493.
  • Noss, E.H., Pai, R.K., Sellati, T.J., Radolf, J.D., Belisle, J., Golenbock, D.T., et al. (2001) Toll-like receptor 2-dependent inhibition of macrophage class II MHC expression and antigen processing by 19-kDa lipoprotein of Mycobacterium tuberculosis. J Immunol 167: 910918.
  • Park, H.D., Guinn, K.M., Harrell, M.I., Liao, R., Voskuil, M.I., Tompa, M., et al. (2003) Rv3133c/dosR is a transcription factor that mediates the hypoxic response of Mycobacterium tuberculosis. Mol Microbiol 48: 833843.
  • Piddington, D.L., Fang, F.C., Laessig, T., Cooper, A.M., Orme, I.M., and Buchmeier, N.A. (2001) Cu,Zn superoxide dismutase of Mycobacterium tuberculosis contributes to survival in activated macrophages that are generating an oxidative burst. Infect Immun 69: 49804987.
  • Russell, D.G. (2001) Mycobacterium tuberculosis: here today, and here tomorrow. Nat Rev Mol Cell Biol 2: 569577.
  • Schnappinger, D., Ehrt, S., Voskuil, M.I., Liu, Y., Mangan, J.A., Monahan, I.M., et al. (2003) Transcriptional adaptation of Mycobacterium tuberculosis within macrophages: insights into the phagosomal environment. J Exp Med 198: 693704.
  • Sherman, D.R., Voskuil, M., Schnappinger, D., Liao, R., Harrell, M.I., and Schoolnik, G.K. (2001) Regulation of the Mycobacterium tuberculosis hypoxic response gene encoding alpha-crystallin. Proc Natl Acad Sci USA 98: 75347539.
  • Shi, L., Jung, Y.J., Tyagi, S., Gennaro, M.L., and North, R.J. (2003) Expression of Th1-mediated immunity in mouse lungs induces a Mycobacterium tuberculosis transcription pattern characteristic of nonreplicating persistence. Proc Natl Acad Sci USA 100: 241246.
  • Stewart, G.R., Snewin, V.A., Walzl, G., Hussell, T., Tormay, P., O'Gaora, P., et al. (2001) Overexpression of heat-shock proteins reduces survival of Mycobacterium tuberculosis in the chronic phase of infection. Nat Med 7: 732737.
  • Stewart, G.R., Wernisch, L., Stabler, R., Mangan, J.A., Hinds, J., Laing, K.G., et al. (2002) Dissection of the heat-shock response in Mycobacterium tuberculosis using mutants and microarrays. Microbiology 148: 31293138.
  • Stewart, G.R., Robertson, B.D., and Young, D. (2003) Tuberculosis: a problem with persistence. Nat Rev Microbiol 1: 97105.
  • Steyn, A.J., Collins, D.M., Hondalus, M.K., Jacobs, W.R., Kawakami, R.P., and Bloom, B.R. (2002) Mycobacterium tuberculosis WhiB3 interacts with RpoV to affect host survival but is dispensable for in vivo growth. Proc Natl Acad Sci USA 99: 31473152.
  • Studer, S., and Narberhaus, F. (2000) Chaperone activity and homo- and hetero-oligomer formation of bacterial small heat shock proteins. J Biol Chem 275: 3721237218.
  • Takeuchi, S., Mandai, Y., Otsu, A., Shirakawa, T., Masuda, K., and Chinami, M. (2003) Differences in properties between human alphaA- and alphaB-crystallin proteins expressed in Escherichia coli cells in response to cold and extreme pH. Biochem J 375: 471475.
  • Ting, L.M., Kim, A.C., Cattamanchi, A., and Ernst, J.D. (1999) Mycobacterium tuberculosis inhibits IFN-gamma transcriptional responses without inhibiting activation of STAT1. J Immunol 163: 38983906.
  • Voskuil, M.I., Schnappinger, D., Visconti, K.C., Harrell, M.I., Dolganov, G.M., Sherman, D.R., and Schoolnik, G.K. (2003) Inhibition of respiration by nitric oxide induces a Mycobacterium tuberculosis dormancy program. J Exp Med 198: 705713.
  • Wayne, L.G., and Sohaskey, C.D. (2001) Nonreplicating persistence of Mycobacterium tuberculosis. Annu Rev Microbiol 55: 139163.
  • Young, D.B., and Garbe, T.R. (1991) Heat shock proteins and antigens of Mycobacterium tuberculosis. Infect Immun 59: 30863093.
  • Yuan, Y., Crane, D.D., and Barry, C.E., (1996) 3rd Stationary phase-associated protein expression in Mycobacterium tuberculosis: function of the mycobacterial alpha-crystallin homolog. J Bacteriol 178: 44844492.
  • Zantema, A., Verlaan-De Vries, M., Maasdam, D., Bol, S., and Van Der Eb, A. (1992) Heat shock protein 27 and alpha Β-crystallin can form a complex, which dissociates by heat shock. J Biol Chem 267: 1293612941.