Enhanced priming of adaptive immunity by a proapoptotic mutant of Mycobacterium tuberculosis

The inhibition of apoptosis of infected host cells is a well-known but poorly understood function of pathogenic mycobacteria. We show that inactivation of the secA2 gene in Mycobacterium tuberculosis, which encodes a component of a virulence-associated protein secretion system, enhanced the apoptosis of infected macrophages by diminishing secretion of mycobacterial superoxide dismutase. Deletion of secA2 markedly increased priming of antigen-specific CD8⁺ T cells in vivo, and vaccination of mice and guinea pigs with a secA2 mutant significantly increased resistance to M. tuberculosis challenge compared with standard M. bovis bacille Calmette-Guérin vaccination. Our results define a mechanism for a key immune evasion strategy of M. tuberculosis and provide what we believe to be a novel approach for improving mycobacterial vaccines.

Mycobacterium tuberculosis infects more than one-third of the world’s population and causes 2–3 million deaths annually (1).The only currently approved vaccine for tuberculosis (TB) is bacille Calmette-Guérin (BCG), which is an attenuated strain of M. bovis. This vaccine has been widely administered in TB-endemic regions, with little apparent impact on the prevalence of adult pulmonary TB, the most common form of disease caused by M. tuberculosis (2). Given the difficulties inherent in antibiotic treatment of TB (i.e., cost and long duration of continuous therapy), the increasing prevalence of active disease due to HIV coinfection, and the emergence of multidrug resistance, there is an urgent need for the development of new and more effective candidate vaccines (3).

Effective host control of M. tuberculosis is critically dependent on both innate and adaptive immune responses, and detailed studies in mouse models show that priming of both CD4⁺ and CD8⁺ T cells is important for long-term control of latent infection and for vaccine-induced protection (4). However, many lines of evidence indicate that M. tuberculosis has evolved mechanisms to evade host immunity, in some cases by inhibiting the priming and effector functions of these T cell subsets (5). One property of M. tuberculosis that has been linked to evasion of host immune responses is its ability to prevent the death by apoptosis that is typically triggered upon infection of phagocytic cells by vacuolar bacterial pathogens (6–8). Whereas macrophages infected by nonpathogenic species of mycobacteria are observed to rapidly undergo apoptosis, virulent mycobacteria such as M. tuberculosis and M. bovis (including BCG) do not efficiently trigger this response (9–12). In vitro studies indicate that the death of infected cells by apoptosis contributes to clearance of the bacteria through activation of as-yet-undefined bactericidal mechanisms (10, 11). In addition, apoptosis of infected cells has been shown to facilitate the transfer of mycobacterial antigens for presentation to T cells through MHC class I and CD1 molecules, which has been designated the “detour pathway” for antigen cross-presentation (13–15). Thus, the inhibition of the apoptotic response of infected host cells by M. tuberculosis is likely to interfere with both innate and adaptive immunity against this pathogen and most likely represents an important immune evasion strategy of this bacterium.

Although previous studies clearly document the antiapoptotic properties of M. tuberculosis, little is known regarding the mechanism of this immune evasion strategy, nor have any specific genes of M. tuberculosis controlling this aspect of virulence been reported. Here we show that the secA2 gene of M. tuberculosis, which encodes a component of a virulence-associated bacterial protein secretion system, is required for prevention of apoptosis of M. tuberculosis–infected macrophages. Our analyses showed that this was attributable to SecA2-dependent secretion of mycobacterial superoxide dismutase A (SodA). Infection with a secA2 deletion mutant of M. tuberculosis not only led to markedly increased levels of apoptosis by infected macrophages, but was also associated with significant enhancement in the priming of naive antigen-specific CD8⁺ T cells in vivo, suggesting increased engagement of the apoptosis-dependent detour pathway for MHC class I–restricted cross presentation. Vaccination of mice or guinea pigs with the secA2-deleted strain generated a level of protective host immunity against challenge with virulent M. tuberculosis that was significantly greater according to several criteria than the protection induced by BCG vaccination. These findings suggest a novel and practical approach to improving live TB vaccine strains by genetic modifications that disrupt their ability to interfere with host cell apoptosis.

Essential role of the secA2 gene of M. tuberculosis in preventing apoptosis of infected macrophages. The secA2 gene (Rv1821) encodes an accessory secretion factor in M. tuberculosis that is required for the export of a small subset of proteins that includes SodA, an M. tuberculosis enzyme that catalyzes conversion of superoxide anions to hydrogen peroxide (16). Whereas deletion of secA2 from M. tuberculosis has no effect on growth in culture, it does result in partial growth attenuation in vivo. This may be explained in part by the loss of SodA secretion, which would be expected to result in a reduced ability of the intracellular bacteria to detoxify superoxide anions generated in host cells in response to infection (17). However, overproduction of superoxide anions is also linked to induction of apoptosis in a variety of experimental systems (18), which suggested that another important function of secreted mycobacterial SodA may be to inhibit the infection-induced apoptosis of host cells. Indeed, when compared with nonpathogenic species of mycobacteria that are known to be proapoptotic, pathogenic mycobacteria such as M. tuberculosis express markedly higher levels of SodA (19), and antisense inhibition of SodA expression in M. tuberculosis results in reduced bacterial growth and higher levels of apoptosis in tissues of infected mice (20).

To directly examine a role for secA2 in the inhibition of apoptosis, transformed human monocyte/macrophage cells (THP-1 cells) or primary mouse BM-derived macrophages (BMMs) (C57BL/6 or BALB/c strains) were infected at an MOI of 10 bacilli per macrophage with WT M. tuberculosis (strain H37Rv), a mutant of H37Rv with a secA2 deletion (ΔsecA2), or a complemented strain of the mutant (ΔsecA2-C). After 72 hours, quantitative assessment of apoptosis by staining for DNA strand breakage (TUNEL) and polycaspase activation showed that ΔsecA2 strongly induced both of these markers of apoptosis in both human and murine macrophages, whereas WT M. tuberculosis did not (Figure 1, A and B, and Supplemental Figure 1; supplemental material available online with this article; doi:10.1172/JCI31947DS1). The enhanced TUNEL staining was fully reversed with ΔsecA2-C, which behaved identically to the WT strain (Figure 1, A and B).

Figure 1

Induction of caspase and SodA-dependent apoptosis in macrophages by infection with M. tuberculosis ΔsecA2. (A) Human THP-1 cells were infected with M. tuberculosis (H37Rv), ΔsecA2, and complemented ΔsecA2 (ΔsecA2-C) strains at an MOI of 10. Cells were stained after 48 and 72 hours for TUNEL or for caspase activation using FLICA probes and analyzed by FACS. Representative histograms of staining with TUNEL or a polycaspase-specific probe (fluorescein-VAD-FMK) at 48 hours are shown (open histogram, uninfected; filled histogram, infected with H37Rv or ΔsecA2 as indicated). Quantitation by FACS of cells positive for TUNEL or for active caspase-8, caspase-9, or polycaspase is summarized. Data shown are from the time points for peak signal detection (48 hours for TUNEL and polycaspase, 72 hours for caspase-8 and 9). (B) TUNEL staining of mouse BMMs 72 hours after infection with indicated bacterial strains. Staining was quantitated by FACS as in A. (C) TUNEL staining of THP-1 cells infected with the indicated bacteria for 48 hours in the absence or presence of caspase-3–specific inhibitor (Z-DEVD-FMK). *P < 0.001 compared to no inhibitor (1-way ANOVA, Bonferroni post-test). (D) Filtrates from cultures of H37Rv, ΔsecA2 mutant, or ΔsecA2-αsodA were analyzed for SodA activity (left). Graph shows mean and range for duplicate samples, with SodA activity normalized to the level of WT (H37Rv). ^†Below detection level. The ΔsecA2-αsodA showed restoration of SodA activity in the culture filtrate and also reversed the TUNEL⁺ cell death induced by ΔsecA2 in THP-1 cells infected for 48 hours at an MOI of 10:1 (right).

Caspases are known to be activated by infection with virulent as well as nonvirulent strains of M. tuberculosis (10, 21), although under certain conditions, induction of macrophage death by M. tuberculosis has been reported to proceed in a caspase-independent manner (22). The involvement of caspases in apoptosis induced by ΔsecA2M. tuberculosis was evaluated by measuring total caspase activity and activities of caspases 8 and 9 at 48 and 72 hours after infection in THP-1 cells (Figure 1A, and data not shown). Caspases 8 and 9 were selected because they are known to be initiators of the extrinsic (receptor-mediated) and intrinsic (mitochondrial) cell death pathways, respectively (23). The 2 caspases were equally activated in cells infected with ΔsecA2 but not in cells infected with either WT M. tuberculosis or the complemented ΔsecA2-C. This suggested the involvement of both extrinsic and intrinsic caspase-dependent pathways for initiation of apoptosis by ΔsecA2. Furthermore, treatment of THP-1 cells with an irreversible specific inhibitor of caspase-3 prior to infection with the ΔsecA2 mutant strongly reduced TUNEL staining, indicating that the macrophage cell death induced by ΔsecA2 was dependent on the activation of this downstream caspase (Figure 1C).

Reversal of proapoptotic phenotype of M. tuberculosis ΔsecA2 by restoration of SodA secretion. In contrast to mycobacterial proteins that undergo signal peptide–dependent secretion by the secA1-dependent pathway, SodA does not possess a recognizable N-terminal signal sequence. We therefore reasoned that extending the N terminus of SodA with a signal sequence from a mycobacterial protein secreted by a secA2-independent pathway might restore SodA secretion in the ΔsecA2 mutant and reverse its apoptosis-inducing phenotype. This was tested by extending the 5′ end of the coding sequence for WT SodA with a sequence encoding the N-terminal signal sequence of Ag85b, a major M. tuberculosis protein that is secreted in a secA1-dependent and secA2-independent manner (16). Expression of this construct in the ΔsecA2 mutant (ΔsecA2-αsodA) restored SodA activity in the culture filtrate and also reversed the induction of apoptosis in infected THP-1 cells (Figure 1D). This result indicated that secretion of SodA was likely to be the major secA2-dependent process involved in the inhibition of host cell apoptosis.

Deletion of secA2 enhanced CD8⁺ T cell priming in vivo. Studies of Salmonella typhimurium and other vacuolar bacteria show that death of infected cells by apoptosis leads to efficient uptake of bacterial antigens by host DCs and cross-presentation to MHC class I–restricted CD8⁺ T cells (8). Because M. tuberculosis antigens are generated within a phagosomal compartment that is sequestered from the classical cytoplasmic pathway of antigen processing, display of these antigens on MHC class I molecules is likely to depend on an indirect route of presentation through bystander DCs, which can take up and process material released from apoptotic infected macrophages via the previously described detour pathway for cross-presentation (13–15). Based on our observation of enhanced apoptosis by the ΔsecA2 mutant in vitro, we predicted that this mutant should also facilitate transmission of antigens to bystander DCs and enhance CD8⁺ T cell priming. To test this in vivo, we expressed a plasmid encoding the M. tuberculosis 19-kDa lipoprotein fused at its C terminus to the H-2K^b–binding sequence of OVA (SIINFEKL) in WT M. tuberculosis (H37Rv-OVA), an attenuated mutant of M. tuberculosis with a deletion of the RD1 region (24) (ΔRD1-OVA), or the ΔsecA2 mutant (ΔsecA2-OVA) (Supplemental Figure 2). Mice were infected with clones of these recombinant mycobacteria expressing comparable levels of the 19-kDa lipoprotein–SIINFEKL fusion protein (Supplemental Figure 2A), and SIINFEKL-responsive CD8⁺ T cells in their spleens were enumerated by IFN-γ ELISPOT. Whereas infection with H37Rv, H37Rv-OVA, or ΔRD1-OVA induced only background levels of IFN-γ spot–forming cells, the ΔsecA2-OVA infection induced a significant, 10-fold increase above background levels (Figure 2A). As another important gauge of CD8⁺ T cell priming, we assessed SIINFEKL-specific cytolytic activity using an in vivo CTL assay. This revealed weak cytolytic activity in mice infected with H37Rv-OVA but a substantial increase in target-specific killing in mice infected with ΔsecA2-OVA (Figure 2B). In contrast to the clear enhancement of T cell responses against an MHC class I–presented epitope in animals primed with the ΔsecA2-OVA strain, IFN-γ ELISPOT responses to crude exogenous mycobacterial proteins contained in M. tuberculosis purified protein derivative (PPD) or the dominant MHC class II–presented epitope of Ag85B (Pep25) were similar in animals infected with WT and ΔsecA2M. tuberculosis (Figure 2C). This indicated that enhancement of T cell priming by the ΔsecA2 mutant was limited mainly or exclusively to the CD8⁺ T cell compartment.

Figure 2

Enhanced CD8⁺ T cell activation in vivo by ΔsecA2. (A) Mice were infected i.v. with the indicated bacterial strains, and spleens were harvested after 7 days for quantitation of SIINFEKL-specific T cells by IFN-γ ELISPOT. SFC, spot-forming cells. (B) CTL activity was measured in vivo by transfer of a mixture of SIINFEKL-pulsed CFSE^low and unpulsed CFSE^high splenocytes into recipient mice 14 days after infection with the indicated bacteria. Relative proportions of CFSE^low and CFSE^high were quantitated by FACS to determine percent specific lysis. (C) CD4⁺ T cell responses were unaffected by deletion of secA2. Seven days after infection with indicated bacteria, responses of splenocytes to purified protein derivative (PPD) and peptide-25 (Pep25) were determined by ELISPOT. (D) Thy1.1⁺ B6.PL mice were injected with CFSE-labeled Thy1.2⁺ OT-I splenocytes, and infected with the indicated bacteria. CD8⁺ T cell activation was assessed 6 days after infection by CFSE dilution. Dot plots show representative mice, and bar graphs show means and SDs for percentages of undivided cells in the transferred population for 2 or 3 mice per bacterial strain. (E) The enhanced CD8⁺ T cell proliferation induced by ΔsecA2 was reversed by reexpression of SodA secretion in the ΔsecA2-αsodA strain. Dot plots show CFSE dilution in the transferred population 6 days after infection with the indicated strains. The bar graph on the right shows means and SDs of the percentages of undivided cells for groups of 4 mice infected with each bacterial strain. All results are representative of at least 2 independent experiments.

To directly visualize the priming of MHC class I–restricted CD8⁺ T cells reactive with SIINFEKL in the context of infection, we used adoptive transfer of CFSE-labeled naive T cells from SIINFEKL/H-2K^b–reactive TCR-transgenic OT-I mice (25). Thy1.1⁺ B6.PL mice were injected with CFSE-labeled Thy1.2⁺ splenocytes from OT-I mice, followed by infection with mycobacteria expressing the 19-kDa lipoprotein–SIINFEKL fusion protein. CD8⁺ T cell activation and proliferation were assessed by dilution of CFSE in the transferred population at 5–7 days after infection (Figure 2D). Negligible proliferation of transferred OT-I T cells was observed in mice infected with H37Rv, and very modest proliferation was seen with H37Rv-OVA infection. In contrast, ΔsecA2-OVA infection induced a massive increase in proliferation of transferred T cells, with the vast majority of cells undergoing 8 or more divisions. The response to ΔsecA2-OVA was most marked in mice that had been adoptively transferred with relatively small numbers (3 × 10⁵) of OT-I T cells, consistent with previous observations on the effect of precursor frequency on limiting CD8⁺ T cell expansion (26). Restoration of secA2 by complementation (ΔsecA2-C-OVA) reversed the effect of secA2 deletion on OT-I T cell priming (Figure 2D). In contrast to the marked differences in OT-I T cell proliferation to infection by secA2-negative versus -positive strains, CD8⁺ T cell responses to lysates of the same strains injected subcutaneously with an adjuvant known to stimulate cross-priming (QS-21) were similar (Supplemental Figure 2C). This confirmed that the SIINFEKL epitope was produced in comparable amounts and in a potentially immunogenic form by all of the different recombinant mycobacterial clones, although efficient presentation of antigen originating from live mycobacteria to OT-I T cells occurred only during infection by ΔsecA2 bacilli.

Since selective reconstitution of secreted SodA activity by expression of αsodA completely reversed apoptosis induction by ΔsecA2 mutants (Figure 1E), we assessed the effect of this on the priming and expansion of OT-I T cells by generating a ΔsecA2-αsodA strain expressing the 19-kDa lipoprotein–SIINFEKL fusion protein (ΔsecA2-αsodA-OVA). Infection with ΔsecA2-αsodA-OVA resulted in markedly reduced expansion of adoptively transferred OT-I T cells compared with ΔsecA2-OVA and was similar in this regard to WT H37Rv-OVA infection (Figure 2E). This indicated a tight linkage among enhanced CD8⁺ T cell priming, the loss of SodA secretion, and the inability of the bacterium to block apoptosis of its infected host cell.

Augmented memory T cell populations following immunization with ΔsecA2. An important benchmark for measuring the efficacy of a vaccine strain is the effect it has on the generation of long-lived memory T cells. Because FACS analysis of endogenous OVA-specific T cells using SIINFEKL-loaded H-2K^b tetramers was not sufficiently sensitive for accurate quantitation (data not shown), we used adoptive transfer of low numbers (5 × 10⁵) of OT-I T cells to boost the precursor frequency of OVA-specific T cells. This enabled accurate quantitation of antigen-specific CD8⁺ T cells by tetramer staining in mice vaccinated by subcutaneous injection with H37Rv-OVA or ΔsecA2-OVA. By following levels of SIINFEKL-loaded H-2K^b tetramer⁺ splenocytes over a period of 20 weeks, we observed that ΔsecA2-OVA vaccinated mice showed an enhanced expansion phase of antigen-specific CD8⁺ T cells that was followed by a sustained maintenance phase characterized by increased levels of CD8⁺ memory T cells (Figure 3). We also measured expression of CD62 ligand (CD62L) and CD44 on the tetramer⁺ cells to distinguish among naive (CD44^lowCD62L^high), central memory (CD44^highCD62L^high), and effector memory (CD44^highCD62L^low) subsets (27). An increase in the long-term expansion of CD8⁺ memory T cells to at least 8 weeks after vaccination with ΔsecA2 was apparent in both the CD62L^high and CD62L^low subsets, with a trend favoring greater and more persistent expansion in the CD62L^high population. We also observed higher expression of IL-7 receptor α chain and lower expression of the programmed death–1 receptor (PD-1) in expanded antigen-specific CD8⁺ T cells in ΔsecA2-vaccinated mice at 20 weeks (data not shown), consistent with the generation of a long-lived and functionally intact population of memory T cells (28, 29).

Figure 3

Augmented memory T cell populations following immunization with ΔsecA2. C57BL/6 mice were adoptively transferred with 5 × 10⁵ OT-I splenocytes 24 hours prior to vaccination by subcutaneous infection with either H37Rv-OVA or ΔsecA2-OVA. The animals were sacrificed 2, 4, 8, and 20 weeks after vaccination, and splenocytes were stained for 5-color fluorescence analysis using antibodies specific to Thy1.2, B220, CD62L, and CD44 plus SIINFEKL-loaded H-2K^b tetramer. (A) FACS analysis of total B220^– events among 1.5 × 10⁶ total lymphocytes, showing proportions of Thy1.2⁺ cells staining with tetramer. (B) Dot plots show expression of CD44 and CD62L on cells gated for tetramer staining as in A from representative mice sacrificed 2 weeks after infection with the indicated bacterial strains. The graphs show percentages of total B220^– lymphocytes staining with tetramer (Tet) and expressing either high or low levels of CD62L at each time point after vaccination. Filled squares represent ΔsecA2-vaccinated mice and open squares, H37Rv-OVA–vaccinated mice. Each symbol represents mean and SD for groups of 2 or 3 mice.

Protective immunity against virulent M. tuberculosis challenge following vaccination with ΔsecA2. In order to determine whether the enhanced T cell priming observed with ΔsecA2 could increase its efficacy as a live vaccine strain, we conducted immunization and challenge studies designed to evaluate potential vaccine efficacy. C57BL/6 mice that were either naive or immunized 8 weeks earlier by subcutaneous inoculation with live ΔsecA2 or BCG (Pasteur strain) were challenged by low-dose aerosol infection with virulent M. tuberculosis (strain Beijing W/HN878). In naive mice, substantial growth in the lungs and dissemination to spleens were detected within 1 month of challenge. In contrast, vaccination with ΔsecA2 and BCG considerably reduced M. tuberculosis bacillary loads in both lungs and spleens of aerosol-challenged mice as compared with naive controls at 1 month after aerosol challenge (Figure 4A). The protection evoked by ΔsecA2 immunization was significantly greater than that obtained with BCG vaccination, as measured by bacterial burden in lung (0.72-log reduction in ΔsecA2 versus BCG vaccinated mice; P < 0.001). Immunization with ΔsecA2 also showed a more prolonged effect on control of M. tuberculosis infection compared with BCG immunization, with reductions in CFU still apparent 5 months after challenge, when BCG immunization no longer showed significant effects. Moreover, the survival of M. tuberculosis–challenged mice vaccinated with ΔsecA2 was significantly prolonged when compared with that of naive mice or mice vaccinated with BCG (Figure 4C). Histopathological examination of the lungs from mice immunized with either BCG or the ΔsecA2 strain showed relatively mild inflammation with small and compact lymphocyte-rich granulomas, compared with naive mice, which had extensive, poorly organized granulomatous lesions (Figure 4B).

Figure 4

Protective immunity against virulent M. tuberculosis challenge in mice following vaccination with ΔsecA2. (A) C57BL/6 mice were vaccinated subcutaneously with saline (naive) or with 1 × 10⁶ BCG or ΔsecA2 and challenged by aerosol 2 months later with 50–100 CFU of virulent M. tuberculosis Beijing/W strain (HN878). Graphs show means and SDs of CFU of M. tuberculosis in lung and spleen at 1, 3, and 5 months after challenge for groups of 5 mice that were either naive (white bars), BCG vaccinated (gray bars), or ΔsecA2 vaccinated (black bars). *P < 0.05 and ^†P < 0.001, compared with unvaccinated group or between bracketed groups; 1-way ANOVA with Tukey’s post-hoc test. (B) Lungs of mice vaccinated and challenged with virulent M. tuberculosis as in A were examined histologically at 1 and 3 months after challenge. More severe, spreading lung lesions with extensive granulomatous pneumonia and consolidation were observed in unvaccinated mice as compared with mice vaccinated with either BCG or ΔsecA2. Original magnification, ×20. (C) C57BL/6 mice were vaccinated subcutaneously with saline or with 1 × 10⁶ BCG or ΔsecA2 and challenged by aerosol 2 months later with 50–100 CFU of virulent M. tuberculosis Erdman strain. Mice were observed daily for survival for 1 year after challenge. Moderate extension of survival was observed in BCG-vaccinated animals (P = 0.006 compared with naive control group; log rank test), and much more pronounced extension of survival was observed in ΔsecA2-vaccinated animals (P = 0.0001 compared with naive; P = 0.0062 compared with BCG).

We also tested the protective capacity of ΔsecA2 vaccination using an aerosol challenge model in guinea pigs, a host species highly susceptible to M. tuberculosis. Animals that were either naive or immunized 2 months previously with ΔsecA2 or BCG were challenged by low-dose aerosol inoculation of M. tuberculosis H37Rv. At 1 and 2 months after challenge, guinea pigs immunized with M. tuberculosis ΔsecA2 showed marked reductions in CFU in the spleens, lungs, and mediastinal lymph nodes relative to naive controls. Although BCG-immunized animals also showed similar reductions in CFU in the lung and spleen, these animals showed much less reduction of bacillary load in the mediastinal lymph nodes than was seen for the ΔsecA2-immunized group (Figure 5A). The results of pathologic evaluation of guinea pig tissues paralleled the bacterial counts in the tissues (Figure 5B; Supplemental Figure 4). Again, the differences in the mediastinal lymph nodes were striking, as these were markedly enlarged in the naive and BCG-immunized groups but appeared normal in size and gross appearance in ΔsecA2-immunized animals. Microscopically, the lymph nodes of both naive and BCG-immunized animals showed marked granulomatous reaction progressing to fibrosis that completely effaced the normal architecture of the nodes by 2 months, while the lymph nodes of ΔsecA2-immunized animals appeared normal at 1 month and showed only small areas of fibrosis at 2 months after challenge. Taken together, these results in 2 different animal models of M. tuberculosis vaccination and challenge indicated that ΔsecA2 was at least equal to BCG in all outcomes examined and significantly exceeded BCG in some measures of efficacy such as prolongation of survival and the ability to restrict extension of pulmonary infection into the regional lymphatic system.

Figure 5

Protective immunity against virulent M. tuberculosis challenge in guinea pigs following vaccination with ΔsecA2. (A) Outbred Hartley guinea pigs were vaccinated intradermally with saline, BCG, or ΔsecA2 (n = 5 animals per group) and challenged by aerosol 2 months later with 10–30 CFU of virulent M. tuberculosis H37Rv. Graphs show SEM of CFU of M. tuberculosis in lung, spleen, and mediastinal lymph nodes at 1 and 2 months after challenge for naive (white bars), BCG-vaccinated (gray bars), or ΔsecA2-vaccinated (black bars) animals (CFU counts obtained from a minimum of 3 and in most instances 5 animals; an exception was 2-month spleen CFU for BCG-vaccinated animals, in which counts were obtained from only 1 animal for technical reasons). (B) Quantitative scoring of histopathology graphed as the SEM of the rank scores for the groups of guinea pigs vaccinated and challenged as described in A. Images show representative low-power (×25) views of H&E-stained sections of mediastinal lymph nodes harvested from each group at 1 or 2 months after challenge. Scale bars: 1 mm. *P < 0.05, **P < 0.01, ^†P < 0.001 compared with unvaccinated group or between bracketed groups; 1-way ANOVA with Tukey’s post-hoc test.

Our results confirm a significant role for inhibition of infected host cell apoptosis in the immune evasion strategy of M. tuberculosis and suggest the secA2-dependent secretion of mycobacterial SodA as a key mechanism involved in this virulence-associated trait. In this regard, virulent mycobacteria appear to have adopted an evasion strategy that is characteristic of many viruses that also produce factors that can delay apoptotic host cell death (30). Whereas the ability of virulent mycobacteria to perform this function has been previously described (9–11, 22), the work reported here is to our knowledge the first demonstration of a specific genetic deletion in M. tuberculosis that curtails the ability of this organism to inhibit apoptosis of infected macrophages. Although our results showed that secA2 was essential for the antiapoptotic properties of M. tuberculosis, it is likely that other mechanisms also cooperate with the secA2-dependent mechanism described here in blocking all of the multiple pathways that can initiate apoptosis. Indeed, our preliminary studies have identified at least 3 additional genes in M. tuberculosis that are required for effective blockade of infected macrophage apoptosis (unpublished observations), although it is currently unclear which pathways for induction or execution of programmed cell death these are inhibiting.

Several advantages may accrue to mycobacteria as a result of delaying or inhibiting macrophage apoptosis, including preservation of a protected growth environment and prevention of engagement of intrinsic microbicidal activities in the host cell (31, 32). The studies presented here also underscore the key role of blockade of apoptosis in preventing optimal stimulation of protective immunity. This effect was particularly evident in our studies of CD8⁺ T cell priming and memory T cell generation, in which WT M. tuberculosis showed remarkably poor immunogenicity when compared with the ΔsecA2 mutant. CD8⁺ T cells have been found by several criteria to be a significant component of antimycobacterial immunity in mouse models (33–36), and they can potentially contribute to containment or eradication of M. tuberculosis via several mechanisms. These include the secretion of protective cytokines such as IFN-γ and TNF, cytolytic activity against infected macrophages, and even direct bactericidal effects on intracellular mycobacteria (4, 37, 38). Although CD4⁺ T cell responses were not measurably enhanced by deletion of secA2 in our experiments (Figure 2C), we have observed that vaccination with M. tuberculosis strains including ΔsecA2 generally prime stronger CD4⁺ T cell responses when compared directly with BCG (data not shown). This raises the possibility that the enhanced protection by ΔsecA2M. tuberculosis compared with BCG observed in the current study could be due at least in part to an augmented CD4⁺ T cell response. However, in our experience so far, the ΔsecA2 strain is unique among M. tuberculosis mutants in its ability to provide a higher level of protection than BCG in animal challenge models (ref. 39 and unpublished observations). This strongly suggests that the apoptosis-promoting and CD8⁺ T cell–augmenting properties of ΔsecA2 M. tuberculosis are likely to be important to its ability to generate significantly increased levels of protective immunity. Further studies will be necessary to determine whether the strong protective effects of immunization with the ΔsecA2 strain can be completely attributed to the enhanced CD8⁺ T cell response or whether modulation of other key components of the adaptive response is also involved.

Interestingly, it has recently been shown that ΔsecA2M. tuberculosis exhibits attenuated growth in isolated mouse macrophages in vitro and that this phenotype is observed even in macrophages that cannot generate high levels of reactive oxygen intermediates because of inactivation of the inducible phagosome oxidase complex (40). This finding is consistent with earlier studies showing that superoxide and other reactive oxygen intermediates contribute little to the antimycobacterial activity of macrophages (41). While this does not contradict our findings on the effects of secA2 deletion on macrophage survival and T cell priming, it does suggest that mycobacterial secretion of SodA is more directly protective of the host cell than of the mycobacterium itself and that the perturbation of other secreted effectors may be responsible for the attenuated growth of ΔsecA2M. tuberculosis in macrophages.

Previous work by Kaufmann and colleagues has demonstrated that modifying BCG to express listeriolysin improves its ability to stimulate protective immunity in mice, and this may also be associated with an increased ability to induce apoptosis of macrophages (42). Our demonstration of the enhancement of long-lived adaptive immunity by the M. tuberculosis ΔsecA2 mutant provides a rational and straightforward basis for improving the immunogenicity of live attenuated mycobacterial strains for use as M. tuberculosis vaccines (43). By combining ΔsecA2 with additional mutations that disable various other immune evasion– and virulence-promoting functions of M. tuberculosis, it may be possible to develop a new generation of safe and effective attenuated mycobacterial vaccine strains that will have greater impact than BCG on the current global TB problem.

View Supplemental data

This work was supported by NIH/NIAID grants AI063537 (PO1 to S.A. Porcelli, W.R. Jacobs Jr., and J. Chan), AI54540 (to M. Braunstein), AI054697 (to I.M. Orme) and by the TB Vaccine Testing and Research Materials Contract (NIH/NIAID HHSN266200400091c) awarded to Colorado State University. The OT-1/Rag1 mouse was obtained from the NIAID exchange program. Flow cytometry studies were facilitated by the FACS Core Resources supported by the AECOM Center for AIDS Research (AI 051519) and the AECOM Cancer Center (CA 13330). S.A. Porcelli also acknowledges support from a Burroughs Wellcome Fund Award in Translational Research and a Career Scientist Award from the Hirschl Trust. We thank John Kim, Mei Chen, Weijun Liu, Amy Yang, Crystal Shanley, and Erin Smith for expert technical assistance with animal vaccine and challenge studies and Mi-Sun Koo and Tatjana Dragic for helpful discussions and critical reading of the manuscript.

Address correspondence to: William R. Jacobs Jr., Howard Hughes Medical Institute and Department of Microbiology and Immunology, Albert Einstein College of Medicine, 1300 Morris Park Avenue, New York, New York, 10461, USA. Phone: (718) 430-2865; Fax: (718) 518-0366; E-mail: jacobs@aecom.yu.edu. Or to: Steven A. Porcelli, Department of Microbiology and Immunology, Albert Einstein College of Medicine, 1300 Morris Park Avenue, New York, New York, 10461, USA. Phone: (718) 430-3228; Fax: (718) 430-8711; E-mail: porcelli@aecom.yu.edu.

Nonstandard abbreviations used: BCG, bacille Calmette-Guérin; BMM, BM-derived macrophage; CD62L, CD62 ligand; RT, room temperature; SodA, superoxide dismutase A; TB, tuberculosis.

Conflict of interest: The authors have declared that no conflict of interest exists.

Reference information: J. Clin. Invest.117:2279–2288 (2007). doi:10.1172/JCI31947

Joseph Hinchey and Sunhee Lee contributed equally to this work.

Steven A. Porcelli and William R. Jacobs, Jr. contributed equally to this work.

See the related article at New TB vaccines: is there a requirement for CD8⁺ T cells?.

1. Dye, C., Watt, C.J., Bleed, D.M., Hosseini, S.M., Raviglione, M.C. 2005. Evolution of tuberculosis control and prospects for reducing tuberculosis incidence, prevalence, and deaths globally. JAMA. 293:2767-2775.
   View this article via: CrossRef PubMed Google Scholar
2. Andersen, P., Doherty, T.M. 2005. The success and failure of BCG — implications for a novel tuberculosis vaccine. Nat. Rev. Microbiol. 3:656-662.
   View this article via: CrossRef PubMed Google Scholar
3. Kaufmann, S.H.E. 2005. Introduction. Rational vaccine development against tuberculosis: “Those who don’t remember the past are condemned to repeat it.”. Microbes Infect. 7:897-898.
   View this article via: CrossRef PubMed Google Scholar
4. Flynn, J.L., Chan, J. 2001. Immunology of tuberculosis. Annu. Rev. Immun. 19:93-129.
   View this article via: CrossRef PubMed Google Scholar
5. Houben, E.N., Nguyen, L., Pieters, J. 2006. Interaction of pathogenic mycobacteria with the host immune system. Curr. Opin. Microbiol. 9:76-85.
   View this article via: CrossRef PubMed Google Scholar
6. Menaker, R.J., Jones, N.L. 2003. Fascination with bacteria-triggered cell death: the significance of Fas-mediated apoptosis during bacterial infection in vivo. Microbes Infect. 5:1149-1158.
   View this article via: CrossRef PubMed Google Scholar
7. Navarre, W.W., Zychlinsky, A. 2000. Pathogen-induced apoptosis of macrophages: a common end for different pathogenic strategies. Cell. Microbiol. 2:265-273.
   View this article via: CrossRef PubMed Google Scholar
8. Yrlid, U., Wick, M.J. 2000. Salmonella-induced apoptosis of infected macrophages results in presentation of a bacteria-encoded antigen after uptake by bystander dendritic cells. J. Exp. Med. 191:613-624.
   View this article via: CrossRef PubMed Google Scholar
9. Keane, J., Remold, H.G., Kornfeld, H. 2000. Virulent Mycobacterium tuberculosis strains evade apoptosis of infected alveolar macrophages. J. Immunol. 164:2016-2020.
   View this article via: PubMed Google Scholar
10. Riendeau, C.J., Kornfeld, H. 2003. THP-1 cell apoptosis in response to mycobacterial infection. Infect. Immun. 71:254-259.
    View this article via: CrossRef PubMed Google Scholar
11. Sly, L.M., Hingley-Wilson, S.M., Reiner, N.E., McMaster, W.R. 2003. Survival of Mycobacterium tuberculosis in host macrophages involves resistance to apoptosis dependent upon induction of antiapoptotic Bcl-2 family member Mcl-1. J. Immunol. 170:430-437.
    View this article via: PubMed Google Scholar
12. Danelishvili, L., McGarvey, J., Li, Y.J., Bermudez, L.E. 2003. Mycobacterium tuberculosis infection causes different levels of apoptosis and necrosis in human macrophages and alveolar epithelial cells. Cell. Microbiol. 5:649-660.
    View this article via: CrossRef PubMed Google Scholar
13. Schaible, U.E., et al. 2003. Apoptosis facilitates antigen presentation to T lymphocytes through MHC-I and CD1 in tuberculosis. Nat. Med. 9:1039-1046.
    View this article via: CrossRef PubMed Google Scholar
14. Winau, F., Hegasy, G., Kaufmann, S.H.E., Schaible, U.E. 2005. No life without death — apoptosis as prerequisite for T cell activation. Apoptosis. 10:707-715.
    View this article via: CrossRef PubMed Google Scholar
15. Winau, F., et al. 2006. Apoptotic vesicles crossprime CD8 T cells and protect against tuberculosis. Immunity. 24:105-117.
    View this article via: CrossRef PubMed Google Scholar
16. Braunstein, M., Espinosa, B.J., Chan, J., Belisle, J.T., Jacobs, R. 2003. SecA2 functions in the secretion of superoxide dismutase A and in the virulence of Mycobacterium tuberculosis. Mol. Microbiol. 48:453-464.
    View this article via: CrossRef PubMed Google Scholar
17. De Groote, M.A., et al. 1997. Periplasmic superoxide dismutase protects Salmonella from products of phagocyte NADPH-oxidase and nitric oxide synthase. Proc. Natl. Acad. Sci. U. S. A. 94:13997-14001.
    View this article via: CrossRef PubMed Google Scholar
18. Kahl, R., Kampkotter, A., Watjen, W., Chovolou, Y. 2004. Antioxidant enzymes and apoptosis. Drug Metab. Rev. 36:747-762.
    View this article via: CrossRef PubMed Google Scholar
19. Harth, G., Horwitz, M.A. 1999. Export of recombinant Mycobacterium tuberculosis superoxide dismutase is dependent upon both information in the protein and mycobacterial export machinery. A model for studying export of leaderless proteins by pathogenic mycobacteria. J. Biol. Chem. 274:4281-4292.
    View this article via: CrossRef PubMed Google Scholar
20. Edwards, K.M., et al. 2001. Iron-cofactored superoxide dismutase inhibits host responses to Mycobacterium tuberculosis. Am. J. Respir. Crit. Care Med. 164:2213-2219.
    View this article via: PubMed Google Scholar
21. Perskvist, N., Long, M., Stendahl, O., Zheng, L. 2002. Mycobacterium tuberculosis promotes apoptosis in human neutrophils by activating caspase-3 and altering expression of Bax/Bcl-xL via an oxygen-dependent pathway. J. Immunol. 168:6358-6365.
    View this article via: PubMed Google Scholar
22. Lee, J., Remold, H.G., Ieong, M.H., Kornfeld, H. 2006. Macrophage apoptosis in response to high intracellular burden of Mycobacterium tuberculosis is mediated by a novel caspase-independent pathway. J. Immunol. 176:4267-4274.
    View this article via: PubMed Google Scholar
23. Boatright, K.M., Salvesen, G.S. 2003. Mechanisms of caspase activation. Curr. Opin. Cell Biol. 15:725-731.
    View this article via: CrossRef PubMed Google Scholar
24. Hsu, T., et al. 2003. The primary mechanism of attenuation of bacillus Calmette-Guerin is a loss of secreted lytic function required for invasion of lung interstitial tissue. Proc. Natl. Acad. Sci. U. S. A. 100:12420-12425.
    View this article via: CrossRef PubMed Google Scholar
25. Hogquist, K.A., Jameson, S.C., Heath, W.R., Howard, J.L., Bevan, M.J., Carbone, F.R. 1994. T cell receptor antagonist peptides induce positive selection. Cell. 76:17-27.
    View this article via: CrossRef PubMed Google Scholar
26. Marzo, A.L., et al. 2005. Initial T cell frequency dictates memory CD8+ T cell lineage commitment. Nat. Immunol. 6:793-799.
    View this article via: CrossRef PubMed Google Scholar
27. Sallusto, F., Geginat, J., Lanzavecchia, A. 2004. Central memory and effector memory T cell subsets: function, generation, and maintenance. Annu. Rev. Immun. 22:745-763.
    View this article via: CrossRef PubMed Google Scholar
28. Barber, D.L., et al. 2006. Restoring function in exhausted CD8 T cells during chronic viral infection. Nature. 439:682-687.
    View this article via: CrossRef PubMed Google Scholar
29. Kaech, S.M., et al. 2003. Selective expression of the interleukin 7 receptor identifies effector CD8 T cells that give rise to long-lived memory cells. Nat. Immunol. 4:1191-1198.
    View this article via: CrossRef PubMed Google Scholar
30. Gillet, G., Brun, G. 1996. Viral inhibition of apoptosis. Trends Microbiol. 4:312-317.
    View this article via: CrossRef PubMed Google Scholar
31. Molloy, A., Laochumroonvorapong, P., Kaplan, G. 1994. Apoptosis, but not necrosis, of infected monocytes is coupled with killing of intracellular bacillus Calmette-Guerin. J. Exp. Med. 180:1499-1509.
    View this article via: CrossRef PubMed Google Scholar
32. Keane, J., Shurtleff, B., Kornfeld, H. 2002. TNF-dependent BALB/c murine macrophage apoptosis following Mycobacterium tuberculosis infection inhibits bacillary growth in an IFN-gamma independent manner. Tuberculosis (Edinb.). 82:55-61.
    View this article via: CrossRef PubMed Google Scholar
33. Mogues, T., Goodrich, M.E., Ryan, L., LaCourse, R., North, R.J. 2001. The relative importance of T cell subsets in immunity and immunopathology of airborne Mycobacterium tuberculosis infection in mice. J. Exp. Med. 193:271-280.
    View this article via: CrossRef PubMed Google Scholar
34. Behar, S.M., Dascher, C.C., Grusby, M.J., Wang, C.R., Brenner, M.B. 1999. Susceptibility of mice deficient in CD1D or TAP1 to infection with Mycobacterium tuberculosis. J. Exp. Med. 189:1973-1980.
    View this article via: CrossRef PubMed Google Scholar
35. Derrick, S.C., Yang, A.L., Morris, S.L. 2005. Vaccination with a Sindbis virus-based DNA vaccine expressing antigen 85B induces protective immunity against Mycobacterium tuberculosis. Infect. Immun. 73:7727-7735.
    View this article via: CrossRef PubMed Google Scholar
36. Huygen, K., et al. 1996. Immunogenicity and protective efficacy of a tuberculosis DNA vaccine. Nat. Med. 2:893-898.
    View this article via: CrossRef PubMed Google Scholar
37. Cho, S., et al. 2000. Antimicrobial activity of MHC class I-restricted CD8+ T cells in human tuberculosis. Proc. Natl. Acad. Sci. U. S. A. 97:12210-12215.
    View this article via: CrossRef PubMed Google Scholar
38. Stenger, S., et al. 1998. An antimicrobial activity of cytolytic T cells mediated by granulysin. Science. 282:121-125.
    View this article via: CrossRef PubMed Google Scholar
39. Sambandamurthy, V.K., et al. 2005. Long-term protection against tuberculosis following vaccination with a severely attenuated double lysine and pantothenate auxotroph of Mycobacterium tuberculosis. Infect. Immun. 73:1196-1203.
    View this article via: CrossRef PubMed Google Scholar
40. Kurtz, S., McKinnon, K.P., Runge, M.S., Ting, J.P., Braunstein, M. 2006. The SecA2 secretion factor of Mycobacterium tuberculosis promotes growth in macrophages and inhibits the host immune response. Infect. Immun. 74:6855-6864.
    View this article via: CrossRef PubMed Google Scholar
41. Chan, J., Xing, Y., Magliozzo, R.S., Bloom, B.R. 1992. Killing of virulent Mycobacterium tuberculosis by reactive nitrogen intermediates produced by activated murine macrophages. J. Exp. Med. 175:1111-1122.
    View this article via: CrossRef PubMed Google Scholar
42. Grode, L., et al. 2005. Increased vaccine efficacy against tuberculosis of recombinant Mycobacterium bovis bacille Calmette-Guerin mutants that secrete listeriolysin. J. Clin. Invest. 115:2472-2479.
    View this article via: JCI PubMed Google Scholar
43. Sambandamurthy, V.K., et al. 2002. A pantothenate auxotroph of Mycobacterium tuberculosis is highly attenuated and protects mice against tuberculosis. Nat. Med. 8:1171-1174.
    View this article via: CrossRef PubMed Google Scholar
44. Manca, C., et al. 2005. Hypervirulent M. tuberculosis W/Beijing strains upregulate type I IFNs and increase expression of negative regulators of the Jak-Stat pathway. J. Interferon Cytokine Res. 25:694-701.
    View this article via: CrossRef PubMed Google Scholar
45. Stover, C.K., et al. 1991. New use of BCG for recombinant vaccines. Nature. 351:456-460.
    View this article via: CrossRef PubMed Google Scholar
46. Braunstein, M., Bardarov, S.S., Jacobs (Jr.), W.R. 2002. Genetic methods for deciphering virulence determinants of Mycobacterium tuberculosis. Methods Enzymol. 358:67-99.
    View this article via: PubMed Google Scholar
47. Yanagisawa, S., Koike, M., Kariyone, A., Nagai, S., Takatsu, K. 1997. Mapping of V beta 11⁺ helper T cell epitopes on mycobacterial antigen in mouse primed with Mycobacterium tuberculosis. Int. Immunol. 9:227-237.
    View this article via: PubMed CrossRef Google Scholar