Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services E-mail this article to a friend Similar articles in this journal Similar articles in ASM journals Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Woodworth, J. S. Articles by Behar, S. M. Search for Related Content PubMed PubMed Citation Articles by Woodworth, J. S. Articles by Behar, S. M.

 Previous Article  |  Next Article 

Infection and Immunity, September 2008, p. 4199-4205, Vol. 76, No. 9
0019-9567/08/$08.00+0     doi:10.1128/IAI.00307-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Bacterial Protein Secretion Is Required for Priming of CD8⁺ T Cells Specific for the Mycobacterium tuberculosis Antigen CFP10

Joshua S. Woodworth,^1,2 Sarah M. Fortune,³ and Samuel M. Behar^1*

Division of Rheumatology, Immunology, and Allergy, Brigham and Women's Hospital and Harvard Medical School,¹ Program in Immunology, Harvard Medical School,² Department of Immunology and Infectious Diseases, Harvard School of Public Health, Boston, Massachusetts 02115³

Received 7 March 2008/ Returned for modification 25 April 2008/ Accepted 25 June 2008

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Mycobacterium tuberculosis infection elicits antigen-specific CD8⁺ T cells that are required to control disease. It is unknown how the major histocompatibility complex class I (MHC-I) pathway samples mycobacterial antigens. CFP10 and ESAT6 are important virulence factors secreted by M. tuberculosis, and they are immunodominant targets of the human and murine T-cell response. Here, we test the hypothesis that CFP10 secretion by M. tuberculosis is required for the priming of CD8⁺ T cells in vivo. Our results reveal an explicit dependence upon the bacterial secretion of the CFP10 antigen for the induction of antigen-specific CD8⁺ T cells in vivo. By using well-defined M. tuberculosis mutants and carefully controlling for virulence, we show that ESX-1 function is required for the priming of CD8⁺ T cells specific for CFP10. CD4⁺ and CD8⁺ T-cell responses to mycobacterial antigens secreted independently of ESX-1 were unaffected, suggesting that ESX-1-dependent phagosomal escape is not required for CD8⁺ T-cell priming during infection. We propose that the overrepresentation of secreted proteins as dominant targets of the CD8⁺ T-cell response during M. tuberculosis infection is a consequence of their preferential sampling by the MHC-I pathway. The implications of these findings should be considered in all models of antigen presentation during M. tuberculosis infection and in vaccine development.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Despite the availability of a vaccine and antibiotic therapy, millions of people die each year from disease caused by Mycobacterium tuberculosis. In response, significant effort is being applied to the design of a more effective prophylactic vaccine that elicits T cells and mediates protective immunity. Substantial progress has been made in identifying target antigens for vaccine development and in delineating the role played by different T-cell subsets during infection. In addition to the critical role served by CD4⁺ T cells, CD8⁺ T cells have an important and potentially unique role in the control of M. tuberculosis infection. The identification of mycobacterial antigens recognized by CD8⁺ T cells elicited by infection makes possible a more detailed assessment of CD8⁺ T-cell function in immunity to tuberculosis (42). Nearly all of the known mycobacterial antigens recognized by CD8⁺ T cells are found in bacterial culture supernatants, and many are known to be actively secreted by the bacterium (42). This shared feature may simply reflect investigator bias in the selection of antigens for study. However, the alternative possibility is that the secretion of these proteins is an important step in their presentation by major histocompatibility complex class I (MHC-I).

The possible relationship between protein secretion and antigen presentation by MHC-I is highlighted by the finding that members of the ESAT6 family of proteins, which includes secreted protein ESAT6, CFP10, and 21 homologues, is overrepresented among antigens recognized by CD8⁺ T cells (4). CFP10 is an immunodominant antigen recognized by CD8⁺ T cells in M. tuberculosis-infected C3H (H-2^k) mice, and as many as 40% of lung CD8⁺ T cells from BALB/c (H-2^d) mice infected via the aerosol route are specific for an epitope shared by the ESAT6-like proteins TB10.3 and TB10.4 (16, 19, 26). Importantly, CD8⁺ T cells recognizing CFP10 are elicited following M. tuberculosis infection in humans (21, 34).

ESAT6 and CFP10 are small secreted proteins encoded within the RD1 locus (31), a nine-gene region that is present in M. tuberculosis and pathogenic Mycobacterium bovis strains, but it is deleted from all BCG strains (1, 8, 15, 22, 28). The RD1 locus is critical for the virulence of the pathogenic mycobacteria, and the loss of RD1 largely accounts for the attenuation of BCG (1, 15, 22, 28). The RD1 deletion encompasses most, but not all, of a genetic locus (Rv3866-Rv3879) known as ESX-1. ESX-1 encodes a specialized protein secretion system required for the secretion of ESAT6 and CFP10 (12, 13, 20). In addition to the ESX-1 genes, proteins encoded outside of the RD1 locus also are required for ESAT6 and CFP10 secretion (11, 23). CFP10/ESAT6 secretion requires the function of a genetically unlinked operon, Rv3616c-Rv3614c (7, 24). Interestingly, Rv3616c encodes a protein, EspA, that is a substrate of the ESX-1 apparatus (11).

Although the molecular mechanisms by which ESX-1 mediates virulence are largely unknown, mycobacteria lacking a functional ESX-1 and unable to secrete CFP10/ESAT6 cause less macrophage (M) cytolysis and appear defective in arresting phagolysomal fusion (12, 37). ESX-1 secretory function is required to elicit T-cell immunity to ESAT6 (2, 3, 25, 28). Recombinant BCG strains that express but do not secrete ESAT6 and CFP10 do not elicit T-cell immunity to ESAT6 (29). Only mice vaccinated with recombinant BCG complemented with the full RD1 region generate an ESAT6-specific T-cell response. These data suggest that mycobacterial antigen secretion is linked to T-cell priming. Likewise, by complementing BCG or Mycobacterium microti with RD1-encoding DNA specifically disrupted at individual genes that abrogated CFP10/ESAT6 secretion but not production, Brodin et al. more specifically associated ESX-1 genes that are required for CFP10/ESAT6 secretion with ESAT6-specific CD4⁺ T-cell priming (3). These studies did not address whether CFP10/ESAT6 secretion is required for CD8⁺ T-cell priming or how ESX-1 function affects T-cell function or the global immune response. Most importantly, the interpretation of these studies is complicated, because a functional ESX-1 system is required for bacterial survival in vivo (13, 22, 36). Thus, it is not clear whether a T-cell response fails to develop because ESAT6 is not secreted or because the bacteria do not replicate.

In this study, we use precise genetic mutations of virulent M. tuberculosis strains that affect the secretion, but not the production, of CFP10 and ESAT6 to specifically address the effect of bacterial protein secretion on both CD4⁺ and CD8⁺ T-cell priming. We demonstrate that the bacterial production of these proteins is not sufficient to prime T cells; the bacterial secretion of CFP10 and ESAT6 is required for generating an antigen-specific CD4⁺ and CD8⁺ T-cell responses to these proteins. Moreover, we carefully address the potential effects of bacterial virulence and persistence, showing that the effect on both CD4⁺ and CD8⁺ T-cell priming is specific to the loss of CFP10 and ESAT6 secretion per se and not an indirect effect of reduced bacterial virulence.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bacteria. Mycobacterial strains were grown to mid-log phase in 7H9 Middlebrook medium containing 10% albumin-dextrose-catalase enrichment and were frozen at –80°C before use. Mutants (Rv0573::Tn, Rv3870::Tn, Rv3874::Tn, espA, espA-pJEB, and espA-pEspA) have been described already (11, 32). See Table 1 for a description of the mutant bacterial strains used in this study.

View this table: [in this window] [in a new window]

TABLE 1. Mutants of M. tuberculosis (H37Rv) used in this study

Mice and in vivo infections. Age-matched female C3H/HeJ (C3H), B10.BR-H2^k H2-T18^a/SgSnJ (B10.BR), and BALB/c mice (from Jackson Laboratories) were used in an approved protocol. Mice were infected via the tail vein with inocula ranging from 10⁵ to 10⁶ CFU and were housed in an animal biosafety level 3 facility (38). The numbers of CFU were determined as described previously (17).

In vitro restimulation assays. CD8⁺ or CD90⁺ T cells were purified from spleen or lung by positive selection using immunomagnetic beads per the manufacturer's protocol (Miltenyi Biotec) (6, 19). The purity of T cells was >90% from spleen and 70 to 85% from lung. Purified T cells were cultured with irradiated naïve syngeneic splenocytes as antigen-presenting cells (APC) and were stimulated with peptides (10 µM unless otherwise stated) or H37Rv sonicate (5 µg/ml) for 48 h at 37°C. Peptides were used in unpurified form (9, 14, 19, 26). Recombinant interleukin-2 was added (100 U/ml). Supernatants were assayed for gamma interferon (IFN-) by enzyme-linked immunosorbent assay. The variability in the IFN- responses between experiments arises from differences in the inoculum, the time point examined, and the frequency of antigen-specific T cells elicited.

Flow cytometry. Spleen and lung cells were stained as described previously (16, 19) with isotype-matched control immunoglobulin G and antibodies specific for mouse CD3, CD4, CD8, and CD19, which were conjugated to fluorescein isothiocyanate, phycoerythrin (PE), PE-Cy5, or peridinin chlorophyll protein. Tetramers were produced using VESTAGSL-loaded H-2K^k or GYAGTLQSL-loaded H-2K^d complexed to streptavidin-PE (SA-PE) (NIH Tetramer Core Facility, Emory University Vaccine Center, Atlanta, GA); their specificities have been reported already (19). Cells were analyzed using a FACSCanto (BD Biosciences) and FlowJo v8.0 (TreeStar Inc.) software. Cells were gated on lymphocyte size and granularity, and CD19-positive events were excluded. The percentage of tetramer-positive staining was determined by gating on CD8⁺ cells and using SA-PE as a control.

Statistics. All data are representative of three to six experiments. Student t test and one-way analysis of variance with Dunnett's post test were performed using GraphPad Prism (www.graphpad.com).

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Priming of ESAT6- and CFP10-specific CD8⁺ and CD4⁺ T cells requires intact ESX-1 genes. M. tuberculosis infection elicits antigen-specific CD8⁺ T cells to amino acid residues 32 to 39 of CFP10 (CFP10_32-39), and CD4⁺ T cells to CFP10_11-25 and ESAT6_53-75, in C3H mice (19). To investigate how the bacterial secretion of CFP10 and ESAT6 affects the priming of T cells specific for these antigens, mice were infected with mutants of H37Rv that cannot secrete CFP10/ESAT6. We compared control strain Rv0573::Tn, which contains a transposon insertion in Rv0573 and has normal CFP10/ESAT6 secretion, to mutant Rv3870::Tn, which contains a transposon insertion in Rv3870 that abolishes the secretion but not the intracellular production of CFP10 and ESAT6 (11, 13). Spleen and lung T cells isolated from mice 3 to 5 weeks following Rv0573::Tn infection secreted IFN- when stimulated with CFP10 and ESAT6 peptides, indicating the in vivo priming of CD8⁺ and CD4⁺ T cells specific for these antigens (Fig. 1A and data not shown). In contrast, T cells from mice infected with Rv3870::Tn failed to generate a response to ESAT6 and CFP10 (Fig. 1A). T cells from mice infected with Rv3874::Tn, which has a transposon insertion in Rv3874 (cfp10) and does not express CFP10 or ESAT6 (11), fail to recognize the CFP10 and ESAT6 peptides, confirming the specificity of these epitopes in C3H mice (Fig. 1A). The requirement for CFP10/ESAT6 secretion in T-cell priming is not peculiar to susceptible C3H mice; the infection of resistant B10.BR (also H-2^k) mice (18) with the mutant M. tuberculosis strains gave similar results (data not shown). Thus, the in vivo priming of IFN--producing CD8⁺ and CD4⁺ T cells specific for CFP10 and ESAT6 requires their secretion by M. tuberculosis.

View larger version (15K): [in this window] [in a new window]

FIG. 1. Priming of CFP10- and ESAT6-specific CD8+ and CD4+ T cells requires intact ESX-1 genes. (A) Splenic CD90+ T cells isolated from mice (n = 3, pooled) 3 weeks postinfection with Rv0573::Tn, Rv3870::Tn, or Rv3874::Tn were restimulated in vitro with peptide or H37Rv sonicate. Bars indicate means ± standard deviations. ***, P < 0.001 compared to results for medium alone. (B) Numbers of splenic CFU from the C3H mice used for panel A. The line indicates the median value. The means were not significantly different. (C) Pooled CD90+ T cells isolated from mice (n = 3) 3 weeks postinfection with the indicated M. tuberculosis strain were analyzed by flow cytometry using the H2-Kk/CFP1032-39 tetramer. Percentages indicate the proportion of CD8+ T cells positive for tetramer binding. (D) H2-Kk/CFP1032-39 tetramer staining of purified CD90+ T cells pooled from mice (n = 3) 3 to 4 weeks postinfection compiled from three experiments from two independent infections. Bars indicate means ± standard errors of the means. *, P < 0.05 compared to results for Rv0573::Tn.

Since the disruption of Rv3870 or Rv3874 leads to bacterial attenuation (11), a possible confounder is that a smaller bacterial burden and reduction in antigen load explain the differences in T-cell priming, although a stable chronic infection is established in vivo following low-dose aerosol inoculation with these and other mutants of the RD1 region (11, 22). Therefore, throughout these studies mice were given 5- to 10-fold higher inoculums of Rv3870::Tn and Rv3874::Tn compared to those of Rv0573::Tn, which is not attenuated, and the numbers of spleen and lung CFU were determined for each experiment. The bacterial burden 3 to 5 weeks after infection with Rv3870::Tn and Rv3874::Tn was similar to, or occasionally greater than, that after infection with the fully virulent Rv0573::Tn control strain (Fig. 1B). No T-cell response to CFP10 and ESAT6 was detected at any of these time points (Fig. 1A and data not shown). Furthermore, T cells from animals infected with each of the M. tuberculosis mutants recognized H37Rv sonicate similarly (Fig. 1A), indicating that the global T-cell response to M. tuberculosis was intact. Therefore, the failure to prime CD4⁺ and CD8⁺ T cells specific for CFP10 and ESAT6 following infection with M. tuberculosis mutants with a defective ESX-1 secretion system cannot be explained by the loss of virulence or immunogenicity.

Heterogeneity exists among CD8⁺ T cells, and not all produce IFN- (16, 27). We considered the possibility that the less virulent mutants qualitatively affect immunity to M. tuberculosis. By using MHC-I tetramers, flow cytometry was used to detect antigen-specific T cells independent of their effector function (e.g., cytokine production). CFP10_32-39-specific CD8⁺ T cells were detected in spleens from Rv0573::Tn-infected, but not Rv3870::Tn-infected, mice (Fig. 1C, D). Rv3874::Tn-infected mice served as a control for tetramer specificity and established that the residual positive events among CD8⁺ T cells from Rv3870::Tn- and Rv3874::Tn-infected mice represent nonspecific staining (Fig. 2C, D). An identical pattern of CFP10_32-39 tetramer staining was observed in lung mononuclear cells isolated from the same animals at both 3 and 5 weeks postinfection (data not shown). Thus, the use of tetramers confirms that the in vivo priming of CFP10-specific CD8⁺ T cells is dependent on CFP10 secretion.

View larger version (15K): [in this window] [in a new window]

FIG. 2. Priming of CFP10- and ESAT6-specific CD8+ and CD4+ T cells requires an intact espA gene. (A) Pooled CD90+ splenic T cells isolated from H37Rv- or espA-infected C3H mice (n = 3) 3 weeks postinfection were restimulated in vitro with peptide or H37Rv sonicate. Bars indicate means ± standard deviations. * and ***, P < 0.05 and P < 0.001, respectively, compared to results for medium alone by one-way analysis of variance. (B) Pooled CD90+ splenic T cells isolated from C3H mice (n = 3) 3 weeks postinfection were stained with the H2-Kk/CFP1032-39 tetramer. Percentages indicate the proportion of CD8+ T cells positive for tetramer binding. (C) H2-Kk/CFP1032-39 tetramer staining of purified CD90+ splenic T cells pooled from mice (n = 3) 3 to 5 weeks postinfection with the indicated M. tuberculosis strain compiled from three experiments from two independent infections. Bars indicate means ± standard errors of the means. The dotted line represents nonspecific staining based on tetramer staining from Rv3874::Tn-infected animals analyzed in parallel. *, P < 0.05 compared to results for H37Rv. (D and E) Spleen (D) or lung (E) CD90+ T cells pooled and isolated from C3H mice (n = 3) infected with espA-pJEB (uncomplemented; open bars) or espA-pEspA (espA plasmid complemented; filled bars) 5 weeks postinfection were restimulated in vitro with peptide or H37Rv sonicate. Bars indicate means ± standard deviations. ***, P < 0.001 compared to results for medium alone. Note that the data presented in panels A to C and panels D and E are from separate experiments.

Priming of CFP10-specific CD8⁺ and CD4⁺ T cells is dependent on EspA. Although espA is unlinked to the ESX-1 locus, it is required for CFP10/ESAT6 secretion (11). The deletion of espA abrogates CFP10/ESAT6 secretion without genetically altering the cfp10- and esat6-containing ESX-1 locus. We used an unmarked espA deletion mutant of H37Rv (espA) to abolish CFP10/ESAT6 secretion, therefore avoiding the local effects of transposon insertion into the ESX-1 locus. We hypothesized that the in vivo priming of CFP10- and ESAT6-specific T cells would be dependent on bacterial EspA expression. H37Rv infection elicited CFP10_32-39-specific CD8⁺ T cells and CFP10_11-25- and ESAT6_53-75-specific CD4⁺ T cells (Fig. 2A and data not shown). In contrast, espA infection failed to elicit CFP10- or ESAT6-specific T cells. As observed following infection with the transposon mutants, the dependence of T-cell priming on espA was maintained at later time points in both spleen and lung tissues and in B10.BR mice, so that even during the chronic phase of espA infection, no T-cell response to CFP10 was detected (data not shown). The adjustment of bacterial inocula during these experiments ensured that a difference in the number of CFU did not account for the disparities in T-cell priming. Tetramer staining of CD8⁺ T cells from H37Rv- and espA-infected C3H mice confirmed that the priming of CFP10_32-39-specific CD8⁺ T cells was dependent upon espA (Fig. 2B, C).

To ensure that the dependence of espA in priming T cells specific for CFP10 and ESAT6 reflected a requirement for EspA protein, the espA mutant was complemented with plasmid DNA encoding espA under the control of its native promoter (espA-pEspA) that restores CFP10 and ESAT6 secretion or an empty plasmid (espA-pJEB) (11). Complementation with espA, but not the control plasmid, reconstituted the in vivo priming of CFP10- and ESAT6-specific T cells (Fig. 2D, E). T cells isolated from mice infected with EspA-sufficient strains (H37Rv and espA-pEspA) or espA-deficient strains (espA and espA-pJEB) responded similarly to whole mycobacterial sonicate, reflecting a similar T-cell immunogenicity overall (Fig. 2A, D, and E). Since the espA mutation is unmarked and distant from the ESX-1 locus, the potential confounding effects of transposons within the ESX-1 locus have been eliminated from the interpretation of these experiments. These experiments reveal a critical role for the EspA- and ESX-1-dependent secretion of CFP10 and ESAT6 in the generation of specific T-cell responses to these antigens in vivo.

Priming of T-cell responses to non-RD1 antigens. We considered whether the disruption of ESX-1 function affects the overall priming of T cells in infected animals. T cells isolated from animals infected with Rv3870::Tn, Rv3874::Tn, espA, and control M. tuberculosis strains were activated similarly when stimulated in vitro with mycobacterial sonicate, indicating that the mutant strains retained the capacity to stimulate a potent immune response (Fig. 1, 2). However, when given as an exogenous antigen, mycobacterial sonicate preferentially enters the MHC-II pathway and more efficiently activates CD4⁺ T cells. To precisely determine whether mutations affecting the ESX-1 secretory apparatus had any global effects on the priming of CD4⁺ or CD8⁺ T cells, we measured the T-cell recognition of other well-defined epitopes. Because no other mycobacterial antigens have been identified in H-2^k mice, we used both MHC-I- and MHC-II-restricted mycobacterial epitopes restricted by H-2^d. The TB10.4 protein contains epitopes recognized by CD4⁺ T cells (TB10.4_74-88 [14]) and CD8⁺ T cells (TB10.4_20-28, [16, 26]). Ag85A is an abundant protein that is secreted by M. tuberculosis, and infection elicits H-2^d-restricted Ag85A_142-161-specific CD4⁺ T cells (9). We used (C3H x BALB/c)F₁ (H-2^k/d) mice to simultaneously measure the T-cell recognition of CFP10 and ESAT6 (H-2^k restricted) and the non-RD1-associated antigens Ag85A and TB10.4 (H-2^d restricted). The infection of (C3H x BALB/c)F₁ mice with H37Rv or espA revealed a pattern of dependence on CFP10/ESAT6 secretion for the priming of CFP10- and ESAT6-specific T cells that was similar to that observed for C3H mice (Fig. 2 and 3A). In contrast, T cells isolated from animals infected with H37Rv or espA produced similar amounts of IFN- after stimulation with Ag85A_142-161, indicating the normal priming of Ag85A-specific CD4⁺ T cells even in the absence of CFP10/ESAT6 secretion (Fig. 3B). Similarly, the priming of TB10.4-specific CD4⁺ and CD8⁺ T cells was independent of espA (Fig. 3B). To verify this result, spleen and lung mononuclear cells from mice infected with H37Rv or espA were stained with TB10.4_20-28-loaded H2-K^d tetramers (19). The frequency of TB10.4-specific CD8⁺ T cells after H37Rv or espA infection was similar and confirmed that the priming of these T cells is independent of espA (Fig. 3C, D). The priming of Ag85A-specific CD4⁺ T cells and TB10.4-specific CD8⁺ T cells was similar in BALB/c mice infected with Rv3870::Tn, Tn3874, espA, and control strains (data not shown). These data indicated that while the priming of T cells specific for CFP10 and ESAT6 requires their secretion, the interruption of CFP10/ESAT6 secretion does not affect the priming of T cells specific for M. tuberculosis proteins that are secreted independently of ESX-1.

View larger version (34K): [in this window] [in a new window]

FIG. 3. Priming of T cells to non-RD1 antigens is independent of CFP10/ESAT6 secretion. (A and B) Pooled (n = 3 to 4 mice) CD90+ splenic T cells isolated from (C3H x BALB/c)F1 H37Rv-infected (filled bars) or empty-plasmid-transformed espA-infected (open bars) mice 5 weeks postinfection were restimulated in vitro with peptide epitopes from RD1-encoded (A) or non-RD1-encoded (B) antigens or H37Rv sonicate (B). Bars indicate means ± standard deviations. ***, P < 0.001 compared to results for medium alone. The dotted line indicates the average amount of IFN- released by T cells cultured with APC but without antigen. (C) Pooled (n = 3 to 4 mice) CD90+ splenic T cells isolated from (C3H x BALB/c)F1 mice 5 weeks postinfection with H37Rv or espA were analyzed by flow cytometry after being stained with the H2-Kd/TB10.420-28 tetramer (left, center) or SA-PE (right). Percentages indicate the proportions of CD8+ T cells positive for tetramer/SA binding. (D) Splenocytes from individual (C3H x BALB/c)F1 mice 3 to 6 weeks postinfection were stained with the H2-Kd/TB10.420-28 tetramer. Mice were infected with H37Rv (left column) or espA or espA-pJEB (right column). Data points represent the percentages of H-2Kd/TB10.420-28 tetramer-positive or CD8+ cells for each subject. Data are compiled from three time points from two independent experiments. To establish the background value, each sample was stained with SA-PE. The dotted line represents the background means plus five standard deviations; n.s., not significant.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
We show that the bacterial secretion of CFP10 and ESAT6 is required for eliciting both CD4⁺ and CD8⁺ T-cell responses to these protein antigens following M. tuberculosis infection in vivo. Our ability to sensitively detect T-cell responses to these antigens was based on the use of defined peptide epitopes in ex vivo assays and the use of MHC-I tetramers to detect antigen-specific CD8⁺ T cells independently of cytokine production. While the specific pathways for the MHC-I and MHC-II presentation of M. tuberculosis antigens may be distinct, our results suggest that bacterial protein secretion is a general requirement for T-cell priming during M. tuberculosis infection and may explain the old observation that killed mycobacterial vaccines are less efficient than live mycobacteria at eliciting protective immunity. Thus, for in vivo priming, which is a dendritic cell (DC) function (38), the bacterial secretion of CFP10 is required for this antigen to be sampled by both the MHC-I and MHC-II pathways.

A confounding factor is that eliminating CFP10/ESAT6 secretion can affect bacterial virulence and persistence, leading to reduced antigen production (11, 22). Our experiments were designed to control for different bacterial loads that could have arisen from the altered virulence by using the intravenous infection route and measuring immunity 3 to 5 weeks postinfection. Under these conditions, non-CFP10/ESAT6-secreting bacteria persist and replicate in vivo, emphasizing that a reduction in the number of bacteria cannot explain the failure to prime T cells. If CFP10 and ESAT6 were rapidly degraded by the espA and Rv3870::Tn mutants, then despite our controlling for bacterial numbers, the intracellular antigen load may differ between wild-type and mutant M. tuberculosis. If this were true, it would be difficult to argue that differences in T-cell priming are due to a failure of protein secretion and not to diminished antigen quantity. Our previous findings showed that the amount of bacteria-associated CFP10 and ESAT6 in H37Rv and espA are similar, and others have shown that the antigenic CFP10 activity in H37Rv and Rv3870::Tn cell pellets is equivalent (15, 18). Thus, it is unlikely that differences in the amounts of intracellular antigen accounts for our observations. Finally, it was unclear whether eliminating CFP10/ESAT6 secretion would affect T-cell priming to other secreted antigens. Importantly, CD4⁺ and CD8⁺ T-cell responses to the secreted antigens TB10.4 and Ag85A were elicited independently of CFP10/ESAT6 secretion. By dissociating the effects of CFP10/ESAT6 bacterial secretion from the priming of T cells specific for such ESX-1-independent antigens, we are able to specifically link CFP10/ESAT6 secretion to the priming of ESAT6- and CFP10-specific CD4⁺ and CD8⁺ T cells.

There are several proposed mechanisms by which M. tuberculosis antigens are acquired by APC, and our data can be interpreted in the context of these models. Since the depletion of DC abrogates T-cell immunity to M. tuberculosis (38), most models of T-cell priming assume a central role for DC in the acquisition of M. tuberculosis antigens and their presentation to naïve T cells in secondary lymphoid tissues. The simplest model is that M. tuberculosis-infected DC prime T cells. This is consistent with T-cell priming occurring after bacterial dissemination to the pulmonary draining lymph nodes, which is known to be mediated by infected DC as they traffic from the lung (5, 41). In the context of this model, our findings suggest that secreted antigens are preferentially sampled by the MHC-I and MHC-II pathways within infected DC. For example, vesicular trafficking could guide secreted M. tuberculosis antigens away from the M. tuberculosis phagosome and toward MHC-I cross-presentation and classical MHC-II presentation pathways. The recent finding that M. tuberculosis escapes from the phagosome into the cytosol provides another model for understanding how MHC-I could sample mycobacterial antigens (39). Electron microscopy data suggest that while H37Rv escapes from the phagosome into the cytosol of infected human DC and M, espA and BCG do not, revealing a requirement for ESX-1 function (39). Proteins secreted by cytosolic bacteria certainly would be accessible to the proteasome for degradation and transport into the ER for MHC-I loading, thus solving the problem of how M. tuberculosis antigens enter the MHC-I pathway. Thus, van der Wel et al. predicted that preventing phagosomal escape would abrogate the priming of CD8⁺ T cells (39). However, our in vivo data argue against this mechanism. Since bacterial escape into the cytosol is dependent upon espA but espA does not affect the priming of TB10.4-specific CD8⁺ T cells, we conclude that escape is not necessary for CD8⁺ T-cell priming in vivo (Fig. 3). Similarly, although BCG is unable to escape from the phagosome, BCG vaccination elicits TB10.4-specific CD8⁺ T cells, which further supports our data (14, 26, 39). Although phagosomal escape may play a role in immune evasion, particularly from CD4⁺ T cells, and may affect MHC-I presentation, our data indicate that the priming of CD8⁺ T cells does not require phagosomal escape, and other mechanisms must exist by which MHC-I samples mycobacterial antigens.

Another hypothesis is that retrograde protein translocation transports microbial proteins from the phagosome into the cytosol. This function can be mediated by endoplasmic reticulum-resident proteins such as sec61 (30), but whether similar proteins exist in the mycobacterial phagosomal membrane is unknown. Alternatively, bacterial factors may lead to the egress of microbial antigens from the phagosome into other cellular compartments, including the cytosol. ESX-1 bears some resemblance to type IV secretion systems in gram-negative bacteria that deliver bacterial proteins into the host cell (10). It is conceivable that ESX-1 directly transfers antigens from the bacterium across the phagosomal membrane and into the cytosol.

Direct DC infection may not be required for the priming of M. tuberculosis-specific T cells. Uninfected DC can acquire viral antigens and prime CD8⁺ T cells, a phenomenon known as cross-priming (35). The detour pathway proposes that DC acquire apoptotic vesicles derived from M. tuberculosis-infected M. These apoptotic vesicles contain bacterial antigens that can be processed by DC and lead to CD8⁺ T-cell priming (33). Indeed, the vaccination of mice with apoptotic vesicles derived from BCG-ovalbumin-infected M activates ovalbumin-specific CD8⁺ T cells (40). Additionally, M. tuberculosis-infected M release exosomes, which may similarly contribute to T-cell priming via DC uptake (33). Our results are consistent with the idea that secreted antigens are preferentially packaged into apoptotic vesicles that facilitate the priming of MHC-I-restricted CD8⁺ T cells by DC.

By using well-defined M. tuberculosis mutants to study the host T-cell response, we leveraged the power of bacterial genetics to inform us about the requirements of antigen presentation and T-cell priming. Coupling the previous genetic and biochemical characterization of EspA as critical for CFP10/ESAT6 secretion with our current finding that EspA is required for priming CFP10- and ESAT6-specific T cells allows us to unambiguously demonstrate that CFP10/ESAT6 secretion is required for T-cell priming to these antigens. Conversely, the T-cell responses provide insight into the function of bacterial genes and the proteins they encode. For example, TB10.4 is an ESAT6-related protein (EsxH) found in the ESX-3 locus (2, 14, 26). Since TB10.4 elicits both CD4⁺ and CD8⁺ T-cell responses independently of EspA, we infer that EspA is not required for TB10.4 secretion or the function of ESX-3. Thus, although espA is unlinked to the ESX-1 locus, a monogamous relationship may exist between EspA and ESX-1.

We have revealed an explicit connection between the secretion of M. tuberculosis antigens and T-cell immunity. Our use of recently defined peptide epitopes has allowed us to show that the disruption of ESX-1 abrogated the CD4⁺ and CD8⁺ T-cell responses only to proteins that are dependent upon ESX-1 for secretion. T-cell priming to antigens secreted in an ESX-1-independent manner was unaffected. We believe that mechanisms exist for antigen entry into the MHC-I pathway that do not require phagosomal escape. Thus, the apparent bias for secreted proteins as targets of the immune system during M. tuberculosis infection may be rooted in a specific biological requirement for antigen presentation and T-cell priming, and the implications of these findings should be considered in all models of antigen processing and presentation in M. tuberculosis infection.

 
We appreciate the technical assistance of Daniel Shin and Erdeta Bani and the advice provided by Darren Higgins and his laboratory.

This work was supported by NIH grant R01 AI47171.

The authors do not have any conflicting financial interests.

 
* Corresponding author. Mailing address: Division of Rheumatology, Immunology, and Allergy, Brigham and Women's Hospital, Smith Building Room 516C, One Jimmy Fund Way, Boston, MA 02115. Phone: (617) 525-1033. Fax: (617) 525-1010. E-mail: sbehar{at}rics.bwh.harvard.edu

Published ahead of print on 30 June 2008.

Editor: J. L. Flynn

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Behr, M. A., M. A. Wilson, W. P. Gill, H. Salamon, G. K. Schoolnik, S. Rane, and P. M. Small. 1999. Comparative genomics of BCG vaccines by whole-genome DNA microarray. Science 284:1520-1523.[Abstract/Free Full Text]
2
2. Brodin, P., M. I. de Jonge, L. Majlessi, C. Leclerc, M. Nilges, S. T. Cole, and R. Brosch. 2005. Functional analysis of early secreted antigenic target-6, the dominant T-cell antigen of Mycobacterium tuberculosis, reveals key residues involved in secretion, complex formation, virulence, and immunogenicity. J. Biol. Chem. 280:33953-33959.[Abstract/Free Full Text]
3
3. Brodin, P., L. Majlessi, L. Marsollier, M. I. de Jonge, D. Bottai, C. Demangel, J. Hinds, O. Neyrolles, P. D. Butcher, C. Leclerc, S. T. Cole, and R. Brosch. 2006. Dissection of ESAT-6 system 1 of Mycobacterium tuberculosis and impact on immunogenicity and virulence. Infect. Immun. 74:88-98.[Abstract/Free Full Text]
4
4. Brodin, P., I. Rosenkrands, P. Andersen, S. T. Cole, and R. Brosch. 2004. ESAT-6 proteins: protective antigens and virulence factors? Trends Microbiol. 12:500-508.[CrossRef][Medline]
5
5. Chackerian, A. A., J. M. Alt, T. V. Perera, C. C. Dascher, and S. M. Behar. 2002. Dissemination of Mycobacterium tuberculosis is influenced by host factors and precedes the initiation of T-cell immunity. Infect. Immun. 70:4501-4509.[Abstract/Free Full Text]
6
6. Chackerian, A. A., T. V. Perera, and S. M. Behar. 2001. Gamma interferon-producing CD4⁺ T lymphocytes in the lung correlate with resistance to infection with Mycobacterium tuberculosis. Infect. Immun. 69:2666-2674.[Abstract/Free Full Text]
7
7. Cleary, A. M., S. M. Fortune, M. J. Yellin, L. Chess, and S. Lederman. 1995. Opposing roles of CD95 (Fas/APO-1) and CD40 in the death and rescue of human low density tonsillar B cells. J. Immunol. 155:3329-3337.[Abstract]
8
8. Converse, S. E., and J. S. Cox. 2005. A protein secretion pathway critical for Mycobacterium tuberculosis virulence is conserved and functional in Mycobacterium smegmatis. J. Bacteriol. 187:1238-1245.[Abstract/Free Full Text]
9
9. Denis, O., A. Tanghe, K. Palfliet, F. Jurion, T. P. van den Berg, A. Vanonckelen, J. Ooms, E. Saman, J. B. Ulmer, J. Content, and K. Huygen. 1998. Vaccination with plasmid DNA encoding mycobacterial antigen 85A stimulates a CD4⁺ and CD8⁺ T-cell epitopic repertoire broader than that stimulated by Mycobacterium tuberculosis H37Rv infection. Infect. Immun. 66:1527-1533.[Abstract/Free Full Text]
10
10. DiGiuseppe Champion, P. A., and J. S. Cox. 2007. Protein secretion systems in Mycobacteria. Cell. Microbiol. 9:1376-1384.[CrossRef][Medline]
11
11. Fortune, S. M., A. Jaeger, D. A. Sarracino, M. R. Chase, C. M. Sassetti, D. R. Sherman, B. R. Bloom, and E. J. Rubin. 2005. Mutually dependent secretion of proteins required for mycobacterial virulence. Proc. Natl. Acad. Sci. USA 102:10676-10681.[Abstract/Free Full Text]
12
12. Gao, L. Y., S. Guo, B. McLaughlin, H. Morisaki, J. N. Engel, and E. J. Brown. 2004. A mycobacterial virulence gene cluster extending RD1 is required for cytolysis, bacterial spreading and ESAT-6 secretion. Mol. Microbiol. 53:1677-1693.[CrossRef][Medline]
13
13. Guinn, K. M., M. J. Hickey, S. K. Mathur, K. L. Zakel, J. E. Grotzke, D. M. Lewinsohn, S. Smith, and D. R. Sherman. 2004. Individual RD1-region genes are required for export of ESAT-6/CFP-10 and for virulence of Mycobacterium tuberculosis. Mol. Microbiol. 51:359-370.[CrossRef][Medline]
14
14. Hervas-Stubbs, S., L. Majlessi, M. Simsova, J. Morova, M. J. Rojas, C. Nouze, P. Brodin, P. Sebo, and C. Leclerc. 2006. High frequency of CD4⁺ T cells specific for the TB10.4 protein correlates with protection against Mycobacterium tuberculosis infection. Infect. Immun. 74:3396-3407.[Abstract/Free Full Text]
15
15. Hsu, T., S. M. Hingley-Wilson, B. Chen, M. Chen, A. Z. Dai, P. M. Morin, C. B. Marks, J. Padiyar, C. Goulding, M. Gingery, D. Eisenberg, R. G. Russell, S. C. Derrick, F. M. Collins, S. L. Morris, C. H. King, and W. R. Jacobs, Jr. 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. USA 100:12420-12425.[Abstract/Free Full Text]
16
16. Kamath, A., J. S. Woodworth, and S. M. Behar. 2006. Antigen-specific CD8+ T cells and the development of central memory during Mycobacterium tuberculosis infection. J. Immunol. 177:6361-6369.[Abstract/Free Full Text]
17
17. Kamath, A. B., J. Alt, H. Debbabi, and S. M. Behar. 2003. Toll-like receptor 4-defective C3H/HeJ mice are not more susceptible than other C3H substrains to infection with Mycobacterium tuberculosis. Infect. Immun. 71:4112-4118.[Abstract/Free Full Text]
18
18. Kamath, A. B., J. Alt, H. Debbabi, C. Taylor, and S. M. Behar. 2004. The major histocompatibility complex haplotype affects T-cell recognition of mycobacterial antigens but not resistance to Mycobacterium tuberculosis in C3H mice. Infect. Immun. 72:6790-6798.[Abstract/Free Full Text]
19
19. Kamath, A. B., J. Woodworth, X. Xiong, C. Taylor, Y. Weng, and S. M. Behar. 2004. Cytolytic CD8+ T cells recognizing CFP10 are recruited to the lung after Mycobacterium tuberculosis infection. J. Exp. Med. 200:1479-1489.[Abstract/Free Full Text]
20
20. Kampmann, B., C. Hemingway, A. Stephens, R. Davidson, A. Goodsall, S. Anderson, M. Nicol, E. Scholvinck, D. Relman, S. Waddell, P. Langford, B. Sheehan, L. Semple, K. A. Wilkinson, R. J. Wilkinson, S. Ress, M. Hibberd, and M. Levin. 2005. Acquired predisposition to mycobacterial disease due to autoantibodies to IFN-. J. Clin. Investig. 115:2480-2488.[CrossRef][Medline]
21
21. Lewinsohn, D. M., L. Zhu, V. J. Madison, D. C. Dillon, S. P. Fling, S. G. Reed, K. H. Grabstein, and M. R. Alderson. 2001. Classically restricted human CD8+ T lymphocytes derived from Mycobacterium tuberculosis-infected cells: definition of antigenic specificity. J. Immunol. 166:439-446.[Abstract/Free Full Text]
22
22. Lewis, K. N., R. Liao, K. M. Guinn, M. J. Hickey, S. Smith, M. A. Behr, and D. R. Sherman. 2003. Deletion of RD1 from Mycobacterium tuberculosis mimics bacille Calmette-Guerin attenuation. J. Infect. Dis. 187:117-123.[CrossRef][Medline]
23
23. MacGurn, J. A., S. Raghavan, S. A. Stanley, and J. S. Cox. 2005. A non-RD1 gene cluster is required for Snm secretion in Mycobacterium tuberculosis. Mol. Microbiol. 57:1653-1663.[CrossRef][Medline]
24
24. MacGurn, J. A., and J. S. Cox. 2007. A genetic screen for Mycobacterium tuberculosis mutants defective for phagosome maturation arrest identifies components of the ESX-1 secretion system. Infect. Immun. 75:2668-2678.[Abstract/Free Full Text]
25
25. Majlessi, L., P. Brodin, R. Brosch, M. J. Rojas, H. Khun, M. Huerre, S. T. Cole, and C. Leclerc. 2005. Influence of ESAT-6 secretion system 1 (RD1) of Mycobacterium tuberculosis on the interaction between mycobacteria and the host immune system. J. Immunol. 174:3570-3579.[Abstract/Free Full Text]
26
26. Majlessi, L., M. J. Rojas, P. Brodin, and C. Leclerc. 2003. CD8⁺-T-cell responses of Mycobacterium-infected mice to a newly identified major histocompatibility complex class I-restricted epitope shared by proteins of the ESAT-6 family. Infect. Immun. 71:7173-7177.[Abstract/Free Full Text]
27
27. Precopio, M. L., M. R. Betts, J. Parrino, D. A. Price, E. Gostick, D. R. Ambrozak, T. E. Asher, D. C. Douek, A. Harari, G. Pantaleo, R. Bailer, B. S. Graham, M. Roederer, and R. A. Koup. 2007. Immunization with vaccinia virus induces polyfunctional and phenotypically distinctive CD8+ T cell responses. J. Exp. Med. 204:1405-1416.[Abstract/Free Full Text]
28
28. Pym, A. S., P. Brodin, R. Brosch, M. Huerre, and S. T. Cole. 2002. Loss of RD1 contributed to the attenuation of the live tuberculosis vaccines Mycobacterium bovis BCG and Mycobacterium microti. Mol. Microbiol. 46:709-717.[CrossRef][Medline]
29
29. Pym, A. S., P. Brodin, L. Majlessi, R. Brosch, C. Demangel, A. Williams, K. E. Griffiths, G. Marchal, C. Leclerc, and S. T. Cole. 2003. Recombinant BCG exporting ESAT-6 confers enhanced protection against tuberculosis. Nat. Med. 9:533-539.[CrossRef][Medline]
30
30. Rapoport, T. A. 2007. Protein translocation across the eukaryotic endoplasmic reticulum and bacterial plasma membranes. Nature 450:663-669.[CrossRef][Medline]
31
31. Renshaw, P. S., P. Panagiotidou, A. Whelan, S. V. Gordon, R. G. Hewinson, R. A. Williamson, and M. D. Carr. 2002. Conclusive evidence that the major T-cell antigens of the Mycobacterium tuberculosis complex ESAT-6 and CFP-10 form a tight, 1:1 complex and characterization of the structural properties of ESAT-6, CFP-10, and the ESAT-6*CFP-10 complex. Implications for pathogenesis and virulence. J. Biol. Chem. 277:21598-21603.[Abstract/Free Full Text]
32
32. Sassetti, C. M., and E. J. Rubin. 2003. Genetic requirements for mycobacterial survival during infection. Proc. Natl. Acad. Sci. USA 100:12989-12994.[Abstract/Free Full Text]
33
33. Schaible, U. E., F. Winau, P. A. Sieling, K. Fischer, H. L. Collins, K. Hagens, R. L. Modlin, V. Brinkmann, and S. H. Kaufmann. 2003. Apoptosis facilitates antigen presentation to T lymphocytes through MHC-I and CD1 in tuberculosis. Nat. Med. 9:1039-1046.[CrossRef][Medline]
34
34. Shams, H., P. Klucar, S. E. Weis, A. Lalvani, P. K. Moonan, H. Safi, B. Wizel, K. Ewer, G. T. Nepom, D. M. Lewinsohn, P. Andersen, and P. F. Barnes. 2004. Characterization of a Mycobacterium tuberculosis peptide that is recognized by human CD4+ and CD8+ T cells in the context of multiple HLA alleles. J. Immunol. 173:1966-1977.[Abstract/Free Full Text]
35
35. Shen, L., and K. L. Rock. 2006. Priming of T cells by exogenous antigen cross-presented on MHC class I molecules. Curr. Opin. Immunol. 18:85-91.[CrossRef][Medline]
36
36. Stanley, S. A., S. Raghavan, W. W. Hwang, and J. S. Cox. 2003. Acute infection and macrophage subversion by Mycobacterium tuberculosis require a specialized secretion system. Proc. Natl. Acad. Sci. USA 100:13001-13006.[Abstract/Free Full Text]
37
37. Tan, T., W. L. Lee, D. C. Alexander, S. Grinstein, and J. Liu. 2006. The ESAT-6/CFP-10 secretion system of Mycobacterium marinum modulates phagosome maturation. Cell. Microbiol. 8:1417-1429.[CrossRef][Medline]
38
38. Tian, T., J. Woodworth, M. Skold, and S. M. Behar. 2005. In vivo depletion of CD11c+ cells delays the CD4+ T cell response to mycobacterium tuberculosis and exacerbates the outcome of infection. J. Immunol. 175:3268-3272.[Abstract/Free Full Text]
39
39. van der Wel, N. N., D. Hava, D. Houben, D. Fluitsma, M. van Zon, M. Brenner, and P. J. Peters. 2007. M. tuberculosis and M. leprae translocate from the phagolysosome to the cytosol in myeloid cells. Cell 129:1287-1298.[CrossRef][Medline]
40
40. Winau, F., S. Weber, S. Sad, J. de Diego, S. L. Hoops, B. Breiden, K. Sandhoff, V. Brinkmann, S. H. Kaufmann, and U. E. Schaible. 2006. Apoptotic vesicles crossprime CD8 T cells and protect against tuberculosis. Immunity 24:105-117.[CrossRef][Medline]
41
41. Wolf, A. J., B. Linas, G. J. Trevejo-Nunez, E. Kincaid, T. Tamura, K. Takatsu, and J. D. Ernst. 2007. Mycobacterium tuberculosis infects dendritic cells with high frequency and impairs their function in vivo. J. Immunol. 179:2509-2519.[Abstract/Free Full Text]
42
42. Woodworth, J. S., and S. M. Behar. 2006. Mycobacterium tuberculosis-specific CD8+ T cells and their role in immunity. Crit. Rev. Immunol. 26:317-352.[Medline]

---

Infection and Immunity, September 2008, p. 4199-4205, Vol. 76, No. 9
0019-9567/08/$08.00+0     doi:10.1128/IAI.00307-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services E-mail this article to a friend Similar articles in this journal Similar articles in ASM journals Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Woodworth, J. S. Articles by Behar, S. M. Search for Related Content PubMed PubMed Citation Articles by Woodworth, J. S. Articles by Behar, S. M.