Delayed protection by ESAT-6–specific effector CD4⁺ T cells after airborne M. tuberculosis infection

ESAT-6 is a secreted protein implicated in M. tuberculosis pathogenesis (17) and is a key antigen in humans, primates, and mice, and vaccination with ESAT-6 induces protective CD4⁺ T cell immunity (18–22). To study the CD4⁺ T cell response to M. tuberculosis, tg mice for an MHC class II–restricted TCR specific for ESAT-6_3-15 in the context of I-A^b were generated. To isolate a TCR that is functional during in vivo infection, CD4⁺ T cells specific for ESAT-6 were harvested from mice that were aerosol infected with M. tuberculosis, and CD4⁺ T cell hybridomas were generated. The specificity of one ESAT-6–specific hybridoma, clone 7, is shown in Fig. 1 A. The TCR used by clone 7 CD4⁺ hybridoma cells was composed of a TCR β chain generated from Vβ10b-Dβ1-Jβ2.4 gene segments and a TCR α chain generated from Vα11.2-Jα27 gene segments (unpublished data). TCR α and β chain cDNAs were cloned into the VA hCD2 cassette vector, which uses the human CD2 promoter to drive the expression of cDNAs in T cells (23).

Because the TCR used to generate ESAT-6–specific TCR tg mice (referred to hereafter as C7 TCR tg mice) was isolated from a CD4⁺ T cell, an increased frequency of this lineage in tg animals was expected. In both thymocytes and splenocytes, an increased percentage of CD4⁺ T cells with a concomitant decrease in the frequency of CD8⁺ T cells was observed (Fig. 1 B). The majority of CD4⁺ thymocytes and splenocytes expressed the transgene-encoded TCR α and β chain (Fig. 1 C).

To confirm the specificity of these cells, CD4⁺ T cells were purified from C7 TCR tg mice or B6 control animals and were stimulated with titrating amounts of ESAT-6 peptide. Although both CD4⁺ T cell populations proliferated similarly to anti-CD3/CD28 stimulation (not depicted), only C7 TCR tg CD4⁺ T cells proliferated in response to ESAT-6 peptide (Fig. 1 D). In summary, C7 TCR tg mice contain CD4⁺ T cells that express a TCR specific to ESAT-6_3-15–I-A^b complexes. This TCR was isolated from a CD4⁺ T cell clone responding to a live M. tuberculosis infection.

To determine whether C7 TCR tg CD4⁺ T cells respond to ESAT-6 antigen generated during aerosol infection of mice, 10⁴ naive C7 TCR tg CD4⁺ T cells (on a RAG-1–deficient [RAG ^−/−] background) were adoptively transferred into B6 recipients that were aerosol infected 1 d later with 100 CFU M. tuberculosis. The data in Fig. 2 A show that at this T cell dosage, naive C7 CD4⁺ T cells (congenically marked with CD90.1) are not detectable in uninfected mice. In contrast, marked expansion of C7 CD4⁺ T cells is noted in the pulmonary LNs (pLNs) of mice infected 12 d earlier (Fig. 2, A and B). Analysis of pLNs, spleen, and lung between days 12 and 18 of infection demonstrates that the frequency of C7 CD4⁺ T cells increases in the lung and spleen (Fig. 2 C). Staining for intracellular TNF-α and IFN-γ demonstrates that responding C7 CD4⁺ T cells acquired Th1 effector cell functions (Fig. 2 D). The differentiation of responding cells into Th1 effector cells is further supported by their expression of T-bet, as measured by intracellular staining (Fig. 2 E). In contrast, naive C7 CD4⁺ T cells did not differentiate into regulatory T cells because the transcription factor Foxp3 was not expressed (unpublished data).

These experiments demonstrate that naive C7 CD4⁺ T cells are activated in vivo in the setting of pulmonary M. tuberculosis infection and that they traffic from draining LNs to the lung. Furthermore, these studies demonstrate that naive C7 TCR tg CD4⁺ T cells differentiate into Th1 effector cells, as is observed with polyclonal ESAT-6–specific T cell responses (24).

The ESAT-6–specific CD4⁺ T cell response to M. tuberculosis has been measured in functional ways via ELISA, ELISPOT, or intracellular cytokine staining in mice with a complex TCR repertoire (24, 25). One limitation of relying on these methods is that the frequency of antigen-specific cells needs to be high enough to detect by fluorimetric analysis, and these cells need to produce cytokines to be detected. Accordingly, naive cells cannot be studied because these cells need to expand and differentiate into cytokine-producing cells before they can be visualized. Using these methods, M. tuberculosis–specific effector CD4⁺ T cells are first detectable in the pLNs ∼14 d after infection, and these cells reach peak numbers in the lungs between 21 and 28 d after infection (24–26). Because the delay in recruitment of effector cells to the lung is not a general characteristic of pulmonary infections, we wanted to determine the kinetics of priming of ESAT-6–specific CD4⁺ T cells (27, 28).

To understand when naive ESAT-6–specific CD4⁺ T cells are first presented antigen after aerosol infection with M. tuberculosis, 10⁵ CFSE-labeled naive C7 TCR tg.RAG^−/− CD4⁺ T cells (CD90.1) were transferred into B6 recipients (CD90.2), and recipient mice were infected 1 d later with 100 CFU M. tuberculosis. At days 7, 10, and 13 after infection, pLNs were recovered and the frequency and phenotype of C7 CD4⁺ T cells were determined. At day 7 after infection, the frequency of C7 CD4⁺ T cells in the pLNs was low and similar to that seen in uninfected animals. C7 CD4⁺ T cells at day 7 had a naive phenotype. These cells were CFSE bright, CD69 negative, CD44 intermediate, and CD62L high (Fig. 3 A and not depicted). The naive phenotype of these cells, with low surface expression of the early activation marker CD69, indicates that C7 CD4⁺ T cells had not been presented antigen at the day 7 time point. By day 10 after infection, the frequency of C7 CD4⁺ T cells among CD4⁺ T cells increased compared with day 7, and every mouse examined contained either recently activated cells (expressing CD69) or divided cells, as indicated by CFSE dilution. At day 13 after infection, the frequency of the C7 CD4⁺ T cells increased to ∼6% of total CD4⁺ T cells. These cells were mostly CFSE negative and displayed an effector/memory-like cell-surface phenotype, expressing high levels CD44 and low levels of CD62L (Fig. 3 A and not depicted). The temporal change in frequency of C7 CD4⁺ T cells among the CD4⁺ T cells in the pLNs is shown in Fig. 3 B. These experiments indicate that the priming of naive ESAT-6–specific CD4⁺ T cells is delayed compared with other respiratory infections and occurs between days 7 and 10 after infection.

Mice immunized with M. tuberculosis proteins plus adjuvant, BCG variants, or M. tuberculosis are, in comparison to naive controls, protected from challenge with M. tuberculosis. These immunization protocols induce complex populations of M. tuberculosis–specific CD4⁺ and CD8⁺ T cells that contribute to protection. The CD4⁺ T cell populations are mixed and may include Th1, Th17, and suppressive regulatory T cell populations (14–16). In one report, in vitro–generated monoclonal Th1 cells (specific to ovalbumin) were shown to provide ∼1 log of protection in animals infected with M. tuberculosis expressing the model antigen ovalbumin (29).

Because ESAT-6 is a native M. tuberculosis antigen and a target for vaccine development (30), we wanted to understand if a pure population of ESAT-6–specific C7 Th1 cells could protect mice from infection. B6 hosts received 10⁷ in vitro–generated C7 Th1 cells (or control Th1 cells) and were then infected with 100 CFU M. tuberculosis 1 d later. Before transfer, the differentiation status of the cells was verified by analysis of CD44 and T-bet expression (Fig. 4 A). 24 d after infection, the number of bacteria in the lungs was determined. Surprisingly, C7 Th1 cells decreased the number of viable bacteria by a factor of 100, whereas animals that received control Th1 cells contained similar numbers of bacteria as mice that received no cells (Fig. 4 B). Next, we sought to determine if fewer C7 Th1 cells could protect animals from M. tuberculosis challenge. Graded numbers of C7 Th1 cells (10⁵, 10⁶, or 10⁷) were transferred into B6 recipients, and bacterial numbers were determined 16, 30, and 90 d after infection. These results show that the degree of protection correlates with the dose of C7 Th1 cells transferred. Administration of as few as 10⁵ cells was sufficient to provide significant protection that was sustained for up to 90 d after infection (Fig. 4, C and D). Interestingly, although the day 30 numbers of bacteria were maintained until day 90 after infection in animals that received low doses of cells, bacterial numbers increased in animals that received 10⁷ C7 Th1 cells (Fig. 4 D).

These studies demonstrate that ESAT-6–specific Th1 cells can greatly reduce the numbers of M. tuberculosis in the lungs. Compared with control mice, animals that received 10⁷ C7 Th1 cells had a 99% reduction in the number of viable bacteria early after infection.

Use of C7 TCR tg cells in the previous experiments demonstrated that naive ESAT-6–specific CD4⁺ T cells were primed in the pLNs between 7 and 10 d after infection, and these cells differentiated into Th1 effector cells and trafficked to the lungs between 12 and 14 d after infection (Figs. 2 and 3). These observations are consistent with other studies that have measured endogenous T cell responses to M. tuberculosis and demonstrated that antigen-specific T cell responses were detected in the pLNs 14 d after infection, and reached peak numbers in the lungs ∼28 d after infection, at which time in vivo bacterial growth was controlled (24–26).

Our experiments with in vitro–generated C7 Th1 cells demonstrated that these cells greatly reduced the bacterial numbers in the lungs as early as day 16 after infection. In the ensuing experiments, we wanted to look earlier than day 16 after infection to determine when C7 Th1 cells manifest protection. 10⁷ C7 Th1 cells were transferred into mice that were infected 1 d later with M. tuberculosis, and bacterial numbers in the lungs were determined early after infection. Before infection, ∼5 × 10⁵ C7 Th1 cells were recovered from either spleen or lungs, and these cells were functional, producing both IFN-γ and TNF-α after ex vivo stimulation (Fig. 5, A–C).

To our surprise, despite the high numbers of C7 effector cells in the lungs, similar numbers of bacteria were recovered from animals that received C7 Th1 cells and from control animals during the first 7 d of infection. In contrast, between days 7 and 10 after infection, while bacterial growth continued in control mice, growth was controlled in animals that received C7 Th1 cells. Interestingly, the number of bacteria remained constant until day 30 after infection despite the presence of C7 Th1 cells capable of producing cytokines in the lungs (Fig. 5, D and E).

The finding that C7 Th1 cells were in the lungs before infection, but did not control bacterial replication until after day 7 following infection, led us to investigate whether these cells were activated during the first week of infection. Unlike our studies using naive C7 cells, we could not determine, using the activation markers CD69 and CD25, when C7 Th1 effector cells were presented antigen in the lung, because ∼60% of C7 Th1 cells expressed these activation markers regardless of whether these cells were harvested from naive or infected animals (Fig. S1, available). This result is consistent with studies of memory CD8 T cells in the lung and suggests that CD69 cannot be used to monitor presentation of cognate antigen to antigen-experienced T cells in the lung (31). In addition, expansion of C7 Th1 cells was not evident in infected animals between days 4 and 13 after infection, because similar numbers of C7 Th1 cells were recovered from naive and infected mice (Fig. S1). As a third attempt to understand if C7 Th1 cells were activated during the first 7 d of infection, message levels for ifnγ and one target gene, nos2, were measured in infected mice that received C7 Th1 cells. Similar levels of these transcripts were observed from days 1 to 7 after infection, whereas both transcripts were induced by day 18 after infection. This suggests that C7 Th1 cells were not activated to produce high levels of IFN-γ during the first week of infection. Collectively, these data do not support the hypothesis that C7 Th1 cells are activated during the first 7 d of infection. However, these negative data also do not rule out the possibility that some C7 Th1 cells (perhaps very small numbers) “see” antigen and are activated during the first week of infection.

In summary, the results show that effector ESAT-6–specific Th1 cells greatly reduce the number of viable M. tuberculosis in the lungs and that control of bacterial growth occurs after day 7 following infection. This delay in protection is unlikely to be caused by a lack of ESAT-6 synthesis during the first week of infection, because ESAT-6 transcript was detectable by at least day 5 after infection in the lungs of infected mice, and ESAT-6–deficient M. tuberculosis showed reduced growth in the lungs (compared with wild-type M. tuberculosis) during the first 5 d of infection, demonstrating that ESAT-6 is synthesized very early after infection (32, 33).

We generated mice tg for a TCR specific to the dominant M. tuberculosis antigen ESAT-6 to investigate the ability of Th1 cells to provide protective immunity. Our analysis of adoptively transferred C7 TCR tg CD4⁺ T cells revealed that priming of naive ESAT-6–specific TCR tg cells occurred 7–10 d after aerosol infection with a low dose of M. tuberculosis. Our experiments demonstrated that adoptively transferred Th1 cells specific for ESAT-6 can provide a high level of protection, as measured up to 90 d after infection, but that in vivo mycobacterial growth is unimpaired for the first week of infection despite large numbers of pathogen-specific T cells in the lungs.

The delay in priming of naive ESAT-6–specific T cells is not unexpected and is consistent with the finding that M. tuberculosis–specific effector T cells are not present at high frequencies in the lungs of mice until ∼3 wk after infection (24, 25). Our results are also consistent with a recent study (34) that monitored the priming of TCR tg CD4⁺ T cells specific to the M. tuberculosis antigen Ag85B. One notable difference is that priming of Ag85B-specific monoclonal CD4⁺ T cells occurred at day 12 after aerosol infection, 2–4 d later than the C7 TCR tg cell response documented in this study. Differences in priming between naive Ag85B and ESAT-6–specific clones may result from differences in the expression patterns of these two proteins (32), differences in the presentation of these antigens, or differences in the sensitivity of the TCRs used by Ag85B TCR tg cells and C7 TCR tg cells for their cognate antigens. Direct comparative studies will be required to distinguish between these possibilities.

Our studies with in vitro–generated Th1 cells revealed two surprising findings. One is how well monoclonal ESAT-6–specific Th1 cells were able to protect mice from M. tuberculosis infection. The second is the fact that despite high numbers of C7 Th1 effector cells in animals before infection, bacterial replication was not controlled until 1 wk after infection.

The 2-log reduction in the number of bacilli in animals that received C7 Th1 cells is greater than that observed after BCG immunization and comparable to prime-boost vaccination methods (35, 36). By using adoptively transferred C7 Th1 cells, our experiments demonstrate that monoclonal T cells specific to ESAT-6 can provide a level of early protection that is comparable to that provided by mixed populations of M. tuberculosis–specific CD4⁺ and CD8⁺ effector T cells after vaccination. Thus, complex populations of T cells are not required to control infection, a result that suggests that adoptive T cell transfer may provide a therapeutic option for drug-resistant M. tuberculosis infections.

It is presumed that improved vaccines can be designed to protect individuals from M. tuberculosis infection by eliciting robust antigen-specific T cell responses. This view is supported by the observation that after aerosol infection in mice, accumulation of M. tuberculosis CD4⁺ T cells in the lungs coincides with the control of bacterial growth (6, 14, 37). Several hypotheses for a lack of sterilizing immunity to M. tuberculosis can be proposed, including a delayed recruitment of effector T cells to the lungs. Our ability to track adoptively transferred M. tuberculosis–specific C7 Th1 cells allowed us to determine that approximately a half million of these cells were in the lungs before infection and allowed us to definitively rule out this hypothesis.

What accounts for the delayed protection by C7 Th1 cells? Given the low dose infection used in our studies (100 bacteria) and the slow doubling time of M. tuberculosis during the first several days of infection, very few cells may be infected, and the probability of C7 Th1 cells “finding” an infected cell may be low. Although several cell types, including macrophages and DCs, are infected by day 14 after infection (38), no information exists on the frequency or phenotype of infected cells in the lungs during the first week of infection (although presumably alveolar macrophages are the first cells infected). A recent study using multiphoton microscopy found that liver-resident macrophages, Kupffer cells, were the first cells infected with BCG after i.v. infection (39). Recruitment of uninfected Kupffer cells and blood-derived monocytes to infected cells occurred 2 wk after infection.

If similar delays in cell recruitment occur in the lungs after M. tuberculosis infection, it could take 7–10 d until lung resident cells (i.e., macrophages, DCs, and effector/memory T cells) home to infected cells. Before day 7 after infection, C7 Th1 cells may randomly encounter infected cells, whereas after day 7, migration within the lungs to infected sites could be directed by inflammatory signals. Our studies with naive C7 CD4⁺ T cells demonstrated that T cell priming in pLNs also occurred only after day 7 following infection. A similar mechanism may account for the delayed priming of naive CD4⁺ T cells. After the first week of infection, DCs and other APCs may home to infected sites within the lung and traffic to draining LNs to prime naive T cells. This hypothesis requires further investigation.

On the other hand, if C7 Th1 cells home to infected cells within the lungs promptly after infection, it would suggest defective collaboration between CD4⁺ effector cells and infected cells: infected cells may be unable to present antigen to activate effector CD4⁺ T cells, or infected cells may be unable to control bacterial replication after stimulation by IFN-γ and other factors provided by effector CD4⁺ T cells. In vitro studies support both hypotheses, but in vivo studies to answer these important questions are necessary (40–45).

The findings reported in this paper have important implications for understanding immunity to M. tuberculosis. In this study, we show that high numbers of ESAT-6–specific Th1 cells provided substantial protection to mice after low dose aerosol infection with M. tuberculosis. However, this protection was delayed until after day 7 following infection, and bacteria persisted in these animals thereafter. This finding may help explain why vaccination against M. tuberculosis has been difficult: immediately after infection, M. tuberculosis resides in cells that CD4⁺ T cells cannot help.

To generate ESAT-6–specific CD4⁺ T cell hybridomas, B6 mice were aerosol infected with ∼100 CFU M. tuberculosis Erdman, and 6 wk after infection, CD4⁺ splenocytes were purified and stimulated in the presence of irradiated T cell–depleted splenocytes (APCs) and 0.5 μg/ml of ESAT-6 protein (the expression vector used to make ESAT-6 protein, pMRLB7, was provided by Mycobacteria Research Laboratories at Colorado State University). 2 d later, activated CD4⁺ T cells were fused with the tumor cell line BW5147. Antigen-specific CD4⁺ T cell hybridomas were selected by screening for IL-2 production in response to ESAT-6 protein and ESAT-6_1-20 peptide. TCR chains were identified by a combination of surface staining with a panel of TCR Vα and Vβ chain monoclonal antibodies (BD Biosciences), and by RT-PCR amplification and DNA sequencing. Hybridoma clone 7 was chosen to generate TCR tg mice because it expressed only one in-frame TCR α chain, and monoclonal antibodies were available for both TCR chains. The TCR chains conferring the specificity of clone 7 hybridoma cells were identified as Vβ10b-Dβ1.1-Jβ2.4 and Vα11.2-Jα27. TCR α chain variable region (VJ) and β chain variable region (VDJ) cDNAS were amplified with the following primers: Vβ10b-Dβ1.1-Jβ2.4, (5′ primer) 5′-tgctgccctcgaggatcgacgaattcacaggggccatgcagaggaa-3′ and (3′ primer) 5′-agcaggttccggattctggatgtttgg-3′; and Vα11.2-Jα27, (5′ primer) 5′-tgctgcctcgaggatcgacgaattcgacccaactatgggctgta-3′ and (3′ primer) 5′-tttctcagatcttctagcaccgatagtcgg-3′.

VJ and VDJ cDNAS were subcloned into pBluescript vectors that contained TCR α or β chain constant regions. Complete TCR α and β chains were excised and cloned into the VA hCD2 vectors (pBluescript and VA hCD2 were provided by E. Huesby, University of Massachusetts Medical School, Worcester, MA) (23). DNA fragments containing the TCR cDNAs were injected into fertilized B6 oocysts. Founders and their progeny were screened by flow cytometry for surface expression of Vβ10b and Vα11.1/.2. TCR tg founder number 27 was selected and used in all experiments presented in this paper.

B6 mice were purchased from the Jackson Laboratory. All animal procedures were approved by the Memorial Sloan-Kettering Institutional Animal Care and Use Committee. C7 TCR tg mice were bred to B6.CD90.1 mice to obtain C7 TCR tg.CD90.1 mice, and to RAG^−/−.CD90.1 mice to generate C7 TCR tg.RAG^−/−.CD90.1 mice. C7 TCR tg mice were maintained at the animal facility in the Memorial Sloan-Kettering Research Animal Resource Center. For adoptive transfer experiments, C7 TCR tg CD4⁺ T cells were isolated from the spleen and LNs of TCR tg mice. Cells were injected i.v. into naive recipients 1 d before infection. In some experiments cells were labeled with 5 μM CFSE before transfer.

3 × 10⁶ purified C7 TCR tg CD4⁺ T cells were cultured with 12 × 10⁶ irradiated T cell–depleted splenocytes. 5 μg/ml of ESAT-6_1-20 peptide, 10 ng/ml IL-12, and 5 μg/ml of neutralizing anti–IL-4 antibody (R&D Systems) were added at day 0 of culture. At days 2 and 3 of culture, the cells were split 1:2, and 50 U/ml IL-2 was added (R&D Systems). On day 4 of culture, Th1 cell differentiation was confirmed via intracellular expression of T-bet. The same protocol was used to make polyclonal Th1 CD4⁺ T cells, except the CD4⁺ T cells were taken from B6 mice and stimulated in the presence of 1 μg/ml anti-CD3 (145-2C11) and 0.5 μg/ml anti-CD28 (37.51). Th1 cells were washed three times with PBS and injected i.v. into B6 mice.

5 × 10⁴ clone 7 hybridoma cells, BW5147 cells, or purified CD4⁺ T cells harvested from B6 or C7 TCR tg mice were plated with 1.5 × 10⁵ irradiated T cell–depleted splenocytes in 96-well round-bottom plates in the presence of various concentrations of ESAT-6_1-20 peptide. For proliferation experiments, cells were pulsed between days 2 and 3 of culture by the addition of 1 μCi [³H]thymidine per well for 16 h before harvesting cells onto glass-fiber filters and determining the incorporated radioactivity using a liquid scintillation counter (TopCount; PerkinElmer). For IL-2 ELISAs, supernantants were harvested after 2 d of stimulation, and IL-2 levels were measured with a mouse IL-2 ELlSA set (BD Biosciences).

To detect intracellular cytokines 3 × 10⁶ cells from pLNs, spleen, or collagenase-digested and -perfused lungs were plated in 96-well round-bottom plates in the presence of GolgiPlug (BD Biosciences) and 5 μg/ml ESAT-6_1-20 peptide. 5 h later, cells were washed and stained with cell-surface antibodies, and intracellular cytokine (ICC) staining was performed according to the manufacturer's instructions. Cells harvested from M. tuberculosis–infected mice were fixed with 2% paraformaldehyde overnight before removal from the biosafety level 3 facility. These cells were washed and permeabilized, and ICC staining was performed. Intracellular T-bet and isotype control staining (anti–T-bet–PE 4B10 [Santa Cruz Biotechnology, Inc.] and mouse IgG1-PE [BD Biosciences]) were performed using reagents supplied with the mouse regulatory T cell staining kit (eBioscience).

M. tuberculosis Erdman was grown in 7H9 media, and log-phase cultures were diluted to 8 × 10⁶ M. tuberculosis per millimeter and were sonicated before infection with an aerosol exposure system (Glass-Col). 8–10-wk-old B6 mice were infected with a volume of suspension, and exposure time was calibrated to deliver ∼100 CFU per animal. To determine the infection dose, three mice were killed 1 d after infection and lungs were harvested and homogenized in PBS/0.05% Tween-80, and half of the lung homogenate was plated. At various time intervals after infection, the right lung was harvested from individual mice and homogenized in PBS/0.05% Tween-80. Serial dilutions were made in PBS/0.05% Tween-80 and plated onto Middlebrook 7H10 agar (BD Biosciences). After 3–4 wk of incubation at 37°C in a 5% CO₂ atmosphere, mycobacterial colonies were counted and the bacterial numbers in the lungs were calculated.

Total RNA was isolated from lung parenchyma using TRIzol reagent (Invitrogen), and cDNA was synthesized from DNase-treated RNA using oligo(dT) primers and SuperScript II (Invitrogen). RNA was pooled from four mice per group. SYBR green–based real-time PCR analysis was performed using the DynNAmo SYBR green qPCR kit (Finnzymes). ifnγ and nos2 primers were purchased from QIAGEN. Signals were normalized to GAPDH transcript (forward, 5′-ACCACAGTCCATGCCATCAC-3′; reverse, 5′-TCCACCACCCTGTTGCTGTA-3′). To determine levels of ifnγ and nos2, the following equation was used, where n equals the difference in PCR cycle number between ifnγ and GAPDH at which the signals were first detected above background (Ct): 1/2^{n = Ct ifnγ − Ct GAPDH}.

Statistical analysis was performed with the unpaired Student's t test on Prism software. P ≤ 0.05 was considered significant. Error bars denote SDs.

Fig. S1 A shows CD69 and CD25 expression on gated C7 Th1 cells harvested from the lungs of naive and infected mice. Fig. S1 B shows numbers of recoverable C7 Th1 cells from infected mice. Fig. S1 C depicts a real-time PCR analysis of ifnγ and nos2 transcripts from the lungs of infected mice.

We are grateful to Natalya Serbina, Amaraliz Rivera, and Tobias Hohl for helpful discussions and/or comments on the manuscript, and Feng Gao and Ingrid Leiner for technical support.

This research was supported by a National Institutes of Health (NIH) F32 grant (AI074248 to A.M. Gallegos) and NIH grants (AI53417 to M. Glickman and AI0677359 to E. Pamer). We received pMRLB7 as part of a National Institute of Allergy and Infectious Diseases, NIH contract (HHSN266200400091C) entitled “Tuberculosis Vaccine Testing and Research Materials,” which was awarded to Colorado State University.

The authors declare that they have no competing financial interests.