Suboptimal activation of antigen-specific CD4+ effector cells enables persistence of M. tuberculosis in vivo

Abstract

Adaptive immunity to Mycobacterium tuberculosis controls progressive bacterial growth and disease but does not eradicate infection. Among CD4+ T cells in the lungs of M. tuberculosis-infected mice, we observed that few produced IFN-γ without ex vivo restimulation. Therefore, we hypothesized that one mechanism whereby M. tuberculosis avoids elimination is by limiting activation of CD4+ effector T cells at the site of infection in the lungs. To test this hypothesis, we adoptively transferred Th1-polarized CD4+ effector T cells specific for M. tuberculosis Ag85B peptide 25 (P25TCRTh1 cells), which trafficked to the lungs of infected mice and exhibited antigen-dependent IFN-γ production. During the early phase of infection, ∼10% of P25TCRTh1 cells produced IFN-γ in vivo; this declined to <1% as infection progressed to chronic phase. Bacterial downregulation of fbpB (encoding Ag85B) contributed to the decrease in effector T cell activation in the lungs, as a strain of M. tuberculosis engineered to express fbpB in the chronic phase stimulated P25TCRTh1 effector cells at higher frequencies in vivo, and this resulted in CD4+ T cell-dependent reduction of lung bacterial burdens and prolonged survival of mice. Administration of synthetic peptide 25 alone also increased activation of endogenous antigen-specific effector cells and reduced the bacterial burden in the lungs without apparent host toxicity. These results indicate that CD4+ effector T cells are activated at suboptimal frequencies in tuberculosis, and that increasing effector T cell activation in the lungs by providing one or more epitope peptides may be a successful strategy for TB therapy.

Figure 1. Low frequency of IFN-γ-producing endogenous CD4⁺ T cells in lungs of M. tuberculosis-infected mice.

A. Frequency of IFN-γ expression by endogenous, polyclonal CD4⁺ T cells in the lungs of M. tuberculosis-infected mice throughout infection, assayed by intracellular cytokine staining without ex vivo restimulation. Flow cytometry dot plots show lung CD4⁺ cells from a representative mouse at the indicated time point post-infection. Values indicate the proportion of cells expressing IFN-γ in the CD4⁺ population for each mouse. B. Mean frequency of IFN-γ⁺ cells among lung CD4⁺ T cells for each group of 4 mice at each time point post-infection, assayed by intracellular cytokine staining without ex vivo restimulation. Asterisks indicate statistical significance of differences in frequency of T cell activation observed between adjacent time points * p<0.05; ** p<0.005.

Figure 2. P25TCRTh1 cells produce IFN-γ in response to M. tuberculosis Ag85B peptide 25.

A. P25TCRTh1 cells were restimulated in vitro with C57BL/6 splenocytes in the presence or absence of peptide 25 and analyzed by flow cytometry for intracellular IFN-γ. B. CFP-P25TCRTh1 cells traffic to the lung parenchyma of M. tuberculosis-infected mice. Th1 effector cells were transferred on day 25, and lungs were harvested on day 28 postinfection. CFP-P25TCRTh1 cells (light blue-green) are found in interstitial regions with a high density of DAPI-stained nuclei, typical of the aggregates of macrophages, dendritic cells, and lymphocytes observed at this stage of infection. Scale bar: 50 µm. C. On day 18 post-infection, mice infected with either H37Rv (w.t.) or ΔAg85B M. tuberculosis received P25TCRTh1 cells by adoptive transfer. Lung cells were harvested 72 hours later (day 21). Transferred P25TCRTh1 (CD45.2⁺) cells were analyzed by flow cytometry for intracellular IFN-γ. One group of mice received intravenous treatment with Ag85B peptide 25 6 hours prior to lung cell harvest. Flow cytometry dot plots from in vivo experiments show a representative of four mice per group. D. Day 21 post-infection with either H37Rv or ΔAg85B: mean percentage from four individual mice of P25TCRTh1 or endogenous CD4⁺ T cells expressing IFN-γ with or without in vivo administration of Ag85B peptide 25.

Figure 3. Peptide 25-specific T cell activation and fbpB expression decrease during chronic infection.

A. Frequency of IFN-γ production by adoptively transferred P25TCRTh1 (CD45.2⁺) CD4⁺ cells in the lungs of M. tuberculosis-infected mice throughout infection, assayed by intracellular cytokine staining without ex vivo restimulation. Flow cytometry dot plots show lung P25TCRTh1 cells which were adoptively transferred 3 days prior to the indicated time point post-infection. Values indicate the proportion of cells expressing IFN-γ among the CD45.2⁺, CD4⁺ population for each mouse. B. CFSE proliferation profile of naïve P25TCR-tg CD4⁺ T cells transferred into M. tuberculosis-infected wild type recipients on days 11, 17 or 35 post-infection. Mediastinal lymph node cells were isolated 7 days after adoptive transfer (days 18, 24, or 48 post-infection) and analyzed by flow cytometry for CFSE dilution to measure proliferation. Histograms are representative of four individual mice per time point. C. The mean percentage of P25TCRTh1 cells from four individual mice expressing IFN-γ at each time point post-infection is compared with the expression of M. tuberculosis fbpB mRNA as infection progresses to chronic phase. Copy number of fbpB mRNA for four individual mice at each time point was determined by RT-qPCR and is normalized to constitutively expressed 16S rRNA.

Figure 4. Forced expression of fbpB enhances T cell activation and impairs bacterial persistence during chronic infection.

A. Expression of fbpB mRNA, normalized to 16S rRNA by H37Rv and CPE85B throughout in vivo infection, determined by RT-qPCR of bacteria in lungs. Data points indicate the mean (±SEM) of 4 mice per time point. B. Bacterial population size of H37Rv or CPE85B in vitro culture before and after stationary liquid incubation to induce expression of hspXp:fbpB. Columns represent the mean (±SEM) population size of three cultures for each strain. Western blot shows Ag85B protein secreted into culture supernatants during stationary culture. C. Activation of P25TCRTh1 cells in vitro by bone marrow derived DCs infected with H37Rv or CPE85B, measured by IFN-γ ELISA. Columns represent the mean (±SEM) of 3 wells at the indicated infected DC∶T cell ratio. D, E. Activation of P25TCRTh1 cells during mouse infection with H37Rv or CPE85B. P25TCRTh1 cells were transferred into infected mice on either day 18 or 39 post-infection. 3 days after adoptive transfer (day 21 or 42), lung cells were analyzed by flow cytometry for intracellular IFN-γ without ex vivo restimulation. (D) Data points indicate the frequency of cells expressing IFN-γ among CD45.2⁺, CD4⁺ lung cells at each time point. (E) Flow cytometry plots show lung P25TCRTh1 cells from a representative mouse at the indicated time point post-infection. Values indicate the proportion of IFN-γ⁺ cells among CD45.2⁺, CD4⁺ population. F. Bacterial population size throughout mouse infection with H37Rv or CPE85B. Data points indicate the mean (±SEM) of 4 mice per time point.

Figure 5. Forced expression of fbpB impairs M. tuberculosis in a CD4⁺ T cell dependent manner.

A. Survival of C57BL/6 and CD4⁺ T cell-deficient MHCII KO mice after aerosol infection with H37Rv or CPE85B. N≥5 mice for each group. B. Bacterial population size in lungs of mice infected with H37Rv or CPE85B after CD4⁺ T cell depletion with monoclonal anti-CD4 antibody GK1.5. Antibody treatment was started on day 28 post-infection and continued every 6 days until day 50. Data points indicate the mean (±SEM) bacterial burden in 4 mice in each infection group at each time point. Asterisks indicate statistical significance between groups of mice at neighboring time points within one infection group; * p<0.05; ** p<0.005.

Figure 6. Treatment with peptide 25 transiently enhances CD4⁺ T cell responses and reduces bacterial burden.

(A) Frequency of adoptively transferred P25TCRTh1 (top row) or endogenous (bottom row) CD4⁺ T cells producing IFN-γ at various time points after intravenous treatment with synthetic Ag85B peptide 25. Flow cytometry dot plots show lung CD4⁺ cells from a representative mouse at the indicated time point after treatment with peptide 25. Values indicate the proportion of IFN-γ⁺ cells among the CD45.2⁺ or CD45.2⁻, CD4⁺ population for each mouse. Data shown are representative of n≥4 mice per group. (B) Bacterial burden in the lungs of wild type mice treated from day 28 to day 45 post-infection with intravenous Ag85B peptide 25 or OVA peptide control. Data points indicate the final bacterial population size for individual mice in each group after treatment with either peptide. Data shown are representative of n≥4 mice per group.

Figure 7. Schematic diagram of CD4⁺ effector T cell activation at the site of M. tuberculosis infection.

A. During the chronic stage of infection, Ag85B-specific CD4⁺ effector cells are activated at low frequencies, at least part due to low bacterial expression of the antigen gene; bacteria are able to persist due to the low frequency of effector cell activation. B. Administration of epitope peptide occupies previously-empty MHC class II and/or displaces previously-bound peptides and provides antigen for recognition by pre-existing epitope-specific CD4⁺ effector cells, resulting in their activation and consequent reduction of the lung bacterial burden.