Critical role for perforin-, Fas/FasL-, and TNFR1-mediated cytotoxic pathways in down-regulation of antigen-specific T cells during persistent viral infection

Abstract

Viral persistence following infection with invasive strains of lymphocytic choriomeningitis virus (LCMV) can be achieved by selective down-regulation of virus-specific T lymphocytes. High viral burden in the onset of infection drives responding cells into functional unresponsiveness (anergy) that can be followed by their physical elimination. In this report, we studied down-regulation of the virus-specific CD8(+)-T-cell response during persistent infection of adult mice with LCMV, with emphasis on the role of perforin-, Fas/FasL-, or tumor necrosis factor receptor 1 (TNFR1)-mediated cytolysis in regulating T-cell homeostasis. The results reveal that the absence of perforin, Fas-ligand, or TNFR1 has no significant effect on the kinetics of proliferation and functional inactivation of virus-specific CD8(+) T cells in the onset of chronic LCMV infection. However, these molecules play a critical role in the homeostatic regulation of T cells, influencing the longevity of the virus-specific CD8(+)-T-cell population once it has become anergic. Thus, CD8(+) T cells specific to the dominant LCMV NP(396-404) epitope persist in an anergic state for at least 70 days in perforin-, FasL-, or TNFR1-deficient mice, but they were eliminated by day 30 in C57BL/6 controls. These effects were additive as shown by a deficit of apoptotic death of NP(396-404) peptide-specific CD8(+) T cells in mice lacking both perforin and TNFR1. This suggests a role for perforin-, FasL-, and TNFR1-mediated pathways in down-regulation of the antiviral T cell response during persistent viral infection by determining the fate of antigen-specific T cells. Moreover, virus-specific anergic CD8(+) T cells in persistently infected C57BL/6 mice contain higher levels of Bcl-2 and Bcl-XL than functionally intact T cells generated during acute LCMV infection. In the case of proapoptotic factors, Bax expression did not differ between T-cell populations and Bad was below the limit of detection in all samples. As expression of the Bcl-2 family members controls susceptibility to apoptosis, this finding may provide a molecular basis for the survival of anergic cells under conditions of prolonged antigen stimulation.

FIG. 1.

Kinetics of virus replication and virus-specific CD8⁺-T-cell response in spleens of perforin- or FasL (gld)-deficient mutant mice compared to those of B6 congenic control mice following infection with a relatively high dose of LCMV Docile. Analyses were performed to correlate the kinetics of virus replication (A) with the kinetics of virus-specific CD8⁺-T-cell response (B and C). Mice were infected with 2 × 10⁶ PFU of LCMV Docile, and virus titers in the spleens were measured at the time points indicated. The data shown are means ± standard errors of the means for 3 to 5 mice (in log₁₀ PFU/gram of tissue). In addition, total numbers of GP1_33–41 or NP_396–404 peptide-specific CD8⁺ T cells were measured by staining for H-2D^b tetramer (filled circles) or intracellular IFN-γ (open circles) production following stimulation of cells with the appropriate peptide to determine the functional responsiveness of these cells. The data shown are means ± standard errors of the means for 3 to 5 mice (in log₁₀ virus-specific T cells per spleen). In the case of perforin-deficient mice, some variation in the number of NP_396–404 tetramer-positive cells was observed at late time points. The symbols (▴,•) used in the right column of panel C represent values obtained from the number of mice indicated.

FIG. 2.

Kinetics of virus replication and virus-specific CD8⁺-T-cell response in spleens of perforin- or FasL (gld)-deficient mutant mice compared to those of B6 congenic control mice following infection with a low dose of LCMV Docile. Mice were infected with 10² PFU of LCMV Docile, and virus titers in the spleens were measured at the time points indicated (A). The data shown are means ± standard errors of the means for 3 to 5 mice (in log₁₀ PFU/gram of tissue). In addition, total numbers of GP1_33–41 or NP_396–404 peptide-specific CD8⁺ T cells were measured by staining for H-2D^b tetramer (filled circles) or intracellular IFN-γ (open circles) production following stimulation of cells with the appropriate peptide to determine the functional responsiveness of these cells (B and C). The data shown are means ± standard errors of the means for 3 to 5 mice (in log₁₀ virus-specific T cells per spleen). In the case of perforin-deficient mice, some variation in the number of NP_396–404 tetramer-positive cells was observed at late time points. The symbols (▴,•) used in the right column of panel C represent values obtained from the number of mice indicated.

FIG. 3.

Kinetics of virus replication and virus-specific CD8⁺-T-cell response in spleens of perforin- or FasL (gld)-deficient mutant mice compared to those of B6 congenic control mice following infection with LCMV Aggressive. Mice were infected with 10² PFU of LCMV Aggressive, and virus titers in the spleens were measured at the time points indicated (A). The data shown are means ± standard errors of the means for 3 to 5 mice (in log₁₀ PFU/gram of tissue). In addition, total numbers of GP1_33–41, GP2_276–286, or NP_396–404 peptide-specific CD8⁺ T cells were measured by staining for H-2D^b tetramer (filled circles) or intracellular IFN-γ (open circles) production following stimulation of cells with viral epitope peptide to determine the functional responsiveness of these cells (B, C, and D). The data shown are means ± standard errors of the means for 3 to 5 mice (in log₁₀ virus-specific T cells per spleen).

FIG. 4.

Virus-specific cytotoxic activities in spleens of control B6 mice in comparison to those of FasL (gld) mice. B6 control mice (A and B) or FasL (gld) mutant mice (C and D) were infected with 2 × 10⁶ or 10² PFU of LCMV Docile as indicated. Splenocytes isolated on days 3, 6, 9, 15, 30, and 50 postinfection were stimulated in vitro with GP1_33–41 or NP_396–404 peptide-pulsed splenocytes as described in Materials and Methods. The cytolytic activity of restimulated splenocytes cultured at a density of 4 × 10⁶ cells/well was measured in a ⁵¹Cr release assay as described in Materials and Methods. Virus-specific cytotoxic activity was expressed in lytic units (LU) per spleen. The data shown are means ± standard errors of the means for 3 to 5 mice (in log₁₀ virus-specific T cells per spleen).

FIG. 5.

Kinetics of virus replication and virus-specific CD8⁺-T-cell response in spleens of mice deficient in TNFR1, perforin and TNFR1, or perforin and IFN-γ following infection with a relatively high dose of LCMV Docile. Mice were infected with 10⁶ PFU of LCMV Docile, and virus titers in the spleens were measured at the time points indicated (A). The data shown are means ± standard errors of the means for 3 to 5 mice (in log₁₀ PFU/gram of tissue). In addition, total numbers of GP1_33–41 or NP_396–404 peptide-specific CD8⁺ T cells were measured by staining for H-2D^b tetramer (filled circles) or intracellular IFN-γ (open circles) production following stimulation of cells with the appropriate peptide to determine the functional responsiveness of these cells (B and C). The data shown are means ± standard errors of the means for 3 to 5 mice (in log₁₀ virus-specific T cells per spleen).

FIG. 6.

Kinetics of virus replication and virus-specific CD8⁺-T-cell response in spleens of mice deficient in TNFR1, perforin and TNFR1, or perforin and IFN-γ following infection with a low dose of LCMV Docile. Mice were infected with 10² PFU of LCMV Docile, and virus titers in the spleens were measured at the time points indicated (A). The data shown are means ± standard errors of the means for 3 to 5 mice (in log₁₀ PFU/gram of tissue). In addition, total numbers of GP1_33–41 or NP_396–404 peptide-specific CD8⁺ T cells were measured by staining for H-2D^b tetramer (filled circles) or intracellular IFN-γ (open circles) production following stimulation of cells with the appropriate peptide to determine the functional responsiveness of these cells (B and C). The data shown are means ± standard errors of the means for 3 to 5 mice (in log₁₀ virus-specific T cells per spleen).

FIG. 7.

Expression of Bcl-2 family members in virus-specific CD8⁺ T cells during chronic LCMV infection of B6 mice. (A) Spleen cells from naive mice (▴) or mice infected with 2 × 10⁶ PFU (•) or 10² PFU (○) of LCMV Docile analyzed at the indicated time points after infection were triple stained with anti-CD8α; antibody specific to Bcl-2, Bcl-XL, or Bax; and either anti-CD44 (naive mice) or GP1_33–41 tetramer (infected mice). The mean fluorescence of Bcl-2 family expression on gated CD8⁺-CD44^low or GP1_33–41 tetramer-positive T cells was measured by FACS. Bcl-2, Bcl-XL, or Bax expression was calculated as the mean (± standard error of the mean) fluorescence increase in relation to isotype antibody control expression for 3 to 6 mice. Statistical analysis (Student’s t test) comparing GP1_33–41-specific T cells between mice infected with a high or low dose of LCMV Docile was conducted and revealed significant differences (P < 0.01) between Bcl-2 levels at day 9 and Bcl-XL levels at day 30. P values are indicated in the panels where appropriate. (B and C) Bcl-2 and Bcl-XL levels in GP1_33–41-specific T cells are indicated for mice infected with high (B) and low (C) doses of LCMV Docile on days 9 and 30. In these panels, the green line represents GP1_33–41 tetramer-positive T cells, the red line represents CD8⁺-CD44^low T cells in uninfected mice, and the broken blue line represents isotype antibody control staining. Mean fluorescence intensity values for GP1_33–41 tetramer-positive cells from infected mice are indicated in the upper right hand corner of each panel. The mean fluorescence intensities were 20 and 25 for CD8⁺-CD44^low cells and 5 and 16 for antibody isotype controls for Bcl-2 and Bcl-XL staining, respectively.

FIG. 7.

Expression of Bcl-2 family members in virus-specific CD8⁺ T cells during chronic LCMV infection of B6 mice. (A) Spleen cells from naive mice (▴) or mice infected with 2 × 10⁶ PFU (•) or 10² PFU (○) of LCMV Docile analyzed at the indicated time points after infection were triple stained with anti-CD8α; antibody specific to Bcl-2, Bcl-XL, or Bax; and either anti-CD44 (naive mice) or GP1_33–41 tetramer (infected mice). The mean fluorescence of Bcl-2 family expression on gated CD8⁺-CD44^low or GP1_33–41 tetramer-positive T cells was measured by FACS. Bcl-2, Bcl-XL, or Bax expression was calculated as the mean (± standard error of the mean) fluorescence increase in relation to isotype antibody control expression for 3 to 6 mice. Statistical analysis (Student’s t test) comparing GP1_33–41-specific T cells between mice infected with a high or low dose of LCMV Docile was conducted and revealed significant differences (P < 0.01) between Bcl-2 levels at day 9 and Bcl-XL levels at day 30. P values are indicated in the panels where appropriate. (B and C) Bcl-2 and Bcl-XL levels in GP1_33–41-specific T cells are indicated for mice infected with high (B) and low (C) doses of LCMV Docile on days 9 and 30. In these panels, the green line represents GP1_33–41 tetramer-positive T cells, the red line represents CD8⁺-CD44^low T cells in uninfected mice, and the broken blue line represents isotype antibody control staining. Mean fluorescence intensity values for GP1_33–41 tetramer-positive cells from infected mice are indicated in the upper right hand corner of each panel. The mean fluorescence intensities were 20 and 25 for CD8⁺-CD44^low cells and 5 and 16 for antibody isotype controls for Bcl-2 and Bcl-XL staining, respectively.