A temporal role of type I interferon signaling in CD8+ T cell maturation during acute West Nile virus infection

Abstract

A genetic absence of the common IFN-α/β signaling receptor (IFNAR) in mice is associated with enhanced viral replication and altered adaptive immune responses. However, analysis of IFNAR(-/-) mice is limited for studying the functions of type I IFN at discrete stages of viral infection. To define the temporal functions of type I IFN signaling in the context of infection by West Nile virus (WNV), we treated mice with MAR1-5A3, a neutralizing, non cell-depleting anti-IFNAR antibody. Inhibition of type I IFN signaling at or before day 2 after infection was associated with markedly enhanced viral burden, whereas treatment at day 4 had substantially less effect on WNV dissemination. While antibody treatment prior to infection resulted in massive expansion of virus-specific CD8(+) T cells, blockade of type I IFN signaling starting at day 4 induced dysfunctional CD8(+) T cells with depressed cytokine responses and expression of phenotypic markers suggesting exhaustion. Thus, only the later maturation phase of anti-WNV CD8(+) T cell development requires type I IFN signaling. WNV infection experiments in BATF3(-/-) mice, which lack CD8-α dendritic cells and have impaired priming due to inefficient antigen cross-presentation, revealed a similar effect of blocking IFN signaling on CD8(+) T cell maturation. Collectively, our results suggest that cell non-autonomous type I IFN signaling shapes maturation of antiviral CD8(+) T cell response at a stage distinct from the initial priming event.

Figure 1. Effect of blockade of type I IFN signaling on WNV-NY infection.

A. Dose titration of the MAR1-5A3 MAb in mice. Mice (n = 10 per group) were treated with increasing doses of MAR1-5A3 MAb, infected one day later with 10² PFU of WNV-NY, and survival was monitored. All MAR1-5A3 treatment doses shown were statistically different compared to GIR-208 treatment (P<0.05). B. Time course of MAR1-5A3 addition after WNV infection. Mice (n = 5 to 10 per group) were infected with 10² PFU of WNV-NY, treated with 1 mg of MAR1-5A3 or an isotype control (GIR-208 (GIR)) at different times after infection, and survival was determined. Treatment with MAR1-5A3 at days 0, 1, 2, 3, and 4 were statistically different (P<0.004) compared to treatment with GIR-208. C–G. Effect of MAR1-5A3 on viral burden. Mice (n = 5 to 10 per group) were infected with 10² PFU of WNV-NY, treated with 1 mg of MAR1-5A3 or GIR at day 2 or day 4 after infection. (C) Serum, (D) spleen, (E) kidney, (F) brain, and (G) spinal cord were harvested at day six and viral titers were determined by plaque assay or qRT-PCR. Asterisks indicate differences that are statistically significant (*, P<0.05; **, P<0.01, ***, P<0.001).

Figure 2. Effect of treatment of MAR1-5A3 on WNV-specific CD8⁺ T cell responses.

A-B. Mice were infected with 10² PFU of WNV-NY and treated with 1 mg of MAR1-5A3 or GIR-208 (GIR) at day 4 post infection. A. Analysis of the CD8⁺ T cell from the spleen of infected MAb-treated mice (n = 20 to 25 per group). Splenocytes were harvested on day 9 after WNV infection, and intracellular cytokine staining of IFN-γ and TNF-α was analyzed in CD8⁺ T cells after ex vivo restimulation with NS4B peptide. (Top left) Total number of splenic CD8⁺ T cells after infection and treatment with MAbs. (Top right) A representative contour plot showing intracellular IFN-γ levels on CD8⁺ T cells after MAb treatment is shown. The percentage, number, and relative staining of WNV-specific IFN-γ⁺ CD8⁺ T cells (middle panels) or WNV-specific TNF-α⁺ CD8⁺ T cells (bottom panels) are shown. Relative intracellular cytokine staining reflects pooling of data from independent experiments after normalization within a given experiment. B. The levels of intracellular granzyme B in WNV-specific CD8⁺ T cells were assessed by co-staining with D^b-NS4B tetramer and antibodies to granzyme B (n = 6 mice per group). A representative contour plot showing intracellular granzyme B levels on CD8⁺ T cells after MAb treatment is shown. C–D. Mice were infected with 10² PFU of WNV-MAD and treated with 1 mg of MAR1-5A3 or GIR-208 (GIR) at day -1 and +4 relative to infection. The percentage, number, and relative staining of WNV-specific (C) IFN-γ⁺ CD8⁺ T cells or (D) TNF-α⁺ CD8⁺ T cells are shown (n = 5 mice per group). Asterisks indicate differences that are statistically significant (*, P<0.05; **, P<0.01, ***, P<0.001).

Figure 3. Effect of MAR1-5A3 on VV-specific CD8⁺ T cell responses in the spleen.

Mice were infected with 10⁴ PFU of VV (Western reserve strain), treated at day 4 with 1 mg of MAR1-5A3 or GIR-208 MAb, and splenocytes were harvested on at day 9 for intracellular cytokine staining of IFN-γ and TNF-α of CD8⁺ T cells after peptide restimulation with immunodominant A47L (A) or B8R (B) peptides (n = 9 mice per group). Relative intracellular cytokine staining reflects pooling of data from independent experiments after normalization within a given experiment. Asterisks indicate differences that are statistically significant (*, P<0.05; **, P<0.01, ***, P<0.001).

Figure 4. Effect of deletion of BATF3 and loss of CD8-α dendritic cells on CD8⁺ T cell responses after WNV infection.

A. Wild type or BATF3 ^-/- mice were infected with 10² PFU of WNV-NY, spleens were harvested on day 9 after infection, and intracellular IFN-γ responses in CD8⁺ T cells were measured by flow cytometry after ex vivo stimulation with NS4B peptide (n = 5 mice per group). B. Wild type or BATF3 ^-/- mice were treated with MAR1-5A3 or GIR-208 (1 mg per dose) at day 4 after infection with 10² PFU of WNV-NY. Spleens were harvested on day 9 after infection, and intracellular IFN-γ (top panels) and TNF-α (bottom panels) responses in CD8⁺ T cells were measured by flow cytometry after ex vivo stimulation with NS4B peptide (n = 5 mice per group). Relative intracellular cytokine staining reflects pooling of data from independent experiments after normalization within a given experiment. Asterisks indicate differences that are statistically significant (*, P<0.05; **, P<0.01, ***, P<0.001).

Figure 5. Effect of MAR1-5A3 treatment on costimulatory molecule expression of antigen-presenting cells.

Mice were infected with 10² PFU of WNV-NY and treated with 1 mg of MAR1-5A3 or GIR-208 at days 2 or 4 post infection. At days 6 or 9 after infection, CD11b⁺ and CD11c⁺ splenocytes were analyzed for expression of CD80 and CD86 by flow cytometry. A. Gating strategy and representative histograms are shown from animals treated with MAR1-5A3 or GIR-208 at day 4 and harvested at day 9. B-C. Summary of data showing the percentage of CD11c⁺ and CD11b⁺ splenocytes and the mean fluorescence intensity of CD80 and CD86 staining (n = 7 to 9 mice per group) from animals (B) treated at days 2 or 4 and harvested at day 6 or (C) treated at day 4 and harvested at day 9. Asterisks indicate differences that are statistically significant (*, P<0.05; **, P<0.01, ***, P<0.001).

Figure 6. Effect of MAR1-5A3 treatment on serum inflammatory cytokines.

Mice were infected with 10² PFU of WNV and treated with 1 mg of MAR1-5A3 or GIR-208 at day 4 post infection. Serum was harvested at day 6 or 9 after infection and analyzed for the (A) IFN-γ, (B) TNF-α, (C) IL-10, (D) IL-12 p40, (E) IL-17, and (F) IL-18 using a Bio-Plex pro cytokine assay (n = 6 mice per group). Asterisks indicate differences that are statistically significant (*, P<0.05; **, P<0.01, ***, P<0.001).

Figure 7. Effect of MAR1-5A3 treatment on expression of markers of CD8⁺ T cell activation and exhaustion.

A. Wild type mice were infected with 10² PFU of WNV-NY and treated with 1 mg of MAR1-5A3 or GIR-208 at day 4 post infection. Splenocytes were harvested on day 9, co-stained for CD8β and D^b-NS4B tetramer, and the gated cells analyzed by flow cytometry for relative expression of PD-1, CTLA-4, CD44 CD127, CD11a and CD43 (n = 18 mice per group). B. BATF3 ^-/- mice were treated with MAR1-5A3 or GIR-208 at day 4 after infection with 10² PFU of WNV-NY. Spleens were harvested on day 9 after infection, co-stained for CD8β and D^b-NS4B tetramer, and the gated cells analyzed by flow cytometry for relative expression of PD-1 and CTLA-4 (n = 5 mice per group). Relative staining reflects pooling of data from independent experiments after normalization within a given experiment. Asterisks indicate differences that are statistically significant (*, P<0.05; **, P<0.01, ***, P<0.001).