CD8+ T cells use TRAIL to restrict West Nile virus pathogenesis by controlling infection in neurons

Abstract

Previous studies of mice have demonstrated that an orchestrated sequence of innate and adaptive immune responses is required to control West Nile virus (WNV) infection in peripheral and central nervous system (CNS) tissues. Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL; also known as CD253) has been reported to inhibit infection with dengue virus, a closely related flavivirus, in cell culture. To determine the physiological function of TRAIL in the context of flavivirus infection, we compared the pathogenesis of WNV in wild-type and TRAIL(-/-) mice. Mice lacking TRAIL showed increased vulnerability and death after subcutaneous WNV infection. Although no difference in viral burden was detected in peripheral tissues, greater viral infection was detected in the brain and spinal cord at late times after infection, and this was associated with delayed viral clearance in the few surviving TRAIL(-/-) mice. While priming of adaptive B and T cell responses and trafficking of immune and antigen-specific cells to the brain were undistinguishable from those in normal mice, in TRAIL(-/-) mice, CD8(+) T cells showed qualitative defects in the ability to clear WNV infection. Adoptive transfer of WNV-primed wild-type but not TRAIL(-/-) CD8(+) T cells to recipient CD8(-/-) mice efficiently limited infection in the brain and spinal cord, and analogous results were obtained when wild-type or TRAIL(-/-) CD8(+) T cells were added to WNV-infected primary cortical neuron cultures ex vivo. Collectively, our results suggest that TRAIL produced by CD8(+) T cells contributes to disease resolution by helping to clear WNV infection from neurons in the central nervous system.

Figures

Fig 1

Survival and viral titer analysis of wild-type and TRAIL−/− mice infected with WNV. (A) Kaplan-Meier survival curves. Wild-type (n = 40) and TRAIL−/− (n = 29) mice were infected via the subcutaneous route with 102 PFU of WNV and monitored for mortality for 28 days. Survival differences were judged by the log rank test and were statistically significant (P < 0.0001). (B to E) Viremia and WNV tissue burdens in wild-type and TRAIL−/− mice. Infectious WNV levels in sera (B), spleens (C), brains (D), and spinal cords (E) of wild-type and TRAIL−/− mice were measured by viral plaque assay of samples harvested at the indicated time points. Data are expressed as log10 PFU per gram of tissue or ml of serum and reflect results for 8 to 13 mice per time point between days 2 and 10. For viral burden experiments, the horizontal bars represent the mean titers, the dotted lines represent the limits of sensitivity of viral detection, and asterisks indicate statistically significant differences (P < 0.05) between wild-type and TRAIL−/− mice, determined by the Mann-Whitney test.

Fig 2

WNV burden in the CNS after intracranial infection. Wild-type and TRAIL−/− mice were infected with 101 PFU of WNV via the intracranial route. Different regions of the brain were harvested at the indicated time points. (A) Cerebellum; (B) brain stem; (C) cerebral cortex; (D) subcortex; (E) spinal cord. Tissue homogenates were analyzed for viral burden by plaque assay. Data are shown as PFU per gram of tissue for 5 to 8 mice per time point. The dotted line represents the limit of detection. None of the differences achieved statistical significance.

Fig 3

WNV-specific antibody responses in wild-type and TRAIL−/− mice. Wild-type and TRAIL−/− mice were infected with WNV via the subcutaneous route, and serum was collected at the indicated time points. The development of WNV-specific IgM (A) or IgG (B) was determined by ELISA using purified WNV E protein. Data are averages for 5 mice at day 2 and 10 mice each at days 4 through 8. (C) Neutralizing antibody response. Neutralizing titers were determined by a PRNT assay. All samples were serially diluted in duplicate, and data are expressed as the reciprocal PRNT50, the antibody titer that reduced the plaque number by 50%. Data are averages for 10 mice per time point. ns, none of the differences achieved statistical significance.

Fig 4

T cell responses in the spleens of wild-type and TRAIL−/− mice after WNV infection. (A to D) CD8+ T cells. WNV-infected splenocytes from wild-type or TRAIL−/− mice were harvested at day 7 and stimulated ex vivo for 6 h (A and B) with an immunodominant Db-restricted NS4B peptide. Cells were costained for CD8 and intracellular IFN-γ or TNF-α and processed by flow cytometry. Data are presented as percentages (A) or total numbers (B) of positive cells. The percentages (C) and total numbers (D) of antigen-specific CD8+ T cells in uninfected or WNV-infected wild-type and TRAIL−/− mice were determined by binding to a Db-restricted NS4B tetramer. Asterisks indicate statistically significant (P < 0.05) differences between wild-type and TRAIL−/− mice. (E and F) CD4+ T cells. WNV-infected splenocytes from wild-type or TRAIL−/− mice were harvested at day 7 and stimulated ex vivo for 6 h with a stimulatory anti-CD3 antibody. Cells were costained for CD4 and intracellular IFN-γ or TNF-α and processed by flow cytometry. Data are presented as percentages (E) or total numbers (F) of positive cells. Data in panels A to F are representative for 6 to 10 mice from three independent experiments. In panels A, C, and E, the percentages of IFN-γ-, TNF-α-, and NS4B tetramer-positive cells represent the fractions of total gated CD8+ or CD4+ T cells. (G) Regulatory T cells. Splenocytes from WNV-infected wild-type (naïve or infected) and TRAIL−/− mice were harvested at day 7 and stained with antibodies to CD4, CD25, and FoxP3. Data are expressed as percentages of CD4+ T cells that stained positive for CD25 and FoxP3. Data were pooled for 6 mice from 2 independent experiments. (H) Flow cytometry profiles. Representative flow cytometry profiles of intracellular IFN-γ staining (upper panels) and NS4B tetramer staining (bottom panels) are shown for splenic CD8+ T cells from wild-type or TRAIL−/− mice at day 7 after WNV infection.

Fig 5

Accumulation of leukocytes in the CNS of WNV-infected wild-type and TRAIL−/− mice. Wild-type and TRAIL−/− mice were infected with 102 PFU of WNV by the subcutaneous route. Eight days later, brains were harvested and leukocytes were isolated by Percoll gradient centrifugation. (A and B) Percentages (A) and total numbers (B) of microglia (CD11b+ CD45lo) and macrophages (CD11b+ CD45hi). (C and D) Percentages (C) and total numbers (D) of CD8+ T cells, CD4+ T cells, and CD19+ B220+ B cells. (E and F) Number of WNV-specific CD8+ T cells in the brain as judged by NS4B peptide restimulation and intracellular IFN-γ staining (E) or NS4B tetramer staining (F). (G) Number of activated CD4+ T cells in the brain as judged after incubation with a stimulatory anti-CD3 antibody and intracellular IFN-γ staining. Data represent the averages for two independent experiments with 6 mice per group. Asterisks indicate statistically significant (P < 0.05) differences between wild-type and TRAIL−/− mice. (H) Representative flow cytometry profiles of CD3+ CD8+ T cells in the brains of TRAIL−/− mice (left) 8 days after WNV infection. Surface NS4B tetramer staining (middle) and intracellular IFN-γ staining after NS4B peptide restimulation ex vivo (right) are shown for brain CD8+ T cells.

Fig 6

Role of TRAIL in CD8+ T cell-mediated control of WNV infection in the brain. (A and B) Viral yields in different tissues after adoptive transfer of WNV-primed CD8+ T cells. Donor CD8+ T cells (10 × 106 [A] or 3 × 106 [B]) were purified from naïve or WNV-primed wild-type and TRAIL−/− mice and transferred into recipient CD8−/− mice 1 day after infection. The spleens, brains, and spinal cords of recipient CD8−/− mice were harvested on day 10 after initial infection, and WNV was titrated by plaque assay. Data represent results from 3 or 4 mice in two independent experiments. The dotted lines represent the limits of sensitivity of the assay. (C and D) WNV persistence in CNS tissues. Wild-type and TRAIL−/− mice were infected with WNV, brains (C) and spinal cords (D) were harvested on days 15, 21, and 28 after infection, and viral yields were titrated by plaque assay. Results are for 4 to 9 mice per time point. (E and F) Control of WNV infection in primary neurons ex vivo after addition of primed CD8+ T cells. Cortical neurons were infected with WNV at an MOI of 0.001. After 1 h, purified naïve or WNV-primed CD8+ T cells from wild-type or TRAIL−/− mice were added at an E:T ratio of 50:1. After 24 h (E) or 48 h (F), supernatants were harvested and WNV production was measured by plaque assay. Asterisks indicate statistically significant differences (*, P < 0.05; **, P < 0.005; and ***, P < 0.0001). ns, no significant difference.

Fig 7

CD8+ T cells express TRAIL. Splenocytes and brain leukocytes from WNV-infected wild-type mice were harvested at day 7 and stained for surface expression of TRAIL on CD4+ and CD8+ T cells. Data were analyzed by flow cytometry and are expressed as mean fluorescence intensities for the positive cells only. (A) TRAIL expression on total CD4+ and CD8+ T cells in the spleen. (B) TRAIL expression on NS4B tetramer-positive and tetramer-negative CD8+ T cells in the spleen. (C) TRAIL expression on total CD4+ and CD8+ T cells in the brain. (D) TRAIL expression on NS4B tetramer-positive and tetramer-negative CD8+ T cells in the brain. Data were pooled from four mice, and asterisks indicate differences that are statistically significant. (E) Representative flow cytometry profiles after staining with anti-TRAIL antibody on the surfaces of CD4+ and CD8+ T cells from the brains of mice infected with WNV. (F) DR5 expression on neurons. Neurons were left uninfected or infected with WNV (MOI of 0.001) for 24 h. Cells were harvested and stained for surface DR5 and intracellular WNV antigen, counterstained with ToPro3, and analyzed by confocal microscopy. Representative data from multiple images are shown.

Fig 8

Mechanism by which TRAIL inhibits WNV infection in neurons. (A) Recombinant TRAIL does not have a direct antiviral effect on cortical neurons. Primary cortical neurons were treated with increasing doses (0.078 to 20 μg/ml) of recombinant TRAIL (rTRAIL) for 24 h prior to infection with WNV (MOI of 0.001). One hour later, free virus and TRAIL were removed by washing. Supernatants were harvested 24 and 48 h after infection and titrated by plaque assay. Results are representative of three independent experiments performed in triplicate. (B) Postexposure addition of recombinant TRAIL does not confer an antiviral effect against WNV infection of cortical neurons. Neurons were infected with WNV (MOI of 0.001). One hour later, free virus was removed by extensive washing. Subsequently, increasing doses (0.078 to 20 μg/ml) of recombinant TRAIL were added, and supernatants were harvested 24 h later and titrated by plaque assay. Results are representative of three independent experiments performed in triplicate. None of the doses of recombinant TRAIL conferred a statistically significant antiviral effect. (C) Addition of recombinant TRAIL enhances the antiviral effect of primed TRAIL−/− CD8+ T cells. Cortical neurons were infected with WNV at an MOI of 0.001. After 1 h, purified WNV-primed CD8+ T cells from TRAIL−/− mice were added at an E:T ratio of 50:1. At the same time, recombinant TRAIL (10 μg/ml) was added to some of the cell cultures. Supernatants were harvested 48 h later, and WNV production was measured by plaque assay. (D) Addition of neutralizing anti-TRAIL antibody decreases the antiviral effect of primed wild-type CD8+ T cells. Cortical neurons were infected with WNV at an MOI of 0.001. After 1 h, purified WNV-primed CD8+ T cells from wild-type mice were added at an E:T ratio of 50:1. At the same time, a neutralizing anti-TRAIL or isotype control antibody (10 μg/ml) was added to cells. Supernatants were harvested 48 h later, and WNV production was measured by plaque assay. Data in panels C and D were pooled from six replicate samples generated from independent experiments.

Fig 9

FasL and granzyme B expression on CD8+ T cells from the brains of WNV-infected wild-type and TRAIL−/− mice. Brain leukocytes from WNV-infected wild-type mice were harvested at day 8 and stained for surface expression of FasL and intracellular expression of granzyme B. Data were analyzed by flow cytometry, and all results are pooled data from two independent experiments. (A) FasL expression on total CD4+ and CD8+ T cells and on NS4B tetramer-positive CD8+ T cells in the brain. Note that 100% of brain T cells expressed FasL on the surface. Data are expressed as normalized mean fluorescence intensities (MFI) of FasL-positive cells. (B and C) Granzyme B expression on total CD4+ and CD8+ T cells and on NS4B tetramer-positive CD8+ T cells in the brain. Data are expressed as percentages of granzyme B-positive cells (B) and normalized mean fluorescence intensities of the granzyme B-positive cells (C). Data were pooled from nine mice, and none of the differences were statistically significant.