Bystander activated CD8+ T cells mediate neuropathology during viral infection via antigen-independent cytotoxicity

Abstract

Although many viral infections are linked to the development of neurological disorders, the mechanism governing virus-induced neuropathology remains poorly understood, particularly when the virus is not directly neuropathic. Using a mouse model of Zika virus (ZIKV) infection, we found that the severity of neurological disease did not correlate with brain ZIKV titers, but rather with infiltration of bystander activated NKG2D⁺CD8⁺ T cells. Antibody depletion of CD8 or blockade of NKG2D prevented ZIKV-associated paralysis, suggesting that CD8⁺ T cells induce neurological disease independent of TCR signaling. Furthermore, spleen and brain CD8⁺ T cells exhibited antigen-independent cytotoxicity that correlated with NKG2D expression. Finally, viral infection and inflammation in the brain was necessary but not sufficient to induce neurological damage. We demonstrate that CD8⁺ T cells mediate virus-induced neuropathology via antigen-independent, NKG2D-mediated cytotoxicity, which may serve as a therapeutic target for treatment of virus-induced neurological disease.

Fig. 1. ZIKV-infected Ifnar^-/- mice exhibit clinical symptoms of neuropathology associated with immune cell infiltration of the brain.

C57BL/6 and Ifnar^-/- mice were infected with 4 × 10⁵ PFU ZIKV via FTPD. a Serum was collected at 3 dpi and brains were isolated at 7 dpi to quantify ZIKV by plaque assay (n =5, repeated once with similar results). b Representative images of a ZIKV plaque assay of serially diluted serum samples. c Representative images of H&E staining of the cortex of brains isolated from C57BL/6 and Ifnar^-/- mice. Scale bar represents 100 μm. d Representative images of TUNEL staining of sagittal sections of cerebral cortex and cerebellum at ×10 and ×40 magnification. Scale bar represents 100 μm. e Quantification of % TUNEL⁺ nuclei in the cortex and f the # of TUNEL⁺ cells /mm² within the granular layer (n =5, 3). g Representative flow cytometry plots of CD45⁺ cells isolated from PBS-treated (naïve) and ZIKV-infected Ifnar^-/- mice at 7 dpi. See also Supplementary Fig. 1 for gating strategy. h Quantification of brain immune cell populations following PBS or ZIKV infection, represented as the percentage of CD45⁺ cells (n =4, 5). i Comparison of CD8⁺ T cells as a percentage of CD45⁺ cells and j absolute count of CD8⁺ T cells in the brains of PBS and ZIKV-infected mice (n =4, 5). k ZIKV-infected Ifnar^-/- mice were assessed at 7 dpi for clinical symptoms of paralysis (n =5). Data represent mean ± SEM (a, e, f, h, i, j). Statistical significance was determined by two-tailed t-test (a, e, f, i, j) or two-way ANOVA with Sidak’s multiple comparison test (h). See also Supplementary Fig. 1. Source data are provided as a Source Data file.

Fig. 2. Innate immune activation via IL-12/18 treatment of Ifnar^-/- mice reduces ZIKV infection but enhances ZIKV-induced neuropathology.

a Visual representation of differences between C57BL/6 and Ifnar^−/− mice and their susceptibility to ZIKV infection. b Ifnar^−/− mice were infected with PBS or 4x10⁵ ZIKV PFU via FTPD and IP injected with PBS or IL-12/18 at 1 and 2 dpi. Serum was collected at 3 dpi and assessed for IFN-γ production (n =3, 3, 4, 4, repeated three times with similar results) or c viremia by plaque assay (n =9). d At 7 dpi, mice were assessed for clinical symptoms of paralysis (n =15, 11, 22, 14). e Quantification of d (n =15, 11, 22, 14). f ZIKV-infected mice were administered PBS or IL-12/18 at 0-3 dpi and brains were collected at 7 dpi for detection of infectious ZIKV particles (n =5). Dashed lines represent limit of detection. g Representative images of TUNEL staining of the cerebral cortex or h cerebellum. Scale bar represents 100 μm. i Quantification of % TUNEL⁺ nuclei in the cortex or j # of TUNEL⁺ nuclei/mm² of the granular layer. Data represent mean ± SEM (b, c, e, f, i, j) of two (i) or three independent experiments (d, e). Statistical significance was determined by repeated measures (b) or ordinary (e) two-way ANOVA with Tukey’s multiple comparisons test and two-tailed Student’s t test (c, f, i, j). See also Supplementary Fig. 2. Source data are provided as a Source Data file.

Fig. 3. Increased CD8⁺ T cell response in the brain correlates with exacerbated neurological disease in ZIKV-infected mice.

Ifnar^-/- mice were infected with 4x10⁵ PFU via FTPD and administered PBS or IL-12/18 at 0-3 dpi (a, b) or 1 and 2 dpi (c–k). a Representative images of H&E staining of the cortex of brains from naïve and ZIKV-infected mice treated with PBS or IL-12/18. Scale bar represents 100 μm. b Quantification of a. Data represent average scores per brain of two blinded measurements on a scale of 0–3, with 0 being no lymphocyte infiltration and 3 being high levels of infiltration and perivascular cuffing. c Representative flow cytometry plots of CD45⁺ cells isolated from brains of naïve and ZIKV-infected mice at 7 dpi. d–k CD45⁺ cells were analyzed for proportion or total number of d, e lymphocytes, f, g NK1.1⁺, h, i CD8⁺, and j, k CD4⁺ (CD8^-) cells (n =3, 3, 4, 4). Data represents mean ± SEM (b, d–k). Statistical significance was determined by two-way ANOVA with Tukey’s multiple comparisons test (b, d, e, h–k). See also Supplementary Fig. 1. Source data are provided as a Source Data file.

Fig. 4. ZIKV infection induces a population of CD8⁺ T cells with a bystander activated phenotype.

Ifnar^−/− mice were infected with PBS (naïve) or 4x10⁵ PFU ZIKV via FTPD. At 7 dpi, cells from the blood and brain were isolated for flow cytometry analysis. a Representative flow plots of NKG2D and CD25 staining of CD8⁺ T cells in the blood and brain of naïve and ZIKV-infected mice. b Proportion of blood CD8⁺ T cells that are NKG2D⁺(n =3, 5). c MFI of NKG2D on blood CD8⁺ T cells (n =3, 5). d Representative flow cytometry plots of APC-conjugated tetramer and NKG2D expression on CD8⁺ T cells in the naïve spleen and ZIKV-infected brains. e Proportion of CD45⁺ cells in naïve or ZIKV-infected brains that are CD8⁺ T cells and their expression of NKG2D and tetramer staining (n =3). f Counts of CD8⁺ T cells subset on NKG2D and tetramer staining (n =3). Data represent mean ± SEM (b, c, e, f) and represent one of two independent experiments with similar results. Statistical significance was determined by two-tailed t-test (b, c) and two-way ANOVA with Tukey’s multiple comparison test (e, f). Tet tetramer, MFI mean fluorescence intensity. See also Supplementary Fig. 3. Source data are provided as a Source Data file.

Fig. 5. NKG2D blockade and CD8⁺ T cell depletion protect against ZIKV-induced neuropathology.

Ifnar^-/- mice were infected with 4x10⁵ PFU via FTPD and treated with PBS, α-NKG2D, or α-CD8 antibody. a At 7 dpi, mice were assessed for clinical symptoms of paralysis (n =16, 12, 10; data represent three independent experiments). b Quantification of a. c Representative TUNEL staining of cerebral cortex and cerebellum at 7 dpi. Scale bar represents 100 μm. d Quantification of % TUNEL⁺ nuclei in the cortex or e # of TUNEL⁺ nuclei/mm² of the granular layer of the cerebellum (n =9, 5, 5). f Serum was collected at 3 dpi and g brains were collected at 7 dpi for ZIKV quantification via plaque assay (n =5, 3, 5, data represent one of two independent experiments with similar results). Dashed lines represent limit of detection. Data represents mean ± SEM (b, d–f, g). Statistical significance was assessed by one-way ANOVA with Tukey’s multiple comparisons test (b, d–f, g, differences were not significant in f, g). See also Supplementary Fig. 4. Source data are provided as a Source Data file.

Fig. 6. ZIKV infection in mice generates bystander activated CD8⁺ T cells capable of antigen-independent cytotoxicity.

Ifnar^-/- mice were infected with PBS (naïve) or 4x10⁵ PFU ZIKV via FTPD. a At 7 dpi, spleens were harvested and CD8⁺ T cells isolated by negative selection. A cytotoxicity assay was performed by incubating CFSE-labeled YAC-1 cells with spleen CD8⁺ T cells at various effector:target (E:T) ratios (n =4, 5, repeated once with similar results). Each datapoint represents the mean of technical duplicates. b Bulk brain cells from ZIKV-infected mice were pooled, with 5 brains per sample (n =3) and compared to cytotoxicity of CD8⁺ T cells isolated from naïve mouse spleens (n =3, repeated once with similar results). Each datapoint represents the mean of technical duplicates. c Correlation of the proportion of spleen CD8⁺ T cells that were NKG2D⁺ with specific lysis at the 20:1 E:T ratio. d–k Ifnar^-/- mice were infected with PBS (naïve) or ZIKV and treated with PBS or IL-12/18 at 1 and 2 dpi. Brains were harvested for flow cytometry analysis and spleen CD8⁺ T cells were isolated for cytotoxicity assay at 7 dpi. d The proportion of brain CD45⁺ cells that were CD8⁺NKG2D⁺ (n =4, 5, 5, 5). e Proportion of brain CD8⁺ T cells from ZIKV-infected mice that were NKG2D⁺ (n =5). f MFI of NKG2D on brain CD8⁺ T cells from ZIKV-infected mice (n =5). g Count of CD8⁺NKG2D⁺ T cells in the brains of naïve and ZIKV-infected mice treated with PBS or IL-12/18 (n =4, 5, 5, 5). h Proportion of spleen CD45⁺ cells that were CD8⁺NKG2D⁺ (n =4, 5, 4, 5). i Proportion of spleen CD8⁺ T cells that were NKG2D⁺ (n =4, 5, 4, 5). j MFI of NKG2D on spleen CD8⁺ T cells (n =4, 5, 4, 5). k % specific lysis at 10:1 E:T ratios (n =4-5). Data represents mean ± SEM of one of two independent experiments with similar results (a, b, d–k). Statistical significance was determined by two-way ANOVA with Sidak’s multiple comparison test (a, b) or Tukey’s multiple comparison test (d, g–k), linear regression (c), and two-tailed Student’s t test (e, f). MFI mean fluorescence intensity. Source data are provided as a Source Data file.

Fig. 7. ZIKV infection of α-IFNAR-treated C57BL/6 mice promotes bystander activation independent of viral load.

C57BL/6 mice were treated with α-IFNAR antibody at −1, 0, 1, 3, 5 dpi and infected via FTPD with 1.4 × 10⁶ ZIKV PFU. Ifnar^-/- mice were infected via FTPD with 4 × 10⁵ ZIKV PFU. a Viral titer in the serum at 3 dpi (n =4, 5). b Spleen cells were isolated for flow cytometry at 8 dpi and CD45⁺ cells were assessed for % CD8⁺NKG2D⁺ T cells (n =4, 5). c Spleen CD8⁺ T cells were assessed for %NKG2D⁺ or D) NKG2D MFI (n =4, 5). e At 8 dpi, spleen CD8⁺ T cells were isolated via negative selection. A cytotoxicity assay was performed by incubating CFSE-labeled YAC-1 cells with spleen CD8⁺ T cells at various effector:target (E:T) ratios (n =4, 5). Each datapoint represents the mean of technical duplicates. f Weight loss at a % of D0 starting weight (n =4, 5). g Mice were assessed for clinical symptoms of paralysis at 8 dpi (n =8, 14; data represent two independent experiments). h Quantification of clinical scores in g (n =8, 14). i Representative flow cytometry plots of brains of ZIKV-infected Ifnar^-/- and α-IFNAR-treated C57BL/6 mice. j Quantification of i (n =4, 5). k Counts of immune cell populations from i (n =4, 5). l Counts of CD8⁺ and m CD8⁺NKG2D⁺ in ZIKV-infected Ifnar^-/- and α-IFNAR-treated C57BL/6 mice (n =4, 5). Data represents mean ± SEM (a–f, h, j–m) of one of two independent experiments with similar results. Statistical significance was determined by two-tailed Student’s t test (a–d, h, l, m) or repeated measures (f) or ordinary (e, j, k) two-way ANOVA with Sidak’s multiple comparisons test. MFI = mean fluorescence intensity. Source data are provided as a Source Data file.

Fig. 8. Comparison of ZIKV infection and neurological outcomes in Ifnar^-/- and C57BL/6 + α-IFNAR mouse models.

While both Ifnar^-/- and C57BL/6 mice treated with α-IFNAR blocking antibody (WT + α-IFNAR) are susceptible to systemic ZIKV infection, only Ifnar^-/- mice exhibit ZIKV infection and inflammation in the CNS, resulting in neurological symptoms of hindlimb paralysis.