An epidemic Zika virus isolate suppresses antiviral immunity by disrupting antigen presentation pathways

Abstract

Zika virus (ZIKV) has emerged as an important global health threat, with the recently acquired capacity to cause severe neurological symptoms and to persist within host tissues. We previously demonstrated that an early Asian lineage ZIKV isolate induces a highly activated CD8 T cell response specific for an immunodominant epitope in the ZIKV envelope protein in wild-type mice. Here we show that a contemporary ZIKV isolate from the Brazilian outbreak severely limits CD8 T cell immunity in mice and blocks generation of the immunodominant CD8 T cell response. This is associated with a more sustained infection that is cleared between 7- and 14-days post-infection. Mechanistically, we demonstrate that infection with the Brazilian ZIKV isolate reduces the cross-presentation capacity of dendritic cells and fails to fully activate the immunoproteasome. Thus, our study provides an isolate-specific mechanism of host immune evasion by one Brazilian ZIKV isolate, which differs from the early Asian lineage isolate and provides potential insight into viral persistence associated with recent ZIKV outbreaks.

Fig. 1. Reduced CD8 T cell response to ZIKV^BR is associated with virus detection at later time points following infection.

a, b Frequency (a) and number (b) of CD8α^loCD11a^hi CD8 T cells in the peripheral blood of mice at indicated dpi with ZIKV^CDN or ZIKV^BR. Data are representative of two independent experiments with n = 3 mice per group, sampled repeatedly at indicated time points. Data were analyzed by two-tailed, unpaired Student’s t-test at each time point. In a day 5 p = 0.0041, day 7 p = 0.0004, day 10 p = 0.0015, day 14 p = 0.0067, day 20 p = 0.0073, day 30 p = 0.0132, day 40 p = 0.0137, day 50 p = 0.0157. In b day 5 p = 0.0004, day 7 p = 0.0014, day 10 p = 0.0003, day 14 p = 0.0218, day 20 p = 0.0169, day 30 p = 0.0002, day 40 p = 0.0371, day 50 p = 0.0327. c, d Frequency (c) and number (d) of CD11a⁺CD49d⁺ CD4 T cells in the peripheral blood of mice at indicated dpi with ZIKV^CDN or ZIKV^BR. Data are representative of two independent experiments with n = 3 mice per group, sampled repeatedly at indicated time points. Data were analyzed by two-tailed, unpaired Student’s t-test at each time point. In d day 0 p = 0.0407. e Viral RNA was quantified in the spleen using RT-qPCR analysis. Data are presented as PFU equivalents per gram of tissue after comparison to a standard curve of C_t value versus log₁₀(PFU) for each isolate. Mice were sacrificed 12 hpi and 3, 5, and 7 dpi with ZIKV^CDN or ZIKV^BR, or following mock-infection or injection with UV-ZIKV (12 hpi only). Data are representative of three independent experiments, with n = 3 mice per group at each time point. Data were analyzed by two-tailed, unpaired Student’s t-test at each time point, 12 hpi p = 0.001, 5 dpi p = 0.000079, 7 dpi p = 0.00014. f, g Viral burden in the spleen (f) and kidney (g) were quantified via plaque assay 7 dpi with each ZIKV isolate. N.D. indicates no data recorded above the LOD (dotted line). Data are pooled from two experiments with n = 5 mice per group. All data are shown as mean ± SEM and *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. Red squares = ZIKV^CDN-infected mice, blue triangles = ZIKV^BR-infected mice, black squares UV-ZIKV-injected mice. Source data are provided as a Source Data file.

Fig. 2. ZIKV^BR induces a less robust CD8 T cell response that does not respond to the immunodominant Env_294–302 epitope.

a Total spleen cellularity 7 dpi with ZIKV^CDN or ZIKV^BR, p = 0.0146. b–g Representative flow cytometry plots (b), frequency (d), and number (e) of CD8α⁺ T cells 7 dpi with ZIKV^CDN or ZIKV^BR. In d p = 0.0047 and in e, p = 0.0008. Representative flow cytometry plots (c), frequency (f), and number (g) of CD8α^loCD11a^hi CD8 T cells in the spleen 7 dpi with ZIKV^CDN or ZIKV^BR. In f p = 0.0002 and in g, p = 0.0004. h–j Representative flow cytometry plots of H-2D^b Env_294–302 tetramer-positive CD8α^loCD11a^hi CD8 T cells from ZIKV^CDN and ZIKV^BR-infected mice (h), and frequency (i) and number (j) of H-2D^b Env_294–302 tetramer-positive CD8α^loCD11a^hi CD8 T cells in the spleen 7 dpi with ZIKV^CDN or ZIKV^BR. In i p = 0.000019 and in j, p = 0.000045. All data are representative of four independent experiments with n = 3 mice per group and are shown as mean ± SEM. All data were analyzed by a two-tailed, unpaired Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. Red squares = ZIKV^CDN-infected mice, blue triangles = ZIKV^BR-infected mice. Source data are provided as a Source Data file.

Fig. 3. CD8 T cells present a less activated phenotype following ZIKV^BR infection.

a–c Representative flow cytometry plots of CD127 and KLRG1 expression by CD8α^loCD11a^hi CD8 T cells from ZIKV^CDN and ZIKV^BR-infected mice (a), and frequency (b) and number (c) of CD127^loKLRG1^hi CD8α^loCD11a^hi CD8 T cells (circles; TEC) and CD127^hiKLRG1^lo CD8α^loCD11a^hi CD8 T cells (squares; MPC) in the spleen 7 dpi with ZIKV^CDN or ZIKV^BR. In b TEC p = 0.0004 and MPC p = 0.0002. In c TEC p = 0.000049 and MPC p = 0.0103. d–f Representative histogram (d), frequency (e) and number (f) of CD62L⁺ CD8α^loCD11a^hi CD8 T cells in the spleen 7 dpi with ZIKV^CDN or ZIKV^BR. Line on histogram indicates gating strategy used to identify CD62L⁺ cells. In e p = 0.0003 and in f p = 0.01. g–i Representative histogram (g), frequency (h), and number (i) of granzyme B⁺ CD8α^loCD11a^hi CD8 T cells in the spleen 7 dpi with ZIKV^CDN or ZIKV^BR. Shaded histogram indicates isotype control. Line on histogram indicates gating strategy used to identify granzyme B⁺ cells. In h p = 0.0026 and in i, p = 0.0005. All data are representative of two independent experiments with n = 3 mice per group and are shown as mean ± SEM. All data were analyzed by a two-tailed, unpaired Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. Red symbols = ZIKV^CDN-infected mice, blue symbols = ZIKV^BR-infected mice. Source data are provided as a Source Data file.

Fig. 4. Reduced IFN-I production and dendritic cell activation following ZIKV^BR infection do not impact antigen-experienced CD8 T cell division.

a, b Ifna (a) and Ifnb1 (b) mRNA expression in the spleen were analyzed by RT-qPCR 12 hpi with ZIKV^CDN or ZIKV^BR, or mock-infection. Data are expressed as fold change over expression in mock-infected mice 12 hpi. Dotted line indicates a fold change of 1. Data are representative of two independent experiments with n = 3 mice per group, and were analyzed by two-tailed, unpaired Student’s t-test. In a p = 0.0047 and in b p = 0.000004. c Total bioactive IFN-I were analyzed in the serum 12 hpi with ZIKV^CDN or ZIKV^BR using the B16-blue reporter cell line. Dotted line indicates average OD in the serum of mice 12 h post-mock infection. Data are representative of two independent experiments with n = 3 mice per group, and were analyzed by two-tailed, unpaired Student’s t-test, p = 0.000011. d–f Representative flow cytometry plots of CD8α^loCD11a^hi CD8 T cells from ZIKV^CDN- or ZIKV^BR-infected mice treated with PBS 2 and 3 dpi (d), and from ZIKV^BR-infected mice treated with pI:C 2 and 3 dpi. Frequency (e) and number (f) of CD8α^loCD11a^hi CD8 T cells in the spleen 7 dpi with ZIKV^CDN or ZIKV^BR, after treatment with either PBS or pI:C 2 and 3 dpi. Data are representative of two independent experiments with n = 5 mice per group and were analyzed by one-way ANOVA with Tukey’s post-test for multiple comparisons. In e ZIKV^CDN + PBS versus ZIKV^BR + PBS p = 0.0000001, ZIKV^CDN + PBS versus ZIKV^BR + pI:C p = 0.0000041, ZIKV^BR + PBS versus ZIKV^BR + pI:C p = 0.0112. In f ZIKV^CDN + PBS versus ZIKV^BR + PBS p = 0.000000008, ZIKV^CDN + PBS versus ZIKV^BR + pI:C p = 0.000000059. g–i Representative histogram (g), frequency (h), and number (i) of BrdU⁺ CD8α^loCD11a^hi CD8 T cells in the spleen 7 dpi with ZIKV^CDN or ZIKV^BR. Line on histogram indicates gating strategy used to identify BrdU⁺ cells. Data are representative of two independent experiments with n = 3 mice per group, and were analyzed by two-tailed, unpaired Student’s t-test. In i p = 0.0013. j–m Representative histograms and geometric mean fluorescence intensity (gMFI) of CD80 (j, k) and CD86 (l, m) expression on CD3⁻CD19⁻NK1.1⁻ CD11c⁺MHC-II⁺ splenic DCs 2 dpi with ZIKV^CDN or ZIKV^BR, or injection with an equivalent dose of UV-ZIKV. Data are representative of two independent experiments with n = 3 mice per group, and were analyzed by one-way ANOVA with Tukey’s post-test of multiple comparisons. In k ZIKV^CDN versus UV-ZIKV p = 0.0002 and ZIKV^CDN versus ZIKV^BR p = 0.0001. In m ZIKV^CDN versus UV-ZIKV p = 0.000002 and ZIKV^CDN versus ZIKV^BR p = 0.000002. All data are shown as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. Red squares and red lines = ZIKV^CDN-infected mice, blue triangles and blue lines = ZIKV^BR-infected mice, black squares and black lines = UV-ZIKV-injected mice, gray lines = isotype controls. Source data are provided as a Source Data file.

Fig. 5. ZIKV^BR infection suppresses antigen cross-presentation and immunoproteasome induction.

a TCR Vβ subunit usage among CD8α^loCD11a^hi CD8 T cells 7 dpi with ZIKV^CDN or ZIKV^BR. Data are representative of two independent experiments with n = 3 mice per group. b, c Proliferation dye dilution in OT-I CD8 T cells 3 days following co-culture with BMDCs. Prior to co-culture with labeled OT-I CD8 T cells, DCs were infected with ZIKV^CDN or ZIKV^BR, or UV-inactivated ZIKV^CDN or ZIKV^BR, at a MOI of 5 for 6 h, followed by a 4-h incubation with LPS and either OVA_257–264 peptide (b) or OVA protein (c). Data are representative of three independent experiments. d mRNA expression of Ifng in the spleen was analyzed by RT-qPCR 12 hpi with ZIKV^CDN or ZIKV^BR, or mock-infection. Data are expressed as fold change over expression in mock-infected mice 12 hpi. Dotted line indicates a fold change of 1. Data are pooled from two independent experiments with n = 3 mice per group and were analyzed by two-tailed, unpaired Student’s t-test, p = 0.0313. e mRNA expression of Psmb8 in the spleen was analyzed by RT-qPCR 12 hpi with ZIKV^CDN or ZIKV^BR, or mock-infection. Data are expressed as fold change over expression in mock-infected mice 12 hpi. Dotted line indicates a fold change of 1. Data are pooled from two independent experiments with n = 3 mice per group and were analyzed by two-tailed, unpaired Student’s t-test, p = 0.0466. f–h Representative flow cytometry plots of H-2D^b Env_294–302 tetramer-positive CD8α^loCD11a^hi CD8 T cells from ZIKV^CDN-infected mice following treatment with vehicle control or ONX 0914 (f). Frequency (g) and number (h) of H-2D^b Env_294–302 tetramer-positive CD8α^loCD11a^hi CD8 T cells in the spleen 7 dpi with ZIKV^CDN. Mice were treated s.c. with vehicle control or ONX 0914 prior to infection. Data are pooled from two independent experiments with n = 3 (first experiment), n = 4 (ONX 0914 treated, second experiment), or n = 5 mice per group (vehicle-treated, second experiment), and were analyzed by two-tailed, unpaired Student’s t-test. In g p = 0.0034 and in h p = 0.0414. All data are shown as mean ± SEM. *p < 0.05, and **p < 0.01. Red squares and red lines = ZIKV^CDN-infected mice or BMDC, blue triangles and blue lines = ZIKV^BR-infected mice or BMDC, black lines = UV- ZIKV^BR inoculated BMDCs, gray lines = UV-ZIKV^CDN inoculated BMDCs. Source data are provided as a Source Data file.

Fig. 6. ZIKV^BR infection actively suppresses the Env_294–302-specific CD8 T cell response during co-infection.

a, b Ifna (a) and Ifnb1 (b) mRNA expression in the spleen were analyzed by RT-qPCR 12 hpi with ZIKV^CDN or ZIKV^BR, co-infection, injection with UV-ZIKV, or mock-infection. Data are expressed as fold change over expression in mock-infected mice 12 hpi. N.D. indicates no expression was detected by RT-qPCR. In a UV-ZIKV versus ZIKV^CDN + ZIKV^BR p = 0.0082 and ZIKV^BR versus ZIKV^CDN + ZIKV^BR p = 0.0085. In b ZIKV^CDN versus ZIKV^BR p = 0.008 and ZIKV^BR versus ZIKV^CDN + ZIKV^BR p = 0.0045. c Total bioactive IFN-I were analyzed in the serum 12 hpi with ZIKV^CDN or ZIKV^BR, co-infection, injection with UV-ZIKV, or mock-infection using the B16-blue reporter cell line. Dotted line indicates average OD in the serum of mice 12 h post-mock infection. UV-ZIKV versus ZIKV^CDN p = 0.000029, UV-ZIKV versus ZIKV^CDN + ZIKV^BR p = 0.00006, ZIKV^BR versus ZIKV^CDN p = 0.000022, ZIKV^BR versus ZIKV^CDN + ZIKV^BR p = 0.000046. d–f Representative flow cytometry plots of H-2D^b Env_294–302 tetramer-positive CD8α^loCD11a^hi CD8 T cells from ZIKV^CDN-, ZIKV^BR- or co-infected mice (d). Frequency (e) and number (f) of H-2D^b Env_294–302 tetramer-positive CD8α^loCD11a^hi CD8 T cells in the spleen 7 dpi with ZIKV^CDN or ZIKV^BR, or co-infection. In e ZIKV^CDN versus ZIKV^BR p = 0.000007, ZIKV^CDN versus ZIKV^CDN + ZIKV^BR p = 0.0112, ZIKV^BR versus ZIKV^CDN + ZIKV^BR p = 0.000044. In f ZIKV^CDN versus ZIKV^BR p = 0.000016, ZIKV^CDN versus ZIKV^CDN + ZIKV^BR p = 0.0003, ZIKV^BR versus ZIKV^CDN + ZIKV^BR p = 0.0034. All data are representative of two independent experiments with n = 3 mice per group and are shown as mean ± SEM. All data were analyzed by one-way ANOVA with Tukey’s post-test of multiple comparisons. UV-ZIKV group in b was excluded from statistical analysis. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. Red squares = ZIKV^CDN-infected mice, blue triangles = ZIKV^BR-infected mice, inverted blue triangles = ZIKV^CDN + ZIKV^BR co-infected mice, black squares = UV-ZIKV-injected mice. Source data are provided as a Source Data file.