Chikungunya virus infection disrupts MHC-I antigen presentation via nonstructural protein 2

Abstract

Infection by chikungunya virus (CHIKV), a mosquito-borne alphavirus, causes severe polyarthralgia and polymyalgia, which can last in some people for months to years. Chronic CHIKV disease signs and symptoms are associated with the persistence of viral nucleic acid and antigen in tissues. Like humans and nonhuman primates, CHIKV infection in mice results in the development of robust adaptive antiviral immune responses. Despite this, joint tissue fibroblasts survive CHIKV infection and can support persistent viral replication, suggesting that they escape immune surveillance. Here, using a recombinant CHIKV strain encoding the fluorescent protein VENUS with an embedded CD8+ T cell epitope, SIINFEKL, we observed a marked loss of both MHC class I (MHC-I) surface expression and antigen presentation by CHIKV-infected joint tissue fibroblasts. Both in vivo and ex vivo infected joint tissue fibroblasts displayed reduced cell surface levels of H2-Kb and H2-Db MHC-I proteins while maintaining similar levels of other cell surface proteins. Mutations within the methyl transferase-like domain of the CHIKV nonstructural protein 2 (nsP2) increased MHC-I cell surface expression and antigen presentation efficiency by CHIKV-infected cells. Moreover, expression of WT nsP2 alone, but not nsP2 with mutations in the methyltransferase-like domain, resulted in decreased MHC-I antigen presentation efficiency. MHC-I surface expression and antigen presentation was rescued by replacing VENUS-SIINFEKL with SIINFEKL tethered to β2-microglobulin in the CHIKV genome, which bypasses the requirement for peptide processing and TAP-mediated peptide transport into the endoplasmic reticulum. Collectively, this work suggests that CHIKV escapes the surveillance of antiviral CD8+ T cells, in part, by nsP2-mediated disruption of MHC-I antigen presentation.

Fig 1. CHIKV-infected primary joint tissue fibroblasts display decreased MHC-I cell surface expression.

(A) Schematic of the recombinant CHIKV-VENKL. The coding sequence for the VENUS-SIINFEKL chimeric protein was inserted in-frame into the structural ORF of the CHIKV genome with known cleavage motifs flanking SIINFEKL. (B-E) EVA cells were inoculated with PBS (mock) or 10⁶ PFU CHIKV-VENKL. (B) At 24 hpi, the percentages of CD45^- (open circles) and CD45⁺ (open squares) cells among mock-inoculated cells and VENUS⁺ cells were determined. (C) The magnitude and frequency of expression of CD29 on CD45^- cells was determined by flow cytometry. (D) Representative flow cytometry plots of H2-K^b and VENUS expressing cells from mock (black) and CHIKV-VENKL-infected cell populations (VENUS⁺, red; VENUS^-, grey). (E) Quantification of surface expression and frequency of H2-K^b and H2-D^b from mock and CHIKV-VENKL-infected EVA cells. Data are pooled (B), or representative (C, E) from 3–4 independent experiments. P values were determined by unpaired student’s t-test (B) or one-way ANOVA with Tukey’s test for multiple comparisons (C,E). *, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001.

Fig 2. Determinants in nsP2 promote impairment of MHC-I antigen presentation by CHIKV-infected cells.

(A) Schematic depicting the location of nsP2 V-loop mutations (QMS; blue) compared with wildtype nsP2 (ATL; red). (B) WT and Ifnar1^-/- murine embryonic fibroblasts (MEFs) were inoculated with CHIKV-VENKL or CHIKV^QMS-VENKL at an MOI of 0.1 FFU/cell. At 0 (input), 1, 24, 48, and 72 hpi, the amount of infectious virus present in culture supernatants was quantified by focus formation assay. (C-H) EVA cells were mock-inoculated or inoculated with CHIKV-VENKL or CHIKV^QMS-VENKL. At 24 hpi, cells were assessed for cell surface expression of H2-K^b, and SIINFEKL-loaded H2-K^b (H2-K^b-SIINFEKL) by flow cytometry. (C) Representative flow cytometry plots depicting H2-K^b-SIINFEKL and H2-K^b cell surface expression on live, CD45^- and live, CD45^-, VENUS⁺ cells from mock-, CHIKV-VENKL-, and CHIKV^QMS-VENKL-inoculated EVA cells. (D) Percentage of VENUS⁺ cells among live, CD45^-, EVA cells. (E) CD29 cell surface expression on live, CD45^-, VENUS⁺ EVA cells. (F) Quantification of H2-D^b and H2-K^b surface expression on live, CD45^-, VENUS⁺ EVA cells. (G) Percentage of double-positive (H2-K^b+H2-K^b-SIINFEKL⁺) cells on live, CD45^-, VENUS⁺ EVA cells. (H) pMHC-I presentation efficiency on live, CD45^-, VENUS⁺ EVA cells. Data are representative of 4 experiments (n = 12). P values were determined by unpaired student’s t-test (D-H). **, P<0.01; ***, P<0.001; ****, P<0.0001.

Fig 3. WT CHIKV-infected cells inefficiently activate CD8⁺ T cells in vitro.

(A-D) EVA cells were mock-inoculated or inoculated with 10⁶ PFU of CHIKV-VENKL or CHIKV^QMS-VENKL. At 24 hpi, EVA cells were cocultured with a 1:1 mixture of 10⁶ bystander CD8⁺ T cells and 10⁶ OT-I SIINFEKL-specific CD8⁺ T cells for 6 h. Bystander CD8⁺ T cells were distinguished from OT-I CD8⁺ T cells by gating on CD8⁺, K^bOVA^257-264 tetramer positive or negative cells, and both populations were assessed for the dual expression of CD69 and CD25 by flow cytometry. (A) Representative flow cytometry plots. (B) Quantification of CD25⁺CD69⁺ double positive cell frequencies for bystander (black) and OT-I (colored) CD8⁺ T cells. (C) Cell surface expression of CD69 (gMFI). (D) Cell surface expression of CD25 (gMFI). Data are representative of two independent experiments (n = 6). P values were determined by paired student’s t-test. ***, P<0.001; ****, P<0.0001.

Fig 4. CHIKV-infected cells inefficiently display MHC-I antigen in vivo.

(A-H) WT C57BL/6 mice were inoculated with 10³ PFU of CHIKV-VENKL (red) or CHIKV^QMS-VENKL (blue). At 1, 2, and 5 dpi, total cells were isolated from the ipsilateral foot and ankle and evaluated by flow cytometry. (A) Frequency of VENUS⁺ cells among live, CD45^- cells. (B) Frequency of CD29⁺ cells among live, CD45^-, VENUS⁺ cells at 2 dpi. (C-E) Cell surface expression of H2-K^b, CD44, and CD29 among live, CD45^-, VENUS⁺ cells normalized to VENUS^- cells. (F) Representative flow cytometry plots of total H2-K^b and H2-K^b-SIINFEKL staining. (G) Quantification of the frequency of double-positive (H2-K^b+H2-K^b-SIINFEKL⁺) cells among live, CD45^-, VENUS⁺ cells. (H) pMHC-I presentation efficiency among live, CD45^-, VENUS⁺ cells. Data are representative of 3 experiments (n = 12) (A-B) or pooled from 3–4 independent experiments (C-H) (n = 8–13). P values were determined by paired student’s t-test (B) or by one-way ANOVA with Tukey’s multiple comparison test (A, C-E, G-H). *, P<0.05; ****, P<0.0001.

Fig 5. Cell surface, but not intracellular, MHC-I expression is reduced in CHIKV-infected cells.

(A) WT murine embryonic fibroblasts (MEFs) stably overexpressing H2-K^b (H2-K^b+) and H2-K^b-/- MEFs were analyzed for intracellular and cell surface expression of H2-K^b by flow cytometry (ICS = intracellular stain, Surf. = surface stain). (B-C) EVA cells were mock-inoculated or inoculated with 10⁶ PFU of CHIKV-VENKL or CHIKV^QMS-VENKL. At 24 hpi, cells were analyzed for intracellular (ICS) and cell surface (Surf.) expression of H2-K^b by flow cytometry. (B) Representative flow cytometry histograms and (C) quantification of the magnitude (gMFI) and frequency (%) of intracellular (ICS) and cell surface (Surf.) H2-K^b expression on CHIKV-infected cells (VENUS⁺) and VENUS^- cells. Data are representative of 2 independent experiments (n = 6). P values were determined by unpaired student’s t-test. ****, P<0.0001.

Fig 6. Bypassing peptide processing and ER transport restores MHC-I antigen presentation by CHIKV-infected cells.

(A) Schematic of the recombinant CHIKV-SIINFEKLβ₂m viral genome. The coding sequence for the β₂mSIINFEKL chimeric protein was inserted into the CHIKV genome in-frame in the viral structural ORF similar to the VENKL coding sequence. (B-C) EVA cells were mock-inoculated or inoculated with CHIKV-VENKL, CHIKV^QMS-VENKL, or CHIKV-SIINFEKLβ₂m. At 24 hpi, live, CD45^-, E2 (CHIKV)⁺ cells were assessed for cell surface expression of H2-K^b and SIINFEKL-loaded H2-K^b (H2-K^b-SIINFEKL) by flow cytometry. (B) Representative flow cytometry plots depicting the frequency of double-positive (H2-K^b+H2-K^b-SIINFEKL⁺) cells among live, CD45^-, CHIKV⁺ cells. (C) Quantification of the frequency of double-positive (H2-K^b+H2-K^b-SIINFEKL⁺) cells (left) and pMHC-I presentation efficiency (right). (D-E) At 24 hpi, ankle cells from C57BL/6 mice mock-inoculated or inoculated with 10³ PFU CHIKV-VENKL, CHIKV^QMS-VENKL, or CHIKV-SIINFEKLβ₂m were assessed by flow cytometry. (D) Representative flow cytometry plots depicting the frequency of double-positive (H2-K^b+H2-K^b-SIINFEKL⁺) cells among live, CD45^-, CHIKV⁺ cells. (E) Quantification of the frequency of double-positive (H2-K^b+H2-K^b-SIINFEKL⁺) cells (left) and pMHC-I presentation efficiency (right) among live, CD45^-, CHIKV⁺ cells. Data are representative of 2 independent experiments (n = 8). P values were determined by one-way ANOVA with Tukey’s multiple comparison test. **, P<0.01; ***, P<0.001; ****, P<0.0001.

Fig 7. CD8⁺ T cells are activated by CHIKV-infected cells when peptide processing and ER transport are bypassed.

(A-B) EVA cells were inoculated with 10⁶ PFU CHIKV-VENKL or CHIKV-SIINFEKLβ₂m. At 24 hpi, EVA cells were cocultured with a 1:1 mixture of 10⁶ bystander and 10⁶ OT-I SIINFEKL-specific CD8⁺ T cells for 6 h. Bystander CD8⁺ T cells were distinguished from OT-I CD8⁺ T cells by gating on CD8⁺, K^bOVA^257-264 tetramer positive or negative cells, and both populations were assessed for the expression of CD69, CD25, IRF4 and CD8 by flow cytometry. (A) Representative flow cytometry plots. (B) Quantification of CD25, CD69, IRF4 (intracellular), and CD8 cell surface expression (upper graphs) and frequency (lower graphs) for OT-I cells above bystander CD8⁺ T cells. Data are representative of two independent experiments (n = 6). P-values were determined by unpaired student’s t-test. *, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001.

Fig 8. Cells expressing nsP2 inefficiently present antigen and have decreased levels of cell surface MHC-I.

(A-F) EVA cells were co-transfected with the designated WT or mutant pCMV-nsP2 plasmid or an identical control plasmid lacking the methionine start codon for nsP2 (vector) and pCMV-VENKL. At 24 h post-transfection, cells were evaluated by flow cytometry and Western blot. (A, D) Western blot analysis of nsP2 and actin expression within transfected EVA cells (anti-nsP2, red, 90 kDa; anti-actin, green, 42 kDa). (B, E) Representative flow cytometry plots depicting the frequency of double-positive (H2-K^b+H2-K^b-SIINFEKL⁺) cells among live, CD45^-, VENUS⁺ cells. (C, F) Quantification of the frequency of double-positive (H2-K^b+H2-K^b-SIINFEKL⁺) cells and pMHC-I presentation efficiency among live, CD45^-, VENUS⁺ cells. Data are pooled from 3 independent experiments. P values were determined by one-way ANOVA with Tukey’s multiple comparisons test. **, P<0.01; ***, P<0.001; ****, P<0.0001.

Fig 9. Proposed model for CHIKV disruption of MHC-I antigen presentation.

CHIKV infection disables MHC-I antigen presentation at a step prior to β₂m-peptide stabilization of the heavy chain within the ER. Figure made with Biorender.com.