Mucosal Responses to Zika Virus Infection in Cynomolgus Macaques

Abstract

Zika virus (ZIKV) cases continue to be reported, and no vaccine or specific antiviral agent has been approved for the prevention or treatment of infection. Though ZIKV is primarily transmitted by mosquitos, cases of sexual transmission and prolonged viral RNA presence in semen have been reported. In this observational study, we report the mucosal responses to sub-cutaneous and mucosal ZIKV exposure in cynomolgus macaques during acute and late chronic infection. Subcutaneous challenge induced a decrease in the growth factor VEGF in colorectal and cervicovaginal tissues 100 days post-challenge, in contrast to the observed increase in these tissues following vaginal infection. This different pattern was not observed in the uterus, where VEGF was upregulated independently of the challenge route. Vaginal challenge induced a pro-inflammatory profile in all mucosal tissues during late chronic infection. Similar responses were already observed during acute infection in a vaginal tissue explant model of ex vivo challenge. Non-productive and productive infection 100 days post-in vivo vaginal challenge induced distinct proteomic profiles which were characterized by further VEGF increase and IL-10 decrease in non-infected animals. Ex vivo challenge of mucosal explants revealed tissue-specific modulation of cytokine levels during the acute phase of infection. Mucosal cytokine profiles could represent biosignatures of persistent ZIKV infection.

Keywords: Zika virus; ex vivo challenge; immune responses; mucosal tissue; non-human primates.

Figure 1

Effect of ZIKV challenge route on the mucosal cytokine profile. Heat map representing cytokines that are upregulated (orange) or downregulated (blue) in culture supernatants from colorectal, vaginal, cervical or uterine tissues obtained 100 days post-challenge from cynomolgus macaques infected or not following subcutaneous or vaginal challenge and from naïve, non-challenged animals. Differences are shown in Log2 from independent experiments performed at least in duplicate with two animals per group.

Figure 2

Canonical pathways significantly enriched in mucosal tissues post-ZIKV challenge. Heat map representing the activation z-scores determined by directionality (overabundance in orange and underabundance in blue) and the number of cytokines measured in colorectal (CR), vaginal (VG), cervical (CX) or uterine (UT) tissue culture supernatants. Tissues were obtained 100 days post-challenge from cynomolgus macaques infected or not following subcutaneous or vaginal challenge and compared with tissues from naïve, non-challenged animals. The pathways included had a minimum of two analytes associated and a p < 0.05 (right-tailed Fisher exact test).

Figure 3

Post-challenge ZIKV RNA level following exposure via the intravaginal route. (a) Vaginal challenge and sampling schedule. (b) Viraemia profiles across the entire time-course following the first challenge with pairs of cynomolgus macaques initially challenged with 3.1 × 10⁶ (R6, R7), 3.1 × 10⁵ (R8, R10) or 3.1 × 10³ pfu (R14, R15) ZIKV PRVABC59 virus inoculum, and a second challenge after 35 days with all 6 macaques challenged with 10⁶ pfu of the same ZIKV PRVABC59 stock. (c) Detailed kinetic profiles following the second challenge. Lower limit of detection for the assay is 50 RNA copies/mL.

Figure 4

Zika anti-NS1 titers. Zika virus-specific (a) IgM levels, (b) IgG levels expressed as relative units per mL (RU/mL) and (c) IgG levels for the 81 days post-challenge time-course showing bleeding frequencies from the start of the study (day 0).

Figure 5

ZIKV replication in mucosal tissue explants following ex vivo challenge. Tissue explants were cultured for 21 days. Culture supernatants were harvested at days 3, 7, 11, 15 and 21. (a) ZIKA RNA levels and (b) viral titers were measure in the harvested culture supernatants. Data are mean (±SD) of experiments performed at least in duplicate with tissues resected from two animals.

Figure 6

RNAscope in situ hybridization analysis. Vaginal and colorectal explants from naïve animals were cultured for 1 day either unchallenged (negative) or exposed to 10⁵ pfu of ZIKV PRABC59 for 2 h (challenged). Representative infection foci are signaled with red arrows. Nuclei were stained with hematoxylin (blue). Blue bar = 200 μm. SSE: stratified squamous epithelium; LP: lamina propria.

Figure 7

Immunohistochemistry analysis of ex vivo mucosal challenge. Uterine, vaginal and colorectal tissue explants from naïve animals were cultured either unchallenged (negative) or exposed to 10⁵ pfu of ZIKV PRABC59 for 2 h (challenged). Samples were stained with anti-NS-1 antibody. Areas of specific staining are indicated by brown foci. Representative infection foci are signaled with red arrows. Nuclei were stained with hematoxylin (blue).

Figure 8

Modulation of mucosal cytokine profiles induced by ex vivo exposure to ZIKV PRABC59. (a) Heat map representing cytokines that are upregulated (orange) or downregulated (blue) from cynomolgus macaque colorectal, vaginal, cervical or uterine explant culture supernatants harvested 24 h post-exposure to ZIKV or not. Differences are shown in Log2 from independent experiments performed at least in duplicate with two animals per group. (b) Heat map representing the canonical pathway activation z-scores determined by directionality (overabundance in orange and underabundance in blue) and the number of cytokines measured in explant culture supernatants. The pathways included had a minimum of two analytes associated and a p < 0.05 (right-tailed Fisher exact test).