Persistent innate immune dysfunction and ZIKV replication in the gastrointestinal tract during SIV infection in pigtail macaques

Abstract

Mosquito-borne flaviviruses, including dengue (DENV) and Zika (ZIKV) viruses, have caused widespread epidemics in areas with high HIV prevalence, partly due to the expanded geographic range of arthropod vectors. Despite the occurrence of large flavivirus outbreaks in areas with high HIV prevalence, little is known about the effects of flavivirus infection in people living with HIV (PLWH). Here, we use a pigtail macaque model of HIV/AIDS to investigate the impact of simian immunodeficiency virus (SIV)-induced immunosuppression on ZIKV replication and pathogenesis. During acute SIV infection, peripheral ZIKV cellular targets expanded and innate immune activation increased. In vitro, peripheral blood mononuclear cells (PBMC) from SIV infected pigtail macaques were less permissive to ZIKV infection. In vivo, ZIKV viremia was delayed and ZIKV was more persistent in the gastrointestinal tissues of SIV-ZIKV co-infected animals. This persistence was associated with changes in innate cellular (monocytes, neutrophils) recruitment to the blood and tissues, reduced anti-ZIKV immunity, and sustained expression of peripheral inflammatory and innate immune genes. Collectively, these findings uniquely suggest that untreated SIV infection may promote inflammatory cellular innate responses and create a state of persistent immune activation that contributes to prolonged ZIKV viremia and persistence in the gastrointestinal tract. Furthermore, these results suggest that PLWH and other immunocompromised individuals could be at higher risk for prolonged ZIKV infection, potentially extending the window of ZIKV transmission. These insights highlight the importance of including PLWH in strategies for deploying vaccines and treatments against ZIKV.

Keywords: Zika virus; co-infection; innate immunity; nonhuman primate; simian immunodeficiency virus (SIV).

Figure 1

PBMC from SIV-infected PTM are less permissive to in vitro ZIKV infection. (A, B) Peripheral blood mononuclear cells (PBMC) were isolated from pigtail macaques prior to and at 2 and 6 weeks post-SIV infection and infected in vitro with ZIKV Brazil at a multiplicity of infection (MOI) of 2. Cells and supernatant were harvested 4, 24, and 48 hours post-infection. (A) Quantitative real-time PCR (qRT-PCR) for ZIKV RNA in PBMC. (B) Plaque assay for infectious Zika virus. (A, B) Medians with interquartile ranges are shown. Kruskal-Wallis test versus pre-SIV levels, p-values * ≤ 0.05. (C) Frequency of CD16+CD14+ monocytes and macrophages (Mon&Mac) (top panel) and dendritic cells (lower panel) in blood from uninfected and SIV-infected pigtail macaques. Wilcoxon matched-pairs signed rank test, p-values ≤ 0.05 considered significant. (D) Gene expression of PBMC in blood at Week 2 post-SIV (top panel) and Week 6 post-SIV (bottom panel). t-test between each timepoint relative to pre-SIV, p-values *<0.01 shown by orange dots.

Figure 2

ZIKV viremia is delayed and protracted and ZIKV viral burden more persistent in gastrointestinal tissues in SIV-infected macaques. (A) Study design of longitudinal blood and tissue sampling following SIV and ZIKV infections in pigtail macaques. Initially n=7/group were infected with ZIKV; however, 1 animal in the SIV+ group (Z14109) displayed no evidence of ZIKV replication and thus was excluded from all post-ZIKV analysis. (B) Quantitative real-time PCR (qRT-PCR) for ZIKV RNA in longitudinal samples from plasma, peripheral lymph node (PLN), rectal biopsies, and rectal cytobrush until necropsy (Nx). Virus was not detected in any longitudinal cerebrospinal fluid (CSF). (C) ZIKV RNA in tissues collected at necropsy 24-28 DPI. Virus was not detected in brain tissue (brainstem, hippocampus, frontal lobe, parietal lobe, and occipital lobe) of any animal. Gastrointestinal (GI), lymph node (LN). (D) Total viral burden in other lymphoid, GI-training and GI tissues, as described in panel (C). Each point represents an individual animal and/or tissue and medians are shown. Mann-Whitney test between groups, *p-values ≤ 0.05 are considered significant. (E) Scatter plot and Spearman’s correlation analysis of the relationship between the total viral burden in GI-draining lymph nodes and gastrointestinal tissues at necropsy. Spearman’s correlation test p-values are indicated: **p ≤ 0.01.

Figure 3

SIV infection may impair anti-ZIKV immunity. (A) Longitudinal plasma concentrations of anti-ZIKV envelope IgG as determined by ELISA. AUCs were calculated from day 10 to 28. Medians with interquartile ranges are displayed. (B) Zika virus neutralization antibody titers (NT50 values) evaluated at necropsy. The line represents the median and the dotted line represents is the limit of detection. (A, B) Mann-Whitney test comparison between groups.

Figure 4

Post-ZIKV recruitment of CD16+ monocytes and macrophages is dampened in the periphery, but enhanced in tissues in SIV-infected macaques. (A) Frequency of CD16+CD14+ monocytes and macrophages (Mon&Mac) in blood (left panel), rectum (center panel), and peripheral lymph node (right panel) after ZIKV infection. (B) Frequency of neutrophils in blood after ZIKV infection. (C) Concentration of myeloperoxidase (MPO) in plasma as measured by ELISA. (A-C) Medians with interquartile ranges are shown. Mann-Whitney test between groups, p-values * ≤ 0.05.

Figure 5

SIV-ZIKV co-infection induces persistent innate immune activation in PBMC. (A) Heatmap showing the Log2FC expression of 23 genes that were significantly different (p-value <0.01) in at least one timepoint and one group. Log2FC expression for the SIV+ZIKV+ group is relative to pre-SIV (Wk-3). Log2FC expression for the SIV-ZIKV+ group is relative to pre-ZIKV (D-14). Genes’ Log2FCs were clustered using Pearson and Ward.D2. (B) Venn diagram of shared and unique genes that were significantly upregulated post-ZIKV. (C) Line plots of select gene kinetics representing the mean of all SIV+ZIKV+ (red) or SIV-ZIKV+ (black) animals. The log2 normalized counts are plotted at each timepoint. The line represents the mean and the standard error is shown as the confidence interval around the mean. p-values * <0.01 indicates a significant difference between SIV+ZIKV+ and or SIV-ZIKV+ at a specified timepoint.

Figure 6

Immune responses and innate gene signature associated with persistent ZIKV RNA in the gastrointestinal tract. Scatter plot and Spearman’s correlation analysis of the relationship between the gut ZIKV viral burden at necropsy and (A) plasma MPO (AUC, days 1-4) (left panel), plasma FABP2 (AUC, days 1-21) (right panel) or (B) CD16+ monocytes and macrophages (Mon&Mac) in rectum (AUC, days 7-21). (A, B) Spearman’s correlation test p-values are indicated as follows: *p ≤ 0.05,**p ≤ 0.01.