Molecular Approaches for the Validation of the Baboon as a Nonhuman Primate Model for the Study of Zika Virus Infection

Abstract

Nonhuman primates (NHP) are particularly important for modeling infections with viruses that do not naturally replicate in rodent cells. Zika virus (ZIKV) has been responsible for sporadic epidemics, but in 2015 a disseminated outbreak of ZIKV resulted in the World Health Organization declaring it a global health emergency. Since the advent of this last epidemic, several NHP species, including the baboon, have been utilized for modeling and understanding the complications of ZIKV infection in humans; several health issues related to the outcome of infection have not been resolved yet and require further investigation. This study was designed to validate, in baboons, the molecular signatures that have previously been identified in ZIKV-infected humans and macaque models. We performed a comprehensive molecular analysis of baboons during acute ZIKV infection, including flow cytometry, cytokine, immunological, and transcriptomic analyses. We show here that, similar to most human cases, ZIKV infection of male baboons tends to be subclinical, but is associated with a rapid and transient antiviral interferon-based response signature that induces a detectable humoral and cell-mediated immune response. This immunity against the virus protects animals from challenge with a divergent ZIKV strain, as evidenced by undetectable viremia but clear anamnestic responses. These results provide additional support for the use of baboons as an alternative animal model to macaques and validate omic techniques that could help identify the molecular basis of complications associated with ZIKV infections in humans.

Keywords: arboviral diseases; cytokines; immunology and infectious diseases; nonhuman primate (NHP); transcriptomic.

Figure 1

Plasma ZIKV loads in four olive baboons. The baboons were initially infected with the Puerto Rico strain of ZIKV (PRVABC59, 2015) via subcutaneous injection with the virus in either a high dose of 10⁶ TCID₅₀ (n = 2, open symbols) or a low dose of 10⁴ TCID₅₀ (n = 2, closed symbols). After 25 weeks, the baboons were exposed to 10⁴ TCID₅₀ of the divergent ZIKV Uganda strain (ZIKV UG1947) through subcutaneous injection.

Figure 2

ZIKV infection in baboons induces a detectable transient systemic change in the activation state of NK, CD8 T, CD4 T, and B cells. The percentage of (A) NK cells and (B) CD8 T cells expressing CD69 increased immediately after infection with ZIKV PR. (C) The percentage of CD4 T cells expressing CD69 peaked at approximately 2 weeks post-infection. (D) The percentage of CD4 T cells expressing CD154 also increased after infection and peaked at 4 weeks post-infection. No changes in expression were detected after challenge with ZIKV UG. (E) Expression of CD83 on B cells. Activation markers were detected using polychromatic flow cytometry. The baboons were challenged with the virus in either a high dose of 10⁶ TCID₅₀ (n = 2, open symbols) or a low dose of 10⁴ TCID₅₀ (n = 2, closed symbols).

Figure 3

Baboon plasma cytokine levels after ZIKV PR infection. (A) IFN-α, (B) IFN-γ, (C) IP-10/CXCL10, (D) BAFF, (E) MIG/CXCL9, and (F) perforin. Cytokines were measured using Luminex assays. The baboons were challenged with the virus in either a high dose of 10⁶ TCID₅₀ (n = 2, open symbols) or a low dose of 10⁴ TCID₅₀ (n = 2, closed symbols).

Figure 4

Humoral immune responses to ZIKV in exposed baboons. (A) ELISA anti-ZIKV NS1 IgG levels. (B) ELISA cross-reactive anti-Dengue virus GP IgG antibodies. (C) Plaque-reduction neutralization test (PRNT) on longitudinally collected baboon plasma. Baboons were challenged with the virus in either a high dose of 10⁶ TCID₅₀ (n = 2, open symbols) or a low dose of 10⁴ TCID₅₀ (n = 2, closed symbols).

Figure 5

ZIKV-specific AIM assay. Baboon PBMCs were obtained a week before and a week after challenge with ZIKV UG, and were incubated with supernatant of HEK cells transduced with empty lentiviral vector (RPMI), or with supernatant of HEK cells transduced with a lentiviral vector expressing ZIKVprME protein (prME); Staphylococcus enterotoxin A and B (SEA/B) was used as positive control. After 24 hr stimulation supernatants were collected and used to measure (A) IL-6, (B) Perforin, (C) TNF-α with a Luminex assay. (D) Cells were used to measure CD83 expression on CD20 B cells. Two-tailed paired t-test. ns, not significant (p>0.05), * = p< 0.05; ** = p<0.01; *** = p<0.001.

Figure 6

Gene Set Enrichment Analysis (GSEA) of Reactome canonical pathways. Gene sets revealed a pattern of increased expression of genes associated with interferon signaling and antiviral mechanisms between the beginning and peak of infection (day 3 vs 0), then decreased expression of these genes between the peak and resolution of infection (day 15 vs 3). No figure was included for the day 15 vs 0 comparison because no gene sets were significantly enriched. (NES: Normalized Enrichment Score).

Figure 7

Gene Set Enrichment Analysis (GSEA) of Hallmark gene sets. Revealed a pattern of increased expression of genes associated with the interferon alpha and gamma response between the beginning and peak of infection (day 3 vs 0), then decreased expression of these genes between the peak and resolution of infection (day 15 vs 3). (NES: Normalized Enrichment Score).

Figure 8

Heat maps of genes associated with the gene ontology (GO) terms. (A) “negative regulation of viral genome replication,” (B) “response to virus,” and (C) “immune response”. The three GO terms shown here were the only ones found to be significantly enriched by the DAVID Functional Annotation tool. No GO terms were found to be enriched for the day 15 vs 0 comparison. Boxes with an X indicates an FDR > 0.05.