Flavivirus Antibodies Reactive to Zika Virus Detected in Multiple Species of Nonhuman Primates in Kenya,2008–2017

Introduction

Zika virus (ZIKV) is a mosquito-borne flavivirus first isolated from a rhesus macaque in captivity in the Zika Forest, Uganda in 1947 (Dick et al., 1952; Pielnaa et al., 2020) and later isolated from Aedes africanus mosquitoes (Dick et al., 1952) and from a febrile patient in 1952 (MacNamara, 1954). For the past 50 years, there have been reports of sporadic human cases and even major outbreaks of ZIKV in various countries in Asia, Africa, and most recently the Americas (Kazmi et al., 2020).

To date, recent serosurveys have identified ZIKV-specific antibodies in the Kenyan population (Chepkorir et al., 2019; Hunsperger et al., 2020; Kisuya et al., 2019), suggesting current transmission. Despite the widespread detection of circulating antibodies in humans, the amplifying host of ZIKV transmission remains unknown. Since Kenya has a broad range of nonhuman primate (NHP) species that could serve as potential virus reservoir hosts (Buechler et al., 2017), we tested for the presence of antibodies reactive to ZIKV in multiple NHP species to assess their putative role in maintaining ZIKV transmission.

Materials and Methods

We obtained archived sera from the Institute of Primate Research (IPR) in Nairobi, Kenya, collected through convenience sampling between 2008 and 2017. We targeted testing at least 73 sera from each NHP species to detect 5% seroprevalence at 5% precision level. From a list of 1193 sera with at least 1 mL sample volume and accompanying data, we randomly selected 87 of 1018 Olive baboon and 76 of 126 Vervet monkey sera stratified by year and place (county) of collection. Forty-nine Sykes monkey sera were also selected. Aliquots were shipped to the U.S. Centers for Disease Control and Prevention (CDC)-supported Kenya Medical Research Institute (KEMRI) laboratory in Nairobi for testing.

We optimized an enzyme-linked immunosorbent assay (ELISA)-format microneutralization test adapted from methods by Vorndam and Beltran.(Vorndam and Beltran, 2002) In brief, twofold serum serial dilutions (1:20 to 1:80) were tested against titrated wild-type ZIKV strain MR766 (African lineage) on confluent Vero cell monolayer in triplicate, incubated for 3 days, fixed and analyzed by an ELISA to detect virus infected cell by the detection of ZIKV envelope protein using monoclonal antibody 6B6C (CDC, Fort-Collins, Arbovirus Disease Branch) and a secondary goat anti-mouse IgG heavy chain polyclonal antibody conjugated to HRPO (Kirkegaard & Perry Laboratories).

The neutralizing titers were expressed as the reciprocal of the dilution in which there was a reduction in optical density (OD) corresponding to ZIKV control virus OD plus two standard deviations. A twofold dilution of titrated ZIKV MR766 was used as the positive control. The OD values were determined and analyzed using ELISA software (Gen 5; BioTek). All samples with a titer of ≥20 were considered positive, whereas samples with a titer of <20 were negative. The study received ethical approval from the KEMRI Scientific Review unit number 3386. In addition, the activity was reviewed by CDC and was conducted consistent with applicable federal law and CDC policy (no. 2017-318).

Results

We tested 212 sera collected from 87 (41.0%) Olive baboons, 76 (35.8%) Vervet monkeys, and 49 (23.1%) Sykes monkeys (Table 1). Half (51.2%) were male, and 60.7% were adult. Sera had been collected from NHPs in seven counties, with 31.6% collected from a colony at IPR and 31.1% from Kajiado County.

We detected antibodies that neutralized ZIKV in 38 (17.9%) individuals. Antibodies were detected in 15 (19.7%) Vervet monkeys, 16 (18.4%) Olive baboons, and 7 (14.3%) Sykes monkeys (Table). Of the nine noncontiguous years that sera were tested, we identified most specimens with antibodies that specifically neutralized ZIKV in 2012 (61.8%), followed by 2008 (33.3%) and 2015 (33.3%), and last 2009 (20.0%). By county, seropositivity was highest in Machakos (50.0%) followed by Kajiado (19.7%). Seropositivity was variable by year and county of collection. Of the 38 sera that neutralized ZIKV 34.2% (n = 13) were from Kajiado and 55.3% (n = 21) were collected in 2012. Most of the samples (65.8%) had serum titers of 20, 6 (15.8%) had titers of 40, and 7 (18.4%) had titers of >80.

Conclusions

We detected flavivirus antibodies reactive to ZIKV in Olive baboons, Vervet monkeys, and Sykes monkeys in five of the nine noncontiguous years when samples were collected and in five of seven counties where samples were collected. Our results suggest ongoing extensive spatial and temporal flavivirus transmission among the NHP populations in Kenya. This observation is in concordance with similar seropositivity rates in NHP species sampled from South Africa, Tanzania, Zambia, and Gambia (Kazmi et al., 2020). A limitation of our study was that there was insufficient specimen volume to test for other potential cross-reactive flaviviruses, notably yellow fever and dengue viruses. However, these findings highlight the range of NHP species that could serve as reservoir hosts for flaviviruses including ZIKV thus increasing the risk of spillover to humans, and the need for surveillance in human populations within regions where NHPs and humans cohabitate.