Seroprevalence of Zika Virus in Wild African Green Monkeys and Baboons

Abstract

Zika virus (ZIKV) has recently spread through the Americas and has been associated with a range of health effects, including birth defects in children born to women infected during pregnancy. Although the natural reservoir of ZIKV remains poorly defined, the virus was first identified in a captive "sentinel" macaque monkey in Africa in 1947. However, the virus has not been reported in humans or nonhuman primates (NHPs) in Africa outside Gabon in over a decade. Here, we examine ZIKV infection in 239 wild baboons and African green monkeys from South Africa, the Gambia, Tanzania, and Zambia using combinations of unbiased deep sequencing, quantitative reverse transcription-PCR (qRT-PCR), and an antibody capture assay that we optimized using serum collected from captive macaque monkeys exposed to ZIKV, dengue virus, and yellow fever virus. While we did not find evidence of active ZIKV infection in wild NHPs in Africa, we found variable ZIKV seropositivity of up to 16% in some of the NHP populations sampled. We anticipate that these results and the methodology described within will help in continued efforts to determine the prevalence, natural reservoir, and transmission dynamics of ZIKV in Africa and elsewhere. IMPORTANCE Zika virus (ZIKV) is a mosquito-borne virus originally discovered in a captive monkey living in the Zika Forest of Uganda, Africa, in 1947. Recently, an outbreak in South America has shown that ZIKV infection can cause myriad health effects, including birth defects in the children of women infected during pregnancy. Here, we sought to investigate ZIKV infection in wild African primates to better understand its emergence and spread, looking for evidence of active or prior infection. Our results suggest that up to 16% of some populations of nonhuman primate were, at some point, exposed to ZIKV. We anticipate that this study will be useful for future studies that examine the spread of infections from wild animals to humans in general and those studying ZIKV in primates in particular.

Keywords: Flavivirus; Zika virus; Zika virus infection; false-positive reactions; sensitivity and specificity; seroepidemiologic studies.

FIG 1

ZIKV ELISA results by NHP type and prior exposure. (A) Reactivity of the anti-ZIKV antibody capture ELISA to serum/plasma collected from macaques (yellow), AGMs (green), and baboons (blue). Anti-ZIKV antibody titers were compared among different groups within each NHP genus using a two-tailed unpaired t test. For macaques, these data were also used to generate the receiver operating characteristic (ROC) curve. The geographic region from which samples were collected from wild NHPs is indicated in the maps under each column. (B) ROC curve, plotting the true-positive rate (TPR; sensitivity) against the false-positive rate (FPR; 1 specificity) for potential discrimination thresholds. Units are presented here as percentages rather than fractions. Potential discrimination thresholds were found by sorting all of the values from captive macaques and averaging adjacent values. The area under the curve (AUC) indicates the ability of the test to discriminate between primates with ZIKV and those without, with a perfect test having an AUC of 1 and a test with no ability to discriminate having an AUC of 0.5. Both AUC and the significance compared to a test with an AUC of 0.5 were computed using GraphPad Prism (GraphPad Software, Inc., La Jolla CA). The potential discrimination value that maximized the likelihood ratio (TPR/FPR) was 1.445, and this value was adopted as the empirical positivity threshold, depicted by the dashed line in panel A. (C) Data points above the dashed line in panel A were deemed “positive” and used to estimate the prevalence of ZIKV exposure in each population.

FIG 2

Geographic representation of ZIKV detection in wild African primates. Countries where human ZIKV infection has been previously detected are shown in blue, countries where human and nonhuman primate ZIKV infection have both been previously detected are shown in brown, locations in which nonhuman primate ZIKV infection was detected in the present study are shown with red dots, and locations in which nonhuman primate ZIKV infection was not detected in the present study are shown with yellow dots. A review of previous serosurvey studies can be found in reference .