Animals in the Zika Virus Life Cycle: What to Expect from Megadiverse Latin American Countries

Zika virus (ZIKV) was first isolated in 1947 in primates in Uganda, West Africa. The virus remained confined to the equatorial regions of Africa and Asia, cycling between infecting monkeys, arboreal mosquitoes, and occasionally humans. The ZIKV Asiatic strain was probably introduced into Brazil in or around late 2013. Presently, ZIKV is in contact with the rich biodiversity in all Brazilian biomes, bordering on other Latin American countries. Infections in Brazilian primates have been reported recently, but the overall impact of this virus on wildlife in the Americas is still unknown. The current epidemic in the Americas requires knowledge on the role of mammals, especially nonhuman primates (NHPs), in ZIKV transmission to humans. The article discusses the available data on ZIKV in host animals and issues of biodiversity, rapid environmental change, and impact on human health in megadiverse Latin American countries. The authors reviewed scientific articles and recent news stories on ZIKV in animals, showing that 47 animal species from three orders (mammals, reptiles, and birds) have been investigated for the potential to establish a sylvatic cycle. The review aims to contribute to epidemiological studies and the knowledge on the natural history of ZIKV. The article concludes with questions that require urgent attention in epidemiological studies involving wildlife in order to understand their role as ZIKV hosts and to effectively control the epidemic.


Animals as ZIKV hosts
Few studies have focused on the role of animals as hosts for ZIKV. Some authors claim that there is no solid evidence of wild mammals, such as nonhuman primates (NHPs), as reservoirs for ZIKV. Meanwhile, studies have reported ZIKV antibodies in livestock like goats and sheep, rodents [22], and lions and ungulates like Artiodactyla, Perissodactyla, and Proboscidea [23]. In 1971, ZIKV antibodies were detected in primates from the Cercopithecidae family in Nigeria [24]. Several studies suggest that DENV, CHIKV, and ZIKV adapted from an ancestral enzootic transmission cycle involving NHPs and a broad spectrum of species from genus Aedes (Stegomyia, aegypti) as vectors in an urban/peri-urban cycle [25].
ZIKV infection has also been identified in other naturally and experimentally susceptible animal species (Table 1 and Fig 1). Sera from 172 domestic animals and 157 wild rodents were tested for ZIKV in Pakistan, showing that sheep, goats, some rodent species, and one human living in the same area tested positive for ZIKV antibodies [22].
A study in Kenya in 1977 focused on the potential role of livestock (goats, sheep, and cattle) and wild vertebrates (2,424 small mammals, 1,202 birds, 18 reptiles) in maintaining arbovirus transmission. Hemagglutination inhibition assays showed that domestic animals (0.4%), wild birds (0.4%), small wild mammals (5.9%), and reptiles (27.7%) tested positive for ZIKV [29]. Serologic studies should be interpreted carefully in view of possible cross-reactions with other antigenic flavivirus, despite studies suggesting that plaque reduction neutralization test  [26,27], and the list of animal species is described in Table 1. doi:10.1371/journal.pntd.0005073.g001 (PRNT) does not cross react and is the most specific serological test for the proper serological identification of flaviviruses [33,34,35].
Regarding ZIKV infection of sylvatic animals, the presence of positive animals for antibodies does not necessarily means that they are viremic, and they may not be able to transmit the virus to a mosquito, but more studies are required to properly address this issue [36]. In the case of the sylvatic cycle of YFV (also a flavivirus) in the Americas, when monkeys become infected, they present overt clinical signs and a viremia high enough to transmit virus to the mosquito vectors [37].
Unlike humans, wild mammals with ZIKV infection display few clinical signs. In a sentinel study in Uganda in 1947, primates showed only mild pyrexia. All monkeys inoculated by different routes developed neutralizing antibodies by day 14 after inoculation [5]. In the same study, Swiss mice became ill and one animal died following intracerebral inoculation [9]. Such inoculation is not a natural transmission route, and authors point out that some species of wild and laboratory rodents are resistant to some flavivirus infections due to innate genetic resistance [38].
Most primates identified as ZIKV-positive in the wild or in sentinel studies are from Old World species. Phylogenetic analysis shows that humans are more closely related to Old World primate species, especially chimpanzees and orangutans [39]. Diseases that can be transmitted between closely related species often increase the relative risk [40,41]. NHPs thus deserve special attention because of their close relatedness to humans and potential disease exchange [42].
Favoretto et. al. [21], using real-time PCR, showed that 29% (7/24) of the New World primates, Callithrix jacchus and Sapajus libidinosus, in Ceará State in Northeast Brazil were infected with ZIKV. They also showed that the ZIKV genome sequence from monkeys was 100% similar to the ZIKV circulating in humans in South America, suggesting that primates sharing the habitat with humans could act as ZIKV hosts, as in the YFV sylvatic cycle in Brazil.
Besides the use of primates as sentinels in ZIKV studies, some experimental work has been performed with other mammals. Cotton-rats, guinea pigs, and rabbits showed no clinical signs of infection after intracerebral inoculation [9]. An experiment in 1955 aimed to determine the susceptibility of cave bats to ZIKV and showed that these bats are susceptible to ZIKV by intraperitoneal, intradermal, intracerebral, and intrarectal exposure, but not by intranasal exposure [28]. Barr et al. [43] infected cell cultures from different animal species with ZIKV and showed that 17 were susceptible to the virus, developing a cytopathic effect seven days post infection. Some of the cell cultures were from domestic animals and others from Old World wild primates, while nine were from wild animals species found in the Americas: Tabarida brasiliensis, Sylvilagus floridanus, Urocyon cinerorgeneus, Odocoileus hemionus, Procyon lotor, Didelphis virginiana, Dasypus novemcinctus, Marmota monax, and Neovison vison. Most of these animals are peri-domestic and sympatric to mosquito vectors. The authors also argued that with sufficiently high viremia, these animals could serve as hosts. However, they also indicated that the virus strain used in the experiment lacks some characteristics of the ZIKV currently circulating in the field, and that the virus in the laboratory does not mirror natural infection.
Public policy and elimination efforts in the Americas are based mainly on vector control and personal protection measures, so the high number of wild species with the potential to establish a sylvatic cycle would make elimination extremely difficult, if not impossible [18]. We thus need studies on ZIKV in wild and domestic animals in the Americas, both to understand their potential role as hosts in the natural cycle and to target surveillance for enzootic ZIKV transmission.

Biodiversity, animal hosts, and diseases
Human health relates closely to environmental health, defined here as the relationship between the health of domestic animals, wildlife, and the environment. Most etiological agents (60.3%) circulate between animals and humans, and 71.8% of emerging diseases are caused by pathogens originating in wildlife [44]. A recent study associated 2,107 etiological agents with diseases in humans and animals [45].
Recent efforts by the Convention on Biological Diversity and the World Health Organization have addressed scientific and political discussions on the relationship between human health and biodiversity. Such relationships include global concern over the importance of emerging zoonotic diseases originating in wildlife. Environmental changes, including loss of biodiversity, can favor emerging diseases originating from wildlife and act as the source of selective forces in new genetic variations, leading to spillover and infecting humans [46]. This justifies actions to improve knowledge on biodiversity and pathogens and to monitor them to anticipate problems.
The current ZIKV epidemic in Brazil requires understanding of the role of mammals, especially primates, in viral transmission to humans, especially when this interface occurs in fragmented forest areas, as described by Favoretto et al. [21]. Such areas are usually bordered or surrounded by farmland and human settlements and by dense urban and unstructured areas that can increase contact between humans, wildlife, and domestic animals and occasionally promote disease spillover [47,48]. Wild animals, especially primates, can thus be considered sentinels for pathogens of human health concern [48,49]. ZIKV is an example of spillover, because this virus adapted from an ancestral transmission cycle involving NHPs to an urban/ peri-urban cycle, with humans as the main host.
Brazil is a megadiverse country with 357 million hectares of tropical forest and other highly biodiverse biomes [50]. Not surprisingly, Brazil has more primate species than any other country. Its 53 species account for 27% of the world's primates [51].
Some NHP species occupy urban forests due to habitat fragmentation and have close contact with humans and domestic animals. Examples include primates from the Callitrichinae (Callithrix, Leontopithecus, and Saguinus), Cebinae (Cebus), and Atelidae families (Alouatta and Brachyteles) [52]. Favoretto et al. [21] were the first to report ZIKV in NHPs in Northeast Brazil, highlighting that these New World primates can act as potential ZIKV hosts in the Americas. Many questions remain unanswered. Does ZIKV impact the health of NHPs? Are NHPs living in urban fragments of forest more prone to ZIKV infection than those in preserved areas? Can naturally infected neotropical primates transmit ZIKV to mosquito vectors and thus help keep the virus circulating in the Americas? Barr et al. [43] demonstrated the feasibility of infection in cell cultures from other mammalian species like carnivores, armadillos, rodents, and bats, thus raising the possibility of a transmission network shaped by biological and ecological factors. These factors include vector and host density and behavior, virulence, viral load, immunity, genetic variation, climate change, competition between biological communities, and anthropogenic forces like urbanization, sanitation, limited access to health services, poverty, and mistreatment of animals [38].
Considering the current epidemiological scenario with simultaneous circulation of the arboviruses ZIKV, DENV, and CHIKV and the fact that Brazil has a large NHP population, there is an urgent need to answer these questions to evaluate the impact of diseases like Zika on the NHP population in Brazil and elsewhere in the Americas. YFV, another flavivirus that circulates in a sylvatic cycle in the Americas, has a great impact on primate populations, especially those of genus Alouatta [53], which exhibit disease signs after infection and act as sentinel primates for viral circulation and for implementation of control measures like human vaccination campaigns.
The pandemic ZIKV strain differs significantly from the African strain mainly in two regions of the genome. These acquired genetic markers increase its fitness for replication in the human host [4]. Whether these mutations also alter the infectivity in NHPs remains to be determined. The role of wild primates and other mammals in ZIKV epidemiology thus requires urgent investigation.
The complex epidemiological panorama currently experienced in several countries of South America, with the co-circulation of three arbovirus, ZIKV, DENV, and CHIKV, of high impact on public health, highlights the importance of a robust epidemiological surveillance. During 2014, two strains of CHIKV were introduced in Brazil: the Asian and the African (East/Central/South Africa [ECSA]) strains, both transmitted by Ae. aegypti.
As seen during the 2005 CHIKV outbreak in La Réunion Island, where the predominant mosquito species was Ae. albopictus, the viruses quickly acquired an E1-A226V mutation, increasing viral fitness to infect Ae. albopictus, which became the principal vector [54]. The Brazilian CHIKV strains analyzed so far did not display mutations that increase CHIKV transmissibility and persistence in Ae. albopictus [55]. However, the elevated density and wide distribution of Ae. albopictus in Latin America warns the risk of the ECSA strain adapt to this vector [56]. Moreover, the abundance of naïve primate (and maybe other small mammals) species and culicids species in South American forests creates the scenario for the establishment of an enzootic cycle, as seen in Africa and Asia, where there is evidence of a sylvatic CHIKV transmission cycle involving NHPs and mosquitoes [57].
Another relevant issue is the development of diagnostic tests for the detection of ZIKV infection in wild mammals, enabling unequivocal results without cross-reactivity with other flavivirus infections.

Final Comments and Research Perspectives
Despite the growth of epidemiological knowledge in the last century, health interventions still mainly react to emergency events involving specific diseases in the human population, with some mitigation efforts [46]. The current ZIKV epidemic is no exception. We cannot expect to completely block the emergence of diseases, considering vector spread due to our limited capacity to reverse climate change, the globalization of goods and people, and our mode of production and consumption of natural resources. This situation is particularly paradoxical in megadiverse countries like Brazil.
The driving forces in the spread of diseases apply to the ZIKV epidemic, including anthropogenic activities, climatic change, intense human movement, loss of biodiversity, habitat destruction, land use change, introduction of invasive species, urban development, lack of knowledge on the role of animals in maintaining the sylvatic cycle, clinical manifestations, and wildlife trafficking [46].
We need to understand the diversity of pathogens in nature and correlate them with biological communities, pathogenic and genetic characteristics, and anthropic impacts in areas where disease transmission occurs. DENV is a good example of how a combination of environmental changes, genetic characteristics, and human mobility propels the spread of viruses in Brazil. A new lineage of DENV entered in the country through Caribbean through the northern/northeast and spread rapidly to the rest of Brazil, especially through the aerial transportation of humans and/or mosquito vectors [58]. In parallel, this example allows us to suggest that the spread of ZIKV to other biomes in the Americas and outside Brazil may also be related to these factors, and that these should be highlighted.
The ZIKV epidemic illustrates the importance of monitoring and predicting the pathogens arising from wild animals and biodiversity. Based on the above and the results of other studies, we pose several questions and hypotheses that emerge from this discussion and that require investigation: 1. What other wild animals besides primates could be infected by ZIKV in Americas? What is their role in maintaining and transmitting the virus to mosquito vectors? Which species can act as hosts?
2. Does the virus circulate at higher levels in wild animals inhabiting forest fragments adjacent to urban areas? What role do these animals play in maintaining the virus in areas close to humans?
3. Which wild hosts help keep the virus circulating in the Americas?
4. Do neotropical primates play a special role in the ZIKV epidemic?
5. Does ZIKV impact wild animal populations and biodiversity? Does it cause disease and mortality in these animals?
Infectious diseases have important implications for animal and human health and biodiversity. Public health and biodiversity needs are misaligned and need to be rebalanced.
Rather than merely attacking and solving epidemic situations, as in the current ZIKV global health emergency, we need to predict and prevent future emerging diseases. Studies of wild hosts are troublesome and costly, especially when they require long-term monitoring. Funding also needs to be targeted for these studies. Future laboratory, field, and eco-epidemiological research should focus on wildlife hosts to elucidate their role in ZIKV epidemiology in the Americas and enhance the epidemic's control.

Key Learning Points
• A remarkable diversity of wildlife species involving three orders, namely mammals, reptiles, and birds, totaling 47 different species, were investigated for their potential to establish a sylvatic ZIKV cycle.
• A study showed that New World primates in Northeast Brazil were infected and that the ZIKV genome sequence in monkeys was 100% similar to the ZIKV circulating in humans in South America. Studies are thus needed on ZIKV in wild and domestic animals in the Americas, both to understand their potential role as hosts in the natural cycle and to target surveillance for enzootic ZIKV transmission.
• Environmental changes, including loss of biodiversity, can favor emerging diseases originating from wildlife and act as selective forces in new genetic variations leading to spillover and infection in humans. This justifies actions to improve knowledge on biodiversity and pathogens and to monitor them to anticipate problems.
• Driving forces in the spread of diseases apply to the ZIKV epidemic, including anthropogenic activities, climatic change, intense human movement, loss of biodiversity, habitat destruction, land use change, introduction of invasive species, urban development, lack of knowledge on the role of animals in maintaining the sylvatic cycle, clinical manifestations, and wildlife trafficking.
• Future laboratory, field, and epidemiological research should focus on wildlife hosts to elucidate their role in ZIKV epidemiology in the Americas and enhance the epidemic's control.