Zika: the origin and spread of a mosquito-borne virus

1Zika, a flavivirus transmitted mainly by mosquitos in the genus Aedes, was discovered in 1947 in Uganda. From the 1960s to 1980s, human infections were found across Africa and Asia, typically accompanied by mild illness. The first large outbreak of disease caused by Zika infection was reported from the island of Yap (Federated States of Micronesia) in 2007, as the virus moved from south-east Asia across the Pacific. During an outbreak in French Polynesia in 2013–2014, Guillain-Barré syndrome was linked to Zika infection and cases of microcephaly in newborn children were also retrospectively linked to this outbreak. The World Health Organization (WHO) received the first reports of locally-transmitted infection from Brazil in May 2015. In July 2015, health ministry officials from Brazil reported an association between Zika virus infection and Guillain-Barré syndrome in adults. In October 2015, WHO received reports from Brazil of microcephaly in babies whose mothers had been exposed to Zika during pregnancy. At this time, there was no proof of a causal link between Zika infection and these neurological complications.

2In February 2016, as infection moved rapidly through the range occupied by Aedes mosquitos in the Americas, WHO declared that Zika infection associated with microcephaly and other neurological disorders constituted a Public Health Emergency of International Concern (PHEIC). By the start of February 2016, local transmission of Zika infection had been reported from more than 20 countries and territories in the Americas and an outbreak numbering thousands of cases was under way in Cabo Verde in western Africa. Beyond the range of its mosquito vectors, Zika virus infections are expected to be carried worldwide by people as they travel and be transmitted by travellers to sexual partners who have not been to places where the virus is endemic.

3To illustrate the spread of Zika virus and associated neurological complications, we did a literature search in PubMed using “Zika” and “ZIKV” as the search terms and cross-checked our findings for completeness against other published reviews., In addition, we drew on formal notifications to WHO under the International Health Regulations (IHR), which are archived in the WHO Event Information Site (EIS). EIS contains information about public health events of potential international concern notified to WHO as required by the IHR. EIS notifications sometimes contain confidential patient information and therefore are not publicly available. Other details of specific events can be provided by the authors on request.

4The first reported case of Zika virus dates to 1947 when the virus was isolated in samples taken from a captive sentinel rhesus monkey by scientists conducting routine surveillance for yellow fever in the Zika forest of Uganda. The virus was recovered from Aedes (Stegomyia) africanus, caught on a tree platform in the forest. Laboratory infection experiments showed the virus to be neurotropic in mice. The timeline presented in this paper includes numerous serological surveys that purportedly detected antibodies to Zika virus in the 1950s and 1960s in Africa and Asia. Because serological (antibody detection) tests for Zika cross-react with antibodies stimulated by other viral infections, the presence of Zika virus is ideally confirmed by the detection of viral nucleic acids by polymerase chain reaction (PCR) testing or by virus isolation. A chronological map of the presence of Zika in those countries for which there is evidence of autochthonous transmission by mosquitos is presented in Fig. 1. The map excludes the many countries from which imported Zika infections have been reported. The country-by-country spread of Zika virus infections, from the earliest published report in 1947 to January 2014 is summarized in Table 1. Reports received by WHO in 2015 and up to 1 February 2016 are summarized in Table 2 (available at: http://www.who.int/bulletin/volumes/94/9/15-171082).

5As shown in Table 1, Zika virus was first isolated from rhesus monkeys. The first human cases were detected in Uganda and the United Republic of Tanzania in 1952 in a study demonstrating the presence of neutralizing antibodies to Zika virus in sera. During the same year, there was confirmation of the presence of the virus in humans in India. Throughout the 1950s, the presence of neutralizing antibodies against Zika virus was identified by serological surveys done in Egypt, Malaysia, Mozambique, Nigeria, the Philippines and Viet Nam.

6Between 1960 and 1999, Zika virus was being detected in mosquitos and sentinel rhesus monkeys used for field research in a narrow band of countries that stretch across equatorial Africa (Table 1). Altogether, the virus was isolated from more than 20 mosquito species, mainly in the genus Aedes. Sporadic human cases were identified, mostly by serological methods, but such cases were rare and the disease was regarded as benign. No deaths or hospitalizations were reported and seroprevalence studies consistently indicated widespread human exposure to the virus.,,– Molecular studies of viruses later mapped the disease as it moved from Uganda to West Africa and Asia in the second half of the 20th century., Between 1969 and 1983, the known geographical distribution of Zika expanded to equatorial Asia, including India, Indonesia, Malaysia and Pakistan, where the virus was detected in mosquitos. As in Africa, sporadic human cases occurred but no outbreaks were detected and infections in humans continued to be regarded as rare, with mild symptoms. Seroprevalence studies in Indonesia, Malaysia and Pakistan indicated widespread population exposure.– Researchers later suggested that the clinical similarity of Zika infection with dengue and chikungunya may be one reason why the disease was so rarely reported in Asia.

7In 2007, Zika virus spread from Africa and Asia to the Pacific island of Yap, in the Federated States of Micronesia, and caused the first large outbreak in humans. Before this event, only 16 cases of human Zika virus disease (including one experimental infection) had been documented worldwide (Table 3). House-to-house surveys done among the island’s population of 7391 people identified 185 cases of suspected Zika virus infections. The sample for the survey was 200 of 1276 total households. Of the 185 suspected cases, 49 (26%) were confirmed by PCR, or a specific neutralizing antibody response to Zika virus in the serum and 59 (32%) were classified as probable (patients with immunoglobulin M (IgM) antibody against Zika virus who had a potentially cross-reactive neutralizing antibody response). An estimated 73% (95% confidence interval: 68–77) of Yap residents older than three years were infected with Zika virus. No deaths, hospitalizations or haemorrhagic complications were reported.–

8Although wind-blown mosquitoes can travel distances of several hundred kilometres over the open ocean, introduction of the virus by travel or trade involving an infected person or an inadvertently imported mosquito is considered the most likely source of this outbreak, especially as no monkeys were present on the island., The outbreak on Yap Island showed that Zika virus could rapidly cause more than a hundred confirmed and probable cases. In the absence of any evidence that viral mutations were the reason for this change in epidemic behaviour, other explanations should be considered. The first is a lack of population immunity. Regular exposure to infection in Africa and Asia may have prevented the large outbreaks eventually seen in Pacific Islands and in the Americas. The second possible explanation is that cases of Zika virus infection were historically misattributed – due to clinical similarities – to other pathogens, including malaria, dengue and chikungunya, and the frequent co-circulation of many infectious agents in these settings. The third possibility is underreporting, due to lack of surveillance systems, the often mild nature of the disease in adults and the many other causes of rare, but serious, neurological complications in adults and in utero.

9Between 2010 and 2015 sporadic cases of Zika virus infection were reported from several countries in south-east Asia.-, These cases included an Australian traveller returning from Indonesia in 2012, a Canadian traveller returning from Thailand in 2013, two German travellers; one returning from Thailand in 2013, and one from Malaysian Borneo in 2014, and a Finnish traveller returning from the Maldives in 2015. We may infer that mosquito-borne transmission of Zika virus was ongoing in the places that these travellers had visited. The presence of infection was also reaffirmed in Africa (Cameroon).

10In 2012, researchers published genetic sequences of Zika virus strains collected in Cambodia, Malaysia, Nigeria, Senegal, Thailand and Uganda, and constructed phylogenetic trees to assess the relationships between them. Two geographically-distinct lineages of the virus, African and Asian, were identified, as well as multiple strains within each lineage. Analysis of viral samples from Yap Island strengthened previous epidemiological evidence that the outbreak on Yap Island originated in south-east Asia.,,, Between 2013 and 2014, the virus caused outbreaks in four other groups of Pacific islands: French Polynesia; Isla de Pascua (Chile); the Cook Islands; and New Caledonia.– The outbreak in French Polynesia generated thousands of suspected infections and was intensely investigated. The results of retrospective investigations were reported to WHO on 24 November 2015 and 27 January 2016. These reports indicated a possible association between Zika virus infection and congenital malformations and severe neurological and autoimmune complications. In particular, an increase in the incidence of Zika infection towards the end of 2013 was followed by a rise in the incidence of Guillain-Barré syndrome., This finding challenged previous assumptions that Zika infection causes only mild illness in humans.,,

11During the 2013–14 outbreak of Zika virus in French Polynesia, two mothers and their newborn babies had Zika virus infection confirmed by PCR done on serum samples collected within four days of birth. Because they were able to show evidence of infection in the first week of life, the researchers concluded that Zika virus had been acquired by transplacental transmission or during delivery. During the same outbreak, 1505 asymptomatic blood donors were reported to be positive for Zika virus by PCR. These findings alerted authorities to the risk of post-transfusion Zika fever. In October 2015, Brazilian health ministry officials reported to WHO the results of a review of 138 clinical records of patients with a neurological syndrome, detected between March and August. Of the 138 cases, 58 (42%) presented neurological syndromes with a previous history of viral infection. Of the 58 cases, 32 (55%) had symptoms consistent with Zika or dengue infection.

12On 1 February 2016 the Director-General of WHO declared that the recent association of Zika infection with clusters of microcephaly and other neurological disorders constituted a PHEIC. Before this declaration, three epidemiologic alerts were issued by the Pan American Health Organization (PAHO) and WHO. The first epidemiological alert had been issued on 7 May 2015. Six months later – 17 November 2015 – PAHO and WHO had issued a second epidemiological alert asking PAHO Member States to report congenital microcephaly and other central nervous system malformations under the International Health Regulations. On 1 December 2015, PAHO and WHO had issued another alert to the association of Zika virus infection with neurological syndrome and congenital malformations in the Americas. The alert included guidelines for laboratory detection of the virus. On 15 January 2016, the United States of America issued interim travel guidance for pregnant women which, “out of an abundance of caution”, advised pregnant women in any trimester to consider postponing travel to areas with ongoing local transmission of the virus, or to take precautions against mosquito bites if they must travel.

13Results of previous studies in which Zika virus was confirmed by serological surveys need to be interpreted with some caution, as they used different serological methods with varying degrees of specificity. These methods include hemagglutination inhibition, complement fixation and enzyme-linked immunosorbent assays, in addition to more specific sero-neutralization tests. The fact that Zika virus cross-reacts immunologically with dengue, other flaviviruses and chikungunya, which frequently co-circulate with Zika and have the same vectors, complicates the interpretation of serological results. Nonetheless, the results of serological surveys, supported by the isolation of Zika virus from mosquitoes and vertebrate hosts, suggest that the virus is endemic in several African and Asian countries.

14Travellers who returned with Zika infections from Malaysia, the Maldives, Senegal and Thailand in 2012, were sentinels for virus circulation that may otherwise have gone undetected. Reasons why the virus may not always be detected include widespread immunological protection in the local population, the asymptomatic or mild and self-limiting nature of most Zika infections, clinical symptoms that overlap with dengue and chikungunya, the weakness of surveillance systems and the lack of specific diagnostic tests in many settings.

15Therefore, additional research is needed to understand whether population immunity in endemic countries will confer protection against the epidemic strains of Zika currently circulating in Latin America and the Caribbean.

16Zika virus infection in humans appears to have changed in character while expanding in geographical range. The change is from an endemic arbovirus causing mild illness across equatorial Africa and Asia. From 2007 onwards, Zika virus caused large outbreaks in previously unexposed populations, and from 2013 onwards, outbreaks linked with neurological disorders including Guillain-Barré syndrome and congenital malformations, for reasons that are not yet known. The future transmission of Zika infection is likely to coincide with the global distribution of Aedes vectors. Person-to-person transmission, both vertically, from mother to fetus, and horizontally through sexual transmission, is also expected to continue, and we anticipate that infections will be carried widely by international travel.