Prior to 1999, West Nile virus (WNV) was a bit player in the screenplay of global vector-borne viral diseases. First discovered in the West Nile District of Uganda in 1937, this Culex sp.-transmitted virus was known for causing small human febrile outbreaks in Africa and the Middle East.
Prior to 1995, the last major human WNV outbreak was in the 1950s in Israel. The epidemiology and ecology of WNV began to change in the mid-1990s when an epidemic of human encephalitis occurred in Romania. The introduction of WNV into Eastern Europe was readily explained by bird migration between Africa and Europe.
The movement of WNV from Africa to Europe could not, however, predict its surprising jump across the Atlantic Ocean to New York City and the surrounding areas of the United States. This movement of WNV from the Eastern to Western Hemisphere in 1999, and its subsequent dissemination throughout two continents in less than ten years is widely recognized as one of the most significant events in arbovirology during the last two centuries. This paper documents the early events of the introduction into and the spread of WNV in the Western Hemisphere.
There are over 500 registered arthropod-borne viruses (arboviruses). Arboviruses are composed of virus members of the Flaviviridae (of which West Nile virus, WNV, is one), Togaviridae, Bunyaviridae, Rhabdoviridae, Reoviridae, and Orthomyxoviridae families. While some of these viruses (e.g., dengue virus, DENV) have global distribution, many have distinct geographic ranges. For example, yellow fever virus (YFV) is essentially a virus of equatorial Africa and Central and South America, even though the YFV mosquito vector can be found throughout the world.
The epidemic encephalitic alphaviruses — eastern, western, and Venezuelan equine encephalitis viruses (EEEV, WEEV, and VEEV) — are essentially Western Hemisphere viruses. The arthrogenic alphaviruses are limited to the Eastern Hemisphere and Australia. Much of these geographic limitations are due to the complex arboviral life cycles that require specific reservoir- and amplifying-hosts, endemic-, epidemic-, and bridge-vectors, represented in unique ecological habitats.
Epidemic arboviruses in the United States (U.S.) during the late 1900s were relatively non-existent. The last major human epidemic of St. Louis encephalitis virus (SLEV) occurred in the mid to late 1970s, although smaller outbreaks of SLEV continued to occur periodically throughout the U.S. after that. SLEV and WNV are members of the Japanese encephalitis virus (JEV) serocomplex and all three viruses are very closely related both genetically and antigenically.
While EEEV caused occasional infections in horses, emus, and humans, these outbreaks were very small and limited to areas where the ecosystem supported the EEEV transmission cycle. Major epidemics- and even smaller outbreaks- of WEEV had stopped altogether. The most important cause of mosquito-borne viral encephalitis in humans in the USA at this time was LaCrosse virus (LACV).
The geographic range of this bunyavirus in the U.S. included the mid-Atlantic states, upper Midwest, and Southwest Louisiana. The LACV is transmitted by the mosquito Ae. triseriatus, so its distribution was limited to well-defined geographic niches containing hardwood forests that supported mosquito breeding in treeholes. Because of this dearth of arboviral activity in the U.S., both federal and state laboratories responsible for arboviral diagnosis, prevention, and control struggled to remain in existence.
A variety of arboviruses cause human encephalitis. Diagnosis of human infection is not easy, and is based upon: the presence or absence of antiviral IgM (as measured in IgM antibody capture-ELISA, MAC-ELISA) in acute-phase serum or cerebrospinal fluid (CSF) specimens; a virus-specific four-fold or higher IgG titer rise or fall from acute- to convalescent-phase paired serum specimens; and/or direct demonstration of infectious virus, viral antigen, or viral RNA in serum, CSF, or tissue specimens. Specimens must be obtained from an individual with symptoms clinically compatible with encephalitis. Serodiagnosis of human flaviviral encephalitis is even more problematic because of the close antigenic relationships among all flaviviruses.
Because the flavivirus envelope protein expresses epitopes that are shared by all flaviviruses, an individual infected with SLEV, for example, produces antiviral antibody that cross-reacts with other flaviviruses as distantly related as DENV and YFV. Quite significant serological cross-reactions are observed among virus members of the same serocomplex (e.g., SLEV, WNV and JEV). Differentiation of human SLEV, WNV, and JEV infections require use of more sophisticated virus-specific serological tests such as the plaque-reduction neutralization test (PRNT).
Of course these infections are more easily diagnosed if virus or pieces of virus are available for analysis. Flaviviral antigens can be differentiated in a simple immunofluorescence assay using virus-infected cells and virus-specific monoclonal antibodies (MAbs). Viral nucleic acid can be positively identified in polymerase chain reaction (PCR) assays using primers that amplify RNA from only one virus.