Serologic evidence of West Nile virus transmission, Jamaica, West Indies - Dispatches

Emerging Infectious Diseases, July, 2003 by Alan P. Dupuis, II, Peter P. Marra, Laura D. Kramer

In spring 2002, an intensive avian serosurvey was initiated in Jamaica, Puerto Rico, and Mexico. We collected >1,600 specimens from resident and nonresident neotropicai migratory birds before their northerly migrations. Plaque reduction neutralization test results indicated specific neutralizing antibodies to West Nile virus in 11 resident species from Jamaica.

**********

West Nile virus (WNV) is maintained in nature between birds and Culex species mosquitoes (1,2). Unlike other viruses maintained in bird and mosquito transmission cycles (for example, St. Louis encephalitis, western equine encephalomyelitis, and eastern equine encephalomyelitis), the WNV strain responsible for the current epizootic in the Western Hemisphere is associated with a high number of avian deaths (3). Although crows and other corvids appear to be especially susceptible to disease (4), WNV has been documented in >190 bird species, including neotropical migratory species and exotic zoo specimens (5).

Birds have been implicated in spreading WNV during migratory events in Europe, Asia, Africa, and the Middle East (6-9). Because of the apparent ease of infecting a multitude of avian hosts, WNV can potentially be introduced during annual migratory events in the Western Hemisphere. Predicting the incursion of WNV into the tropics is complicated by our incomplete knowledge regarding geographic connectivity in populations of migratory birds between winter and summer. We do not know where most species of migratory birds from North America spend their winter, where wintering birds spend their summer, or the routes they use while in transit. Evidence from mark-and-recapture efforts suggests that birds from the northeastern United States tend to winter in the southeastern United States and Greater Antilles (e.g., Puerto Rico, Jamaica) whereas birds from the western United States migrate to Mexico and Central America (10-12). In response to the potential introduction of WNV in tropical America during the fall migrations, we established a network of monitoring sites on the overwintering grounds of neotropical migratory birds in Jamaica, Puerto Rico, and Mexico.

The Study

The primary goal of the monitoring system was to obtain a large number of blood specimens from birds belonging to as many species as possible across the Caribbean. At seven study sites in Jamaica, Puerto Rico, and the Yucatan Peninsula of Mexico, we erected 15 mist nets (12-m) daily for 3 to 4 weeks during January through March 2002. We collected 1,619 avian blood specimens, which represented 98 species, 25 families, and eight orders. In Jamaica, 542 samples were collected from Westmoreland, Manchester, and St. Catherine Parishes; 649 samples were collected at Roosevelt Roads Naval Station in Puerto Rico; and 430 samples were collected from the states of Yucatan and Campeche in Mexico (Table 1). At the time of capture, all migratory birds were banded with an aluminum U.S. Fish and Wildlife Service band and 3-5 breast feathers were removed for isotope analysis. Resident birds had outer rectrices cut. All birds were evaluated for age and gender if possible, bled with microcapillary tubes from the brachial vein, and released. Blood was added immediately to BA-1 medium, consisting of M199 medium with Hank's salts, 1% bovine albumin, TRIS base (tris [hydroxymethyl] aminomethane), sodium bicarbonate, 20% fetal bovine serum (FBS), and antibiotics. The samples were placed in a cooler on ice packs until storage in a -20[degrees]C freezer. Samples were sent on dry ice or hand-carried on ice packs to the Arbovirus Laboratories, Wadsworth Center, New York State Department of Health for serologic analysis and virus isolation attempts.

Specimens were screened at 1:100 for antibodies against flaviviruses by using an indirect enzyme-linked immunosorbent assay (ELISA) (13). Samples with a P/N ratio >2.0 were tested further by a plaque reduction neutralization test (PRNT) for Ilheus virus (ILHV), St. Louis encephalitis virus (SLEV), and WNV, as described (14). The particular virus strains used for the PRNTs were ILHV (original), SLEV 59268 Parton, and WNV (3100365), an isolate from a pool of Culex sp. mosquitoes collected in Staten Island, New York. The indirect ELISA was chosen to screen the samples in order to take advantage of its ability to detect antibodies against a wide range of flaviviruses. PRNT was used as a confirmatory assay to differentiate among recognized flaviviruses (15,16). Briefly, serial dilutions of test samples were mixed with an equal amount of virus suspension containing 200 PFU/0.1 mL and incubated at 37[degrees]C for 1 h. We then added 0.05 mL of each virus-diluted blood sample onto 1 well of a 12-well tissue culture plate containing confuent monolayers of African green monkey kidney cells (Vero). The plate was incubated for 1 h at 37[degrees]C, after which an agarose overlay was added and incubation was continued. When virus plaques became visible, we added a second overlay containing neutral red and counted plaques. The antibody titer reported is the dilution of serum that inhibited 90% of the test virus inoculum. For virus isolation attempts using confluent Vero cell monolayers, 0.1 mL of each serum specimen was added onto one well of a six-well tissue culture plate, incubated for 1 h, rinsed with phosphate-buffered saline, and then the cells were refed with minimum essential medium containing 2% FBS. Cells were monitored once a day for 5 days for cytopathic effect. Cell cultures showing any abnormal cell morphology were then blind passaged after 5 days.

ELISA results indicated 34 of 1,413 serum specimens tested from the three study sites contained immunoglobulin G antibody against a flavivirus. Of the 34 reactive samples representing 20 bird species, 27 were collected in Jamaica (26 residents, 1 migrant), 5 in Puerto Rico (3 residents, 2 migrants), and 2 in Mexico (1 resident, 1 migrant). PRNTs on the Jamaican bird samples indicated 18 WNV infections, 3 SLEV infections, 5 undetermined flavivirus infections (positive results for two or more viruses without a fourfold difference in antibody titer); one additional reactive serum sample was negative for the three viruses tested. Results on the serum samples collected in Puerto Rico indicated one WNV infection in a migratory bird and one SLEV infection in a resident bird; three additional reactive serum samples were negative for all three viruses tested. In Mexico, we found evidence of WNV infection in one migrant bird and SLEV infection in one resident bird (Table 2). Virus isolation attempts were negative for all 1,603 specimens tested (16 samples destroyed). Negative isolation results were not entirely unexpected, considering that birds are viremic for a short period of time (17) and maintaining a proper cold chain (i.e., temperatures) to preserve virus is difficult when working in the tropics.