West Nile Virus Infection of Birds, Mexico

West Nile virus (WNV) has caused disease in humans, equids, and birds at lower frequency in Mexico than in the United States.We hypothesized that the seemingly reduced virulence in Mexico was caused by attenuation of the Tabasco strain from southeastern Mexico, resulting in lower viremia than that caused by the Tecate strain from the more northern location of Baja California.During 2006-2008, we tested this hypothesis in candidate avian amplifying hosts: domestic chickens, rock pigeons, house sparrows, greattailed grackles, and clay-colored thrushes.Only great-tailed grackles and house sparrows were competent amplifying hosts for both strains, and deaths occurred in each species.Tecate strain viremia levels were higher for thrushes.Both strains produced low-level viremia in pigeons and chickens.Our results suggest that certain avian hosts within Mexico are competent for effi cient amplifi cation of both northern and southern WNV strains and that both strains likely contribute to bird deaths.I n Mexico, West Nile virus (WNV; family Flaviviridae, genus Flavivirus) was fi rst isolated in 2003 from a common raven (Corvus corax) carcass in Tabasco (southeast Mexico) (1).According to fi ndings of WNVneutralizing antibodies in horses from the coastal states of eastern Mexico and in resident birds in the Yucatan Peninsula, the virus had spread to Mexico at least 1 year earlier (1-3).In the United States and Canada, morbidity and mortality rates for WNV infection are high among humans, horses, and birds; but in Mexico and other regions of Latin America, the health effects of this virus remain unknown (4).Low numbers of cases in humans, equids, and birds in Mexico have been reported, primarily from the northern border with the United States, where isolated WNV strains (e.g., Tecate) were genetically related to the North American 2002 strain circulating in the southwestern United States (5).The paucity of reported WNV cases in Mexico might be the result of multiple factors involved in local virus ecology.The interactions of amplifying hosts, vectors, and virus strains in Mexico, combined with external factors such as climate, habitat, and circulation of interfering fl aviviruses, may result in relatively low levels of transmission and disease.Virus-host interactions in Mexico, including susceptibility and competence of candidate amplifying hosts, remain unknown.Assessment of the response of various avian species to WNV infection could elucidate aspects of the transmission ecology in tropical ecosystems and provide insight for potential surveillance strategies.To address knowledge gaps regarding transmission and to investigate whether the apparently low prevalence of WNV disease in Mexico could result from reduced virulence of WNV strains from Mexico, during 2006-2008 we experimentally infected birds.