387 Am. J. Trop. Med. Hyg., 60(3), 1999, pp. 387–391 Copyright  1999 by The American Society of Tropical Medicine and Hygiene EASTERN EQUINE ENCEPHALITIS VIRUS IN BIRDS: RELATIVE COMPETENCE OF EUROPEAN STARLINGS (STURNUS VULGARIS) NICHOLAS KOMAR, DAVID J. DOHM, MICHAEL J. TURELL, AND ANDREW SPIELMAN Harvard School of Public Health, Boston, Massachusetts; United States Army Medical Research Institute of Infectious Diseases, Fort Detrick, Frederick, Maryland Abstract. To determine whether eastern equine encephalitis (EEE) virus infection in starlings may be more ful- minant than in various native candidate reservoir birds, we compared their respective intensities and durations of viremia. Viremias are more intense and longer lasting in starlings than in robins and other birds. Starlings frequently die as their viremia begins to wane; other birds generally survive. Various Aedes as well as Culiseta melanura mosquitoes can acquire EEE viral infection from infected starlings under laboratory conditions. The reservoir com- petence of a bird is described as the product of infectiousness (proportion of feeding mosquitoes that become infected) and the duration of infectious viremia. Although starlings are not originally native where EEE is enzootic, a starling can infect about three times as many mosquitoes as can a robin. Precise knowledge of the identity of the avian reservoir of eastern equine encephalitis (EEE) virus may facilitate ef- forts to protect the public health. Serologic evidence sug- gests that numerous species of birds are exposed to this vi- rus, particularly those residing near swamps in which the Culiseta melanura vector mosquito breeds. 1–3 This virus has been isolated from a similarly broad array of hosts. Indeed, the virus replicates in virtually all wild birds that have been infected experimentally. 4–10 Diverse species of birds appear to contribute to the natural transmission cycle of EEE virus. The intensity of transmission of EEE virus depends in part on the vertebrate reservoir competence (ability to sustain infection and present infectious agents to vectors) of local birds for this agent. Candidate reservoir birds would be those that roost abundantly in or near swamps during the late sum- mer season of virus amplification. 11 Species that roost com- munally are especially strong candidate reservoirs because of their tendency to return nightly to stable roost sites. 12 In Massachusetts, these birds mainly include European starlings (Sturnus vulgaris), common grackles (Quiscalus quiscula), red-winged blackbirds (Agelaius phoeniceus), and American robins (Turdus migratorius). Starlings were introduced from Europe in 1890 into New York City 13 and rapidly expanded their range throughout North America. These gregarious birds became abundant in the northeastern United States in the 1920s, shortly before the first recognized epidemic of human EEE in 1938. 14 Reservoir competence of starlings has not previously been evaluated. Because starlings are not native to North America, EEE virus infection may be more fulminant in these birds than in other birds, a condition that may modify reservoir compe- tence. Accordingly, we compared the intensity and duration of viremia of EEE virus in starlings to those in such other likely candidate reservoirs as robins, grackles, and red-wing blackbirds. MATERIALS AND METHODS Source of virus. The EEE virus strain MA92-1406 was provided by Dr. Barbara Werner (Massachusetts Department of Public Health, Jamaica Plain, MA). This virus originated from a 1992 pool of Cs. melanura collected in Massachu- setts, and was passed once in Vero cells prior to use. The EEE virus strain FL91-4679 was provided by Dr. Carl Mitchell (Division of Vector-Borne Infectious Diseases, Centers for Disease Control and Prevention, Fort Collins, CO). This virus originated from a 1991 pool of Aedes al- bopictus collected in Florida and was passed once in Vero cells prior to use. The EEE virus strain MA93-SF52 origi- nated from a 1993 pool of Cs. melanura collected in Mas- sachusetts and was passed once in a Japanese quail (Coturnix coturnix). No differences were noted among the strains in their ability to infect birds or in the duration or intensity of the resulting viremias. Source of birds. Birds were captured in Massachusetts by means of mist nets 15 or baited funnel traps. Bird species cap- tured included starling, grackle, red-winged blackbird, robin, swamp sparrow (Melospiza georgiana), song sparrow (Me- lospiza melodia), brown-headed cowbird (Molothrus ater), northern cardinal (Cardinalis cardinalis), mourning dove (Zenaida macroura), and domestic pigeon (Columba livia). This activity was authorized by U.S. Fish and Wildlife Ser- vice Scientific Collecting Permit PRT-719506 and Massa- chusetts Division of Fisheries and Wildlife Scientific Col- lecting Permit for Birds #121.94SCB, #167.95SCB, and #118.96SCB. Captive birds were maintained in stainless steel monkey cages or guinea pig cages with food and water provided ad libitum. Passeriform birds were fed moistened Sportsmen’s Pride Kennel’s Choice dog food containing 21% protein (Sunshine Mills, Inc., Red Bay, AL) and mixed seeds, supplemented with mealworms. Columbiform birds were fed mixed seeds. Some passerine birds were held tem- porarily (up to two months) in a flight room (2  4 m) prior to placement within standard cages. The maintenance and care of experimental animals complied with the National In- stitutes of Health guidelines for the humane use of labora- tory animals. Source of mosquitoes. The Yale strain of Cs. melanura, 16 provided by Dr. John Edman (University of Massachusetts, Amherst, MA), was used to infect starlings. Our own strain was colonized in 1995 from about 200 blood-engorged fe- males collected from Kingston, MA. They were permitted to oviposit on distilled water in standard 30  30  30 cm 3 mosquito cages (BioQuip, Inc., Gardena, CA) at 22–24°C and 70% relative humidity. Larvae were reared in polysty- rene mouse cages containing distilled water, with daily food supplement comprised of two parts liver powder, one part