Year: 2001
Type: letter
Abstract: Platt et al. (2000) reported that the sera drawn from 24% of 63 bats captured in Costa Rica and Ecuador neutralized different serotypes of dengue virus (DEN) at titers ranging from 1:20 to 1:80. From this they implied that, along with humans, bats may be involved in the natural transmission of DEN. It is important that the proposition of bat involvement in DEN transmission be thoroughly investigated, because if correct, it is a departure from the well-established and longstanding understanding of the transmission of this medically important arbovirus (Monath 1994, Gubler 1997). Below, I discuss four concerns I have with the publication by Platt et al. (2000). I am not arguing that involvement of bats in DEN transmission is an impossibility; rather, my main points are that the report is incomplete and the data may have been misinterpreted. First, the lack of testing for cross-reactivity to heterologous flaviviruses may have resulted in reporting false positive results. Interpretation of flavivirus serology is not straightforward. Neutralizing antibodies can cross-react with flaviviruses within and among antigenic complexes (Innis 1997). It is, therefore, critical that the bat sera be tested for cross-reaction to heterologous bat-infecting flaviviruses including, but not limited to, Rio Bravo, Montana myotis, and Saint Louis encephalitis (SLE) viruses. The large percentage of positive sera with low titers is consistent with the possibility that the bats were infected with a flavivirus other than DEN. The lack of cross-reaction to yellow fever virus (YF), the absence of SLE or DEN isolates from an undefined number and species of mosquitoes, and the supposition that a heterologous response would be broadly reactive constitute incomplete analyses and indirect arguments that are not substitutes for carrying out the appropriate cross-neutralization tests. Second, the blood-feeding behavior of Aedes aegypti (L.) is not consistent with bat involvement in DEN transmission. The authors imply that bats roosting in houses are bitten by Ae. aegypti, which contract and transmit DEN infections; but no evidence was provided that this occurs. Analyses of blood meals in Ae. aegypti collected weekly for 1–2 yr in and around houses in Thailand and Puerto Rico, respectively, revealed that females fed almost exclusively and preferentially on humans (≥96%, Scott et al. 2000). Among the >3,000 engorged Ae. aegypti assayed by an ELISA, no specimens contained fresh red, nonreacting blood meals that might have been taken from an insectivore. Laboratory and field studies indicate that Ae. aegypti obtain a significant fitness advantage when they restrict their diet to human blood (reviewed in Harrington et al. 2001). Demonstrating that Ae. aegypti will feed on bats in the laboratory is not a substitute for demonstrating that bat-Ae. aegypti interactions are common under natural circumstances. Ae. aegypti were previously shown to feed readily on bats in the laboratory (Gordon 1922, Kumm 1932), as well as on a variety of other vertebrates (Christophers 1960) from which they seldom take blood in the field. Third, information necessary to demonstrate bat involvement in DEN transmission and alternative explanations for results were not fully considered. To demonstrate bat involvement in DEN transmission, it will be necessary to definitively demonstrate that bats have specific antibody reactions to DEN, that bats can be infected from Ae. aegypti, and that viremic bats are infective to Ae. aegypti. Results from field studies will need to show that Ae. aegypti take bat blood meals frequently enough to be a meaningful component in natural DEN transmission. The possibility should be ruled out that bats are infected orally by eating infected mosquitoes and do not contribute to virus transmission. Kumm (1932) examined the role of two species of South American bats and Ae. aegypti in the transmission of another flavivirus, YF. He reported "that bats would be of importance as reservoirs of YF in nature only if they could be infected by Ae. aegypti and if normal Ae. aegypti could later obtain the virus from them. This did not happen." Fourth, contrary to the statement that "little is known about the maintenance of DEN in populated areas during interepidemic periods" it is well established that human infections, virus transmission, and Ae. aegytpi development continue year round in places where DEN is endemic (Halstead et al. 1969, Rigau-Perez et al. 1994, Scott et al. 2000). The ecologies of DEN and Ae. aegypti argue against the notion that virus infections in bats are needed to maintain DEN during interepidemic periods (Focks et al. 1995). Taken together, the concerns reviewed above indicate that publication of the bat serologic results and the conclusion that bats may be involved in DEN transmission were premature. More substantive supporting data and consideration of existing published reports are needed. I thank D. M. Watts, J. D. Edman, L. C. Harrington, and S. B. Halstead for discussion regarding this topic and for reviewing the manuscript. Publication of this manuscript was supported by National Institutes of Health grants (AI)-42332 and (AI)-22119.
Source: Journal of Medical Entomology
Author: Thomas W. Scott
Institution: University of California, Davis
Language: English
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
Estimated APC paid: $4,328
OpenAlex ID: W2177977947