Search

A review of 60 years of research on mosquito-borne arboviruses in the Western U.S.A. revealed a number of instances when serendipity influenced the development of new concepts or novel approaches to solve ecological or epidemiological problems. Eight such events were selected as examples. The need for effective mosquito traps to collect live mosquitoes to be tested for virus infection posed design problems and also led to the use of CO2 (dry ice) as a mosquito attractant. This research also led to identification of Culex tarsalis as a primary target for vector control programs in the western U.S.A. Attendance at a movie led to development of fluorescent dusts to mark mosquitoes for studies of their numbers, life tables and movements. Knowledge of vector-virus associations was used to influence state legislative action to provide funding for vector control and further discovery of vector-virus associations. Derivation of the term "Arbovirus" started as laboratory jargon and evolved into being the classification for over 500 vector-borne viruses. Sociobiological changes resulting from the use of television and air conditioning fortuitously decreased exposure of California residents to vector attack. These two changes were introduced into households in California in the early 1950s. The prime time when C. tarsalis, the primary vector of encephalitis, bites people is around sundown which also is primetime for television watching. These sociological changes are a valuable adjunct to vector control programs.

Nearly 80,000 immature and adult mosquitoes in three genera were collected in high-elevation (> 1,000 m) areas of California (68,229), Nevada (3,721), Oregon (5,918), and Washington (1,629) during 1990-1992 and tested for virus as adult males or females in 1,799 pools. Collections comprised primarily alpine Aedes in the Aedes communis (De Geer) group of the subgenus Ochlerotatus. Thirteen strains of Jamestown Canyon (JC) virus were recovered by plaque assay in Vero cell culture from three members of the Ae. communis group: 10 from Aedes tahoensis Dyar, 2 from Aedes cataphylla Dyar, and 1 from Aedes hexodontus Dyar. All isolates came from collections made in Alpine, Sierra, Tulare, or Tuolumne counties in the Sierra Nevada of California. Vertical transmission of JC virus in all three mosquito species was demonstrated by the isolation of virus from adult males or females reared from field-collected larvae or pupae. The prevalence of infected Ae. tahoensis was significantly higher in field-collected adult females than in reared adult males and females in Alpine County, which indicated that JC virus was being amplified by horizontal transmission. This study further incriminated Ae. tahoensis, Ae. cataphylla, and Ae. hexodontus as natural vectors of JC virus in California and greatly extended the known geographical range of this virus in the Sierra Nevada.

Selected mosquito species from Central Valley, coastal, and alpine habitats of California were evaluated for their vector competence for Northway (NOR) virus. Culiseta incidens Thomson, Culiseta inornata (Williston), and Anopheles freeborni Aitken were the only competent vectors when fed virus. Aedes sierrensis (Ludlow), as well as alpine snow pool Aedes (i.e., Ae. cataphylla Dyar, Ae. hexodontus Dyar, Ae. increpitus Dyar and Ae. tahoensis Dyar), Ae. melanimon Dyar, Ae. washinoi Lanzaro & Eldridge, Culex erythrothorax Dyar, and Cx. tarsalis Coquillett were highly refractory to peroral infection. Alpine snow pool species were poor vectors even following parenteral infection, but 21-35% of intrathoracically inoculated Ae. sierrensis from several lower elevation localities and 39-46% of parenterally infected Cx. tarsalis were able to transmit NOR virus per os. Perorally infected Cs. inornata transmitted NOR virus vertically to their F1 adult progeny at a low rate (minimal filial infection rate, 1:248).

Mosquitoes collected from coastal, inland valley, and alpine locations in California were evaluated for their experimental vector competence for two viruses in the California serogroup (Bunyaviridae:Bunyavirus). Aedes squamiger, a coastal salt marsh mosquito, was an efficient vector of a California encephalitis (CE)-like virus isolated from its habitat (89% of the pledget-fed females became infected and 61% transmitted virus). Aedes dorsalis, a coastal mosquito, and Ae. melanimon, an inland valley mosquito, were competent vectors of prototype CE virus (98% and 100% of the pledget-fed females became infected and 56% and 30%, respectively, transmitted virus). Aedes squamiger and Ae. dorsalis transmitted both viruses vertically to one or more of 20 of their progeny. Culiseta inornata was susceptible to infection with both viruses, but 5% or less transmitted virus perorally. Alpine mosquitoes, Ae. cataphylla, Ae. increpitus, and Ae. tahoensis, became infected with both CE and CE-like viruses, but 3% or less transmitted virus. All species of mosquitoes were more efficient vectors of both viruses following intrathoracic inoculation than following pledget feeding, suggesting the presence of mesenteronal barriers.

The susceptibility to infection with a Panamanian isolate of St. Louis encephalitis (SLE) virus was evaluated in Culex quinquefasciatus, Haemagogus equinus, Mansonia dyari, Sabethes cyaneus and Deinocerites pseudes. When fed on blood-virus suspensions using the hanging drop or pledget technique, the median infective dose (ID50) of SLE virus for colonized strains of Cx. quinquefasciatus and Hg. equinus was 10(2.8) plaque forming units (PFU)/2 microliters and 10(3.5) PFU/2 microliters, respectively. The susceptibility of F1 and F2 Cx. quinquefasciatus females was similar to the colonized strain by the same technique. When fed on viremic chicks, the ID50 for both Cx. quinquefasciatus and Hg. equinus was 10(0.5) PFU/2 microliters. These results and successful transmission trials indicate that both of these species have the potential to serve as vectors of SLE virus in Panama. Because of their poor feeding response, complete susceptibility profiles were not obtained for the other 3 species tested. They appear to be less susceptible than Cx. quinquefasciatus and Hg. equinus, although Ma. dyari was readily infected after engorging on a blood-virus suspension with a titer of 10(3.0) PFU/2 microliters.

Eight previously untyped Bunyamwera serogroup bunyaviruses that had been isolated from mosquitoes collected in California and Oregon between 1969 and 1985, were identified by cross-neutralization tests. Four viruses from Anopheles freeborni and a virus from Aedes sierrensis collected in Butte County in the Central Valley of California in 1970-71 were shown to belong to the Northway serotype. The existence of a Northway serotype virus in California had been inferred from previous serologic surveys of deer and horses, but these are the first Northway serotype viruses to be identified from the contiguous United States or from An. freeborni. This is also the first isolation of any virus from Ae. sierrensis. Two viruses from Culiseta inornata collected in Umatilla County, Oregon in 1969 may represent a new serotype in the Bunyamwera serocomplex of the Bunyamwera serogroup. The name "Stanfield" is proposed for this serotype, which is closely related to the Cache Valley and Northway bunyavirus serotypes. A virus from Ae. taeniorhynchus collected in San Diego County, California in 1985 was confirmed to belong to the Main Drain serotype. Previously, Main Drain serotype viruses in California have been associated principally with Culicoides variipennis.

Ten species of mosquitoes became sensitive to CO2 following intrathoracic (i.t.) inoculation of California encephalitis (CE) virus. These included field-collected Aedes melanimon, Aedes nigromaculis and Culiseta incidens and laboratory-colonized strains of Aedes dorsalis, Aedes triseriatus, Anopheles freeborni, Culex peus, Culex pipiens pipiens, Culex pipiens quinquefasciatus and Culex tarsalis. Another California serogroup virus, Jerry Slough (= Jamestown Canyon) (JS), also induced CO2 sensitivity in nearly 100% of infected female Ae. dorsalis, Ae. melanimon and Cx. tarsalis. Sensitivity to CO2 developed in Ae. dorsalis and Cx. tarsalis within 3 days after i.t. inoculation of CE virus, remained at a high level through day 10 after inoculation and decreased gradually with time until by 30 days post-inoculation nearly all mosquitoes were nonsensitive. In contrast, Ae. dorsalis infected transovarially with CE virus were not CO2 sensitive. There was homologous but no heterologous interference to the development of CO2 sensitivity when Ae. dorsalis infected transovarially with CE virus were inoculated with CE or JS viruses. Two other bunyaviruses, Main Drain and Turlock, produced CO2 sensitivity in 3-10% and 0%, respectively, of infected Ae. dorsalis, Ae. melanimon and Cx. tarsalis. Two toga-viruses, St. Louis encephalitis and western equine encephalomyelitis, did not induce CO2 sensitivity in Ae. dorsalis and Cx. tarsalis.

CEV is primarily associated with Ae. melanimon as its maintenance host and vector in California. A few species of lagomorphs and rodents will develop a transient viremia when infected. Infection of domestic mammals and man is frequent but rarely results in a clinical disease. Extensive field and laboratory studies lead to the conclusion that CEV probably is a mosquito virus that is partially adapted to a narrow range of vertebrate hosts but is not reliant on vertebrates for its continuous existence. JCV has many similarities to CEV in California in that it is dependent on one mosquito, Cu. inornata, as its vector and, as far as is known, the same group of mammalian species as for CEV serve as hosts for JCV.

Titers of Turlock (TUR) and Hart Park (HP) viruses were reduced to undetectable levels when virus was mixed with a triturated suspension of uninfected (normal) 4th instar larvae of Culex tarsalis prior to plaque assay in cell culture. There was a linear relationship between the number of larvae in the pool and the titer of virus recoverable. Virus was undetectable when 1,000-10,000 PFU of either agent was added to pools that contained 25 or more larvae. Suspensions of up to 25 adult male or female Cx. tarsalis had little effect on detectable viral titers while pupal suspensions had an intermediate effect. The inhibitory effect of normal larval extracts on viral infectivity could be counteracted by use of polycations or a high pH buffer. A similar reduction in titer of TUR virus was observed with extracts of larvae of Aedes melanimon or Anopheles franciscanus. Larval extracts of Cx. tarsalis similarly reduced titers of California and St. Louis encephalitis viruses but not western equine encephalomyelitis virus. These findings may have significant bearing on the interpretation of transovarial transmission attempts in which pooled larvae are assayed for virus.

In three cases of encephalitis in humans that occurred in the area where the newly described California virus was isolated from mosquitoes, serological evidence seemed to indict the California virus as the etiological agent. In the case of an infant with very severe disease, the serological evidence was convincing; the evidence was almost as strong in the case of a seven-year-old boy; the results in an adult were equivocal. Inapparent infection in man is quite common as indicated by neutralization tests on the sera of nearly 600 residents of California, but encephalitic manifestations of infection are extremely rare. In Kern County, California, where the virus was discovered, approximately 11 per cent of the population has been infected. Infection rates are higher in adults than in very young children. Absence of neutralizing antibodies from 64 specimens of blood from persons in Japan, Washington, and other states supports the specificity of the neutralization test in man and suggests that this virus is absent or is not being similarly transmitted in some areas. Serological evidence from serial bleedings of two sick horses suggested, but did not definitely establish, that this virus leads to a naturally acquired encephalomyelitis in horses. Serological tests with the viruses of western equine and St. Louis encephalitis did not lead to any other etiological diagnosis in the sick animals studied. Results of neutralization tests on the sera of eight horses and three cows in Kern County suggested extremely high infection rates, and an immunity rate of 18 per cent was noted in rabbits and ground squirrels. In the natural biological cycle rabbits and ground squirrels are suspected as the possible counterpart of birds in the St. Louis and western equine virus cycles. There is no evidence from field or laboratory to indicate that birds become infected with the California virus. Sera from 33 mammals other than man were collected from Northern California and Washington. All were free from neutralizing antibodies, again supporting the specificity of positive findings from Kern County.