Experimental passage of St. Louis encephalitis virus in vivo in mosquitoes and chickens reveals evolutionarily significant virus characteristics.

St. Louis encephalitis virus (SLEV; Flaviviridae, flavivirus) was the major cause of epidemic flaviviral encephalitis in the U.S. prior to the introduction of West Nile virus (WNV) in 1999. However, outbreaks of SLEV have been significantly more limited then WNV in terms of levels of activity and geographic dispersal. One possible explanation for these variable levels of activity is that differences in the potential for each virus to adapt to its host cycle exist. The need for arboviruses to replicate in disparate hosts is thought to result in constraints on both evolution and host-specific adaptation. If cycling is the cause of genetic stability observed in nature and arboviruses lack host specialization, then sequential passage should result in both the accumulation of mutations and specialized viruses better suited for replication in that host. Previous studies suggest that WNV and SLEV differ in capacity for both genetic change and host specialization, and in the costs each accrues from specializing. In an attempt to clarify how selective pressures contribute to epidemiological patterns of WNV and SLEV, we evaluated mutant spectra size, consensus genetic change, and phenotypic changes for SLEV in vivo following 20 sequential passages via inoculation in either Culex pipiens mosquitoes or chickens. Results demonstrate that the capacity for genetic change is large for SLEV and that the size of the mutant spectrum is host-dependent using our passage methodology. Despite this, a general lack of consensus change resulted from passage in either host, a result that contrasts with the idea that constraints on evolution in nature result from host cycling alone. Results also suggest that a high level of adaptation to both hosts already exists, despite host cycling. A strain significantly more infectious in chickens did emerge from one lineage of chicken passage, yet other lineages and all mosquito passage strains did not display measurable host-specific fitness gains. In addition, increased infectivity in chickens did not decrease infectivity in mosquitoes, which further contrasts the concept of fitness trade-offs for arboviruses.

Cellular localization of Saint Louis encephalitis virus replication.

Replication of Saint Louis encephalitis (SLE) virus was inhibited when PS cells were treated with actinomycin D, daunomycin or cordycepin during the first 9 hr after infection. Autoradiography of SLE virus-infected pulse labelled cells demonstrated that viral RNA synthesis is localized within the nuclear area. Nuclei purified from cells after 12 hr of infection contained heterogeneous 20 S to 26 S viral RNA but no SLE virus genome sized 43 S RNA. Later during infection, nuclei isolated from infected cells contained large amounts of 43 S and 20 S to 26 S RNAs. The 43 S viral RNA present in cells late in infection could not be removed with 1% Tween 80: Nonidet P 40. Purified nuclei isolated from cells early in infection supported the synthesis of 43 S virion RNA in the absence of cytoplasmic factors. The cytoplasmic membrane fraction prepared from cells early in infection contained heterogeneous 10 S to 26 S RNA species; later during infection these membranes contained viral 43 S, 26 S to 30 S and 4 S RNA. These results suggest that the nucleus is an important site of early viral synthesis.