Abstract:
African horsesickness, a disease of equines caused by African horsesickness virus (AHSV), is often fatal, although the pathogenic effect in different animals is variable. Current AHSV vaccines are live attenuated viruses generated by serial passage in cell culture. This process affects virus plaque size, which has been considered an indicator of AHSV virulence (Erasmus, 1966; Coetzer and Guthrie, 2004). The most likely AHSV proteins to be involved in viral virulence and attenuation are the outer capsid proteins, VP2 and VP5, due to their role in attachment of viral particles to cells and early stages of viral replication. Nonstructural protein NS3 may play an equally important role due to its function in release of viral particles from cells. Two viruses were obtained for this study, AHSV-4(1) and AHSV-4(13). The thirteenth passage virus, AHSV-4(13), originated from the primary isolate AHSV-4(1). The three most variable AHSV proteins are VP2, VP5 and NS3. The question of sequence variation of these proteins between AHSV-4(1) and AHSV-4(13) arising during the attenuation process was addressed. The subject of plaque size variation between these viruses was also investigated. Some of the sequence variation observed in NS3, VP2 and VP5, between AHSV-4(1) and AHSV-4(13), occurred in protein regions that may be involved in virus entry into and exit from cells. The sequence information also indicated that AHSV-4(1) and AHSV-4(13) consist of genetically heterogeneous viral pools. The plaque size of AHSV-4(1) was variable, with small to relatively large plaques, whereas the plaques of AHSV-4(13) were mostly large. During serial plaque purification of AHSV-4(1) plaque size increased and became homogenous in size. No sequence variation in NS3 or VP5 of any of the plaque variants could be linked to variation or change in plaque size. NS3 and VP5 have a possible role in the AHSV virulence phenotype, and exhibit cytotoxic properties in bacterial and insect cells. As these proteins have not been studied in mammalian cells, an aim of this study was to express them in Vero cells and investigate their cytotoxic and membrane permeabilization properties within these cells. The NS3 and VP5 genes of AHSV-4(1) and AHSV-4(13) were successfully inserted into a mammalian expression vector and transiently expressed in Vero cells. The transfection procedure was optimized using eGFP, but expression levels were still low. When NS3 and VP5 were expressed, no obvious signs of cytotoxicity were observed. Cell viability and membrane integrity assays were performed and expression of NS3 and VP5 in Vero cells had no detectable effect on cell viability or membrane integrity. Low expression levels may have resulted in protein levels too low to cause membrane damage or affect cell viability. As Vero cells support AHSV replication, low levels of NS3 and VP5 may not be cytotoxic in these cells. NS3 was further investigated by expressing an NS3-eGFP fusion protein in Vero cells. Putative localization with membranous components and possible perinuclear localization of the fusion protein was observed. These observations may be confirmed with more sensitive microscopic techniques for a better assessment of the localization.
