PS 6-81 - Testing the underpinnings of the trade-off hypothesis of virulence using in vitro ranavirus infections

Abstract: Viral replication is commonly assumed to be the primary determinant of virulence because pathogen replication necessarily exploits host resources. However, this ignores the ubiquitous innate and adaptive host immune responses that can lead to significant inflammation, pathology, and even death. To predict even the direction of virulence evolution, let alone the optimal level of virulence, per the trade-off hypothesis, it is first necessary to understand the relative importance of resource- and immune-mediated virulence. The objective of this study was to test a key assumption of the trade-off hypothesis that virulence is primarily resource-mediated as opposed to immune-mediated. We used ranaviruses, viruses of ectothermic vertebrates that that vary substantially in virulence, from asymptomatic infections to mass die-off events. Five strains of ranavirus strains (wildtype Frog Virus 3 [FV3], chimeric Common Midwife Toad Virus and FV3 [CMTV-FV3], and three recombinant FV3 strains) were characterized in vitro in a Xenopus laevis kidney epithelial cell line. Cell cultures were exposed to one of the five strains or the control, and samples were collected on days 1, 3, and 5 post-exposure. Viral replication rates were estimated in a Bayesian framework by quantifying viral titers over time with plaque assays (infectious virions) and quantitative polymerase chain reaction (qPCR) assays (viral genomes). Gene expression using qPCR for a suite of interferon-stimulated genes were used to quantify host innate immune responses for each viral strain. Lastly, flow cytometry using a live-dead cell assay was used to quantify proportions of live and apoptotic cells, used as a proxy for virulence (i.e., cell death) in vitro. Proportions of live and apoptotic cells varied significantly across viral strains, suggesting that virulence is at least in part determined by viral identity. Furthermore, contrary to the assumptions of the trade-off hypothesis, the relative virulence of the strains was not associated with viral replication rates. Gene expression results to follow will be used along with viral replication rates to determine the relative effect of each predictor on virulence through a multiple linear regression. The results from this experiment highlight the need for interdisciplinary studies of virulence that simultaneously measure pathogen and host traits to bridge the gap between empirical and theoretical studies of virulence evolution.