Viral fitness predicts the magnitude and direction of perturbations in the infected host transcriptome

Abstract

Overall, these results reveal the high information content of viral fitness, and suggest its potential use to predict differences in genomic profiles of infected hosts.. CC-BY-NC-ND 4.0 International license not peer-reviewed) is the author/funder. It is made available under aThe copyright holder for this preprint (which was . http://dx.doi.org/10.1101/206789 doi: bioRxiv preprint first posted online 3 Fitness is a complex parameter often used by evolutionary biologists and ecologists to quantitatively describe the reproductive ability and evolutionary potential of an organism into a particular environment 1,2 . Despite this apparently simple definition, measuring fitness is difficult and most studies only measure one or more fitness components (e.g., survival to maturity, fecundity, number of mates, or number of offspring produced) as proxies to total fitness 1,2 . In the field of virology, it has become standard to measure fitness by growth-competition experiments in mixed infections with a reference strain 3,4 . With this experimental set up, fitness is just the relative ability of a viral strain to produce stable infectious progeny in a given host (cell type, organ, individual, or species) when the available resources have to be shared with a competitor 5 .Regardless its limitations, at least, this approach provides a metric for ranking viral strains according to their performance in a particular environment/host. Such a fitness measure has been pivotal for quantitatively understanding many virus evolution processes: the effect of genetic bottlenecks and accumulation of deleterious mutations in RNA virus populations 6-8 , the rates and dynamics of adaptive evolution into novel hosts 9, , the pleiotropic cost of host range expansion [10][11][12] , the cost of genome complexity 13,14 , the cost of antiviral escape mutations [15][16][17] , the topography of adaptive fitness landscapes [18][19][20] , and the role of robustness in virus evolution [21][22][23] .But differences in viral fitness should also matter in genome wide studies seeking to understand the mode of action of the viruses (i.e., the precise way they interact with their hosts).Even though it has been argued that an integrative systems biology approach to viral pathogenesis would result in a better understanding of pathogenesis and in the identification of common targets for different viruses, therefore serving as a guide to a more rational design of therapeutic drugs [24][25][26][27][28][29][30] , pioneering studies have ignored the high genetic variability of viruses and subsequent differences in fitness and in mode of action. Experimental evidences support that even single nucleotide substitutions have significant effects on viral fitness, regardless they are synonymous or nonsynonymous, or they affect coding or non-coding genomic regions [31][32][33][34][35][36][37] . A common trend among all these studies is that, whenever fitness is evaluated in the standard host, the distribution of mutational effects is highly skewed towards deleterious...