Parallel evolution and enhanced virulence uponin vivopassage of an RNA virus inDrosophila melanogaster

Abstract

Virus evolution is strongly affected by antagonistic co-evolution of virus and host. Host immunity positively selects for viruses that evade the immune response, which in turn may drive counter-adaptations in host immune genes. We investigated how host immune pressure shapes virus populations, using the fruit fly Drosophila melanogaster and its natural pathogen Drosophila C virus (DCV), as a model. We performed an experimental evolution experiment in which DCV was serially passaged for 10 generations in three fly genotypes differing in their antiviral RNAi response, using wild-type flies and flies in which the endonuclease gene Dicer-2 was either overexpressed or inactivated. All evolved virus populations replicated more efficiently in vivo and were more virulent than the parental stock. The number of polymorphisms increased in all three host genotypes with passage number, which was most pronounced in Dicer-2 knockout flies. Mutational analysis showed strong parallel evolution, as every lineage accumulated mutations in a specific region of the VP3 capsid protein in a host genotype-independent manner. The parental tyrosine at position 95 of VP3 was substituted with either one of five different amino acids in 14 out of 15 lineages. However, no consistent amino acid changes were observed in the viral RNAi suppressor gene 1A, nor elsewhere in the genome in any of the host backgrounds. Our study indicates that the RNAi response restricts the sequence space that can be explored by viral populations. Moreover, our study illustrates how evolution towards higher virulence can be a highly reproducible, yet unpredictable process.

VILOCA: Sequencing quality-aware haplotype reconstruction and mutation calling for short- and long-read data

“…Unlike SNVs, which capture individual nucleotide changes, examining co-occurring mutations and local haplotypes provides insights into multiple mutations occurring (locally) on the same genetic background. This piece of information is particularly valuable when analyzing viral populations with complex epistatic interactions affecting viral phenotypes such as virulence [38], viral fitness [39], drug resistance [2,3], and immune escape [40,41].…”