Evolutionary History and Attenuation of Myxoma Virus on Two Continents

Abstract

The attenuation of myxoma virus (MYXV) following its introduction as a biological control into the European rabbit populations of Australia and Europe is the canonical study of the evolution of virulence. However, the evolutionary genetics of this profound change in host-pathogen relationship is unknown. We describe the genome-scale evolution of MYXV covering a range of virulence grades sampled over 49 years from the parallel Australian and European epidemics, including the high-virulence progenitor strains released in the early 1950s. MYXV evolved rapidly over the sampling period, exhibiting one of the highest nucleotide substitution rates ever reported for a double-stranded DNA virus, and indicative of a relatively high mutation rate and/or a continually changing selective environment. Our comparative sequence data reveal that changes in virulence involved multiple genes, likely losses of gene function due to insertion-deletion events, and no mutations common to specific virulence grades. Hence, despite the similarity in selection pressures there are multiple genetic routes to attain either highly virulent or attenuated phenotypes in MYXV, resulting in convergence for phenotype but not genotype. The text-book example of the evolution of virulence is the attenuation of myxoma virus (MYXV) following its introduction as a biological control into the European rabbit populations of Australia and Europe in the 1950s. However, the key work on this topic, most notably by Frank Fenner and his colleagues, occurred before the availability of genome sequence data. The evolutionary genetic basis to the major changes in virulence in both the Australian and European epidemics is therefore largely unknown. We provide, for the first time, key details on the genome-wide changes that underpin this landmark example of pathogen emergence and virulence evolution. By sequencing and comparing MYXV genomes, including the original strains released in the 1950s, we show that (i) MYXV evolved rapidly in both Australia and Europe, producing one of the highest rates of evolutionary change ever recorded for a DNA virus, (ii) that changes in virulence were caused by mutations in multiple genes, often involving losses of gene function due to insertions and deletions, and that (iii) strains of the same virulence were defined by different mutations, such that both attenuated and virulent MYXV strains are produced by a variety genetic pathways, and generating convergent evolution for phenotype but not genotype.