Emergence of a virulent wildlife disease: using spatial epidemiology and phylogenetic methods to reconstruct the spread of amphibian viruses

Abstract

Ranavirus infection has caused severe disease and mass mortality in UK common frogs for more than twenty years resulting in serious declines in some populations. The pathogen has been studied since 1992. These studies generated two valuable resources exploited in this thesis: an archive of tissues and virus isolates and a database of reports from citizen scientists on ranavirus-consistent mortality. The previous studies yielded modest evidence suggesting that introductions from North America initiated ranavirus emergence in the UK, though little else was known about the pattern of introduction or spread. This thesis conducts a more detailed investigation, extending existing knowledge of ranavirus diversity and spread through molecular epidemiology and phylogenetics, an in vivo infection experiment, and in silico models. Non-lethal sampling protocols for ranavirus screening were assessed in a controlled setting and shown to be as effective as traditional protocols. The database of citizen science reports was utilised in spatio-temporal models of the spread of ranavirus disease, finding that ranavirus infection is spreading by transmission between ponds but that new outbreaks are also correlated with both human population density and regional temperatures. The first whole genome sequence from a UK ranavirus is presented. Analysis of the genome shows that it is an isolate of the ranavirus type species, FV3, on the basis of its near identical genome arrangement and a ‘supergene’ phylogenetic analysis. An unexpected finding was evidence for recent lateral transfer of host DNA into the FV3 genome. A candidate gene survey of European ranaviruses revealed considerable diversity that may explain the variation in virulence and host range in Spain. Two proposed new species of Ranavirus are described there - one highly virulent, the other seemingly asymptomatic – and the previously described CMTV is shown to be a likely cause of catastrophic decline across multiple hosts. A lack of monophyly among Spanish ranaviruses and the spatial pattern of incidence suggest recent introduction(s). Together, the evidence presented in this thesis underlines the key role that humans have played in the spread of this group of virulent wildlife pathogens in two European countries.