The diversity and distribution of multihost viruses in bumblebees

Abstract

The bumblebees (genus Bombus) are an ecologically and economically important group in decline. Their decline is driven by many factors, but parasites are believed to play a role. This thesis examines the factors that influence the diversity and distribution of multihost viruses in bumblebees using molecular and modelling techniques. In Chapter 2, I performed viral discovery to isolate new multihost viruses in bumblebees. I investigated factors that explain prevalence differences between different host species using co-phylogenetic models. I found that related hosts are infected with similar viral assemblages, related viruses infect similar host assemblages and related hosts are on average infected with related viruses. Chapter 3 investigated the ecology of four of the novel viruses in greater detail. I applied a multivariate probit regression to investigate the abiotic factors that may drive infection. I found that precipitation may have a positive or negative effect depending on the virus. Also, we observe a strong non-random association between two of the viruses. The novel viruses have considerably more diversity than the previously known viruses. Chapter 4 investigated the effect of pesticides on viral and non-viral infection. I exposed Bombus terrestris colonies to field realistic doses of the neoticotinoid pesticide clothianidin in the laboratory, to the mimic pulsed exposure of crop blooms. I found some evidence for a positive effect of uncertain size on the infection rate of pesticide exposed colonies relative to non-pesticide exposed colonies, a potentially important result. Chapter 5 explored the evolution of avirulent multihost digital organisms across fluctuating fitness landscapes within a discrete sequence space. Consistent with theory, I found that evolution across a fluctuating discrete landscape leads to a faster rate of adaptation, greater diversity and greater specialism or generalism, depending on the correlation between the landscapes. A large range of factors are found to be important in the distribution of infection and diversity of viruses, and we find evidence for abiotic, biotic and anthropogenic factors all playing a role.