Viral diversity of bat communities in human-dominated landscapes in Mexico

Using integrative epidemiologic techniques, we studied the changing relationships (beta and phylogenetic beta diversity) of multihost systems and virus associations in bat communities in fragmented landscapes from Chiapas, Campeche and Greater Mexico City. We combined computing applications, molecular detection, and nucleotide sequencing of coronaviruses, hantaviruses, paramyxoviruses and pegiviruses with ecological and phylogenetic analyses. A total of 22 viruses were discovered in 1,067 samples from 42 bat species, representing an estimated 78% of all viral richness in the system. Based on 17 virus genotypes discovered with an equal sampling effort, a total viral richness of 23 genotypes was estimated using a Chao2 statistic model. Using a residual model, we categorized host species and habitat types that are prone to harboring higher viral richness. Positive relationships were found between phylogenetic host diversity and both viral diversity (r = 0.41, p < 0.05) and viral richness (r = 0.51, p < 0.05). The beta diversity (the rate of change) of viral communities was explained by host beta diversity (r = 0.86, p < 0.05). To understand the change in viral and host communities, we partitioned beta diversity in nestedness (species loss) and turnover (compositional dissimilarity) components. In Chiapas, the host beta diversity was explained by the nestedness of species composition, while the phylogenetic host diversity was explained by turnover of the host lineages. Campeche showed a high phylogenetic host nestedness and low host turnover. Beta-diversity and beta-phylogenetic diversity indicated that patterns of local species assemblages and regional abiotic features in human-dominated landscapes are significant drivers of viral community composition. Our study represents the first effort in Mexico to study the relationship between viral diversity in bat communities in modified landscapes to understand host-virus relationships.