Effect of landscape fragmentation on bat population dynamics and disease persistence in Uruguay

Abstract

The transmission of pathogens into novel host species, a process known as spillover, requires a series of conditions to align in space and time. A series of imperfect barriers prevent the jump of pathogens from one species to others. These may include the distribution and abundance of the primary host, the survival of the pathogen in the environment and the susceptibility of the recipient host to the pathogen. Only when permissive conditions align in time and space can the spillover occur. Spillovers may be relatively rare events and the understanding of the dynamics of the barriers is constrained by the ability of detecting and analyzing such events. Systems where spillover does not occur, despite apparent presence of all required conditions, provide an opportunity to understand barriers preventing inter-species transmission. Vampire bat-borne rabies in Uruguay provides such an opportunity. Despite large and stable livestock density, presence of vampire bats, and circulation of the virus in close proximity, the country did not experience livestock rabies outbreaks until 2007. Here we combined historical review, field sampling, and statistical and mathematical modeling to understand the factors driving the emergence of rabies in Uruguay in 2007 and the previous absence of the disease. Our results suggest that rabies outbreaks in the country are spatially and temporally associated with fragmentation of grasslands. We showed that proposed increased connectivity among colonies, in response to fragmentation, is sufficient to explain longer persistence of the virus in the bat colonies, allowing more opportunities for virus transmission to livestock. We showed that connectivity has a strong effect on rabies persistence and that reproductive seasonality and population turnover have only marginal effects compared to connectivity. As connectivity driven by shared feeding areas might not be detectable by genetic analyses of the bats, we proposed the use of a widespread virus persistently infecting bats as a marker to trace connectivity across colonies. Combined, the results presented here provide tools that can be applied to intervene and apply countermeasures to prevent spillover.