How a multi-host pathogen is affected by the trophic structure of its host community?

Background/Question/Methods

The increasing number of zoonotic diseases spilling over from a range of wild animal species represents a particular concern for public health. While classic epidemiology framework mainly focuses on specialist pathogens in a simplified ecological context, i.e. one host-one vector disease system, recent works highlighted the influence of host species diversity and community structure on disease dynamic, especially in the current context of biodiversity decline. However, despite their influence on pathogen transmission is barely studied, interactions between species, typically trophic interactions, are a key component of ecological communities through structuring their diversity and composition as well as shaping the spill-over routes of pathogen transmitted by predation.

Here we explored the reciprocal relationships between pathogen dynamics, community structure and topology of trophic network by integrating a multi-species population dynamic model within a multi-host species epidemiological model. Taking the advantage of these theoretical results, we investigated this relationship using an extensive data set of Mycobacterium ulcerans (the environmentally-persistent agent of Buruli ulcer, a skin disease endemic in many tropical countries) prevalence in its fresh-water aquatic host community through time and space in both Cameroon and French Guiana.

Results/Conclusions

Our model simulations showed contrasting effects of trophic network structure on pathogen dynamics. Beyond the number of interactions and the network size, the aggregation of interactions into the network seems to be particularly determinant for pathogen prevalence in host carrier species. Investigation of M. ulcerans prevalence data revealed that trophic level of host carrier species influences the probability to carry the bacillus, arguing for a role of trophic structure in the transmission of the disease agent. To our knowledge our model is the first to investigate the influence of complex community dynamics on pathogen transmission. We believe that its investigation can improve our understanding of generalist pathogen dynamics in its natural host communities, and so could help in anticipating the response of environmentally-persistent and zoonotic diseases to biodiversity loss and local community alterations.