Interaction type influences ecological network structure more than local abiotic conditions: evidence from endophytic and endolichenic fungi at a continental scale

Pierre-Luc Chagnon, Jana M. U’Ren, Jolanta Miadlikowska, François Lutzoni, A. Elizabeth Arnold

Understanding the factors that shape community assembly remains one of the most enduring and important questions in modern ecology. Network theory can reveal rules of community assembly within and across study systems and suggest novel hypotheses regarding the formation and stability of communities. However, such studies generally face the challenge of disentangling the relative influence of factors such as interaction type and environmental conditions on shaping communities and associated networks. Endophytic and endolichenic symbioses, characterized by microbial species that occur within healthy plants and lichen thalli, represent some of the most ubiquitous interactions in nature. Fungi that engage in these symbioses are hyperdiverse, often horizontally transmitted, and functionally beneficial in many cases, and they represent the diversification of multiple phylogenetic groups. We evaluated six measures of ecological network structure for >4100 isolates of endophytic and endolichenic fungi collected systematically from five sites across North America. Our comparison of these co-occurring interactions in biomes ranging from tundra to subtropical forest showed that the type of interactions (i.e., endophytic vs. endolichenic) had a much more pronounced influence on network structure than did environmental conditions. In particular, endophytic networks were less nested, less connected, and more modular than endolichenic networks in all sites. The consistency of the network structure within each interaction type, independent of site, is encouraging for current efforts devoted to gathering metadata on ecological network structure at a global scale. We discuss several mechanisms potentially responsible for such patterns and draw attention to knowledge gaps in our understanding of networks for diverse interaction types.