Unravelling ant-symbiont network topology across Europe

Long-term associations between different species are key drivers in community composition in all ecosystems. Understanding the ecological and evolutionary drivers of these symbiotic associations is challenging because of the diversity of species and interaction types hosted in natural ecological networks. Here, we compiled the most complete database on natural ant-symbiont networks in Europe to identify the drivers of bipartite network topology. These ant-symbiont networks host an unrivalled diversity of symbiotic associations across the entire mutualism-antagonism continuum, of which the most diverse types of symbionts are (1) trophobionts: mutualistic aphids and scale insects (2) myrmecophiles: commensalistic and parasitic arthropods, and (3) social parasites: parasitic ant species. These diverse ant-symbiont networks provide a unique opportunity to tease apart ecological and evolutionary drivers. To do so, we dissected network topology and asked what determines host specificity and which host factors drive symbiont species richness and facilitate host switching for the different types of symbionts. We found an unexpectedly high number of 701 obligate symbionts associated with European ants. Symbiont type explained host specificity and the average relatedness of the targeted host species. Social parasites were associated with few, but phylogenetically highly related hosts, whereas trophobionts and myrmecophiles interacted with a higher number of hosts across a wider taxonomic distribution. Colony size, host range and habitat type predicted total symbiont richness, where ants hosts with larger colony size or larger distribution range contained more symbiont species. However, we found that different sets of host factors affected diversity in the different types of symbionts. Ecological factors, such as colony size, host range and niche width predominantly drive myrmecophile species richness, whereas evolutionary factors, such as host phylogeny and biogeography, mainly determine richness of mutualistic trophobionts and social parasites. Lastly, we found that hosts with a common biogeographic history support a more similar community of symbionts. Phylogenetic related hosts also shared more trophobionts and social parasites, but not myrmecophiles. Taken together, these results suggest that ecological and evolutionary processes drive host specificity and symbiont richness in large-scale ant-symbiont networks, but these drivers may shift in importance depending on the type of symbiosis. Our findings highlight the potential of well-characterized bipartite networks composed of different types of symbioses to identify candidate processes driving community composition.