Positive selection has shaped the evolution of Argentine ant immune genes both in native and introduced supercolonies.

The highly invasive Argentine ant (Linepithema humile) started its colonisation from the species' native range in South America approximately 150 years ago and has since become one of the major pests in the world. We investigated how the shifts into new ranges have affected the evolution of Argentine ants' immune genes. To the best of our knowledge, this is the first broadscale population genetic study focusing on ants' immune genes. We analysed comprehensive targeted-seq data of immune and non-immune genes containing 174 genes from 18 Argentine ant supercolonies covering the species' native and introduced ranges. We predicted that the immune gene evolution of introduced supercolonies differs from that of the native supercolonies and proposed two different, non-mutually exclusive hypotheses for this: 1) the enemy release hypothesis and 2) the higher pathogen pressure hypothesis - both of which seem to explain the observed evolutionary patterns on their behalf. Our results show that the introduced supercolonies were targeted by weaker selection than natives, but positive selection was evident among supercolonies of both ranges. Moreover, in some cases, such as the antiviral RNAi genes, introduced range supercolonies harboured a higher proportion of positively selected genes than natives. This observation was striking, knowing the recent demographic history and the detected generally lower selection efficacy of introduced supercolonies. In conclusion, it is evident that pathogen pressure is ubiquitous and strongly affects the immune gene evolution in Argentine ants.

Antiviral immune response reveals host-specific virus infections in natural ant populations.

Hosts can carry many viruses in their bodies, but not all of them cause disease. We studied ants as a social host to determine both their overall viral repertoire and the subset of actively infecting viruses across natural populations of three subfamilies: the Argentine ant (Linepithema humile, Dolichoderinae), the invasive garden ant (Lasius neglectus, Formicinae) and the red ant (Myrmica rubra, Myrmicinae). We used a dual sequencing strategy to reconstruct complete virus genomes by RNA-seq and to simultaneously determine the small interfering RNAs (siRNAs) by small RNA sequencing (sRNA-seq), which constitute the host antiviral RNAi immune response. This approach led to the discovery of 41 novel viruses in ants and revealed a host ant-specific RNAi response (21 vs. 22 nt siRNAs) in the different ant species. The efficiency of the RNAi response (sRNA/RNA read count ratio) depended on the virus and the respective ant species, but not its population. Overall, we found the highest virus abundance and diversity per population in Li. humile, followed by La. neglectus and M. rubra. Argentine ants also shared a high proportion of viruses between populations, whilst overlap was nearly absent in M. rubra. Only one of the 59 viruses was found to infect two of the ant species as hosts, revealing high host-specificity in active infections. In contrast, six viruses actively infected one ant species, but were found as contaminants only in the others. Disentangling spillover of disease-causing infection from non-infecting contamination across species is providing relevant information for disease ecology and ecosystem management.