Peter Prentis

While the unique symbiotic relationship between anemonefishes and sea anemones is iconic, it is still not fully understood how anemonefishes can withstand and thrive within the venomous environment of their host sea anemone. In this study, we used a proteotranscriptomics approach to elucidate the proteinaceous toxin repertoire from the most common host sea anemone, Entacmaea quadricolor. Although 1251 different toxin or toxin-like RNA transcripts were expressed in E. quadricolor tentacles (0.05% of gene clusters, 1.8% of expression) and 5375 proteins were detected in milked venom, only 4% of proteins detected in venom were putative toxins (230), and they only represent on average 14% of the normalised protein expression in the milked venom samples. Thus, most proteins in milked venom do not appear to have a toxin function. This work raises the perils of defining a dominant venom phenotype based on transcriptomics data alone in sea anemones, as we found that the dominant venom phenotype differs between the transcriptome and proteome abundance data. E. quadricolor venom contains a mixture of toxin-like proteins of unknown and known function. A newly identified toxin protein family, Z3, rich in conserved cysteines of unknown function, was the most abundant at the RNA transcript and protein levels. The venom was also rich in toxins from the Protease S1, Kunitz-type and PLA2 toxin protein families and contains toxins from eight venom categories. Exploring the intricate venom toxin components in other host sea anemones will be crucial for improving our understanding of how anemonefish adapt to the venomous environment.

Feral pigs ( Sus scrofa ) are a destructive and widespread invasive pest in Australia. An understanding of feral pig movement is required to develop management strategies to control feral pigs in Australia. Because landscape structure can have a strong influence on animal movement, it is important to determine how landscape features facilitate or impede the movement of feral pigs. Consequently, we conducted a landscape genetic analysis of feral pig populations in the Herbert region of far north Queensland, Australia, to determine management units and provide recommendations to better inform feral pig population control strategies. Using microsatellite data obtained from 256 feral pig samples from 44 sites, we examined feral pig population structure at multiple spatial scales for univariate and multivariate landscape resistance surfaces to determine the optimal spatial scale and to identify which of the nine landscape features tested impede or facilitate feral pig gene flow. Only weak genetic structure was found among the 44 sampling sites, but major waterways were identified as a minor barrier to gene flow, and an isolation by distance model was supported. We also found that highways facilitated gene flow across the study area, and this suggests that they may act as movement corridors or indicate translocation of feral pigs. Additionally, incorporating a second spatial scale enhanced the ability of our landscape genetics analysis to detect the influence of landscape structure on gene flow. We identified three management units based on natural barriers to gene flow and future targeted control should be undertaken in these management units to deliver sustained reduction of feral pig populations in the Herbert region. This study demonstrates how a landscape genetic approach can be used to gain insight into the ecology of an invasive pest species and be used to develop population control strategies which utilise natural barriers to movement.