The roles of balancing selection and recombination in the evolution of rattlesnake venom

Schield, Drew R.

drew.schield@colorado.edu

Department of Ecology and Evolutionary Biology

University of Colorado

Boulder, Colorado USA

The origin of snake venom involved the duplication of non-venom genes and the recruitment of these duplicates into venom systems. For decades, studies have predicted that directional positive selection has governed this process. Venom composition varies substantially across snake species, and venom phenotypes are thought to be locally adapted to prey, leading to coevolutionary interactions between predator and prey. Here, prey may evolve mechanisms to resist snake venom, and in turn snakes may evolve novel venom components to kill prey more effectively. Distinct from venom origins, contemporary patterns of snake venom evolution may therefore be driven by fundamentally different selection regimes, yet population-level patterns of selection have been only rarely investigated. Here we use whole-genome data from 68 rattlesnakes to test hypotheses about selection and the drivers of genomic diversity and differentiation in major venom gene regions. Our results indicate that selection has resulted in long-term maintenance of genetic diversity within and between species in multiple venom gene families. Our findings are inconsistent with a dominant role of directional positive selection, and instead support a role of long-term balancing selection in shaping venom evolution. We also detect rapid decay of linkage disequilibrium due to high recombination rates in venom regions, suggesting that venom genes have reduced selective interference with nearby loci, including other venom paralogs. Our results characterize a new example of long-term balancing selection that drives trans-species polymorphism – for which there are few examples—and help explain major evolutionary forces underlying a widely studied adaptation.