Populations that enter a novel environment face strong and novel selection pressures. Adaptation to novel conditions is of interest to evolutionary biologists, conservation biologists, and ecologists in the context of human-associated environmental change on Earth. How often does adaptation occur using the same alleles across populations and environments? Understanding this parallelism in the genomic basis of adaptation may help in predicting the impact of selection in natural populations. Laboratory evolution systems provide an opportunity to study the effect of selection on the extent and degree of genomic parallelism during adaptation. We address this using laboratory adaptation to novel resources in a pest, the red flour beetle Tribolium castaneum (Ancestral resource- Wheat, Novel resources- Corn, Finger Millet, and Sorghum; most to least sub-optimal). Specifically, we ask whether parallelism is influenced by the founding population size or the strength of selection (different for different resources). After 35-60 generations of evolution, populations exhibit early stages of adaptation and ecological divergence: increased fitness in the new resources, as well as significant assortative mating across populations, adapted to distinct resources. To determine the genetic basis of adaptation, I sequenced the pooled genomic DNA of 20 adults from ancestral and evolved lines and determined the change in allele frequencies over time (for standing genetic variation), as well as new variants representing mutations during evolution. Overall, we found very few new mutations, indicating adaptation largely via SGV. Compared to the ancestor, evolved lines have many more alleles at extreme frequencies, possibly due to rapid purging of deleterious alleles and fixation of beneficial alleles during adaptation. Evolved lines also have reduced nucleotide diversity than the ancestor. The magnitude of these two effects is broadly correlated with the initial selection imposed in each resource (i.e., strongest in corn, which is most suboptimal). Interestingly, allele frequencies changed parallelly both within (i.e., across replicate populations) and between resources, suggesting similar underlying mechanisms of adaptation to different habitats. I also found evidence suggesting large selective sweeps, some involving entire chromosomes. We are now analyzing candidate genes under selection to predict the functional basis of an increase in fitness in novel resources. We are also performing reciprocal transplant experiments with ancestral stocks and evolved lines in all resources to ask if there are any fitness-related trade-offs associated with adaptation to novel environments. In conclusion, rapid adaptation to various resources can occur in this pest beetle using standing genetic variation and the same large effect alleles across populations.
Genomic Parallelism in Pest Beetle Adaptation to Novel Environments