Population transcriptomics reveals weak parallel genetic basis in repeated marine and freshwater divergence in nine-spined sticklebacks

Abstract

© 2020 John Wiley & Sons Ltd. The degree to which adaptation to similar selection pressures is underlain by parallel versus non-parallel genetic changes is a topic of broad interest in contemporary evolutionary biology. Sticklebacks provide opportunities to characterize and compare the genetic underpinnings of repeated marine-freshwater divergences at both intra- and interspecific levels. While the degree of genetic parallelism in repeated marine-freshwater divergences has been frequently studied in the three-spined stickleback (Gasterosteus aculeatus), much less is known about this in other stickleback species. Using a population transcriptomic approach, we identified both genetic and gene expression variations associated with marine-freshwater divergence in the nine-spined stickleback (Pungitius pungitius). Specifically, we used a genome-wide association study approach, and found that ~1% of the total 173,491 identified SNPs showed marine-freshwater ecotypic differentiation. A total of 861 genes were identified to have SNPs associated with marine-freshwater divergence, but only 12 of these genes have also been reported as candidates associated with marine-freshwater divergence in the three-spined stickleback. Hence, our results indicate a low degree of interspecific genetic parallelism in marine-freshwater divergence. Moreover, 1,578 genes in the brain and 1,050 genes in the liver were differentially expressed between marine and freshwater nine-spined sticklebacks, ~5% of which have also been identified as candidates associated with marine-freshwater divergence in the three-spined stickleback. However, only few of these (e.g., CLDND1) appear to have been involved in repeated marine-freshwater divergence in nine-spined sticklebacks. Taken together, the results indicate a low degree of genetic parallelism in repeated marine-freshwater divergence both at intra- and interspecific levels.