Genes that affect Atlantic salmon growth in hatchery do not have the same effect in the wild

Abstract

Summary Dissecting the genetic mechanisms of phenotypic traits that influence fitness in diverse environments provides the important first step towards understanding the robustness of the observed genotype–phenotype associations, the role of genotype‐by‐environment interaction (GEI) shaping fitness trade‐offs and maintaining genetic variation of quantitative traits. However, the molecular basis of complex traits in vertebrates has rarely, if ever, been studied simultaneously in natural and controlled laboratory environments. To evaluate whether the same genomic regions affect the growth of juvenile Atlantic salmon in wild and hatchery conditions, we mapped QTLs affecting the size‐at‐age of juvenile parr after the first summer utilizing the same mapping (family) material to avoid confounding effects of different genetic background. We identified three significant QTLs for fork length in the hatchery that were undetectable in the natural environment, while two QTLs detected in the wild were not observed when fish were reared in hatchery conditions. Altogether, four individual markers showed significant (P < 0·05) genotype‐by‐environment interactions. The allelic effects for three QTLs observed in the hatchery were in the same direction in both environments, while for two QTLs the alleles associated with a larger body size in the wild showed the opposite (albeit non‐significant) trend in the hatchery environment. Our results indicate that the growth of juvenile salmon in two contrasting environments is controlled by different genetic mechanisms that are most likely influenced by multiple processes, including food availability, inter‐ and intraspecific competition, and trade‐offs between the costs and benefits of individual movement in the wild. Furthermore, the findings of the study imply that a substantial proportion of growth‐related QTLs reported by earlier studies in a hatchery environment may represent QTLs specific to farmed conditions and hence have no effect on fish growth in the wild. For comprehensive understanding of the molecular, physiological and ecological mechanisms influencing complex traits, it is therefore crucial to evaluate the genotype–phenotype relationships under natural conditions, rather than relying solely on laboratory experiments.