Within-Generation Polygenic Selection Shapes Fitness-Related Traits across Environments in Juvenile Sea Bream.

Carine Rey,Nicolas Bierne,Pierre-Alexandre Gagnaire,Bruno Guinand,François Bonhomme,Audrey Darnaude,Franck Ferraton

doi:10.3390/genes11040398

Abstract

Understanding the genetic underpinnings of fitness trade-offs across spatially variable environments remains a major challenge in evolutionary biology. In Mediterranean gilthead sea bream, first-year juveniles use various marine and brackish lagoon nursery habitats characterized by a trade-off between food availability and environmental disturbance. Phenotypic differences among juveniles foraging in different habitats rapidly appear after larval settlement, but the relative role of local selection and plasticity in phenotypic variation remains unclear. Here, we combine phenotypic and genetic data to address this question. We first report correlations of opposite signs between growth and condition depending on juvenile habitat type. Then, we use single nucleotide polymorphism (SNP) data obtained by Restriction Associated DNA (RAD) sequencing to search for allele frequency changes caused by a single generation of spatially varying selection between habitats. We found evidence for moderate selection operating at multiple loci showing subtle allele frequency shifts between groups of marine and brackish juveniles. We identified subsets of candidate outlier SNPs that, in interaction with habitat type, additively explain up to 3.8% of the variance in juvenile growth and 8.7% in juvenile condition; these SNPs also explained significant fraction of growth rate in an independent larval sample. Our results indicate that selective mortality across environments during early-life stages involves complex trade-offs between alternative growth strategies.

Highlights

Understanding how species adapt to heterogeneous environments is a central objective in evolutionary biology and ecology [1,2,3,4]
The small allele frequency changes generated by polygenic selection represent a major challenge for ecological genomics studies that search for single locus signatures of selection in molecular data [13,16]
Since we found slopes of opposite signs, we tested the interaction between individual standardized length and habitat type for the condition factor response using an analysis of covariance (ANCOVA) in R, which allows fitting different slopes and intercepts in each habitat using lm(formula = Condition ~ Habitat × Std length)

Summary

Introduction

Understanding how species adapt to heterogeneous environments is a central objective in evolutionary biology and ecology [1,2,3,4] This basic question has important ramifications for our understanding of diversification and extinction and has gained practical importance for predicting species resilience in the face of rapid global change [5,6]. The main obstacle to establishing such connections occurs when selection affects complex quantitative traits that are themselves encoded by many genes [12,13]. In these situations, genome-wide association (GWA) studies between complex traits and single nucleotide polymorphisms (SNPs) usually lack the power to detect loci with small individual effects on phenotype [14,15]. The small allele frequency changes generated by polygenic selection represent a major challenge for ecological genomics studies that search for single locus signatures of selection in molecular data [13,16]

Methods

Results

Discussion

Conclusion