From Saltwater to Freshwater; Intraspecific Differences in Population Structure and Morphology in Threespine Sticklebacks

The threespine stickleback Gasterosteus aculeatus is a primarily marine fish that has invaded freshwater repeatedly in postglacial times. After establishment in fresh water environments sticklebacks show rapid and parallel diversification of several traits. Loss or reduction of armour plates after a few generations in freshwater is typical, leading to a striking separation into three morphs; full, partial and low plated. Recently, the genetic basis for the trait was found to be related to variation at the Ectodysplasin (Eda) locus, where the allele frequency for the low plated morph is approximately 1% in the ocean (potential standing genetic variation). Thus, studying systems that cross salinities, and ideally, having several lateral plated morphs, might give insights into the evolutionary mechanisms and processes required for the reduction of plates. Threespine sticklebacks in the Chignik River system in Alaska are polymorphic for lateral plates. In total, 900 fish were sampled from six localities in this watershed, including a estuarine lagoon, two rivers, and two lakes, the most distant being 50 km upstream from the ocean. All fish were genotyped for 12 neutral microsatellites and 8 potential quantitative trait loci (QTLs), where one of the QTLs targets the Eda locus. The fish were also characterized using geometric morphometric methods for description of overall shape, and a set of morphological traits were measured (total length, number of lateral plates on both sides). From these data we will determine if and how phenotypic traits and allele frequencies vary across environments that differ in salinity (brackish and freshwater) and in physical characteristics (river vs. lake, benthic vs. limnetic). Using the neutral microsatellites we will test for gene flow and if this is correlated with the Eda-genotype through the system. According to the theory of standing genetic variation in the ocean, we expect there to be a higher variation in plate morphs and Eda alleles where the gene flow is highest. The frequency of the low plated morph tends to increase with distance into freshwater. We hypothesize that the shape analysis and the QTLs will group the fish according to their morph but that neutral markers will group the fish according to geography (distance) rather than phenotype.