Understanding how organisms function at the level of gene expression is becoming increasingly important for both ecological and evolutionary studies. It is evident that the diversity and complexity of organisms are not dependent solely on their number of genes, but also the variability in gene expression and gene interactions. Furthermore, slight differences in transcription control can fundamentally affect the fitness of the organism in a variable environment or during development. In this study, triploid and diploid Chinook salmon (Oncorhynchus tshawytscha) were used to examine the effects of polyploidy on specific and genome-wide gene expression response using quantitative real-time PCR (qRT-PCR) and microarray technology after an immune challenge with the pathogen Vibrio anguillarum. Although triploid and diploid fish had significant differences in mortality, qRT-PCR revealed no differences in cytokine gene expression response (interleukin-8, interleukin-1, interleukin-8 receptor and tumor necrosis factor), whereas differences were observed in constitutively expressed genes, (immunoglobulin (Ig) M, major histocompatibility complex (MHC) -II and b-actin) upon live Vibrio anguillarum exposure. Genome-wide microarray analysis revealed that, overall, triploid gene expression is similar to diploids, consistent with their similar phenotypes. This pattern, however, can subtly be altered under stress (for example, handling, V. anguillarum challenge) as we have observed at some housekeeping genes. Our results are the first report of dosage effect on gene transcription in a vertebrate, and they support the observation that diploid and triploid salmon are generally phenotypically indistinguishable, except under stress, when triploids show reduced performance.

Using different classes of genetic markers can provide insight into the role of selection, as well as a broader context for identifying population differentiation. We used nine microsatellite loci and polymorphisms at eight gene loci (major histocompatibility complex (MHC) classes I and II, growth hormones 1 and 2, transferrin, and immunoglobin heavy-chain) to determine population structure in six coastal populations (Vancouver Island, VI) and five interior populations (Fraser River, FR) of Chinook salmon (Oncorhynchus tshawytscha) in British Columbia, Canada. FST and [Formula: see text] values for specific VI gene loci were significantly higher than those for the FR and VI microsatellite loci or the FR gene loci. Pairwise microsatellite FST values were correlated with geographic distance across regions, but not using the gene locus marker data. Neighbor-joining cluster analyses showed one VI population as particularly divergent based on the gene locus data, while the VI and FR microsatellite locus and the FR gene locus analyses yielded no anomalous population divergence. The VI MHC class II marker FST values were exceptionally high, indicative of probable directional selection acting on MHC class II. Our results are consistent with local adaptation in Chinook salmon, but the nature of the local adaptation likely differs among regions.

Rainbow trout Oncorhynchus mykiss in coastal areas of North America occur as two divergent migratory types: steelhead that migrate to salt water and return to their natal river to spawn and resident rainbow trout that either do not migrate or migrate locally within the freshwater system. We extracted DNA from sympatrically occurring steelhead and rainbow trout collected from five major drainages in British Columbia. We used three types of genetic markers to test for differentiation within the sympatric population: microsatellite DNA markers, gene intron polymorphisms (restriction fragment length polymorphisms [RFLPs] and fragment size polymorphisms), and major histocompatibility complex (MHC) exon polymorphisms (RFLPs). The migratory and resident forms were genetically differentiated in only one sympatric population based on microsatellite data, but no significant differentiation was found using the combined gene locus marker data. Overall, most of the observed allele variation at all marker types was attributable to among-drainage variance, while migratory life history contributed a nonsignificant component based on hierarchical analysis of molecular variance. Interestingly, the migratory and resident forms were genetically differentiated in two different populations based on the gonadotropin and p53 intron polymorphisms, while no significant genetic differentiation was found between the two migratory types at any of the MHC exon or growth hormone intron-length polymorphisms. Furthermore, the neighbor-joining cluster dendrogram based on the microsatellite markers reflected geographic patterns, while the neighborjoining dendrogram using gene intron and MHC markers did not. Our data indicate that the migratory and resident life histories have probably evolved independently in different drainages in British Columbia. The disagreement between the results of the analyses using the microsatellite and gene locus markers probably reflects differences in selection, mutation, or both, acting at these two types of marker loci. * Corresponding