The evolution of adaptive phenotypic divergence requires heritable genetic variation. However, it is underappreciated that trait heritability is molded by developmental processes interacting with the environment. We hypothesized that the genetic architecture of divergent functional traits was dependent on age and foraging environment. Thus, we induced plasticity in full-sib families of Arctic charr (Salvelinus alpinus) morphs from two Icelandic lakes by mimicking prey variation in the wild. We characterized variation in body shape and size at two ages and investigated their genetic architecture with quantitative trait locus (QTL) analysis. Age had a greater effect on body shape than diet in most families, suggesting that development strongly influences phenotypic variation available for selection. Consistent with our hypothesis, multiple QTL were detected for all traits and their location depended on age and diet. Many of the genome-wide QTL were located within a subset of duplicated chromosomal regions suggesting that ancestral whole genome duplication events have played a role in the genetic control of functional morphological variation in the species. Moreover, the detection of two body shape QTL after controlling for the effects of age provides additional evidence for genetic variation in the plastic response of morphological traits to environmental variation. Thus, functional morphological traits involved in phenotypic divergence are molded by complex genetic interactions with development and environment.

Chromosomal rearrangements (CRs) can play an important role in evolutionary diversification by preserving linkage among favourable alleles through reduced recombination and/or by reducing hybrid fitness due to genomic incompatibilities. Our goal was to determine to what extent CRs contribute to known patterns of genetic variation in Arctic charr (Salvelinus alpinus). To address this goal, we compared genetic linkage maps to identify whole arm CRs and smaller scale structural variants (SVs) such as translocations/transpositions and inversions found in groups of populations that reflect the temporal sequence of geographic isolation events. If CRs contribute to genetic differentiation, we expected that CRs would be specific to glacial lineages, geographic clusters of populations within lineages, and sympatric morphs. We detected fusions and fissions of whole chromosome arms and SV involving translocations/transpositions of the sex-determining gene (sdY) and inversions. Several CRs were shared across populations from the Arctic and Atlantic glacial lineages, Canadian and Icelandic populations within the Atlantic lineage, between two Icelandic populations and sympatric morphs within Icelandic populations, suggesting that their origin predates geographic isolation in glacial refugia. Other CRs were specific to single populations, which suggests a more recent origin of these variants in refugia, during post-glacial recolonization and/or in contemporary populations. Thus, CRs contribute relatively little to known patterns of genetic differentiation at different geographic scales but represent a pool of standing genetic variation for evolution.

Understanding the mechanisms that underlie the adaptive response of ectotherms to rising temperatures is key to mitigate the effects of climate change. We assessed the molecular and physiological processes that differentiate between rainbow trout (Oncorhynchus mykiss) with high and low tolerance to acute thermal stress. To achieve our goal, we used a critical thermal maximum trial in two strains of rainbow trout to elicit loss of equilibrium responses to identify high and low tolerance fish. We then compared the hepatic transcriptome profiles of high and low tolerance fish relative to untreated controls common to both strains to uncover patterns of differential gene expression and to gain a broad perspective on the interacting gene pathways and functional processes involved. We observed some of the classic responses to increased temperature (e.g., induction of heat shock proteins) but these responses were not the defining factors that differentiated high and low tolerance fish. Instead, high tolerance fish appeared to suppress growth-related functions, enhance certain autophagy components, better regulate neurodegenerative processes, and enhance stress-related protein synthesis, specifically spliceosomal complex activities, mRNA regulation, and protein processing through post-translational processes, relative to low tolerance fish. In contrast, low tolerance fish had higher transcript diversity and demonstrated elevated developmental, cytoskeletal, and morphogenic, as well as lipid and carbohydrate metabolic processes, relative to high tolerance fish. Our results suggest that high tolerance fish engaged in processes that supported the prevention of further damage by enhancing repair pathways, whereas low tolerance fish were more focused on replacing damaged cells and their structures.

Higher water temperatures and pathogens are both significant factors that negatively affect the welfare of teleost fish. In aquaculture, compared to natural populations, these problems are especially exacerbated, as the animals have relatively limited mobility, and the higher density promotes faster spread of infectious diseases. Because of the potential harm these stressors can inflict, methods that can limit the damage of these stressors are particularly valuable. As a method of interest, early-life thermal preconditioning of animals demonstrated some potential for effective improvements in thermotolerance. However, the potential effects of the method on the immune system via the heat-stress model have not been explored. In this experiment, juvenile-stage thermal preconditioned rainbow trout (Oncorhynchus mykiss) were subjected to a secondary thermal challenge, animals were collected and sampled at the time of lost equilibrium. The effects of preconditioning on the general stress response was assessed by measuring the plasma cortisol levels. In addition, we also examined hsp70 and hsc70 mRNA levels in the spleen and gill tissues, as well as IL-1β, IL-6, TNF-α, IFN-1, β2m, and MH class I transcripts via qRT-PCR. No changes in CTmax were observed between the preconditioned and control cohorts upon the second challenge. IL-1β and IL-6 transcripts were generally upregulated with increased temperature of the secondary thermal challenge, whereas IFN-1 transcripts were upregulated in the spleen, but downregulated in the gills, along with MH class I. The juvenile thermal preconditioning produced a series of changes in transcript levels for IL-1β, TNF-α, IFN-1, and hsp70 but the dynamics of these differences were inconsistent. Finally, analysis of plasma cortisol levels presented significantly lower cortisol levels in the pre-conditioned animals compared to the non-pre-conditioned control cohort.

Biodensity is a major factor affecting the production and welfare of farmed fishes. Arctic charr (Salvelinus alpinus) (average mass 176.9 ± 3.9 g) were held at biodensities of 30, 60, 90, 120, and 150 kg/m3 (4 replicates per treatment) during a 91 day study which examined key growth, stress physiology, and welfare parameters. During experimentation fish were fed to near satiety, and a random subsample of 20 fish (5 per replicate tank) were collected from each treatment every 21 days. Biodensity was found to have no significant effect on mortality rates or physical fin damage. Growth rates were lower in charr reared at the highest biodensities (120, and 150 kg/m3), while feed efficiency was negatively affected at both the highest (120, and 150 kg/m3) and lowest (30 kg/m3) biodensities. Plasma cortisol indicated that Arctic charr are more stressed at lower biodensities, but was not correlated with growth or feed efficiency measures. The results support an optimal biodensity range for charr culture between 60 and 90 kg/m3 to optimize production and welfare.

We have generated a high-density, high-throughput genotyping array for characterizing genome-wide variation in Arctic charr (Salvelinus alpinus). Novel single nucleotide polymorphisms (SNPs) were identified in charr from the Fraser, Nauyuk and Tree River aquaculture strains, which originated from northern Canada and fish from Iceland using high coverage sequencing, reduced representation sequencing and RNA-seq datasets. The array was designed to capture genome-wide variation from a diverse suite of Arctic charr populations. Cross validation of SNPs from various sources and comparison with previously published Arctic charr SNP data provided a set of candidate SNPs that generalize across populations. Further candidate SNPs were identified based on minor allele frequency, association with RNA transcripts, even spacing across intergenic regions and association with the sex determining (sdY) gene. The performance of the 86,503 SNP array was assessed by genotyping Fraser, Nauyuk and Tree River strain individuals, as well as wild Icelandic Arctic charr. Overall, 63,060 of the SNPs were polymorphic within at least one group and 36.8% were unique to one of the four groups, suggesting that the array design allows for characterization of both within and across population genetic diversity. The concordance between sdY markers and known phenotypic sex indicated that the array can accurately determine the sex of individuals based on genotype alone. The Salp87k genotyping array provides researchers and breeders the opportunity to analyze genetic variation in Arctic charr at a more detailed level than previously possible.

Antiparasitic drugs such as emamectin benzoate (EMB) are relied upon to reduce the parasite load, particularly of the sea louse Lepeophtheirus salmonis, on farmed salmon. The decline in EMB treatment efficacy for this purpose is an important issue for salmon producers around the world, and particularly for those in the Atlantic Ocean where widespread EMB tolerance in sea lice is recognized as a significant problem. Salmon farms in the Northeast Pacific Ocean have not historically experienced the same issues with treatment efficacy, possibly due to the relatively large population of endemic salmonid hosts that serve to both redistribute surviving lice and dilute populations potentially under selection by introducing naïve lice to farms. Frequent migration of lice among farmed and wild hosts should limit the effect of farm-specific selection pressures on changes to the overall allele frequencies of sea lice in the Pacific Ocean. A previous study using microsatellites examined L. salmonis oncorhynchi from 10 Pacific locations from wild and farmed hosts and found no population structure. Recently however, a farm population of sea lice was detected where EMB bioassay exposure tolerance was abnormally elevated. In response, we have developed a Pacific louse draft genome that complements the previously-released Atlantic louse sequence. These genomes were combined with whole-genome re-sequencing data to design a highly sensitive 201,279 marker SNP array applicable for both subspecies (90,827 validated Pacific loci; 153,569 validated Atlantic loci). Notably, kmer spectrum analysis of the re-sequenced samples indicated that Pacific lice exhibit a large within-individual heterozygosity rate (average of 1 in every 72 bases) that is markedly higher than that of Atlantic individuals (1 in every 173 bases). The SNP chip was used to produce a high-density map for Atlantic sea louse linkage group 5 that was previously shown to be associated with EMB tolerance in Atlantic lice. Additionally, 478 Pacific louse samples from farmed and wild hosts obtained between 2005 and 2014 were also genotyped on the array. Clustering analysis allowed us to detect the apparent emergence of an otherwise rare genotype at a high frequency among the lice collected from two farms in 2013 that had reported elevated EMB tolerance. This genotype was not observed in louse samples collected from the same farm in 2010, nor in any lice sampled from other locations prior to 2013. However, this genotype was detected at low frequencies in louse samples from farms in two locations reporting elevated EMB tolerance in 2014. These results suggest that a rare genotype present in Pacific lice may be locally expanded in farms after EMB treatment. Supporting this hypothesis, 437 SNPs associated with this genotype were found to be in a region of linkage group 5 that overlaps the region associated with EMB resistance in Atlantic lice. Finally, five of the top diagnostic SNPs within this region were used to screen lice that had been subjected to an EMB survival assay, revealing a significant association between these SNPs and EMB treatment outcome. To our knowledge this work is the first report to identify a genetic link to altered EMB efficacy in L. salmonis in the Pacific Ocean.

Arctic charr have a circumpolar distribution, persevere under extreme environmental conditions, and reach ages unknown to most other salmonids. The Salvelinus genus is primarily composed of species with genomes that are structured more like the ancestral salmonid genome than most Oncorhynchus and Salmo species of sister genera. It is thought that this aspect of the genome may be important for local adaptation (due to increased recombination) and anadromy (the migration of fish from saltwater to freshwater). In this study, we describe the generation of a new genetic map, the sequencing and assembly of the Arctic charr genome (GenBank accession: GCF_002910315.2) using the newly created genetic map and a previous genetic map, and present several analyses of the Arctic charr genes and genome assembly. The newly generated genetic map consists of 8,574 unique genetic markers and is similar to previous genetic maps with the exception of three major structural differences. The N50, identified BUSCOs, repetitive DNA content, and total size of the Arctic charr assembled genome are all comparable to other assembled salmonid genomes. An analysis to identify orthologous genes revealed that a large number of orthologs could be identified between salmonids and many appear to have highly conserved gene expression profiles between species. Comparing orthologous gene expression profiles may give us a better insight into which genes are more likely to influence species specific phenotypes.

A salmon louse (Lepeophtheirus salmonis salmonis) genetic linkage map was constructed to serve as a genomic resource for future investigations into the biology of this important marine parasitic copepod species, and to provide insights into the inheritance patterns of genetic markers in this species. SNP genotyping of 8 families confirmed the presence of 15 linkage groups based upon the assignment of 93,773 markers. Progeny sample size weight adjusted map sizes in males (with the exception of SL12 and SL15) ranged in size from 96.50cM (SL11) to 134.61cM (SL06), and total combined map steps or bins ranged from 143 (SL09) to 203 (SL13). The SL12 male map was the smallest linkage group with a weight-averaged size of 3.05cM with 6 recombination bins. Male:female specific recombination rate differences are 10.49:1 and represent one of the largest reported sex-specific differences for any animal species. Recombination ratio differences (M:F) ranged from 1.0 (SL12) to 29:1 (SL15). The number of markers exhibiting normal Mendelian segregation within the sex linkage group SL15 was extremely low (N=80) in comparison to other linkage groups genotyped [range: 1459 (SL12)10206 markers (SL05)]. Re-evaluation of Mendelian inheritance patterns of markers unassigned to any mapping parent according to hemizygous segregation patterns (models presented) identified matches for many of these markers to hemizygous patterns. The greatest proportion of these markers assigned to SL15 (N increased to 574). Inclusion of the hemizygous markers revised SL15 sex-specific recombination rate differences to 28:1. Recombination hot- and coldspots were identified across all linkage groups with all linkage groups possessing multiple peaks. Nine of 13 linkage groups evaluated possessed adjacent domains with hot-coldspot transitional zones. The most common pattern was for one end of the linkage to show elevated recombination in addition to internal regions. For SL01 and SL06, however, a terminal region with high recombination was not evident while a central domain possessing extremely high-recombination levels was present. High levels of recombination were weakly coupled to higher levels of SNP variation within domains, but this association was very strong for the central domains of SL01 and SL06. From the pooled paternal half-sib lots (several virgin females placed with 1 male), only 1 or two surviving family lots were obtained. Surviving families possessed parents where both the male and female possessed either inherently low or high recombination rates. This study provides insight into the organization of the sea louse genome, and describes large differences in recombination rate that exist among individuals of the same sex, and between the sexes. These differences in recombination rate may be coupled to the capabilities of this species to adapt to environmental and pharmaceutical treatments, given that family survivorship appears to be enhanced when parents have similar recombination levels.

Diploidization, which follows whole genome duplication events, does not occur evenly across the genome. In salmonid fishes, certain pairs of homeologous chromosomes preserve tetraploid loci in higher frequencies toward the telomeres due to residual tetrasomic inheritance. Research suggests this occurs only in homeologous pairs where one chromosome arm has undergone a fusion event. We present a linkage map for Arctic charr (Salvelinus alpinus), a salmonid species with relatively fewer chromosome fusions. Genotype by sequencing identified 19,418 SNPs, and a linkage map consisting of 4508 markers was constructed from a subset of high quality SNPs and microsatellite markers that were used to anchor the new map to previous versions. Both male- and female-specific linkage maps contained the expected number of 39 linkage groups. The chromosome type associated with each linkage group was determined, and 10 stable metacentric chromosomes were identified, along with a chromosome polymorphism involving the sex chromosome AC04. Two instances of a weak form of pseudolinkage were detected in the telomeric regions of homeologous chromosome arms in both female and male linkage maps. Chromosome arm homologies within the Atlantic salmon (Salmo salar) and rainbow trout (Oncorhynchus mykiss) genomes were determined. Paralogous sequence variants (PSVs) were identified, and their comparative BLASTn hit locations showed that duplicate markers exist in higher numbers on seven pairs of homeologous arms, previously identified as preserving tetrasomy in salmonid species. Homeologous arm pairs where neither arm has been part of a fusion event in Arctic charr had fewer PSVs, suggesting faster diploidization rates in these regions.

The effects of marker assisted selection for spawning date in males on embryonic developmental rate, growth related traits and precocious male maturation and phenotypic and genetic associations among traits were investigated in a commercial strain of rainbow trout (Oncorhynchus mykiss). The relationship between spawning date category of the sire (early or late) and developmental rate as measured by hatching time was inconsistent among families. However, two half-sib families produced by males with an allelic complement associated with late spawning date had higher rates of precocious male maturation and faster developmental rates than two families produced by males with earlier spawning date alleles. QTL analysis suggests that some of the co-variation between spawning date and developmental rate has a genetic basis as two QTL for developmental rate as measured by hatching time co-localized to the same markers known to be associated with spawning date in this strain. We observed no association between the spawning date category of the sire and progeny length at 61 days post-fertilization (alevin stage), body weight at 381 days (juveniles), body weight at 614 days and specific growth rate, which suggests that selection for early or late spawning date alleles will have little effect on growth traits. Within families, faster developmental rates conferred a size advantage at 61 days post-fertilization but not at 381 and 614 days. Moreover, no correlation between developmental rate and the propensity of early male maturation was detected within families. Males that matured at 2 years were significantly heavier at both 381 and 614 days and had higher specific growth rates than immature males and females. Likewise, body weight at 614 days and family were significant predictors for the propensity for early sexual maturation in males. This indicates that selection for greater body size will likely result in higher rates of unwanted precocious male maturation. Together, these results suggest that co-variation among economically important fitness traits may present potential trade-offs impacting selection goals.Relevance: Addresses feasibility of using molecular approaches in selective breeding programs. (C) 2016 Elsevier B.V. All rights reserved.

Catfish represent 12% of teleost or 6.3% of all vertebrate species, and are of enormous economic value. Here we report a high-quality reference genome sequence of channel catfish ( Ictalurus punctatus ), the major aquaculture species in the US. The reference genome sequence was validated by genetic mapping of 54,000 SNPs, and annotated with 26,661 predicted protein-coding genes. Through comparative analysis of genomes and transcriptomes of scaled and scaleless fish and scale regeneration experiments, we address the genomic basis for the most striking physical characteristic of catfish, the evolutionary loss of scales and provide evidence that lack of secretory calcium-binding phosphoproteins accounts for the evolutionary loss of scales in catfish. The channel catfish reference genome sequence, along with two additional genome sequences and transcriptomes of scaled catfishes, provide crucial resources for evolutionary and biological studies. This work also demonstrates the power of comparative subtraction of candidate genes for traits of structural significance.

Background Circannual rhythms in vertebrates can influence a wide variety of physiological processes. Some notable examples include annual reproductive cycles and for poikilotherms, seasonal changes modulating growth. Increasing water temperature elevates growth rates in fishes, but increases in photoperiod regime can have similar influences even at constant temperature. Therefore, in order to understand the dynamics of growth in fish it is important to consider the background influence of photoperiod regime on gene expression differences. This study examined the influence of a declining photoperiod regime (winter solstice) compared to an increasing photoperiod regime (spring equinox) on white muscle transcriptome profiles in fast and slow-growing rainbow trout from a commercial aquaculture strain. Results Slow-growing fish could be characterized as possessing transcriptome profiles that conform in many respects to an endurance training regime in humans. They have elevated mitochondrial and cytosolic creatine kinase expression levels and appear to suppress mTOR-signaling as evidenced by elevated TSC2 expression, and they also have elevated p53 levels. Large fish display a physiological repertoire that may be consistent with strength/resistance physiology having elevated cytoskeletal gene component expression and glycogen metabolism cycling along with higher PI3K levels. In many respects small vs. large fish match eccentric vs. concentric muscle expression patterns, respectively. Lipid metabolic genes are also more elevated in larger fish, the most notable being the G0S2 switch gene. M and Z-line sarcomere remodelling appears to be more prevalent in large fish. Twenty-three out of 26 gene families with previously reported significant SNP-based growth differences were detected as having significant expression differences. Conclusions Larger fish display a broader array of genes showing higher expression, and their profiles are more similar to those observed in December lot fish (i.e., an accelerated growth period). Conversely, small fish display gene profiles more similar to seasonal growth decline phases (i.e., September lot fish). Overall, seasonal timing was coupled to greater differences in gene expression compared to differences associated with fish size.

We tested whether genes differentially expressed between large and small rainbow trout co-localized with familial QTL regions for body size. Eleven chromosomes, known from previous work to house QTL for weight and length in rainbow trout, were examined for QTL in half-sibling families produced in September (1 XY male and 1 XX neomale) and December (1 XY male). In previous studies, we identified 108 candidate genes for growth expressed in the liver and white muscle in a subset of the fish used in this study. These gene sequences were BLASTN aligned against the rainbow trout and stickleback genomes to determine their location (rainbow trout) and inferred location based on synteny with the stickleback genome. Across the progeny of all three males used in the study, 63.9% of the genes with differential expression appear to co-localize with the QTL regions on 6 of the 11 chromosomes tested in these males. Genes that co-localized with QTL in the mixed-sex offspring of the two XY males primarily showed up-regulation in the muscle of large fish and were related to muscle growth, metabolism, and the stress response.

All-female lines of fish are created by crossing sex reversed (XX genotype) males with normal females. All-female lines avoid the deleterious phenotypic effects that are typical of precocious maturation in males. To determine whether all-female and mixed sex populations of rainbow trout (Oncorhynchus mykiss) differ in performance, we compared the growth and gene expression profiles in progeny groups produced by crossing a XX male and a XY male to the same five females. Body weight and length were measured in the resulting all-female (XX) and mixed sex (XX/XY) offspring groups. Microarray experiments with liver and white muscle were used to determine if the gene expression profiles of large and small XX offspring differ from those in large and small XX/XY offspring. We detected no significant differences in body length and weight between offspring groups but XX offspring were significantly less variable in the value of these traits. A large number of upregulated genes were shared between the large XX and large XX/XY offspring; the small XX and small XX/XY offspring also shared similar expression profiles. No GO category differences were seen in the liver or between the large XX and large XX/XY offspring in the muscle. The greatest differences between the small XX and small XX/XY offspring were in the genes assigned to the "small molecule metabolic process" and "cellular metabolic process" GO level 3 categories. Similarly, genes within these categories as well as the category "macromolecule metabolic process" were more highly expressed in small compared to large XX fish.

Construction of genetic linkage map is essential for genetic and genomic studies. Recent advances in sequencing and genotyping technologies made it possible to generate high-density and high-resolution genetic linkage maps, especially for the organisms lacking extensive genomic resources. In the present work, we constructed a high-density and high-resolution genetic map for channel catfish with three large resource families genotyped using the catfish 250K single-nucleotide polymorphism (SNP) array. A total of 54,342 SNPs were placed on the linkage map, which to our knowledge had the highest marker density among aquaculture species. The estimated genetic size was 3,505.4 cM with a resolution of 0.22 cM for sex-averaged genetic map. The sex-specific linkage maps spanned a total of 4,495.1 cM in females and 2,593.7 cM in males, presenting a ratio of 1.7 : 1 between female and male in recombination fraction. After integration with the previously established physical map, over 87% of physical map contigs were anchored to the linkage groups that covered a physical length of 867 Mb, accounting for ∼90% of the catfish genome. The integrated map provides a valuable tool for validating and improving the catfish whole-genome assembly and facilitates fine-scale QTL mapping and positional cloning of genes responsible for economically important traits.

High-throughput RNA sequencing was used to compare expression profiles in two Arctic charr (Salvelinus alpinus) families post-seawater exposure to identify genes and biological processes involved in hypo-osmoregulation and regulation of salinity tolerance. To further understand the genetic architecture of hypo-osmoregulation, the genomic organization of differentially expressed (DE) genes was also analysed. Using a de novo gill transcriptome assembly we found over 2300 contigs to be DE. Major transporters from the seawater mitochondrion-rich cell (MRC) complex were up-regulated in seawater. Expression ratios for 257 differentially expressed contigs were highly correlated between families, suggesting they are strictly regulated. Based on expression profiles and known molecular pathways we inferred that seawater exposure induced changes in methylation states and elevated peroxynitrite formation in gill. We hypothesized that concomitance between DE immune genes and the transition to a hypo-osmoregulatory state could be related to Cl− sequestration by antimicrobial defence mechanisms. Gene ontology analysis revealed that cell division genes were up-regulated, which could reflect the proliferation of ATP1α1b-type seawater MRCs. Comparative genomics analyses suggest that hypo-osmoregulation is influenced by the relative proximities among a contingent of genes on Arctic charr linkage groups AC-4 and AC-12 that exhibit homologous affinities with a region on stickleback chromosome Ga-I. This supports the hypothesis that relative gene location along a chromosome is a property of the genetic architecture of hypo-osmoregulation. Evidence of non-random structure between hypo-osmoregulation candidate genes was found on AC-1/11 and AC-28, suggesting that interchromosomal rearrangements played a role in the evolution of hypo-osmoregulation in Arctic charr.

Background: Growth in fishes is regulated via many environmental and physiological factors and is shaped by the genetic background of each individual. Previous microarray studies of salmonid growth have examined fish experiencing either muscle wastage or accelerated growth patterns following refeeding, or the influence of growth hormone and transgenesis. This study determines the gene expression profiles of genetically unmanipulated large and small fish from a domesticated salmonid strain reared on a typical feeding regime. Gene expression profiles of white muscle and liver from rainbow trout (Oncorhynchus mykiss) from two seasonal spawning groups (September and December lots) within a single strain were examined when the fish were 15 months of age to assess the influence of season (late fall vs. onset of spring) and body size (large vs. small).Results: Although IGFBP1 gene expression was up-regulated in the livers of small fish in both seasonal lots, few expression differences were detected in the liver overall. Faster growing Dec. fish showed a greater number of differences in white muscle expression compared to Sept. fish. Significant differences in the GO Generic Level 3 categories 'response to external stimulus', 'establishment of localization', and 'response to stress' were detected in white muscle tissue between large and small fish. Larger fish showed up-regulation of cytoskeletal component genes while many genes related to myofibril components of muscle tissue were up-regulated in small fish. Most of the genes up-regulated in large fish within the 'response to stress' category are involved in immunity while in small fish most of these gene functions are related to apoptosis.Conclusions: A higher proportion of genes in white muscle compared to liver showed similar patterns of up-or down-regulation within the same size class across seasons supporting their utility as biomarkers for growth in rainbow trout. Differences between large and small Sept. fish in the 'response to stress' and 'response to external stimulus' categories for white muscle tissue, suggests that smaller fish have a greater inability to handle stress compared to the large fish. Sampling season had a significant impact on the expression of genes related to the growth process in rainbow trout.