Identification of Hypoxia-Specific Biomarkers in Salmonids Using RNA-Sequencing and Validation Using High-Throughput qPCR.

Identifying early gene expression responses to hypoxia (i.e., low dissolved oxygen) as a tool to assess the degree of exposure to this stressor is crucial for salmonids, because they are increasingly exposed to hypoxic stress due to anthropogenic habitat change, e.g., global warming, excessive nutrient loading, and persistent algal blooms. Our goal was to discover and validate gill gene expression biomarkers specific to the hypoxia response in salmonids across multi-stressor conditions. Gill tissue was collected from 24 freshwater juvenile Chinook salmon (Oncorhynchus tshawytscha), held in normoxia [dissolved oxygen (DO) > 8 mg L-1] and hypoxia (DO = 4‒5 mg L-1) in 10 and 18° temperatures for up to six days. RNA-sequencing (RNA-seq) was then used to discover 240 differentially expressed genes between hypoxic and normoxic conditions, but not affected by temperature. The most significantly differentially expressed genes had functional roles in the cell cycle and suppression of cell proliferation associated with hypoxic conditions. The most significant genes (n = 30) were selected for real-time qPCR assay development. These assays demonstrated a strong correlation (r = 0.88; P < 0.001) between the expression values from RNA-seq and the fold changes from qPCR. Further, qPCR of the 30 candidate hypoxia biomarkers was applied to an additional 322 Chinook salmon exposed to hypoxic and normoxic conditions to reveal the top biomarkers to define hypoxic stress. Multivariate analyses revealed that smolt stage, water salinity, and morbidity status were relevant factors to consider with the expression of these genes in relation to hypoxic stress. These hypoxia candidate genes will be put into application screening Chinook salmon to determine the identity of stressors impacting the fish.

Discovery and validation of candidate smoltification gene expression biomarkers across multiple species and ecotypes of Pacific salmonids.

Early marine survival of juvenile salmon is intimately associated with their physiological condition during smoltification and ocean entry. Smoltification (parr-smolt transformation) is a developmental process that allows salmon to acquire seawater tolerance in preparation for marine living. Traditionally, this developmental process has been monitored using gill Na+/K+-ATPase (NKA) activity or plasma hormones, but gill gene expression offers the possibility of another method. Here, we describe the discovery of candidate genes from gill tissue for staging smoltification using comparisons of microarray studies with particular focus on the commonalities between anadromous Rainbow trout and Sockeye salmon datasets, as well as a literature comparison encompassing more species. A subset of 37 candidate genes mainly from the microarray analyses was used for TaqMan quantitative PCR assay design and their expression patterns were validated using gill samples from four groups, representing three species and two ecotypes: Coho salmon, Sockeye salmon, stream-type Chinook salmon and ocean-type Chinook salmon. The best smoltification biomarkers, as measured by consistent changes across these four groups, were genes involved in ion regulation, oxygen transport and immunity. Smoltification gene expression patterns (using the top 10 biomarkers) were confirmed by significant correlations with NKA activity and were associated with changes in body brightness, caudal fin darkness and caudal peduncle length. We incorporate gene expression patterns of pre-smolt, smolt and de-smolt trials from acute seawater transfers from a companion study to develop a preliminary seawater tolerance classification model for ocean-type Chinook salmon. This work demonstrates the potential of gene expression biomarkers to stage smoltification and classify juveniles as pre-smolt, smolt or de-smolt.

Salmonid gene expression biomarkers indicative of physiological responses to changes in salinity and temperature, but not dissolved oxygen.

An organism's ability to respond effectively to environmental change is critical to its survival. Yet, life stage and overall condition can dictate tolerance thresholds to heightened environmental stressors, such that stress may not be equally felt across individuals and at all times. Also, the transcriptional responses induced by environmental changes can reflect both generalized responses as well as others that are highly specific to the type of change being experienced. Thus, if transcriptional biomarkers specific to a stressor, even under multi-stressor conditions, can be identified, the biomarkers could then be applied in natural environments to determine when and where an individual experiences such a stressor. Here, we experimentally challenged juvenile Chinook salmon (Oncorhynchus tshawytscha) to validate candidate gill gene expression biomarkers. A sophisticated experimental design manipulated salinity (freshwater, brackish water and seawater), temperature (10, 14 and 18°C) and dissolved oxygen (normoxia and hypoxia) in all 18 possible combinations for 6 days using separate trials for three smolt statuses (pre-smolt, smolt and de-smolt). In addition, changes in juvenile behaviour, plasma variables, gill Na+/K+-ATPase activity, body size, body morphology and skin pigmentation supplemented the gene expression responses. We identified biomarkers specific to salinity and temperature that transcended the multiple stressors, smolt status and mortality (live, dead and moribund). Similar biomarkers for dissolved oxygen were not identified. This work demonstrates the unique power of gene expression biomarkers to identify a specific stressor even under multi-stressor conditions, and we discuss our next steps for hypoxia biomarkers using an RNA-seq study.

Transcriptional shifts during juvenile Coho salmon (Oncorhynchus kisutch) life stage changes in freshwater and early marine environments.

There is a paucity of information on the physiological changes that occur over the course of salmon early marine migration. Here we aim to provide insight on juvenile Coho salmon (Oncorhynchus kisutch) physiology using the changes in gene expression (cGRASP 44K microarray) of four tissues (brain, gill, muscle, and liver) across the parr to smolt transition in freshwater and through the first eight months of ocean residence. We also examined transcriptome changes with body size as a covariate. The strongest shift in the transcriptome for brain, gill, and muscle occurred between summer and fall in the ocean, representing physiological changes that we speculate may be associated with migration preparation to feeding areas. Metabolic processes in the liver were positively associated with body length, generally consistent with enhanced feeding opportunities. However, a notable exception to this metabolic pattern was for spring post-smolts sampled soon after entry into the ocean, which showed a pattern of gene expression more likely associated with depressed feeding or recent fasting. Overall, this study has revealed life stages that may be the most critical developmentally (fall post-smolt) and for survival (spring post-smolt) in the early marine environment. These life stages may warrant further investigation.

Transcriptome response of Atlantic salmon (Salmo salar) to competition with ecologically similar non-native species.

Non-native species may be introduced either intentionally or unintentionally, and their impact can range from benign to highly disruptive. Non-native salmonids were introduced into Lake Ontario, Canada, to provide recreational fishing opportunities; however, the establishment of those species has been proposed as a significant barrier to the reintroduction of native Atlantic salmon (Salmo salar) due to intense interspecific competition. In this study, we compared population differences of Atlantic salmon in transcriptome response to interspecific competition. We reared Atlantic salmon from two populations (LaHave River and Sebago Lake) with fish of each of three non-native salmonids (Chinook salmon Oncorhynchus tshawytscha, rainbow trout O. mykiss, and brown trout S. trutta) in artificial streams. We used RNA-seq to assess transcriptome differences between the Atlantic salmon populations and the responses of these populations to the interspecific competition treatments after 10 months of competition in the stream tanks. We found that population differences in gene expression were generally greater than the effects of interspecific competition. Interestingly, we found that the two Atlantic salmon populations exhibited similar responses to interspecific competition based on functional gene ontologies, but the specific genes within those ontologies were different. Our transcriptome analyses suggest that the most stressful competitor (as measured by the highest number of differentially expressed genes) differs between the two study populations. Our transcriptome characterization highlights the importance of source population selection for conservation applications, as organisms with different evolutionary histories can possess different transcriptional responses to the same biotic stressors. The results also indicate that generalized predictions of the response of native species to interactions with introduced species may not be appropriate without incorporating potential population-specific response to introduced species.

fullfact: an R package for the analysis of genetic and maternal variance components from full factorial mating designs.

Full factorial breeding designs are useful for quantifying the amount of additive genetic, nonadditive genetic, and maternal variance that explain phenotypic traits. Such variance estimates are important for examining evolutionary potential. Traditionally, full factorial mating designs have been analyzed using a two-way analysis of variance, which may produce negative variance values and is not suited for unbalanced designs. Mixed-effects models do not produce negative variance values and are suited for unbalanced designs. However, extracting the variance components, calculating significance values, and estimating confidence intervals and/or power values for the components are not straightforward using traditional analytic methods. We introduce fullfact - an R package that addresses these issues and facilitates the analysis of full factorial mating designs with mixed-effects models. Here, we summarize the functions of the fullfact package. The observed data functions extract the variance explained by random and fixed effects and provide their significance. We then calculate the additive genetic, nonadditive genetic, and maternal variance components explaining the phenotype. In particular, we integrate nonnormal error structures for estimating these components for nonnormal data types. The resampled data functions are used to produce bootstrap-t confidence intervals, which can then be plotted using a simple function. We explore the fullfact package through a worked example. This package will facilitate the analyses of full factorial mating designs in R, especially for the analysis of binary, proportion, and/or count data types and for the ability to incorporate additional random and fixed effects and power analyses.

Risk assessment of inbreeding and outbreeding depression in a captive-breeding program.

Captive-breeding programs can be implemented to preserve the genetic diversity of endangered populations such that the controlled release of captive-bred individuals into the wild may promote recovery. A common difficulty, however, is that programs are founded with limited wild broodstock, and inbreeding can become increasingly difficult to avoid with successive generations in captivity. Program managers must choose between maintaining the genetic purity of populations, at the risk of inbreeding depression, or interbreeding populations, at the risk of outbreeding depression. We evaluate these relative risks in a captive-breeding program for 3 endangered populations of Atlantic salmon (Salmo salar). In each of 2 years, we released juvenile F(1) and F(2) interpopulation hybrids, backcrosses, as well as inbred and noninbred within-population crosstypes into 9 wild streams. Juvenile size and survival was quantified in each year. Few crosstype effects were observed, but interestingly, the relative fitness consequences of inbreeding and outbreeding varied from year to year. Temporal variation in environmental quality might have driven some of these annual differences, by exacerbating the importance of maternal effects on juvenile fitness in a year of low environmental quality and by affecting the severity of inbreeding depression differently in different years. Nonetheless, inbreeding was more consistently associated with a negative effect on fitness, whereas the consequences of outbreeding were less predictable. Considering the challenges associated with a sound risk assessment in the wild and given that the effect of inbreeding on fitness is relatively predictable, we suggest that risk can be weighted more strongly in terms of the probable outcome of outbreeding. Factors such as genetic similarities between populations and the number of generations in isolation can sometimes be used to assess outbreeding risk, in lieu of experimentation.

Consequences of farmed-wild hybridization across divergent wild populations and multiple traits in salmon.

Theory predicts that hybrid fitness should decrease as population divergence increases. This suggests that the effects of human-induced hybridization might be adequately predicted from the known divergence among parental populations. We tested this prediction by quantifying trait differentiation between multigenerational crosses of farmed Atlantic salmon (Salmo salar) and divergent wild populations from the Northwest Atlantic; the former escape repeatedly into the wild, while the latter are severely depleted. Under common environmental conditions and at the spatiotemporal scale considered (340 km, 12 000 years of divergence), substantial cross differentiation had a largely additive genetic basis at behavioral, life history, and morphological traits. Wild backcrossing did not completely restore hybrid trait distributions to presumably more optimal wild states. Consistent with theory, the degree to which hybrids deviated in absolute terms from their parental populations increased with increasing parental divergence (i.e., the collective environmental and life history differentiation, genetic divergence, and geographic distance between parents). Nevertheless, while these differences were predictable, their implications for risk assessment were not: wild populations that were equally divergent from farmed salmon in the total amount of divergence differed in the specific traits at which this divergence occurred. Combined with ecological data on the rate of farmed escapes and wild population trends, we thus suggest that the greatest utility of hybridization data for risk assessment may be through their incorporation into demographic modeling of the short- and long-term consequences to wild population persistence. In this regard, our work demonstrates that detailed hybridization data are essential to account for life-stage-specific changes in phenotype or fitness within divergent but interrelated groups of wild populations. The approach employed here will be relevant to risk assessments in a range of wild species where hybridization with domesticated relatives is a concern, especially where the conservation status of the wild species may preclude direct fitness comparisons in the wild.