Skip to main content
NCBI home page
PLOS One logoLink to PLOS One
. 2026 Oct 9;21(10):e0360513. doi: 10.1371/journal.pone.0360513

Gene transcriptional profiling of osmoregulatory stress response in salmonids: Development and assessment of high-throughput multispecies assays

Nicolas Tugui 1, Shahinur S Islam 2,3,*, Daniel D Heath 1,2,*
Editor: Arnar Palsson4
PMCID: PMC13652239  PMID: 42853848

Abstract

Osmoregulatory stress is a critical factor in the welfare and performance of salmonids under captive rearing for aquaculture and conservation, and characterizing cellular responses to that stress may foster better husbandry. One approach for early detection of osmotic stress is gene transcriptional profiling, where a suite of qPCR assays designed for related-function genes is used to determine patterns of transcriptional responses to stress. This study describes the development, assessment and application of 28 Taqman qPCR assays for use on the OpenArray® platform as an osmoregulatory stress transcriptional profiling (OSTP) chip. The assays include key genes involved in osmoregulation (twelve), stress response (four), immune function (five), growth/metabolism (four) and endogenous controls (three). Here, using gill and liver tissues from eight salmonid species across four genera, the OSTP chip’s performance was evaluated. While results showed both tissue- and species-specific transcriptional differences, the OSTP chip has considerable utility for characterizing molecular stress responses. As a test application of the OSTP chip, we performed a 24-hour saltwater challenge with Chinook salmon and applied the chip to control and challenged fish. This test confirmed the OSTP chip’s capacity to detect acute osmotic stress-induced molecular genetic signatures. This OSTP chip will be useful for aquaculture practices by fostering better husbandry as well as providing a valuable tool for future studies of the osmoregulatory response of salmonid fishes to environmental stress.

1. Introduction

Osmoregulation in anadromous salmonids is a well-explored aspect of their physiology [1], with extensive research targeting their remarkable ability to adapt to a large range of salinity levels [2]. These species generally display exhibit complex mechanisms for regulating ion balance and water uptake [3], which is key for maintaining osmotic balance while transitioning between freshwater and marine environments through their life cycle [4]. Specifically, juveniles adapt to seawater before migrating to the ocean (smolting; [5]) and upon returning to spawn, salmonids undergo the reverse process of freshwater adaptation [6]. This adaptability involves complex interactions between ion transport, hormone regulation, and cellular osmoregulatory mechanisms [7]. Through decades of study, researchers have characterized key genes and pathways involved in osmoregulation [8], thus providing a solid foundation for molecular genetic analyses of the salmonid osmoregulatory coping mechanisms [9].

Our extensive knowledge of the dynamics of osmoregulation in wild and captive populations of salmonids underscores that the processes can differ depending on environmental context [10], since the fish experience different stressors in the wild versus captivity [11]. One obvious example is freshwater rearing, where the fish are maintained in freshwater for their entire life cycle, which is increasingly common in commercial aquaculture [12]. In enhancement hatchery programs, salmonids are reared in hatcheries and later released into the wild at the smolt stage [13]. As the goal of those programs is to improve the conservation status and productivity of wild salmon stocks, smolt quality and fitness are of high priority [14]. This emphasizes the need for a quick and reliable advanced molecular genetic method to quantify and monitor salmonid osmoregulatory stress in captive and wild populations to enhance captive rearing and management, and conservation protocols, and ultimately, improve animal welfare.

In response to the challenges posed by stress in captive and wild environments, fish culturists, conservationists, and fishery managers are increasingly turning to molecular genetic tools for fish health monitoring [15]. Transcriptional profiling, which is the measurement of transcription of multiple related-function genes, allows broad assessment of candidate gene network responses to environmental stimuli [16]. This approach is powerful because different gene expression pathways may be triggered depending on the genetic background of the fish, and many specific gene functions remain uncertain in non-model organisms such as most fish species [17]. By measuring the transcription patterns of genes involved in osmoregulatory stress, researchers can identify the molecular signatures indicative of stress before fish health is compromised [18]. Such a proactive approach enables fish culturists to implement targeted interventions to modify rearing conditions to alleviate osmoregulatory stress and promote fish welfare [19]. Furthermore, transcriptional profiling provides valuable insights into the adaptive response of salmonids to stress, shedding light on the molecular mechanisms underlying this form of plasticity [20].

Here, we described the development and assessment of a panel of gene transcription TaqmanTM qPCR assays designed for the OpenArray® nanofluidic platform, targeting multiple genes associated with osmoregulatory and related stress responses in salmonid fishes. The panel included assays for osmoregulatory function (twelve), stress response (four), immune function (five), growth and metabolism (four) and three endogenous control genes. This advanced technology offers rapid, high-throughput, and low-cost analyses, providing a valuable tool for monitoring fish health and coping mechanisms and fostering better husbandry [21]. However, while gene expression assays can provide valuable molecular data, they may not always be the fastest approach compared to more immediate physiological markers such as plasma chloride levels. We explored the utility of the OSTP chip by applying it to a controlled experimental acute stress test (24-hour saltwater challenge). The OSTP chip not only advances our basic understanding of the osmoregulatory stress response but also has significant implications for improving aquaculture practices, paving the way for more effective stress management and healthier fish. Future developments may include expanding the assay panel to cover additional stress-related pathways, coupled with commercialization of the protocols for “fee for service” agreements.

2. Methods

2.1. Candidate gene selection

In selecting candidate genes for the OSTP chip, we targeted gene function categories relevant to osmoregulatory stress, ensuring coverage of key physiological responses. We included three endogenous control genes: Elongation factor 1-alpha (EF1A; [22]), Ribosomal protein L13A (RPL13A; [23]), and Glucose-6-phosphate dehydrogenase (G6PD; [24]) to serve as internal normalization factors (Supplementary S1 Table). Normalization in this context adjusts for technical variation such as differences in RNA extraction efficiency, cDNA synthesis yield, and overall reaction efficiency [25]. Specifically, each target gene’s expression is measured relative to one or more stable reference genes, ensuring that the observed differences in mRNA abundance reflect true biological variation rather than technical artifacts. To evaluate the stability of these candidate housekeeping genes, we used RefFinder, which integrates four commonly used methods—Delta CT, BestKeeper, NormFinder, and geNorm. These algorithms compare variation in cycle threshold (CT) values across experimental conditions to identify the most stable reference gene(s). The stability ranking indicated that RPL13A (1.19) was the most stable assay, followed by EF1A (1.41), while G6PD (9.00) did not perform reliably across treatments. Consequently, G6PD was excluded from further analyses. Such variability in certain housekeeping genes is not unexpected, as many traditional reference genes can be downregulated under specific stressors (e.g., osmotic or thermal stress; [26]). For example, G6PD may fluctuate in response to oxidative stress or shifts in metabolic demands [27], making it unsuitable under experimental conditions. For the initial chip performance assessment, we used raw CT values. The ΔCT method was employed specifically for the 24-hour challenge experiment. Further details on verifying endogenous controls and excluding less stable genes—a recognized best practice in transcription studies [28]—are provided in the section describing the 24-hour challenge data.

The 25 target genes selected for osmoregulatory stress detection were primarily based on differential gene expression studies conducted under varying salinity conditions in salmonids. Osmoregulatory gene functions include ion transporters, osmolyte synthesis enzymes, and other proteins for maintaining osmotic balance [29,30]; we selected twelve genes within those functions ([31–38]; see Supplementary S1 Table). We also selected four genes associated with the general stress response ([39–42]; see Supplementary S1 Table) since osmoregulatory stress can activate other stress-related responses, such as the hypothalamic-pituitary-interrenal axis and cortisol production. We included five immune function genes as osmoregulatory stress can impact the immune system, affecting susceptibility to pathogens ([43–47]; see Supplementary S1 Table). Finally, four genes related to growth and metabolism were included ([48–50]; see Supplementary S1 Table), as stress can influence metabolic processes and growth rates, helping to identify markers of stress-induced growth inhibition. These 28 TaqmanTM qPCR gene transcription assays were designed to be printed on OpenArray® plates, resulting in the OSTP chip.

2.2. Chip design and optimization

The target candidate genes had been previously used for qPCR analyses by the GEN-FISH (the Genomic Network for Fish Identification, Stress and Health project; https://gen-fish.ca/) co-investigators, and all had shown measurable responses to stress in different studies (see details in [21]. The primers and probes for the assays were designed to target conserved regions within the selected gene coding regions (to ensure broad taxonomic application within the Salmonidae group) by aligning all available salmonid sequences for each gene using the NCBI sequence (http://www.ncbi.nlm.nih.gov/Genbank/) database. The goal was to develop qPCR assays that would work across a broad range of salmonid species as possible to maximize the utility of the OpenArray® OSTP chip for transcriptional profiling. More specifically, the Atlantic salmon ortholog (or orthologs from other salmonids) were searched for using BLASTn (NCBI) for sequence identification, alignments, phylogenetic tree building, and primer/probe design using Geneious Prime software (version 7.1.5; [51]). A search query of 500 was used. All sequences from the Salmo, Oncorhynchus, Salvelinus, and Coregonus spp., as well as Esox lucius (as a closely-related non-salmonid species) and Danio rerio (as a well-annotated model species) were selected and aligned using the MUSCLE aligner. The extracted sequences contained only the coding regions (CDS). An initial alignment including all these sequences was used to generate a phylogenetic tree using the Geneious Tree Builder with the default settings (i.e., Tamura-Nei, Neighbor-joining). The cluster of sequences that grouped with the Atlantic salmon ortholog (or other salmonid orthologs) were extracted. These target sequences were realigned and used to design primers and probes based on a consensus sequence. It is possible that some assays may amplify paralogs; however, because the assays were intended to function as biomarkers, their response to stress was the primary consideration. The functional gene categories are listed in Supplementary S1 Table. The primer and probe sequences used for assay design are listed in Supplementary S2 Table. The accession numbers and species used in the final alignments for primer/probe design are also listed in Supplementary S2 Table.

Primers (forward and reverse) were designed to be 16–22 bp, with three or fewer consecutive identical nucleotides and to generate amplicons less than 150 bp. The primers were designed with a target of 55% GC content (range: 45–65%) to ensure stable binding. Probes were designed using similar criteria to ensure gene sequence specificity and sensitivity but to function across multiple salmonid species effectively. The probes were designed with a melting temperature between 56–62°C to yield efficient detection for the OpenArray® TaqmanTM qPCR platform. All primers and TaqmanTM probes for the 25 candidate and three endogenous control gene assays were developed using Primer Express Software v3.0 (Thermo Fisher Scientific) for use on the OpenArray® chip platform. Both the primers and probes underwent in silico species testing via BLAST searches to confirm their broad applicability across salmonid species. The OSTP chip was printed using OpenArray® technology, which allows high throughput multiplexing of assays on a single chip, enhancing efficiency and consistency across samples.

2.3. OSTP chip assessment

To assess the performance of the OpenArray® OSTP chip among salmonid species, we applied it to eight diverse species. The eight species represent four genera: Arctic char (Salvelinus alpinus, AC), Brook Trout (Salvelinus fontinalis, BRKT), Lake Trout (Salvelinus namaycush, LT), Atlantic Salmon (Salmo salar, AS), Bloater (Coregonus hoyi, BLR), Lake Whitefish (Coregonus clupeaformis, LW), Chinook Salmon (Oncorhynchus tshawytscha, CS), Rainbow Trout (Oncorhynchus mykiss, RT). These species were selected due to their significance in sport and commercial fisheries, as well as their ecological and conservation importance.

Gill and liver tissues from three individuals from each species were provided by collaborators and shipped to the Great Lakes Institute for Environmental Research (GLIER), University of Windsor, for RNA extraction and cDNA synthesis. Different collaborators under the GEN-FISH project used the same Standard Operating Procedure (SOP) (e.g., euthanasia, sample preservation, etc.), making the samples are comparable, despite coming from different labs. The fish were all captive and sampled at rest, experiencing no obvious stress. RNA extractions from both gill and liver tissues used a TRIzol-based method (Invitrogen, Carlsbad, CA, USA) to provide high-quality RNA. Briefly, the tissue samples were homogenized using 1.0 mm glass beads and TRIzol, followed by phase separation with chloroform to isolate the RNA. Post-separation, the aqueous phase containing RNA was precipitated using isopropanol and washed with 70% ethanol to remove impurities. Selected RNA samples were used to estimate the RNA Integrity Number (RIN) on a 2100 Bioanalyzer to ensure the samples were suitable for subsequent analyses. The remaining RNA sample concentrations were measured using a NanoVue spectrophotometer (General Electric Company, Boston, MA, USA), and samples that fell below 500 ng/μL were re-extracted to ensure sufficient yield. DNase I (Ambion, Thermo Fisher Scientific, Waltham, MA, USA) was used to eliminate residual DNA. cDNA was then synthesized using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Burlington, ON, Canada), following the manufacturer’s protocol. cDNA samples were stored at −20°C until they were used for qPCR.

qPCR assays were conducted using the TaqmanTM OpenArray® OSTP chip which had each assay printed in duplicate, resulting in six qPCR assays for each species for each gene (3 samples x 2 replicates). The QuantStudio 12K Flex PCR System (Thermo Fisher Scientific) was used to perform qPCR and generate initial CT values through 40 cycles of amplification. Each 5.0 µL qPCR reaction mixture consisted of cDNA (1.2 µL per sample), ddH2O (1.3 µL) and TaqmanTM OpenArray® PCR Master Mix 2.5 µL (Applied Biosystems, Burlington, ON, Canada). The TaqmanTM OpenArray® chips were loaded using the OpenArray® AccuFill System to reduce inter-assay variation. The qPCR started with an initial denaturation step at 95°C for 10 minutes, followed by 40 cycles of denaturation at 95°C for 15 seconds and annealing/extension at 60°C for one minute. A total of 2 chips were used for 96 cDNA samples (3 samples x 2 biological replicates x 8 species x 2 tissues [gill and liver]). We used CT values to assess the performance of the individual assays on the OSTP chip across all eight species. To determine expected qPCR assay performance variation among the eight selected salmonid species, the primer and probe sequences were aligned with each species’ corresponding gene sequences, and the number of nucleotide mismatches between the primer/probe and target species gene sequences were counted to quantify expected assay performance for each of the eight species across the 28 qPCR assays.

2.4. 24-hour saltwater challenge

To assess the OSTP chip’s performance under actual osmotic stress conditions, we performed a 24-hour saltwater challenge using Chinook salmon at the hatchery facilities of Yellow Island Aquaculture Ltd., a commercial salmon farm on Quadra Island, BC, Canada. The fish used in this experiment were 20 months post-fertilization with an average mass of 51.4 g (±0.3 SE) and were being reared in freshwater raceways. Two experimental groups were established: (1) Freshwater control (FW-FW), where fish were transferred between freshwater (FW) tanks, and (2) Saltwater treatment (FW-SW), where freshwater-reared fish were transferred to saltwater (SW) tanks (30 ppt). Fish were acclimated in freshwater tanks for 72 hours before the start of the experiment to recover from transport stress, after which they were transferred to four identical 200 L tanks (two freshwater and two saltwater tanks), the second transfer (after recovery from transport) constituted the acute stressor. Throughout the experiment, dissolved oxygen (DO) levels were maintained at an average of 9.58 mg/L, with an oxygen saturation of 88%. Sampling was conducted at the end of the 24-hour challenge. Six fish per treatment (three per replicate) were euthanized with an overdose of clove oil (CloveOilCorp; 0.4 mL/L clove oil in aerated hatchery water) – the fish were held in the anesthetic bath until they lost equilibrium, after which their spinal cords were severed before taking measurements and tissue samples. Liver and gill samples were collected within 3 minutes of euthanasia and were preserved in high salt buffer (see above) at 4°C for 24 hours before being transferred to −20°C for long-term storage. This experiment was designed to assess transcriptional responses to saltwater exposure, focusing on freshwater-reared Chinook salmon. All animals were treated following the guidelines provided by the Canadian Council on Animal Care during holding and experimentation, and approval was granted by the University of Windsor Animal Utilization Project Proposal (AUPP#23-06).

Gene transcription was quantified by calculating ΔCT values. We then calculated ΔΔCT values to assess the gene transcription response for the saltwater challenge relative to the mean control ΔCT value. To display both up- and downregulation on a single scale, gene expression was expressed as ΔΔCT, where positive values indicate downregulation and negative values indicate upregulation.

2.5. Data analysis

All statistical analyses were conducted in R v4.5.0 [52]. For binary outcomes (amplification success), generalized linear mixed models (GLMMs) were fitted using the glmer() function in the lme4 package [53], specifying a binomial error distribution and a logit link. For continuous outcomes (CT or log-transformed ΔCT values), linear mixed models (LMMs) were fitted with the lmer() function in lme4. Tissue (gill vs. liver) and species were included as fixed effects. Target gene and species were modeled as random effects to account for variability across genes and species.

Model simplification followed a backward stepwise procedure: higher-order interaction and polynomial terms were tested first and removed if non-significant, followed by sequential removal of non-significant lower-order fixed effects. Random effects were assessed by likelihood ratio tests and retained only if they improved model fit. Collinearity among predictors was checked with variance inflation factors (VIFs). Model assumptions (normality, homoscedasticity) were evaluated using residual diagnostics (histograms, QQ plots, and Shapiro–Wilk tests).

For the 24-hour saltwater challenge, transcriptional responses (ΔΔCT) were analyzed relative to the FW-FW group at twenty-four hours, which served as the control condition. A two-way ANCOVA, with treatment (FW-FW vs. FW-SW) and tissue (gill vs. liver) as fixed factors, and body weight (g) included as a covariate, was used. Replicate tank was included as a random effect. Models were fitted with lmer() in lme4, and Type III sums of squares were obtained via Anova () in the car package [54]. Pairwise contrasts were estimated using emmeans, with Tukey adjustments [55]. False discovery rate (FDR) corrections were applied within functional gene categories (osmoregulation, stress response, immune function, growth & metabolism) using the Benjamini–Hochberg method. Statistical significance was determined at P < 0.05, and significant results are displayed by asterisks in the figures. All statistical analyses were run as described above.

Long format pairwise estimates for CT differences are provided in Supplementary S3 Table; GLMM fixed and random effect estimates and odds ratios are provided in Supplementary S4 Table; Model visual diagnostics (histograms and QQ plots; Supplementary S1, S2 Figs) and Shapiro-Wilk tests.

3. Results

3.1. Primer-probe mismatch

We evaluated the match between primer/probe sequences across the eight target species to determine whether the observed species-specific effects on gene transcription could simply result from assay design (Table 1). The overall primer-probe mismatch rates across all genes showed that Chinook salmon (3.04%), followed by Arctic char (3.09%) and Atlantic salmon (3.1%) had the lowest mismatch rates, whereas, Bloater (5.72%) had the highest primer-probe mismatch.

Table 1. Proportion of mismatch nucleotides (%) for the forward and reverse primers and probe sequences across the 8 selected fish species based on NCBI gene alignments. The mismatch percentages are used to evaluate qPCR compatibility for each gene across species. A dash (“- “) indicates that no sequence was available on NCBI for that gene in that species.

Gene

Category		 Gene

Name		 Salvelinus Salmo Coregonus Oncorhynchus
Arctic char Brook trout Lake trout Atlantic salmon Bloater Lake whitefish Rainbow trout Chinook salmon
Endogenous control G6PD 4.8% 6.0% 4.8% 5.1% 14.9% 10.5% 3.7% 3.7%
EF1A 0.0% 0.0% 4.0% 0.0% 0.0% 0.0% 0.0% 8.7%
RPL13A 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
Osmoregulation CFTR 0.0% 6.6% 0.0% 0.0% – 0.0% 4.5% 0.0%
ATP1A1 6.8% 6.8% 1.8% 3.3% 8.4% 8.4% 10.5% 0.0%
ATP1B1 1.8% 0.0% 6.6% 4.5% – 0.0% 5.0% 6.7%
SLC2A11A 0.0% 7.4% 7.4% 0.0% 5.7% 5.6% 0.0% 0.0%
SLC9A1B 5.1% 5.1% 6.8% 4.8% 1.2% 1.6% 8.5% 5.7%
SLC9A2 0.0% 3.6% 4.8% 8.8% 7.4% 1.8% 0% 0.0%
SLC9A3 0.0% 0.0% 0.0% 0.0% 9.3% 2.9% 6.4% 0.0%
SLC12A1 0.0% 1.6% 0.0% 0.0% 10.8% 8.9% 5.5% 0.0%
SLC12A2 0.0% 6.3% 6.0% 0.0% 10.8% 2.1% 7.3% 1.6%
ATP6V1BA 3.6% 1.8% 5.5% 6.8% 7.6% 9.1% 5.5% 5.7%
RHBG 3.8% 1.6% 3.8% 5.5% 9.2% 9.0% 8.1% 1.5%
RHCG1 8.6% 6.4% 6.4% 1.8% 6.4% 6.4% 2.1% 2.1%
Stress response GR1 8.1% 2.8% 6.3% 3.5% 1.8% 1.8% 6.5% 2.7%
GR2 1.0% 1.8% 4.4% 9.4% 7.4% 8.1% 0.0% 7.8%
HDS11B2 0.0% 2.7% 2.7% 6.9% 6.4% 2.7% 0.3% 7.1%
HSP90BA 3.3% 1.6% 3.3% 0.0% 7.6% 7.6% 1.6% 0.0%
Immune Function IL-1β 9.3% 2.3% 4.7% 8.4% 1.1% 3.0% 0.0% 8.7%
IFNγ – – 8.6% 1.4% 5.4% 0.0% 0.0% 0.0%
TNF-α 3.3% 2.7% 5.6% 8.8% – 10.6% 3.3% 3.5%
MX 6.7% 3.4% 3.4% 2.8% 3.5% 3.5% 3.5% 8.9%
TAPBP 7.6% 10.3% 8.5% 1.4% 8.8% 6.8% 3.2% 8.9%
Growth & Metabolism GHR 0.0% 0.0% 0% 0.0% 0.0% 0.0% 0.0% 0.0%
IGF1 7.9% 4.2% 4.2% 0.0% 4.3% 4.3% 0.0% 0.0%
IGFBP1 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
IGFBP2 1.7% 4.7% 2.4% 3.6% 5.1% 5.1% 3.6% 1.8%
Open in a new tab

3.2. Tissue and species-specific performances

Overall, gill-derived cDNA exhibited significantly better amplification success (83.0% of 1344 assays; 1116 successful amplifications; GLMM P < 0.001) and lower CT values (mean CT = 25.8) compared to liver-derived cDNA (63.4% of 1344 assays; 852 successful amplifications; mean CT = 29.5; GLMM P < 0.01) (Tables 2 and 3). Assay success rates for gill and liver samples varied among the eight salmonid species (Table 2; Fig 1). Gill assays were generally more successful across species, with both Chinook salmon and Lake trout achieving the highest success rate of 96.4% (Table 2; Fig 1). For liver assays, Lake trout exhibited the highest success rate (78.6%), followed by Atlantic salmon (75%). In contrast, Arctic char showed the lowest liver assay success rate (42.9%), followed by Rainbow trout (46.4%).

Table 2. Tissue effect comparisons of mean CT values between gill and liver tissues for eight salmonid species (and for all combined species) across 28 genes, with six replicates for both gill and liver cDNA, resulting in 168 assays per tissue per species. qPCR assay results are categorized as “Successful” (5–6 replicates), “Partial Success” (2–4 replicates), or “Failed” (0–1 replicates). Average CT values are reported for successful assays only. P-values for tissue effects were calculated using a logistic generalized linear mixed model (GLMM) for amplification success and a linear mixed model (LMM) for CT values, with adjustments for multiple comparisons using the Benjamini–Hochberg (FDR) method. Significance levels are (* P < 0.05, ** P < 0.01, *** P < 0.001) are shown in bold. Dashed lines (“- “) reflects the absence of data.

Species Tissue Failed Partial Success Successful Proportion Successful Avg_CT P_value (Logistic GLMM) P_value (LMM) Total Assays
Arctic char Gill 24 12 132 78.6% 26.4 0.15 0.11 168
Liver 66 30 72 42.9% 28.8 0.22 168
Atlantic salmon Gill 12 12 144 85.7% 24.1 <0.05 <0.05 168
Liver 42 0 126 75.0% 29.6 0.06 168
Bloater Gill 12 18 138 82.1% 26.1 <0.05 <0.05 168
Liver 36 24 108 64.3% 28 0.07 168
Brook trout Gill 6 30 132 78.6% 25.8 <0.05 <0.05 168
Liver 24 30 114 67.9% 28.1 0.06 168
Chinook salmon Gill 6 0 162 96.4% 25.1 <0.01 <0.01 168
Liver 36 12 120 71.4% 30.4 <0.05 168
Lake whitefish Gill 18 18 132 78.6% 26.2 0.1 0.1 168
Liver 48 18 102 60.7% 29.7 0.18 168
Lake trout Gill 6 0 162 96.4% 25.4 <0.05 <0.05 168
Liver 18 18 132 78.6% 28.6 <0.05 168
Rainbow trout Gill 30 24 114 67.9% 27.6 0.22 0.18 168
Liver 72 18 78 46.4% 33.3 0.25 168
Gill Total – 114 114 1116 83.0% 25.8 <0.001 <0.001 1344
Liver Total – 342 150 852 63.4% 29.5 <0.01 1344
Overall Total – 456 264 1968 73.20% 27.75 – – 2688
Open in a new tab

Table 3. Mean CT values from replicated TaqmanTM qPCR assays across 28 target genes using cDNA from gill and liver tissues of various salmonid species generated on the OpenArray® platform. Each value represents the mean CT (cycle threshold) ± SEM, with the number of successful qPCR replicates (out of 6, 3 individuals replicated twice) in parentheses. Color coding demonstrates assay performance: red (0–1, failed), yellow (2–4, partial success), and green (5–6, successful amplification).

Gene

Category		 Gene

Name		 Salvelinus Salmo Coregonus Oncorhynchus
Arctic char Brook trout Lake trout Atlantic salmon Bloater Lake whitefish Rainbow trout Chinook salmon
Gill Liver Gill Liver Gill Liver Gill Liver Gill Liver Gill Liver Gill Liver Gill Liver
Endogenous control G6PD 26.4 ± 0.6 (6) 26.8 ± 0.6 (4) 27.6 ± 0.3 (6) 25.7 ± 0.4 (6) 25.4 ± 0.6 (6) 24.6 ± 0.8 (6) 24.1 ± 0.5 (6) 27.6 ± 0.7 (6) 26.1 ± 0.3 (6) 23.0 ± 0.3 (6) 26.0 ± 0.3 (6) 23.7 ± 0.3 (6) 29.0 ± 0.9 (6) 32.3 ± 0.1 (6) 25.1 ± 0.1 (6) 28.0 ± 0.1 (6)
EF1A 25.6 ± 0.1 (6) 28.9 ± 0.5 (6) 25.8 ± 0.4 (6) 26.1 ± 0.1 (6) 25.6 ± 0.3 (6) 26.2 ± 0.5 (6) 24.8 ± 0.2 (6) 26.3 ± 0.4 (6) 25.5 ± 0.9 (6) 26.5 ± 0.1 (6) 25.9 ± 0.4 (6) 27.1 ± 0.4 (6) 27.4 ± 0.3 (6) 28.2 ± 0.4 (6) 24.5 ± 0.7 (6) 26.1 ± 0.9 (6)
RPL13A 21.1 ± 0.2 (6) 25.9 ± 0.5 (6) 20.9 ± 0.7 (6) 19.3 ± 0.9 (6) 19.7 ± 0.7 (6) 19.7 ± 0.4 (6) 19.5 ± 0.6 (6) 19.6 ± 0.9 (6) 20.4 ± 0.1 (6) 20.3 ± 0.4 (6) 20.5 ± 0.2 (6) 21.8 ± 0.7 (6) 22.2 ± 0.1 (6) 21.7 ± 0.8 (6) 19.3 ± 0.5 (6) 18.8 ± 0.6 (6)
Osmoregulation CFTR 27.2 ± 0.1 (6) 0 (0) 27.2 ± 0.6 (6) 29.7 ± 0.8 (6) 24.5 ± 0.1 (6) 0 (0) 24.6 ± 0.8 (6) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 29.4 ± 0.4 (6) 0 (0) 23.0 ± 0.1 (6) 0 (0)
ATP1A1 20.9 ± 0.3 (6) 0 (0) 20.5 ± 0.3 (6) 29.1 ± 0.2 (2) 22.1 ± 0.3 (6) 32.2 ± 0.7 (4) 19.3 ± 0.3 (6) 0 (0) 20.9 ± 0.5 (6) 28.9 ± 0.7 (4) 22.0 ± 0.1 (6) 22.0 ± 0.7 (2) 24.2 ± 0.6 (6) 0 (0) 20.4 ± 0.6 (6) 0 (0)
ATP1B1 22.7 ± 0.8 (6) 27.6 ± 0.8 (6) 21.0 ± 0.4 (6) 21.3 ± 0.7 (6) 20.5 ± 0.2 (6) 21.9 ± 0.2 (6) 20.0 ± 0.5 (6) 22.2 ± 0.7 (6) 20.0 ± 0.4 (6) 22.6 ± 0.4 (6) 22.5 ± 0.7 (6) 24.3 ± 0.9 (6) 23.8 ± 0.3 (6) 25.5 ± 0.2 (6) 20.5 ± 0.2 (6) 22.3 ± 0.4 (6)
SLC2A11A 0 (0) 0 (0) 32.9 ± 0.6 (4) 0 (0) 28.7 ± 0.5 (6) 0 (0) 0 (0) 0 (0) 31. 0.7 (6) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 29.2 ± 0.5 (6) 0 (0)
SLC9A1B 29.3 ± 0.7 (6) 21.6 ± 0.2 (2) 28.8 ± 0.8 (6) 29.7 ± 0.1 (6) 26.8 ± 0.9 (6) 29.6 ± 0.1 (6) 27.1 ± 0.7 (6) 31.1 ± 0.5 (6) 27.9 ± 0.4 (6) 31.2 ± 0.1 (6) 29.2 ± 0.4 (6) 31.3 ± 0.7 (6) 30.0 ± 0.5 (6) 29.6 ± 0.7 (2) 28.5 ± 0.6 (6) 30.4 ± 0.2 (6)
SLC9A2 26.3 ± 0.7 (6) 0 (0) 27.5 ± 0.2 (6) 0 (0) 25.9 ± 0.3 (6) 0 (0) 25.4 ± 0.3 (6) 0 (0) 28.0 ± 0.3 (4) 0 (0) 28.4 ± 0.5 (6) 0 (0) 28.3 ± 0.2 (6) 0 (0) 25.0 ± 0.1 (6) 0 (0)
SLC9A3 32.2 ± 0.4 (6) 0 (0) 21.9 ± 0.4 (6) 0 (0) 20.8 ± 0.1 (6) 31.6 ± 0.8 (2) 21.7 ± 0.1 (6) 0 (0) 31.2 ± 0.1 (6) 36.2 ± 0.3 (4) 34.6 ± 0.2 (6) 0 (0) 27.5 ± 0.1 (6) 0 (0) 21.5 ± 0.3 (6) 25.1 ± 0.3 (6)
SLC12A1 29.9 ± 0.3 (4) 31.2 ± 0.5 (2) 30.2 ± 0.6 (4) 27.0 ± 0.3 (6) 28.0 ± 0.9 (6) 28.1 ± 0.2 (6) 28.6 ± 0.6 (6) 29.8 ± 0.1 (6) 30.2 ± 0.7 (6) 31.2 ± 0.6 (2) 30.1 ± 0.4 (2) 0 (0) 0 (0) 0 (0) 29.5 ± 0.5 (6) 26.5 ± 0.6 (6)
SLC12A2 22.3 ± 0.9 (6) 29.9 ± 0.2 (6) 23.1 ± 0.9 (6) 27.8 ± 0.6 (6) 21.2 ± 0.3 (6) 27.0 ± 0.6 (6) 23.8 ± 0.9 (6) 28.8 ± 0.4 (6) 23.7 ± 0.5 (6) 30.7 ± 0.4 (6) 24.7 ± 0.8 (6) 31.2 ± 0.2 (6) 25.8 ± 0.6 (6) 30.0 ± 0.9 (4) 20.6 ± 0.8 (6) 26.7 ± 0.2 (6)
ATP6V1BA 25.3 ± 0.3 (6) 28.4 ± 0.2 (6) 24.3 ± 0.5 (6) 24.5 ± 0.7 (6) 23.2 ± 0.7 (6) 24.0 ± 0.8 (6) 23.3 ± 0.3 (6) 25.2 ± 0.6 (6) 22.6 ± 0.9 (6) 23.8 ± 0.8 (6) 24.4 ± 0.1 (6) 26.9 ± 0.5 (6) 27.0 ± 0.4 (6) 27.8 ± 0.3 (6) 22.6 ± 0.9 (6) 24.1 ± 0.4 (6)
RHBG 35.2 ± 0.5 (6) 0 (0) 23.0 + ± 0.2 (6) 27.2 ± 0.8 (6) 23.4 ± 0.1 (6) 27.6 ± 0.4 (6) 24.0 ± 0.1 (6) 28.6 ± 0.3 (6) 33.0 ± 0.4 (2) 0 (0) 34.5 ± 0.3 (6) 0 (0) 0 (0) 0 (0) 23.3 ± 0.3 (6) 26.9 ± 0.5 (6)
RHCG1 21.8 ± 0.9 (6) 0 (0) 20.6 ± 0.1 (6) 30.5 ± 0.6 (2) 18.9 ± 0.2 (6) 29.5 ± 0.6 (6) 18.8 ± 0.4 (6) 0 (0) 18.6 ± 0.2 (6) 29.2 ± 0.5 (4) 19.9 ± 0.7 (6) 0 (0) 25.4 ± 0.6 (6) 0 (0) 18.7 ± 0.2 (6) 0 (0)
Stress

response		 GR1 26.4 ± 0.5 (6) 29.8 ± 0.3(4) 26.9 ± 0.7 (6) 27.3 ± 0.6 (6) 26.4 ± 0.4 (6) 28.1 ± 0.9 (6) 26.6 ± 0.1 (6) 28.3 ± 0.4 (6) 26.6 ± 0.8 (6) 27.1 ± 0.3 (6) 27.0 ± 0.2 (6) 29.2 ± 0.4 (6) 29.1 ± 0.7 (6) 28.8 ± 0.4 (6) 25.7 ± 0.8 (6) 26.8 ± 0.8 (6)
GR2 26.6 ± 0.4 (6) 32.1 ± 0.5 (6) 26.4 ± 0.4 (6) 25.9 ± 0.3 (6) 25.1 ± 0.3 (6) 26.8 ± 0.4 (6) 25.7 ± 0.3 (6) 27.1 ± 0.5 (6) 25.7 ± 0.1 (6) 25.7 ± 0.1 (6) 26.4 ± 0.9 (6) 29.3 ± 0.2 (6) 28.7 ± 0.2 (6) 28.8 ± 0.8 (6) 24.7 ± 0.9 (6) 26.1 ± 0.4 (6)
HDS11B2 25.3 ± 0.5 (6) 30.9 ± 0.8 (6) 27.5 ± 0.3 (6) 31.4 ± 0.7 (6) 24.9 ± 0.6 (6) 29.8 ± 0.1 (6) 26.8 ± 0.7 (6) 30.0 ± 0.1 (6) 26.8 ± 0.3 (6) 29.3 ± 0.8 (6) 27.8 ± 0.3 (6) 31.3 ± 0.7 (6) 26.3 ± 0.1 (6) 27.5 ± 0.3 (6) 25.1 ± 0.4 (6) 26.8 ± 0.9 (6)
HSP90BA 26.1 ± 0.7 (6) 26.9 ± 0.1 (6) 26.9 ± 0.7 (6) 22.9 ± 0.3 (6) 26.6 ± 0.8 (6) 24.4 ± 0.5 (6) 26.2 ± 0.2 (6) 23.5 ± 0.6 (6) 26.2 ± 0.7 (6) 25.5 ± 0.3 (6) 26.9 ± 0.2 (6) 24.1 ± 0.2 (6) 30.5 ± 0.9 (6) 31.2 ± 0.1 (6) 26.4 ± 0.2 (6) 24.4 ± 0.1 (6)
Immune

function		 IL-1β 30.5 ± 0.1 (6) 0 (0) 29.2 ± 0.1 (6) 29.5 ± 0.3 (2) 26.9 ± 0.4 (6) 31.4 ± 0.5 (2) 27.8 ± 0.3 (4) 0 (0) 31.0 ± 0.2 (6) 0 (0) 30.2 ± 0.7 (4) 31.4 ± 0.8 (2) 29.6 ± 0.9 (2) 0 (0) 25.5 ± 0.7 (6) 30.9 ± 0.4 (4)
IFNγ 34.1 ± 0.6 (2) 0 (0) 31.2 ± 0.4 (2) 33.1 ± 0.2 (2) 28.5 ± 0.7 (6) 29.6 ± 0.3 (6) 29.8 ± 0.2 (6) 30.1 ± 0.6 (6) 31.5 ± 0.2 (4) 0 (0) 32.5 ± 0.1 (6) 0 (0) 31.0 ± 0.3 (2) 0 (0) 25.7 ± 0.9 (6) 29.4 ± 0.2 (4)
TNF-α 0 (0) 0 (0) 0 (0) 0 (0) 29.6 ± 0.3 (6) 31.2 ± 0.1 (6) 30.6 ± 0.4 (6) 32.4 ± 0.7 (6) 31.7 ± 0.8 (6) 36.4 ± 0.6 (2) 31.3 ± 0.7 (6) 32.6 ± 0.9 (2) 32.7 ± 0.5 (4) 0 (0) 28.2 ± 0.4 (6) 32.1 ± 0.7 (6)
MX 28.1 ± 0.6 (6) 30.9 ± 0.2 (2) 28.2 ± 0.2 (6) 28.2 ± 0.7 (6) 24.7 ± 0.8 (6) 27.4 ± 0.5 (6) 25.6 ± 0.3 (6) 29.9 ± 0.4 (6) 25.7 ± 0.1 (6) 28.7 ± 0.4 (6) 25.0 ± 0.1 (6) 29.3 ± 0.1 (6) 30.2 ± 0.1 (6) 31.7 ± 0.3 (2) 23.1 ± 0.1 (6) 26.7 ± 0.5 (6)
TAPBP 25.2 ± 0.7 (6) 27.5 ± 0.6 (4) 26.5 ± 0.7 (6) 26.5 ± 0.4 (6) 26.0 ± 0.4 (6) 28.2 ± 0.3 (6) 29.3 ± 0.1 (6) 27.8 ± 0.2 (6) 29.3 ± 0.4 (6) 32.1 ± 0.1 (6) 29.2 ± 0.7 (6) 32.4 ± 0.7 (6) 26.4 ± 0.9 (6) 28.3 ± 0.5 (6) 24.5 ± 0.8 (6) 27.9 ± 0.1 (6)
Growth & Metabolism GHR 28.4 ± 0.5 (6) 29.8 ± 0.2 (6) 27.9 ± 0.1 (6) 25.1 ± 0.2 (6) 26.5 ± 0.6 (6) 24.2 ± 0.1 (6) 25.9 ± 0.5 (6) 24.1 ± 0.3 (6) 25.9 ± 0.1 (6) 23.1 ± 0.7 (6) 29.6 ± 0.9 (6) 27.6 ± 0.2 (6) 30.1 ± 0.3 (6) 26.2 ± 0.4 (6) 26.2 ± 0.7 (6) 23.5 ± 0.8 (6)
IGF1 29.1 ± 0.3 (6) 25.7 ± 0.9 (6) 26.1 ± 0.2 (6) 23.4 ± 0.3 (6) 26.0 ± 0.1 (6) 19.8 ± 0.3 (6) 30.2 ± 0.8 (6) 22.2 ± 0.1 (6) 30.7 ± 0.3 (6) 28.2 ± 0.4 (6) 34.2 ± 0.6 (6) 28.9 ± 0.5 (6) 30.8 ± 0.2 (4) 23.5 ± 0.2 (6) 25.9 ± 0.1 (6) 22.2 ± 0.3 (6)
IGFBP1 0 (0) 27.4 ± 0.8 (6) 28.5 ± 0.6 (2) 25.7 ± 0.5 (6) 32.1 ± 0.3 (6) 25.7 ± 0.6 (6) 0 (0) 21.4 ± 0.6 (6) 33.2 0.7 (4) 25.7 ± 0.6 (6) 0 (0) 28.9 ± 0.2 (6) 0 (0) 28.9 ± 0.4 (6) 28.5 ± 0.4 (6) 19.5 ± 0.6 (6)
IGFBP2 0 (0) 0 (0) 30.4 ± 0.8 (4) 24.0 ± 0.2 (4) 0 (0) 29.3 ± 0.8 (6) 31.7 ± 0.4 (4) 23.6 ± 0.9 (6) 0 (0) 22.5 ± 0.8 (6) 31.8 ± 0.3 (2) 26.7 ± 0.8 (6) 0 (0) 0 (0) 0 (0) 0 (0)
Open in a new tab

Fig 1. Comparison of qPCR success rates between gill and liver tissues across eight salmonid fish species.

Fig 1

Open in a new tab

Comparison of qPCR success rates across eight fish species (Arctic char, Atlantic salmon, Bloater, Brook trout, Lake trout, Chinook salmon, Lake whitefish, and Rainbow trout) between two tissue types: gill (white bars) and liver (grey bars). The x-axis represents species, and the y-axis shows qPCR success rates expressed as the percentage of assays that produced a CT value. Success was defined as the percentage of assays that yielded a CT value out of a total of 2688 assays (8 species × 28 genes × 6 replicates × 2 tissues). Error bars represent the standard error of the mean (SEM) calculated from replicate success rates for each species and tissue. Significance levels are displayed above the bars: *P < 0.05, **P < 0.01, ***P < 0.001.

Across species, higher mean CT values generally corresponded to fewer successful replicates, with partial success (2–4 replicates) or failed amplification (0–1 replicate) also appearing in the table (Tables 2 and 3). With gill cDNA, 10 osmoregulatory genes (CFTR, SLC2A11A, SLC9A2, SLC9A3, SLC12A1, SLC12A2, ATP1A1, ATP1B1, RHBG, and RHCG1) showed substantially lower CT (≥2 units) in gill relative to liver, often also associated with higher amplification success in gill and partial or failed amplification in liver (Tables 2 and 3). Three immune genes (IL-1β, TNF-α, and MX) showed mean CT values favoring gill and success rates exceeding 70% across species. These patterns were consistent across the osmoregulation and immune gene categories (Tables 2 and 3). However, liver cDNA showed higher transcript abundance (≥2 CT difference) than gill cDNA for all four growth & metabolism genes (GHR, IGF1, IGFBP1, and IGFBP2), often with > 90% amplification success in liver tissue (Tables 2 and 3). In contrast, none of the osmoregulatory, immune, or stress-related targets or the endogenous control assays showed higher transcription in liver. We found no significant differences in the performance of the endogenous control gene assays between gill and liver (Tables 2 and 3).

3.3. Chinook salmon 24-hour saltwater challenge

The 24-hour saltwater challenge with Chinook salmon revealed distinct transcriptional responses in gill and liver tissues, with significant changes observed across osmoregulatory, stress, immune, and growth/metabolism gene categories. Relative transcription data, presented as ΔΔCT, highlighted tissue-specific dynamics in transcriptional activity. Additionally, ΔCT values (calculated as the mean CT of gill minus the mean CT of liver) were used to compare average transcription levels between these tissues under experimental conditions, rather than reflecting baseline expression differences in unstressed fish.

In gill tissue, among osmoregulatory genes, SLC12A2 showed significant upregulation in gene transcript abundance (Fig 2), whereas, CFTR, SLC9A2, and ATP6V1BA showed significant downregulation in gene transcript abundance. Other osmoregulatory genes exhibited no significant transcriptional differences. In the stress response category, GR1 and GR2 showed significant upregulation in gene transcript abundance (Fig 2). Immune-related and growth and metabolism genes did not exhibit significant transcriptional changes in gill tissue (Fig 2). In liver tissue, only GHR and IGF1 exhibited significant upregulation among the growth and metabolism genes. No other genes among any other gene function category showed significant up- or downregulation in transcript abundance in liver tissue (Fig 2).

Fig 2. Relative transcriptional responses (ΔΔCT) in gill and liver tissues of Chinook salmon following a 24-h saltwater challenge.

Fig 2

Open in a new tab

Bar graphs showing relative transcription response (displayed as ΔΔCT values) in gill (white bars) and liver (grey bars) tissue after a 24-hour saltwater challenge in Chinook salmon. The 25 gene assays are categorized into osmoregulation (12 genes), stress response (four genes), immune function (five genes), and growth & metabolism (four genes). Bars represent the average ΔΔCT values relative to control fish transcription, with error bars displaying standard error of the mean (SEM). Significant differences between tissues are indicated by an asterisk (*). Some gene assays did not provide data and are marked as not available (NA).

All genes across different biological function categories showed negative ΔCT values (except for SLC2A11A, SLC9A2, and SLC12A1 in the osmoregulatory gene category, and IGF1, IGFBP1, and IGFBP2 in the growth and metabolism category), indicating that transcription levels were more highly expressed in gill than in liver tissue (Fig 3).

Fig 3. Differences in gene expression (ΔCT) between gill and liver tissues in Chinook salmon following a 24-h saltwater challenge.

Fig 3

Open in a new tab

Bar graph showing the ΔCT differences in gene expression between gill and liver tissues in Chinook salmon following a 24-hour saltwater challenge. The data presented here are ΔCT values. The 25 gene assays are grouped into functional categories—osmoregulation (12 genes), stress response (four genes), immune function (five genes), and growth & metabolism (four genes). Positive ΔCT values indicate higher relative expression in liver, while negative ΔCT values indicate higher relative expression in gill. Bars represent the calculated ΔCT values for each gene, with gene categories.

4. Discussion

The overall performance of the OSTP chip assays varied across tissues and species and generally, gill-derived cDNA performed better than liver-derived cDNA in both the eight at-rest species experiment and the 24-hour saltwater challenge. This is not surprising given that many of the assays were designed for osmoregulatory genes, which are often more actively expressed in the gill as it is a primary site for osmoregulation in fish. Liver cDNA samples exhibited reduced performance for assays targeting immune function and osmoregulation genes, although some assays within these functional categories worked well, depending on the species. Although the liver is known to be involved in fish immune function [56], the reduced success for some of our qPCR assays suggests that our specific gene assays may have limited applicability to liver tissue, although immune challenge applications should be explored. Overall, our study suggests that gill tissue can be used non-lethally and readily sampled, making this tissue particularly useful for conservation applications rather than using lethal liver tissue, which requires animal sacrifice [5,16].

Species-level differences in chip performance were also observed (e.g., Arctic char and Rainbow trout showed fewer successful qPCR amplifications, whereas Chinook salmon, Lake trout, and Atlantic salmon showed more), but no consistent genus-level effects were obvious. Sequence mismatch between primer/probe and target template sequences may have contributed to the variability in assay performance among the eight target species. Moreover, the qPCR amplification success rate among species might vary due to different lab protocols [19]; however, different collaborators under the GEN-FISH project used the same SOP/protocol to minimize handling effects (e.g., the method of euthanasia or sample preservation). These findings emphasize the importance of considering tissue- and species-specific variability in transcriptional profiling. While the OSTP chip has demonstrated significant utility, further refinement and expansion could enhance its applicability, improving performance for underperforming tissues and optimizing assay compatibility across diverse salmonid species.

Our study reveals that the qPCR assays exhibited significant tissue-specific transcriptional differences between gill and liver cDNAs across all eight species, emphasizing the specialized functions of these organs in the physiological response to stress. These findings align with a study on Atlantic salmon, Brown trout, Chinook salmon, Rainbow trout, Brook trout, Arctic char, Lake whitefish, and Bloater, which observed similar trends in stress- and osmoregulation-related gene expression in gill tissue across these same eight salmonid species [21]. Additionally, this cited study [21] observed lower amplification success for immune-related genes in liver and muscle tissues compared to gills [21]. In contrast, our study showed more consistent expression across tissues, which may reflect differences in the applied stressors or species-specific responses. However, the exclusive expression of SLC2A11A and SLC9A2 in gills highlights their pivotal roles in ion transport and osmotic regulation, essential for maintaining ionic balance in aquatic environments [57,58]. Additionally, the elevated expression of immune genes in gill aligns with their function as frontline barriers against environmental pathogens, reinforcing localized immune responses [59].

Conversely, liver-derived cDNA demonstrated stronger transcription of metabolism and growth-related genes. These findings align with a study on the liver-specific expression of IGFBP paralogs in Atlantic salmon, which emphasizes their role in modulating IGF activity during stress and growth regulation [48]. It also emphasized the pivotal role of IGFBP1 and IGFBP2 in regulating IGF bioavailability under stress and catabolic conditions. Similarly, temperature-dependent modulation of GHR and IGFBP genes has been demonstrated in Atlantic salmon, with increased GHR transcription in muscle at elevated temperatures (17–19°C), suggesting lipolysis to meet energy demands [60]. These findings align with a study on Atlantic salmon, which reported increased IGFBP1b expression at intermediate temperatures (15–17°C), highlighting the liver’s role in regulating IGF bioavailability under environmental challenges [60]. In contrast, our study observed consistently high GHR expression in the liver, reinforcing its systemic role in energy regulation and metabolic adaptation. Prolonged thermal stress was also shown to reduce IGF1 and IGF2 mRNA levels in both liver and muscle, reflecting a shift from growth to energy mobilization [48]. Together, these results emphasize the interconnected roles of GHR and IGFBP-related genes in liver tissue in coordinating growth and metabolism in response to environmental stress.

Interestingly, a subset of genes involved in cellular homeostasis and stress response, such as GR1, GR2, HSD11B2, and HSP90BA, maintained comparable expression levels in both gill and liver tissues. These genes exhibited consistent transcription levels across tissues, emphasizing their essential role in maintaining cellular stability and mitigating damage from environmental stressors. This pattern aligns with findings from a comparative study on eight species of salmonids, where similar expression profiles were observed for these genes, further supporting their function in cellular stress management [21]. This consistent pattern of expression suggests a fundamental role in maintaining cellular stability and mitigating damage from environmental stressors across different tissues [19]. The ability of the OSTP chip to capture gene transcription in both gill and liver tissue suggests its potential for broad applications in other tissue types, including mucus swabs, representing an emerging non-invasive method for gene transcription profiling [21]. By facilitating the detection of both tissue-specific and cross-tissue transcriptional responses, the OSTP chip enables a comprehensive evaluation of gene transcription patterns.

While the OSTP chip exhibited generally robust performance across tissues and species in our broad at-rest assessment, its practical application is important for showcasing its utility as a transcriptional profiling tool. To assess its effectiveness in detecting molecular responses under physiological stress, we performed a 24-hour saltwater challenge with Chinook salmon and used the OSTP chip to profile their transcriptional responses to acute salinity stress. As observed in the multi-species analysis, liver-derived cDNA showed lower qPCR success compared to gill-derived cDNA, with only 20 of 25 assays yielding successful amplifications in liver versus 23 of 25 in gill. The saltwater challenge elicited significant responses in osmoregulatory and stress-related genes in gill tissue, reflecting an adaptive response to maintain ion balance under acute salinity stress. Overall, for saltwater-challenged Chinook salmon, 20 and 23 assays were successful in liver and gill, compared to 18 and 22 successful assays, respectively, in resting fish. By comparing the resting and challenged outcomes, our findings highlight the distinct transcriptional roles of gill and liver tissues in responding to acute salinity stress. The study further emphasizes the utility of the OSTP chip in detecting molecular responses to physiological challenges, with key insights into the roles of osmoregulatory, stress, and metabolic pathways during short-term stress responses.

Transcriptional profiling using the OSTP chip described here provides a framework for characterizing how salmonid fishes cope with osmoregulatory challenges. While our OSTP chip was conceived for applications in captive-reared fish to provide a tool to improve rearing practices and enhance fish performance [61], it also has value for wild populations, where it can be used to monitor responses to changing environmental conditions [62]. As part of the GEN-FISH project, we also designed tissues- and stressor-specific chips (see [63–66]) to foster the assessment of the health of salmonid fish under changing environments. Multi-scale controlled laboratory and field validation of these custom chips will be useful approaches for real-world applications to key environmental stressors. By targeting key genes involved in osmoregulation, stress, immune, and growth/metabolism, the OSTP chip allows extrapolation to possible impacts on survival, growth, and performance related to osmoregulatory stress. Although our OSTP chip was designed for, and performs better in, gill cDNA, it is clear that the chip can also provide valuable data using other tissues, which will be useful to measure stress responses in the wild. Furthermore, the OSTP chip could be used to characterize the remarkable osmoregulatory plasticity of salmonid species relative to other teleosts, providing valuable insights into their ability to adapt to diverse salinity conditions. Thus, the OSTP chip will be useful for aquaculture practices by fostering better husbandry as well as providing a valuable tool for future studies of the osmoregulatory response of salmonid fishes to environmental stress.

Supporting information

S1 Table. List of candidate genes and their function.

(PDF)

pone.0360513.s001.pdf (86.5KB, pdf)
S2 Table. Primer and probe sequences.

(PDF)

pone.0360513.s002.pdf (119.6KB, pdf)
S3 Table. Pairwise comparisons of mean CT values among eight salmonid species.

(PDF)

pone.0360513.s003.pdf (74.5KB, pdf)
S4 Table. Effects of species and tissues on amplification success.

(PDF)

pone.0360513.s004.pdf (61.4KB, pdf)
S1 Fig. Histogram of model residuals.

(PDF)

pone.0360513.s005.pdf (47.8KB, pdf)
S2 Fig. QQ plot of residuals.

(PDF)

pone.0360513.s006.pdf (82.2KB, pdf)

Acknowledgments

The authors would like to thank Drs. John and Ann Heath and staffs of Yellow Island Aquaculture Ltd. for technical assistance. Yellow Island Aquaculture Ltd. provided logistical support for this project. The authors would also like to thank Jonathon LeBlanc and Shelby Toews for technical assistance.

Data Availability

Metadata for the qPCR amplification success and the 24-h saltwater challenge experiment have been submitted to Figshare: https://doi.org/10.6084/m9.figshare.31879918.

Funding Statement

This study was funded by Genome Canada. The funders provided financial support for the study, including data collection and analysis.

References

  • 1.Sundell KS, Sundh H. Intestinal fluid absorption in anadromous salmonids: Importance of tight junctions and aquaporins. Front Physiol. 2012;3:388. doi: 10.3389/fphys.2012.00388 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.McCormick SD. Ontogeny and evolution of salinity tolerance in anadromous salmonids: Hormones and heterochrony. Estuaries. 1994;17(1):26. doi: 10.2307/1352332 [DOI] [Google Scholar]
  • 3.Houde ALS, Schulze AD, Kaukinen KH, Strohm J, Patterson DA, Beacham TD, et al. Transcriptional shifts during juvenile Coho salmon (Oncorhynchus kisutch) life stage changes in freshwater and early marine environments. Comp Biochem Physiol Part D Genomics Proteomics. 2019;29:32–42. doi: 10.1016/j.cbd.2018.10.002 [DOI] [PubMed] [Google Scholar]
  • 4.Brijs J, Sandblom E, Sundh H, Gräns A, Hinchcliffe J, Ekström A, et al. Increased mitochondrial coupling and anaerobic capacity minimizes aerobic costs of trout in the sea. Sci Rep. 2017;7:45778. doi: 10.1038/srep45778 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Houde ALS, Günther OP, Strohm J, Ming TJ, Li S, Kaukinen KH, et al. Discovery and validation of candidate smoltification gene expression biomarkers across multiple species and ecotypes of Pacific salmonids. Conserv Physiol. 2019;7(1):coz051. doi: 10.1093/conphys/coz051 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Crozier LG, Hendry AP, Lawson PW, Quinn TP, Mantua NJ, Battin J, et al. Potential responses to climate change in organisms with complex life histories: evolution and plasticity in Pacific salmon. Evol Appl. 2008;1(2):252–70. doi: 10.1111/j.1752-4571.2008.00033.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Sakamoto T, Uchida K, Yokota S. Regulation of the ion-transporting mitochondrion-rich cell during adaptation of teleost fishes to different salinities. Zoolog Sci. 2001;18(9):1163–74. doi: 10.2108/zsj.18.1163 [DOI] [PubMed] [Google Scholar]
  • 8.Alexandrou MA, Swartz BA, Matzke NJ, Oakley TH. Genome duplication and multiple evolutionary origins of complex migratory behavior in Salmonidae. Mol Phylogenet Evol. 2013;69(3):514–23. doi: 10.1016/j.ympev.2013.07.026 [DOI] [PubMed] [Google Scholar]
  • 9.Mark Shrimpton J. Seawater to freshwater transitions in diadromous fishes. Fish Physiology. Vol. 32. Academic Press; 2012. pp. 327–93. doi: 10.1016/b978-0-12-396951-4.00007-4 [DOI] [Google Scholar]
  • 10.Shrimpton JM, Bernier NJ, Iwama GK, Randall DJ. Differences in measurements of smolt development between wild and hatchery-reared juvenile coho salmon (Oncorhynchus kisutch) before and after saltwater exposure. Can J Fish Aquat Sci. 1994;51(10):2170–8. doi: 10.1139/f94-218 [DOI] [Google Scholar]
  • 11.Näslund J, Johnsson JI. Environmental enrichment for fish in captive environments: Effects of physical structures and substrates. Fish Fisheries. 2014;17(1):1–30. doi: 10.1111/faf.12088 [DOI] [Google Scholar]
  • 12.Tamario C, Sunde J, Petersson E, Tibblin P, Forsman A. Ecological and evolutionary consequences of environmental change and management actions for migrating fish. Front Ecol Evol. 2019;7. doi: 10.3389/fevo.2019.00271 [DOI] [Google Scholar]
  • 13.McCormick SD, Shrimpton JM, Nilsen TO, Ebbesson LO. Advances in our understanding of the parr–smolt transformation of juvenile salmon: A summary of the 10th International Workshop on Salmon Smoltification. J Fish Biol. 2018;93(3):437–9. doi: 10.1111/jfb.13799 [DOI] [PubMed] [Google Scholar]
  • 14.Chittenden CM, Biagi CA, Davidsen JG, Davidsen AG, Kondo H, McKnight A, et al. Genetic versus rearing-environment effects on phenotype: Hatchery and natural rearing effects on hatchery- and wild-born coho salmon. PLoS One. 2010;5(8):e12261. doi: 10.1371/journal.pone.0012261 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Semeniuk CAD, Jeffries KM, Li T, Bettles CM, Cooke SJ, Dufour BA, et al. Innovating transcriptomics for practitioners in freshwater fish management and conservation: Best practices across diverse resource-sector users. Rev Fish Biol Fisheries. 2022;32(3):921–39. doi: 10.1007/s11160-022-09715-w [DOI] [Google Scholar]
  • 16.Jeffries KM, Teffer A, Michaleski S, Bernier NJ, Heath DD, Miller KM. The use of non-lethal sampling for transcriptomics to assess the physiological status of wild fishes. Comp Biochem Physiol B Biochem Mol Biol. 2021;256:110629. doi: 10.1016/j.cbpb.2021.110629 [DOI] [PubMed] [Google Scholar]
  • 17.Teffer AK, Bass AL, Miller KM, Patterson DA, Juanes F, Hinch SG. Infections, fisheries capture, temperature, and host responses: Multistressor influences on survival and behaviour of adult Chinook salmon. Can J Fish Aquat Sci. 2018;75(11):2069–83. doi: 10.1139/cjfas-2017-0491 [DOI] [Google Scholar]
  • 18.Miller KM, Li S, Kaukinen KH, Ginther N, Hammill E, Curtis JMR, et al. Genomic signatures predict migration and spawning failure in wild Canadian salmon. Science. 2011;331(6014):214–7. doi: 10.1126/science.1196901 [DOI] [PubMed] [Google Scholar]
  • 19.Eissa N, Wang H. Transcriptional stress responses to environmental and husbandry stressors in aquaculture species. Rev Aquacult. 2014;8(1):61–88. doi: 10.1111/raq.12081 [DOI] [Google Scholar]
  • 20.Connon RE, Jeffries KM, Komoroske LM, Todgham AE, Fangue NA. The utility of transcriptomics in fish conservation. J Exp Biol. 2018;221(Pt 2):jeb148833. doi: 10.1242/jeb.148833 [DOI] [PubMed] [Google Scholar]
  • 21.Islam SS, Heath DD, Dixon B, Karpowicz P, Vuu K, LeBlanc J, et al. Development of multi-species qPCR assays for a nanofluidic transcriptional profiling chip to assess responses of salmonids to changing environments. Can J Fish Aquat Sci. 2026;83:1–12. doi: 10.1139/cjfas-2024-0294 [DOI] [Google Scholar]
  • 22.Olsvik PA, Lie KK, Jordal A-EO, Nilsen TO, Hordvik I. Evaluation of potential reference genes in real-time RT-PCR studies of Atlantic salmon. BMC Mol Biol. 2005;6:21. doi: 10.1186/1471-2199-6-21 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Sun S, Wang Z, Yuan D, Ni M, Xu H, Wang W, et al. Selection of stable reference genes for RT–qPCR in Salmo trutta. Aquacult Rep. 2022;26:101290. doi: 10.1016/j.aqrep.2022.101290 [DOI] [Google Scholar]
  • 24.Kortner TM, Valen EC, Kortner H, Marjara IS, Krogdahl Å, Bakke AM. Candidate reference genes for quantitative real-time PCR (qPCR) assays during development of a diet-related enteropathy in Atlantic salmon (Salmo salar L.) and the potential pitfalls of uncritical use of normalization software tools. Aquaculture. 2011;318(3–4):355–63. doi: 10.1016/j.aquaculture.2011.05.038 [DOI] [Google Scholar]
  • 25.Gong Z, Zhang J, Chen Q, Li H, Zhang Z, Duan Y, et al. Comprehensive screening and validation of stable internal reference genes for accurate qRT-PCR analysis in Holotrichia parallela under diverse biological conditions and environmental stresses. Insects. 2024;15(9):661. doi: 10.3390/insects15090661 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Akbarzadeh A, Ming TJ, Schulze AD, Kaukinen KH, Li S, Günther OP, et al. Developing molecular classifiers to detect environmental stressors, smolt stages and morbidity in coho salmon, Oncorhynchus kisutch. Sci Total Environ. 2024;951:175626. doi: 10.1016/j.scitotenv.2024.175626 [DOI] [PubMed] [Google Scholar]
  • 27.Pinho N, Bombaça AC, Wiśniewski JR, Dias-Lopes G, Saboia-Vahia L, Cupolillo E, et al. Nitric oxide resistance in Leishmania (Viannia) braziliensis involves regulation of glucose consumption, glutathione metabolism and abundance of pentose phosphate pathway enzymes. Antioxidants. 2022;11(2):277. doi: 10.3390/antiox11020277 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Chapman JR, Waldenström J. With reference to reference genes: A systematic review of endogenous controls in gene expression studies. PLoS One. 2015;10(11):e0141853. doi: 10.1371/journal.pone.0141853 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Evans DH, Piermarini PM, Choe KP. The multifunctional fish gill: dominant site of gas exchange, osmoregulation, acid-base regulation, and excretion of nitrogenous waste. Physiol Rev. 2005;85(1):97–177. doi: 10.1152/physrev.00050.2003 [DOI] [PubMed] [Google Scholar]
  • 30.Stewart HA, Noakes DLG, Cogliati KM, Peterson JT, Iversen MH, Schreck CB. Salinity effects on plasma ion levels, cortisol, and osmolality in Chinook salmon following lethal sampling. Comp Biochem Physiol A Mol Integr Physiol. 2016;192:38–43. doi: 10.1016/j.cbpa.2015.11.011 [DOI] [PubMed] [Google Scholar]
  • 31.Nilsen TO, Ebbesson LOE, Madsen SS, McCormick SD, Andersson E, Björnsson BT, et al. Differential expression of gill Na+,K+-ATPase alpha- and beta-subunits, Na+,K+,2Cl- cotransporter and CFTR anion channel in juvenile anadromous and landlocked Atlantic salmon Salmo salar. J Exp Biol. 2007;210(Pt 16):2885–96. doi: 10.1242/jeb.002873 [DOI] [PubMed] [Google Scholar]
  • 32.Shrimpton JM, Sentlinger AMC, Heath JW, Devlin RH, Heath DD. Biochemical and molecular differences in diploid and triploid ocean-type chinook salmon (Oncorhynchus tshawytscha) smolts. Fish Physiol Biochem. 2007;33(3):259–68. doi: 10.1007/s10695-007-9138-5 [DOI] [Google Scholar]
  • 33.Ivanis G, Esbaugh AJ, Perry SF. Branchial expression and localization of SLC9A2 and SLC9A3 sodium/hydrogen exchangers and their possible role in acid-base regulation in freshwater rainbow trout (Oncorhynchus mykiss). J Exp Biol. 2008;211(Pt 15):2467–77. doi: 10.1242/jeb.017491 [DOI] [PubMed] [Google Scholar]
  • 34.Roberts D, Madliger CL, Mokdad AI, Firth BL, Pitcher TE. Evaluating the effects of probiotic supplementation and structural enrichment prior to transportation on the stress physiology of reintroduced juvenile Atlantic salmon. Environ Biol Fish. 2025;108(12):2039–69. doi: 10.1007/s10641-025-01691-x [DOI] [Google Scholar]
  • 35.Ocalewicz K, Gurgul A, Dobosz S, Jasielczuk I, Szmatoła T, Semik-Gurgul E, et al. Transcriptome profiling reveals differences between rainbow trout eggs with high and low potential for gynogenesis. Genes. 2025;16(7):803. doi: 10.3390/genes16070803 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Tipsmark CK, Madsen SS, Seidelin M, Christensen AS, Cutler CP, Cramb G. Dynamics of Na(+),K(+),2Cl(-) cotransporter and Na(+),K(+)-ATPase expression in the branchial epithelium of brown trout (Salmo trutta) and Atlantic salmon (Salmo salar). J Exp Zool. 2002;293(2):106–18. doi: 10.1002/jez.10118 [DOI] [PubMed] [Google Scholar]
  • 37.Wu S, Huang J, Li Y, Zhao L, Liu Z. Analysis of yellow mutant rainbow trout transcriptomes at different developmental stages reveals dynamic regulation of skin pigmentation genes. Sci Rep. 2022;12(1):256. doi: 10.1038/s41598-021-04255-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Nawata CM, Hung CC, Tsui TK, Wilson JM, Wright PA, Wood CM. Ammonia excretion in rainbow trout (Oncorhynchus mykiss): Evidence for Rh glycoprotein and H -ATPase involvement. Physiol Genomics. 2007;31(3):463–74. doi: 10.1152/physiolgenomics.00061.2007 [DOI] [PubMed] [Google Scholar]
  • 39.Aluru N, Vijayan MM. Hepatic transcriptome response to glucocorticoid receptor activation in rainbow trout. Physiol Genomics. 2007;31(3):483–91. doi: 10.1152/physiolgenomics.00118.2007 [DOI] [PubMed] [Google Scholar]
  • 40.McCormick SD, Regish A, O’Dea MF, Shrimpton JM. Are we missing a mineralocorticoid in teleost fish? Effects of cortisol, deoxycorticosterone and aldosterone on osmoregulation, gill Na+,K+ -ATPase activity and isoform mRNA levels in Atlantic salmon. Gen Comp Endocrinol. 2008;157(1):35–40. doi: 10.1016/j.ygcen.2008.03.024 [DOI] [PubMed] [Google Scholar]
  • 41.Kusakabe M, Yada T, Young G, Regish AM, McCormick SD. Expression of corticoid-regulatory genes in the gills of Atlantic salmon (Salmo salar) parr and smolt and during salinity acclimation. J Fish Biol. 2025. doi: 10.1111/jfb.70119 [DOI] [PubMed] [Google Scholar]
  • 42.Garcia de la Serrana D, Macqueen DJ. Insulin-like growth factor-binding proteins of teleost fishes. Front Endocrinol (Lausanne). 2018;9:80. doi: 10.3389/fendo.2018.00080 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Fast MD, Johnson SC, Jones SRM. Differential expression of the pro-inflammatory cytokines IL-1beta-1, TNFalpha-1 and IL-8 in vaccinated pink (Oncorhynchus gorbuscha) and chum (Oncorhynchus keta) salmon juveniles. Fish Shellfish Immunol. 2007;22(4):403–7. doi: 10.1016/j.fsi.2006.06.012 [DOI] [PubMed] [Google Scholar]
  • 44.Sun B, Skjæveland I, Svingerud T, Zou J, Jørgensen J, Robertsen B. Antiviral activity of salmonid gamma interferon against infectious pancreatic necrosis virus and salmonid alphavirus and its dependency on type I interferon. J Virol. 2011;85(17):9188–98. doi: 10.1128/JVI.00319-11 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Haugland O, Mercy IS, Romøren K, Torgersen J, Evensen O. Differential expression profiles and gene structure of two tumor necrosis factor-alpha variants in Atlantic salmon (Salmo salar L.). Mol Immunol. 2007;44(7):1652–63. doi: 10.1016/j.molimm.2006.08.015 [DOI] [PubMed] [Google Scholar]
  • 46.Landis ED, Palti Y, Dekoning J, Drew R, Phillips RB, Hansen JD. Identification and regulatory analysis of rainbow trout tapasin and tapasin-related genes. Immunogenetics. 2006;58(1):56–69. doi: 10.1007/s00251-005-0070-5 [DOI] [PubMed] [Google Scholar]
  • 47.Nygaard R, Husgard S, Sommer AI, Leong JA, Robertsen B. Induction of Mx protein by interferon and double-stranded RNA in salmonid cells. Fish Shellfish Immunol. 2000;10(5):435–50. doi: 10.1006/fsim.1999.0249 [DOI] [PubMed] [Google Scholar]
  • 48.Shrimpton JM, McCormick SD. Environmental and endocrine control of gill corticosteroid receptor number and affinity in Atlantic salmon (Salmo salar) during smolting. Aquaculture. 2003;222(1–4):83–99. doi: 10.1016/s0044-8486(03)00104-2 [DOI] [Google Scholar]
  • 49.Fukada H, Ozaki Y, Pierce AL, Adachi S, Yamauchi K, Hara A, et al. Salmon growth hormone receptor: Molecular cloning, ligand specificity, and response to fasting. Gen Comp Endocrinol. 2004;139(1):61–71. doi: 10.1016/j.ygcen.2004.07.001 [DOI] [PubMed] [Google Scholar]
  • 50.Breves JP, Phipps-Costin SK, Fujimoto CK, Einarsdottir IE, Regish AM, Björnsson BT, et al. Hepatic insulin-like growth-factor binding protein (igfbp) responses to food restriction in Atlantic salmon smolts. Gen Comp Endocrinol. 2016;233:79–87. doi: 10.1016/j.ygcen.2016.05.015 [DOI] [PubMed] [Google Scholar]
  • 51.Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, et al. Geneious basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics. 2012;28(12):1647–9. doi: 10.1093/bioinformatics/bts199 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Team R. R language definition. Vienna, Austria: R Foundation for Statistical Computing; 2025. [Google Scholar]
  • 53.Bates D, Maechler M, Bolker B, Walker S, Christensen RH, Singmann H, et al. Package ‘lme4’. Convergence. 2015;12(1):2. [Google Scholar]
  • 54.Fox J, Friendly GG, Graves S, Heiberger R, Monette G, Nilsson H, et al. The car package. R Foundation for Stat Comput. 2007;1109(1431):84039–5. [Google Scholar]
  • 55.Lenth RV. Least-squares means: The R Package lsmeans. J Stat Soft. 2016;69(1). doi: 10.18637/jss.v069.i01 [DOI] [Google Scholar]
  • 56.Mokhtar DM, Zaccone G, Alesci A, Kuciel M, Hussein MT, Sayed RK. Main components of fish immunity: An overview of the fish immune system. Fishes. 2023;8(2):93. doi: 10.3390/fishes8020093 [DOI] [Google Scholar]
  • 57.Aedo JE, Zuloaga R, Aravena-Canales D, Molina A, Valdés JA. Role of glucocorticoid and mineralocorticoid receptors in rainbow trout (Oncorhynchus mykiss) skeletal muscle: A transcriptomic perspective of cortisol action. Front Physiol. 2023;13:1048008. doi: 10.3389/fphys.2022.1048008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Breves JP, Shaughnessy CA. Endocrine control of gill ionocyte function in euryhaline fishes. J Comp Physiol B. 2024;194(5):663–84. doi: 10.1007/s00360-024-01555-3 [DOI] [PubMed] [Google Scholar]
  • 59.Houde ALS, Akbarzadeh A, Günther OP, Li S, Patterson DA, Farrell AP, et al. Salmonid gene expression biomarkers indicative of physiological responses to changes in salinity and temperature, but not dissolved oxygen. J Exp Biol. 2019;222(Pt 13):jeb198036. doi: 10.1242/jeb.198036 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Hevrøy EM, Hunskår C, de Gelder S, Shimizu M, Waagbø R, Breck O, et al. GH-IGF system regulation of attenuated muscle growth and lipolysis in Atlantic salmon reared at elevated sea temperatures. J Comp Physiol B. 2013;183(2):243–59. doi: 10.1007/s00360-012-0704-5 [DOI] [PubMed] [Google Scholar]
  • 61.Sneddon LU, Wolfenden DCC, Thomson JS. Stress Management and Welfare. Fish Physiology. Vol. 35. Academic Press; 2016. pp. 463–539. doi: 10.1016/b978-0-12-802728-8.00012-6 [DOI] [Google Scholar]
  • 62.Debes PV, Normandeau E, Fraser DJ, Bernatchez L, Hutchings JA. Differences in transcription levels among wild, domesticated, and hybrid Atlantic salmon (Salmo salar) from two environments. Mol Ecol. 2012;21(11):2574–87. doi: 10.1111/j.1365-294X.2012.05567.x [DOI] [PubMed] [Google Scholar]
  • 63.Best C, Durhack TC, Chapelsky AJ, Aminot M, Islam SS, Heath DD, et al. Transcriptional profiling provides insights into sublethal thermal stress thresholds in juvenile bull trout. Can J Fish Aquat Sci. 2025;82:1–17. doi: 10.1139/cjfas-2024-0378 [DOI] [Google Scholar]
  • 64.Banousse G, Semeniuk CAD, Garant D, Jeffrey JD, Islam SS, Bernier NJ, et al. Genetic and environmental basis of transcriptional thermal plasticity of brook charr (Salvelinus fontinalis) fry. Can J Fish Aquat Sci. 2025;82:1–15. doi: 10.1139/cjfas-2024-0205 [DOI] [Google Scholar]
  • 65.Howarth A, Islam SS, Firth BL, Heath DD, Cooke SJ. Transcriptomic responses to thermal and angling stress in wild brook trout from a southern Ontario stream. Conserv Physiol. 2025;13(1):coaf084. doi: 10.1093/conphys/coaf084 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Dang X, Chen Y, Islam SS, Heath DD, Dixon B. QuantStudio™ 12K flex OpenArray™ chip as a tool for high-throughput immunogene expression analysis in chinook salmon Oncorhynchus tshawytscha. Comp Immunol Rep. 2026;10:200276. doi: 10.1016/j.cirep.2026.200276 [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Table. List of candidate genes and their function.

(PDF)

pone.0360513.s001.pdf (86.5KB, pdf)
S2 Table. Primer and probe sequences.

(PDF)

pone.0360513.s002.pdf (119.6KB, pdf)
S3 Table. Pairwise comparisons of mean CT values among eight salmonid species.

(PDF)

pone.0360513.s003.pdf (74.5KB, pdf)
S4 Table. Effects of species and tissues on amplification success.

(PDF)

pone.0360513.s004.pdf (61.4KB, pdf)
S1 Fig. Histogram of model residuals.

(PDF)

pone.0360513.s005.pdf (47.8KB, pdf)
S2 Fig. QQ plot of residuals.

(PDF)

pone.0360513.s006.pdf (82.2KB, pdf)

Data Availability Statement

Metadata for the qPCR amplification success and the 24-h saltwater challenge experiment have been submitted to Figshare: https://doi.org/10.6084/m9.figshare.31879918.

Articles from PLOS One are provided here courtesy of PLOS

RESOURCES