Skip to main content
Download PDF

Transcriptional profiles reveal physiological mechanisms for compensation during a simulated marine heatwave in Yellowtail Kingfish (Seriola lalandi)

BMC Genomics volume 26, Article number: 230 (2025) Cite this article

Abstract

Background

Changing ocean temperatures are already causing declines in populations of marine organisms. Predicting the capacity of organisms to adjust to the pressures posed by climate change is a topic of much current research effort, particularly for species we farm or harvest. To explore one measure of phenotypic plasticity, the physiological compensations in response to heat stress as might be experienced in a marine heatwave, we exposed Yellowtail Kingfish (Seriola lalandi) to sublethal heat stress, and used the transcriptome in gill and muscle, benchmarked against heat shock proteins and oxidative stress indicators, to characterise the acute heat stress response (6 h after the initiation of stress), and the physiological compensation to that response (24 and 72 h after the initiation of stress).

Results

The heat stress experiments induced elevations in heat shock proteins, as measured in blood, demonstrating the sublethal stress level. The initial response (6 h) to heat stress included the expected cellular stress response. Exposure of 24 h or more led to altered transcriptomic patterns for protein degradation, membrane transporters, and primary metabolism. In the muscle, numerous transcripts with mitochondrial function had altered abundance. There was a profound change to the regulation of transcription, as well as numerous transcripts with differential exon usage, suggesting that this may be a mechanism for conferring physiological resilience to heat stress.

Conclusions

These results demonstrate the processes involved in acclimation to heat stress in this species, and the utility of using the transcriptome to assess plasticity. It also showed that differential exon usage may be an important mechanism for conferring plasticity. Future work should investigate the role of genome regulation, and alternative splicing in particular, on conferring resilience to temperature changes.

Graphic Abstract

Peer Review reports

Introduction

Anthropogenic release of carbon dioxide, along with other greenhouse gases, is resulting in an increase in global air and sea surface temperatures [1]. Additional temperature increases are expected over the coming decades. The pressures may be especially acute for some marine species, which have more narrow temperature ranges than their terrestrial counterparts, and thus, may be near their thermal maxima [2, 3]. While there is an overall warming trend, short term extreme warming can also occur via marine heatwaves, where temperature increases by a few degrees for a period of weeks to months [4,5,6]. The sudden temperature increases caused by marine heatwaves could alter the physiology of marine organisms, particularly for those with narrow temperature tolerances or living close to their thermal maximum [2, 6]. Indeed, climate change is already changing the distribution of many marine organisms, with many migrating poleward or into deeper water [2, 7]. In addition to the profound ecological consequences [2, 8, 9], these temperature changes pose a risk for fishing and aquaculture [10, 11].

Understanding the physiological response to warming will help to predict the response of individual species to climate change. As fish are ectothermic, increasing temperature increases rates of metabolism and oxygen consumption [12]. At high temperatures, oxygen demand may exceed oxygen availability in body fluids, resulting in oxidative stress, a breakdown of molecular repair, and cessation of protein synthesis [13]. Simultaneously, at high temperatures, oxygen solubility in seawater decreases, further exacerbating oxygen stress [2, 9, 14]. Sufficient temperature stress results in damaged proteins, DNA and lipids [15]. In response to thermal stress, there is a rapid transcription of heat shock proteins (HSP)’s, frequently coupled by exonucleases degrading other transcripts [15]. If this heat related damage persists, cells may undergo proteolysis, apoptosis and inhibition of cell division [15]. In addition, the oxygen limitation induced by high temperature may result in anaerobic metabolism, with further energetic costs [13]. The process is energetically costly, and results in a depletion of ATP.

At relatively warm but sub-lethal (< Tcritical) temperatures, organisms may divert resources away from reproduction and other critical functions [13]. These temperature exposures can lead to energetic depletion, and are known as capacity impacts [16]. At the pejus temperature, all physiological processes proceed normally. A common assumption is that organisms with the greatest flexibility in both their critical temperature and pejus temperature will have the greatest capacity to survive the coming decades, through both acclimation (an ability as the ability to adjust to environmental change(s) over a defined temporal scale occurring within the lifespan of an individual) and plasticity (the ability of an organism with a given genotype to change its behaviour, morphology and/or physiology in response to its present environment) (e.g [17]).

The ability of an organism to adapt to changing environments will depend, in part, to their phenotypic plasticity. Plasticity is defined as changes in the phenotype, despite a constant genotype, that occur in response to environmental stimuli [18]. Plasticity has physiological costs [18, 19], but is favoured in heterogenous environments. It also allows organisms to make physiological adjustments to changing environments [17, 18]. Plasticity may be mediated via the regulation of the transcriptome; e.g. via alternative splicing or epigenomic regulation. The heat stress response in fish has previously been shown to have phenotypic plasticity. For example, numerous studies have shown that prior exposure to warm temperatures results in higher basal transcription levels of HSP’s as well as higher critical temperatures (e.g [20, 21]). Previous studies have also shown that changes in gene expression contribute to adaptive diversity [22]. While many current ecological studies focus on traits as fixed entities (e.g [23]), examining the plasticity of these traits may be a better predictor of an organisms’ capacity to cope with climate change. For example, the delta smelt (Hypomesus transpacificus) has a limited capacity to modulate the transcriptome and is less tolerant of thermal pressures relative to invasive fish that inhabit the same environment, perhaps contributing to declining populations [24].

However, physiological adjustments and compensation to temperature stress may result in physiological costs that impair the health of individuals, populations, and even ecosystems. Sublethal heat stress in fish results in: (i) the release of stress hormones; (ii) metabolic alterations, such as induction of gluconeogenesis, and elevated plasma glucose levels; (iii) a diversion of energy away from growth and reproduction [25]. Previous studies have found that zebrafish exposed to more variable temperatures during embryological development had increased thermal tolerance, but slower growth [26]. As these processes are energetically costly, plasticity may result in lipid depletion and reduce the quality of the fishes as prey [2]. Also, sublethal heat stress slows protein synthesis in fish, which is an important energy store in these organisms [16, 27].

The response of fishes to heat stress will be species specific and will depend on their physiological tolerance and life history [16]. Acclimation (here defined as a physiological adjustment within a single organism’s lifespan) to stress response is possible, and these changes can be measured in the transcriptome or proteome [15]. Individuals exposed to higher or more variable ambient temperatures have higher basal transcription levels of heat shock proteins, suggesting the potential for thermal plasticity [15]. However, other physiological adjustments to heat stress are yet to be quantified [28].

The transcriptome has been used to characterise the response to heat stress [16, 24, 29]. However, in most studies, tissues are collected at a single time period after exposure, and thus measure the initial response, but not the compensatory mechanisms used for recovery and subsequent acclimation to new temperatures. To better understand how sublethal heat stress could impact the physiology and ecological resilience of fish, we exposed yellowtail kingfish (Seriola lalandi) to elevated temperatures similar to those experienced at the onset of a marine heatwave and measured a range of responses. This fish is distributed globally in tropical and subtropical environments, and thus has a wide temperature tolerance [30]. However, recent studies have found different temperature optima depending on the organisms’ thermal history [30,31,32,33], suggesting that it is a good model for thermal plasticity. We evaluated responses in two tissues to allow a holistic assessment of the impacts of temperature increases on the physiology of the fish. We evaluated responses in two tissues to allow a holistic assessment of the impacts of temperature increases on the physiology of the fish. As animals often use differential splicing to maintain a tissue specific proteome [34], transcriptomes were synthesised for each of the tissues. The gill was chosen for study because of high metabolic activity, and because of the physiological importance as a site of oxygen uptake, acid base regulation and nitrogenous waste excretion [20]. The muscle was chosen because of metabolic importance. We hypothesised that the transcriptomic changes, whether measured via differential transcript abundance or differential exon utilisation, could be used to infer the mechanisms by which the fish adjust their physiologies in the face of a sublethal heat stress. The results give an indication of the capacity of this species in particular and provide additional evidence for general stress responses to both extreme events and a warming ocean.

Materials and methods

Animal husbandry

All work with fish was conducted under a permit from Western Australia’s Department of Primary Industries and Regional Development Animal Ethics Committee (permit number: AEC 22-2-06). This work resulted from tissues shared from a subset of fish in a larger study. A total of 90 juvenile yellowtail kingfish (less than 1 year old, approximately 262 ± 2.2 g) were sourced from the DPIRD commercial nursery (Perth, Western Australia) for use in the experiments. These were F1 fish whose parents were wild caught locally (offshore of Rottnest Island, WA, Australia) in 2017. Fish were anesthetised with benzocaine (50 mg·L − 1, ADSI Pty Ltd) and individually radio frequency identification tagged in the proximal dorsal muscle and weighed. They were randomly distributed into six 600 L tanks, with flow through of 8 L per minute. Three tanks were maintained at ambient temperature (20o C − 20.1 ± 0.1) and three were heated to 27 o C (27.5 ± 0.5) within 120 min by setting the heating/cooling system to 27 o C. These temperatures were maintained for the duration of the experiment. Oxygen was directly injected into the incoming water lines for both tanks, and the water was at full saturation for the duration of the experiment. Fish were fed a temperature dependent maintenance ration– with the fish in the 20 ° C tank receiving 3.5% of their body weight per day, and the fish in the 27 ° C tank receiving 4% of their body weight per day. At 6 h, 24 h, 3 days, 8 days and 18 days, five fish from each tank were collected and rapidly euthanised with an overdose of anaesthetic (benzocaine 100 mg.L− 1, ADSI Pty Ltd., Australia). Blood was collected to measure protein oxidation, and gill (the central arch of the second gill arch on the left side) and muscle (the left proximal dorsal muscle) collected for transcriptomic profiling. Tissue was collected for RNA extractions at 6 H, 24 H, and 72 H and stored in RNAlater, kept at 4o C overnight then transferred to -20 o C for storage until processing.

Protein bioassays

Whole blood was collected from the caudal vein of the sampled fish and stored in 1 mL foetal tubes (K3E EDTA Mini collect®, Greiner Bio-One, Austria). The whole blood was stored at 4 °C overnight prior to centrifugation at 3000 ×g for 10 min. The plasma was collected at stored at -80 °C for plasma for heat shock protein (Hsp70) determination. The concentration of Hsp70 in plasma was assayed in duplicate using an enzyme-linked immunosorbent assay (ELISA) kit (Fish Hsp70 ELISA E16327Fh, Cusabio Biotech Co., Ltd., Texas, USA) following the manufacturer’s instructions. A standard curve for the HSP 70 was run, and an R2 of 0.99963 was obtained. Statistical analysis (ANOVA) was performed in R [35].

RNA extractions

RNA was extracted at the Australian Centre for Disease Preparedness (ACDP) in Geelong, VIC, Australia following procedures similar to those used previously [36, 37]. Only the individuals sampled at the beginning of the heat stress exposure (e.g. 6 h, 24 h, 72 h) were used for transcriptome profiling due to the cost of the analysis and our hypothesis that the early time points would be the most dynamic. Briefly, preserved tissues were placed in a tube with silicon carbide beads and 700 µl QIAzol, and tissue was disrupted using a Omni Bead Ruptor. Gill tissues were agitated at 5.5 m/s for 30 s and then immediately placed into ice. The Qiagen miRNeasy Mini kit was used for the rest of the extraction as per manufacturer’s instructions. RNA quality was ascertained using a Bioanalyzer RNA Nano chip (Agilent). RINs were consistently above 8.0 and often above 8.5 for gill. For muscle, the process was modified such that tissues were agitated at 5.5 m/s for 30 s, placed into ice for 2 min, then agitated at 5.5 m/s for another 30 s. RNA quality was typically between 7.0 and 9.0 RIN for muscle. RNA concentration was also measured using the Bioanalyzer. Yield varied but was always sufficient to produce the 75 ng per µl required for sequencing. The Bioanalyzer did not indicate any trace DNA contamination for either tissue.

Sequencing

RNA sequencing was performed at the Ramaciotti Centre for Genomics (University of New South Wales, Sydney, NSW Australia). Briefly, approximately 2 µg total RNA were used in an Illumina stranded mRNA prep ligation using the Illumina Stranded mRNA prep kit with UDIs. These libraries were sequenced on a NovaSeq s2 1 × 100 bp Flowcell with PhiX spike-in, using the NovaSeq 6000 S2 100 cycle kit. Each RNA prep resulted in two libraries, which each had approximately 20 million reads. On average, each treatment yielded 40 million reads. Reads were uploaded to NCBI’s short read archive and can be accessed using BioProject number PRJNA1086548.

Assembly and mapping

Reads were trimmed for quality using trimmomatic (version 0.39) [38] and the parameters “CROP:96 HEADCROP:12 LEADING:28 TRAILING:28 SLIDINGWINDOW:4:28”. Trimmed reads were assembled into a de novo transcriptome using Trinity (version 2.13.2) [39, 40] with the added parameter “--min_kmer_cov 2”. A separate assembly using identical parameters was created for muscle and gill to allow for tissue-specific splice variants. The completeness of the transcriptome was assessed using BUSCO (version 5.2.2) against the actinopterygii lineage [41], and the transcriptome assembly checked with the Trinity stats function. Reads were mapped to the transcriptome using salmon [42] (version 1.8.0). An operational annotation to both the nr and uniprot databases was created using Diamond (version 2.0.15) [43]. Transcripts were also mapped to the subspecies (Yellowtail Amberjack– Seriola lalandi dorsalis) with annotation in Ensembl for pathway analysis.

Differential abudance

Differential abundance was calculated using the DESeq2 package in Bioconductor (version 3.2) for R [35, 44, 45]. Transcripts were considered differentially abundant if they had a log2 fold change in abundance with an absolute value greater than one and a false discovery rate corrected p value of less than 0.05. Heat maps and clustering for visualisation were created with the pheatmap package, which uses Euclidean distances by default [46], and Venn diagrams were created with Venny [47]. All scripts used for input into R are available in CSIRO’s data access portal. GO term and Pathway analysis were conducted using DAVID (version 6.8) [48,49,50] with the whole genome as the background.

Differential exon usage

Differential Exon usage (henceforth DEU) was determined using the SuperTranscript concept [51] in the Trinity toolkit. SuperTranscripts were generated using the “Trinity_gene_splice_modeler.py” function, and used the STAR aligner (version 2.7.11b) [52] to map reads to SuperTranscripts. DEU was calculated using the DEXSeq package (version 3.4) [53] in Bioconductor from R. To better understand the functional significance of the differentially utilised transcripts, reads were also mapped to the coding sequences of the yellowtail amberjack using the Salmon algorithm and DEU was again calculated using DEXSeq (53).

Results

The results first anchor the sublethal heat stress by describing the impact of the heat stress on the mortality and growth (described in Sect. 3.1), and the evidence of heat stress (described in Sect. 3.2). We then use the transcriptome (the assembly is described in Sect. 3.3) of fish collected after 6, 24, 72 H exposure to continuous heat stress to characterise physiological adjustments in the gill and the muscle (described in Sect. 3.4 and subsequent sections).

Fish mortality and growth

No mortality was recorded in any of the treatments, except for one individual which jumped from the tank (data not shown). There was no difference in growth between treatments throughout the month they were held at high temperatures (Supplementary Table S1, t-test, p = 0.5), which is consistent with our goal of sub-lethally stressing the fish and the maintenance rations they were provided.

Indications of heat stress response in blood proteins

HSP 70 levels in blood, measured via ELISA in plasma, were higher in fish from the warm tanks than in those from the ambient tanks (t-test, p < 0.001), Fig. 1. These increases in HSP persisted throughout the exposure period.

Fig. 1
figure 1

HSP 70 levels, presented in arbitrary units, in blood in fish from warm or ambient tanks. Mean values (n = 4) and ± 1 standard deviations are shown. 6 h, 24 h, 4 days, 8 days and 18 days

Full size image

Assembly results for gill and muscle transcriptome

For the muscle extractions, 28 of the 30 RNA preps yielded sufficient quality material for sequencing– one replicate from the 6 h ambient treatment and one replicate from the 72 h ambient treatment had to be abandoned because of low RNA quality (data not shown). All RNA preparations from gill tissue were of sufficient quality to be used in sequencing. Illumina High throughput sequencing resulted in approximately 2.4 billion reads (approximately 40 million per treatment), with an average quality score above 36 before trimming. Further details about the library size and trimming are available in Supplemental Table S2. On average, 99% of reads could be used in the assembly. Trinity assemblies were performed iteratively to produce the most complete assembly with the highest proportion of reads mapping. The best quality assembly was obtained when gill and muscle were assembled separately, and min-kmer was set to 2. High quality transcriptome assemblies were obtained for both tissues. The gill assembly contained 282,295 contigs and was 85% complete according to BUSCO analysis, whereas the muscle assembly contained approximately 141,121 contigs and was 79% complete according to the BUSCO analysis. In both tissues, greater than 95% of reads mapped to the assemblies. Additional information about the quality of the assemblies is provided in Supplemental Table S3. When reads were aligned to the coding sequences of the Yellowtail Amberjack, the mapping percentages are lower -approximately 89%, so all functional analyses were conducted with the Trinity assembly.

Differential transcript abundance

The overall transcriptomic profiles were very similar, regardless of temperature or time point (see similarity matrix in Supplemental Figure S1). All profiles were 90% similar in gill (Supplemental Figure S1A) and at least 85% similar in muscle (Supplemental Figure S1B). The number of differentially abundant transcripts with heat treatment varied with tissue and time point, (Supplemental Figures S2 and 3; Supplemental Table S4). Gill had a more dynamic transcriptome than muscle; there were more transcripts with altered abundance and the magnitude of change was greater in gill than in muscle. The identity of the differentially abundant transcripts varied with time, with very little overlap (Supplemental Figure S2). Full lists of differentially abundant transcripts are provided in Supplemental Table S5.

When clustering is performed on these differentially abundant to determine relatedness between samples, some trends emerge in both gill (Fig. 2) and muscle (Fig. 3). In gill, “warm” treatments always form a distinct cluster relative to “ambient” treatments, however, timepoints only cluster with the 6 h differentially abundant transcripts (Fig. 2a). With the other differentially abundant transcript lists, the timepoints are interspersed (Fig. 2b and c). In muscle, the warm and ambient transcriptomic profiles are not consistently separated. At 6 h (Fig. 3a), the warm treatment form a distinct cluster. The same pattern is seen for the 72 h heat exposed fish using the transcripts with altered abundance at 24 h (Fig. 3b).

Fig. 2
figure 2

Heat maps showing clustering patterns amongst the differentially abundant gill transcriptome for individual fish by differentially abundant transcripts. Transcripts that are differentially abundant in the 6 h warm treatment relative to the 6 h ambient treatment are shown in panel A, those that are differentially abundant in the 24 h warm treatment relative to the 24 h ambient are shown in panel B, and those that are differentially abundant in the 72 h warm treatment relative to the 72 h ambient are shown in panel C. The colour in the heat map represents the relative abundance of each transcript as TPM (scaled to the range for each transcript), with those transcripts most increased in abundance shown as red, and those most decreased in abundance shown in blue. Treatments are shown in the colour bars at the top dendogram, and the labels at the bottom of the plot show individual tissues, with the notation tissue_tank number_replicate number

Full size image
Fig. 3
figure 3

Heat maps showing clustering patterns amongst the muscle transcriptome for individual fish by differentially abundant transcripts. Transcripts that are differentially abundant in the 6 h warm treatment relative to the 6 h ambient treatment are shown in panel A, those that are differentially abundant in the 24 h warm treatment relative to the 24 h ambient are shown in panel B, and those that are differentially abundant in the 72 h warm treatment relative to the 72 h ambient are shown in panel C. The colour in the heat map represents the relative abundance of each transcript as TPM (scaled to the range for each transcript), with those transcripts most increased in abundance shown as red, and those most decreased in abundance shown in blue. Treatments are shown in the colour bars at the top dendogram, and the labels at the bottom of the plot show individual tissues, with the notation tissue_tank number_replicate number

Full size image

Differential exon utilisation (DEU)

In addition to the heat stress causing differences in transcript abundance, we measured DEU (e.g. transcripts that are spliced differently) in transcripts from gill and muscle in fish exposed to heat stress when compared to those exposed to ambient temperatures (Fig. 4). In contrast to the patterns observed for transcript abundance, we observed more DEU at later time periods (Supplemental Figure S3). We also observed many more transcripts with DEU in gill than in muscle, particularly at the two early time points (Fig. 4, supplemental Figure S3). There is also more individual variability in differential exon utilisation in the muscle than there is in the gill (Fig. 4). Full lists transcripts with differential exon usage are provided in Supplemental Table S6.

Fig. 4
figure 4

Transcripts with differential exon usage, coloured on a relative scale as depicted in the colour bar to the left of each panel. Panel A depicts the transcripts with differential exon usage in the gill after 6 H exposure to heat stress; Panel B depicts the transcripts with differential exon usage in the muscle after 6 H exposure to heat stress; Panel C depicts the transcripts with differential exon usage in the gill after 24 H exposure to heat stress; Panel D depicts the transcripts with differential exon usage in the gill after 24 H exposure to heat stress; Panel E depicts the transcripts with differential exon usage in the gill after 72 H exposure to heat stress; Panel F depicts the transcripts with differential exon usage in the muscle after 72 H exposure to heat stress. Patterns in individual animal replicates are plotted

Full size image

In an attempt to better understand the functional significance of the differential exon usage, reads were also mapped to the coding sequences from Yellowtail Amberjack. By contrast, as shown in Supplemental Table S7, there were far fewer differentially spliced transcripts when reads are mapped to the coding sequences of this closely related species.

Altered physiological pathways

We also used the enriched clusters function in DAVID [43, 44] to identify altered physiological pathways and enriched GO terms for differentially abundant transcripts and transcripts with DEU. Kegg pathways are plotted in Figs. 5 and 6, GO terms are plotted in Supplemental Figures S5-8; InterPro terms are plotted in supplemental Figure S9-10). Full gene lists containing read counts and differential analysis are available in Supplemental Table S4 for differential transcript abundance and in Table S5 for Differential Exon Usage. The transcriptomic profiles are grouped by physiological process for enhanced readability, as summarised in Table 1. A brief overview of the results is provided here; additional details are available in the Supplementary Materials.

Fig. 5
figure 5

Physiological pathways with altered transcriptomic profiles in the gill. Rows denote individual KEGG pathways that were significantly enriched (p < 0.05). Each column represents aggregated enrichment results across treatments. The colour scale shows enrichment score. The number denotes the time elapsed (in hours) from the beginning of the heat exposure; DEU is differential exon usage; ITA is increased transcript abundance; DTA is decreased transcript abundance. 72gillITA is not plotted because no KEGG terms were significantly enriched

Full size image
Fig. 6
figure 6

Physiological pathways with altered transcriptomic profiles in muscle. Rows denote individual KEGG pathways that were significantly enriched (p < 0.05). The colour scale shows enrichment score. Each column represents aggregated enrichment results across treatments. The number denotes the time elapsed (in hours) from the beginning of the het exposure; DEU is differential exon usage; ITA is increased transcript abundance; DTA is decreased transcript abundance. 24muscITA is not plotted because no KEGG terms were significantly enriched

Full size image

As summarised in Table 1, transcripts known to be involved in the cellular stress response were modulated in both tissues. Four transcripts encoding HSP70 and 3 encoding HSP 90 had increased abundance in the gill tissues at six hours exposure to thermal stress, and 3 transcripts encoding HSP70 had increased abundance in the muscle, although far fewer were had increased abundance at later time points, and the response is more variable (Fig. 7; supplemental Figure S9). Heat shock proteins also showed DEU, as did transcripts for RNA degradation (Figs. 5 and 6).

Transcripts governing “the regulation of transcription” were greatly modulated in abundance in both tissues at all time points, and many of these also showed DEU (Figs. 5 and 6). Specifically, the spliceosome functional group had 9 and 6 enriched transcripts at 6 h in gill and muscle, respectively. Five transcripts involved in both RNA splicing, and mRNA processing were upregulated at 6 H in the gill. As many as 24 transcripts involved in RNA binding were enriched in the muscle (Figs. 5 and 6; supplemental Figure S9). At 72 h, these transcripts in these functional categories are still modulated; however abundance can increase or decrease depending on the tissue and specific transcript.

There were also changes in the abundance and exon utilisation of transcripts related to different cell signalling pathways in the gill, but not the muscle, at all time points (Fig. 5; Supplemental Figures S5 and S9). Some of the enriched categories include Pleckstrin, 5 transcripts involved in regulation of small GTPase mediated signal transduction, G-protein coupled receptor signalling pathway, G-protein coupled receptor binding (Supplemental Figures S5 and S9).

Transcripts encoding transmembrane proteins had increased abundance and differential exon utilisation in the gill at all timepoints (Fig. 5). These transcripts were also increased in abundance in the muscle at 72 H. There was some variability in the response, with some transcripts encoding transmembrane proteins showing a decrease in abundance. Notably, oxygen binding proteins show DEU in the gill at 24 h, and transcripts mapped to the haemoglobin complex showed DEU in the gill at 24 H (Supplemental Figure S5). The InterPro term for Haemoglobin is also enriched amongst the DEU transcripts in the gill 24 h into heat exposure (Supplemental Figure S9).

Numerous transcripts with roles in energetics and metabolism, such as lactate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase, had increased abundance in the gill, but not the muscle. Mitochondrial transcripts were also modulated (Supplemental Figures S6, S8, S10). Although the abundance of transcripts involved in energetics is relatively stable following heat stress, many transcripts show DEU in muscle, such as many as 13 transcripts involved in the pathways “Glycolysis / Gluconeogenesis”, 7 transcripts involved in starch and sucrose metabolism, 5 involved in galactose metabolism, 7 transcripts involved in fructose and mannose metabolism, and 10 transcripts in the composite groupings of carbon metabolism (Fig. 6).

Transcripts encoding proteases had increased abundance after 24 h in both tissues (Figs. 5 and 6). Transcripts encoding proteins involved in proteolytic activity are upregulated after 6 h heat exposure in muscle and after 72 h exposure to heat stress in both gill and muscle (Fig. 6). For example, 7 and 4 peptidases are increased in abundance in the gill and muscle at 72 h. In addition, at 72 h, the GO analysis suggests that the major changes relate to repair of damaged proteins (e.g. protein deubiquitination, transfer of glycosyl groups are upregulated, chaperone binding is has decreased abundance) (Supplemental Figure S6).

Transcripts encoding structural proteins also had increased abundance after 6 h in both tissues (Figs. 5 and 6), and were DEU most notably in the muscle after 72 h (Fig. 6). Structural proteins, including ankyrin, actin and tropomyosin, have increased abundance in the gill and muscle transcriptome 6 h after exposure to heat stress (Figs. 5 and 6). However, many had decreased transcript abundance 72 h after heat exposure. These transcripts, for InterPro terms “Actins”, “tubulin”, “Myogenic basic muscle-specific protein” and “tropomyosin” in particular, had DEU in the muscle transcriptome (Fig. 6, Supplemental Fig. 10). Notably, structural proteins such as actin and tropomyosin are among the very few transcripts with DEU when reads are mapped against the de novo assembled transcriptome as well as against the yellowtail amberjack coding sequences (data not shown).

Transcripts in other pathways showed different transcriptomic profiles in each tissue, as summarised in Table 1. The abundance and exon usage of transcripts with roles in cell signalling, oxidative stress, as well as fatty acid metabolism and steroid biosynthesis were modulated in the gill (Figs. 5 and 6), but not in the muscle.

Fig. 7
figure 7

Read abundance of representative Heat Shock Proteins. Panel A shows heat shock protein 70 (XP_023285174.1); Panel B shows HSP90 (XP_028253067.1); Data are shown as a box plot, with the center bar showing the median, the upper and lower edge of the box the 75th and 25th percentiles, the whiskers showing the 10th and 90th percentiles, and dots showing outliers. To best show differential transcript abundance, gill and muscle are frequently plotted with different scales. An asterisk denotes significant difference in read abundance between ambient and warm treatments. Other heat shock protein isoforms are shown in Supplemental Figure S4

Full size image

This pathway is also amongst enriched Gene Ontology terms: the initial change in the transcriptome shows an increase in chaperone proteins such as the HSP’s (the unfolded protein binding and protein folding). Some heat shock proteins are still upregulated in the gill transcriptome 24 h after heat stress began; although fewer than in the 6 h, and they are not increased in abundance to the same degree. After 72 h continuous exposure to heat stress, some transcripts encoding isoforms of HSP are still upregulated in the gill. Similar trends are observed in the InterPro Pathways 6 h after initiation of heat stress only (Supplemental Figure S9).

DEU was also measured amongst the Heat Shock Proteins. Transcripts encoding isoforms of HSP 70, 71, and 90 showed DEU in the gill (Supplemental Table 5). HSPs were amongst the most enriched InterPro DEU pathways in the gill after 6 h and 72 h exposure to heat stress (Supplemental Figure S9), although not KEGG pathways (Figs. 5 and 6) or GO terms (Supplemental Figs. 7–8). The KEGG pathway for RNA degradation was enriched amongst DEU transcripts in muscle (Fig. 6), suggesting activation of the cellular heat stress response.

The pattern of heat shock protein transcript abundance in the muscle is slightly different. While HSPs have increased abundance in muscle after 6 h exposure to heat stress, as shown via the GO terms (for unfolded protein binding and protein folding; Supplemental Figure S6),and InterPro terms, they are not the most upregulated transcripts (Supplemental Table S4). There were fewer HSP’s with DEU in muscle.

Table 1 Summary of transcriptomic results and pathways altered in each tissue
Full size table

Discussion

In this study, Yellowtail Kingfish were exposed to a sublethal temperature challenge representing a marine heatwave. To characterise the processes involved in acclimation to elevated temperatures, changes in the transcriptome, measured via both differential transcript abundance and differential exon usage, were characterised in the gill and muscle at 6, 24 and 72 h into the heat stress experiment. The transcriptomic and HSP results show not only the acute, sublethal stress response to elevated temperature stress in Yellowtail Kingfish, but also acclimation processes and compensatory pathways inducted as the fish compensates for the energetic costs associated with the acute temperature stress. We only found a transcriptome profile corresponding to moderate heat stress, which is consistent with the lack of mortality and similar growth across treatments (Table S1). Transcriptomic patterns of differential transcript abundance in the muscle and gill initially (6 h post exposure) show a classic temperature stress response, with multiple isoforms of the heat shock proteins upregulated. As the animal acclimated to elevated temperatures, the transcriptomic patterns shift to repair of damaged macromolecules, and in the muscle, altered metabolic pathways. The kinetics of change were different than for differential transcript abundance, particularly in the muscle, with the most profound changes observed in later time points. Intriguingly, the transcripts with differential exon usage frequently encode proteins that are known to be temperature sensitive (e.g. junction proteins in the gill and cytoskeletal proteins in the muscle) [15].

The gill transcriptome was larger than that of the muscle, which is expected given its diverse roles in oxygen transport, acid-base metabolism, and excretion of nitrogenous waste [20]. Previous studies have also found it to be more transcriptionally active than muscle [15]. Consistent with previous studies, we found different transcriptomic patterns in the gill and muscle [15], as well as different patterns of DEU (Table 1).

These responses provide insight into the mechanisms of acclimation (here defined as reversible phenotypic change in response to an environmental trigger) [54] to temperature stress. The transcriptomic changes reported in this study illustrate the processes used in fish to acclimate to heat stress, and provide insight into the potential for energetic costs, as well as suggesting targets for study of the potential plasticity and thermal resilience. Although we do not know how the changes in exon usage translate to protein structure and function, we hypothesise that DEU could be important in conferring mechanisms for plasticity and acclimation (e.g [55]).

The acute stress response

The initial changes in the transcriptome measured in each tissue as measured in this study are consistent with the acute stress response to a temperature challenge [15, 56]. In blood, we measured a consistent increase in HSP70 levels in fish from the 27° C tank relative to the 20° C tank, including at 24 h, where there is an unexplained decrease in protein levels in both tanks. The difference between tanks remains consistent. The transcriptomic profile is different (and varies depending on the individual transcript), where frequently the HSP 70 transcripts have increased abundance at initial time points but not at later time points, suggesting that the protein, once synthesised, is persistent. The induction of heat shock proteins that we measured using both ELISA and transcriptome profiling is a well-documented stress response in fish [25]. The increased abundance of molecular chaperones, such as HSPs, reduces the degree of structural unfolding under heat stress and allow for the repair of damaged macromolecules [56]. If the cellular damage can not be repaired, further stress responses, including cell cycle arrest and apoptosis, are induced [56]. Consistent with our findings, previous studies have found that HSPs are upregulated even after mild heat stress, ubiquitin was upregulated following moderate heat stress, and that apoptosis related proteins were upregulated following extreme heat stress [20]. Most of the changes in the transcriptome were measured via differential transcript abundance, with comparatively few transcripts showing differential exon usage (relative to the two later time periods). Although the heat stress proteins are small, the re-arranging of the overall transcriptome can alter other physiological processes and as a consequence, is physiologically costly [15], such that even organisms that activate the CSR and repair damaged proteins have fitness costs. The other transcriptomic changes we measured suggest the possibility of metabolic costs, as described below.

Compensatory responses to heat stress

Following the well characterised cellular stress response [15, 29] that was induced at early time points, we observed a change in transcriptional pathways (both via differential transcript abundance (DTA) and differential exon utilisation (DEU)) that may be indicative of the compensatory responses. Initial time points showed increases in HSP’s and a rearrangement of transcriptional machinery. This induction of the spliceosome may have led to the large number of transcripts with DEU we measured at later time points. Rapid induction of transcription factors may be a response to environmental stress and may be “primed” in organisms from variable environments [29]. Alternative splicing has been shown previously to be an important response to heat stress in fishes [57]. Previous studies have shown that heat stress in catfish induces alternative splicing [58]. RNA binding genes, which modulate RNA degradation, were differentially spliced, showing the impact of heat stress on transcriptional processes [58]. Similar processes, including upregulation of the splicing factor prpf38b, were invoked in redside dace (Clinostomus elongatus) exposed to an acute thermal stress [57]. These splice variants may be the mechanism through which plasticity in conserved [18], and may result in adaptation to new niches (e.g [59].

At the 24- and 72-hour time points in our study, cell signalling, protein degradation, energy metabolism, and mitochondrial pathways were all altered. This is consistent with the molecular changes measured in other studies (e.g [16]. Alterations in signalling and energetic pathways were measured via both DTA and DEU. Temperature stress is well known to cause damage to macromolecules [60], which may explain the alterations in protein degradation pathways we observed. The changes in amino acid synthesis pathways, translation, and in the ribosome may reflect a cessation of protein synthesis, a known response to intracellular oxygen limitation initiated by thermal stress [13]. Heat stress has previously been shown to slow protein synthesis, which is an energy store in fish [16, 27]. Previous studies have found alterations to protein ubquination following temperature stress [61]. The changes in the mitochondrial transcripts observed may reflect the organisms changing mitochondrial density and function [13]. Mitochondrial efficiency declines at high temperatures, possibly due to inhibition of enzymes with roles in substrate oxidation [12]. Alternatively, these responses may be due to the energetic costs of the acute heat stress response, even in sublethal exposures.

As energy is shunted to the heat stress response (as reviewed by [56]), the transcriptome reflects changes in the fish’s energetics as well as diversion away from other metabolic processes. In the muscle, energetic changes were measured both as changes in transcript abundance and changes in exon usage. These trends are observed in the findings of previous studies as well. A study that examined the change in the transcriptome in sea horses collected from the southern edge of their range 30 min after exposure to a temperature increase found induction of HSPs, as well as a switch to anaerobic metabolism, decrease in muscle growth, and induction of apoptosis [56, 59]. Although the induction of apoptosis suggests a more severe response to temperature stress than measured in this study, the changes measured in the sea horses also demonstrate that decreased muscle growth may be a consequence of heat exposure. Changes in the abundance of transcripts with metabolic function– including both lipids and carbohydrates- were also measured in reef fish exposed to temperature stress for 28 days [62]. A recent study compared the proteomes of fish exposed to elevated temperatures for 28 days [63]. Because of the difference in kinetics between the transcriptome and proteome, early time points were not sampled. They also observed a distinct proteome at different temperatures, with proteins involved in chaperoning, energy and metabolism, cytoskeleton and inflammatory responses induced [63], similar functional groups to those that were modulated by heat stress in this study. Although induction of transcripts for lactate dehydrogenase is a commonly used marker for thermal stress, we only found it induced in the gill at the 6 h time point, which is consistent with some previous studies [24]. Again, this may be due to the fact that the water was oxygenated, decreasing the likelihood of anaerobic metabolism.

Differential exon usage (DEU) in thermally sensitive proteins

Alternative splicing generates multiple transcripts from a single gene, and thus increases the diversity of proteins and phenotypes available to a single genotype [34]. Although the process can be non-functional, alternative splicing has been shown to result in differences in the proteome [34]. Temperature changes lead to changes in protein structure by altering the hydrogen bonds and ionic interactions and hydrophobic interactions that give these molecules their shape [15]. Recent reviews indicate that alternative splicing has been shown to be temperature responsive (reviewed by [34]). Ectothermic animals, including fishes, would be especially susceptible to temperature induced changes in protein structure, and consequently, function. Micromolecular adaptations– potentially including the changes in exon usage we measured in this study– have the potential to allow for slight confirmational changes to molecules without changing function [55], and may be important in allowing organisms to inhabit variable environments or to migrate across the thermocline.

In the muscle, we measured DEU in transcripts for metabolic pathways, particularly gluconeogenesis, and in motor pathways, at the 24 and 72 h pathways. Some of the cytoskeletal proteins in which DEU was measured (such as actin, tubulin, myogenic basic muscle-specific protein and tropomyosin) have been identified in previous studies as being particularly sensitive to thermal stress [63]. For instance, actin had increased abundance in the proteome of thermally-stressed Sea Bream (Sparus aurata) [64]. The cytoskeleton has been shown in other taxa to be important in conferring phenotypic plasticity under different environmental conditions [65]. Changes to energetic pathways are an important component of the cellular stress response (reviewed in [15, 56]). We could not relate the differentially exon usage to different Open Reading Frames due to the design of the SuperTranscript algorithm [51]. When reads were aligned to the coding sequence of a closely related species, there were far fewer transcripts with differential exon usage (e.g. Table S6). While this difference could relate to the quality of the different assemblies, we instead hypothesise that many of the changes in exon utilisation relate to non-coding portions of the transcript. The differential exon usage may not change the structure of the protein product but may instead modulate transcriptional efficiency or transcript stability [59]. The much lower rates of mapping to the genome than our assembled transcriptome support our hypothesis of differential utilisation of non-coding portions of the genome. Although we do not know how the changes in exon usage change protein structure and function, physiological pathways, and ultimately, resilience, we would hypothesise that these changes could increase the fish’s performance under high temperature or decrease the susceptibility to oxygen limitation. Similarly, we would hypothesise that the DEU we measured with respect to oxygen binding in the gill (Supplemental Figure S5) could confer a functional benefit in high temperature, low oxygen waters. Other temperature induced structural changes to RNA (that were mediated via RNA editing, not splice variation) were confirmed at the protein level [66]. Future studies may want to examine differential exon usage as a mechanism conferring phenotypic plasticity under heat stress using long-read technology, which can be more readily related to the final protein product [67].

Other heat induced changes

Heat stress is known to alter membrane permeability [15] which could explain the DEU measured in the junction proteins in the gill. There could be alternative explanations, aside from acclimation to temperature, for some of the changes we measured. Previous studies have measured changes in cortisol levels and osmoregulation in heat stressed fish [54, 68]. Even fish that are seemingly tolerant of temperature changes had variable ion concentrations in blood [69]. Hypothetically, altered ionic regulation could be driving the alterations in membrane transporters we measured in the gill (Fig. 5).

The study species chosen, yellowtail kingfish (Seriola lalandi), has a global distribution in coastal tropical and subtropical environments [31]. Although it is an established aquaculture species in Japan, farming is being established elsewhere in the world, including Australia [32]. Its wide thermal tolerance makes it an attractive species for aquaculture, especially given the pressures of climate change. Previous studies have found that each population has specific physiological changes that suit its environment. For instance, studies conducted in Mexico and acclimated to 25 degrees as juveniles [30, 31], found an optimal temperature for growth at 26 degrees. Previous studies conducted with this stock of fish found a lower optimal temperature 22.8 degrees [32]. Previous studies conducted with fish reared in NSW, Australia also found a lower optimal temperature of 20 degrees [33]. As a consequence, we hypothesise that the sudden change from 20 to 27 degrees would cause thermal stress, especially since these fish had been derived from individuals acclimated to temperatures between 18 and 22, but also that the fish would acclimate to this stress. Previous environmental exposures to higher temperatures, at or above 27 degrees, in sea cage culture in Western Australia caused mortality (G. Partridge, Harvest Road Oceans, Pers. Comms), therefore this temperature was selected as thermal stress point.

Even though the changes we measured suggest sublethal stress, no changes in growth were measured in fish maintained over the full 30 day study, heat stress could still compromise the fitness of the individuals. The complex and non-linear nature of physiological responses to stressors such as heat makes vulnerability difficult to predict [60]. None the less, consistent with physiological pathways we have measured in the altered transcriptomic profiles we describe above, previous studies have found depletion of cellular energy reserves in heat exposed animals [60]. Previous studies have also measured longer recovery times and higher basal metabolic rates in high temperature acclimated fish [54]. Non-essential physiological functions can be suppressed during thermal stress events [29]. Elevated temperatures cause decreased or negative growth rates, decreased aerobic scope, decreased feeding and swimming, and altered respiration at elevated temperatures [68]. It has been suggested that the energetic depletion associated with temperature stress leads to capacity impacts, whereby altered patterns of migration, reduced growth and reproductive output, and increased susceptibility to disease result from reallocation of metabolic resources [16]. There is also the possibility that sublethal temperature stress alters the epigenome, leading to thermal tolerance in fish [70], although the evidence for this is strongest in larval fish.

Plasticity and the capacity to acclimate to temperature stress

Previous studies have indicated that reduced plasticity in the transcriptome and lower induction of HSPs following thermal stress suggests a reduced capacity for acclimation [62]. Fish with little transcriptomic response to heat stress are thought to be at the edge of their thermal tolerance [62] or to have comparatively little capacity to make physiological changes to adjust to climate stressors [23]. Previous studies have hypothesised that genes with differential transcript abundance or differential exon usage are important in conferring resilience following exposure to a sublethal stressor, including heat stress [57, 71]. An organisms ability to modulate its transcriptome and proteome may determine its ability to withstand temperature stress [56], and as a consequence, may confer its ability to survive in a changing world.

Here, we found that heat stress led not only to differences in transcript abundance, hypothetically indicating differential gene expression, but also DEU. Along with differences in gene expression, alternative splicing (measured as differential exon usage) confers regulatory variation in the genotype [59]. Some studies suggest that alternative splicing diverges more quickly in populations than differences in gene expression [59]. Recent reviews have suggested that alternative splicing may have the potential to facilitate rapid change during evolution [34]. Although we do not know if changes in exon usage change protein structure and alter physiological pathways, or alternatively alters the transcriptional efficiency of existing proteins, we hypothesise that this ability to even subtly alter proteins is important in maintaining the function of these proteins in a variable environment (e.g [55, 72]). and in conferring plasticity. We propose this ability to not only make different proteins under stress, but to make proteins differently, as potentially important, if not well understood, facet of phenotypic plasticity. This facet of the transcriptome should be included in more studies, especially those that are concerned with mechanisms of resilience. In future predictions of the potential of organisms to change under climate pressures, DEU should be measured along with DTA to get a more wholistic view of the organisms potential for plasticity. This may be most efficiently pursued using long-read technology to better identify the functional significance of differential exon usage (e.g [67, 73]. Future studies should also confirm whether the differential exon usage is translated to differences in protein structure and function.

Conclusions

There would be numerous ecophysiological consequences for fish (and presumably, other organisms) experiencing repeated exposures to temperature stress, including lower growth rates, and decreased energy reserves, that would impact the fitness of the individual/ population, especially in the face of multiple stressors (reviewed by [74]). Our results also show how the transcriptome, measured as both differences in abundance and differences in exon usage, can be used to predict species tolerance to thermal stress. Previous studies have shown that organisms with reduced thermal tolerance have fewer transcriptomic responses in response to a thermal challenge [23], suggesting that they have less of a capacity for phenotypic buffering and energetic compensation. Our study was performed using sublethal temperature stress– i.e. a temperature that did not cause mortality or growth restriction (Table S1) in the yellowtail kingfish. By examining both the initial response to the heat stress and the subsequent compensatory mechanisms, including differential exon usage, we have a better understanding of the transcriptomic response to heat stress. We need to better understand the mechanisms underlying environmental stress tolerance so that we can predict species responses to climate change [24, 56]. Studies of the plasticity of populations to climate variables can help us predict the dynamics of future populations [74]. Understanding these mechanisms will be important from and aquaculture perspective to ensure that we are selecting strains that are likely to be able to survive the coming decades [57]. It will also be important that any strains used for assisted evolution [75] are likely to be resilient to variable environments.

Data availability

Reads were uploaded to NCBI’s short read archive and can be accessed using BioProject number PRJNA1086548.

References

  1. IPCC, Summary for P, Zhai P, Pirani A, Connors SL, Péan C, Berger S, Caud N, Chen Y, Goldfarb L, Gomis MI, Huang M, Leitzell K, Lonnoy E, Matthews JBR, Maycock TK, Waterfield T, Yelekçi O, Yu R. B. Zhou editor. The Physical Science Basis Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change Cambridge University.; 2021.

  2. Pinsky ML, Selden RL, Kitchel ZJ. Climate-Driven shifts in Marine species ranges: scaling from Organisms to communities. Ann Rev Mar Sci. 2020;12:153–79.

    Article  PubMed  Google Scholar 

  3. Sunday JM, Bates AE, Dulvy NK. Thermal tolerance and the global redistribution of animals. Nat Clim Change. 2012;2(9):686–90.

    Article  Google Scholar 

  4. Smale DA, Wernberg T, Oliver ECJ, Thomsen M, Harvey BP, Straub SC, et al. Marine heatwaves threaten global biodiversity and the provision of ecosystem services. Nat Clim Change. 2019;9(4):306–.

    Article  Google Scholar 

  5. Oliver ECJ, Lago V, Hobday AJ, Holbrook NJ, Ling SD, Mundy CN. Marine heatwaves off eastern Tasmania: Trends, interannual variability, and predictability. Prog Oceanogr. 2018;161:116–30.

    Article  Google Scholar 

  6. Holbrook NJ, Sen Gupta A, Oliver ECJ, Hobday AJ, Benthuysen JA, Scannell HA, et al. Keeping pace with marine heatwaves. Nat Reviews Earth Environ. 2020;1(9):482–93.

    Article  Google Scholar 

  7. Pecl GT, Araújo MB, Bell JD, Blanchard J, Bonebrake TC, Chen I-C et al. Biodiversity redistribution under climate change: impacts on ecosystems and human well-being. 2017;355(6332):eaai9214.

  8. Sanford E, Sones JL, Garcia-Reyes M, Goddard JHR, Largier JL. Widespread shifts in the coastal biota of northern California during the 2014–2016 marine heatwaves. Sci Rep. 2019;9(1):4216.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Edgar GJ, Stuart-Smith RD, Heather FJ, Barrett NS, Turak E, Sweatman H et al. Continent-wide declines in shallow reef life over a decade of ocean warming. Nature. 2023.

  10. Doubleday ZA, Clarke SM, Li X, Pecl GT, Ward TM, Battaglene S, et al. Assessing the risk of climate change to aquaculture: a case study from south-east Australia. Aquaculture Environ Interact. 2013;3(2):163–75.

    Article  Google Scholar 

  11. Hobday AJ, Poloczanska ES, Matear R. Implications of Climate Change for Australian Fisheries and Aquaculture: A preliminary assessment. available from http://www.cmar.csiro.au/climateimpacts/reports.htm; 2008.

  12. Ekstrom A, Sandblom E, Blier PU, Dupont Cyr BA, Brijs J, Pichaud N. Thermal sensitivity and phenotypic plasticity of cardiac mitochondrial metabolism in European perch, Perca fluviatilis. J Exp Biol. 2017;220(Pt 3):386–96.

    PubMed  Google Scholar 

  13. Portner HO. Climate variations and the physiological basis of temperature dependent biogeography: systemic to molecular hierarchy of thermal tolerance in animals. Comp Biochem Physiol a-Molecular Integr Physiol. 2002;132(4):739–61.

    Article  CAS  Google Scholar 

  14. Schmidtko S, Stramma L, Visbeck M. Decline in global oceanic oxygen content during the past five decades. Nature. 2017;542(7641):335–9.

    Article  PubMed  CAS  Google Scholar 

  15. Somero GN. The cellular stress response and temperature: function, regulation, and evolution. J Experimental Zool Part a-Ecological Integr Physiol. 2020;333(6):379–97.

    Article  CAS  Google Scholar 

  16. Akbarzadeh A, Gunther OP, Houde AL, Li S, Ming TJ, Jeffries KM, et al. Developing specific molecular biomarkers for thermal stress in salmonids. BMC Genomics. 2018;19(1):749.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  17. Donelson JM, Gaitan-Espitia JD, Hobday AJ, Mokany K, Andrew SC, Boulter S, et al. Putting plasticity into practice for effective conservation actions under climate change. Nat Clim Change. 2023;13(7):632–47.

    Article  Google Scholar 

  18. Schlichting CD, Smith H. Phenotypic plasticity: linking molecular mechanisms with evolutionary outcomes. Evol Ecol. 2002;16:189–211.

    Article  Google Scholar 

  19. Hendry AP. Key questions on the role of phenotypic plasticity in Eco-evolutionary Dynamics. J Hered. 2016;107(1):25–41.

    Article  PubMed  Google Scholar 

  20. Logan CA, Somero GN. Effects of thermal acclimation on transcriptional responses to acute heat stress in the eurythermal fish Gillichthys mirabilis (Cooper). Am J Physiol Regul Integr Comp Physiol. 2011;300(6):R1373–83.

    Article  PubMed  CAS  Google Scholar 

  21. Bugg WS, Yoon GR, Schoen AN, Laluk A, Brandt C, Anderson WG, et al. Effects of acclimation temperature on the thermal physiology in two geographically distinct populations of lake sturgeon (Acipenser fulvescens). Conserv Physiol. 2020;8(1):coaa087.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Sandoval-Castillo J, Gates K, Brauer CJ, Smith S, Bernatchez L, Beheregaray LB. Adaptation of plasticity to projected maximum temperatures and across climatically defined bioregions. Proc Natl Acad Sci U S A. 2020;117(29):17112–21.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  23. Komoroske LM, Jeffries KM, Whitehead A, Roach JL, Britton M, Connon RE, et al. Transcriptional flexibility during thermal challenge corresponds with expanded thermal tolerance in an invasive compared to native fish. Evol Appl. 2020;14(4):931–49.

    Article  Google Scholar 

  24. Komoroske LM, Connon RE, Jeffries KM, Fangue NA. Linking transcriptional responses to organismal tolerance reveals mechanisms of thermal sensitivity in a mesothermal endangered fish. Mol Ecol. 2015;24(19):4960–81.

    Article  PubMed  CAS  Google Scholar 

  25. Iwama GK, Vijayan MM, Forsyth RB, Ackerman PA. Heat shock proteins and physiological stress in fish. Am Zool. 1999;39(6):901–9.

    Article  CAS  Google Scholar 

  26. Schaefer J, Ryan A. Developmental plasticity in the thermal tolerance of zebrafish Danio rerio. J Fish Biol. 2006;69(3):722–34.

    Article  Google Scholar 

  27. Liu B, Qian S-B. Translational reprogramming in cellular stress response. 2014;5(3):301–5.

  28. Morley SA, Peck LS, Sunday JM, Heiser S, Bates AE, Algar A. Physiological acclimation and persistence of ectothermic species under extreme heat events. Glob Ecol Biogeogr. 2019;28(7):1018–37.

    Article  Google Scholar 

  29. Jeffries KM, Fangue NA, Connon RE. Multiple sub-lethal thresholds for cellular responses to thermal stressors in an estuarine fish. Comp Biochem Physiol Mol Integr Physiol. 2018;225:33–45.

    Article  CAS  Google Scholar 

  30. Larios-Soriano E, Re-Araujo AD, Diaz F, Lopez-Galindo LL, Rosas C, Ibarra-Castro L. Effects of recent thermal history on thermal behaviour, thermal tolerance and oxygen uptake of Yellowtail Kingfish (Seriola lalandi) juveniles. J Therm Biol. 2021;99:103023.

    Article  PubMed  CAS  Google Scholar 

  31. Abbink W, Blanco Garcia A, Roques JAC, Partridge GJ, Kloet K, Schneider O. The effect of temperature and pH on the growth and physiological response of juvenile yellowtail kingfish Seriola lalandi in recirculating aquaculture systems. Aquaculture. 2012;330–333:130–5.

    Article  Google Scholar 

  32. Donohue CG, Partridge GJ, Sequeira AMM. Bioenergetic growth model for the yellowtail kingfish (Seriola lalandi). Aquaculture. 2021;531.

  33. Pirozzi I, Booth MA. The routine metabolic rate of mulloway (Argyrosomus japonicus: Sciaenidae) and yellowtail kingfish (Seriola lalandi: Carangidae) acclimated to six different temperatures. Comp Biochem Physiol Mol Integr Physiol. 2009;152(4):586–92.

    Article  Google Scholar 

  34. Wright CJ, Smith CWJ, Jiggins CD. Alternative splicing as a source of phenotypic diversity. Nat Rev Genet. 2022;23(11):697–710.

    Article  PubMed  CAS  Google Scholar 

  35. Team RC. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria2022 [Available from: https://www.R-project.org/

  36. Hook SE, Mondon J, Revill AT, Greenfield PA, Stephenson SA, Strzelecki J, et al. Monitoring sublethal changes in fish physiology following exposure to a light, unweathered crude oil. Aquat Toxicol. 2018;204:27–45.

    Article  PubMed  CAS  Google Scholar 

  37. Hook SE, Mondon J, Revill AT, Greenfield PA, Smith RA, Turner RDR, et al. Transcriptomic, lipid, and histological profiles suggest changes in health in fish from a pesticide hot spot. Mar Environ Res. 2018;140:299–321.

    Article  PubMed  CAS  Google Scholar 

  38. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30(15):2114–20.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  39. Haas BJ, Papanicolaou A, Yassour M, Grabherr M, Blood PD, Bowden J, et al. De novo transcript sequence reconstruction from RNA-seq using the Trinity platform for reference generation and analysis. Nat Protoc. 2013;8(8):1494–512.

    Article  PubMed  CAS  Google Scholar 

  40. Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, et al. Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat Biotechnol. 2011;29(7):644–U130.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  41. Simão FA, Waterhouse RM, Ioannidis P, Kriventseva EV, Zdobnov EM. BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics. 2015;31(19):3210–2.

    Article  PubMed  Google Scholar 

  42. Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C. Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods. 2017;14(4):417–.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  43. Buchfink B, Xie C, Huson DH. Fast and sensitive protein alignment using DIAMOND. Nat Methods. 2015;12(1):59–60.

    Article  PubMed  CAS  Google Scholar 

  44. Love MI, Huber W, Anders S. Moderated estimation of Fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550.

    Article  PubMed  PubMed Central  Google Scholar 

  45. Huber W, Carey VJ, Gentleman R, Anders S, Carlson M, Carvalho BS, et al. Orchestrating high-throughput genomic analysis with Bioconductor. Nat Methods. 2015;12(2):115–21.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  46. Kolde R, _pheatmap. Pretty Heatmaps_. R package version 1.0.12 2019 [Available from: https://CRAN.R-project.org/package=pheatmap

  47. Oliveros JC. VENNY. An interactive tool for comparing lists with Venn Diagrams. 2007.

  48. Huang DW, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc. 2009;4(1):44–57.

    Article  PubMed  CAS  Google Scholar 

  49. Sherman BT, Hao M, Qiu J, Jiao X, Baseler MW, Lane HC et al. DAVID: a web server for functional enrichment analysis and functional annotation of gene lists (2021 update). (1362–4962 (Electronic)).

  50. Huang daW, Sherman Bt Fau -, Lempicki RA, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. 2022(1750–2799 (Electronic)).

  51. Davidson NM, Hawkins ADK, Oshlack A. SuperTranscripts: a data driven reference for analysis and visualisation of transcriptomes. Genome Biol. 2017;18(1):148.

    Article  PubMed  PubMed Central  Google Scholar 

  52. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29(1):15–21.

    Article  PubMed  CAS  Google Scholar 

  53. Anders S, Reyes A, Huber W. Detecting differential usage of exons from RNA-seq data. Genome Res. 2012;22(10):2008–17.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  54. Mackey TE, Hasler CT, Durhack T, Jeffrey JD, Macnaughton CJ, Ta K et al. Molecular and physiological responses predict acclimation limits in juvenile brook trout (Salvelinus fontinalis). J Exp Biol. 2021;224(16).

  55. Somero GN. The Goldilocks Principle: a unifying perspective on biochemical adaptation to Abiotic stressors in the Sea. Ann Rev Mar Sci. 2022;14:1–23.

    Article  PubMed  Google Scholar 

  56. Dong YW, Liao ML, Han GD, Somero GN. An integrated, multi-level analysis of thermal effects on intertidal molluscs for understanding species distribution patterns. Biol Rev Camb Philos Soc. 2022;97(2):554–81.

    Article  PubMed  Google Scholar 

  57. Thorstensen MJ, Turko AJ, Heath DD, Jeffries KM, Pitcher TE. Acute thermal stress elicits interactions between gene expression and alternative splicing in a fish of conservation concern. J Exp Biol. 2022;225(12).

  58. Tan S, Wang W, Tian C, Niu D, Zhou T, Jin Y, et al. Heat stress induced alternative splicing in catfish as determined by transcriptome analysis. Comp Biochem Physiol Part D Genomics Proteom. 2019;29:166–72.

    Article  CAS  Google Scholar 

  59. Singh P, Borger C, More H, Sturmbauer C. The role of alternative splicing and Differential Gene expression in Cichlid Adaptive Radiation. Genome Biol Evol. 2017;9(10):2764–81.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  60. Madeira C, Mendonca V, Leal MC, Flores AAV, Cabral HN, Diniz MS, et al. Environmental health assessment of warming coastal ecosystems in the tropics - application of integrative physiological indices. Sci Total Environ. 2018;643:28–39.

    Article  PubMed  CAS  Google Scholar 

  61. Madeira D, Mendonça V, Dias M, Roma J, Costa PM, Diniz MS, et al. Physiological and biochemical thermal stress response of the intertidal rock goby Gobius paganellus. Ecol Ind. 2014;46:232–9.

    Article  CAS  Google Scholar 

  62. Veilleux HD, Ryu T, Donelson JM, Ravasi T, Munday PL. Molecular response to Extreme Summer temperatures differs between two genetically differentiated populations of a coral reef fish. Front Mar Sci. 2018;5.

  63. Madeira D, Araujo JE, Vitorino R, Costa PM, Capelo JL, Vinagre C, et al. Molecular plasticity under Ocean Warming: proteomics and Fitness Data provides clues for a better understanding of the Thermal Tolerance in Fish. Front Physiol. 2017;8:825.

    Article  PubMed  PubMed Central  Google Scholar 

  64. Madeira D, Araújo JE, Vitorino R, Capelo JL, Vinagre C, Diniz MS. Ocean warming alters cellular metabolism and induces mortality in fish early life stages: a proteomic approach. Environ Res. 2016;148:164–76.

    Article  PubMed  CAS  Google Scholar 

  65. Popovic I, Riginos C. Comparative genomics reveals divergent thermal selection in warm- and cold-tolerant marine mussels. Mol Ecol. 2020;29(3):519–35.

    Article  PubMed  Google Scholar 

  66. Birk MA, Liscovitch-Brauer N, Dominguez MJ, McNeme S, Yue Y, Hoff JD, et al. Temperature-dependent RNA editing in octopus extensively recodes the neural proteome. Cell. 2023;186(12):2544–e5513.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  67. Gehring NH, Roignant JY. Anything but ordinary - emerging splicing mechanisms in eukaryotic gene regulation. Trends Genet. 2021;37(4):355–72.

    Article  PubMed  CAS  Google Scholar 

  68. Chadwick JG, McCormick SD. Upper thermal limits of growth in brook trout and their relationship to stress physiology. J Exp Biol. 2017;220(Pt 21):3976–87.

    Article  PubMed  Google Scholar 

  69. Zhang Y, Loughery JR, Martyniuk CJ, Kieffer JD. Physiological and molecular responses of juvenile shortnose sturgeon (Acipenser brevirostrum) to thermal stress. Comp Biochem Physiol Mol Integr Physiol. 2017;203:314–21.

    Article  CAS  Google Scholar 

  70. Metzger DCH, Schulte PM. Persistent and plastic effects of temperature on DNA methylation across the genome of threespine stickleback (Gasterosteus aculeatus). Proc Biol Sci. 2017;284:1864.

    Google Scholar 

  71. Thorstensen MJ, Baerwald MR, Jeffries KM. RNA sequencing describes both population structure and plasticity-selection dynamics in a non-model fish. BMC Genomics. 2021;22(1):273.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  72. Fields PA, Dong Y, Meng X, Somero GN. Adaptations of protein structure and function to temperature: there is more than one way to ‘skin a cat’. J Exp Biol. 2015;218(Pt 12):1801–11.

    Article  PubMed  Google Scholar 

  73. Fernandes JCR, Acuna SM, Aoki JI, Floeter-Winter LM, Muxel SM. Long non-coding RNAs in the regulation of Gene expression: physiology and disease. Noncoding RNA. 2019;5(1).

  74. Palumbi SR, Evans TG, Pespeni MH, Somero GN. Present and future adaptation of marine species assemblages: DNA-based insights into climate change from studies of physiology, genomics, and evolution. Oceanography. 2019;32:82–93.

    Article  Google Scholar 

  75. van Oppen MJ, Oliver JK, Putnam HM, Gates RD. Building coral reef resilience through assisted evolution. Proc Natl Acad Sci U S A. 2015;112(8):2307–13.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

Ondrej Hlinka (CSIRO IM&T) provided valuable assistance with the bioinformatics. Tom Walsh (CSIRO Environment) provided valuable discussion. This manuscript was greatly improved by constructive comments made by two anonymous reviewers.

Funding

Funding for this research was provided by CSIRO’s Interchange project, the Oceans and Atmosphere’s Business Unit, and the Environment Business Unit.

Author information

Authors and Affiliations

  1. CSIRO Environment, Hobart, TAS, Australia

    Sharon E. Hook & Alistair J. Hobday

  2. CSIRO Health and Biosecurity, Geelong, VIC, Australia

    Ryan J. Farr & Jenny Su

  3. DPIRD Marine Fish, Aquaculture Research and Development, Fremantle, WA, Australia

    Catherine Wingate, Lindsey Woolley & Luke Pilmer

  4. School of Molecular Sciences, University of Western Australia, Perth, WA, Australia

    Catherine Wingate

Authors
  1. Sharon E. Hook
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Ryan J. Farr
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Jenny Su
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Alistair J. Hobday
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Catherine Wingate
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Lindsey Woolley
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Luke Pilmer
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

S.E.H.– Conceptualization; Formal analysis; Funding acquisition; Writing– original draft; Writing– review & editing. R. J. F.– Investigation; Methodology; Writing– review & editing. J. S. - Investigation; Writing– review & editing. A. J. H.– Conceptualization; Funding acquisition; Writing– review & editing. C. W.e– Resources; Formal analysis; L. W.– Resources, Conceptualization, Writing– review & editing. L. P.– Resources, Writing– review & editing.

Corresponding author

Correspondence to Sharon E. Hook.

Ethics declarations

Ethics approval and consent to participate

The study was approved by the Ethics committee at Western Australia’s Department of Primary Industries and Regional Development (permit number: AEC 22-2-06).

Consent for publication

Not Applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Supplementary Material 3

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Hook, S.E., Farr, R., Su, J. et al. Transcriptional profiles reveal physiological mechanisms for compensation during a simulated marine heatwave in Yellowtail Kingfish (Seriola lalandi). BMC Genomics 26, 230 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12864-025-11283-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12864-025-11283-9

Keywords

BMC Genomics

ISSN: 1471-2164

Contact us