ABSTRACT
Chronic exposure to high temperatures may leave freshwater fishes vulnerable to opportunistic pathogens, particularly during early life stages. Lake sturgeon, Acipenser fulvescens, populations within the northern expanse of their range in Manitoba, Canada, may be susceptible to high temperature stress and pathogenic infection. We acclimated developing lake sturgeon for 22 days to two ecologically relevant, summer temperatures (16 and 20°C). Individuals from both acclimation treatments were then exposed to 0, 30 and 60 µg ml−1 bacterial lipopolysaccharides (endotoxins), as an immune stimulus, for 48 h and sampled 4 and 48 h during trial exposures and following a 7 day recovery period. We then measured whole-body transcriptional (mRNA) responses involved in the innate immune, stress and fatty acid responses following acute exposure to the bacterial endotoxins. Data revealed that overall levels of mRNA transcript abundance were higher in 20°C-reared sturgeon under control conditions. However, following exposure to a bacterial stimulus, lake sturgeon acclimated to 16°C produced a more robust and persistent transcriptional response with higher mRNA transcript abundance across innate immune, stress and fatty acid responses than their 20°C-acclimated counterparts. Additional whole-animal performance metrics (critical thermal maximum, metabolic rate, cortisol concentration and whole-body and mucosal lysozyme activity) demonstrated acclimation-specific responses, indicating compromised metabolic, stress and enzymatic capacity following the initiation of immune-related responses. Our study showed that acclimation to 20°C during early development impaired the immune capacity of developing lake sturgeon as well as the activation of molecular pathways involved in the immune, stress and fatty acid responses. The present study highlights the effects of ecologically relevant, chronic thermal stress on seasonal pathogen susceptibility in this endangered species.
INTRODUCTION
Globally, temperature change, flow alteration and extreme weather events have led to decreased productivity and imperiled fish populations in freshwater systems (Milly et al., 2008; Reid et al., 2019; Dudgeon, 2019). In response to environmental changes, fishes may show plasticity to alterations in their thermal environment (Earhart et al., 2022); however, this phenotypic plasticity may potentially decrease their ability to compensate during other physiological challenges. Additional stressors such as the effects of pathogenic infection may shape developmental and evolutionary trajectories for population and species dynamics as phenotypes change and mortality occurs (Dittmar et al., 2014; Schade et al., 2014; Strowbridge et al., 2021). These environmental changes are often accompanied by alterations in pathogen populations, abundance and virulence, which could leave freshwater species more vulnerable to the compounding stressors of temperature and disease (Marcos-López et al., 2010; Paull and Johnson, 2011; Dittmar et al., 2014; Miller et al., 2014). However, until recently, little research has focused on the effects of these combined stressors under laboratory conditions, especially in species of conservation concern (Bugg et al., 2021a). As host–pathogen–environment interactions may determine survival outcomes for fish in wild environments (Jeffries et al., 2014; Teffer et al., 2022), it is important to study the interactions of these stressors in fishes of conservation concern.
Sturgeons are members of the most critically endangered group of animals on the planet (IUCN, 2022) and are often reared in hatcheries for subsequent release to enhance wild populations. Recent research has indicated that in both wild and hatchery environments, sturgeons are susceptible to a variety of fungal, viral and bacterial pathogens (Coleman et al., 2018; Fujimoto et al., 2018; Mugetti et al., 2020; Brocca et al., 2022; Costábile et al., 2022; Stilwell et al., 2022; Soto et al., 2022) with the potential for thermal stress to compromise immune capacity, as revealed by full transcriptome studies (Penny et al., 2023; Bugg et al., 2023 preprint). This may be exacerbated among populations in the northern range for these species, such as lake sturgeon (Acipenser fulvescens) in Manitoba, Canada, as they may be more susceptible to the effects of increasing temperatures and inhabit areas where temperatures are projected to increase by 2.1–3.4°C by 2050 (Manitoba Hydro, 2015). Currently, temperatures rise above 20°C during the summer (May to October) for many rivers throughout Manitoba, a critical period for early development in lake sturgeon (Bugg et al., 2020, 2021b).
Phenotypic development during early life history is a critical period involving a remarkable shift of biological function, organ development, narrowing energy reserves and often high levels of mortality (Sifa and Mathias, 1987; Wieser, 1991; Rombough, 1994). During this period, fish (and particularly sturgeons) must cope with environmental conditions due to limited swimming ability (Kopf et al., 2014; Brandt et al., 2021), making them particularly vulnerable to environmental perturbations, especially changes in temperature. Additionally, at hatch and during early development, many fishes have a limited adaptive immune response, leaving them reliant on their innate immune systems until later life stages (Chantanachookhin et al., 1991; Petrie-Hanson and Ainsworth, 2001; Magnadottir et al., 2006; Reyes-López et al., 2018). Thus, with limited mobility and immune capacity, exposure to thermal stress may compromise the defenses of developing larval freshwater species and leave them susceptible to the effects of opportunistic viral, bacterial and fungal pathogens, which are pervasive throughout freshwater ecosystems. Sturgeons are likely even more susceptible to opportunistic pathogenic infections in early development as their adaptive immune development is slower than that of many other fish species (Gradil et al., 2014a,b). While previous research has demonstrated that prolonged periods at temperatures of 20°C and above can be thermally stressful and have negative physiological consequences for developing lake sturgeon (Bugg et al., 2020; Bugg et al., 2023 preprint), there has been little evaluation of the immune capabilities of this species in early development or the immune capacity of sturgeons under thermal stress (Bugg et al., 2021a). Ultimately, in this critical period of early development, the ability to functionally sustain organismal responses against multiple stressors likely determines survival (Dittmar et al., 2014), and may influence population-level outcomes for northern sturgeon.
Innate immunity relies on two specific mechanisms to respond to pathogens: the detection of the immune stimulus (pathogen detection), resulting in an intracellular immune-stimulating cascade, and the subsequent transcriptional initiation, activating the production of immune-related compounds (immune response). The ability of the innate immune system to detect pathogens largely relies on pattern recognition receptors, such as toll-like receptor 4 (TLR4), which detects gram-negative bacteria by their outer lipopolysaccharide (LPS) structure and other associated peptidoglycans (Magnadottir, 2006; Amarante-Mendes et al., 2018). Once activated, TLR4 can respond through two different activation pathways: myeloid differentiation primary response 88 (MyD88)-dependent activation or toll-like receptor adaptor molecule 1 (TICAM-1) signaling, both of which induce the transcription factor nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) to translocate into the nucleus and initiate the transcription of mRNAs coding for pro-inflammatory cytokines and other immune compounds (Fig. 1A; Vaure and Liu, 2014; Srivastava et al., 2017; Amarante-Mendes et al., 2018). Immune-responsive proteins such as cytokines (TNFα, IL-1β and IL-8, involved in immune signaling, the inflammation response and macrophage recruitment; Turnbull and Rivier, 1999), antibodies (IgM; Lobo, 2016; Yu et al., 2020), complement activators (C3; Holland and Lambris, 2002) and enzymes (lysozyme; Saurabh and Sahoo, 2008) initiate a variety of immunomodulatory activities to suppress and destroy bacteria both inside and outside the body, but can also have detrimental impacts on survival if overexpressed (Anderberg et al., 2021). Additionally, the activation of innate immune responses can further induce changes in long-chain fatty acids (Fig. 1B) either through mitochondrial β-oxidation, limited by the mitochondrial transporter carnitine palmitoyltransferase I (CPT1; Coccia et al., 2014; Norambuena et al., 2015), or through the formation of immune precursors via cytokine-induced phospholipase A2 (PLA2)-mediated cleavage (Okamura et al., 2021; Nguyen et al., 2022). However, many of these innate immune mechanisms may be ultimately compromised by the effects of chronic thermal stress, associated allostatic load (i.e. physiological stress caused by abiotic or biotic environmental factors; Samaras et al., 2018), and the energetic costs related to elevation of the stress response (Schreck, 2010; Schreck and Tort, 2016).
Under environmental conditions with limited allostatic load, infection, triggering of the immune response and the detection of circulating cytokines by the hypothalamus can stimulate the hypothalamic-pituitary–interrenal (HPI) axis with an increased release of adrenocorticotropin (ACTH) from the pituitary, which ultimately increases the production of cortisol (Shintani et al., 1995; Turnbull and Rivier, 1999; Fig. 1C). This synthesis of cortisol is rate limited by steroidogenic acute regulatory protein (StAR; Stocco et al., 2005). Once cortisol enters the circulation and then cells, cytosolic glucocorticoid receptors (GRs) can then bind to it with the help of molecular chaperones and members of the general stress response (HSP70 and HSP90) (Hutchinson et al., 1994; Bamberger et al., 1996; Bekhbat et al., 2017). This complex acts as a transcription factor, transcriptionally regulating further cellular signaling mechanisms and physiological responses critical to the maintenance of homeostasis and organismal survival (Marchi and van Eeden, 2021). However, if stressful conditions are present before pathogen exposure, sub-lethal thermal thresholds may be exceeded, increasing GRs and cortisol levels. In these situations, increased GRs and cortisol levels may interfere with immune-stimulating transcription factors such as NF-κB (Bekhbat et al., 2017; Jeffries et al., 2018). Ultimately, this may result in the inactivation of the cytokine response, immunosuppression and subsequent infection by opportunistic pathogens (Tort, 2011; Alfonso et al., 2021; Aversa-Marnai et al., 2022), to which fish at early life stages are particularly vulnerable.
The goal of this study was to investigate the innate immune capacity of developing lake sturgeon when exposed to bacterial LPS under thermal stress. Using a transcriptional profiling approach (Jeffrey et al., 2020), we targeted genes involved in the innate immune, stress and fatty acid pathways responsive to bacterial infection. This strategy was then paired with metrics to assess the immune responses and physiology of developing lake sturgeon exposed to bacterial LPS following acclimation to ecologically relevant temperatures of 16 and 20°C. We tested the hypothesis that chronic thermal stress will impact energetic balance, transcriptional plasticity and whole-animal physiology, and that it will specifically alter the activation of the innate immune system and related processes. Thus, we predicted that a heightened stress response associated with acclimation to increased temperatures would reduce energetic stores and elevate glucocorticoid (cortisol) and receptor levels (GR1), impairing the transcriptional activation of the innate immune system and downstream processes following exposure to bacterial LPS. We also predicted that the immune responses of 16°C-acclimated sturgeon would involve a greater transcriptional response than their 20°C-acclimated counterparts, following exposure to the bacterial stimuli. Finally, we predicted that these compounding impacts from both elevated temperature and immune stimulus would extend to measurements of lysozyme activity and whole-organism physiology [hepatosomatic index (HSI), critical thermal maximum (CTmax) and metabolic rates].
MATERIALS AND METHODS
Lake sturgeon husbandry
In May 2021, gametes from wild-caught female and male lake sturgeon (Acipenser fulvescens Rafinesque 1817) were harvested from individuals at the Pointe du Bois Generating Station on Winnipeg River, MB, Canada (50°17′52″N, 95°32′51″W). Once collected, eggs and sperm were transported to the University of Manitoba animal holding facility. Immediately upon arrival, fertilization was carried out using the eggs from three females and the diluted sperm from six males to produce three maternal families. After 1 min of fertilization, embryos were washed 3 times with dechlorinated water and immediately placed on mesh egg mats. Once the eggs had adhered to the egg mats, flow of well-oxygenated water at 12°C was applied over them (Earhart et al., 2020b).
Post-hatch, larvae were transferred to three 9 l flow-through aquaria with aeration and bio-balls as substrate, with each maternal family initially reared in a separate tank. Starting at 13 days post-fertilization (dpf), the rearing temperature was increased by 1°C day−1 until 16°C. Once temperatures reached 16°C at 17 dpf, developing sturgeon were transferred into five replicate 9 l flow-through aquaria. Each aquarium had 100 sturgeon larvae from each maternal family (n=300 sturgeon per replicate tank to reduce stress and mortality by decreasing rearing density; Aidos et al., 2020). Beginning at 19 dpf, freshly hatched Artemia (Artemia International LLC, Fairview, TX, USA) were provided as a starting diet before yolk sac absorption had been completed, and tank substrate was removed over a 7 day period (Earhart et al., 2020a). Lake sturgeon were fed freshly hatched Artemia to satiation 3 times daily until LPS challenges began (59–63 dpf).
At 33 dpf, acclimation began by reducing the stocking density of sturgeon further. For each acclimation temperature (16 and 20°C), four replicate 9 l tanks were filled with approximately 170 sturgeon each, evenly distributed from the five replicate initial rearing tanks for a total of 680 sturgeon per acclimation temperature. Sturgeon remained in these tanks until they were moved to the LPS trial experimental setup at 52 and 56 dpf for 20 and 16°C treatments, respectively. Throughout initial rearing and acclimation, mortality and rearing temperature was monitored at least twice daily. All animals in this study were reared and sampled under guidelines established by the Canadian Council for Animal Care and approved by the animal care committee at the University of Manitoba under protocol #F15-007.
Acclimation
Acclimation and experimental trials were staggered by 4 days to make accumulated thermal exposure similar across the two treatment groups (degree days at the beginning of the trials: 16°C, 950; 20°C, 980; a difference of approximate 2 developmental days at 16°C). In addition to temperature acclimation, developing lake sturgeon were also acclimated to the environment of the LPS trial experimental tanks for 1 week prior to experimentation (beginning at 56 and 52 dpf for 16 and 20°C treatments, respectively), to avoid possible stress-related effects of handling and transfer (Bugg et al., 2021a). Acclimation for the 20°C treatment began at 34 dpf, increasing the temperature 1°C day−1 until 20°C was reached at 37 dpf; this temperature was maintained until the end of the study. LPS trials began 22 days later (59 dpf; Fig. 2), with a 48 h LPS exposure (61 dpf) and ended following the 7 day recovery period (68 dpf). For sturgeon acclimated to 16°C, this temperature was maintained from 16 dpf until the beginning of the LPS trials (63 dpf), throughout the 48 h trial (65 dpf), and during the 7 day recovery period (72 dpf).
Each LPS trial experimental tank was 30×25×7 cm L×W×H with drainage holes cut into the side at the 3.3 l volume mark to allow for water to flow out of the unit during acclimation. A total of approximately 75 sturgeon (selected from the four acclimation tanks) were stocked into each of six experimental tanks for each acclimation temperature (16 and 20°C), with duplication for each of the LPS treatments (i.e. two replicate tanks for each concentration of 0, 30 and 60 μg ml−1 LPS for each temperature, n=12 tanks total). Throughout experimental tank acclimation, developing sturgeon were provided with flow-through water and aeration. During the first 6 days of acclimation, sturgeon were fed Artemia to satiation, and were fasted for 24 h before the initiation of LPS trials. Throughout acclimation and LPS trials, water temperature was recorded every 15 min by HOBO Water Temperature Pro v2 Data Loggers (Onset Computer Corporation, Bourne, MA, USA).
LPS trials and sampling
The LPS trials conducted on developing lake sturgeon were based on previously established protocols (Dalmo et al., 2000; Novoa et al., 2009; Bugg et al., 2021a). Developing lake sturgeon were exposed to concentrations of 0, 30 and 60 μg ml−1 of LPS from Pseudomonas aeruginosa (extracted to >97% purity by phenol extraction; Sigma-Aldrich, St Louis, MO, USA) for 48 h. Each tank was dosed once with the designated concentration of LPS diluted with ultrapure water. Following introduction of LPS into the experimental tanks, each tank was monitored every 15 min for the first 8 h and then at least every 2 h for the following 40 h. As flow-through water was removed for the duration of the trials, ammonia was also measured every 6 h to monitor any potential accumulation and found to be below 1 mg l−1 in all tanks throughout the study. Following the 48 h LPS exposure, flow-through water was then returned, and fish remained in these flow-through units for 7 days to recover from their LPS exposure.
An additional 40 fish from each treatment with surviving fish (20 replicates per experimental tank) were sampled pre-trial, at the end of the 48 h trial and 7 days post-trial during recovery for collection of skin mucus samples. Sturgeon were euthanized as described above, carefully patted dry to remove excess water, and then mucus was collected using Puritan PurFlock Ultra Flocked swabs (Puritan, Guilford, ME, USA). Prior to collecting the sample, each swab was individually weighed (to 0.0001 g). Swabs were then firmly rubbed across the dorsal and ventral surfaces of each sturgeon with the mucus from four sturgeon collected with one sample swab in order to amass an adequate mucus sample for detection of lysozyme activity based on results from preliminary trials (samples >0.25 mg performed well for enzyme detection). Following mucus collection, each swab was reweighed to determine the amount of mucus collected in the sample. Next, the stem of the swab was snipped off and the head of the swab was place in a CryoELITE cryogenic vial (DWK Life Sciences, Millville, NJ, USA) and immediately flash frozen in liquid nitrogen, after which it was stored at −80°C until analysis for lysozyme activity.
Post-sampling processing
Whole-body samples of sturgeon collected both prior to and during LPS trials were homogenized to measure the mRNA transcript abundance, cortisol concentration and lysozyme activity from each individual sturgeon sampled. Each whole fish was individually homogenized using a pestle and mortar in liquid nitrogen. All homogenized samples were then returned to storage at −80°C until further analysis.
Primer design
Primers were designed to target genes involved in the innate immune, stress and fatty acid responses of lake sturgeon to the combined effects of elevated temperature and bacterial infection (Table 1). Many primers were sourced from previous studies conducted on the lake sturgeon immune response (MyD88, IL-1β; Bugg et al., 2021a), stress response (StAR, GR1, HSP70, HSP90a; Bugg et al., 2020; Earhart et al., 2020a) and fatty acid metabolism response (PLA2, CPT1; Yoon et al., 2022). Additionally, primers for other targets in immune-responsive pathways (TLR4, TICAM-1, NF-κB, TNFα, IL-8, C3, Lysozyme-C, IgM) and potential reference genes (RPL13a, eEF1A1 and RPL4) were designed from lake sturgeon (Thorstensen et al., 2022 preprint) and white sturgeon, Acipenser transmontanus (Doering et al., 2016), tissue-specific transcriptomes. All results from transcriptome searches were aligned against publicly available transcripts using NCBI BLAST (Johnson et al., 2008), with primers designed over conserved regions between the query and the search result(s). Original transcript sources, results from related species with highly conserved regions, percentage identities and accession numbers for each publicly available transcript from NCBI BLAST results are listed in the Supplementary Materials and Methods.
RNA extraction, cDNA synthesis and qPCR
Total RNA was extracted from the whole-body homogenates of developing lake sturgeon from all treatment groups using RNeasy Plus Mini Prep Kits (Qiagen, Germantown, MD, USA) following the manufacturer's instructions. Whole-body homogenates were additionally homogenized in 500 µl of lysis buffer for 5 min at 50 Hz using a TissueLyser II (Qiagen). Total concentration, integrity and purity of RNA for each sample were assessed using a Nanodrop One (Thermo Fisher Scientific, Waltham, MA, USA) and gel electrophoresis. Synthesis of cDNA was conducted using a SuperScript IV First-Strand Synthesis System with ezDNase Enzyme (Quantbio, Beverly, MA, USA) from 500 ng of total RNA following the manufacturer's instructions. Genomic DNA was first removed using 1 µl of ezDNAse Enzyme prior to cDNA synthesis. Synthesis of cDNA was then performed using 1 µl of 50 ng µl−1 random hexamers to anneal to the template RNA, followed by reverse transcription using 1 µl of SuperScript IV Reverse Transcriptase. Synthesis was conducted using a SimpliAmp Thermal Cycler (Thermo Fisher Scientific) with cycling conditions of one cycle of 23°C for 10 min, one cycle of 55°C for 10 min and one cycle of 88°C for 10 min, with a final hold at 4°C.
Real-time quantitative polymerase chain reactions (qPCR) for each gene of interest (TLR4, MyD88, TICAM-1, NF-κB, TNFα, IL-8, IL-1β, C3, Lysozyme-C, IgM, StAR, GR1, HSP70, HSP90a, PLA2 and CPT1) and potential reference genes (RPL13a, eEF1A1, RPL7, RPS6, β-actin and RPL4) were conducted using 5 µl of PowerUp SYBR Green Master Mix (Applied Biosystems, Bedford, MA, USA), 0.1–0.025 µl of each 100 µmol l−1 forward and reverse primer, 5–10 µl 1:10 nuclease-free water-diluted cDNA per well, with additional nuclease-free water adjusted for each assay to bring the total volume of each well to 12–16 µl based on the amount of cDNA included. Assays for potential reference genes RPS6 and RPL7 included 0.1 µl each of forward and reverse primer per well, while all other assays included 0.025 µl each of forward and reverse primers per well. All assays used 5 µl 1:10 diluted cDNA per well and had a total reaction volume of 12 µl except for the StAR assay, which used 10 µl of 1:10 diluted cDNA per well and a total volume of 16 µl.
As potential reference genes were not stable across acclimation treatments and LPS exposure concentrations, NORMA-Gene, a robust method for qPCR normalization based on the mRNA transcript abundance of target genes, was used for normalization of mRNA transcript abundance, inputting the abundance of all 22 assayed genes to produce the lowest theoretical variance in normalization (Heckmann et al., 2011). Post-normalization, abundance was then analyzed after applying the 2−ΔΔCt method as described by Livak and Schmittgen (2001). The mRNA transcript abundance of all target genes was then normalized to the abundance of the pre-trial 16°C (negative) control group.
Cortisol and lysozyme analysis
Cortisol was assayed in whole-body homogenates as previously described in lake sturgeon (Earhart et al., 2020a), while lysozyme assays were conducted using EnzChekTM Lysozyme Assay Kits as per the manufacturer's instructions (Invitrogen, Thermo Fisher Scientific; see Supplementary Materials and Methods for further details).
Post-trial metabolism and thermal tolerance
During the 7 day recovery from LPS trials, measurements of both the metabolic rates (routine and forced maximum) and CTmax of sturgeon surviving the trials were taken. Metabolic rates were measured 6 days post-trial, while the CTmax of the same fish were measured the following day, 7 days post-trial.
Whole-body metabolic rate (ṀO2) was measured by intermittent flow respirometry (Loligo Systems, Viborg, Denmark) following previously established protocols with some modifications (Yoon et al., 2021). The respirometry system consisted of 16 borosilicate metabolic chambers with oxygen sensor spots (PreSens, Regensburg, Germany) in two water baths (400×225×140 mm L×W×H) on which fiber optic cables were situated in order to read oxygen saturation as percentage air saturation at 1 Hz using a Witrox 4 Oxygen Meter (Loligo Systems). Water temperature was regulated by room temperature while the respirometry system was connected to AutoResp 4.1 on a PC (Loligo Systems). The oxygen probe was calibrated with 0% (2% sodium sulfite) and 100% dissolved oxygen (fully air saturated). The tubing was non-oxygen permeable (Tygon). The volume of the chambers and tubing was 44.60±3.58 ml and 3.51±0.90 ml (mean±s.d.), respectively. The ratio between the metabolic chamber and fish mass was 80.8±22.7 (mean±s.d.). Before the experiment, fish were fasted overnight (approximately 12 h), and at each sampling point, fish (n=8 per treatment) were haphazardly taken from rearing tanks and chased for 15 min to induce forced maximum metabolic rate (FMR). Then, fish were immediately placed into chambers, and ṀO2 was measured for the next 6 h. A black curtain was hung around the respirometry chamber to minimize visual disturbance, while lights were on for the duration of the experiment.
The measurement cycle for FMR consisted of 60 s waiting and 300 s measurement; that for routine metabolic rate (RMR) consisted of 300 s flushing, 60 s waiting and 300 s measurement. ṀO2 slopes with r2≥0.9 were used to ensure linearity in oxygen consumption (96.2% of all slopes; Chabot et al., 2021). Before each experiment, ṀO2 was measured for 15 min without fish to assess background respiration, and background respiration data were used to linearly interpolate over each ṀO2 recording session (Rogers et al., 2016). Then, all ṀO2 data were corrected by subtracting all corresponding background respiration data. The average ratio between background respiration and RMR was 5.3±5.8% (mean±s.d.). Maximum metabolic rate (MMR) was chosen as the highest ṀO2 during the first three measurements, whereas RMR was estimated by averaging ṀO2 of the last 2 h, excluding the first 4 h of acclimation. Because we measured RMR, we chose to report the difference (ΔṀO2) and ratio (MMR/RMR) between routine and maximum metabolic rate as metabolic scope, both of which are analogous to absolute and factorial aerobic scope in the literature (Halsey et al., 2018).
After measurement of metabolic rates, CTmax was measured following previously established protocols with some modification (Bugg et al., 2020). At the end of metabolic rate measurements, the 8 sturgeon were transferred from metabolic chambers directly into the CTmax arena by placing them into individually labeled experimental units (9.5×5×4 cm L×W×H) with mesh-screened sides to allow water to flow through each unit; Yusishen et al., 2020) and well-oxygenated circulating water held at the respective acclimation temperature. An additional 16 fish, for which metabolic rate had not been measured, were added to each CTmax trial for a total of 24 fish per trial. The position of all fish in the experimental units in the CTmax arena was assigned via a random number generator. Temperature in the recirculating water bath was heated by approximately 0.3°C min−1 using a Isotemp recirculating heater (Fisher Scientific, Hampton, NH, USA) until fish were unable to right themselves after a physical disturbance. When sturgeon were unable to right themselves, the final CTmax temperature was recorded, the fish was euthanized, and mass and length measurements were recorded as described above.
Statistical analysis
Physiological data collected throughout acclimation, including body mass, length, condition factor, HSI and energy density, were analyzed using a two-factor ANOVA including acclimation treatment and developmental time (dpf) in the model as main effects. Importantly, our initial data analysis indicated that rearing tank did not influence the results and thus it was removed from the analysis.
Differences in mortality between acclimation treatments and LPS concentrations throughout trials were assessed using Cox proportional hazards models using the ‘survival’ and ‘survminer’ R packages (https://CRAN.R-project.org/package=survival; https://cran.r-project.org/web/packages/survminer/index.html), with covariates of both acclimation treatment and LPS concentration included in the model. A pairwise comparison was then conducted to compare mortality across both concentrations and acclimation temperatures, using the ‘pairwise_survdiff’ function in the ‘survminer’ package, as well as a Bonferroni correction to correct for the effects of multiple comparisons. Assumptions of the hazard model was evaluated using the ‘cox.zph’ function in the ‘survival’ package. Data collected from each mortality (body mass, total length, condition factor, HSI) were analyzed to determine whether there was a relationship with time to mortality using a Spearman's correlation to identify which physiological metrics were most indicative of time to survival in sturgeon exposed to LPS. Mortalities were only apparent in the 20°C acclimation treatment exposed to 60 μg ml−1 LPS; thus, these data are representative of mortality under these conditions (n=142).
The mRNA transcript abundance of all target genes, as well as whole-body cortisol, lysozyme and mucosal lysozyme activity, were analyzed using three-factor ANOVA to investigate changes in mRNA transcript abundance across treatments; this included acclimation treatment, LPS concentration and time in the model as main effects. A subset of the total mRNA transcript abundance data, including negative control samples following acclimation, 48 h exposure to 30 μg ml−1 LPS and a 7 day recovery, was then analyzed using two-factor ANOVA with acclimation treatment and gene as fixed effects in the model to focus on the sub-lethal tolerance thresholds of LPS exposure.
Principal component analysis (PCA) was conducted using the ‘factomineR’ (Le et al., 2008) and ‘factoextra’ (https://cran.r-project.org/package=factoextra) packages in R, including the subset of data with negative control samples following acclimation, 48 h exposure to 30 μg ml−1 LPS and a 7 day recovery (n=48). Contributions and vector directions of the variance in mRNA transcript abundance observed in PCA were illustrated using a contributions plot and variable plot, with only target genes that exceeded the expected average contributing towards variance from the overall PCA included.
Semi-partial Spearman's correlation was used to investigate the relationship between the mRNA transcript abundance of each gene and that of other studied genes of sturgeon sampled during LPS trials, using the ‘ppcor’ package (Kim, 2015). This analysis was used to control for the effect of both LPS concentration and exposure time during calculation of Spearman's correlation coefficients and was conducted individually for each acclimation temperature (16 and 20°C, n=80 and 61, respectively) to highlight the differences in mRNA transcript abundance relationships between the two acclimation treatments. The difference between the two values between acclimation treatments was then calculated (ρ16°C−ρ20°C) and is presented as delta rho (Δρ). All values are reported as estimated Spearman's rho (ρ).
Metabolic rate and CTmax of developing lake sturgeon during recovery from LPS exposure were analyzed with two-factor ANOVA, including both LPS concentration and acclimation temperature as well as their interactions in the model as fixed effects. Additionally, correlative relationships between the CTmax of each individual sturgeon and each metabolic metric were analyzed using Spearman's correlations. For all ANOVA Shapiro–Wilk's and Levene's tests were used to assess normality of data and homogeneity of variance along with graphical investigations. If assumptions of either normality or homogeneity were violated, a ranked, log or square root transformation was applied to the dataset. Following evaluation of main and interactive effects, post hoc tests were performed with Tukey's HSD tests from the ‘multcomp’ package (Hothorn et al., 2008). All statistical analyses were performed using R 4.0.0 (http://www.R-project.org/), with a significance level (α) of 0.05.
RESULTS
Effects of acclimation temperature
Throughout acclimation, sturgeon reared at 20°C were larger than their 16°C counterparts with an interaction of both temperature and time (P<0.005). However, sturgeon from the two acclimation treatments had similar condition and energy density, with an effect of time on both metrics (P<0.0001). Following 14 days of acclimation, 20°C-acclimated sturgeon had lower hepatosomatic indices than 16°C-acclimated sturgeon, and this effect persisted throughout acclimation (Temperature:Time interaction, P<0.005). Physiological data are shown in Fig. S1.
In-trial mortality and physiological relationships
During trials, there was complete mortality (100%) in the 20°C-acclimated, 60 µg ml−1 LPS treatment group. Additionally, there were 2 mortalities (1.3%) in the 16°C-acclimated, 60 µg ml−1 LPS treatment group. There were no mortalities in any of the other treatment groups.
Physiological metrics of body mass, length and HSI demonstrated significant correlations with the time to mortality of 20°C-acclimated sturgeon exposed to 60 µg ml−1 LPS (P<0.005), in which all sturgeon perished. Condition factor (K) demonstrated no significant relationship (P>0.1), while the metric with the strongest correlation with time to mortality was HSI (ρ=0.36; P<0.0001). Body mass and length demonstrated similar correlative relationships (ρ=0.28 and ρ=0.26, respectively; P<0.005). These correlative relationships can be found in Fig. S2.
LPS-induced molecular modifications
ANOVA
Three-factor ANOVA
There was an effect of acclimation temperature, either individual or interactive, on the mRNA transcript abundance of all studied genes (P<0.05; Figs S3–S6) across pathogen detection, immune response, fatty acid response and stress response. These effects of temperature extended to physiological factors, including whole-body cortisol concentration (Temperature:LPS and LPS:Time, P<0.0005), whole-body lysozyme activity (Temperature, LPS and Time, P<0.05) and mucosal lysozyme activity (Temperature, P<0.005; Fig. 3). There were interactive effects of LPS exposure, temperature and time on the mRNA transcript abundance of genes MyD88, GR1, HSP70, HSP90a, Lysozyme-C, IL-1β and PLA2 (P<0.05) and combined effects of Temperature:Time and LPS:Time on IL-8, TICAM-1, CPT1, StAR, C3, NF-κB, TNFα and TLR4 (P<0.05; further ANOVA data can be found in Table S1, Figs S3–S6).
Multiple comparison tests for 16°C-acclimated sturgeon revealed that all pathogen detection genes (TLR4, MyD88, TICAM-1 and NF-κB) demonstrated an upregulation over the time course of 30 µg ml−1 LPS exposure (similar to patterns in 16°C-acclimated 60 µg ml−1 LPS-exposed sturgeon), while a similar upregulation was not observed in their 20°C-acclimated counterparts (P<0.05) or in 0 µg ml−1 LPS-exposed sturgeon (handling controls). Immune-responsive genes demonstrated more nuanced changes in abundance, with C3 and TNFα increasing for only 16°C-acclimated sturgeon throughout the time course (P<0.05), while IL-1β, IL-8 and Lysozyme-C had similar responses in 16 and 20°C. However, IgM was higher in 20°C-acclimated sturgeon across sampling points (P<0.05). Fatty acid-responsive genes, CPT1 and PLA2, were more responsive to LPS in 16°C-acclimated sturgeon than in 20°C-acclimated fish (P<0.05), with higher peak mRNA transcript abundance following exposure and persistent increases in abundance of PLA2 mRNA (P<0.05). Finally, stress-responsive genes StAR and GR1 were upregulated throughout the exposure to LPS and recovery at 16°C, while their abundance was not modified by LPS in 20°C-acclimated sturgeon (P<0.05). Chaperones HSP70 and HSP90a were upregulated across both acclimation temperatures during LPS exposure, with a higher magnitude of induction for HSP70 in 16°C-acclimated sturgeon (P<0.05). Detailed results for all three-factor ANOVA analyses are provided in Table S1.
Two-factor ANOVA
Overall, there was a strong effect of acclimation temperature on mRNA transcript abundance for each analyzed time point (P<0.0001; Fig. 4). Prior to the beginning of LPS trials, acclimation to 20°C increased mRNA transcript abundance of genes across biological processes (Fig. 4A; P<0.05) as compared with 16°C-acclimated sturgeon. The mRNA transcript abundance of both MyD88 and NF-κB, involved in pathogen detection, increased approximately 1.5-fold in 20°C- as compared with 16°C-acclimated sturgeon (P<0.05). Additionally, C3 and Lysozyme-C mRNA abundance, involved in the immune response, were increased 4.8-fold and 2.4-fold, respectively, when compared with 16°C-acclimated sturgeon (P<0.05). Mitochondrial fatty acid transporter CPT1 mRNA abundance throughout acclimation also increased 3.1-fold in 20°C-acclimated sturgeon versus 16°C-acclimated sturgeon (P<0.05). Finally, GR1, HSP70 and HSP90a, all of which are involved in the endocrine stress response, increased their transcript abundance 2.2-, 2.0- and 1.9-fold, respectively, in 20°C- compared with 16°C-acclimated sturgeon (P<0.05).
However, following 48 h exposure to 30 µg ml−1 LPS, 16°C-acclimated sturgeon had higher levels of mRNA transcript abundance across biological processes, when compared with their 20°C-acclimated counterparts at the same time point and LPS exposure (Fig. 4B; P<0.05). Intracellular signaling molecules MyD88 and TICAM-1 were 78.5% and 48.8% more highly induced in 16°C-acclimated lake sturgeon, respectively, when compared with 20°C-acclimated sturgeon (P<0.05). While all immune-responsive genes were quantitatively higher in mRNA abundance for 16°C-acclimated sturgeon (with the exception of IgM), only IL-1β demonstrated significantly higher transcript abundance, 2.2-fold higher in 16°C-acclimated sturgeon than in 20°C-acclimated sturgeon (P<0.05). In contrast to the observed trend of the other immune-responsive genes, IgM was 58% higher in 20°C-acclimated sturgeon (P<0.05). Fatty acid-responsive genes PLA2 and CTP1 were much more highly induced in 16°C-acclimated sturgeon, 5.6- and 2.5-fold, respectively, when compared with their 20°C-acclimated counterparts (P<0.05). For stress-responsive genes, only HSP70 had higher mRNA transcript abundance between acclimation treatments, with 2-fold higher abundance in 16°C-acclimated sturgeon.
Following a 7 day recovery, these increased mRNA responses in 16°C-acclimated lake sturgeon observed during LPS exposure persisted, with higher levels of mRNA transcript abundance across pathogen detection, fatty acid and stress response processes when compared with 20°C-acclimated sturgeon (Fig. 4C; P<0.05). Pathogen detection components of the toll-like receptor signaling complex TLR4, MyD88 and TICAM-1 were all upregulated in 16°C as compared with 20°C-acclimated lake sturgeon, 1.9-, 1.5- and 2.6-fold, respectively (P<0.05). Fatty acid-responsive genes PLA2 and CPT1 showed 5.5- and 1.5-fold higher mRNA abundance, respectively, in 16°C-acclimated sturgeon when compared with 20°C-acclimated sturgeon (P<0.05). Stress-responsive genes StAR, GR1 and HSP70 were also all elevated in 16°C- versus 20°C-acclimated lake sturgeon, 2-, 1.4- and 2.3-fold, respectively (P<0.05).
PCA
PCA (Fig. 5A) demonstrated separation between 16 and 20°C acclimation treatments across their responsive trajectories to acclimation, 48 h LPS exposure and 7 day recovery. Principal component 1 (PC1) along the x-axis explained 43.6% of the variation, while PC2 on the y-axis explained 13.1%. As developing sturgeon responded to LPS exposure, the two acclimation treatments separated out from the left to the right side of the y-axis, but with different responsive trajectories, as 16°C-acclimated sturgeon moved to the bottom right-hand quadrant, while those acclimated to 20°C moved to the upper left. Following recovery, these responses also differed, with 16°C-acclimated sturgeon moving to the upper right quadrant and 20°C-acclimated sturgeon regressing to the upper left. Thus, the mRNA transcriptional responses of the acclimation treatments had different trajectories, but also differences in their magnitude, with 16°C-acclimated sturgeon moving further across the axes than 20°C-acclimated sturgeon in response to LPS exposure.
The contribution of variables to observed variation in the PCA was distributed across biological processes, with at least one gene from each contributing more than the average expected value to the overall variation observed (Fig. 5B). Genes passing this average expected contribution threshold were TLR4, MyD88 and TICAM-1 (involved in pathogen detection), Lysozyme-C and IL-8 (involved in the immune response), HSP70, StAR and GR1 (involved in the stress response), and CPT1 (involved in fatty acid responses). Variable plots further demonstrate the response trajectory, which the variation in mRNA transcript abundance of these genes contributed to the PCA, with all genes over the expected contribution threshold moving to the righthand side of the y-axis (Fig. 5C).
Correlative relationships
The relationships between mRNA transcript abundance for lake sturgeon differed according to acclimation treatments throughout acclimation, LPS exposure and recovery, with higher Spearman's correlation coefficients across 16°C-acclimated sturgeon when comparing genes across biological processes (Fig. 6A). Transcript abundance of many genes involved in the downstream responses to pathogen detection of LPS (i.e. IL-8, C3, Lysozyme-C, PLA2, CPT1, GR1, HSP70 and HSP90a) demonstrated strong correlative relationships (ρ≥0.6) with that of intracellular immune signaling molecules (MyD88 or TICAM-1) for 16°C-acclimated sturgeon. In contrast there were no relationships as strong as these for the above genes in 20°C-acclimated sturgeon, and only one as strong when comparing the relationships of every studied gene (IL-8 to IL-1β). The largest differences in correlative relationships between the treatments can be found in the transcript abundance of stress-responsive genes, GR1, HSP70 and HSP90a (Fig. 6B), while the relationship of StAR expression with that of every other gene was relatively consistent across the acclimation treatments. There were also inconsistencies between the acclimation treatments in the relationship between C3 and Lysozyme expression and that of fatty acid-responsive genes PLA2 and CPT1, and for C3 with elements of pathogen detection.
Post-trial metabolism and thermal tolerance
While there was an interactive effect of acclimation temperature and LPS (P<0.05) on CTmax, there was no compromise in thermal tolerance by the effects of LPS exposure in either acclimation treatment (Fig. S7). In contrast, several metabolic traits were suppressed following exposure to LPS, but only in 16°C-acclimated sturgeon (Fig. 7). Importantly, measurement of metabolic rate prior to CTmax trials did not impact the resulting CTmax for sturgeon from any treatment (P>0.05). There was no effect of LPS exposure or acclimation temperature on the routine metabolic rate of lake sturgeon; however, there were effects on FMR, ΔṀO2 and metabolic scope, all of which demonstrated significant interactions between acclimation treatment and LPS concentration (P<0.01). Further, multiple comparisons revealed specific differences between acclimation treatments and across LPS concentrations.
In 16°C-acclimated sturgeon, FMR was reduced in both 30 and 60 µg ml−1 LPS, by 38.5% and 40.9%, respectively, compared with control sturgeon (P<0.01; Fig. 7B). Also, FMR for 20°C-acclimated sturgeon exposed to 30 µg ml−1 LPS was 30.3% higher than that of 16°C-acclimated sturgeon (P<0.05). Multiple comparisons for ΔṀO2 indicated a decrease of 73.6% and 88.6% for 16°C-acclimated lake sturgeon following exposure to LPS concentrations of 30 and 60 µg ml−1, respectively, when compared with controls acclimated to the same temperature (P<0.001; Fig. 7C). This again resulted in higher ΔṀO2, similar to FMR, between the acclimation treatments exposed to 30 µg ml−1 LPS (P<0.05), with a 55.5% higher ΔṀO2 in 20°C-acclimated sturgeon than in their 16°C counterparts. Finally, metabolic scope also decreased in LPS-exposed sturgeon in the 16°C acclimation treatment, by 41.6% and 54.7% for 30 and 60 µg ml−1 LPS treatments when compared with that of handling control fish, respectively (P<0.005; Fig. 7D). However, there was no change in FMR or ΔṀO2, or metabolic scope for 20°C-acclimated fish.
There was a significant interaction between acclimation treatment and LPS concentration on the CTmax of sturgeon following the 7 day recovery period from LPS trials (P<0.05; Fig. S7). Sturgeon acclimated to 20°C and exposed to 30 µg ml−1 LPS had a CTmax 0.44°C higher than control sturgeon not exposed to LPS (P<0.001). There were no differences in CTmax for 16°C-acclimated sturgeon across LPS exposure concentrations. Across acclimation temperatures, sturgeon acclimated to 20°C had CTmax values 2.71 and 3.13°C higher than those of 16°C-acclimated sturgeon, for handling control and 30 µg ml−1 LPS, respectively (P<0.0001). There was no correlation between CTmax for individual fish with any of their respective measured metabolic metrics.
DISCUSSION
In the current study, we investigated the effects of temperature on the innate immune responses of lake sturgeon during early life. Increased environmental temperatures influenced the mRNA transcript abundance of every measured endpoint. Further, when fish were also challenged with LPS, elevated temperature dampened their innate immune capacity, as observed across innate immune-, fatty acid- and stress-responsive biological processes. Acclimation temperatures used in the current study were approximately 2–3°C below maximum sustained summer temperatures for this population of lake sturgeon (Bugg et al., 2020, 2021b). These results suggest the presence of seasonal sub-lethal thermal thresholds on innate immunity for wild populations of lake sturgeon throughout Manitoba, which may be especially vulnerable to the effects of pathogens during early development (Clouthier et al., 2020). While developing sturgeon may be plastic in the face of thermal changes (Bugg et al., 2020; Penman, 2021), this plasticity accompanied by the induction of the glucocorticoid stress response and decrease in energy reserves may diminish their capacity to mount an effective immune response against opportunistic pathogens.
Effects of acclimation temperature and in-trial mortality
While there was no difference in condition or energy density prior to the LPS trials, length and mass were greater and HSI was less for 20°C-acclimated sturgeon when compared with their 16°C-acclimated counterparts. The HSI showed the strongest relationship to time to mortality during LPS exposure. As HSI is an indication of fatty acid and glycogen stores accrued during development (Chellappa et al., 1995; Rossi et al., 2017; Morrison et al., 2020), these findings suggest that the effects of increasing temperature compromise the acquisition and allocation of these energy reserves (Yoon et al., 2022), contributing to increases in the susceptibility of developing sturgeon to pathogenic stressors. Interestingly, there was no difference in whole-body energy density of lake sturgeon from different acclimation treatments during development, while HSI and growth differed between acclimation treatments. This suggests that increasing temperature influenced energy partitioning in developing sturgeon, either into storage in the liver or into somatic growth (Post and Parkinson, 2001). This metabolic trade off, to increase somatic growth at higher temperatures, may ultimately have immunocompromising effects (Kim et al., 2019). Research in pallid sturgeon, Scaphirhynchus albus, and white sturgeon, A. transmontanus, suggested that increased rearing temperatures resulted in high levels of mortality (50–60%) once a pathogen was introduced, while there was limited mortality (<10%) in lower temperature acclimation treatments (Coleman et al., 2018; Stilwell et al., 2022). These findings demonstrate thermal thresholds for pathogen-induced mortality, which is potentially associated with the energy allocation of developing sturgeon and may have contributed to the mortality associated with elevated temperatures in the current study.
Impacts of thermal acclimation on mRNA abundance
Prior to the LPS trials, lake sturgeon acclimated to 20°C had increased mRNA transcript abundance of genes involved in pathogen detection, immune responsiveness, fatty acid responses and stress responses. Increased transcript abundance of pathogen detection (MyD88 and NF-κB) and immune-responsive genes (C3 and Lysozyme-C) may indicate a stress-responsive role for their induction, as observed in channel catfish, Ictalurus punctatus, and large yellow croaker, Larimichthys crocea (Small and Bilodeau, 2005; Sun et al., 2017). Interestingly, mRNA abundance of CPT1 (involved in fatty acid oxidation; Coccia et al., 2014) and genes involved in the HPI axis and general stress response (GR1, HSP70 and HSP90a) were upregulated in response to warm acclimation. Together, these findings, paired with a decrease in HSI for 20°C-acclimated sturgeon, suggest that accumulated thermal stress results in an activation of the glucocorticoid stress response, and an increase in the oxidation of fatty acids, similar to observations made in thermally stressed Atlantic salmon (Norambuena, et al., 2015). Although there were increases in immune, energy production and stress response mRNA transcript abundance in the 20°C treatment following acclimation, these did not result in enhanced survival during LPS trials, and likely represent thermal stress-responsive mechanisms.
LPS-induced transcriptional modifications
Sturgeon acclimated to 16°C had stronger activation of pathogen-detection mechanisms (MyD88 and TICAM-1), immune-responsive transcripts (IL-1β, IL-8, C3, TNFα and Lysozyme-C), fatty acid responses (PLA2 and CPT1) and stress-responsive mechanisms (HSP70) following LPS exposure, as demonstrated through the increase in mRNA transcript abundance of associated genes, when compared with fish acclimated to 20°C. Interestingly, IgM expression was not highly induced at either temperature following LPS exposure, suggesting that sturgeon may not be able to strongly upregulate this primarily adaptive immune response and may be relying on innate immune mechanisms at this developmental stage instead. Together, these responses suggest that 16°C-acclimated sturgeon produced a stronger stimulation of the immune signaling cascade (Deguine and Barton, 2014; Tanekhy, 2014) and may exhibit a higher capacity to respond with both innate immune and energetically intensive processes (Angosto and Mulero, 2014; Arnemo et al., 2017), and that acclimation to 20°C disrupted the magnitude and timing of the innate immune responses induced by LPS exposure at the level of the transcriptome (Bennoit and Craig, 2020). These stronger observed responses in 16°C-acclimated sturgeon represent a larger physiological capacity to effectively counter pathogenic infection, compared with sturgeon acclimated to 20°C. Further, following a 7 day recovery from LPS exposure, elevated mRNA responses in 16°C-acclimated sturgeon were sustained, remaining higher than those of their 20°C-acclimated counterparts across pathogen detection, fatty acid- and stress-responsive processes, indicating that chronic thermal stress may impede long-term physiological processes.
Although the abundance of many mRNA transcripts was largely different between acclimation treatments, there was an induction of IL-8, Lysozyme-C and C3 transcripts across both temperature treatments, which is consistent with other studies of the sturgeon immune response (Li et al., 2017; Lou et al., 2018; Valipour et al., 2018; Hohne et al., 2021). These innate immune mechanisms may play a crucial role in pathogen defense in sturgeon, especially during early development (Magnadottir, 2006; Wang et al., 2009; Huber-Lang et al., 2018). However, the induction of these transcripts did not improve survival at 20°C during the LPS trials. This result emphasizes the importance of peripheral immune-responsive mechanisms in pathogen defense (e.g. unstudied processes at the receptor, complement, cytokine and antimicrobial peptide levels) in developing fish.
Effects of temperature on lysozyme and HPI axis activity
Both whole-body lysozyme and mucosal lysozyme were elevated in 16°C when compared with their 20°C-acclimated counterparts. As lysozyme is a key innate immune enzyme involved in bacterial defense (Saurabh and Sahoo, 2008), decreases in lysozyme activity demonstrate that the effects of temperature extend to protein-level responses. Further, both 16 and 20°C-acclimated sturgeon exposed to 60 μg ml−1 LPS increased cortisol concentrations at the 4 h time point, confirming that HPI axis-related responses are activated in lake sturgeon following pathogenic stimulus (Haukenes et al., 2008; Bugg et al., 2021a).
Overall, lake sturgeon acclimated to 20°C were less transcriptionally responsive than their 16°C counterparts (PCA analysis) and suffered a breakdown of their transcriptional relationships regulating pathogen detection and downstream processes (correlative analysis), resulting in a modification of the magnitude of the transcriptional response of various immune-related processes (Bennoit and Craig, 2020). Interestingly, the GR response of the 16°C-acclimated sturgeon following LPS stimulus was upregulated to the same levels as in 20°C-acclimated sturgeon under control conditions, suggesting that 20°C-acclimated sturgeon are already stimulating the HPI axis and may be chronically stressed, resulting in their inability to respond to additional stressors such as LPS. In contrast, 16°C-acclimated individuals were able to respond through elevation of cortisol, GR and HSP70. Further, HPI axis-related co-chaperones involved in the thermal stress response were also upregulated in 20°C- when compared with 16°C-acclimated sturgeon. Chronic elevation of a stress response is energetically costly, consistent with decreased HSI as well as the suppression of transcriptional activation (Tort, 2011; Alfonso et al., 2021), and likely impacted the capacity of lake sturgeon to physiologically respond to LPS when acclimated to elevated temperatures.
Post-trial metabolism and thermal tolerance
It is expected that activation of immune responses is energetically costly, which can increase metabolic rate following pathogenic infection (Martin and Krol, 2017; Bennoit and Craig, 2020; Polinski et al., 2021). Our results contrast those from similar studies exposing mosquitofish (Gambusia holbrooki) and zebrafish (Danio rerio) to an immune stimulus, where, 1 week following exposure, they demonstrated increased metabolic scope and RMR, respectively (Bonneaud et al., 2016; Bennoit and Craig, 2020). In sockeye salmon, Oncorhynchus nerka, a purportedly minimal metabolic cost and a reportedly highly energetically efficient innate immune response to infection (Polinski et al., 2021) was observed. However, in our study, LPS exposure and induction of immune-responsive mechanisms did not increase RMR, but instead resulted in a depression of FMR and metabolic scope only in 16°C-acclimated sturgeon exposed to LPS. This finding may indicate that innate immune activation in lake sturgeon is more energetically costly when compared with that in more recently derived fish lineages or results in a limitation based on energy constraints during this early life stage. Further, it remains unclear why the aerobic capacity of sturgeon acclimated to 20°C was not altered, but it may suggest that their ability to manipulate their aerobic capacity was limited at this higher temperature. Thus, there may be a constraint on energy allocation under elevate temperatures which may be influenced by the increase in other routine activities including growth, as demonstrated by decreased HSI. This decrease in metabolic capacity may show a cost–benefit relationship to its activation, with complete mortality observed at 20°C in 60 μg ml−1 LPS. While there is limited research on the metabolic cost of immune activation in fishes, our results suggest that increased energetic costs, due to increased temperatures, may compromise the induction of the immune system in developing lake sturgeon.
Study limitations
This study measured the mRNA abundance of transcripts representing a broad array of genes across biological pathways that contribute to innate immune capacity. However, many of these molecular responses are involved in numerous pathways. Further, in these responses there are many other genes at play, and their mRNA expression is not necessarily directly reflective of protein abundance. Further, sturgeons are also ancestral species, exhibiting differences in their cortisol stress response (Penny et al., 2023) and responses to physiological stressors (Kieffer et al., 2001; Haukenes et al., 2008) when compared with more recently derived teleosts, and this could extend to cytokine-level responses (Li et al., 2017; Lou et al., 2018; Jiang et al., 2018). Thus, direct comparisons of sturgeons with teleosts should be approached with caution because they diverged hundreds of millions of years ago (Du et al., 2020).
The use of whole animals as opposed to targeted immune tissues such as the head kidney could dilute observations of immune-related gene expression. However, we chose to use whole-body samples for two reasons: (1) we aimed to investigate the interaction between LPS and temperature exposure on immune capacity across multiple biological processes, including stress- and fatty acid-responsive gene expression, which would not be possible if we took a tissue-specific approach; and (2) we aimed to investigate these responses in a critical early developmental window where high mortality often occurs, temperatures are elevated and sturgeon are likely reliant on their innate immune responses. This limited the size of the fish used in this study.
Additionally, in the current study, we chose to use an environmental exposure of LPS, which may result in variability of an individual's exposure dosage when compared with an intraperitoneal injection. However, as previous experiments demonstrated this variation to be minor in Atlantic halibut, Hippoglussus hippoglossus (especially over a 48 h exposure; Dalmo et al., 2000), we chose to use an environmental exposure to LPS to illicit a comprehensive whole-organism response (Anderson and Siwicki, 1994), which may more accurately reflect how lake sturgeon would encounter and respond to pathogens in the wild. This was especially important for examining mucosal lysozyme activity, which is a potential non-lethal method of measuring an immune response in juvenile lake sturgeon. Further, the LPS used in the study had a purity of >97%, leaving the possibility (albeit low) of potential impurities that could have impacted the immune responses observed in this study. Despite these potential limitations, the collective evidence from this study suggests that elevated temperatures have a profound impact on the immune capacity of developing lake sturgeon.
Ecological and management implications
Lake sturgeon are exposed to thermal stress and environmentally pervasive pathogens in both hatchery and wild environments, especially during early development. As climate change effects will likely exacerbate cross-species pathogen transmission, it is important to understand how pathogens affect phenotypic development during early life and its impacts on the recruitment of lake sturgeon.
Our data suggest that the effects of increasing temperatures will reduce the immune capacity of lake sturgeon during early development, when they experience high mortality (Sifa and Mathias, 1987; Wieser, 1991; Rombough, 1994). This study demonstrated the effects of 20°C for a period of 22 days; however, wild sturgeon from this population are likely exposed to this temperature, and higher, for prolonged durations, upwards of 50 days in the summer (Bugg et al., 2020). Therefore, further research should focus on the long-term effects of increasing temperatures during early life on pathogen burden, virulence and physiological condition with the goal of evaluating the impacts on health and survival of fishes to effectively manage wild northern populations that may be most threatened by increasing temperatures.
Acknowledgements
The authors thank North South consultants for their assistance in capture of wild spawning adult lake sturgeon. We also would like to thank the staff of the University of Manitoba animal holding facility for their assistance in the care and maintenance of fish. This work could not have been completed without the assistance of lab members Jess MacPherson, Ian Bouyoucos, Tyler Edwards and Jenna Drummond in the spawning and initial rearing of juvenile lake sturgeon. In addition to the above, we also would like to thank Frauke Fhermann, Alaina Taylor, Morgan Anderson, Kaitlynn Weisgerber, Theresa Mackey and Matt Thorstensen for their assistance in early rearing of sturgeon and/or data collection. In addition to their other contributions, we would also like to thank Tyler Edwards and Jenna Drummond for their assistance in conducting CTmax and metabolic rate measurements. We would also like to thank Kyra Shewchuk for performing the RNA extractions for this experiment. Finally, we would like to thank Madison Earhart for her assistance in figure aesthetics and stimulating discussion throughout the production of the manuscript, as well as Miri E. Seo for her willingness to share her artwork, which is included in Fig. 1C. The University of Manitoba campuses are located on original lands of Anishinaabeg, Cree, Oji-Cree, Dakota and Dene peoples, and on the homeland of the M’etis Nation. We recognize that water supplied for our fish at the University of Manitoba campuses is sourced from the Shoal Lake 40 First Nation.
Footnotes
Author contributions
Conceptualization: W.S.B., G.R.Y., A.N.S., A.M.W., K.M.J., W.G.A.; Methodology: W.S.B., G.R.Y., A.N.S., A.M.W., W.G.A.; Validation: G.R.Y., A.N.S., A.M.W.; Formal analysis: W.S.B.; Investigation: W.S.B., G.R.Y., A.N.S., K.M.J., W.G.A.; Resources: W.G.A.; Data curation: W.S.B., G.R.Y., A.N.S., A.M.W.; Writing - original draft: W.S.B., A.M.W.; Writing - review & editing: W.S.B., G.R.Y., A.N.S., A.M.W., K.M.J., W.G.A.; Visualization: W.S.B.; Supervision: K.M.J., W.G.A.; Project administration: W.S.B., W.G.A.; Funding acquisition: K.M.J., W.G.A.
Funding
This work was supported by Natural Sciences and Engineering Research Council of Canada Manitoba Hydro Industrial Research Chair awarded to W.G.A., Natural Sciences and Engineering Research Council of Canada Discovery grant awarded to K.M.J. and University of Manitoba Graduate fellowship awarded to W.S.B. and G.R.Y.
Data availability
All relevant data can be found within the article and its supplementary information.
References
Competing interests
The authors declare no competing or financial interests.