Getting to the heart of anatomical diversity and phenotypic plasticity: fish hearts are an optimal organ model in need of greater mechanistic study

Natural selection has produced many vertebrate 'solutions' for the cardiac life-support system, especially among the approximately 30,000 species of fishes. For example, across species, fish have the greatest range for central arterial blood pressure and relative ventricular mass of any vertebrate group. This enormous cardiac diversity is excellent ground material for mechanistic explorations. Added to this species diversity is the emerging field of population-specific diversity, which is revealing that cardiac design and function can be tailored to a fish population's local environmental conditions. Such information is important to conservation biologists and ecologists, as well as physiologists. Furthermore, the cardiac structure and function of an individual adult fish are extremely pliable (through phenotypic plasticity), which is typically beneficial to the heart's function when environmental conditions are variable. Consequently, exploring factors that trigger cardiac remodelling with acclimation to new environments represents a marvellous opportunity for performing mechanistic studies that minimize the genetic differences that accompany cross-species comparisons. What makes the heart an especially good system for the investigation of phenotypic plasticity and species diversity is that its function can be readily evaluated at the organ level using established methodologies, unlike most other organ systems. Although the fish heart has many merits as an organ-level model to provide a mechanistic understanding of phenotypic plasticity and species diversity, bringing this potential to fruition will require productive research collaborations among physiologists, geneticists, developmental biologists and ecologists.

Sex-specific differences in swimming, aerobic metabolism and recovery from exercise in adult coho salmon (Oncorhynchus kisutch) across ecologically relevant temperatures

Adult female Pacific salmon can have higher migration mortality rates than males, particularly at warm temperatures. However, the mechanisms underlying this phenomenon remain a mystery. Given the importance of swimming energetics on fitness, we measured critical swim speed, swimming metabolism, cost of transport, aerobic scope (absolute and factorial) and exercise recovery in adult female and male coho salmon (Oncorhynchus kisutch) held for 2 days at 3 environmentally relevant temperatures (9°C, 14°C, 18°C) in fresh water. Critical swimming performance (U _crit) was equivalent between sexes and maximal at 14°C. Absolute aerobic scope was sex- and temperature-independent, whereas factorial aerobic scope decreased with increasing temperature in both sexes. The full cost of recovery from exhaustive exercise (excess post-exercise oxygen consumption) was higher in males compared to females. Immediately following exhaustive exercise (i.e. 1 h), recovery was impaired at 18°C for both sexes. At an intermediate time scale (i.e. 5 h), recovery in males was compromised at 14°C and 18°C compared to females. Overall, swimming, aerobic metabolism, and recovery energetics do not appear to explain the phenomenon of increased mortality rates in female coho salmon. However, our results suggest that warming temperatures compromise recovery following exhaustive exercise in both male and female salmon, which may delay migration progression and could contribute to en route mortality.

Limits on performance and survival of juvenile sockeye salmon (Oncorhynchus nerka) during food deprivation: a laboratory-based study

Long-distance migrations can be energetically demanding and can represent phases of high mortality. Understanding relationships between body condition and migratory performance can help illuminate the challenges and vulnerabilities of migratory species. Juvenile anadromous sockeye salmon (Oncorhynchus nerka) may migrate over 1000 km from their freshwater nursery habitats to estuary and ocean feeding grounds. During the period corresponding to the seaward migration of sockeye salmon, we held smolts in the laboratory to ask the following: (i) Does non-feeding migration duration influence prolonged swim performance and survival? (ii) What are the relationships between individual body condition and swim performance and survival? Wild sockeye salmon were intercepted during their migration and held without food for up to 61 days to represent the non-feeding freshwater migration and the extremes of poor estuary habitat. We conducted 40 sets of prolonged swim trials on 319 fish from 3 treatment groups that represented entrance to the marine environment on (i) an average,(ii) a delayed and (iii) a severely delayed migration schedule. Experimentally controlled freshwater migration duration did not impact swim performance or survival. Swim performance decreased concomitant with condition factor, where smolts with a Fulton's condition factor of <0.69 were less likely (<50% probability) to complete the swim test (90 min swim test, at ~0.50 m/s). Survival of salmon smolts in the laboratory was less likely at energy densities of less than 3.47 MJ/kg. Swim performance decreased much sooner than survival, suggesting that swim performance, and therefore condition factor, may be a good indicator of survival of migratory smolts, as fish with reduced swim performance will likely be predated. These two relationships, one more ecologically relevant and one more clinical, help reveal the limits of long-distance migration for juvenile salmon and can be used to determine population-specific starvation risk associated with various freshwater and marine habitat conditions.

Reproductive energy expenditure and changes in body morphology for a population of Chinook salmon Oncorhynchus tshawytscha with a long distance migration

Energetic demands of a long freshwater migration, extended holding period, gamete development and spawning were evaluated for a population of stream-type Chinook salmon Oncorhynchus tshawytscha. Female and male somatic mass decreased by 24 and 21%, respectively, during migration and by an additional 18 and 12% during holding. Between freshwater entry and death after spawning, females allocated 14% of initial somatic energy towards gonad development and 78% for metabolism (46, 25 and 7% during migration, holding and spawning, respectively). Males used only 2% of initial somatic energy for gonad development and 80% on metabolic costs, as well as an increase in snout length (41, 28 and 11% during migration, holding and spawning, respectively). Individually marked O. tshawytscha took between 27 and 53 days to migrate 920 km. Those with slower travel times through the dammed section of the migration corridor arrived at spawning grounds with less muscle energy than faster migrants. Although energy depletion did not appear to be the proximate cause of death in most pre-spawn mortalities, average final post-spawning somatic energy densities were low at 3·6 kJ g^-1 in females and 4·1 kJ g^-1 in males, consistent with the concept of a minimum energy threshold required to sustain life in semelparous salmonids.

Behavioural responses of Pacific salmon to chemical disturbance cues during the spawning migration

Many fish that are exposed to a threat release disturbance cues, which are chemicals that alert conspecifics to the presence of the threat. The release of disturbance cues has been well demonstrated in various species of laboratory-reared fish. Migratory fish species often exhibit increased cortisol levels and are exposed to numerous stressors during their migrations, which could trigger the release of disturbance cues. We tested the responses of wild migrating sockeye salmon (Oncorhynchus nerka) and pink salmon (O. gorbuscha) to the odours of disturbed and undisturbed conspecifics to determine whether these fish release disturbance cues following exposure to a simulated stressor. Furthermore, we tested the responses of sockeye salmon to water-borne cortisol, following evidence from past studies that this chemical is excreted through the gills of stressed fish, and speculation that endogenous correlates of stress might function as disturbance cues. We found that sockeye salmon avoid the odour of disturbed conspecifics, whereas pink salmon do not. Avoidance occurred in both female and male sockeye salmon, and was associated with an increase in plasma cortisol levels in females, but not in males. We also found no behavioural response to water-borne cortisol, which suggests this chemical does not act as an exogenous disturbance cue in sockeye salmon. Avoidance of disturbed conspecifics could limit exposure to risks during the sockeye salmon spawning migration, but could also delay the rate of migration and thereby accrue reproductive costs.

Oxygen uptake in Pacific salmon Oncorhynchus spp.: when ecology and physiology meet

Over the past several decades, a substantial amount of research has examined how cardiorespiratory physiology supports the diverse activities performed throughout the life cycle of Pacific salmon, genus Oncorhynchus. Pioneering experiments emphasized the importance of aerobic scope in setting the functional thermal tolerance for activity in fishes. Variation in routine metabolism can have important performance and fitness consequences as it is related to dominance, aggression, boldness, territoriality, growth rate, postprandial oxygen consumption, life history, season, time of day, availability of shelter and social interactions. Wild fishes must perform many activities simultaneously (e.g. swim, obtain prey, avoid predators, compete, digest and reproduce) and oxygen delivery is allocated among competing organ systems according to the capacity of the heart to deliver blood. For example, salmonids that are simultaneously swimming and digesting trade-off maximum swimming performance in order to support the oxygen demands of digestion. As adult Pacific salmonids cease feeding in the ocean prior to their home migration, endogenous energy reserves and cardiac capacity are primarily partitioned among the demands for swimming upriver, sexual maturation and spawning behaviours. Furthermore, the upriver spawning migration is under strong selection pressure, given that Pacific salmonids are semelparous (single opportunity to spawn). Consequently, these fishes optimize energy expenditures in a number of ways: strong homing, precise migration timing, choosing forward-assist current paths and exploiting the boundary layer to avoid the strong currents in the middle of the river, using energetically efficient swimming speeds, and recovering rapidly from anaerobic swimming. Upon arrival at the spawning ground, remaining energy can be strategically allocated to the various spawning behaviours. Strong fidelity to natal streams has resulted in reproductively isolated populations that appear to be locally adapted physiologically to their specific environmental conditions. Populations with more challenging migrations have enhanced cardiorespiratory performance. Pacific salmonids are able to maintain aerobic scope across the broad range of temperatures encountered historically during their migration; however, climate change-induced river warming has created lethal conditions for many populations, raising conservation concerns. Despite considerable research examining cardiorespiratory physiology in Pacific salmonids over the last 70 years, critical knowledge gaps are identified.

Dorsal hump morphology in pink salmon (Oncorhynchus gorbuscha)

Mature male Pacific salmon (Genus Oncorhynchus) develop a dorsal hump, as a secondary male sexual characteristic, during the spawning period. Previous gross anatomical studies have indicated that the dorsal humps of salmon are mainly composed of cartilaginous tissue (Davidson [1935] J Morphol 57:169-183.) However, the histological and biochemical characteristics of such humps are poorly understood. In this study, the detailed microstructures and components of the dorsal humps of pink salmon were analyzed using histochemical techniques and electrophoresis. In mature males, free interneural spines and neural spines were located in a line near to the median septum of the dorsal hump. No cartilaginous tissue was detected within the dorsal hump. Fibrous and mucous connective tissues were mainly found in three regions of the dorsal hump: i) the median septum, ii) the distal region, and iii) the crescent-shaped region. Both the median septum and distal region consisted of connective tissue with a high water content, which contained elastic fibers and hyaluronic acid. It was also demonstrated that the lipid content of the dorsal hump connective tissue was markedly decreased in the mature males compared with the immature and maturing males. Although, the crescent-shaped region of the hump consisted of connective tissue, it did not contain elastic fibers, hyaluronic acid, or lipids. In an ultrastructural examination, it was found that all of the connective tissues in the dorsal hump were composed of collagen fibers. Gel electrophoresis of collagen extracts from these tissues found that the collagen in the dorsal hump is composed of Type I collagen, as is the case in salmon skin. These results indicate that in male pink salmon the dorsal hump is formed as a result of an increase in the amount of connective tissue, rather than cartilage, and the growth of free interneural spines and neural spines.

Transcriptomic responses to high water temperature in two species of Pacific salmon

Characterizing the cellular stress response (CSR) of species at ecologically relevant temperatures is useful for determining whether populations and species can successfully respond to current climatic extremes and future warming. In this study, populations of wild-caught adult pink (Oncorhynchus gorbuscha) and sockeye (Oncorhynchus nerka) salmon from the Fraser River, British Columbia, Canada, were experimentally treated to ecologically relevant 'cool' or 'warm' water temperatures to uncover common transcriptomic responses to elevated water temperature in non-lethally sampled gill tissue. We detected the differential expression of 49 microarray features (29 unique annotated genes and one gene with unknown function) associated with protein folding, protein synthesis, metabolism, oxidative stress and ion transport that were common between populations and species of Pacific salmon held at 19°C compared with fish held at a cooler temperature (13 or 14°C). There was higher mortality in fish held at 19°C, which suggests a possible relationship between a temperature-induced CSR and mortality in these species. Our results suggest that frequently encountered water temperatures ≥19°C, which are capable of inducing a common CSR across species and populations, may increase risk of upstream spawning migration failure for pink and sockeye salmon.

Migration timing of female kokanee salmon Oncorhynchus nerka: diel patterns and effects of maturation state

Diel patterns of migration and migration speed were compared between reproductive timing phenotypes in female kokanee salmon Oncorhynchus nerka. Females of varying degrees of reproductive maturation were captured on their migration route to the Meadow Creek Spawning Channel (British Columbia, Canada), were tagged with passive-integrated transponders (PIT tags) and were subsequently monitored with stationary receivers. Females showed crepuscular migration timing, with approximately equal detections at dawn and dusk. In particular, peaks of movement were associated with the appearance of the sun over the mountains in the east and the disappearance of the sun over the mountains in the west. Over 25 m, migration speed was 1·0 body lengths (measured as fork length; L(F)) s(-1) and did not depend on maturation state. Over 3 km, migration speed was much slower (0·2-0·3 L(F) s(-1)) than over the short distance, with less mature females migrating more slowly than more mature females. Less mature females appeared to be in less of a hurry to reach breeding areas compared with more mature females.

Environmental and lunar cues are predictive of the timing of river entry and spawning-site arrival in lake sturgeon Acipenser fulvescens

The associations were quantified between daily and interannual variation in the timing of a closed population of lake sturgeon Acipenser fulvescens migration and arrival at spawning sites with stream environmental and lunar covariates. Spawning data were gathered from 1262 fish in Black Lake, Michigan 2001 to 2008 and by video monitoring 2000 to 2002. Sex-specific variation in responses to external cues was also tested. Results showed that a greater number of individuals initiated migration from lake to riverine habitats at dawn and dusk relative to other times of the day. Current and lagged effects of water temperature and river discharge, and periods in the lunar cycle were important variables in models quantifying movements into the river and timing of adult arrival at spawning sites. Different suites of covariates were predictive of A. fulverscens responses during different periods of the spawning season. The timing of initiation of migration and spawning, and the importance of covariates to the timing of these events, did not differ between sexes. Stream flow and temperature covaried with other variables including day length and the lunar cycle. Anthropogenic disruption of relationships among variables may mean that environmental cues may no longer reliably convey information for Acipenseriformes and other migratory fishes.

Fitness and behavioral correlates of pre-stress and stress-induced plasma cortisol titers in pink salmon (Oncorhynchus gorbuscha) upon arrival at spawning grounds

Semelparous Pacific salmon (Onchorynchus spp.) serve as an excellent model for examining the relationships between life history, behavior and individual variation in glucocorticoid (GC) stress hormone levels because reproductive behaviors are highly variable between individuals and failure to reproduce results in zero fitness. Pink salmon (O. gorbuscha) were intercepted upon arrival at the spawning grounds across three time periods. Pre-stress and stress-induced plasma cortisol concentrations were assessed in relation to behavior, longevity and reproductive success. Results revealed differences between sexes and with arrival time. The study period marked a year of high reproductive success and only nine females (12% of sample) failed to spawn. Female pre-spawn mortalities were characterized by significantly elevated stress-induced cortisol concentrations and decreased longevity as well as pre-stress cortisol above the normal range in pink salmon from the study area. Interestingly, reproductive behaviors were only associated with pre-stress cortisol levels. For females, aggression and mate interaction time were reduced in individuals with elevated pre-stress cortisol concentrations. In males, a similar negative relationship between pre-stress cortisol concentration and mate interaction time was detected. The observed behavioral correlations are likely a factor of social status where dominant individuals, known to have higher reproductive success, are characterized by lower cortisol levels relative to subordinate conspecifics. Findings show both elevated pre-stress and stress-induced cortisol concentrations at arrival to the spawning grounds to be associated with reduced survival.

Influences of sex and activity level on physiological changes in individual adult sockeye salmon during rapid senescence

A noninvasive biopsy protocol was used to sample plasma and gill tissue in individual sockeye salmon (Oncorhynchus nerka) during the critical life stage associated with spawning-arrival at a spawning channel through senescence to death several days later. Our main objective was to characterize the physiological changes associated with rapid senescence in terms of the physiological stress/cortisol hypersecretion model and the energy exhaustion model. Salmon lived an average of 5 d in the spawning channel, during which time there were three major physiological trends that were independent of sexual status: a large increase in plasma indicators of stress and exercise (i.e., lactate and cortisol), a decrease in the major plasma ions (i.e., Cl(-) and Na(+)) and osmolality, and a decrease in gross somatic energy reserves. Contrary to a generalized stress response, plasma glucose decreased in approximately 2/3 of the fish after arrival, as opposed to increasing. Furthermore, plasma cortisol levels at spawning-ground arrival were not correlated with the degree of ionoregulatory changes during rapid senescence. One mechanism of mortality in some fish may involve the exhaustion of energy reserves, resulting in the inability to mobilize plasma glucose. Sex had a significant modulating effect on the degree of physiological change. Females exhibited a greater magnitude of change for gross somatic energy, osmolality, and plasma concentrations of Cl(-), Na(+), cortisol, testosterone, 11-ketotestosterone, 17,20beta-progesterone, and estradiol. The activity level of an individual on the spawning grounds appeared to influence the degree of some physiological changes during senescence. For example, males that received a greater frequency of attacks exhibited larger net decreases in plasma 11-ketotestosterone while on the spawning grounds. These results suggest that rapid senescence on spawning grounds is influenced by multiple physiological processes and perhaps behavior. This study provides some of the first data to look at sex differences in senescence in Pacific salmon.