Regulatory, Integrative and Comparative Physiology

β-Receptors and stress protein 70 expression in hypoxic myocardium of rainbow trout and chinook salmon

A. K. Gamperl, M. M. Vijayan, C. Pereira, A. P. Farrell


We examined the in vivo effect of acute hypoxemia on myocardial cell-surface (sarcolemmal) β-adrenoreceptor density (Bmax) and binding affinity (K D) and on stress protein 70 (sp70) expression by exposing rainbow trout (Oncorhynchus mykiss; 2.1–2.7 kg) to hypoxic water (3 mg/l O2) at 15°C for 6 h. This degree of hypoxia was the minimum O2 level that these trout could tolerate without losing equilibrium and struggling violently. Hypoxic exposure reduced arterial PO2 ( PaO2 ) from 98 to 26 mmHg and arterial oxygen content ( CaO2 ) from 10.8 to 7.4 vol/100 vol, but did not elevate epinephrine and norepinephrine levels above 10 and 30 nM, respectively. Despite the substantial reduction in blood oxygen status, the Bmax andK D of myocardial cell-surface β-adrenoreceptors were unaffected by 6 h of hypoxic exposure. In addition, acute hypoxemia did not increase myocardial sp70 expression. The failure of short-term hypoxia to decrease trout myocardial β-adrenoreceptor density clearly contrasts with the established hypoxia-mediated downregulation shown for mammals. To further investigate the influence of low PO2 on salmonid myocardial β-adrenoreceptors, binding studies were performed on the spongy (continuously exposed to deoxygenated venous blood) and compact (perfused by oxygenated blood supplied by the coronary artery) myocardia of chinook salmon. The spongy myocardium has adapted to its microenvironment of continuous low PO2 by having 14% more cell-surface β-adrenoreceptors compared with the compact myocardium. There was no tissue-specific difference inK D and no evidence of sexual dimorphism in Bmax orK D. We conclude from our studies that the salmonid heart is well adapted for sustained performance under hypoxic conditions. We found that wild chinook salmon had 2.8× more cell-surface β-adrenoreceptors compared with hatchery-reared rainbow trout. This difference suggests a significant degree of plasticity exists for fish myocardial β-adrenoreceptors. The signals underlying such differences await further study, but are not likely to include moderate hypoxia and sexual dimorphism.

  • Oncorhynchus mykiss
  • Oncorhynchus tshawytscha
  • heart
  • hypoxia
  • temperature
  • catecholamines
  • cortisol

in mammals, in vivo or in vitro exposure of cardiac myocytes to hypoxia results in a 35–65% decrease in cell-surface β-adrenoreceptor density (2, 3, 37, 45). This loss of β-adrenoreceptors blunts the inotropic and chronotropic responsiveness of the myocardium to catecholamines. In nature, fish often encounter water in which oxygen content/partial pressure has been substantially reduced through biological oxygen demand, eutrophication, or anthropogenic activity. Because significant reductions in blood oxygen content and partial pressure are associated with environmental hypoxia (5), it is possible that fish cardiac β-adrenoreceptor density and/or binding affinity are also significantly affected by hypoxic exposure. Hypoxemia-mediated decreases in fish myocardial cell-surface β-adrenoreceptor density (Bmax) and/or binding affinity (K D) could have deleterious effects on the ability of circulating catecholamines to protect or enhance cardiac performance following exposure to additional physiological or environmental stressors (9, 17, 18). At present, there is no information on the effects of hypoxemia on cardiac β-adrenoreceptors in fish.

In all organisms, exposure to unfavorable conditions causes the increased synthesis of a class of proteins termed stress or heat shock proteins (38, 47), which protect the cells from subsequent or more severe stressors. For example, in mammalian cardiac tissue brief ischemia increases stress protein levels approximately threefold (25, 28), and myocytes with elevated stress protein 70 (sp70) expression show significantly improved survival when exposed to chronic hypoxia ( PO2 15–38 mmHg; 20). Because stress proteins are thought to play a vital role in cellular protection, evaluation of sp70 expression in trout myocardium following short-term hypoxemia may provide additional insights into the hypoxia tolerance of the salmonid heart. Stress proteins have not been measured in fish cardiac tissues.

To investigate whether hypoxemia directly effects myocardial β-adrenoreceptor density/binding affinity and sp70 expression in salmonids, we used two experimental approaches. In the first study, mature rainbow trout were exposed to either normoxic (9 mg/l O2) or hypoxic (3 mg/l O2) water for 6 h, and hearts were subsequently assayed for cell surface β-adrenoreceptor density/binding affinity and sp70 expression. This level of hypoxia was a significant insult to the fish (based on reductions in blood O2 status), but was not so severe as to provoke large increases in circulating catecholamines. In preliminary experiments, it was determined that 3 mg/l O2 was the minimum water oxygen level that our fish could tolerate without losing equilibrium and struggling violently. These behaviors were avoided because they are associated with large increases in circulating catecholamines, and high levels of catecholamines alone can downregulate β-adrenoreceptors in fish tissues (19, 41).

In the second study, we compared cell-surface β-adrenoreceptor density and binding affinity in the compact and spongy myocardia of chinook salmon (Fig.1). In the salmonid heart there are two types of myocardia: a spongy myocardium, which is exposed continuously to a hypoxic microenvironment because it is perfused by the venous blood that percolates through its trabecular sinusoids, and a compact myocardium, which is perfused with highly oxygenated arterial blood supplied by the coronary artery. This morphological comparison would reveal to what degree cardiac β-adrenoreceptors had adapted to the hypoxic microenvironment faced by the spongy myocardium. Previous studies on fish hearts have shown that the spongy myocardium contains mitochondria with higher oxidative enzyme activities and other specializations, compared with the compact myocardium (see Ref. 43, for review). In addition, Tota (43) suggests that these differences in mitochondrial physiology are adaptive for a microenvironment characterized by a low and changing PO2 . Based on data from in vivo and in vitro mammalian models of hypoxia, one would predict that the spongy myocardium of fishes would have a reduced β-adrenoreceptor density/binding affinity (2, 3, 37, 45), unless it had adapted to its permanently hypoxic environment (1). To perform this morphological comparison, very large fish (>5 kg) were needed to ensure that a sufficient number of discrete tissue punches (2 mm diam) were obtained from both myocardial tissue subtypes in each fish. Spawning chinook salmon were used because rainbow trout were unavailable in the required size range.

Fig. 1.

Medial section through the alcohol-preserved ventricle of a 10-kg chinook salmon (Oncorhynchus tshawytscha). Note location of and the proportion of ventricle occupied by the compact (C) and spongy (S) myocardium. Arrows indicate coronary arteries. Original magnification, ×5.


Short-term hypoxia. Mature rainbow trout (Oncorhynchus mykiss, 2.1–2.7 kg) were anesthetized in buffered MS-222 (0.1 g/l), fitted with dorsal aortic cannulas (PE-50; 37), and placed into 50-liter round tubes (1 m long by 25 cm diam) supplied with aerated dechlorinated freshwater (14–15°C) at 8 l/min. After 24 h of recovery, trout were continuously exposed to either normoxic water (∼20 kPa, 9 mg/l O2;n = 9) or hypoxic water (∼6 kPa, 3 mg/l O2;n = 9) for 6 h. In the hypoxic fish, the oxygen content of the inflowing water was slowly reduced to the desired level over a 30-min period. Water oxygen content was manipulated by bubbling a controlled mixture of air and nitrogen through a gas exchange column (90 cm in length, 7.5 cm diam) filled with Tri-Packs. Water oxygen content (mg/l) was monitored with a YSI oxygen meter (model 50) and was converted to partial pressure on the basis of calibrations obtained with a thermostatted Radiometer O2 Electrode (model E5046–0).

Before the onset of hypoxia (time 0) and 0.5, 1, 2, 4, and 6 h after the desired level of hypoxia was reached, 0.8-ml blood samples were taken for analysis of arterial PO2 ( PaO2 ), arterial O2 content ( CaO2 ), hematocrit (Hct), norepinephrine, epinephrine, and cortisol. Following the removal of each blood sample, an equal volume of saline was injected into the fish to restore blood volume. Measurements of PaO2 , CaO2 , and Hct allowed for determination of the degree of hypoxic insult. Plasma levels of norepinephrine, epinephrine, and cortisol were monitored because these hormones have all been shown to affect β-adrenoreceptors in fish (33).

At the end of 6 h, normoxic and hypoxic trout were killed by a blow to the head. The heart was quickly (<30 s) removed, allowed to beat for 30–60 s in cold (0–2°C)N-tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid (TES)-buffered teleost saline (23) to remove erythrocytes from the ventricular lumen, and quickly frozen in liquid nitrogen. Hearts were stored at −70°C prior to β-adrenoreceptor and sp70 assays. Plasma for norepinephrine, epinephrine, and cortisol measurements was obtained by centrifuging blood at 10,000g for 30 s immediately on collection. Before the plasma for catecholamine measurements was frozen in liquid nitrogen, an anti-oxidant solution (10% by volume; 0.1 mM sodium metabisulfite, 0.3% EDTA, 0.3% ascorbic acid) was added. Plasma samples were also stored at −70° C prior to catecholamine and cortisol analysis.

Hematological parameters. PaO2 was measured using a thermostatted Radiometer O2 electrode (E5046–0), and CaO2 was measured using a Oxycon blood oxygen content analyzer (Cameron Instruments). Plasma cortisol was measured on 25-μl samples using a commercially available radioimmunoassay kit (Intermedico Diagnostics, Ontario). Plasma norepinephrine and epinephrine were assayed on alumina-extracted samples using a Waters (Waters Chromatography Division of Millipore, Mississauga, Canada) high-performance liquid chromatography system (model 510 delivery system, model 460 electrochemical detector and data module/integrator, reverse-phase plasma catecholamine column) and a commercially prepared mobile phase (Chromsystems Instruments and Chemicals, Germany). 3,4-Dihydroxybenzylamine was used as an internal standard for all plasma samples and catecholamine standards. Hct was determined on 20-μl blood samples following 2 min of centrifugation at 10,000 g.

Compact versus spongy myocardium.Large male (8.8 ± 0.6 kg, n = 8) and female (9.2 ± 0.4 kg, n = 8) chinook salmon (Oncorhynchus tshawytscha) were netted from a 13°C spawning channel at Robertson Creek Hatchery (Dept. of Fisheries and Oceans Canada, Port Alberni, Canada) in October, 1995. The salmon were quickly killed by a blow to the head, and the hearts were processed as above. To ensure that the fish were in good condition, only individuals able to maintain themselves in the water current (∼50 cm/s) were selected for β-adrenoreceptor measurements.

Cardiac β-adrenoreceptors. Cell-surface (sarcolemmal) Bmax andK D were determined using the tissue punch technique of Wilkinson et al. (48), as modified for fish hearts by Gamperl et al. (16). Briefly, β-adrenergic binding was accomplished by incubating 2 mm diameter × 350 μm thick (∼1 mg) myocardial tissue punches in varying concentrations (0.05–3.0 nM) of the hydrophilic β-adrenoreceptor ligand [3H]CGP-12177 (CGP), with or without the competitive antagonist timolol (10−5 M) for 2 h. Following removal of the tissue punches from the incubation medium and two washes in fresh TES-buffered saline, tissue punches were placed into scintillation vials, and radioactivity was quantified in a scintillation counter at an efficiency of ∼42% for3H. Radioactivity [counts per minute (cpm)] measured in punches incubated with [3H]CGP + timolol was subtracted from cpm for punches incubated with [3H]CGP alone to yield specific binding. Saturation binding curves were analyzed, and binding parameters (Bmax,K D) were determined using the method of Zivin and Waud (49). To allow for the expression of Bmax as femtomoles per milligram protein, the protein content of representative punches was measured using the Bio-Rad protein assay. In the rainbow trout, myocardial tissue punches consisted of approximately equal proportions of compact and spongy myocardium. In the chinook salmon, tissue punches were taken from both the inner spongy and the outer compact myocardium of each fish. These punches were then incubated separately.

Western blot for sp70. Frozen samples of trout ventricle (100–200 mg), containing both compact and spongy myocardium, were homogenized in 50 mM tris(hydroxymethyl)aminomethane buffer (pH 7.5) using a tissue tearer (model 985–370, Biospec Products). The homogenate was measured for protein content using the bicinchoninic acid method (40), while an aliquot of the homogenate was mixed with an equal volume of 2× Laemmli buffer (26), boiled for 8 min (100°C), cooled, and stored at −70°C prior to sp70 determination on a Western blot according to Forsyth et al. (11). Briefly, myocardial proteins were separated on a sodium dodecyl sulfate-polyacrylamide gel electrophoresis (40 μg protein/lane) using the discontinuous buffer system of Laemmli (26). The electrophoresed proteins were transferred to nitrocellulose sheets (Bio-Rad, 0.2 μm pore size), blocked with skim milk, and incubated with primary and secondary antibodies (11). The primary antibody was a polyclonal rabbit antibody raised against rainbow trout sp70 (11), whereas the secondary antibody was alkaline phosphatase conjugated with anti-rabbit immunoglobulin G (GIBCO-BRL). Distinct sp70 bands were detected using a chromogenic substrate (330 mg/l nitroblue tetrazolium, 167 mg/l 5-bromo-4-chloro-3-indolyl-phosphatep-toluidine salt) in alkaline buffer according to Blake et al. (4). The intensity of the resultant bands was determined on a Scanjet IIP using SigmaGel software (Jandel Scientific). Band intensity was normalized in relationship to the maximum intensity of the bands after background subtraction and was expressed as relative units.

Statistical analyses. Hematological variables in the hypoxic/normoxic trout were compared using a two-way repeated- measures analysis of variance (ANOVA). When time-dependent effects were identified, contrasts were used to determine which time points were significantly different from control values (time 0). One-way ANOVAs were used to identify significant differences between1) Bmax,K D, and sp70 levels in hypoxic and normoxic trout;2) morphometric variables [body mass; ventricle mass; relative ventricular mass (RVM)] in male and female chinook salmon;3) morphometric variables in chinook salmon and rainbow trout; and 4) Bmax andK D values in the compact and spongy myocardium of male and female chinook salmon. Pairedt-tests were used to determine if the spongy and compact myocardium of chinook salmon had different values of β-adrenoreceptor density and binding affinity. All statistical analyses were performed using the SAS statistical package (SAS Institute). P < 0.05 was used as the level of significance. Values in Table 1, Figs. 2-4, 6, and 7, and throughout the text are expressed as means ± SE.


Short-term hypoxia in rainbow trout. Blood oxygen content and partial pressure were significantly reduced by the experimental protocol (Fig. 2). PaO2 decreased from 100 to ∼26 mmHg within 30 min after the onset of hypoxia and remained at this level for the remainder of the experiment. Blood CaO2 decreased by 32% (10.8 to 7.4 vol/100 vol) during hypoxia. Hct increased significantly by 20% (from 32.6 to 38.8%) after 30 min and remained at elevated levels for the duration of hypoxia exposure. The increase in Hct resulted from a release of erythrocytes from the spleen and/or erythrocyte swelling, and these mechanisms would have helped maintain CaO2 despite the severe hypoxia.

Fig. 2.

Effect of 6 h of hypoxia (water O2content 3 mg/l) on hematological variables in rainbow trout (Oncorhynchus mykiss). * Significant difference (P < 0.05) between normoxic and hypoxic trout (n = 9). + Within group differences in values from those measured attime 0.

Control plasma levels of norepinephrine and epinephrine in both groups were ∼6 and 3 nM, respectively. In hypoxic fish, norepinephrine and epinephrine were significantly elevated, compared with normoxic fish (Fig. 3). For both catecholamines, maximum levels [norepinephrine (30 nM) and epinephrine (10 nM)] were reached after 30 min of exposure to 3 mg/l O2 water. Catecholamine levels remained modestly elevated during the first 4 h of hypoxic exposure, but returned to control levels by 6 h. Control levels of cortisol averaged 60 and 75 ng/ml for the normoxic and hypoxic fish, respectively (Fig. 3). Hypoxia had no significant effect on plasma cortisol levels. However, repeated blood sampling caused an increase in plasma cortisol of ∼60 ng/ml during the 6-h experiment.

Fig. 3.

Plasma levels of stress hormones in rainbow trout during 6 h of exposure to hypoxia (water O2content 3 mg/l) or normoxia (water O2 content 9 mg/l). * Significant difference (P < 0.05) between normoxic and hypoxic trout (n = 9). + Within group differences in values from those measured attime 0.

Cell-surface β-adrenoreceptor Bmax andK D were both numerically higher in the hypoxic fish (Bmax by 23%;K D by 20%), but these differences were not statistically significant (P = 0.07 andP = 0.27, respectively; Fig.4). Constitutive levels of sp70 were found in the myocardium of normoxic and hypoxic trout. However, no significant difference in sp70 expression between hypoxic (318.6 ± 36.7 relative units) and normoxic (290.7 ± 65.2 relative units) trout was evident (Fig. 5).

Fig. 4.

Effect of 6 h of hypoxic exposure on the density (Bmax) and binding affinity (K D) of rainbow trout myocardial cell-surface β-adrenoreceptors. Hypoxic treatment had no significant effect (P> 0.05) on myocardial β-adrenoreceptors.

Fig. 5.

Representative Western blot of sp70 expression in myocardium of rainbow trout (Oncorhynchus mykiss) exposed to normoxia (control) or hypoxia (water O2 content 3 mg/l) for 6 h. Myocardial sp70 expression (determined using SigmaGel software) in hypoxic and normoxic fish was not significantly different (n = 8;P > 0.05).

Compact versus spongy myocardium in chinook salmon. There were no sex differences in either the Bmax or theK D values for β-adrenoreceptors in the compact or spongy myocardium (P = 0.43; Fig.6). Therefore, the data for male and female chinook salmon were pooled for comparisons. Cell-surface β-adrenoreceptors were significantly greater (14%) in the spongy myocardium (P = 0.02; Fig. 6).K D values for the compact and spongy myocardium were not significantly different (P = 0.64). Although male and female chinook salmon had a similar body mass, the male salmon had significantly higher values for ventricle mass (17%) and RVM (24%) than the female salmon (Table 1).

Fig. 6.

Myocardial cell-surface β-adrenoreceptor Bmax andK D in the compact and spongy myocardium of chinook salmon (Oncorhynchus tshawytscha). Sex had no effect on β-adrenoreceptor binding characteristics. * Significant difference (P < 0.05) in β-adrenoreceptor Bmax between compact and spongy myocardium.

View this table:
Table 1.

Morphological variables for spawning male and female chinook salmon and for mature hatchery-reared rainbow trout

Rainbow trout versus chinook salmon.RVM was ∼1.8× greater in the wild chinook salmon (0.161%) compared with hatchery-reared rainbow trout (0.091%) (Table 1). The chinook salmon had 2.8× the number of cell-surface β-adrenoreceptors, whereas theK D values were similar [chinook salmon (0.26 nM) and rainbow trout (0.21 nM); Fig. 7].

Fig. 7.

Comparison of β-adrenoreceptor binding characteristics between normoxic rainbow trout (Oncorhynchus mykiss; n = 8) acclimated to 8°C (16) and 14°C (present study;n = 9), and chinook salmon (Oncorhynchus tshawytscha) sampled from Robertson Creek (present study, 13°C,n = 16). In all groups, the Bmax andK D of myocardial β-adrenoreceptors was measured using the myocardial punch technique of Gamperl et al. (16). To obtain mean values for Bmax andK D in chinook salmon, data from all groups were pooled.


Hypoxia and cardiac β-adrenoreceptors. A short-term (6 h) hypoxic signal that reduced PaO2 to 26% of its normal value failed to decrease trout cardiac β-adrenoreceptors in vivo. This finding suggests that salmonid myocardial β-adrenoreceptors are resistant to hypoxia-mediated downregulation in situations where catecholamines are not greatly elevated. This result contrasts with expectations based on most avian and mammalian models of myocardial hypoxia. For example, 2 h of hypoxia ( PO2 ∼23 mmHg) significantly reduced cell-surface β-adrenoreceptors in cultured rat (−29%; 37) and chick (−62%; 29) ventricular myocytes. In addition, both chronic (5 wk) hypobaric hypoxia ( PO2 95 mmHg) in rats (45) and chronic (2 wk) hypoxemia in young lambs (2,3) caused a 45–55% decrease in ventricular cell-surface β-adrenoreceptors.

The reasons underlying the difference between trout and avian/mammalian hearts are unclear. However, we are confident that both the duration and level of hypoxia used in our acute experiment were sufficiently severe to provoke a cellular response, if it existed. Our hypoxic period was three times longer than that necessary to downregulate myocardial β-adrenoreceptors in avian and mammalian in vitro models (2 h: 29, 37) and 12 times longer than that required to significantly alter trout erythrocyte β-adrenoreceptor density in vitro (30 min; 34). Furthermore, the reduction in CaO2 in our trout (32%) was more than twice that necessary to reduce β-adrenoreceptor density in lambs exposed to 2 wk of hypoxemia (2,3). PaO2 in our fish was ∼26 mmHg, and the PO2 of the venous blood that perfuses the spongy myocardium (which represents 70% of ventricular mass) is estimated to have been ∼8 mmHg, based on previous measurements of venous PO2 in rainbow trout exposed to aquatic hypoxia (22, 46). This level of myocardial hypoxia is comparable to, or lower than, the levels of PO2 known to cause downregulation of β-adrenoreceptors in cultured chick and rat cardiac myocytes (29, 37). In addition, data from saline-perfused isolated hearts (with a perfused coronary circulation) (10) suggest that in vivo myocardial function is compromised in trout if venous PO2 falls to 10 mmHg.

The resistance of trout ventricular β-adrenoreceptors to hypoxia-mediated downregulation was probably not related to the relatively minor increases in circulating catecholamine levels that accompanied the first 4 h of hypoxemia in this study. Two lines of evidence support this conclusion: 1) acute hypoxia significantly reduced cell-surface β-adrenoreceptors in cultured rat (37) and chick (29) ventricular myocytes in the absence of adrenergic stimulation; and 2) the levels of circulating catecholamines in our hypoxic trout (10–30 nM) were approximately five times those measured in lambs where mild chronic hypoxemia (2 wk) significantly downregulated β-adrenoreceptors (3).

The inability of 6 h of hypoxia to downregulate trout ventricular β-adrenoreceptors may be related, instead, to the fact that rainbow trout are exposed routinely to aquatic hypoxia and there is a need for this species to retain its main mechanism for cardiac stimulation under such conditions. Circulating catecholamines ameliorate or diminish the negative inotropic effects of hypoxia/acidosis on fish cardiac performance (9, 17, 18). Clearly, the maintainance of myocardial β-adrenoreceptor density during short-term hypoxia may be important in this regard, whereas a hypoxic downregulation of these receptors would be counterproductive. In addition, fish may further protect their myocardial tissue through the reflex bradycardia that accompanies environmental hypoxia (13, 22). Indeed, this reflex bradycardia may offset some of the energetic costs associated with myocardial β-adrenergic-mediated protection/stimulation of cardiac performance during hypoxia. In contrast, in mammals and birds myocardial hypoxemia can be regarded as an unusual situation, and the downregulation of β-adrenoreceptors may represent a “safeguard” against the stimulation of myocardial activity at a time when there is insufficient energy production by oxidative processes.

The hypothesis that fish myocardial cell-surface β-adrenoreceptor density is not diminished during moderate hypoxemia, because of the vital role that β-adrenoreceptors play in maintaining cellular performance, is not without experimental support. Reid and Perry (36) have shown that in vivo exposure of red blood cells to chronic (48 h) hypoxia (arterial PO2 ∼ 40 mmHg), in the absence of large increases in plasma catecholamines, does not affect the number or binding affinity of cell-surface β-adrenoreceptors. In fish, catecholamine stimulation of red blood cells during hypoxia is regarded as important for maintaining oxygen delivery to the tissues. For example, β-adrenoreceptor-mediated alkalization of red blood cells during hypoxemia increases hemoglobin affinity for oxygen (32), thereby protecting or enhancing blood oxygen transport.

β-Adrenoreceptors in the spongy myocardium. We further suggest that the slightly elevated density of β-adrenoreceptors in the spongy myocardium, compared with the compact myocardium, reflects the adaptational strategy of a cardiac tissue that is continously exposed to a hypoxic microenvironment. This speculation is consistent with Tota (43) who suggested that elevated mitochondrial enzyme activities in the spongy myocardium of fishes are adaptive for a microenvironment characterized by a low and changing PO2 . Furthermore, this interpretation is supported by data obtained from humans who have resided at high altitude (3,600–4,000 m) for generations. These individuals are continuously exposed to hypobaric hypoxia, but they show no adrenergic desensitization and have normal chronotropic responsiveness to isoproterenol (i.e., similar to sea level natives) (1).

We have difficulty ascribing the similarity of myocardial β-adrenoreceptor density in spongy and compact myocardia to factors that may have masked a large hypoxia-mediated downregulation. Differences in the levels of circulating catecholamines to which the two tissues were exposed are unlikely to be a factor. Although the gill is an important site of catecholamine removal from the blood (31), catecholamine levels in postbranchial blood (which perfuses the compact myocardium) are similar to those in the prebranchial (venous) blood (24). Furthermore, elevated levels of catecholamines in the venous blood, as might be expected in these wild fish, are predicted to decrease β-adrenoreceptors density. It is also unlikely that differences in tension development between the two tissues can explain why the spongy myocardium had 14% more β-adrenoreceptors. According to Tota (43), the compact myocardium develops greater tension during contraction than the spongy myocardium. Also, the mammalian left ventricle, which develops significantly greater pressure than the right ventricle, often has significantly more β-adrenoreceptors (e.g., see Ref. 3). These data, therefore, suggest that the spongy myocardium would have fewer β-adrenoreceptors than the compact myocardium if tension development was a significant factor. One limitation of the method used to quantify cell-surface β-adrenoreceptors is that it is not sensitive to differences in myocyte size, and therefore changes in surface area/volume relationships. However, because the length and cross-sectional area of myocytes in the spongy and compact myocardium of mature rainbow trout are not different (Rodnick, unpublished data), the higher density of β-adrenoreceptors measured in the spongy myocardium was unlikely to be influenced by differences in cell surface/volume relationships between the two tissues.

Plasticity of fish myocardial β-adrenoreceptors. Despite the apparent resistance of trout ventricular β-adrenoreceptors to hypoxia-mediated downregulation, we obtained convincing evidence for substantial plasticity in the density of myocardial cell-surface β-adrenoreceptors in salmonids. The ventricle of wild chinook salmon had 2.8 times more β-adrenoreceptors compared with hatchery-reared rainbow trout (Fig. 7). In addition, since ventricular mass was seven times greater in the chinook salmon and myocyte size increases substantially with ventricular mass (8a), it is probable that we underestimated the difference in cell-surface β-adrenoreceptor Bmax. Although an explanation for this novel finding awaits further study, data from fish hearts and from other fish tissues make it possible to suggest which signals are likely to influence fish myocardial β-adrenoreceptor density.

Reid and Perry (36) have shown that chronic (48 h) in vivo hypoxic exposure ( PaO2 ∼ 40 mmHg) does not affect the number or binding affinity of red blood cell-surface β-adrenoreceptors, provided there is no large increase in plasma catecholamine levels. This observation is entirely consistent with our results for the hypoxic trout myocardium. In contrast, experiments with trout red blood cells also show that β-adrenoreceptors can be reduced by as much as 50% if plasma catecholamine levels approach 100 nM during moderate levels ( PwO2 > 50 mmHg) of chronic hypoxia or as a result of other stressors (19, 41). Together, these data suggest that elevated plasma catecholamines, but not hypoxemia, are likely to affect fish myocardial β-adrenoreceptors. However, in vitro studies with fish red blood cells may reveal a further layer of complexity. Short-term (30–60 min) hypoxia increases cell-surface β-adrenoreceptor Bmax in trout ( PO2 15 mmHg) by 65% (34) and in carp ( PO2 35 mmHg) to detectable levels (700 β-adrenoreceptors per cell; 30). Thus the response of fish myocardial β-adrenoreceptors to very short (<60 min) bouts of hypoxemia may be considerable.

Cortisol also influences fish red blood cell and hepatocyte cell-surface β-adrenoreceptors. In vivo (10 days) exposure of red blood cells to elevated cortisol levels (∼ 100 ng/ml) significantly increased the intracellular pool of β-adrenoreceptors by 21%, and these intracellular β-adrenoreceptors were rapidly (<30 min) recruited to the cell surface during in vitro hypoxic exposure (33). Trout hepatocytes appear to be even more sensitive to cortisol since chronic (11–12 days) in vivo exposure to 130 ng/ml cortisol directly increased cell-surface β-adrenoreceptor density by 250% (35). In view of the in vitro and in vivo sensitivity of fish tissues to cortisol, the elevated cortisol levels (60–120 ng/ml, Fig. 3) in our experiments are a potential concern in terms of masking a hypoxia-mediated downregulation. However, it may be difficult to perform in vivo hypoxia studies and keep cortisol levels closer to the normoxic level of 10 ng/ml for unstressed fish (15). Further studies are needed to resolve this issue and it may be necessary to turn to in vitro studies as have been used for fish red blood cells and mammalian/avian myocytes. Chronically elevated cortisol levels in the wild chinook salmon may be a contributing factor to the heightened myocardial β-adrenoceptor density compared with rainbow trout.

Comparison with earlier studies on fish myocardial β-adrenoceptors enable us to better define factors that have already been established as important signals. Gamperl et al. (16) showed that the Bmax andK D values for [3H]CGP binding to ventricular punches were not different between mature male and female rainbow trout. In the present study, although RVM in spawning chinook salmon showed sexual dimorphism, no such effect was evident for myocardial β-adrenoreceptor Bmax orK D (Table 1, Fig. 6). The 24% greater RVM in male chinook salmon is consistent with the 19–35% sexual dimorphism in RVM shown for various populations of rainbow trout (20). Although an improved cardiac performance is associated with the larger ventricular mass (12), the lack of a sexual dimorphism for cardiac β-adrenoreceptors suggests that gonadal steroid hormones do not modulate fish cardiac function through alterations in β-adrenoreceptor Bmax/K D.

Keen et al. (23) identified an inverse relationship between myocardial β-adrenoreceptor Bmax and temperature for trout. Rainbow trout acclimated to 8°C had nearly three times (2.8×) the number of cell surface β-adrenoreceptors as individuals acclimated to 18°C. The influence of temperature is further highlighted by comparing our data with those of Gamperl et al. (16) (Fig. 7). The density of myocardial β-adrenergic receptors was significantly higher (70%) in trout at 8°C (16) than in trout at 14°C (present study). Although different techniques were used to assess receptor densities (homogenates vs. tissue punches), both approaches suggest that cell-surface β-adrenoreceptor Bmax in rainbow trout increases by 11% with each 1°C decrease in water temperature. Such a significant temperature dependency could greatly diminish the ability of catecholamines to stimulate cardiac performance at high water temperatures. Indeed, Keen et al. (23) showed that the 180% reduction in cell-surface β-adrenoreceptor Bmax between 8 and 18°C trout was concomitant with a decreased sensitivity to catecholamines. In addition, Farrell et al. (8) have shown that in situ maximum cardiac performance at high acclimation temperatures is not improved by increasing the perfusate epinephrine concentration from 30 to 200 nM.

sp70 Expression. The presence of sp70 in the myocardium of normoxic fish is not surprising since most fish tissues have constitutive (“prestress”) levels of stress proteins [brain, gill, and skeletal muscle (7); liver and kidney (11)], and the production of stress proteins is ubiquitous among organisms (47). This is the first study to either measure sp70 expression in the fish myocardium or to examine whether stress protein synthesis is enhanced in fish hearts following a hypoxic insult. Short-term hypoxic exposure did not induce myocardial sp70 expression, an observation that contrasts with findings for the mammalian myocardium. Knowlton et al. (25) showed that brief (5 min and 4 × 5 min) myocardial ischemia increased sp70 mRNA levels by 1.8× and 2.3×, respectively, and that increases in sp70 expression were evident within 2 h after reperfusion. Furthermore, sp70 expression improves the survival of rat cardiac myocytes on exposure to 20 h of hypoxia (21), and sp70 expression may contribute to the myocardial protection seen in models of brief ischemia (27,28). Similar levels of sp70 were not measured in hypoxic and normoxic trout because the primary antibody failed to detect inducible forms of sp70. The trout-specific primary antibody used in this study has been used successfully by other authors to measure heightened sp70 expression in salmonid cells (6, 11, 44). As with the β-adrenoreceptors, the resistance of myocardial sp70 expression to short-term hypoxemia may be related to the hypoxia tolerance of fish tissues. For example, Currie and Tufts (6) showed that while trout red blood cells increased the synthesis of sp70 by 2.5× within 2 h of exposure to a 15°C increase in temperature, no increase in sp70 expression was associated with a similar period of anoxic exposure. The possibility that myocardial sp70 expression is increased with exposure to chronic (days or wk) or more severe hypoxia or during recovery from hypoxia should be investigated.


Both the failure of short-term hypoxia to affect trout myocardial cell-surface β-adrenoreceptor density or sp70 expression and the 14% greater cell-surface β-adrenoreceptor density in the spongy myocardium of chinook salmon suggest that the salmonid heart is well adapted for sustained performance during moderate hypoxemia. Our finding that short-term hypoxia does not downregulate myocardial β-adrenoreceptors clearly contrasts with results from mammalian models. In mammals, experimentally induced myocardial hypoxia/ischemia results in the loss of cell-surface β-adrenoreceptors, the induction of stress protein expression, and diminished cardiac function. Our data suggest that the β-adrenergic system in the spongy myocardium of salmonids has adapted to its microenvironment of continuous low PO2 and this idea is indirectly supported by data from humans residing at high altitude. Further studies are needed to directly test whether the maintenance or upregulation of myocardial β-adrenoreceptors has adaptive value in vertebrates that are continuously exposed to hypoxia.

It is clear that a significant degree of plasticity exists in the density of myocardial cell-surface β-adrenoreceptors in salmonids. Although we have identified numerous factors that may be important in mediating such differences (e.g., cortisol, catecholamines, temperature), the influence of rearing environment (hatchery vs. wild), life history, and species on the β-adrenergic system of fishes is largely unknown. Indeed, these factors may have contributed significantly to the 2.8-fold difference in myocardial β-adrenoreceptor density between our chinook salmon and rainbow trout.


We thank Bill Bennett, Glenn Rasmussen, and the staff of the Robertson Creek Hatchery for assistance in collecting chinook salmon hearts. We are grateful to Drs. Brian McKeown, Leah Bendell-Young, Chris Kennedy, and Dave Randall for the use of equipment and to Dr. George Iwama for providing the primary antibody for trout sp70. The assistance of Nia Whitely in the analysis of plasma catecholamines was much appreciated. We thank the reviewers of this manuscript for constructive comments.


  • Address for reprint requests: A. K. Gamperl, Dept. of Biological Sciences, Simon Fraser Univ., Burnaby, British Columbia, Canada V5A 1S6.

  • This research was supported through a Natural Sciences and Engineering Research Council of Canada (NSERC) operating grant to A. P. Farrell and a British Columbia Science Council Grant to M. M. Vijayan. A. K. Gamperl was the recipient of an NSERC postdoctoral fellowship.

  • Received 27 June 1997; accepted in final form 14 October 1997.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 8a.
  10. 9.
  11. 10.
  12. 11.
  13. 12.
  14. 13.
  15. 14.
  16. 15.
  17. 16.
  18. 17.
  19. 18.
  20. 19.
  21. 20.
  22. 21.
  23. 22.
  24. 23.
  25. 24.
  26. 25.
  27. 26.
  28. 27.
  29. 28.
  30. 29.
  31. 30.
  32. 31.
  33. 32.
  34. 33.
  35. 34.
  36. 35.
  37. 36.
  38. 37.
  39. 38.
  40. 39.
  41. 40.
  42. 41.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
View Abstract