INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

IN MAMMALS, in vivo or in vitro exposure of cardiac myocytes to hypoxia results in a 35-65% decrease in cell-surface -adrenoreceptor density (2, 3, 37, 45). This loss of -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 -adrenoreceptor density and/or binding affinity are also significantly affected by hypoxic exposure. Hypoxemia-mediated decreases in fish myocardial cell-surface -adrenoreceptor density (B_max) 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 -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 -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 -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 -adrenoreceptors in fish tissues (19, 41).

In the second study, we compared cell-surface -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 -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 -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.

View larger version (98K): [in this window] [in a new window]   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.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

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 (Pa_O2), arterial O₂ content (Ca_O2), 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 Pa_O2, Ca_O2, 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 -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 -adrenoreceptor and sp70 assays. Plasma for norepinephrine, epinephrine, and cortisol measurements was obtained by centrifuging blood at 10,000 g 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. Pa_O2 was measured using a thermostatted Radiometer O₂ electrode (E5046-0), and Ca_O2 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 -adrenoreceptor measurements.

Cardiac -adrenoreceptors. Cell-surface (sarcolemmal) B_max and K_D were determined using the tissue punch technique of Wilkinson et al. (48), as modified for fish hearts by Gamperl et al. (16). Briefly, -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 -adrenoreceptor ligand [³H]CGP-12177 (CGP), with or without the competitive antagonist timolol (10⁵ 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% for ³H. 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 (B_max, K_D) were determined using the method of Zivin and Waud (49). To allow for the expression of B_max 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-phosphate p-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 between 1) B_max, 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) B_max and K_D values in the compact and spongy myocardium of male and female chinook salmon. Paired t-tests were used to determine if the spongy and compact myocardium of chinook salmon had different values of -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.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Short-term hypoxia in rainbow trout. Blood oxygen content and partial pressure were significantly reduced by the experimental protocol (Fig. 2). Pa_O2 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 Ca_O2 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 Ca_O2 despite the severe hypoxia.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Effect of 6 h of hypoxia (water O2 content 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 at time 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.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Plasma levels of stress hormones in rainbow trout during 6 h of exposure to hypoxia (water O2 content 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 at time 0.

Cell-surface -adrenoreceptor B_max and K_D were both numerically higher in the hypoxic fish (B_max by 23%; K_D by 20%), but these differences were not statistically significant (P = 0.07 and P = 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).

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Effect of 6 h of hypoxic exposure on the density (Bmax) and binding affinity (KD) of rainbow trout myocardial cell-surface -adrenoreceptors. Hypoxic treatment had no significant effect (P > 0.05) on myocardial -adrenoreceptors.

View larger version (74K): [in this window] [in a new window]   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 B_max or the K_D values for -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 -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).

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Myocardial cell-surface -adrenoreceptor Bmax and KD in the compact and spongy myocardium of chinook salmon (Oncorhynchus tshawytscha). Sex had no effect on -adrenoreceptor binding characteristics. * Significant difference (P < 0.05) in -adrenoreceptor Bmax between compact and spongy myocardium.

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 -adrenoreceptors, whereas the K_D values were similar [chinook salmon (0.26 nM) and rainbow trout (0.21 nM); Fig. 7].

View larger version (21K): [in this window] [in a new window]   Fig. 7.   Comparison of -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 and KD of myocardial -adrenoreceptors was measured using the myocardial punch technique of Gamperl et al. (16). To obtain mean values for Bmax and KD in chinook salmon, data from all groups were pooled.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Hypoxia and cardiac -adrenoreceptors. A short-term (6 h) hypoxic signal that reduced Pa_O2 to 26% of its normal value failed to decrease trout cardiac -adrenoreceptors in vivo. This finding suggests that salmonid myocardial -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 -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 -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 -adrenoreceptors in avian and mammalian in vitro models (2 h: 29, 37) and 12 times longer than that required to significantly alter trout erythrocyte -adrenoreceptor density in vitro (30 min; 34). Furthermore, the reduction in Ca_O2 in our trout (32%) was more than twice that necessary to reduce -adrenoreceptor density in lambs exposed to 2 wk of hypoxemia (2, 3). Pa_O2 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 -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 -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 -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 -adrenoreceptors (3).

The inability of 6 h of hypoxia to downregulate trout ventricular -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 -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 -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 -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 -adrenoreceptor density is not diminished during moderate hypoxemia, because of the vital role that -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 -adrenoreceptors. In fish, catecholamine stimulation of red blood cells during hypoxia is regarded as important for maintaining oxygen delivery to the tissues. For example, -adrenoreceptor-mediated alkalization of red blood cells during hypoxemia increases hemoglobin affinity for oxygen (32), thereby protecting or enhancing blood oxygen transport.

-Adrenoreceptors in the spongy myocardium. We further suggest that the slightly elevated density of -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 -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 -adrenoreceptors density. It is also unlikely that differences in tension development between the two tissues can explain why the spongy myocardium had 14% more -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 -adrenoreceptors (e.g., see Ref. 3). These data, therefore, suggest that the spongy myocardium would have fewer -adrenoreceptors than the compact myocardium if tension development was a significant factor. One limitation of the method used to quantify cell-surface -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 -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 -adrenoreceptors. Despite the apparent resistance of trout ventricular -adrenoreceptors to hypoxia-mediated downregulation, we obtained convincing evidence for substantial plasticity in the density of myocardial cell-surface -adrenoreceptors in salmonids. The ventricle of wild chinook salmon had 2.8 times more -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 -adrenoreceptor B_max. 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 -adrenoreceptor density.

Reid and Perry (36) have shown that chronic (48 h) in vivo hypoxic exposure (Pa_O2 ~ 40 mmHg) does not affect the number or binding affinity of red blood cell-surface -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 -adrenoreceptors can be reduced by as much as 50% if plasma catecholamine levels approach 100 nM during moderate levels (Pw_O2 > 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 -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 -adrenoreceptor B_max in trout (PO₂ 15 mmHg) by 65% (34) and in carp (PO₂ 35 mmHg) to detectable levels (700 -adrenoreceptors per cell; 30). Thus the response of fish myocardial -adrenoreceptors to very short (<60 min) bouts of hypoxemia may be considerable.

Cortisol also influences fish red blood cell and hepatocyte cell-surface -adrenoreceptors. In vivo (10 days) exposure of red blood cells to elevated cortisol levels (~ 100 ng/ml) significantly increased the intracellular pool of -adrenoreceptors by 21%, and these intracellular -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 -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 -adrenoceptor density compared with rainbow trout.

Comparison with earlier studies on fish myocardial -adrenoceptors enable us to better define factors that have already been established as important signals. Gamperl et al. (16) showed that the B_max and K_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 -adrenoreceptor B_max or K_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 -adrenoreceptors suggests that gonadal steroid hormones do not modulate fish cardiac function through alterations in -adrenoreceptor B_max/K_D.

Keen et al. (23) identified an inverse relationship between myocardial -adrenoreceptor B_max and temperature for trout. Rainbow trout acclimated to 8°C had nearly three times (2.8×) the number of cell surface -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 -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 -adrenoreceptor B_max 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 -adrenoreceptor B_max 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 -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.

Perspectives

It is clear that a significant degree of plasticity exists in the density of myocardial cell-surface -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 -adrenergic system of fishes is largely unknown. Indeed, these factors may have contributed significantly to the 2.8-fold difference in myocardial -adrenoreceptor density between our chinook salmon and rainbow trout.