Ca²⁺-ATPase ACTIVITY AND Ca²⁺ UPTAKE BY SARCOPLASMIC RETICULUM IN FISH HEART: EFFECTS OF THERMAL ACCLIMATION

Sarcoplasmic reticulum (SR) participates in the contraction and relaxation of cardiac muscle by a Ca²⁺-induced Ca²⁺ release mechanism and an ATP-dependent resequestration process, respectively. Ryanodine-sensitive Ca²⁺ efflux channels are confined to the junctional and corbular SR, while thapsigargin- and cyclopiazonic-acid-inhibitable Ca²⁺-pumping ATPase molecules are localized both in junctional and in nonjunctional SR (Feher et al. 1988; Jorgensen et al. 1993). The significance of SR in the contraction–relaxation cycle of cardiac muscle varies greatly among different vertebrate classes, among different species within the same phylogenetic group and during the ontogenetic development of an individual. The SR is well developed in mammalian and avian cardiac myocytes, where it is thought to be the major source and sink of the activator Ca²⁺. In ectothermic vertebrates, such as lizards, frogs and fishes, and in newborn mammals, cardiac SR is relatively sparse, and contraction and relaxation are more directly dependent on sarcolemmal Ca²⁺ fluxes (Fabiato, 1982; Bossen and Sommer, 1984; Vornanen, 1996a,b).

Electron microscopic findings suggest that there are marked differences in the amount of cardiac SR among fish species (Santer, 1985), although rigorous morphometric analyses of SR in teleost species are still lacking. Species-specific differences are also evident in the inhibition of contraction by ryanodine: in crucian carp heart, ryanodine has no effect on ventricular contraction (Vornanen 1996a); in rainbow trout ventricle, ryanodine slightly reduces the force of contraction, especially at high experimental temperatures and at low contraction frequencies (Keen et al. 1994); in the atrium of tuna heart, ryanodine exerts a clear negative inotropic effect (Keen et al. 1992). These findings suggest that the more active fish have a higher Ca²⁺-handling capacity in the cardiac SR than the less active species. Furthermore, recent studies suggest that thermal acclimation may alter the excitation–contraction coupling process of the fish heart. Morphometric analyses of perch heart suggest that SR is better developed in cold-acclimated than in warm-acclimated fish (Bowler and Tirri, 1990), and functional studies in thermally acclimated trout have shown that contraction is more sensitive to ryanodine inhibition in cold-acclimated than in warm-acclimated animals (Keen et al. 1994). However, to our knowledge, the effects of thermal acclimation on the activity of Ca²⁺-ATPase and the Ca²⁺ uptake velocity of cardiac SR have not been directly determined in the fish heart. Therefore, the aim of the present study was to measure the Ca²⁺-ATPase activity and Ca²⁺ uptake velocity of cardiac SR in two fish species, crucian carp and rainbow trout, species that differ in activity pattern and in acclimation response to changing ambient temperature. Furthermore, the function of fish cardiac SR was compared with that of newborn and adult rats to give a more comprehensive picture of the Ca²⁺-handling capacity of fish heart SR. The results show that Ca²⁺ uptake by cardiac SR is slower in fish than in mammalian ventricle and that it is significantly faster in trout than in crucian carp heart. Acclimation to cold increases SR Ca²⁺ uptake activity in trout, but decreases it in crucian carp.

Crucian carp, Carassius carassius L. (N=120), were collected in May and June (1995, 1996, 1997) from local ponds within 7 km of the campus area of the University of Joensuu in eastern Finland. The fish were randomly divided in two groups and were acclimated at 4°C or 24°C in the laboratory for more than 4 weeks. The body masses of warm-acclimated and cold-acclimated carp were 34.4±2.4 g (N=68) and 29.8±1.4 g (N=52) (mean ± S.E.M., P=0.12), respectively. Rainbow trout [Oncorhynchus mykiss (Walbaum)] (N=33), were obtained from a local fish farm. The trout were acclimated at 4 or 17°C for a minimum of 4 weeks. The mean body masses of warm-acclimated and cold-acclimated trout were 227.9±31.5 g (N=13) and 210.4±36.7 g (N=15) (P=0.72), respectively. During the acclimation, the fish were held in 500 l stainless-steel aquaria with a continuous circulation of aerated tap (ground) water. The fish were fed five times a week with commercial fish pellets (Ewos, Sweden). Photoperiod was a constant 12 h:12 h light:dark cycle. The rats were Sprague-Dawley strain. Adult rats were 3–6 months in age, neonates were used within 12 h of birth. Fish were stunned by a blow on the head and killed by cutting the spine. Rats were killed by cervical dislocation under light anaesthesia before the hearts were removed.

The amount of ventricular tissue needed for the determination of SR Ca²⁺-ATPase was approximately 50 mg. A portion of rainbow trout ventricle was sufficient for one preparation, while whole ventricles from 3–4 crucian carp had to be pooled for one homogenate. A small piece from the apex of an adult rat ventricle and a whole ventricle from a newborn rat were used for the preparation of one homogenate. The Ca²⁺-dependent ATPase activity of SR was determined from unfractioned ventricular homogenates using two different methods. The Ca²⁺-ATPase activity of SR can be selectively inhibited at high (28 mmol l⁻¹) Ca²⁺ concentration, which saturates the low-affinity inhibitory binding site on the enzyme. This feedback inhibition phenomenon was exploited to determine SR Ca²⁺-ATPase activity according to the procedure developed by Simonides and van Hardeveld (1990) for skeletal muscle homogenates (method 1). In the second assay procedure (method 2), SR Ca²⁺-ATPase was specifically inhibited with 20 μmol l⁻¹ thapsigargin (TG) (Sagara and Inesi, 1991).

Ventricular muscle was minced with scissors and homogenised in 10 volumes of cold sucrose–histidine buffer containing (in mmol l⁻¹): sucrose, 200; L-histidine, 40; EDTA, 1; NaN₃, 10; pH 7.8) with a glass homogenisor (Heidolph) using three 10 s periods each separated by a 30 s interval. The final dilution of homogenates was 1:30. In method 1, the Ca²⁺-ATPase activity of SR was determined as the difference in the rate of ATP splitting in the presence of low (1 mmol l⁻¹; pCa 5.0) and high (28 mmol l⁻¹; pCa 1.6) CaCl₂ concentrations (Simonides and Van Hardeveld, 1990). The final incubation medium contained (in mmol l⁻¹): Hepes, 20; KCl, 200; MgCl₂, 15; NaN₃, 10; EGTA, 1; Na₂ATP, 5; and CaCl₂, 1 or 28; at pH 7.5. Furthermore, Triton X-100 (0.005%) was included to make sealed membrane vesicles leaky. The enzyme reaction was initiated by adding 0.1 ml of the homogenate and terminated after 10 min of incubation at 25°C with acid phosphomolybdate reagent. Samples were centrifuged at 1000 g for 5 min to sediment the tissue, and liberated inorganic phosphate was determined according to Atkinson et al. (1973). In method 2, the activity of SR Ca²⁺-ATPase was determined as the TG-sensitive portion of the total Ca²⁺-activated ATPase activity in the low-Ca²⁺ (1 mmol l⁻¹) medium. SR homogenates were incubated for 10 min in 1 ml of low-Ca²⁺ medium in the presence and absence of 20 μmol l⁻¹ TG at 25°C. The released inorganic phosphate was determined in the protein-free supernatant as described above. Homogenate protein concentration was determined using the method of Lowry (Lowry et al. 1951).

The effect of TG (20 μmol l⁻¹) on myofibrillar Ca²⁺-ATPase was also determined. Ventricular muscle (approximately 150 mg) was collected from 10–12 crucian carp, and myofibrils were purified as described previously (Vornanen, 1996a). Purified myofibrils were suspended in a buffer of low ionic strength containing 45 mmol l⁻¹ imidazole and 50 mmol l⁻¹ KCl at pH 7.0. Myofibrillar Ca²⁺/Mg²⁺-ATPase activity was determined in buffer solution containing (in mmol l⁻¹): imidazole, 45; KCl, 50; EGTA, 5; MgCl₂, 5; Na₂ATP, 3; and CaCl₂, 5.137; to give a free [Ca²⁺] of 100 μmol l⁻¹ (calculated according to Fabiato, 1988) at pH 7.0. The difference in ATPase activity in the presence and absence of Ca²⁺ was taken as myofibrillar Ca²⁺/Mg²⁺-ATPase activity.

ATP-dependent Ca²⁺ uptake in the SR was measured spectrophotometrically in crude ventricular homogenates using a Ca²⁺-selective fluorescent dye, Fura-2 (Hove-Madsen and Bers, 1993b; Kargacin and Kargacin, 1994). Ventricular homogenates were prepared in a medium containing (in mmol l⁻¹): Hepes, 20; KCl, 100; and MgCl₂, 4; pH 7.0. Ca²⁺ uptake was measured in a buffer containing (in mmol l⁻¹): KCl, 100; Hepes, 20; MgCl₂, 4; oxalate, 10; Na₂ATP, 1.25; and creatine phosphate, 1.25 (pH 7.0). Creatine phosphokinase (0.4 units ml⁻¹) was included in the medium to regenerate ATP. Fura-2 was added to a final concentration of 2 μmol l⁻¹. Ca²⁺ uptake was initiated by the addition of 1 mmol l⁻¹ CaCl₂ stock solution to give a final concentration of 2.5 or 5 μmol l⁻¹. The uptake medium and all other constituents were mixed by turning the capped cuvette (4 ml) quickly upside down. Fluorescence was measured using a Shimadzu RF-5000 fluorimeter at room temperature (22°C). The excitation wavelength alternated between 340 and 380 nm at a frequency of 0.5 Hz. Emission was monitored at 510 nm. The emission signal was integrated for 2 s at various times following Ca²⁺ addition. In all steps of fluorescence measurement, special care was taken to avoid Ca²⁺ contamination from glassware and chemicals. Glassware was routinely soaked in 1 mmol l⁻¹ EGTA and thoroughly rinsed in Millipore water (resistance >17 MΩ) before use. Hepes (Sigma), oxalate (Fluka), MgCl₂(Fluka), CaCl₂ and KCl (Merck) were of the highest quality available. Creatine phosphate, creatine phosphokinase and Na₂ATP were obtained from Sigma. Fura-2 (Sigma) was dissolved in dimethyl sulphoxide at 1 mmol l⁻¹ and stored frozen in small samples.

The temperature coefficient (Q₁₀) of TG-sensitive Ca²⁺ uptake was determined for cold-acclimated trout, cold-acclimated crucian carp and adult rat. To this end, the Ca²⁺ uptake velocity was first measured at room temperature, and the temperature of the thermostatted cuvette compartment of the fluorimeter was then either lowered to 12°C (trout and carp) or raised to 32°C (rat) for Ca²⁺ uptake determinations at the second temperature. Q₁₀ values were calculated and used to determine the Ca²⁺ uptake velocity at the physiological body temperature of the animal. The body temperatures of warm-acclimated trout and warm-acclimated carp were close to the experimental temperature, and any small deviation was corrected using the Q₁₀ values of the cold-acclimated fish.

All results are given as mean ± S.E.M. Differences between species, acclimation temperature and age groups were compared by one-way analysis of variance (ANOVA) and post-hoc Student–Newmann–Keuls tests. Before statistical analysis, the data were log_e-transformed to fulfil the assumptions of the parametric test. The differences were considered to be significant at P<0.05.

The crude homogenate contains a number of cation-activated ATPases including those of the mitochondria, myofibrils and sarcoplasmic reticulum. Although TG is a highly specific inhibitor of SR Ca²⁺-ATPase, without effect on plasma membrane Ca²⁺-ATPase or Na⁺/K⁺-ATPase (Lytton et al. 1991), the experimental conditions should be selected to optimize the expression of SR Ca²⁺-ATPase and to minimize background activities. Under the experimental conditions used in this study, approximately 80% of the total ATPase was due to the background ATPase activities. Since mitochondrial ATPase activity was abolished by 10 mmol l⁻¹ azide, the background activity probably originates from the myofibrils. Therefore, it was necessary to ascertain that actomyosin ATPase does not contribute to the TG-sensitive activity. At 20 μmol l⁻¹, TG neither inhibited nor activated the Ca²⁺/Mg²⁺-ATPase of purified myofibrils, suggesting that the TG-sensitive ATPase is not contaminated by actomyosin ATPase (Fig. 1A). The effectiveness of TG as an inhibitor of SR Ca²⁺-ATPase in cardiac homogenates was examined by incubating homogenates in the presence of different TG concentrations. Half-maximal inhibition occurred at 1.6 μmol l⁻¹ and maximal inhibition was obtained at a TG concentration of 20 μmol l⁻¹ (Fig. 1B). As a lipophilic compound, TG shows a high level of nonspecific binding, which may reduce its effectiveness as an inhibitor of SR Ca²⁺-ATPase. Therefore, the inhibition of total Ca²⁺-dependent ATPase activity by 20 μmol l⁻¹ TG was tested at different homogenate dilutions. The inhibition of SR Ca²⁺-ATPase was independent of the dilution of the homogenate within the range used in the present experiments (Fig. 1C). The above findings suggest that SR Ca²⁺-ATPase can be confidently assayed in crude cardiac homogenates as the TG-sensitive ATPase activity.

Fig. 2 shows Ca²⁺-ATPase activities of SR from trout, carp and rat determined as ATPase activity inhibitable by 20 μmol l⁻¹ TG (Fig. 2A) or high concentration of Ca²⁺ (Fig. 2B). The TG-sensitive Ca²⁺-ATPase activity was greatest in adult rat, and the corresponding activities of cold-acclimated trout, warm-acclimated trout, warm-acclimated carp, cold-acclimated carp and newborn rat were 76, 58, 43, 28 and 23%, respectively, of that of the adult rat at 25°C. Similar activities and species order were obtained when SR Ca²⁺-ATPase was determined as the Ca²⁺-inhibitable portion of the total ATPase activity. However, unlike Ca²⁺ uptake rates, the differences in ATPase activities between acclimation groups were not statistically significant (see below).

Fura-2 was used to monitor ATP-dependent and oxalate-supported Ca²⁺ uptake of the SR continuously in crude cardiac homogenates. As Ca²⁺ is transported into the SR, the cytosolic [Ca²⁺]_free declines and the total Ca²⁺ uptake by the SR can be calculated from [Ca²⁺]_free and the known concentrations of Ca²⁺ buffers in the uptake medium (Fig. 3). Accumulation of Ca²⁺ within SR results in the formation of calcium oxalate crystals, which effectively block Ca²⁺ efflux (Palade, 1987). Ruthenium Red was not used to block Ca²⁺ leakage since, at 30 μmol l⁻¹, it caused variable inhibition of Ca²⁺ uptake both in fish and in mammalian preparations. The contribution of mitochondrial Ca²⁺ uptake to Fura-2 fluorescence is excluded since [Ca²⁺]_free in the uptake medium was less than 3 μmol l⁻¹, which is below the threshold concentration for mitochondrial Ca²⁺ uptake (Hove-Madsen and Bers, 1993a). Fig. 3A shows changes in Fura-2 fluorescence due to Ca²⁺ uptake by various cardiac preparations. It is evident that the rate of Ca²⁺ uptake is fastest in adult rat, intermediate in trout and relatively slow in carp and newborn rat. The oxalate-supported uptake of Ca²⁺ was completely inhibited when homogenates were pretreated with 20 μmol l⁻¹ TG for 3 min (Fig. 3B).

Since [Ca²⁺]_free varied slightly from experiment to experiment, the Ca²⁺ uptake rate was calculated at a constant [Ca²⁺]_free of 0.4 μmol l⁻¹. SR Ca²⁺ uptake velocity, measured using Fura-2 at room temperature (22°C), was greatest in cold-acclimated trout, and the values for adult rat, warm-acclimated trout, newborn rat, warm-acclimated carp and cold-acclimated carp were 93, 56, 24, 21 and 14%, respectively, of the uptake rate of cold-acclimated trout (Fig. 4). Therefore, the species order for SR Ca²⁺ uptake was: cold-acclimated trout ≈adult rat > warm-acclimated trout > newborn rat ≈ warm-acclimated carp > cold-acclimated carp (where > indicates a statistically significant difference). Furthermore, there was good correlation between Ca²⁺-ATPase activity and Ca²⁺ uptake rate (Fig. 5), with a slope close to 1.0 for the percentage values (not shown). The major difference between ATPase activity and Ca²⁺ uptake velocity in absolute terms is partly because Ca²⁺-ATPase activity was determined under conditions that allow maximal enzyme activity, while the rate of Ca²⁺ uptake was measured at a Ca²⁺ concentration close to the K_m of the SR Ca²⁺ pump.

Since the physiological body temperatures of the animals were markedly different, the measured Ca²⁺ uptake velocities were transformed to the respective body temperature of the species. Q₁₀ values were 1.75, 1.64 and 1.79 for cold-acclimated carp, cold-acclimated trout and adult rat (N=2 for each), respectively. When corrected to the body temperature of the animal, the relative rates of SR Ca²⁺ uptake were 100, 26, 19, 18, 11 and 2% for adult rat, newborn rat, cold-acclimated trout, warm-acclimated trout, warm-acclimated carp and cold-acclimated carp, respectively.

The small body mass of many fish species, the small relative size of fish hearts and the relatively low content of SR in fish cardiac myocytes make species comparisons with respect to SR Ca²⁺-handling capacity difficult. Quantitative estimates of the Ca²⁺-ATPase activity and Ca²⁺ uptake rate of SR between different animal species are further complicated by the unpredictable loss of SR membranes and the variable inactivation of the ATPase during the long isolation procedures (Feher and LeBolt, 1990). The recent discovery of specific inhibitors for SR Ca²⁺-ATPase and the use of Ca²⁺-specific fluorescent indicators have enabled Ca²⁺-ATPase activity and SR Ca²⁺ uptake to be analysed directly in isolated cardiac myocytes or unfractioned skeletal muscle homogenates, thus avoiding many of these problems. We have used these novel tools to study the effects of thermal acclimation on fish cardiac SR and to compare Ca²⁺ handling in the SR of fish and mammalian hearts.

In the present study, SR Ca²⁺-ATPase activity and Ca²⁺ uptake rate of fish and mammalian hearts were determined at the same experimental temperature, which allows direct species comparisons of SR Ca²⁺-handling capacity. It is evident that, in crucian carp and newborn rat heart, SR Ca²⁺ uptake is relatively slow, while in rainbow trout and adult rat heart SR Ca²⁺ uptake is much faster. These findings agree well with previous observations on both mammalian and fish hearts. It is generally accepted that, in the newborn rat ventricle, the SR is both structurally and functionally weakly developed, while in the adult rat heart the SR is the major source and sink of the activator Ca²⁺ (Fabiato, 1982; Bers, 1985; Vornanen, 1996b). In our experiments using Fura-2, the SR Ca²⁺ uptake rate of newborn rat heart was approximately one-quarter of that in the adult rat, which is in good agreement with previous findings on the amount of SR Ca²⁺-ATPase mRNA (Lompre et al. 1991) and the rate of ⁴⁵Ca²⁺ uptake in the developing rat heart (Vetter et al. 1995). The SR Ca²⁺ uptake rates of trout and carp cardiac homogenates are also consistent with previous functional studies on intact trout and carp cardiac preparations. In agreement with the low Ca²⁺-handling capacity of cardiac homogenates, the contraction of crucian carp ventricle is insensitive to the SR inhibitor ryanodine (Vornanen, 1996a). In contrast to the carp, the SR of trout ventricle seems to be involved in contractile regulation (Keen et al. 1994) as also suggested by the higher Ca²⁺ uptake rate of trout cardiac SR.

Since the body temperatures of fish and mammals are quite different, the real differences in SR Ca²⁺ uptake rate between ecto- and endothermic vertebrates might be greater than is evident from experiments performed at room temperature. When corrected to the physiological body temperatures, it is evident that the rate of Ca²⁺ uptake is much higher in mammalian than in teleost hearts. The Ca²⁺ uptake rate of trout SR would be approximately 20% of the activity of the SR in adult rat heart. The corresponding value for crucian carp heart would be only 2–11%. However, comparisons of Ca²⁺ uptake rates in vitro may exaggerate the true differences in the rates of SR Ca²⁺ accumulation during cardiac contraction in vivo. SR Ca²⁺ uptake will occur during sarcolemmal depolarization, when Ca²⁺ efflux through Na⁺/Ca²⁺ exchanger is impeded. The duration of the action potential in adult rat heart is less than 100 ms, while it is closer to 500 ms in fish ventricles (Vornanen, 1989). Thus, despite its relatively slow Ca²⁺ uptake rate, the fish cardiac SR may accumulate significant amounts of Ca²⁺ during the long-lasting depolarization if the Ca²⁺-storing capacity of SR is adequate. The true differences in Ca²⁺-handling capacity of cardiac SR between vertebrate species can only be evaluated by experiments using intact cardiac myocytes conducted at physiological body temperatures. The present results support the hypothesis that efficient SR Ca²⁺ uptake is associated with high in vivo heart rates (mammals), which necessitate high rates of contraction and relaxation during the brief systolic period.

Temperature acclimation changed the activity of SR Ca²⁺ handling in both crucian carp and rainbow trout. The temperature-induced changes in the activity of the Ca²⁺-sensitive ATPase, the TG-sensitive ATPase and TG-sensitive Ca²⁺ uptake were qualitatively similar, although the differences were statistically significant only in the case of Ca²⁺ uptake rate. It is possible that thermal acclimation changes not only the maximum rate of Ca²⁺ sequestration but also the Ca²⁺ affinity of the SR Ca²⁺-ATPase, which could explain the differences between Ca²⁺-ATPase activities (measured at optimal Ca²⁺ concentration) and Ca²⁺ uptake rates (measured at a Ca²⁺ concentration close to the K_m of the enzyme). However, we cannot exclude the possibility that thermal acclimation also affects passive Ca²⁺ leakage across SR membranes.

The effects of thermal acclimation in the two teleost species were opposite: an enhancement of SR Ca²⁺ uptake in trout and a reduction in Ca²⁺-handling capacity of the SR in carp after acclimation to cold temperature. It is notable that, in the trout heart, there is complete thermal compensation of SR Ca²⁺ uptake velocity. However, the physiological significance of this acclimation response is not clear. It has been argued that fish cardiac SR may be unable to store Ca²⁺ (Keen et al. 1992), since Ca²⁺ release channels tend to remain open at low ambient temperatures (Sitsapesan et al. 1991). The present findings agree with previous functional studies showing increased sensitivity of trout ventricle to ryanodine inhibition after acclimation to cold (Keen et al. 1994). Similarly, the low Ca²⁺ uptake rate of cold-acclimated carp ventricle agrees with the reduced cardiac activity in cold-acclimated crucian carp, expressed as a long relaxation time, a slow heart rate and low myosin-ATPase activity of the heart (Matikainen and Vornanen, 1992; Vornanen, 1994, 1996a). Therefore, several kinds of evidence suggest that, in species that remain active at low ambient temperatures (rainbow trout, perch), adaptation to cold increases the volume and Ca²⁺-handling efficiency of the SR (Bowler and Tirri, 1990; Keen et al. 1994), while in species that become inactive or dormant in the cold (crucian carp), the limited Ca²⁺-handling capacity is depressed after acclimation to low ambient temperature.

In conclusion, the locomotor activity and cardiac performance of fish vary greatly. Over the entire spectrum of species, there are approximately tenfold differences in heart rate, cardiac output and maximum ventral aortic pressure (Farrell, 1996). The present study suggests that there are cellular differences in cardiac Ca²⁺ regulation between teleost fish: ‘athletic’ fish species (such as trout) have a higher Ca²⁺-handling capacity in the cardiac SR than less active species (such as the crucian carp). Furthermore, acclimation to cold in active species is associated with an increased Ca²⁺-handling capacity of the SR, which may allow partial compensation for the detrimental effects of cold on cardiac contractility. The inverse thermal compensation in Ca²⁺ uptake rate of crucian carp cardiac SR is probably associated with its unusual ecophysiological adaptations. Crucian carp are specialized to inhabit small shallow ponds which become anoxic in winter. Under these conditions positive thermal compensation with associated increase in cardiac energy consumption would not be beneficial for the survival of the fish.

This study was supported by the Academy of Finland (project no. 7641).