Effect of β-adrenergic stimulation on the relationship between membrane potential, intracellular [Ca²⁺] and sarcoplasmic reticulum Ca²⁺ uptake in rainbow trout atrial myocytes

In the mammalian heart, β-adrenergic stimulation has both inotropic and lusitropic effects. The inotropic effect is due to phosphokinase A(PKA)-dependent effects on both l-type Ca²⁺ current(Hartzell et al., 1991; Hove-Madsen et al., 1996; Hove-Madsen and Tort, 1998; Jurevicius and Fischmeister,1996) and the SR function, while stimulation of SR Ca²⁺uptake through a PKA-dependent phosphorylation of phospholamban(Chu et al., 2000; Mattiazzi et al., 1994) has been reported to account for the lusitropic effect(Li et al., 2000). Indeed, the effect of β-adrenergic agonists and PKA-dependent phosphorylation on SR Ca²⁺ uptake has been examined in both intact and permeabilized myocytes, measuring the effects on Ca²⁺ sparks(Gomez et al., 1996; Zhou et al., 1999; Ziolo et al., 2001), caffeine(Caf) induced Ca²⁺ transients(Li et al., 1998) or oxalate stimulated Ca²⁺ uptake in the SR(Mattiazzi et al., 1994).

In the teleost heart, the effect of β-adrenergic stimulation on I_Ca has recently been characterized(Hove-Madsen and Tort, 1998; Vornanen, 1998), with a two-to threefold stimulation of I_Ca. This is similar to values obtained in mammals, but much smaller that the seven- to tenfold stimulation reported in frog (Fischmeister and Shrier,1989). Based on the stimulatory effect of β-adrenergic stimulation of I_Ca, it has been suggested that sarcolemmal Ca²⁺ influx is sufficient to activate contraction under these conditions (Vornanen, 1998).β-adrenergic stimulation does, however, also affect other important factors such as myofilament sensitivity and SR function, which critically determine the amplitude and kinetics of cell shortening(Freestone et al., 2000; Li et al., 2000).

With respect to the SR function, the short duration of the Ca²⁺transient in the mammalian cardiac myocyte and the simultaneous regulation of the intercellular [Ca²⁺] by several mechanisms have made it difficult to characterize the relationship between[Ca²⁺]_i and the rate of Ca²⁺ uptake in the SR in intact mammalian cardiac myocytes. In contrast to this, long depolarizations in lower vertebrate myocytes lead to a tonic cell shortening,a maintained outward current (Hove-Madsen et al., 1998, 2000) and presumably an intracellular [Ca²⁺] that is maintained elevated. Furthermore, a voltage dependent Ca²⁺ uptake by the SR takes place during long depolarizations in these cells and can be used to estimate the average SR Ca²⁺ uptake rate (Hove-Madsen et al., 1998, 2000). Using simultaneous measurements of intracellular [Ca²⁺] and whole membrane current,the present study has taken advantage of these properties of trout cardiac myocytes in order to (1) characterize the relationship between membrane potential, [Ca²⁺]_i and the SR Ca²⁺ uptake rate under basal and phosphorylating conditions, (2) use this relationship between cytosolic [Ca²⁺] and SR Ca²⁺ uptake to calculate the SR Ca²⁺ uptake during a Ca²⁺ transient induced by a 200 ms depolarization, and (3) determine the relative contribution of SR Ca²⁺ uptake to total Ca²⁺ transport in the absence and presence of β-adrenergic stimulation.

Trouts Oncorhynchus mykiss Walbaum were purchased from a local fish farm and acclimated in tanks at the facilities at the Universitat Autònoma de Barcelona to the water temperature at the fish farm(16°C). Atrial myocytes were obtained by enzymatic digestion of the heart at 18–20°C as described previously(Hove-Madsen et al., 1999). The cell isolation procedure and the experimental protocols were approved by the local ethical committee (DARP no. 1017).

Whole membrane current was measured using a software driven patch-clamp amplifier (EPC-9, Heka, Germany) or an Axopatch 200, (Axon, California, USA). After seal formation, the cell was placed in front of one of eight capillaries containing the desired extracellular solution. Standard internal and external solutions were used to eliminate Na⁺ and K⁺ currents. The pipette solution contained (in mmol l^–1): CsCl 100, MgATP 3.1; MgCl₂ 1; sodium phosphocreatine 5; Li₂GTP 0.42;EGTA 0.025 (no EGTA was included when using fluo-3 loaded cells); Hepes 10;tetraethyl ammonium (TEA) 20. The pH was adjusted to 7.2 with CsOH. The external medium contained (in mmol l^–1): NaCl 107; CsCl 20;MgCl₂ 1.8; NaHCO₃ 4; NaH₂PO₄ 0.8;CaCl₂ 1.8; Hepes 10; glucose 5; pyruvate 5. The pH was adjusted to 7.4 with NaOH. The Na⁺ current was eliminated with 1 μmol l^–1 tetrodotoxin (TTX). Whole membrane current was measured in the ruptured or the perforated patch configuration with 200–250 μg ml^–1 amphotericin B (Figs 4, 5, 6, 7) at room temperature(20°C). The pipette resistance was 2–5 MΩ, seal resistance was 2–20 GΩ, and the access resistance was 6.5±1.2 MΩ for ruptured patches. When using amphotericin B, equilibration of the preparation by continuous stimulation at 0.2 Hz was started when access resistance had decreased to ∼20 MΩ, and the access resistance was 11.6±3.1 MΩ at the end of the experiments. Cells showing a sudden drop in access resistance were discarded.

The cells were stimulated continuously with a 200 ms depolarization from–80 mV to 0 mV every 5 s. This protocol allowed evaluation of I_Ca and Caf induced Na⁺–Ca²⁺ exchange(NCX) current (I_NCX) under steady state conditions in the absence and presence of β-adrenergic stimulation. β-adrenergic stimulation was produced by exposing cell to 1 μmol l^–1 isoproterenol(ISO). The Caf induced I_NCX allows determination of the SR Ca²⁺ content in the absence and presence of ISO as described previously (Hove-Madsen et al.,1998; Negretti et al.,1995; Varro et al.,1993). The basic experimental protocol outlined in Fig. 1 was used to determine the effect of β-adrenergic stimulation on Ca²⁺ uptake in the SR. First the SR Ca²⁺ content was cleared with a brief exposure to 10 mmol l^–1 caffeine (Caf). Then Ca²⁺ uptake was induced by depolarizing the cell to different membrane potentials for 3 or 10 s (labeled Load in Fig. 1). The amount of Ca²⁺ accumulated in the SR was determined from the I_NCX elicited by exposing the cell to 10 mmol l^–1Caf after repolarizing the cell to –80 mV (labeled Caf in Fig. 1). The SR Ca²⁺uptake rate was calculated by dividing the Ca²⁺ accumulated by the SR with the duration of the depolarization. For conversion of the time integrals of Ca²⁺ carrying currents to Ca²⁺concentrations, a conversion factor of 15.4 pF pl^–1 was used(Hove-Madsen and Tort, 1998)and the Ca²⁺ accessible cell volume was considered to equal the non-mitochondrial cell volume, which in trout corresponds to 55% of the total cell volume (Vornanen,1998).

The intracellular Ca²⁺ concentration[Ca²⁺]_i was measured with the Ca²⁺ indicator fluo-3. Cells were incubated with 5 μmol l^–1 fluo-3 AM for 10–15 min at room temperature. After incubation cells were washed and left for 30 min at room temperature before starting experiments. Fluo-3 was excited at 488±1 nm and fluorescence emission was measured using a 530 nm bandpass filter. The bandpass width was ±20 nm. A sampling interval of 10 ms was used to achieve a reasonable signal-to-noise ratio. This sampling interval was necessary as the cell volume of trout atrial myocytes is 2–4 pl, which is about ten times less than the cell volume of mammalian cardiac myocytes. Calibration of the fluo-3 signal in each cell was done by subtraction of background fluorescence and measurement of the maximal fluorescence at the end of each experiment. As long depolarizations lead to a tonic elevation of [Ca²⁺]_i and contraction in trout cardiac myocytes (Hove-Madsen et al., 1998, 2000), maximal fluorescence was obtained by depolarizing the cell for 3 s to increasingly more positive potentials until an irreversible cell contracture was achieved. The fluo-3 signal measured at irreversible cell contracture was taken as the maximal fluorescence since it also coincided with a large and irreversible increase in leak, and it was similar to the fluorescence obtained when leak was induced by squeezing the cell with the patch pipette. [Ca²⁺]_i was then calculated from[Ca²⁺]_i=F×K_d/(F_max–F)where F is the measured net fluorescence, F_max is the maximal net fluorescence and K_d is the dissociation constant for Ca²⁺ for fluo-3(Trafford et al., 1999). A K_d for fluo-3 in cells of 1100 nmol l^–1(Harkins et al., 1993) was used in the present study.

Unless otherwise stated, cells served as their own control when the effects of β-adrenergic stimulation were examined. For statistical analysis of the data, Student's t-test for paired samples was used to compare two experimental conditions. Values are given as means ± s.e.m. Analysis of variance (ANOVA) was used to evaluate the effect of β-adrenergic stimulation on the relationships between membrane potential, [Ca²⁺]_i and SR Ca²⁺ uptake (Figs 3, 6 and 9). While SR Ca²⁺uptake does not depend directly on the membrane potential, it does depend on the [Ca²⁺]_i concentration, and since the[Ca²⁺]_i concentration depends on the membrane potential,the Boltzmann equation was used to empirically fit the relationship between[Ca²⁺]_i or SR Ca²⁺ uptake and membrane potential (Figs 3 and 9). Similarly, the relationship between SR Ca²⁺ uptake and [Ca²⁺]_i has been described by a Hill equation (Hove-Madsen and Bers, 1993) and this equation was therefore used to fit the data in Fig. 6C.

Fig. 2A shows typical NCX currents elicited by Caf after a 10 s depolarization to –70, –30 and –10 mV. Current traces obtained in the absence and the presence of 1μmol l^–1 ISO are shown for each membrane potential. The amount of Ca²⁺ released from the SR was determined from the time integrals of the Caf induced NCX currents as shown in Fig. 2B. Fig. 3 summarizes the dependency of the SR Ca²⁺ accumulation on the membrane potential in the absence and the presence of ISO. The lines represent empirical fits of data from six cells with a Boltzmann equation. This fit gave slopes of 22.2 and 23.8 in the absence and the presence of ISO, respectively, and a basal SR Ca²⁺ loading of 14 μmol l^–1 and 20 μmol l^–1, respectively. K_0.5 was –12 mV in control and –38 mV in the presence of 1 μmol l^–1ISO while the maximal SR Ca²⁺ load was 38.8 and 39.2 amol pF^–1 or μmol l^–1 in control and ISO,respectively. This corresponds to a maximal SR Ca²⁺ content of 1086 and 1097 μmol l^–1 non-mitochondrial volume in control and ISO, respectively. While ISO changed the relationship between voltage and SR Ca²⁺ uptake significantly (P<0.001, N=6),causing a shift in K_0.5 of –26 mV, it had no significant effect on basal or maximal SR Ca²⁺ uptake. Fitting of data points from individual experiments showed a significant difference between the average K_0.5 in the absence and the presence of ISO (P<0.01), while ISO did not significantly change the slope,the basal or the maximal SR Ca²⁺ loading.

These results show that ISO stimulates Ca²⁺ accumulation, so it was important to determine whether the stimulation was a result of an increased sarcolemmal Ca²⁺ entry or a direct stimulation of SR Ca²⁺ uptake. Therefore, the free [Ca²⁺]_i and the whole membrane current resulting from a long depolarization were measured simultaneously. First, the steady state effect of ISO on ionic current and[Ca²⁺]_i was established using continuous stimulation with 200 ms depolarizations from –80 to 0 mV at a frequency of 0.2 Hz. As shown in Fig. 4, ISO increased both I_Ca (Fig. 4A) and Ca²⁺ transient(Fig. 4B) significantly by 91±29 (P<0,01, N=7) and 41±10%,respectively (P<0.01, N=7). Furthermore, when fitting the decaying phase of the Ca²⁺ transient with a single exponential, ISO decreased the half life of the Ca²⁺ transient significantly from 151±12 to 111±6 ms (P<0.001, N=7).

Fig. 5 shows superimposed fluo-3 traces (Fig. 5A) and ionic currents (Fig. 5B)elicited by exposure to 10 mmol l^–1 Caf (right panels) after a 3 s depolarization to +10 mV (left panels), before (black traces) and after stimulation with 1 μmol l^–1 ISO (gray traces). The membrane potential is indicated in the panel above the traces and the period of Caf application is shown. Notice that both the[Ca²⁺]_i and the whole membrane currents are permanently elevated throughout the 3 s depolarization to +10 mV. Also notice that except from the first few hundred ms, ISO reduced the [Ca²⁺]_iwhile it stimulated the outward current during depolarization. This observation was consistent in all seven cells examined. Fig. 5C shows that the effect of ISO on the Caf induced ionic current and Ca²⁺ transient was reversible. This was also true for the effects of ISO on the whole membrane current and the Ca²⁺ transient elicited with long depolarizations or with repetitive 200 ms depolarizations (data not shown). Fig. 6A shows recordings of the[Ca²⁺]_i during depolarization to different membrane potentials and caffeine applications in the absence (left panels) and the presence of 1 μmol l^–1 ISO (right panels). The membrane potential is given in the top of the two bars above the recordings, and the absence (–) or presence of 10 mmol l^–1 Caf (+) in the bath solution in the bottom bar. Fig. 6B compares the average Ca²⁺ transient during depolarization (black bars) and the peak Caf induced Ca²⁺ transient(white bars) after depolarization in the absence (left panel) or the presence of 1 μmol l^–1 ISO (right panel). The membrane potential during the depolarization is given below the bars. Notice that there was no significant difference between the average Ca²⁺ transient during depolarization and the peak Ca²⁺ transient elicited by Caf after depolarization in control conditions, while the Caf induced Ca²⁺transient was significantly bigger than the average Ca²⁺ transient during the preceding depolarization in the presence of ISO(P<0.001, N=7). This suggest that ISO directly stimulates SR Ca²⁺ uptake, and Fig. 6C shows the relationship between SR Ca²⁺ uptake(dCa/dt) and the average [Ca²⁺]_i during depolarization without (filled circles) and with 1 μmol l^–1 ISO (open circles). The lines represent fits of the data points with a Hill equation. This gave Hill slopes of 1.74 and 2.24 without and with ISO, respectively,while the V̇_max values were unchanged at 418 and 435 μmol l^–1 s^–1,respectively. K_0.5 was shifted from 1.27 μmol l^–1 in the control to 0.80 μmol l^–1 with 1 μmol l^–1 ISO. Fitting of data points from individual experiments showed that there was a significant difference between the average K_0.5 in the absence and the presence of ISO(P<0.05), while ISO did not significantly affect the Hill slope or the V̇_max.

Using this relationship between cytosolic [Ca²⁺] and SR Ca²⁺ uptake rate, the total SR Ca²⁺ uptake during the Ca²⁺ transients shown in Fig. 4B can be calculated by first calculating the SR Ca²⁺uptake rate during the Ca²⁺ transient, and then integrating this signal to obtain the total SR Ca²⁺ uptake during the Ca²⁺ transient. Similarly the total sarcolemmal Ca²⁺extrusion can be calculated from the time integral of the tail currents in Fig. 4A. Fig. 7A shows the total SR Ca²⁺ uptake (labeled SR), total sarcolemmal Ca²⁺extrusion (labeled SL) and total Ca²⁺ transport (labeled TOT)without (black bars) and with 1 μmol l^–1 ISO (gray bars). Notice that ISO significantly increased all three parameters. When normalizing the SR Ca²⁺ uptake to the total Ca²⁺ transport, it constituted 35±6% and 41±4% of the total Ca²⁺eliminated from the cytosol without and with 1 μmol l^–1ISO, respectively.

To examine whether the stimulatory effect of ISO was able to compensate for a lowering of the extracellular [Ca²⁺], the effect of ISO applied in the presence of a low extracellular [Ca²⁺] (0.3 or 0.5 mmol l^–1) was compared with control values obtained in the absence of ISO with 1.8 mmol l^–1 extracellular Ca²⁺. Fig. 8A shows representative current traces elicited by a 200 ms depolarization in control conditions and with ISO + low Ca²⁺. On average, ISO increased the I_Caamplitude by 67±40% (N=5). Fig. 8B shows the corresponding current traces elicited by a rapid Caf application after a 10 s depolarization. Notice that SR Ca²⁺ uptake was strongly reduced in the presence of ISO + low Ca²⁺. Fig. 9 compares the relationship between the membrane potential during a 10 s depolarization and the amount of Ca²⁺ taken up by the SR in control conditions and with ISO + low Ca²⁺. The relationship between SR Ca²⁺uptake and membrane potential was significantly affected by application of ISO+ low Ca²⁺ (P<0.001, N=5) resulting in a reduced uptake at membrane potentials above –30 mV. Thus while ISO could overcome the effect of lowering extracellular Ca²⁺ on I_Ca it was unable to prevent a strong reduction in SR Ca²⁺ uptake at physiologically relevant membrane potentials.

The aim of the present work was to characterize the relationship between the intracellular [Ca²⁺] and the SR Ca²⁺ uptake rate and to examine the effect of β-adrenergic stimulation on SR Ca²⁺uptake and Ca²⁺ removal from the cytosol in trout atrial myocytes. To do this, the perforated-patch configuration was used to apply physiological or long depolarizations to trout atrial myocytes.

This experimental approach is based on the following assumptions: (1) SR Ca²⁺ uptake can be quantified as the amount of Ca²⁺released from the SR by Caf, (2) SR Ca²⁺ uptake takes place throughout a long depolarization, (3) the average SR Ca²⁺ uptake rate during a depolarization can be calculated because uptake is uniform throughout the depolarization. With respect to the first assumption, it has been shown that Ca²⁺ released from the SR by Caf can be used as a measure of the SR Ca²⁺ content in mammalian and trout cardiomyocytes (Hove-Madsen et al., 1998, 2000; Trafford et al., 1999). Since K is substituted with Cs and 15 mmol l^–1 extracellular NaCl is replaced with CsCl in the present experiments, SR Ca²⁺ uptake could on the one hand be underestimated, because Cs has been shown to decrease SR Ca²⁺ uptake in mammalian cardiomyocytes(Kawai et al., 1998), and on the other hand be overestimated, because of the 10% reduction in the extracellular [Na⁺]. The effect of Cs on SR Ca²⁺ uptake is, however, not expected to differ in the absence and presence of ISO. Regarding the second assumption, Ca²⁺ uptake has been shown to increase proportionally to the length of the depolarization as long as the SR Ca²⁺ content is not near saturation(Hove-Madsen et al., 1998),suggesting that SR Ca²⁺ uptake takes place throughout the depolarizations used in the present study. Finally, Fig. 5 shows that both the Ca²⁺ transient and the outward current remain elevated throughout a 3 s depolarization so that the average SR Ca²⁺ uptake rate can be calculated as the amount of Ca²⁺ released by Caf divided by the duration of the preceding depolarization.

When measuring SR Ca²⁺ uptake as a function of membrane potential, β-adrenergic stimulation affects neither the maximal SR Ca²⁺ uptake nor the SR Ca²⁺ uptake at rest. It does,however, significantly shift the relationship between SR Ca²⁺uptake and membrane potential by –26 mV, so that the SR can more efficiently remove Ca²⁺ from the cytosol during the repolarization of the action potential and during diastole. This in turn may contribute to the faster decay of the Ca²⁺ transient elicited by a standard 200 ms depolarization observed in Fig. 4.

The effect of β-adrenergic stimulation on the relationship between SR Ca²⁺ uptake and membrane potential could either be due to a direct effect on SR Ca²⁺ uptake or a consequence of its stimulatory effect on I_Ca and the intracellular Ca²⁺ transient (see Fig. 4). In this respect, the voltage dependency of I_Ca is also shifted towards more negative membrane potentials by β-adrenergic stimulation(Brum et al., 1984) and this could potentially account for the observed shift in the relationship between membrane potential and SR Ca²⁺ uptake. β-adrenergic stimulation does, however, also stimulate SR Ca²⁺ uptake through phospholamban phosphorylation (Bassani et al., 1995; Li et al.,2000; Mattiazzi et al.,1994), and PKA-dependent phosporylation of phospholamban has been reported to shift the K_d for SR Ca²⁺ uptake towards lower [Ca²⁺] (Mattiazzi et al., 1994). The fact that β-adrenergic stimulation lowers the average [Ca²⁺]_i during long depolarizations and shifts the K_0.5 for the relationship between cytosolic[Ca²⁺] and SR Ca²⁺ uptake from 1.27 to 0.8 μmol l^–1 does therefore suggest that its stimulatory effect on SR Ca²⁺ loading is due to a stimulation of the SR Ca²⁺uptake rather than being a consequence of I_Ca stimulation.

The SR Ca²⁺ uptake in trout atrial myocytes is comparable to values previously reported in trout ventricular myocytes(Hove-Madsen et al., 1998),and the uptake rate is within the range reported in intact or permeabilized mammalian cardiac myocytes at room temperature(Balke et al., 1994; Wimsatt et al., 1990). Also, a 40% reduction of the K_0.5 for SR Ca²⁺ uptake by ISO is similar to values reported in mammals(Mattiazzi et al., 1994). The absolute value for K_0.5 in control is similar to a value previously reported for permeabilized trout ventricular myocytes(Hove-Madsen et al., 1998),while values for both control and ISO are 2–3 times higher than values reported in several mammalian studies(Balke et al., 1994; Hove-Madsen and Bers, 1993; Mattiazzi et al., 1994). These differences may reflect species dependent differences in the SR Ca²⁺ ATPase, but could also be related to the calibration of the fluorescent Ca²⁺ indicator. The present work uses a K_d for fluo-3 of 1.1 μmol l^–1 obtained from in situ calibration in frog skeletal muscle fibers(Harkins et al., 1993), while other studies have often used in vitro calibration of the fluorescent dyes (Balke et al., 1994; Trafford et al., 1999). Using a K_d for fluo-3 of 400 nmol l^–1 from in vitro calibration would give K_0.5 values of 0.46 μmol l^–1 in controls and 0.29 μmol l^–1 with ISO. These values are similar to those reported in mammals (Balke et al., 1994; Trafford et al., 1999),suggesting that differences in dye calibration could at least partially account for the differences in K_0.5 in the present study and values measured in mammals.

While the present results do show that β-adrenergic stimulation causes a direct stimulation of the SR Ca²⁺ uptake, they do not exclude an indirect effect of β-adrenergic stimulation on SR Ca²⁺ uptake with more physiological stimulation pulses. Thus, the contribution of I_Ca to the elevation of the [Ca²⁺]_i during a 3 s or a 10 s depolarization is expected to be significantly smaller than its contribution during a 200 ms depolarization. In agreement with this,I_Ca is not reduced when the extracellular solution is switched from control to ISO plus low extracellular [Ca²⁺] while the SR Ca²⁺ uptake during a 10 s depolarization is drastically reduced at all membrane potentials examined (see Figs 8 and 9). This agrees with previous results, which suggested that the maintained outward current and tonic contraction during long depolarizations in trout cardiac myocytes(Hove-Madsen et al., 2000; Hove-Madsen and Tort, 2001)are due to reverse mode Na⁺–Ca²⁺ exchange.

Furthermore, the fact that ISO has a stimulatory effect on I_Caeven when the extracellular Ca²⁺ is lowered while SR Ca²⁺ uptake during long depolarizations is strongly reduced under these conditions (see Figs 8, 9), suggests that ISO do not have any stimulatory effect on I_NCX.

In this respect the average Ca²⁺ transient is smaller in the presence of ISO, while the corresponding outward current is larger in Fig. 5. Assuming that the outward current is reverse mode I_NCX, this suggests that under these conditions the amplitude of reverse mode I_NCX during long depolarizations is being determined by the [Ca²⁺]_irather than being the amplitude of reverse mode I_NCX that determines the [Ca²⁺]_i. Therefore, ISO not only stimulates SR Ca²⁺ uptake as shown in Figs 2, 3, 5, 6, 7, but the SR Ca²⁺uptake relative to reverse mode NCX does in fact become faster in the presence of ISO. The relatively faster SR Ca²⁺ uptake in turn leads to a reduction in [Ca²⁺]_i, an increase in the transsarcolemmal Ca²⁺ gradient and as a consequence a larger reverse mode I_NCX as observed in Fig. 5. Thus, the apparent stimulation of reverse mode I_NCX in Fig. 5 is more likely to be an indirect effect of ISO on SR Ca²⁺ uptake. The absence ofβ-adrenergic stimulation of the I_NCX agrees with data from the toad heart (Ju and Allen,1999). It would also be in agreement with the absence, in both the trout and mammalian NCX protein (Nicoll et al., 1990; Xue et al.,1999), of the putative PKA-dependent phosporylation site described in frog (Fan et al., 1996). It is, however, contrary to results from the frog heart, where a putative PKA-dependent phosporylation site on the NCX has been described(Fan et al., 1996) and a PKA-dependent inhibition of NCX activity has been reported in both frog and shark cardiac myocytes (Fan et al.,1996; Woo and Morad,2001). The present results also show that the importance of SR Ca²⁺ uptake relative to sarcolemmal Ca²⁺ extrusion by the NCX is unchanged with β-adrenergic stimulation in trout atrial myocytes. This is contrary to conclusions reached from measurements of SR function in tissue preparations (Shiels and Farrell, 1997) and the difference is most likely related to the indirect nature of the measurements in the tissue preparation. Also, the stimulation of l-type Ca²⁺ current by ISO in isolated trout myocytes has been taken as evidence for a larger contribution of sarcolemmal Ca²⁺ entry to the activation of contraction in the presence of β-adrenergic stimulation(Vornanen, 1998). This conclusion is not easily compatible with an unchanged relative contribution of forward mode I_NCX to relaxation with β-adrenergic stimulation found in the present study, since it indirectly implies a similar contribution of sarcolemmal Ca²⁺ entry at steady state.

Finally, it should be noticed that the present results were obtained at room temperature, while the acclimation temperature was 16°C and the preferred temperature for rainbow trout is around 14°C. We cannot rule out that this may have consequences for the measured effects of β-adrenergic stimulation on SR Ca²⁺ content and uptake rates, but we recently found that neither SR Ca²⁺ uptake nor SR Ca²⁺ release were significantly changed when the experimental temperature was lowered from 21 to 7°C (Hove-Madsen et al.,2001). Furthermore, it is not clear that a difference between acclimation and experimental temperature of 4°C would selectively alter the SR Ca²⁺ uptake under basal or phosporylating conditions. An unequivocal answer to this question will, however, have to await further experimentation that directly addresses it.

In conclusion, the present study has characterized the relationship between the membrane potential, cytosolic [Ca²⁺] and SR Ca²⁺uptake rate in trout atrial myocytes in the absence and presence ofβ-adrenergic stimulation. The results show that β-adrenergic stimulation shifts the relationship between membrane potential and SR Ca²⁺ uptake by –26 mV and decreases the K_0.5 for SR Ca²⁺ uptake from 1.27 to 0.8μmol l^–1, while it has no effect on resting or maximal SR Ca²⁺ loading. This may contribute to the faster decay of the Ca²⁺ transient with β-adrenergic stimulation in these cells. Furthermore, β-adrenergic stimulation does not alter the importance of SR Ca²⁺ uptake relative to Ca²⁺ extrusion by the I_NCX, and the characterization of the dependency of SR Ca²⁺ uptake on membrane potential and [Ca²⁺]_ishould be useful for quantitative comparisons of SR Ca²⁺ uptake and Ca²⁺ extrusion by the I_NCX under specific experimental conditions.

This work was supported by a grants from The Spanish Ministry of Education and Culture (Incorporación de Doctores) to L. Hove-Madsen and Generalitat de Catalunya (FPI grant) to A. Llach, (MAR97-402-C02) to L. Tort,and the NSERC of Canada to G. F. Tibbits.