INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

IN MAMMALIAN VENTRICULAR MYOCYTES, most of the Ca²⁺ that activates contraction is released from the sarcoplasmic reticulum (SR) through Ca²⁺-induced Ca²⁺ release (6, 13, 32), whereas direct activation of contraction by Ca²⁺ flux across the sarcolemma occurs in the amphibian heart (6, 14, 17, 26). In contrast, the information about mechanisms involved in the excitation-contraction (E-C) coupling in the heart of other lower vertebrates is sparse and largely based on studies of the heart in vivo or in situ or in multicellular preparations (see Refs. 10 and 37 for reviews). Ultrastructural studies (30, 39, 40) have shown that cardiac myocytes from teleosts are similar to amphibian myocytes in having small diameters ranging from 2 to 8 µm. This suggests that diffusion in these myocytes may not limit a direct activation of the myofilaments by Ca²⁺ fluxes across the sarcolemma, and it has been shown that Ca²⁺ flux through L-type Ca²⁺ channels is sufficient to activate contraction in frog ventricular myocytes (17, 26). In agreement with this, recent studies show that L-type Ca²⁺ current in both carp and trout ventricular myocytes may account for a significant amount of the total Ca²⁺ transient (39, 40), and several studies using multicellular preparations have failed to show any effect of inhibition of the SR function with ryanodine under physiological conditions (9, 22). However, inhibition of the SR function with ryanodine has been reported to affect contraction in some species (12, 24). In particular, ryanodine inhibits postrest potentiation of contraction (12, 18, 25), which is attributed to increased SR Ca²⁺ release in the mammalian heart (6). Furthermore, inhibition by ryanodine of both postrest contractions and steady-state contractions increases with increasing temperatures, being insignificant below 15°C (18, 25). This confirms ultrastructural data showing that teleost myocytes do posses an SR (30). However, in contrast to mammalian cells, trout ventricular myocytes do not have transverse tubules, but both the SR and the myofibrils are located superficially beneath the sarcolemma (30, 40). Moreover, measurements of Ca²⁺ uptake in crude homogenates (11) as well as ryanodine binding to SR vesicles (35) confirm the presence of a functional SR in the trout heart. Thus the present information about the functional significance of the SR in the teleost heart is ambiguous in multicellular preparations. At the cellular level, information is practically absent and the relative importance of the SR and transsarcolemmal Ca²⁺ fluxes in the E-C coupling in lower vertebrates still relies on extrapolations from the amphibian heart, which still serves as a general model for lower vertebrates. The aim of the present study was therefore to characterize SR Ca²⁺ transport and load in trout ventricular myocytes. We have used a methodology similar to that used in mammalian cells (21, 27). However, in contrast to previous reports from mammals that have focused on steady-state SR Ca²⁺ content, we have designed specific stimulation protocols that also permit a quantification of the maximal SR Ca²⁺ content and SR Ca²⁺ uptake rates. Some of this work has been presented in abstract form (23).

	METHODS

Top Abstract Introduction Methods Results Discussion References

Cell isolation. Rainbow trout were obtained from a commercial trout farm and were kept in tanks at 16°C with a 12:12-h light-dark photoperiod. Myocytes were obtained by enzymatic digestion using a protocol similar to that described for frogs (16). In brief, rainbow trout (Oncorhynchus mykiss) were killed by stunning and decapitation. The heart was rapidly excised, and a cannula was inserted into the ventricle. The heart was then rinsed for 5-10 min with nominally Ca²⁺-free Tyrode solution containing (in mM) 125 NaCl, 1.8 MgCl₂, 4 NaHCO₃, 0.8 NaH₂PO₄, 10 HEPES, 5 glucose, and 5 pyruvate. The pH was adjusted to 7.4 with NaOH. The flow rate during the rinse was gradually increased from <1 ml/min to 4 ml/min. The heart was then perfused two times for 25 min at 20°C with a nominally Ca²⁺-free Tyrode solution containing 50 µM EGTA and 37.5 µM Ca²⁺, 0.1 mg/ml collagenase (Yakult), 0.1 mg/ml trypsin (Sigma, St. Louis, MO), and 0.5 mg/ml BSA. After digestion, the ventricle was cut open in a nominally Ca²⁺-free Tyrode solution containing vitamins, amino acids, penicillin, and 1 mg/ml BSA (16). The tissue was gently agitated, and the supernatant was filtered through a nylon filter (100 µm). The remaining tissue was resuspended, agitation was repeated, and the whole suspension was filtered. After filtration, cells were washed one time and Ca²⁺ was gradually increased to 750 µM. The cells were then stored at 6°C until use. Cells could be stored at this temperature up to 1 wk without any visible changes in contraction or Ca²⁺ current. However, cells were only used within 48 h after isolation for patch clamp experiments, and SR Ca²⁺ uptake in permeabilized myocytes was measured only on the day of cell isolation.

Measurement of free Ca²⁺ in cell suspensions. The free Ca²⁺ in a myocyte suspension was measured using a combination of Ca²⁺-selective electrodes and indo 1 fluorescence that allowed in situ calibration of the indo 1 signal (19, 21). Briefly, Ca²⁺ was calculated from the indo 1 fluorescence emission measured at 465 nm (F465) and at the isosbestic wavelength (FIB). Figure 1A shows the continuous measurement of F465 during an experiment. The downwards deflection in the fluorescence signal is due to Ca²⁺ additions to the cell suspension. To follow temporal changes in FIB, we performed indo 1 emission scans throughout the whole experiment (asterisks and letters in Fig. 1, A and B). At the end of each experiment, a saturating dose of thapsigargin (TG) was added to the cell suspension and a Ca²⁺ titration was done. This allowed us to determine FIB and to use the fluorescence ratio FIB/F465 to calculate the free Ca²⁺ in the cell suspension. Figure 1B shows indo 1 emission scans performed before (A) and after the TG addition (B-F). The free Ca²⁺ in the cell suspension was measured simultaneously with a Ca²⁺-selective minielectrode that allowed us to obtain a ratio calibration for indo 1 for each individual experiment (Fig. 1C). The data were fit with a Hill equation, and the parameters maximum and minimum fluorescence and binding constants K_0.5 and n_H were then used to calculate the free Ca²⁺ in the cell suspension from the FIB/F465 ratio.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Measurement of free Ca2+ in myocyte suspensions. A: continuous measurement of indo 1 fluorescence emission at 465 nm (F465). Downward deflections in F465 are due to Ca2+ additions. Emission scans are marked with asterisks and letters. B: emission scans performed before (A) and after addition of a saturating dose of thapsigargin (TG; B-F). Isosbestic fluorescence (FIB) and wavelength () were determined at crossing of scans. C: fluorescence ratio FIB/F465 obtained from scans in B was plotted against free Ca2+, which was measured simultaneously with Ca2+-selective electrodes. Points were fit with a Hill equation, and resulting parameters were used to calculate free Ca2+ in cell suspension.

Measurement of SR Ca²⁺ uptake in trout ventricular myocyte suspensions. Isolated myocytes were washed in a KCl buffer and suspended in an intracellular medium containing (in mM) 140 KCl, 10 NaCl, 10 HEPES, 1 Na₂ATP, 10 phosphocreatine, 0.1 Na₂GTP, 10 glucose, 10 potassium oxalate, 0.006 indo 1, and 0.020 ruthenium red. The pH was adjusted to 7.2 using KOH. Cells were permeabilized with digitonin (0.3-1 µM), and Ca²⁺ uptake was induced by addition of 10 µM Ca²⁺ to the cell suspension. When Ca²⁺ uptake was inhibited with TG, the drug was always added at low free-Ca²⁺ concentrations. The SR Ca²⁺ uptake was calculated from the free Ca²⁺ measured in the cell suspension using indo 1, as described in Measurement of free Ca²⁺ in cell suspensions. A detailed description of the method has previously been given (21) and is summarized in Fig. 2, A and B. In experiments in which the temperature was lowered from 20°C to 10 or 5°C, the pH was adjusted to 7.2 at the beginning of the experiment. The Ca²⁺ concentration in the cell suspension was measured online, and a calibration of the electrode response was done at the end of the experiment. Because only one calibration can be done, it was done at either 20°C or the corresponding low temperature, and the data were pooled.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Temperature dependency of sarcoplasmic reticulum (SR) Ca2+ uptake. A: Ca2+ addition to cell suspension at arrow caused a transient increase in free [Ca2+]. B: total Ca2+ in indo 1 accessible volume was calculated (as sum of free [Ca2+] and Ca2+ bound to oxalate, indo 1, and cellular protein) from measured free [Ca2+] in A. Its time derivative (dCa/dt) was plotted against time. C: SR Ca2+ uptake rate (dCa/dt) was plotted against corresponding free [Ca2+] and fit with a Hill equation (solid lines). D: SR Ca2+ uptake rate determined at 1 µM free Ca2+ (dashed lines in C) at 10 and 5°C was normalized to corresponding uptake rate at 20°C. * Significant difference from 20°C (P < 0.01).

Caf contractures. Caffeine (Caf) contractures were evoked by exposure of trout ventricular myocytes to 10 mM Caf. After formation of a gigaseal, the myocyte was lifted up from the bottom of the experimental chamber and placed in front of one of several parallel polyethylene tubes (280 µm ID) containing the experimental solutions. Use of a relatively high flow rate (between 1 and 2 cm/s) allowed an effective change of the external solution in less than a few hundred milliseconds (see Maximal contraction). Ionic currents were measured using a software-driven patch-clamp amplifier (EPC-9; Heka). Unless otherwise stated, a standard holding potential of 80 mV was used. In experiments used to determine the maximal SR Ca²⁺ content and uptake rates (Figs. 9 and 10), cells were repolarized to 30 mV after SR Ca²⁺ loading to avoid net gain or loss of Ca²⁺ from the SR before the Caf exposure.

Solutions. Internal medium consisted of the following (in mM): 100 CsCl, 3.1 Na₂ATP, 4 MgCl₂, 5 sodium phosphocreatine, 0.42 Na₂GTP, 0.025 EGTA, 10 HEPES, and 20 tetraethylammonium. The pH was adjusted to 7.2 with CsOH. External medium consisted of the following (in mM): 107 NaCl, 20 CsCl, 1.8 MgCl₂, 4 NaHCO₃, 0.8 NaH₂PO₄, 1.8 CaCl₂, 10 HEPES, 5 glucose, and 5 pyruvate. The pH was adjusted to 7.4 with NaOH. Sodium current was eliminated with 3 µM TTX.

Maximal contraction. Maximal contraction and resting cell length was measured from individual video frames. This allowed us to determine the lag time between Caf exposure and the onset of the Caf contracture that varied between 100 and 300 ms, depending on the flow rate. All chemicals were from Sigma unless otherwise stated.

Statistics and data analysis. Student's t-test for paired and unpaired samples was used to evaluate statistical significance, and fitting of data with exponential or Hill equations was done using SigmaPlot (Jandel Scientific, San Rafael, CA) software.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Net SR Ca²⁺ uptake is not abolished by low temperatures in trout. SR Ca²⁺ uptake was induced in permeabilized myocytes by addition of 10 µM Ca²⁺ to a cell suspension at 20°C. This increased the free Ca²⁺ transiently from a baseline around 0.1 µM to a peak around 1.5 µM. After the peak, the free Ca²⁺ slowly returned to the baseline (Fig. 2A). This represents removal of Ca²⁺ from the cell suspension by an intracellular compartment that is not accessible to indo 1. In the presence of 20 µM ruthenium red and 10 mM oxalate, the total Ca²⁺ removed from the cell suspension corresponds to the Ca²⁺ taken up by the SR. Using binding constants of K₁ = 0.42 µM, n₁ = 1.27 nmol Ca²⁺/mg cell protein and K₂ = 79 µM, n₂ = 4.13 nmol Ca²⁺/mg cell protein for Ca²⁺ binding to cellular protein (20), Ca²⁺ binding to indo 1 (K_d = 0.25 µM), and Ca²⁺ binding to oxalate (K_d = 4 mM), we could calculate the total Ca²⁺ in the indo 1 accessible volume (Ca²⁺_t) for a given measured free Ca²⁺ (see Ref. 21). Thus the Ca²⁺ uptake rate in the SR is represented by the change in Ca²⁺_t with time (dCa/dt in Fig. 2B). In each of 10 experiments, the SR Ca²⁺ uptake (dCa/dt) was plotted against the free Ca²⁺ and fit with a Hill equation, as shown in Fig. 2C, giving a maximum uptake rate (V_max) of 4.4 ± 0.8 nmol Ca²⁺ · mg cell protein¹ · min¹, an n_H of 2.5 ± 0.2, and a K_0.5 of 0.87 ± 0.11 µM at 20°C. After at least two Ca²⁺ additions at 20°C (Fig. 2, circles and squares), the temperature was gradually lowered to 10 or 5°C (Fig. 2, triangles). As shown in Fig. 2A, lowering the temperature had only a small effect on the baseline Ca²⁺ but slowed the return of the free Ca²⁺ to the baseline after a Ca²⁺ addition. The temperature effect was reversible, because a return to 20°C restored the Ca²⁺ uptake rate (Fig. 2). To estimate the effect of temperature on the SR Ca²⁺ uptake rate, we normalized the uptake rate at 5 (n = 6) or 10°C measured at 1 µM free Ca²⁺ (n = 5) to the corresponding uptake rate at 20°C and 1 µM free Ca²⁺. When the data of each individual experiment were fitted with a Hill equation, only the V_max was changed significantly (see Fig. 2C). As shown in Fig. 2D, the uptake rate at 1 µM free Ca²⁺ was reduced by 50% at 10°C and 63% at 5°C, giving an average Q₁₀ of 1.6. The average reduction of the V_max obtained from the Hill fits was similar to the reduction of the Ca²⁺ uptake at 1 µM free Ca²⁺.

Passive Ca²⁺ leak from SR. The aforementioned experiments suggest that not only do trout ventricular cells possess an SR capable of accumulating Ca²⁺, but the SR also exhibits a net Ca²⁺ uptake at temperatures at which the SR from intact mammalian ventricular myocytes exhibits a net Ca²⁺ release due to an increased Ca²⁺ leak from the SR. To examine whether the measured reduction in net SR Ca²⁺ uptake rate at low temperatures was due to an increase in passive Ca²⁺ leak from the SR or due to a decrease in SR Ca²⁺ uptake, we examined the effect of a lowered temperature on passive Ca²⁺ leak from the SR in permeabilized myocytes. To measure the passive Ca²⁺ leak, the SR was first loaded with Ca²⁺ by several Ca²⁺ additions to the cell suspension. The SR Ca²⁺ pump was then blocked by addition of a saturating dose of 5 µM TG to the cell suspension. As shown in Fig. 3A, addition of TG caused the onset of a continuous increase in the free Ca²⁺ concentration. Calculation of the change in Ca²⁺_t with time in 15 experiments yielded the passive Ca²⁺ leak rates depicted in Fig. 3B (measured at a free Ca²⁺ of 0.9 ± 0.1 µM). When normalized to the Ca²⁺ uptake rate at the corresponding temperature, the passive Ca²⁺ leak amounts to 15% of the SR Ca²⁺ uptake rate at all the temperatures examined. Thus the passive Ca²⁺ leak from the SR decreases at lower temperatures, and the decrease in net SR Ca²⁺ uptake is due to a decrease in the Ca²⁺ uptake rate rather than an increase in the passive Ca²⁺ leak.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Effect of temperature on passive Ca2+ leak from SR. A: at left arrow, 10 µM total Ca2+ is added to cell suspension to load SR with Ca2+. After return of free [Ca2+] ([Ca2+]f) to baseline, a saturating dose of TG was added at right arrow. B: passive Ca2+ leak from SR was calculated from measured increase in [Ca2+]f as time derivative of calculated total Ca2+. Calculated dCa/dt from 5 experiments is shown as a function of temperature. * Significant difference from 20°C (P < 0.05).

Titration of SR Ca²⁺ pumps with TG. If it is assumed that TG binds to the SR Ca²⁺ pump with a 1:1 stoichiometry (29) and that the K_d for TG binding is much lower than the concentration of SR Ca²⁺ pumps, then the number of SR Ca²⁺ pumps should be equal to the total amount of TG added to the cell suspension when inhibition of the SR Ca²⁺ uptake is one-half maximal. We therefore examined the inhibition of the SR Ca²⁺ uptake at nonsaturating TG concentrations. Figure 4A shows the transient increase in free Ca²⁺ after addition of 10 µM Ca²⁺ to the cell suspension in the absence and presence of 500 nM TG. TG caused a small dose-dependent increase in the baseline and a dose-dependent reduction in the SR Ca²⁺ uptake after a Ca²⁺ addition. When Ca²⁺-dependent SR Ca²⁺ uptake was fitted with a Hill equation, only the V_max was changed significantly by TG (Fig. 4B). In 10 experiments with TG ranging from 0.1 to 2 µM, the V_max was significantly reduced by 20-90% of the control value. On the contrary, the average K_0.5 and n_H were not significantly different from the corresponding values in the absence of TG.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Titration of SR Ca2+ pump sites with TG. A: 2 experiments showing superimposed traces of Ca2+ transient elicited by Ca2+ addition to cell suspension in absence (Con) and presence of a nonsaturating dose of TG. B: Ca2+ dependency of calculated SR Ca2+ uptake rate (calculated from A) in presence (TG) and absence (Con) of a nonsaturating dose of TG. Solid line represents a fit of data with a Hill equation. Arrows indicate SR Ca2+ uptake rates at 1 µM free Ca2+ in the 2 conditions. C: uptake rate at 1 µM free Ca2+ in presence of TG was normalized to corresponding uptake rate in Con and plotted against total TG added. Normalized data were fit with a Hill equation (solid line).

SR Ca²⁺ content measured from Caf contractures. Because the results from permeabilized myocytes showed that trout ventricular myocytes possess an SR that is potentially capable of participating in the beat-to-beat regulation of contraction, it was important to confirm the results from permeabilized cells in intact cells. We therefore used the contracture and the inward Na⁺/Ca²⁺ exchange current (I_NCX) elicited by exposure of the cells to 10 mM Caf to estimate the SR Ca²⁺ content. Figure 5A, left, shows the time course of a Caf contracture elicited after loading of the SR with Ca²⁺ for 5 min. Figure 5A, right, shows a relaxed trout ventricular myocyte (Fig. 5Aa) and the same myocyte at the peak contracture during exposure to 10 mM Caf (Fig. 5Ab). The average cell length was 176 ± 11 µm at rest and 111 ± 9 µm at the peak of the Caf contracture (n = 8). The time course of the Caf-induced I_NCX in the continued presence of Caf was faster than that reported in mammalian cells (8, 27, 28), with an average duration of 1.07 ± 0.04 s (n = 45). Figure 5, B and D, shows the inward current elicited by Caf and the L-type Ca²⁺ current elicited by a 200-ms depolarization to 0 mV used to load the SR, respectively. The charge carried by the inward current and the Ca²⁺ current in Fig. 5, C and E, was obtained as the time integral of the respective currents. Assuming that the inward current during the Caf contracture is due to Ca²⁺ extrusion by the NCX, the total Ca²⁺ released from the SR can be calculated from the charge carried by the I_NCX. Using a surface-to-volume ratio of 1.15/µm for trout myocytes, a specific capacitance of 1.66 µF · cm², and 55% nonmitochondrial cell volume (39, 40), a cell capacitance of 30.4 ± 3.4 pF and a total charge of 65 ± 9 pC or 674 ± 93 amol Ca²⁺ gives a steady-state SR Ca²⁺ content of 750 µmol Ca²⁺/l nonmitochondrial cell volume (µM), whereas a charge carried by the L-type Ca²⁺ current of 2.90 ± 0.37 pC or 15.0 ± 1.9 amol Ca²⁺ corresponds to 17 µM. Notice that the Ca²⁺ carried by the Caf-induced I_NCX is 40- to 50-fold larger than that carried by the L-type Ca²⁺ current.

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Caffeine (Caf) induces an inward ionic current and cell contracture. A: cell length was measured from individual video images and plotted against time (left). Relaxation of Caf-induced cell shortening was fit with a single exponential (solid line), giving a time constant of 0.52 s. Video images of relaxed (a) and maximally contracted (b) trout ventricular myocyte are shown at right. Calibration bar represents 25 µm. B: exposure of cell to 10 mM Caf induced a transient inward ionic current. C: time integral of inward current in B. D: L-type Ca2+ current elicited by a 200-ms depolarization to 0 mV. E: time integral of L-type Ca2+ current, which was considered to be area delimited by current trace and current at end of depolarization (area below solid line in D).

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Caf-induced current is due to Ca2+ extrusion by Na+/Ca2+ exchanger. A: Ca2+ release from SR was induced by exposure to 10 mM Caf in presence (Con) or absence of Na+ and Ca2+ (0 Na+ and 0 Ca2+, respectively), containing Li+ instead of Na+ and 5 mM EGTA instead of Ca2+. A 5-s depolarization to 30 mV was used to obtain a moderate SR Ca2+ load. B: membrane current corresponding to experimental protocol in A. C: superimposed traces of current elicited by Caf in presence and absence of extracellular Na+ and Ca2+. In 4 cells, no Na+/Ca2+ exchange current (INCX) was seen in absence of extracellular Na+ and Ca2+.

View larger version (23K): [in this window] [in a new window]   Fig. 7.   Dependency of Caf-induced INCX on SR Ca2+ loading conditions. A: SR Ca2+ loading protocol. First brief Caf exposure was done after continuous stimulation for at least 5 min (SS). Second Caf exposure was done 30 s after the first without any stimulation (Rest). Last Caf exposure was done 30 s after the second, with 24 intermittent stimulation pulses (24P). B: inward currents elicited by Caf after 3 different SR Ca2+ loading protocols in A. C: summary of dependency of charge carried by inward current on SR Ca2+ loading conditions (n = 14). * Significant difference (P < 0.0001) from SS; $ significant difference from Rest (P < 0.002). D: maximal cell shortening was normalized to resting cell length and plotted against charge carried by corresponding inward current. , Obtained with experimental protocol shown in A; , obtained from data in Fig. 8.

Net Ca²⁺ loss from SR does not occur during moderate rest periods. In mammalian cells, the SR loses Ca²⁺ during rest, although the net leak from the SR is species dependent. To examine the net SR Ca²⁺ loss during rest periods in trout ventricular myocytes, the SR was first loaded with Ca²⁺ using 40 standard depolarizations at a frequency of 0.125 Hz. Caf was then applied after 0, 120, and 320 s without stimulation. Figure 8A shows the Caf-induced I_NCX elicited after rest periods of 0 and 320 s. Figure 8B summarizes the SR Ca²⁺ content after different rest periods (given in seconds) in eight myocytes, and Fig. 8C shows the average cell contracture after the same rest intervals. Notice that trout ventricular myocytes did not exhibit any significant decay in SR Ca²⁺ content or cell contracture at rest periods up to 5 min.

View larger version (18K): [in this window] [in a new window]   Fig. 8.   Effect of rest periods on SR Ca2+ content. A: inward current elicited by Caf after 320 s of continuous stimulation. Left: current elicited immediately after last stimulation pulse. Right: current elicited 320 s after last stimulation pulse. B: summary of SR Ca2+ content as a function of rest period (given in seconds below each bar). SR Ca2+ content was evaluated as charge carried by INCX elicited by Caf in 8 cells. C: summary of Caf-induced cell shortening as a function of rest period in 8 cells. Cell shortening was normalized to resting length in absence of Caf.

Maximal SR Ca²⁺ content. To estimate the maximal SR Ca²⁺ load, an SR loading protocol using increasingly longer depolarizations to +50 mV before exposure to Caf at a holding potential of 30 mV was used. It should be noticed that depolarizations to +50 mV induced a cell contracture that relaxed on repolarization to 30 mV before Caf exposure (data not shown). Figure 9A shows loading protocols with 2-, 8-, and 16-s depolarizations to +50 mV, and the corresponding current traces are shown in Fig. 9B. Figure 9C summarizes the dependency of the charge density of the I_NCX induced by Caf exposure on the length of the loading pulse. When the length was increased from 16 to 32 s, an increase in the charge density was still seen, suggesting that the charge density resulting from a 32-s depolarization does not necessarily load the SR maximally. However, when the loading period was prolonged to 64 s, spontaneous contractions were observed in three out of six cells, and the SR Ca²⁺ content was lower than with the 32-s loading period (not shown). Thus a maximal Caf-induced I_NCX of 5.02 ± 0.57 pC/pF or 52.0 ± 5.9 amol Ca²⁺/pF (n = 5) was obtained with a loading period of 32 s.

View larger version (13K): [in this window] [in a new window]   Fig. 9.   Effect of loading period on SR Ca2+ content. A: stimulation protocol for SR Ca2+ loading. A short Caf exposure at a holding potential of 30 mV was used to evaluate effect of duration of a preceding depolarization to +50 mV. B: membrane current in response to stimulation protocols in A. C: summary of dependency of SR Ca2+ content on duration of a depolarization to +50 mV (n = 5).

SR Ca²⁺ uptake rates. The steady-state SR Ca²⁺ content measured as the Ca²⁺ released during a Caf contracture is 2.5 times larger than that reported for mammalian cells (27, 8), and the maximal SR Ca²⁺ content is 2-10 times larger than reported maximal SR Ca²⁺ contents in mammals (20, 27). However, as shown, the number of beats necessary to load the SR maximally is also larger than that found in mammals, in which fewer than 10 stimulations are generally enough to fully load the SR (5). This could suggest that the SR is capable of accumulating large amounts of Ca²⁺ but that either the rate of Ca²⁺ accumulation is too slow or the Ca²⁺ release mechanism does not permit the SR to participate actively in the activation of contraction. We therefore examined whether we could measure SR Ca²⁺- uptake rates in intact trout myocytes that were high enough to allow a significant Ca²⁺ reaccumulation between two contractions. To do this, we used stimulation protocols that produced a prolonged elevation of the free intracellular Ca²⁺ to concentrations that allowed a maximal Ca²⁺ uptake. First, the SR was emptied by a brief Caf exposure, and the cells were then depolarized to potentials between +10 and +100 mV for 3 s to load the SR. The membrane potential was then repolarized to 30 mV before the cell again was exposed briefly to 10 mM Caf. Figure 10A, left, shows this stimulation protocol, and Fig. 10A, right, shows the corresponding current trace with a loading potential of +10 mV. Assuming that the SR Ca²⁺ content after the second Caf exposure was largely due to SR Ca²⁺ uptake during the 3-s depolarization, the SR Ca²⁺-uptake rate can be calculated. Figure 10B shows I_NCX at 30 mV after loading potentials of +10, +30, +60, and +80 mV, i.e., the part of the current trace in Fig. 10A where the cell is exposed to Caf. Figure 10C summarizes the dependency of the calculated SR Ca²⁺ uptake rates on the test potential (values were obtained from 5-13 cells). At 100 mV, the total Ca²⁺ taken up by the SR was 36.6 ± 3.8 amol/pF, giving a near-maximal SR Ca²⁺ uptake rate of 12.2 ± 1.2 amol Ca²⁺ · pF¹ · s¹ or 417 µM/s.

View larger version (15K): [in this window] [in a new window]   Fig. 10.   Effect of membrane potential on SR Ca2+ uptake. A: left shows stimulation protocol used to evaluate SR Ca2+ uptake rates. SR Ca2+ content was evaluated at a holding potential of 30 mV after a preceding depolarization for 3 s to potentials between +10 and +100 mV (Vtest). Immediately before this, SR Ca2+ content had been discharged. Right panel shows membrane current with a Vtest of +10 mV. B: typical inward currents elicited by Caf at 30 mV with 4 different Vtest values given below each trace. C: summary of dependency of SR Ca2+ uptake rate on Vtest in 4-6 cells. SR Ca2+ uptake rate was calculated assuming that uptake was constant over the 3-s depolarization.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

SR Ca²⁺ content in cardiac myocytes. In lower vertebrates, the SR has generally been considered a poorly developed organelle of minor importance in the E-C coupling (see Refs. 10 and 37 for reviews). This is based mainly on morphological and ultrastructural data showing that heart cells from lower vertebrates are long, thin cells lacking T tubules and with an SR that is poorly developed compared with mammals (30, 39, 40). Cardiac myocytes from frogs are well characterized and appear to represent an extreme, having cell diameters of 2-3 µm, showing a lack of Ca²⁺-induced Ca²⁺ release in mechanically skinned fibers (14) and no effect of ryanodine (5), and having a very low oxalate-supported SR Ca²⁺ uptake in crude homogenates (11). The results of the present work are contrary to this general assumption. Thus the steady-state SR Ca²⁺ content in trout obtained after >5 min of continuous stimulation (Figs. 5 and 7) was two to five times larger than values reported in mammalian myocytes at room temperature with use of a similar methodology (8, 27, 34). Furthermore, when using a stimulation protocol specifically designed to progressively saturate the SR with Ca²⁺, we obtained a maximal SR Ca²⁺ content in trout ventricular myocytes of 52 amol Ca²⁺/pF. This is six to ten times higher than the steady-state values reported in mammals (8, 27).

Quantification of SR Ca²⁺ uptake rates. When assessing the importance of the SR in the beat-to-beat regulation of contraction, it is, however, the SR Ca²⁺ uptake rate that determines whether the SR is able to resequester Ca²⁺ between beats. In this respect, it has previously been reported that in teleosts the SR does not appear to participate in the regulation of contraction at physiological temperatures (<15°C) and beating frequencies (>0.25 Hz) (24, 25). Results from oxalate-supported Ca²⁺ uptake in crude homogenates (11) also suggest that a maximal Ca²⁺ uptake rate, which is 10 times less than that in rat at 37°C, is too slow to allow the SR to reaccumulate Ca²⁺ between beats at 15°C. In contrast to this, ryanodine turns postrest potentiation into rest decay and reduces steady-state contraction by 50% at 25°C (18). Using permeabilized myocytes, we obtained maximal SR Ca²⁺ uptake rates that are two- to fourfold smaller than in mammalian cells with use of the same technique at room temperature (20). However, the trout ventricular cell suspensions with an average protein content ~1 mg/ml are three- to fivefold more diluted than the mammalian suspensions. This may cause an underestimation of the Ca²⁺ uptake rate, because it is calculated from the measured free Ca²⁺, which in turn will decrease more slowly at lower cell densities.

1

1

1

1

1

1

Passive Ca²⁺ buffering in trout. A maximal measurable SR Ca²⁺ uptake rate of 417 µM/s is similar to that reported in mammalian myocytes with use of a different technique (2). A direct comparison to mammals is, however, complicated, because the physiological temperature range is different and the Ca²⁺ uptake rate in mammals is about three times larger at their body temperature (41). Furthermore, it may be argued that the passive Ca²⁺ buffering is different in trout ventricular myocytes. Considering information about myofibrillar volume in trout, which is similar to values reported for mammals (6, 40), we do not, however, expect large differences between trout and mammals with respect to Ca²⁺ binding to the myofilaments. Furthermore, Fig. 7D shows that in the presence of Caf, a total charge of 0.2-0.4 pC/pF (corresponding to 2-4 amol of Ca²⁺/pF) is needed to activate a cell shortening of 10-15% of the resting cell length. Because Caf is known to sensitize the myofilaments, this value should represent a lower estimate of the total Ca²⁺. Assuming a Ca²⁺-accessible cell volume of 70%, as reported in mammals (8), the total Ca²⁺ would be required to increase by 55-109 µM. This value is within the expected increase in total Ca²⁺ in mammalian myocytes (4, 8, 20), suggesting that passive Ca²⁺ buffering in trout is indeed within the same range as that reported in mammals. In trout, the Ca²⁺-accessible cell volume is, however, <70%. Thus the nonmitochondrial volume has been reported to amount to 55% of the cell volume (40). Because of the superficial position of the myofilaments, it may even be argued that the relevant Ca²⁺ accessible volume is limited to the 40% that the myofibril occupies. With the two latter assumptions, the total Ca²⁺ would be required to increase by 69-139 or 96-191 µM to activate a 10 to 15% cell shortening, respectively.

Removal of Ca²⁺ from cytosol. This then suggests that a maximal SR Ca²⁺ uptake rate of 12.2 amol Ca²⁺ · pF¹ · s¹ should be more than sufficient to remove an expected total Ca²⁺ transient of 2-4 amol Ca²⁺/pF between two consecutive beats at physiological heart rates ranging from 0.5 to 1.5 Hz (15). Indeed, if the SR is working near V_max, it should be able to resequester 60-120% of the total Ca²⁺ transient during a 200-ms plateau phase of the action potential, where Ca²⁺ extrusion by the NCX is reduced or absent due to the depolarized membrane potential.

1

1

Contribution of SR Ca²⁺ to contraction. The fact that the maximal SR Ca²⁺ uptake rate is sufficiently high to allow the SR to reuptake a considerable amount of the total Ca²⁺ transient does not necessarily imply that the SR contributes with a large fraction of the total Ca²⁺ that activates contraction. Indeed, the SR has to be able to liberate and retain the accumulated Ca²⁺.

Perspectives