INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

INFLUX of Ca²⁺ through the cardiac L-type Ca²⁺ channel is the primary pathway for the triggering of Ca²⁺ release from the sarcoplasmic reticulum (SR) in mammalian cardiac myocytes (7). The localized nature of the Ca²⁺-release process has been demonstrated by confocal measurements of "Ca²⁺ sparks," which are thought to reflect Ca²⁺ release from a cluster of ryanodine receptors either spontaneously at rest or as triggered by activation of nearby Ca²⁺ channels (5, 6, 18). This could imply that Ca²⁺ signaling occurs within microdomains over very short distances (12 nm; Ref. 25) and may be largely insensitive to Ca²⁺ concentration in the global cytosolic pool of the intracellular compartment. Previously, we tested this hypothesis by examining Ca²⁺ signaling in rat ventricular myocytes in which the diffusion distance of free Ca²⁺ was reduced to <50 nm by buffering cytosolic Ca²⁺ with Ca²⁺ chelators (2 mM fura 2 and 14 mM EGTA) (1, 30). Under such conditions, although cytosolic free Ca²⁺ concentration ([Ca²⁺]_i) was reduced and the current generated by the Na⁺/Ca²⁺ exchanger was suppressed, effective Ca²⁺ cross signaling between dihydropyridine (DHP) and ryanodine receptors persisted such that the activation of Ca²⁺-channel current (I_Ca) triggered a normal Ca²⁺ release (total Ca²⁺ release  140 µM; Ref. 1) from the SR and the released Ca²⁺, in turn, inactivated the Ca²⁺ channel. Such Ca²⁺ cross signaling provides for an efficient, locally controlled, negative feedback circuit and suggests that high concentrations of intracellular Ca²⁺ buffers can be used to study Ca²⁺ signals in the microdomain surrounding the DHP and ryanodine receptors without significant interference from the global cytosolic Ca²⁺ concentrations.

The ()-isomer of BAY K 8644 is known to increase L-type Ca²⁺-channel current (I_Ca) and contraction (16, 23). The increase in I_Ca results from both enhanced open probability and a shift in the modal gating of the channel (13). In control myocytes, BAY K 8644 (at 0.1-1.0 µM) shifts the voltage dependence of activation and inactivation of I_Ca by 10 to 15 mV, accelerates both the activation and inactivation rates, and significantly slows the deactivation rate (23). On the other hand, when the L-type Ca²⁺ channels are phosphorylated by cAMP-dependent protein kinase (PKA), BAY K 8644 significantly slows the inactivation rate of I_Ca (33, 34).

The effect of BAY K 8644 on cardiac excitation-contraction (E-C) coupling has been examined in detail in papillary muscle and isolated cardiomyocytes from dog and ferret heart at 30-37°C (21, 22). These results show that BAY K 8644 not only abolishes rested-state potentiation by increasing the leak of Ca²⁺ from the SR stores but also suppresses Ca²⁺ release for a given I_Ca and SR Ca²⁺ load, suggesting that these effects may reflect an altered state of the ryanodine receptors due to binding of BAY K 8644 to DHP receptors.

The aim of this study was to examine the regulatory role of Ca²⁺-channel gating in Ca²⁺ cross signaling between DHP and ryanodine receptors by quantifying the voltage, Ca²⁺, and drug dependence of the amplification factor of the Ca²⁺-release mechanism. The experiments were conducted at room temperature on rat ventricular myocytes dialyzed with 2 mM fura 2 and 200 µM cAMP to limit the Ca²⁺ diffusion distance to <50 nm, thereby suppressing Ca²⁺ signaling outside Ca²⁺ microdomains (1). The addition of 200 µM cAMP to the dialyzing solution was required to prevent rundown of I_Ca and keep intact the Ca²⁺ content of SR during the long experimental periods (>20 min).

Our studies show that the amplification factor represents a unique and inherent property of DHP-/ryanodine-receptor complex. The exponential voltage dependence of the amplification factor and its suppression by BAY K 8644 are consistent with a scheme in which the rate of inactivation of the Ca²⁺ channel serves as a regulator of Ca²⁺-induced Ca²⁺ release.

A preliminary report of this study has already appeared (2).

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Single ventricular myocytes. Adult rat ventricular myocytes were isolated as described previously (24). Briefly, rats were deeply anesthetized with pentobarbital sodium (50 mg/kg ip), and hearts were excised quickly and perfused at 7 ml/min in a Langendorff apparatus, first with Ca²⁺-free Tyrode solution composed of (in mM) 137 NaCl, 5.4 KCl, 10 HEPES, 1 MgCl₂, and 10 glucose, pH 7.3, at 37°C for 8 min, then with Ca²⁺-free Tyrode solution containing collagenase (0.5-0.6 U/ml) and protease (0.55 U/ml) for 15 min, and finally with Tyrode solution containing 0.2 mM CaCl₂ for 8 min. The ventricle of the digested heart was then cut into several sections and subjected to gentle agitation to dissociate cells. The freshly dissociated cells were stored at room temperature in Tyrode solution containing 0.2 mM CaCl₂ and were used for 10 h after isolation.

Current recording. Ca²⁺ current was measured in the whole cell configuration of the patch-clamp technique using a Dagan 8900 amplifier (Dagan, Minneapolis, MN). The patch electrodes, made of borosilicate glass capillaries, were fire polished to have a resistance of 1.5-3.0 M when filled with the internal solution composed of (in mM) 110 CsCl, 30 tetraethylammonium chloride (TEA-Cl), 10 HEPES, 5 Mg-ATP 0.1 Li-GTP, 0.2 cAMP, and 2 K₅-fura 2 and titrated to pH 7.4 with CsOH.

Intracellular Ca²⁺ activity. [Ca²⁺]_i (20-150 nM) was measured ratiometrically with fura 2 (11) as previously described (7). For reliable measurements of [Ca²⁺]_i, we required that the fluorescence ratio of cells observed in vivo be the same as that in vitro and that the background fluorescence be minimized by blocking off the electrode (7). Under these conditions we estimated that the absolute resting Ca²⁺ activity could be measured with an accuracy of 10-20 nM and that the sensitivity to changes was 2-5 nM. High time resolution was achieved using a vibrating mirror that alternated between two wavelengths of excitation (335 and 405 nm) at 1,200 Hz (8). The amount of released Ca²⁺ (0-200 µM) was also estimated on the basis of the verified assumptions that the intracellular concentration of fura 2, after equilibration periods in excess of 8 min, approached the dye concentration of 2 mM in the patch pipette (see Figs. 1 and 2 of Ref. 1). These calculations were performed using a custom-made computer program that operated on the measured fluorescence in pCLAMP format and produced calibrated files of [Ca²⁺]_i, concentration of total fura 2 ([fura 2]_tot), and concentration of Ca²⁺ bound to fura 2([Ca²⁺-fura 2]) and their time derivatives.

Materials. Collagenase (type A) was purchased from Boehringer Mannheim (Indianapolis, IN), protease (type XIV, Pronase E) and Mg-ATP were from Sigma (St. Louis, MO), thapsigargin and TTX were from Calbiochem (La Jolla, CA), and K₅-fura 2 was from Molecular Probes (Eugene, OR). ()-BAY K 8644 was purchased from Research Biochemicals International (Natick, MA).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Amplification factor. We have previously shown that Ca²⁺ cross signaling between the L-type Ca²⁺ channel and the ryanodine receptor persists in highly Ca²⁺-buffered ventricular myocytes (1, 30). Under such conditions the gain of the Ca²⁺-release process (amplification factor) was determined by measuring the net transfer of Ca²⁺ and Ba²⁺ through the L-type Ca²⁺ channel, using fura 2 in millimolar concentrations as the dominant divalent buffer and effective indicator of both cations. Because Ba²⁺ does not cause Ca²⁺ release, it was possible to calibrate the fluorescence signal in terms of the equivalent cation charge and determine the extent to which the entry of Ca²⁺ was amplified by the release of Ca²⁺ from the SR. Figure 1, A and B, shows that the amplification factor may also be determined by using cells in which the SR stores are depleted of Ca²⁺ by thapsigargin. The rise of cytosolic Ca²⁺ in such cells, therefore, would be determined solely by the influx of Ca²⁺ through the Ca²⁺ channel. In control myocytes, on the other hand, both Ca²⁺ coming through the L-type Ca²⁺ channels and Ca²⁺ released from the SR contribute to the fura 2 signal (Fig. 1A). Thus amplification factor can be calculated using the equation

(1)

where [Ca²⁺-fura 2] indicates the rise in cytosolic Ca²⁺ bound to fura 2 (a value mostly equivalent to the magnitude of Ca²⁺ release from the SR gated by I_Ca); Q_Ca is the Ca²⁺ charge carried by I_Ca; and subscripts L and D refer, respectively, to fully Ca²⁺-loaded and Ca²⁺-depleted SR. As indicated in Fig. 1, A and B, the Ca²⁺ charge carried by the Ca²⁺channel was quantified by the integration to the time when intracellular Ca²⁺ transients reached 90% of peak value. I_Ca activated by test potentials of 10 mV from V_h of 60 mV triggered Ca²⁺ release with an average amplification factor of 24.0 ± 4.4 (n = 5).

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Measurement and voltage dependence of amplification factor. A and B: calculation of amplification factor by comparison of Ca2+ channel current (ICa) and intracellular Ca2+ transients measured with intact (A) and depleted (B) sarcoplasmic reticulum (SR) Ca2+ stores. This rationale (top) is based on the assumption that the increase in Ca2+ bound to fura 2 ([Ca2+-fura 2]; bottom) is normally due to combined effects of Ca2+ influx as ICa (QCaL; middle left) and triggered release of Ca2+ from SR, but, after depletion of SR by thapsigargin, it is due only to ICa (QCaD; middle right). Amount of charge influx during ICa required to trigger Ca2+ release from SR was obtained by integrating ICa up to time when [Ca2+-fura 2]L reached 90% of peak value. After exposure of myocytes to thapsigargin, overall Ca2+ influx through L-type Ca2+ channel was responsible for [Ca2+-fura 2]D. [fura 2], fura 2 concentration; DHP, dihydropyridine; RYR, ryanodine receptor. C and E: voltage dependence of Ca2+ charge carried by ICa (QCa; C) and intracellular Ca2+ transient ([Ca2+-fura 2]; E) measured under control conditions (loaded SR; ) and after depletion of SR by thapsigargin (depleted SR; ). D: [Ca2+-fura 2] is linearly related to QCa when SR is empty () but not when SR is full (). Voltage dependence of amplification factor in F was calculated from Eq. 1 using data from fully loaded SR from C and D and slope of regression line from E. Test potentials were given every 10 s from a holding potential (Vh) of 60 mV.

Suppression of amplification factor by BAY K 8644. To further characterize the amplification factor and determine its dependence on modifiers of Ca²⁺-channel activity, we used BAY K 8644. This drug is a racemic mixture in which only the ()-isomer behaves as the Ca²⁺-channel agonist (16, 23). In cAMP-dialyzed ventricular myocytes, BAY K 8644 significantly enhanced I_Ca, altered the kinetics of its inactivation, and developed the large and slowly inactivating tail currents generally observed with this drug (Fig. 2, A and B). Intracellular Ca²⁺ transients, however, were either not significantly altered or were somewhat suppressed in the presence of the drug.

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Effect of ()-BAY K 8644 (BayK; 1 µM) on ICa and Ca2+ release in a whole cell patch-clamped rat ventricular myocyte. A and B: traces of ICa (middle) and magnitude of intracellular Ca2+ transients ([Ca2+]i; bottom) elicited by test pulses from Vh of 60 mV (schematics, top) in absence (A) and presence (B) of ()-isomer of BayK. BayK enhanced peak amplitude of ICa. Rate of inactivation of ICa in presence of BayK was slower than that of control under such protein kinase A (PKA)-phosphorylated conditions. Deactivation rate was also markedly slowed by BayK. BayK reduced size of intracellular Ca2+ transients gated by ICa. Current traces are shown after leak subtraction by P/6 protocol. C-F: voltage dependence of magnitude and time course of ICa (C and D) and intracellular Ca2+ transients (E and F) in absence () and presence () of BayK (1 µM). In absence of BayK, rate of inactivation of ICa (1/, inverse inactivation time constant) has essentially the same bell-shaped voltage dependence as ICa. BayK abolishes this bell-shaped relationship and causes ICa to inactivate at a rate that is only slightly retarded by increasing depolarization. Intracellular Ca2+ transients, however, developed with almost the same bell-shaped voltage relationship. Surprisingly, BayK did not increase, or rather slightly decreased, intracellular Ca2+ transients. Rate of Ca2+ release also was not changed by BayK. Each point represents a mean ± SE of 4 experiments. [Ca2+]i and d[Ca2+]i/dt, magnitude and rate of rise of intracellular Ca2+ transients, respectively.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Suppression of amplification factor by BayK at different potentials (C). , Control; , measurements after addition of BayK. Amplification factor was calculated (Eq. 1) from values of QCa (A) and [Ca2+-fura 2] (B) from slope of regression lines (compare Fig. 2D) obtained after addition of thapsigargin (not shown). Inset: sample traces show effects of BayK on membrane current (top) and Ca2+ signal (bottom) at 10 (left) and +30 mV (right).

Does BAY K 8644 reduce the Ca²⁺ content of the SR? In some experiments, BAY K 8644 appeared to increase the basal [Ca²⁺]_i at 60 mV (see e.g., Fig. 2B). To further explore and quantify this effect we specifically examined the effect of BAY K 8644 at the onset of a new V_h. Figure 4A shows that changing V_h from 90 to 60 mV in control myocytes caused a small but slow rise in basal [Ca²⁺]_i. This small increase in basal [Ca²⁺]_i occurred without activation of noticeable inward current (Fig. 4A), most likely caused by a small, sustained increase in open probability of Ca²⁺ channels, leading to partial activation of the ryanodine receptors. Consistent with this idea, a reduction of the extracellular Ca²⁺ or its replacement by Ba²⁺, and addition of 100 µM Cd²⁺, prevented the depolarization-induced rise in [Ca²⁺]_i (3). Such small increases in [Ca²⁺]_i appeared to partially deplete the SR Ca²⁺ content, as indicated by a decrease in the magnitude of the caffeine-triggered Ca²⁺ release (compare Fig. 4, A and B).

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Resting Ca2+ activity was increased by depolarization of Vh from 90 to 60 mV, and addition of BayK augmented this increase. A-D show timing of changes in Vh and application of caffeine (5 mM; hatched bar), membrane current displayed with high gain, and a tracing of [Ca2+]i. A: Vh was changed from 90 to 60 mV and caffeine was applied 4 s later. B: Vh was kept at 90 mV. The same protocol was repeated in presence of BayK (1 µM) (C and D). Typical recordings were obtained from same myocyte.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Summary of change of basal [Ca2+]i (A) and magnitude of caffeine-gated intracellular Ca2+ transient (B) measured at Vh of 60 and 90 mV in absence (control) and presence of BayK (1 µM). Data are means ± SE of 4-6 experiments. Individual symbols correspond to different experiments. * P < 0.05, ** P < 0.01 vs. control at Vh of 90 mV. # P < 0.05, ## P < 0.01 vs. BayK at Vh of 90 mV. ++ P < 0.01 vs. control at Vh of 60 mV.

View larger version (26K): [in this window] [in a new window]   Fig. 6.   Effect of BayK (1 µM) on ICa, intracellular Ca2+ transient, and amplification factor at a Vh of 90 mV. A: change of ICa elicited by test pulses from a Vh of 90 to 10 mV given every 10 s. Fast Na+ current (INa) component was not completely blocked by TTX at a Vh of 90 mV. Current traces recorded at respective times (a-c) indicated in B are superimposed. Traces are shown after leak subtraction by P/6 protocol. B: change of ICa amplitude by BayK. BayK was applied during period indicated by shaded bar. C: change of rate of inactivation of ICa (1/) by BayK. D: change of magnitude of Ca2+ release ([Ca2+-fura 2]) at respective times (a-c) indicated in B. E: change of basal [Ca2+]i as well as intracellular Ca2+ transient by BayK. F: change of amplification factor by BayK.

Effects of elevation of extracellular Ca²⁺ concentration. To investigate whether the actions of BAY K 8644 were directly related to the enhancement of I_Ca by the drug, we examined the I_Ca-enhancing effects of elevation of the extracellular Ca²⁺ concentration ([Ca²⁺]_o). [Ca²⁺]_o levels were switched from 2 to 10 mM in <50 ms for only one depolarizing voltage-clamp step to avoid changing the Ca²⁺ content of the SR significantly. Under control conditions, the rapid change of [Ca²⁺]_o from 2 to 10 mM enhanced the peak amplitude of I_Ca (Fig. 7A), accelerated the rate of inactivation of I_Ca (Fig. 7B), and augmented the intracellular Ca²⁺ transients (Fig. 7C). In the presence of BAY K 8644 (1 µM), even though the addition of 10 mM Ca²⁺ further enhanced I_Ca (Fig. 7D) and the intracellular Ca²⁺ transients (Fig. 7F), the rate of inactivation of I_Ca was not significantly changed (Fig. 7E). Numerical evaluation showed that the amplification factor (Eq. 1) measured at 20 mV was insensitive to [Ca²⁺]_o-dependent changes in I_Ca in both control (26.4 ± 2.0 at 2 mM Ca²⁺ vs. 28.0 ± 4.7 at 10 mM Ca²⁺, n = 3) and BAY K 8644-treated myocytes (13.1 ± 4.2 at 2 mM Ca²⁺ vs. 11.6 ± 2.0 at 10 mM Ca²⁺, n = 3). Thus the enhancement of the amplitude of I_Ca by BAY K 8644 was not by itself sufficient to suppress the amplification factor.

View larger version (24K): [in this window] [in a new window]   Fig. 7.   Comparison of BayK effect with extracellular Ca2+ concentration at 2 and 10 mM. Extracellular Ca2+ concentration was changed from 2 to 10 mM just before third episode for only one episode to prevent a change of SR Ca2+ content. Test pulses were given every 10 s. In absence of BayK, increase in extracellular Ca2+ augmented magnitude of ICa (A), accelerated its inactivation as seen from normalized currents (B), and increased [Ca2+]i (C). In presence of BayK, ICa (D) and [Ca2+]i (F) are also augmented by increase of extracellular Ca2+; however, rate of inactivation of ICa was unchanged and remained slow (E).

View larger version (21K): [in this window] [in a new window]   Fig. 8.   Voltage dependence of ICa (A), its rate of inactivation (1/; B), intracellular Ca2+ transient (C), and gain of Ca2+-release system (D) measured at extracellular Ca2+ concentrations of 2 and 10 mM. Gain was measured as intracellular Ca2+ transient relative to integral of ICa ([Ca2+-fura 2]L/QCaL).

BAY K 8644 effect on Ca²⁺-induced inactivation of I_Ca. Figure 9 examines the dependence of the rate of inactivation of I_Ca on the rate of release of Ca²⁺ from the SR. Consistent with our previous findings (1), in control solutions the rate of inactivation of I_Ca [defined as the inverse of the time constant for the decay of I_Ca (1/)] depended linearly on the magnitude and rate of release of Ca²⁺ (Fig. 9, open circles). Transient elevation of [Ca²⁺]_o under control conditions induced faster Ca²⁺ release and enhanced the rate of inactivation of I_Ca (compare open circles and squares). The rate of inactivation of I_Ca recorded from six different cells, when plotted as a function of the rate of rise of intracellular Ca²⁺ transient (d[Ca²⁺]_i/dt) gave a linear regression line (Fig. 9, open symbols; r = 0.797). Surprisingly, in the presence of BAY K 8644, the rate of inactivation of I_Ca was no longer dependent on the magnitude of Ca²⁺ release from the SR (Fig. 9, filled symbols). These results suggest that BAY K 8644 renders the L-type Ca²⁺ channels less sensitive to Ca²⁺ release from the SR. This property of BAY K 8644 may be responsible, in part, for some of the well-known Ca²⁺-channel modifying properties of the drug.

View larger version (14K): [in this window] [in a new window]   Fig. 9.   Relationship between rate of inactivation of ICa (1/) and d[Ca2+]i/dt. Plotted values are from 6 myocytes measured with 2 and 10 mM Ca2+ before and after application of BayK. Experiments were carried out with rapid exchange of solution as described in Fig. 7. Data obtained in absence of BayK under control conditions were fitted by a regression line [1/ = (14.0 × 106 M1 × d[Ca2+]i/dt) + 29.5 s1; r = 0.797]. This relationship was abolished in presence of BayK.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The major finding of this study is that the amplification factor of I_Ca-induced Ca²⁺ release depends on the membrane potential as well as on gating properties of Ca²⁺ channels. Despite the steep voltage dependence of the amplification factor at negative potentials (40 to 10 mV), the gain of the Ca²⁺-induced Ca²⁺-release mechanism appears to be fairly constant at membrane potentials spanning the plateau of the cardiac action potential (0 to 30 mV). The most unexpected finding was that BAY K 8644, while enhancing I_Ca, suppressed the amplification factor at all potentials and shifted its voltage dependence. The suppression of the amplification factor in the presence of BAY K 8644 was accompanied by decreased Ca²⁺ sensitivity of the inactivation kinetics of the L-type Ca²⁺ channel to the released Ca²⁺. In the present paper, the amplification factor was used to provide a quantitative measure of the Ca²⁺-induced Ca²⁺ release and to elucidate the mechanism of its previously reported voltage dependence (17, 18, 27) and suppression by BAY K 8644 (21). The loss of the Ca²⁺-dependent inactivation of the fully phosphorylated channel, in the presence of BAY K 8644, may be responsible, in part, for the decreased gain of the Ca²⁺-induced Ca²⁺-release mechanism.

Amplification factor. The uniqueness of the amplification factor as a parameter of Ca²⁺-signaling cascade was explored by examining its voltage, Ca²⁺, and drug dependence. The steep voltage dependence of the amplification factor at negative voltages (Fig. 1F) may be related to the larger driving force for Ca²⁺ influx encountered at more negative potentials (17, 27). This characteristic may reflect the effectiveness of the Ca²⁺ influx through the L-type Ca²⁺ channels in gating ryanodine receptors before they are buffered by the endogenous Ca²⁺ buffers. Considering the findings, shown in Fig. 9, that BAY K 8644 significantly altered the Ca²⁺-dependent inactivation of the L-type Ca²⁺ channels and reduced the amplification factor (Fig. 3), the steep voltage dependence of the amplification factor at negative potentials may reflect the tight coupling between the gating of single Ca²⁺ channels and the ryanodine receptors. At test potentials positive to 10 mV, the amplification factor was significantly smaller and almost constant at ~25. The maximal open probability of the L-type Ca²⁺ channels at 0 mV may to some extent saturate the Ca²⁺-signaling mechanism and thereby stabilize the amplification factors, producing a high degree of biological safety for E-C coupling. It is not clear as yet whether the high values of gain encountered at negative voltages play a physiological role in the regulation of the Ca²⁺-release mechanism. It should be noted, however, that high amplification factors are consistent with the idea that the spontaneous opening of the L-type Ca²⁺ channels, even when occurring at extremely low probability at negative potentials (60 to 90 mV), do activate Ca²⁺ sparks (9).

Effect of Bay K 8644 on amplification factor. BAY K 8644 not only decreased the amplification factor but also enhanced the leak of Ca²⁺ from the SR at a V_h of 60 mV, thus partially depleting the SR. This property of BAY K 8644 was quantified by examining the magnitude of Ca²⁺ release by caffeine at 60 and 90 mV (Figs. 4 and 5). Such voltage dependence of the caffeine-induced Ca²⁺ release and its modification by the drug may have resulted from a negative shift of the open probability of the L-type Ca²⁺ channels. The shift would have also increased the spontaneous opening of the single Ca²⁺ channels, leading to localized activation of the ryanodine receptors (27). Using confocal Ca²⁺ imaging, we have confirmed that the slow global rise in [Ca²⁺]_i on change of V_h from 90 to 60 mV (Figs. 4 and 5) is caused by the higher frequency of spontaneous activation of multiple Ca²⁺ sparks at 60 versus 90 mV (9). It has been previously suggested that BAY K 8644 (~0.1 µM) interacts with sarcolemmal Ca²⁺ channels and, via a functional link to the ryanodine receptors, depletes the SR of Ca²⁺ in ferret ventricular myocytes even in the absence of the extracellular Ca²⁺ at very negative membrane potentials (22, 29). In our hands, in rat ventricular myocytes, the leak of Ca²⁺ from the SR by BAY K 8644 depended on the extracellular Ca²⁺ and V_h. We found that replacement of Ca²⁺ by Ba²⁺ and V_h of 90 mV helped maintain the Ca²⁺ content of the caffeine-sensitive Ca²⁺ pools. This discrepancy may be related to the different species and experimental approaches used to obtain the two sets of data.

Voltage dependence of amplification factor. The voltage dependence of the amplification factor may result from the voltage dependence of the single-channel Ca²⁺ current (18, 27) and/or may reflect the kinetics of opening of single channels (20). The idea that the gain factor of the Ca²⁺-induced Ca²⁺-release mechanism may be regulated by the kinetics of Ca²⁺ channels is supported by the finding that the ascending limb of the amplification factor (Figs. 1E, 3C, and 8D) and the activation range of I_Ca (Figs. 2C and 8A) have similar voltage dependence and are shifted to the same extent when exposed to BAY K 8644 or higher [Ca²⁺]_o. In sharp contrast, the nearly linear voltage dependence of the unitary currents of the Ca²⁺ channel (14, 20, 26) appears to deviate markedly from the strongly nonlinear voltage dependence of the amplification factor (Figs. 1E, 3C, and 8D). Furthermore, the amplification factor does not approach zero as the driving force for the Ca²⁺ current decreases at very positive potentials (>60 mV). Our data, therefore, suggest that brief openings of Ca²⁺ channels might be relatively more effective in causing release than longer lasting openings. This is consistent with the idea that a brief rise of Ca²⁺ near the ryanodine receptors is more effective than a sustained increase of Ca²⁺ in gating the release channel (5, 10, 27). The low efficacy of long opening in gating the release channels is also supported by our data wherein large increases of Ca²⁺ current, due to BAY K 8644-induced modal shift in gating of Ca²⁺ channels (13, 34), are not as effective in triggering Ca²⁺ release (Fig. 3).

Suppression of Ca²⁺-dependent inactivation of L-type Ca²⁺ channels. The most striking finding of this study was that BAY K 8644 suppressed the dependence of inactivation of I_Ca on Ca²⁺-release from the SR (Fig. 9) in cells dialyzed with high concentrations of cAMP. Although cAMP-dependent phosphorylation by forskolin is known to decrease the effect of resting [Ca²⁺]_i on the magnitude of I_Ca in guinea pig cells (35), the present results show that the inactivation kinetics of I_Ca in rat ventricular cells with 200 µM cAMP remain quite sensitive to SR Ca²⁺ release, except in the presence of BAY K 8644 (Fig. 9). A possible Ca²⁺ binding site responsible for the Ca²⁺-dependent inactivation of the cardiac L-type Ca²⁺ channels has been suggested to be located near the E-F hand domain of the cytosolic carboxy terminal close to IVS6 of the ₁-subunit (31). The IVS6 appears to be also one of the hot spots for the binding of DHPs (15). Because serine 1,928 in the carboxy tail of ₁-subunit may serve as a PKA-dependent phosphorylation site, it is possible that the binding of BAY K 8644 to ₁-subunit of the PKA-phosphorylated cardiac L-type Ca²⁺ channel can modify the Ca²⁺ binding domain of the channel in a manner that blunts the sensitivity of the Ca²⁺-dependent inactivation site to Ca²⁺.

3