cAMP-dependent phosphorylation sites and macroscopic activity of recombinant cardiac L-type calcium channels

The involvement of cAMP-dependent phosphorylation sites in establishing the basal activity of cardiac L-type Ca²⁺ channels was studied in HEK 293 cells transiently cotransfected with mutants of the human cardiac ₁ and accessory subunits. Systematic individual or combined elimination of high consensus protein kinase A (PKA) sites, by serine to alanine substitutions at the amino and carboxyl termini of the ₁ subunit, resulted in Ca²⁺ channel currents indistinguishable from those of wild type channels. Dihydropyridine (DHP)-binding characteristics were also unaltered. To explore the possible involvement of nonconsensus sites, deletion mutants were used. Carboxyl-terminal truncations of the ₁ subunit distal to residue 1597 resulted in increased channel expression and current amplitudes. Modulation of PKA activity in cells transfected with the wild type channel or any of the mutants did not alter Ca²⁺ channel functions suggesting that cardiac Ca²⁺ channels expressed in these cells behave, in terms of lack of PKA control, like Ca²⁺ channels of smooth muscle cells.     

Stimulation of cardiac L-type calcium channels by extracellular ATP

The co-release of ATP with norepinephrine from sympatheticnerve terminals in the heart may augment adrenergic stimulation ofcardiac Ca²⁺ channel activity. To test for a possibledirect effect of extracellular ATP on L-type Ca²⁺ channels,single channels were reconstituted from porcine sarcolemma into planarlipid bilayers so that intracellular signaling pathways could becontrolled. Extracellular ATP (2-100 µM) increased the openprobability of the reconstituted channels, with a maximal increase of~2.6-fold and an EC₅₀ of 3.9 µM. The increase in open probability was due to an increase in channel availability and adecrease in channel inactivation rate. Other nucleotides displayed arank order of effectiveness of ATP > ,-methylene-ATP > 2-methylthio-ATP > UTP > adenosine5'-O-(3-thiotriphosphate) >> ADP; adenosine had no effect.Several antagonists of P2 receptors had no impact on the ATP-dependentincrease in open probability, indicating that receptor activation wasnot required. These results suggest that extracellular ATP and othernucleotides can stimulate the activity of cardiac L-typeCa²⁺ channels via a direct interaction with the channels.

     

Biochemical analysis of L-type calcium channels from skeletal and cardiac muscle

The development of specific pharmacological agents that modulate different types of ion channels has prompted an extensive effort to elucidate the molecular structure of these important molecules. The calcium channel blockers that specifically modulate the L-type calcium channel activity have aided in the purification and reconstitution of this channel from skeletal muscle transverse tubules. The L-type calcium channel from skeletal muscle is composed of five subunits designated alpha 1, alpha 2, beta, gamma, and sigma. The alpha 1-subunit is the pore-forming polypeptide and contains the ligand binding and phosphorylation sites through which channel activity can be modulated. The role of the other subunits in channel function remains to be studied. The calcium channel components have also been partially purified from cardiac muscle. The channel consists of at least three subunits that have properties related to the subunits of the calcium channel from skeletal muscle. A core polypeptide that can form a channel and contains ligand binding and phosphorylation sites has been identified in cardiac preparations. Here we summarize recent biochemical and molecular studies describing the structural features of these important ion channels.     

Gamma 1 subunit interactions within the skeletal muscle L-type voltage-gated calcium channels

Voltage-gated calcium channels mediate excitationcontraction coupling in the skeletal muscle. Their molecular composition, similar to neuronal channels, includes the pore-forming alpha(1) and auxiliary alpha(2)delta, beta, and gamma subunits. The gamma subunits are the least characterized, and their subunit interactions are unclear. The physiological importance of the neuronal gamma is emphasized by epileptic stargazer mice that lack gamma(2). In this study, we examined the molecular basis of interaction between skeletal gamma(1) and the calcium channel. Our data show that the alpha(1)1.1, beta(1a), and alpha(2)delta subunits are still associated in gamma(1) null mice. Reexpression of gamma(1) and gamma(2) showed that gamma(1), but not gamma(2), incorporates into gamma(1) null channels. By using chimeric constructs, we demonstrate that the first half of the gamma(1) subunit, including the first two transmembrane domains, is important for subunit interaction. Interestingly, this chimera also restores calcium conductance in gamma(1) null myotubes, indicating that the domain mediates both subunit interaction and current modulation. To determine the subunit of the channel that interacts with gamma(1), we examined the channel in muscular dysgenesis mice. Cosedimentation experiments showed that gamma(1) and alpha(2)delta are not associated. Moreover, alpha(1)1.1 and gamma(1) subunits form a complex in transiently transfected cells, indicating direct interaction between the gamma(1) and alpha(1)1.1 subunits. Our data demonstrate that the first half of gamma(1) subunit is required for association with the channel through alpha(1)1.1. Because subunit interactions are conserved, these studies have broad implications for gamma heterogeneity, function and subunit association with voltage-gated calcium channels.     

Characterization of metal ion-binding sites in bacteriorhodopsin

We have investigated the effects of the binding of various metal ions to cation-free bacteriorhodopsin ("blue membrane"). The following have been measured: shift of the absorption maximum from 603 to 558 nm (blue to purple transition), binding isotherms, the release of H+ upon binding, and the decay of the deprotonated intermediate of the photocycle, M412. We find that all cations of the lanthanide series, as well as the alkali and alkali earth metals earlier investigated, are able to bring about the absorption shift, whereas Hg2+ and Pt4+ are not. Sigmoidal spectroscopic titration curves and nonsigmoidal binding curves suggest that there are two high affinity sites for cations in bacteriorhodopsin. Binding to the site with the second highest affinity is responsible for the absorption shift. Divalent cation binding to blue membrane causes release of about six protons, whereas higher numbers of protons are released by trivalent cations, suggesting that the shift of absorption maximum involves proton release from carboxyl group(s). The metal ion bound to this site must be surrounded by carboxyl oxygen atoms acting together as a multidentate ligand with a specific geometry because multivalent ions are effective only when capable of octahedral coordination. Lanthanide ions dramatically inhibit M412 decay at pH above 6.3, an effect probably due to binding to lipid phosphoryl groups.     

Regulation of cloned cardiac L-type calcium channels by cGMP-dependent protein kinase

We have studied the effect of 8-bromo-cyclic GMP (8-Br-cGMP) on cloned cardiac L-type calcium channel currents to determine the site and mechanism of action underlying the functional effect. Rabbit cardiac alpha(1C) subunit, in the presence or absence of beta(1) subunit (rabbit skeletal muscle) or beta(2) subunit (rat cardiac/brain), was expressed in Xenopus oocytes, and two-electrode voltage-clamp recordings were made 2 or 3 days later. Application of 8-Br-cGMP caused decreases in calcium channel currents in cells expressing the alpha(1C) subunit, whether or not a beta subunit was co-expressed. No inhibition of currents by 8-Br-cGMP was observed in the presence of the protein kinase G inhibitor KT5823. Substitutions of serine residues by alanine were made at residues Ser(533) and Ser(1371) on the alpha(1C) subunit. As for wild type, the mutant S1371A exhibited inhibition of calcium channel currents by 8-Br-cGMP, whereas no effect of 8-Br-cGMP was observed for mutant S533A. Inhibition of calcium currents by 8-Br-cGMP was also observed in the additional presence of the alpha(2)delta subunit for wild type channels but not for the mutant S533A. These results indicate that cGMP causes inhibition of L-type calcium channel currents by phosphorylation of the alpha(1C) subunit at position Ser(533) via the action of protein kinase G.     

Role of gelsolin in the actin filament regulation of cardiac L-type calcium channels

The actin cytoskeleton is an important contributor to themodulation of the cell function. However, little is known about theregulatory role of this supermolecular structure in the membrane eventsthat take place in the heart. In this report, the regulation of cardiacmyocyte function by actin filament organization was investigated inneonatal mouse cardiac myocytes (NMCM) from both wild-type mice andmice genetically devoid of the actin filament severing protein gelsolin(Gsn/). Cardiac L-type calcium channel currents(I_Ca) wereassessed using the whole cell voltage-clamp technique. Addition of theactin filament stabilizer phalloidin to wild-type NMCM increasedI_Ca by 227% overcontrol conditions. The basalI_Ca ofGsn/ NMCM was 300% higher than wild-type controls. Thisincrease was completely reversed by intracellular perfusion of theGsn/ NMCM with exogenous gelsolin. Further, cytoskeletal disruption of either Gsn/ or phalloidin-dialyzedwild-type NMCM with cytochalasin D (CD) decreased the enhancedI_Ca by 84% and 87%, respectively. The data indicate that actin filament stabilization by either a lack of gelsolin or intracellular dialysis with phalloidin increase I_Ca,whereas actin filament disruption with CD or dialysis ofGsn/ NMCM with gelsolin decreaseI_Ca. We concludethat cardiac L-type calcium channel regulation is tightly controlled byactin filament organization. Actin filament rearrangement mediated by gelsolin may contribute to calcium channel inactivation.

     

Calcicludine binding to the outer pore of L-type calcium channels is allosterically coupled to dihydropyridine binding

How dihydropyridines modulate L-type voltage-gated Ca2+ channels is not known. Dihydropyridines bind cooperatively with Ca2+ binding to the selectivity filter, suggesting that they alter channel activity by promoting structural rearrangements in the pore. We used radioligand binding and patch-clamp electrophysiology to demonstrate that calcicludine, a toxin from the venom of the green mamba snake, binds in the outer vestibule of the pore and, like Ca2+, is a positive modulator of dihydropyridine binding. Data were fit using an allosteric scheme where dissociation constants for dihydropyridine and calcicludine binding, KDHP and KCaC, are linked via the coupling factor, alpha. Nine acidic amino acids located within the S5-Pore-helix segment of repeat III were sequentially changed to alanine in groups of three, resulting in the mutant channels, Mut-A, Mut-B, and Mut-C. Mut-A, whose substitutions are proximal to IIIS5, exhibits a 4.5-fold reduction in dihydropyridine binding and is insensitive to calcicludine binding. Block of Mut-A currents by calcicludine is indistinguishable from wild-type, indicating that KCaC is unchanged and that the coupling between dihydropyridine and calcicludine binding (i.e., alpha) is disrupted. Mut-B and Mut-C possess KDHP values that resemble that of the wild type. Mut-C, the most C-terminal of the mutant channels, is insensitive to calcicludine binding and block. KCaC values for the Mut-C single mutants, E1122A, D1127A, and D1129A, increase from 0.3 (wild type) to 1.14, 2.00, and 20.5 microM, respectively. Together, these findings suggest that dihydropyridine antagonist and calcicludine binding to L-type Ca2+ channels promote similar structural changes in the pore that stabilize the channel in a nonconducting, blocked state.     

Interactions between H+ and Ca2+ near cardiac L-type calcium channels: evidence for independent channel-associated binding sites.

Monovalent and divalent ions are known to affect voltage-gated ion channels by the screening of, and/or binding to, negative charges located on the surface of cell membranes within the vicinity of the channel protein. In this investigation, we studied gating shifts of cardiac L-type calcium channels induced by extracellular H+ and Ca2+ to determine whether these cations interact at independent or competitive binding sites. At constant pHo (7.4), Cao-induced gating shifts begin to approach a maximum value (approximately equal to 17 mV) at concentrations of extracellular calcium of > or = 40 mM. A fraction of the calcium-dependent gating shift could be titrated with an effective pKa = 6.9 indicating common and competitive access to H+ and Ca2+ ions for at least one binding site. However, if pHo is lowered when Cao is > or = 40 mM, additional shifts in gating are measured, suggesting a subpopulation of sites to which Ca2+ and H+ bind independently. The interdependence of L-channel gating shifts and Cao and pHo was well described by the predictions of surface potential theory in which two sets of binding sites are postulated; site 1 (pKa = 5.5) is accessible only to H+ ions and site 2 (pKa = 6.9) is accessible to both Ca2+ and H+ ions. Theoretical computations generated with this model are consistent with previously determined data, in which interactions between these two cations were not studied, in addition to the present experiments in which interactions were systematically probed.     

Comparative analysis of the kinetic characteristics of L-type calcium channels in cardiac cells of hibernators.

An undefined property of L-type Ca2+ channels is believed to underlie the unique phenotype of hibernating hearts. Therefore, L-type Ca2+ channels in single cardiomyocytes isolated from hibernating versus awake ground-squirrels (Citellus undulatus) were compared using the perforated mode of the patch-clamp technique, and interpreted by way of a kinetic model of Ca2+ channel behavior based upon the concept of independence of the activation and inactivation processes. We find that, in hibernating ground-squirrels, the cardiac L-type Ca2+ current is lower in magnitude when compared to awake animals. Both in the awake or hibernating states, kinetics of L-type Ca2+ channels could be described by a d2f1(2)f2 model with an activation and two inactivation processes. The activation (or d) process relates to the movement of the gating charge. The slow (or f1) inactivation is associated with movement of gating charge and is current-dependent. The rapid (or f2) inactivation is a complex process which cannot be represented as a single-step conformational transition induced by the gating charge movement, and is regulated by beta-adrenoceptor stimulation. When compared to awake animals, the kinetic properties of Ca2+ channels from hibernating ground-squirrels differed in the following parameters: (1) pronounced shift (15-20 mV) toward depolarization in the normalized conductance of both inactivation components, and moderate shift in the activation component; (2) 1.5-2-fold greater time constants; and (3) two-fold greater activation gating charge. Thus, L-type Ca2+ channels apparently switch their phenotype during the hibernating transition. Stimulation of beta-adrenoceptors by isoproterenol, reversed the hibernating kinetic- (but not amplitude-) phenotype toward the awake type. Therefore, an aberrance in the beta-adrenergic system can not fully explain the observed changes in the L-type Ca2+ current. This suggests that during hibernation additional mechanisms may reduce the single Ca2+ channel-conductance and/or keep a fraction of the cardiac L-type Ca2+ channel population in a non-active state.     

Anomalous L-type calcium channels of rat spinal motoneurons

Single channel patch-clamp recordings show that embryonic rat spinal motoneurons express anomalous L-type calcium channels, which reopen upon repolarization to resting potentials, displaying both short and long reopenings. The probability of reopening increases with increasing voltage of the preceding depolarization without any apparent correlation with inactivation during the depolarization. The probability of long with respect to short reopenings increases with increasing length of the depolarization, with little change in the total number of reopenings and in their delay. With less negative repolarization voltages, the delay increases, while the mean duration of both short and long reopenings decreases, remaining longer than that of the openings during the preceding depolarization. Open times decrease with increasing voltage in the range -60 to +40 mV. Closed times tend to increase at V > 20 mV. The open probability is low at all voltages and has an anomalous bell-shaped voltage dependence. We provide evidence that short and long reopenings of anomalous L-type channels correspond to two gating modes, whose relative probability depends on voltage. Positive voltages favor both the transition from a short-opening to a long-opening mode and the occupancy of a closed state outside the activation pathway within each mode from which the channel reopens upon repolarization. The voltage dependence of the probability of reopenings reflects the voltage dependence of the occupancy of these closed states, while the relative probability of long with respect to short reopenings reflects the voltage dependence of the equilibrium between modes. The anomalous gating persists after patch excision, and therefore our data rule out voltage-dependent block by diffusible ions as the basis for the anomalous gating and imply that a diffusible cytosolic factor is not necessary for voltage-dependent potentiation of anomalous L-type channels.     

L-type calcium channel blockade attenuates morphine withdrawal: in vivo interaction between L-type calcium channels and corticosterone

Both opioids and calcium channel blockers could affect hypothalamic-pituitary-adrenal (HPA) axis function. Nifedipine, as a calcium channel blocker, can attenuate the development of morphine dependence; however, the role of the HPA axis in this effect has not been elucidated. We examined the effect of nifedipine on the induction of morphine dependency in intact and adrenalectomized (ADX) male rats, as assessed by the naloxone precipitation test. We also evaluated the effect of this drug on HPA activity induced by naloxone. Our results showed that despite the demonstration of dependence in both groups of rats, nifedipine is more effective in preventing of withdrawal signs in ADX rats than in sham-operated rats. In groups that received morphine and nifedipine concomitantly, naloxone-induced corticosterone secretion was attenuated. Thus, we have shown the involvement of the HPA axis in the effect of nifedipine on the development of morphine dependency and additionally demonstrated an in vivo interaction between the L-type Ca2+ channels and corticosterone.     

Amiloride analogs inhibit L-type calcium channels and display calcium entry blocker activity

Three structural classes of commonly used amiloride analogs, molecules derivatized at the terminal guanidino-nitrogen, the five-position pyrazinoyl-nitrogen, or di-substituted at both of these positions, inhibit binding of the L-type Ca2+ channel modulators diltiazem, gallopamil, and nitrendipine to porcine cardiac sarcolemmal membrane vesicles. The rank order of inhibitory potencies among the various derivatives tested is well defined with amiloride being the least potent. Saturation binding studies indicate that inhibition of ligand binding results primarily from effects on Kd. Ligand dissociation measurements suggest that amiloride derivatives do not associate directly at any of the known sites in the Ca2+ entry blocker receptor complex. In addition, these compounds do not compete at the "Ca2+ coordination site" within the channel. However, studies with inorganic and substituted diphenylbutylpiperidine Ca2+ entry blockers reveal that amiloride analogs interact at a site on the channel where metal ions bind and occlude the pore. Photolysis experiments performed with amiloride photoaffinity reagents confirm that a specific interaction occurs between such probes and the channel protein. Upon photolysis, these agents produce concentration- and time-dependent irreversible inactivation of Ca2+ entry blocker binding activities, which can be protected against by either verapamil or diltiazem. 45Ca2+ flux and voltage-clamp experiments performed with GH3 anterior pituitary cells demonstrate that amiloride-like compounds inhibit L-type Ca2+ channels directly. Moreover, these compounds block contraction of isolated vascular tissue in pharmacological assays. Electrophysiological experiments indicate that they also inhibit T-type Ca2+ channels in GH3 cells. Taken together, these results demonstrate unequivocally that amiloride analogs display significant Ca2+ entry blocker activity in both ligand binding and functional assays. This property, therefore, can seriously complicate the interpretation of many in vitro and in vivo studies where amiloride analogs are used to elicit inhibition of other transport systems (e.g. Na-Ca and Na-H exchange).     

L-type voltage-operated calcium channels, N-methyl-D-aspartate receptors and neuronal nitric-oxide synthase form a calcium/redox nano-transducer within lipid rafts

Cytosolic calcium plays a leading role in the control of neuronal excitability, plasticity and survival. This work aims to experimentally assess the possibility that lipid rafts of the plasma membrane can provide a structural platform for a faster and tighter functional coupling between calcium and nitric-oxide signaling in neurons. Using primary cerebellar granule neurons (CGN) in culture this hypothesis has been experimentally assessed with fluorescence resonance energy transfer imaging, preparations of lipid rafts-enriched membrane fragments and western blotting. The results obtained in this work demonstrated that major calcium entry systems of the plasma membrane of CGN (L-type calcium channels and N-methyl-D-aspartate receptors) and nitric-oxide synthase are separated by less than 80 nm from each other within lipid rafts-associated sub-microdomains, suggesting a new role of lipid rafts as neuronal calcium/redox nano-transducers.     

中枢神经系统L-型电压门控钙通道的功能调控与脑缺血

中枢神经系统L 型电压门控钙通道 (L typevoltage gatedcalciumchannels ,L VGCCs)由α1C(D)亚基和辅助亚基组成。α1C亚基的C 端包含多个功能结构域 ,可分别与钙调素、钙调蛋白酶、cAMP依赖性蛋白激酶、Src家族酪氨酸蛋白激酶 (Srcfamilyproteintyrosinekinases ,SrcPTKs)等相互作用 ,从而参与L VGCCs的功能调控。SrcPTKs介导的两种钙通道———L VGCCs和N 甲基 D 天冬氨酸 (N methyl D aspartate ,NMDA)受体的对话可能是缺血性脑损伤的重要机制     

Calmodulin kinase determines calcium-dependent facilitation of L-type calcium channels

A dynamic positive feedback mechanism, known as 'facilitation', augments L-type calcium-ion currents (ICa) in response to increased intracellular Ca2+ concentrations. The Ca2+-binding protein calmodulin (CaM) has been implicated in facilitation, but the single-channel signature and the signalling events underlying Ca2+/CaM-dependent facilitation are unknown. Here we show that the Ca2+/CaM-dependent protein kinase II (CaMK) is necessary and possibly sufficient for ICa facilitation. CaMK induces a channel-gating mode that is characterized by frequent, long openings of L-type Ca2+ channels. We conclude that CaMK-mediated phosphorylation is an essential signalling event in triggering Ca2+/CaM-dependent ICa facilitation.     

Coactivation of capacitative calcium entry and L-type calcium channels in guinea pig gallbladder

We have evaluated the presence of capacitative Ca(2+) entry (CCE) in guinea pig gallbladder smooth muscle (GBSM), including a possible relation with activation of L-type Ca(2+) channels. Changes in cytosolic Ca(2+) concentration induced by Ca(2+) entry were assessed by digital microfluorometry in isolated, fura 2-loaded GBSM cells. Application of thapsigargin, a specific inhibitor of the Ca(2+) store pump, induced a transient Ca(2+) release followed by sustained entry of extracellular Ca(2+). Depletion of the stores with thapsigargin, cyclopiazonic acid, ryanodine and caffeine, high levels of the Ca(2+)-mobilizing hormone cholecystokinin octapeptide, or simple removal of external Ca(2+) resulted in a sustained increase in Ca(2+) entry on subsequent reapplication of Ca(2+). This entry was attenuated by 2-aminoethoxydiphenylborane, L-type Ca(2+) channel blockade, pinacidil, and Gd(3+). Accumulation of the voltage-sensitive dye 3,3'-dipentylcarbocyanine and direct intracellular recordings showed that depletion of the stores is sufficient for depolarization of the plasma membrane. Contractility studies in intact gallbladder muscle strips showed that CCE induced contractions. The CCE-evoked contraction was sensitive to 2-aminoethoxydiphenylborane, L-type Ca(2+) channel blockers, and Gd(3+). We conclude that, in GBSM, release of Ca(2+) from internal stores activates a CCE pathway and depolarizes plasma membrane, allowing coactivation of voltage-operated L-type Ca(2+) channels. This process may play a role in excitation-contraction coupling in GBSM.     

Characterization of the binding sites for the interactions between FKBP12 and intracellular calcium release channels

FKBP12, an FK506 binding protein, interacts with type 1 ryanodine receptor (RyR1) and modulates its calcium channel activity. However, there are many opposing reports of FKBP12's interaction with other related calcium channels, such as type 1 IP(3) receptor and type 3 ryanodine receptor (IP(3)R1 and RyR3). In addition, the involvement of the prolyl-dipeptide motif in the calcium channels and the corresponding binding residues in FKBP12 remain controversial. Through pulldown assays with recombinant proteins, we provide biochemical evidence of the interaction between FKBP12 and RyR1, RyR3 and IP(3)R1. Using NMR chemical shift mapping, we show that the important binding residues in FKBP12 are located in its hydrophobic FK506 binding region. Consistently, we demonstrate that FK506 can competitively inhibit the interaction between FKBP12 and the dipeptide motifs of the calcium channels. We believe our results shed lights on the binding mechanism of calcium channel-FKBP12 interaction.     

Physiologic gating properties of unitary cardiac L-type Ca channels

The contraction of adult mammalian ventricular cardiomyocytes is triggered by the influx of Ca²⁺ ions through sarcolemmal L-type Ca²⁺ channels (LCCs). However, the gating properties of unitary LCCs under physiologic conditions have remained elusive. Towards this end, we investigated the voltage-dependence of the gating kinetics of unitary LCCs, with a physiologic concentration of Ca²⁺ ions permeating the channel. Unitary LCC currents were recorded with 2 mM external Ca²⁺ ions (in the absence of LCC agonists), using cell-attached patches on K-depolarized adult rat ventricular myocytes. The voltage-dependence of the peak probability of channel opening (Po vs. Vm) displayed a maximum value of 0.3, a midpoint of −12 mV, and a slope factor of 8.5. The maximum value for Po of the unitary LCC was significantly higher than previously assumed, under physiologic conditions. We also found that the mean open dwell time of the unitary LCC increased twofold with depolarization, ranging from 0.53 ± 0.02 ms at −30 mV to 1.08 ± 0.03 ms at 0 mV. The increase in mean LCC open time with depolarization counterbalanced the decrease in the single LCC current amplitude; the latter due to the decrease in driving force for Ca²⁺ ion entry. Thus, the average amount of Ca²⁺ ions entering through an individual LCC opening (∼300-400 ions) remained relatively constant over this range of potentials. These novel results establish the voltage-dependence of unitary LCC gating kinetics using a physiologic Ca²⁺ ion concentration. Moreover, they provide insight into local Ca²⁺-induced Ca²⁺ release and a more accurate basis for mathematical modeling of excitation-contraction coupling in cardiac myocytes.