QUARTERLY Regulation of Ca 2+ release from internal stores in cardiac and

It is widely accepted that Ca2+ is released from the sarcoplasmic reticulum by a specialized type of calcium channel, i.e., ryanodine receptor, by the process of Ca2+-induced Ca2+ release. This process is triggered mainly by dihydropyridine receptors, i.e., L-type (long lasting) calcium channels, directly or indirectly interacting with ryanodine receptor. In addition, multiple endogenous and exogenous compounds were found to modulate the activity of both types of calcium channels, ryanodine and dihydropyridine receptors. These compounds, by changing the Ca2+ transport activity of these channels, are able to influence intracellular Ca2+ homeostasis. As a result not only the overall Ca2+ concentration becomes affected but also spatial distribution of this ion in the cell. In cardiac and skeletal muscles the release of Ca2+ from internal stores is triggered by the same transport proteins, although by their specific isoforms. Concomitantly, heart and skeletal muscle specific regulatory mechanisms are different.

teins of the contractile apparatus, leading finally to muscle contraction (Zot & Potter, 1987).The functional link between electrical stimulation of the muscle and its contraction is called the excitation-contraction (E-C) coupling.First experiments related to muscle contraction revealed that heart muscle preparations, in contrast to skeletal muscle, could not contract for a long time if Ca 2+ was absent from the bath solution (Ringer, 1883;Rich et al., 1988;Bers, 1991).These observations suggest that the contraction-relaxation cycles of these two muscle types are regulated by different E-C coupling mechanisms.Additional differences between the two types of muscle were revealed by experiments in which the voltage (E m ) dependence of Ca 2+ current (I Ca ), changes in [Ca 2+ ] i transient, charge movement and muscle shortening were determined.Briefly, in cardiac muscle cells the E m -dependence of [Ca 2+ ] i transient and contraction are very similar to the E m -dependence of I Ca and are bell-shaped, in contrast to a sigmoidal shape of the curve reflecting the charge movement in skeletal muscle cells (Schneider & Chandler, 1973;Bers, 1991).These observations clearly indicated that, in skeletal muscle cells the intramembrane charge movement, is the factor triggering the increase in [Ca 2+ ] i , while in cardiac muscle cells this increase is due to the influx of Ca 2+ through the plasma membrane.Considering the mechanisms of E-C coupling in skeletal and cardiac muscle cells, it is surprising that such very homologous proteins developed different mechanisms of Ca 2+ release from intracellular stores.

SPATIAL RELATIONSHIP BETWEEN DIHYDROPYRIDINE AND RYANODINE RECEPTORS
Muscle cell membrane depolarization induces conformational changes in dihydropyridine receptor (DHPR), i.e., L-type (long lasting) Ca 2+ channels located in the plasma mem-brane.These conformational changes are translated into signal amplification by direct interactions with sarcoplasmic reticulum (SR) membrane receptors for the plant alkaloid ryanodine receptor (RyR), as in skeletal muscle cell, or through the Ca 2+ influx, as in cardiac muscle cells (Sutko et al., 1997).The spatial relationship between DHPR and RyR skeletal muscle differs from that in cardiac or smooth muscles as well as in neuronal cells.In different cell types, DHPR and RyR are localized in junctions of membranes, called triads, dyads or peripheral couplings; the latter representing the junctions between the plasma and SR membranes (Flucher & Franzini-Armstrong, 1996).The characteristic features of these junctions are the "feet", i.e., electron-dense structures spanning the narrow junctional gap (about 16 nm) separating SR membranes and transverse tubules (Langer & Peskoff, 1997).In the case of skeletal muscles, the foot structures form a semicrystalline array in the plane of the membrane, consisting of RyR connected with tetrads of DHPR (Fig. 1) (Flucher & Franzini-Armstrong, 1996).In cardiac muscles, these structures are less organized (Fig. 1) and their number differs significantly from that of RyR channels in skeletal muscles (Sun et al., 1995).In skeletal muscles, the ratio of DHPR to RyR ranges from 0.6 to 2.1, while in cardiac muscles, from 0.1 to 0.3 (Sutko & Airey, 1996).It is worth stressing that these two channels are not the major components responsible for the formation of junctions between SR membranes and transverse tubules which can be formed even in the absence of DHPR and RyR in the cell (Flucher et al., 1992;Takekura et al., 1995a).On the other hand, co-expression of DHPR and RyR in CHO cells failed to induce the formation of the junction between SR and plasma membranes (Takekura et al., 1995b).This suggests that other "docking" proteins are required for formation of the junctional connections.
The observations cited above suggest that differences in E-C coupling between skeletal and cardiac muscles are related to different spatial organization of their respective Ca 2+ transport systems.In skeletal muscles direct interaction between DHPR and RyR is responsible for releasing Ca 2+ from internal stores while in cardiac muscles a Ca 2+ -induced Ca 2+ release (CICR) mechanism exists under the local control of Ca 2+ current through the L-type channels (Franzini-Armstrong & Protasi, 1997).

CALCIUM SPARKS
A small increase in [Ca 2+ ] i in cardiac muscle cells causes opening of the calcium channels located in SR membrane in a process called calcium-induced calcium release, which was observed in skinned skeletal and cardiac muscle preparations (Endo et al., 1970;Fabiato, 1985aFabiato, , 1985b)).The release of Ca 2+ from SR is triggered by the opening of a small number of DHPR channels and opening of a similar number of RyR channels (Fig. 2), as theoretically foreseen by Stern & Lakatta (1992), and experimentally confirmed by Cheng et al. (1993).The cardiomyocytes loaded with Fluo-3 showed characteristic points of brightness, called "sparks".These sparks, as it was later shown, arise at the sites adjacent to transverse tubules and provide both analog and digital gain for E-C coupling (Shacklock et al., 1995;Cheng et al., 1996;Isenberg et al., 1996).

DHPR CALCIUM CHANNELS
Biochemical, molecular and immunoprecipitation studies revealed that DHPR present in excitable cells consist of complexes of five protein subunits.The main part of the channel is formed by the a 1 subunit (200 kDa) composed of four motifs, each containing six putative transmembrane segments.At least six genes encoding different isoforms of a 1 subunit have been identified.This variety is responsible for characteristic electrophysiological parameters of voltage-dependent calcium channel (VDCC) (Jones, 1998).In addition to a 1 subunit, VDCC consists of b (60 kDa), g (33 kDa), and a 2 and d subunits (170 kDa) which are connected by disulfide bonds (Jones, 1998;Mitterdorfer et al., 1998).Recently, a novel VDCC-associated protein of 700 kDa has been identified in mammalian cardiomyocytes as a target protein for protein kinase A (PKA)-mediated phosphorylation.This novel protein exhibits some homology to AHNAK-encoded protein, which interacts with phospholipase C. It has been hypothesized that this protein plays a physiological role in transduction of cardiac b-adrenergic signal (Haase et al., 1999).
The most abundant isoforms of DHPR in skeletal and cardiac muscles are L-type calcium channels consisting of a 1s and a 1c subunits, respectively.The a 1 subunit is the receptor for Ca 2+ antagonists and can function as a voltage-gated channel.It was observed that a 1s subunit was able to interact with RyR channels, influencing their activity (Leong & MacLennan, 1998b).Additionally, in skeletal muscle, a short amino-acid sequence of RyR has been identified, which binds to loops I-III of DHPR channel (Leong & MacLennan, 1998a;Grabner et al., 1999;Zhu et al., 1999;Avila & Dirksen, 2000).The DHPR channel activity is also regulated by protein kinase-mediated phosphorylation (McDonald et al., 1994;Strauss et al., 1997;He et al., 2000;Boixel et al., 2000).Moreover, it was shown that, in addition to phosphorylation of a 1 subunit, phosphorylation of DHPR b subunit mediated by PKA may be also partially responsible for modifying the channel activity (Bunemann et al., 1999).Among other protein kinases involved in regulation of DHPR channel activity cyclic GMP-dependent protein kinase was found to decrease the L-type Ca 2+ channel current (Schulz et al., 1994;Haddad et al., 1995), by phosphorylating the serine residue at position 533 in a 1c subunit (Jiang et al., 2000).Involvement of cyclic GMP in regulation of DHPR channel activity is probably related to the effect of nitric oxide which activates guanylate cyclase (McDonald & Murad, 1996).

Molecular cloning
The functional RyR exists as a homotetramer and has a quatrefoil shape about 50 nm in diameter.The center of the quatrefoil contains a pore with an internal diameter of 1 to 2 nm, connected to four radial channels in the peripheral portion of each monomer (Wagenknecht et al., 1989).The use of cryo-electron microscopy of purified RyR, in combination with three-dimensional image reconstructions, allowed creation of a three-dimensional model of the RyR channels (Radermacher et al., 1994;Serysheva et al., 1995;Sharma et al., 1998;Sharma et al., 2000b).It was also possible with the aid of mass displacement in the image analysis to distinguish between two functional states of RyR: open and closed.These states were related to measured conductance of RyR incorporated into lipid bilayers (Murayama & Ogawa, 1997;Murayama et al., 1999).
In mammalian tissues three different genes encode three isoforms of RyR: RyR1, RyR2, and RyR3.cDNA corresponding to the ryr1 gene encoding the RyR1 isoform characteristic for skeletal muscle has been cloned  AP, action potential; CSQ, calsequestrin; PL, plasma membrane; SR, sarcoplasmic reticulum; TC, terminal cisterne; T-tubules, transversal tubules (modified from Wrzosek, 1999) (Takeshima et al., 1989;Zorzato et al., 1990).Rabbit isoform at a protein level has a predicted of 565 223 Da and only slightly differs from its human counterpart with predicted of 563 584 Da (97% identity).The ryr2 gene encoding the RyR2 isoform was cloned from rabbit heart (Otsu et al., 1990;Nakai et al., 1990) and its predicted amino-acid sequence was published later (Tunwell et al., 1996).Much less is known about the RyR3 isoform because it is found at lower concentrations than RyR1 or RyR2 (Murayama & Ogawa, 1997).Partial cDNA corresponding to RyR3 form and then full-length homolog from rabbit brain were cloned (Giannini et al., 1992;Hakamata et al., 1992).
Although three isoforms of RyR are highly homologous proteins, with 67-70% identity, there are three regions in their molecules, D1, D2, and D3, where the amino-acid sequences significantly diverge (Takeshima et al., 1989;Zorzato et al., 1990;Otsu et al., 1990;Hakamata et al., 1992;Sorrentino & Volpe, 1993).In skeletal muscles, region D2 of RyR1 (amino acids 1303-1406), containing charged aminoacid residues, is thought to be essential for E-C coupling (Yamazawa et al., 1997).Recently, region D2 has been overexpressed as a glutathione S-transferase fusion protein and crystallized (Kim et al., 1999).Moreover, molecular imaging of RyR1 and RyR3 isoforms clearly revealed that region D2 is missing in RyR3 isoform (Sharma et al., 2000b).Comparison of reconstructed images of the three RyR isoforms showed that all isoforms are highly conserved with respect to the three-dimensional organization, in which a large part of protein mass (about 80%) forms a cytoplasmic assembly.Despite of similarities in structure, all the three isoforms of RyR differ, however, in the way they are regulated by endogenous and exogenous factors.

Regulation of RyR by endogenous factors
The most important regulator of RyR is Ca 2+  which not only is transported by the channel but also modulates its activity.Interestingly, in the absence of Ca 2+ , RyR conductance for monovalent ions is even higher than for Ca 2+ (Tinker & Williams, 1992;Coronado et al., 1994;Liu et al., 1998).Ca 2+ has both activating and inhibiting effects on Ca 2+ release from SR vesicles, the activity of a single channel, and ryanodine binding to RyR (Meissner, 1986;Michalak et al., 1988;Ashley & Williams, 1990).Ryanodine binding to RyR is absolutely Ca 2+ dependent.RyR channels contain two different Ca 2+ binding sites, with high and low affinity for the cation.Differences in binding constants for these sites were determined using trivalent cations, La 3+ and Tb 3+ .Tb 3+ , each of which, when it binds to proteins, changes its luminescence, thus they were found to be sensitive probes for displacing Ca 2+ from calcium-binding sites.RyR has two binding sites for Tb 3+ , one at low concentrations, Tb 3+ activated, and the second, which at high concentration, inhibited the channel activity of RyR reconstituted in planar lipid bilayer.Part of Tb 3+ was found occluded and it was impossible to remove it from protein using EGTA or high concentration of Ca 2+ (Hadad et al., 1994).
Several putative binding sites for Ca 2+ in RyR have been proposed on the basis of analysis of primary structure of RyR, point directed mutagenesis, modification with dicyclohexylcarbodiimide, and by using specific antibodies directed to the Ca 2+ -binding domains ( The bell-shaped dependence of RyR activity on Ca 2+ concentration is affected by increased concentrations of Mg 2+ , which is another divalent cation influencing the RyR channel activity (Lamb & Stephenson, 1994;Murayama et al., 2000;Lamb, 2000).Mg 2+ competes with Ca 2+ for both activatory and inhibitory binding sites (Meissner et al., 1997;Murayama et al., 2000).RyR isoforms are affected by Mg 2+ in a different manner.This could be explained by the differences between RyR1 and RyR2 in the mechanism of Ca 2+ release from SR stores (Lamb, 2000).Zn 2+ , which is known to be a catalytic and structural element of many proteins involved in the metabolic regulation and gene expression in mammalian cells, was found to modulate also RyR activity (Berg & Shi, 1996) by binding to activatory and inhibitory sites for Ca 2+ (Xia et al., 2000).
It is well documented that not only divalent but also monovalent cations modulate RyR function (Meissner, 1994; Zucchi & Ronca-Testoni, 1997).For example, H + exerts a profound effect on RyR activity and CICR processes (Meissner & Henderson, 1987;Michalak, 1988;Kentish & Xiang, 1997).Changes in pH in skeletal muscle cells are observed during prolonged activity and fatigue (Fitts, 1994), while in cardiac muscle cells, during ischemia (Mohabir et al., 1991;Xu et al., 1996).The effects of H + on cytoplasmic and luminal domains of RyR from rabbit skeletal muscle incorporated into planar lipid bilayer were also investigated (Harrison & Bers, 1987;Rousseau & Pinkos, 1990;Ma & Zhao, 1994;Laver et al., 2000), revealing a difference in time dependence of recovery of the channels (Ma & Zhao, 1994;Laver et al., 2000).Moreover, it was found that ionic composition and ionic strength could affect competitive binding of Mg 2+ to the Ca 2+ activatory and inhibitory binding sites, as well as binding of anions to specific regulatory sites on RyR molecules (Ma et al., 1993 During intensive exercise the concentration of lactate in skeletal muscle cells increases and can reach up to 30 mM.It has been shown that this metabolite may affect RyR activity independently of H + , although acidification of cytoplasm is a feature concomitant with lactate production (Favero et al., 1995).Recently, the results illustrating the effects of lactate on depolarization-induced Ca 2+ release in mechanically skinned skeletal muscle fibers were published.These data revealed, however, that lactate had an only minimal effect on the contractile apparatus and depolarization-induced Ca 2+ release, and was not a major factor in muscle fatigue (Dutka & Lamb, 2000).
It has been observed using SR vesicles that ATP enhances the release of Ca 2+ even in nanomolar Ca 2+   (Blazev & Lamb, 1999a;1999b).Therefore, it has been suggested that an in-crease in adenosine concentration may be at least partially responsible for fatigue of the muscle (Blazev & Lamb, 1999a).Inorganic phosphate (P i ), one of the important metabolites in a muscle cell, is able to influence the contraction-relaxation cycle of cardiac and skeletal muscles.It has been proposed that P i exerts its effects by interaction with the actin-myosin binding function or by modulating earlier events of E-C coupling, namely Ca 2+ release from SR stores.In addition P i could be a factor contributing to the muscle fatigue (Fitts, 1994) (Balog et al., 2000).Moreover, in skinned skeletal muscle fibers, P i enhanced CICR (Balog et al., 2000).On the basis of these data it has been suggested that P i acts on RyR channels via a mechanism distinct from that exerted by adenine nucleotides, and can be an endogenous modulator of the skeletal muscles under fatigue conditions (Balog et al., 2000).
In addition to adenine nucleotides and P i , cADP-ribose (cADPR), a metabolite of b-NAD + , has been also recognized as an endogenous regulator of RyR2 and RyR3 channels in many cell types (Galione et al., 1991;Galione, 1992;Hua et al., 1994;Rakovic et al., 1999).RyR1 isoform of the channel was not activated by cADPR (Sonnleitner et al., 1998).Studies on intact cardiac cells revealed that endogenous cADPR plays a role in regulation of E-C coupling (Iino et al., 1997).It was shown that in guinea pig ventricular cells antagonists of cADPR inhibited arrhythmogenic oscillations of intracellular Ca 2+ and [Ca 2+ ] i oscillations evoked by overloading of SR stores by Ca 2+ in the presence of isoproterenol or ouabain (Rakovic et al., 1999).As was shown by Noguchi et al. (1997), cADPR binds to immunosuppressant FK506 binding protein 12.6 (FKBP) (see next paragraph) within a RyR molecule.Recently, it has been observed that, in neuronal cells NG108-15, cADPR can act as a direct agonist of RyR channels and, in addition, can indirectly interact with L-type voltage-activated Ca 2+ channels (Hashii et al., 2000).
RyRs are integral proteins of SR membrane.About 20% of their molecule is buried in membrane but, surprisingly, the effects of lipids and lipid derivatives on RyR activity are not well documented.Only controversial data related to the action of fatty acids on Ca 2+ release from SR stores are available (Cheah, 1981;Messineo et al., 1984;Zucchi & Ronca-Testoni, 1997).Arachidonic acid, for example, has been found to induce Ca 2+ release from skeletal and cardiac muscles and to inhibit binding of ryanodine to SR membranes, however, a single channel open probability of RyR remained unaffected (Damron & Bond, 1993;el-Hayek et al., 1993;Zucchi & Ronca-Testoni, 1997).The observations of Uehara et al. (1996) suggest that arachidonic acid directly modifies the structure of the ryanodine binding site in RyRs.Recently sphingosine, a long-chain amino-alcohol which is a component of sphingolipids, was found to modulate, in a noncompetitive manner, ryanodine binding to RyRs from skeletal and cardiac muscles, by interacting with a site between Arg4475 and the C-terminus of RyR channels (Needleman et al., 1997;Sharma et al., 2000a).

Protein-protein interaction in regulation of RyR activity
There are many cellular proteins exerting various direct or indirect effects on the activity of RyR channels.Some of them are regu-lated by Ca 2+ or by phosphorylation-dephosphorylation processes (Shoshan-Barmatz & Ashley, 1998;MacKrill, 1999).Some of these proteins interact only with specific isoforms of RyR (Leong & MacLennan, 1998b;MacKrill, 1999).These proteins were divided into a few categories, depending on their cellular location, role and interactions, and were characterized not only as regulators of RyRs but also as other Ca 2+ release channels, the activity of which is regulated by inositol 1,4,5,-trisphosphate (MacKrill, 1999).
Calsequestrin (CSQ), a Ca 2+ storage protein located in the lumen of SR membranes, especially in their terminal cisterns, was overexpressed in mouse cardiac muscles (Jones et al., 1998).Cardiomyocytes from these hearts were by 50-100% larger than in normal animals and had significant electrophysiological and electromechanical defects.In addition, transgenic mice overexpressing cardiac CSQ developed with age extensive electrical remodeling of ionic current, and upregulation of Ca 2+ /Na + exchanger; it has been found that Ca 2+ signaling pathways are the primary site of the defect (Knollmann et al., 2000).The transgenic animals overexpressing CSQ, were characterized by impaired Ca 2+ signaling, resulting from reduced co-ordination and decreased frequency of Ca 2+ sparks; the latter could not be restored by increased calcium current, but with caffeine, which was found to alter the sensitivity of RyR channels to Ca 2+ (Wang et al., 2000).The transgenic mice overexpressing CQS and characterized by heart failure and premature death exhibited also alterations in b-adrenergic receptor signaling preceding the development of heart failure (Cho et al., 1999).Recently, it has been shown that not only the amount of CSQ, but also phosphorylation-dephosphorylation processes, can influence Ca 2+ homeostasis in striated muscles (Szegedi et al., 1999).
Annexin VI, a Ca 2+ and phospholipid binding protein, was shown to regulate RyR2 activity, at the nanomolar concentration range.

Annexin VI increased the mean open time and open probability of RyR2 channels (Diaz-Munoz et al., 1990
).Moreover, targeted overexpression of annexin VI in transgenic mice caused altered signaling, attributed to the probable effects of annexin VI on plasma membrane Ca 2+ -ATPase or Na + /Ca 2+ exchanger (Gunteski-Hamblin et al., 1996).
Sarcalumenin (150 kDa) and a histidine-rich Ca 2+ binding protein (160 kDa) were identified in the lumen of SR from skeletal muscle cells (Orr & Shoshan-Barmatz, 1996; Shoshan-Barmatz et al., 1996b), and were found to be phosphorylated by SR casein kinase II.ATP required for phosphorylation of these two proteins in the lumen of SR is probably transported via a voltage-dependent anion channel (Shoshan- Barmatz et al., 1996a).Recently, a cardiac isoform of sarcalumenin of 130 kDa, was also identified, and found to be phosphorylated by casein kinase II.Phosphorylation of sarcalumenin resulted in modification of its interaction with RyR (Hadad et al., 1999).
Triadin and junctin are examples of integral membrane proteins that interact with RyR channels and with CSQ (MacKrill, 1999).Triadin from rabbit skeletal muscles was shown, using the ligand overlay and crosslinking analysis, to bind RyR and a1 subunit of DHPR (Caswell et al., 1991).Triadin is localized in the junctional terminal cisterns of SR in rabbit skeletal muscles (immunofluorescence and microscopy analysis).This protein consists of 706 amino acids, mostly basic residues (pI 10.18) (Knudson et al., 1993a;1993b).Using surface plasmon resonance spectroscopy and overlay assay, a functional interaction of cytoplasmic part of triadin with skeletal muscle RyR was demonstrated.This interaction can be important in the control of E-C coupling in skeletal muscles (Groh et al., 1999), since it has been shown that C-terminus of triadin interacts with both CSQ and the luminal portion of RyR (Guo et al., 1996a).Using triadin antibodies that did not discriminate between isoforms, three triadin isoforms of 35 kDa, 40 kDa and 92 kDa, respectively, were identified in rabbit cardiac SR vesicles, as also predicted by cDNA cloning (Guo et al., 1996b).Recently triadin 1 was identified as the predominant isoform expressed in mammalian myocardium (Kobayashi & Jones, 1999).
By using iodide labeled CSQ overlay assays, three major CSQ binding proteins were identified in cardiac SR vesicles (Mitchell et al., 1988).The 26 kDa protein was purified, cloned and named junctin.An identical protein was identified in skeletal muscles (Jones et al., 1995).It has been suggested that interactions between triadins, junctin and CSQ work as a link, regulator and sensor of Ca 2+ in the lumen of SR, and also as a link between RyR and DHPR (Jones et al., 1995).
The larger part of RyR channel which is exposed to the cytoplasmic site of the cell interacts with many accessory proteins; these proteins modulate the RyR activity directly or through the phosphorylation-dephosphorylation processes (Shoshan-Barmatz & Ashley, 1998;MacKrill, 1999).One of among them is calmodulin (CaM), which is a ubiquitous Ca 2+ binding protein for which three binding sites on RyR have been identified: PM1, PM2 and PM3 (Menegazzi et al., 1994;Wagenknecht et al., 1997).The modulation of RyR activity by CaM was shown to be biphasic; at the nanomolar Ca 2+ concentration range CaM stimulates RyR activity; the biphasic profile of stimulation by CaM is probably a consequence of different binding of CaM at different Ca 2+ concentrations (Fuentes et al., 1994;Tripathy et al., 1995;Ikemoto et al., 1995).
Sorcin is a 22 kDa EF-hand Ca 2+ binding protein (Zamparelli et al., 2000), which undergoes Ca 2+ -dependent translocation from cytoplasm to membrane.Sorcin is associated with RyR channels and was determined to be present in cardiac myocytes by co-immunoprecipitation (Meyers et al., 1995).Recently, it has been observed that sorcin, at the nanomolar concentration range, may inhibit RyR2 activity, while phosphorylation of sorcin, mediated by PKA, reduces this inhibition.It was also shown that sorcin interacted with the C-terminus of DHPR from both skeletal and cardiac muscles.It has been proposed that interactions between sorcin and RyR and DHPR can form a platform of interchannel communication during E-C coupling (Lokuta et al., 1997;Meyers et al., 1998).
The S100 family of EF-hand proteins, localized predominantly in the cytosol of muscle cells, was recognized as potent modulators of RyR1 activity in skeletal muscles.It was shown that S100 interacted directly with RyR1 in three putative binding sites.One of them overlaps with the CaM binding site (Schafer & Heizmann, 1996;Treves et al., 1997), leading to speculations about the role of S100 proteins in the cardiac muscle contraction-relaxation cycle (Remppis et al., 1996).FKBP belong to a family of proteins which bind immunosuppressant drugs, FK506 and rapamycin.Among several existing isoforms, the predominant FKBP12 of 12 kDa is engaged in mediating the immunosuppressive actions of the drug.FKBP12 is also involved in catalysis of peptidylpropyl-cis-trans-isomerization, and probably is involved in protein folding.Binding of FK506 and/or rapamycin to FKBP12 inhibits its isomerase activity and alters its association with other proteins (Marks, 1996).In addition, it has been shown that FKBP12 binds to cardiac RyR isoform and modulates its channel activity (Timerman et al., 1993), with a stoichiometry of one FKBP12 molecule per one monomer of functional tetramer of RyR (Wagenknecht et al., 1997).RyR2 channels from heart muscle interact with a specific isoform of FKPB12, FKBP12.6, while the skeletal isoform (RyR1), interacts with both FKBP12 and FKBP12.6 (Timerman et al., 1996).The selectivity of FKBP12.6 for RyR2 isoform is related to differences in three amino acids between FKBP12 and FKBP12.6 (Xin et al., 1999).Phosphorylation of FKPB12.6 by PKA dissociated this protein from RyR2.This dissociation regulates the channel open probability of RyR.Using cosedimentation and coimmunoprecipitation methods a macromolecular complex composed of RyR2, FKBP12.6,PKA, the protein phosphatases PP1 and PP2A, and anchoring protein mAKAP, has been identified (Marx et al., 2000).This complex is probably involved in regulation of RyR channel activity by PKA; in failing heart this complex is hyperphosphorylated by PKA leading to abnormal activity of RyR (Marx et al., 2000).

REACTIVE OXYGEN SPECIES
RyR channels are under the constant influence of reactive oxygen species, such as superoxide radical anion, singlet oxygen, hydrogen peroxide, and hydroxyl radicals that are by-products of oxidative metabolism, and nitric oxide as a product of nitric oxide synthase.Production of reactive oxygen species is increased during inflammation, aging, endotoxic shock, and ischemia-reperfusion of heart and other tissues (Kourie, 1998).Recently, oxidative and nitrosilation processes were shown to exert a modulatory role on RyR channel activity (Zable et al., 1997;Xu et al., 1998;Zhang et al., 1999;Haarmann et al., 1999;Suko et al., 1999;Hart & Dulhunty, 2000).In addition, it has been found that oxidation of sulfhydryl groups overrides Mg 2+ inhibition of CICR in skeletal muscles (Donoso et al., 2000).Oxidative stress may affect the binding of CaM to RyR, changing in this way the activity of the channel (Porter Moore et al., 1999; Suko et al., 2000).

CONCLUDING REMARKS
Ca 2+ liberated from internal cellular stores (SR) regulates the contraction of mammalian cardiac and skeletal muscles.The systems involved in Ca 2+ regulation in these muscles consist of proteins that form of similar pattern.The only differences are in tissue-specific isoforms, their distribution, and spatial organization.Endogenous compounds which regulate the activity of RyR and DHPR channels by direct or indirect interactions, display different regulatory abilities.For example, Mg 2+ , which in the case of skeletal muscles inhibits RyR activity at the micromolar concentration range, has no effect on RyR activity in cardiac muscles.The differences in amino-acid composition between DHPR and RyR from cardiac and skeletal muscle form a basis for differences in regulation of opening of RyR channels.As a consequence, the release of Ca 2+ from SR is fully responsible for the contraction of skeletal muscles, but only in part in heart muscle (Chiesi et al., 1994;Wrzosek, 1999).However, as suggested by Eisner et al. (1998), the temporary inhibition or activation of RyR channels produces only a transient effect on systolic Ca 2+ .It seems that more important for E-C coupling are changes in Ca 2+ compartmentalization within the cell.The disturbances of these processes may lead to development of pathological states related to impaired Ca 2+ homeostasis.