Identification of calcium release-triggering and blocking regions of the II-III loop of the skeletal muscle dihydropyridine receptor.

In an attempt to identify and characterize functional domains of the rabbit skeletal muscle dihydropyridine receptor α subunit II-III loop, we synthetized several peptides corresponding to different regions of the loop: peptides A, B, C, C1, C2, D (cf. Fig. 1). Peptide A (Thr-Leu) activated ryanodine binding to, and induced Ca release from, rabbit skeletal muscle triads, but none of the other peptides had such effects. Peptide A-induced Ca release and activation of ryanodine binding were partially suppressed by an equimolar concentration of peptide C (Glu-Pro) but were not affected by the other peptides. These results suggest that the short stretch in the II-III loop, Thr-Leu, is responsible for triggering SR Ca release, while the other region, Glu-Pro, functions as a blocker of the release trigger. A hypothesis is proposed to account for how these subdomains interact with the sarcoplasmic reticulum Ca release channel protein during excitation-contraction coupling.

In an attempt to identify and characterize functional domains of the rabbit skeletal muscle dihydropyridine receptor ␣ 1 subunit II-III loop, we synthetized several peptides corresponding to different regions of the loop: peptides A, B, C, C1, C2, D (cf. Fig. 1

). Peptide A (Thr 671 -Leu 690 ) activated [ 3 H]ryanodine binding to, and induced
Ca 2؉ release from, rabbit skeletal muscle triads, but none of the other peptides had such effects. Peptide A-induced Ca 2؉ release and activation of ryanodine binding were partially suppressed by an equimolar concentration of peptide C (Glu 724 -Pro 760 ) but were not affected by the other peptides. These results suggest that the short stretch in the II-III loop, Thr 671 -Leu 690 , is responsible for triggering SR Ca 2؉ release, while the other region, Glu 724 -Pro 760 , functions as a blocker of the release trigger. A hypothesis is proposed to account for how these subdomains interact with the sarcoplasmic reticulum Ca 2؉ release channel protein during excitation-contraction coupling.
The electrical signal elicited at the T-tubule 1 membrane is transmitted to the sarcoplasmic reticulum (SR) to induce Ca 2ϩ release, which in turn leads to muscle contraction (1)(2)(3)(4)(5)(6)(7)(8). According to the current widely accepted view, upon T-tubule depolarization a portion of the dihydropyridine receptor (DHPR), the voltage-sensing protein in the T-tubule, undergoes a conformational change to make contact with the ryanodine receptor (RyR) to open its Ca 2ϩ release channel (9 -13). The idea that the cytoplasmic loop linking Repeats II and III of the ␣ 1 subunit of the DHPR, the so-called II-III loop, may play an essential role in this process has emerged from an earlier finding that this portion of the DHPR is the critical determinant of the skeletal muscle-type Ca 2ϩ current (14). This view has been further supported by recent findings that the expressed II-III loop (both skeletal and cardiac isoforms) enhanced the ryanodine binding to the skeletal muscle RyR (15). The site important for activation of ryanodine binding was localized in the region encompassing residues Glu 666 -Glu 726 (16), which contains the phosphorylatable serine 687 (17). Furthermore, a recent study with dysgenic myotubes expressing the chimeric (skeletal/cardiac) DHPR has shown that the critical determinant of the skeletal muscle-type Ca 2ϩ transient is localized in the stretch of residues Glu 726 -Pro 742 (18). In this study, using synthetic peptides corresponding to different regions of the II-III loop of rabbit skeletal muscle DHPR ␣ 1 subunit, we identified the region responsible for triggering Ca 2ϩ release and another region for blocking the release. The implication of these findings on the E-C coupling mechanism is discussed.

EXPERIMENTAL PROCEDURES
Preparation-Triad-enriched microsomal fractions (triads) were prepared from rabbit back paraspinous and hind leg skeletal muscles by differential centrifugation as described previously (19). Microsomes from the final centrifugation were resuspended in a solution containing 0.3 M sucrose, 0.15 M potassium gluconate, proteolytic enzyme inhibitors (0.1 mM phenylmethylsulfonyl fluoride, 10 g/ml aprotinin, 0.8 g/ml antipain, 2.0 g/ml soybean trypsin inhibitor), 20 mM MES, pH 6.8, to a final concentration of 20 -30 mg/ml, frozen immediately in liquid N 2 , and stored at Ϫ70°C.
Synthesis of II-III Loop Peptides-Peptides were synthesized on an Applied Biosystems model 431 A synthesizer employing Fmoc (N-(9fluorenyl)methoxycarbonyl) as the ␣-amino protecting group. Peptides were cleaved and deprotected with 95% trifluoroacetic acid. Purification was carried out by reversed-phase high pressure liquid chromatography using a Rainin Instruments preparative C8 column.
Ca 2ϩ Release Assays-Triads (1 mg/ml) were incubated in a solution containing 0.15 M KCl, 1 mM Mg-ATP, an ATP-regenerating system, 20 mM MES, pH 6.8 (Solution A) for 6 -7 min to load the SR moiety with Ca 2ϩ . Then, 1 volume of Solution A was mixed with 1 volume of Solution B containing 0.15 M KCl, 20 mM MES, pH 6.8, and various concentrations of peptides. The Ca 2ϩ concentration in both solutions was buffered at 3 M using an EGTA-calcium buffer (4.22 mM CaCl 2 , 5 mM EGTA). The time course of SR Ca 2ϩ release was monitored in a stopped flow apparatus (Bio-Logic SFM-3) using 10 M arsenazo III as a Ca 2ϩ indicator (20). Twenty to twenty-five traces (each representing 1000 data points) of the arsenazo III signal were averaged for each experiment. The arsenazo III signal was converted to nanomoles of Ca 2ϩ released per mg of protein by determining the ⌬ arsenazo III signal/⌬ [Ca 2ϩ ] coefficient from a Ca 2ϩ calibration curve. Time courses of Ca 2ϩ release were determined at different peptide concentrations. Curves were fitted by a single exponential function, y ϭ A(1 Ϫ e Ϫkt ), where y is the amount of Ca 2ϩ released at time t, A is the final amount of Ca 2ϩ released at an infinite time, and k is the rate constant of release.
Ryanodine Binding Assay-Rabbit skeletal muscle triads (0.5 mg/ml) or porcine cardiac microsomes (1.0 mg/ml) were incubated in 0. 2). The specific binding was calculated as the difference between the binding in the absence (total binding) and in the presence (nonspecific binding) of 10 M unlabeled ryanodine (21). Experiments were carried out in duplicate; each datum point is obtained by averaging the duplicates. Nonspecific binding was Ͻ10% of total binding.

RESULTS AND DISCUSSION
In an attempt to identify the subdomains of the II-III loop of the ␣ 1 subunit of the DHPR that play important roles in exci-tation-contraction coupling, we synthetized several peptides corresponding to different regions of the loop as shown in Fig.  1 and investigated the effect of each of those synthetic peptides on [ 3 H]ryanodine binding to, and Ca 2ϩ release from, rabbit skeletal muscle triads. Fig. 2A depicts the extent of ryanodine binding activation/inhibition (expressed as percent of control) induced by various concentrations of these peptides. Of all the peptides investigated up to a concentration of 50 M, only peptide A produced significant activation of ryanodine binding. Increasing concentrations of peptide A progressively increased ryanodine binding to a maximal level (about 230% of control). However, peptides B, C, C1, C2, and D produced virtually no effect on the ryanodine binding. Mirroring the ryanodine binding experiments ( Fig. 2A), only peptide A induced an appreciable SR Ca 2ϩ release (Fig. 2B). Thus, at a maximally activating concentration (20 M, see the inset to Fig. 2B), peptide A induced a significant amount of Ca 2ϩ release from SR. However, equimolar concentrations of all the other peptides induced virtually no Ca 2ϩ release.
Peptide A produced no appreciable effects on ryanodine binding to microsomes isolated from porcine cardiac muscle (percent of control: at 20 M peptide A, 102 Ϯ 7 (n ϭ 3); at 50 M, FIG. 1. The location and amino acid sequence of the various  synthetic peptides (A, B, C, C1, C2, and D) encompassing different regions of the II-III loop of the ␣ 1 subunit of the rabbit skeletal muscle dihydropyridine receptor.

FIG. 2. Effects of various synthetic peptides of the II-III loop on [ 3 H]ryanodine binding (A) and SR calcium release from skeletal muscle triads (B).
A, 50 g of SR triads were incubated with 8 nM  11 (n ϭ 4)). This is in agreement with the recent report that the expressed II-III loop activates the skeletal muscle RyR but not the cardiac RyR isoform (15).
Under the same conditions as above, in which peptide A produced significant activation, the whole II-III loop expressed in Escherichia coli (22) had virtually no effects on ryanodine binding nor induced Ca 2ϩ release, unless 5 mM AMP was added as done in the original study by Meissner and co-workers (15). This suggests that there might be an inhibitory domain counteracting the peptide A region within the II-III loop. Indeed, as shown by the experiments in Fig. 3, A and B, the presence of 50 M peptide C, but not the other peptides (B, C1, C2, or D), produced significant suppression of the activation of ryanodine binding induced by 50 M peptide A (Fig. 3A). Again mirroring the ryanodine binding experiments, an equimolar concentration (20 M in this case) of peptide C produced significant inhibition of SR Ca 2ϩ release induced by peptide A. However, peptides B, C1, C2, and D had no effect. It is particularly interesting that neither peptide C1 nor C2, which represent the two subdomains of peptide C, had any Ca 2ϩ release blocking effect by themselves. This indicates that both C1 and C2 subdomains must be linked to exert the blocking function.
Several important new properties of the II-III loop of the DHPR are revealed in this study. Most importantly, we could localize the critical site for activating the RyR/Ca 2ϩ release channel to peptide A (Thr 671 -Leu 690 ), which represents approximately one-third of the recently reported 61-residue ryanodine binding activating peptide of the II-III loop (16). Another important aspect of this study is the finding of peptide C, which antagonized the effect of peptide A on Ca 2ϩ release or ryanodine binding. These results suggest that there are at least two functionally important subdomains in the II-III loop: an activator that is responsible for the stimulation of the RyR/Ca 2ϩ release channel in E-C coupling and a blocker that antagonizes the activator. These results suggest an intriguing hypothesis as follows. In the resting state, the putative signal receptor site in the RyR is occupied by the blocker domain of the loop. Upon depolarization, the blocker domain (corresponding to peptide C) is removed from the site; then the activator domain (corresponding to peptide A) is allowed to interact with the site to trigger SR Ca 2ϩ release. In the present study, the activation of SR Ca 2ϩ release by peptide A was produced presumably by competitive binding with the blocker domain to the signal receptor (in the case of coupled RyR) or by direct binding (in the case of uncoupled RyR). Peptide C1 (Phe 725 -Gly 743 ) used in the present study covers the 17-residue (Glu 726 -Pro 742 ) region reported to be a critical determinant for the skeletal muscle-type regulation (18), which requires a physical contact of the II-III loop to the RyR (11). On this basis, we tentatively propose that the C1 subdomain may behave like a hinge for this blocker/ activator exchange operation.