INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Depolarization of the skeletal muscle plasma membrane results in the spread of the action potential over the surface and transverse-tubule membranes. Transverse-tubule depolarization effects a structural change in the dihydropyridine receptor/L-type Ca²⁺ channel that results in the release of Ca²⁺ from the sarcoplasmic reticulum (SR) via the high-conductance Ca²⁺ release channel/ryanodine receptor protein (RyR1). RyR1 is regulated in a complex fashion by numerous endogenous effectors and, in the absence of other modulators, is activated by Ca²⁺ concentrations in the nano- to micromolar range and inhibited by Ca²⁺ concentrations in the micro- to millimolar range. Thus, it has been concluded that the RyR1 has a high-affinity divalent binding site, which when occupied by Ca²⁺ will activate the channel (A-site), and a low-affinity divalent binding site, which when occupied by Ca²⁺ will inhibit the channel (I-site) (Meissner, 1994).

RyR1 channel opening is inhibited by physiological concentrations of Mg²⁺ (Endo, 1977; Meissner, 1994). Mg²⁺ could potentially be a competitive antagonist at the A-site on the RyR1, be an agonist at the I-site, or inhibit RyR1 channel opening via an independent inhibitory site. Laver et al. (1997a) and Meissner et al. (1997) have suggested that Mg²⁺ inhibits RyR1 channel opening via its interaction with both Ca²⁺ binding sites and that the extent of the interaction of Mg²⁺ with the two regulatory sites depends on the Ca²⁺ concentration. Although the regulation of the RyR1 by Ca²⁺ and Mg²⁺ has been extensively studied, and the Ca²⁺ affinity of the A- and I-sites estimated (Zucchi and Ronca-Testoni, 1997; Meissner, 1994), the Mg²⁺ affinity of these sites has been reported only for frog RyRs (Murayama et al., 2000). In addition, the apparent affinity of each of these sites is dependent on the conditions under which the measurements are made (Fruen et al., 1996; Meissner et al., 1997). Thus, until the affinities of the two binding sites for Ca²⁺ and Mg²⁺ are determined under identical conditions, the potential physiological role of Mg²⁺ at each site cannot be concluded.

Malignant hyperthermia (MH) is a pharmacogenetic disorder originating primarily from mutations in the RyR1. Although in the human population there are 24 known RyR1 MH mutations (McCarthy et al., 2000; Jurkatt-Rott et al., 2000), the primary defect in porcine MH is a single point mutation (Arg⁶¹⁵Cys) in the RyR1 (Fujii et al., 1991). RyR1 from MH-susceptible (MHS) individuals exhibits a decreased sensitivity to inhibition by high concentrations of Ca²⁺ (Mickelson et al., 1988, 1990; Shomer et al., 1993; Richter et al., 1997). A greater sensitivity to Ca²⁺ activation has also been reported (Shomer et al., 1993; Herrmann-Frank et al., 1996; Richter et al., 1997). Although the Mg²⁺ regulation of these channels also appears to be altered, most studies have focused on the interaction of Mg²⁺ with the low-affinity I-site (Mickelson et al., 1990; Laver et al., 1997a; Owen et al., 1997). In their comparison of RyR1 from normal and MHS pigs, Laver et al. (1997b) reported that in the presence of 1 µM Ca²⁺, a Ca²⁺ concentration where they had previously found Mg²⁺ inhibition via the A- and I-sites to be equally important (Laver et al., 1997a), a higher Mg²⁺ concentration was required to half-maximally inhibit MHS channels compared with normal channels. Although the reported decrease in the Mg²⁺ sensitivity of the I-site contributed to the diminished Mg²⁺ inhibition, the possible involvement of the A-site, via an increased Ca²⁺ or decreased Mg²⁺ affinity, cannot be excluded.

The application of caffeine to skeletal muscle can trigger SR Ca²⁺ release and muscle contraction (Herrmann-Frank et al., 1999). MHS skeletal muscle is more sensitive to caffeine-induced contracture than skeletal muscle from normal individuals, and this enhanced caffeine sensitivity is integral to the clinical diagnosis of MH (Jurkatt-Rott et al., 2000). However, whether the MHS RyR1 itself is more sensitive to caffeine remains controversial (Shomer et al., 1994; Herrmann-Frank et al., 1996). Shomer et al. (1994) reported that the MHS RyR1 is no more sensitive to caffeine than the normal RyR1 and suggested that the increased caffeine sensitivity of MHS muscle may be secondary to an elevated resting myoplasmic Ca²⁺ concentration or altered Ca²⁺ regulation of the RyR1. Although it has been reported recently that the enhanced caffeine sensitivity of MHS muscle is mediated by an increase in the resting myoplasmic Ca²⁺ concentration (Lopez et al., 2000), the effect of caffeine on the Ca²⁺ and Mg²⁺ affinities of the A- and I-sites of the MHS and normal RyR1 have not been rigorously examined.

We have now estimated the Ca²⁺ and Mg²⁺ affinities of the normal and MHS RyR1 A- and I-sites under identical conditions in the presence and absence of caffeine. Compared with the normal RyR1, the MHS RyR1 I-site has a lower apparent affinity for both Ca²⁺ and Mg²⁺. In contrast, compared with the normal RyR1, the MHS RyR1 A-site has a higher apparent affinity for Ca²⁺ but a lower apparent affinity for Mg²⁺. In addition, caffeine increased the Ca²⁺ affinity of the MHS and normal RyR1 A-sites to a similar extent. However, caffeine increased the Mg²⁺ affinity of the normal RyR1 A-site but not of the MHS RyR1 A-site. Thus, the MH mutation has opposite effects on the Ca²⁺ and Mg²⁺ affinities of the RyR1 A-site that would greatly enhance the sensitivity of the MHS RyR1 to Ca²⁺ activation in intact muscle.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

[³H]Ryanodine binding

Isolation of SR vesicles

[³H]Ryanodine binding

Single-channel studies

The RyR1 was purified from SR membrane vesicles as described previously (Shomer et al., 1993). Muller-Rudin planar lipid bilayers were formed by painting a lipid mixture (phosphatidylethanolamine, phosphatidylserine, and phosphatidylcholine in a 5:3:2 ratio by weight, 50 mg/ml dissolved in n-decane) across a 250-µm aperture in a Delrin cup. The cis chamber was connected to the headstage input of an Axoclamp 200B patch clamp amplifier (Axon Instruments, Foster City, CA). The trans chamber was held at virtual ground. Data was filtered at 2 kHz with an eight-pole Bessel filter, recorded at 4.5 kHz, and stored on a Jazz disk drive (Iomega, Roy, UT). Recording solution consisted of symmetric 100 mM KCl, 10 mM HEPES, pH 7.4, 1 mM EGTA. The Mg²⁺ and Ca²⁺ concentrations were adjusted by adding small aliquots of concentrated EGTA, CaCl₂, and MgCl₂ (Brooks and Storey, 1992). Single-channel data were collected using a pulsing protocol in which the potential was held at 0 mV for 4 s between steps of 2-s duration to +70 mV (CLAMPEX program, pClamp software, Axon Instruments, Foster City, CA). Only those channels that had a conductance of at least 700 pS were used (Shomer et al., 1994). Single-channel open probability (P_o) was calculated from at least 50 2-s sweeps using FETCHAN and PSTAT analysis programs (pClamp software, Axon Instruments). When two channels were present in the bilayer, indicated by current amplitudes of twice the expected magnitude, P_o was estimated as the average P_o of the two channels, calculated as [P_{o,level 1} + (P_{o,level 2} × 2)]/2. Bilayers in which three channels had incorporated were dealt with similarly; recordings were not made from bilayers containing more than three channels.

Analysis

The Ca²⁺ and Mg²⁺ affinities of the A- and I-sites of RyR1 were estimated according to the model of Murayama et al. (2000). The fraction of A-sites bound with Ca²⁺ (_A) and the fraction of I-sites not bound with Ca²⁺ or Mg²⁺ (1  _I) were expressed as

(1)

(2)

where K_A,Ca, K_A,Mg, K_I,Ca, and K_I,Mg are the apparent affinity constants for Ca²⁺ and Mg²⁺ of the A- and I-sites, respectively. n_A,Ca, n_A,Mg, n_I,Ca, and n_I,Mg are the Hill coefficients for Ca²⁺ and Mg²⁺ of the A- and I-sites, respectively.

The above parameters were determined in a three-step procedure. First, K_A,Ca, n_A,Ca, K_I,Ca, and n_I,Ca were determined from the Ca²⁺ dependence of SR vesicle [³H]ryanodine binding (B) in the absence of Mg²⁺ according to Eq. 3:

(3)

B_peak in these equations indicates the maximal SR vesicle [³H]ryanodine binding under the conditions of the experiment. Then, the concentration dependence of Mg²⁺ inhibition of SR vesicle [³H]ryanodine binding was measured at a Ca²⁺ concentration much greater than K_A,Ca. At this Ca²⁺ concentration, Ca²⁺ is bound to the A-site and the inhibition by Mg²⁺ occurs via the I-site. K_I,Mg and n_I,Mg were determined by fitting the data from these experiments with Eq. 4 and including the values previously obtained for K_A,Ca,n_A,Ca,K_I,Ca,andn_I,Ca:

(4)

Finally, the concentration dependence of Mg²⁺ inhibition of SR vesicle [³H]ryanodine binding was measured at a Ca²⁺ concentration near K_A,Ca. At this Ca²⁺ concentration, competitive inhibition by Mg²⁺ at the A-site predominates. K_A,Mg and n_A,Mg were determined by fitting the data from these experiments with Eq. 5, using values previously obtained for K_A,Ca, n_A,Ca, K_I,Ca, n_I,Ca, K_I,Mg, and n_I,Mg:

(5)

The model is depicted diagrammatically in Fig. 1 using parameters derived in the presence of 1 mM Mg²⁺. The Ca²⁺ dependence of the fraction of channels activated, i.e., SR vesicle [³H]ryanodine binding (solid line), is the product of the fraction of channels with the A-site bound with Ca²⁺ (f_A, dashed and dotted line) and the fraction of channels with the I-site free of both Ca²⁺ and Mg²⁺ (1  f_I, dashed line). As a competitive antagonist of Ca²⁺ for binding to the A-site, Mg²⁺ effectively increases K_A,Ca, shifts the Ca²⁺ dependence of f_A to higher Ca²⁺ concentrations, and increases the Ca²⁺ required for channel activation. Because Mg²⁺ is an agonist at the I-site, Mg²⁺ will decrease the fraction of channels with the I-site free of both Ca²⁺ and Mg²⁺, resulting in a reduction in the number of channels available for activation. Thus, it is clear that the diagram is drawn as would occur in the presence of a Mg²⁺ concentration somewhat below the K_I,Mg (at very low Ca²⁺ concentrations, 1  f_I > 0.5). It should be pointed out that in the presence of Mg²⁺, differences in the Ca²⁺ dependence of MHS and normal RyR1 channel activation could potentially occur via an increase in the Ca²⁺ affinity of the A-site, by a decrease in the Mg²⁺ affinity of the A-site, or both.

View larger version (21K): [in this window] [in a new window]   FIGURE 1   Diagrammatic representation of the model described in Materials and Methods used to derive the Ca2+ and Mg2+ affinities of the RyR1 A- and I-sites. The Ca2+ dependence of channel activation () is the product of the fraction of channels with Ca2+ bound to the A-site (fA,  · · ) and the fraction of channels with the I-site free of both Ca2+ and Mg2+ (1-fI,   ). Because Mg2+ is a competitive antagonist with Ca2+ at the A-site, Mg2+ will effectively increase the KA,Ca thereby shifting the Ca2+ dependence of channel activation to higher Ca2+ concentrations. As an agonist at the I-site, Mg2+ will decrease the fraction of channels with the I-site free of Ca2+ and Mg2+ and thus decrease the number of channels available for activation. Note that the parameters illustrated were derived in the presence of 1 mM Mg2+.

In an initial experiment, to determine whether MHS and normal RyR1s differ in their sensitivities to inhibition by Mg²⁺, the Mg²⁺ concentration dependence of MHS and normal SR [³H]ryanodine binding was compared in the presence of 10 µM Ca²⁺. In this experiment the inhibitory effect of Mg²⁺ could not be attributed to its action at a single site. Therefore, the half-inhibitory (IC₅₀) concentrations were determined using the Hill equation. Curve fitting was performed using SigmaPlot 5.0 (SPSS, Richmond, CA) software. All data are expressed as mean ± SEM. Comparisons between muscle types or treatments performed were made via two-sample t-tests with the level of significance set at p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The model used to estimate the Ca²⁺ and Mg²⁺ affinities of the RyR1 A- and I-sites (Murayama et al., 2000) is dependent on the assumption that Mg²⁺ is a competitive inhibitor with Ca²⁺ at the A-site and is an agonist at the I-site. This assumption was confirmed in single-channel studies (Fig. 2). Thus, if channels were activated by low concentrations of Ca²⁺ (near K_A,Ca), the subsequent addition of a low concentration of Mg²⁺ should compete with Ca²⁺ for the A-site and decrease the mean single-channel percent open time. Under these conditions, increasing concentrations of Ca²⁺ would effectively compete with Mg²⁺ for the A-site and increase the single-channel percent open time. As shown in Fig. 2 A, a normal RyR1 channel activated by 300 nM cis Ca²⁺ had a mean single-channel percent open time of 4.19. The addition of 50 µM Mg²⁺ to the cis chamber decreased the percent open time to 0.92. Increasing the Ca²⁺ concentration in the cis chamber to 3 µM increased the single-channel percent open time to 5.06. Similar results were obtained in all six experiments, although the sensitivity of the channels to Ca²⁺ and Mg²⁺ varied (Fig. 2 C). Thus, these single-channel experiments are consistent with the hypothesis that Mg²⁺ can act as a competitive antagonist with Ca²⁺ at the A-site.

View larger version (54K): [in this window] [in a new window]   FIGURE 2   Effects of Ca2+ and Mg2+ in combination on the single-channel activity of the RyR1. Single-channel activity of normal channels was recorded as described in Materials and Methods. Recording solution contained 100 mM KCl, 10 mM HEPES (pH 7.4), 2 mM EGTA, and CaCl2 and MgCl2 to obtain the indicated ionized Ca2+ and Mg2+ concentrations. Single-channel currents were obtained with voltage steps from 0 mV to +70 mV. The solid line indicates the closed state of the channels; openings are upward. (A) (a) Single channel activated by 300 nM cis Ca2+. Percent open time (%OT) = 4.19, mean open time (OT) = 1.0 ms, and mean closed time (CT) = 8.4 ms; (b) Addition of 50 µM cis Mg2+ decreased the %OT to 0.92 (OT = 0.9 ms, CT = 21.4 ms); (c) Increasing cis Ca2+ to 3 µM reactivated the channel (%OT = 5.06, OT = 1.2 ms, CT = 8.8 ms). (B) (a) Single channel activated by 300 µM cis Ca2+ (%OT = 1.23, OT = 1.0 ms, CT = 16.6 ms); (b) Addition of 50 µM cis Mg2+ decreased %OT to 1.05 (OT = 0.9 ms, CT = 17.2 ms); (c) Increasing cis Ca2+ to 400 µM further decreased the %OT to 0.51 (OT = 1.0 ms, CT = 32.5 ms). Each color bar in C and D represents a different experiment. In all experiments, the addition of 50 µM Mg2+ to the cis chamber reduced the mean single-channel percent open time when the free Ca2+ concentration was either 300 nM, n = 6 (C) or 300 µM, n = 6 (D). (C) When the Ca2+ concentration was increased from 300 nM to 3 µM in the presence of Mg2+, the %OT increased. (D) When the Ca2+ concentration was increased from 300 µM to 400 µM in the presence of Mg2+, the %OT decreased.

Single-channel studies also confirmed the assumption that at Ca²⁺ concentrations sufficient to saturate the A-site, Mg²⁺ interacts with the I-site. Thus, the single normal channel in Fig. 2 B, activated by 300 µM Ca²⁺ had a mean single-channel percent open time of 1.23. The addition of 50 µM Mg²⁺ to the cis chamber lowered the percent open time to 1.05. However, in contrast to experiments in 300 nM Ca²⁺, this channel could not be reactivated by the subsequent addition of Ca²⁺. Indeed, the additional 100 µM Ca²⁺ added to the cis chamber further reduced the percent open time to 0.51. Similar results were obtained with all six experiments (Fig. 2 D). Although the extent of inhibition was variable, in no case did increasing cis Ca²⁺ increase the percent open time of Mg²⁺-inhibited channels. Thus, when the Ca²⁺ concentration is greater than that required to maximally activate the RyR1 (i.e., when the A-site is in the Ca²⁺ bound state), Mg²⁺ is an agonist at the low-affinity I-site.

The maximal [³H]ryanodine binding (i.e., in 500 mM KCl, 10 µM Ca²⁺, and 6 mM ATP) for 9 normal (11.6 ± 1.1 pmol/mg protein) and 11 MHS (9.5 ± 0.8 pmol/mg protein) SR preparations were not significantly different. Both MHS and normal SR exhibited the characteristic bell-shaped Ca²⁺ dependence of [³H]ryanodine binding. However, compared with normal SR, MHS SR [³H]ryanodine binding was more sensitive to Ca²⁺ activation and less sensitive to inhibition by Ca²⁺ (Fig. 3). The K_A,Ca, n_A,Ca, K_I,Ca, and n_I,Ca of the RyR1 determined for both muscle types according to Eq. 3 are presented in Table 1. The MHS K_A,Ca was significantly smaller than the normal K_A,Ca; in contrast, the MHS K_I,Ca was more than 1.7-fold greater than the normal K_I,Ca. Thus, compared with the normal RyR1, the MHS RyR1 A-site had a higher apparent affinity for Ca²⁺ whereas the I-site had a lower apparent affinity for Ca²⁺.

View larger version (19K): [in this window] [in a new window]   FIGURE 3   Comparison of the Ca2+ dependence of [3H]ryanodine binding to normal and MHS SR vesicles. [3H]Ryanodine binding to normal () and MHS () skeletal muscle SR vesicles was determined as described in Materials and Methods. Media contained 100 mM KCl, 10 mM HEPES (pH 7.4), and Ca-EGTA buffer set to provide the indicated free Ca2+ concentrations. Data are expressed as percentages of the maximal [3H]ryanodine binding capacity of the SR preparations. Solid lines are based on fits to Eq. 3 (see Materials and Methods). Means ± SEM are of seven independent experiments performed in duplicate (seven different SR vesicle preparations).

To determine whether the normal and MHS RyR1 also differ in their sensitivity to Mg²⁺ inhibition, the Mg²⁺ dependence of SR vesicle [³H]ryanodine binding was determined in the presence of 10 µM Ca²⁺, a Ca²⁺ concentration that is more than 30 times the K_A,Ca. As shown in Fig. 4, in the presence of 10 µM Ca²⁺ the concentration dependence of Mg²⁺ inhibition of [³H]ryanodine binding to MHS SR was shifted to significantly higher Mg²⁺ concentrations compared with that of normal SR. The IC₅₀ values, derived from the Hill equation, for MHS and normal SR were 652.0 ± 46.5 µM and 304.9 ± 49.6 µM Mg²⁺, respectively. Thus, [³H]ryanodine binding to MHS SR appears to be less sensitive to inhibition by Mg²⁺ than is [³H]ryanodine binding to normal SR.

View larger version (23K): [in this window] [in a new window]   FIGURE 4   Inhibition of [3H]ryanodine binding to normal and MHS SR by Mg2+. [3H]Ryanodine binding to normal () and MHS () skeletal muscle SR vesicles was determined as described in Materials and Methods. Media contained 100 mM KCl, 10 mM HEPES (pH 7.4), and 2 mM EGTA. Ca2+ and Mg2+ concentrations were adjusted to maintain an ionized Ca2+ concentration of 10 µM and various ionized Mg2+ concentrations as indicated. Data are expressed as percentages of the maximal [3H]ryanodine binding capacity of the SR vesicle preparations; solid lines are based on fits to the Hill equation. Means ± SEM are of five independent experiments preformed in duplicate (five different SR vesicle preparations).

From the experiments described above, we conclude that compared with the normal RyR1, the MHS RyR1 A-site has a higher Ca²⁺ affinity, the I-site has a lower Ca²⁺ affinity, and the MHS channels are less sensitive to inhibition by Mg²⁺. However, because Mg²⁺ inhibition occurs via its binding to both regulatory sites, and there are significant differences in the Ca²⁺ affinities of both sites, it is not possible, from the data presented in Fig. 4, to identify the mechanism responsible for the decreased inhibition of the MHS RyR1 by Mg²⁺. The MHS RyR1 I-site may have a reduced Mg²⁺ affinity as suggested by Laver et al. (1997b), the MHS RYR1 A-site may have a reduced Mg²⁺ affinity, or both. Furthermore, an increased affinity of the MHS RyR1 A-site for Ca²⁺, with no change in the Mg²⁺ affinity, could also result in a decreased competitive inhibition of the MHS RyR1 by Mg²⁺. To distinguish between these possibilities, we estimated the affinities of the normal and MHS RyR1 A- and I-sites for Mg²⁺.

We determined the Mg²⁺ dependence of the inhibition of [³H]ryanodine binding in the presence of 300 µM Ca²⁺, a concentration of Ca²⁺ that is 1000-fold greater than the RyR1 K_A,Ca (Table 1). At this Ca²⁺ concentration the A-sites should be fully occupied by Ca²⁺, and inhibition of [³H]ryanodine binding by Mg²⁺ should occur primarily via the I-site. Fitting this data with Eq. 4, using the previously determined Ca²⁺ affinities and Hill coefficients, allowed us to estimate the Mg²⁺ affinity of the I-sites of both MHS and normal RyR1 (Fig. 5; Table 1). This analysis indicated that the K_I,Mg of the MHS RyR1 I-site was approximately twice that of the normal RyR1 I-site. However, the relative Mg²⁺/Ca²⁺ affinities for MHS and normal RyR1 were both ~1 (Table 2), indicating the MHS mutation decreases the affinity of the I-site for both divalent ions in a similar fashion.

View larger version (24K): [in this window] [in a new window]   FIGURE 5   Mg2+ dependence of [3H]ryanodine binding to normal and MHS SR in the presence of various concentrations of Ca2+. [3H]Ryanodine binding to normal (A) and MHS (B) skeletal muscle SR was determined as described in Materials and Methods. Media contained 100 mM KCl, 10 mM HEPES (pH 7.4), and 2 mM EGTA. The concentrations of MgCl2 and CaCl2 were adjusted to maintain the ionized Ca2+ concentration of 300 nM (), 10 µM (), or 300 µM () and Mg2+ as indicated in the figure. Data for 10 µM Ca2+ () are replotted from Fig. 3 for comparison with 300 nM and 300 µM Ca2+. Data are expressed as percentages of the [3H]ryanodine binding in the absence of Mg2+; solid lines are based on fits to Eq. 5 (see Materials and Methods). Means ± SEM are of five independent experiments preformed in duplicate (five different SR vesicle preparations).

Next, we examined the concentration dependence of the inhibition of SR [³H]ryanodine binding by Mg²⁺ in the presence of 300 nM Ca²⁺, a Ca²⁺ concentration near the K_A,Ca. At this Ca²⁺ concentration, Mg²⁺ will inhibit RyR1 channel opening primarily via its action at the RyR1 A-site. Fitting these data in Fig. 5 with Eq. 5 provided values for K_A,Mg. As shown in Table 1, the MHS K_A,Mg was approximately twice the value for the normal RyR1. Consequently, in contrast to its effect on the I-site, the MHS mutation altered the apparent affinity of the A-site for Ca²⁺ and Mg²⁺ in opposite ways, increasing the Ca²⁺ affinity and decreasing the Mg²⁺ affinity. As a result, the selectivity of the MHS RyR1 A-site for Ca²⁺ over Mg²⁺ was ~3.5-fold greater than the normal RyR1 A-site (Table 2).

If the model described in the methods and derived parameters in Table 1 are valid, it should be possible to predict the Ca²⁺ dependence of [³H]ryanodine binding to SR vesicles in the presence of various concentrations of Mg²⁺. Therefore, we determined the Ca²⁺ dependence of ryanodine binding to normal and MHS SR vesicles in the presence of 100 µM and 500 µM Mg²⁺ and fit the data according to Eq. 5 using the parameters given in Table 1 (Fig. 6). If the conditions are established such that Mg²⁺ inhibition occurs primarily as a result of its binding to the A-site, a shift in the activation side of the Ca²⁺ dependence curve to higher Ca²⁺ concentrations, with no change in the inactivation side of the curve would be expected. Thus, 100 µM Mg²⁺, a Mg²⁺ concentration near the K_A,Mg increased the Ca²⁺ concentration required to activate normal and MHS SR vesicle [³H]ryanodine binding compared with experiments performed in the absence of Mg²⁺. Half-activating Ca²⁺ concentrations (EC₅₀) of 0.50 ± 0.03 µM and 0.34 ± 0.03 µM were derived for normal and MHS SR, respectively (compare with K_A,Ca in Table 1). In contrast, the Ca²⁺ dependence of RyR1 inhibition was not significantly altered in either muscle type. The IC₅₀ values in the presence of 100 µM Mg²⁺ were 344.7 ± 22.6 µM for normal and 416.3 ± 36.1 µM for MHS SR (compare with K_I,Ca in Table 1). Thus, low concentrations of Mg²⁺ (100 µM Mg²⁺) affect [³H]ryanodine binding primarily via competition with Ca²⁺ for the A-site on the RyR1.

View larger version (29K): [in this window] [in a new window]   FIGURE 6   Ca2+ dependence of [3H]ryanodine binding to normal and MHS SR in the presence or absence of various concentrations of Mg2+. [3H]Ryanodine binding to normal (A) and MHS (B) skeletal muscle SR was determined as described in Materials and Methods. Media contained 100 mM KCl, 10 mM HEPES (pH 7.4), and 2 mM EGTA. The concentrations of MgCl2 and CaCl2 were adjusted to maintain an ionized Mg2+ concentration of either 0 mM (, ), 0.1 mM (, ) or 0.5 mM (, ) and Ca2+ as indicated in the figure. Data are expressed as percentages of the maximal [3H]ryanodine binding capacity of the SR vesicle preparations; solid lines are based on fits to Eq. 5 (see Materials and Methods) and the parameters presented in Table 1. Data for 0 Mg2+ (, ) are replotted from Fig. 1 for comparison with 0.1 and 0.5 mM Mg2+. Means ± SEM are of five to seven independent experiments performed in duplicate (five to seven different SR vesicle preparations).

The lower-affinity I-site has a similar affinity for Ca²⁺ and Mg²⁺ (Table 1). Therefore, a Mg²⁺ concentration near the K_I,Ca should inhibit ryanodine binding by acting at both the A- and I-sites. Accordingly, 500 µM Mg²⁺ shifted both the Ca²⁺ dependence of activation and decreased the maximal extent of Ca²⁺ activation (Fig. 6). This concentration of Mg²⁺ increased the Ca²⁺ EC₅₀ for both types of SR (MHS: 5.86 ± 1.06 µM; normal: 6.36 ± 1.32 µM). Likewise, the Ca²⁺ IC₅₀ was also increased for both MHS (740 ± 8 µM) and normal (630 ± 13 µM) SR.

The fitted lines in Fig. 6 derived from Eq. 5 and using the parameters presented in Table 1 appear to fit the data well and adequately describe the combined effects of Ca²⁺ and Mg²⁺ on ryanodine binding to both MHS and normal SR. Thus, the data presented in Fig. 6 support the validity of both the model and the derived parameters.

Although MHS skeletal muscle fibers are more sensitive to caffeine-induced contraction, Shomer et al. (1994) suggested that an increased Ca²⁺ sensitivity of the MHS RyR1 channel, rather than increased caffeine affinity, underlies the greater responsiveness of MHS muscle to caffeine. Therefore, we examined the effects of 5 mM caffeine on the Ca²⁺ and Mg²⁺ affinities of the normal and MHS RyR1 A-and I-sites (Table 1). Caffeine did not significantly alter the Ca²⁺ or Mg²⁺ affinity of the I-site of either the MHS or normal RyR1. Caffeine had a similar effect on the MHS and normal A-site Ca²⁺ affinity, increasing the apparent affinity approximately 3-4-fold. In contrast, caffeine had divergent effects on the Mg²⁺ affinity of the MHS and normal A-site. The caffeine-induced increase in the Mg²⁺ affinity (3.9-fold) of the normal RyR1 A-site was similar to the increase in the Ca²⁺ affinity (4.1-fold). In contrast, caffeine did not significantly increase the affinity of the MHS RyR1 A-site for Mg²⁺. Thus, 5 mM caffeine increased the Ca²⁺ affinity of the A-site of both channel types to a similar extent. However, caffeine increased the selectivity of the MHS but not the normal A-site for Ca²⁺ over Mg²⁺ (Table 2).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In agreement with previous reports (Mickelson et al., 1988; Shomer et al., 1993; Herrmann-Frank et al., 1996), [³H]ryanodine binding to SR from pigs with the Arg⁶¹⁵Cys RyR1 MHS mutation was more sensitive to Ca²⁺ activation and less sensitive to Ca²⁺ inhibition than SR from normal individuals (Fig. 3; Table 1). An increased sensitivity of Ca²⁺ activation has also been reported for SR isolated from individuals with the RyR1 Gly²⁴³⁴Arg human MH mutation (Richter et al., 1997) as well as individuals with positive MH contracture tests but unknown and probably variable genotypes (Valdivia et al., 1991). Therefore, we conclude that the porcine Arg⁶¹⁵Cys mutation, its human homolog, and likely other MH mutations as well, increase the sensitivity of the RyR1 to activation by Ca²⁺ and decrease the sensitivity of the RyR1 to inhibition by Ca²⁺.

Consistent with Mickelson et al. (1990), we found that Mg²⁺ is a less effective inhibitor of [³H]ryanodine binding to MHS than to normal SR (Fig. 4). Mg²⁺ has also been shown to be less effective at inhibiting MHS RyR1 channel opening (Laver et al., 1997b) and Ca²⁺ release in mechanically peeled MHS muscle fibers (Owen et al., 1997).

Because Mg²⁺ competes with Ca²⁺ for binding to the two divalent cation regulatory sites on RyR1, and the extent of Mg²⁺ inhibition at each site is dependent on the Ca²⁺ concentration (Figs. 5 and 6), the relative magnitude of the Mg²⁺ effect at each site has been difficult to assess. This is of particular importance as it relates to the molecular basis of MH, as it is now clear that RyR1 channels with the Arg⁶¹⁵Cys mutation are more sensitive to activation by Ca²⁺ and less sensitive to inhibition by both Ca²⁺ and Mg²⁺. Although the decreased sensitivity of the MHS RyR1 to inhibition by Mg²⁺ has been extensively studied, it has not been clear whether the decreased inhibition of the MHS RyR1 by Mg²⁺ is due solely to a decreased affinity of the MHS RyR1 I-site for Mg²⁺ or whether alterations in the affinity of the MHS RyR1 A-site for divalent cations also plays a role. Either an increase in the affinity of the A-site for Ca²⁺, or a decreased affinity of the A-site for Mg²⁺ would enhance the ability of Ca²⁺ to compete with Mg²⁺ for the A-site and activate the RyR1. We show here that in addition to a decreased affinity of the MHS RyR1 I-site for Mg²⁺, alterations in the MHS RyR1 A-site contribute to the decreased Mg²⁺ inhibition; i.e., both an increased Ca²⁺ affinity and a decreased Mg²⁺ affinity of the MHS RyR1 A-site contribute to the decreased Mg²⁺ inhibition of the MHS RyR1.

Although the increased sensitivity of MHS skeletal muscle to caffeine-induced contracture is integral to the clinical diagnosis of MH (Jurkatt-Rott et al., 2000), the mechanistic basis for the differential response of normal and MHS muscle to caffeine has been unclear (Shomer et al., 1994; Herrmann-Frank et al., 1996). Shomer et al. (1994) reported no difference in the apparent affinity of the MHS and normal RyR1 for caffeine and suggested the increased caffeine sensitivity of MHS muscle may be due to an increased resting myoplasmic Ca²⁺ concentration and/or alterations in the Ca²⁺ affinity of the RyR1. Recent measurements of intracellular Ca²⁺ in MHS skeletal muscle fibers are consistent with the former hypothesis (Lopez et al., 2000). However, the effects of caffeine on the Ca²⁺ and Mg²⁺ affinity of normal and MHS RyR1 have not been thoroughly examined. The data presented suggest that the increased Ca²⁺ affinity of the MHS RyR1 A-site does indeed contribute to the increased caffeine sensitivity of MHS skeletal muscle. Because caffeine increased the Ca²⁺ affinity of the A-site of both the normal and MHS RyR1 to a similar extent (3-4-fold) the Ca²⁺ affinity of the MHS RyR1 A-site remained significantly higher than the normal RyR1 A-site (Table 1). This alone could increase the sensitivity of MHS skeletal muscle to caffeine-induced contraction. However, because the extent of the caffeine-induced increase in the affinity of the normal RyR1 A-site for Ca²⁺ and Mg²⁺ were similar, the selectivity of the A-site for Ca²⁺ over Mg²⁺ did not change. In contrast, caffeine did not alter the affinity of the MHS A-site for Mg²⁺; thus, caffeine further increased the selectivity of the MHS A-site for Ca²⁺ over Mg²⁺ (Table 2). Thus, both an increased Ca²⁺ affinity of the MHS RyR1 A-site and a greater selectivity of the MHS A-site for Ca²⁺ over Mg²⁺ contribute to the enhanced caffeine sensitivity of MHS skeletal muscle.

The extent of RyR1 channel activation reflects the product of the fraction of channels with Ca²⁺ bound to the A-site (f_A) and the fraction of channels with the I-site free of both Ca²⁺ and Mg²⁺ (1  f_I). Mg²⁺ shifts the Ca²⁺ dependence of f_A to higher Ca²⁺ concentrations, and decreases 1  f_I at all Ca²⁺ concentrations (Murayama et al., 2000). Consequently, Mg²⁺ decreases the maximal Ca²⁺ activation of the RyR1 and shifts the Ca²⁺ dependence of activation to higher Ca²⁺ concentrations (see Fig. 8. of Murayama et al., 2000). The Arg⁶¹⁵Cys RyR1 mutation opposes the effects of Mg²⁺ on RyR1. Thus, in a manner similar to caffeine, in the presence of Mg²⁺, the MH mutation shifted the Ca²⁺ dependence of f_A to lower Ca²⁺ concentrations (Fig. 1). However, in contrast to caffeine, at low Ca²⁺ concentrations, the MH mutation also increased 1  f_I (Fig. 1). As a result, maximal Ca²⁺ activation of the MHS RyR1 is increased and the Ca²⁺ dependence of activation is shifted to lower Ca²⁺ concentrations compared with the normal RyR1.

The concentration of free Mg²⁺ present in the myoplasm is sufficient to inhibit activation of the RyR1 by Ca²⁺ in intact muscle (Endo, 1977). Indeed, based on the parameters given in Table 1, and using Eqs. 1 and 2, at a resting Ca²⁺ concentration of 0.1 µM and 1 mM Mg²⁺ (Konishi, 1998) the I-sites of normal and MHS RyR1 are predicted to be partially occupied (~86% and 72%, respectively) by Mg²⁺. Mg²⁺ is predicted to occupy >99% of the A-sites of both normal and MHS RyR1. Thus, at physiological levels of Mg²⁺ both MHS and normal RyR1 channels can be partially activated (~15% and 30%, respectively) by increasing the Ca²⁺ concentration. If the channels are to be maximally activated, the Mg²⁺ inhibition at the I-site must be removed. At the same time, Ca²⁺ must replace Mg²⁺ occupying the A-site. However, the resting myoplasmic Ca²⁺ concentration is not sufficient to activate RyR1 channel opening. Therefore, if maximal SR Ca²⁺ release is to occur, the Ca²⁺ sensitivity of the RyR1 A-site must be increased concurrent with the removal of the Mg²⁺-dependent inhibition of the RyR1. The increased Ca²⁺ affinity of the A-site could depend in part on other endogenous effectors of the RyR1, such as ATP (Meissner, 1994) and calmodulin (Fruen et al., 2000), or alternatively on the interaction of the RyR1 with the dihydropyridine receptor.

The single point mutation in the porcine MHS RyR1 results in an increased sensitivity of the muscle to voltage activation (Gallant et al., 1982; Dietze et al., 2000). Although the mechanism by which this occurs is unclear, Dietze et al. (2000) suggested the MHS mutation alters the equilibrium for a voltage-independent transition of the RyR1 from the closed to the open state. The association of the increased voltage sensitivity of MHS SR Ca²⁺ release with the enhanced sensitivity of the MHS RyR1 to activation by Ca²⁺ raises the possibility that endogenous effectors may modulate the voltage-independent transition. Thus, the 3.5-fold increase in the selectivity of the MHS RyR1 A-site (K_A,Mg/K_A,Ca, Table 2) for Ca²⁺ over Mg²⁺ might provide the basis for the enhanced sensitivity of MHS muscle to voltage activation.

The Arg⁶¹⁵Cys mutation likely alters the affinity of the A- and I-sites via a conformational change transmitted over a significant distance because the mutation is a substantial distance along the primary sequence from putative locations of the A- (near amino acid 3885 of RyR3) (Chen et al., 1998) and I-sites (between amino acids 3726 and 5037) of RyR1 (Du and MacLennan, 1999). The proposal that Arg⁶¹⁵ is not located in either of the RyR1 Ca²⁺/Mg²⁺-binding sites is supported by the observation that the human MH mutation, Gly²⁴³⁴Arg, has a similar effect on the Ca²⁺ sensitivity of activation and inhibition of the RyR1 (Richter et al., 1997). Although it is difficult to envision how residues 615 and 2434 could form part of both the high- and low-affinity Ca²⁺-binding sites, the mechanism by which these and other MH mutations alter RyR1 function will remain a matter of speculation until the relationship of the primary sequence to the tertiary structure of the RyR1 is resolved in detail.

In summary, we have determined the Ca²⁺ and Mg²⁺ affinities of the A- and I-sites of both the MHS and normal RyR1. Although the I-site displayed no preference for Ca²⁺ over Mg²⁺, the affinity of the MHS RyR1 I-site for these ions was reduced nearly twofold compared with the normal RyR1. The A-site of the normal RyR1, however, had ~75-fold higher affinity for Ca²⁺ compared with Mg²⁺, whereas the MHS A-site had more than a 250-fold greater preference for Ca²⁺ over Mg²⁺. This significant increase in Ca²⁺ selectivity over Mg²⁺ may contribute not only to the increased voltage sensitivity of MHS skeletal muscle but also to the increased sensitivity of MHS muscle to caffeine and other pharmacological activators.