Dantrolene-Induced Inhibition of Skeletal L-Type Ca2+ Current Requires RyR1 Expression

Malignant hyperthermia (MH) is a pharmacogenetic disorder most often linked to mutations in the type 1 ryanodine receptor (RyR1) or the skeletal L-type Ca2+ channel (CaV1.1). The only effective treatment for an MH crisis is administration of the hydantoin derivative Dantrolene. In addition to reducing voltage induced Ca2+ release from the sarcoplasmic reticulum, Dantrolene was recently found to inhibit L-type currents in developing myotubes by shifting the voltage-dependence of CaV1.1 channel activation to more depolarizing potentials. Thus, the purpose of this study was to obtain information regarding the mechanism of Dantrolene-induced inhibition of CaV1.1. A mechanism involving a general depression of plasma membrane excitability was excluded because the biophysical properties of skeletal muscle Na+ current in normal mouse myotubes were largely unaffected by exposure to Dantrolene. However, a role for RyR1 was evident as Dantrolene failed to alter the amplitude, voltage dependence and inactivation kinetics of L-type currents recorded from dyspedic (RyR1 null) myotubes. Taken together, these results suggest that the mechanism of Dantrolene-induced inhibition of the skeletal muscle L-type Ca2+ current is related to altered communication between CaV1.1 and RyR1.


Introduction
In skeletal muscle, depolarization of the transverse tubule network causes conformational rearrangements within the sarcolemmal L-type Ca 2+ channel (Ca V 1.1) that produce a signal which is transmitted to the type 1 ryanodine receptor (RyR1) in the sarcoplasmic reticulum (SR) membrane via a transient protein-protein interaction [1]. is "orthograde" signal gates RyR1, enabling the Ca 2+ efflux from the SR into the myoplasm which ultimately initiates contraction. In addition, RyR1 produces a "retrograde" signal that enhances Ca V 1.1 P o [2,3] and accelerates Ca V 1.1 activation kinetics [3][4][5]. Like orthograde coupling, retrograde coupling is thought to be propagated via protein-protein contracts between RyR1 and Ca V 1.1 [5][6][7].
Malignant hyperthermia (MH) is a fulminant pharmacogenetic disorder in which the vast majority of identi�ed causative mutations are found in the genes encoding RyR1 [8,9] or Ca V 1.1 [10][11][12][13]. MH crises are triggered by heat, depolarizing muscle relaxants, or halogenated anaesthetics [14]. Following exposure to one of these triggers, MHsusceptible individuals enter a potentially lethal hypermetabolic crisis. e only effective treatment for an MH crisis is administration of the hydantoin derivative Dantrolene, which has substantially reduced MH-related mortality since its clinical introduction in the late 1970s [15]. Despite the therapeutic success of Dantrolene, the mechanism(s) by which it ameliorates MH crises is (are) not clear. ere is general agreement that one effect of Dantrolene is to stem aberrant Ca 2+ efflux from the SR into the myoplasm that occurs during MH crises [15]. Dantrolene and its more soluble analogue azumolene have also been shown to reduce store-operated [16,17] and voltage-triggered Ca 2+ entry [18,19] into muscle from the extracellular space. e major route of voltage-triggered Ca 2+ entry into myotubes is the L-type Ca 2+ current conducted by Ca V 1.1 [19,20]. In myotubes, Dantrolene reduces such Ca 2+ entry by shiing the voltage dependence of Ca V 1.1 activation to more depolarizing potentials [19]. Despite the aforementioned effects of Dantrolene on L-type current in mammalian muscle, the precise mechanism by which Dantrolene alters Ca V 1.1 channel activity has not been investigated. In this study, I have sought to determine whether the previously described depolarizing shi was a consequence of a Dantrolene-induced depression in membrane excitability or a modi�cation of bidirectional communication between RyR1 and Ca V 1.1. In order to investigate the former possibility, the skeletal muscle Na + current was employed as an assay to gauge membrane excitability. A general depression of membrane excitability appeared an unlikely explanation as Dantrolene had little effect on the biophysical properties of the Na + current. To probe the latter mechanism, L-type Ca 2+ currents were recorded from dyspedic (RyR1 null) myotubes. In these experiments, no Dantrolene-induced effects on the L-type current were observed, indicating a requirement for RyR1 expression in Dantrolene-induced inhibition of Ca V 1.1.

Patch-Clamp
Recording of Skeletal Muscle Na + and L-Type Ca 2+ Currents. Pipettes were fabricated from borosilicate glass and had resistances of ∼2.0 MΩ when �lled with a standard internal solution containing (mM): 140 Cs-aspartate, 10 Cs 2 -EGTA, 5 MgCl 2 , and 10 HEPES, pH 7.4 with CsOH. In order to record skeletal Na + currents, the bath solution contained (mM): 140 tetraethylammonium (TEA)-Cl, 5 NaCl, 10 CaCl 2 , and 10 HEPES, pH 7.4 with TEA-OH. When recording L-type Ca 2+ currents, the bath solution contained (mM): 145 TEA-Cl, 10 CaCl 2 , and 10 HEPES, 0.002 tetrodotoxin; pH 7.4 with TEA-OH. −P/4 and P/4 subtraction were employed to correct for linear current components while recording Na + and L-type Ca 2+ currents, respectively. Electronic compensation was used to reduce the effective series resistance and the time constant for charging the linear cell capacitance. Na + currents were �ltered at 10 kHz and digitized at 50 kHz. L-type Ca 2+ currents were �ltered at 2 kHz and digitized at 10 kHz. Cell capacitance was determined by integration of the current transient evoked from −80 mV to −70 mV using Clampex 8.0 (Molecular Devices, Foster City, CA). All current-voltage ( -) curves were �tted using the following equation: where is the current for test potential , rev is the reversal potential, max is the maximum inward (either Na + or Ca 2+ ) conductance, is the half-maximal activation potential, and is the slope factor. Conductance-voltage ( -) relationships for either Na + or L-type Ca 2+ currents were derived from the -data using: where the rev value for Na + or Ca 2+ in each cell was taken from (1). e average conductance values were subsequently �t with the following equation: where is the conductance for test potential , max is the maximal cation conductance, 1/2 act is the half-maximal activation potential, and is the slope factor. Steady-state inactivation curves for Na + currents were �t by the following equation: where is the current amplitude for test potential to 0 mV following a 300 ms prepulse to potential , max is current amplitude evoked by a test depolarization to 0 mV following a 300 ms prepulse to −110 mV, 1/2 act is the half-maximal inactivation potential, and is the slope factor.

Pharmacology.
Dantrolene (Sigma no. D9175) was dissolved in dry DMSO to make a 20 mM stock solution, diluted to 10 M, and sonicated just prior to use. Myotubes were exposed to Dantrolene in the bath solution (∼25 ∘ ) for 10 to 30 minutes. Dantrolene was stored and used in the dark.

Analysis.
Figures were made using the soware program SigmaPlot (versions 7.0 and 11.0, SSPS Inc., Chicago, IL). All data are presented as mean ± SEM. Statistical comparisons were by unpaired, two-tailed t-test, with considered signi�cant.

Dantrolene Does Not Affect the Fast Skeletal Muscle Na +
Current. e whole-cell patch clamp technique [22] was employed to test directly whether Dantrolene affects the fast skeletal muscle Na + current in cultured myotubes (which is conducted by a combination of Na + channel isoforms) [23]. With 5 mM external Na + as the charge carrier, myotubes ) and Dantrolene-treated ( ; ) myotubes. Currents were evoked at 0.1 Hz by test potentials ranging from −70 mV through +40 mV in 10 mV increments. Peak current amplitudes were normalized by linear cell capacitance (pA/pF). Smooth -curves were �t by (1) (see "Section 2") with the following respective parameters for control and Dantrolene-treated groups: max 3 5 ± 43 and 355 ± 44 nS/nF; / −9.8 ± .7 and −7.4 ± .3 mV; .9 ± .3 and 7. ± . mV. (d) comparison of conductance-voltage relationships for control and Dantrolene-treated myotubes. e average normalized conductance values (derived from -data using (2); see Section 2) were �t by (3) with the following respective parameters for control and Dantrolene-treated groups: / −9.9 ± .8 and −5.9 ± 3. mV; .9 ± .3 and 7.9 ± . mV. roughout, data are given as mean ± SEM, with the numbers in parentheses indicating the number of myotubes tested. For all the data given, the calculated average voltage error was <5 mV.
produced robust, rapidly-activating and -inactivating inward currents (Figure 1(a)). Myotubes exposed to Dantrolene (10 M) for greater than 10 minutes also produced Na + current with similar amplitude and kinetics (Figure 1(b)). As shown in Figure 1(c), the -relationships obtained in the absence and presence of Dantrolene displayed no signi�cant differences in average peak current density (−9.8 ± .3 pA/pF, versus − .4 ± . pA/pF, , resp.; . 5). �ikewise, �tting of the average conductance values (derived from the -data using the reversal potential for each individual cell; see Section 2) revealed little difference in the voltage-dependence of activation ( / act −9.9 ± .8 mV for untreated versus −5. ± 3. mV for Dantrolene-treated, resp.; . 5 t-test; Figure 1(d)). Next, the voltage protocol illustrated at the inset of Figure 2(a) was employed to determine whether Dantrolene in�uences steady-state inactivation of the fast Na + current. Speci�cally, 300 ms prepulses (ranging from −110 mV to −10 mV) were applied immediately prior to a 20 ms test depolarization to 0 mV. Steady-state inactivation was normalized to the measured current amplitude evoked by the test depolarization to 0 mV following a prepulse to −110 mV. As shown in Figures 2(a)-2(c), steady-state inactivation of the fast Na + current in the presence of Dantrolene was nearly indistinguishable from that observed in untreated myotubes ( /2 inact −4 .2 ± . ; versus −44.6 ± .6; , resp.; . 5). Superimposition of the steady-state inactivation curves with conductance-voltage relationships revealed minimal window current with no obvious differences in between control and Dantrolene-treated groups (Figure 2(c)).
In the representative dantrolene-treated cells shown in Figures 1 and 2, there appears to be a reduction in tail current amplitude. In both cases, tail current is re�ective of the activation phase of L-type Ca 2+ current and a lesser contribution of residual Na + current. Analysis of these tail currents indicated that there is a slight, but not quite significant, difference in the peak amplitude of the tail currents at more depolarized test potentials (− 2. ± −2. pA/pF; versus 4.3 ± .6 pA/pF; 2 , resp.; . 5; Figure 3). When a contribution from residual Na + current is considered, this difference may be more substantial.
e ability of Dantrolene to alter the recovery of the fast Na + current from inactivation was tested using the voltage protocol illustrated at the top of Figure 4, in which 20 ms reference pulses evoked by depolarizations from −80 mV to 0 mV were followed by test pulses from −80 mV to 0 mV at intervals increasing in 5 ms increments. Recovery from inactivation was assessed as the fraction of the fast Na + current evoked by the test pulse relative to that evoked by the
For this reason, the conductance-voltage relationships for control and Dantrolene-treated dyspedic myotubes conductance relationships were derived using (2) and were found to be similar ( Figure 5(d)).
In an earlier work, Szentesi and colleagues [25] reported that Dantrolene slows inactivation of L-type current in adult rodent �bres. To investigate whether such an effect of Dantrolene on Ca V 1.1 inactivation requires RyR1, inactivation of the L-type current was quanti�ed in control and Dantrolenetreated dyspedic myotubes as the value (fraction of the peak current remaining at the end of the 200 ms test depolarization to + mV). As summarized in Figure 5(e), the values for control ( .9 ± . , 4) and Dantrolene-treated ( . 9 ± . 4, ) dyspedic myotubes were not signi�cantly different ( . , unpaired t-test). Taken together, the inability of Dantrolene to modulate the skeletal L-type current in the absence of RyR1 supports the idea that these previously observed effects of Dantrolene on the L-type current [19,25] in normal muscle resulted from altered conformational coupling between RyR1 and Ca V 1.1.

Discussion
It is widely accepted that Dantrolene inhibits excitationcontraction (EC) coupling without greatly affecting the ability of skeletal muscle �bers to conduct action potentials [26]. However, recent evidence demonstrating that Dantrolene inhibits skeletal muscle L-type Ca + current and/or charge movement attributed to Ca V 1.1 [19,25,27] has raised the possibility that Dantrolene may produce inhibition of Ca V 1.1 by altering the membrane environment. In order to probe depression of plasma membrane excitability as the mechanism of L-type channel inhibition, the effects of Dantrolene on various facets of the skeletal muscle Na + current were investigated. us, exposure to Dantrolene produced no signi�cant effect on the average peak current density, voltage dependence of activation, voltage dependence of inactivation, or recovery from inactivation of the Na + current in developing myotubes (Figures 1-3). e inability of Dantrolene to modulate these parameters of the Na + current in this preparation is consistent with the idea that the Dantrolene-induced inhibition of Ca V 1.1 observed in previous studies [19,25,27] most likely was a consequence of Dantrolene interacting with the EC coupling apparatus at plasma membrane-SR junctions, rather than a general nonspeci�c depression of plasma membrane excitability.
e observation that Dantrolene did not greatly alter the amplitude, voltage dependence, or inactivation kinetics of L-type Ca + currents in dyspedic myotubes ( Figure 5) lends support to the idea that the inhibition of L-type current by dantrolene is unlikely a consequence of a direct interaction of Dantrolene with Ca V 1.1 channels. One caveat to this interpretation is that dantrolene may selectively interact with high P states of Ca V 1.1 that only occur with the in�uence of RyR1 [2,3]. In either case, the data presented in Figure 5 indicate that expression of RyR1 is necessary for Dantrolene's inhibitory effects on Ca V 1.1 and that Dantrolene alters conformational coupling between the two channels. Figure 6(a) shows a simpli�ed model of �normal� bidirectional coupling between Ca V 1.1 and RyR1 aer [2] where the green arrow and the red arrow represent orthograde (i.e., skeletal-type EC coupling) and retrograde (i.e., modulation of L-type current by RyR1) signaling, respectively. e precise mechanism of Dantrolene-induced inhibition of orthograde coupling is area of controversy. A body of evidence points to a mechanism in which Dantrolene interferes with orthograde signaling by inhibiting Ca 2+ release from the SR. is view is based largely on the observations that radiolabelled Dantrolene binds to RyR1 directly [28,29] and inhibits RyR1mediated Ca 2+ efflux from SR vesicles [30,31]. However, other investigators have found little effect of Dantrolene on the single channel properties of RyR1 in reconstituted lipid bilayers [18,25]. . ± .6 and . ± 2. mV. �o signi�cant ( > . 5, unpaired t-test) differences were observed amongst the �t parameters. (d) Comparison of conductance-voltage relationships for control and Dantrolene-treated dyspedic myotubes. e average conductance values (derived from -data using (2); see Section 2). (e) Inactivation summary. 2 = fraction of the peak current remaining at the end of the 200 ms test depolarization to +30 mV. (a) A simpli�ed model illustrating bidirectional coupling between RyR1 and Ca V 1.1, the skeletal muscle L-type Ca 2+ channel [2]. e red arrow represents orthograde signal transmitted from Ca V 1.1 to RyR1 that engages SR Ca 2+ release (i.e., EC coupling). e green arrow represents the retrograde communication transmitted from RyR1 to Ca V 1.1 that enhances channel P 0 and modi�es L-type current activation kinetics. Panels B and C show two possible mechanisms by which Dantrolene alters Ca V 1.1 function. (b) Binding of Dantrolene to RyR1 causes conformational changes in RyR1 that reduce SR Ca 2+ release (represented by small red arrow) and shis the activation of L-type currents to more depolarizing test potentials by altering retrograde contacts between RyR1 and Ca V 1.1 (represented by horizontal green arrow). (c) Alternatively, dantrolene-induced inhibition of SR Ca 2+ release (again represented by small red arrow) and alterations in L-type current activation (again represented by horizontal green arrow) may be a consequence of disruption of interactions between RyR1 and Ca V 1.1 that are critical for bidirectional communication between the two channels.
Although my present observations do not resolve the issue of how Dantrolene inhibits orthograde coupling, they do provide support for the idea that Dantrolene's effect on retrograde signaling stems from a junctional interaction of Dantrolene with the intact (i.e., RyR1-containing) EC coupling macromolecular complex. ese �ndings are incorporated into the two hypothetical models shown in Figures 6(b) and 6(c). In the �rst model ( Figure 6(b)), Dantrolene would alter Ca V 1.1 gating indirectly by inducing allosteric rearrangements in RyR1. is model is based on the notion that agents that alter the activity of orthograde coupling by altering the functional state of RyR1 may affect retrograde signaling and vice-versa. For example, application of a high concentration of ryanodine (≥200 M) not only attenuates EC coupling by locking RyR1 in a nonconducting state [32,33], but also causes hyperpolarizing shis in skeletal muscle L-type current activation [34,35] and in charge movement [35]. Similarly, Dantrolene-induced allosteric changes in the myoplasmic region of RyR1 may affect Ca V 1.1 in such a way that causes a depolarizing shi in L-type current activation.
Another potential mechanism for the inhibition of Ca V 1.1 gating is that Dantrolene could disrupt the proteinprotein interactions between RyR1 and the Ca V 1.1 heteromultimer that support bidirectional signaling (illustrated in Figure 6(c)). Within the context of this model, Dantrolene not only would impair EC coupling by blocking the transient interaction between the Ca V 1.1 channel and RyR1 but also would cause inhibition of the L-type current by removing the in�uence of RyR1 on Ca V 1.1 gating.
In summary, the anti-MH drug Dantrolene has little effect on the biophysical properties of the Na + current or Ltype Ca 2+ current in developing skeletal muscle harvested from normal mice or mice lacking RyR1, respectively. ese results indicate (1) that a general depression of plasma membrane excitability seems not to be responsible for inhibition of skeletal L-type Ca 2+ current by Dantrolene, and (2) that RyR1 expression is necessary for the effect(s) of Dantrolene on the L-type current. us, this study reveals useful information towards the mechanism of Dantrolene's effect on Ca 2+ currents mediated by Ca V 1.1.