Pathogenic CaV1.1 Variants Cause Hyperpolarizing Shift of Depolarization-Induced Ca2+ Release in Malignant Hyperthermia Susceptibility Authors: Takashi Murayama, Takuro Numaga-Tomita, Hirotsugu Miyoshi, Tsuyoshi Ikeda, Keiko Mukaida, Takuya Kobayashi, Nagomi Kurebayashi, Tsutomu Nakada, Taichiro Tomida, Satomi Adachi-Akahane, Yasuo M. Tsutsumi, Mitsuhiko Yamada, Takashi Sakurai Published: 2026 ABSTRACT Malignant hyperthermia (MH) is a potentially fatal genetic disorder triggered by certain anesthetics, leading to excessive calcium release in skeletal muscle. While most MH-associated mutations affect the type 1 ryanodine receptor, a Ca2+ release channel in the sarcoplasmic reticulum, variants in CaV1.1—the L-type calcium channel critical for excitation–contraction coupling—have also been implicated. Here, we investigate four novel CaV1.1 variants identified in MH-susceptible individuals (A560T, S879P, F1161L, and D1382V) using a reconstituted Ca2+ release platform in HEK293 cells. Two variants, S879P and F1161L, increase sensitivity to extracellular potassium of depolarization-induced Ca2+ release (DICR), enhance caffeine responsiveness, and shift voltage sensing toward more hyperpolarized potential. Interestingly, enhanced caffeine sensitivity is abolished by loss of DICR at rest. These findings suggest that hyperpolarizing shift of DICR underlies Ca2+ dysregulation in S879P and F1161L CaV1.1-linked MH. Our findings uncover a mechanistic link between CaV1.1 variants and MH susceptibility, highlighting the pathological role of DICR. Introduction Malignant hyperthermia (MH) is a life-threatening pharmacogenetic disorder characterized by skeletal muscle rigidity, elevated body temperature, and hypermetabolism, occurring in response to inhalational anesthetics or depolarizing muscle relaxants1. MH has been associated with elevated skeletal muscle resting intracellular Ca2+ ([Ca2+]i) and is directly linked to defects in RYR1, CACNA1S, and other genes that encode essential proteins for Ca2+ release from the sarcoplasmic reticulum (SR) in skeletal muscle2. RYR1 encodes the type 1 ryanodine receptor (RyR1), a Ca2+-release channel in the SR membrane, whereas CACNA1S encodes the α1 subunit (CaV1.1) of the voltage-dependent L-type calcium channel in the transverse tubule (T-tubule). RyR1 and CaV1.1 are physically coupled to each other at the triad junction where the terminal cisterna of the SR is closely apposed to the T-tubule3,4. In skeletal muscle excitation–contraction (E–C) coupling, depolarization of the T-tubule membrane activates CaV1.1 to open the RyR1 channel5,6. This process is known as depolarization-induced Ca2+ release (DICR). DICR requires additional components, such as β1a, Stac3, and junctophilins, which are necessary for the formation of specific machinery in the triad junction7,8. The RyR1 channel can also be directly activated by Ca2+ binding, a phenomenon referred to as Ca2+-induced Ca2+ release (CICR)9. CICR is thought to be implicated in the pathogenesis of MH9–11. MH-associated variants have been predominantly identified in the RYR1 gene. MH variants are thought to hyperactivate the RyR1 channel, which triggers rapid and uncontrollable Ca2+ release from the SR via CICR in response to inhalational anesthetics under resting conditions12,13. In contrast, MH-associated variants in the CACNA1S gene account for only ~1% of the MH-susceptible patients14,15. Functional characterization of CaV1.1 variants, including R174W16,17, R1086H18, and T1354S19, has been conducted using CaV1.1-null myotubes. These variants commonly show increased sensitivity to caffeine-induced Ca2+ release (i.e., CICR) compared with the wild type (WT). However, the underlying mechanism of how CaV1.1 variants increase CICR activity remains unclear. We have recently identified several novel CaV1.1 variants from MH-susceptible patients20,21. In this study, we functionally analyzed four novel CaV1.1 variants (A560T, S879P, F1161L, and D1382V) using a reconstituted DICR platform in HEK293 cells22,23. We demonstrate that two variants (S879P and F1161L) exhibit hyperpolarizing shift of DICR, concomitant with the corresponding shift in the charge movement. Interestingly, the cells expressing these variants show enhanced sensitivity to caffeine-induced Ca2+ release in a DICR-dependent manner. Our findings provide novel mechanistic insights into MH pathogenesis associated with CaV1.1 variants, highlighting the role of DICR in Ca2+ dysregulation. Results Characterization of CACNA1S variants used in this study We selected four novel CACNA1S variants (A560T, S879P, F1161L, and D1382V) identified in patients who exhibited MH or MH-like events20,21 (Supplementary Table 1). All the variants were heterozygous except for F1161L. Two patients (#163 and #248) also carried RyR1 variants (D2431H and M2208K), both of which are currently classified as variants of unknown significance. Three variants (A560T, S879P, and D1382V) have been reported in the single nucleotide polymorphism database (dbSNP) of the National Center for Biotechnology Information (NCBI) (Supplementary Table 2). Pathogenicity prediction using five computational tools (REVEL, CADD, MutationTaster, SIFT, and PolyPhen-2) suggested that all variants are likely pathogenic. These variants are located within the transmembrane segments of the channel domain (Fig. 1a). The affected residues are conserved among L-type Ca2+ channels (CaV1.1, CaV1.2, CaV1.3, and CaV1.4), and some are also conserved across other classes of voltage-gated Ca2+ channels (Supplementary Fig. 1). Among them, one variant NM_001256789.3(CACNA1F):c.1936G>A (p.Ala646Thr) corresponding to A560T was found in CACNA1F gene which is associated with congenital stationary night blindness 2A24.Fig. 1DICR activity of reconstituted cells expressing CaV1.1 variants.a Schematic drawing of location of MH-associated variants (A560T, S879P, F1161L, and D1382V) in CaV1.1. Four voltage sensing domains (VSD) are indicated in Roman numerals (I–IV). b Schematic drawing of measurement of DICR in the reconstituted platform. HEK293 cells expressing RyR1 and R-CEPIA1er were infected with baculovirus carrying essential components (CaV1.1 WT or variants, β1a, Stac3, JP2) to reconstitute the DICR machinery. Kir2.1 was also transduced to hyperpolarize the membrane potential. Note that CaV1.1 carries a Ca2+ non-conducting mutation (N617D) in this experiment. DICR was triggered by depolarization of the membrane potential with high [K+]o solution and measured by a reduction in R-CEPIA1er fluorescence. c Typical results of time-lapse R-CEPIA1er fluorescence measurements at 5-s intervals for WT (left), A560T (center), and F1161L (right) CaV1.1 cells. High [K+]o solution ranging from 5 to 61 mM (symbols shown in WT) was applied 30 s after starting (arrowheads). d Resting [Ca2+]ER in cells expressing WT and CaV1.1 variants. Resting R-CEPIA1er fluorescence (F0) was normalized to Fmax, which was determined by high Ca2+ and ionomycin solution (see Supplementary Fig. 3). Data are means ± SD (n = 9). e [K+]o dependence of R-CEPIA1er fluorescence in WT and Cav1.1-variant cells, which was corrected by the resting [Ca2+]ER in WT cells (F/F0,WT). Data are means ± SD (n = 9). f EC50 values for [K+]o. Data are means ± SD (n = 9). *p < 0.05 compared with WT. Note that the images in a and b were created by PowerPoint software. DICR activity of reconstituted cells expressing CaV1.1 variants. a Schematic drawing of location of MH-associated variants (A560T, S879P, F1161L, and D1382V) in CaV1.1. Four voltage sensing domains (VSD) are indicated in Roman numerals (I–IV). b Schematic drawing of measurement of DICR in the reconstituted platform. HEK293 cells expressing RyR1 and R-CEPIA1er were infected with baculovirus carrying essential components (CaV1.1 WT or variants, β1a, Stac3, JP2) to reconstitute the DICR machinery. Kir2.1 was also transduced to hyperpolarize the membrane potential. Note that CaV1.1 carries a Ca2+ non-conducting mutation (N617D) in this experiment. DICR was triggered by depolarization of the membrane potential with high [K+]o solution and measured by a reduction in R-CEPIA1er fluorescence. c Typical results of time-lapse R-CEPIA1er fluorescence measurements at 5-s intervals for WT (left), A560T (center), and F1161L (right) CaV1.1 cells. High [K+]o solution ranging from 5 to 61 mM (symbols shown in WT) was applied 30 s after starting (arrowheads). d Resting [Ca2+]ER in cells expressing WT and CaV1.1 variants. Resting R-CEPIA1er fluorescence (F0) was normalized to Fmax, which was determined by high Ca2+ and ionomycin solution (see Supplementary Fig. 3). Data are means ± SD (n = 9). e [K+]o dependence of R-CEPIA1er fluorescence in WT and Cav1.1-variant cells, which was corrected by the resting [Ca2+]ER in WT cells (F/F0,WT). Data are means ± SD (n = 9). f EC50 values for [K+]o. Data are means ± SD (n = 9). *p < 0.05 compared with WT. Note that the images in a and b were created by PowerPoint software. DICR activity of cells expressing CaV1.1 variants Since CaV1.1 functions as a voltage sensor in skeletal muscle E–C coupling, we initially assessed the DICR activity of the CaV1.1 variants using our recently developed reconstituted DICR platform22. HEK293 cells stably expressing WT RyR1 and R-CEPIA1er, a genetically-encoded endoplasmic reticulum (ER) Ca2+ indicator25, were seeded in 96-well plate and infected with a mixture of baculovirus (CaV1.1 WT or variants, β1a, Stac3, junctophilin-2, and Kir2.1) to reconstitute the DICR machinery (see “Methods”). A Ca2+-impermeable mutation (N617D)26 was introduced in CaV1.1 to avoid contamination of Ca2+ influx. Surface biotinylation assay revealed that total and plasma membrane expression levels of CaV1.1 in four variants were comparable to those in WT (Supplementary Fig. 2). The reconstituted cells were incubated with Krebs solution containing 5 mM [K+]o and DICR was triggered by depolarization of the membrane potential with high (8–61 mM) [K+]o solutions (Fig. 1b). The reduction in R-CEPIA1er fluorescence in each well was captured at 5-s intervals using FlexStation3 fluorometer, which corresponds to the average from the population available in the well (2–3 thousand cells). The cells expressing WT CaV1.1 exhibited a [K+]o-dependent monophasic decrease in fluorescence intensity (Fig. 1c). Similar responses were observed with A560T. F1161L, in contrast, showed lower initial fluorescence intensity than WT, suggesting a reduced resting [Ca2+]ER. We therefore quantified resting [Ca2+]ER by normalizing baseline fluorescence to the maximum fluorescence intensity of R-CEPIA1er in the presence of ionomycin/Ca2+22 (Supplementary Fig. 3). Resting [Ca2+]ER was significantly lower in F1161L compared with WT (Fig. 1d). Reduction in resting [Ca2+]ER was also observed with S879P (Fig. 1d). A sigmoidal [K+]o dependence of DICR activity was obtained by the ratio of the average fluorescence intensity between the first (F0) and the last (F) 25 s, corrected by resting [Ca2+]ER in WT cells (Fig. 1e). In contrast to the isolated muscle cells27, voltage-dependent inactivation was not observed at higher [K+]o in our platform22. This may be due to absence of γ1 subunit which contributes to voltage-dependent inactivation28,29. The EC50 for [K+]o was significantly lower in S879P (8.0 ± 0.5 mM) and F1161L (8.5 ± 0.3 mM), but not A560T (11.5 ± 0.7 mM) or D1382V (12.4 ± 0.3 mM), compared with WT (12.0 ± 0.3 mM) (Fig. 1f). These results suggest that the voltage sensitivity of DICR activity is substantially enhanced in S879P and F1161L. Caffeine-induced Ca2+ release of cells expressing CaV1.1 variants Increased sensitivity to RyR1-activating agents, such as caffeine or halothane, is a hallmark used in the clinical diagnosis of MH2. We assessed the caffeine sensitivity of the reconstituted HEK293 cells expressing CaV1.1 variants (Fig. 2a). Application of caffeine (0–20 mM) resulted in an initial rapid drop in [Ca2+]ER, followed by gradual recovery toward the respective steady-state levels (Fig. 2b). The initial phase is thought to involve a regenerative Ca2+ release through a CICR mechanism by the released Ca2+, and the steady-state phase is determined by a balance between Ca2+ release via RyR1 and Ca2+ uptake by the SERCA Ca2+ pump. Since our slow (5-s intervals) measurement is difficult to accurately detect the initial phase, we instead used the steady state (last 25 s, F) to determine caffeine dependence of the Ca2+ release by normalizing to the initial phase without caffeine (initial 25 s, F0) (Fig. 2c). Interestingly, the EC50 for caffeine was slightly but significantly lower in S879P (1.8 ± 0.2 mM) and F1161L (2.0 ± 0.2 mM) compared with WT (2.4 ± 0.1 mM) (Fig. 2d). No significant differences were observed in A560T (2.5 ± 0.1 mM) and D1382V (2.7 ± 0.4 mM). Thus, S879P and F1161L exhibited MH-like phenotypes in our reconstituted system. Dysgenic myotubes lacking CaV1.1 have been shown to cause increased RyR1-mediated Ca2+ leak to reduce resting SR Ca2+ 30. To determine whether the observed decrease in [Ca2+]ER in the S879P and F1161L variants mirrors the reduction seen in the absence of CaV1.1, we measured caffeine-induced Ca2+ release of the reconstituted cells lacking CaV1.1. Unexpectedly, the resting [Ca2+]ER was not reduced and caffeine dependence was comparable to that of fully reconstituted cells (Supplementary Fig. 4). Thus, our reconstituted cells did not reproduce orthograde CaV1.1 signals regulating RyR1-mediated Ca2+ leak 30. Phenotypes of S879P and F1161L cells may be caused by distinct mechanisms from that in the dysgenic myotubes. Since A560T and D1382V showed no differences in DICR (Fig. 1) or caffeine sensitivity (Fig. 2), we mainly focused on S879P and F1161L for further experiments.Fig. 2Caffeine-induced Ca2+ release in reconstituted cells expressing CaV1.1 variants.a Schematic drawing of caffeine-induced Ca2+ release in the reconstituted platform. b Typical results of time-lapse R-CEPIA1er fluorescence measurement at 5-s intervals for WT (left), A560T (center), and F1161L (right) CaV1.1 cells. Caffeine solution ranging from 0 to 20 mM (symbols shown in WT) was applied 30 s after starting (arrows). c Caffeine dependence of R-CEPIA1er fluorescence in WT and Cav1.1-variant cells, corrected by resting [Ca2+]ER in WT cells (F/F0,WT). Data are means ± SD (n = 9). d EC50 values for caffeine. Data are means ± SD (n = 9). *p < 0.05 compared with WT. Note that the image in a was created by PowerPoint software. Caffeine-induced Ca2+ release in reconstituted cells expressing CaV1.1 variants. a Schematic drawing of caffeine-induced Ca2+ release in the reconstituted platform. b Typical results of time-lapse R-CEPIA1er fluorescence measurement at 5-s intervals for WT (left), A560T (center), and F1161L (right) CaV1.1 cells. Caffeine solution ranging from 0 to 20 mM (symbols shown in WT) was applied 30 s after starting (arrows). c Caffeine dependence of R-CEPIA1er fluorescence in WT and Cav1.1-variant cells, corrected by resting [Ca2+]ER in WT cells (F/F0,WT). Data are means ± SD (n = 9). d EC50 values for caffeine. Data are means ± SD (n = 9). *p < 0.05 compared with WT. Note that the image in a was created by PowerPoint software. Activation of DICR at rest contributes to increased caffeine sensitivity S879P and F1161L cells exhibited lower resting [Ca2+]ER (Fig. 1d). We hypothesized that activation of DICR in CaV1.1-variant cells causes Ca2+ release to reduce the [Ca2+]ER under a resting membrane potential. To test this hypothesis, we hyperpolarized the membrane potential, which is expected to reduce the activation of CaV1.1. Reduction of [K+]o in Krebs solution from 5 mM to 1 mM successfully hyperpolarized the membrane potential by 36 mV (from −70 ± 3 mV to −106 ± 4 mV) (Fig. 3a). Under those conditions, resting [Ca2+]ER in the CaV1.1-variant cells was comparable with that in WT cells (Fig. 3b). Again, the [K+]o dependence was clearly shifted leftward in S879P and F1161L (Fig. 3c), and the EC50 for [K+]o was significantly lower in S879P (4.5 ± 0.4 mM) and F1161L (5.2 ± 0.5 mM) compared with WT (8.1 ± 1.2 mM) (Fig. 3d).Fig. 3Effect of DICR activity on caffeine sensitivity.a Resting membrane potential of the reconstituted HEK293 cells in 5 mM and 1 mM [K+]o Krebs solution. Data are means ± SD (n = 8). b Resting [Ca2+]ER at 1 mM [K+]o solution in WT (black), S879P (red), and F1161L (blue) CaV1.1 cells. Data are means ± SD (n = 9). c [K+]o dependence of R-CEPIA1er fluorescence at 1 mM [K+]o solution in WT and Cav1.1-variant cells, corrected by resting [Ca2+]ER in WT cells (F/F0,WT). Data are means ± SD (n = 9). d EC50 values for [K+]o at 1 mM [K+]o solution. Data are means ± SD (n = 9). e Caffeine dependence of R-CEPIA1er fluorescence at 1 mM [K+]o solution in WT and CaV1.1-variant cells, corrected by resting [Ca2+]ER in WT cells (F/F0,WT). Data are means ± SD (n = 9). f EC50 values for caffeine at 1 mM [K+]o solution. Data are means ± SD (n = 9). g, h EC50 values for caffeine at 5 mM [K+]o (normal) Krebs solution in the reconstituted HEK293 cells without β1a (g) or Stac3 (h). *p < 0.05 compared with WT. Effect of DICR activity on caffeine sensitivity. a Resting membrane potential of the reconstituted HEK293 cells in 5 mM and 1 mM [K+]o Krebs solution. Data are means ± SD (n = 8). b Resting [Ca2+]ER at 1 mM [K+]o solution in WT (black), S879P (red), and F1161L (blue) CaV1.1 cells. Data are means ± SD (n = 9). c [K+]o dependence of R-CEPIA1er fluorescence at 1 mM [K+]o solution in WT and Cav1.1-variant cells, corrected by resting [Ca2+]ER in WT cells (F/F0,WT). Data are means ± SD (n = 9). d EC50 values for [K+]o at 1 mM [K+]o solution. Data are means ± SD (n = 9). e Caffeine dependence of R-CEPIA1er fluorescence at 1 mM [K+]o solution in WT and CaV1.1-variant cells, corrected by resting [Ca2+]ER in WT cells (F/F0,WT). Data are means ± SD (n = 9). f EC50 values for caffeine at 1 mM [K+]o solution. Data are means ± SD (n = 9). g, h EC50 values for caffeine at 5 mM [K+]o (normal) Krebs solution in the reconstituted HEK293 cells without β1a (g) or Stac3 (h). *p < 0.05 compared with WT. We examined the caffeine sensitivity of WT and CaV1.1-variant cells under a 1 mM [K+]o condition. Caffeine dependence was almost the same between WT and the variants (Fig. 3e), and no significant differences were observed in the EC50 values (Fig. 3f). This suggests that activation of DICR may contribute to the increased caffeine sensitivity of the CaV1.1-variant cells. We tested this possibility by a different approach. A lack of essential components (β1a or Stac3) results in complete loss of DICR activity31,32, and this is the same in our reconstituted platform22. Removal of β1a (Fig. 3g) or Stac3 (Fig. 3h) totally abolished the increased sensitivity to caffeine even under normal (5 mM) [K+]o conditions. Taken together, these findings suggest that the increased caffeine sensitivity of S879P and F1161L is associated with activation of DICR under resting conditions. Properties of charge movement and the Ca2+ current in CaV1.1 variants We next investigated the mechanism underlying the altered voltage dependence of DICR in the CaV1.1 variants. Intramembranous charge movements reflect activation of the voltage sensor in CaV1.1, and correlate with E–C coupling33,34. We therefore measured gating charge movements using the patch–clamp technique (Fig. 4a). Representative gating currents elicited by 20-ms depolarizations from −100 mV to +80 mV are shown in Fig. 4b. No Ca2+ currents were detected because of the introduction of a Ca2+-impermeable mutation (N617D). The maximum gating charge (Qmax) was significantly lower with the S879P and F1161L variants than in WT (Fig. 4c). Notably, the voltage dependence of the charge movement for S879P and F1161L was substantially shifted toward hyperpolarized potentials (Fig. 4d). Fitting the data to the Boltzmann equation yielded lower V0.5 values for the CaV1.1 variants (−43 ± 6 mV and −44 ± 8 mV for S879P and F1161L, respectively) compared with WT (−15 ± 6 mV) (Fig. 4e), with no significant changes in the slope factor (Fig. 4f). A560T and D1382V cells, in contrast, showed no changes in voltage dependence of the charge movement (Supplementary Fig. 5). Qmax was not changed in A560T but significantly reduced in D1382V. Thus, hyperpolarizing shift of the charge movement in S879P and F1161L cells closely parallels the shift of DICR in the CaV1.1 mutants.Fig. 4Gating charge movement in CaV1.1 MH mutants.a Schematic illustration of gating current measurements using pore-dead N617D-mutant CaV1.1 channels. b Representative traces of test pulses and gating currents recorded from WT, S879P, and F1161L CaV1.1 channels. c Maximum gating charge (Qmax) obtained by fitting the Q–V relationship with a Boltzmann equation. d Voltage dependence of gating charge movement. Q was measured by integrating the ON-gating current over the first 2 ms of each test pulse, normalized to Qmax, and plotted as Q/Qmax against membrane potential. Curves represent Boltzmann fits. e, f Half-activation voltage (V0.5, e) and slope factor (k, f) derived from the Boltzmann fits shown in (d). Data are means ± SD (WT, n = 7; S879P, n = 9; F1161L, n = 7). Gating charge movement in CaV1.1 MH mutants. a Schematic illustration of gating current measurements using pore-dead N617D-mutant CaV1.1 channels. b Representative traces of test pulses and gating currents recorded from WT, S879P, and F1161L CaV1.1 channels. c Maximum gating charge (Qmax) obtained by fitting the Q–V relationship with a Boltzmann equation. d Voltage dependence of gating charge movement. Q was measured by integrating the ON-gating current over the first 2 ms of each test pulse, normalized to Qmax, and plotted as Q/Qmax against membrane potential. Curves represent Boltzmann fits. e, f Half-activation voltage (V0.5, e) and slope factor (k, f) derived from the Boltzmann fits shown in (d). Data are means ± SD (WT, n = 7; S879P, n = 9; F1161L, n = 7). We next measured Ca2+ currents of the CaV1.1 variants. For this purpose, we used CaV1.1 channels with no N617D mutation (Fig. 5a). Representative Ca2+ currents elicited by 20-ms depolarizations from −80 mV to +80 mV are shown in Fig. 5b. The peak current densities of S879P and F1161L were not significantly different from that of WT (Fig. 5c). In contrast to the charge movements, the voltage dependence of the currents was similar between WT and the CaV1.1 variants (Fig. 5d), with a V0.5 of ~20 mV (Fig. 5e). Interestingly, current activation was fourfold slower in the F1161L variant than in WT or S879P (Fig. 5f).Fig. 5Ionic currents and coupling efficiency in CaV1.1 MH mutants.a Schematic illustration of ionic current measurements using normal CaV1.1 channels. b Representative current traces evoked by test pulses (from −80 to 80 mV) in WT, S879P, and F1161L channels. Note that typical currents evoked by test pulses from −40 to +60 mV are shown. c Peak current density measured at +30 mV. d Current–voltage relationships fitted with a Boltzmann function. e Half-activation voltage (V0.5) derived from the fits in (d). f Activation time constants (τ) at +30 mV, obtained by single-exponential fitting. Data are means ± SD (WT, n = 5; S879P, n = 5; F1161L, n = 5). g Representative traces of test pulses and tail currents used to assess coupling efficiency. h Total gating charge movement determined by integrating ON-gating currents. i Peak tail current density upon repolarization to −50 mV. j Coupling efficiency calculated as the ratio of peak tail current to total gating charge (nA/pC). Data are means ± SD (WT, n = 10; S879P, n = 11; F1161L, n = 10). Ionic currents and coupling efficiency in CaV1.1 MH mutants. a Schematic illustration of ionic current measurements using normal CaV1.1 channels. b Representative current traces evoked by test pulses (from −80 to 80 mV) in WT, S879P, and F1161L channels. Note that typical currents evoked by test pulses from −40 to +60 mV are shown. c Peak current density measured at +30 mV. d Current–voltage relationships fitted with a Boltzmann function. e Half-activation voltage (V0.5) derived from the fits in (d). f Activation time constants (τ) at +30 mV, obtained by single-exponential fitting. Data are means ± SD (WT, n = 5; S879P, n = 5; F1161L, n = 5). g Representative traces of test pulses and tail currents used to assess coupling efficiency. h Total gating charge movement determined by integrating ON-gating currents. i Peak tail current density upon repolarization to −50 mV. j Coupling efficiency calculated as the ratio of peak tail current to total gating charge (nA/pC). Data are means ± SD (WT, n = 10; S879P, n = 11; F1161L, n = 10). The S879P and F1161L mutants exhibited reduced gating charge (Fig. 4c) despite showing peak current densities comparable with WT channels (Fig. 5c). Because charge movement was initially assessed using channels harboring the Ca2+-impermeable N617D mutation, we considered that this mutation might influence the results. To test this possibility, we simultaneously measured both parameters in channels with no N617D mutation. To evaluate the efficiency of coupling between channel gating and pore opening, we quantified the gating charge generated during depolarization to the reversal potential (Erev) of CaV1.1, and the peak tail current elicited upon repolarization to −50 mV (Fig. 5g). Even in the absence of the N617D mutation, the S879P and F1161L mutants displayed significantly reduced gating charge movements (Fig. 5h), while the average peak amplitudes of the tail currents remained comparable between all three channel variants (Fig. 5i). The coupling efficiency was calculated as the ratio of the peak tail current to the total gating charge (nA/pC). Both mutants exhibited significantly higher coupling efficiency than WT (Fig. 5j). These results support the conclusion that the two mutants exhibit reduced charge movement without changing the current density. Discussion In this study, we investigated the functional properties of four CaV1.1 variants identified in MH-susceptible patients: A560T, S879P, F1161L, and D1382V. We demonstrated that two variants (S879P and F1161L) exhibited a substantial hyperpolarizing shift of DICR (Fig. 1), and increased caffeine sensitivity (Fig. 2) in a reconstituted DICR platform using HEK293 cells. Notably, the increased caffeine sensitivity was abolished by hyperpolarizing the membrane potential or by removal of the essential components for DICR (Fig. 3). These results suggest that this hyperpolarizing shift of DICR under resting conditions may contribute to the increased caffeine sensitivity, and consequently, to MH pathogenesis. We propose a possible mechanism underlying MH pathogenesis by the CaV1.1 variants (Fig. 6). In WT controls, DICR activity is negligible under resting conditions. Inhalational anesthetics, such as isoflurane or halothane, bind to RyR1 to activate the channel35. However, they do not trigger regenerative Ca2+ release, since the CICR activity of WT RyR1 is low36,37 (Fig. 6, left). In MH patients carrying RyR1 variants, inhalational anesthetics can trigger regenerative Ca2+ release due to high CICR activity by the gain-of-function RyR1 variants (Fig. 6, center). In contrast, in MH patients carrying CaV1.1 variants, DICR is partially activated even under resting conditions. This may slightly increase Ca2+ levels in the junctional space ([Ca2+]JS), a highly specialized nano-environment38,39. Under these conditions, inhalational anesthetics can trigger regenerative Ca2+ release (Fig. 6, right).Fig. 6Proposed mechanism of pathogenesis of MH by CaV1.1 mutations.Left In WT controls, Ca2+ release from RyR1 is negligible under resting conditions (upper panel). Inhalational anesthetics, such as isoflurane, activate RyR1, but Ca2+ release is still low (lower panel). Center In patients carrying RyR1 MH variants, Ca2+ release activity is increased by gain-of-function of RyR1, but does not trigger regenerative Ca2+ release under resting conditions (upper panel). Inhalational anesthetics further activates RyR1 to trigger regenerative Ca2+ release (lower panel). Right In patients carrying CaV1.1 MH variants, DICR may occur under resting conditions, which increases Ca2+ in the junctional space (upper panel). Inhalational anesthetics activate RyR1, which can trigger regenerative Ca2+ release under such higher Ca2+ in the junctional space (lower panel). Proposed mechanism of pathogenesis of MH by CaV1.1 mutations. Left In WT controls, Ca2+ release from RyR1 is negligible under resting conditions (upper panel). Inhalational anesthetics, such as isoflurane, activate RyR1, but Ca2+ release is still low (lower panel). Center In patients carrying RyR1 MH variants, Ca2+ release activity is increased by gain-of-function of RyR1, but does not trigger regenerative Ca2+ release under resting conditions (upper panel). Inhalational anesthetics further activates RyR1 to trigger regenerative Ca2+ release (lower panel). Right In patients carrying CaV1.1 MH variants, DICR may occur under resting conditions, which increases Ca2+ in the junctional space (upper panel). Inhalational anesthetics activate RyR1, which can trigger regenerative Ca2+ release under such higher Ca2+ in the junctional space (lower panel). The proposed pathogenic mechanism by CaV1.1 variants may also explain pathogenesis by depolarizing muscle relaxants, such as succinylcholine. Succinylcholine binds to nicotinic acetylcholine receptors to trigger persistent small depolarizations. This, in turn, inactivates voltage-gated Na+ channels to suppress action potentials40. However, such depolarizations may also activate DICR to increase [Ca2+]JS. This might trigger regenerative Ca2+ release in patients carrying gain-of-function variants on the RyR1 channel or associated proteins41,42. Thus, an increase in [Ca2+]JS by DICR may be an important factor for underlying mechanism of MH pathogenesis. The S879P and F1161L variants largely shifted the voltage dependence of the charge movements to more negative potential (Fig. 4). Several CaV1.1 variants from MH-susceptible patients have been characterized using dysgenic myotubes: R174W16,17, R1086H18, and T1354S19. Myotubes expressing these CaV1.1 variants exhibited increased sensitivity to caffeine, suggesting that these variants show an MH phenotype. However, no existing variants showed clear alterations in charge movements16–19. Thus, the two variants discovered in this study have novel phenotypes. How do the two variants shift the voltage dependence of the charge movements? Both the variant residues are located in the alpha-helix of the transmembrane segments: S879 in S3 of the third voltage-sensing domain (VSDIII), and F1161 in S2 of the fourth VSD (VSDIV) (Fig. 1). Recent cryo-electron microscopy structures of the CaV1.1 complex43,44 have revealed that these two residues are in close proximity to the voltage-sensing S4 segment (Supplementary Fig. 6). Substitution of S879 to a proline can result in the termination of the helix structure by proline, potentially causing a collapse at the position between S3 and S4. F1161 forms a hydrophobic space with L1127 and I1128 in S1 and may interact with K1245 in S4. Substitution of the phenylalanine with leucine disrupts this space and interaction, altering the helix positions around S4. Recent studies using mutagenesis or voltage-clamp fluorometry have demonstrated that VSDIII is exclusively responsible for E–C coupling45,46. Hyperpolarizing shift of DICR and charge movement in S879P (Figs. 1 and 4), a variant located on VSDIII, can well be explained by alterations of VSDIII voltage sensor. On the other hand, it remains unclear how F1161L, a variant located on VSDIV, causes similar hyperpolarizing shift to S879P (Figs. 1 and 4). Further experiments are needed to address this question. In contrast to the charge movements, S879P and F1161L exhibited no substantial changes in the amplitude and the voltage dependence of Ca2+ currents (Fig. 5). Existing evidence suggests that VSDI and VSDIV are critical for controlling the current properties: VSDI is thought to be primarily involved in slow Ca2+ current activation47,48, whereas VSDIV determines the voltage dependence49,50. F1161L showed markedly slower activation of Ca2+ currents (Fig. 5), suggesting that VSDIV might also contribute to the activation of the Ca2+ current. We show that Qmax is significantly reduced in the S879P and F1161L variants compared with WT (Fig. 4). This finding was further confirmed by the coupling efficiency data (Fig. 5). The reduction in Qmax was also observed in D1382V (Supplementary Fig. 5). Since total and plasma membrane expressions of all four variants were comparable to those of WT (Supplementary Fig. 2), this may be attributed to alterations of some intrinsic properties of CaV1.1. Interestingly, reduced Qmax has been reported with CaV1.1 variants associated with hypokalemic periodic paralysis (HypoPP), such as R528H, R528G, R897S, R900G, and R900S51,52. Most HypoPP mutations target arginine residues in the voltage-sensing S4 segment53. Further experiments will be required to clarify the underlying mechanism. In this study, we analyzed CaV1.1 variants using our reconstituted DICR platform. Although this platform has been shown to recapitulate functions of CaV1.122, it is outside the context of skeletal muscle. Indeed, in our platform voltage-dependent inactivation of Ca2+ release was not observed22, probably due to lack of γ1 subunit which contributes to voltage-dependent inactivation28,29. It is necessary to validate these results in a skeletal muscle context, such as primary cultured myotubes and living animals. We detected no changes in DICR (Fig. 1) or caffeine-induced Ca2+ release (Fig. 2) in A560T and D1382V. However, these residues are located within the transmembrane segments and might affect the conformation of the transmembrane helices (Supplementary Fig. 6). It is possible that these CaV1.1 variants have some effects that were not detected by our platform. Further characterizations will address these questions. Materials and methods Genetic testing The gene panel study for malignant hyperthermia was approved by the Ethics Committee of Hiroshima University (Approval No. E2015-9151-25; last updated on December 23, 2024). Written informed consent was obtained from each participant or their legal guardian prior to genetic testing. For certain cases involving the use of cryopreserved muscle tissue, an opt-out procedure was applied in accordance with institutional guidelines. Genomic DNA was extracted from EDTA-stabilized peripheral blood samples or cryopreserved skeletal muscle tissue using the QIAamp® DNA Mini Kit (QIAGEN GmbH, Germany), in accordance with the manufacturer’s instructions. A gene panel analysis targeting malignant hyperthermia-related genes, including RYR1, CACNA1S, and 23 additional associated genes, was performed using whole-exome sequencing (WES) of genomic DNA. Variants were filtered based on the following criteria: a Rare Exome Variant Ensemble Learner (REVEL) score ≥ 0.75, a Combined Annotation Dependent Depletion (CADD, GRCh37-v1.7) score ≥ 20, and a minor allele frequency (MAF) < 0.001 in the Genome Aggregation Database (gnomAD v4.0 Exome). Baculovirus production Baculovirus (BV) for essential components in skeletal muscle E–C coupling (CaV1.1, β1a, JP2, Stac3, and Kir2.1) and R-CEPIA1er were produced as described previously22. MH-associated CaV1.1 variants were constructed by inverse PCR (primer pairs listed in Supplementary Table 3) using mouse CaV1.1 with an exon 29 insertion50 as a template. Each cDNA was cloned into the modified donor vector (pFastBac1-VSVG-CMV-WPRE). BV was produced in Sf9 cells using the Bac-to-Bac system (LifeTechnologies). BV titers were immunologically determined using an anti-gp64 antibody (AcV5, Santa Cruz Biotechnology)54. Titers were 1–2 × 108 pfu/mL for all viruses. The BV solutions were supplemented with 2.5% DMSO, quickly frozen in liquid nitrogen, and stored at −80 °C until use55. Surface biotinylation assay RyR1-HEK293 cells were seeded on 6-well plates at a density of 5 × 105 cells/well in 1.5 mL of culture media. One day after seeding, the culture media were replaced with 1.5 mL of the media containing 1 µg/mL doxycycline (for induction of RyR1) and BV solutions for WT or CaV1.1 variants, β1a, JP2, Stac3, and Kir2.1 (30 µL each). After 24 h, cells were rinsed twice with ice-cold phosphate-buffered saline containing 1 mM CaCl2 and 1 mM MgCl2 (PBS+) and incubated for 30 min at 4 °C in 1 mL of PBS+ containing 0.5 mg/mL sulfo-NHS-SS-biotin (Cayman Chemical, MI, USA) to specifically label surface proteins. Then, cells were further incubated for 10 min at 4 °C with 100 mM glycine with PBS+. The cells were rinsed twice with PBS + , harvested, and lysed with 100 μL of Buffer A (0.15 M NaCl, 20 mM MOPS, pH 7.4, 10% glycerol) supplemented with 1% n-dodecyl-β-D-maltopyranoside (DDM) and a protease inhibitor cocktail. After centrifugation at 20,000×g for 5 min, the supernatant was diluted with 400 μL of Buffer A and incubated overnight at 4 °C with Dynabeads Streptavidin C1 magnetic beads. The beads were washed four times with 500 μL of Buffer A containing 0.2% DDM, and proteins bound to the beads were eluted by SDS sample buffer containing 100 mM dithiothreitol at 37 °C for 30 min. The eluted proteins were separated on 3–15% polyacrylamide gels and transferred to PVDF membranes. Membranes were probed with primary antibodies against CaV1.1 (sc-514685, Santa Cruz Biotechnology) and β-actin (sc-47778, Santa Cruz Biotechnology) as a loading control, followed by HRP-labeled anti-mouse IgG (04-18-18, KPL). Positive bands were detected by chemiluminescence using ImmunoStar LD (Fujifilm Wako Chemicals) as a substrate. DICR assay The DICR assay was carried out as described previously22. Briefly, stable HEK293 cells expressing WT RyR1 were seeded on 96-well clear-bottom black microplates (#3603; Corning) at a density of 3 × 104 cells/well in 100 µL culture media supplemented with baculovirus solution for R-CEPIA1er. One day after seeding, 100 µL culture media containing 2 µg/mL doxycycline (for induction of RyR1) and baculovirus solutions for WT or CaV1.1 variants, β1a, JP2, Stac3, and Kir2.1 (2 µL each) were added to each well. The multiplicity of infection (MOI) was 7.5–15 for each BV (depending on the virus titer). After 24 h, the culture media were replaced with 81 µL normal Krebs solution (140 mM NaCl, 5 mM KCl, 2 mM CaCl2, 1 mM MgCl2, 11 mM glucose, 10 mM HEPES, pH 7.4), and the microplate was placed in a FlexStation3 fluorometer preincubated at 37 °C. Signals from R-CEPIA1er, which was excited at 560 nm and emitted at 610 nm, were captured from the center of the well at 1.5 mm diameter (corresponding to 2–3 thousands cells) at 5-s intervals for 150 s. Additionally, 30 s after starting, 54 µL of the high [K+]o or caffeine solution was applied to the cells. The fluorescence change was expressed as F/F0, where the average fluorescence intensity over the final 25 s (F) was normalized to that over the initial 25 s (F0). To obtain EC50 values for extracellular K+ ([K+]o) and caffeine, the data were fitted to the following equation:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$F/{F}_{0}=1-A\times \frac{{x}^{n}}{\left({x}^{n}+{K}^{n}\right)}$$\end{document}F/F0=1−A×xnxn+Knwhere F/F0 represents the fluorescence ratio of R-CEPIA1er at a given [K+]o or caffeine concentration (x), A is the minimal attainable F/F0 value, K is the EC50 value, and n is the Hill coefficient. Based on the shapes of the [K+]o- and caffeine-dependent curves, the Hill coefficients did not differ substantially between WT and mutant CaV1.1 channels. Therefore, fixed values of n = 6 and n = 2 were used for [K+]o-induced and caffeine-induced Ca2+ release, respectively. For determination of resting [Ca2+]ER, cells were incubated with 81 µL normal Krebs solution, and 54 µL Krebs solution containing 50 mM Ca2+ and 50 µM ionomycin was added 30 s after starting (the final concentrations of Ca2+ and ionomycin were 20 mM and 20 µM, respectively) to obtain maximum fluorescence intensity of R-CEPIA1er (Fmax). R-CEPIA1er fluorescence was captured every 10 s for 5 min. The [Ca2+]ER was determined by normalizing the average fluorescence intensity of the initial 25 s (F0) to that of the last 25 s (Fmax). Electrophysiology Cells stably expressing WT RyR1 were transduced with baculovirus encoding CaV1.1 (WT, S879P, or F1161L MH mutants), β1a, Stac3, JP2, and Kir2.1. Cells were replated on the glass coverslips 4–6 h after viral transduction. Then, 24 h after baculovirus transduction, the ionic and gating currents of CaV1.1 were recorded in the whole-cell configuration at room temperature with a patch–clamp amplifier (HEKA EPC10). The currents were low-pass filtered at 1 kHz and sampled at 100 kHz. Patch pipettes (3–4 MΩ) were fabricated from borosilicate glass capillaries (Kimax-51; Kimble Glass) and coated with Sylgard 184 (Dow Corning Toray Company). The series resistance was maintained below 7 MΩ and routinely compensated by approximately 70% using the amplifier. All whole-cell currents were measured with an intracellular solution (145 mM D-glutamate, 2 mM MgCl2, 10 mM EGTA, 2 mM MgATP, and 10 mM HEPES-CsOH, pH 7.4). The extracellular bath solution contained 10 mM CaCl2, 145 mM TEA-Cl, 5.5 mM glucose, and 10 mM HEPES (pH 7.4, adjusted with TEAOH). Ionic currents of CaV1.1 were recorded using the following pulse protocol: the membrane potential (Vm) was stepped from a holding potential of −80 mV to test potentials ranging from −80 to +80 mV in 10 mV increments, with each step lasting 300 ms and delivered at 1-s intervals. The liquid junction potential (LJP) of +20 mV was corrected. Peak current amplitudes at each Vm were measured and plotted against the respective membrane potentials. The resulting current–voltage (I–V) curve from each patch was normalized to the maximum inward current amplitude and fitted using the following equation:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$I\left({V}_{m}\right)={G}_{\max }({V}_{m}-{V}_{{rev}})/(1+exp ({V}_{0.5}-{V}_{m})/k),$$\end{document}IVm=Gmax(Vm−Vrev)/(1+exp(V0.5−Vm)/k),where I(V) is the peak current density at the command potential V, Gmax is the maximum conductance, Vrev is the reversal potential, V0.5 is the half-activation potential, and k is a slope. The activation kinetics of ionic currents through CaV1.1 were analyzed as follows: the first 100 ms of the channel current at +30 mV was fitted to a single exponential function:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$I\left(t\right)={y}_{0}+A\ast exp \left(-\left(t-{t}_{0}\right)/t\right),$$\end{document}It=y0+A*exp−t−t0/t,where I(t) is a current at time t, t0 and y0 are the horizontal and vertical axis offset, respectively, A is the channel current amplitude, and t is the time constant. The voltage dependence of gating charge movement was recorded using cells expressing either CaV1.1 WT or MH mutants with a pore-dead mutation to block ionic currents. The membrane potential was stepped from a holding potential of −80 mV to −100 mV for 100 ms, followed by test potentials ranging from −100 to +80 mV in 10 mV increments for 20 ms at 1-s intervals. No online leak correction was applied. Offline leak correction was performed by subtracting the steady-state current at the end of each test pulse, which was assumed to represent the linear leak current at that voltage. “ON” gating currents were then integrated for 2 ms following each potential change and normalized to cell capacitance. The gating charge (Q) at each Vm was plotted and fitted to the following Boltzmann equation:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Q={Q}_{max }* (1/(1+exp (({V}_{0.5}-{V}_{m})/k))),$$\end{document}Q=Qmax*(1/(1+exp((V0.5−Vm)/k))),where Qmax is the maximum Q, V0.5 is the half-maximum Vm, and k is a slope factor. The coupling efficiency of recombinant CaV1.1 channels was analyzed in cells expressing CaV1.1 WT or MH mutants, following methods described by Hulme et al.56 and Kashihara et al.57. Gating charge movement was measured by applying a series of test pulses from a holding potential of −100 mV to potentials between +60 and +80 mV in 2 mV increments for 20 ms at 3-s intervals, integrating the ON gating charge for 2 ms from the beginning of each test pulse. Ca2+ tail currents were recorded after repolarization to −50 mV from each test pulse. Statistics and reproducibility Sample sizes were described in each figure legend. Data are presented as the individual data points or the means ± SD depending on the experiments. Unpaired Student’s t-test was used for comparisons between two groups. One-way analysis of variance (ANOVA) followed by Dunnett’s test was performed to compare three or more groups. Statistical analyses were performed using Prism 9 (GraphPad Software Inc., La Jolla, CA, USA). P < 0.05 was considered as statistically significant. Reporting summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Supplementary information Supplementary Information Description of Additional Supplementary files Supplementary Data Reporting Summary Supplementary Information Description of Additional Supplementary files Supplementary Data Reporting Summary