Mice with R2509C-RYR1 mutation exhibit dysfunctional Ca²⁺ dynamics in primary skeletal myocytes

The skeletal muscle contracts when depolarization is transmitted to the dihydropyridine receptor in the T-tubule, and Ca²⁺ is released through type 1 ryanodine receptor (RYR1) from the sarcoplasmic reticulum (SR; Meissner, 1994; Iino, 1999; Rios, 2018; Woll and Van Petegem, 2022). RYR1 is a monomeric protein molecule consisting of ∼5,000 amino acids that form a tetramer; this protein functions as a giant Ca²⁺ release channel with Ca²⁺-induced Ca²⁺ release (CICR) properties (Schneider, 1994; Endo, 2009; Rios, 2018). Mutations in RYR1 have been reported to induce severe muscle diseases, such as malignant hyperthermia (MH; MIM accession no. 145600) and central core disease (CCD; MIM accession no. 117000), the latter of which is characterized by muscle weakness (Treves et al., 2008; Allard, 2018; Jungbluth et al., 2018).

MH is induced by inhalational anesthetics, such as halothane and isoflurane, as well as by depolarizing muscle relaxants. Typical symptoms of MH are elevated body temperature (up to ∼42°C) and enhanced skeletal muscle contraction. MH-related mutations in RYR1 are thought to cause massive Ca²⁺ release from the SR due to acceleration of CICR. MH is a relatively rare disease, occurring in only one in tens of thousands of cases of anesthesia; however, once it occurs, it can be fatal unless treated promptly (Sinha et al., 2017). It has also been reported that MH mutations in RYR1 are involved in severe heat stroke (Wappler et al., 2001; Davis et al., 2002). Analyses of the RYR1 gene in patients with MH and related muscle diseases have identified >300 different mutations located throughout the molecule, including in the N-terminal, central, and C-terminal regions (Lanner et al., 2010; Pancaroglu and Van Petegem, 2018).

The x-ray crystal structure of the RYR1 N-terminal region was resolved in 2010 (Tung et al., 2010). In 2015, three research groups (Efremov et al., 2015; Yan et al., 2015; Zalk et al., 2015) reported its protein structure at the near-atomic level using cryo-electron microscopy. In the following year, the structure of the channel in its open state was reported (des Georges et al., 2016), improving the understanding of its 3-D structure. To understand the detailed pathogenesis of MH, the structure and function of the mutants must also be comprehensively analyzed.

A heterologous expression system in HEK293 cells has been used for functional analysis of MH-related mutations in RYR1 (Tong et al., 1997; Tong et al., 1999). We previously performed functional analysis using HEK293 cells overexpressing a disease mutant of RYR1 that causes MH syndrome in humans (Murayama et al., 2015; Murayama et al., 2016; Yamazawa et al., 2020). We reported that the Y523S mutation in the N-terminal region (Murayama et al., 2015; Yamazawa et al., 2020) and R2508C mutation in the central region (Murayama et al., 2016) lead to extremely severe phenotypes that reduce the stability of the closed state of the channel, thereby causing Ca²⁺ leakage from the SR.

To fully understand the mechanisms by which MH occurs in humans, valid animal models are needed. Since the turn of the century, studies using knock-in/knockout mouse models with RYR1 mutations, corresponding to human MH mutations, have been reported (Chelu et al., 2006; Yang et al., 2006; Yuen et al., 2012; Lopez et al., 2018). We previously generated an MH mouse model (R2509C-RYR1 mice) carrying a p.R2509C mutation in RYR1 using the CRISPR/Cas9 system (Yamazawa et al., 2021). In R2509C-RYR1 heterozygous (R2509C-Het) mice, MH and MH-like episodes were induced by volatile anesthetics as well as by an increase in environmental temperature (Yamazawa et al., 2021). As in humans with MH, MH responses in R2509C-Het mice were characterized by sustained skeletal muscle contractions, resulting in hyperthermia, hyperventilation, rhabdomyolysis, and, ultimately, death. Additionally, R2509C-RYR1 homozygous (R2509C-Hom) mice died in utero (Yamazawa et al., 2021), as did knock-in homozygous mice with R163C (Yang et al., 2006) and Y524S (Chelu et al., 2006) RYR1 mutations. Therefore, no experiments on intracellular Ca²⁺ responses have been performed using skeletal muscle cells from R2509C-Hom mice. In the present study, we generated primary cultured skeletal myocytes to investigate Ca²⁺ dynamics in R2509C-Hom mice. We found that Ca²⁺ homeostasis was markedly disrupted in myocytes from R2509C-Hom mice, accompanied by sarcomere shortening.

All animal experiments were performed in accordance with the regulations and guidelines of the Institutional Animal Care and approved by the Ethics Committees of The Jikei University School of Medicine. Mice were housed in isolator cages, fed with food and water ad libitum, and kept in a controlled environment with 12-h light/12-h dark cycles, 23–25°C temperature, and 50–60% relative humidity under specific pathogen-free conditions in the animal facilities of The Jikei University of Medicine. Homozygous mutant neonates were obtained by mating heterozygotes, and their littermates were used as controls. The genotypes of all neonates and embryos used in the experiments were determined using PCR analysis.

Primary cultured myocytes at embryonic days 17–18 were prepared using a previously described procedure with some modifications (Rando and Blau, 1994; Yamazawa et al., 1996; Yamazawa et al., 1997). Briefly, the forelimbs and hindlimbs were removed from mice embryo, and the bones were dissected. The muscle was cut into small fragments and enzymatically dissociated with 2 mg/ml collagenase (Worthington Biochemical) and 2 mg/ml dispase (FUJIFILM Wako Pure Chemical Corporation) at 37°C for 40–50 min. The fragments were passed through a 40-μm cell strainer, and the suspension was subjected to low-speed centrifugation. The pellet was resuspended in Ham’s Nutrient Mixture (F10), supplemented with 20% (v/v) FBS, penicillin (100 U/ml), streptomycin (100 U/ml), and 5 ng/ml recombinant human fibroblast growth factor-basic (Nacalai Tesque Inc.). Myoblasts were differentiated into myotubes/myocytes using Dulbecco’s modified Eagle’s medium (Nacalai Tesque Inc.) containing 2% horse serum. After 2–5 d of culture in differentiation medium, the myocytes were used in experiments. In optical heating experiments, 100 µM N-benzyl-N-nitroso-p-toluenesulfonamide (FUJIFILM Wako Pure Chemical Corporation) was added to HEPES-Krebs solution (140 mM NaCl, 5 mM KCl, 2 mM CaCl₂, 1 mM MgCl₂, 11 mM glucose, and 5 mM HEPES, pH 7.4) to suppress the movement of skeletal myocytes upon activation.

Skeletal muscle cells isolated from mice were prepared based on a previously described procedure, with some modifications (Yamazawa et al., 2021). Briefly, WT and R2509C-Het mice (8–20 wk old) were anesthetized through intraperitoneal injection of an anesthetic mixture (0.75 mg/kg medetomidine, 4 mg/kg midazolam, and 5 mg/kg butorphanol) before euthanasia. Flexor digitorum brevis (FDB) muscles were dissected and incubated with 2 mg/ml collagenase (Worthington Biochemical) in the HEPES-Krebs solution containing 2 mg/ml BSA (A8806; Merck) for 2 h at 37°C. Following incubation, single cells were separated by gentle trituration in HEPES-Krebs solution. Isolated single FDB fibers were seeded onto a laminin (Sigma-Aldrich)-coated or iMatrix-511 (T311; Takara Bio)-coated cover slip of a 35-mm glass-bottom dish. Next, 100 µM N-benzyl-N-nitroso-p-toluenesulfonamide, in HEPES-Krebs solution, was applied to suppress the movement of FDB fibers upon activation.

Cultured primary myocytes were incubated with 4 µM Cal-520-AM (AAT Bioquest) or fura-2-AM (Thermo Fisher Scientific) dye in HEPES-Krebs solution, containing 1 mg/ml BSA, for 30 min at room temperature (RT; 23–26°C). The cells were then washed thrice with HEPES-Krebs solution to remove excess dye; after 30 min of de-esterification, fluorescence images were obtained with a 20× objective lens (NA 0.75, UPlanSApo; Olympus) of an inverted microscope (DMI6000B; Leica) equipped with a complementary metal-oxide-semiconductor camera (ORCA-Flash4.0; Hamamatsu Photonics), at the rate of one frame every 0.5–2 s, using a MetaFluor v7 imaging system (Molecular Devices). The fluorescence wavelength was switched using a fluorescence filter exchanger (LAMBDA 10-3; Sutter Instrument). For Cal-520, it was excited at 480 ± 9 nm, and fluorescence was measured at 520 ± 14 nm. For fura-2, it was excited at alternating wavelengths of 340 (340 ± 13) nm and 380 (387 ± 6) nm, and the fluorescence was measured at 520 ± 14 nm for both wavelengths. Regions of interest corresponding to individual cells were selected and the mean fluorescence intensity (F) of each region of interest minus the background intensity was calculated for each frame.

We used the F340/F380 ratio (R; the value of F at an excitation wavelength of 340 nm divided by the value at an excitation wavelength of 380 nm) to estimate intracellular Ca²⁺ concentration ([Ca²⁺]_cyt), as described previously (Murayama et al., 2016; Yamazawa et al., 2020).

To measure the cell temperature, we used ERthermAC, a photostable fluorescent molecular thermometer that selectively targets the sarco-endoplasmic reticulum (SR/ER). Cultured primary myocytes were loaded with 4 µM Cal-520-AM and 200 nM ERthermAC in HEPES-Krebs solution, containing 1 mg/ml BSA, for 30 min at RT. ERthermAC was excited at 556 ± 10 nm, and fluorescence was measured at 617 ± 37 nm. Excitation light and emission filters for Cal-520 and ERthermAC were exchanged alternately.

Drugs were applied to the myocytes or FDB fibers using a homemade 6-channel puffing pipette (Yamazawa et al., 2020), which changed the solution surrounding the cells within ∼1 s.

The microscopy system used for optical heating has been described previously (Oyama et al., 2022). Briefly, confocal images were obtained using an inverted microscope (IX73; Olympus) equipped with a Nipkow confocal scanner unit (CSU-X1; Yokogawa Electric), a dichroic mirror (Di01-T405/488/561; Semrock), a 20× objective lens (UPlanSApo, NA = 0.75; Olympus), and an EM-CCD camera (iXon Ultra; Andor Technology). For imaging of Cal-520, a 488-nm laser light (Vortran Laser Technology) and an emission filter (FF01-520/35; Semrock) were used. The local temperature was increased using an infrared laser (KPS-STD-BT-RFL-1455-05-CO; Keopsys) with an IR-LEGO mini (SIGMAKOKI). The change in local temperature was calculated through thermal quenching of 10 µg/ml Alexa Fluor 555 conjugated to 10-kD dextran (D34679; Thermo Fisher Scientific) in HEPES-Krebs solution, as previously described (Oyama et al., 2015). The dye was excited using a 561-nm laser light (Cobolt), and the fluorescence was captured through an emission filter (FF01-617/73; Semrock). Microscopic experiments were performed at 24°C. To inhibit the activity of sarco- and endoplasmic reticulum Ca²⁺-ATPase (SERCA), the myocytes were incubated in HEPES-Krebs solution containing 1 mg/ml BSA, 4 µM Cal-520AM, and 20 µM cyclopiazonic acid (CPA), for 30 min at 24–26°C, and then observed 60 min later in HEPES-Krebs solution containing 20 µM CPA.

Microscopic images were analyzed using the ImageJ software (National Institutes of Health). The fluorescence intensity of Cal-520 was measured in an area of 139 µm². To minimize artifacts caused by movement and drifting focus due to contracture during heating, the area at the center of the cell was selected. The maximum changes in the relative intensity of Cal-520 (ΔF_max/F₀) were calculated from the maximum and minimum F/F₀ for 5 s during heating, where F is the fluorescence intensity at an arbitrary time and F₀ is the averaged intensity for 1.2 s after the initiation of recording (i.e., 10 s before beginning the heating). If the maximum F/F₀ – 1 > 1 – minimum F/F₀, then ΔF_max/F₀ = maximum F/F₀ – 1. If the maximum F/F₀ – 1 < 1 – minimum F/F₀, then ΔF_max/F₀ = minimum F/F₀ – 1. The background intensity without excitation was subtracted.

To measure the change in local temperature, relative changes in Alexa Fluor 555 intensity were calculated as ΔF/F₀ = (F_heating − I_back)/(F_after − I_back), where F_heating is the averaged intensity for 1 s at the end of the heating period, F_after is the averaged intensity for 1 s at ∼20 s after the end of heating, and I_back is the background intensity when the excitation light is turned off. The ΔF/F₀ of Alexa Fluor 555 was converted to ΔT using the relationship between ΔT and ΔF/F₀ (Oyama et al., 2015).

Primary cultured myocytes were fixed with 4% paraformaldehyde in PBS (T900; Takara Bio) and permeabilized with 0.3% Triton X-100 (X100; Sigma-Aldrich) in PBS. After rinsing with PBS, the cells were incubated with 2% BSA (01281-84; Nacalai Tesque Inc.) in PBS for 1 h at RT. For immunofluorescence staining, the cells were incubated overnight at 4°C with mouse–anti-myosin heavy chain antibody (MF20, 1:200; 14-6503-82; Thermo Fisher Scientific). After washing, the cells were incubated with Alexa Fluor 546 rabbit anti-mouse IgG (H+L; 1:1,000; A-11060; Thermo Fisher Scientific) and 4′,6-diamidino-2-phenylindole dye (1:1,000; 340-07971; Dojindo Laboratories) for 1 h at RT. Images of the primary cultured cells were captured using an LSM 880 confocal laser-scanning microscope (Carl Zeiss) equipped with a 20× objective (α Plan-Apochromat, NA 0.8; Carl Zeiss).

Myoblasts and myotubes were lysed with Pro-Prep protein extraction solution (iNtRON Biotechnology). The extracted proteins were separated on 3–15% polyacrylamide gels and transferred to PVDF membranes. Membranes were probed with the primary antibodies against myosin heavy chain (MHC; MF20; Thermo Fisher Scientific), RYR1 (F1; Santa Cruz Biotechnology), and GAPDH (6C5; Santa Cruz Biotechnology), followed by HRP-labeled anti-mouse IgG (04-18-18; Kirkegaard & Perry Laboratories). Positive bands were detected by chemiluminescence using ImmunoStar LD (FUJIFILM Wako Pure Chemical Corporation) as a substrate.

Transmission electron microscopy (TEM) experiments were performed based on a previously described procedure (Yamazawa et al., 2016), with slight modifications. Briefly, primary skeletal muscle culture cells were fixed with 2% glutaraldehyde (Nacalai Tesque Inc.) in 0.1 M phosphate buffer for 1 h at 4°C and post-fixed with 1% osmium tetroxide (Nisshin EM Co., Ltd.) in the same buffer at 0°C for 30 min. The tissues were first dehydrated with a graded series of ethanol, further with 100% ethanol for 10 min three times, and embedded in Epok 812 (Oken). Sections of 60-nm thickness were cut with a diamond knife and mounted on grids. After staining with uranyl acetate and lead citrate, sections were examined using an H-7500 electron microscope (Hitachi) at 80 kV.

The unpaired Student’s t test was used for comparisons between two groups. A one-way ANOVA followed by Tukey’s test was performed to compare three or more groups. Multiple groups were compared using the Steel test. Mann–Whitney U test was used for comparisons of two independent samples. These tests were performed using Prism v7 and v9 (GraphPad Software, Inc.).

Fig. S1 shows the morphology of R2509C-Hom embryos. Fig. S2 shows the distribution of the lengths of A-band in WT and R2509C mutant primary myocytes. Fig. S3 shows spontaneous Ca²⁺ transients in WT and R2509C mutant primary myocytes. Fig. S4 shows the time-course of heat-induced Ca²⁺ release in WT and R2509C primary myocytes. Fig. S5 shows temperature measurement of the SR in WT and R2509C-Het FDB-isolated fibers. Fig. S6 shows ERthermAC staining in primary myocytes. Video 1 shows spontaneous contraction in WT myocytes. Video 2 shows spontaneous contraction in R2509C-Het myocytes. Video 3 shows spontaneous contraction in R2509C-Hom myocytes. Video S4 shows spontaneous Ca²⁺ transients in WT myocytes. Video 5 shows spontaneous Ca²⁺ transients in R2509C-Het myocytes. Video 6 shows spontaneous Ca²⁺ transients in R2509C-Hom myocytes.

To characterize skeletal muscle cells from R2509C-Hom mice, primary cultures were prepared from WT and R2509C mouse skeletal muscle at embryonic days 17−18 (E17.5). We initially attempted to prepare primary-cultured skeletal muscle cells from neonatal R2509C mice; however, R2509C-Hom fetuses showed abnormal deaths between E17 and E19, with subcutaneous hemorrhage, subcutaneous effusion, and translucent skeletal muscles (see Fig. S1). These phenotypes were distinct from those in homozygous RYR1 knockout mice that have been reported to die in the neonatal period (dyspedic mice; see Takeshima et al., 1994). Indeed, we successfully prepared cultured skeletal muscle cells from neonatal dyspedic mice at E19.5 (Yamazawa et al., 1996).

Under the present culture conditions, primary myotubes/myocytes from WT and R2509C (R2509C-Het and R2509C-Hom) mice developed multiple nuclei within 3 d of differentiation induction (Fig. 1 A). The morphology of the cells was similar. Interestingly, primary myocytes from WT (Video 1) and R2509C-Het (Video 2) mice showed spontaneous contractions, whereas those from R2509C-Hom mice (Video 3) did not. Fluorescence immunostaining with anti-myosin heavy chain antibodies revealed a similar level of expression of MHC in WT, R2509C-Het, and R2509C-Hom myocytes (Fig. 1 B). Also, Fig. 1 C presents representative Western blots of MHC in WT, R2509C-Het, and R2509C-Hom myocytes up to 4 d after differentiation. There was no significant difference in average MHC expression levels between WT and R2509C myocytes from day 0–4 (P > 0.05). Therefore, it is unlikely that the absence of spontaneous contractions in R2509C-Hom myocytes occurred because of loss of myosin molecules.

The morphology of primary myocytes was further investigated after 3 d of differentiation using TEM (Figs. 2 and S2). Sarcomere structures were preserved with clear Z-lines in all three types of primary myocytes (Fig. 2 A); however, the I-bands were not detected in R2509C-Hom myocytes, indicating the occurrence of sarcomere shortening. Indeed, the sarcomere length (SL) in R2509C-Hom myocytes (1.7 ± 0.3 µm) was significantly shorter than that in the WT (2.2 ± 0.3 µm) or R2509C-Het (2.3 ± 0.4 µm) myocytes (P < 0.0001 and 0.0001, respectively; Fig. 2 C). To confirm the contribution of RYR1 mutants to the marked shortening of SL in R2509C-Hom myocytes, we tested the effects of a novel RYR1 inhibitor, 6,7-(methylenedioxy)-1-octyl-4-quinolone-3-carboxylic acid (Cpd1; Mori et al., 2019; Yamazawa et al., 2021). Cpd1 (2 μM) was continuously applied to the culture media from the onset of differentiation. The SL in Cpd1-treated R2509C-Hom myocytes (2.0 ± 0.3 µm) was significantly longer than that in the R2509C-Hom myocytes (1.7 ± 0.3 µm; Fig. 2, B and C). To further investigate the distribution of SL, we have analyzed the histograms, as shown in Fig. 2 D. It is reasonable to assume that the distribution shows a single peak in WT, because myocytes are in the resting state. Indeed, the Gaussian fitting provided the mean ± SD as 2.2 ± 0.3 µm. The data in R2509C-Het were similarly fitted by a single Gaussian, providing the value as 2.3 ± 0.3 µm. In contrast, R2509C-Hom and R2509C-Hom+Cpd1 were not fitted well by a single Gaussian (Fig. 2 D). Therefore, these data sets were analyzed using two Gaussians (discussed below in Discussion); one of the peak values was fixed at 2.2 µm in both groups, and the other peak values were derived as 1.6 ± 0.2 and 1.8 ± 0.2 µm in R2509C-Hom and R2509C-Hom+Cpd1, respectively.

It was previously shown in MH mouse models that dyshomeostasis of [Ca²⁺]_cyt in skeletal myocytes is among the most common features of MH (Chelu et al., 2006; Yang et al., 2006). Therefore, we investigated whether and how Ca²⁺ transients are altered in R2509C-Hom myocytes. Fluorescence Ca²⁺ imaging showed that spontaneous Ca²⁺ transients occurred similarly as in WT (Fig. S3 A and Video 4) and R2509C-Het (Fig. S3 B and Video 5) myocytes but not in R2509C-Hom (Fig. S3 C and Video 6) myocytes (Fig. S3 D). We measured the resting intracellular Ca²⁺ concentration ([Ca²⁺]_cyt) using fura-2, a ratiometric fluorescent Ca²⁺ indicator, in myocytes differentiated from primary myoblasts from WT and R2509C-RYR1 mice (at day 2−4 after the onset of differentiation). The average resting [Ca²⁺]_cyt in WT myocytes (103.7 ± 40.7 nM) was close to the reported value of ∼100 nM (see Harkins et al., 1993; Royer et al., 2008). Resting [Ca²⁺]_cyt was significantly higher in R2509C-Het myocytes (124.5 ± 42.0 nM), and highest in R2509C-Hom myocytes (186.4 ± 71.4 nM; Fig. 3 A and Table 1). Therefore, the data for R2509C-Hom myocytes were further analyzed by varying the days of differentiation (Fig. 3 B). In addition, the large distributions of values in R2509C-Hom myocytes may be related to the variance of individual cellular functions involved in Ca²⁺ homeostasis. We found that [Ca²⁺]_cyt was elevated in R2509C-Hom myocytes at day 2 (176.6 ± 54.3 nM) after differentiation, as compared to those in WT (103.7 ± 40.7 nM) and R2509C-Het (124.5 ± 42.0 nM) myocytes (P < 0.0001 and 0.0001, respectively). Resting [Ca²⁺]_cyt increased with the days of differentiation, and the mean value was significantly high at day 4 of differentiation (222.5 ± 100.3 nM). To confirm the contribution of RYR1 mutants to the marked increase in [Ca²⁺]_cyt, we tested the effects of Cpd1. Cpd1 (2 μM) was continuously applied to the culture media from the onset of differentiation. In R2509C-Hom myocytes, a significant decrease in resting [Ca²⁺]_cyt by Cpd1 was observed at day 3 (179.5 ± 63.2 and 140.6 ± 52.7 nM for myocytes in the absence and presence of Cpd1, respectively) and day 4 (222.5 ± 100.3 and 151.4 ± 73.5 nM for control and Cpd1, respectively; Fig. 3 B and Table 1). In addition, Western blot analysis of RYR1 was performed to examine the relationship between RYR1 expression level and increase in resting [Ca²⁺]_cyt in R2509C-Hom myocytes. As shown in Fig. 3 C, the expression level of RYR1 tended to increase depending on the differentiation days.

We next examined how excitation–contraction coupling is altered in mutant myocytes. Electrical stimulation induced significant Ca²⁺ transients (responses >10% of the baseline fluorescence value, i.e., F₀) in WT, R2509C-Het, and R2509C-Hom myocytes without apparent differences in the response rate (>90%). However, the peak amplitude of Ca²⁺ transients (F_max/F₀) was reduced in the order of WT ≥ R2509C-Het > R2509C-Hom (Fig. 4 A, Table 2), although the amplitude for R2509C-Hom may be somewhat underestimated due to an increase in F₀ in R2509C-Hom myocytes (see Fig. 3 A). Similarly, in KCl-induced depolarization, the peak amplitude of Ca²⁺ transients (F_max/F₀) at KCl concentrations above 20 mM was significantly reduced in the order of WT $≅$ R2509C-Het > R2509C-Hom (Fig. 4 B and Table 2). To further characterize the functional properties of depolarization-induced Ca²⁺ transients, we plotted KCl dose-response curves. In each curve, the amplitude of Ca²⁺ transients was normalized by the maximum response to obtain the EC₅₀ value (Fig. 4 C). The order of the sensitivity to KCl stimulation was R2509C-Hom > R2509C-Het > WT (Fig. 4 C and Table 2). Subsequently, we characterized caffeine-induced Ca²⁺ transients as an index for CICR activity (Fig. 4, D and E, and Table 2). Significant Ca²⁺ transients were observed at and above 3 mM in WT myocytes, whereas the corresponding caffeine concentrations for R2509C-Het and R2509C-Hom were 1 and 0.3 mM, respectively. As followed for the response to electrical stimulation, there were no significant differences between groups in the reaction rate (>90%) in the presence of 10 or 20 mM caffeine. Although the absolute peak amplitude (F_max/F₀) varied in the order of WT ≥ R2509C-Het > R2509C-Hom, the caffeine sensitivity was found to vary in the order of R2509C-Hom > R2509C-Het > WT (Fig. 4, D and E, and Table 2).

We previously reported that Ca²⁺ transients are induced by infrared laser-based micro-heating in HEK293 cells expressing various RYR1 mutants as well as in isolated FDB fibers prepared from WT and R2509C-Het mice (Oyama et al., 2022). We found that mild heating (which did not affect [Ca²⁺]_cyt in WT muscle cells) significantly increased [Ca²⁺]_cyt in R2509C-Het muscle cells, demonstrating the heat-hypersensitivity of RYR1. Based on these results, we proposed a heat-mediated Ca²⁺ release mechanism through RYR1 from the SR, i.e., heat-induced Ca²⁺ release (HICR) that is likely to play a crucial role in the progression of MH at the cellular level (Oyama et al., 2022). Therefore, we next applied a heat pulse (ΔT = 11 ± 2°C from the base temperature T₀ = 24 ± 1°C; central value ± range) to primary myocytes from WT, R2509C-Het, and R2509C-Hom mice after loading with the fluorescent Ca²⁺ indicator Cal-520 (Fig. 5, A–F). An infrared laser beam focused with an objective lens formed a concentric temperature gradient around cells (Oyama et al., 2022). The temperature gradient obtained depends on the laser power used. In the present study, the maximum temperature rise was 13°C at the focal point, with a minimum of 10°C or less at the end of the field of view. The duration of heating was controlled by a mechanical shutter. The temperature gradient was formed within 0.1 s of the opening of the shutter, and then was eliminated within 0.1 s by heat diffusion when the shutter was closed (Itoh et al., 2014; Oyama et al., 2020). The change in temperature was calculated with temperature-sensitive fluorescent dextran (Oyama et al., 2015).

Then, we performed time-dependent analyses in WT, R2509C-Het, and R2509C-Hom myocytes. No significant differences were observed in Ca²⁺ level among the before (F₀), during (14–15 s), and after (19–20 s) heating in WT (Fig. 5 G). In R2509C-Het, [Ca²⁺]_cyt significantly increased during heating (Fig. 5 H), while in R2509C-Hom, it decreased during heating and increased after heating (Fig. 5 I). The fluorescence intensity of Cal-520 before heating was increased in the order of WT ≤ R2509C-Het < R2509C-Hom (Fig. 5 J), which is consistent with the order for the resting [Ca²⁺]_cyt (Fig. 3 A). Although little or no change in [Ca²⁺]_cyt was detected during heating in WT myocytes (Fig. 5, A and B; see thick black line), a substantial increase in [Ca²⁺]_cyt was detected in R2509C-Het myocytes (Fig. 5, C and D; see thick red line). The maximum changes in the relative fluorescence intensity of Cal-520 during heating (ΔF_max/F₀) were greater in R2509C-Het myocytes than in WT or R2509C-Hom myocytes (Fig. 5 K and Fig. S4). This finding is consistent with our previous result obtained in isolated R2509C-Het FDB fibers (Oyama et al., 2022).

Why did R2509-Het cause a heat-induced Ca²⁺ release, whereas R2509C-Hom did not, and rather caused a decrease in [Ca²⁺]_cyt? We recently reported that the ER/SR is a major Ca²⁺ source of the heat-induced Ca²⁺ bursts in HEK293 cells expressing WT and mutant RYR1 (Oyama et al., 2022), and that ER/SR Ca²⁺ is substantially depleted in cells expressing severe RYR1 mutation, such as R2509C. Subsequently, we tested whether the decrease in [Ca²⁺]_cyt during heating in R2509C-Hom myocytes is due to heat activation of SERCA. In R2509C-Hom myocytes, inhibition of SERCA by 20 μM CPA significantly suppressed the decrease in Ca²⁺ levels during heating (Fig. 6, A–D). The results suggest that ER/SR Ca²⁺ in R2509C-Hom myocytes were severely depleted, and heating caused Ca²⁺ uptake rather than Ca²⁺ release.

Finally, we investigated whether an increase in the cellular temperature is associated with the increase in [Ca²⁺]_cyt upon application of anesthesia. In a previous study, we reported that inhalation of isoflurane increases the rectal temperature in R2509C-Het mice, and that the exposure to isoflurane increases the [Ca²⁺]_cyt in isolated FDB fibers (Yamazawa et al., 2021). At the cellular level, Meizoso-Huesca et al. (2022) demonstrated that heat production is closely related to Ca²⁺ leakage through RYR1 in the skeletal muscles, using a fluorescent molecular thermometer (Meizoso-Huesca et al., 2022). In the present study, we used ERthermAC, a small-molecule fluorescent dye (Kriszt et al., 2017), to perform the SR-targeted thermometry in living cells (Oyama et al., 2022). This dye is highly photostable and can be used with Cal-520 for intracellular Ca²⁺ imaging (e.g., Hou et al., 2017; Oyama et al., 2022). Striation patterns following ERthermAC treatment represent the morphology of the SR along myocytes, indicating successful staining by the dye (Fig. 7 A). We found that application of 1% isoflurane using a puffer pipette to R2509C-Het FDB fibers increased the fluorescence intensity of Cal-520 (Fig. 7 B, right top) and that of ERthermAC decreased (Fig. 7 B, right bottom). Note that the fluorescence intensity of ERthermAC decreased in response to an increase in temperature (as in Krisz et al., 2017; Hou et al., 2017; Oyama et al., 2022). WT FDB fibers showed no apparent decrease in the fluorescence intensity of Cal-520 or ERthermAC (Fig. 7 B, left), except for a gradual time-dependent reduction because of the photobleaching effect (Fig. S5). The results suggest that the prominent reduction in the fluorescence intensity of ERthermAC in R2509C-Het FDB fibers (Fig. 7 C) is not an artifact caused by application of isoflurane, but rather, at least partially, is due to the isoflurane-induced temperature increase in the cells.

We compared several aspects of Ca²⁺ dynamics (resting Ca²⁺, KCl, caffeine) in primary cultured myocytes from recently generated MH model mice (R2509C mice). R2509C-Hom myocytes exhibited markedly increased resting [Ca²⁺]_cyt, reduced depolarization evoked and caffeine-induced Ca²⁺ release, and lacked spontaneous Ca²⁺ transients. R2509C-Hom myocytes also exhibited significantly shortened sarcomeres, consistent with elevated resting [Ca²⁺]_cyt. In addition, MH is induced by inhaled anesthetics to increase body temperature; however, the actual relationship between heat and Ca²⁺ transients remains unclear. Consequently, we measured [Ca²⁺]_cyt and cellular temperature simultaneously in the present study.

There was no significant difference in the SL between WT and R2509C-Het myocytes, and the value was within the range reported for resting skeletal myocytes (i.e., 2.0–2.5 µm; Goldspink, 1968). In contrast, the SL of R2509C-Hom myocytes was abnormally shortened compared to those for WT or R2509C-Het myocytes. These findings are consistent with the notion that sarcomeres in R2509C-Hom myocytes are contracted even under resting conditions. Excessive shortening of SL has been reported around the core in the leg muscles of aged Y522S heterozygous mice (Boncompagni et al., 2009), which is consistent with our results.

We also analyzed the mechanisms by which SL was shortened to ∼1.7 µm in R2509C-Hom myocytes (Fig. 2). Samples for TEM imaging were chemically treated with glutaraldehyde, followed by dehydration (see Materials and methods). This protocol has been reported to reduce the size of cells by ∼10% (Wyffels, 2001). If this is the case, SL in living R2509C-Hom myocytes can be calculated within the ∼1.3–1.7 μm range, which is the physiologically relevant A-band length in vertebrate striated muscles (cf. Suzuki et al., 2005; Udaka et al., 2008; Shimomura et al., 2016). The short SL in R2509C-Hom myocytes indicates that abnormal contractions occur in these cells, even in the resting state. Such abnormal contractions can likely be induced by enhanced Ca²⁺ leakage from the SR, resulting in a sustained increase in [Ca²⁺]_cyt (Fig. 3 and Tables 1 and 2). Extensively elevated [Ca²⁺]_cyt may cause the disorganization of myofibrillar structures in the skeletal muscles (Fig. S1), resulting in early death of R2509C-Hom embryos (see above).

The higher Ca²⁺ levels in R2509C-Het and R2509C-Hom myocytes than in WT (Fig. 3 A and Table 1) suggest that SL is equal to that in WT, or shortened due to tension generation according to the pCa-tension curve. R2509C-Het was well fit by a Gaussian, showing the peak and the SD as 2.3 µm (Fig. 2 D), which is quite similar to the case in WT, and their distributions are not significantly different (Fig. 2 C). Therefore, we can conclude that Ca²⁺ level distribution in R2509C-Het myocytes was in the region of relaxation. However, R2509C-Hom and R2509C-Hom+Cpd1 were not fit well by a single Gaussian. Therefore, we switch to the other possibility that the distribution of Ca²⁺ level is centered at the overlapping region between the relaxed and the partially activated state; hence, there are two respective distributions. These findings are consistent with those of previous studies (Balnave and Allen, 1996; Glass et al., 2018); i.e., in mouse FDB fibers, active force through actomyosin interaction is generated from 150 nM [Ca²⁺]_cyt, and [Ca²⁺]_cyt for half-maximal activation is 314 nM (Balnave and Allen, 1996; Glass et al., 2018). The SL of a myocyte in the group in the relaxed state should be similar to that in the WT that is also in the relaxed state. Therefore, the distributions of R2509C-Hom and R2509C-Hom+Cpd1 were analyzed based on two Gaussians, where one of the peaks was fixed at 2.2 μm. According to this analysis, the shorter peak is at 1.6 μm in R2509C-Hom, and it is recovered slightly to 1.8 μm in R2509C-Hom+Cpd1 (Fig. 2 D).The remarkable differences in the SL distributions of R2509C-Het and R2509C-Hom are consistent with the less significant difference of the resting [Ca²⁺]_cyt between WT and R2509C-Het myocytes when compared with that between R2509C-Het and R2509C-Hom (Fig. 3 A). It is also noticeable that the population at 2.2 μm that corresponds to the sarcomere in the relaxed state is higher in R2509C-Hom+Cpd1 than in R2509C-Hom. The recovery of SL to the longer distributions strongly suggests that the Ca²⁺ level, and in turn the tension level, was lowered in the presence of Cpd1.

It has been reported that the expression of RYR1 mRNA, which is not expressed in pre-differentiated myoblasts, is induced at ∼24 h after the onset of differentiation, with the expression level reaching the maximum at day 4 (Aley et al., 2010). Given this previous finding, the abrupt increase in the resting [Ca²⁺]_cyt in R2509C-Hom myocytes at day 4 is likely coupled with an increase in RYR1 expression and Ca²⁺ leakage from the SR. This prediction is supported by our previous result on R2508C RYR1-mediated Ca²⁺ leakage from the ER in HEK293 cells expressing mutant RYR1 (Murayama et al., 2016). This indicated that the increase in resting [Ca²⁺]_cyt in R2509C-Het and R2509C-Hom myocytes is caused by enhanced Ca²⁺ leakage through RYR1 from the SR.

Notably, F_max/F₀ in R2509C-Hom myocytes were unexpectedly low relative to those in WT or R2509C-Het myocytes, although it is underestimated because F₀ is increased. The results suggest that Ca²⁺ concentration in the SR lumen of R2509C-Hom myocytes was markedly reduced, as observed in the ER lumen with HEK293 cells overexpressing R2508C mutants (Murayama et al., 2016). In addition, the F_max/F₀ induced by 10 or 20 mM caffeine reportedly represents the maximal response of Ca²⁺ released from the SR/ER (Murayama et al., 2015; Murayama et al., 2016; Yamazawa et al., 2020). Therefore, we calculated the F_max/F₀ values under electrical stimulation and under maximal caffeine. As summarized in Table 2, the ratio values of WT (0.51: 3.59/7.01) and R2509C-Het (0.34: 2.42/7.04) myocytes were comparable, whereas the ratio values (0.70: 1.53/2.19) in R2509C-Hom myocytes were markedly greater than the others; therefore, the sensitivity to depolarization was likely increased in these cells. The increased sensitivity to depolarization in R2509C-Hom myocytes was also indicated by the EC₅₀ values determined from the dose–response curve for KCl (Fig. 4, B and C; and Table 2). These findings are consistent with those of a previous study showing increased sensitivity to K⁺-induced depolarization in myotubes from homozygous R163C mice (Yang et al., 2006).

In R2509C-Hom myocytes, fluorescence intensity was lower during heating and higher after heating than before heating (Fig. 5, E and F; and Fig. S4), coupled presumably with the heat-induced Ca²⁺ release mechanism, as demonstrated in our previous study using R2508C HEK293 cells (Oyama et al., 2022). CPA, a SERCA inhibitor, inhibited the decrease in [Ca²⁺]_cyt during heating in R2509C-Hom myocytes (Fig. 6 D). This result may have been caused by thermal activation of the SERCA, as demonstrated in our previous studies (Tseeb et al., 2009; Itoh et al., 2014; Oyama et al., 2022). This is because [Ca²⁺]_cyt is increased in R2509C-Hom myocytes coupled with decreased [Ca²⁺] in the SR lumen; therefore, [Ca²⁺]_cyt is lowered via heating-induced activation of SERCA. Hence, the differential Ca²⁺ response in R2509C-Het versus R2509C-Hom myocytes may be attributable to a difference in Ca²⁺ concentration between the cytoplasm and SR lumen.

The fluorescence intensity of ERthermAC started to decrease with a delay after onset of the isoflurane-induced Ca²⁺ increase (Fig. 7 B). A similar type of delay was observed in our previous studies using HeLa cells (Suzuki et al., 2007; Takei et al., 2014). The delay of a temperature increase following an increase in [Ca²⁺]_cyt may reflect the time for SERCA to be activated for its thermogenesis, as efficient heat release by the Ca²⁺-ATPase should occur only after [Ca²⁺]_SR is lowered to a certain value (de Meis et al., 2005). In the present study, this hypothesis was directly tested in R2509C-Hom myocytes, in which Ca²⁺ concentration in the SR was significantly reduced compared to that in WT and R2509C-Het myocytes (Fig. 3 and Table 1). We consider that in R2509C-Hom myocytes, the thermogenic activity of SERCA is immediately increased after onset of the isoflurane-induced Ca²⁺ increase and, therefore, the temperature increases with a shorter delay.

However, this experimentation could not be performed in primary myocytes because of the pre-differentiated myoblasts (MF20-negative cells) present at high density, which caused a strong background signal overlapping that of primary myocytes (Fig. 1 B and Fig. S6). Further studies are required to resolve this technical issue by developing (1) a method of obtaining differentiated myoblasts from primary myocytes with a sufficiently homogeneous quality and (2) a fluorescent probe that specifically targets the SR in living skeletal muscles.

In summary, we analyzed morphological and functional changes in myocytes from R2509C-RYR1 mice. In R2509C-Hom myocytes, Ca²⁺ release by depolarization or caffeine was decreased and resting [Ca²⁺]_cyt was increased, indicating a Ca²⁺ leak and a reduced SR Ca²⁺ content. In R2509C-Hom myocytes, sarcomeres were remarkably shortened to the A-band length, even in the resting condition, which was associated with abnormal Ca²⁺ homeostasis, particularly an increase in the resting [Ca²⁺]_cyt. In R2509C-Het myocytes, heat sensitivity for intracellular Ca²⁺ release via the heat-induced Ca²⁺ release mechanism was increased, and the cellular temperature was elevated in the presence of isoflurane. These findings suggest that (1) the abnormal morphological and functional changes underlie the embryonic lethality of homozygotes, and (2) environmental heat stress or exposure to an inhalation anesthetic, or both, accelerates the progression of MH in heterozygotes.

Eduardo Ríos served as editor.

We would like to thank Dr. Togo Shimozawa (The University of Tokyo) for critical reading of the manuscript, and Ms. Ikue Hiraga (Juntendo University) for her technical assistance. We thank Dr. Young-Tae Chang (Pohang University of Science and Technology, Gyeongsangbuk, Korea) for provision of ERthermAC.

This work was supported by JSPS KAKENHI Grant Numbers 19K07306, 20H04511, 21K09755 22H05055 (to T. Yamazawa), 22H05053 (to M. Suzuki), 22H05054 (to K. Oyama), 21K06789 (to F. Kobirumaki-Shimozawa), 20H03421, 21K19929 (to N. Fukuda), 19H03404, 22H02805 (to T. Murayama), 19K07105 (to N. Kurebayashi), 19H03198 (to K. Oyama, T. Yamazawa, and M. Suzuki), and 22H03198 (to N. Fukuda, F. Kobirumaki-Shimozawa, and T. Yamazawa), and the Platform Project for Supporting Drug Discovery and Life Science Research (BINDS; Basis for Supporting Innovative Drug Discovery and Life Science Research) from AMED under Grant Number JP20am0101080 (to T. Murayama and N. Kurebayashi, support No. 2533), Vehicle Racing Commemorative Foundation (#6237, 6303 to T. Murayama), and The Naito Foundation (to F. Kobirumaki-Shimozawa), and an Intramural Research Grant for Neurological and Psychiatric Disorders from the National Center of Neurology and Psychiatry (2-5 to T. Murayama).

The authors declare no competing financial interests.

Author contributions: T. Yamazawa, K. Oyama, N. Fukuda, and M. Suzuki designed the research study; Y. Tsuboi, T. Yamazawa, F. Kobirumaki-Shimozawa, T. Murayama, N. Kurebayashi, Y. Manome, E. Kikuchi, and T. Tachibana performed the experiments; S. Noguchi, T. Inoue, Y.U. Inoue, S. Mori, R. Ishida, H. Kagechika, and I. Nishino contributed new reagents/analytic tools; Y. Tsuboi, K. Oyama, M. Suzuki, N. Fukuda, and T. Yamazawa analyzed the data. All authors were involved in drafting this work.