Transient Receptor Potential Cation Channels and Calcium Dyshomeostasis in a Mouse Model Relevant to Malignant Hyperthermia

The animals used in this study were housed in pathogen-free conditions with free access to food and water and 12-h light-and-dark cycles. All experiments were undertaken with approval from the United Kingdom Home Office, the University of California at Davis Institutional Animal Care and Use Committee, or the Mount Sinai Hospital Institutional Animal Care and Use Committee. For single fiber and in vivo studies, a mixture of male and female mice between 12 and 16 weeks of age were used, and for myotube studies, myoblasts were isolated from 3- to 5-week-old neonatal mice. All experiments were terminal. For in vivo experiments, the malignant hyperthermia animals were allowed to succumb to isoflurane anesthesia when this was used; otherwise they and all wild-type animals were euthanized by cervical dislocation. For muscle collection, all animals were euthanized by cervical dislocation immediately before dissection. Because homozygous animals were viable, the animals were obtained and allocated to homozygous and heterozygous experimental groups by selective breeding of homozygous females to either homozygous or wild-type C57BL/6J males. The animals were used as available in the breeding colony when they reached the appropriate age. Studies on wild-type mice were done on 3- to 5-month-old C57BL/6J animals purchased from Charles River Laboratories (USA) or the St. James’s animal facility (United Kingdom) as needed.

The experiments were conducted at room temperature (~23°C): (1) in vitro using flexor digitorum brevis muscle fibers from 3- to 5-month-old C57BL/6J (wild-type), heterozygous RYR1-p.G2435R, and homozygous RYR1-p.G2435R mice: single fibers were obtained by enzymatic digestion as described previously²¹ ; (2) in vivo, in vastus lateralis fibers in wild-type and RYR1-p.G2435R mice anesthetized with ketamine/xylazine (100/5 mg/kg body weight) as described previously²² ; and (3) in vitro using myotubes differentiated from myoblasts from C57BL/6 (wild-type), RYR1-p.G2435R heterozygous, and RYR1-p.G2435R homozygous mice as described in detail previously.⁹ 

Double-barreled calcium and sodium selective microelectrodes were prepared as described previously.^14,22  Each ion-selective microelectrode was individually calibrated before and after the determination of intracellular calcium concentration and intracellular Na⁺ concentration as described before.²²  Only those Ca²⁺-selective microelectrodes with a linear relationship between pCa 3 and 7 (Nernstian response, 29.5 and 30.5 mV/pCa unit at 23 and 37°C, respectively) were used experimentally. The Na⁺-selective microelectrodes gave virtually Nernstian responses at free sodium concentrations between 100 and 10 mM. However, although at concentrations between 10 and 1 mM Na⁺, the electrodes had a sub-Nernstian response (40 to 45 mV), their response was of a sufficient amplitude to be able to measure intracellular sodium concentration. The sensitivity and response of the calcium- and sodium-selective microelectrodes were not directly affected by any of the drugs used in the present study.

Microelectrode recordings were performed as described previously.²²  Single isolated adult flexor digitorum brevis muscle fibers (in vitro) or vastus lateralis muscle fibers (in vivo) from wild-type and RYR1-p.G2435R heterozygous and homozygous mice were impaled with either double-barreled Ca²⁺ or double-barreled Na⁺ -selective microelectrodes, and their potentials were recorded via a high-impedance amplifier (WPI Duo 773 electrometer; WPI, USA). The potential from the 3 M potassium chloride microelectrode was subtracted electronically from the potential of either the calcium electrode or the sodium electrode to produce a differential calcium-specific potential or sodium-specific potential that represents the intracellular calcium concentration or intracellular sodium concentration, respectively. All electrode potentials were filtered (30 to 50 kHz) to improve the signal-to-noise ratio and stored in a computer for further analysis. The data from any given electrode impalement were accepted if the 3 M potassium chloride microelectrode potential was greater than or equal to −80mV, and we were able to make a stable recording of both the 3 M potassium chloride microelectrode potential and the calcium or sodium-specific potential for 60 s. These criteria resulted in rejection of data from 20% of in vivo impalements and 30 to 35% of in vitro impalements. There was no genotype effect on rejection rate.

When the Ca²⁺ indicator Fura2 is exposed to manganese, its fluorescence at its isobestic emission wavelength is quenched. Because TRPC3, TRPC6, Orai1, Orai2, Orai3, and slow voltage-gated Ca²⁺ channels are permeable to manganese as well as Ca²⁺, the rate of manganese quench can be used to determine the rate of Ca²⁺ entry into a cell.

We used manganese quench to assess sarcolemmal divalent cation entry in myotubes or isolated single adult muscle fibers loaded with 5 μM Fura2-AM for 30 min at 37°C. The solution containing the Fura2-AM was washed off, and the cells were maintained for 25 min at room temperature to allow deesterification and equilibration of the intracellular Fura2. The cells were observed through a 40× 1.3 NA objective on a Nikon Eclipse T2000 epifluorescence microscope using an excitation wavelength of 360 ± 5 nm (the isosbestic wavelength of Fura2 at which the fluorescence of the dye is independent of the Ca²⁺-free state vs. the Ca²⁺-bound state), and the emission signal was measured at 510 ± 40 nm. Images were captured with 2 × 2 binning at a rate of 5 frames/s using an intensified 12-bit digital intensified charge-coupled device (ORCA-ER) and IPLab software. Regions of interest in individual cells were analyzed using ImageJ software (National Institutes of Health), with the data exported to GraphPad Prism 7. Prism was used to fit linear regression models (y = mx + c) for the basal signal in imaging buffer and then independently in manganese-containing solutions with and without treatment drugs. The specific rate of Fura2 quenching induced by manganese entry was calculated by subtracting the basal slope of decline of fluorescence over time from the slope during the application of manganese buffer (i.e., net slope) and is expressed as arbitrary fluorescence units/s. The net slope was calculated in the absence and presence of treatment drugs. Cells with a positive net slope after the addition of manganese buffer were excluded from analysis because this indicated a technical problem with the measurement. For the same reason the slope of the quench signal was normalized to 0 when the net slope was positive after treatment with a cation channel antagonist.

Gastrocnemius muscles from all genotypes were dissected, minced, and homogenized in modified radioimmunoassay precipitation assay buffer. Total protein concentration was determined using the bicinchoninic acid method (Thermo Scientific, USA). Samples of whole gastrocnemius homogenate were prepared as described by Altamirano et al.²¹  and incubated overnight at 4°C with primary antibodies: rabbit anti–transient receptor potential cation channel 3 (TRPC3), dilution 1:2,500 (ab51560; Abcam, USA), rabbit anti-TRPC6, dilution 1:2,500 (ab62461; Abcam), human anti-actin, dilution of 1:5,000 (SC8432; Santa Cruz, USA). All of these antibodies have been validated previously by different research groups and our laboratory. The resolved bands were detected with a Storm 860 imaging system (GE Bio-Sciences, USA). The protein levels were quantified using myImageAnalysis software (Thermo Fisher Scientific, USA) and normalized to β-actin.

The mammalian Ringer solution used for experiments using muscle fibers had the following composition: 140 mM NaCl, 5 mM KCl, 2 mM CaCl₂, 1 mM MgCl₂, 5 mM glucose, and 10 mM HEPES, pH 7.4. The imaging buffer used for the experiments with myotubes contained 133 mM NaCl, 5 mM KCl, 2 mM CaCl₂, 1 mM MgCl₂, 5.5 mM glucose, and 10 mM HEPES, pH 7.4. In the manganese buffer, the CaCl₂ and MgCl₂ were substituted with 0.5 mM MnCl₂. The Ca²⁺-free solution had the following composition: 140 mM NaCl, 5 mM KCl, 2 mM MgCl₂, 1 mM EGTA, 5 mM glucose, and 10 mM HEPES, pH 7.4. 1-Oleoyl-2-acetyl-sn-glycerol, gadolinium, SAR7334, and hyperforin solutions were made by adding the desired concentration of the reagent to the physiologic or manganese-containing solutions.

Neither randomization nor blinding of the investigator performing the test was used for these studies. No specific power calculation was conducted before experimentation, but the rationale for the sample sizes used was based on our previous studies using this and other animal models of malignant hyperthermia with the same and different pharmacologic interventions. All values of intracellular ion concentrations are expressed as means ± SD, with N_micerepresenting the number of mice used to collect muscle fibers or used for the in vivo experiments and N_fibers representing the number of myotubes or muscle fibers in which measurements were carried out. Statistical analysis for studies done on muscle fibers both in vitro and in vivo was performed using one-way between-genotypes ANOVA with Tukey post hoc test for multiple measurements to determine significance. The results of Western blots were analyzed using Student’s unpaired independent t test. Normal distribution of the data was verified by histogram analysis. For studies of manganese quench in myotubes, the results were not normally distributed and therefore are expressed as the median and interquartile range. These results were analyzed using a Kruskal–Wallis test with Dunn’s multiple comparison to determine significance. P < 0.05 was accepted as the minimum confidence level for significance for all comparisons, but the P values obtained for each comparison are reported. With the exception of the myotube manganese quench studies, in which positive quench data were excluded or normalized to zero as described above, and muscle fiber data, in which the criteria for a successful measurement was not met (based on either a depolarized membrane potential or lack of stable electrode recording for 60 s), no outlying data were removed from the data collected before or after analysis. All statistical analyses were performed using GraphPad Prism 7 software.

Malignant hyperthermia is characterized by an intracellular Ca²⁺ and Na⁺ dyshomeostasis.^14,19,22,23  Consequently, intracellular calcium concentration or intracellular sodium concentration was measured in quiescent single muscle fibers isolated from wild-type, RYR1-p.G2435R heterozygous, and RYR1-p.G2435R homozygous cells using double-barreled ion-specific microelectrodes. The intracellular calcium concentration was 120 ± 2 nM (N = 8) in wild-type mice, whereas it was 156 ± 11 nM (N = 11) in RYR1-p.G2435R heterozygous and 275 ± 15 nM (N = 10) in homozygous (P < 0.001 for both; fig. 1A) mice. Likewise, intracellular sodium concentration was more elevated in RYR1-p.G2435R heterozygous and homozygous fibers than that observed in wild-type fibers. The intracellular sodium concentration was 8 ± 0.1 mM (N = 8) in wild-type fibers, whereas it was 10 ± 1 mM (N = 9) in RYR1-p.G2435R heterozygotes and 14 ± 0.6 mM (N = 11) in RYR1-p.G2435R homozygotes, respectively (P < 0.001 for both; fig. 1A). The resting membrane potentials were not different among genotypes (−82 ± 1.2 mV [N = 29] in resting wild-type muscle fibers, −82 ± 1.3mV [N = 31] in RYR1-p.G2435R heterozygous muscle fibers, and −1 ± 2.3 mV [N = 34] in RYR1-p.G2435R homozygous muscle fibers).

As previously reported, similar differences for intracellular calcium concentration and intracellular sodium concentration were obtained in the in vivo measurements carried out on the vastus lateralis from wild-type and malignant hyperthermia mice.⁹  In a new data set carried out for this study, intracellular calcium concentration in wild-type mice was 122 ± 3 nM (N = 15) compared with 144 ± 9 nM (N = 14) and 259 ± 31 nM (N = 18) in RYR1-p.G2435R heterozygous and RYR1-p.G2435R homozygous muscles, respectively (fig. 1B). Intracellular sodium concentration was 8 ± 0.1 mM (N = 16) in wild-type mice compared with 9.5 ± 1 mM (N = 16) and 12.9 ± 1.5 mM (N = 13) in RYR1-p.G2435R heterozygous and RYR1-p.G2435R homozygous muscles, respectively (P < 0.001 compared with wild-type muscles; fig. 1B). Taken together, the results demonstrate that there is a comparable intracellular calcium and sodium overload in both isolated RYR1-p.G2435R muscle cells and intact muscle fibers compared with wild-type muscle fiber and that resting intracellular ion recordings under in vitro conditions recapitulate those found in vivo.

To directly investigate the effect of diacylglycerol on intracellular calcium and intracellular sodium concentrations, single muscle fibers were exposed to 1-oleoyl-2-acetyl-sn-glycerol, a membrane-permeable diacylglycerol analog that has been shown to activate transient receptor potential cation channel subfamily C member 3 (TRPC3) and member 6 (TRPC6) channels.²⁴  Incubation of muscle fibers in 1-oleoyl-2-acetyl-sn-glycerol (100 μM) for 5 min produced an elevation of intracellular calcium concentration in all genotypes. In wild-type mice, the intracellular calcium concentration was increased by 60% (from 123 ± 3 nM [N = 11] to 196 ± 27 nM [N = 13], P < 0.001); in RYR1-p.G2435R heterozygous fibers, it was enhanced by 99% (from 153 ± 11 nM [N = 12] to 304 ± 45 nM [N = 11], P < 0.001); and in RYR1-p.G2435R homozygous fibers, it was elevated by 113% (from 251 ± 25 nM [N = 12] to 534 ± 64 nM [N = 13], P < 0.001; fig. 2A).

Because the permeability of TRPCs is cation-specific but not Ca²⁺-specific, their activation could also increase sodium entry and intracellular sodium concentration. Application of 1-oleoyl-2-acetyl-sn-glycerol (100 μM) significantly elevated intracellular sodium concentration in all genotypes, but its effect was greater in RYR1-p.G2435R muscle cells than in wild-type muscle cells (fig. 2B). In wild-type mice, incubation with 1-oleoyl-2-acetyl-sn-glycerol elevated intracellular sodium concentration by 18% (from 8 ± 0.1 mM, N = 10 to 9.4 ± 0.7 mM, N = 9, P < 0.001), in RYR1-p.G2435R heterozygous by 42% (from 10 ± 0.5 mM, N = 10 to 14.2 ± 0.7 mM, N = 9, P < 0.001), and in RYR1-p.G2435R homozygous by 50% (from 13.9 ± 0.5 mM, N = 9 to 20.9 ± 1.5 mM, N = 10, P < 0.001). These elevations of intracellular calcium and sodium concentrations induced by 1-oleoyl-2-acetyl-sn-glycerol were not affected by the voltage-gated Ca²⁺ channel inhibitor nifedipine (10 μM) nor associated with changes in resting membrane potential in either wild-type or RYR1-p.G2435R muscle fibers (data not shown).

In a different set of experiments, we explored the contribution of the Ca²⁺ influx on intracellular calcium concentration elevation elicited by 1-oleoyl-2-acetyl-sn-glycerol. Single wild-type and RYR1-p.G2435R muscle fibers were incubated for 5 min in Ca²⁺-free solution before 1-oleoyl-2-acetyl-sn-glycerol (100 μM) treatment. This exposure significantly reduced intracellular calcium concentration in all genotypes but had a greater effect in RYR1-p.G2435R mice than in wild-type mice (fig. 3). In wild-type fibers, Ca²⁺-free solution reduced intracellular calcium concentration from 123 ± 4 nM (N = 10) to 98 ± 6 nM (N = 12, P < 0.001), whereas in RYR1-p.G2435R heterozygous fibers, it was reduced from 161 ± 26 nM (N = 14) to 136 ± 6 nM (N = 12, P < 0.001), and in RYR1-p.G2435R homozygous fibers, it decreased from to 256 ± 27 nM (N = 12) to 176 ± 17 nM (N = 14, P < 0.001). In addition, the observed rise in intracellular calcium concentration elicited by 1-oleoyl-2-acetyl-sn-glycerol was completely inhibited in both wild-type and RYR1-p.G2435R muscle fibers in the absence of extracellular Ca²⁺ (fig. 3).

To gain further insight into molecular mechanisms for the increased sarcolemmal Ca²⁺ and Na⁺ entry responsible for the elevation in intracellular calcium and sodium concentrations after exposure to 1-oleoyl-2-acetyl-sn-glycerol, we measured intracellular calcium concentration and intracellular sodium concentration in isolated single wild-type and RYR1-p.G2435R muscle fibers before and after incubation in gadolinium and then in the presence of gadolinium, after exposure to 1-oleoyl-2-acetyl-sn-glycerol (100 μM). Pretreatment with 25 μM gadolinium significantly lowered intracellular calcium concentration and intracellular sodium concentration by 15% and by 8%, respectively, in wild-type fibers; by 20% by 15%, respectively, in RYR1-p.G2435R heterozygous fibers; and by 27% and 29%, respectively, in RYR1-p.G2435R homozygous fibers (figs. 4, A and B). Furthermore, gadolinium pretreatment prevented any significant increase in intracellular calcium and sodium concentrations during exposure to 1-oleoyl-2-acetyl-sn-glycerol in all genotypes (figs. 4, A and B).

Because gadolinium is a nonspecific sarcolemmal cation blocker, SAR7334, a newer more specific blocker of TRPC6 and TRPC3 channels,²⁵  was used to further characterize the mechanism for the 1-oleoyl-2-acetyl-sn-glycerol–mediated increase in intracellular calcium concentration. Single isolated wild-type and RYR1-p.G2435R muscle fibers were incubated in 1 μM SAR7334 and then exposed to 1-oleoyl-2-acetyl-sn-glycerol (100 μM). Pretreatment with SAR7334 significantly reduced the resting intracellular calcium concentration by 14% in wild-type fibers, by 19% in RYR1-p.G2435R heterozygous fibers, and by 34% in RYR1-p.G2435R homozygous fibers (fig. 5A). Furthermore, SAR7334 also prevented the 1-oleoyl-2-acetyl-sn-glycerol–mediated elevation of intracellular calcium concentration in all three genotypes (fig. 5A).

Having observed inhibition of the 1-oleoyl-2-acetyl-sn-glycerol–induced increase in intracellular calcium concentration by SAR7334 in wild-type and RYR1-p.G2435R single muscle fibers, we then tested the activity of this TRPC3/6 channel blocker during a malignant hyperthermia episode. We therefore investigated the effect of SAR7334 on intracellular calcium concentration in malignant hyperthermia muscle in vivo during exposure to 2% halothane. The intracellular calcium concentration was measured simultaneously in the right and left vastus lateralis muscles in wild-type and homozygous RYR1-p.G2435R mice before and after the addition of 2% halothane to their inspired gas (fig. 5B). The right vastus lateralis muscle was pretreated locally with 1 μM SAR7334, and the left muscle served as the control. Figure 5B shows a typical in vivo experiment carried out in a homozygous RYR1-p.G2435R mouse. The data are reported as nanomolar (nM). Before drug treatment, the intracellular calcium concentration was very similar in muscles from both legs (265 ± 9 nM in the left leg vs. 275 ± 9 nM in the right leg). Ten minutes after the application of 1 μM SAR7334 on the exposed vastus lateralis fibers of the right leg, the intracellular calcium concentration was reduced from 265 ± 9 to 187 ± 11 nM (fig. 5B). Inhalation of halothane elicited an elevation of intracellular calcium concentration in the untreated leg to 1,373 ± 81 nM, whereas in the SAR7334-treated leg, the initial increase in intracellular calcium concentration was reduced by 62% (526 ± 99 nM), after which intracellular calcium concentration declined to 453 ± 32nM (fig. 5B). After 20 min of inhalation of halothane, the rectal temperature reached 42°C, and the mouse died.

Hyperforin is known to elevate the intracellular concentration of Ca²⁺ through the activation of di-acyl-glycerol–sensitive TRPC6 channels without activating the other transient receptor potential cation channel isoforms (TRPC1, TRPC3, TRPC4, TRPC5, and TRPC7).²⁶  Hyperforin (5 μM) caused a prominent sustained rise in intracellular calcium concentration that was greater in RYR1-p.G2435R fibers (homozygous greater than heterozygous) than wild-type muscle fibers (fig. 6A). In wild-type muscle fibers, intracellular calcium concentration was enhanced from 124 ± 3 nM (N = 9) to 151 ± 10 nM (N = 10) upon incubation with hyperforin, whereas in RYR1-p.G2435R heterozygous fibers, it increased from 164 ± 22 nM (N = 10) to 234 ± 31 nM (N = 10), and in RYR1-p.G2435R homozygous fibers, it was elevated from 269 ± 37 nM (N = 10) to 441 ± 41 nM (N = 10; fig. 6A). Removal of extracellular Ca²⁺ reduced pre-exposure intracellular calcium concentration and blocked the hyperforin effect in all genotypes (fig. 6C). Similar to its effect on intracellular calcium concentration, hyperforin elevated intracellular sodium concentration in wild-type fibers from 8 ± 0.1 mM (N = 15) to 9.5 ± 0.6 mM (N = 10; P < 0.001), in RYR1-p.G2435R heterozygous fibers from 10 ± 0.6 mM (N = 15) to 13.5 ± 1 (N = 10; P < 0.001), and in RYR1-p.G2435R homozygous fibers from 14.5 ± 1.2 mM (N = 11) to 20.3 ± 1.6 mM (N = 13; P < 0.001; fig. 6B).

When the buffer was changed from imaging buffer to manganese quench buffer, baseline quench of Fura2 fluorescence was observed in all three genotypes (fig. 7A for representative traces reported as arbitrary fluorescence units) with the median quench rate being greater in malignant hyperthermia–homozygous (−0.23 arbitrary fluorescence units/s; interquartile range, −0.13 to −0.38 arbitrary fluorescence units/s) versus malignant hyperthermia–heterozygous (−0.18 arbitrary fluorescence units/s; interquartile range, −0.11 to −0.32 arbitrary fluorescence units/s) and wild-type mice (−0.19 arbitrary fluorescence units/s; interquartile range, −0.12 to −0.32 arbitrary fluorescence units/s; fig. 7B). Qualitatively similar results were observed in isolated adult single muscle fibers (data not shown).

We next determined the effect of sarcolemmal cation blockers 25 μM gadolinium (a nonspecific sarcolemmal cation entry blocker) and 250 nM SAR7334 (a transient receptor potential cation channel specific blocker) on the rate of manganese quench of the Fura2 fluorescence signal. Treatment with 25 μM gadolinium profoundly blocked manganese quench to near baseline levels in all three genotypes (P < 0.0001, N = 65 to 96; fig. 7B). The reduction in the median quench rate was 88% in wild-type myotubes (N = 65), 100% in heterozygous RYR1-p.G2435R myotubes (N = 72), and 95.6% in homozygous RYR1-p.G2435R myotubes (N = 96; fig. 7B). After treatment with 250 nM SAR7334 in wild-type myotubes, the change in median quench rate was not significant (8%, P > 0.999, N = 49), whereas in heterozygous RYR1-p.G2435R myotubes and in homozygous RYR1-p.G2435R myotubes, there was a significant decrease (45%, P < 0.005, N = 53 and 69%, P < 0.001, N = 35 for heterozygous and homozygous RYR1-p.G2435R myotubes, respectively) in the median manganese quench rate (fig. 7B).

Western blot analysis showed that TRPC3 and TRPC6 were significantly increased in a gene–dose–dependent manner in RYR1-p.G2435R muscles (heterozygous [P < 0.01] and homozygous [P < 0.001]) compared with wild-type mice. Figure 8A shows representative Western blot analysis of the expression of TRPC3 and TRPC6 in wild-type, heterozygous RYR1-p.G2435R, and homozygous RYR1-p.G2435R muscle; β-actin was used as a loading control. Densitometric analysis of five independent Western blots for each transient receptor potential cation channel protein is shown in figure 8B.

The major findings of the present study are as follows.

• (1) We confirmed our previous finding⁹  that RYR1-p.G2435R muscle cells have chronically elevated intracellular resting calcium and sodium concentrations (greater in homozygous than in heterozygous mice) and now show that this is associated with elevated manganese quench indicating increased sarcolemmal cation influx.

• (2) Compared with wild-type muscle cells, incubation of RYR1-p.G2435R muscle cells with 1-oleoyl-2-acetyl-sn-glycerol or the TRPC6 activator hyperforin significantly increased intracellular calcium and sodium concentrations (greater in homozygous than heterozygous muscle cells).

• (3) Increases in intracellular calcium and sodium concentrations induced by 1-oleoyl-2-acetyl-sn-glycerol are dependent on extracellular Ca²⁺ and are blocked by the nonspecific sarcolemmal cation channel blocker gadolinium, as well as the more specific TRPC3 and TRPC6 channel blocker SAR7334.

• (4) SAR7334 treatment also significantly reduced resting intracellular calcium and sodium concentrations, blocked manganese quench, and reduced the elevation of intracellular calcium in RYR1-p.G2435R fibers after exposure to halothane.

• (5) The observed changes in the above RYR1-p.G2434R responses were accompanied by an increase in the protein expression of TRPC3 and TRPC6 in RYR1-p.G2435R (homozygous greater than heterozygous) skeletal muscles compared with wild-type muscles.

For some time, it has been thought that the mechanism behind the increased resting intracellular calcium in malignant hyperthermia susceptible muscle is simply due to sarcoplasmic reticulum leak and that the massive increase in muscle intracellular calcium during an malignant hyperthermia episode comes from and is maintained by sarcoplasmic reticulum Ca²⁺ stores. Based on our studies in other animal models relevant to malignant hyperthermia, we have previously suggested that sarcolemmal cation entry channels are, at the very least, significant contributors to the chronically elevated intracellular calcium in quiescent malignant hyperthermia muscles and myotubes^21,22  and that they play a similarly significant role during a fulminant malignant hyperthermia episode.²² 

A primary finding was that in addition to increased resting intracellular calcium, resting intracellular sodium is also elevated in RYR1-p.G2435R muscle cells (homozygous greater than heterozygous) compared with wild-type muscle cells, both in vitro and in vivo.^21,22  This suggests that at least one source of Ca²⁺ could be the extracellular pool and not the sarcoplasmic reticulum. In this study, we confirmed this hypothesis, showing that incubation of muscle with gadolinium, a potent nonspecific inhibitor of sarcolemmal cation influx, significantly reduced the intracellular levels of sodium and calcium and blocked manganese quench, a monitor of sarcolemmal cation entry.

Di-acyl-glycerol is a second messenger involved in key cellular processes in skeletal muscle and other excitable cells.²⁷  In addition to its role as a lipid second messenger to recruit protein kinases C, it also activates the di-acyl-glycerol–sensitive TRPC3/6 subgroup of transient receptor potential cation channels in the sarcolemma.^24,28  Here we utilized the membrane-permeable di-acyl-glycerol analog, 1-oleoyl-2-acetyl-sn-glycerol to determine whether the pathway for sarcolemmal sodium and calcium influx was through TRPC3 and TRPC6.^29,30  1-Oleoyl-2-acetyl-sn-glycerol induced an elevation in the resting intracellular calcium and sodium in all three genotypes, and this was greater in RYR1-p.G2435R (greater in homozygous than heterozygous) muscle than wild-type muscle (figs. 2, A and B). The elevation of intracellular calcium and the increase rate of manganese quench elicited by 1-oleoyl-2-acetyl-sn-glycerol was blocked either by removal of extracellular Ca²⁺ or by blocking sarcolemmal Ca²⁺ and Na⁺ entry with gadolinium or SAR7334 (figs. 3, 4, and 6). These data rule out any possibility that 1-oleoyl-2-acetyl-sn-glycerol increased intracellular calcium by releasing Ca²⁺ from intracellular stores and confirmed that the source of the increase was through TRPC3/6, whose expression we showed was upregulated in RYR1-p.G2435R compared with wild-type muscles.

Incubation of wild-type and RYR1-p.G2435R muscle fibers with SAR7334, a much more specific blocker of TRPC6 and TRPC3 channels than gadolinium,²⁵  reduced resting calcium in RYR1-p.G2435R cells (in homozygous cells less than heterozygous cells, which have less than wild-type cells) and blocked the effects of 1-oleoyl-2-acetyl-sn-glycerol. Local application of SAR7334 directly to the superficial muscle fibers in vivo reduced resting intracellular calcium in homozygous RYR1-p.G2435R fibers and partially inhibited the halothane-induced increase in intracellular calcium in RYR1-p.G2435R muscle cells. This finding in particular demonstrates that during the malignant hyperthermia episode, there is an initial sarcoplasmic reticulum–supported increase in intracellular calcium that appears to be quickly partially depleted, as demonstrated by the rapid fall in intracellular calcium from ~750 to ~400 nM in the SAR7334-treated leg during the same time intracellular calcium rose to 1,373 ± 81 nM and then plateaus in the nontreated leg. All of these data together support the hypothesis that sarcolemmal Ca²⁺ entry plays a critical role in maintaining the new intracellular calcium steady state that has been observed during a malignant hyperthermia crisis. Furthermore, the observed immediate elevation of intracellular calcium in the SAR7334-treated leg is most probably mediated by Ca²⁺ release from the sarcoplasmic reticulum, and the decline in the intracellular calcium in this leg shows that that this level cannot be maintained by sarcoplasmic reticulum Ca²⁺ cycling because of Ca²⁺ extrusion and blockade of sarcolemmal Ca²⁺ entry.

Consistent with the 1-oleoyl-2-acetyl-sn-glycerol-mediated elevation of intracellular calcium and sodium in muscle cells plus the ability of the transient receptor potential cation channel-specific blocker SAR7334 to block both the pharmacologic and halothane-induced increase in Ca²⁺, we also demonstrated that hyperforin, an activator of TRPC6 mediated inward cationic current,^26,31,32  mimicked the action of 1-oleoyl-2-acetyl-sn-glycerol, and its effect was greater in RYR1-p.G2435R (greater in homozygous than in heterozygous mice) than wild-type muscle cells. As observed with 1-oleoyl-2-acetyl-sn-glycerol, removal of extracellular Ca²⁺ also eliminated the hyperforin-elicited increase in resting intracellular calcium.³¹  The enhanced cation flux observed across the sarcolemma in RYR1-p.G2435R cells is consistent with the cell boundary theorem stating that the long-term steady state of the free intracellular calcium concentration inside cells must be driven by changes with the outside through fluxes across the plasmalemma.^22,33 

In summary, these results demonstrated that skeletal muscles of mice expressing the RYR1-p.G2435R mutation have an elevated resting intracellular calcium and sodium and a significantly enhanced, extracellular Ca²⁺ dependent response to 1-oleoyl-2-acetyl-sn-glycerol and hyperforin. We further showed that the 1-oleoyl-2-acetyl-sn-glycerol response could be prevented by TRPC3 and TRPC6 blockers. These results provide further support to our working hypothesis that in response to partial depletion of sarcoplasmic reticulum calcium stores caused by malignant hyperthermia causing mutations in RyR1, TRPC3 and TRPC6 channels play a critical role in producing the intracellular calcium and sodium overload both at rest and during the malignant hyperthermia crisis.

Supported by National Institute of Arthritis, Musculoskeletal and Skin Diseases grant Nos. 1R01AR068897-01A1 (Bethesda, Maryland; to Drs. Allen, Hopkins, Rafael Lopez, and Liu) and 2P01 AR-05235 (to Drs. Allen, Rafael Lopez, and Hopkins), by Medical Research Council grant No. N002407/1 (London, United Kingdom; to Drs. Kaura, Hopkins, and Allen), and AFM-Téléthon-France grant No. 21543 (Evry, France; to Dr. Rafael Lopez).

The authors declare no competing interests.