Ca2+ Influx via the Na+/Ca2+ Exchanger Is Enhanced in Malignant Hyperthermia Skeletal Muscle*

Background: Dysregulation of Ca2+ homeostasis have been described in malignant hyperthermia (MH). Results: Na+/Ca2+ exchanger (NCX3) reverse mode activity is enhanced in MH muscles and it contributes to resting intracellular calcium concentration and Ca2+ transients induced by high [K+]e and by halothane. Conclusion: NCX3 reverse mode activity is increased in MH muscle. Significance: Understanding mechanisms influencing Ca2+ dynamics in MH muscle. Malignant hyperthermia (MH) is potentially fatal pharmacogenetic disorder of skeletal muscle caused by intracellular Ca2+ dysregulation. NCX is a bidirectional transporter that effluxes (forward mode) or influxes (reverse mode) Ca2+ depending on cellular activity. Resting intracellular calcium ([Ca2+]r) and sodium ([Na+]r) concentrations are elevated in MH susceptible (MHS) swine and murine muscles compared with their normal (MHN) counterparts, although the contribution of NCX is unclear. Lowering [Na+]e elevates [Ca2+]r in both MHN and MHS swine muscle fibers and it is prevented by removal of extracellular Ca2+ or reduced by t-tubule disruption, in both genotypes. KB-R7943, a nonselective NCX3 blocker, reduced [Ca2+]r in both swine and murine MHN and MHS muscle fibers at rest and decreased the magnitude of the elevation of [Ca2+]r observed in MHS fibers after exposure to halothane. YM-244769, a high affinity reverse mode NCX3 blocker, reduces [Ca2+]r in MHS muscle fibers and decreases the amplitude of [Ca2+]r rise triggered by halothane, but had no effect on [Ca2+]r in MHN muscle. In addition, YM-244769 reduced the peak and area under the curve of the Ca2+ transient elicited by high [K+]e and increased its rate of decay in MHS muscle fibers. siRNA knockdown of NCX3 in MHS myotubes reduced [Ca2+]r and the Ca2+ transient area induced by high [K+]e. These results demonstrate a functional NCX3 in skeletal muscle whose activity is enhanced in MHS. Moreover reverse mode NCX3 contributes to the Ca2+ transients associated with K+-induced depolarization and the halothane-triggered MH episode in MHS muscle fibers.


Malignant hyperthermia (MH) is potentially fatal pharmacogenetic disorder of skeletal muscle caused by intracellular Ca 2؉ dysregulation. NCX is a bidirectional transporter that effluxes (forward mode) or influxes (reverse mode) Ca 2؉ depending on cellular activity. Resting intracellular calcium ([Ca 2؉ ] r ) and sodium ([Na
] e . These results demonstrate a functional NCX3 in skeletal muscle whose activity is enhanced in MHS. Moreover reverse mode NCX3 contributes to the Ca 2؉ transients associated with K ؉ -induced depolarization and the halothane-triggered MH episode in MHS muscle fibers. The plasma membrane Na ϩ /Ca 2ϩ exchanger (NCX) 3 is a bidirectional electrogenic (3 Na ϩ /1Ca 2ϩ ), and voltage-sensitive reversible ion countertransporter, which is mainly responsible for Ca 2ϩ extrusion in a variety of cells (1,2). The activity and direction of NCX transport is driven by the membrane potential (V m ) as well as the transmembrane electrochemical gradients of Na ϩ and Ca 2ϩ (2). This transporter operates as a high capacity and low affinity system for Ca 2ϩ transport, extruding Ca 2ϩ against its transmembrane electrochemical gradient coupled to Na ϩ influx (forward mode), or transporting Ca 2ϩ into cells coupled to Na ϩ efflux (reverse mode) (2).
Three mammalian isoforms of the NCX protein have been discovered (NCX1, NCX2, and NCX3) (3)(4)(5). Only the NCX1 and NCX3 isoforms are expressed in skeletal muscle and are localized in the transverse tubules (t-tubules) and in the sarcolemma (4,6,7). NCX1 is expressed at high levels during embryonic development and postnatal maturation (8) then its expression is gradually reduced as the NCX3 gene becomes more highly expressed and predominates in adult skeletal muscle (8).
Na ϩ -dependent Ca 2ϩ fluxes have been reported in amphibian skeletal muscle (9,10), and human myotubes (7). NCX has been observed in sarcolemmal fractions isolated from mammalian muscles (11) and in t-tubules isolated from amphibian muscles (12,13). In addition, the reverse mode of the exchanger seems to mediate the enhancement of contraction that takes place after external Na ϩ withdrawal in phasic and tonic amphibian skeletal muscle fibers (14 -17). However, the role of the NCX in skeletal muscle is still controversial and its role in the MHS skeletal muscle fibers is totally unknown.
MH is a life-threatening pharmacogenetic syndrome, which occurs when susceptible individuals are exposed to triggering agents, such as halogenated inhalation anesthetics and/or depolarizing muscle relaxants (18). MH susceptibility is associated with a dysfunction of regulation of intracellular resting Ca 2ϩ concentration ([Ca 2ϩ ] r ) (19 -21) and is characterized by the occurrence of a robust increase in [Ca 2ϩ ] r in response to exposure to halothane (19,22,23). In more than 70% of humans with MH susceptibility (MHS) the trait is linked to one of Ͼ185 mutations within RYR1 located on ch19q13.1, the gene that encodes the type 1 skeletal isoform of the ryanodine receptor (RYR1), the primary Ca 2ϩ release channel in the sarcoplasmic reticulum (SR) (24). In addition, mutations in CACNA1S on ch1q31-32, which codes for Ca V 1.1 have also been found in 2% of MHS individuals (25)(26)(27).
In the present study, we assessed the NCX activity in skeletal muscle cells from MHN and MHS swine and mice by measuring [Ca 2ϩ ] r and [Na ϩ ] r in response to external sodium withdrawal. In addition, we studied the contribution of NCX on the Ca 2ϩ transient associated with sustained membrane depolarization induced by high [K ϩ ] e and during an in vivo MH episode elicited by halothane in murine MHS muscle fibers. In doing so, we demonstrated that the activity of NCX3 is enhanced in MHS muscle cells and that its reverse mode contributes to the Ca 2ϩ transients elicited by K ϩ depolarization and halothane-induced elevation on intracellular Ca 2ϩ in murine muscle fibers.

EXPERIMENTAL PROCEDURES
Biological Preparation-Experiments were conducted: (i) in vitro, using intact intercostal muscle biopsies obtained from 16 MHS Poland China (RYR1 R615C-Homozygous) and 8 MHN Yorkshire swine. MH susceptibility was determined by PCR amplification of RYR1 exon 10 in genomic DNA and halothane challenge as previously described (28). Muscle biopsy specimens were removed under general anesthesia using non-triggering agents (thiopental 15 mg/kg for induction, NO 2 /O 2 /fentanyl to anesthetized the animal). After surgical removal, intact muscle biopsy samples were placed in a temperature-controlled dish (37°C) bubbled with a mixture of 95% O 2 and 5% CO 2 , then dissected free of adipose and connective tissue with the aid of a stereo microscope. They were further divided into several small bundles of four to six muscle fibers for the Ca 2ϩ measurements. (ii) In vitro, Flexor digitorum brevis (FDB) muscle fibers from 3-4-month-old C57BL/6 MHN and congenic heterozygous RYR1 R163C C57BL/6 MHS mice (29) were obtained by enzymatic digestion. FDB muscles were incubated with 400 units/ml of collagenase type 2 (Worthington, Lakewood, NJ) in DMEM supplemented with penicillin/streptomycin (1 time) for 90 min at 37°C and 5% CO 2 . Muscles were then washed in DMEM and single fibers were mechanically dissociated with progressively smaller fire-polished Pasteur pipettes. Isolated fibers were seeded in ECM-coated 96-well plates in DMEM supplemented with 10% fetal bovine serum and penicillin/ streptomycin (1 time) and incubated at 37°C and 5% CO 2 . Fibers were used for experimentation 20 -24 h after isolation. (iii) In vivo, in vastus lateralis fibers in anesthetized MHN and MHS mice as described previously (18). The institutional animal care committees at the Instituto Venezolano de Investiga-ciones Científicas, the Harvard Medical Area, and the University of California, Davis, approved all experimental protocols.
Ca 2ϩ and Na ϩ Selective Microelectrodes-Double-barreled Ca 2ϩ selective and Na ϩ selective microelectrodes were prepared as described previously (19). The calcium ionophore II, ETH-129 (Fluka, Sigma), and the Na ϩ ionophore I, ETH-227 (Fluka, Sigma), were used to back-fill the Ca 2ϩ and Na ϩ -selective microelectrodes, respectively. Each ion-selective microelectrode was individually calibrated as described previously (25,30). After making measurements of [Ca 2ϩ ] r and [Na ϩ ] r all microelectrodes were recalibrated, and if the two calibration curves did not agree within 3 mV, data from that microelectrode were discarded. The drugs used in the present study did not interfere with Ca 2ϩ microelectrode sensitivity in the relevant ranges studied (from pCa6 to pCa7 and from 1 to 30 mM Na ϩ ).
Recording of V m , [Ca 2ϩ ] r , and [Na ϩ ] r in Vitro-Measurements of V m , [Ca 2ϩ ] r , and [Na ϩ ] r , were performed on intact intercostal muscle bundles obtained from MHN and MHS swine. The intact muscle bundles were mounted horizontally in a Plexiglas temperature-controlled chamber (International Plastic, Miami, FL) (36 Ϯ 1.5°C) that was placed on the stage of an upright microscope. Each tendon was fastened to a stainless steel hook connected to micromanipulators and the sarcomere length was adjusted to 2.2 m using a laser diffraction technique (31). The muscle preparation was perfused continuously with bicarbonate-buffered swine physiologic solution (see below for composition), bubbled with a mixture of 95% O 2 and 5% CO 2 , pH 7.4. Individual muscle fibers were impaled with either a double-barreled Ca 2ϩ or Na ϩ selective microelectrodes and V m , V CaE , and V NaE potentials were recorded via high impedance amplifier (WPI FD-223 or WPI Duo 773 electrometer; WPI, Sarasota, FL). The potential from the V m barrel (3 M KCl) was subtracted electronically from V CaE or V NaE , to produce a differential Ca 2ϩ -specific (V Ca ) or Na ϩ specific (V Na ) potential that represents [Ca 2ϩ ] r or [Na ϩ ] r . V m , V Ca , and V Na were filtered with a low-pass filter  to improve the signal-to-noise ratio and stored in a computer for further analysis.
Recording of V m , [Ca 2ϩ ] r , and [Na ϩ ] r in Vivo-[Ca 2ϩ ] r and [Na ϩ ] r were measured in vivo in MHN and MHS mice as described previously (19). In brief, mice were anesthetized with 100 mg/kg of ketamine and 5 mg/kg of xylazine, intubated, and ventilated with air using a mouse ventilator (Harvard Minivent, M-845, Holliston, MA). The mice were kept euthermic (37-37.5°C) using of a heating pad with an automated feedback loop (ATC-1000 temperature controller, WPI, Sarasota, FL). The vastus lateralis was then exposed surgically and determinations of [Ca 2ϩ ] r or [Na ϩ ] r were made in its superficial fibers. Several muscle fibers in each leg were impaled to measure ion concentrations for each condition in each leg (control condition, after test drug application and after halothane exposure). Halothane (1.5%) was administered using a vaporizer (Kent Scientific Corporation, Torrington, CT) connected to the ventilator in such a way as to prevent backpressure on the vaporizer. The procedure to record V m , V CaE , and V NaE and to obtain and process the specific potential for Ca 2ϩ (V Ca ) or Na ϩ (V Na ) for later analysis was identical to that described above for in vitro experiments.
Glycerol Treatment-It is well established that glycerol treatment can be used to osmotically disrupt the transverse tubules in skeletal muscle (32,33). Briefly, intact MHN and MHS swine muscle fibers were incubated in 350 mM glycerol solution for 1 h, and then returned to normal swine physiologic solution (32). Destruction of the transverse tubular system occurs not in the glycerol solution itself, but occurs only after the muscle fibers have been returned to normal physiologic solution due to the osmotic shock produced by glycerol withdrawal. The osmotic shock produces a physical and functional dissociation of the t-tubules from the surface membrane (detubulation) (32,33).
Ca 2ϩ Imaging-Isolated FDB fibers were loaded in a solution of Fluo4-AM (50 g vial) dissolved in 20% pluronic acid in DMSO (15 l), and diluted in murine physiological solution to a final concentration of 10 M. Dye loading was done at room temperature for 30 min and muscle cells were then washed once with normal murine physiological solution and then transferred to the stage of an IX71 inverted microscope. To prevent motion artifact due to muscle contraction during K ϩ depolarization the perfused solution was supplemented with 10 M N-benzyl-p-toluene sulfonamide (Sigma).
The experimental protocol used was the following: single muscle fibers were perfused with normal potassium (5 mM K ϩ ) solution for 1 min, then with high potassium (60 mM K ϩ ) solution for 3 min, and then returned to normal (5 mM K ϩ ) solution for another 1 min. To study the contribution of NCX3 reverse mode on K ϩ depolarization-induced Ca 2ϩ transients, muscle fibers were preincubated with 1 M YM-244769 for 5 min and then used at the same depolarization protocol as described above (1 M YM-244769 was added to all the solutions).
For measurements of Ca 2ϩ transients, fibers were illuminated at 494 nm to excite Fluo-4 with a DeltaRam wavelengthselectable light source. Fluorescence emission at 510 nm was captured from individual muscle fibers at 20 frames/s using a Cascade Evolve 512 camera (Photometrics, Tucson, AZ). The images were acquired using the Easy Ratio Pro software (PTI). Fluorescence data (F) were normalized to the fluorescence base line (F o ) of each individual fiber and expressed as (F Ϫ F 0 )/F 0 or ⌬F/F 0 . The data were analyzed using Prism 6 software (GraphPad Software, Inc.). Peak amplitude and area under the curve within the evoked responses was calculated from the number of fibers indicated in the bar graphs. Rates of decay (k) were calculated adjusting the curve to a single exponential decay function (y(t) ϭ y 0 ⅐ e Ϫkt ).
Myotubes Culture and Transfection-Primary myoblasts were isolated from the hindlimb and forelimb muscles of neonatal MHS (R163C Het) mice and their MHN (C57BL/6) littermates (34). Myoblasts were differentiated for 3 days into myotubes by withdrawal of growth factors as described previously (34). Myoblasts were placed in DMEM low glucose 5% horse serum and 1ϫ penicillin/streptomycin/glutamine for 1 day and the transfected with either scramble FITC siRNA or NCX3 siRNA (a pool of 3 target-specific 19 -25-nucleotide siRNAs, Santa Cruz Biotechnology) and scramble FITC siRNA (25 nM) with DharmaFECT 1 transfection reagent (Thermo Scientific) for 1 day at 37°C in DMEM low glucose 5% horse serum without antibiotics according to the manufacturer's instructions. Media was changed after 1 day (with antibiotics) and myotubes were differentiated for another day (total 3 days of differentiation). Myotubes were then loaded with 5 M Fura2-AM for 30 min at room temperature and calcium transients were assessed with high potassium solution as described above. FITC fluorescence channel was used to select transfected myotubes to carry out the experiments. [Ca 2ϩ ] r was assessed by Ca 2ϩ selective microelectrodes as described above. Total myotubes lysates were analyzed by standard Western blot technique 48 h after transfection (see below).
Measurement of SR Ca 2ϩ Loading in MHN and MHS Adult Fibers and Myotubes-Adult FDB fibers were loaded with Fluo-4AM and incubated with N-benzyl-p-toluene sulfonamide to avoid contraction artifacts. SR Ca 2ϩ loading was measured using the area under the curve following a 30-s exposure to 20 mM caffeine in Ca 2ϩ -free media (ϩ1 mM EGTA). Myotubes were loaded with Fluo-4AM. SR Ca 2ϩ loading was evaluated by measuring the area under the curve of the Ca 2ϩ transient evoked by a 60-s exposure to 5 M ionomycin in Ca 2ϩ -free media (ϩ1 mM EGTA). To test the effect of YM-244769 on SR Ca 2ϩ loading both adult FDB fibers and myotubes were incubated for 5 min with 1 M YM-244769 and this concentration was maintained constant during exposure to either caffeine or ionomycin.
Western Blot-Quadriceps muscles were dissected and minced to smaller pieces with a pair of scissors and then homogenized with an electric homogenizer in modified RIPA buffer (150 mM NaCl, 50 mM Tris, pH 7.4, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 5 mM EDTA, 2 mM EGTA, 1ϫ Roche Complete protease inhibitor). Myotubes were scraped in modified RIPA buffer. Lysates were incubated for 30 min on ice and then spun down by centrifugation at 16,000 ϫ g for 20 min at 4°C. Protein concentrations were determined using the BCA method (Thermo Scientific). Proteins (50 g) were separated using SDS-PAGE 4 -15% gradient gels (Bio-Rad) and transferred to PVDF membranes. The PVDF membranes were blocked with SEA Blocking Buffer, 0.1% Tween 20 (Thermo Scientific), incubated overnight at 4°C with NCX3 (Alpha Diagnostic International, 1:1000) and GAPDH (Santa Cruz Biotechnology, 1:5000) antibodies in blocking buffer, and then washed with PBS, 0.1% Tween 20 (PBS-T). After washing they were incubated with the secondary antibodies in blocking buffer (IRDye 680 and 800 nm, Li-COR Biosciences) for 1 h, then washed with PBS-T and quantified with Odyssey Imaging System (Li-COR Biosciences).
Equilibrium Potential Calculations-Equilibrium potentials were calculated by Nernst equation (Equation 1), using the results obtained in this work, as follows.
Solutions-Swine physiologic solution had the following composition (in mM): 130 NaCl, 5 KCl, 2.5 CaCl 2 , 1 MgCl 2 , 18 NaHCO 3 , 1 NaH 2 PO 4 , and 5 glucose. The pH was 7.4 when it was aerated with a mixture of 5% CO 2 and 95% O 2 . Low or Na ϩ -free solution was prepared by partial or total withdrawal of [Na ϩ ] e , which was replaced by an equivalent amount by N-methyl-D-glucamine (NMG) hydrochloride to maintain osmolarity. Ca 2ϩ -free solution was prepared by omitting the Statistical Analysis-All values are expressed as mean Ϯ S.E. and n is equivalent to the number of independent measurements. Statistical analysis was performed using unpaired t test, or one-way analysis coupled with Tukey's t test for multiple measurements to determine significance (p Ͻ 0.05).
Removal of Extracellular Ca 2ϩ Prevents [Ca 2ϩ ] r Increases Induced by Na ϩ Depletion-In a different set of experiments we explored the contribution of the Ca 2ϩ influx on [Ca 2ϩ ] r elevation elicited by Na ϩ withdrawal. MHN and MHS swine muscle fibers were incubated for 5 min prior to Na ϩ withdrawal in Ca 2ϩ -free solution. This exposure significantly reduced [Ca 2ϩ ] r in both genotypes, but had a greater effect in MHS than MHN muscles (Fig. 2). In MHN fibers Ca 2ϩ -free solution reduced [Ca 2ϩ ] r from 116 Ϯ 1 to 90 Ϯ 1 nM and in MHS it was reduced from 348 Ϯ 3 to 146 Ϯ 2 nM. In the absence of extracellular Ca 2ϩ the observed raise in [Ca 2ϩ ] r elicited by Na ϩ withdrawal was completely inhibited in both MHN and MHS muscle fibers (88 Ϯ 1 nM in MHN and 148 Ϯ 1 nM in MHS muscle fibers) (Fig. 2).
T-tubule Disruption Abrogates [Ca 2ϩ ] r Increases Induced by Na ϩ Withdrawal-To determine the fraction of NCX activity that is restricted to the t-tubules, MHN and MHS muscle fibers  were detubulated by a transient osmotic shock with glycerol.
[Ca 2ϩ ] r was measured during and after hypertonic glycerol treatment and then again after exposure of the detubulated muscle fibers to Na ϩ -free solution. After glycerol treatment and return to normal swine physiologic solution, [Ca 2ϩ ] r in both MHN and MHS fibers rose slightly (Fig. 3). Interestingly, in MHN-glycerol-treated fibers the elevation of [Ca 2ϩ ] r elicited by Na ϩ withdrawal was reduced to 146 Ϯ 2 nM (382 Ϯ 5 nM, non-treated) and in MHS-glycerol-treated fibers was reduced to 589 Ϯ 9 nM (1,836 Ϯ 8 nM, non-treated) (Fig. 3).
Effects KB-R7943 on [Ca 2ϩ ] r in Swine Skeletal Muscle-To clarify mechanisms involved in causing elevation of [Ca 2ϩ ] r elicited by exposure to Na ϩ -free media, the effect of KB-R7943 was studied in MHN and MHS muscle fibers. KB-R7943 acts as a nonspecific blocker of the reverse mode of the NCX3 by modifying the Na ϩ -dependent binding (35,36). Preincubation of MHN muscle fibers with 10 M KB-R7943 reduced [Ca 2ϩ ] r from 118 Ϯ 1 to 99 Ϯ 1 nM, and attenuated the increase in [Ca 2ϩ ] r associated with exposure to Na ϩ -free solution (142 Ϯ 2 versus 382 Ϯ 5 nM, in the absence of KB-R7943) (Fig. 4). Similarly, KB-R7943 pretreatment of MHS fibers lowered [Ca 2ϩ ] r from 370 Ϯ 5 to 234 Ϯ 3 nM and diminished the increase in [Ca 2ϩ ] r associated with exposure to Na ϩ -free solution from 1,836 Ϯ 8 (non-KB-R7943 treated) to 533 Ϯ 11 nM (Fig. 4).

Effects of KB-R7943 and YM-244769 on [Ca 2ϩ ] r in Murine Muscles in Vivo-
To determine whether blockade of reverse mode NCX3 with KB-R7943 would modify [Ca 2ϩ ] r in murine MHN and MHS quiescent muscle fibers at rest and during halothane exposure, KB-R7943 (10 M, in murine physiologic solution) was applied locally to superficial fibers of the vastus lateralis muscle for 5 min and then [Ca 2ϩ ] r was determined. Incubation with KB-R7943 reduced [Ca 2ϩ ] r to 104 Ϯ 1 nM in MHN fibers (Fig. 5, inset) and to 224 Ϯ 4 nM in MHS fibers ( Fig.  5) but had no effect on V m in either genotype. In addition, KB-R7943 decreased the elevation of [Ca 2ϩ ] r evoked by halothane in MHS muscle fibers from 1,826 Ϯ 16 to 830 Ϯ 12 nM (Fig. 5).
Because KB-R7943 has several targets in the muscle cells, we tested YM-244769 that has high affinity for blocking the reverse mode of NCX3 (37). Exposure of fibers to YM-244769 (1 M) for 5 min did not modify [Ca 2ϩ ] r in MHN fibers (119 Ϯ 1 nM, Fig. 5, inset) but reduced [Ca 2ϩ ] r to 275 Ϯ 4 nM in MHS fibers (Fig. 5), with no effect V m in either group. Treatment with YM-244769 also decreased the elevation of [Ca 2ϩ ] r observed in MHS muscle fibers caused by halothane from 1,826 Ϯ 16 to 1,505 Ϯ 20 nM (Fig. 5).

Effects of KB-R7943 and YM-244769 on [Na ϩ ] r in Murine
Muscle in Vivo-[Na ϩ ] r was measured on the superficial fibers of vastus lateralis muscles in MHN and MHS murine before and after local application of KB-R7943 or YM-244769 and after exposure to 1.5% halothane vapor in their inspired gas. Local treatment with 10 M KB-R7943 reduced the [Na ϩ ] r in MHS murine muscle from 15 Ϯ 0.2 to 13 Ϯ 0.2 mM (Fig. 6), but had no effect in MHN (Fig. 6, inset). Local treatment with 1 M YM-244769 did not modify [Na ϩ ] r in either genotype (Fig. 6).
Equilibrium Potentials-Using the V m , [Ca 2ϩ ] r , and [Na ϩ ] r found in the present study, the Nernst potential for Ca 2ϩ (E Ca 2ϩ ) and Na ϩ (E Na ϩ ), as well as the reversal or equilibrium potentials for NCX (E NCX ) were calculated as described under "Experimental Procedures" and shown in Table 1. We found a difference in the NCX reversal potential of 27 and 23 mV between MHN versus MHS from both swine and murine muscle fibers, respectively.

YM-244769 Reduces the Amplitude/Area of the Ca 2ϩ Transient Induced by High Potassium in Murine MHS Muscle Fibers-
To determine the contribution of the reverse mode of NCX during sustained muscle depolarization, single muscle fibers were exposed to high [K ϩ ] e in the presence or absence of YM-244769. Fig. 7A shows representative fluorescence traces of MHN and MHS murine single fibers loaded with Fluo-4 AM and then exposed to 60 mM [K ϩ ] e in the presence or absence of 1 M YM-244769. It can be noticed that Ca 2ϩ transient from the MHS muscle fiber is greater (peak and area under the curve) than MHN (Fig. 7A), a finding that is consistent with previous observations (34,38). YM-244769 treatment reduced both the peak amplitude (⌬F/F 0 max) and the area under the curve in MHS fibers, without any significant effect in MHN fibers (Fig. 7,  B and C). Fig. 7D shows normalized Ca 2ϩ transients (normalization 0 to 1) recorded from single fibers isolated from MHN and MHS with and without YM-244769. YM-244769 also increased the rate of decay of the fluorescence signals in MHS fibers without a significant effect in MHN (Fig. 7E).

YM-244769 Does Not Affect SR Ca 2ϩ Loading in MHN or MHS Adult Fibers and
Myotubes-There was no difference in SR Ca 2ϩ loading in MHN or MHS adult FDB fibers after pretreatment with 1 M YM-244769 and then exposed to caffeine Ϫ69 Ϫ59 (Fig. 7F) or in MHN or MHS myotubes after pre-treatment with 1 M YM-244769 and perfused with ionomycin (Fig. 7G).

NCX3 Expression Is Not Modified in MHS Skeletal
Muscle Cells-We measured the level of NCX3 expression by Western blot analysis of total lysates from adult quadriceps muscles and differentiated myotubes. NCX3 expression was similar between MHN and MHS samples (p ϭ 0.57 for quadriceps muscles and p ϭ 0.59 for myotubes samples) (Fig. 8).
siRNA Knockdown of NCX3 Reduces [Ca 2ϩ ] r and the Ca 2ϩ Transients Induced by High Potassium in MHS Myotubes-MHN and MHS myotubes were transfected with 25 nM NCX3-siRNA or scramble-FITC-siRNA and then 48 h later the expression levels were assessed by Western blot. Fig. 9A shows that NCX3 siRNA significantly reduced expression of the NCX3 protein in both MHN and MHS myotubes compared with scramble-transfected myotubes with a high transfection efficiency (Fig. 9B). NCX3 knockdown (48 h) significantly reduced the [Ca 2ϩ ] r in MHS myotubes compared with scramble-transfected MHS myotubes (188 Ϯ 6 versus 270 Ϯ 8 nM, p Ͻ 0.001) (Fig. 9C). No significant differences in [Ca 2ϩ ] r were observed in MHN myotubes transfected with NCX3 siRNA compared with scramble-transfected MHN myotubes. Similarly, we did not observe a significant difference in [Ca 2ϩ ] r between scramble-transfected and non-transfected MHN and MHS myotubes (Fig. 9C, dashed horizontal lines represent the non-transfected [Ca 2ϩ ] r levels).
We have previously shown that MHS myotubes had a larger Ca 2ϩ transient after exposure to elevated [K ϩ ] e compared with MHN cells (34,38). To assess contribution of the NCX3 in the Ca 2ϩ response to high [K ϩ ] e , MHN and MHS myotubes were transfected with NCX3 or scramble-siRNA for 48 h and then loaded with Fura-2AM. FITC fluorescence was used to track the transfected myotubes. Fig. 9D shows representative Fura-2AM fluorescence traces in transfected myotubes. NCX3 knockdown significantly reduced the amplitude and the transient area in MHS myotubes (Fig. 9, E and F). However, the effect of NCX3 knockdown in MHN myotubes induced a slightly but significant increase in both the amplitude and area of the evoked transients (Fig. 9, D-F). These results suggest that the forward mode NCX is the predominant in MHN myotubes,  Quadriceps muscles were dissected from 3-5-month-old mice and cultured differentiated myotubes from MHN and MHS mice were homogenized and NCX3 protein levels were assessed by Western blot and normalized by GAPDH levels. Data are expressed as mean Ϯ S.E., n are indicated in the graph bars. ns, no significant difference; unpaired t test.
contributing to the extrusion of intracellular Ca 2ϩ during prolonged depolarization (Fig. 9, D-F). These results are supported by the fact that YM-244769 blocks NCX3 in the reverse mode and did not modify the Ca 2ϩ transients in MHN (Fig. 7).

DISCUSSION
It is generally accepted that the main pathways of intracellular Ca 2ϩ removal in skeletal muscle are: (i) Ca 2ϩ uptake via sarcoplasmic reticulum Ca 2ϩ -ATPase (39), (ii) Ca 2ϩ extrusion via the plasma-membrane Ca 2ϩ -ATPase (39), and (iii) Ca 2ϩ extrusion by NCX (9,10,16). NCX is an electrogenic carriermediated transport process that can operate in the forward or reverse mode (2). Perturbation of any of these mechanisms can change the resting intracellular Ca 2ϩ regulation in muscle cells.
The present results confirm not only that [Ca 2ϩ ] r is elevated in MHS compared with MHN muscle fibers (19,21,22), but also that [Na ϩ ] r is also higher in MHS than MHN (19). [Na ϩ ] is important in modulating intracellular Ca 2ϩ concentrations in excitable cells through the NCX. Normally in excitable cells [Na ϩ ] r is primarily regulated by the Na ϩ ,K ϩ -ATPase (2), the NCX, and the amiloride-sensitive Na ϩ /H ϩ exchanger (40). An elevated [Na ϩ ] r would shift the balance of fluxes through NCX to favor more Ca 2ϩ influx resulting in an elevation of [Ca 2ϩ ] r . The elevation of [Na ϩ ] r found in MHS muscle could be related either to an increased influx or decreased efflux of Na ϩ . The fact that Na ϩ ,K ϩ -ATPase expression levels were similar in vastus lateralis from MHS patients who underwent a diagnostic contracture test for susceptibility to malignant hyperthermia (41) and that the values of the V m in MHN and MHS muscle are similar, suggests that observed dysfunction in [Na ϩ ] r would most probably be associated with an increase in Na ϩ influx rather than a decrease in efflux. In addition, the fact that elevation of [Na ϩ ] r in murine MHS muscle cells could be attenuated by local application of agents that block Orai1 and nonselective cation entry through transient receptor potential channels 3/6 (TRPC3/6) suggests that the elevation is mediated by an increase in influx of Na ϩ and Ca 2ϩ through these channels (19). This potential mechanism is supported by a number of electrophysiological studies that demonstrate that TRPCs allow permeation of Na ϩ as well as Ca 2ϩ (42,43).
Here we provide evidence that a MH-RYR1 mutation enhances the reverse mode function of NCX3 in swine and murine skeletal muscle fibers. Withdrawal of external Na ϩ ions, which activates the NCX reverse mode (2) caused an increase of [Ca 2ϩ ] r in both MHN and MHS swine skeletal muscle not mediated by membrane depolarization, but was more pronounced in MHS muscles. The elevation of [Ca 2ϩ ] r was dependent on [Na ϩ ] e . This increase in [Ca 2ϩ ] r was reversed in both groups when Na ϩ was reintroduced into the physiological solution. The rise in [Ca 2ϩ ] r was not associated with membrane depolarization in either MHN or MHS muscle fibers (data not shown). In MHS muscle cells the elevation of [Ca 2ϩ ] r elicited by extracellular Na ϩ withdrawal produced sarcomere oscillations, which were not seen in MHN cells. This lack of muscle activation in MHN muscle appears to be related to the magnitude of [Ca 2ϩ ] r elevation, which fails to reach the mechanical threshold in MHN cells (44,45). The absence of muscle activation in MHN muscles is in agreement with previous studies where the effect of extracellular Na ϩ withdrawal had been studied in diverse muscle preparations (15,46,47).
When external Na ϩ was replaced with lithium, a classical non-selective NCX blocker (48), a reduction in [Ca 2ϩ ] r was observed in both groups (data not shown) supporting the hypothesis that the elevated myoplasmic Ca 2ϩ that occurred in the NMG solution was due to the activation of the reverse mode of NCX, rather than a direct effect of NMG on the intracellular Ca 2ϩ . The increase in [Ca 2ϩ ] r induced by Na ϩ -free solution has been shown to be enhanced with the blockade of plasma-membrane Ca 2ϩ -ATPase with 25 M carboxyeosin (49,50) or sarcoplasmic reticulum Ca 2ϩ -ATPase with 200 nM 2,5-di(tert-butyl)-1,4-benzohydroquinone (51,52). These pharmacological treatments also potentiate the observed increase in [Ca 2ϩ ] r in swine muscle fibers from both genotypes (data not shown), suggesting that plasma-membrane Ca 2ϩ -ATPase as well as sarcoplasmic reticulum Ca 2ϩ -ATPase partially buffer the Ca 2ϩ influx via the NCX. The source of [Ca 2ϩ ] r elevation caused by the Na ϩ -free solution in MHN and MHS swine muscle fibers appears to be extracellular Ca 2ϩ , because incubation of muscle in Ca 2ϩ -free solution blocks the elevation of [Ca 2ϩ ] r in both genotypes.
Previous work has shown that glycerol treatment disrupts excitation-contraction coupling in skeletal muscle by disconnection of the transverse tubular system from the extracellular space (32,33). Here we show that detubulation of MHN and MHS swine fibers induces a small increase in [Ca 2ϩ ] r in both types of muscle fibers (16% MHN and 10% MHS, respectively). This increment may be related to the change in muscle volume, water moving out the cell, upon exposure to hypertonic solution, because striated muscle behaves as a perfect osmometer, decreasing its volume in proportion to the tonicity of the bathing medium (53,54). The increase in [Ca 2ϩ ] r provoked by Na ϩfree solution after muscle detubulation is reduced by 62% in MHN and 68% in MHS compared in non-detubulated muscle fibers. This reduction in the elevation of [Ca 2ϩ ] r supports the idea that NCX activity in MHN and MHS muscle fiber appears to be predominantly across the t-tubule rather than the surface sarcolemma. These results agree with previous studies showing that localization of NCX is higher in the t-tubular system than in the sarcolemma in skeletal and cardiac muscles (6,55).
In the present study KB-R7943 significantly decreases the elevation of [Ca 2ϩ ] r provoked by Na ϩ withdrawal in MHN (63%) and MHS (71%) swine muscle fibers suggesting that the observed elevation of intracellular [Ca 2ϩ ] was due to a Ca 2ϩ influx mediated by the presence of functional NCX3 in its reverse mode, which appears to be enhanced in MHS muscle cells. Furthermore, this NCX3 blocker reduces [Ca 2ϩ ] r by 16 and 37% in MHN and MHS swine muscle, and 13 and 31% in MHN and MHS murine muscles, respectively. However, these reductions in [Ca 2ϩ ] r must be considered with caution due to the lack of selectivity of KB-R7943, because there are evidences that show that it also inhibits voltage-gated Na ϩ and Ca 2ϩ channels, the inward rectifying K ϩ channels in cardiac cells (56), the transient receptor potential channels (TRPCs) in HEK293 cells (57), and RyR1 in skeletal muscle and HEK293 cells (58).
On the other hand, YM-244769 is a potent and highly selective NCX blocker that preferentially inhibits the reverse mode of NCX3 (IC 50 ϭ 18 nM). Furthermore, it is well established that up to 1 M (dose used in the present study) does not affect the forward mode (37). YM-244769 did not modify [Ca 2ϩ ] r in MHN but reduced it by 16% in murine MHS muscle. These results suggest that the reverse mode of NCX3 may play a minor role in the dysregulation of resting intracellular Ca 2ϩ observed in quiescent MHS muscle cells. The fact that the effect of YM-244769 on [Ca 2ϩ ] r was small compared with KB-R7943 (16 versus 37%) is consistent with the fact that KB-R7943 may be blocking several Ca 2ϩ entry pathways as described above. We have shown in vivo that exposure of MHS muscle fibers to halothane caused a robust elevation on intracellular [Ca 2ϩ ] (19,22). Local application of KB-R7943 and YM-244769 reduced the elevation of [Ca 2ϩ ] r induced by halothane in vivo by 55 and 18%, respectively, in MHS murine muscle fibers. This difference in the reduction of intracellular [Ca 2ϩ ] upon halothane exposure (55 versus 18%) may be related to the specificity of the agents regarding their pharmacological effects. Taken together, these data strongly support the hypothesis that sarcolemmal Ca 2ϩ entry mediated by the reverse mode of NCX3 (reported here) in combination with Gd 3ϩ -or GsTMX4-sensitive pathways (19) plays an important role in maintaining the new steady-state for [Ca 2ϩ ] r observed during an MH episode triggered by halothane.
Previous reports have shown that skeletal muscle fibers can contract after electrical stimulation or under whole cell voltage clamp with short step depolarization in low extracellular Ca 2ϩ concentrations (59,60). However, the importance of extracellular Ca 2ϩ both in the presence (61,62) and absence of Ca 2ϩ buffers (63) for maintenance of muscle K ϩ contractures has also been well demonstrated. However, our data provide evidence that extracellular Ca 2ϩ influx mediated by the reverse mode of NCX3 plays a role for the maintenance of Ca 2ϩ transients induced by elevated [K ϩ ] e in MHS muscle fibers. In fact, the peak amplitude and area were reduced and duration of the Ca 2ϩ transient elicited by [K ϩ ] e was shortened (increased rate of decay of peak Ca 2ϩ ) in the presence of YM-244769. These pharmacological effects are not mediated by a reduction in SR Ca 2ϩ loading in either MHN or MHS fibers. Our experiments in FDB fibers challenged with caffeine and in myotubes challenged with ionomycin showed that YM-244769 did not modify SR Ca 2ϩ loading. The fact that we did not observe a significant reduction in the peak Ca 2ϩ transient amplitude, in the area under the curve or rate of decay in MHN fibers in the presence of YM-244769 suggest that the reverse mode of NCX3 plays little or no role in MHN muscle fibers during sustained depolarization. Our pharmacologic observations are supported by our findings using siRNA knockdown of NCX3 in MHS myotubes, which both reduced the [Ca 2ϩ ] r and the amplitude/area of the Ca 2ϩ transient induced by high extracellular potassium confirming the role of Ca 2ϩ entry mediated by NCX3 at rest and during sustained depolarization.
The increased NCX3 reverse mode activity in MHS murine muscle cells suggest a compensatory mechanism to overcome the lower SR Ca 2ϩ loading described in MHS muscle cells (19,25,64) as a result of RyR1-MH leakiness (21). The reduced SR Ca 2ϩ levels are likely to promote an increased Ca 2ϩ influx through Orai and TRPC channels (19), and possibly other Ca 2ϩ channels (34). Our previously published data indicated that depolarization with 60 mM K ϩ produced an average area under the curve of the Ca 2ϩ transient Ϸ2-fold greater in myotubes from heterozygous MHS R163C compared with WT myotubes (34). Extracellular Ca 2ϩ plays a significant physiological role for the maintenance of Ca 2ϩ transients induced by elevated [K ϩ ] e in WT and R163C MHS myotubes, with a more pronounced effect in the later (34).
It is very well established that the direction in which NCX works is determined largely by the relative electrochemical driving forces for Na ϩ and Ca 2ϩ ions and the resting membrane potential. If the driving force for three Na ϩ into the cell is larger than that for one Ca 2ϩ into the cell (3:1 Na ϩ /Ca 2ϩ stoichiometry), the exchanger will transport Na ϩ ions into the cell and take Ca 2ϩ ions out (forward mode). On the other hand, if under another set of conditions (altered membrane potential, ionic gradients, or post-translational modifications), the driving force for three Na ϩ into the cell is less than for one Ca 2ϩ into the cell, then the Ca 2ϩ electrochemical gradient will become the dominant inward driving force (reverse mode). The present results show that in MHS muscle the NCX reverse potential (E NCX ) shifts to a more negative value. In swine it shifts from Ϫ42 mV in MHN to Ϫ69 mV in MHS and in mice from Ϫ36 mV in MHN to Ϫ59 mV in MHS muscle cells. These shifts in E NCX (ϩ27 and ϩ23 mV) observed in MHS muscle cells imply a major Ca 2ϩ influx via reverse mode of NCX with muscle depolarization. The net effect will be an increase in the intracellular Ca 2ϩ concentration during muscle activation.
In summary, these results provide evidence that both swine and murine muscle fibers possess a functional NCX, which can mediate Ca 2ϩ entry when the Na ϩ electrochemical gradient is reduced (reverse mode). This Ca 2ϩ entry via the reverse mode was greater in MHS than MHN muscles, and is sensitive to NCX3 blockers like KB-R7943 and YM-244769. Furthermore, the reverse mode of NCX3 contributes to the overall Ca 2ϩ elevation observed during sustained depolarization induced by high [K ϩ ] e and MH episode elicited by halothane.