Sodium Channel in Human Malignant Hyperthermia 

Studies were conducted on muscle from patients referred for diagnosis of MH susceptibility. Muscle fascicles were isolated from biopsies of the vastus lateralis and mounted in a tissue bath (37 degrees Celsius) containing modified Krebs solution. Twitches were electrically elicited by supramaximal stimulation (0.2 Hz; 2-ms pulse), and the preparations were equilibrated 30–60 min before testing. [16] The protocol of the North American MH Group [17,18] was followed, as previously published. [19] The contracture cutpoints used for a positive diagnosis by the North American protocol were more than 0.7 g to halothane (3% in the gas phase) in any one of three strips tested or more than 0.3 g to caffeine (2-mM concentration during dose response) in any one of three strips tested. In some cases, the time to a 0.2-g contracture after a 1-micro Meter bolus of ryanodine, [20] or the concentration of halothane causing a 0.2-g contracture using the European MH Group protocol, [21] were determined.

The persons in the present study were a typical sampling of referrals for MH diagnosis to our laboratories (in Philadelphia, Pennsylvania, and Bethesda, Maryland) and were not among the very rare families with sufficient numbers of family members tested to yield conclusive evidence for linkage to a specific chromosomal site. Therefore, it is probable that this sampling contains some persons who have a mutation on chromosome 19q13.1 and some who would have other mutations causing MH.

Small biopsy specimens were immediately frozen in liquid nitrogen. The RNA was extracted from 100 mg (wet weight) of these specimens using TRI Reagent (Molecular Research Center, Cincinnati, OH) according to the manufacturer's protocol. RNA samples were negative for degradation, as assessed by electrophoresis on agarose-formaldehyde gels; and the absorbance ratios at 260 and 280 nm were between 1.65 and 2.00. Oligo T¹⁵was used for RNA reverse transcription to select for poly A mRNA, using standard methods. [22] The cDNA was amplified by the polymerase chain reaction (PCR), as previously described for DNA. [5,6] Primers were developed based on sequences obtained from GenBank (Bethesda, MD) for human SkM2 and the human beta¹-subunit (Table 1). For quantitative analysis of the poly A mRNA, we constructed mimics using the PCR MIMIC Construction Kit (Clontech, Palo Alto, CA) and conducted competitive reverse transcription-PCR. A series of tenfold dilutions of mimic was established in separate PCRs, and the results of each reaction were examined on agarose gels. A second series of twofold dilutions was conducted with concentrations of mimic, based on the outcome of the tenfold dilutions. The PCR products were applied to agarose (1.5%) gels mixed with Reso-LOW (Enprotech, Natick, MA) in a horizontal electrophoresis chamber. Biomarker (low; Bioventures, Murfreesboro, TN) was applied for reference. After electrophoresis the gels were stained with ethidium bromide and photographed under ultraviolet light. At least two restriction enzyme sites were tested and the appropriate products verified to confirm that the amplified bands matched the beta¹-subunit and SkM2. Because comparisons are made to twofold increasing amounts of mimic, the amount of poly A mRNA per sample can be estimated with an optimum precision of about twofold using this method. The primers used to amplify the cDNA targets were bridged by an intron, allowing an internal detection of DNA amplification in each PCR reaction. DNA contamination was not obvious in any of the cDNA specimens.

Human (vastus lateralis) or equine (semimembranosus) muscle strips previously unexposed to halothane or caffeine and that remained after diagnostic testing were mounted in the tissue bath (at 37 degrees Celsius) and the muscle was directly stimulated supramaximally, as described previously. After the 30–60 min equilibration period, d-tubocurarine (20 micro Meter) was added to remove any influence of cholinergic nerve terminals. The preparations were exposed to tetrodotoxin (citrate buffer; Sigma Chemical Co., St. Louis, MO) in twofold increasing concentrations (50, 100 200, 400, 800, 1,600, and 3,200 nM). After each addition of tetrodotoxin the twitch height was monitored until it was close to a stable height (10–15 min) before the next concentration of tetrodotoxin was added. The average twitch height immediately before adding the next dose was used to evaluate twitch depression by the previous dose, and the percentage decrease was relative to the twitch height immediately before the first tetrodotoxin addition.

All values are presented as the means +/- SD (number of values). Paired analyses were used when two conditions were examined on the same subject or muscle strip, but otherwise grouped analyses were used. In those cases in which the variances in the groups being compared were not equal, a Mann-Whitney rank sum test was used. All comparisons were two tailed. Whenever more than two groups were compared, a one-way analysis of variance was used. The actual test used is indicated for each comparison.

The control participants were 31 +/- 12 yr old and the MH+ participants were 23 +/- 13 yr old (P = 0.122 by grouped t test). The outcomes of the contracture tests and clinical histories for each participant are indicated in Table 2. The outcome of the European MH Group halothane protocol and the ryanodine test corresponded to the outcome of the North American MH Group protocol in all but one case. One of the MH+ participants in the present study (Table 2; MH+#11) was MH+ by both the North American MH Group protocol and the European MH Group halothane protocol and yet exhibited a time to 0.2 g contracture greater than 20 min by the ryanodine test, which is typical of the MH- population. [20]

Cell culture studies [9] suggest that the functional expression of the SkM2 protein is at lower levels in MH+ than in MH- muscle. Therefore we assayed mRNA from six MH- participants (Table 2; MH-#1–6) and seven MH+ participants (Table 2; MH+#1–7) by reverse-transcription PCR to determine if the level of the transcript was also decreased for this alpha-subunit in intact mature muscle. The amounts of RNA/mg tissue (Table 3) recovered from our samples correspond with published values for human skeletal muscle [23] and were not different for MH- and MH+ muscle. To ensure that the cDNA preparations were similar whether derived from MH- or MH+ muscle, we also measured the abundance of the beta¹-subunit of the sodium channel. The means that the beta¹-subunit mRNA, which was derived from the same cDNA as the values for SkM2, was not significantly different between the MH- and MH+ groups (Table 3), suggesting that the reverse transcription was comparable for RNA extracts from both groups.

The mean value for SkM2 for the seven participants from the MH+ group was approximately 20-fold lower than for the MH- group (Table 3). The values of SkM2 for the MH susceptible participants were so low compared with the MH- participants that the variances were unequal between the groups and a Mann-Whitney rank sum test had to be used to test significance. Comparing the ratios of SkM2:beta, so that each participant had an internal control, revealed that the six MH- participants tested had ratios between 0.5 and 200. Six of the MH+ participants had SkM2:beta ratios between 0.008 and 0.2. The seventh MH+ participant (Table 2; MH+#7) had a ratio of SkM2/beta¹of 1.6:1, and this value was higher than two of the lowest values for the MH- group (0.5:1 and 1:1), suggesting that levels of SkM2 transcript are not always completely suppressed in the MH+ population. This finding corresponds with the known heterogeneity of the MH disorder. If this participant is excluded from the MH+ group, then the mean +/- SD becomes 0.112 +/- 0.076 (n = 6) for the ratio of SkM2:beta. Therefore this patient is approximately 20 SD values from the mean and can be considered an outlier. Excluding this person from the calculation to obtain a better estimate of the degree to which the SkM2 transcript can be suppressed yields a mean +/- SD of 0.016 +/- 0.020 (n = 6) for the MH+ SkM2 transcript. This means that a 115-fold level of suppression of SkM2 occurred in the six remaining MH+ participants compared with those who were MH-. This MH+ participant is included in the tetrodotoxin twitch depression data presented subsequently and exhibited a decreased expression of functional SkM2, suggesting that other factors (such as post-translational modification) may also lead to decreased levels of functional protein.

We expanded the population examined by testing the functional expression of sodium channels in 12 MH-participants (Table 2; MH-#5–16) and six MH+ participants (Table 2; MH+#7–12). Two of the MH- participants (Table 2; MH-#5 and #6) and one MH+ participant (Table 2; MH+#7) were examined for both functional expression and mRNA expression. An initial twitch height greater than 1.5 g was observed in all cases. A 100-nM concentration of tetrodotoxin was used to block the sodium current through human SkM1 by about 90%[13] and thereby to estimate the relative contribution of SkM1 sodium current to the muscle twitch (Figure 1(A)). We have also indicated the outcome of adding 50-, 200- and 400-nM concentrations of tetrodotoxin (Table 4). A large component of the twitch remained in MH- muscle at the 100-nM concentration of tetrodotoxin (Table 4), suggesting that about 50% of the muscle twitch in MH-muscle is mediated through a second type of sodium channel. Based on the mRNA analysis (Table 3) and the known greater resistance to tetrodotoxin than with SkM1, we identified this current as mediated through SkM2. In most cases this current (judged by twitch height) was effectively blocked by a 400-nM concentration of tetrodotoxin.

A slight decrease in twitch height can be observed over time even in some preparations not exposed to tetrodotoxin. The decrease in twitch height over 20 min was examined in untreated preparations, because this is the approximate time from the first addition of tetrodotoxin to the point when the twitch height is calculated for the 100-nM addition. In five preparations from five MH- participants, the mean twitch height at time 20 min (3.0 +/- 0.6 g [n = 5]) did not significantly decrease (P = 0.20, by two-tailed paired t test) from the height at time 0 (3.2 +/- 0.5 g [n = 5]). In five preparations from four MH+ participants, the twitch height did not significantly decrease (P = 0.62, by two-tailed paired t test), because the twitch height values at both times were 2.5 +/- 1.3 g (n = 5). Therefore compensating for any decrease in twitch height unassociated with tetrodotoxin blockade would not greatly change the values at the 50- or 100-nM concentrations in Table 4or in Figure 2.

It is also possible that patient age or initial twitch height might affect twitch blockade by tetrodotoxin. To determine whether patient age affects the tetrodotoxin dose response, we compared three age groups of healthy patients. The three groups were 9–17 yr old (13 +/- 4 yr [n = 3]); 22–34 yr old (27 +/- 6 yr [n = 3]), and 40–55 yr old (45 +/- 7 yr [n = 4]). The decreases in the twitch heights by tetrodotoxin were not significantly different among the three age groups at 50, 100, 200, or 400 nM tetrodotoxin (one-way analyses of variance comparing the three groups were conducted at each concentration). The maximum decrease at 100 nM tetrodotoxin was observed in the youngest group and resulted in a twitch height of 45 +/- 9%(n = 3). To determine whether the initial twitch height has any influence on the tetrodotoxin dose-response curve, we compared two extremes of initial twitch height. The first group consisted of five strips with a twitch height between 2.2 and 2.9 g (2.5 +/- 0.3 g [n = 5]) and the second group consisted of five strips with a twitch height between 4 and 6.4 g (5.2 +/- 1.1 g [n = 5]). The percentage twitch remaining at 50 nM tetrodotoxin was slightly, but significantly (by two-tailed grouped t test), higher for muscle bundles with the smaller size twitches (87 +/- 8%[n = 5]) than for muscle bundles with the larger size twitches (74 +/- 9%[n = 5]). However, the sensitivity to inhibition was not significantly different between the groups at the 100-, 200-, and 400-nM concentrations of tetrodotoxin. Therefore patient age and the initial twitch height do not significantly change the tetrodotoxin dose-response curves and do not alter the blockade at the 100-nM concentration of tetrodotoxin.

The initial twitch heights for the persons in the MH+ group (4 +/- 2.6 g [n = 6]) were similar to those for the MH- group (3.7 +/- 1.6 g [n = 12], P = 0.77). Unlike the results in MH- muscle, MH+ muscle exhibited a large inhibition of the twitch height after 100 nM tetrodotoxin (Table 4, Figure 1(B and C), Figure 2). The low levels of twitch remaining after 100 nM tetrodotoxin (range, 5–33%) suggest that functional SkM2 channels were present only in relatively low amounts in MH+ human muscle compared with MH- human muscle. Because about 10% of SkM1 currents may not be blocked by this concentration of tetrodotoxin, [13] the contribution of SkM2 to the twitch is potentially even lower than indicated in Table 4, as shown in Figure 2.

A third sodium current, as judged by tetrodotoxin sensitivity, was sometimes observed in human skeletal muscle. The twitch associated with this current was not decreased by even very high concentrations of tetrodotoxin. For example, the twitch height of preparations treated with 400 nM tetrodotoxin is not decreased when the concentration of tetrodotoxin is subsequently increased to 10 micro Meter, which corresponds to results of other investigators. [14,15] Therefore the sodium current sustaining the twitch in a 1.6-micro Meter concentration of tetrodotoxin was considered to be the third type of sodium current (highly tetrodotoxin resistant). [14,15] This current supported only a small fraction of the total twitch and was observed in 4 of 12 MH- and in 4 of 6 MH+ participants. The one MH+ patient having the highest percentage of twitch remaining (33%) at the 100-nM concentration of tetrodotoxin also had 12% of the twitch resistant to 10 micro Meter of tetrodotoxin. Therefore the highest level of twitch attributable to SkM2 would only be 21% in this patient. This tetrodotoxin-resistant channel has been associated with denervated muscle [10,14,15] but occurs frequently in equine semimembranosus skeletal muscle (in four of five horses examined) and can support a large fraction (33 +/- 14%[n = 4]) of the muscle twitch in that species. We determined that this current is not blocked by a concentration of tetrodotoxin as high as 100 micro Meter in equine muscle.

The present study using biopsy sections of skeletal muscle from MH-susceptible persons confirmed the same deficiency in functional expression of a sodium channel protein with a low tetrodotoxin sensitivity (presumably SkM2) originally reported in cell culture. [9] We extended these studies to show that the level of SkM2 mRNA can be suppressed in at least some, but not necessarily all, persons. Defects in sodium channel expression occur in all human MH+ muscle that we have tested, even in the MH+ participant in which the ryanodine contracture test failed. The altered ratio of sub-types of the sodium channel in MH+ muscle probably is not due to a change in the total number of functioning sodium channels in the sarcolemma, because the peak Na sup + currents [8] in cell cultures is equal to that of controls.

We have shown a deficiency in levels of transcript and/or functional expression of a specific alpha-subunit of the sodium channel (SkM2) by using three different model systems (patch-clamp in cell culture, muscle twitch, mRNA analysis). The identity has been suggested to be SkM2 by the mRNA analysis, which is a highly specific method of detection, and intermediate sensitivity of the twitch to blockade by tetrodotoxin. The observation that SkM2 is transcribed and functions in normal vastus lateralis muscle is significant for basic muscle physiology, because previous studies suggested the occurrence of only one type of functional sodium channel in healthy innervated skeletal muscle. [10,24] The presence of SkM2 was missed possibly because of the animal models and methods used to distinguish the channels and the lack of an identified muscle disorder for comparison, such as MH, in which one of the two channel subtypes is absent or greatly reduced. Originally two types of Na sup + channels were distinguished in skeletal muscle by being either blocked completely by, or completely resistant to, a 1-micro Meter concentration of tetrodotoxin. [14,15] The tetrodotoxin-resistant channel was absent in healthy muscle and up-regulated by denervation. [14,15] Contrasting the dose response of the blockade of twitch by tetrodotoxin in healthy and MH muscle in the present study shows that two different channel subtypes, rather than one, are inhibited by submicromolar concentrations of tetrodotoxin, as previously assumed.

Measurement of the tetrodotoxin block of currents due to recombinant cDNAs for SkM1 and SkM2 expressed in heterologous systems has found a Kⁱfor human SkM1 of 25 nM [25] and for SkM2 of 1.3 micro Meter. [26] Based on the present study, the estimate for the K sub i of human SkM1 from expressed cDNA is close to the estimated Kⁱin MH+ muscle, based on twitch depression (slightly less than 50 nM). A second level of tetrodotoxin sensitivity, presumably due to SkM2, in intact fully mature muscle apparently has a greater sensitivity to blockade than recombinant SkM2 in cell culture, because the estimated K sub i for this second level of twitch suppression in MH- muscle would be about 150 nM. The greater tetrodotoxin resistance of SkM2 in cell culture relative to biopsy sections of skeletal muscle may have led to SkM2 being identified as the tetrodotoxin-insensitive channel. The tetrodotoxin-insensitive channel has now been identified in human muscle and is not antagonized by concentrations of tetrodotoxin as high as 100 micro Meter. Therefore we have distinguished for the first time three subtypes of the sodium channel in skeletal muscle, based on the tetrodotoxin sensitivity of the directly elicited muscle twitch. The twitch associated with the sodium current most sensitive to tetrodotoxin was almost completely blocked by a 100-nM concentration of tetrodotoxin and is presumed to be supported by SkM1. The twitch associated with the second channel was completely blocked by a 400-nM concentration of tetrodotoxin and is presumed to be supported by SkM2, based on the mRNA analysis. The twitch associated with the third channel was not observed in all participants and was completely resistant to tetrodotoxin. This latter channel is probably the tetrodotoxin-resistant channel previously associated with denervation.

Despite the identification of the mutation in the ryanodine receptor associated with MH in swine, the mechanisms by which this mutation leads to the MH syndrome remains unclear. Mutations in the ryanodine receptor may be necessary, but they are not sufficient for MH, because some adult pigs [5] and humans [6] with these mutations do not necessarily exhibit an MH episode on challenge with triggering agents. In addition, young swine do not exhibit MH, [7] suggesting that a progression of events must occur before the syndrome can be triggered in subjects carrying the MH mutation. Human MH can be caused by a mutation in any one of several different genes. [27,28] Because humans with MH have not been reported to have mutations in SkM2, it is most probable that down-regulation of SkM2 occurs as a secondary response to the mutations in the ryanodine receptor or other proteins and is not due to actual mutations in SkM2. Down-regulation may not be unique to the SkM2 protein in MH muscle, because the amounts of dihydropyridine receptor [29,30] and ryanodine receptor [30] protein have also been reported to be reduced in porcine muscle by about 50% and 25%, respectively. However there is some specificity to down-regulation in MH+ muscle, because many proteins are at normal levels in porcine MH muscle. [31] Similar detailed studies have not been conducted in human muscle, although gross analyses have not found significant differences in amounts of protein expression. [32] The decrease in SkM2 appears to be far more dramatic than the decrease in the dihydropyridine and ryanodine receptors, and it is the first protein to be identified as down-regulated in human MH. In contrast to the depression of SkM2 mRMA in six of seven MH+ humans tested in the present study, other investigators found that the mRNA levels were not decreased for the dihydropyridine and ryanodine receptors in MH+ swine. [30] Because we observed one human MH+ with normal levels of SkM2 mRNA, it is possible that down-regulation of the SkM2 protein, either by transcriptional or post-transcriptional mechanisms, may be involved in the final common pathway in the pathogenesis of MH, and this could account for heterogeneity in the disorder.

What is the possible role of down-regulation of SkM2 in the pathogenesis of MH? The studies of Iaizzo et al. [33] observed that halothane caused a decrease in the maximum rate of rise, the peak height, and the maximum rate of fall of action potentials in muscle from MH+ swine at much lower concentrations than in muscle from healthy swine. Greatly elevated myoplasmic Ca²+ concentrations were not required for this action, which is consistent with the possibility that halothane selectively affects the time-dependent Na sup + or K sup + currents. The suggestion of Iaizzo et al. [33] that abnormal inactivation of Na sup + channels may account for the effects of halothane on the rate of repolarization of the action potential was confirmed in a parallel study by Wieland et al. [8] in cell cultures of human skeletal muscle. In cell cultures of skeletal muscle, halothane at low concentrations preferentially inhibits the adult (SkM1) channel and requires much higher concentrations to inhibit SkM2. [34] This latter study found similar effects of halothane on SkM1 and SkM2 in healthy and MH cell cultures but did not address the possibility of differences in ratios between the channel subtypes. Therefore a decrease in the ratio of SkM2 to SkM1 could result in muscle that would be more sensitive to the effects of low concentrations of halothane, which corresponds with the observations of Iaizzo et al. [33] Recently Wieland et al. [9] reported that the ratio of SkM2 to SkM1 was greatly reduced in cell cultures of MH humans compared with healthy persons. The higher proportion of SkM1 in MH muscle [9] could be involved in the delayed inactivation of fast Na sup + currents [8] reported in cell culture. The slowly inactivating component observed in MH+ muscle [8] was less sensitive to voltage-dependent inactivation than the fast component. SkM1 is also less sensitive to voltage-dependent inactivation than is SkM2. [35] Therefore the simplest explanation for a greater sensitivity of MH muscle to low concentrations of halothane could be based on altered ratios of SkM1 and SkM2, as we report in the present study. A mechanism linking a decrease in Na sup + flux, as manifested by a decrease in the peak current of the action potential, [33] with development of a contracture is unclear. However, it has been shown that omission of Na sup + from the bathing medium enhances contractures to halothane and caffeine. [36] Mepivacaine, a local anesthetic inhibiting Na sup + currents, causes contractures of MH+ but not MH- muscle. [37] Therefore the inhibition of Na sup + currents by halothane may play a role in muscle contractures in MH muscle, especially as SkM1 is far more sensitive to halothane than is SkM2.

However the situation could be much more complex, as the distribution of sodium channels along the sarcolemma and in the t-tubules is not uniform. The cat soleus muscle is predominantly composed of slow twitch (type 1) fibers and is highly sensitive to halothane compared with the gracilis muscle (predominantly fast twitch; type 2), originally suggesting that fiber type may be important in MH. [38] Although later studies suggested that fiber type distribution alone was not responsible for MH, [39] these studies of fiber type may be revealing important clues about what makes a muscle sensitive to triggering anesthesia. Antibodies used to map the distribution of sodium channels have identified two subtypes that exist on the sarcolemma of fast- and slow-twitch muscle, one of which shares epitopes with Na sup + channels in cardiac muscle and could be SkM2. [40] The cardiac-like sodium channels also exist in the t-tubules of slow-twitch skeletal muscle. [40] A third type of sodium channel (possibly a post-translational modification of one of the other two) exists in the t-tubule of fast-twitch muscle. [40] The density of sodium channels in the extrajunctional sarcolemma is much greater in fast-twitch than in slow-twitch muscle. [41] In addition to type and density of channel, the changes in the ratio of silent (inactive) sodium channels to active sodium channels could influence the response to triggering agents. Activation of sodium currents along the sarcolemma of the diaphragm with various toxins found populations of ion channels outside the end-plate region that are silent unless activated by batrachotoxin, a lipid-soluble toxin. [42] Fatty acids are elevated in porcine [7] and human [43] skeletal muscle and selectively activate SkM1. [9,13,44] Thus it is possible that a fraction of SkM1 that is silent on the sarcolemma in normal muscle may be shifted to a voltage-activatable state in MH+ muscle in parallel with the down-regulation of SkM2. Corresponding to this, fatty acids activated silent SkM1 channels in cultures from healthy subjects but had no added effect on SkM1 Na sup + currents in cultures from MH+ subjects. [9] Therefore MH may involve substitutions of sodium channel sub-types and activation of normally latent channels.

The expression of MH probably depends on the status of several different systems within muscle. These systems, including Na sup + channels, may be secondarily affected by a primary mutation to different degrees, depending on the subject. The present study suggests that altered expression of sodium channel subtypes places MH in the context of anesthetic-induced muscle rigidity associated with dystonias, in which mutations in the sodium channel (SkM1) have been shown to play a role. [45–47] The percentage of SkM2 based on the tetrodotoxin dose-response curve has provided the clearest separation of MH- and MH+ populations thus far.

The authors thank Dr. Edwin W. Lojeski, Kirsten Erwin, and Linda Tripolitis for technical assistance.