Multiminicore Disease in a Family Susceptible to Malignant Hyperthermia:  Histology, In Vitro Contracture Tests, and Genetic Characterization FREE

Malignant hyperthermia (MH) is a myopathy with various clinical symptoms. The most important sign is a hypermetabolic status triggered by the administration of halogenated anesthetics and depolarizing agents, which can be fatal if dantrolene treatment is not quickly initiated.¹ From a biochemical point of view, the abnormal reaction to anesthesia is conditioned by a failure of calcium homeostasis mainly due to a dysfunction that affects the ryanodine receptor (RYR1).² Genetic analyses have identified so far more than 20 causative mutations of the RYR1 gene located on chromosome 19q13.2.^3- 4 More recently, a mutation that affects the dihydropyridine receptor has been described in malignant hyperthermia–susceptible (MHS) patients.^5- 6

Diagnosis of MH is currently based on the gold standard in vitro contracture test (IVCT) using halothane and caffeine.⁷ Given the high sensitivity (99%) and a satisfactory specificity (93.6%) of these tests,⁸ a patient is considered to be MHS if halothane and caffeine test results are both positive. The occurrence of abnormal MH contracture test results has been reported in various myopathies known to be genetically distinct from MH^9- 10 and in central core disease (CCD), which is genetically linked to MH.¹¹

Multiminicore or minicore disease is considered an autosomal recessive congenital myopathy and is defined morphologically by the presence of multiple zones of sarcomeric disorganization devoid of oxidative activity ("minicores") in muscle fibers.¹² Multiminicore lesions differ from central cores by their lack of selectivity for a particular fiber type and their smaller size. According to clinical features, 4 different subgroups of minicore disease have been distinguished so far, with the most prevalent form (classic minicore disease) characterized by paraspinal muscle weakness, severe scoliosis, and respiratory insufficiency.¹³ Considering the wide clinical spectrum of minicore disease, it has long been suspected to be genetically heterogeneous. Furthermore, it is strikingly different from CCD in many respects.¹⁴ Multiminicores have been reported as transient architectural changes in patients with a mutation in the RYR1 gene,¹⁴ and additional mutations have been reported in patients with the classic minicore disease phenotype.¹⁵ However, the genetic basis of most of the forms remain unknown, and the association between MHS and multiminicores has never been clearly documented. In the present study, we report the results from histological and genetic tests and IVCTs in a large family with MH in which multiminicores have been clearly identified.

Exploration of the family started after the death of the proband (II:4) in Basel, Switzerland. He had been anesthetized twice for a disc herniation. On both occasions, fluorinated anesthetics were used. In 1986, muscular rigidity appeared at the onset of the anesthetic procedure, but this rigidity receded afterward without treatment. In 1987, surgery was again performed, and complications affecting heart rate, PCO₂, and temperature (40.4°C) occurred. Dantrolene was used, but the procedure was fatal.

In the proband's son, MHS had been confirmed from IVCT performed in Basel. Due to the dominant inheritance of the MH trait (Figure 1), these positive IVCT results also confirmed the MH status of the proband.

Results reported in the present study are from 29 members of this MH family (147 members total) who were investigated in Marseille (ie, 6 patients of 10 in the second generation, 19 of 53 in the third, and all 4 patients in the fourth generation). Besides the proband's death, 5 children died in early childhood. Although a potential link with MHS has not been proved, it is noteworthy that sudden infant death has been associated with MHS in previous studies.^16- 17 Two patients (III:36 and IV:4) complained of paraspinal muscle weakness and muscle pain similar to peripheral cramps.

Muscle specimens (1 g) were excised from the left biceps with the patients under local anesthesia. They were used for morphological and biochemical analyses and IVCT as previously described.¹⁸ In 6 patients (III:24, III:27, III:33, III:34, III:36, III:41), ultrathin sections were prepared and double-stained with uranyl acetate and lead citrate.

For conventional histochemical studies, 10-µm-thick frozen sections were processed with hematoxylin-eosin, modified Gomori trichrome, NADH dehydrogenase, succinate dehydrogenase, cytochrome C oxidase, adenosine triphosphatase (preincubation at pH 4.3, 4.6, and 9.4), acid phosphatase, periodic acid–Schiff, Sudan black, nonspecific esterase, and Congo red as previously described.¹⁹ Paraffin sections were stained with hematoxylin-phloxine-saffron and Masson trichrome. Ultrathin sections were examined by electron microscopy [Jeol 100C electron microscope; JEOL (Europe) SA, Croissy sur Seine, France].

On the basis of pathological examination, patients were divided into 3 major subgroups. Group 1 had normal muscle biopsy results. Nonspecific abnormalities without disorganization of intermyofibrillar network, including type II fibers, were found in group 2. In group 3, features of multiminicore lesions were detected with oxidative techniques. In this subgroup, a broad spectrum of focal disruption of sarcomeres was observed. The number of type I and type II fibers showing anomalies was quantified, and occurrence of a type I or type II atrophy or predominance was assessed. Histological features were reviewed independently by 2 neuropathologists (D.F.-B. and J.-F.P.), without knowledge of the results of genetic study and IVCT.

The IVCTs were performed on muscle biopsy specimens as previously described and in accordance with the protocol recommended by the European Malignant Hyperpyrexia Group (EMHG).⁷ Participants were recognized as being MHS if the results of both contracture tests (halothane and caffeine) were positive and recognized as not being susceptible if halothane and caffeine test results were negative. An equivocal status was indicated if only 1 test result was positive. The 0.2-g IVCT threshold imposed by the EMHG for the muscle contracture was used.

Genomic DNA was extracted from whole blood using a guanidine method according to Jeanpierre.²⁰ DNA from family members was typed using the following microsatellite repeat markers: D19S220, D19S422, and RYR1, an intragenic dinucleotide repeat for linkage study to the RYR1 locus. Primer sequences and allele sizes have been previously reported.^21- 22 For each primer pair, 0.4-pmol forward primer was 5′ end-labeled with sodium phosphate P 32 and added to a 25-µM polymerase chain reaction mix containing 50 ng of genomic DNA, 200µM of dNTPs, 1µM of forward and reverse primers, and 0.25 U of Taq polymerase (Oncor, Illkirch, France). Amplification was performed for 25 to 30 running cycles consisting of 30 seconds at 94°C, 30 seconds at annealing temperature, and 1 minute at 72°C. After denaturation with formamide, amplified products were separated on an 8% polyacrylamide and 8M of urea DNA sequencing gel before autoradiography. Consecutive allele numbers were assigned, and the results were documented using Cyrillic software version 2.1 (Cherwell Scientific Publishing, Oxford, England). Two-point linkage analysis was performed using the 5.2 Linkage Package (Cherwell Scientific Publishing) with the following variables: the disease allele frequency was estimated at 1 per 10 000, the penetrance of the index case was taken as 1.0, and the penetrance of the disease was set at 0.98 when based on IVCT diagnosis. The phenocopy rate was set at 0 and 0.02 under the same conditions.

Total genomic DNA from index cases was assayed for 24 previously described RYR1 mutations¹¹ by restriction enzyme analysis using wild or mismatched primers to create or abolish a restriction site except for the mutation G248R, tested as previously described.²³ When a mutation at a restriction site was detected, the presence of the mutation was confirmed by sequencing. The following mutations were tested: C35R, R163C, G248R, G341R, R401H, I403M, Y522S, R552W, R614C, R614L, R2163C, R2163H, R2163P, V2168M, T2206M, T2206R, G2434R, R2435H, R2435L, R2452W, R2454C, R2454H, R2458C, R2458H, T4637A, Y4796C, T4826I, and I4898T.

Total RNA was extracted from a frozen muscle specimen of patient III:36 using a guanidinium thiocyanate-phenol-chloroform method.²⁴ Complementary DNA (cDNA) synthesis was performed using specific primer mixes. The various overlapping fragments that resulted from amplification of the first cDNA strand were sequenced using an ABI 3700 DNA sequencer (AMI Bioscience, Heidelberg, Germany). NlaIII restriction analysis of exon 50 and EarI restriction analysis of exon 53 were respectively used to detect the R2676W and T2787S mutations in the rest of the family and in 100 chromosomes of the general population.

The χ² test was used to compare histological, IVCT, and genetic data. P<.05 was considered statistically significant.

The IVCTs were performed in 29 patients, and the results are given in Table 1. According to the thresholds defined by the EMHG, 17 patients were diagnosed as being MHS and 10 as being nonsusceptible. For the remaining 2 participants, the caffeine but not the halothane test results were abnormal, and the individuals were classified as equivocal.

Of interest, histological analyses of muscle biopsy specimens revealed the presence of multiminicores in 16 (95%) of the 17 MHS patients. Multiminicore lesions were observed in both fiber types in most of the cases (Figure 2). The mean percentage of affected type I fibers was 22% (range, 5%-50%). On average, 14% (range, 5%-50%) of type II fibers were affected (Table 2 and Table 3). Angulated fibers, isolated or grouped, indicating a denervation process were not observed. Ultrathin section and ultrastructural examination, performed in patients showing at least 30% of multiminicores in type I fibers, confirmed typical features of multiminicores in all cases. The remaining MHS patients had no multiminicores and a few necrotic fibers considered nonspecific histological signs. Among the 10 nonsusceptible participants, 3 showed histological anomalies, including 1 with multiminicores. Muscle biopsy specimens were histologically normal in the remaining 7. Atrophy of type II fibers was also identified in the 2 participants who were diagnosed as being caffeine equivocal on the basis of IVCT. None of the participants had any central cores associated with multiminicores. Overall, taking into account the patients with multiminicores, χ² tests clearly indicated a significant link (P<.001) between MHS and the presence of multiminicores (χ² = 26.5) in this family.

Genetic analyses were performed in 19 patients, including 13 MHS patients, 5 nonsusceptible participants, and 1 equivocal participant. The lod score value obtained (z = 3.2 at 𝚯 = 0) indicated a linkage between the MH trait and the RYR1 locus in chromosome 19q13.2 (Table 4). In addition, a significant link (χ² = 19.0, P<.001) was calculated between the presence of multiminicores and the D19S220-RYR1-D19S4224:5:8 haplotype. As a matter of consequence, χ² tests indicated a significant link (χ² = 13.4, P<.003) between abnormal IVCT results and the RYR1 locus (Table 5).

Although none of the already known mutations could be identified, sequencing of the entire coding cDNA obtained from a muscle biopsy specimen of patient III:36 led to the identification of 2 missense heterozygous mutations. A 8026C→T transition in exon 50 was responsible for the substitution of an arginine by a tryptophane residue at position 2676 of RYR1 (R2676W). An 8160C→G transversion in exon 53 resulted in the substitution of a threonine by a serine residue (T2787S). Haplotyping data (Table 4) and restriction analysis (not shown) indicated that both mutations are carried by the 4:5:8 allele and segregate perfectly with the MHS trait in the family.

This study is the first report to our knowledge of a significant link between the presence of multiminicores and MHS in a large family, with multiminicores identified in 95% of the MHS family members. Of interest, genetic analysis findings indicated that the MH trait and the presence of multiminicores are significantly associated with the presence of a mutated RYR1 gene with 2 newly described mutations.

In that respect, this family strikingly differs from what has been previously reported regarding minicore disease. Multiminicores in skeletal muscle fibers have been associated so far with minicore disease, a congenital myopathy usually considered an autosomal recessive disease and characterized by neonatal hypotonia and nonprogressive or slowly progressive generalized muscle weakness.¹³ Also, mild histological changes that resemble minicores have been recently reported in siblings with CCD.²⁵ It is noteworthy that a recessive mutation in the RYR1 gene has been recently identified in a transient form of minicore disease.¹⁴

The present family is different in many respects from what has been reported so far in MH family members, susceptible or not susceptible (ie, various states of myofibril disorganization, ranging from a simple disorganization to core formation).^26- 27 First, the high occurrence (95%) of multiminicores is remarkable. Among the 534 individuals investigated in Marseille, France, thus far (November 2002) for the diagnosis of MHS, multiminicores have been identified in a few isolated cases (2.6%) but never systematically in several members of the same family as in the present case.¹⁸ Second, the relative number of fibers with cores varied among patients, but it was significantly higher than the percentage incidentally found in other MHS patients, with up to 50% of both fiber types affected in the most severe cases. Third, multiminicore lesions were not associated with type I fiber hypotrophy in contrast to what has been previously reported in congenital myopathies in general and particularly in minicore disease.^14- 15 In keeping with this difference, no clinical muscle involvement was reported by the patients included in the present study except 2 who had paraspinal muscle weakness and muscle pain similar to peripheral cramps (III:36 and IV:4). Multiminicore lesions have been related to a variety of conditions, including denervation process, RYR1 mutations, and mutations in the selenoprotein gene.¹⁵ In the present family, a denervation process was ruled out as a causative factor of multiminicore lesions because of lack of angulated fibers (isolated or grouped). In addition, none of the patients with multiminicore lesions had the clinical signs previously reported in either minicore disease or rigid spine muscular dystrophy, both conditions associated with multiminicore lesions.^14- 15

The genetic investigation revealed a linkage to the RYR1 locus on chromosome 19q13.2, and 2 amino acid substitutions in the same copy of the RYR1 gene were identified. The R2676W and the T2787S substitutions were the only changes disclosed after entire sequencing of the cDNA, indicating that one or both changes might have a causative role in the MHS phenotype in this family. Both changes segregated with the MHS phenotype and affected conserved amino acids in vertebrate RYR1. None of the substitutions were found in 100 chromosomes of the general population. The nature of the R2675W mutation, which involves an arginine residue, and its localization at the vicinity of the MH domain II provide evidence for its pathogenic role. Five of the 13 positions that have been reported so far to be mutated in the MH domain II (exon 39-46) corresponded to an arginyl residue. The threonine-to-serine 2787 change seems more conservative in terms of amino acid physicochemical properties. A pathogenic role of such a threonine-to-serine mutation has been clearly documented in other diseases, but its pathogenic role is more difficult to assess in the present case. It is noteworthy that other mutations identified in the MH domain II, even those involving arginyl residues, are not usually associated with central core lesions.^28- 30 Therefore, one could speculate that the T2787S mutation represents an aggravating factor, leading to the histological features observed in MHS patients in this family.

Although mutations in RYR1 are clearly causative of impaired calcium handling, the relationship (if any) between histological changes and abnormal calcium concentration has never been clearly documented. In the present study, the combined occurrence of multiple histological anomalies, the high percentage of MHS patients, and the genetic results might be in favor of such a causative link. As a matter of comparison, data from CCD patients are of interest. Unlike other MHS patients, the calcium release channel in CCD patients is abnormally permeable to calcium even under basal conditions,³¹ and the resting calcium level is increased in HEK-293 cells expressing RYR1 with the CCD mutation.³² It is therefore reasonable to assume that central cores and by analogy multiminicore formation might result from deleterious effects of a constantly elevated calcium level in myoplasm. However, caution should be taken, and it seems unlikely that the existence of multiminicores would be systematically associated with MHS.

Although the diagnosis of MHS is performed on muscle biopsy specimens, samples are not systematically screened for histological abnormalities. The link between dominantly transmitted multiminicores and MHS has been seen because of a multidisciplinary investigation of patients. The combined use of IVCT, histopathological analysis, and genetic testing afforded evidence toward a significant link between dominantly transmitted multiminicores and MH, as it has already been described for CCD.^11,33

In conclusion, we found a remarkable high occurrence of multiminicore lesions in an MHS family. Genetic analysis findings indicated that this association is linked to novel R2656W and T2787S substitutions present on the same allele of the RYR1 gene. Although we cannot formally exclude 1 of these 2 changes as being a rare polymorphism, we hypothesized that the R2656W substitution would be the candidate mutation responsible for MHS and that the secondary aggravating T2787S mutation would explain the high occurrence of histological multiminicores in MHS patients in this family. This would highlight the role played by individual genetic background in the large spectra of histological features associated with RYR1 mutations in MHS patients. An alternative explanation is that multiple mutations must be coexpressed to exhibit the MHS phenotype.³⁴

Taken together, these results indicate that multiminicore lesions are observed in MHS patients with neither clinical signs related to multiminicore disease nor histological features of congenital myopathies. In addition, multiminicores may be secondary to mutations in the RYR1 gene. As a consequence, these patients must be distinguished from patients with multiminicore disease and from other MHS patients for whom multiminicore are not observed. Further functional studies will be necessary to shed additional light on the respective or synergistic participation of the 2 identified amino acid changes, and this would help provide a better understanding of the mechanisms that lead to the formation of histological entities such as central cores and multiminicores in muscles and their association with the calcium release channel and calcium handling.

Corresponding author: Patrick J. Cozzone, PhD, Centre de Résonance Magnétique Biologique et Médicale, UMR CNRS No. 6612, Faculté de Médecine de la Timone, 27 Boulevard Jean Moulin, 13005 Marseille, France (e-mail: patrick.cozzone@medecine.univ-mrs.fr).

Accepted for publication April 15, 2003.

Author contributions: Study concept and design (Drs Guis, Kozak-Ribbens, and Cozzone); acquisition of data (Drs Guis, Figarella-Branger, Monnier, Kozak-Ribbens, Lunardi, Cozzone, and Pellissier); analysis and interpretation of data (Drs Guis, Figarella-Branger, Monnier, Bendahan, Kozak-Ribbens, Mattei, Lunardi, Cozzone, and Pellissier); drafting of the manuscript (Drs Guis, Figarella-Branger, Bendahan, and Cozzone); critical revision of the manuscript for important intellectual content (Drs Figarella-Branger, Monnier, Bendahan, Kozak-Ribbens, Mattei, Lunardi, Cozzone, and Pellissier); statistical expertise (Drs Bendahan, Mattei, and Cozzone); obtained funding (Drs Guis, Kozak-Ribbens, Lunardi, and Cozzone); administrative, technical, and material support (Drs Figarella-Branger, Monnier, Lunardi, Cozzone, and Pellissier); study supervision (Drs Guis, Kozak-Ribbens, and Cozzone).

This work was supported by grants from Programme Hospitalier de Recherche Clinique (Marseille), Association Française Contre les Myopathies (Paris), Institut National de la Santé et de la Recherche Médicale (Paris), and Centre National de la Recherche Scientifique (Paris).

We thank Albert Urwyler, MD, PhD, and Kathrin Censier, MD, PhD, for their helpful comments about the MH family and M. Ramadan, MD, for the information related to the anesthetic procedure.