Screening of the Entire Ryanodine Receptor Type 1 Coding Region for Sequence Variants Associated with Malignant Hyperthermia Susceptibility in the North American Population

We performed genetic analysis of 30 MH/MH-susceptible individuals. The study was approved by the institutional review boards of all collaborating institutions. Patients had been referred for CHCT because of development of signs of MH during anesthesia or a positive family history of MH. In 28 at-risk family members, the diagnosis of MHS was established according to the North American CHCT protocol, which requires exposure of muscle strips to caffeine and halothane.2Testing with each drug was performed on at least 3 muscle fascicles, and individuals were considered MH positive if any one of the three exposures exceeded the diagnostic threshold. The threshold values for a positive test were greater than 0.7 g contracture response to 3% halothane or greater than 0.2 g in response to 2 mm caffeine. CHCT testing was performed at the Department of Anesthesiology, Uniformed Services University of the Health Sciences, Bethesda, Maryland, and the Wake Forest University, Winston-Salem, North Carolina. A single individual was diagnosed with MHS on the basis of abnormal muscle contracture results before standardization of the CHCT in North America. Two patients who did not undergo CHCT had clinical episodes of MH, one of them fatal.

The frequency of each RYR1  mutation identified in the North American MH/MH-susceptible cohort was estimated in 100 healthy unrelated American population controls. All of these 100 samples were previously collected for other studies and were made available for this study without personal information.

To facilitate the screening and to capture the effect of splicing mutations, we used muscle tissue because RYR1  is most abundantly expressed in skeletal muscle. Total RNA was extracted from skeletal muscle biopsy tissue using the RNA-wiz reagent (Ambion, Austin, TX). Three micrograms of total RNA was used for reverse transcription reaction to synthesize complementary DNA. Avian Myeloblastosis Virus Reverse Transcriptase (Promega, Madison, WI) was used in the reaction as specified in the manufacturer’s protocol. Complementary DNA was subsequently used for amplification reactions. Genomic DNA was extracted from muscle tissue and anticoagulated blood using the Wizard Genomic DNA Purification Kit (Promega).

Complementary DNA (cDNA) and genomic DNA served as templates for PCR. The entire 15,117-nucleotide-long coding region of the RYR1  gene was amplified as 57 overlapping fragments, each approximately 350 nucleotides long. Primers used for this purpose were designed using OLIGO software (Molecular Biology Insights, Cascade, CO). Intronic primers for amplification from genomic DNA were designed for each exon within the three mutational hot spots (exons 2–17, exons 39–46, and exons 90–104), and subsequently for exons 34, 51, and 67 in which novel mutations were detected during screening. All primer sequences are available on request. Amplification from genomic DNA was performed using an optimal procedure designed for each exon. Amplification was performed in a total volume of 20–25 μl with 1 μl single-stranded cDNA or 50–100 ng genomic DNA, 0.5 μm of each primer, 125 μm of each dNTP, 1.5 mm of MgCl², 10 mm Tris-HCl (pH 8.3), 50 mm KCl, and 0.6 U Taq  DNA polymerase (Applied Biosystems, Foster City, CA).

Denaturing high-performance liquid chromatography (DHPLC) detects sequence mismatches by separation of heteroduplexes and homoduplexes of double-stranded DNA molecules, based on a difference in denaturing characteristics.18,19DHPLC analysis was performed on an automated WAVE Nucleic Acid Fragment Analysis System (Transgenomic, Omaha, NE) according to the manufacturer’s protocol. Briefly, 5–7 μl of each PCR product was denatured at 95°C for 5 min and then gradually reannealed by decreasing the sample temperature from 95° to 25°C over a period of 45 min to promote the formation of heteroduplexes. To identify homozygous mutation, the patient’s sample was mixed with a control sample. The samples were subsequently eluted by acetonitrile gradient consisting of a mixture of buffer B (0.1 m triethylamine acetate) and buffer A (25% acetonitrile). Temperature for heteroduplex molecules was determined by using the DHPLC melting algorithm (Wavemaker Software; Transgenomic). The 57 overlapping PCR fragments of each MH/MH-susceptible individual covering the complete RYR1  coding sequence were screened by DHPLC. Five of the 57 PCR fragments had abundant GC nucleotide content that prevented reliable use of DHPLC methodology; these fragments were analyzed by sequencing alone. Two researchers interpreted DHPLC results independently. Samples with abnormal peaks or peaks at different retention times were further analyzed by sequencing.

Each sample showing an abnormal DHPLC peak pattern was sequenced in both directions using a fluorescent Big DyeTerminator Cycle Sequencing protocol on an automated ABI 310 Genetic Analyzer (Applied Biosystems, Foster City, CA). The samples were purified with QIAquick PCR Purification Kit (Qiagen, Valencia, CA) before sequencing. To further verify the detected RYR1  sequence variants, appropriate exons were amplified from the same patient’s genomic DNA, and nucleotide sequences were determined. All nucleotide and codon numbers used in this analysis refer to the wild-type cDNA sequence of the RYR1  gene.6 RYR1  mutation numbering is based on the same reference sequence. To establish population frequency of RYR1  mutations we encountered in the North American MH/MH-susceptible cohort, we studied the appropriate RYR1  segments by DHPLC testing or restriction enzyme analysis or both in 100 unrelated controls.

Of 30 individuals enrolled in the study, two were diagnosed with MH after a clinical episode (US848 and AC), and in 28, the MHS status was confirmed by unequivocal CHCT results (table 1). Each MH/MH-susceptible individual was characterized by DHPLC screening of muscle biopsy tissue–derived RNA. Twenty-one of these individuals revealed unusual elution profiles in one of the PCR produced fragments that were distinct from those of the wild type samples: They showed double peaks or atypical configurations suggesting heteroduplex formation (fig. 1). Subsequent sequencing of these abnormal fragments determined the exact sequence alteration and led to the identification of the underlying RYR1  mutation. We identified nine previously reported mutations in 12 individuals (table 1). The Arg163Cys substitution was detected in two individuals, and the Gly2434Arg substitution was detected in three individuals.

In addition, nine previously unknown single-nucleotide substitutions leading to a predicted amino acid change were identified in nine MH/MH-susceptible individuals. In each of the nine individuals with newly identified RYR1  mutations, the MHS status was confirmed by CHCT (table 1). All newly identified amino acid substitutions were screened in 100 unrelated North Americans by DHPLC profiling or restriction analysis of the affected exons amplified from the genomic DNA. None of the control samples contained any of the nine newly identified mutations. The original residues were highly conserved through the RYR1  evolution (fig. 2). Six of the nine affected residues were also conserved across the RYR species, RYR1, RYR2, and RYR3 (fig. 2).

While screening for mutations, numerous neutral sequence variants in the coding region of the RYR1  gene were identified. Four of them, Pro1785Leu, Gly2059Cys, Gln3756Glu, and Val4849Ile, are alterations leading to a predicted amino acid substitution, but they were previously detected and proven to be DNA polymorphisms because they were found in control subjects.8,13,20The Val4849Ile change in homozygous and heterozygous state was previously reported in association with central core disease in two families but was also found in 1 of 50 French controls.13,21Because the Val4849Ile substitution is located in the C-terminal mutational hot spot corresponding to a transmembrane region of RYR1, alters a highly conserved amino acid, and has been identified in an MH-susceptible individual (BG16) with strongly positive CHCT results (table 1), we have conducted an extensive search for the presence of this substitution among 50 North American white, 50 Asian, and 50 African-American individuals with negative results. Above data suggest that the Val4849Ile substitution may be pathogenic. On the contrary, the Pro1785Leu, Gly2059Cys, and Gln3756Glu substitutions seen in control individuals in this and other studies8,20are most likely nonpathogenic.

The nine remaining MH-susceptible individuals did not show any alterations of the DHPLC profiles throughout the entire RYR1  coding sequence, although the majority had convincingly positive results in the CHCT studies (table 1). The rate of RYR1  mutation detection in the studied MH/MH-susceptible cohort is therefore 70% (21 of 30).

All previously identified RYR1  mutations were located within one of three mutational hot spot areas of the gene (fig. 3). Of the newly identified mutations, two are located in the central transmembrane domain, and four are located in the C-terminal transmembrane domain (fig. 3). However, three novel mutations, Ser1728Phe, Gly2733Asp, and Arg3348His, were detected outside the mutational hot spots. The Ser1728Phe and Gly2733Asp mutations are in close proximity of the central mutational hot spot, whereas Arg3348His is at equal distances from the central and C-terminal regions (fig. 3). This suggests that RYR1 contains other critical domains that cause MHS if mutated and that the entire coding sequence must be analyzed to identify all MH-causing mutations.

We analyzed a previously described22North American family with fatal MH episodes in three family members (fig. 4). A surviving family member (B2) underwent contracture testing and was diagnosed with MHS (table 1). Screening for RYR1  mutations in this individual was performed with the use of genomic DNA extracted from blood lymphocytes. Sequencing of all exons within the mutational hot spot regions of the RYR1  gene revealed the Lys4876Arg mutation. Ten additional members of this family were available for genetic evaluation. In these individuals, DNA was extracted from blood lymphocytes, and exon 101 containing Lys4876 was amplified and analyzed by DHPLC. Four of the 10 at-risk members of this family showed the presence of the Lys4876Arg mutation (fig. 4, subjects B1, B4, C1, and C3).

Although defects in the RYR1  gene are considered the primary cause of MHS, genetic linkage studies indicate that only 50% of studied affected families show linkage to the RYR1  locus,9whereas RYR1  mutations are found in only approximately 22–25% of the individuals with positive CHCT results.10,11There are significant technical difficulties in establishing linkage and finding mutations in a gene spanning a 159,000-nucleotide genomic DNA segment and transcribing a 15,117-nucleotide-long RNA molecule.6To complicate the matter further, the majority of RYR1  mutations are rare and unique to a particular family,5and in addition, multiple silent DNA polymorphisms are interspersed with disease-associated mutations.5,8Most RYR1  mutations known to cause MH/MHS are clustered in three mutational hot spots. Because of the enormous size of the gene, the current mutation-detection methodology settles for screening the mutational hot spots only by sequencing or restriction analysis, which leaves the possibility for the presence of nonrecognized mutations in other areas of the gene.

The above circumstances determined the strategy of this study. We performed a complete screening of the RYR1  RNA derived from the biopsied muscle tissue in 30 MH/MH-susceptible individuals using DHPLC, a highly sensitive mutation-detection methodology.23Muscle tissue is the optimal source of RNA because RYR1  is preferentially expressed in the skeletal muscle. Analysis of RNA extracted from muscle but not lymphocytes proved to be a successful strategy of mutation identification in dystrophin.24Most importantly, RNA analysis allows to significantly reduce the number of DHPLC runs, in the case of RYR1  from 106 to 57 per patient. With the approach based on screening of the entire RYR1  gene, we hoped to obtain direct and complete estimate of mutations associated with MH/MHS in the North American population and to observe the distribution of the mutations within the gene. Although the number of MH and MSH individuals we studied is modest, our results indicate that screening of the entire RYR1  coding sequence with DHPLC followed by sequencing is highly efficient: This methodology significantly increased the mutation-detection rate from 22–25% to 70%. Based on our experience, we concur with other investigators involved in studies of large genes such as dystrophin and RYR2  in that DHPLC is a highly sensitive, rapid, and efficient method for mutation screening.18,19,23With all these improvements, there are nine MH/MH-susceptible individuals in which RYR1  mutations have not been found. There may be several explanations for this: (1) involvement of other genes; (2) possible defects in the regulatory and untranslated regions that have not been screened in the mutation-negative individuals; (3) incorrect results of contracture testing; and (4) several additional mutations could be missed because of less-than-ideal DHPLC sensitivity: Most stringent assessments indicate that the lower limit of sensitivity may be between 90 and 98%.19,23,25 

Although functional studies are needed to conclusively define their exact pathogenic effects, the nine newly identified RYR1  missense mutations induce structural and functional damage as suggested by their cosegregation with the MH/MHS status, absence among 100 healthy controls, and their location in areas of high degree of evolutionary conservation. We also consider Val4849Ile substitution as a candidate mutation for MHS because it was associated with central core disease and MHS in three independent studies, including ours, and absent among 250 control individuals of different ethnic backgrounds that we and others have studied.21However, we emphasize that caution must be taken in determining whether newly discovered amino acid changes in the RYR1  gene are candidate mutations for MH/MHS or simply polymorphic codons. Ultimately, functional studies in animal or cell culture models will provide evidence for the pathogenic role of these mutations.

Analysis of the RYR1  coding sequence in its entirety allowed us to identify three novel candidate mutations located outside of the known mutational hot spots that otherwise would not have been detected. The Arg3348His mutation is markedly distant from the mutational hot spots but close to the putative calmodulin binding site.15A recent study described a calmodulin binding site–linked Pro3527Ser RYR1  mutation that caused central core disease.26Our data indicate that RYR1  mutations are distributed throughout the gene, and therefore, it is important to screen the entire coding sequence, particularly the region between the central and C-terminal mutational hot spots, to reliably determine the cause of MHS.

Genetic testing of a large MH-affected family illustrates the effectiveness of the proposed strategy of screening for MHS. The strategy consists of using the remnants of muscle tissue of individuals with positive CHCT results for RNA extraction or using genomic DNA for screening for RYR1  mutations. When the mutation is found and identified, detection-mutation testing based on DHPLC alone, sequencing of a single RYR1  exon amplified from genomic DNA, or restriction analysis alone can be offered to each family member. Almost half of the individuals with positive CHCT results had new mutations in RYR1 , and only two were recurrent mutations. Therefore, because of heterogeneity, the North American MH-susceptible population is likely to have higher mutation diversity or allelic heterogeneity than other more stable populations. A schematic representation of the known RYR1  mutations, including the novel mutations identified in this study (fig. 3), indicates that some mutations are preferentially present in the North American population. Regionally distributed RYR1  mutations have previously been described: The G341R mutation is most frequent in Ireland and England, whereas R614C is more common in mainland Europe, and V2168M occurs with a high frequency in the MH-susceptible population of Switzerland.5,27,28Of several MH/MHS mutations previously identified in the United States, V2214I, A2367T, A2350T, D2431N, and G2347del, only A2350T was found in a single individual in Europe.11,29–31 

In conclusion, the results of this study indicate that defects in the RYR1  gene cause MH/MHS in at least 70% of MH/MH-susceptible individuals, and new mutations can be identified outside the known mutational hot spots. DHPLC methodology is highly accurate, rapid, and efficient in detecting mutations in the entire RYR1  gene and may be used as primary method for mutation screening in MH patients and CHCT-confirmed MH-susceptible individuals and subsequently for mutation detection in members of MH-affected families. The ability to identify mutations in the entire RYR1  coding region would further increase the sensitivity of a genetic test.