To determine whether malignant hyperthermia (MH) susceptibility in a Canadian pedigree is associated with a mutation in the ryanodine receptor subtype 1 (RYR1) gene, the complete RYR1 transcript obtained from the leukocytes of one MH-susceptible family member was sequenced, using a newly developed protocol.
RNA was extracted from leukocytes and converted into complementary DNA. Overlapping fragments of RYR1 complementary DNA were amplified by the polymerase chain reaction and used for double-strand sequencing to find a single mutation likely to be causal of MH susceptibility. Inheritance of the mutation in the family was studied by restriction endonuclease analysis and/or sequencing of genomic DNA and compared to available caffeine halothane contracture test data. The mutation was introduced into rabbit RYR1 complementary DNA, the complementary DNA was expressed in human embryonic kidney line 293 cells, and Ca2+ release by the mutant Ca2+ release channel was measured following the addition of caffeine and halothane.
A novel arginine 328 to tryptophan mutation in RYR1 was detected by direct sequencing of the RYR1 transcript from leukocytes of one MH-susceptible individual. A causal role for this mutation in MH is indicated by cosegregation of the mutation with the MH-susceptible phenotype within the family and by the demonstration that the mutant channel has increased sensitivity to both caffeine and halothane.
The feasibility of using complete RYR1 transcripts from leukocytes for sequence analysis offers an efficient and noninvasive method for scanning RYR1 for novel mutations.
MOLECULAR genetic and linkage studies have shown unambiguously that the skeletal muscle ryanodine receptor (RyR1) plays a crucial role in the pathogenesis of malignant hyperthermia (MH) as well as of the related myopathy, central core disease (CCD). Though defects in another gene could influence predisposition to MH/CCD, linkage has been demonstrated only to the genes encoding the α subunit of the sodium channel, SCN4A , and the α-1 subunit of the L-type calcium channel, CACNA1S , in a few families. 1–4By contrast, linkage to RYR1 has been established for many MH/CCD families from various populations. 5,6Reevaluation of a recent analysis of 20 large MH-susceptible (MHS) families, 7based on the assumption that a single recombinant in a family could be explained by the less than 100% accuracy of the Caffeine–Halothane Contracture Test (CHCT), showed that more than 85% of the families investigated could be linked to RYR1 . 8More stringent selection of MHS patients increased the level of mutation identification in RYR1 to 70% of MHS index cases, even if only a part of the RYR1 gene was screened. 9
The total number of RYR1 mutations associated with MHS or CCD now exceeds 50. 9–12The mutation frequencies vary, depending on the MH population investigated, 6and many mutations have been found in only a single MH family. Neomutations in the RYR1 gene have been revealed in approximately 10% of CCD families investigated. 13In light of the high level of allelic heterogeneity in RYR1 , it is clear that molecular genetic testing cannot yet be reduced to the automated detection of a few frequent mutations, but, rather, must involve a complete RYR1 sequence scan.
The search for novel mutations in the RYR1 gene is hampered by the large size of the gene, approximately 160 kb, and its complex structure: the 15 kb RYR1 transcript is built up through the splicing of 106 exons with an average size of 144 base pairs (bp). 14The majority of known MH and CCD mutations, however, occur within three regions of the RYR1 gene: MH/CCD region 1 covers exons 2–17, region 2 covers exons 39–46, and region 3 covers exons 100–104. Consequently, mutation screening is usually limited to these regions of the gene, which comprise less than 30% of the RYR1 transcript. 8,9,13,15
In the accompanying article, 16we reported a method for the amplification of the complete RYR1 transcript from muscle samples as eight overlapping polymerase chain reaction (PCR) fragments. Direct sequencing of these fragments allowed the efficient identification of all polymorphisms present in the transcript. Taking advantage of the fact that B lymphocytes also express RYR1 , 17we demonstrated that the method could be used with some modifications for the amplification and analysis of complete RYR1 transcripts from leukocytes isolated from fresh clinical blood samples.
Here, we report the detailed characterization of a novel Arg328Trp mutation in RYR1 that was detected by the direct sequencing of the RYR1 transcript from leukocytes of one MHS individual. The causative role for this mutation in MH was confirmed by cosegregation of the mutation with the positive MH phenotype within the family and by functional analysis.
Materials and Methods
Following Institutional Review Board (University of Toronto, Toronto, Ontario, Canada) approval and written and informed consent, 30 members, representing four generations, of one Canadian family were available for this study (fig. 1). The proband of the family (301) had a perioperative reaction in 1978, diagnosed clinically as an MH crisis. Although the original medical records were destroyed, remaining records indicate that he experienced generalized body rigidity, an elevated temperature, and an elevated creatine phosphokinase concentration. Five of his relatives (203, 204, 206, 303, 306) underwent CHCT in accordance with the standards of the North American Malignant Hyperthermia Group protocol. 18The CHCT results are listed in table 1.
Because muscle samples were no longer available from any of the family members tested, this family was chosen for analysis by RYR1 sequencing from leukocyte messenger RNA.
Blood collection from individual 303, RNA extraction from leukocytes, and subsequent RYR1 transcript sequencing were performed exactly as described in the accompanying article. 16Genomic DNA was also extracted from blood using QIAamp DNA Blood Midi Kit (QIAGEN Inc., Mississauga, Ontario, Canada).
To screen for the presence of the C982T mutation in genomic DNA, a 395-bp fragment of the RYR1 gene spanning the part of exon 11 in which C982 is located and part of the adjacent intron 11 was PCR amplified from genomic DNA samples using forward (5′-GGAGTGGGAGCCACCTGCGCT-3′) and reverse (5′-CCACAGTCCTGAGGCCACATGCTGC-3′) primers. PCR was performed in 50 μl of 1X PCR Gold Buffer (Applied Biosystems, Foster City, CA) with 150 ng genomic DNA, 5% dimethyl sulfoxide, 0.3 μm of each primer, 0.2 mm deoxynucleotide triphosphates, 2.5 mm MgCl2, and 1.5 units of AmpliTaq Gold polymerase (Applied Biosystems). Cycling parameters were as follows: denaturation at 95°C for 10 min, five cycles of touchdown from 70°C to 60°C with denaturation at 95°C for 15 s, and annealing–extension at 70°C to 60°C for 30 s followed by 25 cycles of 95°C for 15 s, annealing at 72°C for 30 s, and a final extension step at 70°C for 10 min, held at 4°C.
Polymerase chain reaction products were purified using QIAquick PCR purification kit (QIAGEN Inc.) and screened for the presence of the C982T mutation, either by direct sequencing or by restriction fragment length polymorphism pattern analysis after digestion with the restriction endonuclease Aci I (New England BioLabs, Beverly, MA). The restriction fragments were resolved in 10% polyacrylamide gel and stained with SYBR Green 1 Nucleic Acid Gel Stain (Roche Diagnostics GmbH, Basel, Switzerland).
The functional effect of the C982T mutation was studied by comparing the effect of caffeine and halothane on Ca2+release of wild-type and mutant RyR1 proteins expressed in human embryonic kidney line 293 cells essentially as described. 19,20Briefly, caffeine and halothane sensitivity was measured by intracellular calcium fluorescence photometry at 340/380 nm using Fura-2-AM indicator; dose–response curves were constructed by measuring peak amplitudes of the fluorescence ratio (340/380 nm), which, after background subtraction, were normalized against a maximal response produced by 10 mm caffeine. Data normalization was necessary to compare individual samples to the response of halothane due to different samples expressing ryanodine receptor proteins at different levels. For each drug concentration, data from six independent experiments were averaged and expressed as mean ± SD. Linear regression analysis was performed using Origin software (Microcal Software Ltd., Northampton, MA). Data were compared using the unpaired Student t test, with statistical significance accepted at P < 0.05. 20
Genetic linkage analysis was performed using LINKAGE (Jurg Ott, Rockefeller University, New York, NY) version 5.1 and FASTLINK (Alejandro Schäffer, National Institutes for Health Research, Bethesda, MD) packages. 21,22The genetic model that was used assumed autosomal dominant inheritance, an affected frequency of 1:10,000, near-complete penetrance of 0.98, a low phenocopy rate of 0.02, and a recombination fraction of 0. Statistical significance favoring linkage was accepted for logarithm of odds scores of 3.0 or greater.
We determined whether MH susceptibility in this Canadian MH pedigree is associated with a mutation in the RYR1 gene by sequencing the RYR1 transcript from the leukocytes of one MHS family member. Direct sequencing of the complete RYR1 transcript revealed the presence of three polymorphisms: two synonymous G/A substitutions and one novel mutation—a heterozygous C982T transition resulting in the substitution of Arg328 with Trp (R328W) (table 2). As in two thirds of other known MH mutations in RYR1 , 23the C982T mutation occurs at a C residue in a CpG dinucleotide. Transcript heterozygosity for the C982T transition in this MHS patient (303) was confirmed by direct sequencing of a 395-bp PCR-amplified fragment spanning part of exon 11 and the adjacent part of intron 11 in genomic DNA (fig. 2A).
Direct sequence analysis of the corresponding RYR1 region from genomic DNA of the proband (301) and the three family members diagnosed as MHS by CHCT (203, 204, 206) showed the presence of the C982T transition in a heterozygous state. The mutation was absent in the one family member (306) who tested MH negative by the CHCT.
To extend our study to additional members of the family, we developed a simple diagnostic assay based on the fact that the C982T base substitution in RYR1 leads to the loss of an Aci I restriction site in the sequence of exon 11. In individuals with the normal allele, Aci I endonuclease digestion cleaves the 395-bp PCR fragment containing two constant Aci I restriction sites into three fragments of 173, 137, and 85 bp; the same fragment with the C982T base substitution is cleaved by AciI endonuclease into fragments of 258 and 137 bp. In the heterozygote, four bands of 258, 173, 137, and 85 bp form a characteristic pattern that is diagnostic (fig. 2B).
Screening for the presence of the R328W mutation within the four generations and 28 individuals that form the family (fig. 1) revealed six heterozygous individuals (102, 304, 401, 403, 406, 408) and16 homozygous normal family members (including three presumed normal spouses). The mutation was absent from 95 unrelated healthy individuals (190 normal chromosomes), indicating that the R328W mutation is not a benign polymorphism.
Arg328 is located in a region of RyR1 protein that is conserved across vertebrate species (fig. 2C). Further evidence for a causative role of the R328W mutation in MH was sought in functional studies. The R328W substitution was incorporated into the corresponding position in a rabbit RYR1 cDNA. Wild-type and mutant cDNAs were expressed in HEK-293 cells, and intracellular Ca2+release, induced by caffeine or halothane, was measured by photometry. 19The R328W mutant Ca+2release channel was significantly more sensitive to caffeine (ED50= 0.80 mm, where ED50refers to the concentration of caffeine or halothane required to reach the half-maximal Ca2+release response) than wild-type RYR1 (ED50= 1.40 mm) and more sensitive to halothane (ED50= 0.16 mm) than wild-type RYR1 (ED50= 0.4 mm) (fig. 3). The higher sensitivity of the expressed mutant protein to caffeine and halothane indicated the pathogenic potential of the R328W mutation.
The Arg328Trp mutation was not present in other unrelated Canadian individuals who had been diagnosed as MHS by the CHCT, nor is it listed among previously reported MH causative mutations. 8,24On the basis of its association with MH and its functional characteristics, we conclude that the C982T transition is a private MH mutation, restricted, as yet, to this Canadian pedigree.
In this study, we identified a novel mutation, R328W, in a large MH family by direct sequencing of the complete RYR1 transcript amplified from mRNA obtained from peripheral blood leukocytes, thus demonstrating that blood can be used as an alternative source of RYR1 mRNA for genetic studies. In this family, we were unable to demonstrate genetic linkage to a satisfactory level of statistical significance (lod score favoring linkage = 1.50) because the power of genetic linkage was limited by the small number of subjects phenotyped by CHCT and the small number of informative meioses. Phenotyping was offered to members of the family, but further testing was declined. Sensitivity analysis in this pedigree showed that odds favoring linkage might have been observed if phenotyping had been performed on eight members of the youngest generation in the pedigree. However, discordance between genotype and CHCT is also likely, given the known limitations in specificity of contracture testing to rule out MHS. 25,26However, the perfect correlation between the presence of the R328W mutation and positive CHCT results within the family as well as the absence of the mutation in a control population of 190 normal chromosomes lends support to the causal nature of the mutation for MHS in this Canadian pedigree.
Further support to the causal nature of this mutation comes from functional studies that demonstrated that the expressed mutant protein was significantly more sensitive to caffeine and to halothane than wild-type RyR1. The effect of the R328W mutation on the sensitivity of RyR1 to caffeine and to halothane was comparable to the effect of several other RyR1 mutations, now recognized as causal of MH. 19
The C982T transition is located within MH/CCD region 1, one of the three mutational hot spots in the RYR1 gene. 27Alignment of the amino acid sequence of the human RyR1 protein with sequences from rabbit, mouse, and pig, as well as with orthologs in bullfrog and fish, shows the conservation of Arg328 across vertebrate species (fig. 2C). Alignment of the same protein sequence with other ryanodine receptor isoforms shows that Arg328 is highly conserved between RyR1 and RyR2 proteins. Arginine is replaced in RyR2 protein with lysine, but this is a conservative substitution. In our MH pedigree, the hydrophilic, positively charged residue, arginine, is replaced by a hydrophobic tryptophan residue, a substitution that is likely to have a critical effect on RyR1 function and, thus, be the cause of MH susceptibility within this pedigree. Accordingly, we believe that the location of the mutation in a mutational hot spot in RYR1 previously linked to MHS, and the altered Ca2+release channel function is sufficient evidence to recommend that the R328W mutation in RYR1 be included in the European MH group list of RYR1 mutations considered to be causal of MH sensitivity. 24
The search for a noninvasive method for MH diagnosis has led to the proposal that alterations in Ca2+homeostasis can be studied either in cultured skeletal muscle cells 28or in lymphoblastoid cells naturally expressing RYR1 . 29Such diagnostic methods, however, do not identify causal mutations. The procedure described here is compatible with the isolation and sequencing of RYR1 transcripts from cultures of small numbers of human cells. Selective sequencing of the 1- to 2-kb PCR fragments corresponding to three MH domains in the transcript offers an even more economical alternative to exon-by-exon scanning of genomic DNA. Regardless of the origin of the tissue, transcript analysis offers a cost-effective way for a complete mutation scan in the RYR1 coding sequence. While this approach cannot identify mutations at the exon–intron junctions, it can readily detect splicing aberrations that result from such mutations. As yet, there is no indication that such mutations are associated with MH.
We believe that the search for MH-associated mutations within an MH family should be limited neither to known MH mutations nor to the MH/CCD regions of the RYR1 gene because several MH- or CCD-linked mutations have now been identified outside these regions. 30,31Direct sequencing of the RYRI transcript offers the highest potential for finding all MH mutations. Until now, sequencing of the entire RYR1 coding sequence has not been feasible because of the cost and the complexity of the gene, as well as the fact that muscle tissue is not always available. With the protocol that we have described, sequencing of the entire RYR1 transcript from a sample of blood is now feasible and could be a method of choice for genetic testing that is compatible with one of the developing “noninvasive” methods of diagnosis of MH status.