|Home | About | Journals | Submit | Contact Us | Français|
The skeletal muscle ryanodine receptor plays a crucial role in excitation–contraction (EC) coupling and is implicated in various congenital myopathies. The periodic paralyses are a heterogeneous, dominantly inherited group of conditions mainly associated with mutations in the SCN4A and the CACNA1S genes. The interaction between RyR1 and DHPR proteins underlies depolarization-induced Ca2+ release during EC coupling in skeletal muscle. We report a 35-year-old woman presenting with signs and symptoms of a congenital myopathy at birth and repeated episodes of generalized, atypical normokalaemic paralysis in her late teens. Genetic studies of this patient revealed three heterozygous RYR1 substitutions (p.Arg2241X, p.Asp708Asn and p.Arg2939Lys) associated with marked reduction of the RyR1 protein and abnormal DHPR distribution. We conclude that RYR1 mutations may give rise to both myopathies and atypical periodic paralysis, and RYR1 mutations may underlie other unresolved cases of periodic paralysis with unusual features.
The skeletal muscle ryanodine receptor (RYR1) gene on chromosome 19q13.1 encodes the principal sarcoplasmic reticulum (SR) calcium release channel (RyR1), which plays a crucial role in excitation– contraction (EC) coupling by releasing calcium from the SR following muscle activation. RYR1 mutations have been implicated in a wide range of conditions (for review ). Among these are malignant hyperthermia susceptibility (MHS)  and various congenital myopathies including “core myopathies” with central cores (Central Core Disease, CCD) , multiple cores (Multi-minicore disease, MmD) with external ophthalmoplegia  and, rarely, Centronuclear Myopathy (CNM) . Whilst MHS is dominantly inherited, CCD involves both dominant and recessive inheritance. MmD with external ophthalmoplegia is associated with recessive inheritance and quantitative defects of RyR1 protein expression, due to epigenetic allele silencing  or compound heterozygosity for RYR1 null and missense mutations [7,8].
The periodic paralyses are a genetically heterogeneous group of rare, dominantly inherited conditions characterized by episodic weakness and divided into hyperkalaemic periodic paralysis (HYPP, OMIM 170500), mainly caused by SCN4A missense mutations [9,10], and hypokalaemic periodic paralysis (HOKPP, OMIM170400), predominantly associated with mutations in the CACNA1S gene , which encodes the α1S subunit of the dihydro-pyridine receptor (DHPR) or L-type calcium channel in skeletal muscle. The close functional relationship between the DHPR and RyR1 is reflected by their co-localization and linkage within the triad , convergence of associated clinical phenotypes, and findings of abolished EC coupling in animal models lacking either of the two proteins [13,14]. Whilst MHS has been reported in one French family harbouring a dominant CACNA1S mutation , prior studies failed to provide a definitive link between RYR1 mutations and periodic paralysis .
We report a 35-year-old patient with Multi-minicore disease (MmD) with external ophthalmoplegia and additional episodes of atypical periodic paralysis associated with compound heterozygosity for a nonsense and missense mutations in RYR. Detailed neurophysiological and immunohistochemical studies in this patient suggested a dramatic reduction in RyR1 protein expression with resulting marked disruption of the EC coupling machinery.
This 35-year-old female was the oldest child of a healthy non-consanguineous couple of Irish ancestry. Two younger brothers were healthy and her deceased sister had suffered from psychiatric problems. There was no preceding family history of neuromuscular disorders. She presented at birth with hypotonia, respiratory impairment and feeding difficulties following an uneventful pregnancy and normal vaginal delivery at term. Her motor development was delayed and she was unable to walk before 30 months of age. By age 12 she had difficulties rising from supine or squatting positions, managing stairs, was unable to hop or jump, and suffered frequent falls. From the age of 17 years, she had several episodes of spontaneous hip dislocations. Following birth of her first child at the age of 26 years, she experienced increasing problems with reduced walking distance and difficulties with both overhead and fine manual tasks. Exercise-induced myalgia and muscle stiffness prompted by cold, sometimes with weakness affecting the hands, were frequent complaints. There was no cognitive, bulbar or cardiorespiratory involvement.
In addition, from the age of 18 she experienced repeated episodes of severe paralysis initially affecting the legs and then rapidly spreading to the upper limbs lasting up to several days. These episodes were typically preceded by a sensation of dizziness, facial and acral paraesthesia, temporal pain and a visual aura comprising flashing lights, blurred vision and diplopia. Respiratory, bulbar and sphincter functions were usually unimpaired. One episode each was triggered by immersion in a cold swimming pool and another followed administration of lignocaine for a minor surgical procedure, which also prompted marked hyperpyrexia. Her episodes improved following commencement of a potassium-sparing, high sodium diet.
She had four general anaesthetics for surgical procedures and did not experience any MH reaction. Whilst only non-triggering anaesthetic agents were used for the last two anaesthetics, no further details could be obtained regarding the first two anaesthetics administered many years ago in different hospitals. She did not have any in vitro contracture testing to formally assess for malignant hyperthermia susceptibility (MHS).
On a recent examination at 32 years of age, there was facial weakness with incomplete eye closure and limitation of extraocular movements mainly on abduction and upgaze. She had slight nasal escape on speaking and a high arched palate, with weak palatal movements but no dysphonia. There was wasting of the deltoids and distal muscles in the upper limb. There was marked axial weakness on trunk and neck flexion (MRC2), moderate limb girdle weakness (MRC3/4) more pronounced in the shoulder than the hip girdle muscles with additional distal weakness in the upper limbs pronounced in the first dorsal interosseoi (MRC3). There was distal joint hyperlaxity, particularly affecting the hands. She had a hyperlordotic stance but there was no kyphoscoliosis. No action or percussion myotonia was evident. Forced vital capacity was 3.05 l. ECG was normal and an echocardiograph showed no structural or functional abnormalities.
Muscle MRI of the lower limb performed as previously described  showed diffuse involvement with relative sparing of the rectus femoris compared to the vasti, and the anterior compartment compared to the gastrocnemius and the soleus within the lower leg, a pattern suggestive of involvement of the RYR1 gene [4,17].
Concentric needle EMG showed myopathic changes in the left biceps and suggested chronic neurogenic changes in the small hand muscles; there was no electrophysiological evidence of myotonia. Cold immersion testing prompted profound weakness of the hands but normally preserved CMAP responses in the right ADM both pre- and post-cooling, suggesting cold-induced E–C uncoupling; EMG of the cooled small hand muscles showed fibrillations and fasciculations with relatively large motor units. An exercise test for periodic paralysis  performed at 21 years was within normal limits. The long exercise test performed at the age of 30 years was negative but demonstrated an unexpected increase in CMAP amplitude and area post-exercise. Oral potassium loading resulted in profound paralysis more than 12 h after the loading but potassium levels were normal when measured during the episode. A forearm ischaemic exercise test and routine laboratory investigations including CK were normal.
SCN4A and CLCN1 mutations were excluded by genomic DNA sequencing. The entire coding regions (exon 1–106) of the RYR1 gene including splice sites were screened at both genomic DNA and RNA levels. The entire coding sequence of the CACNA1S (NM_000069.2) and CACNB1 (NM_000723.3) gene were also screened by using cDNA synthesized from skeletal muscle as template.
The patient had two muscle biopsies in early childhood and in her teens which suggested a diagnosis of MmD. However, on review of her earlier biopsies appearances were less suggestive of typical MmD associated with mutations in the selenoprotein N (SEPN1) gene [19,20] and because of the diagnostic uncertainty repeat muscle biopsy samples were obtained from the quadriceps (rectus femoris) at 30 years of age.
Cryostat sections were stained with haematoxylin and eosin (H&E), periodic acid Schiff (PAS), nicotinamide dehydrogenase tetrazolium reductase (NADH-TR), succinate dehydrogenase (SDH) and cytochrome oxidase (COX), according to standard procedures . Immunohistochemistry was done by double staining with primary antibodies: mouse anti-RyR1 monoclonal antibody (1:500, Abcam) and goat anti-DHPR polyclonal antibody (1:1000, Santa Cruz). Sections were incubated at room temperature for 1 h, and then washed in phosphate buffered saline (PBS) and incubated 30 min at room temperature with anti-mouse biotinylated secondary antibody (1:200, Amersham). After further washes in PBS, sections were incubated at room temperature for 30 min with streptavidin-conjugated Alexa-488 (1:1000, Molecular Probes) and donkey anti-goat Alexa-594 (1:1000, Molecular Probes). Sections were viewed with a Leica fluorescent DMR microscope (Leica, Germany). Images were captured digitally using Metamorph software (Universal Imaging Inc.) with auto-exposure settings. To further investigate the distribution of the DHPR and RyR1 proteins, sections were also examined by high magnification confocal microscope (Zeiss LSM 710, Germany).
Western blotting of RyR1 protein extracted from patient’s muscle, and routine histopathological and immunohistochemical staining were performed as described previously [21,22]. Band densities are relative to the respective desmin band, used as a control of protein loading. The densitometric analyses were performed using Image J software.
To determine the functional consequences of the p.Arg2939Lys mutation more directly, RyR1-R2939K was constructed, expressed in HEK 293 cells, and crude membrane fractions were prepared as previously described . [3H]Ryanodine binding was used as a measurement of RyR1 activity . Specific [3H]ryanodine binding to membrane fractions was determined for 20 h at 24 °C in 0.25 M KCl, 20 mM imidazole, pH 7 solution containing 5 mM Mg2+, 5mM AMPPCP (a nonhydrozyableATP analog), 5 mM glutathione (reduced) to mimick physiological conditions, 1.5–2.5 nM [3H]ryanodine and the indicated concentrations of caffeine and free Ca2+ .
Voltage clamp studies were performed following expression of wild-type and mutant RyR1-R2939K in RyR1-null myotubes from dyspedic mice as described previously . The whole-cell perforated voltage clamp technique was used to simultaneously measure L-type Ca2+ currents (L-currents) and intracellular Ca2+ transients in expressing myotubes as described previously .
An apparently homozygous nucleotide variation of c.8816G>A was identified in exon 57 of the RYR1 cDNA synthesized from patient’s skeletal muscle biopsy . This maternally inherited missense mutation affects the last nucleotide of exon 57 and caused a substitution of p.Arg2939Lys. The p.Arg2939Lys mutation was not found neither in >200 normal control chromosomes from Caucasian populations nor among more than 50 previously reported MHS/CCD individuals of both Caucasian and Asian origin in whom the entire RYR1 coding sequence had been analyzed [27,28]. Furthermore, the c.8816G>A nucleotide change, apparently homozygous in muscle, was heterozygous at the genomic DNA level (Fig. 1), a discrepancy previously reported in MmD with ophthalmoplegia [4,6,8]. The substituted arginine residue is strictly conserved in all three RyR isoforms and throughout species (Fig. 2). Further sequencing of the RYR1 gene from lymphocyte-derived gDNA from this patient eventually revealed an additional heterozygous c.6721 C>T substitution resulting in a p.Arg2241X stop mutation, and an additional heterozygous c.2122G>A change resulting in a p.Asp708Asn substitution on the same allele. mRNA produced from the allele bearing these mutations was likely affected by nonsense-mediated decay, which would account for the inability to amplify this allele from cDNA. In particular, the heterozygous c.2122 G>A (p.Asp708Asn) substitution was undetectable at RNA level suggesting that both this substitution and p.Arg2241X originate from the same allele that was not transcribed. The p.Asp708Asn substitution changes a charged polar to an uncharged polar amino acid and concerns a highly conserved residue. However, the fact that this substitution was not transcribed at mRNA level makes a pathogenetic role unlikely. Maternal inheritance was excluded for the c.6721 C>T (p.Arg2241X) and the c.2122G>A (p.Asp708Asn) substitutions but no DNA was available from her father and siblings for further segregation studies.
Muscle biopsy performed at 30 years showed increased variability in fibre size and increase in internal nuclei in up to 25% of fibres. Internal nuclei were sometimes central and often multiple (Fig. 3A, a and c). There was no excess endomysial connective tissue. There was predominance of hypotrophic type 1 fibres (Fig. 3A, d). Core-like structures of varying size were apparent with oxidative enzyme stains (Fig. 3A, b) and confirmed on electron microscopy (Fig. 3A, e and f).
Double immunohistochemical labelling of control muscle with monoclonal antibodies to RyR1 and DHPR proteins demonstrated nearly identical distribution pattern of the two proteins as shown by the yellow staining in the merged image (Fig. 3B, f). However, in the patient’s muscle reduced signal from the RyR1 protein (Fig. 3B, b and c) and abnormal accumulation of DHPR protein in a subpopulation of fibres (Fig. 3B, a and c) were observed. On further examination of biopsy sections by confocal microscope, unevenness of stain with the RyR1 antibody was noted in regions corresponding to areas of distinct DHPR aggregation (Fig. 4).
In order to quantify RyR1 protein in this patient we performed Western blot studies on patient’s skeletal muscle. RyR1 protein expression levels assessed in Western blotting analysis revealed a marked reduction of RyR1 protein content in the skeletal muscle biopsy from the patient compared to that from a normal control individual (Fig. 3C). This is in keeping with what we have reported recently in other patients with MmD and ophthalmoplegia . On densitometric analysis, there was only about 20% residual RyR1 protein expressed in skeletal muscle from the patient, less than expected due to nonsense-mediated RNA decay of the allele harbouring the p.Arg2241X substitution alone.
[3H]Ryanodine binding is a sensitive method to evaluate regulation of RyR1 by its endogenous and exogenous effectors . As shown in Fig. 5A, membrane fractions isolated from HEK 293 cells expressing wild type (WT) RyR1 and RyR1-R2939K showed a comparable bimodal Ca2+-dependence of [3H]ryanodine binding. We also probed the effects of caffeine, an exogenous ligand known to augment RyR1 activation . Fig. 5B shows that caffeine increased to the same extent [3H]ryanodine binding to membranes containing WT-RyR1 and RyR1-R2939K. The results suggest that the R2435K mutation did not affect two characteristic functional properties of RyR1, as both Ca2+ dependence and activation by caffeine were not altered.
Since the R2939K mRNA was the only RYR1 transcript detected in the skeletal muscle of the patient, we compared EC coupling in dyspedic myotubes following homozygous expression of either wild type RYR1 or R2939K. The whole-cell perforated patch clamp technique was used to simultaneously assess the magnitude and voltage dependence of depolarization-induced Ca2+ release (upper traces in Fig. 6A; “orthograde” coupling) and L-type Ca2+ current density (lower traces in Fig. 6A; “retrograde” coupling). Peak L-current density (Gmax) and the voltage dependence of channel activation (VG1/2) were not significantly different between WT- and R2939K-expressing myotubes (Fig. 6B and Table S1). Moreover, the magnitude (Fmax) and voltage dependence (VF1/2) depolarization-induced Ca2+ release (Fig. 6C and Table S1) and DHPR–RyR1 co-localization (Supplementary figure) were also not significantly different in WT- and R2939K-expressing myotubes. Thus, the results presented in Figs. 4 and and55 indicate that the conservative R2939K mutation identified in the patient does not directly alter RyR1 channel function.
The distribution of wasting and weakness, histopathologic features and radiological findings on muscle MRI in the patient reported in this paper are identical to those of other patients with Multi-minicore disease (MmD) with external ophthalmoplegia attributed to homozygosity , compound heterozygosity [4,8] or monoallelic expression of heterozygous RYR1 missense mutations due to haploinsufficiency associated with either epigenetic allele silencing  or nonsense-mediated RNA decay . In addition, the patient in this report exhibited clinical features suggestive of atypical periodic paralysis, not previously reported to be associated with mutations in the RYR1 gene.
Despite suggestive symptoms, the precise nature of the periodic paralysis in our patient was difficult to define and her presentation was clearly divergent from that seen in other forms of genetically resolved periodic paralysis. Although clinical features, in particular exertional myalgia, paralytic attacks precipitated by cold or K+-loading and improvement of symptoms with a kaliuretic, were most suggestive of a form of hyperkalaemic periodic paralysis (HYPP), there was no overt myotonia and the long and short exercise tests were negative. Most importantly, measurement of serum K+-levels during a paralysis attack induced by K+-administration were normal, suggesting normokalaemic periodic paralysis, a variant of HYPP, by definition.
Regarding the mutations identified, several lines of evidence indicate their pathogenecity: (1) the c.6721 C>T substitution leads to a premature stop codon, (2) no polymorphisms have been previously reported at position 2939 and the amino acid residue is preserved across different RyR isoforms and species, (3) compound heterozygosity for RYR1 nonsense and missense mutations corresponds to findings in other patients with MmD and external ophthalmoplegia , (4) the patient’s clinical presentation matches that associated with monoallelic expression of paternally inherited missense mutations on a haploinsufficient background due to epigenetic silencing of the maternal allele  and (5) other potential causes of the periodic paralysis phenotype in this patient (e.g. mutations in CACNA1S and SCN4A genes) were excluded.
It seems unlikely that the co-occurrence of MmD and atypical periodic paralysis in the same patient might be attributed to “double-trouble” due to an additional mutation in an as yet unidentified gene, considering the combined clinical, neurophysiological and immunohistochemical evidence for the RyR1-R2939K mutation altering proper RyR1 long-term expression and stoichiometry of the EC coupling machinery in vivo. In particular, the combination of profound weakness but preserved compound muscle action potentials (CMAPs) in the hand muscles following cold immersion testing is unusual in the context of other forms of periodic paralysis and suggests a temperature dependent, functionally relevant disruption of EC coupling in the patient, in spite of normal calcium dependence of ryanodine binding (Fig. 5) and voltage-gated calcium release following transient expression in myotubes (Fig. 6). Altered DHPR–RYR1 stoichiometry in the patient’s muscle is supported by reduced RyR1 expression (Fig. 3), as well as uneven and significant fiber-to-fiber variability in RyR1/DHPR abnormal distribution (Figs. 3 and and4).4). Although episodes of atypical periodic paralysis have not been previously described in other cases of RYR1-related MmD with external ophthalmoplegia, this might be partly explained by a relatively low penetrance of this particular trait. While our patient presented with features of a congenital myopathy at birth, paralytic episodes did not develop until age 19 years and responded favourably to commencement of a potassium sparing diet. Since many of the patients with MmD with external ophthalmoplegia reported to date were infants or young children , the periodic paralysis trait may not have developed at the time those cases were reported. In addition, despite the lack of previously reported paralytic episodes in MmD with external ophthalmoplegia, the marked sensitivity of muscle weakness to cold in these individuals  points to a possible continuum between the congenital myopathy phenotype and the manifestations of atypical periodic paralysis.
Our results indicate abnormal RyR1 protein expression and DHPR–RyR1 distribution in the quadriceps of the patient in vivo, but, rather unexpectedly, normal RyR1 function and RyR1/DHPR co-localization in vitro following transient homozygous expression of R2939K in dyspedic myotubes. This apparent discrepancy could be explained by a slow time-dependent quantitative, rather than qualitative, effect of the mutation, expressed homozygously on a background of already reduced RyR1 protein due to nonsense-mediated RNA decay of the second allele harbouring an allelic RYR1 nonsense mutation. Specifically, rather than disrupting intrinsic RyR1 channel activity, the homozygous p.Arg2939Lys mutation may alter the long-term stability of RyR1–DHPR interaction by affecting RyR1 junctional localization and/or degradation, but unfortunately we were unable to obtain more stable muscle cell lines to investigate this hypothesis further. Although the p.Arg2939Lys mutation is a relatively conservative substitution, the presence of a lysine residue at this site could potentially introduce a novel ubiquitination site that may alter the protein’s long-term stability and/or localization . An effect of the p.Arg2939Lys substitution on long-term protein stability is also supported by the observation of in vivo RyR1 protein reduction to only about 20% of the normal amount, much less than can be explained by nonsense-mediated RNA decay of the allele harbouring the p.Arg2241X substitution alone. Impaired stability of the DHPR/RyR1 complex has also been implicated in the zebrafish relatively relaxed mutant, a model of MmD due to a nonsense mutation in the piscine fast muscle ryanodine receptor isoform ryr1b . Zebrafish relatively relaxed mutants harbouring this mutation display similar findings of markedly reduced DHPR/RyR1 complexes associated with an E–C uncoupling defect, further supporting a causative role of the RYR1 gene in the corresponding human phenotype. As evidenced by the asymptomatic carrier mother of the patient, heterozygosity for R2939K does not appear to be associated with a clinical phenotype.
It remains uncertain whether the association of a congenital myopathy and an atypical form of normokalaemic paralysis observed in our patient is unique or may also apply to other RYR1 mutations. Episodes of hyperkalaemic periodic paralysis have been described previously in a genetically unresolved patient with an eye movement disorder, interpreted as supranuclear gaze palsy, but possibly caused by external ophthalmoplegia in the context of an MmD phenotype . More recently, a causative association between periodic paralysis and a RYR1 substitution, c.7025A>G, was initially postulated in a patient with both MHS and periodic paralysis but later rejected because of identification of the same change in other unrelated patients with MHS but without periodic paralysis . However, as RYR1 screening in these patients was performed at the genomic and not muscle-derived cDNA level, potential monoallelic vs. biallelic RYR1 expression in muscle potentially explaining the discrepancy between phenotypes could not therefore be excluded. “Unmasking” of mutations due to monoallelic RYR1 allele expression has emerged as an important mechanism in the pathogenesis of MmD and ophthalmoplegia [6,8]. Some RYR1 mutations may therefore behave as dominant with regard to MHS, but recessives with regard to congenital myopathy (e.g. MmD) and possible other phenotypes (e.g. atypical periodic paralysis), an hypothesis further supported by the lack of myopathic and paralytic symptoms in the asymptomatic carrier parent of our patient.
Whilst mutations in the SCN4A and the CACNA1S gene have been firmly established as the most common cause of hyper- and hypokalaemic forms of periodic paralyses, a substantial proportion of patients presenting with classic clinical features of these conditions remain genetically unresolved. In a large cohort of 226 patients from 56 families with periodic paralysis , almost 40% had no confirmed SCN4A or CACNA1S mutation. Interestingly, mild fixed weakness between episodes without defined myopathic changes other than the vacuolar pathology commonly observed with CACNA1S mutations and less frequently with SCN4A mutations is suggestive of an associated myopathy in this subgroup of genetically unresolved cases.
Our findings indicate a new, RYR1-related form of periodic paralysis and suggest the RYR1 gene is a candidate for genetically unresolved cases of periodic paralysis, particularly those with symptoms divergent from genetically resolved forms of periodic paralysis and additional features of a congenital myopathy. The characteristic muscle MRI found in RYR1 related diseases could be helpful in raising the possibility of involvement of the RYR1 gene.
We would like to thank the financial support from the Muscular Dystrophy Association USA, the National Institute of Health (AR044657 to R.T.D. and AR018687 to G.M.), and the NSCAG support to the Dubowitz Neuromuscular Centre. HJ has been supported by a grant from the Guy’s and St. Thomas’ Charitable Foundation and REL by a National Institutes of Health Dental and Craniofacial Training Grant (T32DE07202).
Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.nmd.2009.12.005.