Introduction
Clinical myotonia impairs muscle relaxation after voluntary intense contraction. Myotonia congenita (MC) is an inherited myotonia due to mutations in the
CLCN1 gene encoding the skeletal muscle ClC-1 chloride channel (Koch et al.
1992; George et al.
1993). Loss-of-function mutations of ClC-1 channel reduce the sarcolemmal chloride conductance, which, in turn, increases sarcolemma excitability and causes a delayed relaxation manifesting as a clinical and electrical myotonia (Imbrici et al.
2015). From the clinical view, MC patients usually describe muscle stiffness after initiating a forceful movement (Lossin and George
2008). Both dominant and recessive inheritance patterns are found in MC families. Becker myotonia congenita, the recessive form, is typically more severe and has an earlier onset than the dominant one, Thomsen myotonia congenita (TMC). TMC often has a wider range of presentations, including subclinical to moderately severe forms. Consequently, these two entities may be distinguished by inheritance pattern, age at onset, and phenotype (Lossin and George
2008; Heatwole et al.
2013).
To date, more than 200 pathogenic mutations have been reported in the
CLCN1 gene, being widely distributed across the 23 exons or within introns (Pusch et al.
1995; Lossin and George
2008; Mazón et al.
2012; Raja Rayan et al.
2012; Brugnoni et al.
2013; the Human Gene Mutation Database). A number of mutations resulting in premature stop codons are not expected to yield functional proteins, but these may variously affect phenotypes (Richardson et al.
2014). Splicing mutations have also been reported, causing out-of-frame mRNA transcripts that do not produce functional ClC-1 (Ulzi et al.
2014). A number of missense mutations have been functionally characterized by measuring chloride currents generated by mutant channels heterologously expressed in cell lines. These studies have been critical to better understand the relationship between ClC-1 channel structure and function.
Experimental studies have demonstrated that
CLCN1 gene mutations can lead to a positive shift of the activation curve, a reduced chloride ion permeation, an increased cation permeability, an inverted voltage dependence, or a defect in protein trafficking (Imbrici et al.
2015). All these alterations reduce the activity of ClC-1 channel mutants, leading to a reduced sarcolemmal chloride conductance. Functional studies have also revealed possible differences between recessive and dominant mutations. The ClC-1 channel is a homodimer with each subunit forming a single pore; the two parallel pores can gate independently (fast gates), while a common slow gate can close both pores together (Saviane et al.
1999). This peculiar structure may explain the two MC inheritance traits (Pusch et al.
1995). A recessive mutation is expected to induce loss of function of the sole mutated subunit. The coexpression of the recessive mutation with the wild-type ClC-1 results at maximum in a 50 % reduction of the sarcolemmal chloride conductance, which is not enough to cause myotonia. The presence of the recessive mutation in homozygosity or two mutations in compound heterozygosity is required to reduce the sarcolemmal chloride conductance by >50 % and to induce myotonia. In contrast, a dominant mutation is expected to exert an adverse effect on the associated wild-type subunit (the so-called dominant-negative effect), which is sufficient to reduce the sarcolemmal chloride conductance by >50 %, so inducing myotonia. Although dominant mutations may show full penetrance, a dominant inheritance pattern with incomplete penetrance was observed in some pedigrees (Plassart-Schiess et al.
1998). The situation may be even more complicated, since some mutations may be recessive in some pedigrees or dominant in others, suggesting that background modifying factors may greatly contribute to the variability of myotonia.
We report, herein, the clinical, molecular, and functional study of four individuals belonging to three different families affected by recessive MC. In addition, we performed a RT-PCR quantification of selected ion channel subunits expression in muscle biopsies of two MC patients.
Discussion
Myotonia congenita (MC) is a musculoskeletal disorder whose identity firstly emerged with the seminal studies conducted by Bryant and colleagues, who described deficient muscle chloride conductance (Bryant
1969; Bryant and Morales-Aguilera
1971). The first
CLCN1 mutations for both autosomal recessive and dominant MC were discovered in the early 1990s (Koch et al.
1992; George et al.
1993). Although this specific form of myotonia is quite rare, researchers have learnt much more about the clinical, pathogenic, and molecular genetic aspects through various studies performed over the last 20 years (Imbrici et al.
2015). While it is well established that MC pathology is sustained by mutations in the
CLCN1 gene, a number of studies have highlighted a great clinical variability, even among patients carrying the same mutation. To date, more than 200 mutations have been detected in this gene, either in the dominant or in the recessive form of MC (Lossin and George
2008). It has been hypothesized that the clinical variability may be attributed to several factors as different expressivity, incomplete penetrance, impact of mutant alleles on wild-type channel proteins, allelic expression, or intrinsic variability of channel dysfunction (Duno et al.
2004).
Clinical myotonia can stem from many different causes. Natural history, inheritance trait, and disease aspects are relevant clues to distinguish different clinical entities. Age at onset, specific symptoms (i.e., predominant/transient weakness), and temperature effects may help in MC differential diagnosis, as well as neurophysiological or molecular results (Fournier et al.
2004; Lossin and George
2008; Heatwole et al.
2013). In this study, based on these criteria, MC was diagnosed in four patients. The molecular analysis confirmed the presence in these patients of four variants in the
CLCN1 gene, one of which has never been reported and two others have not been functionally characterized yet.
The p.G190S mutation in exon 5 is a known mutation, first described in a large consanguineous Arab family (Shalata et al.
2010), in which heterozygous individuals were asymptomatic or mildly affected, whereas homozygous individuals were severely ill. This mutation also appears widespread in Italy and has already been functionally characterized (Ulzi et al.
2012; Brugnoni et al.
2013; Desaphy et al.
2013). Its main effect consists in a dramatic shift of the open probability voltage dependence toward very positive voltages, resulting in nearly zero chloride current within the physiological voltage range of sarcolemma (Desaphy et al.
2013). It occurs indeed in a well-conserved motif in ClC-1 helix D, which is thought to play a critical role in the chloride ion pathway (Fahlke et al.
2001). In this cohort of MC patients, p.G190S was detected in three individuals from two unrelated families. The absence of neuromuscular disorders in their relatives suggests a recessive mode of inheritance; accordingly, these patients were compound heterozygous carrying also the p.T82A or p.R453W ClC-1 variants. Surprisingly, the latter variants displayed chloride currents very similar to WT channels in mammalian cell lines, in terms of amplitude, kinetics, and voltage dependence, thereby leaving their pathogenicity an open question.
The T82 residue is located at the N terminus in the intracellular side of the channel, far from the conducting pathway and the dimer interface. The residue is quite conserved among species, but shows variability among human CLC protein isoforms (see Online Resource 2). This variant was recently reported in two other Italian individuals and was predicted to be benign using the MutPred software (Brugnoni et al.
2013; Ulzi et al.
2014). Altogether, these observations argue for a likely weak role of T82 in ClC-1 function and, consequently, in the MC pathogenesis. The R453 residue is relatively more conserved among CLC proteins (see Online Resource 2). It is located in the extracellular loop between L and M segments on the extracellular side of the channel. The p.R453W was also previously reported and predicted as possibly disease causative (Brugnoni et al.
2013). Nevertheless, we found no effect of p.R453W on heterologously expressed chloride current properties. Importantly, p.G190S did not appear to exert any dominant-negative effect on p.T82A or p.R453W in coexpression studies, because the chloride currents generated in cells coexpressing p.G190S and the allelic mutant were similar to the computed sum of chloride currents measured in cells transfected with p.G190S alone or allelic mutant alone. Like p.T82A and p.R453W, other MC variants have been shown to produce chloride currents very similar to WT, including p.F167L and p.R105C (Desaphy et al.
2013). In addition, the quantification of
CLCN1 gene transcript in the muscle biopsy of family 1 proband suggests that changes in ClC-1 expression are likely not involved in the determination of myotonia, at least for p.G190S and p.T82A mutations. The mechanism by which such variants contribute to the clinical manifestation of myotonia remains unclear.
The p.G270V mutation was found homozygous in a patient with a positive family history. As previously mentioned, considering that two recessive mutations must be present in a single individual to induce myotonia and that myotonic symptoms were referred only in patient’s mother and maternal grandmother, we hypothesize that his father could have been an asymptomatic carrier of p.G270V, whereas his mother and maternal grandmother may harbor p.G270V associated with another
CLCN1 mutation. Being the proband relatives unavailable for molecular analysis, we are not able, up to date, to establish whether the p.G270V is a cause of a dominant or of a recessive form of MC in this family. However, the early onset and severity of symptoms could suggest a recessive MC. To our knowledge, this is the first report of the p.G270V mutation. The mutation is located in exon 7 and occurs in a well-conserved motif of the transmembrane G helix, close to the conducting pathway (see Online Resource 2). Other neighboring mutations have been associated with myotonia, including p.C271R, p.V273M, p.G276D, p.C277R, and p.C277Y (Fialho et al.
2007; Weinberger et al.
2012; Brugnoni et al.
2013). The last two have been shown to profoundly disrupt ClC-1 channel function (Weinberger et al.
2012). Using MutPred software (
http://mutpred.mutdb.org/), p.G270V was scored with a 0.78 probability to be deleterious. Accordingly, p.G270V drastically shifts the channel voltage dependence in tsA201 cells, which likely accounts for a dramatically reduced chloride conductance and consequent myotonia in the homozygous family 3 proband.
One of the recurrent themes regarding MC is the variable clinical presentation. Among the various hypotheses to explain such a variability, one possibility encompasses the expression of disease modifiers in myotonic patients. Using RT-PCR, we analyzed the expression of selected ion channel subunits involved in action potential propagation and/or previously linked to muscle excitability disorders. Although quantification of ion channel transcripts was performed in only two MC patients, the results suggest that myotonia might be associated with changes in expression of voltage-dependent Na
+ and K
+ channels, and ATP-sensitive K
+ channels. The β1 subunit (
SCN1B gene) may affect the membrane surface expression and voltage dependence of the Na
+ channel α-subunit and possibly of the repolarizing K
+ voltage-dependent channels, thereby modulating cell excitability (Desaphy et al.
2001; Brackenbury and Isom
2011; Marionneau et al.
2012). Strikingly, the auxiliary subunit of voltage-gated K
+ channels, MiRP2, is totally lacking in muscle biopsies of myotonic patients. Coassembly with MiRP2 modifies the voltage dependence of Kv3.4, converting the channel to a subthreshold-activating channel that contributes to skeletal muscle resting potential (Abbott et al.
2001). The KATP channels link metabolism to muscle activity and may exert a significant myoprotective action under prolonged muscle activity, which may occur during myotonia due to delayed relaxation (Tricarico et al.
2006; Flagg et al.
2010). Although the limited number of analyzed biopsies impedes a generalization of the results, we hypothesize that extending such analysis to a larger number of samples and of exploratory genes may provide relevant information to improve our understanding of the myotonia etiopathogenesis and may help in the identification of appealing druggable targets.
In conclusion, the three new
CLCN1 variants can be added to the growing database of MC-associated mutations. Functional studies support the pathogenicity of p.G270V, whereas the mechanisms by which p.T82A and p.R453W may cause the disease remained elusive. Other studies are necessary to definitely classify these two variants as pathogenic mutations (Tang and Chen
2011). In addition, a possible identification of disease modifiers in MC muscle biopsies could help to elucidate the disease mechanisms and broaden therapeutic options. The therapy for patients suffering from MC is at the moment purely symptomatic, consisting in the use of sodium channel blockers such as mexiletine. It is expected that the understanding of the various disease mechanisms linked to
CLCN1 mutations could help the development of targeted drugs with improved efficacy and tolerability.