• Congenital muscular dystrophies
In 2001, we identified the first mutations in the selenoprotein N gene, SEPN1, in patients with a form of CMD of the "rigid spine muscular dystrophy"-type (RSMD) (Moghadaszadeh et al, 2001). Mutations in the same gene were thereafter shown to cause the classical form of multiminicore myopathy with zones of sarcomeric disarray (Ferreiro et al, 2002); some rare cases of Mallory-body like desminopathy (Ferreiro et al, 2004); and some cases of congenital fiber type disproportion (Clarke et al, 2006). Although patient’s muscle biopsy analyses show the histopathological consequences of these mutations to be very variable, clinical homogeneity of these forms was found, marked by predominantly axial weakness and contractures leading to breathing difficulties, spine rigidity and very early scoliosis. These traditionally distinct disorders should therefore be considered as a unique entity, selenoprotein-related myopathy.
Thanks to the antibodies developed in collaboration with Ulla Wewer (Copenhagen, Denmark), we have shown that selenoprotein N (SelN) is a glycoprotein strongly anchored in the membrane of the endoplasmic reticulum. This protein, whose function is still unknown, is more abundant in foetal muscle than in adult muscle; also, it is expressed more in human myoblasts than in myotubes (Petit et al, 2003). A murine model with complete selenoprotein N deficiency, developed by Alain Lescure (CNRS, Strasbourg, France), should lead to progress in the comprehension of the function of SelN and the physiopathology of selenoprotein-related myopathy.
Selenocysteine (Sec) insertion into selenoproteins is dependent upon the presence of a secondary structure in the 3’ untranslated region of the mRNA, termed SECIS (Selenocysteine Insertion Sequence). In the presence of the SECIS, a UGA codon is recoded to function as a Sec codon. We have identified a unique case of RSMD presenting with a mutation in the non-coding SECIS, leading to a drastic reduction of mRNA and SelN protein levels in the patient’s skin fibroblasts (Allamand et al, 2006). We have also demonstrated the feasibility of repair of a mutation in SEPN1 affecting the selenocysteine-coding UGA codon, using an innovative approach in collaboration with Alain Lescure (CNRS, Strasburg). This is based on the transfection of a mutated selenocysteine tRNA in order to force the recognition of the mutated selenocysteine codon. We demonstrated that this approach allows re-expression of the normal protein in cultured fibroblasts derived from a patient carrying a homozygous nonsense mutation in the selenocysteine codon (Rederstorff et al, 2008).
Complete or partial deficiencies in collagen VI have been observed in muscle biopsies or skin fibroblasts of patients presenting with Ullrich congenital muscular dystrophy (UCMD); this observation constitutes one of the important milestones specific to 2004. Mutations in the COL6A1, COL6A2, and COL6A3 genes, encoding collagen VI subunits, are responsible for UCMD and the related Bethlem myopathy. It is now clear that UCMD and Bethlem constitute a clinical continuum and are now considered as one entity called “collagen type VI-related muscle disorders” (Lampe and Bushby, 2005). In close collaboration with the Functional Unit of Cardio- and Myogenetic (Dr Pascale Richard, GH Pitié-Salpêtrière, Paris), we are pursuing the identification and characterization of the genetic defects in this relatively common congenital muscle disease (Giusti et al, 2005; Squarzoni et al, 2006; Pepe et al, 2006). We are also developing (in part in collaboration with the group of Gillian Butler-Browne, UMRS787, Institut de Myologie) cellular and animal models to further elucidate the pathophysiological mechanisms underlying the disease in order to identify and test novel therapeutic approaches. The recent discovery of the importance of glycosylation anomalies of alpha dystroglycan, which can be detected by immunostaining or Western Blot, has led both to the revision of the diagnosis of a large number of patients and to the discovery of mutations of the Fukutin-related protein (FKRP) gene. Patients typically demonstrate high levels of serum creatine kinase and often pseudohypertrophy, in particular of the calves (Mercuri et al, 2003). Some FKRP mutations induce rapidly progressive congenital muscular dystrophies with or without mental retardation and cerebellar cysts (Brockington et al, 2001; Quijano et al, 2002; Quijano et al, 2005). In particular, we have described founder mutations in Tunisian and Algerian families, associated with a congenital muscular dystrophy carrying an extremely severe prognosis (Louhichi et al, 2004). Different mutations cause only mild forms of limb girdle dystrophy, with the development of cardiac damage at adulthood; these forms can be confused with Becker muscular dystrophy. At present, these patients are followed at the Institute of Myology in order to better characterise the progression of the disease and the cardiac damage associated with the FKRP mutations. Other genes encoding glycosyltransferases or associated proteins (such as POMT1, POMT2, POMGnT1, LARGE, FCMD are also responsible for abnormal O-glycosylation pattern of the alpha-dystroglycan in patients with congenital muscular dystrophy of variable severity and mental retardation (Van Reeuwijk et al, 2006; Yanagisawa et al, 2007; Manya et al, 2008). • Centronuclear congenital myopathies (CNM)
Centronuclear myopathies (CNM) are congenital myopathies, most often with early onset, characterised by the presence of a variable number of muscle fibres with a central nucleus. CNMs are transmitted in an autosomal dominant (and more rarely recessive) mode, but numerous sporadic cases have been observed. With the Muscular Diseases Department of the Institute of Myology, we undertook a programme of careful follow-up of CNM patients and their family members, in order to obtain a precise clinical and morphological characterisation of this disease and to better define sub-groups for genetic studies (
Jeannet et al, 2004).
We gathered large families with a homogeneous clinical and morphological phenotype, and a positional cloning strategy allowed us to identify a locus common to 3 families. Within locus, several heterozygous missense mutations have been identified in the
dynamin 2, a large GTPase involved in endocytosis and membrane trafficking. Expression of DNM2 mutations in cell models proved their causality in the disease (
Bitoun et al, 2005). Since the identification of the first mutations in the DNM2 gene, sequencing of this gene led us to identify 13 heterozygous mutations in 39 CNM families covering the entire clinical spectrum (
Bitoun et al, 2007). A revision of diagnostic and morphological criteria in DNM2-related CNM is in progress, and has already shown a specific pattern of muscle involvement detectable by muscle imaging, providing an important new tool to orientate the molecular diagnosis (
Fisher et al, 2006). In vivo and in vitro studies of the physiopathological mechanisms associated with the dynamin 2 mutations are now in progress.
• Evaluation of the effect of gentamicin and negamycin on stop codon readthrough
In collaboration with the teams of Jean-Pierre Rousset (CNRS, Orsay, France) and Ryoichi Matsuda (Tokyo, Japan), two antibiotics, gentamicin and negamycin, were tested in different models in order to determine their capacity to suppress premature termination codons (PTC).
Gentamicin, an aminoglycoside with long-known termination suppression capacity, and negamycin, a dipeptide antibiotic only available in Japan, were first tested in cultured mouse fibroblast cell lines with a dual reporter system. We showed that these molecules do not suppress all stop codons with the same efficiency (
Bidou et al, 2004;
Allamand et al, 2008). Furthermore, in cultured myotubes from a patient presenting a complete deficiency of the α2 chain of laminin due to a homozygous nonsense mutation in the LAMA2 gene, we demonstrated that although negamycin treatment could induce stabilization of the LAMA2 mRNA, it was not sufficient to promote detectable re-expression of the protein, emphasizing that PTC readthrough is a complex and multi-step process (
Allamand et al, 2008).
• Cardiac arrhythmias
We are continuing our longstanding project on cardiac rhythm disorders of genetic origin, which lead to a high risk of ventricular fibrillation and sudden death. Our aim is to improve clinical and genetic characterisation of these syndromes, and to develop better therapeutic management. The project addresses congenital long QT syndromes, Brugada syndrome, and catecholaminergic polymorphic ventricular tachycardia.
Mutations in the KCNQ1- and HERG-encoded potassium channels are the most frequent causes of
long QT syndrome. We have identified mutations in these genes in newborns and young children affected with long QT syndrome with or without functional auriculo-ventricular block (AVB), an important risk factor in cardiac events. HERG mutations are responsible for the most severe forms, with AVB. KCNQ1 mutations were found in children presenting with bradycardia but without AVB, and are associated with a less severe prognosis. Beta-blocker treatment can be insufficient in HERG mutation carriers, especially in young children, in which case implantation of a pacemaker should be considered (
Lupoglazoff et al, 2004;
Villain et al, 2004). These results represent an advance for the diagnosis and early management of patients.
We have shown that weakly penetrating mutations of the potassium channel KCNQ1 (LQT1) could be more frequent in the general population than has previously been suggested, and may be responsible for rhythm disorders which appear when certain medicines are taken, a condition also called
acquired long QT syndrome (
Gouas et al, 2004). We have identified several SNPs in ionic channels that contribute to the variability of cardiac ventricular repolarization in the normal population and could represent susceptibly factors to arrhythmias (
Gouas et al, 2005;
Gouas et al, 2007).
Transmission of KCNQ1 (LQT1) and HERG (LQT2) mutations was studied in 753 nuclear LQT families through the collaborative effort of 5 European centers. A distortion in the inheritance of LQT1 and LQT2 leads to an excess of affected carriers, especially women, with maternal transmission to daughters rather than to sons. This could be linked either to epigenetic modifications or to unknown mechanisms leading to positive selection of LQTS mutation carrying gametes or LQTS mutation carrying embryos. Increased maternal transmission from mothers to daughters contributes to
female predominance in LQTS (
Imboden et al, 2006).
Catecholaminergic polymorphic ventricular tachycardia (CPVT) is a rare arrhythmogenic disorder characterized by recurrent syncopal events and sudden cardiac death at young age triggered by physical stress or emotion, in the absence of structural heart disease. The molecular basis of CPVT has been partly elucidated by the identification of dominant mutations in the ryanodine receptor 2 (RYR2) or recessive mutations in calsequestrin 2 (CASQ2). We identified the first CASQ2 mutations leading to a complete absence of calsequestrin 2 (Postma et al, 2002), and a number of new RYR2 mutations (
Postma et al, 2005). Because sinus bradycardia was found in all the carriers of RYR2 or CASQ2 mutations, sinus bradycardia in a symptomatic child should be further evaluated by a treadmill test (
Postma et al, 2005). We are a reference center for this disease, and the molecular diagnosis is now performed in collaboration with the diagnostic teams of J. Lunardi (Grenoble) and B. Hainque (Paris). In around 40% of the patients, no mutations are found in the RYR2 or CASQ2 genes, implying further genetic heterogeneity.
Mutations of the SCN5A-encoded cardiac sodium channel are less common than those of the potassium channels. They can be responsible for either the long QT syndrome (LQT3), associated with an increase in sodium channel function linked to the perturbed inactivation of the channel (
Keller et al, 2003); or
Brugada’s syndrome, in which mutations induce a loss of function and result in a reduction in the number of active channels (
Keller et al, 2005,
Six et al, 2008). Patients affected by Brugada’s syndrome present a high risk of syncope and ventricular fibrillation. Treatment with hydroquinidine, an inhibitor of the Ito current, seems to prevent or reduce the number of arrhythmias in patients, whether or not they are equipped with a defibrillator (
Hermida et al, 2004). Only 20% of Brugada’s syndrome cases carry a mutation of the SCN5A gene, and research directed toward characterizing other syndrome-associated genes is in progress in our team.
update : april 2008
