Brain Advance Access originally published online on September 28, 2006
Brain 2007 130(2):368-380; doi:10.1093/brain/awl270
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
A novel autosomal recessive limb-girdle muscular dystrophy with quadriceps atrophy maps to 11p13–p12
1 Neurogenetics of Locomotion Laboratory, Centre de Recherche du Centre Hospitalier de l'Université de Montréal Montréal, Quebec, Canada 2 Département de radiologie, Centre Hospitalier de l'Université de Montréal Montréal, Quebec, Canada 3 Hôpital Ste-Justine, Montréal Montréal, Quebec, Canada 4 CHA-Hôpital Enfant-Jésus, Université Laval Quebec, Canada 5 CHUS, Université de Sherbrooke Sherbrooke, Quebec, Canada
Correspondence to: B. Brais, Centre for the Study of Brain Diseases, Centre de Recherche du Centre Hospitalier de l'Université de Montréal, Montréal, Québec, Canada E-mail: Bernard.Brais{at}umontreal.ca
 |
Summary |
|---|
 Top Summary IntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Limb-girdle muscular dystrophies (LGMD) are a heterogeneous group of pathologies. We have identified a cohort of 14 French–Canadian patients from eight different families displaying a novel form of LGMD with an autosomal recessive inheritance. These patients share some features with previously described cases of ‘quadriceps myopathy’ that evolved into an LGMD. All demonstrate quadriceps femoris asymmetrical atrophy. Creatine kinase values were variable from normal to 6000 U/l. Clinical evaluations and MRI studies demonstrate a variable intrafamilial and interfamilial phenotype. Asymmetrical muscle involvement was clinically observed and confirmed by imaging. MRI studies suggest that the hamstrings and the adductor magnus are the first limb muscles to demonstrate fatty infiltration. Muscle pathology shows no sign of active inflammation but increased endomysial connective tissue associated with basal lamina duplication and collagen disorganization. A genome-wide scan using the two largest families uncovered linkage to marker D11S1360 on chromosome 11p12 [multipoint logarithm of the odds (LOD) score of 2.78]. Further genotyping for the eight families confirmed linkage to this new LGMD locus (multipoint LOD score of 4.56). Fine mapping subsequently defined a less than 3.3 cM candidate interval on 11p13–p12. Haplotype analysis of carrier chromosomes suggests that the most frequent mutation may account for up to 81.3% of French–Canadian mutations. In this study, we describe the chromosomal locus of a new form of recessive LGMD with prominent quadriceps femoris atrophy.
Key Words: autosomal recessive limb-girdle muscular dystrophy; quadriceps atrophy; genome-wide scan; linkage analysis
Abbreviations: CK, creatine kinase; EMG, electromyogram; IBM, inclusion body myositis; LGMD, limb-girdle muscular dystrophy; LOD, logarithm of the odds; TRIM, tripartite motif
Received June 27, 2006. Revised August 11, 2006. Accepted August 23, 2006.
|
Introduction |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Limb-girdle muscular dystrophies (LGMD) are defined as a weakness and wasting of the limb-girdle muscles, with typical sparing of the facial muscles (Wicklund, 2003
). The various forms of LGMD have been classified based on their mode of transmission, their mutated gene or chromosomal locus. To date, 7 autosomal dominant forms (LGMD1A–G) and 11 autosomal recessive forms (LGMD2A–K) have been characterized. The different autosomal recessive LGMDs (AR-LGMDs) display some epidemiological or clinical differences that may help their clinical distinction (Bushby, 1999). Four AR-LGMDs have been found to exist more in specific populations: LGMD2G in Brazil, LGMD2H in the Manitoba Hutterites of Canada, LGMD2J in Finland and LGMD2I in Denmark (Zatz et al., 2003; Sveen et al., 2006). Variable muscle involvement is present in the different LGMD. Quadriceps femoris atrophy has been described in LGMD2A, LGMD2B, LGMD2D, LGMD2H and LGMD2I (Fischer et al., 2005), but it is more frequent in the latter three forms. Only in LGMDH is it a constant finding (Table 1).
|
Cases of quadriceps myopathy have been described in the literature at least since 1922 (Bramwell, 1922
). In the oldest reports, the disease seems to have been restricted to the quadriceps muscles with weakness and atrophy not spreading to other muscle groups (Denny-Brown, 1939; Walton, 1956; Turner and Heathfield, 1961; Mümenthaler, 1970; Boddie and Stewart-Wynne, 1974). Later observations suggested that in some cases it evolved into an LGMD (Walton, 1956). In 1974, Boddie and Stewart-Wynne suggested that quadriceps myopathies represent a clinical syndrome with heterogeneous pathological bases rather than a specific entity (Boddie and Stewart-Wynne, 1974). Strikingly, many of the cases described presented biopsy and electromyogram (EMG) findings suggesting both active dystrophic and neurogenic processes (Serratrice and Munsat, 1995). While most of these early described cases appear to be sporadic, some familial cases have been described. In 1970, Mümenthaler described two brothers with a shared quadriceps myopathy (Mümenthaler, 1970) and Furukawa et al. (1997) described a case whose parents were first-degree cousins (Furukawa et al., 1977). In 1973, Espir and Matthews published a study of a dominant quadriceps myopathy affecting a father, his three daughters and one of his brothers (Espir and Matthews, 1973). Sunohara et al. (1990) described four male patients with quadriceps myopathy, all of whom showed a mild and slowly progressive myopathy clinically confined to the quadriceps muscles but who, on careful EMG and histological examinations, had more widespread muscle involvement (Sunohara et al., 1990). Dystrophin testing by immunofluorescence studies and immunoblotting were compatible with a diagnosis of Becker muscular dystrophy. The authors concluded that although the ‘quadriceps syndrome’ may result from a variety of heterogeneous diseases, at least a subset of these patients in fact have Becker muscular dystrophy. The observation that the majority of cases of quadriceps myopathy are men was underlined by Munsat and Serratrice (1995). In 2003, Charniot et al. described the autosomal dominant segregation in a French family of a quadriceps atrophy associated with a severe dilated cardiomyopathy with conduction defects or atrial/ventricular arrhythmias (Charniot et al., 2003). Cardiac involvement preceded neuromuscular disease in all affected patients. They detected an Arg377-to-His mutation in the lamin A/C protein (LMNA). This mutation had been reported in limb-girdle muscular dystrophy type 1B, a slowly progressive LGMD with age-related cardiac conduction disturbances and the absence of early contractures. These latter two families further support the conclusion that quadriceps atrophy can indeed be found in different genetic conditions. In fact, quadriceps atrophy has been mostly associated with sporadic inclusion body myositis (IBM) (Phillips et al., 2001). This sporadic inflammatory myopathy presents with an insidious onset of slowly progressive proximal and distal weakness and atrophy, particularly affecting the quadriceps and forearm muscles (wrists and finger flexors) and ankle dorsiflexors. Unlike this sporadic form, hereditary IBM, which links to 9p12–p11 and is caused by mutations in UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase (GNE), is known to spare the quadriceps entirely (Eisenberg et al., 2001). On the other hand, in autosomal dominant IBM3, caused by mutations in the MYHC2A gene, quadriceps atrophy is a prominent feature (Darin et al., 1998). In conclusion, most authors believe that quadriceps myopathy is not a single entity but may be found in different disorders. This study is the first to describe a group of patients affected by a recessive LGMD associated with prominent and asymmetrical quadriceps femoris atrophy linked to 11p13–p12. |
Material and methods |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Clinical assessment
Familial cases were recruited at the Notre-Dame and Hôtel-Dieu hospitals of the Centre Hospitalier de l'Université de Montréal (CHUM). We identified 14 patients (9 males and 5 females) aged 27–68 years old and belonging to 8 families. Neurological examination was performed by two neurologists experienced in neuromuscular disorders (M.F.R., B.B.). First-degree relatives were examined when possible, and genealogical and geographical data were collected by a research nurse (L.L.). Patients were included in our cohort if they presented with weakness of limb-girdle muscles and atrophy of the medial part of the quadriceps femoris. Serum creatine kinase (CK) levels were measured in all affected individuals and most of their first-degree relatives. Previous EMG results were reviewed and histopathological data were gathered on muscle biopsies on 11 cases. Patients signed an informed consent approved by the CHUM's ethics committee.
MRI
Muscle MRI was performed on a 1.5-tesla Siemens Avanto system (PA, USA). Axial and coronal planes of the thighs and upper arms were obtained, positioned to include part of the pelvic and shoulder girdles, respectively, using conventional T1-weighted spin echo (repetition time TR of 500 ms, echo time TE of 13 ms, with slight variations) and short tau inversion recovery sequences. Fat-saturated T1-weighted images post-intravenous gadolinium injections were obtained for some patients only. The slices were 10 mm thick for the lower extremities and 6–10 mm thick for the upper extremities, with an interslice gap of 6–10 mm. The T1-weighted images were evaluated with regard to degree of muscle atrophy (volume loss) and intramuscular fatty infiltration.
Fluorescent and electron microscopy
A muscle biopsy on case IX-15's quadriceps femoris was obtained for diagnostic purposes in 1999. He signed an informed consent to allow us to review the sample in the context of this project. The biopsied muscle was frozen in liquid nitrogen after removal and stored at –70°C until use. Routine histological and histochemical procedures were done, with staining for H & E, HPS, Gomori trichrome, PAS, ATPase 9.5, ATPase 4.6, ATPase 4.3, NADH, glycerol phosphate dehydrogenase and acidic phosphatases. Isoforms 1, 2 and 3 of dystrophin, 
-sarcoglycan (Novocastra, Newcastle-upon-Tyne, UK) were processed according to the indirect immunoperoxidase technique of Sternberger (1969). DAB was used as a chromogen. Case IX-15's muscle biopsy was processed according to usually prescribed techniques, which involved inserting a portion of the specimen on a pair of forceps for gluteraldehyde fixation. After fixation in 3% gluteraldehyde in cacodylate buffer, specimens were stained with saturated uranyl acetate and embedded in Epon. Numerous semi-thin sections stained by paraphenylene diamine were screened by the Nomarski method (Schindl and Rueker, 1973). Representative thin sections were examined with a Phillips EM 208 S electron microscope (Fei Electron Optics BV, Eindhoven, The Netherlands). No other biopsy was examined in the context of this study, only available reports were reviewed.
Exclusion of the LGMD2H and HIBM loci
Genomic DNA was extracted from blood samples according to previously described methods (Zelinski, 1991). Linkage analysis to the LGMD2H locus was performed using standard techniques as will be described under ‘Linkage analysis’ using four selected polymorphic STR markers (D9S241, D9S170, D9S154 and D9S737) surrounding the TRIM32 gene. Screening for mutations in the GNE gene mutated in hereditary IBM (HIBM) was conducted by sequencing the exons and exon/intron boundaries, as will be described under ‘Sequencing of candidate genes’. Primers were designed using the ExonPrimer tool (http://ihg.gsf.de/), based on the sequence provided by the UCSC genome browser.
Genome-wide scan and fine mapping
A genome-wide scan was performed on six affected cases (VIII-3, VIII-4, VIII-6, IX-10, IX-11 and IX-15) and four unaffected relatives (VIII-1, VIII-5, IX-12 and IX-14) at deCODE Genetics (Reykjavik, Iceland). Genotypes were generated for 500 polymorphic microsatellite markers separated by an average of 8 cM. The main region of interest was fine mapped using known microsatellite markers and markers designed based on the genomic sequence of the region (http://genome.ucsc.edu, May 2004 assembly). Three oligo pairs were designed using Primer3 (Rozen, 2000), based on the genomic sequence (http://genome.ucsc.edu, May 2004 assembly): EEJ11AC19 (forward primer: TTG CTT TCA TAT GGA TGC TGT; reverse primer: TTG CTG CAT TCA CCA ATA GC), FFJ11TA27 (forward primer: TCC ATG ACC TCT GGG AAG G; reverse primer: CAC CAG CTG GAC CTG TCT TA) and GGJ11GA24 (forward primer: GCA AAA ACA TGC TGG TGG T; reverse primer: CAC ACA TCT AGG GCT GGT GA). Oligonucleotide primers were synthesized at Invitrogen (CA, USA). Fragment analysis was performed on a LI-COR Gene ReadIR 4200 (NE, USA).
Linkage analysis
Multipoint autosomal recessive parametric linkage was computed using GENEHUNTER v.2.1 (Kruglyak et al., 1996). Phenocopy number was set to zero. Six liability classes were defined based on the reported age of onset: 0–19 years (18%), 20–29 (29%), 30–39 (65%), 40–49 (82%), 50–59 (94%), and 60 and over (100%). Allele frequencies were considered equal for most markers. Allele frequencies provided by deCODE Genetics and based on genotypes of 186 chromosomes from French–Canadian individuals not participating in this study were used for the markers of the genome scan. One cM was assumed to be equivalent to 1 Mb. Two-point linkage analysis of X-chromosome markers was computed using MLINK of LINKAGE (Lathrop and Lalouel, 1984; Lathrop et al., 1984, 1986). The haplotypes were reconstructed using the MAXPROB method of GENEHUNTER.
Sequencing of candidate genes (TRIM44, TRAF6 and LOC119710)
The entire coding and 30 bp flanking intronic sequences for the three genes were amplified for mutation analysis. Primers flanking exons to be sequenced were designed using Primer3 (Rozen, 2000). The first exon of TRIM44 and the last exon of TRAF6 were divided into three overlapping fragments due to their size. The PCR products and primer pairs were sent to the McGill University and Génome Québec Innovation Centre for forward and reverse sequencing. Sequences were aligned using SeqMan 4.03 (DNAStar, WI, USA) and analysed using Chromas 1.62 (Technelysium Pty Ltd, Australia).
|
Results |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Clinical phenotype
Fourteen patients aged between 27 and 68 and belonging to eight families were examined (Fig. 1). All eight families are of French–Canadian descent. The percentage of affected cases in these families is compatible with a recessive mode of inheritance (36%, 14 cases/39 siblings, 9 males/5 females), though we observed more cases than expected in our 2 largest families (3 out of 4 in family VIII and 5 out of 8 in family IX). In all families, there is no history of a similar condition in parents. All cases show a limb-girdle type of muscular dystrophy with prominent biceps brachii and quadriceps femoris atrophy (Fig. 2A and F). The latter is particularly prominent at the level of the medial distal thigh (Fig. 2F). There is a wide range of age of onset (mean 32.7 years, 11–50 years). Case IX-9 even claimed to be asymptomatic at the age of 68 despite observed weakness on examination and elevated CK of 1649 U/l (Table 2). Contractures are seen in only one patient (7.1%), case V-7, who has been wheelchair bound for the past 4 years: they were present at the level of the elbows, wrists, fingers, hamstrings and ankles. Muscular pain, however, is reported in the majority of our patients (85.7%), with seven patients complaining of myalgia following exercise (50.0%). Mild calf hypertrophy (IV-3, VI-5, VII-3) and facial weakness (IV-3, VII-3) are rarer findings. Scoliosis was not observed in any of our patients. Only VII-3, aged 58, has a known cardiomyopathy with a left ventricular ejection fraction of 20%, which is probably related to his coronary heart disease. Four patients out of fourteen (28.6%) are wheelchair bound. The mean age of wheelchair use is 44.3 years, on average 12 years after diagnosis. In our two largest families, males are clearly more affected than their sisters (Table 2). Furthermore, the four brothers of family IX demonstrate a very variable degree of involvement. The variability of the phenotype extends to the level of the individuals, with clear asymmetrical involvement of the same muscles.
|
|
|
Laboratory findings reveal mildly to moderately elevated serum CK levels in 10 (71.4%) of our patients (mean: 1290, 254–6000 U/l, with a normal range of 37–217) (Table 2). Four patients (28.6%) have normal serum CK: these are V-7, VII-3, VIII-3 and VIII-4. Previous EMG reports were available on eight patients. These showed normal motor and sensitive nerve conductions but documented myopathic changes with small amplitude polyphasic potentials (Table 2). Five patients (35.7%) also showed minor signs of possible neurogenic changes: increased insertional activity in one patient; mild spontaneous activity with positive waves and fibrillations in another patient; and important spontaneous activity with positive waves, fibrillation, and repetitive discharges and high amplitude motor unit potentials associated with reduced recruitment at maximal effort in patients IX-10, IX-11 and IX-15.
MRI
Inspection of muscles revealed a more pronounced atrophy of the biceps brachii and the medial part of the quadriceps femoris (Fig. 2A and F). MRI was performed on four patients. Cases VIII-6, IX-11 and IX-15 were chosen because they are all moderately to severely affected and belong to two different families (VIII and IX): VIII-6, a 37-year-old, severely affected man; IX-11, a 65-year-old man with a somewhat milder phenotype; and IX-15, a 47-year-old man similarly affected. We chose to study case IX-9 as well, the asymptomatic 68-year-old sister of IX-11 and IX-15. She had elevated CKs (1649 U/l) documented since her forties and on examination had mild biceps brachii atrophy and weakness, as well as mild iliopsoas and quadriceps weakness.
In the upper extremities, case IX-9 showed no clear atrophy (Fig. 2B). Her brother, IX-15, displayed normal brachio-radialis and triceps, but a mild atrophy and fatty replacement was noted for the long head of the right biceps brachii with a moderate atrophy and fatty replacement in the long head of the left biceps (Fig. 2C). Patient IX-11 revealed similar findings, with a mild atrophy and fatty replacement in the long head of the left biceps and moderate atrophy and fatty replacement in the long head of the right biceps (Fig. 2E). Individual VIII-6's images revealed a marked atrophy and fatty infiltration of the biceps brachii and brachio-radialis accompanied by a minimal atrophy of the triceps (Fig. 2D). The right side was more affected than the left, which corresponded to the observed asymmetry on physical examination.
In the lower extremities, patient IX-9 displayed no clear involvement of the quadriceps, but an asymmetrical, mild to moderate atrophy and fatty replacement of the hamstring muscles (biceps femoris and semitendinosus) and of the adductor magnus were observed (Fig. 2G). Case IX-15 presented with severe atrophy and fatty replacement of the adductor magnus bilaterally (slightly less severe on the left), of the right vastus medialis and of the short head of the left biceps femoris (Fig. 2H). Moderate atrophy was also noted for the tensor fasciae latae bilaterally. The adductors brevis and longus, iliopsoas, gracilis, sartorius and glutei were normal. His brother, patient IX-11, showed a severe atrophy of the adductor magnus, the hamstrings, the tensor fasciae latae, the gracilis, and the glutei minimus and medius bilaterally (Fig. 2I). Moderate atrophy of the vastus lateralis, intermedius and medialis on the right was also noted, with a mild wasting of the right adductor longus. Finally, case VIII-6 revealed a severe atrophy of the medial part of the thigh (Fig. 2J) which corresponds to a moderate to severe atrophy of the vastus intermedius, the vastus medialis, the vastus lateralis and the semimembranosus. Severe atrophy of the left adductor magnus and gluteus minimus bilaterally was present. Mild atrophy was found of the gluteus maximus and medius, rectus femoris, tensor fasciae latae, and remainder of the hamstrings group. No involvement of the piriformis, iliopsoas, sartorius and gracilis was documented.
Therefore, in the lower extremities, atrophy seems to predominate in the hamstring muscles, the adductor magnus and the medial part (vasti medialis and intermedius) of the quadriceps femoris, while the biceps brachii seem to be affected in most cases in the upper extremities. The clear asymmetry documented clinically in all four patients in the upper and lower extremities was well correlated with the MRI findings.
Muscle histology
At this stage of our research, only case IX-15 has been thoroughly studied. However, previous muscle biopsy reports were available for 11 of our patients (78.6%) (Table 2). Biopsies from nine of these individuals were described by different pathologists as displaying dystrophic changes with variation in fibre sizes, degeneration and regeneration of muscle fibres, internal nuclei, and notable increase in endomysial connective tissue and fibre splitting. In patients IV-3 and V-7, neurogenic changes were seen with minor group atrophy and some angular fibres. Mild focal inflammation was reported only in the muscle of the more severely affected case, VIII-6. Available muscle on patient IX-15 was reviewed by an experienced neuropathologist (Y.R.). Deficiencies in dystrophin, dysferlin and sarcoglycans were excluded by immunohistochemistry and western blots (data not shown). Electron micrographs on IX-15's biopsy revealed focal duplication of the muscle basement membrane (Fig. 3C). Endomysial extracellular matrix is increased and disorganized, as several collagen protofibrils can be seen spread out in a starburst pattern haphazardly in the endomysium without internal spacing and intermingling with the obscured basement membrane (Fig. 3B and C). Strands of dense flocculent material can also be seen next to the basement membrane, further showing a disorganization of the extracellular environment close to the muscle fibre (Fig. 3D). ATPase staining on IX-15's biopsy showed both type 1 and type 2 fibre atrophy without predominance. Similar absence of clear fibre type grouping was observed on the biopsy of patient VIII-6. The few groups of either type I or type II fibres on these two biopsies were small and more akin to a myopathic type of microfascicular grouping (data not shown).
|
Exclusion of HIBM and LGMD2H
To exclude the possibility that our phenotype is in fact a variant form of HIBM, we sequenced the entire GNE gene for mutations in two patients (VIII-3 and IX-10) and one carrier (father of family VIII, VIII-I). No variant was uncovered. Moreover, we tested the entire cohort for linkage to the TRIM32 locus, the mutated gene responsible for LGMD2H (Frosk et al., 2002
). LGMD2H is the recessive LGMD in which quadriceps atrophy is the most consistent feature (Table 1) (Shokeir and Kobrinsky, 1976; Shokeir and Rozdilsky, 1985). A negative logarithm of the odds (LOD) score of –23 was obtained by GENEHUNTER, confirming that our families are not linked to the LGMD2H locus.
Mapping of the LGMD2L locus
Genealogical data on family IX indicated the parents of the affected individuals were first-degree cousins, thereby limiting the chances of two or more mutations within the family and increasing the odds of strong homozygosity at the disease locus (Fig. 1). Individuals IX-10, IX-11 and IX-15 were thus selected for a genome-wide scan along with an unaffected brother (IX-14) and sister (IX-12). Family VIII was comprised of three affected siblings and one unaffected brother. All four siblings, as well as their father, were sampled and selected for the genome scan. A medium-density genome-wide scan was thus performed at deCODE Genetics (Reykjavik, Iceland) using 500 polymorphic microsatellite markers. Genotypes were analysed using GENEHUNTER's MAXPROB multipoint linkage method (Kruglyak et al., 1996). The highest LOD score (2.78) was obtained for marker D11S1360 located on chromosome 11p12. Only one other peak was obtained above a 2.0 threshold, with a LOD score of 2.16 on chromosome 4. Two-point linkage analysis on MLINK did not uncover any linkage to X-linked markers (data not shown). Multipoint LOD scores were positive for a 17 cM region on 11p12, centred on and peaking at D11S1360. To confirm linkage of our families to the 11p12 locus, we tested the entire cohort against all three consecutive markers from the genome scan (D11S1776, D11S1360 and D11S4191). A maximum multipoint LOD score of 3.81 for D11S1360 was obtained on GENEHUNTER for six of the eight families. The remaining two families, VII and XI, showed negative LOD scores for the first set of tested markers.
Fine mapping of the 17 cM region was conducted by saturating it with all known informative microsatellite markers as well as a set of three markers designed based on genomic repeat sequences found in the region (http://genome.ucsc.edu, May 2004 assembly). Genotypes generated with marker D11S4083 were excluded from the analyses, because this marker was found to have a very high mutation rate. Recruited individuals for all eight families were genotyped and linkage analysis was computed by GENEHUNTER. A maximum cumulative LOD score of 4.56 was obtained for markers D11S935–GGJ11GA24 for all eight families, defining a less than 3.3 cM (1.4 Mb) region (Fig. 4). Since GGJ11GA24 is not on the deCODE genetic map, the size of the interval in cM was estimated with the next deCODE marker, D11S4966. Haplotype analysis demonstrated that up to 81.3% of carrier chromosomes share a two-marker haplotype (Fig. 5). Ten out of 16 carrier chromosomes (62.5%) share a three-marker haplotype. One of the chromosomes shared by the affected individuals of family IX, IXb, might be linked to the common haplotype if allele 2 of marker D11S4185 is identical by descent. Family IX is linked to the locus with a LOD score of 1.72. Two chromosomes (12.5%), IV-3b and VI-5b, appear to share a second, distinct three-marker haplotype. Chromosome VII-3b does not share alleles with the previous two haplotypes. The presence of shared haplotypes further supports our hypothesis that a more common LGMD2L mutation is present in the French–Canadian population. Present haplotype data suggest that probably two to three mutations will be found in our cohort. Based on the hypothesis that two distinct recombination events have reshaped the IXb chromosome in the consanguineous IX family (Fig. 5), this would make D11S935 the telomeric flanking marker and D11S4102, the centromeric flanking marker for a 1.4 cM (725 kb) interval.
|
|
Sequencing of the TRIM44, TRAF6 and LOC119710 genes
Intriguingly, TRIM44, a member of the tripartite motif (TRIM) family, lies 0.19 Mb telomerically to D11S935. Because TRIM32 is mutated in LGMD2H (Frosk et al., 2002
), we decided to sequence TRIM44. TRIM32 likely participates in myofibrillar protein turnover, via its ability to ubiquitinate actin and to bind myofibrils (Kudryashova et al., 2005). The function of TRIM44 is still unknown, but as it possesses the characteristic RING domain of ubiquitin ligases, it may potentially play a similar role to TRIM32. We sequenced TRIM44 in one affected patient and one carrier from three families (VIII-6, VIII-5, IX-8, IX-10, VI-5 and VI-6). No variant was found. Eight known genes lie in the less than 3.3 cM candidate interval. We chose to sequence two that we felt were the most promising candidate genes for LGMD2L: TRAF6 and LOC119710. TRAF6, which lies in the middle of our interval, is mainly involved in the immune system as an adapter protein for dendritic cell maturation and cytokine production (Kobayashi et al., 2004). It also mediates activation of NF
B and JNK, both major players in apoptosis (Bharti et al., 2004). Most studies have investigated TRAF6's immunological function in the IL-1 receptor/Toll-like receptor pathway. However, TRAF6, as a tumour necrosis factor (TNF) receptor-associated factor, could also be linked to the development of skeletal muscles, since it has been reported that TNF
inhibits myogenic differentiation through the NFB-dependent destabilization of the MyoD protein activity, which interferes with skeletal muscle regeneration and may contribute to muscle wasting (Langen et al., 2004). Moreover, TRAF6 is part of the neurotrophin pathway through its interaction with the p75 neurotrophin receptor (Roux and Barker, 2002). The receptor activates NFB, Akt and JNK pathways, and TRAF6 has been shown to promote cell survival in this context. The adapter protein, RIP2, which binds the p75 neurotrophin receptor, provides a bifunctional switch for the survival and death of Schwann cells (Khursigara et al., 2001). Hence, mutations in TRAF6 could have a potential impact on the p75 neurotrophin receptor and its associated factors: an imbalance in the RIP2 switch could have explained the minor neurogenic component observed on the EMGs. For these reasons, we chose to sequence this gene. All six exons and exon/intron boundaries were sequenced, as well as the two putative exons contained in the 5' untranslated region. No mutations were detected, but known SNPs were observed. LOC119710 was also sequenced. Little is known about this predicted gene and the protein it encodes. However, GeneNote expression arrays suggest a high expression of the mRNA in both skeletal and cardiac muscles (Yanai et al., 2005). No mutations were detected in the five coding exons and the exon/intron boundaries. At this point, we cannot exclude putative mutations in the regulatory elements or promoters of these two genes, large deletions or post-translational modifications. |
Discussion |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
This study is the first to report a new chromosomal locus linked to a form of variable limb-girdle muscular dystrophy associated with marked, asymmetrical atrophy of the biceps brachii and quadriceps femoris and myalgia. The disease phenotype demonstrates important variability among and between families. CK levels remain largely uninformative, as 28.6% of our cohort exhibited CK values within the normal range (Table 2). Some of the EMG findings, particularly in the more severe cases, raise the possibility of a neurogenic component to the disease. The combination of myopathic and neurogenic findings was also previously described in cases of ‘quadriceps myopathy’ (Boddie and Stewart-Wynne, 1974
). MRI is clearly a very sensitive means to confirm muscle involvement in the milder cases, and it confirms the clinical observations of asymmetrical atrophy and weakness. Inflammatory changes are not a feature of this disease and may only correlate with severity. Case VIII-6, which demonstrates some inflammatory change on biopsy, is clinically and pathologically one of the most affected of our cohort. This LGMD links to a less than 3.3 cM (1.4 Mb) region on 11p13–p12.
We propose to name this new disease LGMD2L, in keeping with the current nomenclature (Bushby and Beckmann, 1995). Its extreme variability of involvement among and between families may have limited its clear identification in the past. The age of onset is one such variable: two patients presented with symptoms before the age of 20, two between 21 and 30, six between 31 and 40, three between 41 and 50, and one remains asymptomatic in her late sixties despite abnormal clinical and MRI evaluations (Fig. 2B and G) and elevated CK values for more than two decades (case IX-9, Table 2). Intrafamilial variability is demonstrated in both families VIII and IX. Patient VIII-6 displays a more severe atrophy than his affected sisters, VIII-3 and VIII-4: symptoms began at an earlier age and his CKs are notably higher than his sisters', whose CK values were found to be normal (Table 2). In family IX, this range in severity of affection is even more pronounced. Patient IX-13 was wheelchair bound by the age of 34, while his brothers IX-10 and IX-11 are still able to walk into their late sixties. Their sister, IX-9, the elder of the affected siblings, is still asymptomatic, displays elevated CK levels, and has a very mild muscular atrophy. This variability also extends to single individuals, as a clear asymmetry of the muscular wasting was noted. Interfamilial and intrafamilial variability in phenotype, as well as an asymmetrical muscular atrophy, seem to be a hallmark of this form of LGMD. By combining clinical and MRI findings, we suggest that the following muscles are the first affected in this LGMD: the long head of the biceps brachii, the hamstring muscles (in particular the semimembranosus) and the adductor magnus. As the disease progresses, atrophy and weakness extend to the vastus medialis and the tensor fasciae latae. The early involvement of the hamstring muscles and adductor magnus in our cohort seems to be also found in a number of AR-LGMDs, most notably LGMD2A and LGMD2I for which MRI studies revealed a severe atrophy of these muscles (Fischer et al., 2005). However, these LGMD2A and LGMD2I cases showed a symmetrical atrophy, while our patients demonstrate a clear asymmetry in muscle wasting. We believe this asymmetry could help in the clinical diagnosis until the mutated gene is uncovered. As shown in Table 1, all previously described recessive LGMD have distinctive features from LGMD2L except LGMD2B. Though it is possible that the important variability has limited the diagnosis of cases affected with LGMD2L in the past, it is likely that some of the cases of ‘quadriceps myopathy’, which were indeed later shown to have progressed into a limb-girdle type of muscular dystrophy, were affected by this disease (Boddie and Stewart-Wynne, 1974; Walton, 1956).
No tubular filamentous inclusions or important signs of inflammation were reported in our participants (with the exception of one demonstrating focal inflammation), thereby excluding the diagnosis of IBM. Clinically, LGMD2L patients did not demonstrate finger flexor weakness or atrophy. While the hallmark of the hereditary form of IBM is a characteristic sparing of the quadriceps muscle, quadriceps myopathies have frequently been associated with the sporadic form of IBM. Inclusion body myositis is further characterized by diminished deep tendon reflexes, dysphagia and atrophy of the proximal limb muscles (Burstein et al., 2005). The absence of inflammation in muscle biopsies seems to rule out the possibility of a new form of hereditary IBM. To further exclude this possibility, the GNE gene was sequenced and found to carry no mutation. An in-depth pathological review of the first and only muscle biopsy we had access to, patient IX-15's biopsy, uncovered a disorganization of the basal lamina and extracellular matrix which may suggest that the mutated protein in LGMD2L is either an extracellular matrix or basal lamina protein or a protein important in ensuring the dynamic relationship between the muscle fibre and the extracellular matrix. Further pathological work should allow a better understanding of the pathophysiology of the disease.
Haplotype analysis helped refine the candidate region to a small interval and demonstrated that up to 81.3% of carrier chromosomes may share parts of an ancestral haplotype (Fig. 5). Ten of the sixteen carrier chromosomes (62.5%) seem to share the common haplotype on three consecutive markers or more, while two (12.5%) share a second haplotype. Surprisingly, the affected cases of the consanguineous family IX, though they share the same two haplotypes for the region, are homozygous only for one allele of the haplotype shared by many families (Fig. 5). We conclude that the two mutations present in these affected cases were not transmitted through the known consanguineous loop and were likely introduced independently in the family through a more a distant relationship. Supposing that the presumed historical recombinations between D11S935 and D11S4185 telomerically, and between D11S4185 and D11S4102 centromerically, have reshaped chromosome IXb, this would make, respectively, D11S935 and D11S4102 the telomeric and centromeric flanking markers for a candidate interval of 1.4 cM (725 kb). It is interesting to note that, despite the relatively young ancestry of the French–Canadian population (
400 years), homozygosity at the locus is rare and the conserved region is surprisingly small. This observation can, however, be explained in two ways. First, most of the cases in our cohort have a family history traceable to the Southwest of the Province of Quebec, a region which, unlike the better studied region of Saguenay-Lac-Saint-Jean (Scriver, 2001), displays more genetic heterogeneity and less of a more proximal consanguinity. Second, if the more major mutation is older than the French colonization that began in 1608 and was introduced more than once in Quebec, this could also explain the relatively small shared ancestral haplotype.
No other LGMD families or, in fact, other neuromuscular hereditary diseases have been linked to this region of chromosome 11p13–p12. Eight genes appear to lie in the candidate region: LOC143458, FLJ45212, COMMD9, FLJ14213, TRAF6, RAG1, RAG2 and LOC119710. COMMD9 has homologues in the mouse and the rat, among other animals, and its protein, COMM domain containing protein 9, has 198 amino acids. The GeneNote expression array data indicates the protein is highly expressed in skeletal muscles (http://bioinfo.weizmann.ac.il). A recent paper has shown that COMM-domain-containing proteins are regulators of NFB; thus, potentially linking them to apoptosis and cachexia (Burstein et al., 2005). The RAG1 and RAG2 protein products are initiators of the V(D)J recombination pathway in immunoglobulins (Oettinger, 1992). Mutations in both RAG genes have been linked to either of the two forms of severe combined immunodeficiency syndromes (Tabori et al., 2004). Four predicted proteins of unknown function are also encoded in this interval: LOC143458, FLJ45212, FLJ14213 and LOC119710. The centromeric half of the interval (734 606 bp) contains no known gene. No genes have been predicted to lie within this region by most bioinformatic programs, with the exception of GenScan, the biostatistical algorithms of which predict four more genes on the centromeric side of LOC119710 in our interval (Burge and Karlin, 1997). While no mutations were found in the coding sequences of TRAF6 and LOC119710, further work will be required to exclude potential mutations in their regulatory and promoter regions.
The identification of the mutated gene responsible for this new recessive LGMD will require the sequencing of other genes in the candidate interval and the recruitment of other LGMD2L cases of different ethnic backgrounds. The variability of the phenotype and the asymmetry of the atrophy are reminiscent of some of the features of facioscapulohumeral muscular dystrophy type 1A (FSHMD1A, OMIM 158900
[OMIM]
). This raises the possibility that as in FSHMD1A a non-coding repeat mutation is responsible for LGMD2L (van der Maarel and Frants, 2005). In FSHMD1A the repeat is variably contracted in affected individuals, leading to an overexpression of genes upstream of the repeats. Although no LSau or hhspm3 repeats, known components of the FSHMD1A D4Z4 repeats, were found in the LGMD2L candidate interval, a similar type of mutation might explain the variability of the phenotype and asymmetry of muscle wasting. The identification of the mutations responsible for LGMD2L should provide further insight into the complex pathways leading to adult onset muscular dystrophies.
|
Acknowledgements |
|---|
We would like to thank Marie-Pierre Dubé for her help in linkage analysis and Dr Louise F. Charron for referring to us the family of LGMD2L. This work was supported by an MDA grant (MDA 4001). B.B. is a chercheur-bousier of the Fonds de recherche en santé du Québec (FRSQ).
|
References |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Bharti AC, Takada Y, Shishodia S, Aggarwal BB. (2004) Evidence that receptor activator of nuclear factor (NF)-kappaB ligand can suppress cell proliferation and induce apoptosis through activation of a NF-kappaB-independent and TRAF6-dependent mechanism. J Biol Chem 279:6065–76.
Boddie HG and Stewart-Wynne EG. (1974) Quadriceps myopathy—entity or syndrome? Arch Neurol 31:60–2.[ISI][Medline]
Bramwell E. (1922;) Observations on myopathy: symmetrical atrophic paresis of the quadriceps muscles of probable myopathic origin. Proc R Soc Med 16: pp. 1–3.
Burge C and Karlin S. (1997) Prediction of complete gene structures in human genomic DNA. J Mol Biol 268:78–94.[CrossRef][ISI][Medline]
Burstein E, Hoberg JE, Wilkinson AS, Rumble JM, Csomos RA, Komarck CM, et al. (2005) COMMD proteins: a novel family of structural and functional homologs of MURR1. J Biol Chem 280:22222–32.
Bushby KM. (1999) Making sense of the limb-girdle muscular dystrophies. Brain 122:1403–20.
Bushby KM and Beckmann JS. (1995) The limb-girdle muscular dystrophies—proposal for a new nomenclature. Neuromuscul Disord 5:337–43.[CrossRef][ISI][Medline]
Charniot JC, Pascal C, Bouchier C, Sebillon P, Salama J, Duboscq-Bidot L, et al. (2003) Functional consequences of an LMNA mutation associated with a new cardiac and non-cardiac phenotype. Hum Mutat 21:473–81.[CrossRef][ISI][Medline]
Darin N, Kyllerman M, Wahlstrom J, Martinsson T, Oldfors A. (1998) Autosomal dominant myopathy with congenital joint contractures, ophthalmoplegia and rimmed vacuoles. Ann Neurol 44:242–8.[CrossRef][ISI][Medline]
Denny-Brown D. (1939) Myopathic weakness of the quadriceps. Proc R Soc Med 32:867–9.
Eisenberg I, Hochner H, Shemesh M, Levi T, Potikha T, Sadeh M, et al. (2001) Physical and transcriptional map of the hereditary inclusion body myopathy locus on chromosome 9p12–p13. Eur J Hum Genet 9:501–9.[CrossRef][ISI][Medline]
Espir ML and Matthews WB. (1973) Hereditary quadriceps myopathy. J Neurol Neurosurg Psychiatry 36:1041–5.[ISI][Medline]
Fischer D, Walter MC, Kesper K, Petersen JA, Aurino S, Nigro V, et al. (2005) Diagnostic value of muscle MRI in differentiating LGMD2I from other LGMDs. J Neurol 252:538–47.[CrossRef][ISI][Medline]
Frosk P, Weiler T, Nylen E, Sudha T, Greenberg CR, Morgan K, et al. (2002) Limb-girdle muscular dystrophy type 2H associated with mutation in TRIM32, a putative E3-ubiquitin-ligase gene. Am J Hum Genet 70:663–72.[CrossRef][ISI][Medline]
Furukawa T, Akagami N, Maruyama S. (1977) Chronic neurogenic quadriceps amyotrophy. Ann Neurol 2:528–30.[CrossRef][ISI][Medline]
Khursigara G, Bertin J, Yano H, Moffett H, DiStefano PS, Chao MV. (2001) A prosurvival function for the p75 receptor death domain mediated via the caspase recruitment domain receptor-interacting protein 2. J Neurosci 21:5854–63.
Kobayashi T, Walsh MC, Choi Y. (2004) The role of TRAF6 in signal transduction and the immune response. Microbes Infect 6:1333–8.[CrossRef][ISI][Medline]
Kruglyak L, Daly MJ, Reeve-Daly MP, Lander ES. (1996) Parametric and nonparametric linkage analysis: a unified multipoint approach. Am J Hum Genet 58:1347–63.[ISI][Medline]
Kudryashova E, Kudryashov D, Kramerova I, Spencer MJ. (2005) Trim32 is a ubiquitin ligase mutated in limb girdle muscular dystrophy type 2H that binds to skeletal muscle myosin and ubiquitinates actin. J Mol Biol 354:413–24.[CrossRef][ISI][Medline]
Langen RC, Van Der Velden JL, Schols AM, Kelders MC, Wouters EF, Janssen-Heininger YM. (2004) Tumor necrosis factor-alpha inhibits myogenic differentiation through MyoD protein destabilization. FASEB J 18:227–37.
Lathrop GM and Lalouel JM. (1984) Easy calculations of LOD scores and genetic risks on small computers. Am J Hum Genet 36:460–5.[ISI][Medline]
Lathrop GM, Lalouel JM, Julier C, Ott J. (1984) Strategies for multilocus linkage analysis in humans. Proc Natl Acad Sci USA 81:3443–6.
Lathrop GM, Lalouel JM, White RL. (1986) Construction of human linkage maps: likelihood calculations for multilocus linkage analysis. Genet Epidemiol 3:39–52.[CrossRef][ISI][Medline]
Mümenthaler M. (1970) Myopathy in neuropathy. In Walton JN, Canal N, Scarlato G (Eds.). Muscle diseases. Vol. 199(Excerpta Medica Foundation, Amsterdam) pp. p. 585–98.
Oettinger MA. (1992) Activation of V(D)J recombination by RAG1 and RAG2. Trends Genet 8:413–6.[ISI][Medline]
Phillips BA, Cala LA, Thickbroom GW, Melsom A, Zilko PJ, Mastaglia FL. (2001) Patterns of muscle involvement in inclusion body myositis: clinical and magnetic resonance imaging study. Muscle Nerve 24:1526–34.[CrossRef][ISI][Medline]
Roux PP and Barker PA. (2002) Neurotrophin signaling through the p75 neurotrophin receptor. Prog Neurobiol 67:203–33.[CrossRef][ISI][Medline]
Rozen S, Skaletsky H. (2000) Primer3 on the WWW for general users and for biologist programmers. In Krawetz S and Misener S (Eds.). Bioinformatics methods and protocols: methods in molecular biology(Humana Press, Totowa, NJ) pp. p. 365–86.
Schindl K and Rueker O. (1973) Differential interference contrast for transmitted-light (Nomarski method). J Biol Photogr Assoc 41:109–11.[ISI][Medline]
Scriver CR. (2001) Human genetics: lessons from Quebec populations. Annu Rev Genomics Hum Genet 2:69–101.[CrossRef][ISI][Medline]
Serratrice GT and Munsat TL. (1995) Overview of the pathogenesis and therapy of amyotrophic lateral sclerosis. Adv Neurol 68:1–5.[ISI][Medline]
Shokeir MH and Kobrinsky NL. (1976) Autosomal recessive muscular dystrophy in Manitoba Hutterites. Clin Genet 9:197–202.[ISI][Medline]
Shokeir MH and Rozdilsky B. (1985) Muscular dystrophy in Saskatchewan Hutterites. Am J Med Genet 22:487–93.[CrossRef][ISI][Medline]
Sternberger LA. (1969) Some new developments in immunocytochemistry. Mikroskopie 25:346–61.[Medline]
Sunohara N, Arahata K, Hoffman EP, Yamada H, Nishimiya J, Arikawa E, et al. (1990) Quadriceps myopathy: forme fruste of Becker muscular dystrophy. Ann Neurol 28:634–9.[CrossRef][ISI][Medline]
Sveen ML, Schwartz M, Vissing J. (2006) High prevalence and phenotype-genotype correlations of limb girdle muscular dystrophy type 2I in Denmark. Ann Neurol 59:808–15.[CrossRef][ISI][Medline]
Tabori U, Mark Z, Amariglio N, Etzioni A, Golan H, Biloray B, et al. (2004) Detection of RAG mutations and prenatal diagnosis in families presenting with either T-B- severe combined immunodeficiency or Omenn's syndrome. Clin Genet 65:322–6.[CrossRef][ISI][Medline]
Turner JW and Heathfield KW. (1961) Quadriceps myopathy occurring in middle age. J Neurol Neurosurg Psychiatry 24:18–21.[ISI][Medline]
van der Maarel SM and Frants RR. (2005) The D4Z4 repeat-mediated pathogenesis of facioscapulohumeral muscular dystrophy. Am J Hum Genet 76:375–86.[CrossRef][ISI][Medline]
Walton JN. (1956) Two cases of myopathy limited to the quadriceps. J Neurol Neurosurg Psychiatry pp. 106–8.
Wicklund MP and Mendell JR. (2003) The limb girdle muscular dystrophies: our ever-expanding knowledge. J Clin Neuromuscul Dis 5:12–28.[CrossRef]
Yanai I, Benjamin H, Shmoish M, Chalifa-Caspi V, Shklar M, Ophir R, et al. (2005) Genome-wide midrange transcription profiles reveal expression level relationships in human tissue specification. Bioinformatics 21:650–9.
Zatz M, de Paula F, Starling A, Vainzof M. (2003) The 10 autosomal recessive limb-girdle muscular dystrophies. Neuromuscul Disord 13:532–44.[CrossRef][ISI][Medline]
Zelinski T. (1991) The use of DNA restriction fragment length polymorphisms in conjunction with blood group serology. Transfusion 31:762–70.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||