Statistics from Altmetric.com
Editor—The autosomal recessive limb-girdle muscular dystrophies (AR-LGMD) are clinical entities characterised by primary and progressive muscle degeneration, mainly at the pelvic and shoulder girdles, with great variability in the clinical course. Some patients present a severe course similar to Duchenne muscular dystrophy, while others maintain the capacity to walk even in adult life.1-3 At least eight autosomal recessive genes have been mapped. The chromosome localisation of these genes and their products, and a brief comment on the clinical course of each type of AR-LGMD are summarised in table 1. Of these mapped genes, six have been cloned: the gene responsible for LGMD2A which encodes calpain 3, a muscle specific protease,30 the genes that cause the known sarcoglycanopathies (LGMD2C-LGMD2F),11 15 20 22 26and, recently, the gene for LGMD2B which encodes a protein called “dysferlin” by the investigators.8
LGMD2C is a Duchenne-like muscular dystrophy particularly prevalent in North Africa,10-12 31 32 but rare in other geographical regions; its prevalence in north eastern Italy was estimated to be 1.72 × 10-6 inhabitants.33 This phenotype, which affects both sexes equally, was first described by Ben Hamidaet al 12 in 93 patients belonging to 28 Tunisian families. A few large kindreds, with many affected persons, have been described. Here we report the results of a clinical and molecular study in a large inbred kindred from the north east of Brazil with LGMD2C, which is unusual because the male patients appear to have a more severe clinical course than the affected females.
The genealogical data from five generations (fig 1) were obtained and confirmed by different family members. The dates of birth, marriage, and death, causes of death, and abortions were documented. A total of 56 subjects, including all living affected persons, were clinically examined. Muscle strength was evaluated in 13 patients according to the manual muscle test (based on the Medical Research Council scale). The diagnosis of AR-LGMD was based on clinical examination, course of the disease, family history, serum creatine kinase (CK) levels, muscle histopathology, and muscle protein and DNA analyses.
A muscle sample was obtained from a biceps biopsy of one male patient (V.5). Immediately after removal, the muscle sample was frozen in liquid nitrogen and stored at −70°C until analysis. Dystrophin was analysed by immunofluorescence (IF) and western blotting (WB) with rabbit polyclonal N-terminal 303-8 and C-terminal monoclonal Dy8/6C5 antibodies.34 35 The amount of dystrophin on WB was estimated by densitometric analysis.36 The patient’s band was compared with a normal control and corrected for the myosin content in the muscle extract. On immunohistochemical staining of frozen muscle sections using double labelling reactions for dystrophin + γ-SG, α-SG + β-SG, and γ-SG + δ-SG, the following antibodies were used: α-SG, monoclonal Ad1/20A618; β-SG, rabbit polyclonal antibody21; γ-SG, rabbit polyclonal antibody13 and monoclonal 35DAG/21B537; and δ-SG, rabbit polyclonal raised against a glutathione S-transferase (GST)-δ-sarcoglycan fusion protein.26
DNA samples from 28 members of the family, including 11 affected persons, were extracted from whole blood (after informed consent) according to the method of Miller et al.38 Microsatellite markers corresponding to genes involved in AR-LGMD were amplified by PCR and the products were visualised on 6.5% denaturing gels, which were dried and exposed tox rays. Two point linkage analysis involving the mutant gene and the microsatellite markers D13S232, D13S115, and D13S143 (at chromosome 13q12) was performed using the MLINK software program.39 An estimated gene frequency of 0.001 for the disease allele and an equal recombination for both sexes were assumed. For characterising the mutation, exon 6 of the γ-SG gene was amplified by PCR from DNA samples of affected patients and sequenced. Primers were purchased from Research Genetics.
The pedigree of this large, white, consanguineous kindred, of Portuguese ancestry, is shown in fig 1. The total number of affected persons, including five subjects who had died, is 20 (11 males and nine females).
According to the patients’ parents, the first clinical manifestations were difficulty in running and climbing stairs and frequent falls. The ages of onset, wheelchair confinement, and death (five patients) are listed in table 2. Physical examination showed that all living affected persons had muscle atrophy (more severe in the older patients) and had lost the ability to walk. Prominence of calves was seen only in the brothers V.4 and V.5. The tendon reflexes were abolished and the facial muscles were spared. Clinical cardiological evaluation performed in all patients and electrocardiogram/echocardiogram examinations, performed in five of them (IV.20, IV.22, IV.34, V.4, and V.5), showed no evidence of cardiac disease. Respiratory function (spirometry) was investigated in these five patients. A mild restrictive lung disease pattern was diagnosed in the children V.4 and V.5 and the three adult patients had a moderate to severe pattern. Intellectual development was normal in all affected persons. Serum CK level in V.5 (the youngest patient) was 340 IU/ml (normal ⩽20). CK levels in the oldest patients (IV.3, IV.26, and IV.34) were normal.
The disease seemed to be more severe, with a more rapid rate of clinical progression, in the affected males than in the female patients. The mean ages of onset were 3.18 (SD 1.08) and 4.56 (SD 1.13) years respectively for male and female patients, and the difference between these mean values is significant at the 5% level (p=0.013). Regarding wheelchair confinement, the mean ages were 13.91 (SD 2.47) years (affected males) and 21.67 (SD 3.32) years (affected females), and the significance is p=0.0002. The manual muscle test performed in 13 patients (table 3) showed that the proximal muscles of the four limbs were much more affected than the distal ones, but no significant sex difference was observed. The degree of contractures in these patients ranged from mild to severe and involved the elbow, hip, and knee joints. The most severe contractures were observed in affected males IV.8, IV.12, IV.20, and IV.49. In addition, IV.8 and IV.20 had contractures of the wrist joints. Scoliosis was seen in only one patient (IV.8). The cause of death of the five dead affected subjects (three men and two women) was pneumonia. IV.35 died some months after our clinical evaluation.
The muscle biopsy from patient V.5 showed histopathological changes of a primary myopathic process, characterised by a marked variation in fibre diameter, round shaped fibres, split fibres, proliferation of endomyseal and perimyseal conective tissue, and fat infiltration. The histoenzymological studies showed a predominance of type I fibres.
IF staining with dystrophin antibodies showed a mosaic pattern of positive and negative fibres, and the IF pattern with antibodies directed at each of the four known SG proteins was negative (fig 2). WB analysis showed a reduction in the amount of dystrophin (about 20% of normal).
Confirmation of linkage (lod score >3) was obtained between the disease gene and D13S232, D13S115, and D13S143 microsatellite markers. The rare D13S232-3 allele (122 bp) was found in a homozygous state in the 11 patients studied. Amplification by PCR followed by sequencing of exon 6 of the γ-SG gene in these persons showed the deletion of a thymine from the span of 521-525 bp (Δ521-T mutation).
The great majority of the patients with severe childhood onset progressive muscular dystrophy had mutations on the X chromosome, with autosomal recessive inheritance involved in only about 5% of the cases.40 In the kindred with LGMD2C described here, this pattern is well illustrated, since the affected persons had unaffected consanguineous parents and comprised approximately 25% of their offspring of both sexes. Because of the lack of reliable information about more remote antecedents, it is difficult to establish the origin of the abnormal recessive gene in this inbred kindred, but it was probably introduced by one of the parents of the sibs I.1/I.5 who were heterozygotes. The genetic homogeneity of LGMD2C in North African populations (where this disease was initially studied), derived from the same ancestral population, suggests a common origin for the mutant gene in that geographical region.10 31 32
The immunohistochemical analyses of SGs in muscle biopsy specimens from patients with any of the types of sarcoglycanopathy have shown deficiencies of all components of the SG complex, suggesting that pathogenic mutations in a single SG gene disturb the organisation of the whole glycoproteic complex and lead to secondary deficiency of the other SGs.11 13 25 36 41 In agreement with these features, our patient V.5 showed absence of the four known components of the SG complex. In addition, dystrophin WB showed a reduced quantity in this patient. In a study on muscle proteins in six types of AR-LGMD (2A-2F), involving 35 patients, Vainzof et al 36 also found reduced quantities of dystrophin in the severe cases, but described a variable IF pattern for the sarcoglycans.
The localisation of the gene for LGMD2C on chromosome 13q12 was made initially by Ben Othmane et al 10 in three Tunisian kindreds based on the observation of segregation of the disease with markers in this chromosome region. Further studies with Algerian31 and Moroccan32 families confirmed the mapping. Linkage disequilibrium between the LGMD2C locus and the rare allele 3 (122 bp) of marker D13S232 was described in nine Tunisian and one Egyptian families, suggesting that the two loci are very close to one another.42 A homozygous deletion of one thymine base from the span of 521-525 bp of the γ-SG gene (Δ521-T mutation) was found in affected persons from three of these kindreds.11 In our study, the two point linkage analysis between the disease gene and each of the markers at 13q12 showed the presence of the D13S232-3 allele in all patients. The occurrence of the Δ521-T mutation associated with this allele in the affected subjects is consistent with the linkage disequilibrium previously reported.42
It was surprising to observe that the disease appeared to be more severe in our male patients, since the mean ages of onset and wheelchair confinement were significantly lower in these patients than in the affected females. The pattern of muscle involvement (table 3) showed no evidence of differences within sexes, but this pattern was investigated only when all affected persons had already lost independent walking (data on different stages of the disease are not available because this family had never been previously studied). Recently, in a large study of facioscapulohumeral muscular dystrophy involving 173 affected persons from 53 families, Zatzet al 43 also described a more severe clinical phenotype in the male patients. In their original description of Duchenne-like muscular dystrophy, Ben Hamidaet al 12 found significant intrafamilial and interfamilial variability in the severity of manifestation of the disease, with loss of independent walking varying between the ages of 10 and 31 years, but without differences within sexes. The Δ521-T mutation was previously described in four Brazilian families (with a total of 14 patients) of Negroid ethnicity13 44 with no biological relationship with the (white) kindred reported here. Three of them manifested the classical (severe) form of LGMD2C. However, in the fourth family, the three affected sibs (a 23 year old woman and two males aged 20 and 14 years respectively) had a mild phenotype with preservation of ambulation, particularly the older sister who was almost asymptomatic at that age, possibly reflecting a slight difference of clinical expression favouring the female sex. These observations suggest that the Δ521-T mutation can lead to a milder phenotype or a more severe form of the disease in one sex in some families.
Further investigations in a larger series of LGMD2C patients are necessary for a complete delineation of the spectrum of variation in the clinical expression of this and other mutations in the γ-SG gene, and for understanding the underlying molecular mechanisms, since they have implications for genetic counselling.
We would like to express our gratitude to the members of the family described in this work for their collaboration. Our special thanks to Drs Maria R Passos-Bueno, Mayana Zatz, Mariz Vainzof, and Eloísa S Moreira for their valuable help in the laboratory work at the Department of Biology, University of São Paulo, São Paulo, Brazil. Drs Zatz and Passos-Bueno also critically read the manuscript. This investigation was supported by grant APQ 0168-2.02/94 from FACEPE (Fundação de Amparo à Ciência e Tecnologia de Pernambuco).
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.