Genotype analysis by using the p13E‐11 probe and other 4q35 polymorphic markers was performed in 122 Italian facioscapulohumeral muscular dystrophy families and 230 normal controls. EcoRI—BlnI double digestion was routinely used to avoid the interference of small EcoRI fragments of 10qter origin that were found in 15% of the controls. An EcoRI fragment ranging between 10 and 28 kb that was resistant to BlnI digestion was detected in 114 of 122 families (93%) comprising 76 familial and 38 isolated cases. Among the unaffected individuals, 3 were somatic mosaics and 7, carrying an EcoRI fragment larger than 20 kb, could be rated as nonpenetrant gene carriers. In a cohort of 165 patients with facioscapulohumeral muscular dystrophy we found an inverse correlation between fragment size and clinical severity. A severe lower limb involvement was observed in 100% of patients with an EcoRI fragment size of 10 to 13 kb (1–2 KpnI repeats left), in 53% of patients with a fragment size of 16 to 20 kb (3–4 KpnI repeats left), and in 19% of patients with a fragment size larger than 21 kb (>4 KpnI repeats left). Our results confirm that the size of the fragment is a major factor in determining the facioscapulohumeral muscular dystrophy phenotype and that it has an impact on clinical prognosis and genetic counseling of the disease. Ann Neurol 1999;45:751–757

Autosomal dominant facioscapulohumeral muscular dystrophy (FSHD) is caused by deletion of most copies of the 3.3-kb subtelomeric D4Z4 repeat array on chromosome 4q. The molecular mechanisms behind the deletion and the high proportion of new mutations have remained elusive. We surveyed 35 de novo FSHD families and found somatic mosaicism in 40% of cases, in either the patient or an asymptomatic parent. Mosaic males were typically affected; mosaic females were more often the unaffected parent of a nonmosaic de novo patient. A genotypic-severity score, composed of the residual repeat size and the degree of somatic mosaicism, yields a consistent relationship with severity and age at onset of disease. Mosaic females had a higher proportion of somatic mosaicism than did mosaic males. The repeat deletion is significantly enhanced by supernumerary homologous repeat arrays. In 10% of normal chromosomes, 4-type repeat arrays are present on chromosome 10. In mosaic individuals, 4-type repeats on chromosome 10 are almost five times more frequent. The reverse configuration, also 10% in normal chromosomes, was not found, indicating that mutations may arise from transchromosomal interaction, to which the increase in 4-type repeat clusters is a predisposing factor. The somatic mosaicism suggests a mainly mitotic origin; mitotic interchromosomal gene conversion or translocation between fully homologous 4-type repeat arrays may be a major mechanism for FSHD mutations.

Chromosomal rearrangements occur more frequently in subtelomeric domains than in other regions of the genome and are often associated with human pathology. To further elucidate the plasticity of subtelomeric domains, we examined the 3.3 kb D4Z4 repeat array on chromosome 4 and its homologue on chromosome 10 in 208 Dutch blood donors by pulsed field gel electrophoresis. These subtelomeric repeats are known to rearrange and partial deletions of this polymorphic array on chromosome 4 are associated with facioscapulohumeral muscular dystrophy (FSHD), an autosomal dominant myopathy. Our results show that mitotic rearrangements occur frequently as 3% of individuals display somatic mosaicism for a repeat expansion or contraction explaining the high variability of subtelomeric repeat array sizes. Translocated 4-type repeat arrays on chromosome 10 and the reverse configuration of 10-type repeat arrays on chromosome 4 are observed in 21% of individuals. The translocated repeat arrays on chromosome 4 tend to be more heterogeneous than 4-type repeats on chromosome 10. The repeat length on chromosome 4 is on average larger than on chromosome 10. But on both chromosomes we observe a multi-modal repeat length distribution with equidistant peaks at intervals of 65 kb, possibly reflecting a higher-order chromatin structure. Interestingly, in as many as six random blood donors (3%) we identified FSHD-sized 4-type repeat arrays. Assuming that these individuals are clinically unaffected, these results imply an incomplete penetrance in the upper range of FSHD alleles. Overall, the observed dynamic characteristics of these homologous domains may serve as a model for subtelomeric plasticity.

The pl3E-l1 probe has been shown to detect DNA rearrangements in sporadic and familial cases of FSHD. Its use, however, has been hampered by the fact that it detects at least two pairs ofEcoRI alleles, one derived from the 4q35 region (D4F104S1), the other from 1Oq26 (DlOF104S2). We have cloned pl3E-ll EcoRI fragments from the 4q35 and 1Oq26 subtelomeric regions and shown the presence of several restriction site differences within the KpnI tandem repeat units. The two loci present a different distribution of restriction sites for the enzyme BlnI which allows differential cleavage of the KpnI units derived from lOq26, leaving intact the 4q35 pair of alleles. This method of differential restriction greatly facilitates the interpretation of Southern blots obtained from affected and unaffected subjects, with an important improvement in reliability for diagnosis and genetic counselling. In addition, this method can be used to investigate the molecular mechanism of the 4q35 rearrangement implicated in the disease and to ascertain whether the rearrangement is because of interchromosomal exchange between 4qter and lOqter KpnI repeats.

No abstract

These data point to the CpGs distal to the D4Z4 array as a critical region reflecting multiple factors affecting the epigenetics of FSHD. Additionally, methylation analysis of this region allows the establishment of a rapid and sensitive tool for FSHD diagnosis.

The autosomal dominant myopathy facioscapulohumeral muscular dystrophy (FSHD) is causally related to a short Eco RI fragment detected by probe p13E-11. This remnant fragment is the result of a deletion of an integral number of tandemly arrayed 3.3 kb repeat units (D4Z4) on 4q35. Despite intensive efforts, no transcribed sequences have been identified within this array. Previously, we have shown that these repeats on 4q35 have been exchanged for a similar highly homologous repeat locus on 10q26 in 20% of the population and that a short chromosome 10-like array on 4q35 also results in FSHD. Here, we describe the hybrid structure of some of these repeat arrays, reflecting additional sub-telomeric instability. In three healthy individuals carrying a 4-like repeat on chromosome 10 or vice versa, one repeat array was shown to consist of hybrid clusters of 4-derived and 10-derived repeat units. Moreover, employing pulsed field gel electrophoresis analysis, we identified two unrelated individuals carrying deletions of a chromosomal segment (p13E-11) proximal to the repeat locus. These deletions were not associated with FSHD. In one of these cases, however, an expansion of the deletion into the repeat array was observed in one of his children suffering from FSHD. These data provide additional evidence for instability of this sub-telomeric region and suggests that the length of the repeat, and not its intrinsic properties, is crucial to FSHD. Moreover, they are in agreement with the hypothesis that FSHD is caused by a position effect in which the repeat structure influences the expression of genes nearby. Therefore, the region deleted proximal to the repeat locus in healthy individuals can be instrumental to refine the critical region for FSHD1.

No abstract