Comprehensive analysis for genetic diagnosis of Dystrophinopathies in Japan

Dystrophinopathies are an X-linked disease caused by mutations in the DMD gene (OMIM 300377) located at Xp21.2. DMD is a large gene, spanning more than 2.2 Mb of genomic DNA, and containing 79 exons and lengthy introns, which produces a 14-kb mRNA transcript [1]. Dystrophinopathies are the most common muscle disease in children which affects one in 3600–6000 live male births [2, 3], and has been classically classified into two forms: Duchenne muscular dystrophy (DMD; OMIM 310200) and Becker muscular dystrophy (BMD; OMIM 300376). DMD shows a severe phenotype clinically characterized by rapid progression in early childhood, and loss of ambulation during the second decade of life, while BMD shows a milder form with patients being ambulant after 16 years of age. Patients with an intermediate phenotype are sometimes referred to as having intermediate muscular dystrophy (IMD). Nevertheless, the classification into the three forms is not always easy.

To date, approximately 70% of mutations found in DMD patients are deletions/duplications of one or more exons, while the remaining 30% are caused by small mutations at the nucleotide level. Multiplex ligation-dependent probe amplification (MLPA), which can examine the duplications and/or deletions of all 79 exons, has been developed and widely used. [4‐6] However, the remaining 30% of patients harboring small mutations are undiagnosed and need to perform sanger sequencing for diagnosis.

Recently, promising mutation-specific molecular therapies have been developed. For instance, exon skipping is expected to be applicable to patients that have one or more exons deletions in DMD. Furthermore, read-through therapy of a nonsense codon to produce full-length dystrophin is applicable to patients with DMD harboring nonsense mutations. Nonsense mutations are present in approximately 15% of all DMD patients [7‐9]. However, the precise analysis of the DMD mutation is required for the application of these molecular therapies to patients.

In Japan, a hospital based dataset of 127 patients harboring small mutations by Takeshima et al. in 2010 [10], and a registry based dataset in Japanese Registry of Muscular Dystrophy (Remudy) of exon deletions and duplications by Nakamura et al. [11] have been compiled. Here, we extend these previous reports not only by increasing the number of the patients (1497 patients), but also by comprehensive genetic analysis of exonic and small nucleotide mutations and immunostaining of dystrophin on muscle biopsies to deduce genotype-phenotype correlations in patients with dystrophinopathies.

In July 2009 through March 2017, a total of 1497 dystrophinopathies patients were registered in Remudy. Among them, 1167, 295, and 35 patients were respectively diagnosed having DMD, BMD, and IMD. The genetic diagnosis was made by MLPA and/or multiplex-PCR in 1092 patients with exon deletion in 901 patients (61% of 1497 dystrophinopathies patients), exon duplications in 188 (13%), and both exon deletions and duplications in 3 (0.2%). All patients harboring single exon deletion were performed sanger sequencing to confirm the deletion.

Among the remaining 405 patients, DNA sequencing revealed small mutations in 371.

One patient was found to have pericentric inversion of the X chromosome by chromosome testing. One patients were diagnosed as having facioscalpulohumeral muscular dystrophy after registration. However, the remaining 32 patients remained genetically undiagnosed.

This study retrospectively evaluated the results of gene analysis for 1497 patients with dystrophinopathies with clinical features and dystrophin immunostaining. Among the 1497 patients, 1092 (73%) were diagnosed by the MLPA method; including 901 patients with exon deletions (901/1497, 61%) and 188 patients with exon duplications (188/1497, 13%).

This is the report of the largest number of patients all previous dystrophinopathies reports from Asian countries [19‐24]. Most of the previously reports from Asian countries [19‐21] demonstrated lower deletion and higher duplication rates. This may be explained by the fact that much smaller numbers of patients were studied including female carriers. Nevertheless, Suh et al. [24], who reported the rates of deletions and duplications respectively, as 71.8%/ and 16.4% among 130 patients in Korea, which is similar not only to our results but also to the results reported from Western countries [25, 26]. In addition, exon deletion and duplication hotspots in our study (Fig. 2) are similar to those in the previous reports [20‐22, 25, 26], suggesting that ethnicity is not a factor to cause any difference in the proportion of exon deletions and duplications in DMD.

Based on our results of exon deletions/duplications, we estimated the applicability of exon skipping therapies (Tables 2 and 3). Skipping of exon 51, which has been approved by FDA, and of exon 53 skipping, which is in clinical trials, cans theoretically be applied to the largest group of the patients with exon deletions, as suggested in previous reports [27].

Among 185 patients (13%) who harbor a nonsense mutation, interestingly, 17 patients showed faint and patchy dystrophin expression pattern on immunohistochemistry (Fig .4 and Additional file 1: Table S2). Nonsense mutations found in patients with BMD patients have reported to be present in exons 27, 29, 31, 37, 49, and 72 [28‐30], all of which are in-frame exons, resulting in dystrophin positivity and milder phenotypes. Similarly 15 of the 17 patients with nonsense mutations and faint and patchy dystrophin staining in this study had mutations in in-frame exons. BMD phenotype in these patients may well be explained by the skipping of the correspondent exon, but cDNA analysis is necessary to be conclusive. From these results, there are 168 patients (12%) who are treated for nonsense codon read-through treatment.

Seven distinct missense variants were identified in 8 patients (Table 1). They may well be pathogenic since they are not reported in the HGVD, ESP6500, or ExAC and are predicted to be possibly or probably damaging in proteins by PolyPhen2. Especially, the 4 variants located in the N-terminus of dystrophin may be more likely to have pathogenic significance as previous reports showed the missense mutations in the actin binding domain of dystrophin, which is in the N-terminus region, induce thermodynamic instability of dystrophin molecules and protein aggregation [31, 32]. Needless to say, however, there still remains a possibility that patients may have unidentified causative mutations in other regions such as deep intronic regions or some missense mutations have an effect on splicing.