Statistics from Altmetric.com
Recently, through homozygosity mapping followed by sequencing of candidate genes in the linkage region, Maruyama et al have discovered that OPTN is a causative gene for amyotrophic lateral sclerosis (ALS).1 They examined a total of 689 Japanese ALS subjects (92 with familial ALS (fALS), including 16 from consanguineous marriages, and 597 with sporadic ALS (sALS)) and identified three types of OPTN mutations.1 The first is a homozygous deletion of exon 5 in two siblings from a consanguineous family. The second is a homozygous nonsense mutation (p.Q398X) in a patient from another consanguineous family. The same homozygous mutation has been found in a sALS subject. The third is a heterozygous missense mutation (p.E478G) in two pairs of siblings from unrelated families. A functional study of p.E478G showed that the mutation lost the NF-κB inhibitory effect of OPTN similar to p.Q398X, supporting its causality.1
To validate the previous result using a different cohort and to further define the spectrum and frequency of the OPTN mutation in ALS, we screened the entire coding region (exons 4–16) and exon–intron boundaries of OPTN mutation in 713 ALS (687 sALS and 26 fALS) patients using a direct sequencing method. Detailed clinical information and experimental methods are described in an online supplementary note. We found 17 kinds of sequence variations (table 1 and online supplementary table 1). All were substitutions of a single nucleotide. There were eight variants in the coding region, and five of them were missense variants.
Among the missense variants, there was a previously reported mutation, c.1433A→G (p.E478G).1 p.E478 was evolutionarily conserved and localised within coiled-coil domain 2 (CCD2). Two missense variants (p.M98K and p.R545Q) have been reported as single nucleotide polymorphisms (SNPs)2 and have also been found in our controls (online supplementary table 1). The remaining two missense variants, c.277G→C (p.A93P) and c.811C→T (p.R271C), were novel. These two variants were not observed in 940 controls. p.Ala93 was evolutionarily conserved (online supplementary figure 1A) and localised within CCD1. Both the PolyPhen and the SMART programs predicted that p.Ala93Pro has a possibly damaging function against the gene product. p.R271C changed a polar amino acid (arginine) to a non-polar amino acid (cysteine) in CCD2. However, p.R271 was not evolutonarily conserved and p.C271 was found in Gallus gallus (online supplementary figure 1B). Furthermore, in silico prediction for the p.R271C was benign. Hence, we considered it unlikely to be pathogenic. Patient 67 had p.A93P (online supplementary figure 2A). Patients 462 and 594 had p.E478G (online supplementary figure 2B); the former was heterozygus and the latter was homozygus for the mutation.
We found a total of nine intronic variants; two of them were novel by comparing our data with SNPs deposited in the dbSNP database and with previous reports from elsewhere (online supplementary table 1). The range of their minor allele frequencies was 0–0.47 in our controls. All these variants were not predicted to affect splicing by in silico analyses.
Thus, in the screening of 713 Japanese ALS patients for OPTN mutations, we found a total of 17 sequence variations. We considered the two missense variants c.277G→C (p.A93P) and c.1433A→G (p.E478G) to be pathogenic. The former was not detected in 940 ethnicity-matched controls, the replaced amino acid was highly conserved among different species and the replacement was predicted to cause a serious structural change. The latter has been reported in a previous paper with functional data supporting its loss of normal function and aberrant intracellular localisation.1 Our data, including its cross-species conservation, in silico prediction and absence in 940 controls, further support its pathogenicity. In addition, we investigated the possibility of a missense mutation for one allele and a deletion for another allele in the two patients with heterozygous missense mutations by real-time quantitative PCR. However, deletions of another allele in both patients were not found (data not shown).
The overall mutational frequency of OPTN in this study was 2/687 in sALS (0.29%). Maruyama et al found 1/597 mutation in Japanese sALS.1 Thus, taken together, the mutational frequency of OPTN in Japanese sALS is 0.23%. The mutational frequency of SOD1 in Japanese sALS is 1/35 (2.9%).3 In addition, the mutational frequency of TARDBP in Japanese sALS was reported to be 0.29%.4 Thus, the mutational frequency of OPTN in Japanese sALS is similar to that of TARDBP, but tenfold less than that of SOD1. Regarding fALS, we identified one homozygous mutation in 26 fALS patients (3.8%). In the previous report by Maruyama et al,1 three types of mutations were found in 92 fALS patients (3.3%). Similarly, mutational frequencies of SOD1 and FUS in Japanese fALS are 20% and 12.2%, respectively.3 5 The mutational frequency of fALS in Japanese is also similar to that of TARDBP, but three- to fivefold less than that of SOD1 and FUS. Overall, the mutational frequency of OPTN in fALS is 3.4%, which is 11 times higher than that in sALS.
In the present study, p.E478G was identified in two unrelated ALS patients: one with sALS (patient 462) and other with fALS (patient 594) (table 1). The mutation was heterozygous in patient 462 and homozygous in patient 594, but the age at onset was 6 years earlier in the heterozygous subject. The detailed family histories for the patients were unavailable because of the BioBank Japan privacy rule. The previous study identified the p.E478G mutation in two pairs of siblings from unrelated families.1 In one family, the parents were unaffected, although the mother died before the age of 40 years from heart disease. Information about the parents was not available for the other family. Based on the family information, the authors speculated that p.E478G was an autosomal dominant trait with incomplete penetrance. They found that p.E478G increased the OPTN immunoreactivity in the cell body and neuritis, and speculated that the increased amount and different cytoplasmic distribution of the mutant protein would disturb neural functions, acting in a dominant negative fashion.1
In conclusion, we identified two types of OPTN mutations in 713 Japanese ALS patients. The mutational frequency of OPTN in Japanese fALS is similar to that of TARDBP, but —three- to fivefold less than that of SOD1 and FUS. Taken together with previous reports,1 3–5 we can conclude that OPTN mutation is not a common cause of ALS in Japanese.
We thank all ALS patients who participated in the BioBank Japan project and all members of the Japanese ALS Association as well as all participating doctors and staff from collaborating institutes for providing DNA samples. We also thank Ms T. Kusadokoro for technical assistance. The DNA samples used for this research were provided by the Biobank Japan which was supported by a grant from the Leading Project of the Ministry of Education, Culture, Sports, Science and Technology, Japan.
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
Funding This work was supported by grants from the Leading Project of the Ministry of Education, Culture, Sports, Science and Technology, Japan.
Competing interests None.
Ethics approval This study was conducted with the approval of the Center for Genomic Medicine, RIKEN, the Institute of Medical Science, and the University of Tokyo.
Provenance and peer review Not commissioned; externally peer reviewed.
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.