Identification of novel mutations in WFS1 and genotype–phenotype correlation in Wolfram syndrome

Abstract

Mutations in the WFS1 gene have been reported in Wolfram syndrome (WS), an autosomal recessive disorder defined by early onset of diabetes mellitus (DM) and progressive optic atrophy. Because of the low prevalence of this syndrome and the recent identification of the WFS1 gene, few data are available concerning the relationships between clinical and molecular aspects of the disease. Here, we describe 12 patients from 11 families with WS. We report on eight novel (A214fsX285, L293fsX303, P346L, I427S, V503fsX517, R558C, S605fsX711, P838L) and seven previously reported mutations. We also looked for genotype-phenotype correlation both in patients included in this study and 19 additional WS patients that were previously reported. Subsequently, we performed a systematic review and meta-analysis of five published clinical and molecular studies of WFS1 for genotype-phenotype correlation, combined with our current French patient group for a total of 96 patients. The presence of two inactivating mutations was shown to predispose to an earlier age of onset of both DM and optic atrophy. Moreover, the clinical expression of WS was more complete and occurred earlier in patients harboring no missense mutation.

Identification of novel mutations of the WFS1 gene in Brazilian patients with Wolfram syndrome

“…Similarly, Hardy et al in a national UK cohort, found no clear correlation between any of the observed mutations and disease severity (15). However, Cano et al studying French patients and combining his results with those of five published studies, including the data of Hardy et al found that the presence of two inactivating mutations predisposed patients to an earlier age of onset of both diabetes mellitus and optic atrophy (24). On the other hand, we observed a milder phenotype in a patient carrying a homozygous missense mutation in exon 5.…”

Genotypic classification of patients with Wolfram syndrome: insights into the natural history of the disease and correlation with phenotype

“…36 We analyzed as mutations (i) c.2020G→A (p.Gly674Arg), considered either a mutation 37 or a polymorphism, 38 and (ii) c.1367G→A (p.Arg456His), considered either a mutation 34 or a polymorphism, [31][32][33]39,40 because both are predicted to produce deleterious proteins by Condel 41 and PolyPhen-2. 42 c.2020G→A is also considered a mutation in the LOVD-WFS1 database.…”

“…We considered as polymorphisms c.*47C→T 33,43,44 and c.*253G→A 37 because they locate to the 3′-untranslated region of WFS1 mRNA, and c.316-37C→T because its effects on splicing have not been indicated. 34 Supplementary Table S2 online lists all the mutations described in the patients.…”

“…The genotyping methods of Cano et al 33 and Chaussenot et al 30 were based on mutation types and differ from each other only in genotype naming. The genotyping method of Rohayem et al 34 was based on the effect of mutations on the function of WFS1.…”

“…31,32 The data are generally consistent with respect to single cohorts of patients, but some contradictions appear in comparisons of results from different cohorts. Cano et al 33 determined that the presence of two inactivating mutations predispose to an earlier onset of DM and OA. Chaussenot et al 30 showed no association between the effect of the genotype in the neurological signs observed in patients with WS.…”