Molecular and in silico analyses of the full‐length isoform of usherin identify new pathogenic alleles in Usher type II patients

• Published in: Human mutation 2007-08, Vol.28 (8), p.781-789
• Main Authors: Baux, David, Larrieu, Lise, Blanchet, Catherine, Hamel, Christian, Ben Salah, Safouane, Vielle, Anne, Gilbert‐Dussardier, Brigitte, Holder, Muriel, Calvas, Patrick, Philip, Nicole, Edery, Patrick, Bonneau, Dominique, Claustres, Mireille, Malcolm, Sue, Roux, Anne‐Françoise
• Format: Article
• Language: English
• Subjects: Adolescent; Adult; Aged; Alleles; Amino Acid Sequence; Amino acids; Child; Computational Biology - methods; Cysteine; DNA Mutational Analysis; Extracellular Matrix Proteins - chemistry; Extracellular Matrix Proteins - genetics; Genotype; Hearing loss; Humans; Middle Aged; Models, Molecular; Molecular Sequence Data; Mutation; Mutation - genetics; Protein Isoforms - chemistry; Protein Isoforms - genetics; Protein Structure, Secondary; Protein Structure, Tertiary; Proteins; USH2A; Usher syndrome; Usher Syndromes - genetics; usherin
• ISSN: 1059-7794; 1098-1004
• DOI: 10.1002/humu.20513

The usherin gene (USH2A) has been screened for mutations causing Usher syndrome type II (USH2). Two protein isoforms have been identified: a short isoform of 1,546 amino acids and a more recently recognized isoform extending to 5,202 amino acids. We have screened the full length by genomic sequencing. We confirm that many mutations occur in the exons contributing solely to the longer form. USH2 is an autosomal recessive disorder and, in contrast to previous studies, both mutations were identified in 23 patients and a single mutation in 2 out of 33 patients. A total of 34 distinct mutated alleles were identified, including one complex allele with three variants and another with two. A total of 27 of these are novel, confirming that most mutations in usherin are private. Many of the mutations will lead to prematurely truncated protein but as there are a substantial number of missense variants, we have used in silico analysis to assess their pathogenicity. Evidence that they are disease‐causing has been produced by protein alignments and three‐dimensional (3D) structural predictions when possible. We have identified a previously unrecognized cysteine rich structural domain, containing 12 dicysteine repeats, and show that three missense mutations result in the loss of one of a pair of the defining cysteine‐cysteine pairs. Hum Mutat 28(8), 781–789, 2007. © 2007 Wiley‐Liss, Inc.