Allelic instability in mitosis: a unified model for dominant disorders.

Recent findings indicate that tandemly repeated triplet sequences in certain disease-causing human genes may render these genes highly unstable not only in meiosis but also in mitosis. Typically, a dominant mutation arises upon expansion in the number of these repeated elements. We have considered how mitotic instability of this sort might affect both phenotypic expression and allele transmission. A model based on these considerations leads to the following predictions: (i) Phenotypic severity among individuals who inherit an unstable allele should be highly variable due to stochastic variation in the stage of its earliest mutagenic expansion. (ii) Strikingly increased severity or decreased age of onset in some offspring should arise because of parental germ-line mosaicism for an expanded or mutant allele. (iii) The magnitude of genetic anticipation should be more strongly correlated with paternal than with maternal age at the time of conception. (iv) Given a child born with a severe phenotype, the recurrence risk for a second severely affected child should be significantly elevated. (v) The severity of phenotype in a child should be positively correlated with that in a parent. Available data on fragile X syndrome, Huntington disease, and myotonic dystrophy are shown to be consistent with the model, and implications for an understanding of achondroplasia and other dominant disorders are discussed.