Abstract
The mechanism of expansion of the (CTG)n repeat in myotonic dystrophy (DM1) patients and the cause of its pathobiological effects are still largely unknown. Most likely, long repeats exert toxicity at the level of nuclear RNA transport or splicing. Here, we analyse cis- and trans-acting parameters that determine repeat behaviour in novel mouse models for DM1. Our mice carry 'humanized' myotonic dystrophy protein kinase (Dmpk) allele(s) with either a (CTG)84 or a (CTG)11 repeat, inserted at the correct position into the endogenous DM locus. Unlike in the human situation, the (CTG)84 repeat in the syntenic mouse environment was relatively stable during intergenerational segregation. However, somatic tissues showed substantial repeat expansions which were progressive upon aging and prominent in kidney, and in stomach and small intestine, where it was cell-type restricted. Other tissues examined showed only marginal size changes. The (CTG)11 allele was completely stable, as anticipated. Introducing the (CTG)84 allele into an Msh3-deficient background completely blocked the somatic repeat instability. In contrast, Msh6 deficiency resulted in a significant increase in the frequency of somatic expansions. Competition of Msh3 and Msh6 for binding to Msh2 in functional complexes with different DNA mismatch-recognition specificity may explain why the somatic (CTG)n expansion rate is differentially affected by ablation of Msh3 and Msh6.