Genetic modifiers of repeat expansion disorders

Abstract

Repeat expansion disorders (REDs) are monogenic diseases caused by a sequence of repetitive DNA expanding above a pathogenic threshold. A common feature of the REDs is a strong genotype-phenotype correlation in which a major determinant of age at onset (AAO) and disease progression is the length of the inherited repeat tract. Over a disease-gene carrier's life, the length of the repeat can expand in somatic cells, through the process of somatic expansion which is hypothesised to drive disease progression. Despite being monogenic, individual REDs are phenotypically variable, and exploring what genetic modifying factors drive this phenotypic variability has illuminated key pathogenic mechanisms that are common to this group of diseases. Disease phenotypes are affected by the cognate gene in which the expansion is found, the location of the repeat sequence in coding or non-coding regions and by the presence of repeat sequence interruptions. Human genetic data, mouse models and in vitro models have implicated the disease-modifying effect of DNA repair pathways via the mechanisms of somatic mutation of the repeat tract. As such, developing an understanding of these pathways in the context of expanded repeats could lead to future disease-modifying therapies for REDs.

Keywords: DNA synthesis and repair; cag repeat; genetic modifier; repeat expansion; somatic DNA expansion; somatic instability.

Figure 1.. Normal MMR vs mechanisms that may lead to expansion of repeats in non-dividing cells.

The figure illustrates the working model based on the current strongest evidence. The MutS and MutL complexes: There are two MutS complexes; MutSα, which contains MSH2/MSH6 or MutSβ comprising MSH2/MSH3. MutSα preferentially recognises and targets mismatches and 1–4 base INDELS while MutSβ targets medium-sized loop repair [91]. There are three MutL endonuclease complexes; MLH1 complexed with either PMS2, MLH3 or PMS1 creating MutLα, MutLγ and MutLβ, respectively. MLH3 is involved in meiotic repair but can compensate a small amount for MutLα while PMS1's role is unclear and cannot compensate for MutLα. When DNA damage occurs, in canonical MMR, the MutL complexes’ endonuclease is thought to create a DNA break in the strand with a mismatch (which is assumed to be the strand carrying an incorrect base). MutL induced breaks are the initiation sites for strand excision performed by exonuclease 1. Canonical and Non-Canonical MMR Hypothesis in non-dividing cells (1) Canonical MMR: (1A) The figure illustrates the reaction of MMR to a small loop out. In canonical MMR mismatches, small loops and insertions and deletions (INDELSs) are resolved by the recruitment of one of two MutS complexes; MutSα or MutSβ. MutSα preferentially recognises and targets mismatches and 1–4 base INDELS while MutSβ targets medium loop repair [33, 91]. This occurs when cells are dividing or the DNA is transcriptionally active causing it to unwind and become single stranded. (1B) Once bound, the MutS complex induces recruitment of MutL endonuclease complexes. MutLα is the principle MutL complex for most MMR. The MutL complex creates a DNA break in the strand with an existing break. (1C) Excision is then performed by exonucleases e.g. exo 1. (1D) There is then faithful repair involving DNA polymerase using the opposite strand as the template strand. (2) Hypothesised Non-Canonical MMR: (2A) Strand separation during transcription (or replication) permits pathogenic repeat sequences to form secondary structures e.g. hairpins, R-loops and cruciforms. This figure illustrates the potential downstream effects of large loops. It is thought that these structures act as the substrate for non-canonical MMR. Large loops of 2–10 bases can only be resolved through recognition by MSH3 in MutSβ. (2B) The MutL complex is then recruited and unlike in canonical MMR, MutL complex erroneously creates a break in the strand opposite the loop. (2C) One hypothesis is that FAN1 nuclease cuts the strand opposite the loop i.e. the complementary strand (though the location of its action has not been fully elucidated). Therefore the strand with the loop is used as the template strand. (2D) There is then erroneous resolution of the loop resulting in elongation of the repeat sequence as Polymerase uses the strand with the loop as the template strand, thus incorporating new repeats into the gene.