MSH2 knock-out human pluripotent stem cells as model for CTG repeat instability in myotonic dystrophy type 1

Human pluripotent stem cells (hPSC), either embryonic stem cells (hESC) or induced pluripotent stem cells (iPSC) are a powerful tool to model repeat instability in myotonic dystrophy type 1 (DM1). The mismatch repair (MMR) pathway and especially its major component, MSH2, have been linked to repeat instability as shown in MSH2 knock–out mouse models and MSH2 knock-down human cell models. We recently constructed a MSH2 knock-out hPSC model for DM1, by means of CRISPR/Cas9 genome editing. The complete absence of MSH2 in our hPSC models, allows us to define the role of MSH2 in repeat instability to a greater extent than in human MSH2 knock-down models.
Repeat instability was measured by PacBio sequencing of long range PCR fragments spanning the repeat, allowing accurate and complete assessment of the repeat length. Our preliminary data shows that the CTG repeat in two of our MSH2-/- hESC models seems to stabilize and even seems to contract compared to the unstable repeat in MSH2+/+ hESC models of the same stem cell line. MSH2+/+ hESC models have a wide repeat size distribution compared to its MSH2-/- line in which the repeat lengths are less heterogeneous and cluster around a particular CTG expansion. Our results suggest that MSH2 drives repeat instability in DM1 hPSCs and that a lack of MSH2 could stabilize the CTG repeat. Analysis of our other MSH2 knock-out hPSC lines could probably strengthen this first observation. In addition, the methylation status of the CCCTC-binding factor (CTCF) site, flanking the CTG repeat, seems to be influenced in our MSH2-/- hESC models as analysed by massive parallel sequencing. After MSH2 knock out, the CTCF1 site loses its methylation, gradually overtime, concomitant with CTG repeat stabilization.
We show that hPSC are excellent models for DM1, as we have reiterated the previously observed role of MSH2 in TNR instability. Differentiation of our MSH2 knock-out hPSC lines into disease-relevant tissues or finding the mechanism behind the observed demethylation will expand the model.