Cortical organoids from congenital DM1 PSCs reveal MBNL-dependent corticogenesis defects and enable preclinical testing of therapeutic compounds

Abstract

Myotonic dystrophy type 1 (DM1) is caused by an expansion of a CTG repeat in the 3′ untranslated region of the DMPK gene, leading to accumulation of toxic CUG-repeat RNAs, sequestration of MBNL proteins and widespread splicing dysregulation. Congenital DM1 (CDM), the most severe form of the disease, is associated with profound muscular and neurodevelopmental defects, yet the mechanisms underlying early human brain involvement remain poorly understood. Here, we generated cortical organoids from patient-derived pluripotent stem cells carrying >1000 CTG repeats, an expansion typically associated with CDM, to model early human neurodevelopment. DM1 molecular and cellular hallmarks were detected at early developmental stages, including nuclear DMPK RNA foci in neural progenitor cells and reduced proliferative capacity. As organoids matured, CDM cultures displayed altered cortical composition, with reduced CTIP2⁺ and SATB2⁺ neuronal populations and increased NFIA⁺/GFAP⁺ glial cells. In parallel, 120-day-old organoids recapitulated splicing abnormalities previously identified in DM1 patient brain tissue. To assess the contribution of MBNL dysfunction, we analyzed cortical organoids derived from MBNL2 and MBNL1/2/3 knockout induced pluripotent stem cells, which reproduced key neurodevelopmental phenotypes observed in CDM organoids, supporting a central role for MBNL loss of function in impaired corticogenesis. Finally, we evaluated the translational relevance of this model using tideglusib and erythromycin, two compounds currently under clinical evaluation in DM1 patients. Both treatments reduced DMPK RNA foci and restored proliferation defects in SOX2⁺ neural progenitors. Together, these findings establish cortical organoids as a robust human model of CDM-associated neurodevelopmental defects, uncover MBNL-dependent mechanisms underlying early corticogenesis impairment and demonstrate the utility of this platform for translational therapeutic discovery in DM1.