Single-Cell Multimodal Profiling Reveals Maternal Diabetes-Nkx2-5 Interaction in Congenital Heart Disease in a Genetic Model Recapitulating Human Embryopathies
The etiological complexity of congenital heart disease (CHD) arises from the interaction between genetic and environmental factors. We hypothesized that environmental exposure could potentiate a genetic susceptibility, leading to an increased incidence of specific cardiac defects. To test this hypothesis, we used betaV59M+ female mice exhibiting severe hyperglycemia (>28 mmol/L) without dyslipidemia, a model we characterized that faithfully recapitulates human diabetic embryopathy. These females were crossed with males haploinsufficient for Nkx2-5, a transcription factor essential for cardiac development and implicated in human CHD. Our results show that the combination of genetic susceptibility and a hyperglycemic maternal environment (GxE) leads to a higher frequency and severity of cardiac malformations, including ventricular septal defects specifically involving the muscular septum, univentricular morphologies, and dilation with thinning of the myocardium. These structural abnormalities were associated with transcriptional changes affecting the contractile identity of the chamber myocardium. Single-nucleus multiomic profiling (snRNA-seq and snATAC-seq) of the E9.5 cardiac region revealed that the interaction between maternal diabetes and Nkx2-5 haploinsufficiency profoundly reshapes transcriptional and chromatin accessibility landscapes. These changes affect not only metabolic pathways, but also cell cycle progression in second heart field (SHF) progenitors, likely contributing to the observed structural cardiac defects. Multiomic integration further identifies a disrupted indirect Nkx2-5-Aldh1a2 regulatory axis in posterior SHF progenitors, while demonstrating that maternal diabetes selectively affects a restricted subset of Nkx2-5-dependent regulatory elements associated with cardiomyocyte maturation. Our findings establish a comprehensive single-cell multiomic resource that will facilitate the integration of human GWAS and other genomic datasets, enabling the identification and functional interpretation of disease-associated loci.