Notch Initiates Endothelial-Mesenchymal-Transition in the Developing Atrioventricular Canal Through Autocrine Activation of Soluble Guanylyl Cyclase

The heart is the first organ to form during embryogenesis, and cardiac malformations are the most frequent congenital birth defects, present in ~1-2% of newborns. Valvuloseptal defects, including atrioventricular canal (AVC) defects, pulmonary atresia/stenosis, and a combination of these referred to as Tetralogy of Fallot are the most common. A key process in valve and septum development is the formation, differentiation and remodeling of cardiac cushion that eventually partitions the atria and ventricles, and ventricular outflow tract, into the dorsal aorta and pulmonary artery. Cardiac cushion development is initiated when cardiac endothelial cells (EC) undergo mesenchymal transformation (EndMT) and invade the underlying extracellular matrix (cardiac jelly). Both endocardium and myocardium play a role in EndMT. Notch activation in EC induces EndMT. At E9.5, Notch signaling induces EndMT in the AVC, where the cardiac cushions eventually develop into heart valves and the cardiac septum.  Mutation and disruptions of the Notch pathway result in cardiac malformations. We have shown that Notch activates nitric oxide (NO) signaling to initiate EndMT in the developing AVC. We used global gene expression analysis to identify the soluble guanylyl cyclase (sGC) 1A3 and 1B3 genes, which form the NO receptor heterodimer, as novel Notch targets. We also showed that Notch signaling induces EC-NO production in a paracrine fashion. Ex vivo AVC explant assays showed that NO-sGC activation contributes to cell migration and invasion at the start of EndMT. Thus we conclude that during AVC EndMT, Notch signaling activates the NO pathway by upregulating the sGC in transforming EC, and secreting a paracrine factor to induce NO synthesis. These findings offer new insight into Notch-driven EndMT in AVC development and may be transferrable to other EndMT pathologies such as cancer. This work was supported by Genome BC, Genome Canada, Heart & Stroke Foundation, and CIHR grant MOP-64354.