PROJECT SUMMARY: The vascular system consists of elaborate networks that develop in combination with close regulation of endothelial cells (ECs). An understanding of such regulation is essential for the development of new treatment strategies aimed at vascular malformations, such as arteriovenous malformations (AVMs) and hereditary hemorrhagic telangiectasia (HHT), caused by mutations in activin receptor-like kinase 1 (ALK1). We previously showed that gene deletion in mice of matrix Gla protein (MGP), an inhibitor of bone morphogenetic proteins (BMPs), causes AVMs in multiple organs similar to HHT. We showed that BMP9/ALK1 signaling induces MGP expression in ECs, where MGP plays an important role in differentiation. BMP9/ALK1 signaling also induces Crossveinless-2 (CV2) with a different induction delay, thereby creating two negative feedback loops. Together, MGP and CV2 regulate BMP9 signaling by a previously unknown mechanism. In cultured ECs, we found oscillations of MGP and CV2 expression that temporally coordinated transition to EC stalk cell phenotype in ECs. This also caused markers of stalk cells to oscillate, whereas tip cell markers were suppressed. Deletion of Mgp abolished the oscillatory behavior. In vivo, MGP and CV2 were seen as “shaping waves” or stripes in the growing retina, and lack of MGP perturbed the vascular networks. Our hypothesis is that MGP and CV2 are regulators of BMP9 signaling and vascular morphogenesis through generation of oscillations or waves of expression. In Aim 1, we will characterize how MGP and CV2 orchestrate EC differentiation in response to BMP9 using oscillations of gene expression. We will relate BMP9-induced stalk cell phenotype to the oscillations, and explore expression profiles of ECs capable of this behavior. We will disrupt the system by deleting the Mgp gene in vitro using established techniques of shRNA, and determine the effect on the waves of inhibitors and stalk cell markers. We will also investigate whether waves of MGP and CV2 can be detected in normal vasculature, with focus on the retina. In Aim 2, we will obtain key information about the role of MGP in retinal vascular networks and AVMs by deleting Mgp, impairing MGP protein function, and modulating the cellular origin. We will modulate potential targets for AVM treatments using the Mgp-/- mice as an AVM model. We will start with modulation of CV2 and use approaches that include crossbreeding with genetically altered mice and transmammary immunoblocking, and subsequently screen other factors in the BMP9 response. Our studies will help identify targets in the BMP9 response system that might be used in designing treatments for AVMs.