Molecular signaling in aortic valve development and congenital aortic valve defect

ABSTRACT The normal aortic valve is tricuspid with three leaflets derived from multiple cell lineages during embryogenesis. Aortic valve patterning is genetically controlled where individual cells in the valve- forming field refine their fates and functions in response to positional and environmental cues. Genetic mutations that alter cell-cell and cell-environmental signals can disrupt the developmental process, leading to anomalous aortic valve, for example, bicuspid aortic valve (BAV). Affecting ~2% of the general population in US, BAV is the most common congenital heart defect. Fusion of two of three leaflets or absence of one leaflet during embryogenesis results in various BAV subtypes. After birth, over half of BAV patients develop calcific aortic valve disease with no effective medicine, while BAV subtypes have varied cardiac complications, which decide the disease outcome. With the International Bicuspid Aortic Valve Consortium (BAVCon) being established to identify the genetic causes of BAV in humans, animal models of BAV are critically needed to elucidate morphogenic and cellular mechanisms of human BAV, as well as molecular signals that control aortic valve patterning in order to identify therapeutic targets for disease prevention. To this end, we have generated two mouse models of BAV with distinct signaling defects and anomalous leaflets. In the first model, knocking out notch receptor 1 (Notch1) in valve endocardial cells (VECs) recapitulates the most common human BAV subtype – fusion of left and right coronary leaflets, which are mainly derived from VECs by epithelial to mesenchymal transformation. This model also reveals that the NOTCH1- TNFa signaling from VECs controls apoptosis of valve mesenchymal cells (VMCs). In the second model, deleting SRY-box transcription factor 17 (Sox17) in VECs results in a rare but more severe type of BAV – absence of non-coronary leaflet, of which VMCs arise predominantly from the second heart field (SHF)-derived cardiomyocytes, and the patterning defect is associated with reduced VEC-VMC PDGFB signaling. Based on these findings, we hypothesize that coordinated VEC-VMC signals control normal aortic valve patterning and their disruption leads to various BAV subtypes, in the context of the origin and location of affected cells. We will test this hypothesis in two Aims. Aim 1 is planned to reveal coordinated VEC-VMC signal networks during normal aortic valve patterning and identify signaling events that are disrupted in various BAV subtypes. Aim 2 is designed to uncover the functions of PDGF signaling in normal aortic valve patterning as well as use it as an example to illustrate how a disrupted signaling event can alters cell fate and function, leading to a specific BAV. Successful accomplishment of these Aims will provide new insights into BAV pathogenesis, with a broad implication in congenital heart valve disease.