Endothelial regulation of extracellular matrix in the aortic valve

Abstract

It has been well known that the extracellular matrix (ECM) is important to the aortic valve function and that mechanical forces and cell-cell communication can regulate ECM remodeling. However, it has not been determined which cells regulate ECM production by valve interstitial cells (VICs). Thus, this study aimed to investigate if the valve endothelial cells (VECs) can influence ECM production and whether there is a differential role of VECs from either surface of the valve. In addition, the study aimed to assess the suitability of adipose derived stem cells (ADSCs) to differentiate into VECs with side-specific characteristics, so that could be used to populate tissue engineered heart valves. Firstly, a reliable method of side-specific isolation and culture of VECs was developed. Consequently, the phenotypic characteristics of aortic VECs (aVECs) and ventricular VECs (vVECs) were investigated in vitro by immunostaining, western blots and protein array. Endothelial markers vWF, CD-31, VE-cadherin and eNOS and the release of cytokines IL-1β, IL-8, and TGF-β were assessed in cultured VECs from either side of the valve. However the distinct phenotypes between aVECs and vVECs in culture were not established. The differential ability of side-specific VECs to regulate the production of ECM components by VICs was observed when VICs were cultured with media containing molecules released from side-specific VECs, in static culture, and when aVECs or vVECs were co-cultured with VICs. The media collected from vVECs increased amount of collagen by VICs while the media from aVECs showed no significant changes in the content of ECM. Interestingly, in the co-culture system, vVECs-VIC co-culturing enhanced the amount of both collagen and glycosaminoglycans (GAGs) whereas aVECs did not affect the ECM proteins. Moreover side-specific VECs were exposed to the oscillatory and laminar shear stresses (flow patterns experienced by aortic and ventricular surface of the valve, respectively). Their distinct responses on the production of ECM by VICs were investigated. Aortic VECs exposed to oscillatory flow had higher content of collagen and GAGs while vVECs did not share this effect. The laminar shear stress on both sides of the valve maintained elastin content (as compared to fresh tissue). ADSCs were also exposed to side-specific patterns of flow to assess their ability to differentiate into side-specific VECs. With respect to these factors, the exposure of ADSCs to the oscillatory shear stress induced the differential expression of NOS III and SMαA, similar to reported differences found between porcine aVECs and vVECs. Both flow patterns also increased the endothelial function of ADSCs by up-taking low-density lipoprotein. In conclusion, this study reveals the unique functional phenotypes between the aortic and ventricular VECs, and the unique communication between VECs and VICs that is mediated by shear stress to regulate the specific production of ECM components. ADSCs appear to have the capacity to express endothelial markers, but a full functional assessment of their communication with VICs remains to be investigated. This information is important to the development of a living tissue engineered heart valve.