Engineering Cardiac Valve Extracellular Matrix Structure and Function for Modeling Pathogenesis and Regeneration

Abstract

Fibrosis of the semilunar cardiac valves clinically manifests as deterioration in late-stage functional performance. Stenotic insufficiencies, due to either congenital or acquired heart valve disease, are a result of irreversible structural aberrations to the normally highly ordered leaflet extracellular matrix (ECM). In particular, excessive remolding and deposition of ECM in the load bearing fibrosa layer of the leaflets over time leads to tissue thickening and eventual calcification. Therefore, we asked what role the fibrosa layer, essential for healthy leaflet mechanical functionality, plays in early-stage valve disease etiology and function. We hypothesized that recapitulating the valvular fibrosa ECM structure and composition will enable the self-assembly of valvular tissue to model acute fibrotic pathogenesis and provide a platform for functional tissue replacement. We tested this hypothesis by engineering both in vitro and in vivo models of semilunar valve tissues. In vitro, aortic valve interstitial cell (AVIC) tissues were engineered on two dimensional, flexible thin films and fibrous scaffolds to mimic fibrosal alignment. Engineered AVIC tissues were acutely exposed to a clinically relevant dose of the drug Pergolide, a known valvulopathogen, in order to determine early, drug-induced changes in fibrotic biomechanical function. AVIC tissues on thin films lost nearly half of their capacity to generate both basal and active tissue tone, indicative of a switch from a reparative, contractile cell phenotype to a more motile and synthetic one as a result of treatment with the drug. These data were supported by a decrease in both tissue alignment and expression of the contractile protein alpha smooth muscle actin along with increased cofilin-actin colocalization in drug-treated tissues. Additionally, AVIC tissues engineered on two dimensional, fibrous scaffolds showed slightly increased biaxial stiffness when treated acutely with the drug, suggestive of the drastic increases in stiffness observed clinically in cases of valve fibrosis. We then utilized our experience building two dimensional valve tissue models in vitro to engineer three dimensional fibrous scaffolds designed to both function immediately upon implantation as well as provide a platform for tissue regeneration in vivo. We developed an automated jet-spinning collection process to manufacture seamless, three dimensional semilunar valve scaffolds, maintaining the fibrosal alignment within leaflets. A novel, two-piece mandrel collection system enabled rapid, scalable, and controlled production of these semilunar “JetValve” scaffolds. We tailored JetValve fiber size, alignment, biaxial stiffness, and composition to recapitulate that of the native pulmonary valve leaflet fibrosa (ovine model). JetValves were functionally tested using an in vitro pulse duplicator system as well as implanted using minimally invasive techniques into the pulmonary valve position in an adult, ovine model for 15 hr. Both in vitro and in vivo tests showed JetValve performance similar to native with both a small closing volume during diastole and minimal pressure gradient across the leaflets during systole. Initial histological examination showed intact, explanted JetValves with non/minimal clotting and full-thickness neutrophil infiltration at 15 hr. Initial explants at one month time points reveal endothelialization of the leaflet surface and early cellular penetration into the scaffold. Taken together, these data demonstrate that engineered semilunar heart valve fibrosa of can serve as both a basis for both early-stage valvular tissue disease modeling as well as the functional basis of scaffolds designed for tissue replacement.