Characterizing Thrombogenic and Cell-Adhesive Properties of Poly(vinyl alcohol) Blended with Resorcinol Diphenyl Phosphate-Clay towards Vascular Graft Manufacture

Poly(vinyl alcohol) (PVA) is an excellent candidate for manufactured vascular grafts to replace occluded arteries due to its mechanical and biocompatibility properties, but its lack of cell-adhesion prevents it from being used in humans [1]. Previously, we found that resorcinol diphenyl phosphate (RDP)-clay can improve cell adhesion within polymer blends [3]. We aim here to characterize the thrombogenic and cell-adhesive properties of PVA/RDP-clay blends on both thin films and hydrogels.<br/>2% PVA (w/w) solutions with each of 0%, 0.1%, 0.5%, 1%, 2%, 3%, 5%, 7% and 10% (w/w) RDP-clay added were spun cast onto silicon wafers. Human umbilical vein endothelial cells expressing green fluorescent protein (HUVEC-EGFP) were plated on the spun-cast wafers at a concentration of 20,000 cells/cm2, then imaged on an EVOS M7000 fluorescent microscope.<br/>Good HUVEC cell adhesion with notable elongated cell morphology and cell-to-cell connections were observed on all spun cast films regardless of RDP-Clay concentration, indicating that in the absence of cross linker, PVA polymer was biocompatible, supporting cell adhesion and proliferation.<br/>Potential thrombogeneity of the spun cast films was assayed by placing the substrate in a 4mg/ml solution of human fibrinogen. When fibrinogen contacts a hydrophobic surface, protein unfolding has been reported [3], which frees the alpha-C domains and exposes interior of the molecule in the absence of thrombin. Soluble fibrin is then recruited, forming large fibers which have been shown [3] to initiate clots in the presence of plasma. While multiple large fibers were observed here on the unfilled PVA films via Atomic Force Microscopy, the number of fibers and their diameters decreased with increasing RDP clay concentration, disappearing completely at a maximum loading of 5% RDP clay. This correlated with a continuous decrease in water contact angle with increasing RDP-Clay concentration, with a critical value between 1 and 2% RDP-clay.<br/>Formation of the PVA hydrogels for the grafts, though, require cross linking with NaOH followed by rehydration of the PVA. In this case, samples with different clay concentrations were produced by dissolving 10% PVA with 0% and 10% RDP-clay, then crosslinking the solution with 15% sodium trimetaphosphate (STMP) and 30% sodium hydroxide (NaOH). The crosslinked solution was poured into 35 mm diameter molds and dried, creating hydrogel discs for cell plating. Discs were sterilized in isopropyl alcohol for 5 minutes, then rehydrated in media. Fibroblasts expressing Green Fluorescent Protein were plated atop the discs at a concentration of 2000 cells/cm^2, then imaged after 12 hours with the EVOS microscope. Without extensive washing of the discs, cells were unable to adhere to the hydrogel due to leaching NaOH from the polymer matrix. However, after hydrogel discs were subjected to daily DI exchanges for 5 days and placed under constant agitation from a magnetic stirrer, high fibroblast adhesion to both 0% and 10% RDP-clay gels was observed. This finding contradicts previous studies that show little endothelial cell adhesion to pure PVA [1], which we believe can be attributed to a lack of extensive hydrogel washing.<br/>Our novel processing method of PVA hydrogels as well as the hemocompatibility of PVA/RDP-clay blend thin films show great promise towards using this material for long-term, implanted vascular grafts in humans.<br/>[1] Ino et al. Society for Biomaterials. 2013;101(8):1549-1559.<br/>[2] Feng et al. MRS Fall Meeting; 2018 Nov 25-30; Boston, USA.<br/>[3] Zhang et al. Acta Biomaterialia. 2017;54:164-174.