Bioengineering a coronary stent with covalently immobilised tropoelastin

This thesis describes the characterisation and development of covalently immobilised recombinant human tropoelastin (TE) on a plasma-activated coating (PAC) as a potential stent coating for the treatment of coronary artery disease. A biomimetic approach was used to create a biocompatible coating with an immobilised human protein to enhance biointegration of an implanted stent. A coating that enhanced endothelialisation while displaying low tbrombogenicity was developed and characterised in vitro and in vivo. Covalent binding of TE to PAC was verified using ELISA and radiolabelled TE. Modulating the gas composition of the PAC, and therefore its mechanical and biological properties, resulted in varying amounts of covalently bound TE. The nitrogen containing PA Cs covalently bound up to 89± 1 % of physisorbed TE. The N2/Ar PAC covalently bound a monolayer of TE and was chosen for further characterisation. The covalent binding capacity of PAC extended for at least a year, retaining 65±1 % of its covalent TE binding capacity. Restoration of the full covalent binding capacity was achieved upon heat treatment of the PAC. TE was shown to support the attachment and proliferation of endothelial cells (ECs) when physisorbed to tissue culture plastic (TCP). This was comparable to other adhesive extracellular matrix proteins, fibronectin and collagen. The morphology and distribution of ECs cultured on 316L SS, PAC and PAC+TE was investigated using reflective, fluorescence and scanning electron microscopy. PAC+TE supported increased endothelial attachment and proliferation compared to uncoated 3 l 6L SS and PAC. An EC phenotype was confirmed on 316L SS, PAC and PAC+TE by immunofluorescent labelling of endothelial cell specific markers, CD3 I and vWF. As the thrombogenicity of blood contacting medical devices is crucial, methodology was developed to test the haemocompatibility of metallic surfaces in vitro. In static adhesion assays using whole heparinised blood, PAC was found to confer low thrombogenicity compared to 3 l 6L SS, and nitrogen again modulated this property. PAC and PAC+TE showed lower thrombogenicity than 316L SS after 60 min incubation. A modified Chandler loop was developed to test the tbrombogenicity of metallic surfaces in the presence of flowing blood. PAC and PAC+TE were again found to display low thrombogenicity, resulting in a 3-fold increase in the time to thrombus formation compared to 3 I 6L SS. This effect correlated with a 65±1 % increase in soluble P-selectin, a platelet activation marker on 3 l 6L SS. No significant platelet activation occurred on PAC or PAC+TE. The low thrombogenicity of PAC was retained for between 3 and 7 months. Furthermore, TE coated 3 I 6L SS displayed lower thrombogenicity than uncoated 316L SS, or fibronectin-or collagen-coated 316L SS. The PAC was translated to a 316L SS laser cut stent for evaluation and in vivo testing. The PAC deposition was altered to coat all surfaces and resisted delamination. In vitro crimping and expansion of the PAC stent showed only the formation of nanocracks, compared to the large scale delamination observed on a commercially available Taxus Liberte stent. The covalent TE binding capacity and non-thrombogenicity of the PAC were maintained on the stent PAC. The endothelialisation of PAC and PAC+TE stents was evaluated in vivo. This study marks the first comparator analysis of bare metal stents (BMS), PAC and PAC+ TE stents in a well-characterised model of rabbit bilateral iliac stenting. PAC and PAC+ TE stents were well tolerated and showed no gross inflammatory response. Cell coverage of stent struts occurred by 7 days post-implantation with endotheJialisation occurring both between the struts and over the struts in all samples. PAC and PAC+ TE showed no difference in the rate of endothelialisation compared to BMS, the standard corrunerciaJly available stents. In further work, covalently immobilised TE was found to be susceptible to proteolytic cleavage by the common blood plasma proteases kallikrein and thrombin, which predominantly cleave TE at its arginine 515 residue. A mutant form of TE, R5 I 5A was shown to resist proteolytic cleavage at the 515 residue and thus retained the C-terminus of the protein which is required for cell attachment. This mutant form of TE retained the equivalent level of covalent binding to PAC and would therefore be a suitable candidate for application to a PAC stent for in vivo evaluation