Fabrication and Characterization of Gelatin-based Polyurethane Vascular Graft for Tissue-engineering Applications

INTRODUCTION Peripheral artery disease and related revascularization procedures are increasing due to the aging population and growing incidence of diabetes mellitus. Up to now, autologous saphenous vein represents the conduit of choice for peripheral by-pass. Synthetic vascular graft in polyethylene terephthalate (Dacron®) and expanded polytetrafluoroethylene (ePTFE) are used when the vein is not available. However they have shown poor patency when used in small-diameter sizes (<6 mm) due to thrombosis and intimal hyperplasia. Attempts have been made to reduce the thrombogenicity of synthetic grafts by seeding the lumen with endothelial cells, therefore, the development of scaffolds that can support cell growth has become an important issue in the field of vascular tissue engineering. The aim of this study was to prepare a composite vascular graft by spray, phase-inversion technique constituted by polyurethane and gelatin able to support cell growth. EXPERIMENTAL METHODS Gelatin-based polyurethane (PU-gel) grafts were fabricated co-spraying polyurethane (Estane®5714F1 2% in THF/DX) and gelatin aqueous solution (2%) to obtain a composite vascular graft. After materials deposition, grafts were placed in dH2O at 4°C o.n. to allow solvents removal. Cross-linking of gelatin (PUCLKgel) was obtained by exposition to glutaraldehyde vapour for 2 h at r. t. Then samples were immersed in 50 mmol/L glycine solution for 1 h to block residual aldehyde groups of glutaraldehyde. Polyurethane graft without gelatin (PU) was used as control. Graft microstructure was investigated by stereo- microscopical observation of Sudan Black stained samples and SEM. Uniaxial tensile tests were carried out to assess grafts mechanical properties. Static tests were performed until failure occurred on ongitudinal and circumferential directions by a computer controlled tensile testing machine (100 N load cell) according to ASTM D412-06a protocol. Six test samples were evaluated for each graft materials and for both directions. For each samples stress-strain data, ultimate tensile strength (UTS), and ultimate elongation (UE) were calculated. Human Mesenchymal Stem Cells (hMSCs) were obtained from bone marrow aspirate. Cells displaying a mesenchymal phenotype (CD34-, CD45-, CD44+, CD105+, CD90+ and CD73+) were used to evaluate the graft capability to support cell adhesion and growth. PU-CLKgel and control grafts were punched to obtain round samples (2 cm2 area). These samples were placed at the bottom of a 24-well plate and preconditioned in culture medium for 30 min. hMSCs at passage 6 (six) were seeded onto the graft luminal surfaces (1.4x104 cells/well). After 24, 48 and 72 h of incubation, cell viability was calculated by XTT assay and cells morphology was evaluated by Giemsa staining. Statistical analysis was performed using StatView(TM) by Student's t test. RESULTS AND DISCUSSION Microscopical observation of PU-CLKgel and control graft stained with Sudan Black did not show any difference at 40X magnification, however SEM analysis at higher magnification (500 and 2500X) evidenced differences between samples. The PU samples feature a microporous structure similar to that previously described relating to a polyurethane-polydimethylsiloxane graft obtained by spray, phase inversion technique [1]. The gelatin-based PU samples showed a fibrillar appearance typically to that owned by hydrogel. From the mechanical point of view, the presence of gelatin produced an increase of UTS (17.5% and 35.5%) and UE (9% and 20.6%) in both circumferential and longitudinal directions, respectively, compared to PU material. Similarly PU-CLKgel showed an increase of UTS and UE with respect to PU material. Qualitative analysis of hMSC adhesion after 24, 48 and 72 h onto the grafts revealed remarkable differences between PU-CLKgel and control graft. hMSCs grown onto PU-CLKgel graft form a monolayer that reached confluence at 72 h and appeared to be similar to the physiological morphology of hMSCs grown on culture plastic. On the contrary cells seeded onto the control graft were not able to undergo appropriate spreading and proliferation, thus displaying a rounded morphology. hMSCs grown onto PU-CLKgel graft showed significantly higher O.D. value then cells seeded onto PU luminal surface at all time-points. Moreover hMSCs onto PU-CLKgel graft showed an increase of O.D. over the time indicating a proliferative activity. CONCLUSION A composite vascular graft was successfully produced by simultaneous spraying a synthetic polymer (Estane®5714F1) and a natural biopolymer (gelatin) in order to obtain a scaffold that combines the mechanical characteristics of polyurethanes with the favourable cell interaction features of the biopolymer fibers.