Improved corneal epithelial cell growth over artificial cornea materials is required to improve device retention within the eye. In this work, varying concentrations of epidermal growth factor (EGF), a potent mitogen for epithelial cells, were immobilized to polydimethylsiloxane (PDMS) substrates, and the cellular response was analyzed. Three methods were developed to bind EGF to PDMS via polyethylene glycol (PEG) tethers. 1) Plasma Modification: EGF was first reacted with homobifunctional NHS2PEG and then bound to allylamine plasma-modified PDMS. 2) Hydrosilylation: PDMS was modified with heterobifunctional allyl-PEG-NBS and then EGF was attached to the surface-bound PEG. 3) Thiol Modification: EGF was first reacted with heterobifunctional NHS-PEG-maleimide and then bound to thiol-modified PDMS. Using Method 1 (Plasma Modification), 40 to 90 ng/cm2 of EGF was bound, however 70% of this was adsorbed even under optimized EGF-PEG reaction conditions. Cells rapidly grew to confluence on these surfaces, and cell counts increased significantly compared to control surfaces. Extracellular matrix protein production was also increased on the EGF-modified surfaces, corresponding to significantly higher levels of cell adhesion observed under a detachment force. Modification by Method 2 (Hydrosilylation) resulted in 10 to 300 ng/cm2 of bound EGF, of which 20% was adsorbed. However, despite increased EGF binding homogeneity, the cell growth was slower on these surfaces than on those prepared by Method 1, and coverage was non-uniform at all EGF concentrations. This is likely due to a higher underlying PEG density, and binding of the PEG and EGF in clusters on the surface. Simultaneous tethering of the cell adhesion peptide YIGSR had no further effect on cell coverage. Using Method 3 (Thiol Modification), 24 to 65 ng/cm2 of EGF was bound, of which 22% was adsorbed. This method enables more homogeneous EGF surface binding than Method 1, with a lower PEG density than Method 2. However, free thiol groups were inhibitory to corneal epithelial cell growth, even in the presence of bound EGF. Defunctionalization of free thiols by reaction with 3-maleimidopropionic acid restored cell growth and morphology on the PDMS, and may hence allow for retention of the proliferative effect of the EGF. These results indicate that while tethering of EGF to PDMS can improve the coverage by corneal epithelial cells, and presents a promising strategy for modification of polymeric artificial cornea materials, the effects are highly dependent on the underlying surface chemistry. / Thesis / Doctor of Philosophy (PhD)

Device anchoring of artificial cornea implants, through tissue integration of stromal tissue, is necessary to ensure long-term success. In this work, the delivery of basic fibroblast growth factor (FGF-2), a key modulator in corneal wound healing, via heparin modified materials was investigated as a means of sustained, soluble growth factor delivery for stimulation of device anchorage. Two materials types, commonly used for ophthalmic applications and currently under investigation for use in artificial cornea applications, were utilized. Poly (dimethyl siloxane) (PDMS) is currently under investigation as the base material for keratoprosthetic devices; dendrimer crosslinked collagen has been examined as the basis for use as a tissue engineered corneal equivalent. PDMS surfaces were modified directly or indirectly, through a poly (ethylene oxide) (PEO) spacer, to contain functionalized reactive NSC groups capable of binding heparin and FGF-2 Surface modifications were characterized with attenuated total reflection Fourier transform infrared spectrophotometer (ATR-FTIR), X-ray photoelectron spectroscopy (XPS) and water contact angles. Heparin coverage was assessed with metachromatic and bioactivity assays. Heparinized collagen gels were crosslinked with 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC), N-hydroxysuccinimide (NHS) and polypropyleneimine octaamine G2 dendrimers. Gel integrity was assessed with water uptake, differential sr::anning calorimetry, and heparin and dendrimer stability. Both materials were exposed to radiolabelled FGF-2 and growth factor immobilization and delivery were quantified. Heparinized PDMS surfaces were capable of binding on average 100 ng/cm2 ofFGF-2, while heparinized collagen gels had higher FGF-2 immobilization, 300 ng, likely attributed to their higher heparin densities and the fact that the bulk gel rather than the surface only was modified. Delivery of FGF-2 from the heparinized materials revealed a first order release profile, with an initial burst of FGF-2, followed by gradual growth factor release. Release rates, over a 2 week period, reached 6.5% and 50%, for 1 day and 3 day FGF-2 exposed heparinized PDMS modified surfaces, while hepruinized dendrimer crosslinked collagen gels released 40%. / Thesis / Master of Applied Science (MASc)