(C-105) Engineering a Retinal Pigment Epithelium Mimetic Using Tunable PEG Hydrogels

     To mimic the ECM, PEG-Nb hydrogels were functionalized with adhesive peptides crosslinked with MMP-degradable peptides via thiol-ene step-growth reactions (Fig. 1). PEG-Nb was synthesized from 20-kDa 8-arm PEG through an amine intermediate to limit degradation to only the MMP-crosslinker. Nb functionalization was ³90% based on ¹H-NMR.

Adhesive peptides and MMP-degradable crosslinker (GKKCGPQGIWGQCKKG (SDL)) were synthesized using a CEM Liberty Blue peptide synthesizer. Cysteine residues are critical for hydrogel incorporation as the thiol partner of the thiol-ene reaction. Peptides were purified using preparatory phase high-performance liquid chromatography to >90% and validated for expected molecular weights using matrix-assisted laser desorption ionization-time of flight mass spectrometry.

            Hydrogels were formed by dissolving 5 wt% PEG and peptides with MMP degradable crosslinker at molar ratios to -ene to reach 80% maximal crosslinking to enable incorporation of adhesive peptides up to 3.6 mM in Dulbecco’s phosphate buffered saline (PBS). Lithium phenyl-2,4,6-triethylbenzolphosphinate photoinitiator was added at 0.05 wt% (Fig. 1). Hydrogel precursor solution was injected between glass slides separated by 1 mm spacers, and placed under UV light (365 nm, ~5 mW/cm²) for 3 minutes. Biopsy punch-cut hydrogel disks were placed into 24-well tissue-culture plates with 1 mL PBS and sterilized under UV-light for 30 minutes. ARPE-19 cells, an immortalized RPE cell line, were seeded on gels at 10,000 cells/cm² in 1 mL media (DMEM:F12, 10% fetal bovine serum, 1% Penicillin/Streptomycin/Amphotericin).  

            ARPE-19 adhesion was imaged over 7 days using phase-contrast microscopy and analyzed using ImageJ to determine adhesive peptide combination(s) supporting robust, rapid monolayer formation.

     ARPE-19 cells exhibited superior adhesion to gels functionalized with either RGD or RGD + YIGSR formulations compared to other gel formulations and scrambled peptide controls. For all gel formulations, the percent confluency slowly increased to a plateau at day 3. All formulations, except RGE, exhibited maximal confluency on day 4 (Fig. 2). By day 7, the adhesive peptide combination of 3.6 mM RGD alone or combinations of 1.8 mM RGD + 1.8 mM YIGSR had the greatest cell confluence of 65% and 51% respectively (Fig. 2). These data show that the addition of RGD promotes greater cell adhesion versus YIGSR and RGE alone. ARPE-19 morphology was also influenced by adhesive peptide functionalization; by day 7, cells seeded on 3.6 mM RGD-functionalized gels were less rounded and smaller than cells seeded on 1.8 mM RGD + 1.8 mM YIGSR hydrogel, which exhibited greater cell surface area and spindle-like morphology (Fig. 3). Cells seeded on the 3.6 mM YIGSR functionalized gels formed smaller cell clusters with 36% confluence at day 7 (Fig. 2). Cells seeded on the negative control, 3.6 mM RGE, exhibited minimal adhesion and a confluence of only 16% at day 7 (Fig. 2). Image analysis results suggest that RGD is crucial to promote ARPE-19 confluence but the addition of YIGSR promotes adoption of epithelial cell morphology. 

            Current efforts are investigating additional combinations of adhesive peptides with the inclusion of another laminin-derived peptide, IKVAV, given the importance of RPE-laminin interactions for the oBRB. These data will provide further insight into how ECM biochemical cues guide ARPE-19 adhesion and function. Following this, the effects of MMP-degradable crosslinkers on RPE adhesion and proliferation will be tested. Integrin, MMP, and gene expression analyses will analyze specific markers found in native oBRB and functional metrics, including transepithelial electrical resistance (TEER) and RPE polarization.