Hydrogel-Mediated Sustained Delivery of Corneal Epithelial Extracellular Vesicles: A Strategy for Enhanced Corneal Regeneration

Abstract

Extracellular vesicles (EVs) derived from corneal epithelial cells have shown great promise in promoting corneal wound healing and stromal regeneration, but they face challenges with rapid clearance from the eye. This study addresses these challenges by developing a biocompatible collagen-hydrogel sustained delivery system. We successfully isolated, purified, and characterized corneal epithelial EVs (CE-EVs), assessed their efficacy in corneal epithelial healing in vitro, and demonstrated their sustained delivery over 10 days followed by an on-demand release through enzymatic degradation of the hydrogel, which mimics the in vivo scenario. To develop a microscale understanding of the EV diffusion inside the hydrogel matrix, we probed the hydrogel network with several model compounds and nanoparticles by using advanced confocal microscopy analyses, followed by fitting our results to established diffusion models. Our findings suggest this innovative approach offers a safe and effective strategy to promote corneal wound healing. This technology has the potential to revolutionize corneal injury treatment and improve patient outcomes. Moreover, the possibility to tailor EV-release kinetics broadens the scope of EV research in clinical practices, as varying short- and long-term release profiles will be required for diverse medical applications.

1

Size and morphology of CE-EVs after purification. (a) Nanoparticle tracking analysis (NTA) showing the average size distribution of the CE-EVs (for n = 9 batches). (b) TEM images of CE-EVs at different magnifications showing the size and shape of the CE-EVs.

2

Protein markers demonstrated by (a) FACS with the MACSPlex kit demonstrating the surface markers of the CE-EVs, (b) Western blot showing the expression of calnexin, CD63, CD81, fibronectin, and collagen in CE-cells and CE-EVs, and (c) enhanced proteins in CE-EV samples compared to CE cells determined with LC–MS proteomics. Error bars in the graph represent the standard deviation of n = 8 batches.

3

(a) Scratch assay of human corneal epithelial cells and (b) results of healing percentages for each group of the scratch assay. Error bars in the graph representing the standard deviation of n = 5 replicates. Significant differences are shown by “*” in the graphs, “*” representing a p-value of <0.05, “**” a p-value of <0.01, and “***” a p-value of <0.001.

4

EV release. (a) Schematic of the EV release through diffusion and then after 35 days through collagenase degradation. (b) Release profile and fitted model of EVs from the hydrogel through diffusion.

5

EVs in the hydrogel. (a) Confocal images of extracellular vesicles (EVs) and Silica nanoparticles (SiNPs) in the hydrogel showing both bulk phase and the edge, (b) fitting of Ogston–Amsden and Clague–Philips diffusion models for the hydrogel network from experimentally determined self-diffusion coefficients (D _g) of various probes within the hydrogel assuming two different collagen fiber radii (R _f) and position of the empirical self-diffusion coefficient obtained from the EV-release experiment, (c) size distribution from the nanoparticle tracking analysis (NTA) of CFSE-labeled EVs, and (d) schematic of the EV-release theory where smaller particles can diffuse out, while larger particles are obstructed by the hydrogel network.