Abstract
Purpose :
The human corneal stroma is a hydrated extracellular matrix (ECM) with 80% water, 13.6% collagen, and 0.9% glycosaminoglycans (GAGs) containing specialized fibroblast-like cells in the matrix. Its collagen is mainly type I collagen fibrils weaving together. The most abundant GAG in the normal cornea is keratan sulfate (KS), chondroitin sulfate (CS), and dermatan sulfate (DS). They form proteoglycans with a core protein in the spaces among the major collagen fibrils. GAGs can absorb and retain large amounts of water. Hyaluronan (HA) is a non-sulfated-GAG with well-known effects on corneal wound healing. We are developing in situ-forming semi-interpenetrating polymer network-based gels to mimic corneal stromal ECM for corneal regeneration and report on our recent data on the biological responses to them in vivo.
Methods :
We mixed CS, DS, HA, or their combinations with type I collagen (Col) at a Col-to-GAG ratio of 2:1 and then used FDA approved NHS ester chemistry to crosslink collagen. The GAGs are therefore physically but not covalently integrated in the gel. The crosslinker for collagen here is a bifunctional PEG succinimidyl ester. The Col-to-crosslinker ratios used were 3:4 or 3:8. Gels without GAGs (Col-PEG) are also used as controls. The mixture is then added to a corneal defect of New Zealand White Rabbits under ambient conditions and allowed to gel over 5 minutes without using a catalyst or light activation. A contact lens and tarsorrhaphy were then applied. The treated cornea was monitored by OCT, topography, slit lamp, and fluorescein staining photography. The cornea was collected for immunofluorescence staining, histology, qPCR, and western blot 7 days later.
Results :
No ocular inflammation or damage was found in the surrounding epithelial, stromal, and endothelial layers for all experimental eyes. Both collagen gels containing CS and HA promoted corneal epithelial wound healing compared to the collagen-only gels, and those containing HA were observed to be the most transparent. Stiffer gels (Col:crosslinker=3:4) facilitated curvatures closer to pre-operative curvatures.
Conclusions :
We found that integration of CS and HA within collagen gels promoted corneal epithelial wound healing compared to collagen gels alone, while stiffer gels more closely facilitated native corneal curvature restoration. Future studies will focus on further optimization of gels in rabbit models.
This is a 2021 ARVO Annual Meeting abstract.