Protective effect of lycium barbarum polysaccharide on cornea epithelial-stromal injury : a 3D in vitro and in vivo study

Abstract

Corneal stromal scarring is associated with the uncontrolled proliferation and differentiation of corneal fibroblast into myofibroblast. Upon epithelial-stromal damage, transforming growth factor-beta 1 (TGF-b1) will be released from epithelial membrane to activate corneal fibroblasts to differentiate into myofibroblasts. Myofibroblasts will express smooth muscle actins and collagen proteins for wound contraction, which results in scar formation. Stromal scarring is the fifth most common global cause of blindness. This results in a lower visionrelated quality of life and imposes economic burden to patients. Yet, current medications for corneal scarring confer side effects to the visual system, thus a novel therapy to minimize scar formation is warranted. Lycium barbarum polysaccharide (LBP), a major ingredient found in Goji berries, has been proven various therapeutic effects to our body including anti-fibrotic effects in liver studies. We aim to investigate whether LBP can demonstrate protective effects to the cornea to reduce the undesirable consequences after injury. This research used both in vitro and in vivo models to determine whether 1) LBP is effective in minimizing the differentiation of corneal fibroblasts into myofibroblasts, 2) LBP can minimize the uncontrolled proliferation of myofibroblasts after stromal injury, 3) LBP can lower the expression of angiogenic factors of corneal fibroblasts, and 4) LBP can promote corneal epithelial wound healing and minimize stromal edema. For the in vitro study, we adopted a collagen type I-based three-dimensional eye-on-a-chip model to mimic the native environment of the stroma, where primary human corneal fibroblasts suspended in collagen type I. TGF-b1 (10 mg/mL) was added to activate corneal fibroblasts. Human corneal epithelial cells were as well cultured using a typical twodimensional model to look for the effect of LBP on re-epithelialization rate. Regarding the in vivo study, a mouse model with alkaline burn was used to assess re-epithelialization and stroma edema. Our in vitro study showed that 2 mg/mL LBP pre-treatment can minimize the expression of pro-fibrotic proteins and cytokines upon TGF-b1 stimulation based on immunocytochemistry results and enzyme-linked immunosorbent assays. Biophysically, pre-treating corneal fibroblasts with LBP resulted in a reduction in hydrogel contraction and stiffness, reflecting that LBP may reduce the formation of contractile fibres. LBP could as well reduce undesirable proliferation of corneal fibroblasts while showing no significant cytotoxicity. Apart from that, LBPpretreated corneal fibroblasts revealed a lower expression of angiogenic factors, suggesting that LBP may lessen the chance of corneal neovascularization. Corneal epithelial cells showed an increased migration rate with LBP pre-treatment, with no obvious changes in viability. Results from in vivo study were in line with the in vitro findings, where 2 mg/mL LBP pre-treatment on mouse cornea showed a reduced opacity with an enhanced re-epithelialization three days after alkaline burn. Cornea stromal thickness was increased after chemical burn. Yet, the stromal thickness was preserved in the LBP pre-treated groups, which is similar to that of the control group, illustrating that LBP may be effective in protecting the cornea from edema and over-hydration. Vimentin expression, an intermediate filament of corneal fibroblasts, was as well reduced in LBP pre-treatment group in the early inflammatory stage after injury. This study demonstrated the protective effect of LBP upon epithelial-stromal injury to minimize fibroblasts differentiation, proliferation and angiogenic factors expression. LBP was shown to promote epithelial wound healing. Hence, we believe that LBP may be a novel topical therapy prior to corneal refractive surgeries to minimize the adverse consequences.