Ex vivo cultivated retinal pigment epithelial cell transplantation for the treatment of rabbit corneal endothelial dysfunction

Abstract Objective Stem cell therapy is a promising strategy for the treatment of corneal endothelial dysfunction, and the need to find functional alternative seed cells of corneal endothelial cells (CECs) is urgent. Here, we determined the feasibility of using the retinal pigment epithelium (RPE) as an equivalent substitute for the treatment of corneal endothelial dysfunction. Methods RPE cells and CECs in situ were obtained from healthy New Zealand male rabbits, and the similarities and differences between them were analyzed by electron microscopy, immunofluorescent staining, and quantitative real-time reverse transcription polymerase chain reaction (qRT-PCR). Rabbit primary RPE cells and CECs were isolated and cultivated ex vivo, and Na+/K+-ATPase activity and cellular permeability were detected at passage 2. The injection of cultivated rabbit primary RPE cells, CECs and human embryonic stem cell (hESC)-derived RPE cells was performed on rabbits with corneal endothelial dysfunction. Then, the therapeutic effects were evaluated by corneal transparency, central corneal thickness, enzyme linked immunosorbent assay (ELISA), qRT-PCR and immunofluorescent staining. Results The rabbit RPE cells were similar in form to CECs in situ and ex vivo, showing a larger regular hexagonal shape and a lower cell density, with numerous tightly formed cell junctions and hemidesmosomes. Moreover, RPE cells presented a stronger barrier and ionic pumping capacity than CECs. When intracamerally injected into the rabbits, the transplanted primary RPE cells could dissolve corneal edema and decrease corneal thickness, with effects similar to those of CECs. In addition, the transplantation of hESC-derived RPE cells exhibited a similar therapeutic effect and restored corneal transparency and thickness within seven days. qRT-PCR results showed that the expressions of CEC markers, like CD200 and S100A4, increased, and the RPE markers OTX2, BEST1 and MITF significantly decreased in the transplanted RPE cells. Furthermore, we have demonstrated that rabbits transplanted with hESC-derived RPE cells maintained normal corneal thickness and exhibited slight pigmentation in the central cornea one month after surgery. Immunostaining results showed that the HuNu-positive transplanted cells survived and expressed ZO1, ATP1A1 and MITF. Conclusion RPE cells and CECs showed high structural and functional similarities in barrier and pump characteristics. Intracameral injection of primary RPE cells and hESC-derived RPE cells can effectively restore rabbit corneal clarity and thickness and maintain normal corneal function. This study is the first to report the effectiveness of RPE cells for corneal endothelial dysfunction, suggesting the feasibility of hESC-derived RPE cells as an equivalent substitute for CECs.