Therapy of corneal endothelial dysfunction with corneal endothelial cell-like cells derived from skin-derived precursors

Abstract

Corneal endothelial dysfunction occurs when corneal endothelial cells (CECs) are dramatically lost and eventually results in vision loss. Corneal transplantation is the only solution at present. However, corneal transplantation requires a fresh human cornea and there is a worldwide shortage of donors. Therefore, finding new functional CECs to replace human CECs is urgent. Skin-derived precursors (SKPs) can be easily acquired and have multiple differential potential. We co-cultured human SKPs with B4G12 cells in serum-free medium and obtained abundant CEC-like cells which had similar morphology and characteristic to human CECs. CEC-like cells exerted excellent therapeutic effect when they were transplanted into rabbit and monkey corneal endothelial dysfunction models by injection method. This protocol enables efficient production of CEC-like cells from SKPs. The renewable cell source, novel derivation method and simple treatment strategy may lead to potential applications in cell replacement therapy for corneal endothelial dysfunction.

Figure 1

Cultivation of skin-derived precursors (SKPs). (A) SKPs were cultured and subcultured in the form of floating spheres. (B) SKPs expressed nestin, fibronectin, and vimentin. Scale bars 100 μm.

Figure 2

Derivation of CEC-like cells from SKPs. (A) During co-culturing, cell morphology began to change to a polygonal CEC-like shape on day 4. After 8 days, the polygonal CEC-like cells were the majority and formed a mosaic monolayer. (B) Immunofluorescent staining showed that the CEC-like cells expressed corneal endothelium major markers Na⁺/K⁺ ATPase and ZO-1. (C) RT-PCR analysis of corneal endothelial markers in CEC-like cells and SKPs. (D) Expression of Pitx2 and FoxC1 during derivation. (E,F) Western blotting analysis and protein levels of Na⁺/K⁺ ATPase alpha 1 and ZO-1 in CEC-like cells compared to SKPs. All samples derived from the same experiment were processed in parallel. One representative experiment out of three is presented. Cropped images were displayed and full-length blots are shown in the Supplementary Figure 4. Results are mean ± S.E.M. *p < 0.05 compared to SKPs.

Figure 3

Characteristic of subcultured CEC-like cells. (A) Subcultured CEC-like cells formed a monolayer of hexagonal and pentagonal cells through 10 days of culture. (B–E) Immunofluorescent staining, Western blotting, and RT-PCR showed that the passaged CEC-like cells continued to express CECs markers. All samples derived from the same experiment were processed in parallel. One representative experiment out of three is presented. Cropped images were displayed and full-length blots are shown in the Supplementary Figure 5. Scale bars 100 μm. Results are mean ± S.E.M. *p < 0.05 compared to SKPs.

Figure 4

Clinical Observations of the CEC-like cell group and control group on day 3 and day 7 in rabbits. (A) Slit-lamp photographs showed that the clarity of the CEC-like cell group corneas significantly improved after injection, while corneal opacity and stromal edema were still serious in the control group. (B) Visante OCT showed significant corneal thickness differences in the CEC-like cell group and the control group. (C) Confocal microscope images confirmed full coverage of polygonal cells on the Descemet’s membrane in the CEC-like cell group. (D) Changes of the corneal thickness during clinical observation. There were significant differences in the corneal thickness between the control group (●) and the CEC-like cell group (○) on days 1, 3, 5, and 7. Results are mean ± S.D. (n = 10 per group) *p < 0.05 compared to control group.

Figure 5

Histological examination of the CEC-like cell group and control group on 3 day and 7 day in rabbits. (A) Alizarin red and trypan blue staining. (B) Immunofluorescent staining of frozen section (blue: DAPI, red: Dil, green: Na⁺/K⁺ ATPase). (C) HE staining (top row:100×; bottown row: 600×). Scale bars: 100 μm.

Figure 6

Clinical Observations of the CEC-like cell group and control group on day 7 and 1 month in monkeys. (A) Slit-lamp photographs showed that the corneas became significantly clearly transparent in the CEC-like cell group after injection, whereas corneal opacity and stromal edema became more and more serious in the control group. (B) Visante OCT showed significant corneal thickness differences in the CEC-like cell group and the control group. (C) Non-contact specular microscopy showed that CEC-like cells were in the form of a polygonal monolayer with a cell density of about 2500–3000 cell/mm² in the CEC-like cell group. (D) Changes of the corneal thickness during clinical observation. (E) Changes of intraocular pressures during clinical observation. Results are mean ± S.D. (n₁ = 3, n₂ = 2).

Figure 7

Three months after surgery in the monkeys. (A, B) The corneas of the CEC-like cell group were still transparent while the corneal opacity and stromal edema were still serious in the control group. (C) Non-contact specular microscopy. (D) B-mode ultrasound. (E) Fundus photography. (F) Immunofluorescent staining of frozen section (blue: DAPI, red: Dil, green: Na⁺/K⁺ ATPase). (G) HE staining and Immunohistochemical staining. Scale bars: 100 μm.