Fig 1.
Histology of excrescences of Descemet’s membrane in patients with FECD.
(A) Representative PAS staining images of a healthy donor cornea (left).Representative PAS staining images of a cornea obtained from a patient with FECD. Excrescences, which are clinically called guttae”, were observed on Descemet’s membrane of the patient with FECD. Scale bar: 50 μm (right). (B) Ultra structural analysis of Descemet’s membrane of non-FECD donor cornea was observed using TEM. Scale bar: 3 μm (left). Ultrastructural TEM analysis of Descemet’s membrane of a cornea obtained from a patient with FECD. Flattened CECs adhered to the excrescences on Descemet’s membrane. Scale bar: 6 μm (right).
Fig 2.
Rabbit corneal endothelium cell (RCEC) injection in the rabbit corneal endothelial dysfunction model with removal of a small central part of Descemet’s membrane.
(A) The corneal endothelium was totally scraped from the Descemet’s membrane with a 20-gauge silicone needle, leaving the remaining Descemet’s membrane intact. These eyes were used as the circular descemetorhexis (CCD) (-) group (left). As the CCD (+) group, a 4-mm diameter of Descemet’s membrane was removed by CCD following total corneal endothelium removal. The gray area indicates the area where the corneal endothelium was removed. The green area indicates the area where Descemet’s membrane was removed. (B) A total of 5.0×105 RCECs, suspended in 200 μl of medium supplemented with 100 μM Y-27632 ROCK inhibitor, was injected into the anterior chamber of the (CCD (-) and CCD (+) groups) (n = 6). Six eyes from which the corneal endothelium was totally removed and Descemet’s membrane remained intact were used as controls. Corneal transparency was restored by intracameral injection of RCECs both in CCD (-) and CCD (+) groups, while control eyes exhibited hazy corneas due to corneal endothelial dysfunction.
Fig 3.
Evaluation of restored corneal endothelium by contact specular microscopy.
Restored corneal endothelium was evaluated using contact specular microscopy 14 days after rabbit corneal endothelium cells (RCECs) injection. A representative image showed that monolayer corneal endothelium was restored in the eyes without circular descemetorhexis (CCD) (-) model) (upper panel). In the eyes with circular descemetorhexis (CCD (+) model), hexagonal monolayer corneal endothelium was observed throughout the center to the periphery. The edge of the CCD was observed, and cell density and morphology were similar for the regenerated RCECs directly on the corneal stroma and on Descemet’s membrane. Cell density is likely to be similar regardless the area (middle panel). No corneal endothelial image was observed in the control eyes, which were not injected RCECs (Fig 3, lower panel).
Fig 4.
Histological assessment of corneal endothelium regenerated by injection of rabbit corneal endothelium cells (RCECs).
Regenerated corneal endothelium was evaluated by immunofluorescent staining 2 weeks after RCECs injection. The eyes without circular descemetorhexis ((CCD) (-) model) showed barrier function-related markers (N-cadherin and ZO-1) and the pump function-related marker Na+/K+-ATPase at the cell-cell border. Expression patterns of those markers were similar in both the central and the peripheral areas. Actin staining showed that restored corneal endothelium had a polygonal contact-inhibited phenotypic pattern in both the central and peripheral areas (1st to 2nd lines). In the eyes with circular descemetorhexis (CCD (+) group), N-cadherin, ZO-1, and Na+/K+-ATPase were expressed at the cell-cell border. The expression patterns of these markers in the center, where Descemet’s membrane was removed, was almost the same as in the peripheral area where Descemet’s membrane was not removed. Actin staining showed that the restored corneal endothelium was morphologically similar in both the central and peripheral areas (3rd to 4th lines). In the control eyes, which were not injected with RCECs, almost no cells were observed on Descemet’s membrane (5th to 6th lines). Scale bar: 50 μm.
Fig 5.
Assessment of the effect of circular descemetorhexis (CCD) on clinical parameters.
(A) Scheimpflug images showed the restoration of an anatomically normal cornea in both the CCD (-) and CCD (+) models at 14 days after surgery. Control eyes showed corneal edema due to corneal endothelial dysfunction. A color map of corneal thickness showed that corneal thickness was thinner in both the CCD (-) and CCD (+) models when compared to the control. (B) The corneal volume was similar, when measured with a PentacamTM instrument at 3, 5, 7, and 10 mm diameters, in both the CCD (-) and CCD (+) models (n = 6). The corneal volume of the control group is not shown, as it was not evaluated with the PentacamTM instrument due to severe corneal edema. N.S. indicates no statistical significance. (C) The central corneal thickness was evaluated with an ultrasound pachymeter for 2 weeks after rabbit corneal endothelium cell (RCEC) injection. The eyes of the CCD (-) and CCD (+) models showed recovery of the central corneal thickness to an almost normal range, whereas this thickness did not recover in the controls. However, recovery of corneal thickness was slower in the CCD (+) model than in the CCD (-) model (n = 6). **P < 0.01, *P < 0.05. (D) Cell density of the regenerated corneal endothelium was determined by analyzing immunofluorescence staining images using ImageJ software. The average cell density of the restored corneal endothelium was similar for the CCD (-) and CCD (+) models in both the central and peripheral areas (n = 6). N.S. indicates no statistical significance. (E) Intraocular pressure (IOP) was evaluated with a Tonovet® instrument, and no abnormal IOP elevation was observed in any of the groups (n = 6).