Andling of RAFT for TransplantationAcellular RAFT constructs were created and trephined into 8.25 mm discs. To demonstrate ease of handling of RAFT for transplantation, we used a Tan EndoGlideTM insertion system that is used clinically to deliver DMEK or DSEK tissue to the anterior chamber. RAFT could be successfully loaded into the Tan EndoGlideTM system (Fig. 2A ), curling inwards in the intended manner that would protect the endothelial layer as it does for DMEK (Fig. 2D). An ex vivo porcine eye model was used to confirm that RAFT could be successfully delivered from the Tan EndoGlideTM to the anterior chamber through a typical 4 mm scleral wound using a pull-through 
technique (Fig. 2E ). After removal of all instruments and injection of an air bubble to position RAFT apposed to the posterior stroma, it is possible to see that RAFT remains fully intact with no signs of tearing after thepurchase SC66 culture of Human Endothelial Cells on RAFTRAFT thickness before cell seeding was assessed using OCT and found to be on average 74.162.04 mm (mean6 SD). The morphology of endothelial cells on tissue culture plastic and on the surface of RAFT was then assessed using light microscopy. The hCECL grew in strict monolayer formation comprising small polygonal cells when cultured on CS/L coated tissue culture plastic (Fig. 3A). hCECs expanded and then passaged (up to passage 3) on FNC coated tissue culture plastic displayed a polygonal morphology typical of human corneal endothelium (Fig. 3B). hCECL and hCECs were seeded at varying densities onto RAFT to determine the MedChemExpress CI-1011 optimum seeding density to produce a confluent monolayer. The background topology of acellular RAFT caused some interference with cell image capture (Fig. 3C inset). However, on closer inspection and in comparison to acellular constructs, cell morphology could still be discerned. When seeding either hCECL (Fig. 3C) or hCECs (Fig. 3D) at 2000 cells/mm2, cells attached within hours and after 24 hours hadPC Collagen for Endothelial TransplantationFigure 8. Transmission electron microscope characterisation of hCECs on RAFT. (A) Representative image showing apical microvilli (AV) on the endothelial surface of cells attached to collagen RAFT (Col). (B) Representative image showing tight junctions (TJs) between adjacent cells on collagen RAFT (Col). (C) Further evidence of tight junctions (TJs) at higher magnification. (D) Anchoring filaments (AF) from the overlying endothelium are seen extending into the collagen substrate (Col). Scale bars A, B 0.5 mm, C, D 0.2 mm. doi:10.1371/journal.pone.0050993.gformed a monolayer on the RAFT surface. The same was true for both hCECL (Fig. 3E) and hCECs (Fig. 3F) seeded at 3000 cells/ mm2 and there was no discernable difference between the cells at this 24-hour time point. The final cell density of hCECs after 4 days culture on RAFT seeded at a density of 2000 cells/mm2 was calculated and found to be on average 1941.2 cells/mm2.Endothelial Cell Marker Expression of Cells on RAFTThe effect of cell density and surface coating on the expression of ZO-1 and Na+ K+ -ATPase was tested using the hCECL. No apparent differences were seen in the pattern of ZO-1 expression in cells seeded onto RAFT at 3000 cells/mm2 on CS/L coating compared with FNC coating (Fig. 5A and B, respectively). Equally, cells seeded at different concentrations, 2000 compared with 4000 cells/mm2, showed comparable expression patterns of Na+/K+ATPase (Fig. 5C and D, respectively). This consistent patt.Andling of RAFT for TransplantationAcellular RAFT constructs were created and trephined into 8.25 mm discs. To demonstrate ease of handling 
of RAFT for transplantation, we used a Tan EndoGlideTM insertion system that is used clinically to deliver DMEK or DSEK tissue to the anterior chamber. RAFT could be successfully loaded into the Tan EndoGlideTM system (Fig. 2A ), curling inwards in the intended manner that would protect the endothelial layer as it does for DMEK (Fig. 2D). An ex vivo porcine eye model was used to confirm that RAFT could be successfully delivered from the Tan EndoGlideTM to the anterior chamber through a typical 4 mm scleral wound using a pull-through technique (Fig. 2E ). After removal of all instruments and injection of an air bubble to position RAFT apposed to the posterior stroma, it is possible to see that RAFT remains fully intact with no signs of tearing after theCulture of Human Endothelial Cells on RAFTRAFT thickness before cell seeding was assessed using OCT and found to be on average 74.162.04 mm (mean6 SD). The morphology of endothelial cells on tissue culture plastic and on the surface of RAFT was then assessed using light microscopy. The hCECL grew in strict monolayer formation comprising small polygonal cells when cultured on CS/L coated tissue culture plastic (Fig. 3A). hCECs expanded and then passaged (up to passage 3) on FNC coated tissue culture plastic displayed a polygonal morphology typical of human corneal endothelium (Fig. 3B). hCECL and hCECs were seeded at varying densities onto RAFT to determine the optimum seeding density to produce a confluent monolayer. The background topology of acellular RAFT caused some interference with cell image capture (Fig. 3C inset). However, on closer inspection and in comparison to acellular constructs, cell morphology could still be discerned. When seeding either hCECL (Fig. 3C) or hCECs (Fig. 3D) at 2000 cells/mm2, cells attached within hours and after 24 hours hadPC Collagen for Endothelial TransplantationFigure 8. Transmission electron microscope characterisation of hCECs on RAFT. (A) Representative image showing apical microvilli (AV) on the endothelial surface of cells attached to collagen RAFT (Col). (B) Representative image showing tight junctions (TJs) between adjacent cells on collagen RAFT (Col). (C) Further evidence of tight junctions (TJs) at higher magnification. (D) Anchoring filaments (AF) from the overlying endothelium are seen extending into the collagen substrate (Col). Scale bars A, B 0.5 mm, C, D 0.2 mm. doi:10.1371/journal.pone.0050993.gformed a monolayer on the RAFT surface. The same was true for both hCECL (Fig. 3E) and hCECs (Fig. 3F) seeded at 3000 cells/ mm2 and there was no discernable difference between the cells at this 24-hour time point. The final cell density of hCECs after 4 days culture on RAFT seeded at a density of 2000 cells/mm2 was calculated and found to be on average 1941.2 cells/mm2.Endothelial Cell Marker Expression of Cells on RAFTThe effect of cell density and surface coating on the expression of ZO-1 and Na+ K+ -ATPase was tested using the hCECL. No apparent differences were seen in the pattern of ZO-1 expression in cells seeded onto RAFT at 3000 cells/mm2 on CS/L coating compared with FNC coating (Fig. 5A and B, respectively). Equally, cells seeded at different concentrations, 2000 compared with 4000 cells/mm2, showed comparable expression patterns of Na+/K+ATPase (Fig. 5C and D, respectively). This consistent patt.
