Access to primary human corneal endothelial cells (HCEC) is limited and donor-derived differences between cultures exacerbate the issue of data reproducibility, whereas cell lines can provide sufficient numbers of homogenous cells for multiple experiments. An immortalized HCEC population was adapted to serum-free culture medium and repeated cloning was performed. Clonally grown cells were propagated under serum-free conditions and growth curves were recorded. Cells were characterized immunocytochemically for junctional proteins, collagens, Na,K-ATPase and HCEC-specific 9.3.E-antigen. Ultrastructure was monitored by scanning and transmission electron microscopy. Two clonal cell lines, HCEC-B4G12 and HCEC-H9C1, could be isolated and expanded, which differed morphologically: B4G12 cells were polygonal, strongly adherent and formed a strict monolayer, H9C1 cells were less adherent and formed floating spheres. The generation time of B4G12 cells was 62.26 ± 14.5 h and that of H9C1 cells 44.05 ± 5.05 h. Scanning electron microscopy revealed that B4G12 cells had a smooth cell surface, while H9C1 cells had numerous thin filopodia. Both cell lines expressed ZO-1 and occludin adequately, and little but well detectable amounts of connexin-43. Expression of HCEC-specific 9.3.E-antigen was found commensurately in both cell lines, while expression of Na,K-ATPase α1 was higher in H9C1 cells than in B4G12 cells. B4G12 cells expressed collagen IV abundantly and almost no collagen III, while H9C1 cells expressed both collagens at reasonable amounts. It is concluded that the clonal cell line B4G12 represents an ideal model of differentiated HCEC, while H9C1 may reflect features of developing or transitional HCEC.

Retinal pigmented epithelial (RPE) cell integrity is critical to the maintenance of retina functions and RPE cells do not proliferate in adults. The activation of RPE results in cell proliferation which may be associated with proliferative retinopathy and choroidal melanoma. Mitogen-activated protein kinase (MAPK) is believed to be a key participant in the response to mitogenic stimuli. We therefore investigated the involvement of the extracellular signal-regulated protein kinase (ERK) 1 and 2 during the induction of RPE cell proliferation. After foetal calf serum (FCS) stimulation activation of the Ras/Raf/ERK signalling pathway was detected by Western blotting and immunochemistry, with specific anti-phosphosignalling protein antibodies. Pharmacological and antisense (AS) oligonucleotide (ODN) strategies were used to analyse the signalling involved in FCS-induced RPE cell proliferation. Activation of the small G protein Ras and, to a lesser extent of Raf-1, the kinase directly downstream from Ras, was necessary to FCS-induced cell proliferation. MEK1/2 and ERK1/2 were activated during cell proliferation. Inhibition of MEK1/2 with UO 126 completely abolished ERK1/2 activation and reduced cell proliferation by 33–43%. ERK1/2 depletion by an AS ODN approach reduced cell proliferation by 27–33%, confirming the role of ERK1/2 in the FCS stimulation of RPE cells. We also investigated the role of PKA/cAMP, one of the major inhibitory pathways of ERK1/2. PKA blockade did not modify ERK1/2 activation or cell proliferation. In contrast, agents that increased cAMP concentration, abolished RPE proliferation, and MEK/ERK activation. Moreover, inhibition of the cAMP-activated small G protein Rap1, partially reversed the inhibitory effects of cAMP on cell proliferation and MEK/ERK activation. The requirement for Ras and ERK1/2, the lack of ERK1/2 regulation by PKA and the cAMP/Rap1 counter-regulatory pathway for ERK-mediated cell proliferation suggest complex regulation of signalling in RPE cells. These data may have important implications for the development of more selective models for retinal anti-proliferative therapies.