Search

Abstract: The goal of our study was to determine whether proteins typical of the human mesothelial cell phenotype, such as mesothelin, HBME-1 (Hector Battifora mesothelial cell-1) protein and calbindin 2, are expressed in the human cornea, especially in endothelial cells. Cryosections and endothelial and epithelial imprints of sixteen human cadaverous corneoscleral discs were used. The presence of proteins was examined using immunohistochemistry and Western blotting, while mRNA levels were determined by qRT-PCR. A strong signal for mesothelin was present in the corneal epithelium, while less intense staining was visible in the endothelium. Similarly, higher and lower mRNA levels were detected using qRT-PCR in the corneal epithelium and endothelium, respectively. HBME-1 antibody strongly stained the corneal endothelium and stromal keratocytes. Marked positivity was present in the corneal stromal extracellular matrix, while no staining was present in the sclera. Calbindin 2 was detected using immunohistochemistry and Western blotting in the corneal epithelium, endothelium and stroma. qRT-PCR confirmed its expression in epithelial and endothelial cells. Three proteins expressed constitutively in mesothelial cells were detected in the human cornea. The possible function of mesothelin in cell–cell contact on the ocular surface is discussed. The presence of HBME-1 protein in the endothelial layer may indicate a still unknown function that could be shared with mesothelial cells of the pleura and peritoneum. The much more pronounced occurrence of calbindin 2 in the corneal epithelium compared to fewer positive endothelial cells explains the higher turnover of epithelial cells compared to the proliferatively inactive endothelium. [ABSTRACT FROM AUTHOR]

Abstract: The aim of this study was to determine if cytokeratins (CKs) 8 and 18 – typical epithelial cell markers – are constitutively expressed in adult human corneal endothelium. Cryosections, paraffin-embedded sections and corneal endothelial imprints obtained from eleven adult human corneal discs not suitable for transplantation were used. Different fixative solutions were applied before indirect immunofluorescent or enzymatic staining was performed with antibodies against CK8 (Chemicon), CK18 (Dako and Sigma) and CK8/18 (Novocastra). Semi-quantitative RT-PCR and Western blotting (mRNA or proteins were isolated from Millicell membranes) were used to determine cytokeratin mRNA and protein levels. Approximately 50% of the corneal endothelial cells were positive for CK8 (Chemicon), CK18 (Sigma) and the CK pair 8/18 (Novocastra) in the endothelium when acetone was used for fixation. Four and 52% CK18-positive cells were observed using immunofluorescent and enzymatic immunohistochemistry, respectively, when the CK18 antibody from Dako was used. No signal was detected when 4% formalin or 10% paraformaldehyde was used as a fixative, irrespective of the antibody used. CK8 and CK18 proteins and mRNAs were detected in the endothelium of all tested corneas by Western blotting or semi-quantitative RT-PCR, respectively. We detected both CK8 and CK18 in the endothelium of all specimens at both the protein and mRNA levels. These results clearly demonstrate that cells of the corneal endothelium express CKs 8 and 18 and share some features with simple epithelia. [Copyright &y& Elsevier]