Roy Joseph

Differential epithelial and stromal protein profiles in keratoconus and normal human corneas.

The purpose of the study was to identify epithelial and stromal proteins that exhibit up- or down-regulation in keratoconus (KC) vs. normal human corneas. Because previous proteomic studies utilized whole human corneas or epithelium alone, thereby diluted the specificity of the proteome of each tissue, we selectively analyzed the epithelium and stromal proteins. Individual preparations of epithelial and stromal proteins from KC and age-matched normal corneas were analyzed by two independent methods, i.e., a shotgun proteomic using a Nano-Electrospray Ionization Liquid Chromatography Tandem Mass Spectrometry [Nano-ESI-LC-MS (MS)(2)] and two-dimensional-difference gel electrophoresis (2D-DIGE) coupled with mass spectrometric methods. The label-free Nano-ESI-LC-MS (MS)(2) method identified 104 epithelial and 44 stromal proteins from both normal and KC corneas, and also quantified relative changes in levels of selected proteins, in both the tissues using spectral counts in a proteomic dataset. Relative to normal corneal epithelial proteins, six KC epithelial proteins (lamin-A/C, keratin type I cytoskeletal 14, tubulin beta chain, heat shock cognate 71 kDa protein, keratin type I cytoskeletal 16 and protein S100-A4) exhibited up-regulation and five proteins (transketolase, pyruvate kinase, 14-3-3 sigma isoform, phosphoglycerate kinase 1, and NADPH dehydrogenase (quinone) 1) showed down-regulation. A similar relative analysis showed that three KC stromal proteins (decorin, vimentin and keratocan) were up-regulated and five stromal proteins (TGF-betaig h3 (Bigh3), serotransferrin, MAM domain-containing protein 2 and isoforms 2C2A of collagen alpha-2[VI] chain) were down-regulated. The 2D-DIGE-mass spectrometry followed by Decyder software analysis showed that relative to normal corneas, the KC corneal epithelium exhibited up-regulation of four proteins (serum albumin, keratin 5, L-lactate dehydrogenase and annexin A8) and down-regulation of four proteins (FTH1 [Ferritin heavy chain protein 1], calpain small subunit 1, heat shock protein beta 1 and annexin A2). A similar relative analysis of stroma by this method also showed up-regulation of aldehyde dehydrogenase 3A1 (ALDH3A1), keratin 12, apolipoprotein A-IV precursor, haptoglobin precursor, prolipoprotein and lipoprotein Gln in KC corneas. Together, the results suggested that the Nano-ESI-LC-MS(MS)(2) method was superior than the 2D-DIGE method as it identified a greater number of proteins with altered levels in KC corneas. Further, the epithelial and stromal structural proteins of KC corneas exhibited altered levels compared to normal corneas, suggesting that they are affected due to structural remodeling during KC development and progression. Additionally, because several epithelial and stromal enzymes exhibited up- or down-regulation in the KC corneas relative to normal corneas, the two layers of KC corneas were under metabolic stress to adjust their remodeling.

Modeling Keratoconus Using Induced Pluripotent Stem Cells

Purpose To model keratoconus (KC) using induced pluripotent stem cells (iPSC) generated from fibroblasts of both KC and normal human corneal stroma by a viral method. Methods Both normal and KC corneal fibroblasts from four human donors were reprogramed directly by delivering reprogramming factors in a single virus using 2A “self-cleaving” peptides, using a single polycistronic lentiviral vector coexpressing four transcription factors (Oct 4, Sox2, Klf4, and Myc) to yield iPSC. These iPS cells were characterized by immunofluorescence detection using of stem cell markers (SSEA4, Oct4, and Sox2). The mRNA sequencing was performed and the datasets were analyzed using ingenuity pathways analysis (IPA) software. Results The generated stem cell-like clones expressed the pluripotency markers, SSEA4, Oct4, Sox2, Tra-1-60, and also expressed pax6. Our transcriptome analysis showed 4300 genes, which had 2-fold change and 870 genes with a q-value of <0.05 in keratoconus iPSC compared to normal iPSC. One of the genes that showed difference in KC iPSC was FGFR2 (down-regulated by 2.4 fold), an upstream target of Pi3-Kinase pathway, was further validated in keratoconus corneal sections and also KC iPSC-derived keratocytes (down regulated by 2.0-fold). Both normal and KC-derived keratocytes expressed keratocan, signature marker for keratocytes. KC iPSC-derived keratocytes showed adverse growth and proliferation and was further confirmed by using Ly2924002, a PI3k inhibitor, which severely affected the growth and differentiation in normal iPSC. Conclusions Based on our result, we propose a model for KC in which inhibition FGFR2-Pi3-Kinase pathway affects the AKT phosphorylation, and thus affecting the keratocytes survival signals. This inhibition of the survival signals could be a potential mechanism for the KC-specific decreased cell survival and apoptosis of keratocytes.