Supplementary MaterialsAdditional file 1: S1. limbal stem cell deficiency (LSCD) for their ocular reconstruction capability. As the most important component of the limbal microenvironment, limbal niche cells (LNCs) play a key role in the direction of stem cell differentiation. In this study, we investigated whether LNCs can induce the transdifferentiation of rat OMECs to corneal epithelial-like cells. Methods We isolated OMECs and LNCs from rats by dispase and collagenase, respectively, to establish a three-dimensional or Transwell coculturing system. NIH-3T3 cells and renewed LNCs were also used as feeder layers in the Transwell system to compare their ability to support the OMECs. The airlift method was utilized for the culture of OMECs to obtain a stratified epithelial sheet. Cocultured OMECs SCH 54292 inhibitor were characterized by reverse-transcription polymerase chain reaction, Western blotting, hematoxylin and eosin staining, and immunohistochemistry. Results The cocultured OMECs showed corneal epithelial-like morphology and expressed the corneal epithelial markers CK12 and Pax6 in most cocultured systems. Furthermore, we found that the expression level of CK12, Pax6, and proliferation marker Ki67 was upregulated when compared with that of other groups by renewing the LNCs in the Transwell system (test if test was used to compare the positive cell rate. and Vim+?cells (Fig.?3a). Double immunofluorescence of Vim and CK12, Np63 or Pax6 in P3 ME-LNCs and DF-LNCs was also assessed to confirm that purified LNCs were obtained from rats. Both P3 ME-LNCs and DF-LNCs were CK12C, Np63C, Pax6C, Vim+, N-cadherin+, Oct4+, and Sox2+, indicating that they had been purified and represented the phenotype of limbal niche cells (Fig. ?(Fig.3b).3b). RT-PCR and Western blot were performed to compare the expression levels of Oct4 and Sox2 between ME-LNCs and DF-LNCs. The expression levels of Oct4 and Sox2 in ME-LNCs were significantly higher than that in DF-LNCs. The relative mRNA level of Oct4 was 1.363??0.054-fold for ME-LNCs compared with DF-LNCs (in DF was even higher than that in ME (cultured in either MESCM or DMEM/F12 supplemented with 10% fetal bovine serum. As a result, LNCs did not interfere with the results of further coculture. The results of RT-PCR and Western blotting regarding Oct4 and Sox2 expression in LNCs indicated that using MESCM for culturing rather than DMEM/F12 supplemented with 10% fetal bovine serum could produce LNCs that expressed more mesenchymal stem cell markers, as previously reported [20]. Three-dimensional cocultured OMECs and LNCs produced spheres owing to the 3D Matrigel [37], and other studies have confirmed that LNCs have the ability to appeal to and SCH 54292 inhibitor aggregate the epithelium [20, 21]. Results of 3D coculturing exhibited that use of SHEM and DF-LNCs could upregulate the expression of CK12 and Pax6, indicating that they are better for transdifferentiation of OMECs to corneal epithelial-like cells. However, MESCM Rabbit Polyclonal to MMP-19 is not suitable for transdifferentiation. We consider these results to be due to the ability of MESCM to maintain the phenotype of stem cells and prevent their differentiation [20, 21, 25, 38]. Furthermore, we also exhibited that maintaining the phenotype of LNCs does not benefit transdifferentiation. We consequently attempted to coculture OMECs and LNCs in the Transwell system to obtain a transplantable epithelium sheet. MESCM failed to support the growth of OMECs in the early period of the study, forcing us to give up this medium SCH 54292 inhibitor in the Transwell system. When we compared the transdifferentiation effect of ME-LNCs and DF-LNCs in the Transwell system, we observed results much like those obtained with the 3D coculturing system, showing that DF-LNCs were more effective than ME-LNCs. Immunofluorescence assay of cultured OMECs, ME, DF, and LEPCs confirmed that CK3 cannot be defined as a cornea-specific marker, whereas higher expression levels of CK12 and Pax6 confirmed the transdifferentiation of OMECs into corneal.