Epithelial-mesenchymal interactions in the cornea: development of a novel 3D culture cornea model and progress towards environmental reprogramming of cornea epithelium

Abstract

The unmet need for corneal epithelial cells for treating human eye diseases makes the cornea important in the cell therapy field. Additionally cornea tissue engineering has become valuable for clinical use, research, and for creating representative models replacing animals for chemical/drug testing. This study initially used qPCR to investigate the expression levels of key markers produced by 2D corneal epithelial cell cultures after wounding in a scratch assay. Then an attempt was made to environmentally reprogram human hair follicle keratinocytes into corneal epithelial cells using limbal epithelial stem cell media. Immunohistochemical and qPCR analysis revealed no changes in signature genes but rather a similarity between HFC and LSC’s when cultured in LSC’s culture conditions. Attention then focused on developing a novel three-dimensional bilayered spheroid cornea model using hanging drop culture. It is widely accepted that cells in 3D culture more closely mimic their in vivo counterparts than 2D cultures, and qPCR and immunofluorescence analysis of 3D spheroids made from cultured rabbit corneal stromal cells revealed that they partially reverted back to a quiescent in vivo phenotype. Coating the spheroids with cultured rabbit limbal epithelial cells produced a bilayered model of the cornea. Multiple iterations were produced incorporating variations in media and cell origin, leading to a cornea model that could be maintained for 10 days, expressed appropriate cytokeratins and other corneal markers including Pax6 and that, upregulated the expression of key cornea signature proteins including Aldh1a1 as a result of epithelial-mesenchymal interactions. Preliminary versions of a human bilayered cornea model were then created from equivalent human cell types. Generally this 3D model displays advantages over other in vitro cornea equivalents and has potential, but needs further refinement. The methodology was also used to coat stromal spheroids with skin keratinocytes, highlighting the possibility of reprogramming the former into corneal epithelial cells through epithelial-mesenchymal interactions.