Abstract
Purpose :
Reconstructed corneal in vitro models are employed to identify hazardous chemicals for eye irritation, replacing animal experiments and could help to reduce the need for corneal donor tissue for transplantation. In this study we tested the hypothesis, whether primary skin cells can be transdifferentiated with retinoic acid to corneal cells in reconstructed models to enable an autologous way to generate tissue engineered corneal epithelium.
Methods :
Human epidermal keratinocytes were isolated from foreskin biopsies in accordance and with the approval of the local ethics committee and the informed consent of the patients or their guardians. Human epidermal keratinocytes were seeded on cell culture inserts with a polycarbonate membrane at a cell density of 5*105 cells/cm2 in EpiLife supplemented with human keratinocyte growth supplement and 1.5 mM CaCl2. Medium was changed after 24 hours to EpiLife air-liquid-interface medium, containing 73 µg/ml L-ascrobic acid 2-phosphate and 10 ng/ml keratinocyte growth factor. Retinoic acid was added to the medium in different concentrations 10-5-10-8 M at indicated time points. For analysis, tissue slides were stained with hematoxylin and eosin or immunohistochemically for keratin 3/12, keratin 1, keratin 14, involucrin and loricrin. Additionally, impedance spectroscopy was performed to evaluate the barrier function by transepithelial electrical resistance.
Results :
Retinoic acid suppressed the epidermal cornification even at low concentrations of 10-7 M. Higher concentrations of retinoic acid or added over a longer period of time also inhibited the formation of stratified layers. 4 days after adding retinoic acid, a decrease in barrier function can be observed by impedance spectroscopy. The addition of 10-5 M retinoic acid after 5 days of culture revealed the reduction of epidermis-specific keratin 1 and loricrin. Cornea specific keratin 3/12, keratin 14 and involucrin did not change.
Conclusions :
The addition of retinoic acid to RHE reduces cornification and keratin 1, showing first signs of transdifferentiation to a corneal epithelium. However, the induction of the cornea specific keratin 3/12 was still missing, indicating that a further stimulus is needed to complete the transition of RHE to corneal epithelium. Nonetheless, retinoic acid can be employed to reduce cornification and skin specific markers in RHE.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.