Abstract
Purpose :
Corneal permeability determines the efficacy of drugs inside the eye. As an alternative to animal testing, we developed and validated a tissue-engineered 3D cornea model (based on human cells) with permeability assays using low and high molecular weight pharmaceuticals.
Methods :
The 3D hemi-cornea model consists of a multilayered epithelium of primary epithelial cells from human donor corneas and a stroma equivalent of 45,000 primary human keratocytes in collagen, embedded in a snapwell insert (0.4 µm pore size) (700 µL type I collagen (10.5 mg/mL), compression to 159 µL∼500 µm thickness). Due to the firm adhesion of the model with the insert wall, the apical and basolateral sides are separated. After 8 days of cultivation at the air-liquid interface, histological examination showed a multilayered, nonkeratinized squamous epithelium on the stroma. The test substances used were FITC (fluorescein isothiocyanate)-dextran (4k Da, 0.3 mg/ml), fluorescein sodium (376 Da, 0.002 %), ofloxacin (3 mg/ml) and amphotericin B (0.5 mg/ml) applied apically to the model. After 24 h, fluorescence and absorption in the apical and basolateral medium were measured using a photometer (TECAN Infnite) and UV/VIS-spectrometer (FLUOstarOmega), respectively. The diffusion was calculated as the ratio of the absorption of the basolateral medium to the absorption of the apical medium. To examine the impact of the epithelium as a diffusion barrier, it was removed with EDTA.
Results :
In the 3D hemi-cornea model, a statistically significant lower permeability was observed for the high-molecular weight FITC-dextran compared to fluorescein sodium. With intact epithelium, a diffusion of 0.97 % ± 0.05 % was found for FITC-dextran and 7.57 % ± 2.06 % for fluorescein sodium. After the removal of epithelial barrier by EDTA, FITC-dextran showed an approximately 4.8-fold (4.69 % ± 1.38 %) and fluorescein sodium a 2.5-fold higher diffusion (18.58 % ± 4.11 %) compared to intact epithelium. While amphotericin B could not be detected in the basolateral medium, ofloxacin showed a diffusion of 86 % with intact and 88.7 % with removed epithelial barrier.
Conclusions :
Our 3D hemi-cornea model represents a promising in vitro system for measuring permeability. The low standard deviation shows a high reproducibility of the data and facilitates testing different formulations of drugs, reducing the need for animal or human tissue.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.