Validation of optical coherence tomography as a tool to identify differentiation key drivers in 3D in vitro conjunctival models

Conjunctival in vitro models present a valuable system to investigate conjunctival tissue homeostasis and pathologies. Combinations of collagen and fibroblasts as a stroma equivalent and the supplementation with serum have been reported to promote the differentiation of epithelial cells. However, how the individual factors affect differentiation of ocular surface cells is insufficiently understood. In this study, we analyzed the effect of serum concentration, a collagen matrix, and fibroblasts on conjunctival differentiation in a 3D in vitro model. For this purpose, we developed a computational analysis pipeline for the quantification of optical coherence tomography (OCT) data sets, allowing a time resolved, non-invasive assessment of conjunctiva epithelium differentiation, including goblet cell density. High-resolution dynamic full-field OCT (D-FFOCT) was employed to verify the identity of goblet cells. Conjunctival markers were further analyzed via histology, real-time quantitative PCR, and ELISA to confirm the data of the OCT analysis pipeline. We found that serum is required to induce epithelial differentiation while higher concentrations of 5 – 10% impaired epithelial development. The culture on a collagen matrix increased conjunctival markers upon stimulation with serum, while the co-culture with fibroblasts increased epithelial stratification. Increased serum concentration resulted in the increased occurrence of goblet cells of up to 20 cells/mm². Altogether, the complementary analyses confirmed the quantified OCT data. Summarized, we identified the combination of serum (3%), collagen, and fibroblasts as a condition resulting in the highest physiological resemblance. Altogether, our study emphasizes the need for fine-tuning of culture conditions for 3D in vitro models.