Tissue Engineered Human Blood-Retinal Barrier-on-a-Chip

Abstract

Purpose : Recent advancements in the field of microfluidics have provided the means to observe tissue and organ-level behaviors in a controlled biomimetic microenvironment. In this study, we recapitulated the human blood-retinal barrier (BRB) in a microfluidic platform using induced pluripotent stem cell (iPSC)-derived retinal pigment epithelium (RPE) and endothelial cells (ECs) and compared the functionality of our model to that of a standard Transwell co-culture system.

Methods : The poly(dimethylsiloxame) (PDMS) based microfluidic chip used in the experiment was composed of two vertically-stacked microchannels separated by a porous polyethylene terephthalate membrane. Developmentally guided differentiation protocols were used to generate RPE and ECs from iPSCs. The following assays were performed to characterize the functionality of the co-culture system in the chip: permeability to FITC-dextran, vascular endothelial growth factor (VEGF) and pigment epithelium-derived factor (PEDF) secretion, RPE pigmentation, and fluid transport across top and bottom channels.

Results : Donor-matched sets of iPSC-derived RPE cells and ECs were successfully generated and co-cultured in our microfluidic device. The top channel was seeded with RPE cells, and the bottom channel was seeded with ECs. The cells were maintained using tissue-specific culture medium continuously perfused by a peristaltic pump, and different flow rates were given to each cell type to more closely mimic the microenvironment of the BRB. The RPE and EC monolayers demonstrated barrier resistance, secreted cytokines in a polarized fashion, and transported water from top channel towards the bottom channel.

Conclusions : Using microfluidic channels to co-culture RPE cells and ECs, we showed that our chip platform is able to model the human BRB as well as a standard Transwell culturing system. In addition, the unique functions offered by the chip, specifically its ability to quantify in-situ the rate of fluid transport across the BRB model, helps pave the way for developing a model of different ocular diseases on the chip, which can serve as an efficient drug discovery platform in the future.

This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.