Related Experiment Video
Updated: Sep 5, 2026

An Acute Retinal Model for Evaluating Blood Retinal Barrier Breach and Potential Drugs for Treatment
Published on: September 13, 2016
A Biomimetic Eye In Vitro Model to Investigate the Blood-Retinal Barrier Pathophysiology
Alessio Esposito1,2, Gabriele Maria Fortunato1,2,3, Mauro Di Stasi1,4
1Research Center "E. Piaggio", University of Pisa, Largo Lucio Lazzarino 1, Pisa55122, Italy.
Abstract:
Existing in vivo models for retinal pathologies are costly, time-consuming, and subject to regulatory challenges, promoting interest in advanced in vitro models that replicate the blood-retinal barrier. Challenges related to vascularization and structural multiscale interactions hinder the development of in vitro models, which are useful to study nutrient and drug exchange between the retina and systemic circulation. To address these issues, we developed a blood-retinal barrier in vitro model for studying retinal pathophysiology. It features an electrospun scaffold (Bruch's membrane) placed between a culture chamber (vitreous humor) and a bioinspired microfluidic network (choroidal vasculature). The network, modelled from indocyanine green angiography of the human retina, is laser-engraved into a 15 mm PDMS disc with channels 70-800 μm wide. PLGA and gelatin electrospun scaffold were investigated to mimic Bruch's membrane. The scaffold is fixed to the microfluidic network, assembled into a culture chamber, and integrated within a bioreactor. A peristaltic pump ensures a consistent flow in the channels. Biological validation included ARPE-19 cells cultured on Transwell, PLGA, or gelatin membranes, HUVECs within membrane-coupled microchannels, and dual-layer constructs. Cells were examined separately and together, allowing evaluation of epithelial, endothelial, and construct responses. All configurations showed progressive increases in metabolic activity and limited LDH release over 28 days. Cell-seeded constructs reduced FITC-dextran transport relative to acellular controls, with stable values between days 21 and 28. Blank-corrected TEER likewise increased during ARPE-19 culture and plateaued by day 21. PLGA constructs showed the greatest electrical and solute-transport resistance, although acellular controls confirmed a substantial material contribution. Confocal microscopy demonstrated HUVEC attachment and F-actin organization along the microchannels, while ARPE-19 monolayers exhibited continuous junction-associated ZO-1 staining that remained stable between days 21 and 28. Overall, the platform supported long-term epithelial-endothelial culture and provided a stable, perfusable system for studying oBRB transport and cellular interactions.

