Transport and fate of aflibercept in VEGF-A165-challenged retinal endothelial cells.

Retinal vessels are at least in part involved in clearing of Fc terminus-containing proteins from the vitreous. In vitro, the Fc fusion protein aflibercept is transported through a monolayer of unchallenged immortalized bovine retinal endothelial cells (iBREC), mediated by the neonatal Fc receptor (FcRn), but part of the Fc fusion protein is also degraded. Aflibercept's target VEGF-A not only enhances the permeability of REC by destabilization of tight junctions (TJs) thereby allowing for paracellular flow, it may also lower the intracellular stability of the Fc fusion protein by changing its binding properties to the FcRn. Therefore, we investigated the transport and fate of aflibercept in VEGF-A165-challenged iBREC. All cell culture media were supplemented with 5% fetal bovine serum (FBS) as its absence results in accumulation of aflibercept in iBREC due to deregulated expression of transport proteins. Early after exposure of a confluent iBREC monolayer cultivated on gold electrodes to 5% FBS, the cell index (CI) - assessed as a measure of barrier function, cell viability and cell adhesion - transiently declined but recovered again within a few hours to high values. These values remained stable for several days associated with a strong expression of the TJ-protein claudin-1, indicative of a functional barrier formed by the iBREC monolayer. Transient changes of the plasma membrane localizations of claudin-5 and vascular endothelial cadherin - both important for regulation of paracellular flow - accompanied the transient reduction of the CI not prevented by VEGF-binding proteins. Treatment of iBREC with 50 ng/ml VEGF-A165 for one day resulted in a strong and persistent decline of the CI associated with a low expression level of the TJ-protein claudin-1; reversion to normal values was complete one day after aflibercept's addition at a final concentration of 250 μg/ml. Expressions of other proteins involved in regulation of paracellular flow or transcellular transport were not significantly changed. More aflibercept passed through the monolayer of iBREC cultivated on permeable membrane inserts pretreated with VEGF-A for one day, but this was not affected by a FcRn-inhibiting antibody. Subcellular localization of aflibercept was hardly changed in VEGF-A-exposed iBREC 3 h after its addition to the cells; inhibition of (non)-lysosomal or proteasomal proteases then only weakly affected the amount of internalized aflibercept. iBREC also internalized VEGF-A which was barely detectable as early as 2 h after addition of aflibercept. In contrast, blocking the tyrosine kinase activity of VEGF receptor(s) did not prevent VEGF-A's uptake. Inhibition of cellular proteases strongly increased the amount of internalized VEGF-A in the absence and presence of the Fc fusion protein. We therefore conclude that a FcRn-mediated transport plays a minor role in aflibercept's passage through a leaky barrier of REC. Even early after addition of aflibercept to VEGF-A-exposed iBREC, the levels of free intracellular VEGF-A are low, as aflibercept likely prevents binding of VEGF-A to its receptor. Interestingly, the growth factor's detrimental effects still persist for nearly one day.