In vitro models that can faithfully replicate critical aspects of kidney tubule function such as directional drug transport are in high demand in pharmacology and toxicology. Accordingly, development and validation of new models is underway. The objective of this study was to characterize physiologic and transport functions of various sources of human renal proximal tubule epithelial cells (RPTECs). We tested telomerase reverse transcriptase 1 (TERT1)-immortalized RPTECs, including organic anion transporter 1 (OAT1)-, organic cation transporter 2 (OCT2)-, or OAT3-overexpressing variants and primary RPTECs. Cells were cultured on transwell membranes in static (24-well transwells) and fluidic (transwells in PhysioMimix T12 organ-on-chip with 2 μL/s flow) conditions. Barrier formation, transport, and gene expression were evaluated. We show that 2 commercially available primary RPTECs were not suitable for studies of directional transport on transwells because they formed a substandard barrier even though they exhibited higher expression of transporters, especially under flow. TERT1-parent, -OAT1, and -OAT3 cells formed robust barriers but were unaffected by flow. TERT1-OAT1 cells exhibited inhibitable para-aminohippurate transport that was enhanced by flow. However, efficient tenofovir secretion and perfluorooctanoic acid reabsorption by TERT1-OAT1 cells were not modulated by flow. Gene expression showed that TERT1 and TERT1-OAT1 cells were more correlated with human kidney than other cell lines but that flow did not have noticeable effects. Overall, our data show that addition of flow to in vitro studies of the renal proximal tubule may afford benefits in some aspects of modeling kidney function but that careful consideration of the impact such adaptations would have on the cost and throughput of the experiments is needed. SIGNIFICANCE STATEMENT: The topic of reproducibility and robustness of complex microphysiological systems is looming large in the field of biomedical research
therefore, uptake of these new models by the end-users is slow. This study systematically compared various renal proximal tubule epithelial cell sources and experimental conditions, aiming to identify the level of model complexity needed for testing renal tubule transport. We demonstrate that although tissue chips may afford some benefits, their throughput and complexity need careful consideration in each context of use.