Efflux Pump Substrates Shuttled to Cytosolic or Vesicular Compartments Exhibit Different Permeability in a Quantitative Human Blood–Brain Barrier Model

Abstract

Representative in vitro blood-brain barrier (BBB) models can support the development of strategies to efficiently deliver therapeutic drugs to the brain by aiding the characterization of their internalization, trafficking, and subsequent transport across the BBB. A collagen type I (COL1) hydrogel-based in vitro BBB model was developed to enable the simultaneous characterization of drug transport and intracellular processing using confocal microscopy, in a way that traditional insert-based in vitro BBB models cannot. Human induced pluripotent stem cells (hiPSCs) were differentiated into cells that exhibited a BBB-like phenotype on COL1 hydrogels, which included the expression of key BBB-specific proteins and low permeability of representative small and large molecule therapeutics. Furthermore, the BBB phenotype observed on the COL1 hydrogel was similar to that previously reported on porous inserts. The intracellular visualization of two small molecule efflux pump substrates within the hiPSC-derived BBB-like cells demonstrated a difference in cytosolic and vesicular accumulation, which complemented permeability measurements demonstrating a difference in transport rate. The easy-to-construct COL1-based hiPSC-derived BBB model presented here is the first in vitro two-dimensional BBB experimental system that enables the simultaneous quantification of cellular permeability and visualization of intracellular processes by utilizing confocal microscopy, which can provide insights regarding the relationship between transport and intracellular trafficking of therapeutic drugs.

Antibody transcytosis across brain endothelial-like cells occurs nonspecifically and independent of FcRn

“…Accordingly, the decreasing serum concentration of exogenous IgG, which is exacerbated in FcRn knockout models or when the IgG lacks FcRn binding, and the presence of endogenous IgG represent experimental hurdles to the assessment of IgG transcytosis across BECs in vivo.Here we investigated the role of FcRn or other IgG-specific receptors in influencing the transcytosis of IgG across BECs in vitro using BEC-like cells derived from human induced pluripotent stem cells (iBECs). We have previously shown that iBECs exhibit permeability of IgG that is more restrictive than that observed for the rat BBB in vivo [24][25][26] , which supports its use as a representative BBB model. To compare the influence of FcRn engagement we assessed the intracellular processing and transcytosis of IgGs lacking human FcRn recognition, and we used human IgG as a control, at varying concentrations and in the presence of excess human IgG.…”

“…To assess the influence of FcRn-mediated processing, we used a comparative IgG-based approach, similar to a previous in vivo report 17 , which exploits the well-characterized stringency of human FcRn to only bind a unique sequence present on the Fc domain of some IgGs 19,20 . Visualization and quantification of transcytosis within the same timeframe was achieved by using our previously developed in vitro BBB model that consists of a monolayer of iBECs on a COL1-based hydrogel 25 . We observed that despite significant differences in lysosomal accumulation that was consistent with preferential FcRn-mediated recycling 19 , all tested IgGs exhibited comparable iBEC permeability.…”

“…COL1 from rat tendon (Corning) was prepared as a hydrogel on 8-chambered glass coverslips based on methods previously described 25 . IMR90-4 (WiCell) human induced pluripotent stem cells were maintained and differentiated to BECs as previously described 25,26,50 . After 8 days of differentiation, cells were dissociated with STEMPRO Accutase (Life Technologies) and suspended as single-cells.…”

“…As an internal control, each solution also contained 2 µM sodium fluorescein, except for 155-kDa dextran which used a Alexa Fluor 647-labeled human IgG as an internal control. Datasets with internal controls that exhibited permeability values 2-fold higher than that previously reported were not analyzed because they were indicative of a non-confluent iBEC monolayer 25 . After aspiration of the culture medium, the analyte-containing solutions were added to each chamber of the 8-chambered glass coverslip.…”

“…Imaging was performed with a Zeiss 710 confocal microscope with a C-Apochromat 40× water immersion objective (NA 1.2). Time-lapse z-stacks that contained both the luminal and abluminal (hydrogel) compartments were captured every 15 minutes using image acquisition parameters previously described 25 . Briefly, time-lapse z-stacks were processed in ImageJ as previously described and the average intensity of the fluorescent analyte within the hydrogel was measured at each time point 25 .…”