Intestinal Cells-on-Chip for Permeability Studies

Abstract

Background: To accurately measure permeability of compounds in the intestine, there is a need for preclinical in vitro models that accurately represent the specificity, integrity and complexity of the human small intestinal barrier. Intestine-on-chip systems hold considerable promise as testing platforms, but several characteristics still require optimization and further development.

Methods: An established intestine-on-chip model for tissue explants was adopted for intestinal cell monolayer culture. A 3D-printed culture disc was designed to allow cell culture in static conditions and subsequent permeability studies in a dynamic environment. Membrane characteristics and standardized read-outs were investigated and compared to traditional permeability studies under static conditions.

Results: By starting cultures outside the chip in conventional wells plates, the new cell disc design could support accurate cell monolayer formation for both Caco-2 and human enteroids. When transferred to the chip with laminar flow, there was accurate detection of barrier integrity (FD4 and Cascade Blue) and permeability (atenolol/antipyrine). Both flow and membrane characteristics had a significant impact on permeability outcomes.

Conclusions: This novel intestinal cell-on-chip system offers large flexibility for intestinal permeability studies, although it still requires validation with more compounds to reveal its full potential.

Keywords: cell monolayer; in vitro model; intestinal absorption; intestinal barrier; intestine-on-chip; permeability.

Figure 1

Schematic overview of the cell-on-chip system. (A) In-house developed 3D-printed disc with porous membrane and rubber ring for cell monolayer culture. Scale bar equals 2 mm. (B) Disc in a carrier for cell seeding and monolayer formation (static culture). (C) The disc with a cell monolayer can be transferred to a chip with dual flow. (D) Complete cell-on-chip system set-up, with eight chips connected to separate apical and basolateral medium reservoirs, with a peristaltic pump to create laminar flow, placed inside an incubator.

Figure 2

Intestinal cell monolayer culture on disc. (A) Brightfield images of epithelial cell monolayer cultures on a disc, derived from Caco-2 and enteroid cells. (B) Growth area coverage and monolayer viability indicated by MTT staining for Caco-2 and enteroid monolayers, compared to a disc with no cells. (C) Cell monolayer morphology assessed by whole-mount immunofluorescent detection of dapi (blue) and actin (orange). (D) HE staining of a cross-section of Caco-2 and enteroid monolayers, visualizing a single epithelial cell layer on the membrane. Scale bar equals 200 µm for (A) and 50 µm for (C,D).

Figure 3

Epithelial barrier integrity. (A) Apparent permeability values (P_app) for dextran-FITC 4 kDa (FD4) in cells-on-chip containing transparent Transwell membranes, determined between 1 and 2 h of perfusion and between 2 and 4 h of perfusion. Each dot represents a single chip. (B) P_app values for FD4 in cells-on-chip with translucent (dashed) and transparent (open) membranes, between 2 and 4 h of perfusion. (C) Cascade Blue 0.5 kDa (CB) permeability in cells-on-chip with translucent and transparent membranes, between 2 and 4 h of perfusion. n = 3–4 per group, with mean and SEM plotted.

Figure 4

Epithelial barrier permeability. (A) Schematic overview of the transport routes for antipyrine (transcellular) and atenolol (paracellular) over the intestinal epithelial barrier. P_app values for (B) antipyrine and (C) atenolol in intestinal cell monolayers on chip, measured between 2 and 4 h of perfusion, for both Caco-2 and enteroid monolayers compared to no-cell control chip, with (D) corresponding antipyrine/atenolol ratios. Cut-off value for a functional intestinal epithelial barrier, i.e., an A/A ratio > 2, is indicated by a dashed line. For all graphs, the dashed bars represent monolayers on translucent membranes, and the open bars represent transparent membranes. n = 3–4 per group, with mean +/− SEM displayed.

Figure 5

Active drug transport and metabolism in enteroid monolayers. (A) Schematic overview of the main routes for carrier-mediated active transport of rosuvastatin (BCRP) and metformin (OCT1) in the intestinal epithelial barrier, and phase I metabolism of midazolam (CYP3A4) and phase II metabolism of 7OH-coumarin (UGTs) by the epithelial cells. (B–D) P_app values for rosuvastatin and metformin in enteroid monolayer cultures, determined between 1 and 2 h of incubation, including inhibitors to block active transport, in (B) static transparent Transwell inserts, (C) chip system with translucent membranes and (D) chip system with transparent membranes. (E) OH-midazolam formation over time in enteroid monolayers in static Transwell cultures. (F,G) 7OH-coumarin-glu formation over time in enteroid monolayers in (F) static Transwell inserters and (G) chip system. The drug cocktail included metformin (50 µM), rosuvastatin (10 µM), midazolam (10 µM) and 7-hydroxycoumarin (12 µM). The inhibitors added were elacridar (5 µM) and ketoconazole (5 µM). n = 4–6 per group, with mean +/− SEM displayed.