Flow-induced mechano-modulation of intestinal permeability on chip

Abstract

The comprehension of the working principles behind the intestinal transepithelial transport is critical in nutrient and drug development research. Within this framework, microfluidic microphysiological platforms are on the verge of overshadowing traditional in vitro systems due to their accuracy in replicating key physiological features of the native tissue. Nevertheless, the effects of fluid mechanical stimuli on the selective permeation characteristics of the gut barrier are still unexplored. This is an indispensable feature for designing more biorelevant organ-on-chip models. Here, an intestine-on-chip platform is conceived to mechanically stimulate cells with three different fluid shear stresses and investigate the relative flow-induced changes of molecule transport alongside the resulting epithelial architecture and barrier functionality. Our results reveal that epithelia grown at lower shears exhibit a ∼1.5 higher and faster paracellular permeability while showing a ∼3 times lower and delayed transcellular uptake compared to layers exposed to higher shear stress. This is corroborated by impedance spectroscopy measurements that display altered tight junctional and bilayer resistance, as well as an increased capacitance of the epithelium in response to variations in mechanical stress within the culture. Taken together, these findings advocate that fluid shear stress can serve as mechano-modulator not only for intestinal transport but also for other epithelial cell lines under physiological circumstances.

Keywords: Gut-on-chip; Microfluidics; Permeability; Shear stress; TEER.

Graphical abstract

Fig. 1

The intestine-on-chip platform. a) The microdevice consists of a basal channel and an apical compartment with a customized integrated microporous membrane with specific slots for gold wires and a final top lid closure. b) Real-time image of the intestine-on-chip. c) The experimental setup: the in-vivo like Caco-2 epithelium reconstructing the villus-crypt architecture is stimulated with three different syringe-driven flows. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 2

Intestinal epithelial morphology. a) Planar view of the epithelial surface cytoskeleton at different stresses. Scale bar 40 μm. b) Average normalized F-Actin fluorescence intensity per unit area inside the chip. c) Orthogonal projections of confocal images standing for cell nuclei (blue) and cytoskeleton (orange). Scale bar 30 μm. d) Average epithelial thickness computed from z-stack images. e) Orthogonal projections of confocal images standing for cell nuclei (blue) and mucus (Magenta). Scale bar 30 μm. f) Average amount of mucus covering the epithelial surface. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 3

Relevant intestinal protein markers are sensitive to shear stress variations. a) Tight junction expression exhibits an increasing trend with shear. b) E-Cadherin formation is highly enhanced at high shear conditions. c) Fluid flow generates more structured E-Cadherin but with no sensitivity to different stresses. d) Increasing Shear stresses upregulate the formation of apical microvilli encoded by Ezrin. e) Fluid flow promotes cellular differentiation in absorptive enterocytes witnessed by high abundance of Alkaline Phosphatase. f) The Sucrase Isomaltase, another marker of the epithelial brush border, does not present significant variations with shear. All fold changes are relative to the unstimulated state.

Fig. 4

Modelling shear stress experienced by cells. a) Base-tip shear stress sensed by adherent cells during the seeding stage. b) Base-tip shear stress sensed by cells when they start to grow, divide and polarize. c) Base-tip shear stress sensed by the confluent intestinal epithelium.

Fig. 5

Intestinal biogeography is correlated to local fluid-dynamics. a) Velocity field streamlines show the generation of local turbulences along the height of membrane pores. b) The intestinal epithelium lines the crypt-like membrane experiencing a heterogenous shear stress distribution (in blue cell nuclei, in orange f-actin). Scale bar 20 μm. c) Shear stress magnitude decreases from the top to the bottom of the membrane pore. d) Intestinal markers are confined to specific regions of the pore: enterocytes and enteroendocrine cells populate the villus region, while Paneth and stem cells are localized at the crypts. All scale bars are 40 μm. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 6

Electrical Impedance Spectroscopy. a) The experimental setup utilized for recording spectra (potentiostat, software and microchip) and the relevant representative equivalent circuit following two-path spectroscopy. b) Characteristic Bode plots of the three different configurations. c) Reconstructed paracellular and transcellular resistances behave discordantly with increasing shear. The total epithelial resistance remains constant regardless the stimulus. d) The bilayer cultured at high shear stress is keener to store molecules or ions.

Fig. 7

Transepithelial transport studies. a) Cumulative curves of 4 kDa dextran transported paracellularly across the cultured epithelia. b) Cumulative curves of 70 kDa transported through a combination of paracellular and transcellular trafficking across the epithelia. c) Cumulative curves of 2000 kDa dextran transported transported across the cultured epithelia. d) Lag time measurements reveal a delayed opening of tight junctions with a simultaneous increased uptake by the cellular membrane at higher stress levels. e) Paracellular apparent permeability decreases with increasing shear, while the transcellular counterpart behaves oppositely.