Direct On-Chip Differentiation of Intestinal Tubules from Induced Pluripotent Stem Cells

Abstract

Intestinal organoids have emerged as the new paradigm for modelling the healthy and diseased intestine with patient-relevant properties. In this study, we show directed differentiation of induced pluripotent stem cells towards intestinal-like phenotype within a microfluidic device. iPSCs are cultured against a gel in microfluidic chips of the OrganoPlate, in which they undergo stepwise differentiation. Cells form a tubular structure, lose their stem cell markers and start expressing mature intestinal markers, including markers for Paneth cells, enterocytes and neuroendocrine cells. Tubes develop barrier properties as confirmed by transepithelial electrical resistance (TEER). Lastly, we show that tubules respond to pro-inflammatory cytokine triggers. The whole procedure for differentiation lasts 14 days, making it an efficient process to make patient-specific organoid tubules. We anticipate the usage of the platform for disease modelling and drug candidate screening.

Keywords: 3D cell culture; directed differentiation; gut-on-a-chip; iPSC; intestinal inflammation; intestinal organoids; microfluidics; organ-on-a-chip.

Figure 1

Seeding and differentiation of hiPSC in the 3-lane OrganoPlate. (A) Artist’s impression of the bottom-side of the 3-lane OrganoPlate containing 40 individual microfluidic chips. The inlay shows an individual chip that is communicating with nine connecting wells. (B) Scheme of a single chip containing three microfluidic channels all having top, middle and bottom inlets (A1, B1 and C1), outlets (A3, B3 and C3) and an observation window (B2). (C) A schematic representation of hiPSC introduction to a microfluidic chip in a top (xz plane) and cross sectional view (yz plane); single iPSCs (in red) are seeded in the top channel and are allowed to adhere for up to 24 h after which media is added and perfusion initiated, cells start to form tubules by first growing against the ECM gel (blue) and then covering the whole channel. (D) Schematic of workflow for directed on plate differentiation of iPSC towards gut tubules.

Figure 2

Characterization of hiPSC-derived human intestinal-like tubules. (A) 3D reconstruction image of an iPSC-derived tubule at Day 4 stained with antibodies for SOX17 (green), FOXA2 (red) or DAPI for DNA (blue). Cells attach and form tubule at the DE stage. (B) Representative images of iPSC tubules stained for Definitive endoderm markers SOX17 and FOXA2 at day 4. (C) Hindgut marker CDX2 at day 7. (D) Gene expression was measured using TaqMan qRT-PCR from hiPSC tubules at different differentiation stages. The following genes were analysed: POU class 5 homeobox 1 (POU5F1, indicating pluripotency); Nanog homeobox (NANOG, indicating Primitive Streak):, forkhead box a2 (FOXA2, Definitive Endoderm), SRY (sex determining region Y)-box 17 (SOX17, Definitive Endoderm), and Homeobox protein CDX-2 (Posterior gut). (E) Representative images of iPSC tubules stained for intestinal markers Lysozyme (LYZ), Villin (VIL) and Chromogranin A (CHGA) at day 28 (green). Nuclei were stained with DAPI (blue) to visualize the overall morphology. Scale bars = 100 µm. (F) Relative gene expression analysis of Intestinal markers: Leucine-rich repeat-containing G-protein coupled receptor 5 (LGR5), Mucin-2 (MUC2), Lysozyme (LYZ), Villin-1 (VIL1), Chromagranin A (CHGA) and sucrase-isomaltase (SI). (G) Relative mRNA expression of MDR1 (P-gp) and CYP3A4 in iPSC-derived intestine-like tubules on day 14, day 28 and day 31, human adult colon organoids and Caco-2 cells. Expression levels were normalized to beta-actin (ACTB), data represented as mean ± SD relative to the expression in undifferentiated miFF1 hiPSC (n = 2, n ≥ 2–3). The Y-axis represents the LOG10 relative quantification (RQ). Caco-2 cells and primary colon organoid were also included to compare gene expression and to follow the differentiation of our model during the different stages. Data are presented as the average of two independent experiments +/− SD (n = 3).

Figure 3

Barrier function of iPSC-derived intestinal-like tubules. (A) Merge of fluorescent and phase contrast images of four microfluidic chips perfused in the top channel with fluorescent molecules (in blue) using 4.4 kDa TRITC-Dextran for 15 min on day 25 of culture. The ‘no cell control’ shows the passage of the fluorescent dye into the adjacent gel and secondary perfusion channel, whereas in the chips with iPSC tubules the fluorescent dye is retained by the tubule. (B) P-app value calculated based on a fluorescent permeability assay using both 4.4 kDa TRITC-dextran and 150 kDa FITC-dextran at day 7 and day 14. Data are represented as mean ± SD. Significance was detected by two-way Anova, n ≥ 12. (C) Transepithelial electrical resistance (TEER) measurements of iPSC-derived intestine-like tubules between day 4 and day 14 of culture (N = 3, n ≥ 110). (D) Percentage of iPSC-derived intestine-like tubules between day 4 and day 14 of culture developing TEER values above 15 Ω cm². Significance was detected by ordinary one-way Anova with Dunett’s multiple comparison test. Data are represented as mean ± SD. (n = 3, n ≥ 110), ns p > 0.05, * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001.

Figure 4

hiPSC-derived gut tubules respond to inflammatory triggers. (A) Gene expression analysis of hiPSC-derived gut tubules for CCL20, IL-8, and IL-6 upon stimulation with cytokine cocktail of TNF-α, IL-1B and IFN-γ. Expression levels are measured upon 24 h, 48 h and 72 h of exposure to the cytokine cocktail. Data are represented as mean ± SD normalized to ACTB expression (n = 2–3). (B) Secretion of IL-8 and (C) secretion of IL-6 by triggered hiPSC-derived gut tubules (blue) and Caco-2 tubules (gray) upon stimulation with the cytokine cocktail. Significance was determined with Multiple t-tests (one per row) and discovery determined using the Two-stage linear step-up procedure of Benjamini, Krieger and Yekutieli, with Q = 1%. Each row was analyzed individually, without assuming a consistent SD. Data are represented as mean ± SD. (n ≥ 5) ns p > 0.05, * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001.