In vitro platform to model the function of ionocytes in the human airway epithelium

Abstract

Background: Pulmonary ionocytes have been identified in the airway epithelium as a small population of ion transporting cells expressing high levels of CFTR (cystic fibrosis transmembrane conductance regulator), the gene mutated in cystic fibrosis. By providing an infinite source of airway epithelial cells (AECs), the use of human induced pluripotent stem cells (hiPSCs) could overcome some challenges of studying ionocytes. However, the production of AEC epithelia containing ionocytes from hiPSCs has proven difficult. Here, we present a platform to produce hiPSC-derived AECs (hiPSC-AECs) including ionocytes and investigate their role in the airway epithelium.

Methods: hiPSCs were differentiated into lung progenitors, which were expanded as 3D organoids and matured by air-liquid interface culture as polarised hiPSC-AEC epithelia. Using CRISPR/Cas9 technology, we generated a hiPSCs knockout (KO) for FOXI1, a transcription factor that is essential for ionocyte specification. Differences between FOXI1 KO hiPSC-AECs and their wild-type (WT) isogenic controls were investigated by assessing gene and protein expression, epithelial composition, cilia coverage and motility, pH and transepithelial barrier properties.

Results: Mature hiPSC-AEC epithelia contained basal cells, secretory cells, ciliated cells with motile cilia, pulmonary neuroendocrine cells (PNECs) and ionocytes. There was no difference between FOXI1 WT and KO hiPSCs in terms of their capacity to differentiate into airway progenitors. However, FOXI1 KO led to mature hiPSC-AEC epithelia without ionocytes with reduced capacity to produce ciliated cells.

Conclusion: Our results suggest that ionocytes could have role beyond transepithelial ion transport by regulating epithelial properties and homeostasis in the airway epithelium.

Keywords: Airway epithelium; FOXI1; Human induced pluripotent stem cells; Ionocytes; Tissue modelling.

Fig. 1

hiPSCs differentiate into lung progenitors in 16 days. A: Diagram of the differentiation protocol, fluorescence activated cell sorting, expansion in 3D organoids and maturation at an ALI to form hiPSC-AECs. Abbreviations: AFE, anterior foregut endoderm; BMP4, bone morphogenetic protein 4; CHIR, CHIR-99021; DAPT, (2 S)-N-[2(3,5-Difluorophenyl)acetyl]-L-alanyl-2-phenyl-glycine 1,1-Dimethylethyl ester; DE, definitive endoderm; FGF7/10, fibroblast growth factor 7/10; IBMX, 3-isobutyl-1-methylxanthine; LP, lung progenitor; PALI, PneumaCult™-ALI Medium; RA, retinoic acid; SB, SB431542; Y, Y-27632. B: Relative mRNA expression of key markers at different time points during differentiation. The control (CTL) is human trachea total mRNA. Filled circles represent individual values and columns are means ± SD (n = 4 independent experiments). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs. D0; one-way ANOVA with Tukey’s post-test. C: Brightfield image of cells after 16 days of differentiation. The scale bar is 1000 μm. D: Immunofluorescence staining showing lung progenitor marker NKX2.1 expression (red) and nuclear marker DAPI (blue) on day 16 of differentiation. The scale bar is 100 μm. E: Flow cytometry panel showing levels of expression of CPM at day 16 of differentiation (red). The population labelled CPM + was sorted for enrichment of NKX2.1-expressing lung progenitors. HiPSCs stained with anti-CPM antibody served as a negative control (blue). F: Enrichment of NKX2.1 mRNA expression after sorting of CPM + cells. Filled circles represent individual values and columns are means ± SD (n = 5 independent experiments). **P < 0.01 vs. pre-sorted; Student’s t-test

Fig. 2

ALI culture induces differentiation towards mature AECs with similar properties to HBECs. A: Brightfield image of lung progenitors in 3D organoid culture. The scale bar is 250 μm. B: Schematic of ALI culture. Organoids were dissociated and cells seeded in Transwell® inserts and cultured with medium on both sides. Once cells were confluent, medium from the top compartment was removed to form an ALI with the apical membrane of cells in contact with air. DAPT was added to the maturation medium in the bottom compartment and the cells cultured for a further 14 days, followed by another 14 days of culture with PALI medium. C: Brightfield image of hiPSC-AECs in a Transwell® insert after establishing an ALI. The scale bar is 500 μm. D: mRNA expression of AEC markers in cells in expansion conditions (D0) and in ALI culture (D28). ALI cultured HBECs were used as a control. Filled circles represent individual values and columns are means ± SD (n = 3 independent experiments); *P < 0.05 vs. D0; one-way ANOVA with Tukey’s post-test. E: Immunocytochemical analysis of mature AEC markers in ALI cultures. The scale bars are 50–100 μm as indicated. F: Immunofluorescence staining of a histological section through an ALI culture showing polarization of the airway epithelium. Cilia at the apical side are labelled with Acetylated tubulin (AcTub) and mature basal cells on the basal side are labelled with CK5. G: R_t measurements of polarized epithelia formed by hiPSC-AECs. Filled circles represent the average of three readings of the same sample and columns are means ± SD (n = 3 independent experiments). H: Ciliary beat frequency (CBF) measurements in hiPSC-AECs and HBECs. Filled circles represent the average of 20 FOVs from one sample and columns are means ± SD (n = 3 independent experiments); Student’s t-test. I: Area covered by cilia in hiPSC-AECs ALI cultures is significantly smaller compared to HBECs. Filled circles represent the average of 20 FOVs from one sample and columns are means ± SD (n = 3 independent experiments); **P < 0.01; Student’s t-test

Fig. 3

hiPSC-derived lung progenitors engraft in an in vivo mouse model of airway injury. A: Schematic of the cell transplantation procedure. Mice were anesthetized and 30 µl of 2% Polydocanol was administered oropharyngeally. After 18 h, mice were anesthetised again and 30 µl of sterile PBS with 1% BSA and 1 million GFP + hiPSC-derived lung progenitors were administered to the back of the throat. Tracheas were harvested at different time points for analysis. B: Representative wholemount immunofluorescence staining showing GFP and DAPI at 7 days after transplantation (left) and GFP, human CK5 and DAPI at 10 days after transplantation (right). The scale bars are 50 μm

Fig. 4

FOXI1 knock-out (KO) in hiPSCs leads to hiPSC-AEC cultures lacking ionocytes. A: FOXI1 gene targeting and differentiation strategy. Following FOXI1 KO in hiPSCs, WT and KO clones were selected from the targeted pool and differentiated in parallel towards AECs. FOXI1 KO cells were not expected to generate ionocytes. Targeted but unedited WT cells served as an isogenic control. B: Each hiPSC line was targeted with a different sgRNA (Strategy #1 for FS13B and Strategy #2 for CF17/NKX2.1-GFP), testing two different genetic backgrounds and two different targeting strategies in the same study. The diagram shows the sgRNA used on each line to target the DNA binding domain in Exon 1, the PAM region highlighted in blue and the indels in the selected KO clones in red. C: Relative mRNA expression of key markers at different time points during the first stages of differentiation. Filled circles represent individual data points and bars are means ± SD (n = 3 independent experiments); two-way ANOVA with Sidak multiple comparison test. The dotted line indicates the level of the normalized reference-gene expression average value. D: Relative mRNA expression of key mature AEC markers of cells in expansion (D0) and after maturation in ALI cultures (D28). Filled circles represent individual data points and bars are means ± SD (n = 3 independent experiments), two-way ANOVA with Sidak multiple comparison test. E: Representative immunocytochemical staining of FOXI1, CFTR and DAPI in mature hiPSC-AECs after 28 days in ALI culture in FOXI1 WT and KO cells. The scale bar is 20 μm. F: Z-stack panel with orthogonal views of FOXI1 WT cells from E. G: Cropped representative Western blot images of FOXI1 expression in WT and KO hiPSC-AECs (right panel), undifferentiated hiPSCs were used as a negative control and MCF7 cells were used as a positive control (left panel). Vinculin was used as a loading control

Fig. 5

Functional assays reveal that FOXI1 KO leads to decreased numbers of ciliated cells in hiPSC-AECs. A: Airway surface liquid (ASL) pH of mature FOXI1 WT and KO hiPSC-AECs. Filled circles represent individual values and bars are means ± SD (n = 6 consists of 3 independent experiments, 2 biological replicates per experiment); Mann-Whitney test. B: Transepithelial resistance (R_t) of mature FOXI1 WT and KO hiPSC-AEC ALI cultures. Filled circles represent the average of 3 technical replicates (measurements) and bars are means ± SD (n = 6 ALIs from 3 independent experiments, 2 biological replicates per experiment). * P < 0.05; Mann-Whitney test. C: Ciliary beat frequency (CBF) of FOXI1 WT and KO hiPSC-AEC ALI cultures. Filled circles represent the average of values obtained from 5–20 FOVs with > 5% of coverage from one sample and bars are means ± SD (n = 3 independent experiments); Student’s t-test. D: Area covered with motile cilia in FOXI1 WT and KO hiPSC-AEC ALI cultures. Filled circles represent the average of up to 20 FOVs from one sample and bars are means ± SD (n = 3 independent experiments); Student’s t-test. E: Flow cytometry analysis of the amount of FOXJ1 + ciliated cells in FOXI1 WT and KO hiPSC-AEC ALI cultures. Gating was performed compared to stained hiPSC controls. Filled circles represent individual values and bars are means ± SD (n = 4 independent experiments); *P < 0.05; Mann-Whitney test. F: Cropped representative Western blot images show the expression of the ciliated cell marker DNAI1 in mature FOXI1 WT and KO hiPSC-AECs. Primary basal cells and HBEC ALI cultures served as negative and positive controls, respectively. Vinculin served as a loading control. G: Representative immunofluorescence staining of FOXJ1, acetylated tubulin (AcTub) and DAPI in FOXI1 WT and KO hiPSC-AECs. The scale bar is 100 μm