Generation and applications of an expandable and mature hiPSC-derived liver organoid

Abstract

Organoids have emerged as a powerful tool for modeling liver diseases, drug screening, and personalized treatments. However, they have a limited capacity to generate functional hepatocytes in a reproducible and efficient manner. Here, we designed a novel method to efficiently and reproducibly establish a protocol for generating functionally mature SB-HEOs (SB431542/BMP4-hepatic endoderm organoids) from hiPSC-derived hepatic endoderm (HE) cells. The maturation of these organoids was confirmed through transcriptome analysis and functional expression detection. We extended this culture system to various biomedical applications. This system can be used to identify hepatotoxicity with DILI drugs, model disease using OA stress, metabolize drugs using liquid chromatography-tandem mass spectrometry, and repopulate FRG mice. These organoids have both expansion and maturation characteristics, high drug metabolism ability to prolong the survival of FRG mice, can accurately identify hepatotoxic and non-hepatotoxic drugs, and mimic metabolic dysfunction-associated steatotic liver disease. Thus, SB-HEO provide a new cellular system for toxicology testing, drug metabolism, modeling liver diseases, and regenerative medicine. Especially benefiting from the high expression of CYP450 activity, SB-HEO shows high potential in the fields of drug testing and regenerative medicine.

Keywords: Cytochrome P450; Drug metabolism; Human-induced pluripotent stem cells; Liver diseases; Liver organoid; Regenerative medicine; Signaling pathway; Toxicology testing.

Graphical abstract

Figure 1

Generation of expandable liver organoids from hiPSC. (A) Schematic of HEO and HLCO generation and expansion. (B) Comparison of hepatic stem/progenitor marker expression in the liver samples. (C) Morphology of SB-HEO after Matrigel embedding (left panel). Representative recovery morphology of organoids after passaging by mechanical dissociation on Days 0 and 3 (right panel). Representative recovery morphology of organoids after thawing for 3 days (right panel). Scale bar: 200 μm. (D) IF images of SB-HEO with EPCAM, CK19, and AFP. Scale bar: 50 μm. (E) Bright-field images of different generations of SB-HEO. Scale bar: 200 μm. (F) RT-qPCR data of HPC gene expression in hiPSC, 2D HLCs, SB-HEO, SB-HEO-DM, and PHHs. The results analyzed by Two-way ANOVA are presented as the mean ± SEM, n = 3. FB-HEO: Stage 1 Protocol 1 + Stage 2 Protocol 1; SB-HEO: Stage 1 Protocol 1 + Stage 2 Protocol 2; SB-HLCO: Stage 1 Protocol 1 + Stage 2 Protocol 2 + Stage 3 Protocol 1; SM-HEO: Stage 1 Protocol 2 + Stage 2 Protocol 3; SM-HLCO: Stage 1 Protocol 2 + Stage 2 Protocol 3 + Stage 3 Protocol 2. CIP (CHIR99021, IDE1, and PD0332991); VDF (Vitamin C, Dihexa, and Forskolin); DNAD (Dihexa, NH4Cl, A8301, and dexamethasone).

Figure 2

Maturation characteristics of differentiated human liver organoids. (A) Representative morphology of organoids in EM and DM. Scale bar: 200 μm. (B) RT-qPCR analysis of hepatocyte markers in hiPSC, 2D HLCs, SB-HEO, SB-HEO-DM, and PHHs. The results analyzed by Two-way ANOVA are presented as the mean ± SEM, n = 3. (C) IF images of SB-HEO with AFP, HNF4A, and ALB. Scale bar: 50 μm. (D) Representative images of ICG uptake and release of SB-HEO-DM. Scale bar: 50 μm. (E) Representative images of periodic acid-Schiff (PAS) stain of SB-HEO and SB-HEO-DM. Scale bar: 50 μm. (F) Secretion of ALB, AAT and urea production data from 2D HLCs, SB-HEO, SB-HEO-DM, and PHHs. The results analyzed by One-way ANOVA are presented as the mean ± SEM, n = 3. (G, H) DM-cultured organoids were enriched in pathways characterized by maturation.

Figure 3

Transcriptome profiles and functional assessment of SB-HEO. (A) Principal component analysis on RNA-seq data. (B) Volcano plot of gene microarray data showing upregulated genes of interest (fold change >2 and P < 0.05) in HLCO and HEO (three biological replicates for each condition). (C) Mulberry plot with genes in KEGG pathways. (D) GO analysis of SB-HEO compared to 2D HLCs. (E) Expression of SB-HEO and 2D HLCs in the specified gene set.

Figure 4

Single-cell analysis of SB-HEO. (A) Clustering of SB-HEO cells. (B) Predicted SB-HEO cell types based on lineage marker expression patterns. (C) Expression of representative cell type-specific markers. (D) Composition of SB-HEOs. (E) Dot plot for marker levels in each cell type. (F) Cell type-specific distribution of SB-HEOs. Most cells in SB-HEOs were positioned near hepatocytes and cholangiocytes in liver tissues.

Figure 5

Toxicological outcomes prediction using SB-HEO. (A) Dose–response curves for cytotoxicity. Each data point in the graph represents the percentage of cell viability compared to a vehicle control and is presented as the mean ± SEM, n = 3. (B) IC₅₀ values from the dose–response curves in A for hepatotoxic and cytotoxic drugs in 2D HLCs, PHHs, SB-HEO, SB-HEO-DM, and HepG2. (C) Cytotoxicity comparison data as the mean ± SEM (n = 3) between toxic versus non-toxic structural analogs in SB-HEO.

Figure 6

Regenerative and inflammatory responses in SB-HEO. (A) Timeline of damage and subsequent recovery post-APAP treatment. (B) Organoid morphological images under the control, APAP treatment, and recovery conditions on the indicated days. Scale bar: 200 μm. (C) The organoid size was assessed every 12 h post-APAP in the groups. The results analyzed by Student's t-test are presented as the mean ± SEM, n = 3. (D) RT-qPCR in APAP-treated damaged and recovered organoids of inflammation-related genes on the indicated days. The results analyzed by Student's t-test are presented as the mean ± SEM, n = 3. (E) ATP levels were measured under each condition. The results are presented as the mean ± SEM (n = 3) and analyzed using Student's t-test. (F) The GSH/GSSG ratios presented as the mean ± SEM (n = 3) were evaluated in each condition. (G) HMGB1 staining of each experimental group. Scale bar: 50 μm. (H) Annexin/PI staining of each experimental group. Scale bar: 50 μm.

Figure 7

Drug screening and steatosis pathology in the human hepatic organoid model. (A) Nile red images and bright-field images of SB-HEO treated with different OA concentrations. Scale bars: BF, 500 μm; Lipid, 200 μm. (B) The relative intensity of Nile red staining in A. (C) The triglyceride content of SB-HEO treated with different OA concentrations. (D) Bright-field images and Nile red staining images of SB-HEO treated with different compounds. Scale bar: BF, 500 μm; Lipid, 200 μm. (E) The relative intensity of Nile red staining in D. (F) The triglyceride content of SB-HEO treated with different compounds. (B, C, E, F) The results expressed as the mean ± SEM (n = 3) were examined by Student's t-test. (G) Correlation plot between patient tissues (normal and NAFLD) and OA (0 and 600 μmol/L) organoids on Day 3 of culture. (H) Heat map showing the expression of dysregulated genes. (I) KEGG analysis revealed pathways enriched in genes highly expressed in OA-organoids. (J) GSEA analysis revealed pathways enriched for genes highly expressed in OA-organoids.

Figure 8

Activity of key liver CYP450 isoforms in SB-HEO, 2D HLCs, PHHs, and HepG2. CYP450 activity is expressed as pg product/μg protein. Data are displayed as the mean ± SEM (n = 3).

Figure 9

SB-HEO highly populate the FRG mice liver parenchyma. (A) Kaplan–Meier survival curves of untransplanted FRG, SB-HEO-transplanted (SB-HEO-FRG) or PHHs (PHH-FRG) mice. (B) Changes in weights of SB-HEO-FRG and PHH-FRG mice. (C) ALT and AST in the untransplanted mice (n = 3), living SB-HEO-FRG mice (n = 5, Since the amount of blood collected was not enough for the experiment, 2 mice were not included in the statistics.), and living PHH-FRG mice (n = 5, Since the amount of blood collected was not enough for the experiment, 3 mice were not included in the statistics.) 2 months post-transplantation. Data are displayed as the mean ± SEM. (D) The serum ALB levels of SB-HEO-FRG mice (n = 5) and PHH-FRG mice (n = 5) were measured three months post-transplantation. (E) Control staining for hALB in mouse liver. Immunostaining of SB-HEO-FRG mice with hALB, CYP2E1, FAH, and GS three months post-transplantation. Scale bar: 50 μm.