Advertisement

High-Fidelity Drug-Induced Liver Injury Screen Using Human Pluripotent Stem Cell–Derived Organoids

  • Author Footnotes
    ∗ Authors share co-first authorship.
    Tadahiro Shinozawa
    Footnotes
    ∗ Authors share co-first authorship.
    Affiliations
    Division of Gastroenterology, Hepatology & Nutrition, Developmental Biology and Center for Stem Cell and Organoid Medicine (CuSTOM), Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
    Search for articles by this author
  • Author Footnotes
    ∗ Authors share co-first authorship.
    Masaki Kimura
    Footnotes
    ∗ Authors share co-first authorship.
    Affiliations
    Division of Gastroenterology, Hepatology & Nutrition, Developmental Biology and Center for Stem Cell and Organoid Medicine (CuSTOM), Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
    Search for articles by this author
  • Author Footnotes
    ∗ Authors share co-first authorship.
    Yuqi Cai
    Footnotes
    ∗ Authors share co-first authorship.
    Affiliations
    Division of Gastroenterology, Hepatology & Nutrition, Developmental Biology and Center for Stem Cell and Organoid Medicine (CuSTOM), Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
    Search for articles by this author
  • Norikazu Saiki
    Affiliations
    Institute of Research, Tokyo Medical and Dental University (TMDU), Tokyo, Japan
    Search for articles by this author
  • Yosuke Yoneyama
    Affiliations
    Institute of Research, Tokyo Medical and Dental University (TMDU), Tokyo, Japan
    Search for articles by this author
  • Rie Ouchi
    Affiliations
    Division of Gastroenterology, Hepatology & Nutrition, Developmental Biology and Center for Stem Cell and Organoid Medicine (CuSTOM), Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
    Search for articles by this author
  • Hiroyuki Koike
    Affiliations
    Division of Gastroenterology, Hepatology & Nutrition, Developmental Biology and Center for Stem Cell and Organoid Medicine (CuSTOM), Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
    Search for articles by this author
  • Mari Maezawa
    Affiliations
    Institute of Research, Tokyo Medical and Dental University (TMDU), Tokyo, Japan
    Search for articles by this author
  • Ran-Ran Zhang
    Affiliations
    Division of Gastroenterology, Hepatology & Nutrition, Developmental Biology and Center for Stem Cell and Organoid Medicine (CuSTOM), Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
    Search for articles by this author
  • Andrew Dunn
    Affiliations
    Division of Gastroenterology, Hepatology & Nutrition, Developmental Biology and Center for Stem Cell and Organoid Medicine (CuSTOM), Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
    Search for articles by this author
  • Autumn Ferguson
    Affiliations
    Division of Gastroenterology, Hepatology & Nutrition, Developmental Biology and Center for Stem Cell and Organoid Medicine (CuSTOM), Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
    Search for articles by this author
  • Shodai Togo
    Affiliations
    Division of Gastroenterology, Hepatology & Nutrition, Developmental Biology and Center for Stem Cell and Organoid Medicine (CuSTOM), Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
    Search for articles by this author
  • Kyle Lewis
    Affiliations
    Division of Gastroenterology, Hepatology & Nutrition, Developmental Biology and Center for Stem Cell and Organoid Medicine (CuSTOM), Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
    Search for articles by this author
  • Wendy L. Thompson
    Affiliations
    Division of Gastroenterology, Hepatology & Nutrition, Developmental Biology and Center for Stem Cell and Organoid Medicine (CuSTOM), Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
    Search for articles by this author
  • Akihiro Asai
    Affiliations
    Division of Gastroenterology, Hepatology & Nutrition, Developmental Biology and Center for Stem Cell and Organoid Medicine (CuSTOM), Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
    Search for articles by this author
  • Takanori Takebe
    Correspondence
    Correspondence Address correspondence to: Takanori Takebe, Cincinnati Children's Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, Ohio 45229-3039.
    Affiliations
    Division of Gastroenterology, Hepatology & Nutrition, Developmental Biology and Center for Stem Cell and Organoid Medicine (CuSTOM), Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio

    Institute of Research, Tokyo Medical and Dental University (TMDU), Tokyo, Japan

    Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, Ohio

    Communication Design Center, Advanced Medical Research Center, Yokohama City University Graduate School of Medicine, Japan
    Search for articles by this author
  • Author Footnotes
    ∗ Authors share co-first authorship.
Open AccessPublished:October 08, 2020DOI:https://doi.org/10.1053/j.gastro.2020.10.002

      Background & Aims

      Preclinical identification of compounds at risk of causing drug induced liver injury (DILI) remains a significant challenge in drug development, highlighting a need for a predictive human system to study complicated DILI mechanism and susceptibility to individual drug. Here, we established a human liver organoid (HLO)–based screening model for analyzing DILI pathology at organoid resolution.

      Methods

      We first developed a reproducible method to generate HLO from storable foregut progenitors from pluripotent stem cell (PSC) lines with reproducible bile transport function. The qRT-PCR and single cell RNA-seq determined hepatocyte transcriptomic state in cells of HLO relative to primary hepatocytes. Histological and ultrastructural analyses were performed to evaluate micro-anatomical architecture. HLO based drug-induced liver injury assays were transformed into a 384 well based high-speed live imaging platform.

      Results

      HLO, generated from 10 different pluripotent stem cell lines, contain polarized immature hepatocytes with bile canaliculi-like architecture, establishing the unidirectional bile acid transport pathway. Single cell RNA-seq profiling identified diverse and zonal hepatocytic populations that in part emulate primary adult hepatocytes. The accumulation of fluorescent bile acid into organoid was impaired by CRISPR-Cas9–based gene editing and transporter inhibitor treatment with BSEP. Furthermore, we successfully developed an organoid based assay with multiplexed readouts measuring viability, cholestatic and/or mitochondrial toxicity with high predictive values for 238 marketed drugs at 4 different concentrations (Sensitivity: 88.7%, Specificity: 88.9%). LoT positively predicts genomic predisposition (CYP2C9∗2) for Bosentan-induced cholestasis.

      Conclusions

      Liver organoid-based Toxicity screen (LoT) is a potential assay system for liver toxicology studies, facilitating compound optimization, mechanistic study, and precision medicine as well as drug screening applications.

      Graphical abstract

      Keywords

      Abbreviations used in this paper:

      BSEP (bile salt export pump), CLF (cholyl-lysyl-fluorescein), Cmax (clinical maximum drug concentration), CYP (cytochrome P), DILI (drug-induced liver injury), FD (fluorescein diacetate), FGF (fibroblast growth factor), hiPSC (human induced pluripotent stem cell), HLO (human liver organoid), iPSC (induced pluripotent stem cell), LoT (liver organoid-based toxicity), MMP (mitochondrial membrane potential), MRP (multidrug resistance-associated protein), NAC (N-acetylcysteine), PSC (pluripotent stem cell), qPCR (quantitative polymerase chain reaction), RA (retinoic acid), ROS (reactive oxygen species), scRNA-seq (single cell RNA sequencing), TCs (training compounds)
      See Covering the Cover synopsis on page 633; See editorial on page 653.

       Background and Context

      Preclinical identification for drug induced liver injury (DILI) represents a major challenge due to complex pathogenesis and highly variable individual susceptibility to drug.

       New Findings

      We established a human liver organoid (HLO) based screening model for analyzing DILI pathology and provided the possibility to assess different susceptibility based on polymorphism at organoid resolution.

       Limitations

      This study was performed in at a single institution with limited compound numbers.

       Impact

      Self-organized HLO with human hepatocyte-like properties including bile transport function predicts known cholestatic DILI drug activity, and have potential to predicts genomic predisposition for DILI.
      Billions of dollars are lost annually from drug development in the pharmaceutical industry due to the failures of drug candidates identified in initial screens, and nearly one-third of drugs are withdrawn from the market.
      • Takebe T.
      • Taniguchi H.
      Human iPSC-derived miniature organs: a tool for drug studies.
      Further, despite the promising efficacy, a failure of drug candidates results in tremendous loss of a patient’s treatment opportunity. Preclinical studies generally consist of an in vitro evaluation as a primary efficacy screen to identify a “hit” compound, followed by safety studies in vitro and in vivo to assess the mechanisms of metabolism and toxicology. This inefficiency can be explained by the substantial lack of physiologically relevant preclinical models in evaluating drug-induced liver injury (DILI) in humans and thus, an urgent need to develop an in vitro screening model for the evaluation of the vast amounts of continuously growing compound libraries.
      Primary hepatocytes are a highly polarized metabolic cell type and form a bile canaliculi structure with microvilli-lined channels, separating the peripheral circulation from the bile acid secretion pathway. The most upstream aspects of DILI include detoxification of drugs or their reactive metabolites by hepatocytes and excretion into bile canaliculi through transporters such as multidrug resistance-associated protein (MRP) transporters, suggesting the need to reconstruct these uniquely organized structures as a crucial property of hepatocytes in vivo for predicting DILI pathology.
      There are, however, considerable differences in drug toxicity profiles between the current simplified culture model with the use of isolated primary human hepatocytes or hepatic cell lines and in vivo physiology, resulting in failed translation of drugs or drug discontinuation, for example, troglitazone, nefazodone, and tolcapone (https://livertox.nlm.nih.gov/index.html). The determination of toxicologic properties thus mainly relies on animals as an essential step toward drug development; however, due to the pronounced differences in physiology between humans and animals, there is a significant lack of fidelity to human outcomes.
      • Yang K.
      • Woodhead J.L.
      • Watkins P.B.
      • et al.
      Systems pharmacology modeling predicts delayed presentation and species differences in bile acid-mediated troglitazone hepatotoxicity.
      ,
      • Leslie E.M.
      • Watkins P.B.
      • Kim R.B.
      • et al.
      Differential inhibition of rat and human Na+-dependent taurocholate cotransporting polypeptide (NTCP/SLC10A1)by bosentan: a mechanism for species differences in hepatotoxicity.
      We recently reported a method to generate human liver organoids (HLOs) from pluripotent stem cells (PSCs) with a potential to model inflammatory diseases.
      • Ouchi R.
      • Togo S.
      • Kimura M.
      • et al.
      Modeling steatohepatitis in humans with pluripotent stem cell-derived organoids.
      However, it remains elusive whether the functional bile canaliculi-like structure, an essential component for modeling defective bile excretion, is formed and can be used toward a drug toxicology analysis. Additionally, while minimizing batch differences, the enhancement of the assay throughput will be critical before being translated into preclinical studies. Herein we developed a reproducible liver organoid protocol using stably expandable foregut cells from human stem cells (ie, induced PSC [iPSC] and embryonic stem cells). We established a live imaging-based dynamic detection assay for bile acid uptake and excretion in modified HLOs that can be prevented by bile acid transporter gene knockout. This assay platform is amenable for large-scale compound screening and annotated 238 drugs with multiplexed readouts. Furthermore, HLOs were shown to model genotype-specific susceptibility
      • Inoue H.
      • Nagata N.
      • Kurokawa H.
      • et al.
      iPS cells: a game changer for future medicine.
      to bosentan-induced cholestasis with multiple iPSC. This robust assay, named liver organoid-based toxicity screen (LoT), provides functional readout developed in HLOs and will facilitate diagnosis, functional studies, drug development, and personalized medicine.

      Materials and Methods

       Human Pluripotent Stem Cells

      Maintenance of the PSCs T (TkDA3) human iPSC clone used in this study was kindly provided by K. Eto and H. Nakauchi, and 12 (1231A3) and 13 (1383D6) were gifted by Kyoto University (Japan). CW10150, CW10027, CW10077, and WD90, WD91, and WD92 were purchased from Coriell Institute for Medical Research (Camden, NJ). iPSC_285.0, iPC_18.4, and iPSC_54.1 were obtain from patient skin fibroblasts/peripheral blood mononuclear cells and reprogrammed into iPSC by Cincinnati Children's Hospital Medical Center pluripotent stem cell core. The human H1 embryonic stem cell clone used in this study was kindly provided by WiCell Institute. Human (h)iPSC lines were maintained as described previously.
      • Takebe T.
      • Enomura M.
      • Yoshizawa E.
      • et al.
      Vascularized and complex organ buds from diverse tissues via mesenchymal cell-driven condensation.
      ,
      • Takebe T.
      • Zhang R.R.
      • Koike H.
      • et al.
      Generation of a vascularized and functional human liver from an iPSC-derived organ bud transplant.
      Briefly, undifferentiated hiPSCs were cultured on Laminin 511E8-flagment (Nippi) coated dishes in Stem Fit medium (Ajinomoto Company) with 100 ng/mL basic fibroblast growth factor (FGF; R&D Systems) at 37°C in 5% CO2 with 95% air.

       Foregut Induction

      The hiPSCs were differentiated into foregut using a previously described method with modifications.
      • Spence J.R.
      • Mayhew C.N.
      • Rankin S.A.
      • et al.
      Directed differentiation of human pluripotent stem cells into intestinal tissue in vitro.
      In brief, hiPSCs were detached by Accutase (Thermo Fisher Scientific Inc) and were seeded on laminin-coated tissue culture plates with 100,000 cells/cm
      • Yang K.
      • Woodhead J.L.
      • Watkins P.B.
      • et al.
      Systems pharmacology modeling predicts delayed presentation and species differences in bile acid-mediated troglitazone hepatotoxicity.
      . Medium was changed to RPMI 1640 medium (Life Technologies) containing 100 ng/mL activin A (R&D Systems) and 50 ng/mL bone morphogenetic protein 4 (BMP4; R&D Systems) at day 1, 100 ng/mL activin A and 0.2% fetal calf serum (Thermo Fisher Scientific Inc) at day 2, and 100 ng/mL activin A and 2% fetal calf serum at day 3. On days 4 to 6, cells were cultured in Advanced DMEM/F12 (Thermo Fisher Scientific Inc) with B27 (Life Technologies) and N2 (Gibco) containing 500 ng/mL FGF4 (R&D Systems) and 3 μmol/L CHIR99021 (Stemgent). Cells were maintained at 37°C in 5% CO2 with 95% air, and the medium was replaced every day. The foregut cells were detached by Accutase and then frozen in Cell Banker 1(Nippon Zenyaku Kogyo Co, Ltd). The foregut cells can be stored in −80°C or liquid nitrogen for long-term storage. The differentiation protocol is illustrated in Figure 1A.
      Figure thumbnail gr1
      Figure 1Generation of HLOs from storable PSC-derived foregut cells. (A) Overview of our differentiation method for liver organoids. (B) Representative image of HLO at day 20. Scale bar = 200 μm. (C) Phase-contrast image of HLO in different culture conditions with 5 factors, which were FGF2, vascular endothelial growth factor (VEGF), epidermal growth factor (EGF), CHIR-99021 (glycogen synthase kinase 3 inhibitor), and A83-01 (transforming growth factor-β inhibitor). Scale bar = 200 μm. (D) The number of HLOs cultured with 5 factors. Conventional: Treated with only RA. (E) The secretion of albumin from HLOs in culture condition with treatment with different factors. (F) Reverse transcription qPCR analysis of representative genes related to hepatic function and pluripotency markers. Undifferentiated iPSCs (iPSC, n = 8), posterior foregut organoid (foregut, n = 8), HLOs (n = 16), and primary liver (human primary hepatocytes, n = 8). (G) The secretion of albumin in HLOs (n = 92) and primary hepatocytes (n = 14). (H) The secretion of complement factors in HLOs (n = 4). (I) CYP3A4 activation by rifampicin treatment (n = 5). (J) CYP1A2 activation by omeprazole treatment (n = 5). (K) The CYP2C9 induction capacity after rifampicin treatment in HLOs and primary hepatocytes (PHH). Bars represent the mean ± SD (n = 7 for HLOs, n = 9 for primary hepatocytes).

       Generation of Human Liver Organoids

      The frozen foregut cells were thawed quickly and then centrifuged at 1200 rpm for 3 minutes. Cells was resuspended in Matrigel (Corning, Inc). A total of 100,000 cells were embedded in 50 μL Matrigel drop on the dishes in organoid formation media with 5 factors for 4 days. After organoid formation, the media was switched to liver specification media for 4 days. After the liver specification step, organoids were harvested from Matrigel by scratching and pipetting. Then organoids were re-embedded in Matrigel on the Ultra-Low Attachment Multiwell Plate (Corning) in liver maturation media for 10 days. Cultures for HLO induction were maintained at 37°C in 5% CO2 with 95% air, and the medium was added every 2 days (see Supplementary Methods).

       Large-Scale Screening With 238 Test Compounds

      Hepatotoxicity library (Enzo SCREEN-WELL): Each drug was diluted into 4 doses (100 μmol/L, 10 μmol/L, 1 μmol/L, and 0.1 μmol/L) and dispersed in 384-well plates using Eppendorf epMotion 5075 liquid handler. Each plate included on-board controls for toxicity (eg, penicillin V, lactic acid) and delivery vehicle (dimethyl sulfoxide). The hiPSCs (TkDA3) were differentiated into liver organoids as described previously. At day 15, embedded liver organoids were replated in floating culture by disrupting Matrigel using gentle pipetting. At day 21, floating cultured liver organoids were seeded at ∼15 organoids per well into 384-well plates (Corning high content imaging plate, Cat. 4681), which contained test compounds.
      Cholyl-lysyl-fluorescein (CLF; 5 nmol/L) was added before imaging at 24 hours. Multichannel fluorescence and bright-field images were acquired by Nikon Ti-E SpectraX Widefield Microscope equipped with high-speed, triggerable Lumencor SpectraX LED light engine under 40× magnification. Imaging was processed in ∼7 minutes per plate. Measurements of image intensity were obtained from each individual organoid (n = ∼10). Cell viability was tested using the CellTiter-Glo luminescent cell viability assay (Promega) and quantified by a BioTek Synergy H1 plate reader after 72 hours. Each experiment was repeated in triplicate. The data were given from 5 independent experiments.

       Image Data Analysis

      Nikon Elements analysis software was used to identify fluorescent pixels expressed within organoids based on intensity gaiting, size, and circularity filtering. Fluorescent data are taken as an intensity average to normalize for any changes in organoid size. Moreover, because each organoid imaged within a well contains its own analyzed fluorescent intensity data, an average and SD of the intensity data across all organoids in a single well is then taken to account for variation in organoid number.
      Although the size of the HLOs was relatively uniform (Supplementary Figure 1), to account for variation in HLO size and number during cell titer glow, a custom image analysis script was developed in MATLAB (MathWorks) using Canny edge detection, watershed segmentation, and filtering based on size and circularity to identify total cellular area per well as the normalizing factor for the luminescent intensity (CellTiter-Glo) data characterizing organoid viability. Pixels were regarded as comprising part of an organoid if a pixel appeared as positive (value of 1 in binary images) in ≥2 sets of processed images. Each well’s CellTiter-Glo value was calculated as luminescent signal/organoid area. All data points reflected on the 2- or 3-dimensional plots were normalized to the control (dimethyl sulfoxide) wells.

       Statistical Analysis

      Statistical significance was determined using the unpaired Student t test, the Dunnett’s test, or one-way analysis of variance with Dunnett’s multiple comparison post hoc test. P < .05 was considered significant.

      Results

       Generation and Characterization of Polarized Liver Organoids From Multiple Human Induced Pluripotent Stem Cells

      We first established a new method to obtain uniform organoids in a large quantity using hiPSC-derived foregut spheroids
      • Ouchi R.
      • Togo S.
      • Kimura M.
      • et al.
      Modeling steatohepatitis in humans with pluripotent stem cell-derived organoids.
      (Figure 1A). Briefly, foregut cells are dissociated into single cells on day 7. In this stage, dissociated foregut cells can be cryopreserved at −80°. Fresh or thawed cells were embedded into Matrigel, followed by 4 days of organoid formation medium composed of 5 factors, which were FGF2, vascular endothelial growth factor, endothelial growth factor, epidermal growth factor, a glycogen synthase kinase 3 GSK inhibitor (CHIR99021), and a transforming growth factor-β inhibitor (A83-01), with ascorbic acid
      • Zhang R.R.
      • Koido M.
      • Tadokoro T.
      • et al.
      Human iPSC-derived posterior gut progenitors are expandable and capable of forming gut and liver organoids.
      (Supplementary Figure 2A and B). On the following day, we exposed the cultures to 4 days of retinoic acid (RA), which enhances the formation of the bile canaliculi and the pericanalicular sheaths.
      • Falasca L.
      • Favale A.
      • Serafino A.
      • et al.
      The effect of retinoic acid on the re-establishment of differentiated hepatocyte phenotype in primary culture.
      Next, we switched to hepatocyte maturation medium (see Materials and Methods). Organoids with intraluminal structure were efficiently produced (Figure 1B), and 5 factors provided the larger number of HLOs than conventional factors, which was only treated with RA (Figure 1C and D). We confirmed that HLOs were formed from multiple iPSC lines (Supplementary Figure 2C) under 5 factors. Furthermore, we demonstrate that HLOs cultured in the organoid formation medium with 5 factors, then RA for 96 hours, have the highest albumin secretion (Figure 1E).

       Gene Expressions and Functional Profiling in Human Liver Organoids

      To profile hepatic lineage in HLOs, the expressions of some representative genes related to hepatic functional liability were analyzed by quantitative polymerase chain reaction (qPCR). The expression of hepatic marker genes was upregulated in HLO, such as albumin (ALB), α-fetoprotein (AFP), cytochrome P (CYP) 450 family 2 subfamily C member 9 (CYP2C9), and CYP7A1 (Figure 1F). Also increased were expressions of genes related to polarity and transporter activities, such as multidrug resistance-associated protein 2 (MRP2), bile salt export pump (BSEP), multiple drug resistance 1 (MDR1), BCRP1, Na+-taurocholate co-transporting polypeptide (NTCP), organic anion transporter 2 (OAT2), concentrative nucleoside transporter 1 (CNT1), and MRP3. Moreover, this was accompanied by progressive decreases in genes related to the undifferentiated state, such as NANOG and OCT4. The gene expression of NANOG, OCT4, ALB, CYP2C9, MRP3, and TDO2 in HLOs generated from 6 different donors were comparable (Supplementary Figure 3A).
      The albumin secretion capacity of HLO is shown in Figure 1G. HLO and primary hepatocytes albumin secretion levels were 488.2 ± 517.5 ng/mL/d and 630.49 ± 292.5 ng/mL/d, respectively, which were not statistically different. Although the albumin secretion levels in HLOs and primary liver were comparable, gene expression of albumin in both were different, presumably due to different cell compositions in HLOs and primary liver.
      In addition, to investigate the difference of HLOs generated from foregut, with or without frozen process, the albumin secretion level in both HLOs were tested and comparable (Supplementary Figure 3B), suggested that the influence of frozen process on the foregut stage was not large for hepatic functions in HLO.
      Hepatocyte-specific proteins, such as complement factors, were also confirmed in HLO culture supernatant by enzyme-linked immunosorbent assay (Figure 1H). Finally, at a functional level, we demonstrate the major CYP induction responses in HLO (ie, CYP3A4 and CYP1A2) after treatment with rifampicin and omeprazole, respectively (Figure 1I and J). Additionally, we tested CYP2C9 inducibility in HLO. After the treatment with rifampicin, CYP2C9 reactivity in HLO was comparable to that of primary hepatocytes (Figure 1K).

       Single-Cell RNA Sequencing Profiling in Human Liver Organoids

      To compare the differentiation status with those of primary hepatocytes, we used single-cell RNA sequencing (scRNA-seq) to measure the transcriptome of HLO-derived cells (5177 cells) in HLOs at day 20 and conducted integrated analysis with primary human liver-derived cells derived from 5 independent donors
      • MacParland S.A.
      • Liu J.C.
      • Ma X.Z.
      • et al.
      Single cell RNA sequencing of human liver reveals distinct intrahepatic macrophage populations.
      (Figure 2A and Supplementary Figure 4). Analysis with t-distributed stochastic neighbor embedding revealed distinct populations consisting of a parenchymal (74.41%) and a nonparenchymal population (25.59%) in HLOs (Figure 2B and C) that express characteristic markers of hepatic stellate cells, portal endothelial cells, and cholangiocytes in primary samples, as seen in our previous method.
      • Ouchi R.
      • Togo S.
      • Kimura M.
      • et al.
      Modeling steatohepatitis in humans with pluripotent stem cell-derived organoids.
      Figure thumbnail gr2
      Figure 2scRNA-seq profiling of hepatic cells in HLOs. (A) Integrated t-distributed stochastic neighbor embedding (tSNE) map of 5177 HLO cells and 8439 primary human liver cells. Each point indicates a single cell, which is colored by cell type (upper and right bottom) or original source. (B) In integrated tSNE map of HLO and primary hepatocytes, the feature plots showed expression of hepatic marker genes. The color bar indicates scaled gene expression level. (C) Integrated tSNE map and clustering of 3852 hepatoblast-like and hepatocyte-like cells in HLOs and 3507 hepatocytes in primary human liver identify zonal characters in A and C. The inset indicates the cell cycle phase of each cell. (D) The feature plots showed expression of 4 zonal pericentral markers of hepatocytic clusters. (E) The table shows the percentage of hepatoblast-like and hepatocyte-like cell of HLO in each cluster.
      Next, we investigated gene expressions related to pericentral, periportal, mesenchyme, and endothelial and compared with those of human primary livers (Supplementary Figure 4A). Of them, although HLO-derived cells include nonparenchymal population, subsets of hepatic cells, termed as hepatocyte-like1, 2 cells, are nearly identical to primary hepatocytes, with the remaining half of cells representing a hepatoblast-like state. In the gene set of mesenchyme and endothelial, the gene expression in HLOs shared similarity to primary liver tissues.
      To further probe the hepatocytic identity, we isolated the hepatocytic population from both HLO and primary liver samples and conducted t-distributed stochastic neighbor embedding analysis. We found 88.99% of hepatocyte-like 2 cells were in the periportal primary hepatocyte population, whereas 27.91% of hepatocyte-like 1 cells were in the pericentral primary hepatocyte population (Figure 2D and E).
      Zone 3 (centrilobular region in the liver) plays a role in drug metabolism and detoxification. Four zonal pericentral markers, recently reported in human scRNA-seq,
      • MacParland S.A.
      • Liu J.C.
      • Ma X.Z.
      • et al.
      Single cell RNA sequencing of human liver reveals distinct intrahepatic macrophage populations.
      are notably expressed at similar levels to primary hepatocytes, which include a key drug metabolism enzyme, CYP2C9 (Figure 1F). Pathway analysis indicated highly enriched gene sets in the pericentral and periportal population in HLOs correlate with lipid/drug metabolism, CYP450, and cholesterol biosynthetic process, respectively. Gene expression levels in these pathways in HLOs were similar to those in primary liver (Supplementary Figure 4B). Collectively, our HLO model harbors diverse and zonal hepatocytic populations that in part emulate primary adult hepatocyte profiles.

       Structural Profiling With Microanatomical Characterization in Human Liver Organoids

      Immunohistochemistry analysis revealed albumin and type IV collagen staining in the epithelial cells of HLO (Figure 3A). In addition, immunostaining of zonula occludens 1, MRP2, BSEP, F-actin, and MDR3 demonstrated that these proteins preferentially localized in the intraluminal region with staining hepatocyte nuclear factor 4a (HNF4a). CYP7A1 was also positively stained in HLO.
      Figure thumbnail gr3
      Figure 3Structural profiling of HLO. (A) Immunostaining for albumin (ALB), collagen IV, zonula occludens 1 (ZO-1), MRP2, E-cadherin (E-cad), BSEP, F-actin, MDR3, CYP7A1, and HNF4a in HLOs. Nuclei were stained with 4′,6-diamidino-2-phenylindole (blue). Scale bars = 50 μm. (B) Transmission electron micrograph of an HLO showing microvilli (V) intraluminal surface, bile canaliculi-like structure (BC), lumen (L), and nuclei (N). Scale bars = 10 μm.
      The bile canaliculus is the smallest intrahepatic secretory channel, and the canalicular lumen consists of a space formed by a modified apical region of the opposing plasma membranes of contiguous hepatocytes.
      • Tsukada N.
      • Ackerley C.A.
      • Phillips M.J.
      The structure and organization of the bile canalicular cytoskeleton with special reference to actin and actin-binding proteins.
      In addition, it is delimited by tight junction complexes, and the microvilli are located on the inside of the canalicular lumen.
      • Tsukada N.
      • Ackerley C.A.
      • Phillips M.J.
      The structure and organization of the bile canalicular cytoskeleton with special reference to actin and actin-binding proteins.
      Similarly, transmission electron microscopy analysis of HLO confirmed the bile canaliculus-like structure in between the hepatocyte-like cells (Figure 3B, left). Transmission electron microscopy analysis also revealed HLO contained microvilli directed toward the lumen (Figure 3B, right). The immunohistochemistry analysis showing that MDR3, MRP2, BSEP, and zonula occludens 1 stained the intraluminal lining suggested that these HLOs have polarized characteristics. Consistent with these anatomical features, qPCR analysis revealed that HLOs had gene expression of BSEP and NTCP (Figure 1F). Therefore, the HLOs contained polarized human hepatocytes separated from the internal lumen surrounded by canaliculi-like structures, which reflects a unique microanatomical architecture resembling in vivo hepatic tissues.

       Bile Acid Producing and Transport Properties in Human Liver Organoids

      Next, to determine bile acid production capacity, we conducted a bile acid enzyme-linked immunosorbent assay on intraluminal fluid collected from organoid cultures. As shown in Figures 1F and 3A, CYP7A1, which catalyzes the initial step in cholesterol catabolism and bile acid synthesis, was expressed and stained positively in HLOs, respectively, suggesting the activation of the bile acid synthesis pathway in HLOs. The level of the total bile acid pool of intraluminal fluid was 26.7 μg/d per 106 cells (∼125 μmol/L in a single HLO with a 200-μm diameter) (Figure 4A), and surprisingly, the bile acid concentration was comparable to that in primary hepatocytes derived from sandwich culture (∼40 μg/d per 106 cells, 10 μmol/L in culture supernatant) in previous reports.
      • Ni X.
      • Gao Y.
      • Wu Z.
      • et al.
      Functional human induced hepatocytes (hiHeps) with bile acid synthesis and transport capacities: a novel in vitro cholestatic model.
      Bile acid excretion is the major determinant of bile flow, therefore, defects in this system may result in impaired bile secretion (cholestasis) associated with various liver disease pathologies .
      Figure thumbnail gr4
      Figure 4Bile transport property in HLO. (A) Total bile acid (TBA) secretion level inside HLOs at day 27. Bars represent the mean ± SEM (n = 4). (B) Sequential images for efflux of FD toward the inside of the organoid from outside. The arrow indicates surface of an organoid. Scale bar = 50 μm. (C) Overview of the method to investigate transporter BSEP activity in organoids. (D) Representative phase-contrast image of BSEP-mutated HLO. Scale bar = 200 μm (left image); scale bar = 50 μm (right image). (E) Representative image of bile acid uptake in BSEP-mutated HLOs, cyclosporine A (CSA), and sitaxentan (SIT), which were BSEP inhibitors after 30 minutes of culture in the presence of CLF. Scale bar = 200 μm. (F) Graph indicates CLF intensity levels in each HLO. The white circles show the medians, box limits indicate the 25th and 75th percentiles, as determined by R software (R Foundation for Statistical Computing); whiskers extend 1.5 times the interquartile range from the 25th and 75th percentiles; and polygons represent density estimates of data and extend to extreme values. BSEPmut, BSEP mutant; SIT100, sitaxentan 100 μmol/L; SIT200, 200 μmol/L. ∗P < .01 on t test; #P < .05 on Dunnett’s test. (G) The viability of HLOs. ∗P < .01 on Dunnett’s test.
      Efflux transport proteins located in the apical (canaliculi) membranes of hepatocytes play an important role in the hepatic elimination of many endogenous and exogenous compounds, including drugs and metabolites.
      • Kock K.
      • Brouwer K.L.
      A perspective on efflux transport proteins in the liver.
      BSEP and MRP2 mediate canaliculi bile salt transport in humans. As the positive expression of key proteins for bile transport in Figure 3A, we next wondered whether the HLO can actively transport bile acid into its lumen. To monitor the dynamics on bile transport activity in HLOs, fluorescein diacetate (FD), which was used to investigate an active transport mechanism from the sinusoidal membrane to the bile canalicular membrane in vitro,
      • Yumoto A.U.
      • Watanabe S.
      • Hirose M.
      • et al.
      Structural and functional features of bile canaliculi in adult rat hepatocyte spheroids.
      was applied. HLOs incubated with FD demonstrated that FD was transported into inside HLOs with changing the intensity sequentially in 45 minutes (Figure 4B). In addition, the fluorescent bile acid CLF and cholylglycylamido-fluorescein, an indicator for bile acid transport activity
      • Keitel V.
      • Burdelski M.
      • Vojnisek Z.
      • et al.
      De novo bile salt transporter antibodies as a possible cause of recurrent graft failure after liver transplantation: a novel mechanism of cholestasis.
      and a bile salt analog,
      • Mork L.M.
      • Isaksson B.
      • Boran N.
      • et al.
      Comparison of culture media for bile acid transport studies in primary human hepatocytes.
      respectively, were found to be reproducibly excreted from outside and accumulated into the intralumen of HLOs (Supplementary Figure 5A and B).
      To determine the specificity deeper, we developed an iPSC line carrying a defunctionalized BSEP allele constructed by the clustered regularly interspaced short palindromic repeat (CRISPR)-Cas9 based gene editing approach (Figure 4C). Although the morphology of BSEP-mutated iPSC-HLO was similar to the control HLO (Figure 4D), the BSEP-mutated HLO failed to accumulate fluorescent bile acid compared with the parental control HLO (Figure 4E). Similarly, the chemical BSEP inhibitor sitaxentan at viable dosing inhibited CLF transport in normal HLOs (Figure 4F and G). These data suggest that HLOs have the ability to uptake bile acid from the outside and efflux them inside the HLO. Thus, HLOs do not merely have the canaliculi-like morphology but also possess bile acid production and secretion activity, suggesting that the bile acids transport pathway is functionally constructed.

       High-Throughput Drug-Induced Cholestasis Evaluation in Human Liver Organoids

      During drug discovery, in the early stages of DILI investigation, functional and quantifiable assays are critical for lead compound generation. However, current models suffer from oversimplified readouts, such as only viability despite multiple factors being involved such as bile transport defects. To address these limitations, we developed a 384 well–based high-speed live imaging platform that we refer to as an HLO-based high throughput toxicity screening (LoT) model (Figure 5A). Our LoT system can evaluate 15 to 20 organoids per well with a unique imaging algorithm (Figure 5B) and has functionally validated 238 marketed drugs, including 32 negative control and 206 reported DILI compounds with 4 different concentrations based on dual readouts: viability and cholestatic function.
      Figure thumbnail gr5
      Figure 5The LoT screen identifies compounds with toxic potential. (A) Schematic of the DILI screen workflow with image of quantification strategy for CLF and cell viability. (B) Representative images of 384-well drug screen plate with quantification of CLF at 24 hours. CLF monofluorescence converted to intensity plot by ImageJ (National Institutes of Health). Right upper image: negative compound (sucrose), right bottom image: positive compounds (CSA). See . (C) Violin plot of CLF intensity in HLO treated with compounds relative to control. (D) Scatter plot of the viability and CLF intensity. Sen: sensitivity (true positive/true positive + false negative), Spe: specificity (true negative/true negative + false positive). Circle dots: the data of DILI positive compounds. Triangle dot: the data of DILI-negative compounds (n = 5). (E) CLF intensity of representative drugs categorized into negative and cholestatic groups.
      In the system, CLF transport inhibition induced by representative cholestatic drugs was observed. First, we screened the viability in response to the treatment at multiple doses comparing with previous reports (Supplementary Table 1). To evaluate the predictability of the LoT system, we extracted the clinical maximum drug concentration (Cmax) value of each compound listed in Supplementary Table 1,
      • Khetani S.R.
      • Kanchagar C.
      • Ukairo O.
      • et al.
      Use of micropatterned cocultures to detect compounds that cause drug-induced liver injury in humans.
      • O’Brien P.J.
      • Irwin W.
      • Diaz D.
      • et al.
      High concordance of drug-induced human hepatotoxicity with in vitro cytotoxicity measured in a novel cell-based model using high content screening.
      • Vorrink S.U.
      • Zhou Y.
      • Ingelman-Sundberg M.
      • et al.
      Prediction of drug-induced hepatotoxicity using long-term stable primary hepatic 3D spheroid cultures in chemically defined conditions.
      and analyzed the data at a concentration closest over Cmax. As a result, the sensitivity and specificity were 88.7% and 88.9% (or 69.0% and 100%), respectively, and provided the comparable or higher values to previous reports on primary hepatocyte-based models
      • Khetani S.R.
      • Kanchagar C.
      • Ukairo O.
      • et al.
      Use of micropatterned cocultures to detect compounds that cause drug-induced liver injury in humans.
      ,
      • Vorrink S.U.
      • Zhou Y.
      • Ingelman-Sundberg M.
      • et al.
      Prediction of drug-induced hepatotoxicity using long-term stable primary hepatic 3D spheroid cultures in chemically defined conditions.
      • Proctor W.R.
      • Foster A.J.
      • Vogt J.
      • et al.
      Utility of spherical human liver microtissues for prediction of clinical drug-induced liver injury.
      • Xu J.J.
      • Henstock P.V.
      • Dunn M.C.
      • et al.
      Cellular imaging predictions of clinical drug-induced liver injury.
      (Supplementary Table 2 and Figure 5C). Interestingly, toxicity induced by indomethacin (Cmax: 8.38 μmol/L) and zileuton (Cmax: 13.12 μmol/L) was successfully detected at 10 μmol/L and 100 μmol/L in the cytotoxicity assay, respectively, while these drugs were previously difficult to be detected in other platforms.
      • Fromenty B.
      Drug-induced liver injury in obesity.
      ,
      • Michaut A.
      • Le Guillou D.
      • Moreau C.
      • et al.
      A cellular model to study drug-induced liver injury in nonalcoholic fatty liver disease: application to acetaminophen.
      Of them, drugs known to induce cholestasis tended to reduce CLF intensity, suggesting that significant bile transport inhibition effects can be modeled in organoid imaging analysis (Supplementary Table 3 and Figure 5D). Together, our LoT system developed here is potentially available for the early drug discovery process due to its throughput and relevance to human data.

       Revisiting Mechanistic Classification of Drug-Induced Liver Injury Compounds by Liver Organoid-Based High-Throughput Toxicity Screening System

      We next set out to establish an in-depth investigation model with the use of organoids. To monitor mitochondrial membrane potential (MMP) mitochondrial health alongside with bile transport, we multiplexed the readouts in the intact organoids. After defining an optimal dose for mechanistic study of 10 United States Food and Drug Administration-approved drugs that does not affect viability (Supplementary Results and Supplementary Figure 6), we quantified the mitochondrial health assessment by MMP. After treatment with training compounds (TCs) for 24 hours, dose-dependent increases in MMP were observed with treatment with tolcapone (2- to 8-fold change, P < .01), diclofenac (7- to 13-fold change, P < .05 or .01), cyclosporine A (3- to 7-fold change, P < .01), and nefazodone (4- to 42-fold change, P < .01) (Figure 6A–C). In addition, troglitazone also increased MMP in HLO (3- to 5-fold change, P < .05), although dose-dependence was not observed. On the other hand, after treatment with bosentan, entacapone, and pioglitazone, increases in MMP were not clearly observed even in multiple doses.
      Figure thumbnail gr6
      Figure 6Mechanistic toxicologic approach using DILI HLO model. (A) Upper: Image of FD transport inhibition after treatment with 9 TCs for 5 minutes. Bottom: Image of MMP on tetramethylrhodamine after treatment with 9 TCs. (B) Quantification of transport inhibition after treatment with TCs. The bars represent the mean ± SD (n = 4–6). ∗P < .05, ∗∗P < .01. (C) Quantification of MMP change after treatment with TCs. The bars represent the mean ± SD (n = 4–5). ∗P < .05, ∗∗P < .01. B and C: bosentan; CON, control sample; CSA, cyclosporine A; DICLO, diclofenac; ENTA, entacapone; NEFA, nefazodone; PIO, pioglitazone; STP, streptomycin; TOL, tolcapone; TRO, troglitazone. (D) Analysis between viability for 72 hours after treatment with drugs and dual-risk parameters, drug-induced cholestasis potential, and mitochondria toxicity potential. Cholestasis and mitochondria toxicity (Mito-tox) indexes were derived from data in B and C. The size of the circles indicates the magnitude of viability decreases. (E) Overview of evaluation of drug-induced cytotoxicity on vulnerable organoid model. Profiling of vulnerable model on lipid accumulation (blue, nuclei; green, lipid; red. F-actin.) (F) ROS production (blue, nuclei; green, ROS) and mitochondria detection (blue, nuclei; red, mitochondria). Image of organoids at 24 hours after drugs treatment. (G) Viability assessment on lipid accumulation-induced vulnerable organoid model. The bars represent the mean ± SD (n = 5–6). ∗P < .05. CON, control; PIO, pioglitazone, STP: streptomycin; TRO, troglitazone.
      Severe manifestations of human DILI are multifactorial and are highly associated with combinations of known mechanisms of DILI such as mitochondrial and BSEP inhibition.
      • Aleo M.D.
      • Luo Y.
      • Swiss R.
      • et al.
      Human drug-induced liver injury severity is highly associated with dual inhibition of liver mitochondrial function and bile salt export pump.
      We further analyzed the relationship among survival, cholestasis, and mitochondrial stress. To benchmark our toxicity assay against the conventional in vitro assay systems, we chose the TCs and their concentrations based on published reports focused on cholestatic toxicity and mitochondrial stress using primary hepatocyte systems.
      • Oorts M.
      • Baze A.
      • Bachellier P.
      • et al.
      Drug-induced cholestasis risk assessment in sandwich-cultured human hepatocytes.
      ,
      • Bort R.
      • Ponsoda X.
      • Jover R.
      • et al.
      Diclofenac toxicity to hepatocytes: a role for drug metabolism in cell toxicity.
      Of note, drugs with dual potency at 24 hours (cholestasis and mitochondrial stress), such as cyclosporine A, troglitazone, and nefazodone, significantly lowered cell viability at 72 hours relative to tolcapone, diclofenac, and bosentan (Figure 6D and Supplementary Figure 6A).
      These data are comparable to clinical data demonstrating that dual potencies were highly associated with the severity of DILI consistent in previous reports.
      • Aleo M.D.
      • Luo Y.
      • Swiss R.
      • et al.
      Human drug-induced liver injury severity is highly associated with dual inhibition of liver mitochondrial function and bile salt export pump.
      Additionally, we also note that entacapone treatment at 130 μmol/L decreased organoid viability (from 85% at 24 hours to 64% at 72 hours). Entacapone requires extensive binding to plasma proteins, mainly to albumin, to induce DILI.
      • Fisher A.
      • Croft-Baker J.
      • Davis M.
      • et al.
      Entacapone-induced hepatotoxicity and hepatic dysfunction.
      It would be interesting to further investigate the mechanisms of entacapone toxicity beyond cholestasis and mitochondrial health using our LoT system, because toxicity mechanisms are ill defined. Taken together, the LoT system is a useful human model system for the major mechanistic classifications of DILI and is a testing platform for further delineating unknown complex mechanisms.
      DILI incidence is known to be often confounded by a number of host factors. For instance, growing evidence suggests that obesity and nonalcoholic fatty liver disease greatly increase the risk of hepatotoxicity in rodents and humans when combined with certain drugs such as acetaminophen.
      • Fromenty B.
      Drug-induced liver injury in obesity.
      ,
      • Michaut A.
      • Le Guillou D.
      • Moreau C.
      • et al.
      A cellular model to study drug-induced liver injury in nonalcoholic fatty liver disease: application to acetaminophen.
      Therefore, it seems important to foresee DILI potential in such a “vulnerable” condition with a patient even in the subclinical phase.
      Here, we established a lipotoxic organoid model by exposure to an unsaturated fatty acid, oleic acid (Figure 6E). At 3 days after oleic acid treatment, lipid accumulation in HLOs was intense (Figure 6E). The oxidation of fatty acids is an important source of reactive oxygen species (ROS), which leads to depletion of adenosine 5ʹ-triphosphate and nicotinamide dinucleotide and induces DNA damage in fatty livers.
      • Browning J.D.
      • Horton J.D.
      Molecular mediators of hepatic steatosis and liver injury.
      Consistent with this, ROS production was enhanced in lipid-treated HLOs (Figure 6F and Supplementary Figure 7A). Additionally, fatty acids induced massive swelling of liver mitochondria (Figure 6F and Supplementary Figure 7B), similar to a published phenotype.
      • Zborowski J.
      • Wojtczak L.
      Induction of swelling of liver mitochondria by fatty acids of various chain length.
      Because this lipotoxic organoid model is a vulnerable condition with increased ROS, HLOs were treated with 2 thiazolidinediones, troglitazone (0–50 μmol/L) and pioglitazone (200 μmol/L) for 24 hours, and cell viability was assessed. We observed massive fragmentation of organoids due to cell death after treatment with troglitazone during lipotoxic conditions (Supplementary Figure 8), confirmed by subsequent cell viability analysis (−40% cell viability compared with control, P < .05) (Figure 6G). Because treatment with pioglitazone alone did not affect viability in HLO with/without the lipotoxic condition, lipotoxic HLO possibly highlights DILI vulnerability with positive and negative predictive power (Figure 6G).
      Next, we investigated whether HLOs can recover from DILI-like conditions using a compound with therapeutic potential. As a proof of concept, we used N-acetylcysteine (NAC), an antioxidant, to inhibit ROS production based on literature reports that intravenous NAC improves survival in patients with nonacetaminophen-related acute liver failure
      • Lee W.M.
      • Hynan L.S.
      • Rossaro L.
      • et al.
      Intravenous N-acetylcysteine improves transplant-free survival in early stage non-acetaminophen acute liver failure.
      and diminished the troglitazone-induced cytotoxicity.
      • Rachek L.I.
      • Yuzefovych L.V.
      • Ledoux S.P.
      • et al.
      Troglitazone, but not rosiglitazone, damages mitochondrial DNA and induces mitochondrial dysfunction and cell death in human hepatocytes.
      As expected, cell viability was significantly improved by NAC, suggesting that NAC rescued cell death in HLOs even in vulnerable conditions (Figure 6G and Supplementary Figure 8). This LoT system potentially serves as a preclinical tool in identifying an effective compound in alleviating the DILI-like condition under multidrug regimens.

       Bosentan-Induced Cholestasis Specific to CYP2C9∗2-Induced Pluripotent Stem Cell Liver Organoids

      Because the drug-induced events are highly variable, an iPSC-based organoid approach is promising for the potential assessment of individual susceptibility.
      • Inoue H.
      • Nagata N.
      • Kurokawa H.
      • et al.
      iPS cells: a game changer for future medicine.
      To determine the clinical relevance of our system, we used pharmacogenomic insights that affect drug-induced bile transport inhibition potential. We genotyped 8 different iPSC lines with/without the well-known susceptibility gene variant (ie, CYP2C9∗2 activity intermediate) to the bosentan-induced DILI and compared the HLO lines’ cholestatic potential. Interestingly, CLF excretion into HLOs was severely impaired (positive rate: CC, 17.1% of positive; CT, 70.8% of positive) by bosentan in CY2C9∗2 carrier HLOs but not in 3 different iPSC-derived HLOs without CYP2C9∗2 (Figure 7A–E) . These results indicate that the organoid-based cholestasis assay both negatively and positively predicts CYP2C9-mediated variation for drug-induced cholestasis as seen in humans.
      • Markova S.M.
      • De Marco T.
      • Bendjilali N.
      • et al.
      Association of CYP2C9∗2 with bosentan-induced liver injury.
      These results indicate that the organoid-based cholestasis assay predicts some aspects of human genetic variation for drug-induced cholestasis.
      Figure thumbnail gr7
      Figure 7Bosentan-induced cholestasis is specific to CYP2C9∗2 HLO. (A) The table indicates the possession of enzyme activity normal(C/C)/intermediate (C/T) alleles, CYP2C9∗2 to bosentan-induced liver injury in iPSC lines, and the ratio of CLF efflux difference in each group. (B) Images of CLF transport activity and inhibition by bosentan in HLOs derived from donor with enzyme intermediate allele. Scale bar = 100 μm. (C) Images of CLF transport activity and inhibition by bosentan in HLO derived from donor with enzyme active allele. Scale bar = 100 μm). (D) CLF intensity levels in HLOs derived from iPSC lines. ∗∗P < .0001, t test; ns, not significant. The top and bottom of the box plot represent the 75th and 25th percentiles, the center line represents the median, and the dots indicate the data from each organoid. (E) CLF intensity levels in organoids from 8 iPSC lines. The data are shown as the Z score in each sample.

      Discussion

      Human adult stem cell-derived liver organoids
      • Huch M.
      • Gehart H.
      • van Boxtel R.
      • et al.
      Long-term culture of genome-stable bipotent stem cells from adult human liver.
      ,
      • Hu H.
      • Gehart H.
      • Artegiani B.
      • et al.
      Long-term expansion of functional mouse and human hepatocytes as 3D organoids.
      restore mature hepatocyte profiles that can repopulate in vivo injured liver. The morphology of HLOs with lumen and bile canaliculi-like structure within cells share dissimilarity and similarity to primary organoids. For example, iPSC-HLO exhibits hollow-like structure with large lumen, whereas primary tissue contains cell chords with chicken wire appearance of canaliculi. Nevertheless, both systems contain the canalicular lumen consisting of a space formed by a modified apical region of the opposing plasma membranes of contiguous hepatocytes
      • MacParland S.A.
      • Liu J.C.
      • Ma X.Z.
      • et al.
      Single cell RNA sequencing of human liver reveals distinct intrahepatic macrophage populations.
      ,
      • Tsukada N.
      • Ackerley C.A.
      • Phillips M.J.
      The structure and organization of the bile canalicular cytoskeleton with special reference to actin and actin-binding proteins.
      and distribution of tight junction complexes and the microvilli located on the inside of the lumen. Gene editing and pharmcoinhibitor assays confirm internalization of the bile acid analogs into HLO uses BSEP, which is a known transporter for bile acid excretion. Given that half of our hepatocyte-like cells in HLO display an immature status by scRNA-seq, PSC-derived HLOs exhibited relatively immature transcriptome signatures at bulk compared with adult liver tissue-derived organoid-based approaches, similar to other published studies.
      • Zachos N.C.
      • Kovbasnjuk O.
      • Foulke-Abel J.
      • et al.
      Human enteroids/colonoids and intestinal organoids functionally recapitulate normal intestinal physiology and pathophysiology.
      ,
      • Dedhia P.H.
      • Bertaux-Skeirik N.
      • Zavros Y.
      • et al.
      Organoid models of human gastrointestinal development and disease.
      Interestingly, the human fetal liver is known to have a functional drug metabolizing enzyme system even in the early gestational period, a unique characteristic that is remarkably specific to humans but not animal species.
      • Rane A.
      • Sjoqvist F.
      Drug metabolism in the human fetus and newborn infant.
      Additionally, our scRNA-seq–identified subsets of the populations in HLO represents zone 1 (periportal region in the liver) and zone 3 (centrilobular region in the liver) hepatocyte-like cells. Differential signatures between immature vs mature zonal hepatocytes in HLO warrant future investigations to pinpoint signaling logics to continue the maturation of hepatocyte-like cells in HLO.
      The major advantages of the LoT assay include the use of patient’s iPSC, the storable feature at foregut stage, assay throughput, and the multiplexed readouts for analyzing interplay between other factors such as mitochondrial stress. As mentioned above, retrospective studies revealed that multiple cell stress potentials were associated with the incidence of DILI
      • Aleo M.D.
      • Luo Y.
      • Swiss R.
      • et al.
      Human drug-induced liver injury severity is highly associated with dual inhibition of liver mitochondrial function and bile salt export pump.
      and that the LoT assay showed comparable results with that study, because cell viability was decreased dependent on dual readouts: mitochondrial and cholestatic stress.
      Oxidative stress plays an important role in cell death and has been linked to the development of cholestatic liver injury.
      • Serviddio G.
      • Pereda J.
      • Pallardo F.V.
      • et al.
      Ursodeoxycholic acid protects against secondary biliary cirrhosis in rats by preventing mitochondrial oxidative stress.
      Hydrophobic bile acids accumulate intracellularly during cholestasis and interfere with normal mitochondrial electron transport, inhibiting the activity of respiratory complexes I and III and consequently reducing adenosine 5ʹ-triphosphate synthesis,
      • Krahenbuhl S.
      • Talos C.
      • Fischer S.
      • et al.
      Toxicity of bile acids on the electron transport chain of isolated rat liver mitochondria.
      resulting in mitochondrial dysfunction–induced apoptosis.
      • Bernardi P.
      The permeability transition pore. Control points of a cyclosporin A-sensitive mitochondrial channel involved in cell death.
      In line with these findings, our correlational analysis of these dual readouts indicated cholestatic stress was the more dominating factor for liver injury compared with mitochondria stress. Notably, different susceptibilities to bosentan that were associated with gene variants were recapitulated by our LoT system using multidonor iPSC-derived HLOs. Owing to difficulty and inefficiency in primary liver tissue sampling,
      • Huch M.
      • Gehart H.
      • van Boxtel R.
      • et al.
      Long-term culture of genome-stable bipotent stem cells from adult human liver.
      particularly from healthy individuals without diseases, iPSC-derived HLOs generated from healthy donors will be a valuable tool for precision studies, as was preliminarily demonstrated here with CYP2C9∗2 carrier identification. More importantly, the recently evolving large-scale iPSC bank with gene-sequencing data will enable an imaginable cohort coupled with genomic stratification.
      • Koido M.
      • Kawakami E.
      • Fukumura J.
      • et al.
      Polygenic architecture informs potential vulnerability to drug-induced liver injury.
      LoT might serve as an exciting strategy for the pharmaceutical industry by providing essential insights to minimize the potential for DILI.
      LoT system is currently an oversimplified model that lacks adaptive immune components. For this reason, to fully predict other types of DILIs, including idiosyncratic DILI and immune-mediated DILI observed in patients, additional modification is crucial. In addition, the present study was conducted at a single institution. To facilitate the application into preclinical toxicology analysis, the 384-well–based LoT screening assay needs to be validated in multiple independent institutions with blinded compound testing to further enhance reproducibility and credibility. Although the LoT assay showed CYP2C9 genetic variation-dependent cholestasis exacerbation within >8 donors, this interpretation cannot exclude the possibility of donor-dependent differences rather than genotype-dependence. To formerly prove the causative relationship, an isogenic base editing approach in iPSC will be highly informative.

      Acknowledgments

      The authors would like to express their sincere gratitude to Drs Jim Wells, Jorge Bezerra, Aaron Zorn, and their laboratory members for their support and excellent technical assistance, and Asuka Kodaka for graphical abstract illustration and Mary Koch for kind administrative/technical assistance.

      CRediT Authorship Contributions

      Takanori Takebe, MD, PhD (Conceptualization: Lead; Data curation: Lead; Formal analysis: Supporting; Funding acquisition: Lead; Project administration: Lead; Supervision: Lead; Writing – review & editing: Lead); Tadahiro Shinozawa, PhD (Conceptualization: Equal; Formal analysis: Lead; Investigation: Lead; Visualization: Equal; Writing – original draft: Lead); Masaki Kimura, PhD (Investigation: Equal; Visualization: Equal); Yuqi Cai, PhD (Investigation: Equal; Methodology: Equal); Norikazu Saiki, PhD (Formal analysis: Supporting; Investigation: Supporting); Yosuke Yoneyama, PhD (Formal analysis: Supporting; Investigation: Supporting); Rie Ouchi, PhD (Investigation: Supporting; Methodology: Supporting); Hiroyuki Koike, PhD (Investigation: Supporting); Mari Maezawa, MS (Investigation: Supporting; Methodology: Supporting); Ranran Zhang, PhD (Investigation: Supporting); Andrew Dunn, PhD (Formal analysis: Supporting); Autumn Ferguson, MS (Investigation: Supporting; Project administration: Supporting); Shodai Togo, PhD (Investigation: Supporting); Kyle Lewis, PhD (Investigation: Supporting); Wendy Thompson, PhD (Investigation: Supporting; Project administration: Supporting); Akihiro Asai, MD, PhD (Investigation: Supporting; Supervision: Supporting).

      Supplementary Methods

       H&E Staining, Immunohistochemistry, Immunofluorescent Staining

      Liver organoids were collected from Matrigel, fixed in 4% paraformaldehyde, and embedded in paraffin or optimum cutting temperature compound for cryosectioning. Briefly, for immunohistochemistry, formalin-fixed, paraffin-embedded tissue sections were deparaffinized and hydrated. For cryosections, organoids were sectioned at 10 μm and mounted on the slides. Sections underwent H&E and immunohistochemical staining or immunofluorescent staining. The following primary antibodies were used: anti-albumin antibody (1:100 dilution, Sigma-Aldrich), anti-type IV collagen antibody (1:200 dilution; eBioscience, San Diego, CA), anti-zonula occludens 1 (ZO-1) antibody (1:200 dilution; BD Transduction Laboratories, San Jose, CA), anti-MRP2 antibody (1:200 dilution; Novus Biologicals, Littleton, CO), and anti-MDR3 antibody (1:50 dilution; clone P3 II-26, Sigma-Aldrich).
      Add primary antibody for overnight at 4°C and then add biotinylated secondary antibody for 1 hour in room temperature. After the reaction, wash 3 times in Tris-buffered saline with 0.1% Tween 20 for 5 minutes. Drain slides and carefully wipe off excess buffer and incubate slide with streptavidin-horseradish peroxidase for 30 minutes. Add 3,3′-diaminobenzidine tetra hydrochloride chromogen solution and incubate for 3minutes. Rinse with distilled water.
      Slides were mounted after staining with hematoxylin. The specimens were observed under bright-field. For whole-mount fluorescent immunohistochemical staining, liver organoids were fixed in 4% paraformaldehyde overnight at 4°C and then incubated for 1 hour in blocking buffer (phosphate-buffered saline [PBS] containing 1% Triton-X 100, 10% fetal calf serum) at room temperature. After the blocking, the organoids were washed 3 times with blocking buffer and incubated with primary antibody diluted with Can GetSignal (TOYOBO, Osaka, Japan) overnight on gentle rotation at 4°C.
      After the reaction, organoids were washed 3 times with blocking buffer and incubated overnight with fluorescent dye-conjugated secondary antibodies on gentle rotation at 4°C. Dilutions for primary and secondary antibodies were: anti-ABCB11 (BSEP 1:200; Atlas Antibodies), anti–E-cadherin (1:200; R&D), anti–F-actin (1/200; Abcam), anti-MDR3 (1/100; Sigma-Aldrich), anti–ZO-1(1:200 dilution; BD Transduction Laboratories), anti-CYP7A1 (1/00; Thermo Fisher scientific), Alexa Fluor 568-conjugated donkey anti-rabbit immunoglobulin G (1:200; Life Technologies), and Alexa Fluor 488-conjugated goat anti-mouse immunoglobulin G (1:200; Life Technologies).
      After the secondary reaction, the organoids were washed 3 times with washing buffer (PBS containing 1% Triton-X 100). Nuclei were stained with NucBlue Live ReadyProbes Reagent (Invitrogen) according to the manufacturer’s instructions, and organoids were washed again 3 times with washing buffer. The specimens were observed under a confocal imaging performed on a Nikon A1Rsi inverted confocal microscope.

       RNA Isolation, Reverse Transcription Quantitative Polymerase Chain Reaction

      RNA was isolated using the RNeasy mini kit (Qiagen, Hilden, Germany). Reverse transcription was performed using the SuperScriptIII First-Strand Synthesis System for reverse transcription PCR (Invitrogen) according to the manufacturer’s protocol. qPCR was performed using TaqMan gene expression master mix (Applied Biosystems) on a QuantStudio 3 Real-Time PCR System (Thermo). All primers and probes information for each target gene were obtained from the Universal ProbeLibrary Assay Design Center (https://qpcr.probefinder.com/organism.jsp).

       Single-Cell RNA Sequencing and Data Analysis

      Liver organoids generated from 5 different donor-derived iPSC lines (TkDA, CW100150, CW10077, CW10027, and 1231A3) were dissociated to single cells by the treatment with trypsin-EDTA (0.05%), phenol red (Gibco) for 10 minutes and washed by 1× PBS at day 30. Then, 17,500 cells each were loaded to a 96-well plate, including beads loaded with adapters containing 1 of 750,000 different barcodes for the scRNA-seq library preps. A total of 5119 cells were recovered for liver organoids. Sequenced reads were processed using the Cell Ranger gene expression pipelines mkfastq and count, starting with demultiplexing and conversion of barcode and read data to fastq files. Raw reads were aligned to the Hg19 genome and filtered, creating gene-barcode matrices. Cell and gene filtering, data integration, scaling data, clustering, and visualization were performed in Seurat v3 toolkit (https://github.com/satijalab/seurat).
      • O’Brien P.J.
      • Irwin W.
      • Diaz D.
      • et al.
      High concordance of drug-induced human hepatotoxicity with in vitro cytotoxicity measured in a novel cell-based model using high content screening.
      Cells that had unique feature counts of <1500 or mitochondrial counts of >50% were removed. The data were normalized using the default global-scaling method.
      Next, we integrated normalized liver organoid scRNA-seq data and 5 independent donor-derived primary human liver scRNA-seq data available in GSE115469
      • Proctor W.R.
      • Foster A.J.
      • Vogt J.
      • et al.
      Utility of spherical human liver microtissues for prediction of clinical drug-induced liver injury.
      and identified anchors as input into downstream analysis. The integrated data set was scaled using a linear transformation. Then, we applied a graph-based clustering to scaled dataset. Visualization was performed by t-distributed stochastic neighbor embedding with determined cluster. The cell type of each cluster was assigned manually using known liver cell transcripts. The raw and processed data of RNA-seq have been deposited in Gene Expression Omnibus (ID GSE141183).

       Protein Secretion Analysis

      For measuring albumin secreted level of organoids, 200 μL of culture supernatant of organoids on Ultra-Low Attachment Multiwell Plate 96-well plates (Corning) was collected and stored at −80°C until use. The supernatant was assayed with Human Albumin ELISA Quantitation Set (Bethyl Laboratories, Inc) according to the manufacturer’s instructions. For measuring total bile acid secreted level in intraluminal organoids, fluid inside organoids was absorbed using microinjection Nanoject II (Drummond Scientific, Broomall, PA), diluted in PBS, and assayed with the Total Bile Acid ELISA Kit (antibodies-online, Inc). To calculate the volume of total bile acid, the molecular weight of cholic acid was used for calculation and comparing with volume in previous report.

       Transmission Electron Microscopy

      For transmission electron microscopy, briefly, organoids were fixed in 3% glutaraldehyde overnight at 4°C, washed in 0.1 mol/L sodium cacodylate buffer, and incubated for 1 hour in 4% osmium tetroxide. They were subsequently washed, dehydrated in ethanol series, and finally embedded in propylene oxide/LX112. Tissue was then sectioned and stained with 2% uranyl acetate, followed by lead citrate. Images were visualized on a Hitachi transmission electron microscope.

       Cholylglycylamido-Fluorescein Assay

      Briefly, organoids were preincubated with a transport buffer (118 mmol/L NaCl, 23.8 mmol/L NaHCO3, 4.83 mmol/L KCl, 0.96 mmol/L KH2PO4, 1.20 mmol/L MgSO4, 12.5 mmol/L HEPES, 5 mmol/L glucose, and 1.53 mmol/L CaCl2, adjusted to pH 7.4) for 30 minutes. Next, organoids were treated by 10 μmol/L fluorescently labeled bile acid (cholylglycylamido-fluorescein; a kind gift from Dr Hofmann) for 1 hour, after then, organoids were washed 3 times with PBS. Images were captured using fluorescent microscopy BZ-X710 (Keyence, Osaka, Japan).

       Evaluation of Bile Transport Inhibition

      Fluorescein diacetate was used for evaluating bile transport activity in organoids. Around day 25, the organoids were rinsed with PBS, and organoids were treated with fluorescein diacetate in the medium. Next, 10 mg/mL fluorescein diacetate (Sigma-Aldrich) in hepatocyte culture medium was added with 20 μmol/L cyclosporine A (CSA; Sigma-Aldrich) for 45 minutes, and images were captured sequentially using fluorescent microscopy BZ-9000 (Keyence).
      For evaluation of bile transport inhibition, 10 mg/mL fluorescein diacetate in hepatocyte culture medium was added after treatment with dimethyl sulfoxide (DMSO), streptomycin as a negative control, tolcapone, diclofenac, bosentan, CSA, troglitazone, nefadozone, entacapone, and pioglitazone, all from Sigma-Aldrich. After 5 minutes’ incubation, the organoids were rinsed 3 times with PBS, and images were captured sequentially using fluorescent microscopy BZ-X710. Analysis was performed by calculating the ratio between the intensities outside and inside the organoids using ImageJ 1.48k software (Wayne Rasband, National Institutes of Health; http://imagej.nih.gov/ij). Changes in brightness or contrast during processing were applied equally across the entire image.

       Genome Editing of Human Induced Pluripotent Stem Cells

      CRISPR-Cas9 was used to induce the nonsense (R1090X, CGA to TGA) mutation in hiPSCs (TkDA3-4). We used a modified pX458 plasmid (Addgene #48138), where a single guide RNA scaffold was modified to improve Cas9 binding,
      • Jang K.J.
      • Otieno M.A.
      • Ronxhi J.
      • et al.
      Reproducing human and cross-species drug toxicities using a liver-chip.
      and SpCas9 was replaced with high-fidelity eSpCas9 (1.1) from another Addgene plasmid #71814.
      • Takebe T.
      • Zhang B.
      • Radisic M.
      Synergistic engineering: organoids meet organs-on-a-chip.
      The iPSCs were passaged as single cells using mTeSR1 and laminin. For gene editing, iPSCs were reverse transfected using TransIT-LT1 (Mirus) according to the manufacturer’s recommended protocol. Briefly, 2 μg of the modified pX458 plasmid containing the single guide RNA target sequence and 2 μg of a phosphorothioated single-stranded DNA oligonucleotide with desired nucleotide changes of the sequence were cotransfected into 2 × 106 iPSCs in a single well of a 6-well plate. At 48 hours after transfection, green fluorescent protein-positive cells were isolated by fluorescence-activated cell sorting and replated at cloning density in mTeSR1/laminin. After 1 to 2 weeks, single clones with stereotypical iPSC morphology were manually excised and expanded for genotyping.

       Mitochondria Toxicity Evaluation

      After being cultured in on Ultra-Low Attachment Multiwell Plates, with a 6-well plate in each culture condition, organoids were picked up and seeded in μ-Slide 8 Well Glass Bottom (Ibidi). For evaluation of MMP, 250 nmol/L TMRM (tetramethylrhodamine, methyl ester, perchlorate; Thermo Fisher Scientific) was added after treatment with DMSO (Sigma-Aldrich), streptomycin (Sigma-Aldrich) as a negative control, tolcapone, diclofenac, bosentan, CSA, troglitazone, nefazodone, entacapone, and pioglitazone (all from Sigma-Aldrich) for 24 hours. After 30 minutes’ incubation, the organoids were rinsed 3 times with PBS, and images were scanned on a Nikon A1 Inverted Confocal Microscope (Japan) using 60× water immersion objectives.
      In the image acquired, mitochondria stained were detected in cells along Z axis of HLO. Arias and intensity of TMRM were calculated as MMP by IMARIS8 (Bitplane AG). For assessment of cholestatic and mitochondrial health, cell viability was measured by using the CellTiter-Glo luminescent cell viability assay (Promega, Mannheim, Germany) per organoid at 24 hours after treatment with drugs and confirmed not to decrease the viability in each dose for avoiding secondary change due to cell damage toward death.

       Analysis of Mitochondrial and Cholestatic Stress

      To study the relationship of cell viability with mitochondrial and cholestatic stress, first, using the formula: Index = −(Drug treatment value − control value) × 100, we determined the indexed value provided from mitochondrial and cholestatic stress assays. At 72 hours after treatment with drugs, the adenosine 5ʹ-triphosphate content per organoid was determined using the CellTiter-Glo luminescent cell viability assay (Promega). These data are shown as Figure 6D using Infogr.am (http://infogr.am). a free, web-based tool.

       Evaluation of Viability in Organoids on Vulnerable Condition

      The experiment was performed as shown in Figure 4E. After being excluded from Matrigel and washed, organoids were treated with 800 μmol/L oleic acid on Ultra-Low Attachment Multiwell Plates 6-well plate (Corning) for 3 days. Next, cells were treated with 50 μmol/L troglitazone, with or without 50 μmol/L NAC, for 24 hours. Cell viability was performed by using the CellTiter-Glo luminescent cell viability assay (Promega). Images were captured sequentially using fluorescent microscopy BZ-9000.

       Lipid-Induced Mitochondria Stress Evaluation

      After being cultured in on Ultra-Low Attachment Multiwell Plates 6-well plates in each culture condition, 20 organoids were picked up and seeded in μ-Slide 8 Well Glass Bottom (Ibidi), and live-cell staining was performed. After 3 days of oleic acid exposure, HLOs were stained all together for mitochondria and F-actin by incubating them with TMRM and SiR-Actin for 60 minutes at 37°C. The following regents or kits were used all together in each staining: BODIPY 493/503 for lipids (Thermo Fisher Scientific), SiR Actin Kit for cytoskeleton (USA Scientific), CellROX Green Reagent for ROS (Fisher Scientific), and TMRM (Thermo Fisher Scientific) for mitochondria. NucBlue Live ReadyProbes Reagent (Invitrogen) was added for nuclear visualization right before observation. HLO was scanned on a NikonA1 Inverted Confocal Microscope, and mitochondria size and number per captured spot were automatically quantified by IMARIS8. Mitochondria size and the number were measured by volume assay and spots assay, respectively. Results are presented as mean ± SD (n = 3).

      Supplementary Results

       Revisiting Mechanistic Classification of Drug-Induced Liver Injury Compounds by Large-Scale Organoid-Based Toxicity System

      Because mitochondrial toxicity plays a central role in DILI in multiple mechanisms associated with the onset of DILI,
      • Pessayre D.
      • Fromenty B.
      • Berson A.
      • et al.
      Central role of mitochondria in drug-induced liver injury.
      it is informative to delineate mitochondrial health in HLO with an index of MMP.
      • Li N.
      • Oquendo E.
      • Capaldi R.A.
      • et al.
      A systematic assessment of mitochondrial function identified novel signatures for drug-induced mitochondrial disruption in cells.
      To monitor MMP mitochondrial health alongside with bile transport, we multiplexed the readouts in the intact organoids. Before testing arrays of compounds, we applied MMP into organoids with FD for bile transport activity instead of fluorescently labeled bile acid CLF and cholylglycylamido-fluorescein, because FD enabled parallel acquisition of MMP. HLO incubated with FD exhibited identical phenotype in shorter dynamics (20–45 minutes) that aligned with MMP activation (Figure 4B). We then annotated 10 United States Food and Drug Administration-approved drugs on FD-based assay and MMP assay.
      First, while we found an optimal dose for mechanistic study of 9 compounds with a predying dose, amiodarone was significantly toxic to HLOs within our tested range (Supplementary Figure 6A); therefore, we excluded amiodarone in further potential DILI assessment study to avoid noise due incurred by cell death-derived secondary change. Nine TCs were classified into 3 types based on DILI mechanism: DILI compounds without cholestasis (class A), DILI compounds with cholestasis (class B), and compounds not reported as DILI compounds (class C) (Supplementary Figure 6B).
      • Oorts M.
      • Baze A.
      • Bachellier P.
      • et al.
      Drug-induced cholestasis risk assessment in sandwich-cultured human hepatocytes.
      To quantify the inhibitory potential for FD excretion, we quantified the fluorescent intensity ratio between outside and inside each organoid by ImageJ (Supplementary Figure 5C). To define optimal timing to measure, we tested the FD transport inhibition by using CSA. At 5 minutes after treatment with FD, a significant decrease (0.4 compared with control) was observed in the group treated with CSA for 24 hours compared with control (DMSO) (Supplementary Figure 5C). With the same method, we evaluated 9 TCs at multiple concentrations, and the efflux of FD was significantly decreased (P < .01 or .05) in class B compounds; bosentan, CSA, troglitazone, and nefazodone which are similar to clinical observations, while this inhibitory effect was not observed in class A and C compounds (Figure 6A and B).

      Supplementary Discussion

      Serious adverse events, including liver failure, are major causes of drug attrition during clinical development or withdrawal of marketed pharmaceuticals. In particular, DILI is a critical challenge in drug development, and one major cause is drug-induced cholestasis by inhibition of transporter activity. Currently, the best choice in pharmaceuticals is sandwich culture using human primary hepatocytes to recapitulate liver functions such as bile transport activity in vitro.
      • Ouchi R.
      • Togo S.
      • Kimura M.
      • et al.
      Modeling steatohepatitis in humans with pluripotent stem cell-derived organoids.
      ,
      • Inoue H.
      • Nagata N.
      • Kurokawa H.
      • et al.
      iPS cells: a game changer for future medicine.
      Although a recent report demonstrated the promise of a hepatocyte-based cholestasis model using transdifferentiated cells from human fibroblasts,
      • Ni X.
      • Gao Y.
      • Wu Z.
      • et al.
      Functional human induced hepatocytes (hiHeps) with bile acid synthesis and transport capacities: a novel in vitro cholestatic model.
      these assay platforms still face reproducibility challenges as well as throughput problems because of variable and limited human hepatocyte sources and the need for complex quantitation algorithms.
      In addition, HepaRG cells, a human hepatoma cell line, are also useful to evaluate cholestatic features, but their low BSEP activity and time-consuming differentiation procedure limit their use.
      • Le Vee M.
      • Noel G.
      • Jouan E.
      • et al.
      Polarized expression of drug transporters in differentiated human hepatoma HepaRG cells.
      More importantly, a lack of essential anatomical structures prevents its practical use for the pharmaceutical applications. Alternatively, iPSC-derived hepatocytes or organoids are considered as an alternate model. For example, iPSC-derived hepatocytes showed the different metabolic enzyme activity
      • Takayama K.
      • Morisaki Y.
      • Kuno S.
      • et al.
      Prediction of interindividual differences in hepatic functions and drug sensitivity by using human iPS-derived hepatocytes.
      and were used to evaluate the 37 DILI compounds after coculture with mice fibroblasts,
      • Ware B.R.
      • Berger D.R.
      • Khetani S.R.
      Prediction of drug-induced liver injury in micropatterned co-cultures containing iPSC-derived human hepatocytes.
      whereas hepatobiliary organoid has been reported to exhibit CYP3A4 inducibility
      • Wu F.
      • Wu D.
      • Ren Y.
      • et al.
      Generation of hepatobiliary organoids from human induced pluripotent stem cells.
      and inflammatory phenotype.
      • Ouchi R.
      • Togo S.
      • Kimura M.
      • et al.
      Modeling steatohepatitis in humans with pluripotent stem cell-derived organoids.
      Despite its promise, whether drug-induced cholestasis can be scaled up for future screening using an iPSC-based model remains unclear.
      In the present study, we reported a simple, robust, and high-throughput system to measure bile transport activity by live fluorescent imaging in the presence of testing compounds. Using the LoT system, we found the informed sensitivity and specificity of the compounds was comparable to previous studies.
      • O’Brien P.J.
      • Irwin W.
      • Diaz D.
      • et al.
      High concordance of drug-induced human hepatotoxicity with in vitro cytotoxicity measured in a novel cell-based model using high content screening.
      • Vorrink S.U.
      • Zhou Y.
      • Ingelman-Sundberg M.
      • et al.
      Prediction of drug-induced hepatotoxicity using long-term stable primary hepatic 3D spheroid cultures in chemically defined conditions.
      • Proctor W.R.
      • Foster A.J.
      • Vogt J.
      • et al.
      Utility of spherical human liver microtissues for prediction of clinical drug-induced liver injury.
      • Xu J.J.
      • Henstock P.V.
      • Dunn M.C.
      • et al.
      Cellular imaging predictions of clinical drug-induced liver injury.
      The limitation of in vitro systems includes relatively higher concentration of DILI compounds compared with the in vivo serum concentration. In this study, we set out to focus on the modest concentration as 0 to 100 μmol/L to determine sensitivity and specificity by referring Cmax and gained insights for mechanistic toxicology as seen in previous studies of primary hepatocyte-based systems.
      • Leslie E.M.
      • Watkins P.B.
      • Kim R.B.
      • et al.
      Differential inhibition of rat and human Na+-dependent taurocholate cotransporting polypeptide (NTCP/SLC10A1)by bosentan: a mechanism for species differences in hepatotoxicity.
      ,
      • Tsukada N.
      • Ackerley C.A.
      • Phillips M.J.
      The structure and organization of the bile canalicular cytoskeleton with special reference to actin and actin-binding proteins.
      Our presented organoid technology and its key functional features are highly reproducible across 10 different donors. Given that there are remarkable advances in 3-dimensional culture
      • Proctor W.R.
      • Foster A.J.
      • Vogt J.
      • et al.
      Utility of spherical human liver microtissues for prediction of clinical drug-induced liver injury.
      or organ-on-chip
      • Jang K.J.
      • Otieno M.A.
      • Ronxhi J.
      • et al.
      Reproducing human and cross-species drug toxicities using a liver-chip.
      culture with the use human primary hepatocytes with nonparenchymal cells, it should also be possible to create next-generation organoids-on-chip platform
      • Takebe T.
      • Zhang B.
      • Radisic M.
      Synergistic engineering: organoids meet organs-on-a-chip.
      that capture donor-to-donor variability in DILI.
      Figure thumbnail fx2
      Supplementary Figure 1The image of HLOs for quantification. Images showed uniformity in HLO size and shape.
      Figure thumbnail fx3
      Supplementary Figure 2Development of robust protocol for HLO generation from multiple pluripotent stem cells. (A) HLO differentiation at day14. Images are shown in the condition specified in the panel, respectively. (B) Gene expressions of alpha-fetoprotein, albumin, Retinol Binding Protein 4, and transthyretin in HLO on different combinations of organoid formation medium. (C) Phase contrast image of HLO generated from embryonic stem cells and 7 iPSC lines (iPSC#1: 72.3, iPSC#2: 1383D6, iPSC#3: 1231A3, iPSC#4: CW10027, iPSC#5: CW10077, iPSC#6: CW10150, and iPSC#7: WD91).
      Figure thumbnail fx4
      Supplementary Figure 3Profiling of HLO from multiple PSCs. (A) Hepatocyte/pluripotency gene expression was similar in HLOs generated from multiple PSCs (n = 3). (B) ALB secretion of HLO was comparable with and without freezing (n = 14 for with frozen; n = 15 freeze HLO).
      Figure thumbnail fx5
      Supplementary Figure 4Profiling of HLOs by scRNA-seq. (A) In this t-distributed stochastic neighbor embedding map of 5177 HLO cells, each point indicates a single cell, which are colored by cell type (upper left) and cell cycle phase (lower left). Heat map indicates marker gene expressions categorized for hepatic, mesenchyme, and endothelial cell types in single cells (right). (B) Pathway enrichment analysis on several types of cell population in HLOs and human primary livers. Heat map indicates gene expressions categorized as lipid metabolic process, drug metabolism-CYP450. and cholesterol biosynthetic process.
      Figure thumbnail fx6
      Supplementary Figure 5Visualization of bile transport function with probes. (A) CLF (green) transport activity on HLOs derived from iPSC. (B) Bile acid uptake by organoids after 30 minutes of culture in the presence of the fluorescent bile acid cholylglycylamido-fluorescein (CGamF). (C) Quantification of FD efflux transport into the organoid. The left image shows the quantified ratio of fluorescein intensity between inside and outside the organoid. The right graph indicates the result of the validation study using control (DMSO), CSA, and streptomycin (STP) as negative control. Bars represent the mean ± SD (n = 4). ∗∗P < .01.
      Figure thumbnail fx7
      Supplementary Figure 6Cell viability at 24 hours after treatment with 10 compounds. (A) Viability assessment on lipid accumulation-induced vulnerable organoid model. AMIO, amiodarone; BOS, bosentan; CON, control sample; DICLO, diclofenac; ENTA, entacapone; NEFA, nefazodone; PIO, pioglitazone; STP, streptomycin; TOL, tolcapone; TRO, troglitazone. Bars represent the mean ± SD (n = 4–6). (B) Classification of the set of 9 TCs referred to Oorts et al, 2016.
      • Oorts M.
      • Baze A.
      • Bachellier P.
      • et al.
      Drug-induced cholestasis risk assessment in sandwich-cultured human hepatocytes.
      Class A represents TCs with known reports on DILI in vivo, whereas those in class B are TCs with reports on drug-induced cholestasis in vivo. The mechanism of toxicity based on literature data is also provided. Class C compounds are generally considered safe.
      Figure thumbnail fx8
      Supplementary Figure 7ROS production and the morphologic changes of mitochondria in lipotoxic liver organoids. (A) The ratio of cells producing ROS to total cells on lipid accumulation-induced vulnerable organoid model by treatment with 800 μmol oleic acid (OA). (B) Image of mitochondria in organoid on vulnerable organoid mod (red, mitochondria; purple, F-actin; blue, nucleus. (C) The number and size of mitochondria on the vulnerable organoid model. Bars represent the mean ± SD (n = 5–6). ∗P < .05.
      Figure thumbnail fx9
      Supplementary Figure 8Lipotoxic organoid model. The morphologic changes in models without lipid accumulation (upper image) and the conditions of lipid and drugs treatment (bottom table) are shown.

      Supplementary Material

      References

        • Takebe T.
        • Taniguchi H.
        Human iPSC-derived miniature organs: a tool for drug studies.
        Clin Pharmacol Ther. 2014; 96: 310-313
        • Yang K.
        • Woodhead J.L.
        • Watkins P.B.
        • et al.
        Systems pharmacology modeling predicts delayed presentation and species differences in bile acid-mediated troglitazone hepatotoxicity.
        Clin Pharmacol Ther. 2014; 96: 589-598
        • Leslie E.M.
        • Watkins P.B.
        • Kim R.B.
        • et al.
        Differential inhibition of rat and human Na+-dependent taurocholate cotransporting polypeptide (NTCP/SLC10A1)by bosentan: a mechanism for species differences in hepatotoxicity.
        J Pharmacol Exp Ther. 2007; 321: 1170-1178
        • Ouchi R.
        • Togo S.
        • Kimura M.
        • et al.
        Modeling steatohepatitis in humans with pluripotent stem cell-derived organoids.
        Cell Metab. 2019; 30: 374-384.e6
        • Inoue H.
        • Nagata N.
        • Kurokawa H.
        • et al.
        iPS cells: a game changer for future medicine.
        EMBO J. 2014; 33: 409-417
        • Takebe T.
        • Enomura M.
        • Yoshizawa E.
        • et al.
        Vascularized and complex organ buds from diverse tissues via mesenchymal cell-driven condensation.
        Cell Stem Cell. 2015; 16: 556-565
        • Takebe T.
        • Zhang R.R.
        • Koike H.
        • et al.
        Generation of a vascularized and functional human liver from an iPSC-derived organ bud transplant.
        Nat Protoc. 2014; 9: 396-409
        • Spence J.R.
        • Mayhew C.N.
        • Rankin S.A.
        • et al.
        Directed differentiation of human pluripotent stem cells into intestinal tissue in vitro.
        Nature. 2011; 470: 105-109
        • Zhang R.R.
        • Koido M.
        • Tadokoro T.
        • et al.
        Human iPSC-derived posterior gut progenitors are expandable and capable of forming gut and liver organoids.
        Stem Cell Reports. 2018; 10: 780-793
        • Falasca L.
        • Favale A.
        • Serafino A.
        • et al.
        The effect of retinoic acid on the re-establishment of differentiated hepatocyte phenotype in primary culture.
        Cell Tissue Res. 1998; 293: 337-347
        • MacParland S.A.
        • Liu J.C.
        • Ma X.Z.
        • et al.
        Single cell RNA sequencing of human liver reveals distinct intrahepatic macrophage populations.
        Nat Commun. 2018; 9: 4383
        • Tsukada N.
        • Ackerley C.A.
        • Phillips M.J.
        The structure and organization of the bile canalicular cytoskeleton with special reference to actin and actin-binding proteins.
        Hepatology. 1995; 21: 1106-1113
        • Ni X.
        • Gao Y.
        • Wu Z.
        • et al.
        Functional human induced hepatocytes (hiHeps) with bile acid synthesis and transport capacities: a novel in vitro cholestatic model.
        Sci Rep. 2016; 6: 38694
        • Kock K.
        • Brouwer K.L.
        A perspective on efflux transport proteins in the liver.
        Clin Pharmacol Ther. 2012; 92: 599-612
        • Yumoto A.U.
        • Watanabe S.
        • Hirose M.
        • et al.
        Structural and functional features of bile canaliculi in adult rat hepatocyte spheroids.
        Liver. 1996; 16: 61-66
        • Keitel V.
        • Burdelski M.
        • Vojnisek Z.
        • et al.
        De novo bile salt transporter antibodies as a possible cause of recurrent graft failure after liver transplantation: a novel mechanism of cholestasis.
        Hepatology. 2009; 50: 510-517
        • Mork L.M.
        • Isaksson B.
        • Boran N.
        • et al.
        Comparison of culture media for bile acid transport studies in primary human hepatocytes.
        J Clin Exp Hepatol. 2012; 2: 315-322
        • Khetani S.R.
        • Kanchagar C.
        • Ukairo O.
        • et al.
        Use of micropatterned cocultures to detect compounds that cause drug-induced liver injury in humans.
        Toxicol Sci. 2013; 132: 107-117
        • O’Brien P.J.
        • Irwin W.
        • Diaz D.
        • et al.
        High concordance of drug-induced human hepatotoxicity with in vitro cytotoxicity measured in a novel cell-based model using high content screening.
        Arch Toxicol. 2006; 80: 580-604
        • Vorrink S.U.
        • Zhou Y.
        • Ingelman-Sundberg M.
        • et al.
        Prediction of drug-induced hepatotoxicity using long-term stable primary hepatic 3D spheroid cultures in chemically defined conditions.
        Toxicol Sci. 2018; 163: 655-665
        • Proctor W.R.
        • Foster A.J.
        • Vogt J.
        • et al.
        Utility of spherical human liver microtissues for prediction of clinical drug-induced liver injury.
        Arch Toxicol. 2017; 91: 2849-2863
        • Xu J.J.
        • Henstock P.V.
        • Dunn M.C.
        • et al.
        Cellular imaging predictions of clinical drug-induced liver injury.
        Toxicol Sci. 2008; 105: 97-105
        • Fromenty B.
        Drug-induced liver injury in obesity.
        J Hepatol. 2013; 58: 824-826
        • Michaut A.
        • Le Guillou D.
        • Moreau C.
        • et al.
        A cellular model to study drug-induced liver injury in nonalcoholic fatty liver disease: application to acetaminophen.
        Toxicol Appl Pharmacol. 2016; 292: 40-55
        • Aleo M.D.
        • Luo Y.
        • Swiss R.
        • et al.
        Human drug-induced liver injury severity is highly associated with dual inhibition of liver mitochondrial function and bile salt export pump.
        Hepatology. 2014; 60: 1015-1022
        • Oorts M.
        • Baze A.
        • Bachellier P.
        • et al.
        Drug-induced cholestasis risk assessment in sandwich-cultured human hepatocytes.
        Toxicol In Vitro. 2016; 34: 179-186
        • Bort R.
        • Ponsoda X.
        • Jover R.
        • et al.
        Diclofenac toxicity to hepatocytes: a role for drug metabolism in cell toxicity.
        J Pharmacol Exp Ther. 1999; 288: 65-72
        • Fisher A.
        • Croft-Baker J.
        • Davis M.
        • et al.
        Entacapone-induced hepatotoxicity and hepatic dysfunction.
        Mov Disord. 2002; 17 (discussion 1397–1400): 1362-1365
        • Browning J.D.
        • Horton J.D.
        Molecular mediators of hepatic steatosis and liver injury.
        J Clin Invest. 2004; 114: 147-152
        • Zborowski J.
        • Wojtczak L.
        Induction of swelling of liver mitochondria by fatty acids of various chain length.
        Biochim Biophys Acta. 1963; 70: 596-598
        • Lee W.M.
        • Hynan L.S.
        • Rossaro L.
        • et al.
        Intravenous N-acetylcysteine improves transplant-free survival in early stage non-acetaminophen acute liver failure.
        Gastroenterology. 2009; 137: 856-864, 864e1
        • Rachek L.I.
        • Yuzefovych L.V.
        • Ledoux S.P.
        • et al.
        Troglitazone, but not rosiglitazone, damages mitochondrial DNA and induces mitochondrial dysfunction and cell death in human hepatocytes.
        Toxicol Appl Pharmacol. 2009; 240: 348-354
        • Markova S.M.
        • De Marco T.
        • Bendjilali N.
        • et al.
        Association of CYP2C9∗2 with bosentan-induced liver injury.
        Clin Pharmacol Ther. 2013; 94: 678-686
        • Huch M.
        • Gehart H.
        • van Boxtel R.
        • et al.
        Long-term culture of genome-stable bipotent stem cells from adult human liver.
        Cell. 2015; 160: 299-312
        • Hu H.
        • Gehart H.
        • Artegiani B.
        • et al.
        Long-term expansion of functional mouse and human hepatocytes as 3D organoids.
        Cell. 2018; 175: 1591-1606.e19
        • Zachos N.C.
        • Kovbasnjuk O.
        • Foulke-Abel J.
        • et al.
        Human enteroids/colonoids and intestinal organoids functionally recapitulate normal intestinal physiology and pathophysiology.
        J Biol Chem. 2016; 291: 3759-3766
        • Dedhia P.H.
        • Bertaux-Skeirik N.
        • Zavros Y.
        • et al.
        Organoid models of human gastrointestinal development and disease.
        Gastroenterology. 2016; 150: 1098-1112
        • Rane A.
        • Sjoqvist F.
        Drug metabolism in the human fetus and newborn infant.
        Pediatr Clin North Am. 1972; 19: 37-49
        • Serviddio G.
        • Pereda J.
        • Pallardo F.V.
        • et al.
        Ursodeoxycholic acid protects against secondary biliary cirrhosis in rats by preventing mitochondrial oxidative stress.
        Hepatology. 2004; 39: 711-720
        • Krahenbuhl S.
        • Talos C.
        • Fischer S.
        • et al.
        Toxicity of bile acids on the electron transport chain of isolated rat liver mitochondria.
        Hepatology. 1994; 19: 471-479
        • Bernardi P.
        The permeability transition pore. Control points of a cyclosporin A-sensitive mitochondrial channel involved in cell death.
        Biochim Biophys Acta. 1996; 1275: 5-9
        • Koido M.
        • Kawakami E.
        • Fukumura J.
        • et al.
        Polygenic architecture informs potential vulnerability to drug-induced liver injury.
        Nature Medicine. 2020; 26: 1541-1548