1521-009X/42/9/1401–1406$25.00 http://dx.doi.org/10.1124/dmd.114.059154DRUG METABOLISM AND DISPOSITION Drug Metab Dispos 42:1401–1406, September 2014Copyright ª 2014 by The American Society for Pharmacology and Experimental Therapeutics

Short Communication

Long-Term Chronic Toxicity Testing Using Human Pluripotent StemCell–Derived Hepatocytes

Received May 20, 2014; accepted June 30, 2014

ABSTRACT

Human pluripotent stem cells (hPSC) have the potential to becomeimportant tools for the establishment of new models for in vitrodrug testing of, for example, toxicity and pharmacological effects.Late-stage attrition in the pharmaceutical industry is to a largeextent caused by selection of drug candidates using nonpredic-tive preclinical models that are not clinically relevant. The currenthepatic in vivo and in vitro models show clear limitations, es-pecially for studies of chronic hepatotoxicity. For these reasons,we evaluated the potential of using hPSC-derived hepatocytes forlong-term exposure to toxic drugs. The differentiated hepatocytes

were incubated with hepatotoxic compounds for up to 14 days,using a repeated-dose approach. The hPSC-derived hepatocytesbecame more sensitive to the toxic compounds after extendedexposures and, in addition to conventional cytotoxicity, evidenceof phospholipidosis and steatosis was also observed in the cells.This is, to the best of our knowledge, the first report of a long-term toxicity study using hPSC-derived hepatocytes, and theobservations support further development and validation of hPSC-based toxicity models for evaluating novel drugs, chemicals, andcosmetics.

Introduction

The liver is one of the most affected organs involved in drug-relatedtoxicity. More than 30% of the terminations of drug developmentprograms are caused by adverse effects in the liver, and liver toxicityrepresents almost 20% of the safety-related, postmarket withdrawals(Ballet, 1997).The pharmaceutical industry currently lacks relevant in vitro

models that can predict human liver toxicity early in the drugdevelopment process. The gold standard of in vitro models forhepatotoxicity is primary human hepatocyte (PHH) cultures, whichis considered the most relevant model for human liver toxicologyapplications currently available (Guillouzo and Guguen-Guillouzo,2008). However, the low availability, large donor variation, and therapid loss of metabolic activity following in vitro culture limit theirusefulness, in particular for detecting chronic toxicity (Guguen-Guillouzo and Guillouzo, 2010). One alternative to the PHH thathas emerged during recent years is the human hepatocellular car-cinoma cell line HepaRG, a bipotent progenitor cell line with theability to differentiate into hepatocyte and biliary lineages (Griponet al., 2002). The differentiated hepatocyte-like cells retain manyof the functions of PHH, and the potential to use HepaRG in

metabolism and toxicity studies has been reported (Aninat et al.,2006; Kanebratt and Andersson, 2008; Antherieu et al., 2010;Andersson et al., 2012). However, further evaluation of HepaRG asa toxicity model is needed, because a direct comparison betweenHepaRG, PHH, and human liver hepatocellular carcinoma cell line(HepG2) indicated a low sensitivity for detection of hepatotoxicdrugs in HepaRG (Gerets et al., 2012). Another extensively usedhepatic cell model is the HepG2 tumor cell line (Sassa et al., 1987).Despite the frequent usage of these cells, the resemblance to PHH islow and the fact that these cells proliferate in culture makes themimpractical for long-term studies (Xu et al., 2004; Gerets et al., 2009;Gerets et al., 2012).Other novel in vitro models have been evaluated for their usefulness

in studies of long-term drug exposure. Micro-patterned cocultures ofPHH and murine stromal fibroblasts displayed a functional stability toenable a 9-day repeated-dose exposure with 65.7% sensitivity to detectdrug-induced liver injury (Khetani et al., 2013). There are also reportsof a canine coculture model of hepatocytes and nonparenchymal stromalcells with metabolic capacities maintained for up to 2 weeks and asuperiority for the classification of drug-induced liver injury com-pounds compared with HepG2 and PHH during 5-day compound ex-posure (Atienzar et al., 2014). However, prediction of liver toxicityusing different cell models needs further investigation because sen-sitivity and specificity vary, a result that could be due to time ofexposure, selection of investigated substances, and the specific cell typeused (Gerets et al., 2012).Human pluripotent stem cells (hPSC) offer a new approach for

generating cells for in vitro models. The ability of unlimited pro-pagation and the potential to differentiate into all cell types in the

This work was supported by SCR&Tox, which is funded by the EuropeanCommission within its seventh framework Programme and Cosmetics Europe;European Cosmetics Association, as part of the SEURAT (toward the replacementof in vivo repeated dose system toxicity) Cluster [Contract HEALTH-F5–2010-266573]; and Systems Biology Research Centre (University of Skövde) undergrants from the Knowledge Foundation [2011/0295].

dx.doi.org/10.1124/dmd.114.059154.

ABBREVIATIONS: Aflatoxin B1, (6aR,9aS)-4-methoxy-2,3,6a,9a-tetrahydrocyclopenta[c]furo[39,29:4,5]furo[2,3-h]chromene-1,11-dione; Amiodarone,2-butyl-3-benzofuranyl-4-[2-(diethylamino)ethoxy]-3,5-diiodophenyl ketone hydrochloride; troglitazone, 5-[[4-[(3,4-dihydro-6-hydroxy-2,5,7,8-tetra-methyl-2H-1-benzopyran-2-yl)methoxy]phenyl]methyl]-2,4-thiazolidinedione; DMSO, dimethyl sulfoxide; EC10, Effective concentration at 10 %mortality; HepaRG, human hepatocellular carcinoma cell line; HepG2, human liver hepatocellular carcinoma cell line; hiPS-HEP, human inducedpluripotent stem cell–derived hepatocyte; hPSC, human pluripotent stem cell; P450, cytochrome P450; PHH, primary human hepatocyte; Ximelagatran,ethyl 2-{[(1R)-1-cyclohexyl-2-[(2S)-2-[({4-[(Z)-N9-hydroxycarbamimidoyl]phenyl}methyl)carbamoyl]azetidin-1-yl]-2-oxoethyl]amino}acetate].

1401

at ASPE

T Journals on June 9, 2020

dmd.aspetjournals.org

Dow

nloaded from

human body make hPSC very attractive (Thomson et al., 1998;Takahashi et al., 2007). Protocols to differentiate human embryonicstem cells and human induced pluripotent stem cells into hepatocyteshave improved immensely in recent years, and now hepatocyte cul-tures with high purity and many characteristics of functional hepato-cytes can be obtained from hPSC (Brolen et al., 2010; Si-Tayeb et al.,2010; Touboul et al., 2010; Yildirimman et al., 2011; Hannan et al.,2013; Ulvestad et al., 2013; Szkolnicka et al., 2014). The fact thatthese cells can be derived in a robust and reproducible fashion froma well-characterized source makes them very attractive as a toxicityassessment model.In this study, we have evaluated the potential of using human

induced pluripotent stem cell–derived hepatocytes to study hepato-toxicity in response to chronic drug exposure. Earlier studies haveshown that in many cases repeated dosing and exposure for 2 weeksare necessary to be able to detect human toxicity of drugs (Olson et al.,2000; Roberts et al., 2014). The hepatocytes used in our study, humaninduced pluripotent stem cell–derived hepatocytes (hiPS-HEP), wereexposed to relevant concentrations of hepatotoxic compounds for 2, 7,and 14 days of repeated dosing. The cell morphology was monitoredduring the entire time period, and the cell viability, as well as theinduction of steatosis and phospholipidosis, was assessed at each timepoint. Importantly, the data show that the hiPS-HEP model is stableenough to enable at least a 2-week study of drug exposure and that itis specific enough to assess hepatotoxic responses on a mechanisticlevel.

Materials and Methods

Cell Culture. hiPS-HEP were obtained from Cellectis AB (Gothenburg,Sweden) and cultured according to the instructions from the manufacturer. Thecells were maintained in 96-well plates, and medium was replaced every secondday during the compound incubation. The human hepatocellular carcinomacell line HepG2 was used as a control cell model. The HepG2 cells were pur-chased from American Type Culture Collection and cultured in Dulbecco’smodified Eagle’s medium (high glucose, GlutaMAX supplement, pyruvate;Gibco/Life Technologies, Stockholm, Sweden) supplemented with 10% heat-inactivated fetal bovine serum (Gibco/Life Technologies), 1% penicillin-streptomycin (Gibco/Life Technologies), and 1% nonessential amino acids (Gibco/Life Technologies).

Materials. Four model compounds known to cause hepatotoxicity were usedin this study. Amiodarone [2-butyl-3-benzofuranyl-4-[2-(diethylamino)ethoxy]-3,5-diiodophenyl ketone hydrochloride], aflatoxin B1 [(6aR,9aS)-4-methoxy-2,3,6a,9a-tetrahydrocyclopenta[c]furo[39,29:4,5]furo[2,3-h]chromene-1,11-dione],and troglitazone [5-[[4-[(3,4-dihydro-6-hydroxy-2,5,7,8-tetramethyl-2H-1-benzopyran-2-yl)methoxy]phenyl]methyl]-2,4-thiazolidinedione] were purchasedfrom Sigma-Aldrich (Stockholm, Sweden). Ximelagatran [ethyl 2-{[(1R)-1-cyclohexyl-2-[(2S)-2-[({4-[(Z)-N9-hydroxycarbamimidoyl]phenyl}methyl)carbamoyl]azetidin-1-yl]-2-oxoethyl]amino}acetate] was provided byAstraZeneca, R&D (Mölndal, Sweden). All compounds were purchased as powderand dissolved in dimethyl sulfoxide (DMSO) to a stock solution concentration250� higher than the top assay concentration (see Table 1 for details oncompounds). EZ4U cell proliferation and cytotoxicity assay was purchased fromBiomedica/Biotech-IgG (Copenhagen, Denmark). The fluorescent probes forthe imaging analysis, Hoechst 33342, HCS LipidTOX green neutral lipids, andHCS LipidTOX red phospholipidosis were acquired from Life Technologies,Invitrogen (Stockholm, Sweden).

Dose-Response Study and Viability Testing. The hiPS-HEP were exposedto various concentrations of the compounds for three different time points asfollows: one single-dose exposure for 2 days and repeated-dose exposures for 7(three repeated doses) and 14 days (six repeated doses), respectively. Eachcompound was half-log diluted in DMSO from a DMSO stock solution andthen further diluted in a two-step procedure: first 25� dilutions in cell culturemedium followed by a 10� dilution in the cell culture plates. The finalconcentration of DMSO was 0.4%. Every sample was assayed in triplicate

wells for each time point. After incubation with the compounds, the cellviability was assessed using the EZ4U cell proliferation and cytotoxicity assayaccording to the instructions from the provider. The absorbance at 492 nm wasmeasured in a FLUOstar Omega plate reader (BMG LABTECH, Ortenberg,Germany), and the samples were blank-corrected and normalized to vehiclecontrol.

Dose-response curves were plotted as percentage of vehicle control usingGraphPad Prism 4 (GraphPad Software Inc., La Jolla, CA), and effective con-centrations at 10 % mortality (EC10) were generated using nonlinear sigmoidaldose-response regression. Because the values were normalized to the vehiclecontrol, constraints of TOP = 100 and BOTTOM = 0 were included in theregression setup.

Assessment of Steatosis and Phospholipidosis. To assess the induction ofsteatosis and phospholipidosis following compound exposure, cells stainedwith fluorescent probes for neutral lipids and phospholipids were measuredusing quantitative image analysis. To avoid nonspecific general cytotoxic ef-fects, the concentration used for each compound was the EC10 from the 14-daydose-response study. The following concentrations were used: amiodarone, 2.5mM;aflatoxin B1, 0.1 mM; troglitazone, 40 mM; ximelagatran, 10 mM. At the lastmedium change, 48 hours before the termination of the experiment, the HCSLipidTOX Red Phospholipidosis substrate was added to the cells togetherwith the compounds. At termination of the study, the cell viability was assessedusing the EZ4U kit, as described above, and the cells were fixed in 4%formaldehyde for 30 minutes. After subsequent washing steps with Dulbecco’sphosphate-buffered saline+/+, Hoechst stain, at a final concentration of 10 mg/ml(1000-fold dilution), was added to the wells and the cells were incubated atroom temperature for 20 minutes. Alternatively, to assess the steatosis, theHCS LipidTOX Green Neutral Lipid was diluted 1000-fold in Dulbecco’sphosphate-buffered saline+/+ and added to the cells. The cells were kept atroom temperature for a minimum of 30 minutes prior to storage at +4°C untilfurther analysis.

Automated Imaging and Analysis. Cell cultures in 96-well imaging plates(BD Falcon, Stockholm, Sweden) were imaged using an automated roboticmicroscope (ImageXpress Micro, from Molecular Devices, Sunnyvale, CA)fitted with a 10� Nikon plan fluor objective (NA 0.3). Image acquisition andprocessing were performed using MetaXpress software (version 5.1, fromMolecular Devices). To accommodate the variable thickness of these culturesand to facilitate image analysis, focal stacks were taken for each wavelengthand processed to produce single images containing only in-focus pixels. Cel-lular nuclei were then identified in each image and classified as positive forsteatosis and phospholipidosis based on multiple thresholds for lipid mor-phology, staining intensity, and the area stained for each cell. Data are reportedas percentage of positive cells in which the total population was defined by thenuclear staining.

Statistical Analysis. To test whether the dose-response curves significantlydiffered at each time point, an F-test was performed using GraphPad Prism 4(GraphPad Software Inc.). A P value # 0.05 was considered statisticallysignificant.

Results

hiPS-HEP Characteristics. The drug metabolism and transporterfunctions in the hiPS-HEP have previously been characterized in detailand compared with cryopreserved PHH cultured for 4 and 48 hours(Ulvestad et al., 2013). Notably, the hiPS-HEP show stable gene andprotein expression as well as functional activity of transporters andcytochrome P450 (P450) during in vitro culture. In the present in-vestigation, we focused on prediction of hepatotoxicity using hiPS-HEPin long-term cultures. At the start of the experiments, the hiPS-HEPformed a highly homogeneous (.90%) monolayer of hepatocyteswith a morphology resembling PHH. The cells displayed typicalpolygonal shape, clear cell-cell borders, and prominent light nucleisurrounded by dark cytoplasm (Fig. 1A). The cells expressed severalhepatocyte-associated markers, such as cytokeratin 18 and hepatocytenuclear factor 4a, and showed ability to store glycogen, as illustratedin Fig. 1, B–D. Notably, the hiPS-HEP exhibited some P450 enzyme

1402 Holmgren et al.

at ASPE

T Journals on June 9, 2020

dmd.aspetjournals.org

Dow

nloaded from

activity at levels comparable to that of PHH cultured for 48 hours.Specifically, the hiPS-HEP showed ;30% of CYP1A and more thanfourfold higher CYP3A activity (Ulvestad et al., 2013). The cellsexpressed these characteristics stably for more than 2 weeks, withoutany sign of proliferation, which allowed for the setup of an extensiverepeated dose testing.hiPS-HEP Sensitivity to Hepatotoxic Compounds. As shown in

Fig. 2, the initial dose-response study demonstrated that the hiPS-HEP were sensitive to several of the compounds tested, indicated bythe drop in viability at increased concentration of amiodarone, afla-toxin B1, and troglitazone. For troglitazone, the viability droppedfrom approximately 100% to near 0% within a narrow concentra-tion range. The hiPS-HEP responded strongly to amiodarone andaflatoxin B1 in a time-dependent manner. A significant (P , 0.001)

shift of the viability curve was observed for 7-day (Fig. 2, blue)and 14-day exposure (Fig. 2, black) compared with the 2-day ex-posure (Fig. 2, red). This indicates that the cells are more sensitiveover long-term, repeated-dose exposures compared with short-term single-dose exposures. The hiPS-HEP were nonresponsiveto ximelagatran. Comparable levels of sensitivity were observedfor HepG2 and hiPS-HEP cells when exposed to the compoundsfor 48 hours (Fig. 2, green), whereas it was not possible to useHepG2 for the long-term treatment due to the high extent of cellproliferation.Induction of Steatosis and Phospholipidosis. To evaluate whether

the hiPS-HEP responded to the compounds in a hepatocyte-specificway, we assessed the compounds’ ability to induce steatosis andphospholipidosis. For these experiments, a subcytotoxic concentration

TABLE 1

Study compounds

CompoundTop Assay Concentration

(mM)Mechanisms of Toxicity Cytochrome P450 Metabolism

Amiodarone 125 Mitochondrial toxicity (Kaufmann et al., 2005) CYP1A1, CYP3A4 (Elsherbiny et al., 2008)Aflatoxin B1 140 DNA adductor causing mutations and reactive oxygen

species formation (Do and Choi, 2007)CYP3A4, CYP1A2 (Guengerich et al., 1996)

Troglitazone 125 Idiosyncratic hepatotoxic, impaired mitochondria function(Jaeschke, 2007)

CYP3A (Kassahun et al., 2001)

Ximelagatran 400 Unknown hepatotoxic mechanism, connected to raisedaminotransferase levels (Lee et al., 2005)

None of the major P450s are involved in the metabolism(Bredberg et al., 2003; Clement and Lopian, 2003)

Fig. 1. (A) Light microscopy image of the hepatocyte morphology (scale bar = 100 mm). (B) Periodic acid Schiff staining for glycogen storage (scale bar = 100 mm). (C)Merged image of cytokeratin 18 (green) and 49,69-diamidino-2-phenylindole (blue) (scale bar = 50 mm). (D) Merged image of hepatocyte nuclear factor 4a (red) and 49,69-diamidino-2-phenylindole (blue) (scale bar = 50 mm).

Long-Term Toxicity Testing Using hPSC-Derived Hepatocytes 1403

at ASPE

T Journals on June 9, 2020

dmd.aspetjournals.org

Dow

nloaded from

(EC10 at day 14) of each compound was selected. As shown in Fig.3A, the cell viability at each time point was high (about 90%).Amiodarone induced phopholipidosis in hiPS-HEP (Fig. 3B) after 7-and 14-day exposure. Representative images of the amiodarone-inducedphospholipidosis, in comparison with the control, are shown in Fig. 3D.Steatosis was detected in cells dosed with troglitazone (Fig. 3C) for2 days, and with increasing number of positively stained cells after7- and 14-day exposures.

Discussion

There is a great need in the pharmaceutical industry today for morepredictive cell-based models for toxicity assessment. In this study, wedemonstrate the potential of using hPSC-derived hepatocytes as an invitro model for long-term hepatotoxicity. The hiPS-HEP representa robust and reproducible population of cells with many hepatic char-acteristics that can be maintained for at least 2 weeks in culture, thusenabling a repeated-dose chronic exposure to hepatotoxic compounds.The hepatocytes respond to the compounds in a dose-dependent man-ner and show increasing sensitivity to amiodarone and aflatoxin B1during long-term exposure. The induction of steatosis and phospho-lipidosis can also be assessed to evaluate the hepatocyte-specific toxicresponse after compound exposure. The hiPS-HEP display a clearphospholipidosis when exposed to amiodarone, and the effect in-creases during the exposure time period. Troglitazone induces steatosisin the cells in a similar manner, with increasing number of positive cellswith time.The expression and activity of several important metabolizing

P450 enzymes and transporter proteins in hiPS-HEP have previouslybeen investigated (Ulvestad et al., 2013). Those studies revealeda high activity of CYP3A in the cells, even higher than in PHHcultured for 48 hours, and importantly, the high activity was stableover time. CYP3A is involved in the metabolism of many drugs and

is therefore of great importance in toxicity and metabolism studies(Guengerich, 1999). The high CYP3A expression in the hiPS-HEPcorresponds well to their sensitivity to the hepatotoxic compoundsused in the present study. The hiPS-HEP respond strongly toamiodarone and alfatoxin B1, both of which are highly dependent onCYP3A activity to generate toxic metabolites (Guengerich et al.,1996; Elsherbiny et al., 2008). The viability dose-response curvesdisplay a significant shift toward lower concentrations during the 7-and 14-day exposure. The other two compounds tested in this study,troglitazone and ximelagatran, behave differently. The hiPS-HEP aresensitive to troglitazone at higher concentrations, and the viabilitydrops from nearly 100% to close to 0% from one concentration to thenext, in the interval of 12.5 mM and 125 mM, making the calculationof the EC10 values difficult. Nevertheless, the results indicate that thetoxicity increases with longer exposure times. We also includedximelagatran, a direct thrombin inhibitor that was withdrawn fromthe market because of elevated liver enzymes (alanine aminotrans-ferase levels) during long-term clinical trials (Lee et al., 2005).Ximelagatran has been investigated in several preclinical studieswithout showing an effect on cellular functions, even at concentra-tions much higher than the plasma concentration during therapeuticdosing (Kenne et al., 2008). The hiPS-HEP model, despite the pos-sibility of longer exposure time, does not show a decrease in cellviability at any concentration of ximelagatran tested, which is in linewith results obtained in other hepatic in vitro systems. The mech-anism behind the hepatotoxicity of ximelagatran is still unresolvedand may involve the action of class II human leukocyte antigens(Kindmark et al., 2008).We observe a similar sensitivity for HepG2 cells and hiPS-HEP to the

compounds during the acute exposure (48-hour time point). A longerincubation time was not possible to perform using the HepG2 cells dueto their high extent of cell proliferation. A comparison with previousstudies conducted on PHH indicates a similar ranking of the sensitivity

Fig. 2. Viability curves of hiPS-HEP exposed to compounds for 48 hours (red), 7 days (blue), and 14 days (black), as well as HepG2 exposed to compounds for 48 hours(green). Viability displayed as mean 6 S.E.M. (n = 3).

1404 Holmgren et al.

at ASPE

T Journals on June 9, 2020

dmd.aspetjournals.org

Dow

nloaded from

to the tested compounds. PHH seem most sensitive to aflatoxin B1,followed by amiodarone and troglitazone in decreasing order (Geretset al., 2012; Chatterjee et al., 2014; Szkolnicka et al., 2014). Thesestudies also demonstrate the large donor-to-donor variability commonlyobserved with PHH. In addition, the differences in experimental setupsand viability readouts complicate the direct comparison of independentstudies. Nevertheless, Szkolnicka et al. (2014) reported an extendedcompound exposure study in PHH over a 7-day time period, in whichan increased sensitivity to amiodarone, similar to our results, wasdetected. However, in contrast to our results, they failed to show a toxiceffect for troglitazone during the shorter exposure periods. The generallack of long-term toxicity studies in PHH illustrates the challenges usingthis model for long-term in vitro culturing.The hiPS-HEP used in this study have several advantages over

existing models in that they are generated in a reproducible way and arepossible to cultivate for long periods of time without loss of hepaticphenotype. However, they are also limited by their incomplete ex-pression of some drug-metabolizing enzymes. Intense research isongoing to improve and refine the differentiation process to generatemore mature hPSC-derived hepatocytes. In parallel with these develop-ments and in combination with the possibility of applying prolongeddrug exposure times, hPSC-derived hepatocytes are expected to become

an important alternative to PHH in future drug discovery and toxicitytesting. The pharmaceutical industry will most likely benefit froma predictive cell-based model, with high resemblance to the in vivosituation, which can be used for long-term toxicity assessment. Sucha model can potentially be a complement to existing preclinical models,and ultimately allow earlier testing using human relevant tests, resultingin a reduction of the attritions of candidate drugs entering phase IIand III clinical trials. Another major advantage with the hPSC-basedplatform for derivation of functional hepatocytes is the possibility tochoose hPSC lines originating from different individuals with diversegenotypes and having the opportunity to assess compound effects onspecifically designed panels of hepatocytes derived from a variety ofdonor lines.In conclusion, we have performed a to-date, unique long-term

chronic toxicity study using hPSC-derived hepatocytes. The cellsshow a time-dependent toxic response to amiodarone, aflatoxin B1,and troglitazone. On a mechanistic level, we also find specific in-duction of phospholipidosis and steatosis in the cells. Although themodel needs to be challenged with additional toxic compounds, thisstudy demonstrates the utility of hPSC-derived hepatocytes in long-term toxicity assessment and opens up novel avenues for chronictoxicity testing in vitro.

Fig. 3. Hepatotoxicity assessment. (A) Cell viability of corresponding cells measured by EZ4U. All samples were normalized to vehicle control and are presented as mean6S.E.M. (n = 3). (B) Analysis of phospholipidosis, mean 6 S.E.M. (n = 3). (C) Analysis of steatosis, mean 6 S.E.M. (n = 3). (D) Example images from the high contentanalysis, phospholipidosis (red), steatosis (green), and nuclear stain 49,69-diamidino-2-phenylindole (blue).

Long-Term Toxicity Testing Using hPSC-Derived Hepatocytes 1405

at ASPE

T Journals on June 9, 2020

dmd.aspetjournals.org

Dow

nloaded from

Acknowledgments

The authors thank Jonathan Björklund and Henrik Palmgren (AstraZenecaR&D, Mölndal, Sweden) for compound preparation of Ximelagatran andsupport with the automated imaging.

Systems Biology Research Center,School of Bioscience, University ofSkövde, Skövde, Sweden (G.H., P.S.,J.S.); Department of ClinicalChemistry and TransfusionMedicine, Institute of Biomedicine,University of Gothenburg,Gothenburg, Sweden (G.H.);Department of Discovery Safety,Drug Safety and Metabolism,AstraZeneca R&D, Mölndal,Sweden (A.-K.S.); Section ofPharmacogenetics, Department ofPhysiology and Pharmacology,Karolinska Institutet, Stockholm,Sweden (I.B., T.B.A., M.I.-S.);Department of Bioscience,Cardiovascular and MetabolicDiseases, AstraZeneca R&D,Mölndal, Sweden (A.S.); NovaHepAB, Gothenburg, Sweden (P.B.);Department of Drug Metabolismand Pharmacokinetics, AstraZenecaR&D, Mölndal, Sweden (T.B.A.);and Cellectis AB, Gothenburg,Sweden (P.S., P.B., J.E.)

GUSTAV HOLMGREN

ANNA-KARIN SJÖGRENISABEL BARRAGAN

ALAN SABIRSHPETER SARTIPY

JANE SYNNERGRENPETTER BJÖRQUIST

MAGNUS INGELMAN-SUNDBERGTOMMY B. ANDERSSON

JOSEFINA EDSBAGGE

Authorship ContributionsParticipated in research design: Holmgren, Sjögren, Andersson, Barragan,

Ingelman-Sundberg, Björquist, Edsbagge.Conducted experiments: Holmgren.Contributed new reagents or analytic tools: Sjögren, Sabirsh, Andersson.Performed data analysis: Holmgren, Sjögren, Sabirsh.Wrote or contributed to the writing of the manuscript: Holmgren, Sartipy,

Synnergren, Sjögren, Ingelman-Sundberg, Björquist, Edsbagge, Andersson,Barragan, Sabirsh.

References

Andersson TB, Kanebratt KP, and Kenna JG (2012) The HepaRG cell line: a unique in vitro toolfor understanding drug metabolism and toxicology in human. Expert Opin Drug Metab Toxicol8:909–920.10.1517/17425255.2012.685159.

Aninat C, Piton A, Glaise D, Le Charpentier T, Langouët S, Morel F, Guguen-Guillouzo C,and Guillouzo A (2006) Expression of cytochromes P450, conjugating enzymes and nuclear receptorsin human hepatoma HepaRG cells. Drug Metab Dispos 34:75–83.10.1124/dmd.105.006759.

Anthérieu S, Chesné C, Li R, Camus S, Lahoz A, Picazo L, Turpeinen M, Tolonen A, Uusitalo J,and Guguen-Guillouzo C, et al. (2010) Stable expression, activity, and inducibility of cyto-chromes P450 in differentiated HepaRG cells. Drug Metab Dispos 38:516–525.10.1124/dmd.109.030197.

Atienzar FA, Novik EI, Gerets HH, Parekh A, Delatour C, Cardenas A, MacDonald J, YarmushML, and Dhalluin S (2014) Predictivity of dog co-culture model, primary human hepatocytesand HepG2 cells for the detection of hepatotoxic drugs in humans. Toxicol Appl Pharmacol275:44–61.10.1016/j.taap.2013.11.022.

Ballet F (1997) Hepatotoxicity in drug development: detection, significance and solutions.J Hepatol 26 (Suppl 2):26–36.

Bredberg E, Andersson TB, Frison L, Thuresson A, Johansson S, Eriksson-Lepkowska M,Larsson M, and Eriksson UG (2003) Ximelagatran, an oral direct thrombin inhibitor, has a lowpotential for cytochrome P450-mediated drug-drug interactions. Clin Pharmacokinet 42:765–777.10.2165/00003088-200342080-00005.

Brolén G, Sivertsson L, Björquist P, Eriksson G, Ek M, Semb H, Johansson I, Andersson TB,Ingelman-Sundberg M, and Heins N (2010) Hepatocyte-like cells derived from human em-bryonic stem cells specifically via definitive endoderm and a progenitor stage. J Biotechnol145:284–294.10.1016/j.jbiotec.2009.11.007.

Chatterjee S, Richert L, Augustijns P, and Annaert P (2014) Hepatocyte-based in vitro model forassessment of drug-induced cholestasis. Toxicol Appl Pharmacol 274:124–136.10.1016/j.taap.2013.10.032.

Clement B and Lopian K (2003) Characterization of in vitro biotransformation of new, orally active, directthrombin inhibitor ximelagatran, an amidoxime and ester prodrug. Drug Metab Dispos 31:645–651.

Do J and Choi D-K (2007) Aflatoxins: detection, toxicity, and biosynthesis. Biotechnol Bio-process Eng 12:585–593.10.1007/BF02931073.

Elsherbiny ME, El-Kadi AO, and Brocks DR (2008) The metabolism of amiodarone by variousCYP isoenzymes of human and rat, and the inhibitory influence of ketoconazole. J PharmPharm Sci 11:147–159.

Gerets HH, Hanon E, Cornet M, Dhalluin S, Depelchin O, Canning M, and Atienzar FA (2009)Selection of cytotoxicity markers for the screening of new chemical entities in a pharmaceuticalcontext: a preliminary study using a multiplexing approach. Toxicol In Vitro 23:319–332.10.1016/j.tiv.2008.11.012.

Gerets HH, Tilmant K, Gerin B, Chanteux H, Depelchin BO, Dhalluin S, and Atienzar FA (2012)Characterization of primary human hepatocytes, HepG2 cells, and HepaRG cells at the mRNAlevel and CYP activity in response to inducers and their predictivity for the detection of humanhepatotoxins. Cell Biol Toxicol 28:69–87. 10.1007/s10565-011-9208-4.

Gripon P, Rumin S, Urban S, Le Seyec J, Glaise D, Cannie I, Guyomard C, Lucas J, Trepo C,and Guguen-Guillouzo C (2002) Infection of a human hepatoma cell line by hepatitis B virus.Proc Natl Acad Sci USA 99:15655–15660.10.1073/pnas.232137699.

Guengerich FP (1999) Cytochrome P-450 3A4: regulation and role in drug metabolism. Annu RevPharmacol Toxicol 39:1–17.10.1146/annurev.pharmtox.39.1.1.

Guengerich FP, Johnson WW, Ueng YF, Yamazaki H, and Shimada T (1996) Involvement ofcytochrome P450, glutathione S-transferase, and epoxide hydrolase in the metabolism of aflatoxinB1 and relevance to risk of human liver cancer. Environ Health Perspect 104 (Suppl 3):557–562.

Guguen-Guillouzo C and Guillouzo A (2010) General review on in vitro hepatocyte models andtheir applications. Methods Mol Biol 640:1–40.10.1007/978-1-60761-688-7_1.

Guillouzo A and Guguen-Guillouzo C (2008) Evolving concepts in liver tissue modeling andimplications for in vitro toxicology. Expert Opin Drug Metab Toxicol 4:1279–1294.10.1517/17425255.4.10.1279.

Hannan NR, Segeritz CP, Touboul T, and Vallier L (2013) Production of hepatocyte-like cellsfrom human pluripotent stem cells. Nat Protoc 8:430–437.

Jaeschke H (2007) Troglitazone hepatotoxicity: are we getting closer to understanding idiosyn-cratic liver injury? Toxicol Sci 97:1–3.

Kanebratt KP and Andersson TB (2008) HepaRG cells as an in vitro model for evaluation ofcytochrome P450 induction in humans. Drug Metab Dispos 36:137–145.10.1124/dmd.107.017418.

Kassahun K, Pearson PG, Tang W, McIntosh I, Leung K, Elmore C, Dean D, Wang R, Doss G,and Baillie TA (2001) Studies on the metabolism of troglitazone to reactive intermediates invitro and in vivo: evidence for novel biotransformation pathways involving quinone methideformation and thiazolidinedione ring scission. Chem Res Toxicol 14:62–70.

Kaufmann P, Török M, Hänni A, Roberts P, Gasser R, and Krähenbühl S (2005) Mechanisms ofbenzarone and benzbromarone-induced hepatic toxicity. Hepatology 41:925–935.10.1002/hep.20634.

Kenne K, Skanberg I, Glinghammar B, Berson A, Pessayre D, Flinois JP, Beaune P, Edebert I,Pohl CD, Carlsson S, and Andersson TB (2008) Prediction of drug-induced liver injury inhumans by using in vitro methods: the case of ximelagatran. Toxicol In Vitro 22:730–746.10.1016/j.tiv.2007.11.014

Khetani SR, Kanchagar C, Ukairo O, Krzyzewski S, Moore A, Shi J, Aoyama S, Aleo M,and Will Y (2013) Use of micropatterned cocultures to detect compounds that cause drug-induced liver injury in humans. Toxicol Sci 132:107–117. 10.1093/toxsci/kfs326

Kindmark A, Jawaid A, Harbron CG, Barratt BJ, Bengtsson OF, Andersson TB, Carlsson S,Cederbrant KE, Gibson NJ, and Armstrong M, et al. (2008) Genome-wide pharmacogeneticinvestigation of a hepatic adverse event without clinical signs of immunopathology suggests anunderlying immune pathogenesis. Pharmacogenomics J 8:186–195.10.1038/sj.tpj.6500458.

Lee WM, Larrey D, Olsson R, Lewis JH, Keisu M, Auclert L, and Sheth S (2005) Hepaticfindings in long-term clinical trials of ximelagatran. Drug Saf 28:351–370.

Olson H, Betton G, Robinson D, Thomas K, Monro A, Kolaja G, Lilly P, Sanders J, Sipes G,and Bracken W, et al. (2000) Concordance of the toxicity of pharmaceuticals in humans and inanimals. Regul Toxicol Pharmacol 32:56–67.10.1006/rtph.2000.1399.

Roberts RA, Kavanagh SL, Mellor HR, Pollard CE, Robinson S, and Platz SJ (2014) Reducingattrition in drug development: smart loading preclinical safety assessment. Drug Discov Today19:341-347.10.1016/j.drudis.2013.11.014.

Sassa S, Sugita O, Galbraith RA, and Kappas A (1987) Drug metabolism by the human hepatomacell, Hep G2. Biochem Biophys Res Commun 143:52–57.

Si-Tayeb K, Noto FK, Nagaoka M, Li J, Battle MA, Duris C, North PE, Dalton S, and Duncan SA(2010) Highly efficient generation of human hepatocyte-like cells from induced pluripotentstem cells. Hepatology 51:297–305.10.1002/hep.23354.

Szkolnicka D, Farnworth SL, Lucendo-Villarin B, Storck C, Zhou W, Iredale JP, Flint O, and HayDC (2014) Accurate prediction of drug-induced liver injury using stem cell-derived pop-ulations. Stem Cells Transl Med 3:141–148.10.5966/sctm.2013-0146.

Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda K, and Yamanaka S (2007)Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131:861–872.10.1016/j.cell.2007.11.019.

Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS, and JonesJM (1998) Embryonic stem cell lines derived from human blastocysts. Science 282:1145–1147.

Touboul T, Hannan NR, Corbineau S, Martinez A, Martinet C, Branchereau S, Mainot S, Strick-Marchand H, Pedersen R, and Di Santo J, et al. (2010) Generation of functional hepatocytesfrom human embryonic stem cells under chemically defined conditions that recapitulate liverdevelopment. Hepatology 51:1754–1765.10.1002/hep.23506.

Ulvestad M, Nordell P, Asplund A, Rehnström M, Jacobsson S, Holmgren G, Davidson L, BrolénG, Edsbagge J, and Björquist P, et al. (2013) Drug metabolizing enzyme and transporter proteinprofiles of hepatocytes derived from human embryonic and induced pluripotent stem cells.Biochem Pharmacol 86:691–702.10.1016/j.bcp.2013.06.029.

Xu JJ, Diaz D, and O’Brien PJ (2004) Applications of cytotoxicity assays and pre-lethal mech-anistic assays for assessment of human hepatotoxicity potential. Chem Biol Interact 150:115–128.10.1016/j.cbi.2004.09.011.

Yildirimman R, Brolén G, Vilardell M, Eriksson G, Synnergren J, Gmuender H, Kamburov A,Ingelman-Sundberg M, Castell J, and Lahoz A, et al. (2011) Human embryonic stem cellderived hepatocyte-like cells as a tool for in vitro hazard assessment of chemical carcinoge-nicity. Toxicol Sci 124:278–290. 10.1093/toxsci/kfr225

Address correspondence to: Dr. Gustav Holmgren, Box 408, Kanikegränd 3A,541 28 Skövde, Sweden. E-mail: Gustav.holmgren@his.se

1406 Holmgren et al.

at ASPE

T Journals on June 9, 2020

dmd.aspetjournals.org

Dow

nloaded from