In vitro inhibition of the bile salt export pump correlates with risk of cholestatic drug-induced liver injury in humans

The in silico identification of human bile salt export pump (ABCB11) inhibitors associated with cholestatic drug-induced liver injury

Drug-induced liver injury (DILI) is one of the major causes of drug attrition and failure. Currently, there is increasing evidence that direct inhibition of the human bile salt export pump (BSEP/ABCB11) by drugs and/or metabolites is one of the most important mechanisms of cholestatic DILI. In the present study, we employ two in silico methods, random forest (RF) and the pharmacophore method, to recognize potential BSEP inhibitors that could cause cholestatic DILI, with the aim of mitigating the risk of cholestatic DILI to some extent. The RF model achieved the best prediction performance, producing AUC (area under receiver operating characteristic curve) values of 0.901, 0.929 and 0.996 for leave-one-out cross-validation, the test set and the external test set, respectively, indicating that the built RF model has a satisfactory identification ability. As a complement to the RF model, the pharmacophore model was also built and was proved to be reliable with good predictive performance based on the internal and external validation results. Further analysis indicates that hydrophobicity, molecular size and polarity are important factors that influence the inhibitory activity of BSEP. Furthermore, the two models are applied to screen FDA-approved small molecule drugs, among which the drugs with the potential risk of cholestatic DILI are reported. In conclusion, the RF and pharmacophore models that we present can be considered as integrated screening tools to indicate the potential risk of cholestatic DILI by inhibition of BSEP.

Predicting drug-induced liver injury: The importance of data curation

Drug-induced liver injury (DILI) is a major issue for both patients and pharmaceutical industry due to insufficient means of prevention/prediction. In the current work we present a 2-class classification model for DILI, generated with Random Forest and 2D molecular descriptors on a dataset of 966 compounds. In addition, predicted transporter inhibition profiles were also included into the models. The initially compiled dataset of 1773 compounds was reduced via a 2-step approach to 966 compounds, resulting in a significant increase (p-value < 0.05) in model performance. The models have been validated via 10-fold cross-validation and against three external test sets of 921, 341 and 96 compounds, respectively. The final model showed an accuracy of 64% (AUC 68%) for 10-fold cross-validation (average of 50 iterations) and comparable values for two test sets (AUC 59%, 71% and 66%, respectively). In the study we also examined whether the predictions of our in-house transporter inhibition models for BSEP, BCRP, P-glycoprotein, and OATP1B1 and 1B3 contributed in improvement of the DILI mode. Finally, the model was implemented with open-source 2D RDKit descriptors in order to be provided to the community as a Python script.

Mechanistic diversity in ATP-binding cassette (ABC) transporters

ABC transporters catalyze transport reactions, such as the high-affinity uptake of micronutrients into bacteria and the export of cytotoxic compounds from mammalian cells. Crystal structures of ABC domains and full transporters have provided a framework for formulating reaction mechanisms of ATP-driven substrate transport, but recent studies have suggested remarkable mechanistic diversity within this protein family. This review evaluates the differing mechanistic proposals and outlines future directions for the exploration of ABC-transporter-catalyzed reactions.

Drug-induced liver injury: recent advances in diagnosis and risk assessment

Idiosyncratic drug-induced liver injury (IDILI) is a rare but potentially severe adverse drug reaction that should be considered in patients who develop laboratory criteria for liver injury secondary to the administration of a potentially hepatotoxic drug. Although currently used liver parameters are sensitive in detecting DILI, they are neither specific nor able to predict the patient's subsequent clinical course. Genetic risk assessment is useful mainly due to its high negative predictive value, with several human leucocyte antigen alleles being associated with DILI. New emerging biomarkers which could be useful in assessing DILI include total keratin18 (K18) and caspase-cleaved keratin18 (ccK18), macrophage colony-stimulating factor receptor 1, high mobility group box 1 and microRNA-122. From the numerous in vitro test systems that are available, monocyte-derived hepatocytes generated from patients with DILI show promise in identifying the DILI-causing agent from among a panel of coprescribed drugs. Several computer-based algorithms are available that rely on cumulative scores of known risk factors such as the administered dose or potential liabilities such as mitochondrial toxicity, inhibition of the bile salt export pump or the formation of reactive metabolites. A novel DILI cluster score is being developed which predicts DILI from multiple complimentary cluster and classification models using absorption-distribution-metabolism-elimination-related as well as physicochemical properties, diverse substructural descriptors and known structural liabilities. The provision of more advanced scientific and regulatory guidance for liver safety assessment will depend on validating the new diagnostic markers in the ongoing DILI registries, biobanks and public-private partnerships.

Structure based classification for bile salt export pump (BSEP) inhibitors using comparative structural modeling of human BSEP

The bile salt export pump (BSEP) actively transports conjugated monovalent bile acids from the hepatocytes into the bile. This facilitates the formation of micelles and promotes digestion and absorption of dietary fat. Inhibition of BSEP leads to decreased bile flow and accumulation of cytotoxic bile salts in the liver. A number of compounds have been identified to interact with BSEP, which results in drug-induced cholestasis or liver injury. Therefore, in silico approaches for flagging compounds as potential BSEP inhibitors would be of high value in the early stage of the drug discovery pipeline. Up to now, due to the lack of a high-resolution X-ray structure of BSEP, in silico based identification of BSEP inhibitors focused on ligand-based approaches. In this study, we provide a homology model for BSEP, developed using the corrected mouse P-glycoprotein structure (PDB ID: 4M1M). Subsequently, the model was used for docking-based classification of a set of 1212 compounds (405 BSEP inhibitors, 807 non-inhibitors). Using the scoring function ChemScore, a prediction accuracy of 81% on the training set and 73% on two external test sets could be obtained. In addition, the applicability domain of the models was assessed based on Euclidean distance. Further, analysis of the protein-ligand interaction fingerprints revealed certain functional group-amino acid residue interactions that could play a key role for ligand binding. Though ligand-based models, due to their high speed and accuracy, remain the method of choice for classification of BSEP inhibitors, structure-assisted docking models demonstrate reasonably good prediction accuracies while additionally providing information about putative protein-ligand interactions.

Integrating Drug's Mode of Action into Quantitative Structure-Activity Relationships for Improved Prediction of Drug-Induced Liver Injury

Drug-induced liver injury (DILI) is complex in mechanism. Different drugs could undergo different mechanisms but result in the same DILI type, while the same drug could lead to different DILI types via different mechanisms. Therefore, predicting a drug's potential for DILI should take its underlying mechanisms into consideration. To achieve that, we constructed a novel approach by incorporating the drug's Mode of Action (MOA) into Quantitative Structure-Activity Relationship (QSAR) modeling. This MOA-DILI approach was examined using a data set of 333 drugs. The drugs were first grouped according to their MOA profiles (positive or negative in each MOA) based on the Tox21 qHTS assays. QSAR models for individual MOA assays were developed and subsequently combined to obtain the MOA-DILI model. A hold-out testing strategy (222 drugs for training and 111 drugs as a test set) was employed, which yielded a predictive accuracy of 0.711. The MOA-DILI model was directly compared with the standard QSAR approach using the same hold-out strategy, and the QSAR model yielded an accuracy of 0.662. To minimize the random chance in splitting training/test sets, the hold-out testing process was repeated 1000 times, and the observed difference in prediction accuracy between MOA-DILI and QSARs was statistically significant (P value <0.0001). Out of 17 MOAs used, four assays (i.e., antioxidant response elements, PPAR-gamma, estrogen receptor, and thyroid receptor assays) contributed most to the improved prediction of the MOA-DILI model over QSARs. In conclusion, the MOA-DILI approach has the potential to significantly improve predictive outcomes and to reveal complex relationships between MOAs and DILI, all of which would be helpful in developing DILI predictive models in drug screening and for risk assessment of industrial chemicals.

Transporter-Mediated Disposition, Clinical Pharmacokinetics and Cholestatic Potential of Glyburide and Its Primary Active Metabolites

Glyburide is widely used for the treatment of type 2 diabetes. We studied the mechanisms involved in the disposition of glyburide and its pharmacologically active hydroxy metabolites M1 and M2b and evaluated their clinical pharmacokinetics and the potential role in glyburide-induced cholestasis employing physiologically based pharmacokinetic (PBPK) modeling. Transport studies of parent and metabolites in human hepatocytes and transfected cell systems imply hepatic uptake mediated by organic anion-transporting polypeptides. Metabolites are also subjected to basolateral and biliary efflux by P-glycoprotein, breast cancer resistance protein, and multidrug resistance-associated proteins, and are substrates to renal organic anion transporter 3. A PBPK model in combination with a Bayesian approach was developed considering the identified disposition mechanisms. The model reasonably described plasma concentration time profiles and urinary recoveries of glyburide and the metabolites, implying the role of multiple transport processes in their pharmacokinetics. Predicted free liver concentrations of the parent (∼30-fold) and metabolites (∼4-fold) were higher than their free plasma concentrations. Finally, all three compounds showed bile salt export pump inhibition in vitro; however, significant in vivo inhibition was not apparent for any compound on the basis of a predicted unbound liver exposure-response effect model using measured in vitro IC₅₀ values. In conclusion, this study demonstrates the important role of multiple drug transporters in the disposition of glyburide and its active metabolites, suggesting that variability in the function of these processes may lead to pharmacokinetic variability in the parent and the metabolites, potentially translating to pharmacodynamic variability.

Using quantitative intravital multiphoton microscopy to dissect hepatic transport in rats

Hepatic solute transport is a complex process whose disruption is associated with liver disease and drug-induced liver injury. Intravital multiphoton fluorescence excitation microscopy provides the spatial and temporal resolution necessary to characterize hepatic transport at the level of individual hepatocytes in vivo and thus to identify the mechanisms and cellular consequences of cholestasis. Here we present an overview of the use of fluorescence microscopy for studies of hepatic transport in living animals, and describe how we have combined methods of intravital microscopy and digital image analysis to dissect the effects of drugs and pathological conditions on the function of hepatic transporters in vivo.

Predicting Drug-Induced Cholestasis with the Help of Hepatic Transporters-An in Silico Modeling Approach

Cholestasis represents one out of three types of drug induced liver injury (DILI), which comprises a major challenge in drug development. In this study we applied a two-class classification scheme based on k-nearest neighbors in order to predict cholestasis, using a set of 93 two-dimensional (2D) physicochemical descriptors and predictions of selected hepatic transporters' inhibition (BSEP, BCRP, P-gp, OATP1B1, and OATP1B3). In order to assess the potential contribution of transporter inhibition, we compared whether the inclusion of the transporters' inhibition predictions contributes to a significant increase in model performance in comparison to the plain use of the 93 2D physicochemical descriptors. Our findings were in agreement with literature findings, indicating a contribution not only from BSEP inhibition but a rather synergistic effect deriving from the whole set of transporters. The final optimal model was validated via both 10-fold cross validation and external validation. It performs quite satisfactorily resulting in 0.686 ± 0.013 for accuracy and 0.722 ± 0.014 for area under the receiver operating characteristic curve (AUC) for 10-fold cross-validation (mean ± standard deviation from 50 iterations).

How Open Data Shapes In Silico Transporter Modeling

Chemical compound bioactivity and related data are nowadays easily available from open data sources and the open medicinal chemistry literature for many transmembrane proteins. Computational ligand-based modeling of transporters has therefore experienced a shift from local (quantitative) models to more global, qualitative, predictive models. As the size and heterogeneity of the data set rises, careful data curation becomes even more important. This includes, for example, not only a tailored cutoff setting for the generation of binary classes, but also the proper assessment of the applicability domain. Powerful machine learning algorithms (such as multi-label classification) now allow the simultaneous prediction of multiple related targets. However, the more complex, the less interpretable these models will get. We emphasize that transmembrane transporters are very peculiar, some of which act as off-targets rather than as real drug targets. Thus, careful selection of the right modeling technique is important, as well as cautious interpretation of results. We hope that, as more and more data will become available, we will be able to ameliorate and specify our models, coming closer towards function elucidation and the development of safer medicine.

Identification of novel MRP3 inhibitors based on computational models and validation using an in vitro membrane vesicle assay

INTRODUCTION

Multidrug resistance-associated protein 3 (MRP3), an efflux transporter on the hepatic basolateral membrane, may function as a compensatory mechanism to prevent the accumulation of anionic substrates (e.g., bile acids) in hepatocytes. Inhibition of MRP3 may disrupt bile acid homeostasis and is one hypothesized risk factor for the development of drug-induced liver injury (DILI). Therefore, identifying potential MRP3 inhibitors could help mitigate the occurrence of DILI.

METHODS

Bayesian models were developed using MRP3 transporter inhibition data for 86 structurally diverse drugs. The compounds were split into training and test sets of 57 and 29 compounds, respectively, and six models were generated based on distinct inhibition thresholds and molecular fingerprint methods. The six Bayesian models were validated against the test set and the model with the highest accuracy was utilized for a virtual screen of 1470 FDA-approved drugs from DrugBank. Compounds that were predicted to be inhibitors were selected for in vitro validation. The ability of these compounds to inhibit MRP3 transport at a concentration of 100μM was measured in membrane vesicles derived from stably transfected MRP3-over-expressing HEK-293 cells with [³H]-estradiol-17β-d-glucuronide (E₂17G; 10μM; 5min uptake) as the probe substrate.

RESULTS

A predictive Bayesian model was developed with a sensitivity of 73% and specificity of 71% against the test set used to evaluate the six models. The area under the Receiver Operating Characteristic (ROC) curve was 0.710 against the test set. The final selected model was based on compounds that inhibited substrate transport by at least 50% compared to the negative control, and functional-class fingerprints (FCFP) with a circular diameter of six atoms, in addition to one-dimensional physicochemical properties. The in vitro screening of predicted inhibitors and non-inhibitors resulted in similar model performance with a sensitivity of 64% and specificity of 70%. The strongest inhibitors of MRP3-mediated E₂17G transport were fidaxomicin, suramin, and dronedarone. Kinetic assessment revealed that fidaxomicin was the most potent of these inhibitors (IC₅₀=1.83±0.46μM). Suramin and dronedarone exhibited IC₅₀ values of 3.33±0.41 and 47.44±4.41μM, respectively.

CONCLUSION

Bayesian models are a useful screening approach to identify potential inhibitors of transport proteins. Novel MRP3 inhibitors were identified by virtual screening using the selected Bayesian model, and MRP3 inhibition was confirmed by an in vitro transporter inhibition assay. Information generated using this modeling approach may be valuable in predicting the potential for DILI and/or MRP3-mediated drug-drug interactions.

Human biology-based drug safety evaluation: scientific rationale, current status and future challenges

INTRODUCTION

Animal toxicity studies used to assess the safety of new candidate pharmaceuticals prior to their progression into human clinical trials are unable to assess the risk of non-pharmacologically mediated idiosyncratic adverse drug reactions (ADRs), the most frequent of which are drug-induced liver injury and cardiotoxicity. Idiosyncratic ADRs occur only infrequently and in certain susceptible humans, but are caused by many hundreds of different drugs and may lead to serious illness. Areas covered: Idiosyncratic ADRs are initiated by drug-related chemical insults, which cause toxicity due to susceptibility factors that manifest only in certain patients. The chemical insults can be detected using in vitro assays. These enable useful discrimination between drugs that cause high versus low levels of idiosyncratic ADR concern. Especially promising assays, which have been described recently in peer-reviewed scientific literature, are highlighted. Expert opinion: Effective interpretation of in vitro toxicity data requires integration of endpoints from multiple assays, which each address different mechanisms, and must also take account of human systemic and tissue drug exposure in vivo. Widespread acceptance and use of such assays has been hampered by the lack of correlation between idiosyncratic human ADR risk and toxicities observed in vivo in animals.

Mechanisms of Drug-Induced Hepatotoxicity

Drug-induced hepatotoxicity (DIH) is a significant cause of acute liver failure and liver transplantation. Diagnosis is challenging due to the idiosyncratic nature, its presentation in the form of other liver disease, and the lack of a definite diagnostic criteria. Generation of reactive metabolites, oxidative stress, and mitochondrial dysfunction are common mechanisms involved in DIH. Certain risk factors associated with a drug and within an individual further predispose patients to DIH.

Biliary excretion of pravastatin and taurocholate in rats with bile salt export pump (Bsep) impairment

The bile salt export pump (BSEP) is expressed on the canalicular membrane of hepatocytes regulating liver bile salt excretion, and impairment of BSEP function may lead to cholestasis in humans. This study explored drug biliary excretion, as well as serum chemistry, individual bile acid concentrations and liver transporter expressions, in the SAGE Bsep knockout (KO) rat model. It was observed that the Bsep protein in KO rats was decreased to 15% of that in the wild type (WT), as quantified using LC-MS/MS. While the levels of Ntcp and Mrp2 were not significantly altered, Mrp3 expression increased and Oatp1a1 decreased in KO animals. Compared with the WT rats, the KO rats had similar serum chemistry and showed normal liver transaminases. Although the total plasma bile salts and bile flow were not significantly changed in Bsep KO rats, individual bile acids in plasma and liver demonstrated variable changes, indicating the impact of Bsep KO. Following an intravenous dose of deuterium labeled taurocholic acid (D4-TCA, 2 mg/kg), the D4-TCA plasma exposure was higher and bile excretion was delayed by approximately 0.5 h in the KO rats. No differences were observed for the pravastatin plasma concentration-time profile or the biliary excretion after intravenous administration (1 mg/kg). Collectively, the results revealed that these rats have significantly lower Bsep expression, therefore affecting the biliary excretion of endogenous bile acids and Bsep substrates. However, these rats are able to maintain a relatively normal liver function through the remaining Bsep protein and via the regulation of other transporters. Copyright © 2016 John Wiley & Sons, Ltd.

Idiosyncratic Drug-Induced Liver Injury (IDILI): Potential Mechanisms and Predictive Assays

Idiosyncratic drug-induced liver injury (IDILI) is a significant source of drug recall and acute liver failure (ALF) in the United States. While current drug development processes emphasize general toxicity and drug metabolizing enzyme- (DME-) mediated toxicity, it has been challenging to develop comprehensive models for assessing complete idiosyncratic potential. In this review, we describe the enzymes and proteins that contain polymorphisms believed to contribute to IDILI, including ones that affect phase I and phase II metabolism, antioxidant enzymes, drug transporters, inflammation, and human leukocyte antigen (HLA). We then describe the various assays that have been developed to detect individual reactions focusing on each of the mechanisms described in the background. Finally, we examine current trends in developing comprehensive models for examining these mechanisms. There is an urgent need to develop a panel of multiparametric assays for diagnosing individual toxicity potential.

Sandwich-Cultured Hepatocytes as a Tool to Study Drug Disposition and Drug-Induced Liver Injury

Sandwich-cultured hepatocytes (SCH) are metabolically competent and have proper localization of basolateral and canalicular transporters with functional bile networks. Therefore, this cellular model is a unique tool that can be used to estimate biliary excretion of compounds. SCH have been used widely to assess hepatobiliary disposition of endogenous and exogenous compounds and metabolites. Mechanistic modeling based on SCH data enables estimation of metabolic and transporter-mediated clearances, which can be used to construct physiologically based pharmacokinetic models for prediction of drug disposition and drug-drug interactions in humans. In addition to pharmacokinetic studies, SCH also have been used to study cytotoxicity and perturbation of biological processes by drugs and hepatically generated metabolites. Human SCH can provide mechanistic insights underlying clinical drug-induced liver injury (DILI). In addition, data generated in SCH can be integrated into systems pharmacology models to predict potential DILI in humans. In this review, applications of SCH in studying hepatobiliary drug disposition and bile acid-mediated DILI are discussed. An example is presented to show how data generated in the SCH model were used to establish a quantitative relationship between intracellular bile acids and cytotoxicity, and how this information was incorporated into a systems pharmacology model for DILI prediction.

Design, synthesis, in vitro characterization and preliminary imaging studies on fluorinated bile acid derivatives as PET tracers to study hepatic transporters

With the aim of identifying a fluorinated bile acid derivative that could be used as [¹⁸F]-labeled Positron Emission Tomography (PET) tracer for imaging the in vivo functioning of liver transporter proteins, and particularly of OATP1B1, three fluorinated bile acid triazole derivatives of cholic, deoxycholic and lithocholic acid (CATD, DCATD and LCATD 4a-c, respectively) were synthesized and labeled with tritium. In vitro transport properties were studied with cell-based assays to identify the best substrate for OATP1B1. In addition, the lead compound, LCATD (4c), was tested as a substrate of other liver uptake transporters OATP1B3, NTCP and efflux transporter BSEP to evaluate its specificity of liver transport. The results suggest that 4c is a good substrate of OATP1B1 and NTCP, whereas it is a poor substrate of OATP1B3. The efflux transporter BSEP also appears to be involved in the excretion of 4c from hepatocytes. The automated radiosynthesis of [¹⁸F]-4c was accomplished in a multi-GBq scale and a pilot imaging experiment in a wild type rat was performed after i.v. administration to assess the biodistribution and clearance of the tracer. PET imaging revealed that radioactivity was primarily located in the liver (t_max=75s) and cleared exclusively through the bile, thus allowing to image the hepatobiliary excretion of bile acids in the animal model. These findings suggest that [¹⁸F]-LCATD 4c is a promising PET probe for the evaluation of hepatic transporters OATP1B1, NTCP and BSEP activity with potential for studying drug-drug interactions and drug-induced toxicity involving these transporters.

Functional human induced hepatocytes (hiHeps) with bile acid synthesis and transport capacities: A novel in vitro cholestatic model

Drug-induced cholestasis is a leading cause of drug withdrawal. However, the use of primary human hepatocytes (PHHs), the gold standard for predicting cholestasis in vitro, is limited by their high cost and batch-to-batch variability. Mature hepatocyte characteristics have been observed in human induced hepatocytes (hiHeps) derived from human fibroblast transdifferentiation. Here, we evaluated whether hiHeps could biosynthesize and excrete bile acids (BAs) and their potential as PHH alternatives for cholestasis investigations. Quantitative real-time PCR (qRT-PCR) and western blotting indicated that hiHeps highly expressed BA synthases and functional transporters. Liquid chromatography tandem mass spectrometry (LC-MS/MS) showed that hiHeps produced normal intercellular unconjugated BAs but fewer conjugated BAs than human hepatocytes. When incubated with representative cholestatic agents, hiHeps exhibited sensitive drug-induced bile salt export pump (BSEP) dysfunction, and their response to cholestatic agent-mediated cytotoxicity correlated well with that of PHHs (r² = 0.8032). Deoxycholic acid (DCA)-induced hepatotoxicity in hiHeps was verified by elevated aspartate aminotransferase (AST) and γ-glutamyl-transferase (γ-GT) levels. Mitochondrial damage and cell death suggested DCA-induced toxicity in hiHeps, which were attenuated by hepatoprotective drugs, as in PHHs. For the first time, hiHeps were reported to biosynthesize and excrete BAs, which could facilitate predicting cholestatic hepatotoxicity and screening potential therapeutic drugs against cholestasis.

Hepatic 3D spheroid models for the detection and study of compounds with cholestatic liability

Drug-induced cholestasis (DIC) is poorly understood and its preclinical prediction is mainly limited to assessing the compound's potential to inhibit the bile salt export pump (BSEP). Here, we evaluated two 3D spheroid models, one from primary human hepatocytes (PHH) and one from HepaRG cells, for the detection of compounds with cholestatic liability. By repeatedly co-exposing both models to a set of compounds with different mechanisms of hepatotoxicity and a non-toxic concentrated bile acid (BA) mixture for 8 days we observed a selective synergistic toxicity of compounds known to cause cholestatic or mixed cholestatic/hepatocellular toxicity and the BA mixture compared to exposure to the compounds alone, a phenomenon that was more pronounced after extending the exposure time to 14 days. In contrast, no such synergism was observed after both 8 and 14 days of exposure to the BA mixture for compounds that cause non-cholestatic hepatotoxicity. Mechanisms behind the toxicity of the cholestatic compound chlorpromazine were accurately detected in both spheroid models, including intracellular BA accumulation, inhibition of ABCB11 expression and disruption of the F-actin cytoskeleton. Furthermore, the observed synergistic toxicity of chlorpromazine and BA was associated with increased oxidative stress and modulation of death receptor signalling. Combined, our results demonstrate that the hepatic spheroid models presented here can be used to detect and study compounds with cholestatic liability.

Evidence-based selection of training compounds for use in the mechanism-based integrated prediction of drug-induced liver injury in man

The current test systems employed by pharmaceutical industry are poorly predictive for drug-induced liver injury (DILI). The ‘MIP-DILI’ project addresses this situation by the development of innovative preclinical test systems which are both mechanism-based and of physiological, pharmacological and pathological relevance to DILI in humans. An iterative, tiered approach with respect to test compounds, test systems, bioanalysis and systems analysis is adopted to evaluate existing models and develop new models that can provide validated test systems with respect to the prediction of specific forms of DILI and further elucidation of mechanisms. An essential component of this effort is the choice of compound training set that will be used to inform refinement and/or development of new model systems that allow prediction based on knowledge of mechanisms, in a tiered fashion. In this review, we focus on the selection of MIP-DILI training compounds for mechanism-based evaluation of non-clinical prediction of DILI. The selected compounds address both hepatocellular and cholestatic DILI patterns in man, covering a broad range of pharmacologies and chemistries, and taking into account available data on potential DILI mechanisms (e.g. mitochondrial injury, reactive metabolites, biliary transport inhibition, and immune responses). Known mechanisms by which these compounds are believed to cause liver injury have been described, where many if not all drugs in this review appear to exhibit multiple toxicological mechanisms. Thus, the training compounds selection offered a valuable tool to profile DILI mechanisms and to interrogate existing and novel in vitro systems for the prediction of human DILI.

Key Challenges and Opportunities Associated with the Use of In Vitro Models to Detect Human DILI: Integrated Risk Assessment and Mitigation Plans

Drug-induced liver injury (DILI) is a major cause of late-stage clinical drug attrition, market withdrawal, black-box warnings, and acute liver failure. Consequently, it has been an area of focus for toxicologists and clinicians for several decades. In spite of considerable efforts, limited improvements in DILI prediction have been made and efforts to improve existing preclinical models or develop new test systems remain a high priority. While prediction of intrinsic DILI has improved, identifying compounds with a risk for idiosyncratic DILI (iDILI) remains extremely challenging because of the lack of a clear mechanistic understanding and the multifactorial pathogenesis of idiosyncratic drug reactions. Well-defined clinical diagnostic criteria and risk factors are also missing. This paper summarizes key data interpretation challenges, practical considerations, model limitations, and the need for an integrated risk assessment. As demonstrated through selected initiatives to address other types of toxicities, opportunities exist however for improvement, especially through better concerted efforts at harmonization of current, emerging and novel in vitro systems or through the establishment of strategies for implementation of preclinical DILI models across the pharmaceutical industry. Perspectives on the incorporation of newer technologies and the value of precompetitive consortia to identify useful practices are also discussed.

Lipid flopping in the liver

Bile is synthesized in the liver and is essential for the emulsification of dietary lipids and lipid-soluble vitamins. It is a complex mixture of amphiphilic bile acids (BAs; which act as detergent molecules), the membrane phospholipid phosphatidylcholine (PC), cholesterol and a variety of endogenous metabolites and waste products. Over the last 20 years, the combined effort of clinicians, geneticists, physiologists and biochemists has shown that each of these bile components is transported across the canalicular membrane of the hepatocyte by its own specific ATP-binding cassette (ABC) transporter. The bile salt export pump (BSEP) ABCB11 transports the BAs and drives bile flow from the liver, but it is now clear that two lipid transporters, ABCB4 (which flops PC into the bile) and the P-type ATPase ATP8B1/CDC50 (which flips a different phospholipid in the opposite direction) play equally critical roles that protect the biliary tree from the detergent activity of the bile acids. Understanding the interdependency of these lipid floppases and flippases has allowed the development of an assay to measure ABCB4 function. ABCB4 harbours numerous mis-sense mutations which probably reflects the spectrum of liver disease rooted in ABCB4 aetiology. Characterization of the effect of these mutations at the protein level opens the possibility for the development of personalized prognosis and treatment.

Early Alterations of Bile Canaliculi Dynamics and the Rho Kinase/Myosin Light Chain Kinase Pathway Are Characteristics of Drug-Induced Intrahepatic Cholestasis

Intrahepatic cholestasis represents 20%-40% of drug-induced injuries from which a large proportion remains unpredictable. We aimed to investigate mechanisms underlying drug-induced cholestasis and improve its early detection using human HepaRG cells and a set of 12 cholestatic drugs and six noncholestatic drugs. In this study, we analyzed bile canaliculi dynamics, Rho kinase (ROCK)/myosin light chain kinase (MLCK) pathway implication, efflux inhibition of taurocholate [a predominant bile salt export pump (BSEP) substrate], and expression of the major canalicular and basolateral bile acid transporters. We demonstrated that 12 cholestatic drugs classified on the basis of reported clinical findings caused disturbances of both bile canaliculi dynamics, characterized by either dilatation or constriction, and alteration of the ROCK/MLCK signaling pathway, whereas noncholestatic compounds, by contrast, had no effect. Cotreatment with ROCK inhibitor Y-27632 [4-(1-aminoethyl)-N-(4-pyridyl) cyclohexanecarboxamide dihydrochloride] and MLCK activator calmodulin reduced bile canaliculi constriction and dilatation, respectively, confirming the role of these pathways in drug-induced intrahepatic cholestasis. By contrast, inhibition of taurocholate efflux and/or human BSEP overexpressed in membrane vesicles was not observed with all cholestatic drugs; moreover, examples of noncholestatic compounds were reportedly found to inhibit BSEP. Transcripts levels of major bile acid transporters were determined after 24-hour treatment. BSEP, Na⁺-taurocholate cotransporting polypeptide, and organic anion transporting polypeptide B were downregulated with most cholestatic and some noncholestatic drugs, whereas deregulation of multidrug resistance-associated proteins was more variable, probably mainly reflecting secondary effects. Together, our results show that cholestatic drugs consistently cause an early alteration of bile canaliculi dynamics associated with modulation of ROCK/MLCK and these changes are more specific than efflux inhibition measurements alone as predictive nonclinical markers of drug-induced cholestasis.

Prediction of Altered Bile Acid Disposition Due to Inhibition of Multiple Transporters: An Integrated Approach Using Sandwich-Cultured Hepatocytes, Mechanistic Modeling, and Simulation

Transporter-mediated alterations in bile acid disposition may have significant toxicological implications. Current methods to predict interactions are limited by the interplay of multiple transporters, absence of protein in the experimental system, and inaccurate estimates of inhibitor concentrations. An integrated approach was developed to predict altered bile acid disposition due to inhibition of multiple transporters using the model bile acid taurocholate (TCA). TCA pharmacokinetic parameters were estimated by mechanistic modeling using sandwich-cultured human hepatocyte data with protein in the medium. Uptake, basolateral efflux, and biliary clearance estimates were 0.63, 0.034, and 0.074 mL/min/g liver, respectively. Cellular total TCA concentrations (Ct,Cells) were selected as the model output based on sensitivity analysis. Monte Carlo simulations of TCA Ct,Cells in the presence of model inhibitors (telmisartan and bosentan) were performed using inhibition constants for TCA transporters and inhibitor concentrations, including cellular total inhibitor concentrations ([I]t,cell) or unbound concentrations, and cytosolic total or unbound concentrations. For telmisartan, the model prediction was accurate with an average fold error (AFE) of 0.99-1.0 when unbound inhibitor concentration ([I]u) was used; accuracy dropped when total inhibitor concentration ([I]t) was used. For bosentan, AFE was 1.2-1.3 using either [I]u or [I]t This difference was evaluated by sensitivity analysis of the cellular unbound fraction of inhibitor (fu,cell,inhibitor), which revealed higher sensitivity of fu,cell,inhibitor for predicting TCA Ct,Cells when inhibitors exhibited larger ([I]t,cell/IC50) values. In conclusion, this study demonstrated the applicability of a framework to predict hepatocellular bile acid concentrations due to drug-mediated inhibition of transporters using mechanistic modeling and cytosolic or cellular unbound concentrations.

Rho-kinase/myosin light chain kinase pathway plays a key role in the impairment of bile canaliculi dynamics induced by cholestatic drugs

Intrahepatic cholestasis represents a frequent manifestation of drug-induced liver injury; however, the mechanisms underlying such injuries are poorly understood. In this study of human HepaRG and primary hepatocytes, we found that bile canaliculi (BC) underwent spontaneous contractions, which are essential for bile acid (BA) efflux and require alternations in myosin light chain (MLC2) phosphorylation/dephosphorylation. Short exposure to 6 cholestatic compounds revealed that BC constriction and dilation were associated with disruptions in the ROCK/MLCK/myosin pathway. At the studied concentrations, cyclosporine A and chlorpromazine induced early ROCK activity, resulting in permanent MLC2 phosphorylation and BC constriction. However, fasudil reduced ROCK activity and caused rapid, substantial and permanent MLC2 dephosphorylation, leading to BC dilation. The remaining compounds (1-naphthyl isothiocyanate, deoxycholic acid and bosentan) caused BC dilation without modulating ROCK activity, although they were associated with a steady decrease in MLC2 phosphorylation via MLCK. These changes were associated with a common loss of BC contractions and failure of BA clearance. These results provide the first demonstration that cholestatic drugs alter BC dynamics by targeting the ROCK/MLCK pathway; in addition, they highlight new insights into the mechanisms underlying bile flow failure and can be used to identify new predictive biomarkers of drug-induced cholestasis.

Case Characterization, Clinical Features and Risk Factors in Drug-Induced Liver Injury

Idiosyncratic drug-induced liver injury (DILI) caused by xenobiotics (drugs, herbals and dietary supplements) presents with a range of both phenotypes and severity, from acute hepatitis indistinguishable of viral hepatitis to autoimmune syndromes, steatosis or rare chronic vascular syndromes, and from asymptomatic liver test abnormalities to acute liver failure. DILI pathogenesis is complex, depending on the interaction of drug physicochemical properties and host factors. The awareness of risk factors for DILI is arising from the analysis of large databases of DILI cases included in Registries and Consortia networks around the world. These networks are also enabling in-depth phenotyping with the identification of predictors for severe outcome, including acute liver failure and mortality/liver transplantation. Genome wide association studies taking advantage of these large cohorts have identified several alleles from the major histocompatibility complex system indicating a fundamental role of the adaptive immune system in DILI pathogenesis. Correct case definition and characterization is crucial for appropriate phenotyping, which in turn will strengthen sample collection for genotypic and future biomarkers studies.

Advantageous use of HepaRG cells for the screening and mechanistic study of drug-induced steatosis

Only a few in vitro assays have been proposed to evaluate the steatotic potential of new drugs. The present study examines the utility of HepaRG cells as a cell-based assay system for screening drug-induced liver steatosis. A high-content screening assay was run to evaluate multiple toxicity-related cell parameters in HepaRG cells exposed to 28 compounds, including drugs reported to cause steatosis through different mechanisms and non-steatotic compounds. Lipid content was the most sensitive parameter for all the steatotic drugs, whereas no effects on lipid levels were produced by non-steatotic compounds. Apart from fat accumulation, increased ROS production and altered mitochondrial membrane potential were also found in the cells exposed to steatotic drugs, which indicates that all these cellular events contributed to drug-induced hepatotoxicity. These findings are of clinical relevance as most effects were observed at drug concentrations under 100-fold of the therapeutic peak plasmatic concentration. HepaRG cells showed increased lipid overaccumulation vs. HepG2 cells, which suggests greater sensitivity to drug-induced steatosis. An altered expression profile of transcription factors and the genes that code key proteins in lipid metabolism was also found in the cells exposed to drugs capable of inducing liver steatosis. Our results generally indicate the value of HepaRG cells for assessing the risk of liver damage associated with steatogenic compounds and for investigating the molecular mechanisms involved in drug-induced steatosis.

Inhibition of bile salt transport by drugs associated with liver injury in primary hepatocytes from human, monkey, dog, rat, and mouse

Interference of bile salt transport is one of the underlying mechanisms for drug-induced liver injury (DILI). We developed a novel bile salt transport activity assay involving in situ biosynthesis of bile salts from their precursors in primary human, monkey, dog, rat, and mouse hepatocytes in suspension as well as LC-MS/MS determination of extracellular bile salts transported out of hepatocytes. Glycine- and taurine-conjugated bile acids were rapidly formed in hepatocytes and effectively transported into the extracellular medium. The bile salt formation and transport activities were time‒ and bile-acid-concentration‒dependent in primary human hepatocytes. The transport activity was inhibited by the bile salt export pump (BSEP) inhibitors ketoconazole, saquinavir, cyclosporine, and troglitazone. The assay was used to test 86 drugs for their potential to inhibit bile salt transport activity in human hepatocytes, which included 35 drugs associated with severe DILI (sDILI) and 51 with non-severe DILI (non-sDILI). Approximately 60% of the sDILI drugs showed potent inhibition (with IC50 values <50 μM), but only about 20% of the non-sDILI drugs showed this strength of inhibition in primary human hepatocytes and these drugs are associated only with cholestatic and mixed hepatocellular cholestatic (mixed) injuries. The sDILI drugs, which did not show substantial inhibition of bile salt transport activity, are likely to be associated with immune-mediated liver injury. Twenty-four drugs were also tested in monkey, dog, rat and mouse hepatocytes. Species differences in potency were observed with mouse being less sensitive than other species to inhibition of bile salt transport. In summary, a novel assay has been developed using hepatocytes in suspension from human and animal species that can be used to assess the potential for drugs and/or drug-derived metabolites to inhibit bile salt transport and/or formation activity. Drugs causing sDILI, except those by immune-mediated mechanism, are highly associated with potent inhibition of bile salt transport.

Role of enterohepatic recirculation in drug disposition: cooperation and complications

Enterohepatic recirculation (EHC) concerns many physiological processes and notably affects pharmacokinetic parameters such as plasma half-life and AUC as well as estimates of bioavailability of drugs. Also, EHC plays a detrimental role as the compounds/drugs are allowed to recycle. An in-depth comprehension of this phenomenon and its consequences on the pharmacological effects of affected drugs is important and decisive in the design and development of new candidate drugs. EHC of a compound/drug occurs by biliary excretion and intestinal reabsorption, sometimes with hepatic conjugation and intestinal deconjugation. EHC leads to prolonged elimination half-life of the drugs, altered pharmacokinetics and pharmacodynamics. Study of the EHC of any drug is complicated due to unavailability of the apposite model, sophisticated procedures and ethical concerns. Different in vitro and in vivo methods for studies in experimental animals and humans have been devised, each having its own merits and demerits. Involvement of the different transporters in biliary excretion, intra- and inter-species, pathological and biochemical variabilities obscure the study of the phenomenon. Modeling of drugs undergoing EHC has always been intricate and exigent models have been exploited to interpret the pharmacokinetic profiles of drugs witnessing multiple peaks due to EHC. Here, we critically appraise the mechanisms of bile formation, factors affecting biliary drug elimination, methods to estimate biliary excretion of drugs, EHC, multiple peak phenomenon and its modeling.

Inhibition of hepatic cytochrome P450 enzymes and sodium/bile acid cotransporter exacerbates leflunomide-induced hepatotoxicity

AIM

Leflunomide is an immunosuppressive agent marketed as a disease-modifying antirheumatic drug. But it causes severe side effects, including fatal hepatitis and liver failure. In this study we investigated the contributions of hepatic metabolism and transport of leflunomide and its major metabolite teriflunomide to leflunomide induced hepatotoxicity in vitro and in vivo.

METHODS

The metabolism and toxicity of leflunomide and teriflunomide were evaluated in primary rat hepatocytes in vitro. Hepatic cytochrome P450 reductase null (HRN) mice were used to examine the PK profiling and hepatotoxicity of leflunomide in vivo. The expression and function of sodium/bile acid cotransporter (NTCP) were assessed in rat and human hepatocytes and NTCP-transfected HEK293 cells. After Male Sprague-Dawley (SD) rats were administered teriflunomide (1,6, 12 mg · kg(-1) · d(-1), ig) for 4 weeks, their blood samples were analyzed.

RESULTS

A nonspecific CYPs inhibitor aminobenzotriazole (ABT, 1 mmol/L) decreased the IC50 value of leflunomide in rat hepatocytes from 409 to 216 μmol/L, whereas another nonspecific CYPs inhibitor proadifen (SKF, 30 μmol/L) increased the cellular accumulation of leflunomide to 3.68-fold at 4 h. After oral dosing (15 mg/kg), the plasma exposure (AUC0-t) of leflunomide increased to 3-fold in HRN mice compared with wild type mice. Administration of leflunomide (25 mg·kg(-1) · d(-1)) for 7 d significantly increased serum ALT and AST levels in HRN mice; when the dose was increased to 50 mg·kg(-1) · d(-1), all HRN mice died on d 6. Teriflunomide significantly decreased the expression of NTCP in human hepatocytes, as well as the function of NTCP in rat hepatocytes and NTCP-transfected HEK293 cells. Four-week administration of teriflunomide significantly increased serum total bilirubin and direct bilirubin levels in female rats, but not in male rats.

CONCLUSION

Hepatic CYPs play a critical role in detoxification process of leflunomide, whereas the major metabolite teriflunomide suppresses the expression and function of NTCP, leading to potential cholestasis.

Role of endoplasmic reticulum stress in drug-induced toxicity

Drug‐induced toxicity is a key issue for public health because some side effects can be severe and life‐threatening. These adverse effects can also be a major concern for the pharmaceutical companies since significant toxicity can lead to the interruption of clinical trials, or the withdrawal of the incriminated drugs from the market. Recent studies suggested that endoplasmic reticulum (ER) stress could be an important event involved in drug liability, in addition to other key mechanisms such as mitochondrial dysfunction and oxidative stress. Indeed, drug‐induced ER stress could lead to several deleterious effects within cells and tissues including accumulation of lipids, cell death, cytolysis, and inflammation. After recalling important information regarding drug‐induced adverse reactions and ER stress in diverse pathophysiological situations, this review summarizes the main data pertaining to drug‐induced ER stress and its potential involvement in different adverse effects. Drugs presented in this review are for instance acetaminophen (APAP), arsenic trioxide and other anticancer drugs, diclofenac, and different antiretroviral compounds. We also included data on tunicamycin (an antibiotic not used in human medicine because of its toxicity) and thapsigargin (a toxic compound of the Mediterranean plant Thapsia garganica) since both molecules are commonly used as prototypical toxins to induce ER stress in cellular and animal models.

ABC gene-ranking for prediction of drug-induced cholestasis in rats

As legacy toxicogenomics databases have become available, improved data mining approaches are now key to extracting and visualizing subtle relationships between toxicants and gene expression. In the present study, a novel “aggregating bundles of clusters” (ABC) procedure was applied to separate cholestatic from non-cholestatic drugs and model toxicants in the Johnson & Johnson (Janssen) rat liver toxicogenomics database [3]. Drug-induced cholestasis is an important issue, particularly when a new compound enters the market with this liability, with standard preclinical models often mispredicting this toxicity. Three well-characterized cholestasis-responsive genes (Cyp7a1, Mrp3 and Bsep) were chosen from a previous in-house Janssen gene expression signature; these three genes show differing, non-redundant responses across the 90+ paradigm compounds in our database. Using the ABC procedure, extraneous contributions were minimized in comparisons of compound gene responses. All genes were assigned weights proportional to their correlations with Cyp7a1, Mrp3 and Bsep, and a resampling technique was used to derive a stable measure of compound similarity. The compounds that were known to be associated with rat cholestasis generally had small values of this measure relative to each other but also had large values of this measure relative to non-cholestatic compounds. Visualization of the data with the ABC-derived signature showed a very tight, essentially identically behaving cluster of robust human cholestatic drugs and experimental cholestatic toxicants (ethinyl estradiol, LPS, ANIT and methylene dianiline, disulfiram, naltrexone, methapyrilene, phenacetin, alpha-methyl dopa, flutamide, the NSAIDs–—indomethacin, flurbiprofen, diclofenac, flufenamic acid, sulindac, and nimesulide, butylated hydroxytoluene, piperonyl butoxide, and bromobenzene), some slightly less active compounds (3′-acetamidofluorene, amsacrine, hydralazine, tannic acid), some drugs that behaved very differently, and were distinct from both non-cholestatic and cholestatic drugs (ketoconazole, dipyridamole, cyproheptadine and aniline), and many postulated human cholestatic drugs that in rat showed no evidence of cholestasis (chlorpromazine, erythromycin, niacin, captopril, dapsone, rifampicin, glibenclamide, simvastatin, furosemide, tamoxifen, and sulfamethoxazole). Most of these latter drugs were noted previously by other groups as showing cholestasis only in humans. The results of this work suggest that the ABC procedure and similar statistical approaches can be instrumental in combining data to compare toxicants across toxicogenomics databases, extract similarities among responses and reduce unexplained data varation.