Drug-induced cholestasis risk assessment in sandwich-cultured human hepatocytes
Introduction
Drug-induced liver injury (DILI) plays a substantial role in clinical trial failures and post-marketing withdrawals. Symptoms of cholestasis occur in about 50% of the DILI cases reported in literature (Lee, 2003). According to Fischer et al., cholestasis is a pathological condition in which the bile secretion and flow is impaired, leading to hepatocytic accumulation of bile acids (BAs) and other cholephiles (Fischer et al., 1996). The subsequent toxicity, exerted by the accumulated BAs, is due to the fact that intracellular BAs can cause a disruption of the mitochondrial ATP synthesis, resulting in apoptosis or necrosis (Maillette de BuyWenniger and Beuers, 2010, Perez, 2009). The alleged role of BA homeostasis disturbances in different forms of hepatotoxicity is consistent with in vivo observations and in vitro data (Chatterjee et al., 2014a, Yamazaki et al., 2013).
The BA homeostasis is tightly regulated and requires the multiplex interplay between different transporters and metabolic enzymes in both liver and intestine. In hepatocytes, BAs are actively taken up by the sinusoidal uptake transporters, namely the sodium taurocholate co-transporting polypeptide (NTCP; SLC10A1) and the members of the organic anion transporting polypeptide (OATP; SLCO family). The excretion into bile canaliculi is facilitated primarily by the bile salt export pump (BSEP; ABCB11). Additionally, sulfated and glucuronidated BAs are transported by the canalicular efflux transporter multidrug resistance-associated protein 2 (MRP2; ABCC2) (Pauli-Magnus and Meier, 2006). The biliary excretion of cholesterol, which is a precursor of BAs, is facilitated by the heterodimer ABCG5/ABCG8, located in the canalicular membrane of the hepatocytes (Yu et al., 2004). At the sinusoidal membrane, BAs are transported back into the blood by the heterodimeric organic solute transporter (OSTα/OSTβ) (Ballatori et al., 2005). Under cholestatic conditions, the sinusoidal efflux transporters, multidrug resistance-associated protein 3 and 4 (MRP3, ABCC3; MRP4, ABCC4) are upregulated as a protective mechanism to transport BAs back into the blood (Alrefai and Gill, 2007, Bohan and Boyer, 2002).
BSEP inhibition has been implicated as the major mechanism of drug-induced cholestasis (DIC) (Kubitz et al., 2012). Examples of currently existing in vitro models to detect DIC rely on the inhibition of BSEP-mediated taurocholic acid transport, either in sandwich-cultured hepatocytes (SCH) or in BSEP overexpressing membrane vesicles (Dawson et al., 2012, Marion et al., 2007, Morgan et al., 2010, Pedersen et al., 2013). However, several recent studies support the recognition that evaluation of BSEP inhibition alone does not sufficiently predict DIC. Indeed other mechanisms (e.g. inhibition of other BA transporters) or an interplay between different mechanisms are involved (Chatterjee et al., 2014b, Fukuda et al., 2013). More specifically, it has been shown that inhibition of MRP4, next to BSEP inhibition, is associated with DIC (Kock et al., 2013). In that respect, sandwich-cultured human hepatocytes (SCHH) represent a suitable in vitro model to investigate the effect of xenobiotics on toxicity caused by accumulating BAs subsequent to disturbed BA disposition. Previous work has indicated that SCHH preserve the disposition pathways and cellular functions involved in hepatocytic BA handling (De Bruyn et al., 2013).
Recently, we successfully introduced and applied an in vitro model based on SCHH relying on the widely recognized mechanism of intracellular BA accumulation associated with DIC. Essentially, the in vitro cholestatic potential of the training compounds (TCs) was expressed by determining drug-induced cholestasis index (DICI) values. The DICI value reflects the relative residual urea formation by hepatocytes co-incubated with a TC at a specific concentration and BAs as compared to hepatocytes treated with TC alone. The ability of this model to identify compounds that may cause cholestasis in human by interfering with BA disposition was illustrated by a correlation between clinical incidence data on cholestasis on the one hand, and an in vitro DICI-based safety margin (SM) on the other hand. This SM was obtained as the lowest incubation concentration (μM) yielding a DICI ≤ 0.80, divided by the total therapeutic peak plasma concentration (Cmax,total, μM) in human (Chatterjee et al., 2014b).
We presently expand on demonstrating the utility and robustness of this in vitro model for early detection of drug candidates with risk for cholestasis by retrospectively evaluating a total of 14 TCs (Table 1 and Supplemental Table 2) of the EU-EFPIA Innovative Medicine Initiative (IMI) project “Mechanism-Based Integrated Systems for the Prediction of Drug-Induced Liver Injury (MIP-DILI)”. For the purpose of the present study, these 14 TCs were first classified according to the presence or the absence of clinical evidence on DIC and DILI (see Table 1).
Section snippets
Materials and methods
The study design applied/used for determination of the DICI values is depicted in Fig. 1.
Effect of the BA mixture on the biochemical functionality of SCHH
A 50-fold concentrated BA mixture was used to co-incubate the SCHH with TCs. Therefore, the effect of this mixture on the capability of the hepatocytes to convert ammonia into urea was measured. In Table 4, the mean (± SD) urea formation of 6 different human batches used throughout this study, when incubated with or without the BA mixture, is shown. It was observed that a 48 h exposure to a 50-fold concentrated BA mixture did not significantly decrease the urea produced by the hepatocytes (day-7
Discussion
We previously developed and applied a hepatocyte-based in vitro model to identify drug candidates that may cause cholestasis by interfering with the BA homeostasis (Chatterjee et al., 2014b). For the purpose of in vitro data presentation and interpretation, the calculation of the so-called DICI was introduced. The DICI informs on the ability of a TC (incubated at a given concentration) to disturb BA homeostasis in vitro. In order to translate the DICI values obtained in vitro into risk for
Transparency document
Acknowledgments
This work was financially supported by the EU-EFPIA Innovative Medicine Initiative (IMI) project #115336-2 “Mechanism-Based Integrated Systems for the Prediction of Drug-Induced Liver Injury (MIP-DILI)”, and by funds of the hepatic disposition research group of the KU Leuven Drug Delivery and Disposition Lab.
References (67)
Clarifying metformin's role and risks in liver dysfunction
J. Am. Pharm. Assoc.
(2010)- et al.
Toxicity and intracellular accumulation of bile acids in sandwich-cultured rat hepatocytes: role of glycine conjugates. Toxicol
In Vitro
(2014) - et al.
Hepatocyte-based in vitro model for assessment of drug-induced cholestasis
Toxicol. Appl. Pharmacol.
(2014) - et al.
The significance of mitochondrial toxicity testing in drug development
Drug Discov. Today
(2007) - et al.
Hepatic levels of bile acids in end-stage chronic cholestatic liver disease
Clin. Chim. Acta
(1996) Ximelagatran: a new oral anticoagulant
Best Pract. Res. Clin. Haematol.
(2004)- et al.
Inhibition of mitochondrial beta-oxidation as a mechanism of hepatotoxicity
Pharmacol. Ther.
(1995) - et al.
Troglitazone-induced intrahepatic cholestasis by an interference with the hepatobiliary export of bile acids in male and female rats. Correlation with the gender difference in troglitazone sulfate formation and the inhibition of the canalicular bile salt
Toxicology
(2001) - et al.
Prediction of drug-induced liver injury in humans by using in vitro methods: the case of ximelagatran. Toxicol
In Vitro
(2008) - et al.
BSEP inhibition: in vitro screens to assess cholestatic potential of drugs. Toxicol
In Vitro
(2012)