Drug Transporters in Drug-Induced Liver Toxicity

Drug-induced liver injury – a major safety concern

Drug-Induced Liver Injury (DILI) is one of the most common and potentially severe drug side effects, which may lead to acute liver failure or develop into chronic liver disease in affected patients or trial volunteers.^1-2 Among the possible underlying mechanisms of DILI, cholestatic liver injury accounts for the majority of such outcomes, where the excessive accumulation of bile acids and salts (BA and BS, respectively) inside the hepatocytes leads to cell toxicity and liver damage.^3-4

A total of 63 drugs were withdrawn from the market between 1964 and 2018 due to toxic liver effects, demonstrating the serious impact of DILI^5-8, and while the related safety and regulatory measures have become increasingly strict and more detailed over the years, 11 of these withdrawals still occurred after 1998.^9-11 As the impact of DILI may have severe, even life-threatening consequences for patients and trial participants, identification of risk factors associated with DILI early in development is an important safety consideration. While post-market discovery of potential DILI-inducing effects of a drug is arguably the most severe scenario, hepatotoxicity is also a common cause for clinical-stage project terminations.^12-16 Any such termination imposes a significant cost for drug developers and their investors in both time and monetary value.

In Vitro Approaches for DILI Prediction

Early detection of potential hepatotoxic properties of drug candidates in development is therefore beneficial, as it can significantly reduce the risk of later-stage clinical failure due to DILI. While hepatotoxicity-assessment assays are usually part of early safety testing using liver-derived cell lines such as HepG2 or HepaRG cells, the in vitro cellular systems commonly applied in these experiments are not suitable for assessing toxic bile accumulation and cholestatic cytotoxicity.^17-18 In contrast, isolated primary hepatocytes grown in sandwich cultures form bile canaliculi and produce significant quantities of bile salts and can therefore be used for assessing cholestatic effects.^17,19 One drawback of this in vitro model, however, is that it is relatively complex and costly, and thus not typically used in early discovery. Furthermore, while preclinical in vivo toxicity studies can be carried out to investigate signs of liver toxicity, these may fail to accurately predict the risk of cholestatic DILI due to species differences. This can be particularly problematic when rodents are used, as BA composition and associated metabolic pathways in rodents are quite different from those found in humans.^20-24

A review of drug withdrawals due to DILI over the past decades has identified transporter involvement in 76% of cholestasis cases.⁵

Transporters Involved in DILI

Bile acids undergo enterohepatic recirculation, where they are excreted from the liver into the bile, then released in the small intestine to aid in digestion of fats from the diet, followed by reabsorption across the gut enterocytes back into the systemic circulation from where they are taken up into hepatocytes. Uptake of bile acids/salts from the blood into the hepatocytes and their excretion into the bile canaliculi are mediated by transporters. Disruption of the physiological function of these transporters by a drug or xenobiotic can lead to the disruption of bile flow, and consequently, drug-induced (cholestatic) liver injury. To aid hepatotoxicity prediction and de-risking, our scientists summarized the key transporters involved in bile homeostasis in a recent review article.²⁵

The transport of BA/BS into hepatocytes is mediated by the interplay of transporters differentially localized in the sinusoidal or the canalicular membranes of hepatocytes [Figure 1]:

• Sinusoidal influx transporters include Sodium taurocholate co-transporting polypeptide (NTCP)/solute carrier 10A1 (SLC10A1), and Organic anion transporting polypeptide 1B1 (OATP1B1)/solute carrier organic anion transporter 1B1 (SLCO1B1), and OATP1B3/SLCO1B3.
• Canalicular transport is mainly mediated by a bile salt export pump (BSEP)/ABCB11.
• Sinusoidal efflux takes place via multidrug resistance-associated proteins MRP2/ABCC2. MRP3/ABCC3 and MRP4/ABCC4 provide the sinusoidal efflux function.
• (OST α/β)/SLC51A/B is a heterodimeric organic solute transporter that is a facilitative transporter and may mediate transport into as well as out of the hepatocyte.

In addition to BA/BS transporters, two phospholipid translocases also contribute to bile secretion and maintaining healthy bile flow. The multidrug resistance protein 3 (MDR3) is a phosphatidylcholine floppase that has also been shown to transport certain drugs that may also act as its inhibitors. ATP8B1 translocates phosphatidylserine from the outer leaflet to the inner leaflet, and the impairment of this function leads to intrahepatic cholestasis. No drugs have been identified to date that inhibit ATP8B1.

The role and relative importance of each of the hepatic BA/BS transporters is reflected in the impact of their inhibition on DILI risk. BSEP is the primary culprit for possible transporter-mediated DILI, with MRP3 and MRP4 following closely on the list of transporters most likely to be involved in drug-induced cholestasis⁵, contributing to over 50% of recent drug withdrawals due to hepatotoxicity [Figure 2]. Inhibition of other transporters, such as NTCP and MDR3 may also be involved in DILI, but at a significantly lesser frequency. Inhibition can be tested using in vitro transporter assay systems that are not only significantly more cost-efficient than cellular models but also lend themselves better for screening setups to enable quick turnaround of data.

High DILI-Risk Transporters

This is in line with the physiological functions of these “high DILI-risk” transporters as summarized in Figure 3. BA efflux is mediated mainly by BSEP and MRP2. BSEP is responsible for the secretion of monovalent BS, which constitutes the majority of human bile secretion, while efflux of glucuronide or sulfate conjugates of monovalent BA/BS is mediated by MRP2 [Figure 3a]. Inhibition of BSEP by xenobiotics disrupts the normal excretion of bile acids into the bile canaliculi, resulting in increased intracellular accumulation of bile salts [Figure 3b]. Inhibitors of BSEP that also inhibit NTCP may mitigate intracellular BA/BS levels by blocking uptake, but if NTCP function is not impacted the intracellular concentration of BA/BS may further increase due to their continued active uptake into the hepatocytes [Figure 3d]. This, in turn, may lead to the development of cholestatic liver damage.

The efflux transporters MRP3 and MRP4 are expressed at relatively low levels and localized to the apical side of the hepatocytes. Under normal physiological conditions they do not significantly contribute to the efflux of BA into the systemic circulation. However, upon the increase of intracellular BA/BS levels, their expression is induced, leading to increased efflux of bile acids from the hepatocytes to the blood and easing cholestatic conditions [Figure 1c]. A similar role has been proposed for MRP2, which, unlike MRP3 and MRP4, resides in the basolateral (canalicular) membrane of hepatocytes and mediates biliary BA/BS secretion and also excretes high amounts of reduced glutathione (GSH), which is a driving force for overall bile secretion. Inhibition of MRP2 can therefore contribute to cytotoxic effects of BA/BS accumulation in hepatocytes via two separate mechanisms.

The hepatotoxicity de-risking strategy proposed by the International Transporter Consortium (ITC) also named BSEP inhibition assessment at the discovery stage as the initial step, with follow-up measures recommended in case BSEP inhibition is observed.²⁶ Indeed, BSEP inhibition has been deemed to have the best predictive value for DILI risk assessment, but multiple studies have also recommended assessing MRP2-3-4 inhibition to support decision making. These, and inhibition of other hepatic transporters, may also be key in understanding the mechanisms behind hepatotoxicity observed in preclinical experiments and addressing whether toxicity observed in a preclinical species would translate to humans.

At Charles River, in vitro transporter interaction assays are available for all hepatic transporters listed above, to provide drug developers with a flexible tool to investigate and de-risk hepatotoxicity. Obtaining data to address the possible risk drug-induced cholestasis at an early stage of discovery or development can be a useful and cost-effective approach to select safe lead molecules during candidate selection, identify and deprioritize (or discontinue) molecules with increased DILI risk, or if possible, minimize DILI risk via selection of appropriate dosage and treatment intervals for a safe application.