Evaluation of the Interplay between Uptake Transport and CYP3A4 Induction in Micropatterned Cocultured Hepatocytes s

Transcription

1 Supplemental material to this article can be found at: X/44/12/ $ DRUG METABOLISM AND DISPOSITION Drug Metab Dispos 44: , December 2016 Copyright ª 2016 by The American Society for Pharmacology and Experimental Therapeutics Evaluation of the Interplay between Uptake Transport and CYP3A4 Induction in Micropatterned Cocultured Hepatocytes s Amanda Moore, Paresh P. Chothe, Hong Tsao, and Niresh Hariparsad Drug Metabolism and Pharmacokinetics, Vertex Pharmaceuticals Incorporated, Boston, Massachusetts Received July 22, 2016; accepted September 16, 2016 ABSTRACT Previously we assessed the inductive response of prototypical inducers in hepatocyte monocultures and the long-term coculture model HepatoPac using cryopreserved hepatocytes from the same donors. We noted that the rifampicin EC 50 generated using the HepatoPac model corresponded better to the EC 50 based on clinical data compared with data generated in the monoculture system. We postulated that there may be differences in the functioning of uptake transporters between the two systems that may have led to the EC 50 difference. In this study, we characterized the functional activity of multiple uptake transporters in the two systems using cryopreserved hepatocytes from the same donors. Our data suggest that uptake transporter activity is higher in HepatoPac compared with the monoculture system. As a follow up to this Introduction In vitro approaches are now regarded as a critical first step in the assessment of the drug drug interaction (DDI) risk of compounds (FDA, 2012). However, despite improvements in predicting DDIs based on in vitro evaluations, unexpected DDIs do occur that could be due to unknown variables such as the involvement of drug transporters, which are not accurately measured with currently available in vitro tools. As early as 2004, Lam and Benet highlighted the interplay between drug transporters and metabolism as well as the limitations of using microsomes for clearance predictions when compounds are substrates of drug transporters (Lam and Benet, 2004). Thereafter, Frassetto et al. (2007) showed that cytochrome P450 isoform 3A4 (CYP3A4) activity, as measured by the erythromycin breath test, was decreased with rifampicin, which is an organic anion transporter (OATP) 1B1 inhibitor, and was increased with lansoprazole, which is an inhibitor of P-glycoprotein. Since the publication of these seminal reports, there have been several articles highlighting the importance of factoring in the interplay between drug transporters and metabolism and subsequent mechanistic models to understand the DDI risk of compounds (Pang et al., 2009; Li et al., 2014). However, there is a lack of understanding regarding the interplay between drug transporters and CYP3A4 induction. Cultured human hepatocytes, in either the two-dimensional or sandwich cultured models, are the mainstay for the assessment of induction-based DDI risk using dx.doi.org/ /dmd s This article has supplemental material available at dmd.aspetjournals.org. study, we measured the intracellular concentrations of rifampicin and bosentan, which are known substrates of uptake transporters; we observed significantly higher intracellular concentrations of both compounds in HepatoPac relative to the monoculture system. This finding equated to lower cytochrome P450 isoform 3A4 (CYP3A4) EC 50 values in the HepatoPac system compared with the monoculture system for both mrna and activity. In parallel, no significant EC 50 shift was observed for carbamazepine and phenytoin, which are not known to be substrates of uptake transporters. Our data suggest that next generation liver models such as HepatoPac may be a useful in vitro tool to quantitatively predict drug drug interactions when it is known that the perpetrator is also a substrate of drug transporters. static models such as the relative induction score (RIS) model (Fahmi et al., 2010; Einolf et al., 2014). Fahmi et al. (2008) used cryopreserved human hepatocytes to measure the CYP3A4 induction of 10 drugs to validate the RIS model and correlated the data to clinically observed DDIs. Two of the compounds, rifampicin and bosentan, are known to be substrates of uptake (OATP1B1/3) transporters (Paul et al., 2005; Treiber et al., 2007). There have been limited publications describing changes in uptake transporter activity with culture time; optimal expression and functionality levels may not be attained during the monocultures short culture period (3 to 5 days) (Ulvestad et al., 2011; Zhu et al., 2014). This was recognized previously by Shou et al. (2008), who attempted to develop a more quantitative model to predict inductionbased CYP3A4 DDIs in the clinic. They concluded that the DDI risk for bosentan when codosed with CYP3A4 substrates such as ethinylestradiol, glyburide, and sildenafil, using induction data generated with the standard monoculture model, is underpredicted because bosentan is a substrate of OATP. There is now a heightened interest in using technologies that maintain their differentiated in-vivo-like phenotype as the need for more predictive in vitro next generation hepatocyte models increases. One such technology is the HepatoPac model in which hepatocytes are cocultured with 3T3 fibroblasts in a precise microscale architecture for at least 9 days before use in an experiment. This extensive recovery period allows the hepatocytes to regain their polarity and form a biliary network that is essential for proper functioning of canalicular and sinusoidal transporters (Khetani and Bhatia, 2008). In contrast, hepatocytes cultured in a conventional monolayer have limited cell cell and cell matrix interactions with reduced ABBREVIATIONS: [ 14 C]TEA, [ 14 C]tetraethylammonium chloride; DDI, drug drug interaction; DMSO, dimethylsulfoxide; DPBS, Dulbecco s phosphate-buffered saline; ESI, electrospray ionization; HBSS, Hanks balanced salt solution; [ 3 H]TCA, [ 3 H]taurocholic acid; MDCK, Madin- Darby canine kidney cells; MRM, multiple reaction monitoring; NTCP, Na + taurocholate cotransporting polypeptide; OATP, organic aniontransporting polypeptide; OCT, organic cation transporter; P450, cytochrome P450 isoform; RIS, relative induction score. 1910

2 Transport and CYP3A4 Induction Interplay in Coculture Model 1911 polarization, bile canaliculi formation, and metabolic activity (Berthiaume et al., 1996). Monocultured hepatocytes remain viable in culture for 3 to 5 days whereas HepatoPac cultures display long-term viability and metabolic function over several weeks. In our earlier publication, we conducted an evaluation of the Hepatopac model as a tool to assess induction-based DDI risk in the clinic (Dixit et al., 2016). During this assessment, we speculated that the potency of rifampicin could be influenced by differences in the activity and expression of transporters in the two culture models, given that the EC 50 generated with the HepatoPac model (0.1 mm) correlated better with the modeled clinical EC 50 of rifampicin which was predicted to be as low as 0.05 mm using Simcyp (Kharasch et al., 2011; Dixit et al., 2016). In contrast, the EC 50 of rifampicin in monocultured or sandwich cultured cryopreserved hepatocytes is in the range of mm (Zhang et al., 2014; Dixit et al., 2016). To expand our understanding of the interplay between drug transporters and CYP3A4 induction, we: 1) compared the activity differences for uptake transporters in the same batch of cryopreserved hepatocytes when cultured in the monoculture and HepatoPac systems; 2) assessed the impact of differences in transporter activity on intracellular concentrations of perpetrator compounds between the two systems; and 3) determined whether differences in intracellular concentrations of perpetrator compounds translates to changes in EC 50 for compounds that are substrates and nonsubstrates for uptake transporters. Materials and Methods Human HepatoPac system and proprietary hepatocyte maintenance and application media were obtained from Ascendance Biotechnology (formerly Hepregen, Medford, MA). HepatoPac and monocultured hepatocytes were made using cryopreserved human hepatocytes (Supplemental Table 1) bought from Thermo Fisher Scientific (Waltham, MA) or BioreclamationIVT (Baltimore, MD). Dulbecco s modified Eagle medium, Williams E medium, cryopreserved hepatocyte recovery medium, Hepatocyte Maintenance and Plating Supplement Pack, nonessential amino acids, penicillin streptomycin, Hanks balanced salt solution (HBSS), Dulbecco s phosphate-buffered saline (DPBS), RNA Later, MagMax-96 Total RNA Isolation Kit, Applied Biosystems High Capacity cdna Reverse Transcription Kit, TaqMan Fast Advanced Master Mix, human GAPDH probe, and CYP3A4 cdna primer were all procured from Thermo Fisher Scientific (Waltham, MA). Fetal bovine serum, rifampicin, phenytoin, carbamazepine, pitavastatin, telmisartan, estrone 3-sulfate, metoprolol, taurocholic acid, and dimethylsulfoxide (DMSO) were purchased from Sigma-Aldrich (St. Louis, MO). Bosentan was obtained from Ava Chem Scientific (San Antonio, TX). Midazolam, 19OH-midazolam, and [ 13 C 3 ]19-hydroxymidazolam were purchased from BD Biosciences (San Jose, CA). [ 3 H]Pitavastatin (20 Ci/mmol), [ 3 H]telmisartan (15 Ci/mmol), [ 3 H]estrone 3-sulfate (50 Ci/mmol), [ 3 H]metoprolol (80 Ci/mmol), and [ 14 C]tetraethylammonium chloride ([ 14 C]TEA; Ci/mmol) were obtained from American Radiolabeled Chemicals Inc. (Saint Louis, MO). [ 3 H]Rifampicin (45Ci/mmol)and[ 3 H]bosentan (3.3 Ci/mmol) were obtained from Moravek Biochemicals (Brea, CA). [ 3 H]Taurocholic acid ([ 3 H]TCA; 15.4 Ci/mmol), LumaPlate, and Ultima Gold scintillation fluid were purchased from PerkinElmer (Waltham, MA). Cell Culture HepatoPac. HepatoPac and 3T3-J2 murine embryonic fibroblast only cultures were acquired from Ascendance Biotechnology and were prepared as previously described elsewhere (Khetani and Bhatia, 2008). Cryopreserved hepatocytes were seeded at a density of 30,000 and 5,000 viable cells/well with fibroblasts in 24-well and 96-well plates, respectively. The 3T3 fibroblast-only cultures were seeded at a density of 90,000 cells/well in 24 well plates. Both HepatoPac and 3T3 fibroblast-only cultures were maintained in proprietary media at Ascendance Biotechnology for 7 to 9 days after seeding before initiating experiments. The cells were cultured in an incubator with 10% CO 2 and 95% relative humidity at 37 C. Cryopreserved Monocultured Human Hepatocytes. Cells were plated in collagen-coated plates at a density of viable cells/well and viable cells/well in 24-well and 96-well formats, respectively. Cells were maintained in Williams E medium containing Hepatocyte Maintenance Supplement Pack and 10% bovine serum for 6 to 8 hours before changing to serum-free medium. The cells were cultured in an incubator with 5% CO 2 and 95% relative humidity at 37 C. MDCK Culture. Madin-Darby canine kidney cells (MDCK) cells were maintained and grown in Dulbecco s modified Eagle medium containing heat inactivated 10% fetal bovine serum, 1% nonessential amino acids, and 1% penicillin streptomycin solution. Cells were seeded at a density of 7,000 cells per well in HTS Transwell-96 System (0.143 cm 2, 1.0 mm PET membrane; Corning Costar, Cambridge MA). The apical and basolateral compartments received 75 ml and 235 ml of the culture media respectively. Cells were used for the permeability experiment on day 4 after seeding. The integrity of the monolayers was assessed before each experiment by measuring the transepithelial electrical resistance. Cultures exhibiting transepithelial electrical resistance values greater than 200 V.cm 2 were used for permeability measurements. Uptake Assay HepatoPac cultures were used 9 to 12 days after seeding, and monocultures were used 24 hours after seeding for uptake experiments. Dosing solutions were prepared using radiolabeled compounds supplemented with unlabeled compounds to achieve the desired dosing concentration. For the uptake transporter assessment, probe substrates were applied at the following concentrations: 0.5 mm [ 3 H]pitavastatin (spiked with mci/ml), 0.5 mm [ 3 H]telmisartan (spiked with mci/ml), 0.5 mm[ 3 H]estrone 3-sulfate (spiked with mci/ml), 18mM[ 14 C]TEA (spiked with 1 mci/ml), and 5 mm[ 3 H]TCA (spiked with 1mCi/ml) for 0, 0.5, 5, and 15 minutes. For measurement of intracellular drug accumulation the cultures were incubated with rifampicin (unlabeled rifampicin spiked with 1 mci/ml), bosentan (spiked with 0.5 mci/ml), and metoprolol (spiked with 0.5 mci/ml) at 0.01, 1, and 10 mm for 15 minutes, and 4, 18, and 24 hours. The cultures were washed with ice-cold HBSS buffer 4 times at the end of each time point, and then the cells were lysed in acetonitrile to measure the intracellular radioactivity. The intracellular concentrations (pmol) of rifampicin, bosentan, and metoprolol were calculated using eq. 1: Intracellular ConcentrationðpmolÞ ¼ ðintracellular DPM=Total DPMÞTotal pmol where intracellular DPM is the radioactivity measured in cell lysates, total DPM is the total radioactivity in the reaction medium, and total pmol is the total concentration of substrate in reaction medium. The cultures were washed 3 times with calcium containing HBSS and were preincubated at 37 C for 10 minutes. Uptake was initiated by adding HBSS buffer (ph 7.4) containing radiolabeled substrate to the cultures. For Na + taurocholate cotransporting polypeptide (NTCP) assessment with [ 3 H]TCA, the dosing solution was made in either InVitroGRO KHB buffer with Na + or a custom formulation sodium-free (2Na + ) KHB buffer (BioreclamationIVT). Cultures were kept at 37 C during the substrate incubations. The uptake reaction was stopped by adding ice-cold DPBS at specified time points. The cultures were washed 3 times with ice-cold DPBS. Acetonitrile was added to lyse the cells, and the lysate was transferred to LumaPlates with a scintillant-coated bottom or scintillation vials containing scintillation cocktail. The LumaPlates were dried using GeneVac EZ-2 Plus (Ipswich, United Kingdom), and the radioactivity of the sample was measured using Top Count NXT (PerkinElmer). Samples collected into scintillation vials were measured using a liquid scintillation counter (HIDEX 300SL; Laboratory Logic, Brandon, FL). Permeability Assessment with MDCK Cultures The assay procedure was conducted using a Tecan Genesis automated liquid handler (Durham, NC). The cell culture insert plate was removed from the basal feeder tray and placed into a new 96-well basolateral plate. The apical side of monolayers was washed 3 times with 75 ml of HBSS (ph 7.4) containing calcium, magnesium, and 0.025% bovine serum albumin. Rifampicin, bosentan, phenytoin, and carbamazepine solutions were prepared in HBSS at a final concentration of 1 mm. Drug solutions were added to the designated donor compartments, and ð1þ

3 1912 Moore et al. fresh HBSS was added to the corresponding receiver compartments for both apical-to-basolateral (A B) and basolateral-to-apical (B A) permeability experiments. The cells were incubated at 37 C with shaking for 60 minutes. Aliquots (50 ml) were removed from the receiver and donor compartments at 0 and 60 minutes, and were transferred into 96-well microtiter plates for liquid chromatography with tandem mass spectrometry analysis. At the end of the permeability experiment, 100 mm lucifer yellow was added to the apical side of the monolayers to confirm the integrity of the cell monolayer during incubation with the test compounds. Metoprolol, a high permeability marker, was used as a control. The apparent permeability coefficient (P app ) of the test article and controls is determined using the following equation: P app ¼ðdQ=dtÞ=ðArea C 0 Þ where dq/dt is the rate of appearance of the compound in the receiver compartment, which is measured as the change in total quantity in the receiver compartment over time (slope of linear regression); area is the surface area of the membrane (0.143 cm 2 ); and C 0 is the initial concentration of a compound in donor compartment. CYP3A4 Induction Assay HepatoPac cultures were treated with test compounds 9 to 12 days after plating. Cells were treated with rifampicin ( mm), bosentan ( mm), phenytoin (1 200 mm), or carbamazepine ( mm) for 72 hours in proprietary application medium supplied by Ascendance Biotechnology. Cultures were maintained in an incubator with 10% CO 2 and 95% humidity at 37 C. Monocultures were treated with test compounds 24 hours after seeding. Cells were treated with rifampicin ( mm), bosentan ( mm), phenytoin (1 200mM), or carbamazepine ( mM) for 72 hours in Williams E medium containing Hepatocyte Maintenance Supplement Pack. Cultures were maintained in an incubator with 5% CO 2 and 95% humidity at 37 C. In both culture systems, the control wells were treated with 0.1% DMSO, fresh dosing solution was applied every 24 hours for 2 to 3 days, and all experiments were performed in triplicate. CYP3A4 Activity Assay After compound treatment, the supernatants were removed, and the cultures were washed with either application medium (HepatoPac) or HBSS (monoculture). Cultures were then incubated with 15 mm midazolam in either application medium (HepatoPac) or HBSS (monoculture) for 30 minutes. We added 100 ml aliquot of supernatant to 100 ml of acetonitrile containing [ 13 C 3 ]19-hydroxymidazolam as an internal standard. Samples were sealed and stored at 220 C until further analysis. Liquid Chromatography with Tandem Mass Spectrometry Analysis We used an API 5500 mass spectrometer (AB Sciex, Foster City, CA) coupled to CTC Analytics PAL (LEAP) autosampler (CTC Analytics AG, Zwingen, Switzerland) and Agilent 1100 Series binary pumps (Agilent Technologies, Palo Alto, CA) operated in electrospray ionization (ESI) mode for the analysis of analytes. The mass transition was used in ESI multiple reaction monitoring (MRM) mode for detection of the metabolite 1ʹ-hydroxymidazolam. The isotope-labeled internal standard transition was used for [ 13 C 3 ]19-hydroxymidazolam. The MRM transitions of m/z , , and were used in the positive ESI mode for rifampicin, bosentan, and carbamazepine, respectively. The MRM transition of m/z was used in the negative ESI mode for phenytoin. The MRM transition of m/z was used for the Vertex synthesized internal standard. Analytes were separated on a Unisol C mm, 5 mm column (Agela Technologies, Newark, DE). The mobile phase (flow rate 0.8 ml/min) consisted of 10 mm ammonium acetate in water (ph 4.0) (A) and acetonitrile/methanol (50/50, v/v) (B). The linear gradient was minutes, 0 70% B; minutes, 70% 70% B; minutes, 70% 99% B; minutes, 99% 99% B. the system was re-equilibrated with initial conditions for another 0.5 minutes. Standard curves were made in matrix and used for concentration calculations. Data were analyzed using Analyst version (SCIEX, Framingham, MA). ð2þ Relative Expression of mrna After compound treatment and CYP3A4 activity incubations, the cultures were washed with HBSS and stored in RNAlater at 220 C until the RNA isolation protocol was performed. The MagMax-96 Total RNA Isolation Kit was used following the manufacturers protocol to isolate the total RNA from each culture well. Concentration and quality of the isolated RNA were measured in the representative samples using a Nanodrop2000 spectrophotometer (Thermo Fisher Scientific). We reverse transcribed 50 ng of total RNA to cdna using the High Capacity cdna Reverse Transcription Kit, according to the manufacturer s protocol. Quantitative polymerase chain reaction reactions were performed using the TaqMan Fast Advanced Master Mix, human GAPDH probe, and cdna primer for CYP3A4. All data were normalized to the expression of the housekeeping gene GAPDH, and the fold change in mrna expression of CYP3A4 over the solvent control was calculated using the DDCt method. Induction Data Analysis and Fitting Data from the concentration response curves were fit to a three-parameter sigmoid (Hill) model, using eq. 3 to calculate the EC 50 and E max values. E ¼ Emax*C^g EC50^g þ C^g The baseline value of induction (E0) was fixed to 1. Data were fitted with GraphPad Prism 5.0 (GraphPad Software, San Diego, CA). Statistical Analysis The data are shown as the mean value with bars indicating the standard error (S.E.) for n $ 3. Data analysis was performed with GraphPad 5.0 using twoway analysis of variance (ANOVA) with either Bonferroni s or Turkey s post hoctest. P # 0.05 was considered statistically significant. Data are represented as P,0.01. Results Basal Expression of Drug Transporter Genes in HepatoPac and Monoculture Systems. As described in Dixit et al. (2016), the basal gene expression of 76 drug disposition genes was evaluated in both HepatoPac and monoculture systems using the same cryopreserved hepatocytes from three donors in both culture models (Supplemental Table 1). Several major drug transporter genes had higher expression in HepatoPac relative to monoculture systems. A.2-fold increase in gene expression was observed in at least one donor for the following genes in HepatoPac compared with monoculture systems: OATP1B3 (Hu1624: 2.8-fold, BPB: 11.1-fold, NON: 16.8-fold), OATP1A2 (NON: 5.2-fold), and organic cation transporter 1 (OCT1) (BPB: 4.4-fold, NON: 6.1-fold). OATP1B1 and NTCP gene expressions were similar in the HepatoPac and monoculture systems in all three donors. Functional Activity of Uptake Transporters in HepatoPac and Monoculture Systems. As mrna expression does not always translate to function (Harwood et al., 2016), we evaluated the transport activity of major hepatic uptake transporters OATP1B1, OATP1B3, OATP1A2, OCT1, and NTCP in the HepatoPac system compared with corresponding activities in the monoculture system in all three donors using relatively specific radiolabeled probe substrates for each transporter. We measured uptake of [ 3 H]pitavastatin (OATP1B1), [ 3 H]telmisartan (OATP1B3), [ 3 H]estrone-3 sulfate (OATP1A2), [ 14 C]TEA (OCT1), and [ 3 H]TCA (NTCP) for 0, 0.5, 5.0, and 15 minutes at 37 C in HepatoPac and monoculture systems (Fig. 1, rows A and B, respectively) (Hagenbuch and Meier, 1994; Breidert et al., 1998; Hirano et al., 2006; Ishiguro et al., 2006; Konig et al., 2006). The uptake of [ 3 H]TCA (NTCP) was performed with and without Na +. To measure active uptake of the substrates, all studies were performed in parallel with a 4 C passive uptake control, and the raw data were normalized to this background value (data not shown). ð3þ

4 Transport and CYP3A4 Induction Interplay in Coculture Model 1913 Fig. 1. Hepatocyte uptake transporter activity in HepatoPac (row A) and monoculture (row B). Cultures were incubated with radiolabeled probe substrates for the respective uptake transporters as described in Materials and Methods. After 0, 0.5, 5, and 15 minutes at 37 C, the amount of radioactivity in the cell lysates was measured. Data shown for OATP1B1 ([ 3 H]pitavastatin), OATP1B3 ([ 3 H]telmisartan), OATP1A2 ([ 3 H]estrone 3-sulfate), OCT1 ([ 14 C]TEA), and NTCP ([ 3 H]TCA) are the mean and standard error of three independent determinations in three individual hepatocyte donors. [ 3 H]TCA uptake was measured in the absence and presence of Na +. All data were normalized by cell number and a 4 C control. As indicated in Fig. 1 and Table 1, the uptake of pitavastatin (OATP1B1), telmisartan (OATP1B3), TEA (OCT1), and TCA (NTCP) was significantly higher in the HepatoPac system compared with monoculture system in all three donors (P,0.01). However, estrone-3 sulfate (OATP1A2) uptake was similar in both HepatoPac and monoculture in donors Hu1624 and BPB and modestly higher (1.9-fold) in donor NON TABLE 1 Uptake transporter activity in HepatoPac and monoculture Uptake of radiolabeled substrates was measured as described in the Materials and Methods section. Data shown are the mean (6 S.E.) of three independent determinations in each culture model in three individual hepatocyte donors. Transporter HepatoPac Uptake Monoculture Hu1624 BPB NON Hu1624 BPB NON pmol/ cells OATP1B1 [ 3 H]Pitavastatin 0 min 1.4 (0.4) 0.0 (0.0) 3.8 (0.1) a 0.8 (0.0) 0.0 (0.0) 0.7 (0.0) 0.5 min 4.4 (0.1) b 2.7 (0.3) b 0.0 (0.0) a 1.2 (0.0) a 0.8 (0.0) 3.1 (0.2) 5 min 13.5 (1.7) b 9.3 (0.7) b 30.4 (1.1) a,b 4.0 (0.3) a 3.7 (0.1) 9.3 (0.3) 15 min 19.8 (0.8) b 18.5 (0.2) b 82.3 (3.6) a,b 9.1 (0.3) 7.0 (0.1) 15.7 (0.4) OATP1B3 [ 3 H]Telmisartan 0 min 6.5 (0.8) b 8.0 (0.5) b 12.4 (1.5) b 1.3 (0.2) 1.3 (0.2) 0.8 (0.0) 0.5 min 16.8 (1.1) b 10.9 (1.8) b 6.1 (1.7) 2.3 (0.2) 2.1 (0.2) 2.7 (0.5) 5 min 42.2 (1.9) b 32.7 (1.4) b 33.1 (1.3) b 6.2 (0.2) 6.1 (0.4) 8.8 (0.6) 15 min 60.2 (2.1) b 62.0 (1.6) b 67.8 (6.6) b 13.2 (0.8) 11.8 (0.6) 14.1 (0.6) OATP1A2 [ 3 H]Estrone 3-sulfate 0 min 2.0 (0.0) 0.2 (0.1) 5.2 (0.2) 1.2 (0.1) 0.1 (0.0) 1.4 (0.0) 0.5 min 4.5 (1.1) 1.4 (0.3) 6.8 (0.5) 2.7 (0.1) 0.9 (0.0) 2.9 (0.2) 5 min 12.6 (0.1) b 7.2 (1.0) 22.9 (0.5) b 8.4 (0.3) 4.7 (0.1) 8.7 (0.9) 15 min 20.1 (0.2) b 15.2 (1.8) b 36.2 (5.3) b 16.1 (1.0) 9.8 (0.2) 19.2 (1.6) OCT1 [ 14 C]TEA 0 min 4.2 (2.0) b 0.0 (0.0) 2.5 (2.5) 0.1 (0.0) 0.1 (0.0) 0.9 (0.2) 0.5 min 0.1 (0.1) 0.0 (0.0) 0.0 (0.0) 0.4 (0.1) 0.3 (0.0) 1.7 (0.3) 5 min ND 20.0 (1.6) b 14.3 (4.8) b 1.9 (0.1) 1.5 (0.1) 3.7 (0.3) 15 min 16.4 (0.5) b 40.1 (2.4) b 27.8 (2.6) b 4.6 (0.2) 4.1 (0.2) 7.6 (0.4) NTCP c [ 3 H]TCA 0 min 70.0 (0.0) b 24.2 (0.0) b 34.9 (0.0) b 8.3 (0.0) 9.4 (0.0) b 5.0 (0.0) b 0.5 min 85.4 (0.0) b 43.4 (0.0) b 50.4 (0.0) b 10.7 (0.0) 16.8 (0.0) b 9.2 (0.0) b 5 min (0.1) b (0.0) b (0.0) b 41.6 (0.0) 54.1 (0.0) b 31.2 (0.0) b 15 min (0.1) b (0.0) b (0.0) b 68.4 (0.0) (0.0) b 66.2 (0.0) b ND, data not determined. a Mean from duplicate determinations. b P, c Na + -dependent uptake.

5 1914 Moore et al. (Table 1). Higher gene expression in HepatoPac correlated to higher activity for uptake transporters OATP1B3 and OCT1. OATP1B1and NTCP also showed higher activity in HepatoPac compared with the monoculture system even though the basal gene expression for this transporter was similar in the two culture models. Intracellular Concentrations of Rifampicin and Bosentan in HepatoPac and Monoculture Systems. Given the differences in transporter activity between the HepatoPac and monoculture systems, we evaluated the intracellular concentrations of actively transported compounds rifampicin and bosentan using the cryopreserved hepatocytes from donor NON. We selected donor NON as it represented the highest OATP1B1 and OATP1B3 uptake activity of the three donors. We measured the total intracellular uptake of [ 3 H]rifampicin, [ 3 H]bosentan, and [ 3 H]metoprolol in both HepatoPac and monoculture systems at 0.01, 1, and 10 mm for 15 minutes, and 4, 18, and 24 hours. [ 3 H]Metoprolol, a highly permeable compound that is not a CYP3A4 inducer, was used as a control for passive diffusion (Larregieu and Benet, 2014). In both models, the total uptake of rifampicin and bosentan increased with time up to 4 hours. Rifampicin at 0.01 mm and bosentan at 0.01 and 1 mm showed significant reduction in the total uptake after 4 hours, which could be due to depletion of compound in the medium. The total uptake of both rifampicin and bosentan was significantly greater (P, 0.01) in HepatoPac compared with monoculture at all three concentrations tested (Fig. 2). In contrast, the uptake of [ 3 H]metoprolol decreased up to 4 hours and then plateaued over the remainder of the 24-hour period at all three concentrations in both HepatoPac and monoculture systems (Fig. 2C). We also did not see any significant loss of radioactivity in the wash buffer from the last wash step, which demonstrated no washout of intracellular compound. Given that HepatoPac is a coculture model with a 3:1 ratio of 3T3 fibroblast cells to hepatocytes, we measured the intracellular levels of all three compounds in 3T3 fibroblast cells alone, maintaining the same cell density as used in the HepatoPac system. Given the low permeability of rifampicin and bosentan (Supplemental Table 2), the measured intracellular levels of rifampicin and bosentan were minimal in the fibroblast cells compared with HepatoPac coculture at all three concentrations (0.01, 1, and 10 mm) (Fig. 2A and B). However, the metoprolol intracellular levels in fibroblast cells were found to be significant and similar to the HepatoPac uptake (Fig. 2C), indicating substantial contribution by fibroblasts in the uptake of metoprolol in the HepatoPac coculture due to its high passive permeability. Similar to metoprolol, the CYP3A4 inducers phenytoin and carbamazepine are also highly permeable compounds and were used Fig. 2. Uptake analysis of rifampicin (A), bosentan (B), and metoprolol (C) in HepatoPac, monoculture, and 3T3 fibroblast only. Cultures were incubated with radiolabeled compounds ([ 3 H]rifampicin, 3 [H]bosentan, or 3 [H]metoprolol) at three concentrations (0.01, 1.0, and 10.0 mm) for 0.25, 4, 18, and 24 hours at 37 C. At each time point the amount of radioactivity in the cell lysate was measured and normalized by cell number. Data shown are the mean and standard error of three independent determinations in a single hepatocyte donor (NON). **P, 0.01.

6 Transport and CYP3A4 Induction Interplay in Coculture Model 1915 as passive diffusion controls in the later described induction studies (Supplemental Table 2). We measured the medium concentration of phenytoin and carbamazepine in cryopreserved hepatocyte cultures over a 24-hour incubation period and observed that both compounds achieved equilibrium with no significant parent loss (data not shown). Impact of Higher Intracellular Concentrations of Rifampicin and Bosentan on CYP3A4 Induction in the HepatoPac and Monoculture Systems. We previously observed a 2- to 20-fold decrease in the EC 50 for rifampicin-mediated induction of CYP3A4 mrna and activity in HepatoPac compared with the monoculture system in three Fig. 3. Dose response and EC 50 of CYP3A4 mrna expression. Fold change of (A) HepatoPac and (B) monoculture mrna expression is plotted relative to a 0.1% DMSO control after treatment with rifampicin, bosentan, phenytoin, or carbamazepine for 72 hours. Data shown are the mean and standard error of three independent determinations in three individual hepatocyte donors. (Rifampicin and phenytoin data via Dixit et al., 2016.) (C) Calculated EC 50 value and standard error plotted for each individual donor. mhu1624; jbpb; dnon.

7 1916 Moore et al. cryopreserved hepatocyte donors. However, phenytoin, a CYP3A4 inducer that is not actively taken up by hepatocytes (Patsalos et al., 2002), only showed a 0.4- to 3.4-fold decrease in the EC 50 value in HepatoPac relative to the monoculture system for CYP3A4 mrna and activity (Dixit et al., 2016). In the current study we followed up on our initial observations by evaluating two additional CYP3A4 inducers, bosentan and carbamazepine (Bertilsson et al., 1997; Paul et al., 2005). Bosentan is an OATP1B1 and OATP1B3 substrate, and carbamazepine served as a passive uptake control (Paul et al., 2005; Treiber et al., 2007). Dose response curves were generated in both the HepatoPac and monoculture systems in the same cryopreserved hepatocyte donors as used in our previous study (Supplemental Table 1) (Dixit et al., 2016). The fold changes in Fig. 4. Dose response and EC 50 of CYP3A4 activity. Fold change of HepatoPac (A) and monoculture (B) activity is plotted relative to a 0.1% DMSO control after treatment with rifampicin, bosentan, phenytoin, or carbamazepine for 72 hours. (Rifampicin and phenytoin data via Dixit et al., 2016.) Data shown are the mean and standard error of three independent determinations in three individual hepatocyte donors. (C) Calculated EC 50 value and standard error plotted for each individual donor. mhu1624; jbpb; dnon.

8 Transport and CYP3A4 Induction Interplay in Coculture Model 1917 CYP3A4 gene expression and activity were determined and fit to the sigmoidal E max model. The EC 50 for bosentan-mediated induction of CYP3A4 mrna in HepatoPac was 0.2 mm in all three donors in comparison with the EC 50 of mm in monoculture, demonstrating a 4- to 4.5-fold decrease in the EC 50 value in HepatoPac compared with the monoculture system. The EC 50 for carbamazepine-mediated induction of CYP3A4 mrna in HepatoPac was mm; in monoculture it was mm. This is a 1.2- to 1.9-fold decrease in the EC 50 value in HepatoPac compared with the monoculture system. The E max values were similar across both culture models for all the compounds tested. Induction of CYP3A4 activity yielded similar results as the mrna data for these compounds (Figs. 3 and 4; Table 2). Monocultures of donor BPB did not reach maximum fold induction at the tested concentrations of bosentan and carbamazepine. Thus, we were unable to fit the CYP3A4 induction response curves for both mrna expression and activity to the sigmoidal E max model and could not calculate an E max or EC 50 value for bosentan and carbamazepine using monocultures of donor BPB. Discussion Despite being the most preferred in vitro model to assess enzymetransporter interplay, cryopreserved hepatocyte monocultures have limited usage due to a brief ex vivo life span and rapid decline of hepatic function, leading to uncertainties with the DDI predictions (Berthiaume et al., 1996; LeCluyse et al., 2012). Next-generation liver models such as HepatoPac aim to create more sophisticated in vitro tools that are more physiologically relevant and maintain a more in-vivo-like phenotype. Their development and use could improve DDI predictions, minimize unexpected clinical results, and aid in managing adverse risks associated with drug interactions. While early evidence for the interplay between transporters and metabolism has been recognized (Lam and Benet, 2004; Frassetto et al., 2007), little attention has been focused on the impact of drug transporters on P450 induction. Recently, Campbell et al. (2015) assessed the induction-based DDI between methadone and anti-hiv drugs using monocultured human hepatocytes and postulated that compared with clinical studies, hepatocytes underreport induction of methadone metabolism, which could potentially be due to in vitro and in vivo differences in the expression and activity of uptake and efflux transporters. This observation is supported by previous reports in which OATP expression and activity was variable over time. Ulvestad et al. (2011) noted a significant and variable decrease in OATP1B1/1B3 activity and/or increase in passive diffusion over a 24-hour culture period. Similarly, Zhu et al. (2014) showed that OATP-mediated uptake declined substantially in cultured human hepatocytes while uptake by OATP1B3 was not measureable at earlier culture periods but became detectable on day 7 and showed culture duration-dependent changes from days 7 to 14. Even though there are no published reports of changes in OCT1 expression and function in cultured human hepatocytes over time, Tchaparian et al. (2011) conducted studies in sandwichcultured rat hepatocytes and noted that Oct1 exhibited a 3- and 4-fold decrease in expression and activity on days 2 and 4, respectively, relative to day 0, whereas the reduction in Ntcp expression and function was greater than 5-fold. Ramsden et al. (2014) showed that active uptake by OATP1B1 (rosuvastatin) and NTCP (TCA) can be measured using the HepatoPac model, but in our present study we make a direct comparison of the function of both transporters in the HepatoPac and monoculture models using cryopreserved hepatocytes from the same donors (Fig. 1; Table 1). In addition to OATP1B1, we also observed significantly higher uptake TABLE 2 E max and EC 50 for CYP3A4 mrna and activity after treatment with rifampicin, bosentan, phenytoin, and carbamazepine for 72 hours in HepatoPac and monoculture systems. The mrna was measured using real-time polymerase chain reaction with Taqman probes. Activity was measured using midazolam as a probe substrate. Concentration response curves for each compound were fit to a three-parameter sigmoid (Hill) model to calculate the Emax and EC50 values. Data shown are the mean (6 S.E.) of three independent determinations in each culture model in three individual hepatocyte donors. HepatoPac Monoculture Emax EC50 (mm) Emax EC50 (mm) CYP3A4 Hu1624 BPB NON Hu1624 BPB NON Hu1624 BPB NON Hu1624 BPB NON mrna Rifampicin 21.1 (1.1) a 16.5 (1.4) a 13.1 (0.3) a 0.1 (0.1) a 0.1 (0.2) a 0.1 (0.1) a 12.4 (0.3) a 24.4 (0.7) b 24.4 (0.6) a 0.2 (0.1) a 0.5 (0.06) b 1.1 (0.1) a Bosentan 7.0 (0.4) 5.3 (0.1) 6.6 (0.3) 0.2 (0.1) 0.2 (0.1) 0.2 (0.1) 8.3 (0.2) ND 6.5 (0.3) 0.8 (0.0) ND 0.9 (0.2) Phenytoin 17.6 (2.0) a 14.5 (1.0) a 14.7 (4.2) a 31.9 (0..1) a 56.9 (0.1) a 70.7 (0.3) a 6.3 (0.3) a 11.5 (1.6) a 25.8 (2.6) a 12.0 (0.1) a 42.6 (0.2) a 55.0 (0.1) a Carbamazepine 7.1 (0.3) 5.1 (0.1) 8.8 (0.3) 16.2 (0.1) 15.6 (0.0) 14.6 (0.4) 9.7 (0.2) ND 9.4 (0.3) b 20.0 (0.0) ND 27.2 (0.0) b Activity Rifampicin 12.3 (0.4) a 4.2 (0.3) a 6.3 (0.2) a 0.1 (0.1) a 0.1 (0.2) a 0.03 (0.1) a 7.6 (1.5) a 4.8 (0.4) b 17.1 (1.0) a 0.2 (0.6) a 1.3 (0.1) b 0.6 (0.1) a Bosentan 18.1 (1.3) 6.5 (0.7) 9.6 (0.9) 0.4 (0.1) 0.2 (0.2) 0.2 (0.2) 7.5 (0.1) 6.2 (0.3) b 10.2 (0.2) 1.6 (0.0) 0.6 (0.1) b 1.8 (0.0) Phenytoin 9.2 (0.5) a 9.0 (1.0) a 8.0 (0.3) a 37.2 (0.1) a 48.1 (0.1) a 32.0 (0.1) a 6.2 (2.6) a 13.2 (15.4) a 12.0 (1.0) a (0.4) a (0.5) a 42.3 (0.1) a Carbamazepine 7.8 (0.2) 4.1 (0.2) 10.7 (0.6) 10.5 (0.1) 10.5 (0.1) 14.2 (0.1) 4.7 (0.2) ND 5.6 (0.2) b 13.1 (0.1) ND 23.0 (0.0) b ND, data not determined. a Dixit et al., b Mean from duplicate determinations.

9 1918 Moore et al. activity for other important uptake transporters such as OATP1B3, OCT1, and NTCP in the HepatoPac compared with the monoculture model in all three cryopreserved hepatocyte donors (Fig. 1; Table 1). Although the model substrates used in the functional assessment were relatively selective to the transporters tested, the possible involvement of other transporters cannot excluded owing to their overlapping substrate specificity (Hirano et al., 2004). Having established that there are functional differences in transportermediated uptake between the two systems, we sought to understand how this would impact the intracellular concentration of compounds that are both substrates of uptake transporters as well as inducers of CYP3A4. Even though it is known that intracellular drug concentrations are fundamentally important to accurately predict DDIs, they are difficult to quantify directly in humans. Typically, unbound plasma drug concentrations are used as a surrogate measure for intracellular drug concentration. However, this may not be accurate for poorly permeable and actively transported drugs, potentially resulting in an underprediction of DDI-risk (Shou et al., 2008; Chu et al., 2013). While multiple compounds are used for validation and calibration of hepatocyte lots for CYP3A4 induction predictions using static models such as RIS (Fahmi et al., 2008), to our knowledge only rifampicin and bosentan are known to be substrates of uptake transporters (Supplemental Table 2). As described earlier, Shou et al. (2008) postulated that the under predictions of DDIs by bosentan with several CYP3A4 substrates (e.g., ethinyl estradiol, glyburide, and sildenafil) were probably due to higher intracellular concentrations of the inducer in hepatocytes compared with plasma given that bosentan is a substrate for OATP1B1 and OATP1B3 (Treiber et al., 2007). Thus, with this information in hand, we assessed the intracellular concentration of rifampicin and bosentan in cryopreserved hepatocytes from the same donor in both the HepatoPac and monoculture models at multiple concentrations and time points (Fig. 2). While Guo et al. (2016) assessed intracellular levels of bosentan in sandwichcultured hepatocytes, they conducted their studies for only 20 minutes and in the presence of 4% bovine serum albumin, which makes comparisons with our data challenging. In line with the higher active uptake via OATP (Fig. 1), we observed significantly higher intracellular levels of rifampicin and bosentan in the HepatoPac model compared with the monoculture model in cryopreserved hepatocytes from the same donor. There was no significant uptake of rifampicin and bosentan in 3T3 fibroblasts alone, indicating that these compounds are actively taken up by hepatic transporters and that fibroblasts did not significantly contribute to the increased intracellular concentrations observed in the HepatoPac model compared with cryopreserved monocultured hepatocytes (Fig. 2). Parallel studies with metoprolol, a highly permeable compound, did not show an increase in intracellular concentration over time. To understand the implications of higher intracellular levels of rifampicin and bosentan in the HepatoPac model on CYP3A4 induction, we conducted induction studies in both HepatoPac and monocultured hepatocyte systems using cryopreserved hepatocytes from the same donors (Supplemental Table 1). As noted earlier, in our previous publication the EC 50 for rifampicin generated with the HepatoPac model (0.1 mm) correlated better with the modeled clinical EC 50 of rifampicin which was predicted to be as low as 0.05 mm using Simcyp (Kharasch et al., 2011; Dixit et al., 2016). In contrast, the EC 50 of rifampicin in cryopreserved monocultured or sandwich-cultured hepatocytes is in the range of mm (Zhang et al., 2014; Dixit et al., 2016). Similarly, we noted a shift in EC 50 generated for bosentan in the HepatoPac system (CYP3A4 mrna EC mm) compared with the monocultured system (CYP3A4 mrna EC mm). Due to the lack of clinical dose-dependent inductive response data for bosentan, we were unable to correlate the inductive response in HepatoPac to a clinical EC 50. In addition to compounds that are actively taken up into hepatocytes, we also included two compounds, phenytoin and carbamazepine, that are highly permeable and are not substrates of hepatic uptake transporters (Supplemental Table 2). In general, unlike for rifampicin and bosentan, the EC 50 shift for phenytoin and carbamazepine was less than 3-fold for both CYP3A4 mrna and activity, with minimal variability between the donors in the HepatoPac system compared with the monocultured system (Fig. 3C, Fig. 4C, and Table 2). Donor BPB revealed the largest donor difference between the models. With cultures in the monoculture system we were unable to achieve a maximum induction response for bosentan and carbamazepine and thus could not confidently calculate the EC 50. However, when BPB was cultured in the HepatoPac system, the inductive response to these compounds was similar to the other two donors used in this study. Overall, the HepatoPac induction data were more uniform and reproducible across donors compared with the monoculture system. The interplay between different cellular functions must be considered to improve the predictive power of induction-based DDIs using in vitro models. We have presented here evidence that HepatoPac has higher transporter activity for OATP1B1, OATP1B3, OCT1, and NTCP and that this high function correlates with higher intracellular concentration and increased induction potency of actively transported compounds. HepatoPac could be a more sensitive and valuable tool for studying compounds with complex DDIs such as those that are substrates for hepatic uptake transporters. Furthermore, studies could be performed at lower, more clinically relevant concentrations, as higher transporter functionality allows a compound to enter the cell, bind to transcription factors such as pregnane X receptor, and induce CYP3A4. More research is needed to explore the full potential and limitations of this coculture model, but the work presented here demonstrates the model s value as a tool for studying compounds with complex DDIs involving interplay between uptake transporters and CYP3A4 induction. Acknowledgments The authors thank Dr. Vaishali Dixit for valuable scientific discussions. Authorship Contributions Participated in research design: Moore, Chothe, Tsao, Hariparsad. Conducted experiments: Moore, Chothe. Performed data analysis: Moore, Chothe, Tsao. Wrote or contributed to the writing of the manuscript: Moore, Chothe, Tsao, Hariparsad. References Berthiaume F, Moghe PV, Toner M, and Yarmush ML (1996) Effect of extracellular matrix topology on cell structure, function, and physiological responsiveness: hepatocytes cultured in a sandwich configuration. FASEB J 10: Bertilsson L, Tybring G, Widén J, Chang M, and Tomson T (1997) Carbamazepine treatment induces the CYP3A4 catalysed sulphoxidation of omeprazole, but has no or less effect on hydroxylation via CYP2C19. Br J Clin Pharmacol 44: Breidert T, Spitzenberger F, Gründemann D, and Schömig E (1998) Catecholamine transport by the organic cation transporter type 1 (OCT1). Br J Pharmacol 125: Campbell SD, Gadel S, Friedel C, Crafford A, Regina KJ, and Kharasch ED (2015) Influence of HIV antiretrovirals on methadone N-demethylation and transport. Biochem Pharmacol 95: Chu X, Korzekwa K, Elsby R, Fenner K, Galetin A, Lai Y, Matsson P, Moss A, Nagar S, Rosania GR, et al.; International Transporter Consortium (2013) Intracellular drug concentrations and transporters: measurement, modeling, and implications for the liver. Clin Pharmacol Ther 94: Dixit V, Moore A, Tsao H, and Hariparsad N (2016) Application of micropatterned cocultured hepatocytes to evaluate the inductive potential and degradation rate of major xenobiotic metabolizing enzymes. Drug Metab Dispos 44: Einolf HJ, Chen L, Fahmi OA, Gibson CR, Obach RS, Shebley M, Silva J, Sinz MW, Unadkat JD, Zhang L, et al. (2014) Evaluation of various static and dynamic modeling methods to predict clinical CYP3A induction using in vitro CYP3A4 mrna induction data. Clin Pharmacol Ther 95: Fahmi OA, Boldt S, Kish M, Obach RS, and Tremaine LM (2008) Prediction of drug-drug interactions from in vitro induction data: application of the relative induction score approach using cryopreserved human hepatocytes. Drug Metab Dispos 36:

10 Transport and CYP3A4 Induction Interplay in Coculture Model 1919 Fahmi OA, Kish M, Boldt S, and Obach RS (2010) Cytochrome P450 3A4 mrna is a more reliable marker than CYP3A4 activity for detecting pregnane X receptor-activated induction of drug-metabolizing enzymes. Drug Metab Dispos 38: Food and Drug Administration (FDA) (2012) Guidance for Industry Drug Interaction Studies Study Design, Data Analysis, Implications for Dosing, and Labeling Recommendations, US FDA website [online] guidances/ucm pdf. Frassetto LA, Poon S, Tsourounis C, Valera C, and Benet LZ (2007) Effects of uptake and efflux transporter inhibition on erythromycin breath test results. Clin Pharmacol Ther 81: Guo C, Yang K, Brouwer KR, St Claire RL, 3rd, and Brouwer KL (2016) Prediction of altered bile acid disposition due to inhibition of multiple transporters: an integrated approach using sandwich-cultured hepatocytes, mechanistic modeling, and simulation. J Pharmacol Exp Ther 358: Hagenbuch B and Meier PJ (1994) Molecular cloning, chromosomal localization, and functional characterization of a human liver Na + /bile acid cotransporter. J Clin Invest 93: Harwood MD, Neuhoff S, Rostami-Hodjegan A, and Warhurst G (2016) Breast cancer resistance protein abundance, but not mrna expression, correlates with estrone-3-sulfate transport in caco-2. J Pharm Sci 105: Hirano M, Maeda K, Shitara Y, and Sugiyama Y (2004) Contribution of OATP2 (OATP1B1) and OATP8 (OATP1B3) to the hepatic uptake of pitavastatin in humans. JPharmacolExpTher311: Hirano M, Maeda K, Shitara Y, and Sugiyama Y (2006) Drug-drug interaction between pitavastatin and various drugs via OATP1B1. Drug Metab Dispos 34: Ishiguro N, Maeda K, Kishimoto W, Saito A, Harada A, Ebner T, Roth W, Igarashi T, and Sugiyama Y (2006) Predominant contribution of OATP1B3 to the hepatic uptake of telmisartan, an angiotensin II receptor antagonist, in humans. Drug Metab Dispos 34: Kharasch ED, Francis A, London A, Frey K, Kim T, and Blood J (2011) Sensitivity of intravenous and oral alfentanil and pupillary miosis as minimal and noninvasive probes for hepatic and firstpass CYP3A induction. Clin Pharmacol Ther 90: Khetani SR and Bhatia SN (2008) Microscale culture of human liver cells for drug development. Nat Biotechnol 26: König J, Seithel A, Gradhand U, and Fromm MF (2006) Pharmacogenomics of human OATP transporters. Naunyn Schmiedebergs Arch Pharmacol 372: Lam JL and Benet LZ (2004) Hepatic microsome studies are insufficient to characterize in vivo hepatic metabolic clearance and metabolic drug-drug interactions: studies of digoxin metabolism in primary rat hepatocytes versus microsomes. Drug Metab Dispos 32: Larregieu CA and Benet LZ (2014) Distinguishing between the permeability relationships with absorption and metabolism to improve BCS and BDDCS predictions in early drug discovery. Mol Pharm 11: LeCluyse EL, Witek RP, Andersen ME, and Powers MJ (2012) Organotypic liver culture models: meeting current challenges in toxicity testing. Crit Rev Toxicol 42: Li R, Barton HA, and Varma MV (2014) Prediction of pharmacokinetics and drug-drug interactions when hepatic transporters are involved. Clin Pharmacokinet 53: Pang KS, Maeng HJ, and Fan J (2009) Interplay of transporters and enzymes in drug and metabolite processing. Mol Pharm 6: Patsalos PN, Fröscher W, Pisani F, and van Rijn CM (2002) The importance of drug interactions in epilepsy therapy. Epilepsia 43: Paul GA, Gibbs JS, Boobis AR, Abbas A, and Wilkins MR (2005) Bosentan decreases the plasma concentration of sildenafil when coprescribed in pulmonary hypertension. Br J Clin Pharmacol 60: Ramsden D, Tweedie DJ, Chan TS, Taub ME, and Li Y (2014) Bridging in vitro and in vivo metabolism and transport of faldaprevir in human using a novel cocultured human hepatocyte system, HepatoPac. Drug Metab Dispos 42: Shou M, Hayashi M, Pan Y, Xu Y, Morrissey K, Xu L, and Skiles GL (2008) Modeling, prediction, and in vitro in vivo correlation of CYP3A4 induction. Drug Metab Dispos 36: Tchaparian EH, Houghton JS, Uyeda C, Grillo MP, and Jin L (2011) Effect of culture time on the basal expression levels of drug transporters in sandwich-cultured primary rat hepatocytes. Drug Metab Dispos 39: Treiber A, Schneiter R, Häusler S, and Stieger B (2007) Bosentan is a substrate of human OATP1B1 and OATP1B3: inhibition of hepatic uptake as the common mechanism of its interactions with cyclosporin A, rifampicin, and sildenafil. Drug Metab Dispos 35: Ulvestad M, Björquist P, Molden E, Asberg A, and Andersson TB (2011) OATP1B1/1B3 activity in plated primary human hepatocytes over time in culture. Biochem Pharmacol 82: Zhang JG, Ho T, Callendrello AL, Clark RJ, Santone EA, Kinsman S, Xiao D, Fox LG, Einolf HJ, and Stresser DM (2014) Evaluation of calibration curve-based approaches to predict clinical inducers and noninducers of CYP3A4 with plated human hepatocytes. Drug Metab Dispos 42: Zhu Q, Xia H, Xia CQ, Yang Q, Doshi U, Li AP, and Liao M (2014) Culture duration-, donor-, and medium-dependent changes in OATP1B3-mediated telmisartan uptake in human hepatocytes. Drug Metab Lett 7: Address correspondence to: Dr. Niresh Hariparsad, Department of Drug Metabolism and Pharmacokinetics, 4022 Vertex 1, 50 Northern Avenue, Boston, MA niresh_hariparsad@vrtx.com