Abstract
Cyclosporine A (CsA) and rifampin are potent inhibitors of organic anion transporting polypeptide (OATP) 1B1 and are widely used to assess the risk for drug-drug interactions. CsA displays preincubation time-dependent, long-lasting inhibition of OATP1B1 in vitro and in rats in vivo, and a proposed mechanism is the trans-inhibition by which CsA inhibits OATP1B1 from the inside of cells. The current study aimed to experimentally validate the proposed mechanism using human embryonic kidney 293 cells stably expressing OATP1B1. The uptake of CsA reached a plateau following an approximate 60-minute incubation, with the cell-to-buffer concentration ratio of 3930, reflective of the high-affinity, high-capacity intracellular binding of CsA. The time course of CsA uptake was analyzed to estimate the kinetic parameters for permeability clearance and intracellular binding. When the OATP1B1-mediated uptake of [3H]estradiol-17β-glucuronide was measured following preincubation with CsA for 5 to 120 minutes, apparent Ki values became lower with longer preincubation. Our kinetic modeling incorporated the two reversible inhibition constants [Ki,trans and Ki,cis for the inhibition from inside (trans-inhibition) and outside (cis-inhibition) of cells, respectively] and estimated Ki,trans value of CsA was smaller by 48-fold than the estimated Ki,cis value. Rifampin also displayed preincubation time-dependent inhibition of OATP1B1, albeit the extent of enhancement was only twofold. The current study provides experimental evidence for the preincubation time-dependent shift of apparent Ki values and a mechanistic basis for physiologically based pharmacokinetic modeling that incorporates permeability clearance, extensive intracellular binding, and asymmetry of Ki values between the inside and outside of cells.
SIGNIFICANCE STATEMENT In vitro data and kinetic modeling support that preincubation time-dependent, long-lasting inhibition of OATP1B1 by CsA can be explained by the extensive intracellular binding and reversible OATP1B1 inhibition intracellularly (trans-inhibition) as well as extracellularly (cis-inhibition). For inhibitors to display time-dependency, the following factors were found important: time to reach a steady-state cellular concentration, trans-inhibition potency relative to cis-inhibition, and the degree of cellular inhibitor accumulation. This study would aid in the accurate prediction of drug-drug interactions mediated by OATP1B1 inhibition.
Introduction
Organic anion transporting polypeptide (OATP) 1B1 mediates the hepatic uptake of various anionic substrates, including many clinically important drugs (Niemi et al., 2011). When drug-drug interactions (DDIs) occur via mechanisms involving OATP1B1, the systemic exposure of OATP1B1 substrate drugs [e.g., 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors (statins), angiotensin II receptor blockers, antidiabetic drugs, antihepatitis C virus drugs] increases by coadministration of the inhibitor drugs, such as cyclosporine A (CsA) and rifampin (Maeda, 2015; Lowjaga et al., 2020). In patients receiving statin therapy, DDIs can increase the systemic statin exposure and the risk of statin-related myotoxicity, including rare but severe rhabdomyolysis (Omar and Wilson, 2002; Neuvonen et al., 2006; Iwaki et al., 2019). To assess the potential risk of DDIs mediated by OATP1B1 inhibition, new chemical entities (NCEs) are routinely evaluated for their inhibition potency against OATP1B1 in vitro at the nonclinical stages.
Cell-based assays for OATP1B1 inhibition typically determine the half inhibitory concentration (IC50) or the inhibition constant (Ki) for OATP1B1, key parameters in the prediction of DDI risk based on the static and dynamic models (Yoshida et al., 2012; Yoshikado et al., 2016; Taskar et al., 2020). In the conventional inhibition assays, a probe substrate of OATP1B1 is coincubated with an NCE (as a potential inhibitor). Several inhibitors, such as CsA, exhibit preincubation time-dependent, long-lasting inhibition of OATP1B1 in vitro (Amundsen et al., 2010; Shitara et al., 2012; Izumi et al., 2015). The inhibitory effect of CsA was sustained even after removing CsA from incubation buffer (Shitara et al., 2012; Furihata et al., 2014) and the inhibitory potency of CsA was enhanced with prolonged preincubation: apparent Ki (Ki,app) value was reduced by 3- to 22-fold compared with that determined without CsA preincubation (Amundsen et al., 2010; Gertz et al., 2013; Izumi et al., 2015). In addition, as shown in Supplemental Fig. 1, in vitro IC50 and/or Ki values obtained after CsA preincubation were comparable to in vivo Ki values estimated by physiologically based pharmacokinetic (PBPK) modeling of clinical DDI data (Yoshikado et al., 2016). Based on these findings, the latest regulatory DDI guidance/guidelines recommends considering the preincubation effects in the inhibition assays for OATP1B1 and OATP1B3 (MHLW, 2019; U.S. FDA, 2020).
Some practical approaches have been proposed to manage the time-dependent inhibition (TDI) of OATP1B1 in the DDI risk assessments (Izumi et al., 2015; Taguchi et al., 2020). However, the underlying mechanism has not been fully understood. A better understanding of the mechanisms contributing to the TDI of OATP1B1 may facilitate the incorporation of the TDI into PBPK modeling and quantitative prediction of OATP1B1-mediated DDIs. Consideration of the underlying mechanisms may also aid in the optimization of in vitro assay conditions for potential inhibitors of OATP1B1. Shitara and Sugiyama proposed that CsA can inhibit OATP1B1 not only from the outside (cis-inhibition) but also from the inside (trans-inhibition) of cells (Shitara and Sugiyama, 2017). Yet, supporting experimental data have been lacking.
This study investigated the TDI of OATP1B1 by CsA in human embryonic kidney (HEK) 293 cells stably expressing OATP1B1 (OATP1B1-HEK) and control cells (control-HEK), employing different assay conditions (Fig. 1). Rifampin was used for comparison as it is an OATP1B1 inhibitor with a modest preincubation time-dependent inhibitory effect (Pahwa et al., 2017; Barnett et al., 2018). The time profiles of cellular uptake and release were measured for CsA and rifampin in OATP1B1-HEK and control-HEK cells and analyzed by cellular kinetic modeling incorporating the trans-inhibition mechanism. The developed cellular kinetic models were also used to identify the factors contributing to the extent of TDI of OATP1B1.
Materials and Methods
Chemicals
[3H]Estradiol-17β-glucuronide ([3H]E2G, 52.9 Ci/mmol) was purchased from PerkinElmer Life Sciences (Boston, MA), and unlabeled E2G and rifampin were purchased from Sigma-Aldrich (St. Louis, MO). CsA was obtained from Toronto Research Chemicals (Toronto, ON, Canada). All other chemicals were of analytical grade and commercially available.
Cell Culture
OATP1B1-HEK and control-HEK were established previously (Izumi et al., 2013). The cells were seeded in the poly-D-lysine-coated 48-well plate (BD Bioscience, San Jose, CA) at the density of 2 × 105 cells/well and cultured for 48 hour at 37°C under 5% CO2 gas in a humidified incubator.
Inhibition of OATP1B1 by CsA and Rifampin in OATP1B1-HEK Cells
Fig. 1 depicts the four different inhibition assay conditions used to respectively capture cis-, trans-, both cis- and trans-inhibition (cis+trans-inhibition), and long-lasting inhibition by CsA or rifampin using OATP1B1-HEK and control-HEK. Using [3H]E2G (0.1 μM) as a probe substrate for OATP1B1 (Izumi et al., 2013), the initial uptake was assessed for 1 minute. Unless noted otherwise, Krebs Henseleit (KH) buffer (118 mM of NaCl, 23.1 mM of NaHCO3, 4.83 mM of KCl, 0.96 mM of KH2PO4, 1.20 mM of MgSO4, 12.5 mM of HEPES, 5.0 mM of glucose, and 1.53 of mM CaCl2, pH 7.4) was used for cell incubation (500 µl/well) and washing (1 ml/well), and incubations were performed at 37°C. For the long-lasting inhibition conditions, 50 µl of KH buffer was collected after the washout period to measure the inhibitor concentrations released from the cells at the designated times (10, 30, 60, and 120 minutes for CsA; 10 and 60 minutes for rifampin). For all the assay conditions, cells were washed with ice-cold KH buffer twice at the end of the substrate uptake and stored at –20°C until sample processing for measurement of the cellular concentrations of the probe substrate.
Uptake and Release Assays for CsA and Rifampin Using OATP1B1-HEK Cells
For the uptake (for CsA and rifampin) and release assays (for CsA), OATP1B1-HEK and control-HEK cells were preincubated with prewarmed KH buffer for 5 minutes at 37°C. For the uptake assay, the preincubation buffer was removed, and, subsequently, the uptake reaction was initiated by adding 500 µl of a prewarmed KH buffer containing CsA or rifampin. At the designated time, 50 µl of assay buffer was collected to measure the drug concentrations in buffer. After removing the residual buffer, cells were immediately washed with 1 ml of ice-cold KH buffer twice. For the release assay for CsA, cells were incubated with CsA for 120 minutes. After removing the incubation buffer and washing cells, 500 µl of prewarmed KH buffer was added, and cells were incubated for up to 30 minutes at 37°C. At the designated times (up to 30 minute), 50 µl of incubation buffer was collected, and cells were washed with 1 ml of ice-cold KH buffer twice. Cells were plated in triplicate wells for sample collection at each timepoint. Both buffer and cellular samples were stored at –20°C until sample processing to measure CsA and rifampin.
Quantification of [3H]E2G, CsA, and Rifampin
To measure cellular radioactivity of [3H]E2G, cells were lysed with 200 μl of 0.1 N NaOH, and the resulting cell lysate was neutralized with 20 μl of 1 N HCl. An aliquot (150 μl) of the neutralized cell lysate was mixed with 2 ml of scintillation fluid (Hionic-Fluor; Perkin Elmer Life Sciences), and the radioactivity was measured by a liquid scintillation counter (LSC; Tri-Carb A5110TR, PerkinElmer Life Sciences). The remaining neutralized cell lysate samples were used to quantify the protein concentrations using a BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA). For buffer samples, 20 μl of the aliquot was mixed with 2 ml of scintillation fluid to measure the radioactivity by the LSC.
To measure unlabeled CsA and rifampin taken up by OATP1B1-HEK and control-HEK, the cells were vigorously mixed and deproteinized using 300 μl of methanol containing an appropriate internal standard, followed by filtration and analysis via liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS). To quantify CsA and rifampin in KH buffer, 50 μl of the aliquot was mixed with 200 μl of methanol containing an appropriate internal standard and injected to LC-MS/MS. Protein assay was performed as described above for cells seeded in extra wells.
The LC-MS/MS system consisted of an AQUITY UPLC I-Class System (Waters Corp.), Xevo TQ-XS (Waters Corp.), and MassLynx Mass Spectrometry Software (version 4.2; Waters Corp.). For the analysis of CsA, chromatographic separation was performed using a Waters Acquity UPLC BEH C18 column (1.7 μm, 2.1 mm i.d., 30 mm; Waters, Milford, MA) at 65°C. The mobile phases were composed of distilled water containing 0.1% formic acid (solvent A) and acetonitrile containing 0.1% formic acid (solvent B), and the initial condition was 98% solvent A at a flow rate of 0.4 ml/min. After maintaining the initial condition for 0.5 minutes, solvent B was linearly increased to 80% over 1.5 minutes with a 1-minute hold at 80% B. For rifampin, the same UPLC column and mobile phases were used. The column temperature and flow rate were set to 40°C and 0.3 ml/min, respectively. The initial condition was 98% solvent A, and B was linearly increased to 80% over 3 minutes, then to 100% over the next 0.01 minutes with a 0.5-minute hold at 100% B. The column was equilibrated with the initial mobile phase before each injection (injection volume, 1 μl). The analytes were detected by electrospray ionization in positive ion mode, and the selected ion monitoring transitions were: 1202.8 > 224.1 for CsA and 823.46 > 791.09 for rifampin.
Kinetic Modeling
Data fitting was performed by a nonlinear least-squares regression method using Napp version 2.31 (Hisaka and Sugiyama, 1998).
Kinetic Analysis of Cellular Uptake and Inhibition in OATP1B1-HEK and Control-HEK cells
The OATP1B1-mediated uptake clearance of a substrate was calculated by subtracting the uptake clearance in control-HEK from that in OATP1B1-HEK. The apparent inhibition constant (Ki,app) value was estimated for CsA and rifampin by examining their inhibitory effects on the OATP1B1-mediated uptake of a probe substrate (% of control), using the following equation: where Ibuffer represents the nominal concentrations of inhibitors in buffer (μM) and Ki,app is an inhibition constant based on the Ibuffer values. Concentration-dependent uptake of a substrate via OATP1B1 was examined with or without an inhibitor and was analyzed by the Michaelis-Menten equation: where v, S, Vmax, and Km represent the uptake velocity (pmol/min/mg protein), substrate concentration (μM), the maximum uptake velocity (pmol/min/mg protein), and Michaelis constant (μM), respectively.
Saturation of Intracellular Binding of CsA in HEK293 Cells
Concentration-dependent uptake of CsA by OATP1B1-HEK and control-HEK was determined at 120 minutes after incubation and analyzed by the following equation to estimate parameters for the intracellular binding: where Icell and Ibuffer represent the measured values for the total cellular and buffer concentration of CsA (µM), respectively. As no active transport process was implicated for CsA in OATP1B1-HEK or control-HEK, the cellular unbound concentration of CsA (a neutral compound) was assumed to be equal to its concentration in buffer (Ibuffer) at the steady-state. The cellular bound concentration of CsA can be obtained by (Icell–Ibuffer) at steady-state. Kd, Bmax, and A represent the dissociation constant (µM), the maximum number of binding sites per cellular volume (µM), and nonsaturable component for intracellular CsA binding, respectively. To calculate Icell from the uptake data, the cellular volume of HEK293 cells was set to 2 µl/mg protein according to a previous report (Shitara and Sugiyama, 2017).
Modeling of Cellular Pharmacokinetics of CsA and Rifampin
For the observed time profiles of the uptake and release of CsA and rifampin determined in OATP1B1-HEK and control-HEK, the clearances via OATP1B1-mediated active uptake (PSact, µl/min/mg protein) and via passive diffusion (PSdif, µl/min/mg protein) were estimated by simultaneously fitting to the following equations: where Vcell, Vbuffer, PA, and fT represent the cellular volume of HEK293 cells (2 μl/mg protein), the volume of incubation buffer (500 μl/well), the protein amount per well (mg protein/well), and intracellular unbound fraction, respectively. PSact was set to zero based on the lack of evidence supporting CsA as a substrate of OATP1B1. Rifampin is known to be an OATP1B1 substrate, and PSact was defined by the following equation:
The intracellular binding of rifampin appeared constant in HEK293 cells in the tested concentration range (data not shown), but nonlinearity was observed for CsA. Therefore, fT (Ibuffer/Icell) of CsA was described by the following equation, derived from Eq. 3.
Mechanism for Time-Dependent Inhibition of OATP1B1
Fig. 2 depicts our proposed mechanism for TDI of OATP1B1, which is inhibited reversibly not only from the outside (cis-inhibition), but also from the inside (trans-inhibition) of cells (Shitara and Sugiyama, 2017). Based on our in vitro results using CsA, the competitive inhibition was applied to cis-inhibition, while the noncompetitive inhibition to trans-inhibition (detailed description in the “Results” section) also for rifampin. The following equation describes OATP1B1-mediated uptake clearance of a probe substrate in the presence of an inhibitor (CLOATP1B1+I): where Ki,cis is the inhibition constant (μM) for cis-inhibition defined by the extracellular unbound concentration, and Ki,trans is the inhibition constant (μM) for trans-inhibition defined by intracellular unbound concentration. As the substrate concentration used in our inhibition assays ([3H]E2G, 0.1 μM) was much lower than the reported Km value (7.0 µM) (Izumi et al., 2013), the changes in OATP1B1-mediated uptake clearance (expressed as % of control) were approximated as follows (Shitara and Sugiyama, 2017):
For the cis-inhibition assay conditions, the Ki,app value (Eq. 1) corresponds to the Ki,cis value as the Icell value is zero. For the trans-inhibition and long-lasting inhibition assay conditions, the Ibuffer was set to zero as the buffer containing an inhibitor was removed before initiating [3H]E2G uptake.
Statistical Analysis
The data were presented as mean ± S.E.M. or SD. The Student’s two-tailed unpaired t test was used for group comparison (GraphPad Prism 9, GraphPad Software, La Jolla, CA). The differences were considered statistically significant when P < 0.05.
Results
Preincubation Time-Dependent Inhibition of OATP1B1 by CsA in OATP1B1-HEK Cells
The inhibitory effect of CsA on OATP1B1-mediated uptake of [3H]E2G was examined in the cis-, trans-, and cis+trans-inhibition conditions using OATP1B1-HEK and control-HEK (Fig. 3 and Table 1). In the cis-inhibition condition, CsA inhibited OATP1B1-mediated uptake of [3H]E2G in a concentration-dependent manner with Ki,app (= Ki,cis) of 0.297 µM (Fig. 2). In the trans- and cis+trans-inhibition conditions, CsA inhibited the OATP1B1 activity more potently than in the cis-inhibition condition. When the preincubation time was varied (10 versus 120 minutes), the inhibitory potency of CsA increased with a longer preincubation time (Fig. 3 and Table 1): Ki,app values for the trans-inhibition conditions, 0.0523 and 0.0169 µM after 10- and 120-minute preincubation with CsA; Ki,app values for the cis+trans-inhibition conditions, 0.0378 and 0.0109 µM after 10- and 120-minute preincubation with CsA.
Inhibition Types for Cis- and Trans-Inhibition of OATP1B1 by CsA
The inhibition types for cis- and trans-inhibition of OATP1B1 by CsA were examined by preparing the Eadie-Hofstee plots, as shown in Fig. 4. In the cis-inhibition condition [cells were coincubated with [3H]E2G (0.02–100 µM) and CsA (0.3 μM); no preincubation with CsA], the Km value increased from 8.52 to 16.9 µM, but there was no change in Vmax (Table 2). In the trans-inhibition condition [the uptake of [3H]E2G (0.02–100 µM) was measured after 60-minute preincubation with CsA (0.01 μM)], the Vmax decreased from 241 to 167 pmol/min/mg protein, but there was no change in Km (Table 2). These results indicated that CsA inhibited OATP1B1 competitively from the outside and noncompetitively from the inside of cells.
Concentration-Dependent Uptake of CsA into HEK293 Cells
The time-dependent uptake of CsA was similar between OATP1B1-HEK and control-HEK, in line with CsA being a nonsubstrate of OATP1B1 (Fig. 5A). The uptake of CsA reached an apparent plateau around 60–120 minutes. The CsA uptake normalized by the medium concentration (µl/mg protein) at 0.1 µM was approximately 3.5-fold greater than that at 10 μM at 120 minutes after incubation.
When the concentration-dependent uptake of CsA was further analyzed at 120 minutes after incubation in OATP1B1-HEK and control-HEK (Fig. 5B), the cellular uptake of CsA appeared to contain both saturable and nonsaturable components, with nearly overlapping profiles between OATP1B1-HEK and control-HEK. The observed data (from OATP1B1-HEK and control-HEK cells) was fitted to a cellular kinetic model incorporating the saturable intracellular binding (Eq. 3). The obtained parameters were Kd of 0.0914 μM, Bmax of 326 μM, and nonsaturable component (A) of 362 (Table 3). Under the linear condition, the fT value and cell-to-medium CsA concentration ratio were calculated to be 0.000254 and 3930, respectively. CsA was reported to strongly bind with a 1:1 stoichiometry to cyclophilin A (Ke et al., 1994), a pharmacological target of CsA ubiquitously expressed in the body (Ryffel et al., 1991). Cyclophilin A accounted for as much as 0.1–0.4% of the total protein (28.6–114 μM) (Marks et al., 1991; Ryffel et al., 1991; Sarris et al., 1992), and the reported Kd values of CsA for cyclophilin A ranged from 0.013 to 0.03 μM (Dalgarno et al., 1986; Quesniaux et al., 1988; Kuglstatter et al., 2011), which were within a few folds difference compared with the currently observed Bmax (326 µM) and Kd (0.0914 µM) values in HEK293 cells, respectively (Table 3).
Kinetic Modeling of Time-Dependent Inhibition of OATP1B1 by CsA in OATP1B1-HEK Cells
Similar profiles were observed for the uptake and release of CsA between OATP1B1-HEK and control-HEK. As such, the data were combined and analyzed by the cellular kinetic model incorporating the saturable intracellular binding of CsA (Fig. 6). The time profiles for the uptake and release of CsA were well captured by the cellular kinetic model using the observed intracellular binding parameters, yielding the estimated PSdif value of 87.8 ± 9.0 μl/min/mg protein.
To estimate Ki,trans of CsA for OATP1B1, the effect of CsA preincubation time on the potentiation of OATP1B1 inhibition was further analyzed in the trans-inhibition condition (Fig. 7). With prolonged preincubation with CsA (for 5 to 120 minutes), the inhibitory effect of CsA on OATP1B1-mediated uptake of [3H]E2G was enhanced, reaching the maximal inhibition with 60-minute or longer preincubation (Fig. 7, A and B). The observed data were well captured by the cellular kinetic model with the estimated PSdif and Ki,trans values of 51.3 μl/min/mg protein and 0.00619 μM, respectively (Table 3): the simulated profiles are shown in solid lines and in good agreement with the observed data in Fig. 7, A and B. The current model for CsA with the optimized parameters (Table 3) well captured the preincubation time-dependent shift of Ki,app (defined with regard to the initial medium concentration) observed under the trans-inhibition conditions (Ki,app,trans) as shown in Fig. 7C.
For long-lasting inhibition assay conditions, the recovery of OATP1B1 activity was monitored after preincubation with CsA (120 minutes), followed by the incubation with CsA-free KH buffer (up to 120 minutes), using OATP1B1-HEK and control-HEK (Fig. 8A).The release of CsA and recovery of OATP1B1 activity were well captured by the cellular kinetic model (simulated profiles in solid and dotted lines in Fig. 8, B and C, respectively) using the estimated parameters in Table 3.
Kinetic Modeling of Time-Dependent Inhibition of OATP1B1 by Rifampin in OATP1B1-HEK Cells
When the time profiles for the uptake of rifampin was examined at 0.1, 0.3, 1, 2, 5, and 10 µM using OATP1B1-HEK and control-HEK, the uptake of rifampin reached the steady-state promptly, within 10 minutes, and the OATP1B1-mediated uptake was fully saturated at 10 µM (Supplemental Fig. 2). Since control-HEK cells were used in this study, fT value could be estimated as well as PSdif, Vmax, and Km values. The obtained data were simultaneously fitted to the cellular kinetic model, yielding the PSdif, Vmax, Km, and fT values of 52.5 μl/min/mg protein, 25.1 pmol/min/mg protein, 0.382 μM, and 0.0311, respectively (Table 3).
The inhibitory effect of rifampin on OATP1B1-mediated uptake of [3H]E2G was examined in the cis-, trans-, and cis+trans-inhibition conditions (Fig. 9A). In the cis-inhibition conditions, the Ki,app value (corresponding to Ki,cis in Fig. 2) of rifampin was 1.16 μM. Trans-inhibition assay conditions after 10- and 60-minute preincubation with rifampin yielded Ki,app values of 1.39 and 1.01 µM, respectively, comparable to those in the cis-inhibition condition. The cis+trans-inhibition assay conditions gave slightly more potent inhibition of OATP1B1 with the Ki,app values of 0.492 and 0.524 µM after 10- and 60-minute preincubation with rifampin, respectively (Table 1). From the relationship between simulated cellular level of rifampin and observed trans-inhibition of OATP1B1 (Fig. 9B), the Ki,trans of rifampin was estimated to be 1.56 μM (Table 3). The Ki,trans and Ki,app values were estimated based on the intracellular unbound concentration and buffer concentration of rifampin, respectively. Intracellular concentration of unbound rifampin is higher than medium concentration due to its active transport mediated by OATP1B1, and Ki,trans (1.56 µM) was slightly larger than Ki,app,trans (1.01 to 1.39 µM).
Long-lasting inhibition of OATP1B1 after 60-minute preincubation with rifampin at 10 and 100 µM was also investigated (Fig. 9C). OATP1B1 activity immediately recovered within 10 minutes after removing rifampin from buffer, but partially at 100 µM. Using the optimized parameters (Table 3), the recovery of OATP1B1 activity in the long-lasting inhibition assay conditions was also well reproduced by the cellular kinetic model (Fig. 9C). At 100 µM of rifampin, the OATP1B1 activity was recovered only partially, which was due to the amount of rifampin that remains in the cells (exhibit trans-inhibition of OATP1B1) even after a 60-minute washout period.
Discussion
In the present study, preincubation time-dependent, long-lasting inhibition of OATP1B1 by CsA and rifampin was characterized in OATP1B1-HEK and control-HEK cells, and the in vitro data were analyzed by kinetic modeling to gain the mechanistic insights and to identify the factors impacting the extent of the preincubation time-dependent inhibition of OATP1B1.
Similar to the previous reports (Pahwa et al., 2017; Barnett et al., 2018), our current results confirmed the differences between CsA and rifampin in terms of preincubation time-dependent shift of Ki,app values. Following 120-minute preincubation in trans- and cis+trans-inhibition assay conditions, the inhibitory potency of CsA on OATP1B1 was markedly enhanced: the Ki,app values of CsA were decreased by 16.5- and 27.2-fold compared with that in cis-inhibition assay conditions, respectively (Fig. 3 and Table 1). Prolonged preincubation (60 minutes or longer) was required for reaching the maximum OATP1B1 inhibition in trans-inhibition assay conditions (Fig. 7). On the other hand, the Ki,app shift of rifampin was modest (only by twofold) in the cis+trans-inhibition assay conditions (Fig. 9A and Table 1) and 10-minute preincubation was sufficient to reach the maximum trans-inhibition (Fig. 9B). Cellular kinetic modeling predicted that CsA uptake into HEK293 cells reached a steady-state after 60-minute or longer incubation (Fig. 5A) and concentration-dependent cellular distribution (Fig. 5B) via high-affinity, high-capacity intracellular binding (possibly with cyclophilin A). The uptake of rifampin in OATP1B1-HEK was approximately twice that in control-HEK, likely from OATP1B1-mediated uptake with a Km value of 0.382 μM (Supplemental Fig. 2 and Table 3). In contrast to CsA, the uptake of rifampin reached a steady-state quickly (within 10 minutes) in both OATP1B1-HEK and control-HEK (Supplemental Fig. 2). These results support the association of intracellular inhibitor concentrations with the enhancement of OATP1B1 inhibition in the trans- and cis+trans-inhibition conditions.
The time profiles for the uptake and release of CsA in HEK293 cells were well captured by the cellular kinetic model incorporating saturable intracellular binding (Fig. 6). Based on the experimental data of OATP1B1 inhibition by CsA (Fig. 4), our kinetic model incorporated the components of competitive cis-inhibition and noncompetitive trans-inhibition (Fig. 2). The model captured the preincubation time-dependency in trans-inhibition (Fig. 7) as well as long-lasting inhibition of OATP1B1 by CsA (Fig. 8). The trans-inhibition potency of CsA (Ki,trans, 0.00619 μM) was much stronger than that of cis-inhibition (Ki,cis, 0.297 μM), and the trans-inhibition was potentiated as preincubation time-dependent intracellular accumulation of CsA. Our current study offers the experimental validation of the proposed trans-inhibition mechanism for OATP1B1 TDI caused by CsA (Shitara and Sugiyama, 2017), differing from the mechanism-based inhibition of drug metabolizing enzymes (Murray, 1997; Lin and Lu, 1998).
Cellular kinetic parameters of CsA and rifampin may provide important insights into the factors impacting the TDI of OATP1B1. For comparison, we calculated the time to reach half maximum intracellular concentration (T1/2,max) as follows:
The calculated T1/2,max values of CsA and rifampin were 106 and 0.849 minutes, respectively. Given the similar PSdif values between CsA and rifampin (Table 3), the large difference in T1/2,max is related to approximately 120-fold lower fT values of CsA under linear conditions than that of rifampin (Table 3). Other factors to consider are the Ki,cis-to-Ki,trans ratio (α), and intracellular-to-buffer unbound drug concentration ratio (Kp,uu). According to Eq. 9, intracellular unbound inhibitor concentration (fT·Icell) can be replaced by Kp,uu·Ibuffer at a steady-state. Then, OATP1B1-mediated uptake clearance is decreased by a factor of (1+Ibuffer/Ki,cis) or (1+ α·Kp,uu·Ibuffer/Ki,cis) in cis- or trans-inhibition, respectively. Therefore, α·Kp,uu is a determinant of the extent of enhancement of OATP1B1 inhibition after preincubation. The Kp,uu of rifampin was 2.25 due to active uptake via OATP1B1, but symmetrical cis- and trans-inhibition of OATP1B1 offered the α value of 0.744, resulting in α·Kp,uu of 1.7. Although Kp,uu was unity for CsA, the contribution of trans-inhibition to the overall OATP1B1 inhibition was much greater than that of cis-inhibition, producing the α (and α·Kp,uu) value of 48. Furthermore, when the OATP1B1-mediated uptake is reduced by 50% in cis+trans-inhibition study, Ibuffer is equal to Ki,app,cis+trans at a steady-state in Eq. 9, and Ki,app,cis+trans can be described as follows:
The derivation of Eq. 11 is given in the supplemental text. Based on this equation, simulations were performed to assess the impact of α and Kp,uu on the steady-state Ki,app,cis+trans for CsA (Supplemental Figs. 3, A and B) and rifampin (Supplemental Figs. 3, C and D): for both drugs, the larger α and Kp,uu values, the smaller Ki,app,cis+trans values. Together, it is likely that the long T1/2,max and large α·Kp,uu of CsA account for notable preincubation time-dependent potentiation of OATP1B1 inhibition compared with rifampin.
The cellular kinetic model of CsA (Fig. 2 and Table 3) was also used for the sensitivity analysis examining the impact of PSdif, fT, Kp,uu, and Ki,trans on the cellular PK and preincubation time-dependent change in Ki,app,cis+trans and Ki,app,trans (Fig. 10 and Supplemental Fig. 4). Simulation results indicated that the smaller PSdif (Fig. 10, A and B) or fT (Supplemental Fig. 4), the longer preincubation time to reach a steady-state in Icell,u and maximum trans- and cis+trans-inhibition. With a sufficiently long time, Icell,u would correspond to Ibuffer (regardless of PSdif or fT values) and Ki,app,cis+trans to Ki,app,trans at a steady-state. The greater Kp,uu values, the higher Icell,u values (Fig. 10C), and the more potent trans- and cis+trans-inhibition (Fig. 10D). In Figs. 10, E and F, the contribution of trans-inhibition relative to cis-inhibition was reduced by increasing Ki,trans to 10 and 48 times the original values (i.e., Ki,cis = 48 × Ki,trans). Overall OATP1B1 inhibition was accounted for by cis- and trans-inhibition, but preincubation time-dependent shift of Ki,app,cis+trans values was less pronounced. The simulation results suggest that T1/2,max and α·Kp,uu are key parameters determining the degree of preincubation time-dependency of OATP1B1 inhibitors and may collectively influence optimal preincubation time in evaluating OATP1B1 inhibition at steady-state. Further investigation is warranted to conclude whether the current findings are applicable to preincubation effect-positive OATP1B1 inhibitors other than CsA.
The preincubation effects have also been observed in other solute carrier (SLC) transporters such as OATP1B3, organic anion transporter 1, organic anion transporter 3, organic cation transporter 1, organic cation transporter 2, multidrug and toxin extrusion protein 1, multidrug and toxin extrusion protein 2K, and Na+-taurocholate cotransporting polypeptide (NTCP) (Gertz et al., 2013; Furihata et al., 2014; Ma et al., 2015; Arakawa et al., 2017; Oh et al., 2018; Omote et al., 2018; Tátrai et al., 2019; Lowjaga et al., 2020). Tátrai et al. comprehensively examined preincubation-dependency for the inhibition of clinically relevant SLC transporters (Tátrai et al., 2019). In their report, in the case of the inhibitors displaying the enhancement by preincubation, it took a long time to reach equilibrium between intracellular concentration and buffer concentrations. In contrast, rapid equilibrium was achieved for preincubation effect-negative inhibitors (Tátrai et al., 2019). Recently, Lowjaga et al. reported that taurolithocholic acid exhibited long-lasting trans-inhibition of NTCP and discussed the physiologic significance of the NTCP trans-inhibition mechanism as a potential defense system to protect hepatocytes from cholestatic bile acids and the hepatitis B and D virus infection (Lowjaga et al., 2020). The inhibition profiles reported with various SLC transporters appear consistent with the TDI mechanism of OATP1B1, and trans-inhibition may be a common phenomenon for SLC transporters with different intracellular retention mechanisms of inhibitors, such as on-target (e.g., cyclophilin A) or off-target intracellular binding, lysosomal trapping, and/or active uptake of inhibitors. The in vitro experimental protocols and kinetic modeling approach employed in the current study may aid in elucidating the TDI mechanism of additional SLC transporters.
Another implication of the current findings may be for the risk assessment of OATP1B1-mediated DDIs. Regulatory DDI guidelines recommend that Ki values for OATP1B1 and OATP1B3 be determined after preincubation with NCEs (MHLW, 2019; U.S. FDA, 2020). Since the T1/2,max values depend on the PSdif and fT values of a given compound (Fig. 10 and Supplemental Fig. 4), preincubation time needs to be optimized for each compound. In the clinical DDIs, the inhibitor drugs are distributed to the liver and expected to access OATP1B1 from extracellular (plasma) and intracellular (liver) sides. In that regard, the use of steady-state Ki,app,cis+trans values may be more relevant to in vivo situations and desirable to avoid false-negative DDI predictions by the static model. However, it should be noted that the unbound concentrations of the inhibitor in the plasma and liver should change with time in vivo, and the use of steady-state Ki,app,cis+trans based on the unbound inhibitor concentration in buffer may overestimate the extent of in vivo DDI. To overcome this difficulty, the mechanistic PBPK models involving cis- and trans-inhibition taking into the account the time-dependent change in the unbound inhibitor concentrations in the blood and liver should be established in the near future.
In conclusion, we experimentally validated the underlying mechanism for preincubation-dependent, long-lasting inhibition of OATP1B1 by CsA, in which the extensive intracellular binding and cis- and trans-inhibition of OATP1B1 were found to be key drivers. OATP1B1 inhibitors with long T1/2,max, and large Ki,cis-to-Ki,trans ratio and Kp,uu could show a significant preincubation effect in the in vitro inhibition assays. This study provided a mechanistic understanding of the TDI of OATP1B1 in vitro, contributing to more accurate, quantitative prediction of transporter-mediated DDIs and developing safer medicines for patients.
Acknowledgments
The authors sincerely thank Soo-Jin Kim (Biome Research Laboratory, Korea Kolmar Holdings Co., Ltd.), Kota Toshimoto (Astellas Pharma, Inc.), and Takashi Yoshikado (Yokohama University of Pharmacy) for invaluable suggestions and technical support.
Authorship Contributions
Participated in research design: Izumi, Sugiyama.
Conducted experiments: Izumi.
Performed data analysis: Izumi, Nozaki.
Wrote or contributed to the writing of the manuscript: Izumi, Nozaki, Lee, Sugiyama.
Footnotes
- Received November 16, 2021.
- Accepted February 14, 2022.
↵This article has supplemental material available at dmd.aspetjournals.org.
Abbreviations
- A
- nonsaturable component for intracellular binding
- CsA
- cyclosporine A
- DDI
- drug-drug interaction
- E2G
- estradiol-17β-glucuronide
- fT
- intracellular unbound fraction
- Ibuffer
- concentration of inhibitor in buffer
- Icell
- total cellular concentration of inhibitor
- Kd
- dissociation constant
- Ki
- inhibition constant
- Ki,app
- apparent Ki value
- Ki,cis
- inhibition constant for cis-inhibition
- Ki,trans
- inhibition constant for trans-inhibition
- Kp,uu
- intracellular-to-buffer unbound drug concentration ratio
- NTCP
- Na+-taurocholate cotransporting polypeptide
- OATP
- organic anion transporting polypeptide
- PBPK
- physiologically based pharmacokinetics
- PSact
- clearance via transporter-mediated active uptake
- PSdif
- clearance via passive diffusion
- SLC
- solute carrier
- T1/2,max
- time to reach half maximum intracellular concentration
- TDI
- time-dependent inhibition
- Copyright © 2022 by The American Society for Pharmacology and Experimental Therapeutics