Characterization of Clofazimine as a Potential Substrate of Drug Transporter

ABSTRACT

In this study, we explored clofazimine (CFZ) as a potential substrate of uptake and efflux transporters that might be involved in CFZ disposition, using transporter gene overexpressing cell lines in vitro. The intracellular concentrations of CFZ were significantly increased in the presence of selective inhibitors of P-gp and BCRP, which include verapamil, cyclosporine-A, PSC-833, quinidine, Ko143, and daunorubicin. In a bidirectional transport assay using transwell cultures of cell lines overexpressing P-gp and BCRP, the mean efflux ratios of CFZ were found to be 4.17 ± 0.63 and 3.37 ± 1.2, respectively. The K_m and maximum rate of uptake (V_max) were estimated to be 223.3 ± 14.73 μM and 548.8 ± 87.15 pmol/min/mg protein for P-gp and 381.9 ± 25.07 μM and 5.8 ± 1.22 pmol/min/mg protein for BCRP, respectively. Among the uptake transporters screened, the CFZ uptake rate was increased 1.93 and 3.09-fold in HEK293 cell lines overexpressing OAT1 and OAT3, respectively, compared to the control cell lines, but no significant uptake was observed in cell lines overexpressing OCT1, OCT2, OATP1B1, OATP1B3, OATP2B1, or NTCP. Both OAT1- and OAT3-mediated uptake was inhibited by the selective inhibitors diclofenac, probenecid, and butanesulfonic acid. The K_m and V_max values of CFZ were estimated to be 0.63 ± 0.15 μM and 8.23 ± 1.03 pmol/min/mg protein, respectively, for OAT1 and 0.47 ± 0.1 μM and 17.81 ± 2.19 pmol/min/mg protein, respectively, for OAT3. These findings suggest that CFZ is a novel substrate of BCRP, OAT1, and OAT3 and a known substrate of P-gp in vitro.

INTRODUCTION

Multidrug-resistant tuberculosis (MDR-TB) is described as resistant to isoniazid and rifampin, the two most potent tuberculosis (TB) drugs. In recent decades, the increased incidence of MDR-TB has been a major focus for TB programs worldwide (1–4). MDR-TB requires the long-term use of second-line anti-TB drugs, which are less effective than first-line drugs and cause a range of adverse drug reactions in patients (5). The effectiveness of the drug depends on various cell processes that regulate the overall response of the drug, such as pharmacokinetics (PK) and pharmacodynamics. Membrane transporters play critical roles in the disposition of many drugs and xenobiotics in cells (6). The dysregulation of membrane transporter activity is one mechanism known to cause toxicity or treatment failure by altering absorption, distribution, metabolism, and excretion (7).

Clofazimine (CFZ) is a lipid-soluble riminophenazine antibiotic (8). Since 1962, it has been prescribed primarily for the treatment of multibacillary leprosy (9). However, this medication has long been neglected due to its side effects, such as skin pigmentation, conjunctivitis, phototoxicity, and gastrointestinal effects (10). Few studies have been conducted to determine the PK of CFZ in humans or the confounding factors affecting it, such as the activity of membrane transporters. An MDR-TB therapy regimen that included CFZ showed 87.4% relapse-free cure among 206 patients (11). These findings were reproduced in other clinical trials done in African countries (12, 13). There, hence, CFZ was listed as a second-line (group B) drug for the management of MDR-TB in 2016 World Health Organization (WHO) MDR-TB treatment guidelines (14, 15). In addition, there is an interesting report that CFZ showed a potential antiviral effect against SARS-CoV-2 by antagonizing RNA replication (50% effective concentration [EC₅₀], 0.31 μM) by the treatment of CFZ alone or in combination with remdesivir (RDV) in vitro (16).

From a clinical study reported 40 years ago, three metabolites (M1, M2, and M3) of CFZ have been identified from urine collected up to 24 h postdose (17, 18). Only 0.03% to 0.41% of the CFZ dose is excreted unchanged in the urine during the first 24 h after administration of 100 to 600 mg (19). CFZ is known to be highly bound to plasma proteins (99.9% to 99.96%) (20, 21) and accumulates extensively in fatty tissues, reticuloendothelial systems, and macrophages (22, 23). All of these properties appear to be responsible for CFZ’s enormous volume of distribution (1,470 L) (24), long half-life (up to 70 days), and limited daily renal excretion in humans (17, 19, 25–27). Approximately 11% to 59% of administered CFZ doses have been reported to be recovered in stool as unchanged, suggesting that CFZ seems to be excreted via the biliary route or incompletely absorbed in the intestine (17, 25). CFZ has also been reported to be accumulated in various tissues and is remarkably concentrated in bile (28), suggesting that biliary excretion is a major route of CFZ elimination and that CFZ is a potential substrate of drug transporters involved in the biliary excretion. According to the Biopharmaceutics Drug Disposition Classification System (BDDCS), CFZ is classified, as a group 2 drug with low solubility and high permeability, may interact with ATP-binding cassette (ABC) and solute carrier (SLC) transporters (29).

Although CFZ has the potential to be a substrate of drug transporters, limited information is available. In a previous study, CFZ was suggested to be a potential substrate of P-gp (30) and to have inhibitory potential on several ABC efflux transporters (7). As far we are concerned, the substrate potential of CFZ for SLC uptake transporters has not been studied. In this study, therefore, we comprehensively evaluated the substrate potential of CFZ for ABC and SLC drug transporters in vitro, which will be helpful to understand the disposition of CFZ mechanistically.

RESULTS

CFZ as a potential ABC transporter substrate.

Validation of the accumulation assay and cell functionality was done using fluorescent probe substrates and the corresponding inhibitors. The fluorescent intensity of the probe substrates (calcein-AM or BODIPY FL prazosin) increased markedly in cell lines overexpressing transporters in the presence of their corresponding inhibitors (verapamil against P-gp, cyclosporine-A against BSEP, MRP2 and daunorubicin against BCRP) relative to the intensity observed in the absence of the inhibitors. These results indicate that our experiment system, which included Lilly Laboratories cell and porcine kidney 1 (LLC-PK1) cell lines overexpressing P-gp and MDCKII cell lines overexpressing MRP2, BSEP, and BCRP, was adequately established (see Fig. S1 in the supplemental material).

The functionality of the bidirectional transport assay system was confirmed experimentally using probe substrates of P-gp and BCRP. [³H]-digoxin was used as a probe substrate for P-gp and had an efflux ratio (ER) of 14.27, comparable with other studies (31). Fluorescent BODIPY FL prazosin had an ER of 5.34 in MDCKII cell lines overexpressing BCRP monolayers, comparable with other studies (32). The representative inhibitors PSC-833 and Ko143 reduced the ERs to 2.24 and 0.58 in the MDCKII cell lines overexpressing P-gp and BCRP monolayers, respectively (Fig. S2).

To evaluate the substrate potential of CFZ for transporters, we used an intracellular accumulation of CFZ in cell lines overexpressing major ABC and SLC transporters and then confirmed the substrate potential in the presence of selective inhibitors or control cell lines. If any of the transporters showed significant uptake change in the presence of inhibitors or compared to the control cell lines used in the screening, then we did further experiments to estimate the transporter kinetics of CFZ for those transporters.

Initial screening indicated that the CFZ concentration in LLC-PK1 cell lines overexpressing P-gp treated with the known inhibitor verapamil was 2.7-fold higher than that in cell lines not exposed to the inhibitor. MDCKII cell lines overexpressing BCRP showed a 1.9-fold higher concentration of CFZ in the presence of the inhibitor Ko143; no significant change was observed in MDCKII cell lines overexpressing MRP2 and BSEP (Fig. 1). Multiple P-gp inhibitors (i.e., cyclosporine-A, quinidine, and PSC-833) were used to confirm the findings from the preliminary screening. Depending on the concentration used, cyclosporine-A, quinidine, and PSC-833 increased the CFZ concentrations 2.36- to 5.43-fold, 3.25- to 3.95-fold, and 1.35 to 10-fold, respectively, in LLC-PK1 cell lines overexpressing P-gp (Fig. 2A to D). Verapamil inhibited CFZ efflux in a concentration-dependent manner (Fig. 2D). CFZ was added at concentrations of 1 μM and 10 μM in the presence of the inhibitors Ko143 and daunorubicin, respectively, to determine whether it is a substrate for BCRP. Ko143 and daunorubicin increased the CFZ concentrations 1.87- to 2.41-fold and 1.74- to 2.23-fold, respectively, in MDCKII cell lines overexpressing BCRP (Fig. 2E, F).

FIG 1.

Screening for the substrate potential of CFZ in cell lines overexpressing ABC transporters. The intracellular accumulation of 10 μM CFZ was measured after 30 min of treatment in the LLC-PK1 cell lines overexpressing P-gp and MDCKII cell lines overexpressing BCRP, MRP2, and BSEP in the absence or presence of known selective inhibitors of those transporters, i.e., 50 μM verapamil against P-gp, 40 μM Ko143 against BCRP, and 40 μM cyclosporine-A against MRP2 and BSEP. The data are expressed as the mean ± standard deviation (SD) (n = 3; *, P < 0.05; **, P < 0.01; ***, P < 0.001).

FIG 2.

Effect of representative inhibitors on the P-gp- and BCRP-mediated CFZ efflux. The intracellular accumulation of 1 or 10 μM CFZ or both concentrations was measured after 30 min of treatment in the LLC-PK1 cell lines overexpressing P-gp in the absence or presence of known selective inhibitors (A) cyclosporine-A, (B) PSC-833, (C) quinidine, and (D) verapamil and MDCKII cell lines overexpressing BCRP in the absence or presence of known selective inhibitors (E) Ko143 and (F) daunorubicin. The data represent the mean ± SD (n = 3; *, P < 0.05; **, P < 0.01; ***, P < 0.001).

The transport kinetics of CFZ were estimated for P-gp and BCRP efflux transporters. CFZ produced an ER of 4.17 ± 0.63 and 3.37 ± 1.2 in the MDCKII cell lines overexpressing P-gp and BCRP monolayers, respectively. The representative inhibitors PSC-833 and Ko143 reduced these ERs to 1.9 ± 0.71 and 1.3 ± 0.49, respectively (Fig. 3A and C). The K_m and maximum rate of uptake (V_max) values for CFZ in MDCKII cell lines overexpressing P-gp and BCRP monolayers were 223.3 ± 14.73 μM and 548.8 ± 87.15 pmol/min/mg protein, respectively (Fig. 3B), and 381.9 ± 25.07 μM and 5.8 ± 1.22 pmol/min/mg protein, respectively (Fig. 3D). The estimated substrate kinetics (K_m and V_max) values of CFZ for P-gp and BCRP are summarized in Table 1.

FIG 3.

Net flux and concentration-dependent basolateral-apical (B-to-A) transport of clofazimine across MDCKII cell lines overexpressing P-gp and BCRP monolayer. The transport was measured at 15, 30, 45, and 60 min. (A) Net flux of 20 μM clofazimine into MDCKII cell lines overexpressing P-gp in the presence or absence of inhibitor. (B) Transport kinetics of clofazimine into MDCKII cell lines overexpressing P-gp. (C) Net flux of 20 μM clofazimine into MDCKII cell lines overexpressing BCRP in the presence or absence of inhibitor. (D) Transport kinetics of CFZ into MDCKII cell lines overexpressing BCRP. In panels B and D, the subtracted values of CFZ passive transport from the total transport are shown. AB and BA represent apical to basal and basal to apical directions, respectively. The data represent the mean ± SD (n = 3; *, P < 0.05; **, P < 0.01; ***, P < 0.001).

TABLE 1.

Substrate kinetic parameters of clofazimine (CFZ) for P-gp, BCRP, OAT1, and OAT3 drug transporters

Transporter	Substrate kineticsa			Efflux ratio (ER)
	Km (μM)b	Vmax (pmol/min/mg protein)c	Clint (μl/min/mg protein)d
MDCKII-P-gp	223.3 ± 14.73	548.8 ± 87.15	2.46 ± 0.13	4.17 ± 0.63
MDCKII-BCRP	381.9 ± 25.07	5.8 ± 1.22	0.02 ± 0.008	3.37 ± 1.2
HEK293-OAT1	0.63 ± 0.15	8.23 ± 1.03	13.12 ± 2.97
HEK293-OAT3	0.47 ± 0.1	17.81 ± 2.19	37.82 ± 4.28

^aSubstrate kinetics were derived from in vitro experiments and normalized with control.

^bK_m, Michaelis-Menten constant. Data are presented as the means ± SDs of the results from at least three independent experiments.

^cV_max, maximum rate of uptake.

^dCL_int, the intrinsic clearance per unit of time was calculated using the equation V_max/K_m.

CFZ as a potential SLC transporter substrate.

The cellular uptake of [³H]-TCA, [³H]-ES, [³H]-PAH, and [³H]-MPP was used to evaluate the function of HEK293 cell lines overexpressing uptake transporters. Cellular uptake rates of the probe substrates were more than 2-fold higher in HEK-OATP1B1, HEK-OATP2B1, HEK-OATP1B3, HEK-OAT1, HEKOAT3, HEK-OCT1, HEK-OCT2, and HEK-NTCP cell lines than in control cell lines. The respective transporter inhibitors (cyclosporine-A for OATPs, probenecid for OAT1, diclofenac for OAT3, verapamil for OCTs, and bromsulfthalein for NCTP) used in the experiments markedly decreased the cellular uptake rates of the probe substrates (Fig. S3). These findings indicated that the cell lines were functioning well.

SLC transporter screening was performed with control cell lines and compared with the uptake rate of CFZ in cell lines overexpressing uptake transporters. Among the transporters screened, including OATP1B1, OATP1B3, OATP2B1, OCT1, OCT2, OAT1, OAT3, and NTCP, the CFZ uptake rate was elevated only in HEK293 cell lines overexpressing OAT1 (1.93-fold increase) and OAT3 (3.09-fold increase) (Fig. 4A and B). Known inhibitors of these transporters (diclofenac, probenecid, and 1-butanesulfonic acid) significantly attenuated the CFZ uptake rate. CFZ uptake rates were reduced by 27%, 68%, and 21% in HEK293 cell lines overexpressing OAT1 and 69%, 56%, and 40% in HEK293 cell lines overexpressing OAT3 treated with diclofenac, probenecid, and 1-butanesulfonic acid, respectively (Fig. 5A and B).

FIG 4.

Screening for the substrate potential of CFZ in cell lines overexpressing SLC transporters. The uptake of 10 μM CFZ for 5 min into (A) HEK293 cell lines overexpressing OATP1B1, OATP1B3, OATP2B1, OAT1, and OAT3 and (B) HEK293 cell lines (FLP-IN) overexpressing OCT1, OCT2, and NTCP. Wild-type HEK293 cell lines were used as the negative control, and respective inhibitors were used for each representative transporter; 50 μM cyclosporine-A (CSA) against OATP1B1 and 1B3, 50 μM bromsulphthalein (BSP) against OATP2B1 and NTCP, 100 μM probenecid (PBA) against OAT1, 50 μM diclofenac (Diclo) against OAT3, and 100 μM verapamil (VPL) against OCT1 and OCT2 were used as representative inhibitors. The data represent the mean ± SD (n = 3; *, P < 0.05; **, P < 0.01; ***, P < 0.001).

FIG 5.

Effect of representative inhibitors on the uptake of CFZ in HEK293 cell lines overexpressing OAT1 and OAT3. Uptake of 10 μM CFZ for 5 min into control cell lines and HEK293 cell lines overexpressing transporters in the presence or absence of multiple inhibitors; 100 μM probenecid (PBA), 100 μM butasulphonic acid (BSA), and 100 μM diclofenac (Diclo). (A) CFZ uptake into HEK293-OAT1; (B) CFZ uptake into HEK293-OAT3. Wild-type HEK293 cell lines were used as the control. The data represent the mean ± SD (n = 3; *, P < 0.05; **, P < 0.01; ***, P < 0.001).

A range of CFZ concentrations (0 to 3 μM) was used to explore substrate kinetic values. Estimated K_m values of CFZ were 0.63 ± 0.15 μM for OAT1 (Fig. 6A) and 0.47 ± 0.1 μM for OAT3 (Fig. 6B). The estimated substrate kinetics (K_m and V_max) values for CFZ in cell lines overexpressing OAT1 and OAT3 are summarized in Table 1. The cellular uptake of CFZ was 2.4- and 4.7-fold higher in the HEK293 cell lines overexpressing OAT1- and OAT3 than in the control cell lines, respectively (Fig. S4).

FIG 6.

Concentration-dependent uptake of clofazimine in HEK293 cell lines overexpressing OAT1 and OAT3. The uptake of CFZ (0 to 3 μM) was measured for 5 min. (A) CFZ net uptake rate into HEK293-OAT1; (B) CFZ net uptake rate into HEK293-OAT3. Passive transport of CFZ in wild-type HEK293 cell lines was subtracted from the total CFZ transport (passive and active transport) in HEK293-OAT1 or OAT3, and the subtracted values are shown. The data represent the mean ± SD (n = 3).

DISCUSSION

In the present study, we evaluated the possible involvement of ABC and SLC membrane transporters in the disposition of clofazimine (CFZ). The results indicate that P-gp is responsible for CFZ efflux, consistent with previous findings (30). CFZ was identified as a novel substrate of BCRP, where representative BCRP inhibitors (Ko143 and daunorubicin) greatly increased its cellular accumulation. Furthermore, our results suggest that CFZ is a novel substrate of OAT1 and OAT3. These uptake transporters are well expressed in the kidney epithelial cell (33, 34).

Several reports describe that approximately 11% to −59% of administered CFZ doses are recovered in the stool as the parent compound, which indicates biliary excretion or poor absorption of CFZ in the intestine (17, 19, 25–27). Because of the high lipophilicity, a larger amount of CFZ was found in the fat tissue (5.3 mg/g) and bile (3.6 mg/g) than in the liver (3.2 mg/g), intestine (2.1 mg/g), kidney (1.4 mg/g), lung (1.2 mg/g), brain (<0.1 mg/g), and other tissues (28). The high concentration in the bile represents the biliary excretion of CFZ. For the biliary excretion of different endogenous and exogenous substances, the efflux transporters P-gp, BCRP, MRP2, and BSEP expressed in the bile canaliculi play vital roles (6, 35, 36). The efflux transporters P-gp and BCRP may be the major transporters involved in the biliary excretion of CFZ according to our in vitro evidence. As these two transporters are expressed in the apical region of the gut, they can reduce the absorption of their substrate (37). A negligible amount of CFZ was found in the brain tissue (28, 38). As P-gp and BCRP are abundant in the blood-brain barrier (BBB) (37), considering our result, these two transporters may play critical roles for the smaller amount of CFZ in the brain. We did not find CFZ as a substrate of MRP2 and BSEP efflux transporters in our study. CFZ was found to be excreted in the urine at 0.03% to 0.41% of a daily dose of 100 to 600 mg (19) and to have a long half-life (22, 25, 39, 40). CFZ accumulates in the kidney and excretes slowly (28). According to our in vitro results, OAT1 and OAT3 expressed in the kidney may be involved in the CFZ uptake.

Transporters are thought to have major impacts on BDDCS class 2 drugs, such as CFZ, due to efflux in the gut and efflux/uptake in the liver (29). Our in vitro results show that CFZ is a substrate for the efflux transporters P-gp and BCRP and the uptake transporters OAT1 and OAT3. As a BDDCS class 2 drug, CFZ is highly likely to be influenced by passive diffusion, as are tetracycline, minocycline, and doxycycline (41). Adding to that, clofazimine accumulates outside the central compartment due to its high lipophilicity, implying that intracellular concentrations at the transporter site are likely to be higher than those in the central compartment (7). Based on structure-activity relationships, pravastatin, acyclovir, and ganciclovir were reported to have high rates of passive diffusion after acting as substrates for SLC uptake transporters (34, 42). CFZ is a cationic drug (10) containing strong electronegative groups such as nitrogen and chlorine, which allows it to serve as a substrate for anion transporters, much like cimetidine (33, 43). Except for OAT1 and OAT3, CFZ was found to have negligible uptake in other uptake transporter overexpressing cells, indicating that OAT1 and OAT3 are major transporters of CFZ into cells.

Pharmacokinetic data on the clinical interactions of CFZ with ABC and SLC transporters has not been studied extensively. Recently, CFZ showed a very promising effect against SARS-CoV2 alone or in combination with remdesivir, where CFZ increased the potency of remdesivir approximately 20 times (16). Whether the mechanism of this synergistic effect is transporter-mediated or not is unknown. Remdesivir is a substrate of P-gp (44), and CFZ is a substrate and inhibitor of P-gp (7). One case report suggested that CFZ may influence the pharmacokinetics of phenytoin—a P-gp substrate. Phenytoin is a known anticonvulsant drug but is also used to treat TB meningitis with convulsions (45). Verapamil (a known P-gp inhibitor) altered the pharmacokinetics of phenytoin (a P-gp substrate) in one animal study (46). This type of effect is determined by the level of transporter expression as well as the drug concentration in the action site. Verapamil suppressed CFZ efflux in a concentration-dependent manner in our study, with a 50% inhibitory concentration (IC₅₀) value of 1.8 μM. These findings suggest that the transporter-mediated drug-drug interaction with CFZ could be clinically significant in TB treatment, where numerous medicines are required to address comorbidity. CFZ is a substrate of the P-gp, BCRP, OAT1, and OAT3 transporters that may aid in the development of more effective treatments.

The metabolites of CFZ could not be included in this study, which is one of the limitations of this study. Another is that we did not evaluate CFZ’s substrate potential for the multidrug and toxic compound extrusion transporter (MATE), which is expressed in the renal brush border membrane and plays critical roles in the excretion of drug-like cimetidine (47).

In summary, four ABC and eight SLC transporters were used to investigate the possible involvement of CFZ as a transporter substrate. In this study, we explored CFZ as a substrate of P-gp, BCRP, OAT1, and OAT3. These findings necessitate more research into human genetic variants in transporter genes to better understand CFZ disposition. This study provides a mechanical understanding of CFZ disposition, which can aid in the detection and prevention of transporter-mediated drug-drug interactions (DDI) of CFZ in vivo. Pharmacokinetics predictions can be made using a PBPK model with these data, along with the impact of a special population, ontogenesis, genetic polymorphism, ethnicity, and drug-drug interactions on drug exposure. Such information, like the data reported here, can help with prioritizing interaction studies, dose optimization studies, and safety/tolerability studies for specific TB medications or combinations of treatments throughout the early stages of clinical development.

MATERIALS AND METHODS

Chemicals and reagents.

Radiolabeled [³H]-estrone-3-sulfate ([³H]-ES; 57.3 Ci/mmol), [³H]-amino hippuric acid ([³H]-PAH; 4.56 Ci/mmol), [³H]-digoxin ([³H]-DG), and [³H]-taurocholic acid ([³H]-TCA; 5 Ci/mmol) were acquired from PerkinElmer (Waltham, MA, USA). Verapamil (VPL), cyclosporine-A (CSA), quinidine (QD), PSC-833, Ko143, probenecid, diclofenac, butanesulfonic acid, calcein-AM, clofazimine (CFZ), Dulbecco’s modified Eagle’s medium with 4.5 g/L glucose (DMEM), and fetal bovine serum (FBS) were purchased from Sigma-Aldrich (Burlington, MA, USA). BODIPY FL prazosin was purchased from Thermo Fisher Scientific (Waltham, MA, USA). Penicillin-streptomycin (PS) and nonessential amino acids (NEAAs) were purchased from Invitrogen Life Technologies (Grand Island, NY, USA).

Cell culture.

Human embryonic kidney 293 (HEK293), Lilly Laboratories cell and porcine kidney 1 (LLC-PK1), and Madin-Darby kidney II (MDCKII) cells were stably transfected as reported previously, including in our recent articles (48–50). Briefly, stably transfected HEK-OCT1, HEK-OCT2, HEK-OAT1, HEK-OAT3, HEK-OATP1B1, HEK-OATP2B1, HEK-OATP1B3, HEK-NTCP, LLC-PK1-P-gp, MDCKII-BCRP, MDCKII-BSEP, and MDCKII-MRP2 cells were harvested in tissue culture flasks using DMEM supplemented with 10% FBS, 1% NEAA, 2 mM l-glutamine, and 100 U/mL PS at 37°C in air supplemented with 5% CO₂.

Functional validation of cell lines overexpressing ABC and SLC transporters.

To validate the function of HEK293 cell lines overexpressing SLC transporters, we followed the method described previously (51). Briefly, we seeded 1.5 × 10⁵ cells/well in poly-d-lysine-coated 24-well culture plates and incubated it at 37°C in air supplemented with 5% CO₂ for 1 day. Before the experiments, cells were washed twice with phosphate-buffered saline (PBS) and incubated in Hanks’ balanced salt solution (HBSS) (Sigma-Aldrich, Burlington, MA, USA) containing 200 nM [³H]-TCA, 17 nM [³H]-ES, 200 nM [³H]-PAH, or 28 nM [³H]-MPP for 5 min at 37°C. The cells were washed three times with ice-cold PBS and lysed using 1% Triton X-100 for 30 min at room temperature on a shaker. Then the radioactivity in the lysate (1 mL) was measured using a liquid scintillation counter (PerkinElmer, Waltham, MA, USA).

Functional validation of LLC-PK1 or MDCKII cell lines overexpressing ABC transporters was done using fluorescent probe substrates and the corresponding inhibitors for the transporters. Cells were seeded in 96-well culture plates (0.8 × 10⁵ cells/well) and incubated at 37°C in air supplemented with 5% CO₂ for 1 day. The cells were washed twice with PBS and preincubated for 10 min in DMEM without FBS and PS. The cells were then incubated in the transport assay medium containing 0.5 μM calcein-AM or BODIPY FL prazosin with or without an inhibitor (verapamil against P-gp, cyclosporine-A against BSEP and MRP2, Ko143 and daunorubicin against BCRP) for 30 min at 37°C. Transport was halted with ice-cold PBS using proper aspiration techniques after 30 min. The fluorescent intensity of the probe substrates within the cells was detected using multimode microplate readers (SpectraMax iD3; Molecular Devices, LLC, San Jose, CA, USA). Cell function and the effects of transporter overexpression on cellular CFZ uptake and efflux were confirmed using the respective probe substrates according to previous reports, including our own (50, 52).

Transport assays for the screening substrate potential of CFZ for transporters.

The ability of CFZ to act as a substrate for ABC efflux transporters was screened using 24-well culture plates (1.5 × 10⁵ cells/well) and incubation at 37°C in air supplemented with 5% CO₂ for 1 day. The cells were washed twice with PBS and preincubated for 10 min in DMEM without FBS and PS. The cells were then incubated in the transport assay medium containing 1or 10 μM CFZ or both concentrations with or without an inhibitor for 30 min at 37°C. We used 1 to 100 μM of verapamil, cyclosporine-A, PSC-833, quinidine in LLC-PK1 cell lines overexpressing P-gp; Ko143 of daunorubicin in MDCKII cell lines overexpressing BCRP; and cyclosporine-A in MDCKII cell lines overexpressing MRP2 and BSEP. Transport was halted with ice-cold PBS using proper aspiration techniques after 30 min. Then the cells were collected, and samples were prepared for liquid chromatography-tandem mass spectrometry (LC-MS/MS) detection according to the reported method (53).

The ability of CFZ to act as a substrate for the HEK293 cell lines overexpressing SLC transporters was screened using poly-d-lysine-coated 24-well culture plates (1.5 × 10⁵ cells/well) and incubation at 37°C in air supplemented with 5% CO₂ for 1 day. Following 10 min preincubation, the cells were incubated in HBSS containing 10 μM CFZ with or without an inhibitor for 5 min at 37°C. We used cyclosporine-A (50 μM) against OATP1B1 and OATP1B3, bromsulphthalein (50 μM) against OATP2B1 and NTCP, probenecid (100 μM) against OAT1, diclofenac (50 μM) against OAT3, and verapamil (100 μM) against OCT1 and OCT2 as representative inhibitors. The cells were washed three times with ice-cold PBS and collected before detection.

The uptake assay was used for the kinetics study for the transporters found to be involved in the CFZ disposition from the screening, with a concentration range of 0 to 3 μM CFZ. Before performing these experiments, the linearity of cellular uptake with different concentrations and overtime was determined individually for each cell line to check the saturation (Fig. S5). Cellular uptake rates were determined by normalizing the incubation time and total protein content of the cell lines. The seeded cells (1.5 × 10⁵ cells/well) were lysed using 1% Triton X-100 for 30 min at room temperature on a shaker, and then protein content was quantified (only for the cells used in the kinetic study) using the Bradford protein assay (54). The net uptake rate was calculated as the difference between the uptake rates of transporter overexpressing cell lines and control cell lines (lacking transporter DNA transfection) for each concentration. To ensure that uptake occurred only with the selected transporters, wild-type cell lines were used as controls, and representative inhibitors were used for each transporter. To ensure data quality, three technical and biological replicates of each experiment were performed.

Bidirectional transport assays.

The bidirectional transport assays were performed to examine transepithelial transport of CFZ for P-gp and BCRP transporters, as described previously (55) with minor modifications. Shortly, the MDCKII cell lines overexpressing P-gp and BCRP were seeded (2.5 × 10⁵ cells/insert) on transwell filter membrane inserts (Costar, Corning, NY, USA) with a surface area of 1.12 cm² and a pore size of 0.4 μM. The cells were grown for 5 days at 37°C in air supplemented with 5% CO₂, and the medium was replaced every other day. The transepithelial resistance of each well was measured using a Millicell-ERS ohmmeter (Millipore, Bedford, MA, USA), and wells with ≥200 Ω resistance, after correction for the resistance obtained in the control blank wells (55), were used in the experiments. Confluent cells were preincubated for 30 min in HBSS containing 0.35 g NaHCO₃ and 1.95 g glucose. Transport measurements were initiated with the addition of 20 μM CFZ with or without an inhibitor to the transport assay medium on either the apical or basal side. Both compartments contained 0.5 mL and 1.5 mL HBSS, respectively. We used 10 μM PSC-833 and Ko143 as representative inhibitors in the MDCKII cell lines overexpressing P-gp and BCRP, respectively. CFZ transport was measured in the apical-basal (AB) and basolateral-apical (BA) directions. The receiver-side solution (300 μL) was withdrawn and replaced with an equal volume of transport assay medium at 15, 30, 45, and 60 min. An efflux ratio (ER) of ≥2 for CFZ for any efflux transporters are considered to have an active efflux (56). To explore the transport kinetics of CFZ for P-gp and BCRP, we used a concentration range of CFZ (0 to 1 mM). The transported concentration of CFZ was quantified using LC-MS/MS analysis. The seeded cells (2.5 × 10⁵ cells/insert) were lysed using 1% Triton X-100 for 30 min at room temperature on a shaker, and then protein content was quantified using the Bradford protein assay (54). Cell function and the effects of transporter overexpression on cellular CFZ efflux were confirmed using the probe substrates [³H]-digoxin and BODIPY FL prazosin, respectively.

Sample preparation and quantification of CFZ.

CFZ concentrations in cells were quantified using LC-MS/MS according to our established method, with minor modification (57). Briefly, after aspiration of the transport assay medium and washing with ice-cold PBS three times, cells were solubilized in 0.1 M sodium hydroxide (49). Then the cells were collected by adding 70% acetonitrile (100 μL) to a 1.5-mL tube, sonicating for 3 s, and centrifuging at 13,000 rpm for 10 min. After collection of supernatant, 100 nM linezolid (20 μL) was added to the supernatant as an internal standard and vortexed for 3 s before being transferred to LC-MS/MS vials for detection. Lysates (120 μL) were analyzed using the QTRAP 5500 LC-MS/MS system (AB Sciex LLC, Framingham, MA, USA). An Atlantic dC₁₈ column (150 by 2.1 mm, 3 μM; Atlantis, Waters, MA, USA) with a mobile phase of water and acetonitrile (30:70 [vol/vol]) containing 0.1% formic acid was used for the separation. The mobile phase was eluted at a flow rate of 300 μL/min with an Agilent Technologies 1200 series pump, and 2 μL of the samples was injected. Electrospray ionization in the positive mode was used to record mass spectra. The turbo ion spray interface was used at 4,000 V and 300°C.

Data and statistical analysis.

Kinetic values (K_m and V_max) were calculated using the Michaelis-Menten equation—V = (V_max × [S])/(K_m + [S]), where V indicates the uptake velocity of the probe substrate (pmol/min/mg protein), [S] is the substrate concentration in the experimental buffer (μM), K_m is the Michaelis-Menten constant (μM), and V_max is the maximum transport rate. The efflux ratio (ER) of CFZ was estimated by using the following equation: basolateral to apical P_app (B-A) CFZ transport/apical to basolateral P_app (A-B) CFZ transport. The fold change is calculated using basolateral to apical (B-A) CFZ transport/apical to basolateral (A-B) CFZ transport and basolateral to apical (B-A) CFZ transport/basolateral to apical (B-A) CFZ transport in the presence of inhibitor. Values obtained for each experiment are expressed as means ± standard deviations. Student’s t test was used, and a P value of <0.05 was considered to be significant. All statistical analyses were conducted using Prism software version 8 (GraphPad Software, San Diego, CA, USA).