Rheumatoid arthritis (RA) is a chronic disease characterised by inflammation of the lining or the synovium of the joints.1 Three general classes of drugs are commonly used in the treatment of RA: non-steroidal anti-inflammatory drugs (NSAIDs), corticosteroids and disease-modifying antirheumatic drugs (DMARDs).2 3 One of the drawbacks is that, while NSAIDs and corticosteroids have a short onset of action, DMARDs can take several weeks or months to display a clinical effect. Moreover, owing to the long-term application of the drugs, many patients with RA demonstrate loss of efficacy over time.4^–7 Resistance to DMARDs may be a multifactorial event including enhanced drug efflux via ABC transporters, impaired drug uptake and drug activation and enhanced drug detoxification.8^–10

Members of the ATP binding casette (ABC) transporter superfamily are associated with a broad spectrum of physiological functions including detoxification, defence against xenobiotics and oxidative stress, absorption, distribution and excretion processes, and lipid metabolism.11 The multidrug resistance phenotype in tumours can be associated with the overexpression of certain ABC transporters, termed MDR proteins. P-glycoprotein (Pgp, MDR1, ABCB1)-mediated multidrug resistance was discovered first and is probably still the most widely observed mechanism in clinical multidrug resistance.12^–15 Two other ABC transporters have been found to participate in multidrug resistance of tumours: the multidrug resistance protein 1 (MRP1, ABCC1) and the breast cancer resistance protein (BCRP, MXR, ABCG2).16^–19

Several in vitro methods have been developed to study interactions of drugs with ABC transporters. Probably the most commonly used is the ATPase assay, performed on membrane vesicles and based on the fact that substrate transport is coupled with ATP hydrolysis. During the assay the amount of inorganic phosphate liberated from ATP is measured in a colorimetric reaction. The assay distinguishes between transported substrates and inhibitors. In the indirect vesicular transport assay, substrates are transported into inside-out vesicles and become trapped. During the assay the effect of a compound on the transport of a labelled reporter substrate is measured. The indirect vesicular transport assay detects only the interaction between the compound and the transporter, but does not reveal the nature of the interaction.

Interaction with the BCRP transporter can also be measured in a cellular dye efflux assay such as the Hoechst assay. Hoechst 33342 becomes fluorescent only in a complex with DNA.20 Cells expressing BCRP actively pump Hoechst 33342 out of the cell, thereby decreasing the intracellular dye concentration as this dye is an excellent substrate of the transporter. Inhibitors of BCRP inhibit the outward transport of Hoechst 33342. Substrates of the transporter compete with the dye, thus also reducing the rate of dye extrusion. Both cases result in an increased intracellular concentration of dye which can be detected as an increase in fluorescence due to the formation of a Hoechst 33342–DNA complex.21

As mentioned above, ABC transporters may have a role in drug resistance to DMARDs. Previous studies have shown that ABCB1 expression is higher in peripheral blood lymphocytes of patients treated with prednisolone and other DMARDs than in controls.22 23 Llorente and colleagues observed a difference in the expression of ABCB1 between patients who did not respond to treatment and those who did.24 Others found ABCC1 overexpression in chloroquine-resistant human CEM T cells.25 When human T lymphocytes were selected in the presence of sulfasalazine, the resistant population overexpressed BCRP and displayed lower sensitivity (fivefold) to leflunomide also.26 27 However, as the authors did not show that the phenomenon can be modelled in a transfected system, they did not explicitly identify BCRP as the only factor responsible for resistance to sulfasalazine and leflunomide.

Leflunomide is an isoxazole derivative DMARD which is a non-cytotoxic proliferation inhibitor of mitogen-stimulated T and B lymphocytes. It inhibits the dihydroorotate dehydrogenase, a key enzyme in the pathway for de novo synthesis of rUMP. Leflunomide is a prodrug that undergoes rapid non-enzymatic conversion to its active form, A771726.28^–30 Leflunomide is used either alone or in combination with methotrexate in the treatment of RA.31 32

BCRP is one of the most important efflux transporters in endothelial and epithelial cells, modulating absorption, distribution, metabolism and excretion properties of drugs and other xenobiotics. BCRP exhibits broad substrate specificity because it transports hydrophobic, anionic and cationic drugs.33^–38 BRCP has been shown to play a pivotal role in defensive barriers by preventing the penetration of drugs into the brain or the placenta and in the protection of stem cells from hypoxia.39^–41

Based on a previous publication suggesting a possible role for BCRP in resistance to leflunomide, in the present study we have focused on the interaction between the BCRP transporter and leflunomide and its metabolite. In vitro membrane-based methods (ATPase assay and vesicular transport assay) and cell-based methods (Hoechst assay and cytotoxicity assay) were designed to study these interactions. Methotrexate, another commonly used drug in RA, is also a substrate of BCRP.42 By using methotrexate as a reporter substrate in the vesicular transport assay, important drug-drug interactions can be revealed between drugs used in combination therapies and sharing a common target.

METHODS

Chemicals

A771726 (teriflunomide) was purchased from Alexis Corporation (Lausen, Switzerland); [³H]-methotrexate (specific activity 49.6 Ci/mmol) was purchased from Moravek Biochemicals (Brea, California, USA); Ko134 was a kind gift from Professor G J Koomen (National Cancer Institute, Amsterdam, The Netherlands);43 BXP-21 antibody was purchased from Abcam (Cambridge, UK); enhanced chemiluminescence (ECL) Western Blotting Detection System was purchased from GE Healthcare (Buckinghamshire, UK); Hank’s balanced salt solution (HBSS, 10×) without phenol red and Advanced RPMI 1640 were purchased from Gibco; Hank’s F12 medium and DMEM medium were purchased from Cambrex; and MTS reagent was purchased from Promega Corporation (Madison, USA). All other compounds were purchased from Sigma, unless otherwise indicated.

Membrane preparation

All BCRP preparations contained the wild-type (482R) form of the BCRP transporter (acc no NM_004827). Recombinant baculoviruses encoding wild-type human BCRP were gifts from Professor B Sarkadi (National Medical Center, Budapest, Hungary). All membrane preparations (BCRP-HAM-Sf9, defBCRP-HAM-Sf9, MDR1-Sf9) were obtained from Solvo Biotechnology (Szeged, Hungary; http://www.solvo.com). The insect membrane vesicle preparations were obtained using recombinant baculoviruses encoding wild-type human BCRP, defective BCRP-K86M mutant (carrying a mutation at a crucial position of the catalytic center of ATP binding and cleavage) and MDR1.21 Sf9 cells were cultured and infected with recombinant baculovirus stocks as described earlier. Purified membrane vesicles from virus-infected Sf9 cells were prepared essentially as described previously.44 Membrane protein content was determined using the BCA method (Pierce Biotechnology, Rockford, Illinois, USA).

Cell lines

HEK293 cells (obtained from the American Type Culture Collection, Rockville, Maryland, USA) and the BCRP overexpressing clone HEK293-BCRP (HEK293 transduction was carried out as described earlier45) were maintained in MIX MEM medium (Hank’s F12:DMEM, 1:1) supplemented with 10% (v/v) heat-inactivated fetal calf serum, penicillin (100 units/ml), streptomycin (100 μg/ml) and L-glutamine (2 mM) in a humidified atmosphere of 5% carbon dioxide in air at 37°C. PLB985 cells, a human myelomonoblastic leukaemia cell line PLB98546 and the BCRP overexpressing clone PLB985-BCRP45 were cultured in Advanced RPMI 1640 medium supplemented with 10% (v/v) heat-inactivated fetal calf serum, penicillin (100 units/ml), streptomycin (100 μg/ml) and L-glutamine (2 mM) in a humidified atmosphere of 5% carbon dioxide in air at 37°C.

Western blotting

Protein expression was confirmed by SDS-PAGE and subsequent western blotting using specific anti-ABCG2 antibody BXP-21, HRP-conjugated anti-mouse secondary antibody and enhanced chemiluminescence (ECL) as described elsewhere.50

Vesicular transport assay

Vesicular transport studies were performed as described elsewhere.50 BCRP-HAM-Sf9 membrane preparation was used for the experiments (SOLVO Biotechnology, Szeged, Hungary). If not stated otherwise in the figure legend, the concentration of methotrexate was 100 μM.

ATPase activity

ATPase activity was measured as described previously.44 The PREDEASY BCRP-HAM-ATPase kit, PREDEASY defBCRP-HAM-ATPase kit and PREDEASY MDR1-ATPase kit (SOLVO Biotechnology) were used for the determination of BCRP-ATPase and MDR1-ATPase activity according to the manufacturer’s instructions.

Dye transport assay

The protocol is a modified version of the one described by Ozvegy et al.21 Accumulation of Hoechst 33342 dye was measured in a fluorescence spectrophotometer (Fluoroskan Ascent Type 374) at 350 nm (excitation)/460 nm (emission) by using PLB985-BCRP in 150 μl of HBSS (1×), pH 7.4. The BCRP-expressing PLB985 cells were preincubated at 37°C in 1× Hank’s solution with drugs for 30 min. The Hoechst dye was added in 50 μl, final concentration 12.5 μΜ. For maximum inhibition of the BCRP protein, 300 nM Ko134 was applied in each experiment. The fluorescence intensities were recorded for 15 min. Using the slope of the curve determined without inhibitors (Rbase), the slope of the curve in the presence of 300 nM Ko134 (Rmax) and the slope of the curve determined for any drug at the given drug concentration (Rdrug), the percentage inhibition of Hoechst 33342 extrusion can be represented by the following formula:

IC₅₀ values were derived from these curves.

Cytotoxicity assay

Cytotoxicity assays were performed by seeding HEK293 and HEK293-BCRP cells at a density of 4000 cells per well in 96-well plates containing the culture medium (200 μl/well). After 24 h, drugs were prediluted in medium and added to the cells at different concentrations. The cells were further incubated with the drug in a humidified tissue culture chamber (37°C, 5% carbon dioxide) for 96 h. Surviving cells were detected by the MTS method (www.promega.com). IC₅₀ values were calculated from dose-response curves (ie, cell survival versus drug concentration) obtained in triplicate experiments.

Data analysis

The vesicular transport assay, ATPase assay and Hoechst assay were run in duplicates. Data are presented as mean (standard deviation, SD).

In the case of the vesicular transport assay, K_I values were determined using the Cheng-Prusoff equation:47

where K_I is the affinity of the inhibitor, IC₅₀ is the concentration of the inhibitor which inhibits 50% of the transport, S is the concentration of the substrate and K_m is the affinity of the substrate. The concentration of substrate was below K_m.

The potencies of drugs to alter ATPase activity were obtained from plots of the rate of ATP hydrolysis as a function of the logarithm of drug concentration by non-linear regression of the general sigmoid dose-response equation:

where v is the response (nmol Pi/min/mg), V_min is the minimal response, V_max is the maximal response, EC₅₀ is the ligand concentration producing 50% of the maximal response (efficacy), A is the actual test drug concentration and the Hill slope is the parameter characterising the degree of cooperativity. EC₅₀ is defined according to the International Union of Pharmacology Committee.48

In the cytotoxicity assay, each concentration was tested in triplicate. IC₅₀ values were calculated from dose-response curves obtained from four separate experiments.

For curve fitting, V_max,K_m, K_I and IC₅₀ calculations, PRISM 3.0 software was used (GraphPad Software, San Diego, California, USA).

RESULTS

Interaction of leflunomide and A771726 with BCRP

To examine whether the drugs interacted with BCRP, their effect on methotrexate transport was investigated. Both leflunomide and A771726 dose-dependently inhibited the BCRP-mediated transport of methotrexate into inside-out vesicles (fig 1A, B). The K_I values, which indicate the potencies of leflunomide and A771726, were 1.86 μM and 0.093 μM, respectively. Thus, leflunomide had a lower affinity than A771726 in the vesicular transport assay. The experiment was repeated at various methotrexate concentrations and the curves were linearised to obtain a Dixon plot.49 This representation (fig 1C, D) shows that there is competitive inhibition between methotrexate and the two drugs investigated.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

ATP-dependent methotrexate transport into the breast cancer resistance protein (BCRP) transporter containing membrane vesicles in the presence of (A) leflunomide and (B) A771726. Dixon-Webb plots of competition studies of methotrexate uptake by (C) leflunomide and (D) A771726. BCRP-mediated uptake of 25 μM (▾), 50 μM (▾) and 100 μM (▪) methotrexate was carried out at 37°C for 4 min. In our experiments generally 2000 cpm corresponds to the 100% value. The amount of radiolabelled compounds retained by the inside-out vesicles was measured at different concentrations of the drugs as indicated in the figure.

Similar results were obtained using the cellular assay system. To test the interaction of drugs with the transporter in a cellular assay, BCRP-overexpressing PLB895-BCRP cells were used (fig 2A). Both drugs inhibited the BCRP-mediated efflux of Hoechst 33342 dye from PLB985-BCRP cells. Figure 2B and C show the concentration-dependent inhibition of the test drugs using equation 1. The IC₅₀ values for leflunomide and its metabolite are within the same order of magnitude (4.53 and 2.87 μM, respectively). A771726 displayed a maximum-type bimodular curve. The IC₅₀ was calculated from a function fitted to the first part of the curve showing a dose-dependent inhibition.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

(A) Breast cancer resistance protein (BCRP) transport protein expression in parental PLB985 and PLB985-BCRP cells. Effect of (B) leflunomide and (C) A771726 on BCRP-mediated Hoechst 33342 dye efflux using PLB985-BCRP cells. Closed squares indicate the inhibitory effect of Ko134, a specific BCRP inhibitor; closed triangles indicate (B) leflunomide or (C) A771725. Cells were incubated in the presence of different concentrations of the drugs at 37°C for 30 min. The reactions were started by adding 12.5 μM Hoechst 33342 and fluorescence intensities were recorded for 15 min.

Leflunomide and A771726 as substrates of BCRP

To determine whether leflunomide and A771726 are substrates or inhibitors of BCRP, ATPase measurements were carried out. Substrates transported at a high rate turn on the ATPase, and thus stimulation of ATPase is indicative of the nature of the interaction.50 51

The potencies of leflunomide and A771726 to modulate basal ATPase activity of BCRP were estimated by ATPase assay using BCRP-containing membrane vesicles. Figure 3 shows the concentration-dependent stimulation of vanadate-sensitive BCRP ATPase activity, where inorganic Pi production was measured. Both compounds stimulated vanadate-sensitive ATPase activity of BCRP in a dose-dependent manner. The calculated EC₅₀ values for leflunomide and A771726 were 3.93 μM and 0.78 μM, respectively. The control and MDR1-containing vesicles could not be stimulated with leflunomide, nor with A771726 even at high concentrations.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Vanadate-sensitive ATPase activity of (A, B) BCRP-HAM-Sf9, (C) defBCRP-HAM-Sf9 and (D) MDR1-Sf9 membrane preparations in the presence of leflunomide (closed squares) and A771726 (closed triangles) at different concentrations in activation (solid lines) and inhibition (dotted lines) experiments.

Resistance of cells overexpressing BCRP to leflunomide and A771726 cytotoxicity

Empty vector-transduced cells (mock) and BCRP-transduced clones (fig 4A) were treated for 4 days with leflunomide and A771726 and their cytotoxic effect was measured by the survival rate of the cells. The results are summarised in table 1 and representative curves for the tested compounds are shown in fig 4B and C. HEK293 and HEK293-BCRP cells were incubated with leflunomide or A771726 at different concentrations with and without 1 μM Ko134 at 37°C for 96 h. Overexpression of the BCRP transporter conferred resistance to leflunomide and A771726 by 20.6-fold and 7.5-fold, respectively, compared with mock-transfected HEK293 cells. This resistance could be reversed by the specific BCRP inhibitor Ko134, underlining the role of BCRP. HEK293-mock and HEK293-BCRP cells were exposed to varying concentrations of leflunomide and A771726 for 96 h. IC₅₀ values of cell viability were calculated from the dose-response curves (ie, cell survival vs drug concentration) obtained in triplicate experiments.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

(A) Expression of breast cancer resistance protein (BCRP) transporter in HEK293-mock and HEK293-BCRP cells. Cytotoxic effect of (B) leflunomide and (C) A771726 on HEK293-mock and HEK293-BCRP cells. Open triangles, HEK293-mock cells; closed triangles, HEK293-mock cells + 1 µM Ko134; open squares, HEK293-BCRP cells; closed squares, HEK293-BCRP cells + 1 µM Ko134. The cells were incubated with the drugs for 96 h in the presence or absence of 1 µM Ko134.

DISCUSSION

Different membrane-based and whole cell-based assays were performed to investigate and characterise the interaction between the transporter BCRP and leflunomide and its metabolite A771726. Both compounds were found to interact with BCRP and are transported substrates of the transporter based on several facts: both compounds activated the BCRP transporter in the ATPase assay, both proved to be competitive inhibitors of methotrexate transport by BCRP and their cytotoxic effects could be reversed in the presence of the BCRP-specific inhibitor Ko134.

BCRP was shown to be important in the defence mechanism of immunocompetent cells such as stem cells or monocyte-derived dendritic cells.45 52 BCRP could play an important role in inflammatory processes because it was shown to be expressed on endothelial cells and macrophages in the synovial sublining of patients with RA. It was suggested that the expression of this ABC transporter is inflammation-dependent rather than drug-induced.53 Moreover, previous studies have indicated that BCRP is important in the absorption and elimination of other DMARDs such as sulfasalazine and methotrexate.54^–56

Methotrexate is a known BCRP substrate which is commonly used in vesicular transport studies.57 This inhibition-type membrane-based assay is suitable for studying the effect of the test drug on the accumulation of a transported substrate. By using a relevant drug molecule as a reporter substrate, potentially important drug-drug interactions can be detected in a simple in vitro system. Both leflunomide and A771726 inhibited the accumulation of methotrexate into inside-out BCRP-HAM-Sf9 vesicles in a dose-dependent manner, but with different affinities (fig 1A, B). Leflunomide inhibited methotrexate transport with higher K_I (1.86 μM) than A771726 (0.093 μM). Furthermore, the Dixon plots (fig 1C, D) showed that the mechanism of action between methotrexate and leflunomide or A771726 is competitive inhibition.

This difference in kinetic parameters was also observed in the ATPase assay. The modulation of baseline ATPase activity may indicate the substrate nature of a compound. Leflunomide and A771726 stimulated the vanadate-sensitive ATPase activity of the BCRP transporter in the ATPase assay with EC₅₀ values of 3.93 μM and 0.78 μM, respectively (fig 3).

As noted earlier, Hoechst 33342 is a membrane permeable dye which is an excellent substrate of BCRP. This cellular assay is suitable for following the interaction between the BCRP transporter and test drugs. Both substrates and inhibitors increase the cellular accumulation of the dye through interaction with BCRP. Leflunomide and its metabolite inhibited the Hoechst efflux with nearly the same IC₅₀ values (4.53 μM and 2.87 μM, fig 2).

We performed 96 h cytotoxicity assays using empty vector-transfected HEK293 cells and BCRP transporter-expressing clones. HEK293-BCRP-expressing cells showed 20.6-fold and 7.5-fold resistance with leflunomide and A771726 compared with mock HEK293 cells. This increased resistance is attributable to the BCRP transporter because it could be reversed by Ko134, a known specific BCRP inhibitor (fig 4). This assay set-up is considered a surrogate transport assay as it studies directly the involvement of a specific transporter and its role in resistance to the drug tested.

In contrast to membrane-based methods, no differences were seen in the affinities of the two compounds in cell-based assays. This is probably due to the different passive permeability parameters of the compounds; for the Hoechst assay and cytotoxicity assay, the drugs have to enter the cells in order to interact with the transporter. A771726 has an open ring structure with a hydroxylic group, so its passive permeability is likely to be lower than that of leflunomide.

It has been suggested previously that other ABC transporters such as ABCB1 and ABCC1 may also contribute to DMARD resistance.22^–25 We have shown that neither leflunomide nor its active metabolite is capable of stimulating the basal vanadate-sensitive activity of ABCB1 (MDR1, Pgp) transporter in the ATPase assay. This indicates that MDR1 is most probably not involved in the development of resistance to leflunomide and A771726. The compounds were also tested for potential interactions with the MRP1 transporter; neither leflunomide nor A771726 showed an interaction with this transporter (data not shown), so the interaction of leflunomide and A771726 with BCRP is highly specific.

Leflunomide is a commonly used new DMARD applied in monotherapy or in combination therapy with NSAIDs. Moreover, it is used to treat active RA in combination with other DMARDs such as methotrexate, sulfasalazine, infliximab, adalimumab, etanercept and anakinra.58^–62 Leflunomide is used in combination therapy with methotrexate in case resistance to methotrexate develops. Several studies have confirmed that the administration of leflunomide with methotrexate can lead to an improvement in the patient’s condition.30 31 This may be due, at least in part, to the competitive inhibition between methotrexate and leflunomide or A771726 leading to increased local concentrations of one or other drug. No unexpected adverse effects were observed and the most common side effect was an increase in the liver transaminase enzyme level which is also observed with leflunomide monotherapy.30 Of patients treated with methotrexate in combination with leflunomide, 56.9% achieved ACR20,63 which might be indirect proof that the efficacy of either drug is increased in the presence of the other, and this may be due to their interaction with BCRP. Our findings may form the basis of new therapeutic approaches such as the local administration of BCRP-reverting agents or inhibitors in combination with DMARDs.