Abstract
The bioavailability and targeted distribution of abacavir (ABC) and zidovudine (AZT) to viral reservoirs may be influenced by efflux transporters. The purpose of this study was to characterize the interaction of these nucleoside reverse transcriptase inhibitors with the Abcg2/Bcrp1 transporter, the murine homolog of human breast cancer resistance protein (BCRP), using a Bcrp1-transfected Madin-Darby canine kidney II cell model. Intracellular accumulation of ABC and AZT was significantly reduced by ∼90% and ∼70%, respectively, in Bcrp1-transfected cells compared with the wild-type cells. Both ABC and AZT showed significantly increased basolateral-to-apical (B-to-A) and decreased apical-to-basolateral (A-to-B) transport in Bcrp1 cells compared with wild-type directional flux. The efflux ratio (ratio of B-to-A to A-to-B) in Bcrp1-transfected cells was 22 for ABC and 11 for AZT. N-(4-[2-(1,2,3,4-tetrahydro-6,7-dimethoxy-2-isoquinolinyl)ethyl]-phenyl)-9,10-dihydro-5-methoxy-9-oxo-4-acridine carboxamide (GF120918) inhibited this difference in accumulation between the two cell variants with an EC50 of 1.32 ± 0.3 μM for ABC and 0.31 ± 0.1 μM for AZT. Potent and highly cooperative inhibition by Ko143 (3-(6-isobutyl-9-methoxy-1,4-dioxo-1,2,3,4,6,7,12,12a-octahydropyrazino[1′,2′:1,6]pyrido[3,4-b]indol-3-yl)-propionic acid tert-butyl ester) was observed with an EC50 of 121 ± 5 nM for ABC and 19.2 ± 1.5 nM for AZT (Hill coefficient ∼3–6). Probenecid, an organic anion inhibitor known to influence AZT biodistribution, had no effect on cellular accumulation in the Bcrp1 model. These studies characterize the Bcrp1-mediated transport of ABC and AZT and show that prototypical BCRP inhibitors GF120918 and Ko143 can inhibit the Bcrp1-mediated transport of these important antiretroviral compounds. The functional expression of BCRP at critical barriers, such as the intestinal enterocytes, brain capillary endothelium, and target lymphocytes, could influence the bioavailability and targeted delivery of these drugs to sanctuary sites.
Antiretroviral therapy (ART) combines three or more anti–human immunodeficiency virus (HIV)-1 compounds and has proven to be effective in reducing viral load and HIV-1-related mortality (Pomerantz and Horn, 2003). Zidovudine (AZT), a nucleoside reverse transcriptase inhibitor (NRTI), was the first drug to be approved for the treatment of HIV-1 (Ezzell, 1987). A number of NRTI, such as emtricitabine, zalcitabine, stavudine, lamivudine (3TC), didanosine, tenofovir, and abacavir (ABC), have since been approved. Standard combination therapy of HIV-1 infection often includes a dual NRTI “backbone” with the addition of a protease inhibitor or a non-NRTI (Werber, 2003). Both ABC and AZT (Fig. 1) are frequent components of combination therapies that have proved to be effective for HIV-1 treatment, for example, indinavir + AZT/stavudine + 3TC (Baker, 1997) and [abacavir + 3TC + AZT (Trizivir)] (Kessler, 2005).
HIV-1 has been shown to penetrate the central nervous system (CNS) and replicate in brain macrophages and microglia (Kure et al., 1991; Bagasra et al., 1996). Moreover, the CNS provides a sanctuary for the virus in the body as a result of insufficient antiretroviral drug delivery through the blood-brain barrier (BBB) (Pialoux et al., 1997). Numerous studies have implicated efflux transport proteins in the BBB in limiting the CNS distribution of some therapeutic agents (Polli et al., 2003; Park and Sinko, 2005). A number of drug efflux transporters have been identified and localized at the BBB, such as P-glycoprotein (P-gp) (Cordon-Cardo et al., 1990), multidrug resistance-associated protein (MRP) (Mohri et al., 2000), organic anion transporting polypeptide (Pizzagalli et al., 2002), and breast cancer resistance protein (BCRP)/ABCG2 (Cooray et al., 2002; Hori et al., 2004).
ABCG2/BCRP is a relatively new member of the ATP-binding cassette transporter superfamily discovered in mitoxantrone-resistant cell lines that do not overexpress P-gp or the MRP (Doyle et al., 1998). BCRP has been localized in the plasma membrane (Minderman et al., 2002) of cells forming various tissue barriers like the small intestines, colon, bile canaliculi, placental syncytiotrophoblast, and the BBB (Jonker et al., 2000; Cooray et al., 2002). The expression of rat Bcrp1 cDNA was found to be approximately 7-fold higher in the brain capillary fraction compared with the small intestine in rats, suggesting that this efflux protein may play a significant role in limiting brain distribution of its substrates (Hori et al., 2004). The cellular and tissue distribution and efflux transport function of BCRP may be critical in the absorption, distribution, and efficacy of several anti-HIV-1 agents and their passage across the BBB and blood-placental barrier into protected sites.
The CNS exposure of AZT and other NRTI has been found to be much lower in the brain compared with the blood concentrations. Active efflux transport out of the CNS is speculated to be a predominant mechanism limiting distribution of these NRTI to the CNS (Sawchuk and Yang, 1999). ABC has also been shown to have lower CNS exposure compared with plasma or whole blood in rats, guinea pigs, monkeys, and humans (Daluge et al., 1997; McDowell et al., 1999; Thomas et al., 2001). Currently, there are no reports that identify efflux transport proteins that may be involved in the absorption and distribution of ABC. Recent studies have shown that AZT has decreased cytotoxicity and anti-HIV-1 activity in CD4+ T cells that overexpress the wild-type and mutant variants of the efflux transporter BCRP (Wang et al., 2003). These BCRP-overexpressing cell lines were also shown to have a reduced accumulation of AZT, which was reversed by the BCRP inhibitor fumitremorgin C (Wang et al., 2003, 2004). However, the involvement of BCRP in the polarized transport of AZT is yet to be reported, and there are no reports regarding the interaction of ABC and BCRP or other efflux transporters.
The objective of this study was to characterize the interaction of ABC and AZT with murine Bcrp1, which is homologous to human BCRP. An in vitro Madin-Darby canine kidney (MDCK) II cell model transfected with Abcg2/Bcrp1 was used to assess this interaction with ABC and AZT. If BCRP is shown to be an important transporter of ABC and AZT at critical barriers, then modulation of BCRP-mediated efflux transport of ABC and AZT may lead to efficacious treatment and improved availability to specific sites of action, such as the CNS sanctuary and the target lymphocytes.
Materials and Methods
Materials. [14C]AZT and [3H]ABC were obtained from Moravek Biochemicals (Brea, CA). AZT and 3′-azido 2′,3′-dideoxyuridine were purchased from Sigma-Aldrich, Inc. (St. Louis, MO). ABC powder was provided by the National Institutes of Health AIDS Research and Reference Reagent program. GF120918 was provided by GlaxoSmithKline (Research Triangle Park, NC). Ko143 (Allen et al., 2002) was donated by Dr. Alfred H. Schinkel (The Netherlands Cancer Institute, Amsterdam, The Netherlands). GF120918 and Ko143 were solubilized in dimethyl sulfoxide (Sigma-Aldrich, Inc.) and diluted to desired concentration in assay buffer (122 mM NaCl, 25 mM NaHCO3, 10 mM glucose, 10 mM HEPES, 3 mM KCl, 1.2 mM MgSO4 · 7H2O, 1.4 mM CaCl2 · H2O, and 0.4 mM K2HPO4). The final concentration of dimethyl sulfoxide in all the solutions (including control group) was less than 0.1%. All the other reagents or solvents used were either analytical or high-performance liquid chromatography (HPLC) grade.
Cell Culture. MDCKII wild-type and Bcrp1-transfected cell lines (Jonker et al., 2000) were provided by Dr. Alfred H. Schinkel (The Netherlands Cancer Institute). Cells used for all our experiments were between passages 5 and 15. The cells were cultured in Dulbecco's modified Eagle's medium (Mediatech, Inc., Herndon, VA) supplemented with 10% fetal bovine serum (SeraCare Life Sciences, Inc., Oceanside, CA), penicillin (100 U/ml), and streptomycin (100 μg/ml) (Sigma-Aldrich, Inc.). All the other cell culture materials were obtained from BD Biosciences (San Jose, CA).
Intracellular Accumulation. The wild-type and Bcrp1-transfected cells were seeded at a density of 2 × 105/well and were grown for 2 to 3 days on 12-well plates (TPP tissue culture plates) to form confluent epithelial monolayers. For the experiment, the growth media were aspirated, and the cells in each well were washed twice with 2 ml of assay buffer, followed by preincubation for 30 min with 1 ml of assay buffer. The accumulation experiment involved incubation of these cells for 180 min at 37°C in an assay buffer (1 ml) containing tracer concentrations of the radiolabeled ABC or AZT. Assay buffer containing radiolabeled drug was aspirated at the end of incubation period, and the cells were washed three times with 1 ml of ice-cold phosphate-buffered saline. The cells were solubilized by adding 1 ml of 1% Triton X-100 (Sigma-Aldrich, Inc.). Total protein concentration in each well was determined by the BCA protein assay kit (Pierce Biotechnology, Inc., Rockford, IL), and the corresponding radioactivity was determined by liquid scintillation counting (LS-6500, Beckman Coulter, Inc., Fullerton, CA). Tracer accumulation in the two cell variants was compared, and the results were expressed as a percentage of control [total amount of radioactivity (dpm) accumulated per milligram of protein in the wild-type cell].
Directional Transport. The wild-type and Bcrp1-transfected cells were seeded at a density of 2 × 105/well and grown for 3 to 4 days on separate Transwell permeable supports (Corning Inc., Corning, NY) until they formed a confluent polarized monolayer. The upper compartment of the Transwell is the apical (A) side, and the lower compartment is the basolateral (B) side. For the transport experiment, growth medium was aspirated, and the cells were washed twice and preincubated with the assay buffer for 30 min. The assay buffer was then replaced with buffer containing tracer quantities of radiolabeled ABC or AZT (<100 nM) on the donor side. Fresh drug-free assay buffer was placed on the receiver side. The assay plates were incubated on an orbital shaker (60 rpm) at 37°C for the entire duration of the experiment except while sampling. Two hundred-microliter samples were drawn from the receiver side at 0, 30, 60, and 90 min and replaced with fresh drug-free assay buffer. Similarly, 200-μl samples were drawn from the donor side at 0 and 180 min and replaced with the assay buffer containing radiolabeled nucleotides at the initial donor concentration. The amount of compound transported was calculated using the specific activities of the radiolabeled nucleotides (0.6 Ci/mmol for ABC and 53 mCi/mmol for AZT) and compared in the wild-type and transfected cells. AZT and ABC directional transport experiments were also conducted in the presence of 50 μM unlabeled AZT or ABC to evaluate the saturability of [14C]AZT and [3H]ABC transport.
Inhibitory Effect of GF120913 and Ko143. Cellular accumulation experiments were conducted in the presence of inhibitors, where assay buffers containing GF120918 (5 μM), Ko143 (200 nM), or probenecid (100 μM) were used to preincubate the cells and for preparing solutions for the assay. The accumulation of ABC was determined at varying concentrations of GF120918 (0.3125, 0.625, 1.25, 2.5, 5.0, and 10 μM) and Ko143 (0, 25, 50, 100, 125, 150, 175, 200, and 400 nM). The accumulation of AZT was determined at varying concentrations of GF120918 (0, 0.005, 0.01, 0.025, 0.05, 0.1, 0.25, 0.5, 1.0, and 5.0 μM) and Ko143 (0, 0.001, 0.01, 0.1, 0.2, 1, 2, 10, 20, 100, and 200 nM). The data were analyzed by a sigmoid Emax model (eq. 1 using WinNonlin software version 5.0.1, Pharsight, Mountain View, CA): where effect is the -fold increase in accumulation, seen in the presence of inhibitor, over the Bcrp1 control (without inhibitor). E0 is the cell-accumulated dpm per milligram protein in the Bcrp1 control, normalized to equal unity. Emax is the -fold increase over Bcrp1 control at maximal inhibition; C is the concentration of the inhibitor in the incubating media; γ (gamma) is the shape factor (Hill coefficient) in the sigmoid model; and EC50 is the concentration of inhibitor at half-maximal inhibition.
Directional Transport Assay by HPLC. The directional transport assay was also conducted using unlabeled ABC or AZT to confirm the transport of intact nucleoside. The assay was performed as described above using unlabeled ABC and AZT at a concentration of 5 μM each. The samples were stored at –20°C until analysis. For analysis, 100-μl samples were spiked with 5 ng of AZT and 3′-azidouridine as internal standards for ABC and AZT, respectively. Then 800 μl of ethyl acetate was added, and the sample was vortexed vigorously for 10 min. The supernatant was dried under a flow of liquid nitrogen, reconstituted in 100 μl of mobile phase, and injected onto an HPLC system. The analysis was performed on a Hypersil-BDS column (C-18, 2.1 × 150 mm, 5 μM; Thermo Electron Corporation, Waltham, MA) maintained at 45°C using a Shimadzu (Kyoto, Japan) column oven (CTO-10Avp). The HPLC system consisted of a Shimadzu pump (LC-10ATvp), flow control valve (FCV-10ALvp), degasser (DGU-20A5), autoinjector (SIL-10ADvp), system controller (SCL-10Avp), and detector (SPD-10Avp). The flow rate of mobile phase [10 mM ammonium phosphate buffer, pH 4.6, and methanol (80:20)] was set at 0.2 ml/min, and UV absorbance was measured at 266 nM.
Permeability Calculations and Efflux Ratio. The effective directional permeabilities (Peff) of the nucleosides were calculated from the permeability equation [eq. 2 using slopes (dQ/dt)] obtained in the initial linear range from the amount transported versus time plots (for up to 90 min), where A is the area of the Transwell membrane, and C0 is the initial donor concentration. The efflux ratio was determined as the ratio of the Peff calculated in the B-to-A direction divided by the Peff in the A-to-B direction.
Statistical Analysis. Statistical analysis was performed using SigmaStat 3.1 (Systat Software, Inc., Point Richmond, CA). Groups were compared using simple one-way analysis of variance, and the Holm-Sidak method was used for the post hoc multiple comparison procedure with a significance level of p < 0.05.
Results
Intracellular ABC and AZT Accumulation. [3H]ABC accumulation was decreased significantly (∼90%; p < 0.001) in the Bcrp1-transfected cells compared with the wild-type cells. GF120918 (5 μM) and Ko143 (200 nM) could significantly increase ABC accumulation in the transfected cells. However, GF120918 significantly increased the observed levels in the Bcrp1-transfected cells, but inhibition did not completely reverse the efflux to reach levels equivalent as those seen in the wild-type cells. Probenecid (200 μM) had no significant effect on ABC accumulation in the Bcrp1-transfected cells and slightly reduced ABC accumulation in wild-type cells (Fig. 2A). [14C]AZT accumulation was decreased significantly (∼70%; p < 0.001) in the Bcrp1-transfected cells compared with the wild-type cells. GF120918 (5 μM) and Ko143 (200 nM) could increase AZT accumulation in the transfected cells to an equivalent level observed in the wild-type cells. Probenecid (200 μM) had no significant effects on AZT accumulation in the Bcrp1-transfected or wild-type cells (Fig. 2B).
Effect-Inhibitor Concentration Relationship: GF120918 and Ko143. Varying GF120918 and Ko143 concentrations were used to evaluate inhibitory effect on the Bcrp1-mediated [3H]ABC and [14C]AZT accumulation in the Bcrp1-transfected MDCKII cells (Fig. 3, A–D). Response (inhibitory effect) was measured as the -fold increase in accumulation observed in the presence of inhibitor compared with the Bcrp1 cells without inhibitor (control). GF120918 exhibited a maximal response at 5 μM with an Emax,EC50, and gamma of 4.1 ± 0.3-fold, 1.3 ± 0.3 μM, and 1.7 ± 0.5, respectively, for ABC (Fig. 3A), and 8 ± 0.5-fold, 0.31 ± 0.1 μM, and 0.75 ± 0.1, respectively, for AZT (Fig. 3B). Ko143 exhibited a maximal response at 200 nM with an Emax,EC50, and gamma of 7 ± 0.3-fold, 121 ± 5 nM, and 6.1 ± 1.4, respectively, for ABC (Fig. 3C), and 6.7 ± 0.21-fold, 19.2 ± 1.5 nM, and 2.73 ± 0.7, respectively, for AZT (Fig. 3D).
BCRP-Mediated Directional Flux of ABC and AZT. Both [3H]ABC and [14C]AZT showed significant differences in directional transport between the wild-type and Bcrp1-transfected cells (Figs. 4, 5, 6, 7). There was a significantly higher B-to-A flux and lower A-to-B flux of [3H]ABC in the Bcrp1-transfected cells compared with the wild-type cells (Fig. 4A). GF120918 and Ko143, known Bcrp1 inhibitors, decreased this difference in directional flux of ABC in the Bcrp1-transfected cell lines (Fig. 5, A and C). From these data, the effective directional permeabilities (Peff) of ABC were calculated as described under Materials and Methods. The efflux ratio of Peff in the Bcrp1-transfected cells was calculated to be approximately 22-fold greater than in the wild-type cells (Fig. 4B). GF120918 and Ko143 could reverse this difference in directional permeabilities in the two cell variants, and Ko143 at 200 nM was found to almost entirely inhibit this difference in directional permeability (Fig. 5, B and D). For [14C]AZT, the B-to-A flux was significantly higher, and the A-to-B flux was significantly lower in the Bcrp1-transfected cells compared with the wild-type cells (Fig. 4C). GF120918 and Ko143 significantly decreased this difference in directional flux of AZT in the Bcrp1-transfected cell lines (Fig. 6, A and C). From these data, the Peff of AZT were calculated as described under Materials and Methods. The efflux ratio of Peff in the Bcrp1-transfected cells was calculated to be approximately 11-fold greater than in the wild-type cells (Fig. 4D). Both GF120918 and Ko143 could almost entirely reverse this difference in directional permeabilities in the two cell variants (Fig. 6, B and D).
Directional Transport of Intact Nucleosides. Directional transport studies using unlabeled ABC and AZT measured by HPLC confirmed the transport of intact nucleosides in the tracer studies. For both unlabeled ABC and AZT, the B-to-A flux was significantly higher and the A-to-B flux was lower in the Bcrp1-transfected cells compared with the wild-type cells (Fig. 7, A and B). When plotted as a percentage of initial donor concentrations over time, there are similar profiles for both radiolabeled and unlabeled ABC and AZT (Fig. 7, A and B), and there was no difference in calculated permeabilities, indicating that for the experimental time and system, the radiolabel was a good indicator of intact ABC and AZT transport.
Discussion
Numerous studies have reported the CNS distribution of ABC, AZT, and other NRTI to be low, and it has been suggested that an active efflux transporter system is partially responsible for limiting CNS distribution of nucleosides (McDowell et al., 1999; Sawchuk and Yang, 1999; Thomas et al., 2001). ABCG2/BCRP is a recently discovered member of the ATP-binding cassette family of transport proteins. This efflux transporter was discovered in mitoxantroneresistant cell lines that do not overexpress P-gp or the MRP (Doyle et al., 1998). BCRP is localized in the plasma membrane of drugresistant tumor cells (Scheffer et al., 2000) and is found in barrier cells of various tissues like the small intestines, colon, bile canaliculi, and placental syncytiotrophoblast (Maliepaard et al., 2001), and it has been recently described in the endothelial cells of the cerebral microvasculature (BBB) (Cooray et al., 2002). These cellular and tissue expression sites indicate that BCRP may be important in the absorption, distribution, and elimination of its substrates. Moreover, recent studies have shown that human BCRP-overexpressing cells showed a diminished intracellular accumulation and anti-HIV activity of the NRTI AZT and 3TC (Wang et al., 2004). The increased resistance of the BCRP-overexpressing MT4 lymphocytes strongly suggested that the NRTI AZT is a substrate of human BCRP (Wang et al., 2004), leading to ineffective intracellular concentrations of AZT and its active metabolites. Therefore, if AZT and other anti-HIV1 nucleosides prove to be avid substrates for BCRP-mediated transport, this efflux transporter could be important in limiting oral absorption, altering distribution to sites of efficacy and toxicity such as the CD4-positive lymphocytes, and finally, influencing BBB penetration, which would limit entry into important viral sanctuary sites such as the CNS.
In the current studies, we used in vitro cell monolayers, both parental MDCKII cells and MDCKII cells transfected with Bcrp1 (Jonker et al., 2000), the murine homolog of human BCRP, to evaluate the role of BCRP in the directional transport of ABC and AZT. Using real-time polymerase chain reaction and Western blot analysis, we observed a significantly higher Bcrp1 gene mRNA and protein expression in transfected cells (data not shown). Importantly, these cells maintained a high expression of Bcrp1 without continuous positive selection via a cytotoxic substrate that could influence the expression of other protective efflux transporters, which could confound the results. The functional activity of this model was further evaluated by examining the cellular accumulation of mitoxantrone, a prototypical Bcrp1 substrate. Mitoxantrone cellular accumulation is greatly diminished in the Bcrp1-transfected cells, and accumulation can be completely restored by inhibition of Bcrp1 using Ko143 (data not shown). Therefore, from the gene expression and functional model validation studies, we can conclude that the differences in cellular accumulation or directional transport of compounds in these two variants of MDCKII cells (wild-type and Bcrp1-transfected cells) can be attributed to the efflux transporter Bcrp1.
There was a significant decrease in the accumulation of ABC and AZT (∼90% and ∼70%, respectively) in the Bcrp1-transfected cell line compared with the parental cells. This decrease in cellular accumulation was effectively reversed in the presence of Bcrp1 inhibitors GF120918 and Ko143 (Fig. 2, A and B). Addition of the potent Bcrp1 inhibitor Ko143 reversed the Bcrp1-mediated transport for both ABC and AZT to yield equivalent cellular accumulation as seen in the parental MDCKII cells. However, GF120918 was able to completely reverse this difference in accumulation of AZT but not for ABC. Ko143 did not significantly enhance AZT accumulation in the parental cells. Meanwhile, there was slight but significant reduction in ABC accumulation in the parental cells, which may be because of an inhibitory effect of Ko143 on its influx. These data indicate that the difference in cellular accumulation of ABC and AZT in these two cell types is mediated by Bcrp1 efflux, which strongly indicates that both ABC and AZT are substrates for Bcrp1. These data are the first reported evidence for the interaction of ABC with a specific efflux transporter or an active transporter of any type.
Given the strong evidence that ABC and AZT are Bcrp1 substrates, it was then useful to examine whether known inhibitors of NRTI active transport at critical sites can also inhibit Bcrp1. Probenecid-sensitive CNS distribution of AZT was previously reported in the rat (Takasawa et al., 1997), adult rhesus monkey (Cretton et al., 1991), and rabbit (Wong et al., 1993; Wang et al., 1997). The influence of probenecid on efflux transport of other NRTI, such as zalcitabine, didanosine, and tenofovir (Takasawa et al., 1997; Gibbs and Thomas, 2002; Mallants et al., 2005), has also been reported. Therefore, we investigated the effect of probenecid on Bcrp1 transport to evaluate whether the previously reported probenecid alterations of the CNS distribution of AZT and other NRTI could be in part mediated through Bcrp1 efflux transport. Our results (Fig. 2, A and B) show that probenecid does not influence the Bcrp1-mediated transport of either ABC or AZT. This finding strongly suggests the previously reported effects of probenecid on the BBB permeability of AZT involve a transport system other than Bcrp1, such as organic anion transporters in the MRP subfamily or the organic anion transport systems in the brain capillary endothelial cells (Sun et al., 2003).
These cellular accumulation studies were followed by directional flux studies for ABC and AZT. Compared with cellular accumulation studies, one of the advantages of the directional flux method is that it provides additional information about whether the compound is transported by a particular active transporter if one knows the localization and orientation of that transporter. The directional transport studies confirmed the role of this efflux protein in the polarized transport of ABC and AZT (Fig. 4). The calculated directional permeabilities, for both ABC and AZT, show that, given the known orientation of Bcrp1 in the transfected MDCKII cells (Mohrmann et al., 2005), there was the expected significant increase in the B-to-A transport over the A-to-B transport in the Bcrp1-transfected cells, which was significantly higher than in the wild-type cells. This Bcrp1-mediated transport of either nucleoside was not saturated even in the presence of a high concentration (50 μM ABC and AZT) of unlabeled nucleoside (data not shown). These concentrations are approximately 20-fold higher than normal therapeutic ABC and AZT concentrations in the plasma (Donnerer et al., 2003).
The directional transport of ABC and AZT in the Bcrp1-transfected cells could be modulated using Bcrp1 inhibitors GF120918 and Ko143, and Ko143 was able to almost completely abolish this transport for both ABC and AZT (Figs. 5C and 6C). However, GF120918 was able to completely abolish Bcrp1-mediated transport of AZT but not ABC (Figs. 5A and 6A). Concentration-dependent effects of these inhibitors on the accumulation of ABC and AZT in the transfected cells show that Ko143 is a more potent inhibitor for both ABC and AZT with an EC50 of approximately 121 ± 5 and 19.2 ± 1.5 nM, respectively, compared with approximately 1.32 ± 0.3 and 0.31 ± 0.1 μM for GF120918 (Fig. 3, A–D). Hill coefficients (gamma values) were much larger for Ko143 than GF120918 for both ABC and AZT (6.1 versus 1.7, 2.73 versus 0.75, respectively), which may imply the existence of cooperative inhibition. The mechanism of such an inhibition is unclear, and it is difficult to speculate based on model-fitted shape factors. However, it may be that Ko143 has a different binding site than GF120918 that is more sensitive to small changes in inhibitor concentration. More interestingly though, may be the possibility that Ko143 can exert its inhibitory effects through another mechanism that is not related to direct binding inhibition, but rather one that includes an intermediary signaling step, which then lends itself to a switch-like behavior, triggered by a minimum concentration of Ko143. It is known that there is cooperativity in the inhibition of the P-gp–mediated daunorubicin transport, and the authors indicate that this could be related to interactions between the homologous halves of the P-gp molecule (Wang et al., 2000). It is known that Bcrp1 transports as a homodimer or multimer, and as such could be subject to similar cooperative mechanisms of inhibition. However, it is not known whether other Bcrp1 substrates will give similar results as the two nucleosides under current study. Moreover, GF120918 did not completely inhibit the Bcrp-1-mediated accumulation of ABC. These results may explain the inability of GF120918 to be able to completely inhibit the difference in cellular accumulation and directional transport. Further studies are needed to evaluate the probable differences in mechanism of action, or binding sites, of these two inhibitors. The higher EC50 observed for ABC with both the inhibitors of Bcrp1 suggests that it may have a greater affinity for the efflux transporter compared with AZT.
Major obstacles for the successful treatment of HIV-1 infection are the emergence of drug-resistant strains and the failure of therapeutic agents, such as the NRTI, to reach the so-called viral sanctuary sites like the CNS (Sawchuk and Yang, 1999). Therefore, the eventual efficacy of ART depends not only on reducing the number of viral particles in the blood but also on the ability of these anti-HIV agents to reach viral reservoirs. P-gp and the MRP subfamily of multidrug resistance proteins have been widely investigated for their influence on the disposition of a diverse array of therapeutic compounds. There is some suggested involvement of MRP4 and MRP5 and not P-gp in AZT efflux (Jorajuria et al., 2004). Meanwhile, there is no direct evidence for the involvement of an efflux transporter limiting the CNS distribution of ABC. Candidate transporters that may influence absorption, distribution, and elimination of both ABC and AZT include some of the traditional organic anion transporters and ABCG2/BCRP.
Our results conclusively show that the directional transport of ABC and AZT is mediated by the efflux transporter Bcrp1. BCRP-mediated efflux may influence the oral bioavailability and possibly prevent the delivery of these important components of ART to the CNS, amniotic fluid, and lymphocytes. These results suggest the need for further investigation to determine the in vivo contribution of BCRP in the oral bioavailability and disposition of ABC, AZT, and other NRTI used in therapy. Other components of ART, like the protease inhibitors, such as ritonavir, saquinavir, and nelfinavir, have been found to be effective inhibitors of this efflux transporter (Gupta et al., 2004). Characterizing the role of BCRP in drug-drug interactions during the coadministration of BCRP substrates and/or inhibitors in combination therapy would be important follow-up studies for both the delivery of these NRTI to target sites and may result in the altered efficacy or toxicity of NRTI in AIDS therapy.
Acknowledgments
We thank GlaxoSmithKline for providing us with GF120918 and Dr. Alfred H. Schinkel from Netherlands Cancer Institute for providing MDCK-Bcrp1 cells and Ko143.
The following reagent was obtained through the National Institutes of Health AIDS Research and Reference Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases (NI-AID): abacavir, catalog no. 4680 from DAIDS, NIAID.
Footnotes
-
The project was supported by National Institutes of Health Grant NS42549. G.P. and N.G. contributed equally to this work.
-
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
doi:10.1124/dmd.106.014274.
-
ABBREVIATIONS: ART, antiretroviral therapy; HIV, human immunodeficiency virus; AZT, zidovudine (3′-azido-3′-deoxythymidine); NRTI, nucleoside reverse transcriptase inhibitor(s); 3TC, lamivudine; ABC, abacavir; CNS, central nervous system; BBB, blood-brain barrier; P-gp, P-glycoprotein; MRP, multidrug resistance-associated protein(s); BCRP, breast cancer resistance protein; MDCK, Madin-Darby canine kidney; GF120918, N-(4-[2-(1,2,3,4-tetrahydro-6,7-dimethoxy-2-isoquinolinyl)ethyl]-phenyl)-9,10-dihydro-5-methoxy-9-oxo-4-acridine carboxamide; Ko143, 3-(6-isobutyl-9-methoxy-1,4-dioxo-1,2,3,4,6,7,12,12a-octahydropyrazino[1′,2′:1,6]pyrido[3,4-b]indol-3-yl)-propionic acid tert-butyl ester; HPLC, high-performance liquid chromatography; A, apical; B, basolateral.
- Received December 8, 2006.
- Accepted April 13, 2007.
- The American Society for Pharmacology and Experimental Therapeutics