Abstract
A synergistic effect of P-glycoprotein (P-gp)/Abcb1a and breast cancer resistance protein (Bcrp)/Abcg2 was reported to limit the brain penetration of their common substrates. This study investigated this based on pharmacokinetics using Mdr1a/1b(−/−), Bcrp(−/−), and Mdr1a/1b(−/−)/Bcrp(−/−) mice. Comparison of the brain- and testis-to-plasma ratios (Cbrain/Cplasma and Ctestis/Cplasma, respectively) of the reference compounds quinidine and dantrolene for P-gp and Bcrp, respectively, indicates that impairment of either P-gp and Bcrp did not cause any change in the efflux activities of Bcrp or P-gp, respectively, at both the blood-brain barrier (BBB) and blood-testis barrier (BTB). The Cbrain/Cplasma and Ctestis/Cplasma of the common substrates erlotinib, flavopiridol, and mitoxantrone were markedly increased in Mdr1a/1b(−/−)/Bcrp(−/−) mice even compared with Mdr1a/1b(−/−) and Bcrp(−/−) mice. Efflux activities by P-gp and Bcrp relative to passive diffusion at the BBB and BTB were separately evaluated based on the Cbrain/Cplasma and Ctestis/Cplasma in the knockout strains to the wild-type strain. P-gp made a larger contribution than Bcrp to the net efflux of the common substrates, but Bcrp activities were also significantly larger than passive diffusion. These parameters could reasonably account for the marked increase in Cbrain/Cplasma and Ctestis/Cplasma in the Mdr1a/1b(−/−)/Bcrp(−/−) mice. In conclusion, the synergistic effect of P-gp and Bcrp on Cbrain/Cplasma and Ctestis/Cplasma can be explained by their contribution to the net efflux at the BBB and BTB without any interaction between P-gp and Bcrp.
It is well accepted that the penetration of xenobiotic compounds into the brain and testis is restricted by the blood-brain barrier (BBB) and blood-testis barrier (BTB), respectively. The BBB is formed by brain capillary endothelial cells, whereas, in addition to endothelial cells, myoid and Sertoli cells form the BTB (Bart et al., 2002; Kusuhara and Sugiyama, 2005). Tight junctions between adjacent cells in the BBB and BTB are highly developed and limit the penetration of substances via the paracellular route. Moreover, drug transporters act as active barriers to limit the tissue penetration of substrates from the blood by extruding them back into the blood in the BBB and BTB and, thereby, modulating pharmacological or adverse reactions. It has been shown that ATP binding cassette (ABC) transporters, which are known to mediate resistance to anticancer drugs and antiviral drugs, are expressed in the BBB and BTB. These include P-glycoprotein (P-gp/MDR1/ABCB1), breast cancer resistance protein (BCRP/ABCG2), multidrug resistance-associated protein (MRP)-1/ABCC1, MRP2/ABCC2, MRP4/ABCC4, and MRP5/ABCC5 (Leggas et al., 2004; Zhang et al., 2004; Lee et al., 2005). In particular, P-gp is a well known transporter that plays a pivotal role in barrier function, and disruption of the Mdr1a gene, a predominant isoform expressed in the barriers, causes accumulation of a number of its substrates (Schinkel, 1999; Scherrmann, 2005).
Recently, we demonstrated that Bcrp also acts as an active barrier in both the BBB and BTB. Disruption of the Bcrp gene causes a significant increase in the accumulation of isoflavonoids, drugs (dantrolene, prazosin, and triamterene), and food-derived carcinogens in the brain and testis without affecting systemic exposure (Enokizono et al., 2007, 2008). According to Adachi et al. (2001), the ratio of the brain-to-plasma ratio in Bcrp(−/−) mice to wild-type mice represents Bcrp activity in the BBB when only Bcrp accounts for the active efflux. However, the ratio showed rather negative correlation to in vitro Bcrp activities (Enokizono et al., 2008). We hypothesized that Bcrp activity at the BBB is underestimated because of P-gp-mediated efflux based on the in vitro finding that some Bcrp substrates were also found to be P-gp substrates (Enokizono et al., 2008). Later, Oostendorp et al. (2009) demonstrated that imatinib, a common substrate of P-gp and Bcrp, exhibits a considerable increase in the brain-to-plasma ratio in mice lacking both P-gp and Bcrp, Mdr1a/1b(−/−)/Bcrp(−/−) mice, and in FVB mice treated with 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], an inhibitor of both Bcrp and P-gp. The increase observed in Mdr1a/1b(−/−)/Bcrp(−/−) mice appears to be a synergistic effect considering the increase observed in Mdr1a/1b(−/−) or Bcrp(−/−) mice. In addition to imatinib, the same synergistic effect was observed for dasatinib, flavopiridol, lapatinib, and prazosin (Chen et al., 2009; Lagas et al., 2009; Polli et al., 2009; Zhou et al., 2009). Because Cisternino et al. (2004) reported an induction of Bcrp mRNA in the brain capillaries from Mdr1a/1b(−/−) mice, induction of Bcrp or P-gp in Mdr1a/1b(−/−) or Bcrp(−/−) mice could be the underlying mechanism. However, this remains controversial because de Vries et al. (2007) reported similar expression of Bcrp in the brain of wild-type and Mdr1a/1b(−/−) mice. The equations for the Kp,brain that we derived by taking the active efflux mediated by both P-gp and Bcrp into consideration could reasonably explain such a synergistic effect in the increase in Mdr1a/1b(−/−)/Bcrp(−/−) mice without consideration of any interplay between P-gp and Bcrp (Kusuhara and Sugiyama, 2009). The present study was undertaken to support this kinetic consideration based on in vivo experiments using wild-type, Mdr1a/1b(−/−), Bcrp(−/−), and Mdr1a/1b(−/−)/Bcrp(−/−) mice. In addition to the BBB, we demonstrated that Bcrp limits the penetration of xenobiotic compounds into the testis (Enokizono et al., 2008). Because P-gp and Bcrp are colocalized on the luminal side of the endothelial cells in the testis and the apical side of the myoid cells in the testis (Melaine et al., 2002; Bart et al., 2004; Lee et al., 2005; Enokizono et al., 2007), it is possible that the synergistic increase in the accumulation of xenobiotic compounds is also observed in the testis. Therefore, the present study also determined the concentrations of drugs in the testis of Mdr1a/1b(−/−)/Bcrp(−/−) mice.
Materials and Methods
Materials.
Erlotinib (Tarceva) was purchased from Toronto Research Chemicals Inc. (North York, ON, Canada), and flavopiridol [Alvocidib, HMR-1275, (−)cis-5,7-dihydroxy-2-(2-chlorophenyl)-8-(4R-(3S-hydroxy-1-methyl) piperidinyl)-4H-1-benzopyran-4-one] was kindly supplied by sanofi-aventis (Bridgewater, NJ). Dantrolene (Dantrium) and mitoxantrone (Novantrone) were purchased from LKT Labs (St. Paul, MN). Quinidine was purchased from Tokyo Kasei (Tokyo, Japan). All other chemicals were commercially available and of reagent grade.
Animals.
Male wild-type FVB, Mdr1a/b(−/−), Bcrp(−/−), and Mdr1a/b(−/−)/Bcrp(−/−) mice, 9 weeks of age, were obtained from Taconic Farms (Germantown, NY). The mice used in the present study were 10 to 18 weeks old and weighed 23 to 36 g. All animals were maintained at a controlled temperature under a 12-h light/dark cycle. Food and water were available ad libitum.
Determination of the Transcellular Transport across Monolayers of Cell Lines Expressing P-gp or Bcrp.
In vitro mouse Bcrp transport experiments were performed as reported previously (Enokizono et al., 2007, 2008). In brief, MDCK II cells were seeded into 24-well Transwell plates (Corning Life Sciences, Lowell, MA) at a density of 1.4 × 105 cells/well and grown for 2 days in Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA) with 10% fetal bovine serum (Sigma-Aldrich, St. Louis, MO) and 1% antibiotic–antimycotic solution (Sigma-Aldrich). The cells were infected with recombinant adenovirus harboring green fluorescent protein (GFP) or mouse Bcrp expression vector at a ×200 multiplicity of infection. Details of the construction of these recombinant adenoviruses have been described previously (Kondo et al., 2004). After 2 days in culture, both GFP- and Bcrp-expressing cells (GFP-MDCK and Bcrp-MDCK, respectively) were used for transport studies.
In vitro transport experiments to determine the transport activity by mouse Mdr1a were conducted with Mdr1a-expressing LLC-PK1 cells (L-Mdr1a) that were established previously (Schinkel et al., 1995, 1996). L-Mdr1a and parent LLC-PK1 cells were seeded in 24-well Transwell plates at a density of 3.2 × 105 and 2.1 × 105 cells/well, respectively, and grown in Medium 199 (Invitrogen) with 10% fetal bovine serum and 1% antibiotic–antimycotic solution. Medium was changed on the 2nd day of culture, and cells were subjected to the transport study on the 4th day.
The cells were preincubated in Krebs–Henseleit buffer (118 mM NaCl, 23.8 mM NaHCO3, 4.83 mM KCl, 0.96 mM KH2PO4, 1.20 mM MgSO4, 12.5 mM HEPES, 5 mM glucose, and 1.53 mM CaCl2, pH 7.4) at 37°C for more than 30 min, and transport experiments were initiated by replacing the medium on one side of the cell monolayer with Krebs–Henseleit buffer containing 3 μM test compounds. At the appropriate times (60, 120, and 180 min), 100-μl aliquots were taken from the opposite side of the cell monolayer and replaced with 100 μl of drug-free buffer. The medium (100 μl) obtained from in vitro transport studies was mixed with 50 μl of acetonitrile for all compounds except mitoxantrone, which was mixed with 50 μl of 20% (w/v) ascorbic acid–saline.
Transport rates were calculated from the slopes of the time profiles of the apical-to-basal and basal-to-apical transport. Flux ratios (Table 1) were obtained by dividing the efflux rates in the basal-to-apical direction by those in the apical-to-basal direction. Flux ratios in transporter-expressing cells were divided by those in control cells to give a corrected flux ratio (CFR), an in vitro index of the P-gp- and Bcrp-mediated efflux.
Determination of the Tissue-to-Plasma Ratio in Mice.
Under urethane anesthesia (1.25 g/kg i.p.), the right jugular vein of the mice was cannulated with a polyethylene tube (PE-10; BD Biosciences, San Jose, CA). Compounds were administered via the cannula by continuous infusion for 120 min. The infusion rates of dantrolene, erlotinib, flavopiridol, mitoxantrone, and quinidine were 2, 4, 32, 8, and 8 μmol/h/kg after priming doses of 1, 2, 8, 2, and 6 μmol/kg, respectively. Blood samples were collected from the left jugular vein at the appropriate time points and centrifuged at 4°C and 9000g for 5 min to obtain plasma. Immediately after the final blood sampling, mice were sacrificed by exsanguination. For mitoxantrone, the plasma samples were transferred to microtubes containing ascorbic acid. For dantrolene, erlotinib, flavopiridol, and quinidine, the brain and testes specimens were homogenized with three volumes of phosphate-buffered saline (pH 7.4), whereas 0.9% saline containing 20% (w/v) ascorbic acid was used to prepare the homogenates containing mitoxantrone, to obtain a 25% homogenate for all compounds. Plasma specimens and tissue homogenates were stored at −80°C until use. Tissue-to-plasma ratios (Cbrain/Cplasma and Ctestis/Cplasma) were obtained by dividing the drug concentrations in the brain and testes by the plasma concentrations at the last sampling point.
Determination of PSP-gp and PSBcrp.
Adachi et al. (2001) demonstrated previously that the Kp,brain is given by the ratio of permeability surface area (PS) products for the uptake (PSbrain-to-blood) and efflux (PSblood-to-brain). The PStissue-to-blood and PSblood-to-tissue in wild-type mice are given by eqs. 1 and 2, respectively, where PSb,inf and PSb,eff represent the PS product for the influx and efflux across the blood-side membrane of the endothelial cells, PSt,inf and PSt,eff represent the PS product for the influx and efflux across the tissue-side membrane of the endothelial cells, and PSBcrp and PSP-gp represent the PS product for the efflux mediated by Bcrp and P-gp on the blood-side membrane of the endothelial cells, respectively. Because the Kp value is given by PSblood-to-tissue/PStissue-to-blood, Kp in wild-type mice [Kp (WT)] is given by eq. 3. The Kp in Mdr1a/1b(−/−), Bcrp(−/−), and Mdr1a/1b(−/−)/Bcrp(−/−) mice are given by eqs. 4, 5, and 6, respectively. The ratios of Kp (R) in Mdr1a/1b(−/−), Bcrp(−/−), and Mdr1a/1b(−/−)/Bcrp(−/−) mice to that in FVB mice are given by eqs. 7, 8, and 9, respectively. The PSP-gp and PSBcrp, intrinsic efflux transport activities at the BBB and BTB relative to passive diffusion, were obtained by using a nonlinear least-squares method by fitting eqs. 7 to 9 to the actual RMdr1a/1b(−/−), RBcrp(−/−), and RMdr1a/1b(−/−)/Bcrp(−/−) by using the MULTI program (Yamaoka et al., 1981). The algorithm used for the fitting was the Damping Gauss Newton Method (Yamaoka et al., 1981).
Quantification of Drugs in the Biological Samples.
Plasma samples were diluted with two volumes of phosphate-buffered saline for all compounds except mitoxantrone. Mitoxantrone was mixed with 0.9% saline containing 20% (w/v) ascorbic acid to obtain 33% diluted plasma. The proteins in these diluted plasma and tissue homogenates were precipitated with two volumes of acetonitrile, and the suspensions were centrifuged twice at 4°C and 5000g for 10 min. The supernatant were evaporated, and the remaining residues were reconstituted in mobile phase and subjected to LC-MS/MS or LC-UV analysis. The media (100 μl) mixed with 50 μl of acetonitrile or 20% (w/v) ascorbic acid–saline in in vitro transport studies were centrifuged at 4°C and 5000g for 5 min. The supernatants were also subjected to analysis by LC-MS/MS or LC-UV. All compounds except mitoxantrone were analyzed in a multiple reaction monitoring mode with an API2000 instrument (Applied Biosystems, Foster City, CA) equipped with an Agilent 1100 series LC system (Agilent Technologies, Santa Clara, CA). The transitions (precursor/product) used for quantification of erlotinib, flavopiridol, and quinidine were 394/278, 402/70, and 325/160, respectively, under positive atmospheric pressure chemical ionization mode, and that of dantrolene was 313/214 under negative atmospheric pressure chemical ionization mode. Mitoxantrone was detected by its absorbance at 655 nm by using an Agilent 1100 series diode array detector. All the compounds was separated by using a Capcell Pak C18 MGII column (3 μm, 3-mm i.d. × 35 mm; Shiseido, Kanagawa, Japan) at room temperature at the flow rate of 0.8 ml/min. Mobile phase A was 0.05% formic acid for all compounds except dantrolene, and 10 mM ammonium acetate was used as mobile phase A for dantrolene. Mobile phase B was acetonitrile for all compounds. For the analysis of erlotinib and quinidine, the fraction of mobile phase B was initially 5%, kept at 5% for 0.5 min, linearly increased to 90% over 2.9 min, kept at 90% for an additional 0.1 min, and finally re-equilibrated at 5% for 1.5 min. For the analysis of flavopiridol, the fraction of mobile phase B was initially 5%, kept at 5% for 1 min, linearly increased to 90% over 2.4 min, kept at 90% for an additional 0.1 min, and finally re-equilibrated at 5% for 1.5 min. For the analysis of mitoxantrone, the fraction of mobile phase B was initially 5%, kept at 5% for 1 min, linearly increased to 75% over 3 min, kept at 75% for an additional 0.5 min, and finally re-equilibrated at 5% for 1.5 min. For the analysis of dantrolene, the fraction of mobile phase B was initially 5%, kept at 5% for 0.5 min, linearly increased to 90% over 2.95 min, kept at 90% for an additional 0.05 min, and finally re-equilibrated at 5% for 1.5 min.
Statistical Analysis.
The presented values are all mean ± S.E. For comparison between genotype groups, log-transformed data were processed by using one-way analysis of variance, followed by a Tukey post hoc test. For comparison of the in vitro transcellular transport between transporter-expressing cells and the corresponding control cells, Student's two-tailed t test was used. Differences were considered significant at P < 0.05. All statistical calculations were performed with SAS software (version 9; SAS Institute, Cary, NC).
Results
Transcellular Transport of Erlotinib, Flavopiridol, Mitoxantrone, Dantrolene, and Quinidine in Cell Lines Expressing Mouse Bcrp or Mdr1a.
Transcellular transport of erlotinib, flavopiridol, and mitoxantrone was determined in the basal-to-apical and apical-to-basal directions across cells expressing Bcrp or Mdr1a (Bcrp-MDCK or L-Mdr1a). In the present study, based on previous knowledge, dantrolene and quinidine were selected as specific substrates of Bcrp and P-gp, respectively. Although dantrolene exhibited a directional transport in the basal-to-apical direction in parent LLC-PK1, the directional transport was unchanged by the expression of P-gp (Fig. 1). The transcellular transport of dantrolene was nondirectional in mock MDCK II cells, but expression of Bcrp induced directional transport in the basal-to-apical direction (Fig. 2). There was no directional transport of quinidine in parent LLC-PK1, but the expression of P-gp induced directional transport in the basal-to-apical direction (Fig. 1). Expression of Bcrp did not affect the transcellular transport of quinidine (Fig. 2).
The time profiles for the transcellular transport of erlotinib, flavopiridol, and mitoxantrone are shown in Figs. 1 and 2. In both Mdr1a- and Bcrp-expressing cells, the permeability of erlotinib and flavopiridol in the basal-to-apical direction was significantly greater than those in the opposite direction. The transcellular transport of mitoxantrone showed directional transport in the basal-to-apical direction in both parent LLC-PK1 and MDCK II cells; however, the expression of both P-gp and Bcrp clearly increased the basal-to-apical transport. The ratios of the permeability in the basal-to-apical direction to those in the opposite direction were higher for L-Mdr1a and Bcrp-MDCK than those for parent LLC-PK1 and GFP-MDCK (Table 1), suggesting that erlotinib, flavopiridol, and mitoxantrone are common substrates of P-gp and Bcrp.
Tissue Distribution in the Brain and Testis of Wild-Type, Mdr1a/1b(−/−), Bcrp(−/−), and Mdr1a/1b(−/−)/Bcrp(−/−) Mice.
The test drugs were administered by continuous infusion and the concentrations in plasma samples during infusion and in brain and testis samples obtained at 2 h after administration were determined. The time profiles of the plasma concentrations are shown in Fig. 3. The plasma concentrations of the test drugs in Mdr1a/1b(−/−), Bcrp(−/−), and Mdr1a/1b(−/−)/Bcrp(−/−) mice were comparable with those in FVB mice except for flavopiridol and dantrolene. Flavopiridol exhibited significantly lower plasma concentrations in Bcrp(−/−) mice and Mdr1a/1b(−/−)/Bcrp(−/−) mice compared with FVB mice only at 1 h, and dantrolene exhibited significantly higher plasma concentrations in Mdr1a/1b(−/−)/Bcrp(−/−) mice compared with FVB mice at 2 h.
Both the Cbrain/Cplasma and Ctestis/Cplasma of dantrolene were significantly increased in Bcrp(−/−) and Mdr1a/1b(−/−)/Bcrp(−/−) mice compared with those in Mdr1a/1b(−/−) and FVB mice (Fig. 4). The Cbrain/Cplasma and Ctestis/Cplasma of dantrolene were slightly higher in Mdr1a/1b(−/−)/Bcrp(−/−) mice than in Bcrp(−/−) mice, and the Ctestis/Cplasma was higher in Mdr1a/1b(−/−) mice than in wild-type mice. The Cbrain/Cplasma and Ctestis/Cplasma of quinidine exhibited a significant increase only in Mdr1a/1b(−/−) and Mdr1a/1b(−/−)/Bcrp(−/−) mice, and there were no significant differences in these parameters between Bcrp(−/−) mice and FVB mice or Mdr1a/1b(−/−) and Mdr1a/1b(−/−)/Bcrp(−/−) mice (Fig. 4).
The Cbrain/Cplasma and Ctestis/Cplasma of erlotinib, flavopiridol, and mitoxantrone were markedly increased in Mdr1a/1b(−/−)/Bcrp(−/−) mice compared with FVB, Mdr1a/1b(−/−), and/or Bcrp(−/−) mice (Fig. 4). Both the Cbrain/Cplasma and Ctestis/Cplasma of erlotinib were significantly higher in both Mdr1a/1b(−/−) and Bcrp(−/−) mice compared with FVB mice. The Cbrain/Cplasma of flavopiridol and mitoxantrone was significantly increased in Mdr1a/1b(−/−) mice compared with FVB mice, whereas there were no significant differences in the Cbrain/Cplasma between Bcrp(−/−) and FVB mice. The Ctestis/Cplasma of flavopiridol and mitoxantrone was significantly increased in both Mdr1a/1b(−/−) and Bcrp(−/−) mice compared with FVB mice.
Determination of PSP-gp and PSBCRP of Erlotinib, Flavopiridol, and Mitoxantrone at the BBB and BTB.
Equations 7 to 9 were fitted to the Ctissue/Cplasma in the knockout mice to wild-type mice (Table 2) to obtain the PSBcrp and PSP-gp of erlotinib, flavopiridol, and mitoxantrone (Table 2). The PS products could reproduce the observed values in both the BBB and the BTB, validating the equations (Table 2). In the BBB, both PSBcrp and PSP-gp of erlotinib, flavopiridol, and mitoxantrone were higher than the passive diffusion, but PSP-gp was higher than PSBcrp. This holds true for the BTB; however, the absolute values of both PSBcrp and PSP-gp were slightly lower than the corresponding parameters at the BBB.
Discussion
The overlap in the membrane localization and substrate specificities of Bcrp and P-gp has suggested their cooperation in the active efflux in the blood–tissue barriers. A considerable increase in the accumulation of common substrates in the brain of Mdr1a/1b(−/−)/Bcrp(−/−) mice compared with that observed in either Mdr1a/1b(−/−) or Bcrp(−/−) mice has been interpreted as synergistic effect. The present study investigated this synergistic effect kinetically by using erlotinib, flavopiridol, and mitoxantrone as test compounds. Because Cisternino et al. (2004) reported an induction of Bcrp mRNA in the brain capillaries from Mdr1a/1b(−/−) mice, quinidine and dantrolene were used as probes for P-gp and Bcrp, respectively, to examine their adaptive regulation in the knockout strain.
Adaptive regulation of P-gp at the BBB and BTB of Bcrp(−/−) mice is negligible because neither the Cbrain/Cplasma nor the Ctestis/Cplasma of quinidine changed in Bcrp(−/−) mice compared with the values in wild-type mice. Although the in vitro transport study could not detect it, dantrolene is likely a P-gp substrate because the Cbrain/Cplasma and Ctestis/Cplasma of dantrolene were slightly higher in Mdr1a/1b(−/−)/Bcrp(−/−) mice compared with Bcrp(−/−) mice. However, the P-gp-mediated efflux (PSP-gp) was 10- to 20-fold lower than the Bcrp-mediated efflux (PSBcrp) both at the BBB and BTB (PSP-gp was 0.4 and 0.5 at the BBB and BTB, whereas PSBcrp was 4 and 8 at the BBB and BTB, respectively), supporting the rationality of dantrolene as a Bcrp probe. Namely, the unchanged Cbrain/Cplasma of dantrolene in Mdr1a/1b(−/−) mice compared with that in wild-type mice suggests that the adaptive regulation of Bcrp is negligible even if it occurs in the BBB in Mdr1a/1b(−/−) mice. On the other hand, the Ctestis/Cplasma of dantrolene was significantly increased in Mdr1a/1b(−/−) mice for some unknown reason, and further studies of this are necessary.
A marked increase was observed in the Cbrain/Cplasma of erlotinib, flavopiridol, and mitoxantrone in Mdr1a/1b(−/−)/Bcrp(−/−) mice, even compared with Mdr1a/1b(−/−) and Bcrp(−/−) mice. In addition, as expected, a marked increase was observed in the Ctestis/Cplasma of erlotinib, flavopiridol, and mitoxantrone in Mdr1a/1b(−/−)/Bcrp(−/−) mice, even compared with Mdr1a/1b(−/−) and Bcrp(−/−) mice (Fig. 4). The impact of the defect of both P-gp and Bcrp on the accumulation of their common substrates in the brain and testis could be reasonably explained by using the PSP-gp and PSBcrp without introducing any interplay between P-gp and Bcrp (Table 2). This holds true for the drugs (dasatinib, flavopiridol, imatinib, lapatinib, and prazosin) for which the brain-to-plasma ratio in triple knockout mice was reported previously (Table 2) (Chen et al., 2009; Oostendorp et al., 2009; Polli et al., 2009; Zhou et al., 2009). It is reasonable that the impact of the defect of Bcrp was not as marked for the common substrates considering that P-gp makes a larger contribution to the net efflux of the common substrates tested than Bcrp at the BBB and BTB (Table 2). However, the fact that Bcrp-mediated efflux is larger than the passive diffusion at the BBB and BTB (Table 2) indicates that Bcrp can play a more significant role when P-gp is unfunctional. Actually, without considering P-gp-mediated efflux, the Cbrain/Cplasma of erlotinib, flavopiridol, and mitoxantrone could be increased 3-, 4-, and 4-fold, respectively, by the defect of Bcrp compared with wild-type mice (the actual increases were only 1.3-, 1.3-, and 1.4-fold). This is the reason Mdr1a/1b(−/−)/Bcrp(−/−) mice show further increase in the Cbrain/Cplasma of the common substrates compared with Mdr1a/1b(−/−) mice. It is worth mentioning that the present study did not confirm that the tissue concentrations reached a plateau, although the mass balance equations were solved under steady-state conditions to obtain eqs. 7 to 9. Therefore, fitting the equations to the observed data shown in Table 2 may underestimate the PS products of P-gp and Bcrp.
Bcrp cooperates with P-gp in the active efflux at both the BBB and BTB because of their overlapped substrate specificity. Such cooperation is undoubtedly important to protect the brain and testis. Even though one transporter is unfunctional by chemicals or shows reduced transport function caused by genetic polymorphisms, the other remained intact and can still limit tissue penetration. Bcrp(−/−) mice did not show any increase in the brain concentrations of Bcrp substrates, such as dehydroepiandrosterone sulfate, pitavastatin, and fluoroquinolones, compared with wild-type mice (Hirano et al., 2005; Lee et al., 2005; Matsushima et al., 2005; Ando et al., 2007). It is possible that transporters other than Bcrp are involved in the efflux of these drugs at the BBB, and, thereby, attenuate the impact of Bcrp dysfunction although Bcrp-mediated efflux is larger than passive diffusion. In fact, P-gp also accepts some anionic Bcrp substrates and fluoroquinolones as substrates (Tsuji et al., 1992; Matsushima et al., 2005; Kitamura et al., 2008). In addition to P-gp, Mrp4 can be an alternative candidate. MRP4 has been demonstrated to limit the brain penetration of anionic drugs at the BBB (Leggas et al., 2004; Belinsky et al., 2007; Ose et al., 2009). Furthermore, Takenaka et al. (2007) found that enhanced toxicity of adefovir is enhanced in mice lacking both Bcrp and Mrp4. Mrp4 may act as an active barrier cooperatively with P-gp and Bcrp, and Bcrp(−/−)/Mrp4(−/−) or Mdr1a/1b(−/−)/Bcrp(−/−)/Mrp4(−/−) mice may exhibit a marked accumulation of their common substrates. Further studies are necessary to uncover such cooperation of xenobiotic transporters at the BBB.
In conclusion, the present study elucidated that the synergistic effect of P-gp and Bcrp on the accumulation of their common substrates in the brain and testis can be explained by their contribution to the net efflux at the BBB and BTB without any direct interaction between P-gp and Bcrp.
Acknowledgments
We thank Dr. Alfred H. Schinkel (The Netherlands Cancer Institute, Amsterdam, The Netherlands) for supplying L-Mdr1a cells and sanofi-aventis (Bridgewater, NJ) for supplying flavopiridol.
Footnotes
This work was supported in part by the Ministry of Education, Culture, Sports, Science, and Technology [Grants 20249008, 20390046] (to Y.S. and H.K., respectively) and by the Kanal Foundation for the Promotion of Medical Science grant.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
doi:10.1124/jpet.109.162321.
-
ABBREVIATIONS:
- P-gp
- P-glycoprotein
- Mdr
- multidrug resistance protein
- Bcrp
- breast cancer resistance protein
- BBB
- blood-brain barrier
- BTB
- blood-testis barrier
- ABC
- ATP binding cassette
- MDCK
- Madin-Darby canine kidney
- GFP
- green fluorescent protein
- L-Mdr1a
- LLC-PK1 cells expressing mouse Mdr1a
- LC
- liquid chromatography
- MS/MS
- tandem mass spectrometry
- CFR
- corrected flux ratio
- PS
- permeability surface area
- Cbrain/Cplasma
- brain-to-plasma ratio
- Ctestis/Cplasma
- testis-to-plasma ratio
- MRP
- multidrug resistance-associated protein
- 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
- HMR-1275
- (−)cis-5,7-dihydroxy-2-(2-chlorophenyl)-8-(4R-(3S-hydroxy-1-methyl) piperidinyl)-4H-1-benzopyran-4-one.
- Received October 5, 2009.
- Accepted March 15, 2010.
- Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics
References
- Adachi et al., 2001.↵
- Ando et al., 2007.↵
- Bart et al., 2002.↵
- Bart et al., 2004.↵
- Belinsky et al., 2007.↵
- Chen et al., 2009.↵
- Cisternino et al., 2004.↵
- de Vries et al., 2007.↵
- Enokizono et al., 2008.↵
- Enokizono et al., 2007.↵
- Hirano et al., 2005.↵
- Kitamura et al., 2008.↵
- Kondo et al., 2004.↵
- Kusuhara and Sugiyama, 2005.↵
- Kusuhara and Sugiyama, 2009.↵
- Lagas et al., 2009.↵
- Lee et al., 2005.↵
- Leggas et al., 2004.↵
- Matsushima et al., 2005.↵
- Melaine et al., 2002.↵
- Oostendorp et al., 2009.↵
- Ose et al., 2009.↵
- Polli et al., 2009.↵
- Scherrmann, 2005.↵
- Schinkel, 1999.↵
- Schinkel et al., 1996.↵
- Schinkel et al., 1995.↵
- Takenaka et al., 2007.↵
- Tsuji et al., 1992.↵
- Yamaoka et al., 1981.↵
- Zhang et al., 2004.↵
- Zhou et al., 2009.↵