Breast cancer resistance protein (BCRP/ABCG2) is a multidrug resistance protein that is a member of the ATP-binding cassette (ABC) transporter family (Doyle et al., 1998; Miyake et al., 1999). BCRP shows broad substrate specificity, including anticancer drugs, such as topotecan, daunomycin or mitoxantrone, and endogenous and exogenous sulfate conjugates, such as estrone 3-sulfate, dehydroepiandrosterone sulfate, E3040S, and 4MUS (Allen and Schinkel, 2002; Suzuki et al., 2003).

In normal human tissues, BCRP has been found to be expressed on the apical membranes of the intestinal epithelium, bile canalicular membranes, and placental syncytiotrophoblasts (Maliepaard et al., 2001). BCRP may play an important role in protecting these tissues against exposure to xenobiotics by extruding them across the apical membrane. Indeed, Bcrp1(-/-) mice were more sensitive to pheophorbide, a breakdown product of chlorophyll, resulting in phototoxic lesions on light-exposed skin, due to marked increase in its plasma concentration (Jonker et al., 2002). The oral availability of topotecan was increased 6-fold in Bcrp1(-/-) mice, and fetal penetration of topotecan was 2-fold higher in Bcrp1(-/-) mice compared with their wild-type counterparts (Jonker et al., 2002). The hepatobiliary excretion of i.v. administered topotecan was reduced 2-fold by oral GF120918, consistent with an excretory role for bile canalicular Bcrp1 (Jonker et al., 2000). Recently, it was demonstrated that the hepatobiliary excretion of the dietary carcinogen, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine, was greatly reduced in Bcrp1(-/-) mice (van Herwaarden et al., 2003).

In addition to the liver, the kidney plays an important role in the elimination of a wide variety of xenobiotics from the body. The mechanisms that contribute to their renal excretion are closely related to the physiological events occurring in the nephrons, i.e., glomerular filtration, secretion, and reabsorption. Since a high level of murine Bcrp1 is found in the renal proximal tubules, it is possible that murine Bcrp1 is also involved in the tubular secretion of organic compounds (Jonker et al., 2002; Zhou et al., 2002).

In the present study, we compared the renal excretion of E3040S and 4MUS, in Bcrp1(-/-) and wild-type mice to investigate the physiologic role of Bcrp1 in the urinary excretion of sulfate conjugates.

Materials and Methods

Animals and Materials. Female Bcrp1(-/-) and wild-type FVB mice (17-23 weeks old) were used in the present study (Jonker et al., 2002). All animals were treated humanely. Mice were housed and handled according to the Guide for the Care and Use of Laboratory Animals as adopted and promulgated by the National Institutes of Health. E3040, E3040S and E3040G were supplied by Eisai Co., Ltd. (Tsukuba, Japan). 4MU, 4MUS, mannitol, and phenacetin were purchased from Sigma-Aldrich (St. Louis, MO). β-Trifluoromethylumbelliferone was obtained from Molecular Probes (Eugene, OR). All other chemicals were analytical grade and commercially available.

Urinary Excretion Experiment. After anesthesia with intraperitoneal sodium pentobarbital (50 mg/kg), the bladder was catheterized for urine collection. The doses of E3040S or 4MUS including 4% mannitol were administered via the tail vein. Mannitol was used to maintain an adequate, constant urinary flow rate (Gotoh et al., 2002). After an intravenous bolus injection (2.1 μmol/kg E3040S or 6.1 μmol/kg 4MUS), E3040S (0.074 μmol/min/kg) or 4MUS (0.24 μmol/min/kg) was infused. Blood samples were collected from the retroorbital sinus at 30, 60, and 90 min after administration and centrifuged. Urine was collected at 0 to 30, 30 to 60, and 60 to 90 min by washing the bladder with 0.2 ml of saline. At the end of the experiment, the kidneys were removed.

Liquid Chromatography/Tandem Mass Spectrometry Analysis. The 10-μl plasma or urine was extracted using a solid phase extraction column (OASIS HLB; Waters, Milford, MA). A 25% (w/v) kidney homogenate was centrifuged at 4°C and 10,000g for 10min, and the supernatant (50 μl) was also extracted by OASIS HLB. The quantification of E3040S and 4MUS was performed by liquid chromatography/tandem mass spectrometry (model TSQ7000; Thermo Quest, San Jose, CA). Phenacetin and β-trifluoromethylumbelliferone were used as internal standards for E3040S and 4MUS quantification, respectively. HPLC analysis of E3040S was performed using an Xterra (3.5 μm, 2.1 × 50 mm) column (Waters). The mobile phase was 10 mM ammonium acetic acid and acetonitrile. Initially, 5% acetonitrile was maintained for 3 min, then a gradient to 60% acetonitrile was started over 1 min, and this was then held for 8 min. The flow rate of the mobile phase was 0.3 ml/min. Then, the HPLC eluate was ionized using the electrospray interface. The following transitions were monitored: 300 to 221 for E3040, 380 to 300 for E3040S, 476 to 300 for E3040G, and 180 to 136 for phenacetin. HPLC analysis of 4MUS was performed using a Mightysil RP-18 GP (3 μm, 2 × 50 mm) column (Kanto-Kagaku, Tokyo, Japan). The mobile phase was 0.1% formic acid and acetonitrile. Initially, 10% acetonitrile was maintained for 1 min, then a gradient of 50% acetonitrile was produced over 1 min, and this was then held for 5 min. The flow rate of mobile phase was 0.25 ml/min. Then, the HPLC eluate was ionized using the electrospray interface. The following transitions were monitored: 255 to 175 for 4MUS, 175 to 119 for 4MU, and 229 to 153 for β-trifluoromethylumbelliferone. The levels of creatinine in plasma and urine were determined using assay kits from Wako Pure Chemicals (Osaka, Japan).

Pharmacokinetic Analysis. Total body clearance (CL_total), renal clearance with respect to circulating plasma (CL_renal,p), and renal clearance with respect to the kidney concentration (CL_renal,k) were calculated from the following equations: CL_total = I/C_p; CL_renal,p = V_urine/C_p; and CL_renal,k = V_urine/C_k, where I, C_p, V_urine, and C_k represent the infusion rate (nmol/min/kg), plasma concentration at 90 min (μM), the urinary excretion rate from 60 to 90 min (nmol/min/kg), and the kidney concentration at 90 min (μM), respectively. The glomerular filtration rate (GFR) was assumed to be equal to the creatinine clearance.

Statistical Analysis. Statistical analysis was performed by Student's t test to identify significant differences between two sets of data.

Results and Discussion

To investigate the role of Bcrp1 in urinary excretion, we examined the urinary excretion of E3040S and 4MUS in wild-type and Bcrp1(-/-) mice. The body weight, urine volume, and creatinine clearance were comparable in Bcrp1(-/-) mice and wild-type mice (Table 1). The plasma concentration after intravenous infusion of E3040S was slightly higher in Bcrp1(-/-) mice than in wild-type mice (Fig. 1A), whereas the urinary excretion of E3040S was significantly reduced in Bcrp1(-/-) mice (Fig. 1B). Little E3040 and E3040G were detected in plasma and urine (data not shown). Pharmacokinetic parameters were estimated from the plasma or kidney concentrations at 90 min and the urinary excretion rate from 60 to 90 min. The total clearance was slightly lower in Bcrp1(-/-) mice than in wild-type mice, but the difference was not significant (Table 2). Assuming that the unbound fraction of E3040S in mouse plasma is similar to that in rat (0.06; Takenaka et al., 1997), the renal clearance of unbound E3040S is much greater than the GFR, suggesting that tubular secretion accounts for the major part of the urinary excretion of E3040S. Both the renal clearance with respect to circulating plasma (CL_renal,p) and that with respect to the kidney concentration (CL_renal,k) were reduced in Bcrp1(-/-) mice compared with wild-type mice (Table 2). Furthermore, the kidney to plasma concentration ratio (K_p,kidney) was increased 2.4-fold in Bcrp1(-/-) mice compared with that in wild-type mice (Table 2). These results suggest that Bcrp1 plays a significant role in the urinary excretion of E3040S. The renal clearance of E3040S accounted for 30% and 56% of the total clearance in Bcrp1(-/-) and wild-type mice, respectively. Since the nonrenal clearance of E3040S is mainly accounted for by the biliary clearance in rats (Takenaka et al., 1997), it is possible that nonrenal clearance is accounted for by the biliary excretion in mice. However, the nonrenal clearance of E3040S was not reduced and slightly, but not significantly, increased in Bcrp1(-/-) mice compared with wild-type mice (6.37 ± 1.30 versus 4.77 ± 1.44 ml/min/kg). This suggests the minor contribution of Bcrp1 to the nonrenal clearance of E3040S. Presumably, there may be other transporters of this compound that are up-regulated to compensate for the Bcrp1 deficiency in these mice. Further investigations are required to identify the nonrenal mechanism for the removal of E3040S from the body.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.

Plasma concentration (A), urinary excretion (B), and kidney concentration (C) of E3040S in Bcrp1(-/-) (open symbols) and wild-type mice (closed symbols). E3040S was infused intravenously at 0.074 μmol/min/kg after a loading dose of 2.1 μmol/kg. Results are mean ± S.E. (n = 6). ^*, p < 0.05; ^**, p < 0.01, significantly different from wild-type mice.

In contrast, knockout of Bcrp1 did not affect the pharmacokinetic profile of 4MUS (Fig. 2). The renal clearance of 4MUS fully accounted for the total clearance in both types of mice and was much higher than GFR. Little conversion to 4MU was detected in plasma and urine (data not shown). All the kinetic parameters listed in Table 3 were comparable in Bcrp1(-/-) mice and wild-type mice. It is likely that the contribution of Bcrp1 to the urinary secretion of 4MUS is naturally very small. Currently, it is unknown whether E3040S and 4MUS are mouse Bcrp1 substrates or not, although human BCRP transports both these compounds (Suzuki et al., 2003). Thus, it is possible that there are species differences in transport selectivity between human BCRP and mouse Bcrp1, and further investigations are required.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2.

Plasma concentration (A), urinary excretion (B), and kidney concentration (C) of 4MUS in Bcrp1(-/-) (open symbols) and wild-type mice (closed symbols). 4MUS was infused intravenously at 0.24 μmol/min/kg after a loading dose of 6.1 μmol/kg. Results are mean ± S.E. (n = 3-4).

Another possibility is that transporters other than Bcrp1 may play a major role in the renal excretion of 4MUS in proximal tubules. Primary active transporters, such as Mrp2 or Mrp4, are localized on the brush-border membranes of rat proximal tubules (for a review, see Russel et al., 2002). Previous results indicate that xenobiotic sulfates are not significantly transported by MRP2 but, rather, stimulate the function of MRP2 (Takenaka et al., 1995; Niinuma et al., 1997). MRP4 was recently identified as a novel organic anion transporter in the brush-border membrane of rat and human proximal tubules (van Aubel et al., 2002). Masereeuw et al. (2003) suggested that MRP4 may play a role in the urinary secretion, but did not reach a final conclusion. Cumulative functional characterization has revealed that MRP4 has broad substrate specificity, including sulfate or glucuronide conjugates of endogenous steroids, dehydroepiandrosterone sulfate or estradiol-17β glucuronide, and sulfated bile acid (Chen et al., 2002; Russel et al., 2002; van Aubel et al., 2002; Zelcer et al., 2003). MRP4 seems to be a good candidate for transporting sulfate conjugates besides Bcrp1. Further investigations are required to see if some other transporters may play a role in the renal excretion of 4MUS.

In conclusion, this is, to our knowledge, the first demonstration of involvement of Bcrp1 in the renal secretion of an organic sulfate, E3040S. However, taking in vivo results of 4MUS into consideration, the renal secretion of organic sulfates cannot be accounted for solely by Bcrp1, and transporters other than Bcrp1 are also involved.