Inhibitory Effects of Herbal Extracts on Breast Cancer Resistance Protein (BCRP) and Structure-Inhibitory Potency Relationship of Isoflavonoids

Drug Metab. Pharmacokinet. 25 (2): 170–179 (2010).

Regular Article Inhibitory Effects of Herbal Extracts on Breast Cancer Resistance Protein (BCRP) and Structure-Inhibitory Potency Relationship of Isoflavonoids Hirofumi TAMAKI1, Hiroki SATOH1, Satoko HORI1,2, Hisakazu OHTANI1 and Yasufumi SAWADA*,1,2 1Laboratory

of Drug Informatics, Graduate School of Pharmaceutical Sciences, The University of Tokyo, Tokyo, Japan 2 Graduate School of Interdisciplinary Information Studies, The University of Tokyo, Tokyo, Japan Full text of this paper is available at http://www.jstage.jst.go.jp/browse/dmpk

Summary: The inhibition of intestinal breast cancer resistance protein (BCRP), which restricts the absorption of xenobiotics, may increase the systemic availability of its substrates. The aim of this study was to evaluate the inhibitory effects of herbal extracts and their constituents on BCRP-mediated transport. The inhibitory effects of 9 herbal extracts and 23 isoflavonoids, including soybean-derived isoflavones, on BCRP-mediated methotrexate (MTX) transport were evaluated using BCRP-expressing membrane vesicles. The structure-inhibitory potency relationship was investigated by multiple factor analysis. Extracts of soybean, Gymnema sylvestre, black cohosh and passion flower and rutin strongly inhibited BCRP-mediated transport of MTX at 1 mg/ml, while inhibition by chlorella, milk thistle and Siberian ginseng extracts was weak. Among the 23 isoflavonoids examined, all of which inhibited BCRP-mediated transport, coumestrol showed the most potent inhibition (IC50＝63 nM). The inhibitory potencies of 6 isoflavonoid glucosides were 10- to 100-fold lower than those of the corresponding aglycones. The addition of a 5-hydroxyl or 6-methoxyl moiety tended to potentiate the inhibition. The inhibitory potency of daidzein was decreased 100-fold by 7-glucuronidation, but was virtually unaffected by 4?-sulfation. Thus, some herbal and dietary supplements and isoflavonoids may increase the systemic availability of BCRP substrates when concomitantly given orally. Keywords: breast cancer resistance protein (BCRP); ATP-binding cassette transporter G2 (ABCG2); dietary supplement; drug interactions; isoflavonoids; structure-activity relationship

tins11,12) and proton pump inhibitors.13,14) In vivo murine studies revealed that intestinal Bcrp restricts the absorption of orally administered drugs, such as topotecan,15) irinotecan,16) nitrofurantoin,17) GV196771, N-methyl-Daspartate receptor antagonist18) and sulfasalazine.19) Recent studies showed that the genetic polymorphism of ABCG2 (c.421CÀA) affects the absorption of BCRP substrate drugs.20–22) These results imply that BCRP plays an important role in restricting the intestinal absorption of various drugs. Various phytochemicals, such as flavonoids, chalcones, terpenoids, isothiocyanates and nonprenylated rotenoids, are known to inhibit BCRP.23–29) Among them, flavonoids

Introduction ATP-binding cassette (ABC) transporters such as Pglycoprotein and multidrug resistance-associated protein 2 restrict intestinal drug absorption,1) and the inhibition of ABC transporters has been shown to increase intestinal absorption of their substrates.2) Breast cancer resistance protein (BCRP, encoded by ABCG2), a member of the ABC transporter family,3–5) is an efflux transporter expressed in intestinal epithelial cells, as well as bile canaliculus, kidney, blood-brain barrier and placenta.6,7) BCRP transports various drugs, including methotrexate (MTX),8) topoisomerase I inhibitors,9) flavopiridol,10) sta-

Received; September 10, 2009, Accepted; December 24, 2009 *To whom correspondence should be addressed: Yasufumi SAWADA, Laboratory of Drug Informatics, Graduate School of Pharmaceutical Science, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan. Tel.＋81-3-5841-2270, Fax.＋81-3-5841-1097, E-mail address: sawada＠mol.f.u-tokyo.ac.jp This study was supported in part by the Foundation for Total Health Promotion and the Mochida Memorial Foundation for Medical and Pharmaceutical Research.

170

Effects of Herbal Extracts and Isoflavonoids on BCRP

have been extensively investigated with regard to the relationship between inhibitory potency and chemical structure by using BCRP-overexpressing cells.30–32) In contrast, little is known about isoflavonoids other than genistein, daidzein and coumestrol, which were reported to be substrates of human BCRP and mouse Bcrp.23,24,33) In rats, the plasma concentration of nitrofurantoin, a Bcrp substrate, was increased by the concomitant oral administration of chrysin, a structural analog of flavone,34) suggesting that some BCRP inhibitors derived from herbs may increase the systemic availability of BCRP substrate drugs. Some plant-derived health foods and herbal supplements contain these compounds at high levels,35) whereas the amounts of these compounds taken from foods are usually limited. In the United States, herbal supplements are used widely, often concomitantly with drugs (18.4% of all prescription users36)). Coadministration of BCRP substrate drugs with herbal supplements may induce drug-herbal supplement interactions. The purpose of the present study was to clarify the inhibitory potencies of various herbal extracts and their constituents on BCRP in order to estimate the likelihood of drug-herbal supplement interactions. The relationship between the chemical structure of isoflavonoids and inhibitory potency towards BCRP was also examined by means of multiple factor analysis.

Materials and Methods Materials: Membrane vesicles isolated from mammalian cells expressing human wild-type BCRP (SBBCRP-M-VT) were purchased from SOLVO Biotechnology (Budapest, Hungary). MTX hydrate and fumitremorgin C (FTC) were purchased from Sigma-Aldrich (St Louis, MO, USA). MTX disodium salt [3?,5?,7-3H(N)] (740 GBq/mmol, ethanol:water＝4:6) was purchased from Moravek Biochemicals, Inc. (Brea, CA, USA). Daidzein, genistein and glycitein (each À98.0% pure) were purchased from Wako Pure Chemical Industries (Osaka, Japan). Genistin and daidzin (each À98.0% pure) were purchased from LC Laboratories (Woburn, MA, USA). Glycitin (Æ98%) was purchased from Nagara Science (Gifu, Japan). Formononetin (91.0%), ononin (99.9%), sissotrin (99.9%), sophoricoside, 4?,7-dimethoxy -5-hydroxyisoflavone (genistein-4?,7-dimethylether, Æ96.0%) and 5-methyl-7-methoxyisoflavone were purchased from ChromaDex (CA, USA). Daidzein-7glucuronide potassium salt, daidzein-4?-sulfate and 3?,4?, 7-trihydroxyisoflavone (98.0%) were purchased from Toronto Research Chemicals (Ontario, Canada). Biochanin A, 4?,7-dimethoxyisoflavone, prunetin (Æ98.0%) and coumestrol (Æ97.0%) were purchased from SigmaAldrich (St Louis, MO, USA). 4?,6,7-Trihydroxyisoflavone and 4?,6,7-trimethoxyisoflavone were purchased from Extrasynthese (Genay, France). Ipriflavone (Æ97.0%) was purchased from Wako Pure Chemical In-

171

dustries. Coumestrin was a kind gift from Dr. Satoshi Morimoto (Department of Medico-Pharmaceutical Sciences, Graduate School of Pharmaceutical Sciences, Kyushu University, Fukuoka, Japan). All other chemicals were commercial products of reagent grade. Herbal extracts: The nine herbal extracts (extracts of soybean, barley, chlorella, milk thistle, Gymnema sylvestre, black cohosh, passion flower and Siberian ginseng, and rutin, which are industrially added to various end products) were kind gifts from Tokiwa Phytochemical Co (Tokyo, Japan). The putative active ingredients of each herb were enriched in the extracts, the extraction methods of which are briefly shown below. An extract from barley flour of Hordeum vulgare was purified, mixed with cyclodextrin and dried. Chlorella pyrenoidosa was extracted with water and the extract was purified, mixed with cyclodextrin and dried (the value of OD260 was 4,150). Soybean extract was made from Glycine max (mainly its hypocotyl), concentrated, purified and dried; the product contained 30.4% isoflavones as glucosides (product name: Isomax-30). An extract of passion flower was made from whole Passiflora incarnata L. and dried. An extract of Siberian ginseng was obtained by extraction with hydrous ethanol from rhizomes of Eleutherococcus senticosus and dried. Rutin was extracted from buds of Sophora japonica L. and dried; its purity was 96.2%. Milk thistle extract from seeds of Silybum marianum was purified and dried; the product contained 31.7% silybin. An extract of black cohosh was obtained by extraction with hydrous ethanol from roots of Cimicifuga racemosa Nutt. and dried. An extract of Gymnema sylvestre R.Br. leaves was obtained by extraction with dilute alcohol and dried; the product contained 10.5% gymnemic acid as nonhydrate. Uptake experiments using BCRP-overexpressing membrane vesicles: MTX uptake of BCRP-expressing membrane vesicles was measured using a rapid filtration technique, following the protocol (version 1.3) provided by the manufacturer. Briefly, BCRP-expressing membrane vesicles were quickly thawed at 379C. Substrate uptake was initiated by adding a preincubated substrate buffer (final concentration: 3-(N-morpholino) propanesulfonic acid (MOPS)-Tris 39.8 mM, KCl 55.7 mM, MgCl2 32.7 mM and [3H]MTX 18.5 kBq, pH 7.0) with 26.7 mM ATP or AMP to 10 ml (containing 50 mg protein) of preincubated membrane vesicles solution (total volume: 75 ml, the ethanol contained in the radioisotope solution amounted to less than 0.3% (v/v)). To evaluate the concentration dependency of BCRPmediated transport, the uptake study was run with various final concentrations of MTX, with ATP or AMP. The mixture was incubated for 2 min at 379C, then 1 ml of ice-cold washing buffer (0.04 M MOPS-Tris, 0.07 M KCl, pH 7.0) was added to stop the uptake reaction. Vesicleassociated [3H]MTX in the stopped reaction mixture was

172

Hirofumi TAMAKI, et al.

separated from free [3H]MTX by rapid filtration through a GF/C filter (Whatman, Maidstone, UK). Filters were further washed twice with 5 ml of ice-cold washing buffer and assayed for radioactivity in 4 ml of liquid scintillation cocktail (Clear-sol I; Nacalai Tesque, Kyoto, Japan). The amount of MTX taken up into vesicles was calculated from the radioactivity. ATP-dependent MTX transport was assessed by subtracting the uptake in the presence of AMP from that in the presence of ATP. The ATP-dependent uptake into control vesicles (SB-M-CTRL; SOLVO Biotechnology) was also measured using the same protocol. To evaluate the effect of the osmolality of extravesicular buffer, uptake of MTX into the vesicles was evaluated after addition of mannitol to the substrate buffer to give 500 or 1,000 mOsmol final osmolality. The incubation time for uptake of MTX was 20 min in this experiment. Inhibition studies with herbal extracts: To evaluate the inhibitory effects of nine herbal extracts, MTX uptake into the vesicles was measured in the presence of herbal extracts. The herbal extracts were dissolved in dimethyl sulfoxide (DMSO) and water. The herbal extract solutions were filtered through 0.22 mm Millex-GV filters (Millipore Corp., Bedford, MA, USA) and diluted with the substrate buffer to give 1 mg/ml final herbal extract concentration and 0.5% (v/v) final DMSO concentration. To examine the concentration dependency of the inhibitory effect of soybean extract on BCRP-mediated transport, the uptake of MTX into the vesicles was measured in the presence of final concentrations of 1, 10, 100 and 1,000 mg/ml soybean extract. In these experiments, substrate buffer containing 0.5% (v/v) DMSO was used as the control. FTC was used as a positive control BCRP inhibitor. The final concentration of MTX in the substrate buffer was 100 mM and the incubation time for determination of uptake of MTX was 2 min. Inhibition studies with isoflavonoids: To evaluate the inhibitory effects of 23 isoflavonoids, MTX uptake into vesicles was measured in the presence of various concentrations of each isoflavonoid. Isoflavonoids were dissolved in DMSO and diluted with the substrate buffer to give 0.5% (v/v) final DMSO concentration. In this experiment, the substrate buffer containing 0.5% (v/v) DMSO was used as the control. The final concentration of MTX in the substrate buffer was 100 mM and the incubation time for determination of uptake was 2 min. Data analysis: To determine the negative logarithm of half-maximal inhibitory concentration of soybean extract and isoflavonoids (pIC50; log[g/ml] for soybean extract and log[mol/l] for isoflavonoids) for the BCRPmediated uptake of MTX, the following equation was fitted to the data for the ATP-dependent uptake in the presence of soybean extract or isoflavonoids by using a nonlinear least-squares regression analysis program (Mlab, Civilized Software Inc., Bethesda, MD, USA).

V/ V 0 ＝

1 log [I]＋ pIC50

1＋10 where V and V0 are the ATP-dependent transport rates of MTX in the presence or absence of inhibitors, respectively, and [I] is the concentration of inhibitor (soybean extract: g/ml and isoflavonoids: mol/l). The effects of functional moieties on the inhibitory potencies of isoflavonoids for BCRP-mediated transport were analyzed by multiple factor analysis based on Hayashi's quantification theory type I,37) using all models which contain some of four items of moieties' position (C5, C6, C7, C4?, Fig. 4), and then the Akaike information criterion (AIC) was assessed for each model to select the best one. In the statistical analysis, the significance of differences between mean values was determined by analysis of variance (ANOVA), followed by Dunnett's test, and a p value of less than 0.05 was considered statistically significant. The significance of differences between the pIC50 values of aglycones and the corresponding glucosides was determined by paired t-test.

Results BCRP-mediated transport of MTX into the vesicles: The amount of MTX taken up by the vesicles markedly decreased as the osmolarity of the extravesicular medium increased (6.85 ml/mg protein/20 min at 380 mOsmol, 6.54 ml/mg protein/20 min at 500 mOsmol and 4.55 ml/mg protein/20 min at 1,000 mOsmol). This result indicates that MTX was actually transported into the intravesicular space. The uptake of MTX into BCRP-expressing membrane vesicles increased linearly for at least 3 min (data not shown), so initial uptake rates were determined at 2 min in the uptake and inhibition studies. The ATP-dependent transport of MTX increased linearly with concentration from 100 mM to 1 mM, and appeared to be saturated at concentrations above 1 mM (Fig. 1), although uptake studies at MTX concentrations higher than 5 mM were not experimentally feasible due to limited solubility of the drug. ATP-dependent uptake was not found in the experiment using control membrane vesicles (in the presence of ATP 0.34±0.01 ml/mg protein at 1 min vs AMP 0.46±0.01 ml/mg protein at 1 min) and also in the experiment with BCRP-expressed membrane vesicles in the presence of 5 mM FTC, the specific inhibitor of BCRP (Fig. 2). From these results, we examined the ATP-dependent uptake of MTX into BCRP-expressing membrane vesicles as an index of BCRP-mediated transport. The Km value of BCRP-mediated transport of MTX was estimated to be 5.78±0.84 mM. This value was similar to previously reported Km values of MTX in vesicular uptake studies at pH 7.4 (5.7 mM,38) 5.21 mM39)). Inhibitory effects of herbal extracts on BCRPmediated transport of MTX: The extracts of soybe-

173

Effects of Herbal Extracts and Isoflavonoids on BCRP

Fig. 1. Concentration-dependent uptake of methotrexate (MTX) into BCRP-expressing membrane vesicles The uptake of MTX into human BCRP-expressing mammalian cell membrane vesicles was measured at 379 C at 2 min in the presence of ATP (closed circles) or AMP (open circles). Each point represents the mean±SEM of three experiments. The dashed line indicates V0×[C], where V0 is the mean of vesicle uptake rate of MTX in the presence of AMP, and [C] is the concentration of MTX (mmol/l). The solid line indicates the best fit to the equation: Vesicle uptake＝Vmax×[C]/(Km＋[C])＋V0×[C], where Km is the Michaelis-Menten parameter of MTX, and Vmax is the maximum ATPdependent vesicular uptake of MTX.

an, gymnema sylvestre, black cohosh and passion flower and rutin at 1 mg/ml almost completely inhibited BCRPmediated transport of MTX (pº0.001; Fig. 2A). The extracts of chlorella, milk thistle and Siberian ginseng moderately inhibited BCRP-mediated transport by 22.3%, 45.4% and 36.4%, respectively ( pº0.01 or 0.001), whereas barley extract was not significantly inhibitory (Fig. 2A). Soybean extract inhibited BCRP-mediated transport of MTX in a concentration-dependent manner (Fig. 2B) with an IC50 value of 16.5 mg/ml. Inhibitory effects of isoflavonoids on BCRPmediated transport of MTX: All 23 isoflavonoids studied, including soybean isoflavones such as daidzein and genistein, inhibited BCRP-mediated transport of MTX in a concentration-dependent manner (Fig. 3). The pIC50 values for these isoflavonoids are shown in Table 1. The influence of the functional moieties of isoflavonoids on the inhibitory potency (pIC50 value) for BCRPmediated transport was quantitatively analyzed by multiple factor analysis (Fig. 4). The model consisting of two items as factors (C5, C7 position, Fig. 4) best described the pIC50 values. The presence of a hydroxyl moiety at the C5 position potentiated the inhibitory effects (category scores of moieties: OH 0.275 vs H －0.206 at the C5 position), while 7-O-glucosidation decreased the pIC50 values (category scores of moieties: O-glucoside －0.653 vs OH 0.363, OCH3 0.363 at the C7 position). The pIC50 values of glucosides, including the C4?O-glucoside sophoricoside, were about 1.0 smaller than

Fig. 2. The effect of various herbal extracts on the ATP-dependent transport of methotrexate (MTX) into BCRP-expressing vesicles The uptake of MTX in the presence of (A) 1,000 mg/ml herbal extracts or (B) 1, 10, 100 or 1,000 mg/ml soybean extract was measured. BCRP membrane vesicles were incubated with 100 mM MTX at 379 C for 2 min in the presence of ATP (closed bars) or AMP (open bars). Each point represents the mean＋SEM of three experiments. The significance of differences from the respective control (ATP or AMP) was determined by ANOVA followed by Dunnett's test (**p º0.01; ***p º0.001)

those of the corresponding aglycones (Table 1). The decrease of pIC50 values due to 7-O-glucosidation was statistically significant ( pº0.001, paired t-test, Fig. 5), whereas methylation at the C6 position significantly increased the pIC50 values by almost 0.3 (glycitein pIC50 value 5.99 vs daidzein 5.62, glycitin 4.76 vs daidzin 4.54, 4?,6,7-trimethoxyisoflavone 6.31 vs 4?,7-dimethoxyisoflavone 5.96; pº0.05, paired t-test; Table 1).

Discussion Among the nine herbal extracts examined, soybean extract most strongly inhibited the BCRP-mediated transport of MTX in a concentration-dependent manner (Fig. 2B). This soybean extract contains 30.4% isoflavones as glucosides, and can be consumed in amounts of up to

174

Hirofumi TAMAKI, et al.

Fig. 3. Concentration-dependent inhibition by isoflavonoids of the transport of methotrexate (MTX) into BCRP-expressing vesicles BCRP membrane vesicles were incubated with 100 mM MTX and the indicated concentrations of isoflavonoids at 379 C for 2 min in the presence of ATP or AMP. ATP-dependent transport was assessed by subtracting the uptake in the presence of AMP from that in the presence of ATP. Each point represents the mean±SEM of three or four experiments. Each line indicates the best fit to the equation given in Materials and Methods .

160 mg/day (30 mg/day as soy isoflavone aglycones, 48 mg/day as isoflavone glucosides) as a dietary supplement in Japan.40) The upper limit for daily intake of soybean isoflavonoids in Japan is 75 mg/day as aglycones, which is similar to the levels in France and Italy (1.6 mg/kg body weight/day as total isoflavonoids including glycosides,41) 80 mg/day as aglycones42)). Assuming administration with a glass of water (250 ml), which volume is derived from typical clinical study protocols,43) the intestinal concentration of soybean extract ([I]2) is estimated to be 632 mg/ml. The US Food and Drug Administration research group proposed a criterion: P-glycoprotein inhibitor drugs that exhibit an [I]2/IC50À10 should be evaluated in vivo to determine whether there is clinically relevant Pglycoprotein inhibition with digoxin, a P-glycoprotein substrate with a narrow therapeutic index, for the possibility of drug interaction.43) Given the IC50 value of soybean extract (16.5 mg/ml), the [I]2/IC50 value is 38.3, which exceeds 10, the cutoff value of the criterion. Although it is necessary to consider that the criterion is not for BCRP but for P-glycoprotein, the concomitant administration of soybean extract with a BCRP substrate drug may therefore increase the absorption of the drug. Some of the examined extracts other than soybean may also strongly inhibit intestinal BCRP-mediated transport. However, the intestinal concentrations of these extracts

after the recommended amounts have been taken are estimated to be lower than the concentration examined (1 mg/ml). Therefore, additional studies are required to examine the inhibitory profiles of these extracts and to estimate the likelihood of drug-herbal extract interactions in the clinical setting. Soybean-derived isoflavonoids, such as daidzein, daidzin, genistein, genistin, glycitein, glycitin and ononin,44) all inhibited BCRP-mediated transport (Fig. 3). Based on the assumption that the isoflavone compositions of soybean extract are identical to those of raw soybean,45) we calculated the concentrations of daidzein, genistein, daidzin, genistin and glycitin in 16.5 mg/ml soybean extract solution and compared them with the corresponding IC50 values. The estimated concentrations of genistein and genistin were comparable to the respective IC50 values (0.15 vs 0.24 mM, 2.1 vs 4.3 mM), while those of daidzein, daidzin and glycitin were more than 10-fold smaller than the respective IC50 values (0.08 vs 2.4 mM, 1.6 vs 29 mM, and 0.20 vs 16 mM). Therefore, genistein and genistin are likely to be primarily responsible for the inhibitory effect of soybean extract. Milk thistle extract contains silybin, a weak inhibitor of BCRP.24) The weak inhibitory effect of the milk thistle extract may be largely accounted for by silybin, because the extract contains 31.7% silybin. Passion flower con-

175

Effects of Herbal Extracts and Isoflavonoids on BCRP

Table 1. Structure of investigated isoflavonoids and pIC50 values for BCRP inhibition Compound

C5

C6

C7

C4?

C3?

pIC50 (log [M])

S.D. 0.04

Daidzein

-H

-H

-OH

-OH

-H

5.62

Daidzin

-H

-H

-O-glucoside

-OH

-H

4.54

0.05

Formononetin

-H

-H

-OH

-OCH3

-H

5.69

0.06

Ononin

-H

-H

-O-glucoside

-OCH3

-H

4.86

0.05

4?,7-Dimethoxyisoflavone

-H

-H

-OCH3

-OCH3

-H

5.96

0.05

Daidzein-7-glucuronide Daidzein-4?-sulfate

-H

-H

-O-glucuronide

-OH

-H

3.37

0.14

-H

-H

-OH

-OSO3

-H

5.41

0.06

Genistein

-OH

-H

-OH

-OH

-H

6.62

0.06

Genistin

-OH

-H

-O-glucoside

-OH

-H

5.37

0.05

Biochanin A

-OH

-H

-OH

-OCH3

-H

6.30

0.07

Sissotrin 4?,7-Dimethoxy-5-hydroxyisoflavone

-OH

-H

-O-glucoside

-OCH3

-H

5.57

0.05

-OH

-H

-OCH3

-OCH3

-H

5.70

0.09

Sophoricoside

-OH

-H

-OH

-O-glucoside

-H

5.38

0.06

Prunetin

-OH

-H

-OCH3

-OH

-H

6.40

0.07

Glycitein

-H

-OCH3

-OH

-OH

-H

5.99

0.05

Glycitin 4?,6,7-Trihydroxyisoflavone

-H

-OCH3

-O-glucoside

-OH

-H

4.79

0.05

-H

-OH

-OH

-OH

-H

5.46

0.06

4?,6,7-Trimethoxyisoflavone

-H

-OCH3

-OCH3

-OCH3

-H

6.31

0.06

3?,4?,7-Trihydroxyisoflavone

-H

-H

-OH

-OH

-OH

5.57

0.07

5-Methyl-7-methoxyisoflavone

-CH3

-H

-OH

-H

-H

6.07

0.06

Ipriflavone

-H

-H

-OCH(CH3)2

-H

-H

5.88

0.05

Coumestrola

—

—

-OH

—

—

7.20

0.05

Coumestrina

—

—

-O-glucoside

—

—

5.40

0.05

a

The structures of coumestrol and coumestrin are shown below.

tains chrysin, which inhibits BCRP,24,31) and other flavonoids.46,47) Although the flavonoid compositions of passion flower extract are unknown, the inhibitory effect of passion flower extract may be due to flavonoids, including chrysin. Inhibition of BCRP by rutin seems reasonable, because it is a glycoside of quercetin, which inhibits BCRP.24) With regard to other extracts, the inhibitory constituents responsible for BCRP inhibition remain to be investigated. Among the 23 isoflavonoids, including soybean-derived isoflavones, coumestrol showed the most potent inhibition (Fig. 3, Table 1). Its pIC50 value was comparable to that of Ko143, a typical potent inhibitor of BCRP (7.20 vs 7.89 log[mol/l]48)). A structure-inhibitory potency relationship analysis of 14 isoflavonoids showed that the predicted pIC50 values of isoflavonoids were well correlated to the observed pIC50 values (Fig. 4). Among five pairs of isoflavonoids with and without a hydroxyl moiety at the C5 position, four isoflavonoids with a hydroxyl moiety at the C5 position exhibited more

potent inhibition than the corresponding isoflavonoids without the hydroxyl group. With regard to the remaining pair, the pIC50 value of 4?,7-dimethoxy-5-hydroxyisoflavone was slightly lower than that of 4?,7-dimethoxyisoflavone (5.70 vs 5.96). As a result of multiple factor analysis, the pIC50 values of an isoflavonoid with the hydroxyl moiety was expected to be larger than that of the corresponding isoflavonoid without the hydroxyl moiety at C5 (Fig. 4), so that genisteins have higher affinity for BCRP than do daidzeins. Ahmed-Belkacem et al. examined the influence of the C5 hydroxyl moiety in flavonoids and reported that the inhibitory potency of chrysin (5,7-dihydroxyflavone) for mitoxantrone efflux from BCRP-overexpressing cells was greater than that of 7-hydroxyflavone.31) The influence of a hydroxyl moiety at the C5 position on the inhibition of BCRP may be similar in isoflavonoids and flavonoids. In all three pairs of isoflavonoids with and without a methoxyl moiety at C6, the inhibitory effects of isoflavonoids with the methoxyl moiety were more potent than

176

Hirofumi TAMAKI, et al.

Fig. 5. Comparison of the pIC50 values for BCRP inhibition between isoflavonoid aglycone and the corresponding glucoside Each set represents the combination of an isoflavonoid aglycone and the corresponding glucoside: open circles, daidzein-daidzin; closed circles, formononetin-ononin; open squares, glycitein-glycitin; closed squares, biochanin A-sissotrin; open triangle, genisteingenistin; closed triangle, coumestrol-coumestrin. The significance of differences in mean pIC50 values between aglycones and glucosides was determined by use of the paired t -test (***p º0.001).

Fig. 4. Predictability of the pIC50 values for BCRP inhibition by multiple factor analysis based on quantification theory type I The observed pIC50 value±S.D. of each isoflavonoid was plotted against the predicted value, calculated by using the best describing equation shown at the bottom of the figure. In the equation, category scores of dummy variables (Xij) are shown as their respective factors. The upper table shows the definition of the dummy variable (Xij) for each category (j) and item (i). Each point represents the pIC50 value of the appropriately numbered compound (1 daidzein, 2 daidzin, 3 formononetin, 4 ononin, 5 4?,7 -dimethoxyisoflavone, 6 genistein, 7 genistin, 8 biochanin A, 9 sissotrin, 10 4?,7-dimethoxy-5-hydroxyisoflavone, 11 prunetin, 12 glycitein, 13 glycitin, 14 4?,6,7-trimethoxyisoflavone).

those of the isoflavonoids without it (Table 1), so that glyciteins have higher affinity for BCRP than do daidzeins. 7-O-Glucosidation of isoflavonoids reduced the inhibitory potency towand BCRP-mediated transport, and the differences in IC50 values between glucosides and the corresponding aglycones were approximately 10- to 100fold (Table 1). Multiple factor analysis revealed that the pIC50 values of 7-O-glucosides can be expected to be approximately 1.0 smaller than those of the corresponding aglycones (Fig. 4). The inhibitory potencies of glucosides tended to be related to those of the corresponding aglycones, although this was not statistically significant (Spearman's rank correlation coefficient＝0.77, p＝ 0.10). As for compounds other than 7-O-glucosides, the pIC50 value of sophoricoside, the 4?-O-glucoside of genistein, was 1.24 smaller than that of genistein.

Although a previous study has shown that some glucosides of flavonoids seem to inhibit BCRP-mediated transport,32) the present study is the first to examine quantitatively and systematically the differences in inhibitory potency between glucosides and the corresponding aglycones. We found that the pIC50 values of biochanin A and genistein were comparable (6.62 vs 6.30), whereas Zhang et al. reported that biochanin A was 9 times more potent than genistein in inhibiting BCRP-mediated transport.30) This discrepancy may be explained by the differences in the experimental systems and substrates used, i.e., Zhang et al. evaluated the increase in the accumulation of mitoxantrone in BCRP-overexpressing cells, while we directly evaluated the inhibition of MTX uptake into inside-out vesicles. Although daidzein-7-glucuronide, the glucuronic acid conjugate of daidzein, inhibited BCRP-mediated transport, its IC50 value was about 100-fold larger than that of daidzein. In an uptake study using membrane vesicles of BCRP-overexpressing cells, the Km value of SN-38 glucuronide was reported to be greater than that of SN-38 (26 vs 4.0 mM49)). Thus, the results overall indicate that glucuronidation may reduce the affinity of a substrate or inhibitor for BCRP. The inhibitory potency of daidzein-4?-sulfate was comparable to that of daidzein (pIC50 value; 5.41 vs 5.62). Some sulfate conjugates are known to be substrates or inhibitors of BCRP.50,51) Estrone-3-sulfate inhibited the BCRP-mediated transport of [3H]estrone-3-sulfate more potent than did estrone (¿80% vs ¿30% at 30 mM), and the inhibitory effect of 17b-estradiol-3-sulfate was com-

Effects of Herbal Extracts and Isoflavonoids on BCRP

parable to that of 17b-estradiol.52) These published observations are consistent with the results of the present study. The mean plasma concentrations of isoflavonoids have been reported to be submicromolar after consumption of soybean products containing abundant isoflavonoids.53) Considering that the plasma free-fractions of isoflavones are relatively low (e.g., º12% for daidzein54)), calculated unbound concentrations of isoflavonoids were lower than the IC50 values for inhibition of BCRP-mediated transport, whereas epithelial cells of the gastrointestinal tract are exposed to high levels of isoflavonoids. Drugisoflavonoid interactions on intestinal BCRP were predicted to be induced by taking more than 0.16 mg genistein or 0.042 mg coumestrol ([I]2/IC50À1043)). More studies are needed to clarify the possibility of clinical interactions between isoflavonoids and the substrates of BCRP. In conclusion, some herbal extracts and isoflavonoids strongly inhibited BCRP, and the inhibitory potencies of isoflavonoids were characterized in relation to the functional moieties. Some herbal or dietary supplements containing these extracts or constituents seem likely to increase the systemic availability of BCRP substrates when concomitantly taken orally.

Acknowledgments: We thank Tokiwa Phytochemical Co., Ltd. for the gifts of herbal extracts.

8)

9)

10)

11)

12)

13)

References 1)

2)

3)

4)

5)

6)

7)

Oostendorp, R. L., Beijnen, J. H. and Schellens, J. H. M.: The biological and clinical role of drug transporters at the intestinal barrier. Cancer Treat. Rev., 35: 137–147 (2009). Lilja, J. J., Niemi, M., Fredrikson, H. and Neuvonen, P. J.: Effects of clarithromycin and grapefruit juice on the pharmacokinetics of glibenclamide. Br. J. Clin. Pharmacol., 63: 732–740 (2007). Doyle, L. A., Yang, W., Abruzzo, L. V., Krogmann, T., Gao, Y., Rishi, A. K. and Ross, D. D.: A multidrug resistance transporter from human MCF-7 breast cancer cells. Proc. Natl. Acad. Sci. U.S.A., 95: 15665–15670 (1998). Miyake, K., Mickley, L., Litman, T., Zhan, Z., Robey, R., Cristensen, B., Brangi, M., Greenberger, L., Dean, M., Fojo, T. and Bates, S. E.: Molecular cloning of cDNAs which are highly overexpressed in mitoxantrone-resistant cells: demonstration of homology to ABC transport genes. Cancer Res., 59: 8–13 (1999). Allikmets, R., Schriml, L. M., Hutchinson, A., Romano-Spica, V. and Dean, M.: A human placenta-specific ATP-binding cassette gene (ABCP) on chromosome 4q22 that is involved in multidrug resistance. Cancer Res., 58: 5337–5339 (1998). Maliepaard, M., Scheffer, G. L., Faneyte, I. F., van Gastelen, M. A., Pijnenborg, A. C., Schinkel, A. H., van De Vijver, M. J., Scheper, R. J. and Schellens, J. H.: Subcellular localization and distribution of the breast cancer resistance protein transporter in normal human tissues. Cancer Res., 61: 3458–3464 (2001). Huls, M., Brown, C. D. A., Windass, A. S., Sayer, R., van den

14)

15)

16)

17)

18)

177

Heuvel, J. J. M. W., Heemskerk, S., Russel, F. G. M. and Masereeuw, R.: The breast cancer resistance protein transporter ABCG2 is expressed in the human kidney proximal tubule apical membrane. Kidney Int., 73: 220–225 (2008). Volk, E. L., Farley, K. M., Wu, Y., Li, F., Robey, R. W. and Schneider, E.: Overexpression of wild-type breast cancer resistance protein mediates methotrexate resistance. Cancer Res., 62: 5035–5040 (2002). Doyle, L. A. and Ross, D. D.: Multidrug resistance mediated by the breast cancer resistance protein BCRP (ABCG2). Oncogene, 22: 7340–7358 (2003). Robey, R. W., Medina-Perez, W. Y., Nishiyama, K., Lahusen, T., Miyake, K., Litman, T., Senderowicz, A. M., Ross, D. D. and Bates, S. E.: Overexpression of the ATP-binding cassette halftransporter, ABCG2 (MXR/BCRP/ABCP1), in flavopiridolresistant human breast cancer cells. Clin. Cancer Res., 7: 145–152 (2001). Matsushima, S., Maeda, K., Kondo, C., Hirano, M., Sasaki, M., Suzuki, H. and Sugiyama, Y.: Identification of the hepatic efflux transporters of organic anions using double-transfected MadinDarby canine kidney II cells expressing human organic aniontransporting polypeptide 1B1 (OATP1B1)/multidrug resistanceassociated protein 2, OATP1B1/multidrug resistance 1, and OATP1B1/breast cancer resistance protein. J. Pharmacol. Exp. Ther., 314: 1059–1067 (2005). Hirano, M., Maeda, K., Matsushima, S., Nozaki, Y., Kusuhara, H. and Sugiyama, Y.: Involvement of BCRP (ABCG2) in the biliary excretion of pitavastatin. Mol. Pharmacol., 68: 800–807 (2005). Breedveld, P., Zelcer, N., Pluim, D., Sonmezer, O., Tibben, M. M., Beijnen, J. H., Schinkel, A. H., van Tellingen, O., Borst, P. and Schellens, J. H. M.: Mechanism of the pharmacokinetic interaction between methotrexate and benzimidazoles: potential role for breast cancer resistance protein in clinical drug-drug interactions. Cancer Res., 64: 5804–5811 (2004). Suzuki, K., Doki, K., Homma, M., Tamaki, H., Hori, S., Ohtani, H., Sawada, Y. and Kohda, Y.: Co-administration of proton pump inhibitors delays elimination of plasma methotrexate in high-dose methotrexate therapy. Br. J. Clin. Pharmacol., 67: 44–49 (2009). Jonker, J. W., Smit, J. W., Brinkhuis, R. F., Maliepaard, M., Beijnen, J. H., Schellens, J. H. and Schinkel, A. H.: Role of breast cancer resistance protein in the bioavailability and fetal penetration of topotecan. J. Natl. Cancer Inst., 92: 1651–1656 (2000). Stewart, C. F., Leggas, M., Schuetz, J. D., Panetta, J. C., Cheshire, P. J., Peterson, J., Daw, N., Jenkins, J. J., Gilbertson, R., Germain, G. S., Harwood, F. C. and Houghton, P. J.: Gefitinib enhances the antitumor activity and oral bioavailability of irinotecan in mice. Cancer Res., 64: 7491–7499 (2004). Merino, G., Jonker, J. W., Wagenaar, E., van Herwaarden, A. E. and Schinkel, A. H.: The breast cancer resistance protein (BCRP/ABCG2) affects pharmacokinetics, hepatobiliary excretion, and milk secretion of the antibiotic nitrofurantoin. Mol. Pharmacol., 67: 1758–1764 (2005). Polli, J. W., Baughman, T. M., Humphreys, J. E., Jordan, K. H., Mote, A. L., Webster, L. O., Barnaby, R. J., Vitulli, G., Bertolotti, L., Read, K. D. and Serabjit-Singh, C. J.: The systemic exposure of an N-methyl-D-aspartate receptor antagonist is limited in mice by the P-glycoprotein and breast cancer resistance protein efflux

178

19)

20)

21)

22)

23)

24)

25)

26)

27)

28)

29)

30)

31)

32)

Hirofumi TAMAKI, et al.

transporters. Drug Metab. Dispos., 32: 722–726 (2004). Zaher, H., Khan, A. A., Palandra, J., Brayman, T. G., Yu, L. and Ware, J. A.: Breast cancer resistance protein (Bcrp/abcg2) is a major determinant of sulfasalazine absorption and elimination in the mouse. Mol. Pharm., 3: 55–61 (2006). Yamasaki, Y., Ieiri, I., Kusuhara, H., Sasaki, T., Kimura, M., Tabuchi, H., Ando, Y., Irie, S., Ware, J., Nakai, Y., Higuchi, S. and Sugiyama, Y.: Pharmacogenetic characterization of sulfasalazine disposition based on NAT2 and ABCG2 (BCRP) gene polymorphisms in humans. Mol. Ther., 84: 95–103 (2008). Sparreboom, A., Loos, W. J., Burger, H., Sissung, T. M., Verweij, J., Figg, W. D., Nooter, K. and Gelderblom, H.: Effect of ABCG2 genotype on the oral bioavailability of topotecan. Cancer Biol. Ther., 4: 650–658 (2005). Zhang, W., Yu, B., He, Y., Fan, L., Li, Q., Liu, Z., Wang, A., Liu, Y., Tan, Z., Fen-Jiang, Huang, Y. and Zhou, H.: Role of BCRP 421CÀA polymorphism on rosuvastatin pharmacokinetics in healthy Chinese males. Clin. Chim. Acta., 373: 99–103 (2006). Imai, Y., Tsukahara, S., Asada, S. and Sugimoto, Y.: Phytoestrogens/flavonoids reverse breast cancer resistance protein/ABCG2-mediated multidrug resistance. Cancer Res., 64: 4346–4352 (2004). Zhang, S., Yang, X. and Morris, M. E.: Flavonoids are inhibitors of breast cancer resistance protein (ABCG2)-mediated transport. Mol. Pharmacol., 65: 1208–1216 (2004). Cooray, H. C., Janvilisri, T., van Veen, H. W., Hladky, S. B. and Barrand, M. A.: Interaction of the breast cancer resistance protein with plant polyphenols. Biochem. Biophys. Res. Commun., 317: 269–275 (2004). Liu, X., Tee, H. and Go, M.: Functionalized chalcones as selective inhibitors of P-glycoprotein and breast cancer resistance protein. Bioorg. Med. Chem., 16: 171–180 (2008). Yoshida, N., Takada, T., Yamamura, Y., Adachi, I., Suzuki, H. and Kawakami, J.: Inhibitory effects of terpenoids on multidrug resistance-associated protein 2- and breast cancer resistance protein-mediated transport. Drug Metab. Dispos., 36: 1206– 1211 (2008). Ji, Y. and Morris, M. E.: Effect of organic isothiocyanates on breast cancer resistance protein (ABCG2)-mediated transport. Pharm. Res., 21: 2261–2269 (2004). Ahmed-Belkacem, A., Macalou, S., Borrelli, F., Capasso, R., Fattorusso, E., Taglialatela-Scafati, O. and Pietro, A. D.: Nonprenylated rotenoids, a new class of potent breast cancer resistance protein inhibitors. J. Med. Chem., 50: 1933–1938 (2007). Zhang, S., Yang, X., Coburn, R. A. and Morris, M. E.: Structure activity relationships and quantitative structure activity relationships for the flavonoid-mediated inhibition of breast cancer resistance protein. Biochem. Pharmacol., 70: 627–639 (2005). Ahmed-Belkacem, A., Pozza, A., Mu ãnoz-Martáƒnez, F., Bates, S. E., Castanys, S., Gamarro, F., Pietro, A. D. and Peá rez-Victoria, J. M.: Flavonoid structure-activity studies identify 6-prenylchrysin and tectochrysin as potent and specific inhibitors of breast cancer resistance protein ABCG2. Cancer Res., 65: 4852–4860 (2005). Katayama, K., Masuyama, K., Yoshioka, S., Hasegawa, H., Mitsuhashi, J. and Sugimoto, Y.: Flavonoids inhibit breast cancer resistance protein-mediated drug resistance: transporter specificity and structure-activity relationship. Cancer Chemother. Phar-

33)

34)

35)

36)

37)

38)

39)

40)

41)

42) 43)

44)

45)

46)

47)

macol., 60: 789–797 (2007). Enokizono, J., Kusuhara, H. and Sugiyama, Y.: Effect of breast cancer resistance protein (Bcrp/Abcg2) on the disposition of phytoestrogens. Mol. Pharmacol., 72: 967–975 (2007). Wang, X. and Morris, M. E.: Effects of the flavonoid chrysin on nitrofurantoin pharmacokinetics in rats: potential involvement of ABCG2. Drug Metab. Dispos., 35: 268–274 (2007). Thompson, L. U., Boucher, B. A., Cotterchio, M., Kreiger, N. and Liu., Z.: Dietary phytoestrogens, including isoflavones, lignans, and coumestrol, in nonvitamin, nonmineral supplements commonly consumed by women in Canada. Nutr. Cancer, 59: 176–184 (2007). Eisenberg, D. M., Davis, R. B., Ettner, S. L., Appel, S., Wilkey, S., Rompay, M. V. and Kessler, R. C.: Trends in alternative medicine use in the United States, 1990–1997: results of a follow-up national survey. JAMA, 280: 1569–1575 (1998). Hayashi, C.: On the prediction of phenomena from qualitative data and the quantification of qualitative data from the mathematico-statistical point of view. Ann. Inst. Statistical Math., 3: 69–98 (1952). Mitomo, H., Kato, R., Ito, A., Kasamatsu, S., Ikegami, Y., Kii, I., Kudo, A., Kobatake, E., Sumino, Y. and Ishikawa, T.: A functional study on polymorphism of the ATP-binding cassette transporter ABCG2: critical role of arginine-482 in methotrexate transport. Biochem. J., 373: 767–774 (2003). Nozaki, Y., Kusuhara, H., Kondo, T., Iwaki, M., Shiroyanagi, Y., Nakayama, H., Horita, S., Nakazawa, H., Okano, T. and Sugiyama, Y.: Species difference in the inhibitory effect of nonsteroidal anti-inflammatory drugs on the uptake of methotrexate by human kidney slices. J. Pharmacol. Exp. Ther., 322: 1162–1170 (2007). Food Safety Commission.: Fundamental concepts in the safety assessment of foods containing soy isoflavones for the purpose of specified health use. http://www.fsc.go.jp/english/evaluationreports/newfoods_sphealth/soy_isoflavones.pdf (accessed 08/03/ 09). AFSSA.: Presentation du rapport sur ``Securite et benefices des phyto-estrogenes apportes l'alimentation-recommandations. (2005). Italian Government.: Gazzetta Ufficiale della Repubblica Italiana 18/7/2002, number 188, annex 2 (2002). Zhang, L., Zhang, Y. D., Strong, J. M., Reynolds, K. S. and Huang, S.-M.: A regulatory viewpoint on transporter-based drug interactions. Xenobiotica, 38: 709–724 (2008). Klejdus, B., Mikelov áa, R., Petrlovaá , J., Potesil, D., Adam, V., Stiborova, M., Hodek, P., Vacek, J., Kizek, R. and Kuban, V.: Evaluation of isoflavone aglycon and glycoside distribution in soy plants and soybeans by fast column high-performance liquid chromatography coupled with a diode-array detector. J. Agric. Food Chem., 53: 5848–5852 (2005). Wang, H. J. and Murphy, P. A.: Mass balance study of isoflavones during soybean processing. J. Agric. Food Chem., 44: 2377–2383 (1996). Medina, J. H., Paladini, A. C., Wolfman, C., de Stein, M. L., Calvo, D., Diaz, L. E. and Pena, C.: Chrysin (5,7-di-OH-flavone), a naturally-occurring ligand for benzodiazepine receptors, with anticonvulsant properties. Biochem. Pharmacol., 40: 2227–2231 (1990). Grundmann, O., Wang, J., McGregor, G. P. and Butterweck, V.:

Effects of Herbal Extracts and Isoflavonoids on BCRP

48)

49)

50)

Anxiolytic activity of a phytochemically characterized Passiflora incarnata extract is mediated via the GABAergic system. Planta Med., 74: 1769–1773 (2008). Xia, C. Q., Liu, N., Miwa, G. T. and Gan, L. S.: Interactions of cyclosporin A with breast cancer resistance protein. Drug Metab. Dispos., 35: 576–582 (2007). Nakatomi, K., Yoshikawa, M., Oka, M., Ikegami, Y., Hayasaka, S., Sano, K., Shiozawa, K., Kawabata, S., Soda, H., Ishikawa, T., Tanabe, S. and Kohno, S.: Transport of 7-ethyl-10-hydroxycamptothecin (SN-38) by breast cancer resistance protein ABCG2 in human lung cancer cells. Biochem. Biophys. Res. Commun., 288: 827–832 (2001). Zamek-Gliszczynski, M. J., Hoffmaster, K. A., Nezasa, K., Tallman, M. N. and Brouwer., K. L. R.: Integration of hepatic drug transporters and phase II metabolizing enzymes: mechanisms of hepatic excretion of sulfate, glucuronide, and

51)

52)

53)

54)

179

glutathione metabolites. Eur. J. Pharm. Sci., 27: 447–486 (2006). Suzuki, M., Suzuki, H., Sugimoto, Y. and Sugiyama, Y.: ABCG2 transports sulfated conjugates of steroids and xenobiotics. J. Biol. Chem., 278: 22644–22649 (2003). Imai, Y., Asada, S., Tsukahara, S., Ishikawa, E., Tsuruo, T. and Sugimoto, Y.: Breast cancer resistance protein exports sulfated estrogens but not free estrogens. Mol. Pharmacol., 64: 610–618 (2003). Kano, M., Takayanagi, T., Harada, K. Sawada, S. and Ishikawa, F.: Bioavailability of isoflavones after ingestion of soy beverages in healthy adults. J. Nutr., 136: 2291–2296 (2006). Csan áady, G. A., Oberste-Frielinghaus, H. R., Semder, B., Baur, C., Schneider, K. T. and Filser, J. G.: Distribution and unspecific protein binding of the xenoestrogens bisphenol A and daidzein. Arch. Toxicol., 76: 299–305 (2002).