An in vitro Evaluation of Cytochrome P450 Inhibition and P-Glycoprotein Interaction with Goldenseal, Ginkgo biloba, Grape Seed, Milk Thistle, and Ginseng Extracts and Their Constituents

• Abstract
• Abbreviations
• Introduction
• Materials and Methods
• Chemicals
• Human liver microsomes (HLM)
• In vitro incubations
• Cytochrome P450 assays
• P-glycoprotein ATPase assay
• Statistical analysis
• Supporting information
• Results and Discussion
• Acknowledgements
• References

Abstract

Drug-herb interactions can result from the modulation of the activities of cytochrome P450 (P450) and/or drug transporters. The effect of extracts and individual constituents of goldenseal, Ginkgo biloba (and its hydrolyzate), grape seed, milk thistle, and ginseng on the activities of cytochrome P450 enzymes CYP1A2, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1, and CYP3A4 in human liver microsomes were determined using enzyme-selective probe substrates, and their effect on human P-glycoprotein (Pgp) was determined using a baculovirus expression system by measuring the verapamil-stimulated, vanadate-sensitive ATPase activity. Extracts were analyzed by HPLC to standardize their concentration(s) of constituents associated with the pharmacological activity, and to allow comparison of their effects on P450 and Pgp with literature values. Many of the extracts/constituents exerted ≥ 50 % inhibition of P450 activity. These include those from goldenseal (normalized to alkaloid content) inhibiting CYP2C8, CYP2D6, and CYP3A4 at 20 μM, ginkgo inhibiting CYP2C8 at 10 μM, grape seed inhibiting CYP2C9 and CYP3A4 at 10 μM, milk thistle inhibiting CYP2C8 at 10 μM, and ginsenosides F₁ and Rh₁ (but not ginseng extract) inhibiting CYP3A4 at 10 μM. Goldenseal extracts/constituents (20 μM, particularly hydrastine) and ginsenoside Rh₁ stimulated ATPase at about half of the activity of the model substrate, verapamil (20 μM). The data suggest that the clearance of a variety of drugs may be diminished by concomitant use of these herbs via inhibition of P450 enzymes, but less so by Pgp-mediated effects.

Abbreviations

HLM: human liver microsomes

Pgp: P-glycoprotein

P_i: inorganic phosphate

Introduction

The use of herbal remedies in the United States has been increasing, with the number of adults reporting using herbs to treat medical conditions increasing from 3 % in 1990 to 12 % in 1997, and estimated out-of-pocket expenditures reaching $5.1 billion [1]. Although classified as dietary supplements in the United States and not regulated as drugs, herbal products can interact with biochemical and physiological processes key to drug therapy, including the complex interplay between drug transporters and enzymes of metabolism. As many patients using prescription medications are concomitantly using botanical supplements, there is considerable risk for adverse herb-drug interactions.

There are numerous examples of herbs and foods affecting the disposition of pharmaceuticals. Among the most well known is grapefruit, which has been shown to affect drug bioavailability by the inhibition of metabolism and/or transport [2], [3]. St. John’s wort (Hypericum perforatum) inhibits CYP enzymes in vitro [4] and, in clinical studies, has been shown to induce transporters and CYP enzymes with concomitant decreases in steady state plasma concentrations of therapeutic agents [5]. Kava, a popular herbal anxiolytic, interacts with CYP enzymes in vitro and alters the pharmacokinetics of concomitantly administered drugs in vivo [6], [7]. Characterization of these processes aids prediction of interactions between herbal drugs and standard pharmaceuticals that may jeopardize the health of the consumer.

The goal of the present work is to investigate the interaction of goldenseal, Ginkgo biloba, grape seed, milk thistle, and ginseng extracts, along with individual constituents of these herbals, with cytochrome P450 enzymes and the efflux transporter P-glycoprotein. Goldenseal is commercially available as an ingredient in over-the-counter supplements and as a tea used for the treatment of urinary tract infections, conjunctivitis, nasal congestion, and upper respiratory tract inflammation [8]. Ginkgo biloba is the most frequently prescribed herbal remedy in Germany and is indicated for the treatment of conditions associated with cerebral vascular insufficiency, including age-related memory loss. Dietary intake of proanthocyanidins, a primary constituent of grape seed extract, has been linked through epidemiologic studies to a decreased risk of coronary heart disease, stroke and cancer [9], [10]. Milk thistle has been employed in the treatment of liver ailments, including jaundice, cirrhosis, and hepatitis [11]. Ginseng, touted for its ability to boost immune function and improve physical and athletic stamina, cognitive function, concentration, and work efficiency, is one of the most popular herbal supplements among American consumers, with annual sales totaling more than $300 million [8]. The popularity of these herbal supplements, combined with the tendency of individuals to combine herbal remedies with prescription medications, underscores the need for better characterization of potential herb-drug interactions.

Materials and Methods

Chemicals

Dried rhizomes of goldenseal [Hydrastis canadensis L. (Ranunculaceae)] were harvested from a cultivated plot in Wisconsin, USA in September 2001 and authenticated by Edward Fletcher of Strategic Sourcing, Inc. (Banner Elk, NC, USA; Lot HYCA 10/7 - 10.28.01-C). Ginkgo biloba L. (Ginkgoaceae) leaf extract, GBE50 (Batch Number GBE50 - 001 003), produced in October 2000 was obtained from Shanghai Xing Ling Sci & Tech Pharmaceutical Co., Ltd. (Shanghai, China). Grape seed [Vitis vinifera (Vitaceae)] extract, ActiVin GSE-2000-S (Lot Number 2001032902F), was provided by Dry Creek Nutrition, Inc. (Modesto, CA, USA). Milk thistle [Silybum marianum (L.) Gaertn. (Asteraceae)] extract was from Indena USA, Inc. (Seattle, WA, USA). Ginseng [Panax ginseng C.A. Mey. (Araliaceae)] root powder extract, EFLA®910 (Batch Number 3 031 978), conforming to the standards of The European Pharmacopoeia was produced in August 2003 and supplied by Flachsmann Canada Ltd. (Brampton, Canada). Hydrastine, berberine, ginkgolide A, ginkgolide B, bilobalide, and catechin were obtained from Sigma-Aldrich (St. Louis, MO, USA). Silybin B was the kind gift of Dr. Nicholas Oberlies of RTI International. Individual ginsenoside standards were purchased from Chromadex (Santa Ana, CA, USA). All other chemicals and reagents used were of the highest commercially available quality.

Human liver microsomes (HLM)

Pooled samples of male human liver microsomes (lot number HHM-247 and HHM-258) were obtained from Tissue Transformation Technologies (Edison, NJ, USA). Microsomes from individual male donors (HMMC-SD118, HMMC-SD124, and HMMC-SD129) were provided by CellzDirect (Pittsboro, NC, USA) and were pooled at RTI International.

In vitro incubations

All extract concentrations reported for in vitro incubations are nominal, as microsomal suspensions contain both an aqueous phase and a lipid membrane phase. Microsomes prepared from human liver were incubated in the presence of aqueous and ethanolic goldenseal extracts or the goldenseal constituents berberine and hydrastine, acid hydrolyzed and non-hydrolyzed Ginkgo biloba extract, grape seed extract, milk thistle extract, ginseng extract or the individual ginsenosides F₁ and Rh₁ and NADPH prior to determination of P450 enzymatic activities. Goldenseal treated incubation tubes were prepared by adding goldenseal extract (aqueous or ethanolic), berberine or hydrastine, dissolved in methanol or acetone, to each tube for a final assay concentration of 1 or 20 μM alkaloids (extracts normalized as the sum of berberine and hydrastine content). Ginkgo biloba, hydrolyzed Ginkgo biloba, grape seed, and milk thistle treated incubation tubes were prepared by adding hydrolyzed or non-hydrolyzed Ginkgo biloba extract or grape seed extract, dissolved in methanol, or milk thistle extract, dissolved in acetonitrile, to each tube for a final assay concentration of 1, 3 (grape seed only), 10 or 30 (grape seed only) μM terpene lactones (Ginkgo biloba), catechin (grape seed) or silybin B (milk thistle). Ginseng treated incubation tubes were prepared by adding ginseng extract or ginsenosides F₁ or Rh₁, dissolved in 15 % acetonitrile in water, to each tube for a final assay concentration of 1 or 10 μM ginsenosides. Control incubation tubes contained no herbal extracts or constituents. Incubations were run in triplicate and solvents were allowed to evaporate prior to initiation of assays. Microsomes and assay buffer were added to control and treated tubes as described below, and the tubes were preincubated at 37 °C for 10 min. NADPH was then added, and the incubations were maintained at 37 °C for an additional 15 min. Following initiation of the reactions by addition of substrate, assays were performed as described below.

Cytochrome P450 assays

Microsomes prepared from human liver and preincubated in the presence of herbal extracts or individual herbal constituents and/or NADPH were assayed for protein concentration, acetanilide hydroxylase activity (CYP1A2), taxol 6α-hydroxylase activity (CYP2C8), tolbutamide hydroxylase activity (CYP2C9), (S)-mephenytoin 4-hydroxylase activity (CYP2C19), dextromethorphan O-demethylase activity (CYP2D6), p-nitrophenol hydroxylase activity (CYP2E1), midazolam 1-hydroxylase activity (CYP3A4), testosterone 6β-hydroxylase activity (CYP3A4), and lauric acid ω-hydroxylase activity (CYP4A) as previously described [7].

P-glycoprotein ATPase assay

Human P-glycoprotein (Pgp), expressed in a baculovirus expression system obtained from BD Biosciences (Bedford, MA, USA), was used for the determination of verapamil-stimulated, vanadate-sensitive ATPase activity, an activity indicative of substrate binding to the active site of Pgp, as described previously [7].

Statistical analysis

The values for the enzyme activities were compared by ANOVA followed by Dunnett’s test. Statistically significant differences were determined at the p < 0.05 level.

Supporting information

The standardization of the extracts used is available as Supporting Information.

Results and Discussion

In order to standardize herbal assay concentrations, each herbal preparation was analyzed for constituents associated with pharmacological activity (see Supporting Information text and figures).

Extraction of goldenseal root powder with water or absolute ethanol resulted in a 27.6 or 15.1 % yield of total extracted material, respectively. Aqueous extracts, prepared to simulate goldenseal teas commercially available to consumers, contained 8.0 and 4.5 % (dry extract weight) of the active components, berberine and hydrastine, respectively, while ethanolic extracts contained 16.6 % berberine and 12.3 % hydrastine. These variations in extract composition, dependent on the method of extraction, highlight the potential for significant variations in efficacy and/or drug interactions in different goldenseal preparations.

Ginkgo biloba extract was standardized to the concentration of terpene lactones. Bilobalide was the terpene lactone present in the highest concentration and accounted for 5.7 % of Ginkgo biloba extract, while ginkgolides A and B accounted for 2.5 and 2.7 % of the extract, respectively.

Proanthocyanidins, including catechin, epicatechin, epigallocatechin, and epigallocatechin gallate, are among the most abundant phenolic compounds in grape seeds and are thought to be the principal pharmacologically active constituents [12]. Grape seed extract utilized in these studies was standardized to catechin content, which constituted 3.1 % of the total extract.

Milk thistle and ginseng extracts were standardized to silybin B and ginsenoside content, respectively. Silybin B accounted for 21.1 % of milk thistle extract. The ginsenosides Rb₁, Rb₂, Rc, Rg₁, Rd, and Re have been reported to account for 90 % of the total ginsenoside content of P. ginseng. These six ginsenosides, along with Rf, accounted for 8.4 % of the ginseng root powder utilized in these studies.

In experiments to determine the effect of goldenseal and its alkaloid constituents on P450 enzymes, human hepatic microsomes were incubated with aqueous or ethanolic extracts of powdered goldenseal root material, normalized to 1 or 20 μM alkaloids (defined as the sum of berberine and hydrastine concentrations), or with individual alkaloids at concentrations of 1 and 20 μM. The activities of CYP1A2, 2C8, 2C19, and 3A4 were not statistically different from control activities following incubation with 1 μM berberine or hydrastine or with aqueous or ethanolic extracts of goldenseal root powder normalized to 1 μM alkaloids (Fig. [1]). The activity of 2C9, however, was increased 35 % following incubation with ethanolic goldenseal extract normalized to 1 μM alkaloids. As the incubation concentration of alkaloids was increased to 20 μM, the inhibition by goldenseal extracts and its constituents also increased. There was a 64 % decrease in the activity of CYP2D6 following incubation with 20 μM berberine, while aqueous and ethanolic extracts of goldenseal normalized to 20 μM alkaloids inhibited this activity by 76 and 78 %, respectively. The activity of CYP3A4, as determined by midazolam 1-hydroxylase activity, was diminished by 27 % following incubation with 20 μM hydrastine. Aqueous and ethanolic extracts of goldenseal normalized to 20 μM alkaloids inhibited this activity more markedly (45 and 77 %, respectively). Inhibition of CYP3A4, as determined by testosterone 6β-hydroxylase activity, following incubation with goldenseal extracts normalized to 20 μM alkaloids or with 20 μM hydrastine, was similar to that observed with midazolam 1-hydroxylase activity. Testosterone 6β-hydroxylase activity decreased 58 and 67 % following incubation with aqueous and ethanolic extracts, respectively, while incubation with hydrastine resulted in a 45 % decrease in the activity of this enzyme. While 20 μM berberine did not inhibit midazolam 1-hydroxylase activity, it did modestly decrease testosterone 6β-hydroxylase activity. CYP2E1 activity was also modestly, but significantly, decreased following incubation of human liver microsomes with 20 μM berberine. The activity of CYP1A2, while unaffected by higher concentrations of goldenseal extracts or berberine, was inhibited 24 % following incubation with 20 μM hydrastine.

Chatterjee and Franklin [13] have also demonstrated inhibition of the activities of P450 enzymes 2C9 (diclofenac hydroxylation), 2D6 (bufuralol hydroxylation), and 3A4 (testosterone 6β-hydroxylation) following incubation of human liver microsomes with goldenseal extract and its methylenedioxyphenyl-containing constituents, berberine and hydrastine. In these studies, inhibition of CYP2C9 and 2D6 was evident with goldenseal extract at final incubation concentrations as low as 60 - 75 μM total alkaloids and increased in a concentration-dependent manner. Similar to findings in the present study, hydrastine isomers did not inhibit the activity of either CYP2C9 or 2D6 at concentration ≤ 20 μM. Preincubation of human hepatic microsomes with berberine at concentrations as low as 10 μM, however, resulted in significant inhibition of CYP2D6 activity. Gurley et al. [14] have demonstrated inhibition of CYP2D6 and CYP3A4 activities following administration of goldenseal to human volunteers. Together these enzymes account for the metabolism of ∼70 % of prescription medications [15], indicating a significant potential for altered efficacy and/or toxicity of therapeutic agents upon concomitant ingestion of goldenseal supplements. In contrast to the present investigations, these authors did not examine the effects of aqueous extracts of goldenseal, such as those that would result from the preparation of goldenseal teas, on enzyme activities and did not investigate the other P450 enzymes reported here.

In order to determine the inhibitory potential of deglycosylated Ginkgo biloba constituents, ginkgo extract was acid hydrolyzed prior to preincubation with human hepatic microsomes. Although flavonoids are present in plant material primarily as glycosides [16], aglycones are present in low concentrations. Additionally, flavonoid glycosides can be deglycosylated in the gut by glycosidases present in microflora [17] or found in intestinal tissue as membrane-bound enzymes on the luminal side of the brush border [18], [19].

The diminution of P450 activities was similar following incubation of human liver microsomes with both acid-hydrolyzed and non-hydrolyzed Ginkgo biloba extract. The activities of CYP2C19, 2D6, and 2E1 (10 μM only) were not significantly different from control activities following incubation with either extract preparation normalized to 1 or 10 μM terpene lactones (Fig. [2]). CYP1A2 and 2C9 activities were also unaltered following preincubation with hydrolyzed and non-hydrolyzed ginkgo extracts normalized to 1 μM terpene lactones. However, as the concentration of terpene lactones was increased to 10 μM, CYP1A2 activity was decreased 26 % in the presence of acid hydrolyzed extract, while this activity was not significantly different from control values when microsomes were incubated with non-hydrolyzed extract. CYP2C9 activity was decreased 16 and 45 % following incubation with 10 μM non-hydrolyzed and hydrolyzed extracts, respectively. CYP2C8 activity was also significantly decreased following incubation with non-hydrolyzed ginkgo extract at both concentrations, with activities diminished 31 and 54 %, respectively, at concentrations of 1 and 10 μM terpene lactones. Inhibition of CYP2C8 also exhibited concentration-dependence following incubation with acid hydrolyzed ginkgo extract, with 57 % diminution of activity at a concentration of 1 μM terpene lactones and complete loss of this activity at a concentration of 10 μM terpene lactones. CYP3A4 activity, as measured by midazolam 1-hydroxylation, was inhibited 46 and 35 % by hydrolyzed and non-hydrolyzed ginkgo extracts, respectively. CYP3A4-catalyzed testosterone 6β-hydroxylase activity was decreased appreciably (48 %) only by 10 μM hydrolyzed extract.

These results are in agreement with the findings of Gaudineau et al. [20] that demonstrated K_i values of 106 ± 24 μg/mL, 14 ± 4 μg/mL, >900 μg/mL, and 127 ± 42 μg/mL for CYP1A2, 2C9, 2D6, and 2E1, respectively. These authors utilized the standardized EGb761 G. biloba extract, reported to contain 24 % flavonoids (glycosidic derivatives of quercetin, kaempferol, isorhamnetin, and myricetin, and traces of the aglycones), 6 % terpenoids (ginkgolides A, B, C, J, and M, and bilobalide), and 0.5 - 1 % organic acids (acetic acid, shikimic acid, p-hydroxybenzoic acid, vanillic acid, kynurenic acid, and ascorbic acid). According to the authors’ calculations, assuming an average molecular weight of 500 for extract constituents, results reported in μg/mL correspond approximately to μM concentrations. Using probe drug cocktails for 1A2, 2D6, 2E1, and 3A4, Gurley et al. [21], [22] found no significant changes on phenotypic metabolic ratios in humans following administration of Ginkgo biloba. Uchida et al. [23] observed a modest increase in CYP2C9 mediated-metabolism of tolbutamide in clinical pharmacokinetic studies with Gingko biloba, but a decrease in CYP3A4-mediated metabolism of midazolam consistent with the in vitro findings of the present work.

Grape seed extract, normalized to 1 μM catechin, did not inhibit the activity of any P450 enzyme assayed, although there was a 30 % increase in CYP2C9 activity (Fig. [3]). However, as the incubation concentration of grape seed extract was increased, the activities of numerous enzymes were inhibited. Following incubation with grape seed extract normalized to 10 μM catechin, the activities of CYP2C9 and 2D6 were decreased 58 and 46 %, respectively, while CYP3A4-catalyzed testosterone 6β-hydroxylase activity was decreased by 31 % and midazolam 1-hydroxylase activity was virtually abolished. In a subsequent study to investigate the precipitous drop in the activity of this enzyme, human hepatic microsomes were incubated with grape seed extract normalized to 3 and 30 μM catechin. Following incubation with grape seed extract normalized to 3 μM catechin midazolam 1-hydroxylase activity was decreased 79 %, however, testosterone 6β-hydroxylase activity was not significantly decreased. Similar to results seen following incubation with grape seed extract normalized to 10 μM catechin, midazolam 1-hydroxylase activity was also abolished at a normalized concentration of 30 μM catechin. Testosterone 6β-hydroxylase activity was diminished 86 % at this extract concentration. No clinical investigations of the effect of grape seed extract on P450-mediated metabolism were found in the literature.

Incubation of human hepatic microsomes with milk thistle extract normalized to 1 μM silybin B did not inhibit the activity of any P450 enzyme assayed (Fig. [4]). However, as the incubation concentration of milk thistle was increased, the activities of CYP2C8 and 3A4 were significantly inhibited. Following incubation with milk thistle extract normalized to 10 μM silybin B, the activity of CYP2C8 was decreased 66 %, while midazolam 1-hydroxylase and testosterone 6β-hydroxylase activities, markers for CYP3A4, were both decreased 43 % by milk thistle extract at this concentration. These results are in agreement with the findings of Sridar et al. [24] demonstrating time- and concentration-dependent inactivation of purified CYP3A4 in the presence of silybin and NADPH. These authors also noted no inhibitory effect of silybin on purified CYP2D6 and 2E1. In in vivo studies, Gurley et al. [25] also found no inhibition of the activities of CYP1A2, 2D6, or 2E1 following repeated administration of milk thistle extract to twelve healthy human volunteers for 28 days. Unlike results seen following in vitro incubation of milk thistle extract with HLM or purified enzyme, the activity of CYP3A4 was also unaltered following repeat administration of milk thistle extract to humans, suggesting that in vivo levels of herbal constituents do not reach inhibitory concentrations [26].

None of the P450 enzyme activities measured was significantly altered following incubation with ginseng extract normalized to 1 or 10 μM ginsenosides (Fig. [5]). However, following oral administration of ginseng extract to humans at dose levels equivalent to the German Commission E daily recommendation, only three ginsenoside metabolites appeared to reach systemic circulation. Along with Compound K, the primary metabolite of protopanaxadiol ginsenosides produced by intestinal bacteria, ginsenosides F₁ and Rh₁, hydrolysis products of the protopanaxatriol ginsenosides, were detectable in plasma samples [27]. Those workers have suggested that these products are responsible for the pharmacological action of ginseng in humans. Therefore, P450 activities were also assayed in the presence of ginsenosides F₁ and Rh₁. When microsomes were incubated with the individual ginsenosides F₁ and Rh₁ at concentrations of 1 and 10 μM, there were no significant alterations in the activities of CYP1A2, 2C8, 2C9, 2C19, 2D6, or testosterone 6β-hydroxylase. Midazolam 1-hydroxylase activities were also unaffected following incubation with individual ginsenosides at a concentration of 1 μM. However, as the concentration of each ginsenoside was increased to 10 μM, midazolam 1-hydroxylase activity was significantly decreased, with F₁ and Rh₁ producing 60 and 54 % decreases, respectively. Using probe drug cocktails for 1A2, 2D6, 2E1, and 3A4, Gurley et al. [21] found no significant changes on phenotypic metabolic ratios in humans following administration of ginseng. Consistent with our findings, Liu et al. [28] found that ginseng extract did not inhibit the activities of these enzymes in human liver microsomes. Ginseng hydrolysis products, however, inhibited 2C9 and 3A4, but with K_i values higher than C_max anticipated for humans. Similar to the findings reported herein, Liu et al. [29] showed that neither the hydrolysis products Rh₁ or F₁ were responsible for this inhibition. Yuan et al. [30] found that repeat administration of ginseng to human volunteers reduced the effectiveness of warfarin, suggesting induction, rather than inhibition of CYP2C9 in those studies.

In addition to modulating the activities of metabolic enzymes, herbal constituents have been demonstrated to interact with transport proteins, including the human multidrug resistance transporter P-glycoprotein. Numerous clinically prescribed drugs, including cardiac glycosides, antihypertensives, and chemotherapeutics, are known substrates for this transporter [31]. Interactions with herbal constituents, therefore, present the potential for alterations in absorption, distribution, and elimination of these therapeutics, indicating possible herb-drug interactions leading to modulation of both drug efficacy and adverse events.

Following incubation with either aqueous or ethanolic goldenseal extracts, Pgp ATPase activity increased in a concentration-dependent manner (Table [1]). Membranes incubated in the absence of either extract liberated 6.2 nmol of inorganic phosphate (P_i)/mg protein/min. As the extract concentration was increased, ATPase activity increased from ca. 8 - 9 nmol P_i/mg protein/min at a concentration of 1 μM alkaloids to ca. 12 - 15 nmol P_i/mg protein/min at 20 μM alkaloids. In comparison, when membranes were incubated in the presence of 20 μM verapamil, a known Pgp substrate, ATPase activity increased to 30.5 nmol P_i/mg protein/min. Further experiments were subsequently conducted, incubating membranes with the goldenseal constituents berberine and hydrastine at concentrations of 1 and 20 μM. Similar to results obtained following incubation of vesicles with goldenseal extracts, both berberine and hydrastine increased Pgp ATPase activity (Table [2]), with substantial increases in activity following incubation particularly with hydrastine (16 nmol P_i/mg protein/min). Increases in Pgp ATPase activity observed with berberine were maximal at a concentration of 1 μM. Results of Maeng et al. [32], demonstrating inhibition of berberine flux by prototypical Pgp substrates including verapamil, daunomycin, and rhodamine 123, further support the conclusion that this alkaloid is a substrate for Pgp. This was not borne out in clinical studies with goldenseal and digoxin, however [33].

There was no detectable Pgp ATPase activity in the presence of acid hydrolyzed or non-hydrolyzed ginkgo extract normalized to 1 μM terpene lactones (Table [3]). As the concentration of each extract was increased to 10 μM terpene lactones, ATPase activity was modestly increased for the non-hydrolyzed extract preparation. In rats, ginkgo markedly decreased, rather than increased, cyclosporin bioavailability [34]. Those authors suggested that inhibition of Pgp may be overwhelmed by the net induction of the transporter caused by repeat administration of ginkgo.

Incubation of membrane vesicles with grape seed extract normalized to 1 μM or 10 μM catechin, or milk thistle extract normalized to 1 μM or 10 μM silybin B, failed to stimulate Pgp ATPase, with activities below the lower limit of detection at both concentrations (data not shown). Although milk thistle extract failed to stimulate ATPase activity in the present study, an interaction between this herb and Pgp cannot be ruled out. Zhang and Morris [35] also reported a lack of Pgp ATPase stimulation following incubation of the milk thistle constituent silymarin with Pgp in vitro, consistent with the lack of effect on digoxin pharmacokinetics in clinical studies [36]. No reports of the effect of grape seed extract on Pgp were found in the clinical literature.

Incubation of vesicles with ginseng extract resulted in a concentration-dependent stimulation of ATPase activity (Table [4]). At a concentration of 10 μM ginsenosides, ginseng extract increased ATPase activity to 10 nmol P_i/mg protein/min. Neither F₁ nor Rh₁ stimulated ATPase activity at a concentration of 1 μM, but Rh₁ at a concentration of 10 μM resulted in the production of 17 nmol P_i/mg protein/min. The concentration-dependent increase in Pgp ATPase activity observed following incubation of membrane vesicles with ginseng extract in the present study is also in agreement with the findings of Choi et al. [37] that showed protopanaxatriol ginsenosides were able to completely inhibit azidopine photolabelling of Pgp and reverse doxorubicin resistance in Pgp overexpressing acute myelogenous leukemia cells.

Despite the popularity of herbal supplements among consumers, limited information concerning potential interactions with clinically-prescribed therapeutics is available. Key clinical information on circulating levels of herbal components are needed to allow extrapolation of data such as that reported here to the clinical setting. However, in the absence of clinical data, the results presented herein can help guide the design and interpretation of subsequent trials in humans. The P450 enzymes investigated in the present work are involved in the metabolism of >90 % of pharmaceutical agents [38]. The present work, however, investigated the inhibition of those enzymes in vitro, but induction of expression of those proteins in vivo following ingestion of herbal supplements may lead to opposite effects in the clinic. Results of the present study, indicating significant inhibition of metabolic enzymes and efflux transport, indicate the potential for alterations in bioavailability and metabolism of pharmaceuticals concomitantly ingested with popular herbal preparations, including goldenseal, Ginkgo biloba, grape seed, milk thistle, and ginseng. The activities of four P450 enzymes, CYP2C8, CYP2C9, CYP2D6, and CYP3A4, were diminished by ≥ 50 % following in vitro incubation with one or more of these herbal extracts and/or their individual constituents. In the present work, goldenseal, along with both Ginkgo biloba and milk thistle, also significantly decreased the activity of CYP2C8. Among the substrates for this enzyme are the thiazolidinediones, a class of oral antidiabetic agents, and the chemotherapeutic paclitaxel. Cognitive dysfunction resulting from chemotherapy with paclitaxel may lead cancer patients to attempt to alleviate these unwanted side effects by self-medicating with Ginkgo biloba, with the potential of attendant increases in blood levels of paclitaxel. These examples illustrate the need for caution when combining herbal supplements and prescription medications and for additional investigations into the in vivo effects of herbal constituents on enzymes of metabolism and transport.

Acknowledgements

The authors are grateful to Kathy Ancheta for her assistance in preparation of the manuscript, and to Dr. Nick Oberlies for his insightful comments in its review. This work was performed under National Institute of Environmental Health Sciences contract N01-ES-25 482.

References

• 1 Eisenberg D M, Davis R B, Ettner S L, Appel S, Wilkey S, Van Rompay M. et al . Trends in alternative medicine use in the United States, 1990 - 1997.  JAMA. 1998;  280 1569-75.
  CrossrefPubMedSearch in Google Scholar
• 2 Lown K S, Bailey D G, Fontana R J, Janardan S K, Adair C H, Fortlage L A. et al . Grapefruit juice increases felodipine oral availability in humans by decreasing intestinal CYP 3A protein expression.  J Clin Invest. 1997;  99 2545-53.
  CrossrefPubMedSearch in Google Scholar
• 3 Mouly S J, Paine M F, Watkins P B. Contributions of CYP3A4, P-glycoprotein, and serum protein binding to the intestinal first-pass extraction of saquinavir.  J Pharmacol Exp Ther. 2004;  308 941-8.
  PubMedSearch in Google Scholar
• 4 Obach R S. Inhibition of human cytochrome P450 enzymes by constituents of St. John’s Wort, an herbal preparation used in the treatment of depression.  J Pharmacol Exp Ther. 2000;  294 88-95.
  PubMedSearch in Google Scholar
• 5 Dresser G K, Schwarz U I, Wilkinson G R, Kim R B. Coordinate induction of both cytochrome P4503A and MDR1 by St John’s wort in healthy subjects.  Clin Pharmacol Ther. 2003;  73 41-50.
  CrossrefPubMedSearch in Google Scholar
• 6 Mathews J M, Etheridge A S, Black S R. Inhibition of human cytochrome P450 activities by kava extract and kavalactones.  Drug Metab Dispos. 2002;  30 1153-7.
  CrossrefPubMedSearch in Google Scholar
• 7 Mathews J M, Etheridge A S, Valentine J L, Black S R, Coleman D P, Patel P. et al . Pharmacokinetics and disposition of the kavalactone kawain: interaction with kava extract and kavalactones in vivo and in vitro .  Drug Metab Dispos. 2005;  33 1555-63.
  CrossrefPubMedSearch in Google Scholar
• 8 Jellin J M, Gregory P, Batz F, Hitchens K. Pharmacist’s Letter/Prescriber’s Letter. Stockton, CA; Natural Medicines Comprehensive Database 2000.
  Search in Google Scholar
• 9 Hertog M G, Kromhout D, Aravanis C, Blackburn H, Buzina R, Fidanza F. et al . Flavonoid intake and long-term risk of coronary heart disease and cancer in the seven countries study.  Arch Intern Med. 1995;  155 381-6.
  CrossrefPubMedSearch in Google Scholar
• 10 Keli S O, Hertog M G, Feskens E J, Kromhout D. Dietary flavonoids, antioxidant vitamins, and incidence of stroke: the Zutphen study.  Arch Intern Med. 1996;  156 637-42.
  CrossrefPubMedSearch in Google Scholar
• 11 Blumenthal M, Busse W R, Goldberg A, Gruenwald J, Hall T, Riggins C W. et al .The complete German Commission E Monographs: Therapeutic Guide to Herbal Medicines. Austin, TX; American Botanical Council 1998: 563-5.
  Search in Google Scholar
• 12 Santos-Buelga C, Scalbert A. Proanthcyanidins and tannin-like compounds in human nutrition.  J Food Sci Agric. 2000;  80 1094-117.
  CrossrefPubMedSearch in Google Scholar
• 13 Chatterjee P, Franklin M R. Human cytochrome P450 inhibition and metabolic-intermediate complex formation by goldenseal extract and its methylenedioxyphenyl components.  Drug Metab Dispos. 2003;  31 1391-7.
  CrossrefPubMedSearch in Google Scholar
• 14 Gurley B J, Gardner S F, Hubbard M A, Williams D K, Gentry W B, Khan I A. et al . In vivo effects of goldenseal, kava kava, black cohosh, and valerian on human cytochrome P450 1A2, 2D6, 2E1, and 3A4/5 phenotypes.  Clin Pharmacol Ther. 2005;  77 415-26.
  CrossrefPubMedSearch in Google Scholar
• 15 Lewis D F, Eddershaw P J, Goldfarb P S, Tarbit M H. Molecular modelling of cytochrome P4502D6 (CYP2D6) based on an alignment with CYP102: structural studies on specific CYP2D6 substrate metabolism.  Xenobiotica. 1997;  27 319-39.
  CrossrefPubMedSearch in Google Scholar
• 16 Hollman P C. Bioavailability of flavonoids.  Eur J Clin Nutr. 1997;  51 Suppl 1 S66-9.
  PubMedSearch in Google Scholar
• 17 Bokkenheuser V D, Shackleton C H, Winter J. Hydrolysis of dietary flavonoid glycosides by strains of intestinal bacteroides from humans.  Biochem J. 1987;  248 953-6.
  PubMedSearch in Google Scholar
• 18 Day A J, DuPont M S, Ridley S, Rhodes M, Rhodes M J, Morgan M R. et al . Deglycosylation of flavonoid and isoflavonoid glycosides by human small intestine and liver beta-glucosidase activity.  FEBS Lett. 1998;  436 71-5.
  CrossrefPubMedSearch in Google Scholar
• 19 Walgren R A, Lin J T, Kinne R K, Walle T. Cellular uptake of dietary flavonoid quercetin 4’-beta-glucoside by sodium-dependent glucose transporter SGLT1.  J Pharmacol Exp Ther. 2000;  294 837-43.
  PubMedSearch in Google Scholar
• 20 Gaudineau C, Beckerman R, Welbourn S, Auclair K. Inhibition of human P450 enzymes by multiple constituents of the Ginkgo biloba extract.  Biochem Biophys Res Commun. 2004;  318 1072-8.
  CrossrefPubMedSearch in Google Scholar
• 21 Gurley B J, Gardner S F, Hubbard M A, Williams D K, Gentry W B, Cui Y. et al . Cytochrome P450 phenotypic ratios for predicting herb-drug interactions in humans.  Clin Pharmacol Ther. 2002;  72 276-87.
  CrossrefPubMedSearch in Google Scholar
• 22 Gurley B J, Gardner S F, Hubbard M A, Williams D K, Gentry W B, Cui Y. et al . Clinical assessment of effects of botanical supplementation on cytochrome P450 phenotypes in the elderly: St John’s wort, garlic oil, Panax ginseng and Ginkgo biloba .  Drugs Aging. 2005;  22 525-39.
  CrossrefPubMedSearch in Google Scholar
• 23 Uchida S, Yamada H, Li X D, Maruyama S, Ohmori Y, Oki T. et al . Effects of Ginkgo biloba extract on pharmacokinetics and pharmacodynamics of tolbutamide and midazolam in healthy volunteers.  J Clin Pharmacol. 2006;  46 1290-8.
  CrossrefPubMedSearch in Google Scholar
• 24 Sridar C, Goosen T C, Kent U M, Williams J A, Hollenberg P F. Silybin inactivates cytochromes P450 3A4 and 2C9 and inhibits major hepatic glucuronosyltransferases.  Drug Metab Dispos. 2004;  32 587-94.
  CrossrefPubMedSearch in Google Scholar
• 25 Gurley B J, Gardner S F, Hubbard M A, Williams D K, Gentry W B, Carrier J. et al . In vivo assessment of botanical supplementation on human cytochrome P450 phenotypes: Citrus aurantium, Echinacea purpurea, milk thistle, and saw palmetto.  Clin Pharmacol Ther. 2004;  76 428-40.
  CrossrefPubMedSearch in Google Scholar
• 26 Gurley B, Hubbard M A, Williams D K, Thaden J, Tong Y, Gentry W B. et al . Assessing the clinical significance of botanical supplementation on human cytochrome P450 3A activity: comparison of a milk thistle and black cohosh product to rifampin and clarithromycin.  J Clin Pharmacol. 2006;  46 201-13.
  CrossrefPubMedSearch in Google Scholar
• 27 Tawab M A, Bahr U, Karas M, Wurglics M, Schubert-Zsilavecz M. Degradation of ginsenosides in humans after oral administration.  Drug Metab Dispos. 2003;  31 1065-71.
  CrossrefPubMedSearch in Google Scholar
• 28 Liu Y, Zhang J W, Li W, Ma H, Sun J, Deng M C. et al . Ginsenoside metabolites, rather than naturally occurring ginsenosides, lead to inhibition of human cytochrome P450 enzymes.  Toxicol Sci. 2006;  91 356-64.
  CrossrefPubMedSearch in Google Scholar
• 29 Liu Y, Ma H, Zhang J W, Deng M C, Yang L. Influence of ginsenoside Rh1 and F1 on human cytochrome P450 enzymes.  Planta Med. 2006;  72 126-31.
  Thieme ConnectPubMedSearch in Google Scholar
• 30 Yuan C S, Wei G, Dey L, Karrison T, Nahlik L, Maleckar S. et al . Brief communication: American ginseng reduces warfarin’s effect in healthy patients: a randomized, controlled trial.  Ann Intern Med. 2004;  141 23-7.
  CrossrefPubMedSearch in Google Scholar
• 31 Rautio J, Humphreys J E, Webster L O, Balakrishnan A, Keogh J P, Kunta J R. et al . In vitro P-glycoprotein inhibition assays for assessment of clinical drug interaction potential of new drug candidates: a recommendation for probe substrates.  Drug Metab Dispos. 2006;  7 7.
  PubMedSearch in Google Scholar
• 32 Maeng H J, Yoo H J, Kim I W, Song I S, Chung S J, Shim C K. P-glycoprotein-mediated transport of berberine across Caco-2 cell monolayers.  J Pharm Sci. 2002;  91 2614-21.
  CrossrefPubMedSearch in Google Scholar
• 33 Gurley B J, Swain A, Barone G W, Williams D K, Breen P, Yates C R. et al . Effect of goldenseal (Hydrastis canadensis) and kava kava (Piper methysticum) supplementation on digoxin pharmacokinetics in humans.  Drug Metab Dispos. 2007;  35 240-5.
  PubMedSearch in Google Scholar
• 34 Yang C Y, Chao P D, Hou Y C, Tsai S Y, Wen K C, Hsiu S L. Marked decrease of cyclosporin bioavailability caused by coadministration of ginkgo and onion in rats.  Food Chem Toxicol. 2006;  44 1572-8.
  PubMedSearch in Google Scholar
• 35 Zhang S, Morris M E. Effects of the flavonoids biochanin A, morin, phloretin, and silymarin on P-glycoprotein-mediated transport.  J Pharmacol Exp Ther. 2003;  304 1258-67.
  PubMedSearch in Google Scholar
• 36 Gurley B J, Barone G W, Williams D K, Carrier J, Breen P, Yates C R. et al . Effect of milk thistle (Silybum marianum) and black cohosh (Cimicifuga racemosa) supplementation on digoxin pharmacokinetics in humans.  Drug Metab Dispos. 2006;  34 69-74.
  PubMedSearch in Google Scholar
• 37 Choi C H, Kang G, Min Y D. Reversal of P-glycoprotein-mediated multidrug resistance by protopanaxatriol ginsenosides from Korean red ginseng.  Planta Med. 2003;  69 235-40.
  Thieme ConnectPubMedSearch in Google Scholar
• 38 Benet L Z, Kroetz D L, Sheiner L B. Pharmacokinetics: the dynamics of drug absorption, distribution, and elimination. In: Hardman JG, editor Goodman & Gilman’s The Pharmacological Basis of Therapeutics. New York, NY; McGraw-Hill 1996: 3-27.
  Search in Google Scholar

Dr. James M. Mathews

Health Sciences Unit

Science and Engineering

RTI International

3040 Cornwallis Road

P.O. Box 12194

Research Triangle Park

NC 27709-2194

USA

Phone: +1-919-541-7461

Fax: +1-919-541-6499

Email: mathews@rti.org

References

• 1 Eisenberg D M, Davis R B, Ettner S L, Appel S, Wilkey S, Van Rompay M. et al . Trends in alternative medicine use in the United States, 1990 - 1997.  JAMA. 1998;  280 1569-75.
  CrossrefPubMedSearch in Google Scholar
• 2 Lown K S, Bailey D G, Fontana R J, Janardan S K, Adair C H, Fortlage L A. et al . Grapefruit juice increases felodipine oral availability in humans by decreasing intestinal CYP 3A protein expression.  J Clin Invest. 1997;  99 2545-53.
  CrossrefPubMedSearch in Google Scholar
• 3 Mouly S J, Paine M F, Watkins P B. Contributions of CYP3A4, P-glycoprotein, and serum protein binding to the intestinal first-pass extraction of saquinavir.  J Pharmacol Exp Ther. 2004;  308 941-8.
  PubMedSearch in Google Scholar
• 4 Obach R S. Inhibition of human cytochrome P450 enzymes by constituents of St. John’s Wort, an herbal preparation used in the treatment of depression.  J Pharmacol Exp Ther. 2000;  294 88-95.
  PubMedSearch in Google Scholar
• 5 Dresser G K, Schwarz U I, Wilkinson G R, Kim R B. Coordinate induction of both cytochrome P4503A and MDR1 by St John’s wort in healthy subjects.  Clin Pharmacol Ther. 2003;  73 41-50.
  CrossrefPubMedSearch in Google Scholar
• 6 Mathews J M, Etheridge A S, Black S R. Inhibition of human cytochrome P450 activities by kava extract and kavalactones.  Drug Metab Dispos. 2002;  30 1153-7.
  CrossrefPubMedSearch in Google Scholar
• 7 Mathews J M, Etheridge A S, Valentine J L, Black S R, Coleman D P, Patel P. et al . Pharmacokinetics and disposition of the kavalactone kawain: interaction with kava extract and kavalactones in vivo and in vitro .  Drug Metab Dispos. 2005;  33 1555-63.
  CrossrefPubMedSearch in Google Scholar
• 8 Jellin J M, Gregory P, Batz F, Hitchens K. Pharmacist’s Letter/Prescriber’s Letter. Stockton, CA; Natural Medicines Comprehensive Database 2000.
  Search in Google Scholar
• 9 Hertog M G, Kromhout D, Aravanis C, Blackburn H, Buzina R, Fidanza F. et al . Flavonoid intake and long-term risk of coronary heart disease and cancer in the seven countries study.  Arch Intern Med. 1995;  155 381-6.
  CrossrefPubMedSearch in Google Scholar
• 10 Keli S O, Hertog M G, Feskens E J, Kromhout D. Dietary flavonoids, antioxidant vitamins, and incidence of stroke: the Zutphen study.  Arch Intern Med. 1996;  156 637-42.
  CrossrefPubMedSearch in Google Scholar
• 11 Blumenthal M, Busse W R, Goldberg A, Gruenwald J, Hall T, Riggins C W. et al .The complete German Commission E Monographs: Therapeutic Guide to Herbal Medicines. Austin, TX; American Botanical Council 1998: 563-5.
  Search in Google Scholar
• 12 Santos-Buelga C, Scalbert A. Proanthcyanidins and tannin-like compounds in human nutrition.  J Food Sci Agric. 2000;  80 1094-117.
  CrossrefPubMedSearch in Google Scholar
• 13 Chatterjee P, Franklin M R. Human cytochrome P450 inhibition and metabolic-intermediate complex formation by goldenseal extract and its methylenedioxyphenyl components.  Drug Metab Dispos. 2003;  31 1391-7.
  CrossrefPubMedSearch in Google Scholar
• 14 Gurley B J, Gardner S F, Hubbard M A, Williams D K, Gentry W B, Khan I A. et al . In vivo effects of goldenseal, kava kava, black cohosh, and valerian on human cytochrome P450 1A2, 2D6, 2E1, and 3A4/5 phenotypes.  Clin Pharmacol Ther. 2005;  77 415-26.
  CrossrefPubMedSearch in Google Scholar
• 15 Lewis D F, Eddershaw P J, Goldfarb P S, Tarbit M H. Molecular modelling of cytochrome P4502D6 (CYP2D6) based on an alignment with CYP102: structural studies on specific CYP2D6 substrate metabolism.  Xenobiotica. 1997;  27 319-39.
  CrossrefPubMedSearch in Google Scholar
• 16 Hollman P C. Bioavailability of flavonoids.  Eur J Clin Nutr. 1997;  51 Suppl 1 S66-9.
  PubMedSearch in Google Scholar
• 17 Bokkenheuser V D, Shackleton C H, Winter J. Hydrolysis of dietary flavonoid glycosides by strains of intestinal bacteroides from humans.  Biochem J. 1987;  248 953-6.
  PubMedSearch in Google Scholar
• 18 Day A J, DuPont M S, Ridley S, Rhodes M, Rhodes M J, Morgan M R. et al . Deglycosylation of flavonoid and isoflavonoid glycosides by human small intestine and liver beta-glucosidase activity.  FEBS Lett. 1998;  436 71-5.
  CrossrefPubMedSearch in Google Scholar
• 19 Walgren R A, Lin J T, Kinne R K, Walle T. Cellular uptake of dietary flavonoid quercetin 4’-beta-glucoside by sodium-dependent glucose transporter SGLT1.  J Pharmacol Exp Ther. 2000;  294 837-43.
  PubMedSearch in Google Scholar
• 20 Gaudineau C, Beckerman R, Welbourn S, Auclair K. Inhibition of human P450 enzymes by multiple constituents of the Ginkgo biloba extract.  Biochem Biophys Res Commun. 2004;  318 1072-8.
  CrossrefPubMedSearch in Google Scholar
• 21 Gurley B J, Gardner S F, Hubbard M A, Williams D K, Gentry W B, Cui Y. et al . Cytochrome P450 phenotypic ratios for predicting herb-drug interactions in humans.  Clin Pharmacol Ther. 2002;  72 276-87.
  CrossrefPubMedSearch in Google Scholar
• 22 Gurley B J, Gardner S F, Hubbard M A, Williams D K, Gentry W B, Cui Y. et al . Clinical assessment of effects of botanical supplementation on cytochrome P450 phenotypes in the elderly: St John’s wort, garlic oil, Panax ginseng and Ginkgo biloba .  Drugs Aging. 2005;  22 525-39.
  CrossrefPubMedSearch in Google Scholar
• 23 Uchida S, Yamada H, Li X D, Maruyama S, Ohmori Y, Oki T. et al . Effects of Ginkgo biloba extract on pharmacokinetics and pharmacodynamics of tolbutamide and midazolam in healthy volunteers.  J Clin Pharmacol. 2006;  46 1290-8.
  CrossrefPubMedSearch in Google Scholar
• 24 Sridar C, Goosen T C, Kent U M, Williams J A, Hollenberg P F. Silybin inactivates cytochromes P450 3A4 and 2C9 and inhibits major hepatic glucuronosyltransferases.  Drug Metab Dispos. 2004;  32 587-94.
  CrossrefPubMedSearch in Google Scholar
• 25 Gurley B J, Gardner S F, Hubbard M A, Williams D K, Gentry W B, Carrier J. et al . In vivo assessment of botanical supplementation on human cytochrome P450 phenotypes: Citrus aurantium, Echinacea purpurea, milk thistle, and saw palmetto.  Clin Pharmacol Ther. 2004;  76 428-40.
  CrossrefPubMedSearch in Google Scholar
• 26 Gurley B, Hubbard M A, Williams D K, Thaden J, Tong Y, Gentry W B. et al . Assessing the clinical significance of botanical supplementation on human cytochrome P450 3A activity: comparison of a milk thistle and black cohosh product to rifampin and clarithromycin.  J Clin Pharmacol. 2006;  46 201-13.
  CrossrefPubMedSearch in Google Scholar
• 27 Tawab M A, Bahr U, Karas M, Wurglics M, Schubert-Zsilavecz M. Degradation of ginsenosides in humans after oral administration.  Drug Metab Dispos. 2003;  31 1065-71.
  CrossrefPubMedSearch in Google Scholar
• 28 Liu Y, Zhang J W, Li W, Ma H, Sun J, Deng M C. et al . Ginsenoside metabolites, rather than naturally occurring ginsenosides, lead to inhibition of human cytochrome P450 enzymes.  Toxicol Sci. 2006;  91 356-64.
  CrossrefPubMedSearch in Google Scholar
• 29 Liu Y, Ma H, Zhang J W, Deng M C, Yang L. Influence of ginsenoside Rh1 and F1 on human cytochrome P450 enzymes.  Planta Med. 2006;  72 126-31.
  Thieme ConnectPubMedSearch in Google Scholar
• 30 Yuan C S, Wei G, Dey L, Karrison T, Nahlik L, Maleckar S. et al . Brief communication: American ginseng reduces warfarin’s effect in healthy patients: a randomized, controlled trial.  Ann Intern Med. 2004;  141 23-7.
  CrossrefPubMedSearch in Google Scholar
• 31 Rautio J, Humphreys J E, Webster L O, Balakrishnan A, Keogh J P, Kunta J R. et al . In vitro P-glycoprotein inhibition assays for assessment of clinical drug interaction potential of new drug candidates: a recommendation for probe substrates.  Drug Metab Dispos. 2006;  7 7.
  PubMedSearch in Google Scholar
• 32 Maeng H J, Yoo H J, Kim I W, Song I S, Chung S J, Shim C K. P-glycoprotein-mediated transport of berberine across Caco-2 cell monolayers.  J Pharm Sci. 2002;  91 2614-21.
  CrossrefPubMedSearch in Google Scholar
• 33 Gurley B J, Swain A, Barone G W, Williams D K, Breen P, Yates C R. et al . Effect of goldenseal (Hydrastis canadensis) and kava kava (Piper methysticum) supplementation on digoxin pharmacokinetics in humans.  Drug Metab Dispos. 2007;  35 240-5.
  PubMedSearch in Google Scholar
• 34 Yang C Y, Chao P D, Hou Y C, Tsai S Y, Wen K C, Hsiu S L. Marked decrease of cyclosporin bioavailability caused by coadministration of ginkgo and onion in rats.  Food Chem Toxicol. 2006;  44 1572-8.
  PubMedSearch in Google Scholar
• 35 Zhang S, Morris M E. Effects of the flavonoids biochanin A, morin, phloretin, and silymarin on P-glycoprotein-mediated transport.  J Pharmacol Exp Ther. 2003;  304 1258-67.
  PubMedSearch in Google Scholar
• 36 Gurley B J, Barone G W, Williams D K, Carrier J, Breen P, Yates C R. et al . Effect of milk thistle (Silybum marianum) and black cohosh (Cimicifuga racemosa) supplementation on digoxin pharmacokinetics in humans.  Drug Metab Dispos. 2006;  34 69-74.
  PubMedSearch in Google Scholar
• 37 Choi C H, Kang G, Min Y D. Reversal of P-glycoprotein-mediated multidrug resistance by protopanaxatriol ginsenosides from Korean red ginseng.  Planta Med. 2003;  69 235-40.
  Thieme ConnectPubMedSearch in Google Scholar
• 38 Benet L Z, Kroetz D L, Sheiner L B. Pharmacokinetics: the dynamics of drug absorption, distribution, and elimination. In: Hardman JG, editor Goodman & Gilman’s The Pharmacological Basis of Therapeutics. New York, NY; McGraw-Hill 1996: 3-27.
  Search in Google Scholar

Dr. James M. Mathews

Health Sciences Unit

Science and Engineering

RTI International

3040 Cornwallis Road

P.O. Box 12194

Research Triangle Park

NC 27709-2194

USA

Phone: +1-919-541-7461

Fax: +1-919-541-6499

Email: mathews@rti.org

• Supplementary Material