Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Mar 1.
Published in final edited form as: Pharm Res. 2024 Feb 1;41(3):557–566. doi: 10.1007/s11095-024-03662-w

Green tea catechins decrease solubility of raloxifene in vitro and its systemic exposure in mice

Victoria O Oyanna 1, Baron J Bechtold 1, Katherine D Lynch 1, M Ridge Call 1, Tyler N Graf 2, Nicholas H Oberlies 2, John D Clarke 1,3
PMCID: PMC10939713  NIHMSID: NIHMS1968650  PMID: 38302834

Abstract

Purpose

Green tea is a widely consumed beverage. A recent clinical study reported green tea decreased systemic exposure of raloxifene and its glucuronide metabolites by 34-43%. However, the underlying mechanism(s) remains unknown. This study investigated a change in raloxifene’s solubility as the responsible mechanism.

Methods

The effects of green tea extract, (−)-epigallocatechin gallate (EGCG), and (−)-epigallocatechin (EGC) on raloxifene’s solubility were assessed in fasted state simulated intestinal fluids (FaSSIF) and fed state simulated intestinal fluids (FeSSIF). EGCG and EGC represent green tea’s main bioactive constituents, flavan-3-gallate and flavan-3-ol catechins respectively, and the tested concentrations (mM) match the μg/mg of each compound in the extract. Our mouse study (n=5/time point) evaluated the effect of green tea extract and EGCG on the systemic exposure of raloxifene.

Results

EGCG (1 mM) and EGC (1.27 mM) decreased raloxifene’s solubility in FaSSIF by 78% and 13%, respectively. Micelle size in FaSSIF increased with increasing EGCG concentrations (>1000% at 1 mM), whereas EGC (1.27 mM) did not change micelle size. We observed 3.4-fold higher raloxifene solubility in FeSSIF compared to FaSSIF, and neither green tea extract nor EGCG significantly affected raloxifene solubility or micelle size in FeSSIF. The mice study showed that green tea extract significantly decreased raloxifene Cmax by 44%, whereas EGCG had no effect. Green tea extract and EGCG did not affect the AUC0-24h of raloxifene or the metabolite-to-parent AUC ratio.

Conclusions

This study demonstrated flavan-3-gallate catechins may decrease solubility of poorly water-soluble drugs such as raloxifene, particularly in the fasted state.

Keywords: absorption, green tea, intestine, pharmacokinetics, solubility

Introduction

Green tea, prepared from the leaves of Camellia sinensis (L.), Kuntze (Theaceae) is one of the most popular beverages worldwide [1,2]. The United States was the leading importer in 2021, accounting for $110.9 million in green tea trade value [3]. Polyphenols make up 30-40% of the dry weight of green tea [1,4,5]. Flavan-3-ol and flavan-3-gallate catechins comprise the largest group of polyphenols and are reported to be the main bioactive constituents in green tea [6]. Green tea and its catechins have been investigated and promoted for their possible protective effects against heart disease, diabetes, COVID-19, and certain cancers [7-10]. These potential medicinal properties likely contribute to the continual growth of the tea market [11].

Co-consumption of green tea or its catechins with conventional medications has been reported to change drug disposition and pharmacodynamics [12-14]. It is crucial to identify and understand the underlying mechanisms of these natural product-drug interactions to reduce or prevent unwanted outcomes, such as reduced efficacy or adverse drug effects. A recent clinical study evaluating the effect of green tea on the pharmacokinetics of raloxifene observed that both acute and chronic consumption of green tea decreased raloxifene area under the plasma concentration versus time curve (AUC0-96h) and maximum plasma concentration (Cmax) by 34-43% and 61-64%, respectively [15]. No change was observed in the half-life of raloxifene, raloxifene-4’-glucuronide, and raloxifene-6-glucuronide or metabolite-to-parent AUC ratio between the baseline and green tea groups [15]. These pharmacokinetic data suggest the interaction occurred in the intestine, but further investigations into the role of intestinal transporters and the gut microbiota did not fully clarify the green tea-raloxifene pharmacokinetic interaction [15]. Thus, the mediating mechanism(s) is (are) still unknown.

Drug solubility in gastrointestinal fluids is an important factor that affects pharmacokinetics because the drug molecule must be soluble to be absorbed [16-18]. Raloxifene is a poorly water-soluble drug with high membrane permeability and is categorized as a Biopharmaceutics Drug Disposition Classification System (BDDCS) class 2 drug [19]. It has a LogP of 5.78 [20] and low oral bioavailability of ~2% [21]. The presence of factors that increase raloxifene solubility in gastrointestinal fluids can increase raloxifene bioavailability. For example, the incorporation of raloxifene into chitosan nanoparticles or nanostructured lipid carriers increased oral bioavailability in rats by 2.6- or 3.2-fold, respectively [22,23]. In addition, optimized nano lipid carriers loaded with raloxifene increased oral bioavailability by 3.85-fold in healthy participants compared to the standard product [24]. In contrast, factors that decrease raloxifene solubility in gastrointestinal fluids may decrease raloxifene absorption. Previous studies reported flavan-3-gallate catechins in green tea decreased solubility of cholesterol, a poorly water-soluble compound, in mixed micelles [25,26]. Mixed micelles are colloidal structures formed between bile acids and phospholipids in the intestinal lumen that help solubilize compounds and facilitate their absorption. This led us to hypothesize that flavan-3-gallate catechins decrease raloxifene’s solubility and contribute to the green tea-raloxifene interaction.

This study investigated the effect of green tea on raloxifene’s solubility and sought to identify the constituents mediating the interaction. The effects of green tea extract, (−) -epigallocatechin gallate (EGCG), and (−)-epigallocatechin (EGC) on raloxifene’s solubility were assessed in fasted state simulated intestinal fluids (FaSSIF) and fed state simulated intestinal fluids (FeSSIF). EGCG and EGC were investigated because they make up a large proportion of the catechins in the green tea extract [27] and represent flavan-3-gallate and flavan-3-ol catechins, respectively. The green tea extract’s effect on solubility in FaSSIF and FeSSIF was also determined for the water-soluble drug nadolol (logP=0.85) [28], to compare the effects of green tea extract on solubility of a poorly water-soluble drug (i.e., raloxifene) versus a water-soluble drug. In addition, an in vivo pharmacokinetic natural product-drug interaction study was completed to compare the effects of EGCG to green tea extract on the oral pharmacokinetics of raloxifene in mice.

Materials and methods

Materials

Raloxifene hydrochloride (cat.# 1598201) was purchased from USP Pharmacopoeia (Rockville, MD). Raloxifene-d4-hydrochloride (cat.# R100002), raloxifene-4′-glucuronide (cat.# R100020), raloxifene-d4-4′-glucuronide (cat.# R100022), raloxifene-6-glucuronide (cat.# R100025), and raloxifene-d4-6-glucuronide (cat.# R100027) were purchased from Toronto Research Chemicals (North York, Ontario). Nadolol (cat.# N1892) and EGCG (cat.# E4143) were purchased from Sigma-Aldrich (St. Louis, MO). EGC (cat.# A12045) was purchased from AdooQ Bioscience (Irvine, CA). Metoprolol tartrate (cat.# S1856) was purchased from Selleckchem (Houston, TX). Green tea extract was prepared as previously described from a characterized commercial product (coded T21) [27]. FaSSIF/FeSSIF powder (cat.# FFF02) containing sodium taurocholate and soy lecithin were purchased from Biorelevant Ltd (London, UK). All other chemicals and reagents were analytical grade.

Solubility studies with green tea extract and catechins

FaSSIF version 1 (pH 6.5) and FeSSIF version 1 (pH 5.0) were prepared using the media preparation tool provided by Biorelevant Ltd [29]. Micelles in the biorelevant media are formed spontaneously by the process of micellization when the concentration of amphiphilic molecules (bile acids) reaches its critical micelle concentration (PMID: 31704344). Raloxifene and nadolol powders were added to the simulated intestinal fluids to obtain desired concentrations of 0.24 and 0.16 mg/mL, respectively. Both concentrations represent the amount of each drug reaching the human intestinal lumen after oral administration, assuming an intestinal fluid volume of 250 mL. Nadolol was completely soluble at this concentration, whereas raloxifene was not. Each mixture was incubated in a shaking incubator for 30 min at 100 rpm and 37°C. Next, the mixture was added to the test articles (i.e., green tea extract, EGCG, and EGC) and incubated for 30 min at 37°C. EGCG and EGC concentrations match the μg/mg concentration of each compound in green tea extract (i.e., EGCG concentrations 0.24, 0.62, and 1.00 mM and EGC concentrations 0.30, 0.78, and 1.27 mM correspond to green tea extract concentrations 1.69, 4.75, and 7.05 mg/mL) [27]. After incubation, the mixture was centrifuged at 250 x g for 10 min at room temperature to remove insoluble particles. The supernatant was analyzed by dynamic light scattering and UHPLC-MS/MS to measure micelle size and drug solubility, respectively. Raloxifene and nadolol were prepared for UHPLC-MS/MS analysis by adding 100 μL of the supernatant to 400 μL of methanol containing internal standards for raloxifene and nadolol (i.e., raloxifene-d4-hydrochloride and metoprolol tartrate, respectively). The sample was homogenized with a vortex mixer and dried in a speedvac at 30°C. The residue was reconstituted in 1 mL of 100% methanol, and the solution was centrifuged at 10,000 x g at room temperature for 10 min. Serial dilution of the supernatant in 50% methanol was performed until the concentration was within the linear range (0.024 – 25 ng/mL) for UHPLC-MS/MS analysis.

Dynamic Light Scattering

The size and distribution of micelles in FaSSIF and FeSSIF were measured before and after incubation with test articles using a Malvern Zetasizer Nano ZS90 dynamic light scattering spectrometer (Malvern Instruments, Worcestershire, UK). The samples were analyzed at 25°C using a material refractive index of 1.59, a dispersant refractive index of 1.33, and a viscosity of 0.89. Measurements were repeated three times for each sample.

Mouse study

Adult male C57BL/6 mice (28 – 36 g) were purchased from Jackson Laboratory (Bar Harbor, ME) and maintained under a standard 12-hour light/dark cycle with free access to food and water. After acclimation, mice (n=10) were randomly allocated to three groups: vehicle, green tea extract (200 mg/kg), and EGCG (12.96 mg/kg). The green tea extract was prepared from the same green tea product used in the solubility experiments and the published clinical study (coded T21) [15]. The dose of raloxifene was selected to reflect the allometrically scaled standard human dose of 60 mg used in the clinical study [15] and other doses used in other rodent studies [23,30,31]. The green tea extract dose was based on a preliminary dose response study which showed 200 mg/kg to produce similar pharmacokinetic observations as the clinical study and the dose of EGCG was calculated to match the 64.96 μg/mg EGCG content in the green tea extract [27]. All mice were fasted for four hours before and after the raloxifene gavage. Test articles were suspended in (2.5% v/v) polyethylene glycol 400 and carboxymethylcellulose (0.75% w/v) by heating at 37°C and homogenizing with a vortex mixer. A single oral dose of raloxifene (20 mg/kg) was administered alone or with test articles at time zero. Vehicle or test articles were administered again via gavage 4 and 8 h after the raloxifene dose. Blood (~40 μL) was collected by sparse sampling (n=5 per time point, two animals comprising one plasma concentration versus time curve/replicate) through the tail vein into heparinized capillary tubes at 0.25, 0.5, 1, 2, 4, 6, 8, 12, and 24 h post-dose of raloxifene. Plasma was separated by centrifugation and stored at −80°C until further analysis. The study protocol was approved and executed following the Institutional Animal Care and Use Committee guidelines at Washington State University. Animal handling, care, and maintenance occurred at the Program of Laboratory Animal Resources facility of Washington State University Health Sciences Spokane.

Bioanalysis

Plasma samples were processed by mixing 3 μL of plasma with 45 μL of 1% formic acid in acetonitrile containing internal standards (1 ng/mL for both raloxifene-d4-hydrochloride and raloxifene-d4-4′-glucuronide, 5 ng/mL for raloxifene-d4-6-glucuronide). Samples were mixed vigorously with a vortex mixer and centrifuged at 15,000 x g for 10 min at 4°C. The supernatant (40 μL) was dried at 30°C, and the residue was reconstituted by vortex-mixing in 30 μL of 50% methanol. The samples were centrifuged at 15,000 x g for 10 min, then 25 μL of supernatant was transferred to an autosampler vial for UHPLC-MS/MS analysis.

Quantitation of raloxifene and raloxifene-glucuronides by tandem mass spectrometry

Raloxifene, raloxifene-6-glucuronide, and raloxifene-4′-glucuronide quantification was adapted from published methods [20,32]. The analytes were separated from matrix components and quantified using a Shimadzu Nexera X2 UHPLC (Shimadzu Corporation, Tokyo, Japan) and a QTRAP 6500 MS/MS system (AB Sciex, Framingham, MA) in positive ionization mode. A gradient of 0.1% formic acid in water (A) and 100% acetonitrile (B) was passed through an ACQUITY UHPLC® BEH C18 column, 2.1 x 50 mm (Waters, Milford, MA) at 40°C. The gradient (0.5 mL/min) was as follows: 0-0.5 min, 20% B; 0.5-3.0 min, 20-50% B; 3.0-3.5 min, 50-95% B; 3.5-4.10 min, 95% B; 4.10-4.11 min, 95-20% B; and 4.11-5.0 min, 20% B. The ion transitions monitored were raloxifene 474.2 → 112.2 m/z, raloxifene-d4-hydrochloride 478.3 → 116.3 m/z, raloxifene-6-glucuronide 650.0 → 474.0 m/z, raloxifene-d4-6-glucuronide 654.0 → 478.3 m/z, raloxifene-4′-glucuronide 650.0 → 474.0 m/z, and raloxifene-d4-4′-glucuronide 654.0 → 478.3 m/z. The calibration standards were linear from 0.195 – 50 ng/mL, 0.058 – 30 ng/mL, and 0.234 – 60 ng/mL, for raloxifene, raloxifene-4′-glucuronide, and raloxifene-6-glucuronide, respectively with r2 = 0.990. The lower limit of quantification was 0.024 ng/mL, 0.058 ng/mL, 0.0234 ng/mL for raloxifene, raloxifene-4′-glucuronide, and raloxifene-6-glucuronide, respectively. Analyst® software (v1.7, AB Sciex) was used for data acquisition and quantification.

Quantitation of nadolol by tandem mass spectrometry

Nadolol quantification was adapted from published methods [33,34]. Nadolol was quantified using a Shimadzu Nexera X2 UHPLC and a QTRAP 6500 MS/MS system in positive ionization mode. A gradient of 0.1% formic acid in water (A) and 100% acetonitrile (B) was passed through an ACQUITY UHPLC® BEH C18 column, 2.1 x 50 mm (Waters, USA) at 40°C. The gradient (0.5 ml/min) was as follows: 0-0.8 min, 5% B; 0.8-2.2 min, 5-95% B; 2.2-2.5 min, 95% B; 2.50-2.51 min, 95-5% B; and 2.51-3.50 min, 5% B. The ion transitions monitored were nadolol 310.1 → 254.2 m/z and metoprolol 268.1 → 116.3 m/z. The lower limit of quantification was 0.244 ng/mL, and a calibration range of 0.0244 – 50 ng/mL was linear with r2 = 0.990. Analyst® software (v1.7) was used for data acquisition and quantification.

Pharmacokinetic analysis

The pharmacokinetics of raloxifene and raloxifene-6-glucuronide were determined via non-compartmental analysis using Phoenix WinNonlin software (v8.3; Certara, Princeton, NJ). AUC in mice was determined using the linear up/log down trapezoidal method. Cmax and time to Cmax (tmax) were obtained directly from the concentration-time profiles. AUC ratios (AUCR) were calculated as the ratio of the AUC of raloxifene or raloxifene-6-glucuronide in the presence or absence of test article. Metabolite-to-parent AUC ratio was calculated by dividing the metabolite AUC by the parent AUC.

Statistical analysis

Statistical analyses were completed using GraphPad Prism (v 9.31; San Diego, CA). In vitro data represent the mean ± standard deviation (SD) of three replicates. One-way analysis of variance with Dunnett’s post-hoc test was used to compare mean AUC and Cmax between the presence and absence of test article on log transformed pharmacokinetic data. A p-value <0.05 was deemed statistically significant.

Results

Effect of test articles on the drug solubility and micelle size

Prior to experiments with drugs and test articles, colloidal mixed micelle size was determined by dynamic light scattering for FaSSIF (66.23 ± 11.11 nm) and FeSSIF (122.2 ± 33.95 nm) and compared to published sizes (10-71 nm and 8-200 nm, respectively) [35-37]. Raloxifene concentrations in FaSSIF decreased with increasing green tea extract (21%, 25%, 30%), EGCG (16%, 22%, 78%), and EGC (10%, 10%, 13%) concentrations (Fig. 1 A-C). Micelle size in FaSSIF increased at 0.62 and 1.0 mM EGCG concentrations (155% and 1510%, respectively), whereas EGC did not change micelle size (Fig. 1 D-E). Raloxifene concentrations in FeSSIF did not change after incubation with green tea extract, EGCG, or EGC (Fig. 2 A-C). Although there was an increase in micelle size in FeSSIF after incubation with the catechins, no significant change was observed compared to the zero concentration (Fig. 2 D-E). Nadolol concentrations were not affected by green tea extract in either the FaSSIF or FeSSIF (Fig. 3).

Fig. 1.

Fig. 1

Open in a new tab

The effect of (A) Green tea extract, (B) EGCG, and (C) EGC on raloxifene’s solubility, and the effect of (D) EGCG and (E) EGC on micelle size after incubation in FaSSIF. EGCG and EGC concentrations represent equimolar concentrations within the green tea extract. Data represents the means and SDs of three replicates. * p < 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0.0001 compared to zero concentration of green tea extract, EGCG, or EGC by 1-way ANOVA Dunnett’s post-test.

Fig. 2.

Fig. 2

Open in a new tab

The effect of (A) Green tea extract, (B) EGCG, and (C) EGC on raloxifene’s solubility, and the effect of (D) EGCG and (E) EGC on micelle size after incubation in FeSSIF. EGCG and EGC concentrations represent equimolar concentrations within the green tea extract. Data represents the means and SDs of three replicates.

Fig. 3.

Fig. 3

Open in a new tab

Effect of green tea extract on nadolol’s solubility after incubation in (A) FaSSIF and (B) FeSSIF. Data represents the means and SD of three replicates.

Pharmacokinetics of raloxifene and raloxifene-6-glucuronide in mice

No overt effects were observed in the mice after administration of either raloxifene or the test articles during the study. Raloxifene and raloxifene-6-glucuronide were quantifiable in plasma at all time points in the absence and presence of orally administered green tea extract and EGCG (Fig. 4). Raloxifene-4′-glucuronide concentrations were low (data not shown), 50% of 12 h and all 24 h time points were not quantifiable. Compared to the vehicle group, green tea extract significantly decreased the geometric mean Cmax of raloxifene and raloxifene-6-glucuronide by 44% and 58%, respectively, whereas EGCG had no effect (Table 1). Green tea extract and EGCG did not significantly affect the geometric mean AUC0-4h or AUC0-24h of raloxifene or raloxifene-6-glucuronide, or the metabolite-to-parent AUC ratio.

Fig. 4.

Fig. 4

Open in a new tab

Geometric mean plasma concentration versus time profiles for (A) raloxifene and (B) raloxifene-6-glucuronide after oral administration of raloxifene to mice. A single oral dose of raloxifene (20 mg/kg) administered alone in vehicle (open black circles) or with green tea extract (200 mg/kg) (green squares) and EGCG (12.96 mg/kg) (red triangles) at 0h and then repeated doses of vehicle, green tea extract, and EGCG at 4 and 8h. Symbols and error bars denote geometric means and SDs, respectively, of in the absence and presence of test articles (n=5 per time point).

Table 1:

Pharmacokinetics of raloxifene and raloxifene-6-glucuronide in the absence and presence of orally administered test articles (n=5 per time point)

Vehicle Green tea extract EGCG
Raloxifene
Cmax (nM) 140 (22) 79 (20)** 168 (30)
AUC0-4h (nM•h) 341 (16) 200 (19) 322 (80)
AUCR0-4h NA 0.59 (0.46 – 0.74) 0.94 (0.44 – 2.01)
AUC0-24h (nM•h) 867 (31) 669 (14) 867 (18)
AUCR NA 0.77 (0.6 – 1) 1.01 (0.65 – 1.55)
tmax (h) 2 (0.25 - 4) 4 (1 - 6) 0.5 (0.25 - 2)
Raloxifene-6-glucuronide
Cmax (nM) 700 (44) 288 (33)** 820 (35)
AUC0-4h (nM•h) 935 (33) 624 (31) 1200 (53)
AUCR0-4h NA 0.67 (0.40 – 1.12) 1.29 (0.8 – 2.07)
AUC0-24h (nM•h) 1930 (24) 1560 (32) 2150 (52)
AUCR0-24h NA 0.81 (0.50 - 1.30) 1.11 (0.65 – 1.91)
tmax (h) 0.25 (0.25 - 0.5) 0.25 (0.25 - 0.5) 0.5 (0.25 - 4)
Metabolite-to-parent 2.74 (30) 3.12 (34) 3.74 (25)
AUC0-4h ratio
Metabolite-to-parent 2.23 (36) 2.33 (24) 2.48 (38)
AUC0-24h ratio
Open in a new tab

Values are geometric mean (CV%). AUCR values are geometric means (90% confidence interval), and tmax are median (range). CV, coefficient of variation; Cmax, maximum plasma concentration; AUC0-4h, area under the plasma concentration-time curve from 0 to 4 h; AUC0-12h, AUC from 0 to 12 h; tmax, time to Cmax; NA, not applicable. ** p < 0.01 compared to vehicle by 1-way ANOVA with Dunnett’s post-test.

Discussion

This study investigated drug solubility as a potential mechanism for the observed clinical pharmacokinetic interaction between raloxifene and green tea [15]. The most common mechanisms implicated in green tea-drug pharmacokinetic interactions are catechin-mediated inhibition or induction of drug metabolizing enzymes and transporters [38]. However, there is growing evidence for a mechanism involving catechin-mediated effects on drug solubility in the gastrointestinal tract [14,39,40]. Human intestinal fluid is a complex matrix that contains colloidal structures such as mixed micelles and laminar vesicles formed between bile acids (e.g., taurocholic acid), phospholipids, and lipids. These colloidal structures help to solubilize both endogenous and exogenous compounds in the intestinal lumen and facilitate their absorption [41]. We investigated the contribution of raloxifene solubility to the green tea-raloxifene interaction using biorelevant simulated intestinal fluid, FaSSIF and FeSSIF, because they form mixed micelles and are formulated to closely replicate the composition of human intestinal fluids in the fasted versus fed state [41-43]. Inclusion of both states is important because improved solubility of poorly water-soluble BDDCS class 2 drugs in the fed state typically increases drug absorption and bioavailability [37,43,44]. The two major mechanisms reported for the effect of catechins on compound solubility involve either direct binding of the catechin to the drug or binding of the catechin to taurocholic acid in mixed micelles.

Direct binding of flavan-3-gallate catechins to drugs has been reported for bortezomib, sunitinib, propericiazine, aripiprazole, lomerizine, and cetirizine [14,40,45,46]. The drugs and catechins were incubated together in an aqueous solution, and the researcher’s observed formation of a cloudy precipitate and decreased drug solubility. To provide more clinically relevant experiments, our study was designed to mimic the composition of human intestinal fluid. The experiments also reflected the amount of raloxifene, green tea extract, or catechin, present in the intestinal lumen of participants in the published clinical study [15]. Insoluble raloxifene particles were present in FaSSIF before addition of green tea extract or catechins due to the low solubility of raloxifene. After incubating raloxifene with green tea extract or catechins and separating the insoluble and soluble fractions, we observed green tea extract and EGCG decreased the solubility of raloxifene in FaSSIF (Fig. 1). In contrast, the flavan-3-ol catechin EGC had no effect on raloxifene solubility, which is consistent with lack of interaction between EGC and propericiazine, aripiprazole, lomerizine, or cetirizine [40,46,47]. The interaction between EGCG and some drugs was reported to involve a hydrogen bond between the nitrogen atom of the piperidine ring in the drug with the hydroxyl group of the gallate moiety [40]. Raloxifene also contains a piperidine ring that may contribute to direct interaction between raloxifene and flavan-3-gallate catechins.

The binding of catechins to taurocholate in mixed micelles is another mechanism that is reported to alter compound solubility. EGCG dose-dependently excluded cholesterol and phosphatidylcholine from mixed micelles without directly binding to cholesterol [48]. The proposed mechanism for this change in cholesterol solubility involves disruption of the micelle structure. It is well-documented that micelle composition and size distribution can affect the solubility of poorly water-soluble compounds [36,49]. EGCG caused a concentration-dependent increase in the diameter of the micelles, along with a corresponding decrease in cholesterol solubility [39]. Consistent with these data, we report that EGCG increased micelle size in FaSSIF (Fig. 1 D-E). Just as the gallate moiety is important for the interaction between catechins and drugs, it has been reported that flavan-3-gallate catechins such as EGCG and several synthetic EGCG derivatives [(−)-gallocatechin gallate (GCg) and (+)-catechin gallate (Cg)] bind to taurocholate and decrease micellar solubility of cholesterol [25]. No interaction between the non-gallate EGC and taurocholic acid was reported [25], which is consistent with our observation that EGC did not change micelle size (Fig. 1 D-E). The interaction between flavan-3-gallate catechins and taurocholate has been ascribed to hydrogen bonds between the carbonyl group and hydrophobic bonds through the aromatic ring of the gallate moiety [25].

Both direct binding of the catechin to the drug or binding of the catechin to taurocholic acid in mixed micelles are possible mechanisms for EGCG-mediated effects on drug solubility in our experiments. Insight into whether one mechanism dominates the interaction can be attained by examining the stoichiometry of each component in our experiments. Sakakibara et al. reported a 1:1 interaction between EGCG and taurocholic acid [25], whereas Ogawa et al. proposed 2:1 stoichiometry [26]. Ohata et al. reported a 1:2 interaction between EGCG and drug (cetirizine or lomerizine) [46]. In our FaSSIF experiments, there was less EGCG compared to taurocholate at the different EGCG concentrations tested (3 to 12.5 times less). In contrast, there was excess EGCG compared to raloxifene (3.2 to 13.5 times more). Based on the stoichiometry reported for EGCG binding to drugs versus binding to taurocholate and the amount of each compound in our experiments, direct binding of EGCG to raloxifene may be the main factor in the green tea-raloxifene interaction in the fasted state, although the interaction between taurocholic acid and EGCG may contribute. Our data in FeSSIF further support direct binding as a major factor in this interaction. FeSSIF increases poorly-water soluble drug solubility because it contains a higher concentration of bile acids and phospholipids compared to FaSSIF [50]. We observed 3.4-fold higher raloxifene solubility in FeSSIF compared to FaSSIF. Neither green tea extract nor EGCG had a significant effect on raloxifene solubility or micelle size in FeSSIF (Fig. 2). Thus, with the higher solubility of raloxifene in FeSSIF, the stoichiometry now shifted to an excess of raloxifene compared to EGCG (1 to 4 times more). These data suggest that excess soluble raloxifene in the fed state may explain the lack of green tea extract or EGCG effects in FeSSIF.

To further distinguish the drug solubility-based mechanism for green tea pharmacokinetic interactions from a transporter-based mechanism, we determined the effects of green tea on the solubility of nadolol. Nadolol is a water-soluble drug that has been reported to have decreased AUC and Cmax after green tea consumption in healthy participants due to inhibition of intestinal uptake transporters [13,51,52]. Green tea extract had no effect on nadolol solubility in either FaSSIF or FeSSIF (Fig. 3), indicating that the mechanism of altered drug solubility by green tea catechins may occur primarily with poorly water-soluble compounds such as raloxifene.

In vivo studies have also demonstrated catechin-mediated effects on drug solubility in green tea-drug interactions. In a myeloma mouse model, the complexation of EGCG with bortezomib reduced bortezomib’s anticancer efficacy [45]. Oral administration of EGCG and sunitinib in rats created sticky semisolid stomach contents and reduced sunitinib AUC0-∞ and Cmax by 52% and 48%, respectively [14]. Intestinal absorption of cholesterol from a test meal in rats decreased from 79% in the control group to 63% in the EGCG group, which was attributed to the effect of EGCG on cholesterol micellar solubility [25,53]. In our study, green tea extract decreased Cmax for raloxifene and raloxifene-6-glucuronide by 44 and 58%, respectively, which is comparable to the decreased Cmax for each compound in the acute phase of the clinical study (61 and 64%, respectively) [15]. Green tea extract also decreased AUC0-4h for raloxifene and raloxifene-6-glucuronide by 41% and 33%, respectively, although it did not reach statistical significance. The different magnitude of change in mice versus humans may reflect differences in intestinal fluid components (e.g., bile acids) between mice and humans. However, EGCG is reported to interact with other bile acids, such as sodium glycocholate, therefore, the interaction between flavan-3-gallate catechins may still occur with other bile acids [26]. The green tea extract pharmacokinetic data in mice closely replicates the clinical in vivo pharmacokinetic interaction between green tea and raloxifene and further supports the interaction occurred in the intestine. No change in parent-to-metabolite AUC ratio indicates green tea did not affect raloxifene metabolism, whereas decreased Cmax and the reduced effect of green tea on raloxifene AUC at the 0-24 h versus the 0-4 h time intervals suggests the rate of raloxifene absorption may have been altered. One limitation of our study was that only partial AUC were captured, precluding determination of other PK parameters, such as V and Cl, or completion of pharmacokinetic modeling. The latter could be useful for in vitro-in vivo extrapolation of the green tea effect on other poorly water-soluble drugs. Further mechanistic in vivo investigation into the perpetrating catechin constituent indicates that EGCG alone, at the concentration evaluated, is insufficient to cause the pharmacokinetic interaction, suggesting a combination of the different flavan-3-gallate catechins may be necessary to decrease raloxifene solubility and systemic exposure. Additional experiments are needed to determine whether this pharmacokinetic interaction still occurs in the fed state in vivo.

Conclusion

In summary, our work identified a change in raloxifene solubility as a mechanism underlying the green tea-raloxifene pharmacokinetic interaction. It also provided further evidence that flavan-3-gallate catechins are the constituents responsible for drug binding and decreased solubility. The use of FaSSIF and FeSSIF in our studies emphasizes the need to utilize biorelevant solutions and clinically translatable drug concentrations in vitro to determine potential drug interactions. Ultimately, patients should apply caution when co-consuming green tea with raloxifene under fasting conditions and future studies should include mechanistic investigation of green tea interactions with other poorly water-soluble compounds such as cyclosporine and itraconazole.

Funding statement

This work was supported by the National Institutes of Health National Center for Complementary and Integrative Health grants R21 AT011101 and U54 AT008909.

Abbreviations:

AUC

area under the plasma concentration versus time curve

BDDCS

biopharmaceutics drug disposition classification system

Cg

(+)-catechin gallate

Cmax

maximum plasma concentration

EGC

(−)-epigallocatechin

EGCG

(−) -epigallocatechin gallate

FaSSIF

fasted state simulated intestinal fluid

FeSSIF

fed state simulated intestinal fluid

Gcg

(−)-gallocatechin gallate

tmax

time to Cmax

UHPLC

ultra-high performance liquid chromatography

Footnotes

Conflicts of Interest

None

Data Availability Statement

All data supporting the findings of this study are included within the paper.

References

  • [1].Harbowy ME, Balentine DA, Davies AP, Cai Y. Tea Chemistry. CRC Crit Rev Plant Sci. 1997;16:415–80. 10.1080/07352689709701956. [DOI] [Google Scholar]
  • [2].Tea Association of the USA Inc. Tea Fact Sheet n.d. https://www.teausa.com/14655/tea-fact-sheet (accessed March 6, 2023).
  • [3].UN Comtrade. Leading green tea importing countries worldwide in 2021 (in million U.S. dollars; ). 2022. [Google Scholar]
  • [4].Lin Y-S, Tsai Y-J, Tsay J-S, Lin J-K. Factors affecting the levels of tea polyphenols and caffeine in tea leaves. J Agric Food Chem. 2003;51:1864–73. 10.1021/jf021066b. [DOI] [PubMed] [Google Scholar]
  • [5].Chaturvedula VSP, Prakash I. The aroma, taste, color and bioactive constituents of tea. Journal of Medicinal Plants Research. 2011;5:2110–24. https://doi.org/ 10.5897/JMPR.9001187. [DOI] [Google Scholar]
  • [6].Yang CS, Pan E. The effects of green tea polyphenols on drug metabolism. Expert Opin Drug Metab Toxicol. 2012;8:677–89. 10.1517/17425255.2012.681375. [DOI] [PubMed] [Google Scholar]
  • [7].Wang Z-M, Chen B, Zhou B, Zhao D, Wang L-S. Green tea consumption and the risk of stroke: A systematic review and meta-analysis of cohort studies. Nutrition. 2023;107:111936. 10.1016/j.nut.2022.111936. [DOI] [PubMed] [Google Scholar]
  • [8].Mhatre S, Srivastava T, Naik S, Patravale V. Antiviral activity of green tea and black tea polyphenols in prophylaxis and treatment of COVID-19: A review. Phytomedicine. 2021;85:153286. 10.1016/j.phymed.2020.153286. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [9].Wada K, Oba S, Tsuji M, Goto Y, Mizuta F, Koda S, et al. Green tea intake and colorectal cancer risk in Japan: the Takayama study. Jpn J Clin Oncol. 2019;49:515–20. 10.1093/jjco/hyz030. [DOI] [PubMed] [Google Scholar]
  • [10].Khan N, Mukhtar H. Tea and health: studies in humans. Curr Pharm Des. 2013;19:6141–7. 10.2174/1381612811319340008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [11].Organic green tea retail sales value in the United States 2021. ∣ Statista n.d. https://www.statista.com/statistics/1378291/retail-sales-value-of-organic-green-tea-in-the-united-states/ (accessed September 16, 2023).
  • [12].Zeng W, Lao S, Guo Y, Wu Y, Huang M, Tomlinson B, et al. The Influence of EGCG on the Pharmacokinetics and Pharmacodynamics of Bisoprolol and a New Method for Simultaneous Determination of EGCG and Bisoprolol in Rat Plasma. Front Nutr. 2022;9:907986. 10.3389/fnut.2022.907986. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Misaka S, Yatabe J, Müller F, Takano K, Kawabe K, Glaeser H, et al. Green tea ingestion greatly reduces plasma concentrations of nadolol in healthy subjects. Clin Pharmacol Ther. 2014;95:432–8. 10.1038/clpt.2013.241. [DOI] [PubMed] [Google Scholar]
  • [14].Ge J, Tan B-X, Chen Y, Yang L, Peng X-C, Li H-Z, et al. Interaction of green tea polyphenol epigallocatechin-3-gallate with sunitinib: potential risk of diminished sunitinib bioavailability. J Mol Med (Berl). 2011;89:595–602. 10.1007/s00109-011-0737-3. [DOI] [PubMed] [Google Scholar]
  • [15].Clarke JD, Judson SM, Tian D, Kirby TO, Tanna RS, Matula-Péntek A, et al. Co-consuming green tea with raloxifene decreases raloxifene systemic exposure in healthy adult participants. Clin Transl Sci. 2023. 10.1111/cts.13578. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [16].Dahan A, Miller JM. The solubility-permeability interplay and its implications in formulation design and development for poorly soluble drugs. AAPS J. 2012;14:244–51. 10.1208/s12248-012-9337-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Holm R, Müllertz A, Mu H. Bile salts and their importance for drug absorption. Int J Pharm. 2013;453:44–55. 10.1016/j.ijpharm.2013.04.003. [DOI] [PubMed] [Google Scholar]
  • [18].Porat D, Dahan A. Active intestinal drug absorption and the solubility-permeability interplay. Int J Pharm. 2018;537:84–93. 10.1016/j.ijpharm.2017.10.058. [DOI] [PubMed] [Google Scholar]
  • [19].Bocci G, Oprea TI, Benet LZ. State of the Art and Uses for the Biopharmaceutics Drug Disposition Classification System (BDDCS): New Additions, Revisions, and Citation References. AAPS J. 2022;24:37. 10.1208/s12248-022-00687-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [20].Du T, Sun R, Li L, Ebuzoeme C, Bui D, Zheng Z, et al. Development and validation of ultra-high-performance liquid chromatography-mass spectrometry method for the determination of raloxifene and its phase II metabolites in plasma: Application to pharmacokinetic studies in rats. J Sep Sci. 2020;43:4414–23. 10.1002/jssc.202000835. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Hochner-Celnikier D Pharmacokinetics of raloxifene and its clinical application. European Journal of Obstetrics & Gynecology and Reproductive Biology. 1999;85:23–9. 10.1016/S0301-2115(98)00278-4. [DOI] [PubMed] [Google Scholar]
  • [22].Wang Z, Li Y. Raloxifene/SBE-β-CD Inclusion Complexes Formulated into Nanoparticles with Chitosan to Overcome the Absorption Barrier for Bioavailability Enhancement. Pharmaceutics. 2018;10:76. 10.3390/pharmaceutics10030076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [23].Murthy A, Rao Ravi P, Kathuria H, Malekar S. Oral Bioavailability Enhancement of Raloxifene with Nanostructured Lipid Carriers. Nanomaterials. 2020;10:1085. 10.3390/nano10061085. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Hosny KM, Bahmdan RH, Alhakamy NA, Alfaleh MA, Ahmed OA, Elkomy MH. Physically Optimized Nano-Lipid Carriers Augment Raloxifene and Vitamin D Oral Bioavailability in Healthy Humans for Management of Osteoporosis. J Pharm Sci. 2020;109:2145–55. 10.1016/j.xphs.2020.03.009. [DOI] [PubMed] [Google Scholar]
  • [25].Sakakibara T, Sawada Y, Wang J, Nagaoka S, Yanase E. Molecular Mechanism by Which Tea Catechins Decrease the Micellar Solubility of Cholesterol. J Agric Food Chem. 2019;67:7128–35. 10.1021/acs.jafc.9b02265. [DOI] [PubMed] [Google Scholar]
  • [26].Ogawa K, Hirose S, Nagaoka S, Yanase E. Interaction between Tea Polyphenols and Bile Acid Inhibits Micellar Cholesterol Solubility. J Agric Food Chem. 2016;64:204–9. 10.1021/acs.jafc.5b05088. [DOI] [PubMed] [Google Scholar]
  • [27].Kellogg JJ, Graf TN, Paine MF, McCune JS, Kvalheim OM, Oberlies NH, et al. Comparison of Metabolomics Approaches for Evaluating the Variability of Complex Botanical Preparations: Green Tea (Camellia sinensis) as a Case Study. J Nat Prod. 2017;80:1457–66. 10.1021/acs.jnatprod.6b01156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [28].McFarland JW, Avdeef A, Berger CM, Raevsky OA. Estimating the Water Solubilities of Crystalline Compounds from Their Chemical Structures Alone. J Chem Inf Comput Sci. 2001;41:1355–9. 10.1021/ci0102822. [DOI] [PubMed] [Google Scholar]
  • [29].Biorelevant.com. Media prep tool n.d. https://biorelevant.com/#media_prep_tool_tab (accessed March 6, 2023).
  • [30].Kalam A, Talegaonkar S, Vohora D. Effects of raloxifene against letrozole-induced bone loss in chemically-induced model of menopause in mice. Mol Cell Endocrinol. 2017;440:34–43. 10.1016/j.mce.2016.11.005. [DOI] [PubMed] [Google Scholar]
  • [31].Yang ZhY, Zhang ZhF, He XB, Zhao GY, Zhang YQ. Validation of a Novel HPLC Method for the Determination of Raloxifene and Its Pharmacokinetics in Rat Plasma. Chromatographia. 2007;65:197–201. 10.1365/s10337-006-0123-4. [DOI] [Google Scholar]
  • [32].Trontelj J, Bogataj M, Marc J, Mrhar A. Development and validation of a liquid chromatography–tandem mass spectrometry assay for determination of raloxifene and its metabolites in human plasma. Journal of Chromatography B. 2007;855:220–7. 10.1016/j.jchromb.2007.05.004. [DOI] [PubMed] [Google Scholar]
  • [33].Rajagopaludu P, Saritha N, Devanna N, Srinivas M. Method Development and Validation of Anabasine and Nornicotine in Human Plasma by LC-MS/MS. J Pharm Res Int. 2021:7–17. 10.9734/jpri/2021/v33i1731301. [DOI] [Google Scholar]
  • [34].Zhang J, Shao B, Yin J, Wu Y, Duan H. Simultaneous detection of residues of β-adrenergic receptor blockers and sedatives in animal tissues by high-performance liquid chromatography/tandem mass spectrometry. Journal of Chromatography B. 2009;877:1915–22. 10.1016/j.jchromb.2009.05.025. [DOI] [PubMed] [Google Scholar]
  • [35].Teleki A, Nylander O, Bergström CAS. Intrinsic Dissolution Rate Profiling of Poorly Water-Soluble Compounds in Biorelevant Dissolution Media. Pharmaceutics. 2020;12:493. 10.3390/pharmaceutics12060493. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [36].Riethorst D, Baatsen P, Remijn C, Mitra A, Tack J, Brouwers J, et al. An In-Depth View into Human Intestinal Fluid Colloids: Intersubject Variability in Relation to Composition. Mol Pharm. 2016;13:3484–93. 10.1021/acs.molpharmaceut.6b00496. [DOI] [PubMed] [Google Scholar]
  • [37].Jamil R, Polli JE. Prediction of In Vitro Drug Dissolution into Fed-state Biorelevant Media: Contributions of Solubility Enhancement and Relatively Low Colloid Diffusivity. Eur J Pharm Sci. 2022;173:106179. 10.1016/j.ejps.2022.106179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [38].Albassam AA, Markowitz JS. An Appraisal of Drug-Drug Interactions with Green Tea (Camellia sinensis). Planta Med. 2017;83:496–508. 10.1055/s-0043-100934. [DOI] [PubMed] [Google Scholar]
  • [39].Raederstorff DG, Schlachter MF, Elste V, Weber P. Effect of EGCG on lipid absorption and plasma lipid levels in rats. J Nutr Biochem. 2003;14:326–32. 10.1016/s0955-2863(03)00054-8. [DOI] [PubMed] [Google Scholar]
  • [40].Ikeda H, Yamanaka M, Takahashi S, Ohata T, Yukawa M, Nakashima R, et al. Drug-Tea Polyphenol Interaction (III) Incompatibility between Aripiprazole Oral Solution and Green Tea. Chem Pharm Bull (Tokyo). 2022;70:230–4. 10.1248/cpb.c21-00746. [DOI] [PubMed] [Google Scholar]
  • [41].Riethorst D, Mols R, Duchateau G, Tack J, Brouwers J, Augustijns P. Characterization of Human Duodenal Fluids in Fasted and Fed State Conditions. J Pharm Sci. 2016;105:673–81. 10.1002/jps.24603. [DOI] [PubMed] [Google Scholar]
  • [42].Flanagan T, Van Peer A, Lindahl A. Use of physiologically relevant biopharmaceutics tools within the pharmaceutical industry and in regulatory sciences: Where are we now and what are the gaps? Eur J Pharm Sci. 2016;91:84–90. 10.1016/j.ejps.2016.06.006. [DOI] [PubMed] [Google Scholar]
  • [43].Dahlgren D, Venczel M, Ridoux J-P, Skjöld C, Müllertz A, Holm R, et al. Fasted and fed state human duodenal fluids: Characterization, drug solubility, and comparison to simulated fluids and with human bioavailability. Eur J Pharm Biopharm. 2021;163:240–51. 10.1016/j.ejpb.2021.04.005. [DOI] [PubMed] [Google Scholar]
  • [44].Wu C, Benet LZ. Predicting Drug Disposition via Application of BCS: Transport/Absorption/ Elimination Interplay and Development of a Biopharmaceutics Drug Disposition Classification System. Pharm Res. 2005;22:11–23. https://doi.org/ 10.1007/s11095-004-9004-4. [DOI] [PubMed] [Google Scholar]
  • [45].Golden EB, Lam PY, Kardosh A, Gaffney KJ, Cadenas E, Louie SG, et al. Green tea polyphenols block the anticancer effects of bortezomib and other boronic acid-based proteasome inhibitors. Blood. 2009;113:5927–37. 10.1182/blood-2008-07-171389. [DOI] [PubMed] [Google Scholar]
  • [46].Ohata T, Ikeda H, Inenaga M, Mizobe T, Yukawa M, Fujisawa M, et al. Drug-tea polyphenol interaction (II) complexation of piperazine derivatives with green tea polyphenol. Thermochim Acta. 2017;653:1–7. 10.1016/j.tca.2017.03.023. [DOI] [Google Scholar]
  • [47].Ikeda H, Tsuji E, Matsubara T, Yukawa M, Fujisawa M, Yukawa E, et al. Incompatibility between propericiazine oral solution and tea-based drink. Chem Pharm Bull (Tokyo). 2012;60:1207–11. 10.1248/cpb.c12-00116. [DOI] [PubMed] [Google Scholar]
  • [48].Kobayashi M, Nishizawa M, Inoue N, Hosoya T, Yoshida M, Ukawa Y, et al. Epigallocatechin gallate decreases the micellar solubility of cholesterol via specific interaction with phosphatidylcholine. J Agric Food Chem. 2014;62:2881–90. 10.1021/jf405591g. [DOI] [PubMed] [Google Scholar]
  • [49].Seedher N, Kanojia M. Micellar solubilization of some poorly soluble antidiabetic drugs: a technical note. AAPS PharmSciTech. 2008;9:431–6. 10.1208/s12249-008-9057-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [50].Fagerberg JH, Bergström CAS. Intestinal solubility and absorption of poorly water soluble compounds: predictions, challenges and solutions. Ther Deliv. 2015;6:935–59. 10.4155/tde.15.45. [DOI] [PubMed] [Google Scholar]
  • [51].Misaka S, Abe O, Ono T, Ono Y, Ogata H, Miura I, et al. Effects of single green tea ingestion on pharmacokinetics of nadolol in healthy volunteers. Br J Clin Pharmacol. 2020;86:2314–8. 10.1111/bcp.14315. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [52].Abe O, Ono T, Sato H, Müller F, Ogata H, Miura I, et al. Role of (−)-epigallocatechin gallate in the pharmacokinetic interaction between nadolol and green tea in healthy volunteers. Eur J Clin Pharmacol. 2018;74:775–83. 10.1007/s00228-018-2436-2. [DOI] [PubMed] [Google Scholar]
  • [53].Muramatsu K, Fukuyo M, Hara Y. Effect of green tea catechins on plasma cholesterol level in cholesterol-fed rats. J Nutr Sci Vitaminol (Tokyo). 1986;32:613–22. 10.3177/jnsv.32.613. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

All data supporting the findings of this study are included within the paper.

RESOURCES