Abstract
This study aimed to evaluate the effects of Schisandra lignan extract (SLE) with short- and long-term pretreatment on regulating rat hepatic and intestinal CYP3A for a comprehensive evaluation of metabolism-based herb-drug interactions. Inhibitory effects of SLE and its major components on rat CYP3A were confirmed in both hepatic and intestinal microsomal incubation systems. After a single dose of SLE pretreatment, higher Cmax and area under the concentration-time curves from zero to infinity (AUC0-∞) values were observed for intragastric midazolam (MDZ), whereas those for the intravenous MDZ were little changed. The mechanism-based inhibition of SLE toward CYP3A was further confirmed in vivo, characterized with a recovery half-life of 38 h. In contrast, SLE long-term treatment enhanced both hepatic (2.5-fold) and intestinal (4.0-fold) CYP3A protein expression and promoted the in vivo clearance of MDZ. When MDZ was coadministered with SLE after a consecutive long-term treatment, the AUC0-∞ value of MDZ was still lower than that of the control group, suggesting a much stronger inducing than inhibiting effect of SLE toward CYP3A. Furthermore, the intragastric administration of SLE exhibited a more intensive regulating effect toward intestinal than hepatic CYP3A, which could be partially explained by the relatively high exposures of lignans in the intestine. In conclusion, this study provides a comprehensive map for showing the complicated effects of SLE and its components on regulating rat CYP3A. The important findings are that SLE possesses a much stronger inducing than inhibiting effect on CYP3A, as well as a more intensive regulating effect on intestinal than hepatic CYP3A.
In recent years, the concomitant use of herbal medicines and/or natural products with synthetic drugs has gained increasing popularity in modern medical treatment regimens around the world. Because of the widespread and indiscriminate use of herbal medicines, the potential metabolic herb-drug interactions that may lead to severe adverse reactions have attracted a great attention from both pharmaceutical industries and regulating authorities. In the current literature, there are many references on the metabolic herb-drug interactions, such as typical studies on St. John's wort (Hypericum perforatum) (Rengelshausen et al., 2005; Xie and Kim, 2005), grapefruit juice (Arayne et al., 2005; Paine et al., 2005), and garlic (Allium sativum) (Hu et al., 2005; van den Bout-van den Beukel et al., 2006).
Schisandra is such an herbal medicine that has been widely used not only in China for the treatments of dyspnea, cough, mouth dryness, spontaneous diaphoresis, nocturnal diaphoresis, nocturnal emission, dysentery, insomnia, and amnesia but also widely used in Japan as a component of Kampo medicines and in the United States as a dietary supplement. More recently, its lignan extract has been widely used along with routinely prescribed agents to treat viral and drug-induced hepatitis in China (Zhu et al., 1999). Because of the widespread use of Schisandra lignan extract (SLE) worldwide, its potential on eliciting metabolic herb-drug interactions has become an important issue for clinical safety consideration. SLE and its components, especially gomisin C, have been identified to be potent CYP3A inhibitors in an in vitro study using the human hepatic microsomal incubation system (Iwata et al., 2004). SLE has also been found to be capable of increasing the oral bioavailability of tacrolimus, a substrate of both CYP3A and P-glycoprotein, in healthy volunteers (Xin et al., 2007). Interestingly, it was found that SLE treatment in hepatocytes induced both CYP3A and CYP2C expression through activating orphan nuclear receptor pregnane X receptor (PXR) (Mu et al., 2006). Thus, it is reasonable to propose that SLE may exert a biphasic effect on regulating CYP3A expression and activity, i.e., short-term inhibition and long-term induction, the same as that observed from St. John's wort (Rengelshausen et al., 2005; Xie and Kim, 2005).
Although previous research has shed a light on understanding the effect of SLE on regulating CYP3A, there are still several critical issues that remain unresolved for predicting the clinical SLE-drug interactions mediated by CYP3A. First, as intestinal CYP3A contributes to a great extent to the first-pass metabolism of many CYP3A substrates (Andersen et al., 2002; Paine et al., 2005; Hao et al., 2007; Kato, 2008), it is necessary to dissect the influence of SLE on regulating intestinal and hepatic CYP3A, and thus for better understanding its differential effect on regulating pharmacokinetic profiles of CYP3A substrates after intravenous and oral administration. Second, the CYP3A-inducing effect of SLE with long-term treatment has not been well studied in vivo, and it is unclear whether the long-term consumption of SLE will result in a rapid clearance of CYP3A substrates. Third, because SLE possesses a biphasic effect on regulating CYP3A, the net effect is unknown when SLE, after long-term consumption, is coadministered with CYP3A substrates. Finally, it is necessary to assess the recovery time span of CYP3A activity after SLE treatment, considering SLE has been proven a mechanism-based inhibitor of CYP3A.
Thus, our study was designed to resolve such critical issues described above and to provide a systematic insight into the effects of SLE on regulating CYP3A expression and activity. The potential inhibitory activity of SLE and its major components on CYP3A was tested in both rat intestinal and hepatic microsome systems using midazolam (MDZ) as the typical substrate (Ghosal et al., 1996; Kotegawa et al., 2002; Marathe and Rodrigues, 2006). The recovery time span of CYP3A after SLE pretreatment was determined in vivo. The differential effect of SLE with short- and long-term treatment on regulating intestinal and hepatic CYP3A has been determined by the pharmacokinetic monitoring of intravenous and intragastric administration of MDZ. The intestinal and hepatic expression of CYP3A was determined using Western blotting assay. The concentrations of major lignan components in livers and intestines were determined for better understanding of their differential effect on modulating intestinal and hepatic CYP3A. Results obtained from this study would be helpful for the better prediction of clinical SLE-drug metabolic interactions mediated by CYP3A, thus providing useful information for safe and effective use of SLE preparation combined with prescription drugs.
Materials and Methods
Chemicals and Reagents.
Schizandrol A, deoxyschizandrin, γ-schizandrin, gomisin C, MDZ, and diazepam were obtained from the National Institute for Control of Pharmaceutical and Biological Products (Beijing, China). Panaxatriol was from the College of Chemistry in Jilin University (Changchun, China). 1-Hydroxylation midazolam (1-OH MDZ) and 4-hydroxylation midazolam (4-OH MDZ) were from Sigma-Aldrich China (Shanghai, China). The purity of all the chemicals above was proven to exceed 99%. SLE powder (containing 10.9% schizandrol A, 2.4% gomisin C, 1.9% deoxyschizandrin, and 1.8% γ-schizandrin) was purchased from Qing Ze Medical and Pharmaceutical Technical Development Co., Ltd. (Nanjing, China). Glucose-6-phosphate (purity 98–100%), NADP+ (purity 97%), and glucose-6-phosphate dehydrogenase (200–400 units/mg protein) were purchased from Sigma-Aldrich China. High-performance liquid chromatography (HPLC)-grade acetonitrile and methanol were obtained from Merck (Darmstadt, Germany). Deionized water was purified using a Milli-Q system (Millipore Corporation, Billerica, MA). Radioimmunoprecipitation assay lysis buffer, phenylmethanesulfonyl fluoride, BCA Protein Assay Kit, and SDS-polyacrylamide gel electrophoresis sample loading buffer were purchased from Beyotime Institute of Biotechnology (Jiangsu, China). Rabbit anti-rat CYP3A2 polyclonal antibody was from Millipore Bioscience Research Reagents (Temecula, CA) (binds specifically to cytochrome P450 CYP3A2 in rat hepatic microsomal fraction). Polyvinylidene difluoride membranes were obtained from Millipore Corporation (Shanghai, China). Horseradish peroxidase-conjugated goat anti-rabbit IgG was purchased from Boster Biological Technology, Ltd. (Wuhan, China). Enhanced chemiluminescence kit was purchased from Pierce Chemical (Rockford, IL). Ethyl acetate and other regents were all analytical grade.
Animals.
Male Sprague-Dawley rats (200–220 g) were obtained from the Academy of Military Medical Sciences (Beijing, China). All the rats were acclimated for at least 1 week before all the experiments and allowed water and standard chow ad libitum. All the animal studies were approved by the Animal Ethics Committee of China Pharmaceutical University.
CYP3A Inhibitory Assay in Vitro.
Microsomes were prepared from male rat livers and intestines through differential centrifugation (Hao et al., 2007). The protein concentrations were determined with a commercially available kit (BCA Protein Assay; Beyotime Institute of Biotechnology). Rat hepatic and intestinal microsomes were stored at −80°C until use.
To investigate IC50 values of SLE and its major components toward hepatic and intestinal microsomal MDZ 4-hydroxylation activity, a typical incubation mixture contained rat microsomal protein (1 mg/ml), NADPH-regenerating system (10 mM glucose-6-phosphate, 10 mM MgCl2, 1 U/ml glucose-6-phosphate dehydrogenase, and 0.5 mM NADP+), the typical CYP3A substrate MDZ (10 μM), SLE or lignan components (schizandrol A, deoxyschizandrin, γ-schizandrin, and gomisin C) at different concentrations, and 100 mM phosphate buffer, pH 7.4, in a final volume of 200 μl. All the incubations were conducted at 37°C for 5 min. To determine time-dependent inactivation, rat hepatic and intestinal microsomes were preincubated with SLE for 10 min in a metabolic system as described above except without substrate MDZ. After that, 10 μl of MDZ solution was added to the mixture to reach a final volume of 200 μl. Reactions were terminated 5 min later by adding cold acetonitrile, and the remaining MDZ 4-hydroxylation activity was determined.
To determine Ki values, schizandrol A (0, 25, 50, and 100 μM), deoxyschizandrin (0, 6.25, 12.5, and 25 μM), γ-schizandrin (0, 3.125, 6.25, and 12.5 μM), and gomisin C (0, 0.125, 0.25, and 0.5 μM) were added to the reaction mixture containing different concentrations of MDZ (5, 10, 20, and 40 μM) in the rat hepatic microsomal incubation systems. The total volume was 200 μl. Reactions were stopped by adding cold acetonitrile. Reaction conditions and sample preparation procedures were the same as that described above for the IC50 assay. All the incubations were performed in triplicate.
Pharmacokinetic Studies of MDZ.
Rats were fasted for 12 h with free access to water before the experiment. MDZ and SLE were suspended in 0.5% carboxymethylcellulose (CMC-Na) for intragastric administrations. To determine the short-term effect on CYP3A, SLE (0, 50, and 150 mg/kg) was intragastrically administered 30 min before the intravenous administration (4 mg/kg) or 5 min before the intragastric administration of MDZ (20 mg/kg). The MDZ-dosing time intervals to SLE pretreatment were designed from the pharmacokinetic and tissue exposure study of lignan components to ensure MDZ at maximal exposures of lignan components. To ascertain whether SLE has a mechanism-based inhibition toward CYP3A in vivo, MDZ was administered intragastrically to rats (4 mg/kg) 6, 24, or 72 h after SLE administration (150 mg/kg).
To determine the long-term effect of SLE on CYP3A, rats were randomly divided into three groups. Rats assigned to the pretreatment group and the coadministered group were gavaged with SLE (150 mg/kg/day) for 14 consecutive days. Rats in the control group were administered with an equivalent volume of 0.5% CMC-Na vehicle. On day 15, rats in the control group and the pretreatment group received the vehicle, whereas those in the coadministered group were given SLE (150 mg/kg). Five minutes later, all of the rats were intragastrically administered MDZ (20 mg/kg). Blood samples were collected from the postocular vein at 0, 5, 15, 30, 45, 60, 90, 120, 180, 240, 300, and 360 min after intragastric administration and at 0, 2, 5, 10, 20, 30, 60, 120, 180, 240, 300, and 360 min after intravenous administration of MDZ. All of the blood samples were immediately centrifuged to obtain plasma, which was immediately stored at −20°C until analysis.
Pharmacokinetics and Hepatic and Intestinal Distribution Assays of SLE Components.
For assessing pharmacokinetic behaviors of the major lignan components, rats were given a single intragastric administration of SLE (150 mg/kg). Blood samples were collected from the postocular vein at 0, 0.03, 0.08, 0.17, 0.33, 0.5, 1, 2, 4, 6, 8, 12, and 24 h. For hepatic and intestinal exposure determinations, another set of rats was given a single intragastric administration of SLE (150 mg/kg). Tissues (livers and intestines including duodenum, jejunum, and ileum) were promptly removed at 0, 0.5, 1, and 5 h after dosing and washed with saline. Each tissue sample was diluted with 4 volumes (v/w) of saline and then homogenized.
CYP3A Protein Expression Assay.
Rats in the pharmacokinetic studies of MDZ after long-term treatment with SLE were euthanized after blood collection. Intestines and livers were immediately excised, washed with ice-cold saline, snap-frozen in liquid nitrogen, and stored at −80°C until analysis. Tissues were homogenized (40 mg of the mucosa of small intestine, 40 mg of liver) in ice-cold radioimmunoprecipitation assay lysis buffer with 0.02 mM phenylmethanesulfonyl fluoride through ultrasonication. Samples were solubilized in lysis buffer for 30 min on an ice-cold plate and then centrifuged at 9000g for 10 min at 4°C. The supernatant was collected and stored at −80°C until use. Protein concentrations were determined with a BCA Protein Assay Kit (Beyotime Institute of Biotechnology) according to the manufacturer's instructions. SDS-polyacrylamide gel electrophoresis sample loading buffer was added into the protein samples. Samples were boiled for 4 min for protein denaturation and then centrifuged at 9000g for 3 min at 4°C. Protein samples equivalent to 60 μg of protein were size-fractionated by electrophoresis on an 8% SDS-polyacrylamide gel at 150 V for 1 h and then transferred to polyvinylidene fluoride membranes for approximately 2.5 h. After blocked in a buffer containing 5% nonfat milk for 1 h, the membranes were then incubated with CYP3A2 polyclonal antibody (1:1000 for intestine samples, 1:1500 for liver samples) for 24 h at 4°C. For reference, β-actin was detected using a polyclonal antibody (1:200). After washing three times with Tris-buffered saline/Tween 20 buffer for 10 min each time, membranes were incubated with secondary antibody horseradish peroxidase-conjugated goat anti-rabbit IgG (1:1500) for 1 h at 37°C and then washed three times with Tris-buffered saline/Tween 20. The signals were detected by enhanced chemiluminescence kit. The CYP3A2 protein band intensity was normalized with that of β-actin.
Liquid Chromatography/Mass Spectrometry Assays.
MDZ and its metabolites were analyzed according to the previous reports (Cummins et al., 2003; Arellano et al., 2007) with slight modifications. Diazepam was used as the internal standard. All the samples were analyzed by HPLC/electrospray ionization/mass spectrometry (Shimadzu, Kyoto, Japan). In brief, a Shim-Pack 250 × 2.0-mm column packed with 5-μm VP-ODS C18 from Shimadzu was used. The column temperature was set at 40°C. The mobile phase delivered at a flow rate of 0.2 ml/min consisted of 2 mM ammonium acetate (A) and acetonitrile (B) with a gradient program as follows: 40% B to 75% B from 0 to 5 min, 75% B to 40% B from 5 to 7 min, and holding 40% B for another 5 min. Mass spectrometry was conducted in the positive-ion mode for the detections of MDZ at m/z 326 ([M+H]+), 1-OH MDZ at m/z 342 ([M+H]+), 4-OH MDZ at m/z 342 ([M+H]+), and diazepam at m/z 285 ([M+H]+). For microsome samples, 400 μl of acetonitrile was added after incubation and then centrifuged at 20,000 rpm for 10 min twice. Plasma samples were extracted with 0.8 ml of ethyl acetate after adding a saturated solution of sodium carbonate. The precipitate was removed by centrifugation at 8000 rpm for 5 min. The supernatant was transferred to an Eppendorf tube and evaporated to dryness by the Thermo Savant SPD 2010 SpeedVac System (Thermo Fisher Scientific, Waltham, MA). The residue was dissolved in 0.1 ml of mobile phase.
Lignan components were analyzed by HPLC/atmospheric pressure chemical ionization/mass spectrometry (Shimadzu) using panaxatriol as the internal standard. The analytical column was a Gemini 150 × 4.6-mm column packed with 5-μm C18 110A (Phenomenex, Torrance, CA). The column temperature was maintained at 40°C. The mobile phase delivered at a flow rate of 0.75 ml/min consisted of water (A) and methanol (B) with a gradient program as follows: 70% B to 85% B from 0 to 5 min, holding 85% B for another 2 min, 85% B to 70% B from 7 to 9 min, and holding for another 3 min. Mass spectrometry was conducted in the positive-ion mode for the detections of schizandrol A at m/z 415 ([M−H2O+H]+), deoxyschizandrin at m/z 417 ([M+H]+), γ-schizandrin at m/z 401 ([M+H]+), gomisin C at m/z 415 ([M−C6H5CO2+H]+), and panaxatriol at m/z 423 ([M−3H2O+H]+). Samples (100 μl) of plasma, liver, and intestine homogenates were extracted with 0.8 ml of ethyl acetate. The supernatant was transferred to an Eppendorf tube and evaporated to dryness by the Thermo Savant SPD 2010 SpeedVac System (Thermo Fisher Scientific). The residue was dissolved in 0.1 ml of mobile phase. Ten microliters of supernatant was injected to liquid chromatography/atmospheric pressure chemical ionization/mass spectrometry for analysis.
Data and Statistical Analysis.
IC50 values were determined by nonlinear regression analysis. Dixon plots were used to determine Ki values of four lignan components. Data are expressed as mean ± S.D. Pharmacokinetic parameters of MDZ, 1-OH MDZ, and 4-OH MDZ were estimated using the computer software DAS package (version 2.0 pharmacokinetic software; Chinese Pharmacological Association, Beijing, China). Comparisons between two groups were performed using Student's t test. For multiple comparisons, one-way analysis of variance followed by post hoc test was performed. The difference was considered to be statistically significant if the probability value was less than 0.05 (P < 0.05).
Results
CYP3A Inhibitory Effect of SLE and Its Major Components in Vitro.
Microsomal MDZ 4-hydroxylation activity test was used to determine the potential inhibitory effect of SLE and its major lignan components on CYP3A. SLE showed a concentration-dependent inhibitory effect on CYP3A activity with an IC50 value of 40 ± 5 μg/ml in rat hepatic microsomes and 25 ± 3 μg/ml in intestinal microsomes, respectively. After preincubation with microsomes in the presence of the NADPH-regenerating system, the IC50 values markedly reduced to 35 ± 5 and 6.25 ± 1.25 μg/ml, respectively, suggesting a mechanism-based inhibitory effect of SLE on CYP3A. For the components tested, deoxyschizandrin and γ-schizandrin showed a moderate inhibitory activity with an IC50 value of 12.5 and 6.25 μM and a Ki value of 4.8 and 5.0 μM, respectively. Schizandrol A showed poor inhibition activity with an IC50 value of 70 μM and a Ki value of 45.0 μM. Gomisin C showed a potent inhibitory effect on CYP3A with an IC50 value of 0.30 μM and a Ki value of 0.06 μM (Table 1).
Pharmacokinetics of MDZ after Short-Term Treatment with SLE.
Pharmacokinetic interaction between SLE and MDZ in vivo was studied in rats. The mean plasma concentration-time profiles of MDZ, 4-OH MDZ, and 1-OH MDZ after intravenous (4 mg/kg) and intragastric (20 mg/kg) administrations of MDZ with or without SLE (0, 50, 150 mg/kg) pretreatment are shown in Fig. 1. The pharmacokinetic parameters obtained are collected in Table 2. No significant differences of the pharmacokinetic parameters for the intravenous administration of MDZ were observed among different groups. In contrast, the area under the concentration-time curves from zero to infinity (AUC0-∞) value of MDZ with intragastric administration significantly increased up to 2.3- and 4.0-fold in rats pretreated with 50 and 150 mg/kg SLE, respectively. Likewise, the Cmax value of MDZ increased to 1.6- and 2.3-fold, whereas the apparent oral total body clearance (where F represents bioavailability) value decreased to 44 and 24% of the control group, respectively. Consequently, the Cmax and AUC0-∞ of MDZ metabolites 4-OH MDZ and 1-OH MDZ were significantly decreased in rats pretreated with SLE, providing direct evidence of the in vivo inhibitory effect of SLE on CYP3A.
To determine the potential mechanism-based inhibition of SLE on CYP3A, the pharmacokinetic changes of MDZ were determined 6, 24, and 72 h after SLE pretreatment. The obtained pharmacokinetic profiles of MDZ and its metabolites are shown in Fig. 2, and the calculated pharmacokinetic parameters are summarized in Table 3. Accompanying the increase of MDZ dosing time intervals to SLE pretreatment, a clear restoring trend toward the control group was observed for the pharmacokinetic profiles and parameters of both MDZ and its metabolites in rats pretreated with SLE. The CYP3A protein recovery half-life was estimated at approximately 38 h by plotting the net increased ratio of the AUC0-∞ of MDZ against MDZ dosing time intervals to SLE pretreatment (Fig. 3).
Pharmacokinetic Profiles and Hepatic and Intestinal Distributions of Lignan Components.
SLE and its major lignan components exhibited potent inhibitory effects on both hepatic and intestinal CYP3A in the microsomal incubation systems. However, pharmacokinetic studies in vivo showed SLE pretreatment significantly inhibited intestinal but not hepatic CYP3A. We hypothesized that such a differential effect might be mainly resourced from the different hepatic and intestinal exposures of lignan components. To examine such a hypothesis, the pharmacokinetic profiles and hepatic and intestinal distributions of four lignan compounds (schizandrol A, deoxyschizandrin, γ-schizandrin, and gomisin C) were determined. As observed in Figs. 4 and 5, although gomisin C, the most potent inhibitor in SLE, exhibited a relatively high plasma exposure, it was not detected in the liver tissues. It is interesting to note that schizandrol A, the poorest inhibitor in SLE, possessed the highest hepatic exposure levels among all the components tested. The concentrations of all the lignan components tested, except the poorest inhibitor schizandrol A, in the intestine were much higher than those in the liver. Such results were in good agreement with the differential effect of SLE with intragastric administration on regulating hepatic and intestinal CYP3A.
Pharmacokinetics of MDZ after Long-Term Treatment with SLE.
The plasma profiles of MDZ, 4-OH MDZ, and 1-OH MDZ obtained after intragastric administration of 20 mg/kg MDZ of the control, the coadministered, and the pretreatment groups are shown in Fig. 6, and the pharmacokinetic parameters are summarized in Table 4. Compared with the control group, rats in the coadministered and the pretreatment group showed lower MDZ AUC0-∞ values (63 and 34%), as well as higher apparent oral total body clearance (CL/F, where F represents bioavailability) values (1.6- and 3.1-fold). Similar changes to those of AUC0-∞ values were observed for Cmax values of MDZ. Pharmacokinetic parameters of 1-OH MDZ in the coadministered group showed no significant differences from those in the control group, whereas significantly higher Cmax and AUC0-∞ values were observed in the pretreatment group. For 4-OH MDZ, rats in the coadministered and the pretreatment group showed higher AUC0-∞ values (1.8- and 2.5-fold), as well as higher Cmax values (1.4- and 1.8-fold), compared with the control group.
Inducing Effects of SLE on Hepatic and Intestinal CYP3A Protein Expression.
After pharmacokinetic research of MDZ and its metabolites in rats with long-term intake of SLE, the hepatic and intestinal CYP3A protein expressions were determined. After 14 days of treatment, there were no significant differences in mean body weight between the control and SLE treatment groups. Compared with the control group, intestinal CYP3A protein level increased approximately by 300% (P < 0.05), whereas the hepatic CYP3A protein level increased approximately by 150% (P < 0.05) after 14 days of pretreatment with 150 mg/kg/day SLE (Fig. 7).
Discussion
The concomitant administration of herbal supplements and synthetic drugs has now become more and more popular. As a result, herb-drug interactions have become a common clinical problem. CYP3A, the most abundant drug-metabolizing enzyme in human liver and the intestinal wall, is responsible for metabolizing a majority of drugs. Case reports, clinical trials, and in vitro studies have shown a number of important pharmacokinetic interaction risks of herbal medicines with prescribed agents medicated by CYP3A (Hu et al., 2005; van den Bout-van den Beukel et al., 2006). Several methods have been published describing the interactions between herbal medicines and synthetic drugs based on in vitro data. Despite some successes, it sometimes failed to predict drug-drug interactions for a variety of reasons (Kanamitsu et al., 2000; Kanazu et al., 2005). Furthermore, some herbal medicines, such as St. John's wort (Rengelshausen et al., 2005; Xie and Kim, 2005), showed differential effects toward cytochrome P450 isoforms between short-term and long-term treatment. Thus, it is difficult to predict the herb-drug interactions in various clinical practices, especially for the herbal compounds characterized with biphasic effects on regulating drug-metabolizing enzymes. For example, we have designed systematic experiments to study the effects of SLE on regulating CYP3A and to evaluate more accurately the potentials for herb-drug interactions in different clinical practices.
In our in vitro study using rat hepatic and intestinal microsomes, SLE exhibited a concentration- and time-dependent inhibitory effect on CYP3A. Among the lignan components tested, gomisin C was the most potent CYP3A inhibitor, whereas deoxyschizandrin and γ-schizandrin showed a moderate and almost equivalent inhibitory potency. The results are generally consistent with the previous report (Iwata et al., 2004), in which gomisin C was identified as a very potent mechanism-based CYP3A inactivator. Considering that the mechanism-based inhibitory effects of lignans in vitro had been previously reported (Iwata et al., 2004) and the present study was mainly aimed to focus on the in vivo biphasic and differential effect of SLE on regulating hepatic and intestinal CYP3A, the in vitro inhibitory effect determinations in our study were designed only for a confirmatory purpose; thus, the detailed mechanism-based inhibition assay of SLE and its components had not been performed.
SLE with a single intragastric administration to rats increased both AUC0-∞ and Cmax values of MDZ and consequently decreased those of its metabolites, 1-OH MDZ and 4-OH MDZ, when MDZ was intragastrically administered 5 min after SLE pretreatment. However, pretreatment with SLE did not alter the intravenous pharmacokinetic behaviors of MDZ and its metabolites' formation. These lines of evidence suggest that SLE with oral ingestion may exhibit a potent inhibitory effect on intestinal rather than hepatic CYP3A. Such results in vivo seem controversial with the in vitro findings. To explain such a discrepancy between in vitro and in vivo, we determined the plasma and tissue exposures of four major lignan components that showed certain inhibitory effects on CYP3A. It was found that the concentrations of most components except schizandrol A, a poor inhibitor of CYP3A, in intestines were much higher than those in livers. Gomisin C, the most potent CYP3A inhibitor among lignan components (Iwata et al., 2004), was not detected in liver samples despite that its total plasma concentrations could reach its reported KI value (0.399 μM). The potential high plasma protein binding of lignan components would further restrict their distribution into livers and thus compromise the inhibitory effects on hepatic enzymes. As evidenced from both in vitro and in vivo data in this study, we conclude that the oral ingestion of SLE is unlikely to induce a significant inhibitory effect on hepatic CYP3A in vivo as a result of poor exposures in liver, which may be caused by poor intestinal absorption and potential high plasma protein binding.
The previous report (Iwata et al., 2004) and the present study in vitro showed that SLE exhibited a mechanism-based inhibition toward CYP3A. It had been well acknowledged that the mechanism-based inhibitor caused irreversible inhibition of an enzyme until this enzyme was newly synthesized. To our knowledge, there have been no reports contributing to the mechanism-based inhibition of SLE study in vivo. Therefore, we have designed an experiment for determining the recovery time course of CYP3A activity in rats after a single intragastric dose treatment with SLE. As a result, inhibition of the enteric CYP3A activity appeared to continue for approximately 3 days, which was consistent with the previous reports (Venkatakrishnan et al., 2001; Greenblatt et al., 2003) that the activity of human enteric CYP3A recovered within 3 days.
It was interesting to note that both aqueous and ethanolic extracts of Schisandra chinensis Baill up-regulated CYP3A expression through PXR activation (Mu et al., 2006). However, the in vivo CYP3A-inducing potency of SLE had not been previously confirmed. Results from the present study provide direct evidence that SLE is a strong modulator of CYP3A expression in rats, as evidenced from both protein expression and activity test in vivo. Unlike inhibition, long-term oral pretreatment with SLE significantly up-regulated the hepatic CYP3A protein level, albeit to a lesser extent compared with that on intestinal CYP3A. Such a discrepancy is not surprising in view that schizandrol A was reported to be a potent PXR activator (Mu et al., 2006) and showed relatively high hepatic exposures as determined from the present study. In combinatorial considerations, we propose that the various lignan components with differential inducing and/or inhibiting potencies may underlie the biphasic regulating effect of SLE on CYP3A. The major limitation of this study is that we have not measured the protein expression of CYP3A23/CYP3A1, which has been reported much more inducible than CYP3A2 in most cases (Huss et al., 1999). However, in view that CYP3A2 is the predominant constitutively expressed CYP3A isoform in adult male rats (Jan et al., 2006) and that it is more active than other CYP3A isoforms (Mahnke et al., 1997), we consider that the up-regulation of CYP3A2 protein levels explains to a great extent the enhanced total CYP3A activity as measured by MDZ in the adult male rats with SLE long-term treatment in this study.
Because SLE possessed a biphasic effect on regulating CYP3A, we were interested in investigating the net effect when SLE was coadministered with MDZ after long-term consumption of SLE. It has been found that the rats in both the coadministered group and the pretreatment group showed lower Cmax and AUC0-∞ values of MDZ and higher Cmax and AUC0-∞ values of 4-OH MDZ compared with the control group, suggesting the net effect was of CYP3A induction. In other words, the induction capacity of SLE is much stronger than that of inhibition on CYP3A. Our studies suggest that SLE appears to exert a biphasic effect on regulating CYP3A. The initial oral dose of SLE would lead to an apparent inhibition on intestinal CYP3A both reversibly and irreversibly, whereas its long-term consumption is more likely to produce a significant CYP3A induction.
In summary, our results showed that SLE with intragastric administration inhibited the activity of intestinal but not of hepatic CYP3A. Its mechanism-based inhibition of CYP3A activity appeared to continue for 38 h in rats. However, the long-term intake of SLE resulted in the induction of both intestinal and hepatic CYP3A protein. In addition, SLE with long-term treatment exerted a much stronger inducing than inhibiting effect on CYP3A. The present findings suggest that it is very critical to assess the potential interactions between SLE remedies and prescribed drugs as CYP3A substrates, thus avoiding clinically unwanted SLE-drug interactions.
Footnotes
This work was supported in part by the National Natural Science Foundation of People's Republic of China [Grants 30630076, 30801422]; and the “eleventh-five years” supporting programs from the Ministry of Science and Technology of People's Republic of China [Grant 2006BAI08B04-05].
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
doi:10.1124/dmd.109.027433
-
- SLE
- Schisandra lignan extract
- PXR
- pregnane X receptor
- MDZ
- midazolam
- 1-OH MDZ
- 1-hydroxylation midazolam
- 4-OH MDZ
- 4-hydroxylation midazolam
- HPLC
- high-performance liquid chromatography
- CMC-Na
- carboxymethylcellulose
- AUC0-∞
- area under the concentration-time curves from zero to infinity.
- Received May 15, 2009.
- Accepted September 3, 2009.
- Copyright © 2009 by The American Society for Pharmacology and Experimental Therapeutics