![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Faculty of Pharmaceutical Sciences, The University of British Columbia, Vancouver, British Columbia, Canada
(Received May 31, 2005; accepted October 27, 2005)
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
). Many consumers believe that herbal products, unlike conventional drugs, are safe and have no side effects (Barnes, 2003). A survey indicated that among the respondents who consumed herbal medicines, 31% ingested them concomitantly with conventional drugs (Abebe, 2002). Therefore, a concern is the potential risk of adverse effects due to herb-drug interactions (Huang et al., 2004).
According to a 2002 survey conducted in the United States, Ginkgo biloba was the third most popular herbal medicine (Barnes et al., 2004). It was used by 21% of adults who consumed herbal medicines. G. biloba extract is used by consumers for the management of a variety of disorders; for example, memory impairment (De Smet, 2002). Many individual chemical constituents are present in G. biloba extract, including terpene trilactones, such as diterpenes (e.g., ginkgolide A, ginkgolide B, ginkgolide C, and ginkgolide J) and sesquiterpenes (e.g., bilobalide), and ginkgo flavonol glycosides (e.g., glycosides of kaempferol, quercetin, and isorhamnetin) (van Beek, 2002).
Previous rodent studies indicated that G. biloba extract, when administered as a single i.p. dose (25 or 50 mg/kg) (Brochet et al., 1999) or fed for 2 weeks via the diet (0.5% or 1% w/w) (Kubota et al., 2004), shortened barbiturate-induced narcosis. G. biloba extract (0.5% w/w in the diet for 2 weeks) has also been shown to reduce the hypotensive action of nicardipine and to alter the pharmacokinetics of this drug, specifically, via a decrease in the maximum plasma concentration and area under the plasma concentration-time curve (Kubota et al., 2003). The effect of G. biloba on the pharmacodynamics and pharmacokinetics of these drugs may be related to induction of cytochrome P450 (P450) enzymes by this herbal medicine. It has been reported that the administration of a G. biloba extract (containing 24.9% flavonol glycosides and 10.6% terpene trilactones) to rats increased hepatic expression of several P450 enzymes, including CYP2B (Shinozuka et al., 2002; Umegaki et al., 2002). However, it is not known whether the flavonols or the terpene trilactones are responsible for the reported effects of G. biloba extract. The levels of the terpene trilactones in the extracts used in the previous studies (Shinozuka et al., 2002; Umegaki et al., 2002) were greater than those in many of the commercially available G. biloba products formulated to contain only 6% terpene trilactones (van Beek, 2002). Therefore, it is not known whether the results from those studies (Shinozuka et al., 2002; Umegaki et al., 2002) can be generalized to other extracts of G. biloba.
The present study was conducted to 1) investigate the effect of multiple extracts of G. biloba containing known amounts of bilobalide, ginkgolides, and flavonol glycosides on CYP2B1, CYP3A23, CYP3A2, and CYP3A18 gene expression in primary cultures of rat hepatocytes, as determined by reverse transcription and real-time polymerase chain reaction (PCR); 2) assess the effect of G. biloba extract on CYP2B- and CYP3A-mediated enzyme activities in cultured hepatocytes, as analyzed by the 7-benzyloxyresorufin O-dealkylation (BROD) assay and the testosterone 6ß-hydroxylation assay, respectively; and 3) determine whether terpene trilactones, such as bilobalide and ginkgolides A, B, C, and J, or flavonol aglycones and glycosides, such as those of kaempferol, quercetin, and isorhamnetin, are responsible for the effect of G. biloba extract on CYP2B1, CYP3A23, CYP3A2, and CYP3A18 expression. Our results indicate that bilobalide and ginkgolide A are novel inducers of CYP2B1 and CYP3A, respectively, and that these chemicals play a distinct role in the modulation of CYP2B1 and CYP3A23 mRNA expression and enzyme activities by G. biloba extract.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
|
Animals and Isolation of Hepatocytes. Adult male Sprague-Dawley rats (225–250 g) were purchased from Charles River (Montreal, QC, Canada) and cared for in our animal care facility as described previously (Kuo et al., 2004). Rats were anesthetized with sodium pentobarbital (60 mg/kg i.p.) and hepatocytes were isolated by a two-step collagenase perfusion method (Seglen, 1993). The hepatic portal vein was cannulated with a 21-gauge (1-inch) Teflon catheter (0.9 x 25 mm; BD Biosciences) and the liver was perfused with calcium-free Liver Perfusion Media at 37°C at a rate of 25 ml/min using a peristaltic pump. The inferior vena cava was severed immediately to allow for the efflux of the perfusate. The superior vena cava was cannulated with a PE-190 tubing (BD Biosciences) and the inferior vena cava was then tied to allow for the efflux of the perfusate through the superior vena cava. Subsequently, the liver was perfused at 37°C for 7 min at a rate of 25 ml/min with calcium-free Liver Perfusion Media followed by perfusion for another 7 min with digest medium (final pH adjusted to 7.5) consisting of Hanks' balanced salt solution supplemented with 2.38 g/l HEPES, 0.35 g/l NaHCO3, 0.05 g/l trypsin inhibitor (type II-S), and 0.5 g/l collagenase (type IV). Perfused liver was excised and placed in a sterile Petri dish containing Hepatocyte Wash Media. Cells were mechanically dispersed using a blunt glass rod. The cell suspension was filtered through sterile Nytex nylon mesh (64-µm pore size; TETCO, Briarcliff Manor, NY) into a 50-ml Falcon tube. The suspension was centrifuged (50g) at 4°C for 5 min and the pellet was suspended in fresh ice-cold Hepatocyte Wash Media. A portion (20 ml) of the suspension was added to ice-cold Percoll solution (26.1 ml of Percoll and 3.9 ml of 10x Dulbecco's phosphate-buffered saline) and mixed gently by inversion (five times). The mixture was centrifuged (50g) at 4°C for 10 min. The supernatant was decanted, and the pellet was washed again in Hepatocyte Wash Media and subsequently suspended in supplemented Williams' medium E containing heat-inactivated fetal bovine serum (10% v/v). Supplemented Williams' medium E contained insulin (1 µM), dexamethasone (100 nM), L-glutamine (2 mM), penicillin (100 units/ml), and streptomycin (100 µg/ml). Cell viability was 80 to 90%, as assessed by trypan blue exclusion.
Hepatocyte Cultures. Culture dishes or microplates were coated with Matrigel at approximately 1 h before use (Schuetz et al., 1988). In the experiments investigating P450 gene expression, 170 µl of Matrigel was applied onto a 60-mm Permanox culture dish. Hepatocyte suspension was diluted to 106 cells/ml in serum-free, supplemented Williams' medium E, and a 3-ml volume was plated onto each culture dish. In experiments investigating P450-mediated enzyme activities, 30 µl of Matrigel (diluted 1:1 with culture medium) was applied onto each well in a 24-well microplate. Each well was loaded with 3 x 105 cells in a volume of 500 µl containing serum-free, supplemented Williams' medium E. Cells were allowed to attach for 4 h in a humidified 37°C incubator with 5% CO2 and 95% air. Subsequently, the medium was decanted to remove unattached cells and the hepatocytes were cultured in serum-free, supplemented Williams' medium E. Culture medium was changed daily.
Treatment of Primary Cultures of Rat Hepatocytes. At 48 h after plating, cultured hepatocytes were treated for 48 h with G. biloba extract (10, 100, or 1000 µg/ml) or culture medium (vehicle control). In other experiments, cultured hepatocytes were treated with bilobalide, ginkgolide A, ginkgolide B, ginkgolide C, ginkgolide J, kaempferol, kaempferol-3-O-rutinoside, quercetin, quercetin-3-O-rutinoside, isorhamnetin, isorhamnetin-3-O-rutinoside, or DMSO (vehicle control) at the concentration indicated in each figure legend. For comparison, cultured hepatocytes were treated with PB (100 µM, positive control for CYP2B1 induction) or DEX (10 µM, positive control for CYP3A23, CYP3A2, and CYP3A18 induction). Culture medium was changed daily.
Isolation of Total RNA and Quantification of Total RNA Concentration. Hepatocytes were harvested at the end of the treatment period. Total cellular RNA was isolated using the TriZol reagent, and total RNA concentration was determined using the RiboGreen RNA Quantitation Kit (Jones et al., 1998) as detailed elsewhere (Chang et al., 2003).
Reverse Transcription and Quantification of Total cDNA Concentration. Reverse transcription was performed and total cDNA concentrations were quantified using the PicoGreen dsDNA Quantitation Kit (Singer et al., 1997) as detailed elsewhere (Chang et al., 2003).
PCR Primers. The sequences of the forward and reverse primers used to amplify CYP2B1 cDNA (GenBank accession number J00719
[GenBank]
) were 5'-CTGTGGGTCATGGAGAGCTG-3' and 5'-TCACACCGGCTACCAACCCT-3', respectively (Li and Kupfer, 1998). The sequences of the forward and reverse primers used to amplify CYP3A2 cDNA (GenBank accession number NM153312) were 5'-TTGATCCGTTGTTCTTGTCA-3' and 5'-GGCCAGGAAATACAAGACAA-3', respectively (Zhang et al., 1996). The sequences of the forward and reverse primers used to amplify CYP3A18 cDNA (gene accession number NM145782) were 5'-CAACTACGGTGATGGCATGT-3' and 5'-CACTCGGTTCTTCTGGTTTG-3', respectively (Mahnke et al., 1997). The sequences of the forward and reverse primers used to amplify CYP3A23 cDNA (GenBank accession number X96721
[GenBank]
) were 5'-GGAAATTCGATGTGGAGTGC-3' and 5'-AGGTTTGCCTTTCTCTTGCC-3', respectively (Mahnke et al., 1997). The specificity of the primers was confirmed by sequencing analysis of the purified amplicons as described previously (Yu et al., 2005).
Real-Time PCR Analysis. Each 20-µl PCR reaction mixture contained 1 unit of Platinum TaqDNA polymerase in 1x PCR reaction buffer (20 mM Tris-HCl, pH 8.4, and 50 mM KCl), 3 mM magnesium chloride (except that 5 mM was used to amplify CYP3A2 cDNA), 1 ng of total cDNA, 200 µM deoxynucleoside-5'-triphosphate mix, 0.2 µM concentration, each, of the forward and reverse primers, 0.25 mg/ml bovine serum albumin, and 2 µl of a 3.3x SYBR Green I solution. The conditions for the amplification of CYP2B1 cDNA were 95°C for 1 s (denaturation), 56°C for 6 s (annealing), and 72°C for 23 s (extension). The conditions for the amplification of CYP3A2 cDNA were 95°C for 1 s, 60°C for 6 s, and 72°C for 17 s. The conditions for the amplification of CYP3A18 cDNA were 95°C for 1 s, 60°C for 6 s, and 72°C for 30 s. The conditions for the amplification of CYP3A23 cDNA were 95°C for 1 s, 60°C for 6 s, and 72°C for 14 s. In all cases, the initial denaturation was performed at 95°C for 5 min. Under these conditions, no primer-dimer formation was detected. However, to prevent any potential contribution to the fluorescence signal due to primer-dimer formation, the real-time DNA thermal cycler (LightCycler; Roche Diagnostics, Mannheim, Germany) was programmed to record fluorescence readings after each cycle at a temperature several degrees lower than the melting temperature of the amplicon. Calibration curves were constructed as described previously (Cheung et al., 2004). The level of gene expression (copies of cDNA) was normalized to the amount of total cDNA used in the amplification, as quantified by the PicoGreen dsDNA Quantitation Kit (Singer et al., 1997). This approach to normalizing gene expression data was previously shown to be comparable to normalizing it to the amount of a "housekeeping" gene, such as cyclophilin (Yu et al., 2005). Results are expressed as fold increase over the vehicle-treated control group.
BROD Assay. Hepatocytes cultured in 24-well microplates were treated for 48 h with G. biloba extract (100 µg/ml), bilobalide (1–5 µg/ml), ginkgolide A (1.1–5 µg/ml), or PB (100 µM, positive control). Control incubations contained culture medium (vehicle for the extract) or DMSO (0.1% final concentration, vehicle for the individual chemicals). At the end of the treatment period, culture medium was removed and the cells were washed with 1x phosphate-buffered saline. The BROD assay was initiated by the addition of a 150-µl mixture containing 7-benzyloxyresorufin (15 µM) and dicumarol (10 µM) dissolved in DMSO (0.1% final concentration) (Donato et al., 1993). The reaction proceeded for 60 min in a 37°C incubator. Subsequently, a 75-µl aliquot of the reaction mixture was transferred to another well containing a 25-µl mixture of ß-glucuronidase (15 Fishman units) and arylsulfatase (120 Roy units) dissolved in 0.1 M sodium acetate (pH 4.5). The microplates were placed in a shaking 37°C incubator for 2 h. Following the addition of ethanol (200 µl), the microplates were centrifuged at 3000 rpm for 10 min. The fluorescence reading of a 200-µl sample of the supernatant was recorded at an excitation wavelength of 530 nm (5-nm bandwidth), an emission wavelength of 580 nm (5-nm bandwith), and a gain of 50 in a Cytofluor 2350 fluorescence microplate reader (Millipore Corporation, Billerica, MA). A standard curve was constructed with 10 to 80 pmol of resorufin.
Testosterone 6ß-Hydroxylation Assay. Hepatocytes cultured in 24-well microplates were treated as described above for the BROD assay, except that the positive control was DEX (10 µM). At the end of the treatment period, culture medium was removed and the cells were washed with 1x phosphate-buffered saline. The testosterone 6ß-hydroxylation assay was performed according to a published method (Nicoll-Griffith et al., 2004), but with minor modifications. Testosterone (200 µM) dissolved in DMSO (0.1% final concentration) was added in a volume of 200 µl. After an incubation period of 60 min in a 37°C incubator, the reaction was stopped with 100 µl of 100% acetonitrile and the mixture was spiked with 20 µl of an internal standard, which was 11ß-hydroxytestosterone (82 µg/ml in methanol containing 0.5% acetic acid). The incubation mixture was transferred to a microcentrifuge tube and the content was centrifuged at 10,000g for 3 min. A 20-µl sample of the supernatant was analyzed by HPLC. Our HPLC system consisted of a Phenomenex Gemini C18 column (150 x 2.0 mm, i.d., 5 µm; Phenomenex, Torrance, CA) linked to a Phenomenex SecurityGuard cartridge (ODS, 4.0 x 2.0 mm, i.d., 5 µm), a Waters model 1525 Binary HPLC pump (Waters Corporation Canada, Mississauga, ON, Canada), a Waters model 717 Plus autosampler, and a Waters model 2487 dual 
absorbance detector. Instrument operation and data acquisition were performed using Waters Breeze software (version 3.20). The levels of 6ß-hydroxytestosterone and 11ß-hydroxytestosterone were determined by reversed-phase HPLC, as modified from a published method (Purdon and Lehman-McKeeman, 1997), with an ultraviolet detector set at 242 nm. The analytes were eluted at ambient temperature at a flow rate of 0.2 ml/min, with a linear gradient initially with 55% methanol containing 0.5% acetic acid, increasing to 90% methanol containing 0.5% acetic acid over 8 min, and staying at 90% for 2 min before reequilibration at the initial condition. A standard curve was constructed with 30 to 5000 pmol of 6ß-hydroxytestosterone.
|
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Comparative Effect of Multiple Extracts of G. biloba on CYP2B1 and CYP3A23 Gene Expression. To determine whether G. biloba extract from another lot or from another manufacturer can also modulate CYP2B1 and CYP3A23 gene expression, primary cultures of rat hepatocytes were treated for 48 h with one of two different lots of G. biloba extract from the same manufacturer (i.e., extract A and extract B, Table 1) or an extract from a different manufacturer (i.e., extract C, Table 1) at a concentration of 10, 100, or 1000 µg/ml. Control hepatocytes were treated with culture medium (vehicle control). As shown in Table 2, all three extracts increased CYP2B1 mRNA levels at a concentration of 10 µg/ml or 100 µg/ml, but not 1000 µg/ml. By comparison, each of the extracts increased CYP3A23 mRNA expression at concentrations of 10, 100, and 1000 µg/ml.
|
Role of Bilobalide in the Modulation of CYP2B1 and CYP3A23 Gene Expression by G. biloba Extract. G. biloba extract contains various individual chemicals, including bilobalide (Table 1). Therefore, to determine whether this compound was responsible for the induction of CYP2B1 and CYP3A23 by G. biloba extract, primary cultures of rat hepatocytes were treated for 48 h with bilobalide (2.8 µg/ml). This concentration of bilobalide was chosen to represent the level present in a CYP2B1-inducing concentration (100 µg/ml; Fig. 1A) of a G. biloba extract (i.e., extract A, Table 1). As shown in Fig. 2A, bilobalide increased CYP2B1 mRNA expression by 7.9 ± 0.5-fold, which was similar to the increase (8.3 ± 1.7-fold) by the extract. In contrast, it did not increase CYP3A23 mRNA levels (Fig. 2B).
|
Differential Effect of Bilobalide on CYP2B1 and CYP3A23 Gene Expression. In a study that quantified the amount of bilobalide in 27 commercial products of G. biloba, the abundance of this compound ranged from 0.8% w/w to 4.6% w/w (Kressmann et al., 2002). In the present study, a concentration-response experiment was conducted whereby primary cultures of rat hepatocytes were treated for 48 h with bilobalide at 1, 2.8, or 5 µg/ml. As shown in Fig. 3A, bilobalide increased CYP2B1 mRNA expression in a concentration-dependent manner, but it had no effect on CYP3A23 mRNA levels.
|
). Therefore, a concentration-response experiment was performed. Primary cultures of rat hepatocytes were treated for 48 h with ginkgolide A at 1.1, 2.5, or 5 µg/ml. As shown in Fig. 3B, ginkgolide A increased CYP3A23 mRNA levels linearly over this range of concentrations, but it did not affect CYP2B1 mRNA expression.
Role of Flavonol Glycosides and Aglycones in the Modulation of CYP2B1 and CYP3A23 Gene Expression by G. biloba Extract. Flavonol glycosides are present in G. biloba extracts (van Beek, 2002). Therefore, primary cultures of rat hepatocytes were treated for 48 h with kaempferol-3-O-rutinoside (1.9 µg/ml), quercetin-3-O-rutinoside (4 µg/ml), isorhamnetin-3-O-rutinoside (0.6 µg/ml), or the vehicle (0.1% DMSO). The concentration of isorhamnetin-3-O-rutinoside was chosen to reflect the level present in a 100 µg/ml concentration of a G. biloba extract (i.e., extract A, Table 1). However, the levels of kaempferol-3-O-rutinoside and quercetin-3-O-rutinoside were not known in extract A. Therefore, we used a concentration that represented the level of all the kaempferol diglycosides or quercetin diglycosides present in the extract. As shown in Table 3, isorhamnetin-3-O-rutinoside, kaempferol-3-O-rutinoside, and quercetin-3-O-rutinoside did not affect CYP2B1 or CYP3A23 mRNA expression.
|
Flavonol glycosides are converted to the corresponding aglycones by the action of intestinal microflora (Walle, 2004). Therefore, primary cultures of rat hepatocytes were treated for 48 h with kaempferol (6.3 µg/ml), quercetin (10.6 µg/ml), isorhamnetin (4.1 µg/ml), or the vehicle (0.1% DMSO). These concentrations were selected to reflect the total levels of each of the flavonols (i.e., the sum of the aglycone and the glycosides). As shown in Table 3, kaempferol, quercetin, and isorhamnetin did not increase CYP2B1 or CYP3A23 mRNA expression.
Effect of G. biloba Extract, Ginkgolides, Bilobalide, and Flavonols on CYP3A2 and CYP3A18 Gene Expression. To determine the effect of G. biloba extract and some of its individual chemical constituents on CYP3A2 and CYP3A18 gene expression, primary cultures of rat hepatocytes were treated for 48 h with a G. biloba extract (i.e., extract A, Table 1; 100 µg/ml), culture medium (vehicle for the extract), ginkgolide A (1.1 or 5 µg/ml), ginkgolide B (0.3 µg/ml), ginkgolide C (1.4 µg/ml), ginkgolide J (0.6 µg/ml), bilobalide (2.8 µg/ml), kaempferol-3-O-rutinoside (1.9 µg/ml), quercetin-3-O-rutinoside (4 µg/ml), isorhamnetin-3-O-rutinoside (0.6 µg/ml), kaempferol (6.3 µg/ml), quercetin (10.6 µg/ml), isorhamnetin (4.1 µg/ml), DEX (10 µM, positive control), or DMSO (0.1%, vehicle for the individual chemicals). As shown in Fig. 4A, G. biloba extract increased CYP3A2 mRNA levels by 4.2 ± 1.2-fold, which was less than the 18 ± 5-fold induction by DEX. The extract also increased CYP3A18 mRNA expression (4.6 ± 1.7-fold), and the extent was similar to that by DEX (3.3 ± 0.7-fold) (Fig. 4B). However, ginkgolide A at the level (1.1 µg/ml) found in the extract had no effect on CYP3A2 or CYP3A18 mRNA levels, but an increase was obtained at a concentration of 5 µg/ml. In contrast, ginkgolide B, ginkgolide C, ginkgolide J, bilobalide, and the flavonols (both the aglycone and the diglycoside) did not influence CYP3A2 or CYP3A18 mRNA expression (data not shown).
|
|
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
; Umegaki et al., 2002). However, those findings may not be applicable to other extracts of G. biloba. Most of the commercially available G. biloba products contain approximately 6% (w/w) of terpene trilactones (Kressmann et al., 2002). The extract used in the previous studies (Shinozuka et al., 2002; Umegaki et al., 2002) contained a greater amount (10.6% w/w) of terpene trilactones. The present study was conducted with G. biloba extracts from different lots and different manufacturers. These extracts contained known amounts of terpene trilactones (6.2–6.6% w/w) and flavonol glycosides (21–24.4% w/w) (Table 1). Our results indicated that G. biloba extracts increased rat hepatic CYP2B1, CYP3A23, CYP3A2, and CYP3A18 gene expression.
CYP2B1 is a major PB-inducible P450 enzyme in rat liver (Waxman, 1999). In the present study, treatment of primary cultures of rat hepatocytes with G. biloba extract increased CYP2B1 gene expression level to approximately one-third of that by 100 µM PB, which is a maximal CYP2B1-inducing concentration (LeCluyse et al., 1999). It also increased CYP2B-mediated BROD to approximately one-quarter of the level by PB. At a concentration of 1000 µg/ml, G. biloba extract did not increase CYP2B1 gene expression. A potential explanation could be that one or more chemicals present in the extract might have suppressed the CYP2B induction response. Another explanation might be general cellular toxicity, although this was not assessed in our hepatocyte cultures. However, the same concentration of the extract resulted in a maximal increase in CYP3A23 gene expression.
Several hepatic CYP3A enzymes, such as CYP3A23, CYP3A2, and CYP3A18, are expressed in male rats (Mahnke et al., 1997; Matsubara et al., 2004). The constitutive expression of hepatic CYP3A23 is low, but it is a major DEX-inducible P450 enzyme (Mahnke et al., 1997). Our results obtained with a G. biloba extract (i.e., extract A, Table 1) showed that it induced CYP3A23 gene expression to a level that was approximately one-third of that by 10 µM DEX, which is a maximal CYP3A23-inducing concentration (Lu and Li, 2001). However, minor differences existed in the magnitude of the increase among the extracts tested. Similar to CYP3A23, CYP3A2 and CYP3A18 are also inducible by DEX (Mahnke et al., 1997). As shown in the present study, when compared with a maximal CYP3A-inducing concentration (10 µM) of DEX (Lu and Li, 2001), G. biloba extract was less effective in elevating CYP3A23 and CYP3A2 mRNA levels, but both were similarly effective in inducing CYP3A18. Consistent with the CYP3A gene expression data, treatment of cultured hepatocytes with G. biloba extract increased testosterone 6ß-hydroxylation, which is mediated by CYP3A23, CYP3A2, CYP3A18, and other CYP3A enzymes (Matsubara et al., 2004). Our cell culture results are in agreement with a previous study indicating that treatment of 5-week-old male Wistar rats with an extract of G. biloba (10, 100, or 1000 mg/kg via intragastric gavage once daily for 5 days) increased hepatic microsomal testosterone 6ß-hydroxylation activity (Umegaki et al., 2002).
Bilobalide is the most abundant terpene trilactone in commercial preparations of G. biloba (Kressmann et al., 2002). According to the German Commission E monograph, the amount of bilobalide in a G. biloba product should be within the range of 2.6 to 3.2% w/w (Blumenthal, 1998). A novel finding from the present study is that bilobalide is responsible for the modulation of CYP2B1 gene expression and CYP2B-mediated BROD by G. biloba extract. Our results also indicated a concentration-dependent relationship on the effect of bilobalide (1–5 µg/ml) on CYP2B1 mRNA and enzyme activity. We tested this range of concentrations because according to a survey of commercial G. biloba products, the bilobalide concentrations were between 0.8 and 4.6 µg/ml (Kressmann et al., 2002). Interestingly, an earlier study reported that the oral administration of bilobalide (30 mg/kg once daily for 4 days) to mice increased hepatic microsomal 7-methoxycoumarin O-demethylation activity (Sasaki et al., 1997). However, it is not known which P450 enzyme(s) is responsible for this PB-inducible activity in mouse liver microsomes.
Another group of terpene trilactones present in G. biloba extracts are the diterpenes, which include ginkgolides A, B, C, and J (van Beek, 2002). According to the German Commission E monograph, the total amount of ginkgolides in a G. biloba product should be within the range of 2.8 to 3.4% w/w (Blumenthal, 1998), although there is no specified amount for the individual ginkgolides. As shown for the first time in the present study, ginkgolide A is partially responsible for the modulation of CYP3A23 gene expression by G. biloba extract. This compound, when determined at a concentration (1.1 µg/ml) that reflected the level present in a CYP3A-inducing concentration (100 µg/ml) of a G. biloba extract (i.e., extract A, Table 1), increased CYP3A23 mRNA expression to approximately one-half of that by the extract. Ginkgolide A increased CYP3A23 mRNA expression and CYP3A-catalyzed testosterone 6ß-hydroxylation over the range of 1.1 to 5 µg/ml, which reflected the concentrations (1.1–3.8%) of ginkgolide A present in a survey of 27 commercial products of G. biloba (Kressmann et al., 2002). Interestingly, ginkgolide A, at the concentration (1.1 µg/ml) present in extract A (Table 1), did not increase CYP3A2 or CYP3A18 mRNA expression. A greater concentration (5 µg/ml) was required for the modulation of these two CYP3A genes by ginkgolide A.
A novel finding from the present study is the differential induction of CYP2B1 and CYP3A23 gene expression by bilobalide and ginkgolide A. The molecular basis for these effects is not known. The current evidence indicates that the constitutive androstane receptor mediates the induction of CYP2B1 (Muangmoonchai et al., 2001), whereas the pregnane X receptor is a regulator of CYP3A23 induction (Zhang et al., 1999). Therefore, a potential explanation for the differential effect of the two compounds on CYP2B1 and CYP3A23 expression is the preferential activation of constitutive androstane receptor and pregnane X receptor by bilobalide and ginkgolide A, respectively. Studies are planned to elucidate the molecular mechanism of CYP2B and CYP3A induction by these naturally occurring chemicals.
Another major group of constituents in G. biloba extracts are the flavonol glycosides, including those of quercetin, kaempferol, and isorhamnetin (van Beek, 2002). However, as demonstrated in the present study, the aglycone and the 3-O-rutinoside of each of these flavonols did not increase CYP2B1, CYP3A23, CYP3A2, or CYP3A18 gene expression. This finding is consistent with the result from a previous study showing a lack of an increase in hepatic microsomal CYP2B-mediated pentoxyresorufin O-dealkylation activity in rats administered quercetin via the diet for 2 weeks at a dosage of 1% w/w (Brouard et al., 1988).
It is not known whether G. biloba extract alters CYP2B-mediated drug clearance in humans, but conflicting data appear to exist for the effect of this herbal medicine on the clearance of drugs metabolized by human CYP3A enzymes. In one study, the oral ingestion of G. biloba extract (60 mg four times daily for 28 days) by healthy human volunteers did not alter the 1-h serum ratio of 1-hydroxymidazolam/midazolam (Gurley et al., 2002), which is used as an in vivo metabolic index for CYP3A activity (Thummel et al., 1994). However, in another study, the oral ingestion of G. biloba extract (120 mg twice daily) for 14 days resulted in a modest but statistically significant reduction in the area under the plasma alprazolam concentration-time curve (Markowitz et al., 2003). This result suggests a CYP3A-inductive effect by G. biloba because alprazolam is metabolized predominantly by CYP3A enzymes (Gorski et al., 1999). In the studies that investigated the effect of G. biloba extract on drug elimination (Gurley et al., 2002; Markowitz et al., 2003), the levels of bilobalide and ginkgolide A in the extracts were not reported. Studies are in progress to determine directly whether G. biloba extract and some of its individual chemical constituents are capable of modulating CYP2B6 and CYP3A4 expression and function in primary cultures of human hepatocytes.
In summary, the present study in primary cultures of rat hepatocytes shows 1) induction of CYP2B1, CYP3A23, CYP3A2, and CYP3A18 gene expression by G. biloba, as shown with multiple extracts containing known amounts of terpene trilactones and flavonol glycosides; 2) bilobalide and ginkgolide A are novel inducers of CYP2B1 and CYP3A, respectively, as demonstrated by real-time PCR and enzyme activity assays; and 3) bilobalide is responsible for the modulation of CYP2B1 expression by G. biloba extract, whereas ginkgolide A contributes partially to the modulation of CYP3A23 expression by this herbal medicine.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
ABBREVIATIONS: P450, cytochrome P450; BROD, 7-benzyloxyresorufin O-dealkylation; DEX, dexamethasone; DMSO, dimethyl sulfoxide; PB, phenobarbital; PCR, polymerase chain reaction.
Address correspondence to: Dr. Thomas K. H. Chang, Faculty of Pharmaceutical Sciences, The University of British Columbia, 2146 East Mall, Vancouver, BC, V6T 1Z3, Canada. E-mail: tchang{at}interchange.ubc.ca
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
