![[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}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Bionovo Inc., Emeryville, California (R.K., L.S., S.C., R.E.S., Y.L.Z., I.C., U.C.); and Clinical Research and Development, Department of Anesthesiology, University of Colorado Health Sciences Center, Denver, Colorado (Y.L.Z., U.C.)
(Received March 7, 2008; Accepted July 29, 2008)
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
. Incubation with pooled human liver microsomes yielded a single metabolite. Its structure was elucidated using tandem mass spectrometry in combination with analysis of the fragmentation patterns. The metabolite resulted from the loss of two hydrogens and rearrangement to the stable 7,4'-dihydroxyflavone. The structure was also confirmed by comparison with authentic standard material. Maximum apparent reaction velocity (Vmax) and Michaelis-Menten constant (Km) for the formation of 7,4'-dihydroxyflavone were 32.5 nmol/g protein/min and 128 µM, respectively. After correction for protein binding (free fraction = 0.84), the apparent intrinsic clearance (CLint) for 7,4'-dihydroxyflavone formation was 0.3 ml/g/min. Liquiritigenin was almost exclusively metabolized by CYP3A enzymes. Comparison of liquiritigenin metabolism in human liver microsomes isolated from 16 individuals showed 9.5-fold variability in metabolite formation (3.4-32.2 nmol/g protein/min). An estrogen receptor luciferase assay indicated that the metabolite was a 3-fold more potent activator of the estrogen receptor β than the parent compound and did not activate the estrogen receptor .
; Moon et al., 2006).
Liquiritigenin is an active ingredient of MF101, an herbal extract currently in clinical trials for the relief of postmenopausal symptoms in women. MF101 is an ethanol/aqueous extract of 22 herbal species used in traditional Chinese medicine (Cvoro et al., 2007). Because of its effects on the estrogen receptor, MF101 belongs to a class of compounds referred to as selective estrogen receptor modulators. Most clinically used estrogen receptor agonists or antagonists similarly affect both estrogen receptor and β. Selective estrogen receptor modulators discriminate between estrogen receptor and β (Riggs and Hartmann, 2003). MF101 is a potent agonist of estrogen receptor β but does not activate estrogen receptor (Cvoro et al., 2007). Most importantly, it is not implicated in tumor formation as a result of estrogen receptor activation (Cvoro et al., 2007). MF101 is an oral drug designed for the treatment of hot flashes and night sweats in perimenopausal and menopausal women. In animal studies, the compound did not adversely alter reproductive hormones or promote tumor formation in the breast or uterus, suggesting that MF101 will not increase the risk of either breast or uterine cancer (Hillerns et al., 2005; Cvoro et al., 2007).
As of today, liquiritigenin metabolism has been studied mainly in rats. In the rat, liquiritigenin is metabolized to five glucuronide and sulfate conjugated metabolites (Shimamura et al., 1990, 1993) that are actively excreted into bile but can also be found in urine (Shimamura et al., 1993, 1994). Nikolic and Van Breemen (2004) studied the oxidative metabolism of liquiritigenin using rat liver microsomes and found six metabolites: 7,3',4'-trihydroxyflavone, a hydroxyl quinone metabolite, two A-ring dihydroxy metabolites, 7,4'-dihydroxyflavone, and 7-hydroxychromone.
However, the human metabolism of liquiritigenin is still largely unknown. As a first step, it was the goal of our study to assess the oxidative metabolism of liquiritigenin by human liver microsomes, to elucidate the structures of the metabolites formed, to identify the cytochrome P450 enzymes involved in human oxidative metabolism of liquiritigenin, to assess interindividual variability of liquiritigenin metabolite formation in human liver microsomes, to evaluate potential interspecies differences, and to test the metabolites' activity as estrogen β-receptor agonists.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
General Procedure for Microsomal Incubations. Microsomal incubations contained liquiritigenin (0.04 mM), 200 mg/l microsomal protein, and the NADPH-regenerating system in phosphate buffer, pH 7.4 (0.1 mol/l). Final volume was either 500 or 1000 µl. Samples were incubated at 37°C for 5 min before initiating the reaction by addition of microsomal proteins. Thereafter, samples were incubated at 37°C for 60 min. Reactions were terminated, and proteins were precipitated by adding a protein precipitation solution [0.2 mol/l ZnSO4/methanol, 30:70 (v/v)] that also contained the internal standard 2',4'-dihydroxychalcone at a concentration of 1 µg/ml. Samples were vortexed on a DS-500 orbital shaker (VWR, West Chester, PA) at 500 rpm for 30 min and were centrifuged at 24,400g at 4°C for 5 min. Supernatants were transferred into HPLC vials for LC/MS analysis.
Quantification of Liquiritigenin and Its Major Metabolite Using LC/MS. An Agilent series 1100 HPLC system, in combination with a mass selective detector, was used for quantitative analysis (all Agilent Technologies, Santa Clara, CA). One hundred microliters of the extracted sample was injected onto an extraction column (4.6 x 12.5-mm Eclipse XDB-C8, 5-µm particle size; Agilent Technologies) with a mobile phase of 20% methanol in 0.1% formic acid at a flow rate of 5 ml/min. A switching valve was activated after 1 min, and the analytes were backflushed from the extraction column onto two sequentially linked analytical columns (4.6 x 250-mm Eclipse XDB-C8, 5-µm particle size; Agilent Technologies). The following gradient was used: methanol/0.1% formic acid [20:80 (v/v) for 1 min, to 75:25 (v/v) in 17 min, to 98:2 (v/v) within 3 min, and 98:2 (v/v) for 4 min], and then the columns were re-equilibrated to starting conditions. The flow rate was set to 1 ml/min, and the column temperature was kept at 65°C.
The mass spectrometer was run in negative scan mode detecting a mass range of m/z = 150 to 500. The drying gas flow was set to 12 l/min, the nebulizer pressure to 50 psi, the drying gas temperature to 350°C, the capillary voltage to 3000 V, and the fragmentor to 100 V. Analytes were quantified in extracted ion mode using the following ions: liquiritigenin (m/z = 255), the metabolite of liquiritigenin (7,4'-dihydroxyflavone, m/z = 253), and 2',4'-dihydroxychalcone (internal standard, m/z = 239, all [M-H]-).
For quantification, areas under the peak were corrected using the internal standard and were compared with a 7,4'-dihydroxyflavone calibration curve. Because extraction recovery depends on the amount of protein in the incubation mixture, two sets of calibration curves were prepared based on two different microsomal protein concentrations (0.2 and 0.5 mg/ml human liver microsomes solution). The metabolite was added at the following final concentrations in calibration samples: 0, 5, 10, 25, 50, 100, 250, 500, 1000, and 1500 nM. Samples were extracted immediately by adding protein precipitation solution containing the internal standard [2',4'-dihydroxychalcone, 0.2 mol/l ZnSO4/methanol, 30:70 (v/v)] as described above.
The assay for 7,4'-dihydroxyflavone had the following key performance parameters. The lower limit of quantitation (LLOQ) was determined as the lowest quantity consistently achieving accuracy 
±20% of the nominal concentration and a precision 20%. Lower limits of detection were 10 nM, LLOQs were 25 nM, and ranges of linear response were 25 to 1500 nM (r2 > 0.9992, n = 6). There were no carryover, matrix interferences, or relevant ion suppression as tested using the procedure proposed by Müller et al. (2002). The analytes (liquiritigenin, 7,4'-dihydroxyflavone, and internal standard) were stable in the extract for at least 48 h at +4°C (temperature in the autosampler). Intraday and interday accuracies and precisions were <10% (except at the lower limit of quantitation, <20%).
Identification of Liquiritigenin Metabolite Structures. For structural identification, metabolites were separated using HPLC and MS spectra (m/z = 100-1600), and MS/MS spectra (m/z = 50-300) were recorded on an Agilent QTOF 6510 instrument (Agilent Technologies). Metabolites were identified based on their accurate mass determination and fragmentation patterns in positive and negative mode. The hypothetical structures of the metabolites were verified by comparison with the authentic, commercially available materials (HPLC retention, exact mass, and MS/MS fragments).
Evaluation of the Time Dependence of Liquiritigenin Metabolite Formation. A solution of liquiritigenin (0.04 mM final concentration) and the NADPH-regenerating system in phosphate buffer was aliquoted into 1.5-ml conical polypropylene tubes with snap-on lids (n = 6 for each time point). After preincubation for 5 min, the reaction was initiated by adding human liver microsomes to each sample for a final protein concentration of 200 mg/l. Samples were incubated for 0 (stopped immediately), 5, 10, 15, 20, 30, 45, 60, 90, or 120 min. An additional set of samples was incubated, but no NADPH had been added (controls). Reactions were terminated by the addition of the protein precipitation/internal standard solution and further processed as described above.
Evaluation of the Dependence of Liquiritigenin Metabolite Formation on the Microsomal Protein Concentration. A solution of liquiritigenin (0.04 mM final concentration) and the NADPH-regenerating system in phosphate buffer was aliquoted into 1.5-ml conical polypropylene tubes with snap-on lids (n = 6 for each protein concentration). After preincubation for 5 min, the reaction was initiated by adding microsomes to a final protein concentration of 0, 10, 50, 100, 250, 500, 1000, and 2500 mg/l. Samples were incubated for 60 min, and samples were extracted as described above.
Estimation of the Apparent Michaelis-Menten Constant (Km) and the Maximum Metabolite Formation Velocity (Vmax). Microsomal incubations were conducted with final liquiritigenin concentrations of 0, 20, 40, 60, 80, 100, 150, 200, 250, 300, 400, and 500 µM (n = 6/concentration). Samples were preincubated for 5 min and the reaction was started by adding human liver microsomes to a final microsomal protein concentration of 500 mg/l. Samples were incubated for 60 min at 37°C. Reactions were stopped by adding 250 µl of protein precipitation/internal standard solution and were extracted as previously described. Incubation mixtures without NADPH incubated for 60 min were used as controls. Apparent Km and Vmax were determined after data fitting using the Enzyme Kinetics Module (version 1.3) of the SigmaPlot software (version 9.0; Systat Software, Inc., San Jose, CA).
Determination of the Nonspecific Binding of Liquiritigenin to Microsomal Protein. Solutions of human liver microsomes (500 mg/l microsomal protein) and 80 µM liquiritigenin were prepared (both in 0.1 M phosphate buffer, pH 7.4). Using a 96-well equilibrium dialyzer plate (molecular weight cut-off 5 kDa; Harvard Apparatus Inc., Holliston, MA), the liquiritigenin solution was dialyzed against the microsomal solution or phosphate buffer only (200 µl/well, n = 6/experiment) at 37°C in a Big Shot II rotator oven (Boekel Scientific, Feasterville, PA) for 24 h. The volume of liquid in each well was checked to ensure that no volume increase or decrease occurred because of osmosis. Seventy-five microliters of ice-cold zinc sulfate (0.2 mol/l)/methanol [3:7 (v/v)] containing 1 µg/l internal standard 2',4'-dihydroxychalcone, and 300 µl of acetonitrile was added to 150 µl of the solution from each well. The samples were vortexed after each addition, shaken for 10 min, and then centrifuged at 24,400g for 5 min. The supernatant was analyzed for liquiritigenin by LC/MS as described above. The fraction of unbound substrate (Fu) was determined by comparison of substrate concentration in wells originally containing the substrate. To confirm that equilibrium was achieved, control wells that contained no microsomal proteins were compared, and equal concentrations of liquiritigenin were found.
Determination of the Human Cytochrome P450 Enzymes Responsible for the Oxidative Metabolism of Liquiritigenin and Interindividual Variability. The following Escherichia coli-expressed human cytochrome P450 enzymes were tested at 25 nM: 1A1, 1A2, 2B6, 2C8, 2C9, 2 C9*2, 2C19, 2D6, 2 D6*2, 2 D6*10, 2 D6*39, 2E1, 3A4, and E. coli control membrane protein extracts. Solutions of liquiritigenin (0.04 mM) and the NADPH-regenerating system in 0.1 M phosphate buffer were preincubated for 5 min, and then P450 enzymes were added to start the reaction. Samples were incubated for 120 min and extracted as described earlier.
|
|
Species-Dependent Differences in Liquiritigenin Oxidative Liver Metabolism. Microsomes from the following species were tested: rhesus monkey, cynomolgus monkey, beagle, minipig, guinea pig, Sprague-Dawley rat, Fischer rat, CD1 mouse, and B6C3F1 mouse. A stock solution of the NADPH-regenerating system and liquiritigenin (0.04 mM) in phosphate buffer (0.1 mol/l) was aliquoted into 1.5-ml conical polypropylene tubes with snap-on lids (n = 6/species). After a 5-min preincubation period, microsomes (0.4 mg/ml protein) were added to each sample and then incubated for 60 min at 37°C prior to being extracted as previously described.
Assessment of the β-Estrogen Receptor Agonist Activity of the Major Liquiritigenin Metabolite. After the major liquiritigenin metabolite generated by human liver microsomes was identified as 7,4'-dihydroxyflavone, its binding to and activation of estrogen receptors were examined and compared with the one of liquiritigenin.
Estrogen Receptor Binding Assay. The relative binding affinity of 7,4'-dihydroxyflavone and liquiritigenin to pure, full-length receptors was determined using the estrogen receptor and β competitor assay kits (Invitrogen, Carlsbad, CA). Fluorescence polarization of fluorophore-tagged estrogen bound to the or β receptor, whereas in the presence of increasing amounts of competitor ligand or extract, it was determined (10 readings/well; 0.02-ms integration time; G factor, 1.1087) using a GENios Pro microplate reader (Tecan, Durham, NC), with fluorescein excitation (485 nM) and emission (530 nM) filters. The vehicle ethanol was used as the negative control. Each 7,4'-dihydroxyflavone and liquiritigenin concentration was tested in triplicate.
Estrogen Receptor Transfection and Luciferase Assay. U2OS cells were grown to 85% confluency, trypsinized from 150-mm plates, centrifuged in 50-ml conical tubes, and resuspended in phosphate-buffered saline containing 0.1% glucose. Cells were aliquoted (500 µl) into 0.4-cm cuvettes with 3 µg of reporter plasmid and 1 µg of estrogen receptor or β expression vectors. Cells were electroporated using a Genepulser II (Bio-Rad, Hercules, CA) and were resuspended in phenol red-free Dulbecco's modified Eagle's medium/F12 media with 4% charcoal-dextran-stripped fetal bovine serum. Cells were plated at 1 ml/well in 12-well plates and treated with varying dilutions of 7,4'-dihydroxyflavone or liquiritigenin overnight for 18 h. Cells were lysed with one freeze-thaw cycle and 200 µl of 1x Reporter Lysis Buffer (Promega, Madison, WI). Activity was determined using the Luciferase Assay System (Promega) in a Veritas luminometer (Turner Designs, Sunnyvale, CA).
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
) after incubation of liquiritigenin with rat liver microsomes. Incubation of the 7,4'-dihydroxyflavone itself yielded minimal amounts of two metabolites with a mol. wt. gain of +16 (not identified).
|
Time and protein dependence of 7,4'-dihydroxyflavone formation by pooled human liver microsomes was established. Liquiritigenin was incubated with microsomes from 0 to 120 min. The reaction was linear up to 60 min. Protein dependence was tested over a microsomal protein concentration range from 0 to 2500 mg/l. The reaction was found linear up to 1000 mg/l protein. Based on these results, incubation times of 60 min and protein concentrations of 500 mg/l were chosen to assess the formation kinetics of 7,4'-dihydroxyflavone by pooled human liver microsomes. The formation of 7,4'-dihydroxyflavone followed Michaelis-Menten kinetics (Fig. 3). The estimated apparent Km was 128 µM, and the apparent Vmax was 32.5 nmol/g protein/min. The unbound intrinsic clearance was calculated: intrinsic clearance (CLint) = Vmax/Km · Fu, where Fu is the unbound fraction of substrate in solution during microsomal incubation. An average Fu of 0.84 was calculated from equilibrium dialysis experiments. Based on these data, the apparent CLint for 7,4'-dihydroxyflavone formation was 0.3 ml/g/min.
|
). When 7,4'-dihydroxyflavone formation was plotted against the testosterone hydroxylation rates, as provided by the manufacturer of the individual human liver microsomes, there was a significant correlation (r2 = 0.896) (Fig. 6; Table 1), further supporting this isozyme's primary role in the oxidative metabolism of liquiritigenin. Interestingly, microsomes of the individuals with the highest cytochrome P450 3A4/5 activity also generated traces of two hydroxylated metabolites. However, again, the amounts were too small for further assessment of their structures. Specific activities of none of the other major cytochrome P450 enzyme activities tested by the manufacturer showed a significant correlation with the individual 7,4'-dihydroxyflavone formation rates (Table 1). In addition to cytochrome P450 3A4/5, the best correlation (r2 = 0.57, statistically not significant) was found for cytochrome P450 2B6; however, incubation of liquiritigenin with the isolated cytochrome P450 enzyme did not produce any detectable 7,4'-dihydroxyflavone. Based on those results, it was concluded that cytochrome P450 3A4/5 enzymes are mainly responsible for the metabolism of liquiritigenin to 7,4'-dihydroxyflavone in the human liver. This was further confirmed by inhibition studies with the specific cytochrome P450 inhibitor ketoconazole and specific CYP3A antibodies based on pooled human liver microsomes. A mean half-maximal inhibition concentration (IC50) of 0.9 nM was estimated for ketoconazole after curve fitting using the SigmaPlot enzyme kinetics module, and the inhibition constant (Ki) was determined to be 0.09 nM. The maximum inhibition of liquiritigenin metabolite formation that could be reached with ketoconazole was 76.4%. Specific CYP3A antibodies (BD Gentest, Woburn, MA) inhibited 7,4'-dihydroxyflavone formation by pooled human liver microsomes by >50%.
|
|
|
Human microsomes were compared with microsomes from different animal species (Fig. 7). The formation rates of 7,4'-dihydroxyflavone ranged from 1.7 to 6.3 nmol/g protein/min, with the exception of the guinea pig, which exhibited the lowest formation rates at 0.8 nmol/g protein/min.
|
The metabolite 7,4'-dihydroxyflavone binds to both estrogen receptors and β (Fig. 8A). However, the IC50 in the competitive binding assay based on fluorophore-tagged estrogen was 10-fold lower for the β- than for the -receptor; 0.59 (0.48-0.71) µmol/l versus 6.2 (1.4-26.3) µmol/l, all median (95% confidence interval). Most importantly, 7,4'-dihydroxyflavone specifically activated the β-receptor [EC50, 0.23 (0.14-0.35) µM, median 95% confidence interval], whereas activation of the estrogen receptor was not different from the vehicle control (Fig. 8B). In comparison, IC50 values of the parent compound liquiritigenin in the competitive estrogen receptor binding tests were 2.8 µM (median, 95% confidence interval, 2.1-3.5 µM) for estrogen receptor and 0.41 µM (median, 95% confidence interval, 0.32-0.50 µM) for estrogen receptor β. Liquiritigenin activated the β-receptor (EC50, 0.69 µM), whereas activation of the estrogen receptor was not different from the vehicle control.
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
) found that rat liver microsomes metabolized liquiritigenin to 7,4'-dihydroxyflavone and several hydroxylated products, including 3'- and 1'-hydroxylated metabolites and two metabolites hydroxylated at the 5, 6, or 8 carbon (the exact position was not identified). In vitro metabolism of liquiritigenin with pooled human liver microsomes, however, yielded almost exclusively the 7,4'-dihydroxyflavone. Traces of two hydroxylated metabolites that could not be fully characterized were detected only when liquiritigenin was incubated with isolated CYP3A4 or after incubation with liver microsomes from an individual with high CYP3A activity. Based on our results, we conclude that 7,4'-dihydroxyflavone is the major metabolite generated by human liver microsomes. Incubation of the metabolite 7,4'-dihydroxyflavone with pooled human liver microsomes under the same conditions yielded only traces of two hydroxylated metabolites, although both structures could not be elucidated because of the low quantities generated.
Although we made an extensive effort to detect potential hydroxylated intermediates in the reaction pathway from liquiritigenin to 7,4'-dihydroxyflavone, none could be detected. Therefore, we hypothesize a quinone methide intermediate rather than a hydroxylation-dehydration mechanism. Because of the very low intrinsic formation clearance of 7,4'-dihydroxyflavone (see also below), we were unable to generate sufficient quantities of the metabolite to confirm its structure by NMR spectroscopy. However, the high-resolution mass spectra, including fragments in MS/MS spectra, and the HPLC retention time of the metabolite were identical to that of authentic 7,4'-dihydroxyflavone material.
Because microsomes are a mixture of different enzymes, only apparent enzyme kinetic parameters could be determined. The apparent maximum formation velocity (Vmax = 32.5 nmol/g protein/min) is low in comparison with most clinically relevant CYP3A substrates. Also, the affinity of the substrate to the microsomal enzymes is relatively low, as indicated by an apparent Km value of 128 µM.
The intrinsic clearance (CLint = Vmax/Km) is a parameter commonly used for quantitative in vitro-in vivo allometric scaling (Ashforth et al., 1995; Iwatsubo et al., 1996; Houston and Carlile, 1997). It has been recognized that the binding of the substrate to microsomes can hamper the prediction of the metabolic clearance in vivo. Thus, correction of the Km with the unbound substrate fraction during incubation (Fu) gives a much better correlation between in vitro drug metabolism and in vivo pharmacokinetics (Houston, 1994; Obach, 1996; Iwatsubo et al., 1997; Austin et al., 2002). The apparent intrinsic formation clearance after correction for nonspecific protein binding of 7,4'-dihydroxyflavone with 0.30 ml/g/min is similar to that of pravastatin with 0.20 ml/g/min (Jacobsen et al., 1999). Microsomal oxidative metabolism of pravastatin is not considered a clinically relevant elimination pathway (Christians et al., 1998). This can also be assumed to be the case for liquiritigenin. However, flavonoids in general are readily conjugated by phase II enzymes (Zamek-Gliszczynski et al., 2006; Zhang et al., 2007). Previous studies demonstrated that liquiritigenin undergoes significant metabolism by conjugation in the rat (Shimamura et al., 1990; 1993), but no other species have been examined.
The only cytochrome P450 enzyme that converted liquiritigenin to 7,4'-dihydroxyflavone in significant amounts was CYP3A4, as demonstrated by incubations with isolated cytochrome P450 enzymes and correlation to specific cytochrome P450 enzyme activities in human liver microsomes from 16 individuals. CYP3A is the most abundant cytochrome P450 enzyme in humans, and it has been estimated that CYP3A4 is the major drug-metabolizing enzyme for nearly 50% of all currently marketed drugs (Shimada et al., 1994; Yan and Caldwell, 2001). This raises the issue of potential drug-drug interactions. A significant cytochrome P450 drug interaction may occur when two or more drugs compete for the same enzyme and when the metabolic reactions catalyzed by this enzyme constitute the major elimination pathway (Rowland and Matin, 1973; Lin and Lu, 1998). Furthermore, a large interindividual variability of metabolite formation by CYP3A enzymes in the human liver is well established (Shimada et al., 1994; Thummel et al., 1994). Based on our results, it is unlikely that drug-drug interactions at cytochrome P450 enzymes will affect liquiritigenin pharmacokinetics. Tsukamoto et al. (2005) isolated several CYP3A inhibitors from licorice, among them the liquiritigenin glycoside (liquiritin), but it remains unclear as to whether liquiritigenin contributes to the CYP3A inhibitory effect of licorice. Our data suggest that clinically relevant competitive drug-drug interactions at microsomal enzymes seem rather unlikely. Because there was no relevant metabolism of liquiritigenin by cytochrome P450 enzymes other than CYP3A4, it is unlikely that drugs that interact with other cytochrome P450 enzymes will modify liquiritigenin elimination or that liquiritigenin will competitively inhibit their metabolism. 7,4'-Dihydroxyflavone showed significant activity as a β-estrogen receptor agonist. Its binding affinity to the estrogen receptor β as estimated based on the IC50 in the competitive estrogen binding assay was similar to that of its parent liquiritigenin; however, binding to the estrogen receptor was lower, indicating better selectivity. The estrogen receptor luciferase assay suggested that the metabolite was a 3-fold more potent activator of the estrogen receptor β than the parent compound, and activation was highly estrogen receptor β-specific. The contribution of both 7,4'-dihydroxyflavone and liquiritigenin to the positive effect of the drug on the relief of postmenopausal symptoms in humans is currently unknown. It was not possible to evaluate the human pharmacokinetics of 7,4'-dihydroxyflavone after formation by cytochrome P4503A enzymes from its parent liquiritigenin because the MF101 preparation contains 7,4'-dihydroxyflavone extracted from plants that cannot be differentiated from the metabolite. The efficacy of MF101 is most likely the result of a combination of multiple ingredients in the herbal drug, including these two key ingredients.
In conclusion, the oxidative metabolism of liquiritigenin in the human liver by cytochrome P4503A enzymes results mostly in the active metabolite 7,4'-dihydroxyflavone. However, the intrinsic formation clearance of the metabolite was relatively low, and based on our data, it is reasonable to assume that, although active, the metabolite does not significantly contribute to the overall biological activity of liquiritigenin. Recent in vivo studies in rats confirmed this conclusion. After oral gavage and i.v. bolus injection, the metabolite was not detectable in most plasma samples and only sporadically in urine and feces (our unpublished data). Phase II metabolism of liquiritigenin could have greater relevance to its excretion in humans. Glucuronidation, sulfation, and glutathione adduction reactions have been shown to be a major elimination pathway for a variety of flavonoids (Zamek-Gliszczynski et al., 2006; Zhang et al., 2007). Ongoing studies in our group are currently focused on phase II metabolism of liquiritigenin as the potential major elimination pathway of liquiritigenin in humans.
|
Footnotes |
|---|
ABBREVIATIONS: liquiritigenin, 2,3-dihydro-7-hydroxy-2-(4-hydroxyphenyl)-(S)-4H-1-benzopyran-4-one; HPLC, high-performance liquid chromatography; LC, liquid chromatography; MS, mass spectrometry; LLOQ, lower limit of quantitation; MS/MS, tandem mass spectrometry; Fu, fraction of unbound substrate; CLint, intrinsic clearance.
Address correspondence to: Dr. René Kupfer, BioNovo Inc., 12635 Montview Blvd., Suite 155, Aurora, CO 80045. E-mail: rene.kupfer{at}bionovo.com
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
