![[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}
|
|
|
|
|
||
National Center for Toxicological Research, Food and Drug Administration, Jefferson, Arkansas
(Received May 9, 2003; accepted August 28, 2003)
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
). St. John's wort has been suggested as an alternative herbal treatment for mild to moderate depression and has been referred to as the "Prozac from the plant kingdom" (Carlo et al., 2001). Several clinical trials have demonstrated the efficacy of St. John's wort to be superior to placebo in treating mild depression (Greeson et al., 2001; Whiskey et al., 2001). Animal and in vitro studies suggest that St John's wort constituents inhibit neuronal reuptake of serotonin and dopamine (Nathan, 1999).
Extracts of St. John's wort standardized for hyperforin content have been correlated in a dose-dependent manner with clinical antidepressive efficacy (Chatterjee et al., 1998a). Cumulative evidence (Müller et al., 1998) suggested that hyperforin may be the key constituent responsible for the antidepressant property of Hypericum. Inhibition of serotonin, norepinephrine, and dopamine reuptake, and down-regulation of cortical-adrenoceptor expression have been proposed as mechanisms for the antidepressive effect (Chatterjee et al., 1998b). Recently, hyperforin was also reported as an inhibitor of cyclooxygenase-1 and 5-lipoxygenase (Albert et al., 2002).
In 1998 to 1999, St. John's wort rose to second place in the market of dietary supplements, with about $400 million in sales in the United States and an estimated $6 billion in Europe (Greeson et al., 2001). Later, it fell to fifth place due, in part, to reports of interactions with several prescription drugs, including indinavir (Piscitelli et al., 2000), cyclosporine (Barone et al., 2000), digoxin (Johne et al., 1999), and warfarin (Yue and Bergqist, 2000). These effects were explained in part by in vitro studies, which showed that hyperforin is an agonist of the human pregnane X receptor system (Moore et al., 2000; Wentworth et al., 2000). The pregnane X receptor controls the induction and expression of several proteins involved in drug metabolism and clearance, including the cytochrome P450 isoform, CYP3A4, and the multidrug transporter, P-glycoprotein (Ekins and Erickson, 2002). Both these proteins have been shown to be induced by hyperforin in human tissues (Moore et al., 2000; Durr et al., 2002). An in vitro study, however, indicated that hyperforin was a competitive inhibitor of CYP3A4 and a noncompetitive inhibitor of CYP2D6 (Obach, 2000).
Hyperforin is highly lipophilic, heat-sensitive, and susceptible to photodegradation, and decomposes quickly in nonpolar solvents, such as hexane (Orth and Schmidt, 2000). The stability and isolation of hyperforin were specifically documented in the related scientific publication (Orth et al., 1999). The quantitative analysis of hyperforin from herb, dietary supplements, and functional foods has been discussed recently (Bilia et al., 2001; Ang et al., 2002). The pharmacokinetic profile of hyperforin in plasma was studied by different analytical methods, including HPLC/MS1 and HPLC/UV with different extractions (Biber et al., 1998; Cui et al., 2002). The long half-life for clearance of hyperforin in chronic-exposure clinical dosing (Bennet et al., 1998) implied important roles of its metabolites in the pharmacology and herbal-drug interactions of St. John's wort. The primary objective of our study was to investigate the in vitro metabolism of hyperforin by rat liver microsomes and to identify the major metabolites, to better characterize the biotransformation of hyperforin for further understanding of its effect mechanisms.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). The yield of hyperforin after the purification was about 0.35% and the purity of isolated hyperforin (>99%) was determined by NMR, HPLC/MS, and HPLC/photodiode array detection (PDA). The pure hyperforin solution was prepared in methanol (25 mg/ml) and stored at –70°C. All operations, including incubations, extractions, and isolations, were performed in the dark or under dim yellow light. Hepatic microsomes from either male or female untreated Sprague-Dawley rats, or from rats pretreated with either phenobarbital or dexamethasone, were purchased from In Vitro Technologies Inc. (Baltimore, MD). Additional rat liver microsomes, derived from Sprague-Dawley rats from the National Center for Toxicological Research colony, were prepared in-house by previously published methods. These were used for the initial control experiments. Glucose 6-phosphate, glucose-6-phosphate dehydrogenase, NADP (Na+ salt), NADH, bovine serum albumin, Brij 58, and MgCl2 were purchased from Sigma-Aldrich (St. Louis, MO). All other reagents used were HPLC grade and obtained from J. T. Baker (Phillipsburg, NJ).
Microsomal Incubations. The hyperforin stock solution was prepared freshly before the assay by mixing 40 µl of pure hyperforin methanol solution (25 mg/ml) with 3 ml of 100 mM potassium phosphate buffer (pH 7.4) containing bovine serum albumin (20 mg/ml) and Brij 58 (0.2 mg/ml). Microsomal stock solutions (10 mg/ml) were also prepared freshly by mixing rat liver microsomes (20 mg/ml) with an equal volume of 100 mM potassium phosphate buffer.
For microsomal incubations to prepare hyperforin metabolites, 125 µl of hyperforin stock solution was mixed with 25 µl of microsomal stock solution (10 mg/ml) and 100 µl of coenzyme stock solution containing 3 mM NADP, 1.5 mM NADH, 25 mM glucose 6-phosphate, 5 U/ml glucose-6-phosphate dehydrogenase, and 10 mM MgCl2 in 100 mM potassium phosphate buffer. The final incubation conditions were 66 mM phosphate buffer (pH 7.4), which contained 1.0 mg/ml microsomal protein, 4 mM MgCl2, 10 mM glucose 6-phosphate, 1.2 mM NADP, 0.6 mM NADH, 2 U/ml glucose-6-phosphate dehydrogenase, 10 mg/ml bovine serum albumin, 0.1 mg/ml Brij 58, 0.7% methanol, and 0.167 mg/ml (0.36 mM) hyperforin. The total volume was 0.25 ml. The concentrations of P450s in incubation systems were 0.81, 0.25, 1.93 and 1.39 nmol/ml in male, female, phenobarbital-induced, and dxamethasone-induced microsomal systems, respectively. Incubations were conducted for 60 min at 37°C in a water bath and stopped by adding 0.75 ml of methanol. After the mixture was vortexed for 2 min and centrifuged for 20 min at 6000g, 50-µl aliquots of the supernatant were used for HPLC analysis. Control and kinetic experiments utilized shorter incubation time and, where necessary, heat-inactivated microsomes (80°C for 5min).
HPLC. HPLC analyses were carried out using a Waters liquid chromatography system (Waters, Milford, MA) consisting of a model 600 pump, model 717 autosampler, and model 996 photodiode array detector. The analyses were performed at ambient temperature (23°C) on a Luna C18 column (150 x 4.6 mm, 3-µm particle size) protected by a Luna C18 guard column (4 x 3 mm) (Phenomenex, Torrance, CA). The flow rate was 0.8 ml/min. Mobile phase A consisted of 0.1% acetic acid and 0.2% triethylamine in water (pH 6.5 ± 0.1), and mobile phase B was 100% acetonitrile (ACN). Separations were conducted using a 45-min linear gradient. The proportion of ACN was increased linearly from 50% to 80% over 35 min, then maintained at 80% for 10 min. UV absorbance was monitored at 290 nm.
Liquid Chromatography-Mass Spectrometry. Liquid chromatography/electrospray ionization mass spectrometry (LC/ESI MS) analysis of the microsomal hyperforin metabolism mixture was performed on a Finnigan TSQ 7000 mass spectrometer (Thermo Finnigan, San Jose, CA) equipped with an HP 1100 HPLC apparatus (Agilent Technologies, Palo Alto, CA). The mass spectrometer was operated in negative-ion electrospray mode and full scans were acquired from m/z 50 to 750 at 1 scan/s. Peaks were resolved using a Luna C8 5-µm particle size, 250 x 2.0 mm HPLC column (Phenomenex). The mobile phase was a H2O/ACN gradient with constant 0.1% formic acid and delivered at 0.2 ml/min (50% ACN to 95% ACN over 30 min, then held for 20 min).
Metabolite Isolation for NMR. All metabolites for NMR characterization were isolated from a large-scale phenobarbital-induced rat liver microsomal incubation. The preparative HPLC isolations were performed using a Luna C18 column, 30 x 9.0 mm, 5-µm particle size (Phenomenex). For isolation of metabolites, the isocratic mobile phase consisted of 70% ACN and 30% H2O (containing 0.1% acetic acid and 0.2% triethylamine). One metabolite (M3) was further purified by using isocratic 80% ACN and 20% H2O (containing 0.1% formic acid and 0.1% triethylamine). The flow rate was 3.0 ml/min and UV absorbance was monitored at 290 nm.
NMR Spectroscopy. NMR spectral analyses were performed on a Bruker Avance 600 MHz NMR spectrometer operating with a superconducting HCN probe (Bruker, Newark, DE). Compounds were dissolved in MeOD and all experiments were run at 300 K. Chemical shifts were defined by assigning the MeOD 1H resonance peak to 3.31 ppm and the 13C resonance peak to 69.0 ppm. 1H 1D, 13C 1D, NOESY, COSY, TOCSY, 1H-13C HSQC, and 1H-13C HMBC NMR experiments were run on all samples. The NOESY experiment had a spectral width of 5681.82 Hz, an acquisition time of 180 ms, 853-ms delay time, 400-ms mixing time, and 378 increments of d2 (Kumar et al., 1980). The COSY experiment had a spectral width of 5681.82 Hz, an acquisition time of 180 ms, 1.39-s delay time, and 512 increments of d2 (Aue et al., 1976). The TOCSY experiment had a spectral width of 5681.82 Hz, an acquisition time of 180 ms, 853-ms delay time, 60-ms mixing time, and 256 increments of d2 (Griesinger et al., 1988). The HSQC experiment had a 1H spectral width of 5682 Hz, a 13C spectral width of 25,000 Hz, an acquisition time of 90 ms, 900-ms delay time, and 256 increments of d2 (Vuister et al., 1991). The HMBC experiment had a 1H spectral width of 5682 Hz, a 13C spectral width of 36,231 Hz, an acquisition time of 90 ms, 800-ms delay time, and 256 increments of d2 (Bax et al., 1986).
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
The hydrophobicity of hyperforin was problematic in achieving adequate substrate concentrations. The addition of bovine serum albumin to the substrate stock solutions maintained good suspension of hyperforin with lower detergent and methanol concentrations. Although bovine serum albumin and Brij 58 can have a profound influence on P450 activities, the metabolic efficiency of this system was much higher than that using dimethyl sulfoxide or Tween 80. Methanol concentrations above 1% in the incubation mixture were inhibitory, as were MgCl2 concentrations greater than 5 mM. The metabolism of hyperforin was relatively slow when microsomes from uninduced male rat liver were used but continued at a linear rate for at least 30 min (data not shown). Hyperforin metabolism proceeded much faster with dexamethasone- and phenobarbital-induced microsomes, and although linear rates were only maintained for 10 min, the metabolites continued to accumulate for at least 60 min. The incubation time of 60 min was therefore used to optimize metabolite production.
Typical incubations using different types of rat liver microsomes are shown in Fig. 3. Despite the long incubation time, the percentage of hyperforin transformation with liver microsomes from male rats was only 7.1 ± 1.0%, and was even less with microsomes from female rats (3.8 ± 2.8%). Both phenobarbital and dexamethasone induction enhanced hyperforin metabolism dramatically. The highest metabolic activity was achieved in dexamethasone-induced microsomes, 66.7 ± 3.7%, whereas the activity in the phenobarbital-induced microsomal system was 56.3 ± 6.6%. Supplementing the NADPH-regenerating system with NADH enhanced the rate of hyperforin metabolism, especially in the two induced microsomal systems. The formation of both M1 and M2 was increased by NADH in the dexamethasone-induced system. In the phenobarbital-induced system, the formation of all four metabolites was enhanced by NADH, and the levels of M1, M2, M3, and M4 were similar (Fig. 4). Thus, phenobarbital-induced rat liver microsomes were selected for large-scale incubation. Metabolites were isolated for further characterization by NMR.
|
|
Identification and Characterization of Metabolites
Hyperforin. The negative-ion LC/ESI MS spectrum of hyperforin displayed an intense ion [M-H]– at m/z 535 (Fig. 2). The proton and carbon NMR signals of hyperforin were assigned in CD3OD (Table 1).
|
Metabolites. Negative-ion LC/MS showed the four major metabolites had the same [M-H]– ion at m/z 551 (Fig. 2), 16 Da more than hyperforin. The results suggested that the four compounds were hyperforin with one oxygen added.
The NMR data showed that each of the four metabolites had nine methyl groups and one hydroxymethyl group, suggesting that the four metabolites were hydroxymethyl derivatives. The proton and carbon NMR signals of the four metabolites were also assigned (Tables 2 and 3). All hydroxymethyl groups in the four metabolites were cis to their vinyl protons determined by the observed crosspeaks between the hydroxymethyl protons and the vinyl protons, as in the NOESY spectra; NOESY crosspeaks between H17 (
5.37 ppm) and the hydroxymethylene H19 ( 3.91 ppm) in M1; between H22 ( 5.20 ppm) and the hydroxymethylene H24 ( 3.78 ppm) in M2; between H32 ( 5.22 ppm) and the hydroxymethylene H34 ( 3.76 ppm) in M3; and between H27 ( 5.31 ppm) and the hydroxymethylene H29 ( 3.77 ppm) in M4. The proposed structures are shown in Fig. 5, named as 19-hydroxyhyperforin, 24-hydroxyhyperforin, 29-hydroxyhyperforin, and 34-hydroxyhyperforin.
|
|
|
The other minor metabolites with [M-H]– ions at m/z 551 or 567 (Fig. 2) were probably hyperforin with one or two oxygens added, and need further identification.
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Hyperforin was easily attacked by various oxidants, unstable in many solvents, and sensitive to light (Orth and Schmidt, 2000). The resulting degradation products have been characterized (Trifunovic et al., 1998; Verotta et al., 2002) and are structurally different from the microsomal hydroxylation products reported here.
As a mixture of tautomers, hyperforin showed a broad shape for most 1H NMR signals and the poor resolution of many 13C NMR lines when dissolved in CDCl3, because of its keto-enol equilibrium of the ß-dicarbonyl skeleton. In CDCl3, hyperforin and its metabolites broke down quickly within a few hours at room temperature. In methanol, hyperforin and the four metabolites exist in only one tautomer, are stable, and presented acceptable NMR data in CD3OD. Therefore, proton and carbon NMR signals of hyperforin and its metabolites were successfully assigned in CD3OD by using 1H NMR, 13C NMR, double quantum-filtered COSY, TOCSY, HSQC, HMBC, and NOESY spectra, with very limit amounts (less than 1 mg).
NADH enhanced the metabolic activity of hyperforin metabolism, especially in phenobarbital- or dexamethasone-induced microsomal systems. This observation suggested that cytochrome b5 may be involved in hyperforin metabolism (Estabrook and Cohen, 1969). Therefore, NADH was added to the NADPH-regeneration system for large-scale incubation.
Characterization of the cytochrome P450 isoforms responsible for the metabolism of drugs and herbal constituents is important for the identification of potential drug-drug or drug-herb interactions (Moore et al., 2000; Gurley et al., 2002). In humans, the CYP3A4 isoform plays a major role in metabolism of a wide spectrum of xenobiotics (Roby et al., 2000) and has been shown to be inducible by hyperforin (Moore et al., 2000).
Rats express two major cytochrome P450 isoforms that are oligomerically related to human CYP3A4, CYP3A1 and CYP3A2 (Nelson et al., 1996). Rat liver microsomes were readily available from commercial sources; thus, rat liver microsomes were the first choice in investigation of hyperforin metabolism. It was reported that CYP3A1 is not constitutively expressed in adult rat liver, whereas hepatic CYP3A2 is constitutively expressed in males but not in mature females (Gonzalez et al., 1986; Gonzalez, 1989). Hepatic expression of CYP3A1 is induced by dexamethasone, whereas expression of both isoforms is induced by phenobarbital (Gonzalez, 1989; Waxman et al., 1992; Choudhuri et al., 1995; Joannard et al., 2000). Both isoforms show similar substrate specificity and are highly dependent on cytochrome b5 and NADH-dependent cytochrome b5 reductase of maximal activity (Miyata et al., 1994). It is therefore probable that CYP3A1 and/or CYP3A2 play a major role in hyperforin hydroxylation in rat liver, since we have shown that the rate of hydroxylation is greater in male rat microsomes than in female, inducible by both dexamethasone and phenobarbital, and dependent upon NADH for full activity. The relative differences in metabolite production shown in Fig. 4 could reflect differences in CYP3A1 and CYP3A2 substrate specificity or could suggest a role for other isoforms such as those of the phenobarbital-inducible CYP2B family.
In preliminary experiments we have found that human liver microsomes also metabolize hyperforin at rates similar to those of uninduced rat liver (Y. Cui, unpublished observations). Further work is required to determine whether the hyperforin-inducible CYP3A isoforms are responsible for hyperforin metabolism in human liver and whether the hydroxylated metabolites are pharmacologically active, accumulate in tissues, or are rapidly eliminated.
In summary, hyperforin is apparently metabolized by rat liver microsomes via the hydroxylation pathway. Four major phase I metabolites were identified as 19-hydroxyhyperforin, 24-hydroxyhyperforin, 29-hydroxyhyperforin, and 34-hydroxyhyperforin. The biotransformation of hyperforin may be catalyzed in part by rat CYP3A isoforms.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Abbreviations used are: HPLC/MS, high performance liquid chromatography/mass spectrometry; PDA, photodiode array detection; P450, cytochrome P450; ACN, acetonitrile; LC/ESI MS, liquid chromatography/electrospray ionization mass spectrometry; NOESY, nuclear Overhauser effect spectroscopy; COSY, correlation spectroscopy; TOCSY, total correlation spectroscopy; HSQC, heteronuclear single quantum correlation; HMBC, heteronuclear multiple-bond correlation spectroscopy. 
Address correspondence to: Catharina Y.W. Ang, HFT-230, Division of Chemistry, National Center for Toxicological Research, Food and Drug Administration, 3900 NCTR Road, Jefferson, AR 72079. E-mail: Cang{at}NCTR.FDA.GOV
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. I. Block Integrative Physician's Perspective, With a Special Consideration of St John's Wort Integr Cancer Ther, September 1, 2005; 4(3): 261 - 265. [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
