![[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}
|
|
|
|
|
||
Department of Medicinal Chemistry and Pharmacognosy, University of Illinois at Chicago/National Institutes of Health Center for Botanical Dietary Supplements Research, College of Pharmacy (D.N., Y.L., L.R.C., R.B.v.B.) and Department of Chemistry (S.G.), University of Illinois at Chicago, Chicago, Illinois; and Institut für Organische Chemie, Technische Universität Dresden, Dresden, Germany (P.S., P.M.)
(Received August 20, 2003; accepted October 30, 2003)
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
). Although some studies have reported that hop extracts are not estrogenic (Fenselau and Talalay, 1973), others have found a high estrogenic effect (see Milligan et al., 2002 and references therein). A possible explanation for this discrepancy might be the types of extracts that were evaluated as well as the different assays used to determine estrogenic properties. Because of potentially high exposure of the human population to hops products, we are investigating the estrogenicity of hop extracts and their active compounds.
Among the possible active constituents, 8-prenylnaringenin (8-PN1; see structure in Fig. 1) has attracted considerable interest. This molecule belongs to the prenylated flavanones group, which also includes isoxanthohumol, 6-prenylnaringenin, and a number of diprenylated analogs (Stevens et al., 1997). Various in vitro studies have identified 8-PN as one the most potent phytoestrogens in hops, with a potency equal to or greater than that of other established phytoestrogens such as genistein or coumestrol (Kitaoka et al., 1998; Milligan et al., 1999, 2000; Zierau et al., 2002). Milligan et al. (2002) found that 8-PN possessed in vivo estrogenic activity, although the activity appeared to be much weaker than estradiol. Recently, Adeoya-Osiguwa et al. (2003) reported that 8-PN and various environmental estrogens affect mammalian sperm function much more than estradiol, suggesting that mechanisms other than estrogen receptor binding might be operative. In addition to estrogenicity, other activities such as antifungal properties of 8-PN have also been reported (Tahara et al., 1994).
|
Little is known about the metabolism of 8-PN or other prenylated flavonoids. The few studies on this topic were carried out by Yilmazer et al. (2001a,b), who studied rat liver metabolism of xanthohumol and identified four phase I metabolites as well as two phase II glucuronidation conjugates. 8-PN has been evaluated as an inhibitor of cytochrome P450 isozymes and found to be a potent inhibitor of the CYP1A2 isoform (Henderson et al., 2000; Miranda et al., 2000). In the present study we investigated in vitro the oxidative biotransformation of 8-PN by pooled human liver microsomes. Metabolites were identified by using liquid chromatography-tandem mass spectrometry as well as by comparison with authentic standards.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). Synthesis of trans 8-PN Alcohol (M2) and trans 8-PN Aldehyde (M4). An aqueous solution of 70% tert-butylhydroperoxide (110 µl, 0.42 mmol) and 8-PN (120 mg, 0.35 mmol) dissolved in tetrahydrofuran (2 ml) was added to a suspension of SeO2 (39 mg, 0.35 mmol) in tetrahydrofuran (3 ml), and the resulting solution was stirred for 23 h. The reaction mixture was separated by using silica gel column chromatography eluted with pentane/ethyl acetate (1:1, v/v). After recrystallization from diethyl ether, trans 8-PN alcohol (30 mg, 24%) and trans 8-PN aldehyde (65 mg, 62%) were obtained as pale yellow powders.
5,7-Dihydroxy-8-[(E)-4-hydroxy-3-methyl-2-butenyl]-2-(4-hydroxyphenyl)-2,3-dihydro-4H-chromen-4-one (M2). Rf 0.12 (pentane/ethyl acetate, 1:1); m.p. 203-205°C; infrared (KBr): 3293 (s), 3026 (w), 2919 (s), 1599 (s), 1232 (s), 1068 (s), 828 (m) cm-1; 1H NMR (DMSO, 500 MHz): 
1.51 (s, 1 H, 4''-CH3), 2.72 (dd, 1 H, 3Jcis = 3.0 Hz, 2J = 17.1 Hz, 3-CH2), 3.14 (d, 2 H, J = 7.2 Hz, 1''-CH2), 3.20 (dd, 2 H, 3Jtrans = 12.6 Hz, 2J = 17.1 Hz, 3-CH2), 3.71 (d, 2 H, J = 5.1 Hz, 5''-CH2), 4.61 (t, 1 H, J = 5.6 Hz, 5''-OH), 5.32 (t, 1 H, J = 7.2 Hz, 2''-CH), 5.42 (dd, 1 H, 3Jcis = 2.9 Hz, 3Jtrans = 12.5 Hz, 2-CH), 5.97 (s, 1 H, 6-CH), 6.79 (d, 2 H, J = 8.5 Hz, 3'- and 5'-CH), 7.31 (d, 2 H, J = 8.5 Hz, 2'- and 6'-CH), 9.57 (OH), 10.79 (OH), 12.11 (OH); 13C NMR (DMSO, 75 MHz): 13.52 (4''-CH3), 20.77 (1''-CH2), 41.95 (3-CH2), 66.38 (5''-CH2), 78.25 (2-CH), 95.32 (6-CH), 101.82 and 106.75 (8- and 4a-C), 115.18 (3'- and 5'-CH), 121.52 (2''-CH), 128.08 (2'- and 6'-CH), 129.22 (1'-C), 134.81 (3''-C), 157.59, 159.77, 161.24, and 164.44 (5-, 7-, 8a-, and 4'-C), 196.74 (4-CO). High-resolution negative ion electrospray mass spectrometry [M-H]- m/z 355.1172 (theoretical 355.1182,-2.7 ppm).
(E)-4-[5,7-Dihydroxy-2-(4-hydroxyphenyl)-4-oxo-3,4-dihydro-2H- chromen-8-yl]-2-methyl-2-butenal (M4). Rf 0.46 (pentane/ethyl acetate, 1:1); m.p. 194-198°C; infrared (KBr): 3443 (m), 3286 (s), 3029 (w), 2975 (w), 1628 (s), 1602 (s), 1238 (m), 1072 (s), 824 (m) cm-1; 1H NMR (DMSO, 500 MHz): 1.58 (s, 3 H, 4''-CH3), 2.72 (dd, 1 H, 3Jcis = 3.0 Hz, 2J = 17.1 Hz, 3-CH2), 3.24 (dd, 1 H, 3Jtrans = 12.7 Hz, 2J = 17.1 Hz, 3-CH2), 3.45 (d, 2 H, J = 7.4 Hz, 1''-CH2), 5.45 (dd, 1 H, 3Jcis = 2.9 Hz, 3Jtrans = 12.6 Hz, 2-CH2), 6.02 (s, 1 H, 6-CH), 6.51 (t, 1 H, J = 7.4 Hz, 2''-CH), 6.77 (d, 2 H, J = 8.5 Hz, 3'- and 5'-CH), 7.29 (d, 2 H, J = 8.5 Hz, 2'- and 6'-CH), 9.33 (s, 1 H, 5''-CHO), 9.60 (OH), 11.09 (OH), 12.13 (OH); 13C NMR (DMSO, 75 MHz): 8.84 (4''-CH3), 22.43 (1''-CH2), 41.83 (3-CH2), 78.59 (2-CH), 95.43 (6-CH), 101.89 and 103.70 (4a- and 8-C), 115.23 (3'- and 5'-CH), 128.24 (2'- and 6'-CH), 128.98 (1'-C), 138.11 (3''-C), 152.66 (2''-C), 157.74, 160.24, 161.83, and 164.46 (5-, 7-, 8a-, and 4'-C), 195.72 (5''-CHO), 196.79 (4-CO). High-resolution negative ion electrospray mass spectrometry [M-H]- m/z 353.1020 (theoretical 353.1025,-1.5 ppm).
Human Liver Microsomes. Pooled human liver microsomes from 15 donors were purchased from In Vitro Technologies (Baltimore, MD) and used as supplied. The cytochrome P450 content was 0.17 nmol of P450/mg of protein.
Microsomal Incubations. A typical incubation mixture (0.4 ml) contained 1 mg/ml of microsomal protein, 10 or 50 µM 8-PN, and 1 mM NADPH in 50 mM phosphate buffer, pH 7.4. The reaction was initiated by addition of the NADPH after a 2-min preincubation of the substrate and the microsomal protein. Incubations were carried out for 60 min at 37°C. The reaction was stopped by chilling the mixture on ice followed by addition of 1.6 ml of a cold mixture of acetonitrile/ethanol (1:1, v/v) to precipitate proteins. Samples were centrifuged, and the supernatant was evaporated to dryness under nitrogen. The residue was redissolved in the mobile phase before LC-MS analysis. Control incubations were carried out either without microsomal protein or without NADPH.
Kinetic Studies. For the determination of kinetic parameters, a slightly different procedure was used so that initial rate conditions were obtained. The incubation mixture (0.4 ml) contained 0.25 mg/ml of microsomal protein, 1 mM NADPH, and various substrate concentrations ranging from 0.5 to 200 µM. Incubations were carried out for 10 min. Reactions were stopped, and the samples were processed as described above. Kinetic parameters were calculated by nonlinear regression analysis using GraphPad 3.0 software (GraphPad Software, Inc., San Diego, CA).
LC-MS Analysis. Reversed-phase HPLC separations were carried out using a Discovery 2.1 x 100 mm C18 column (5 µm particle size) (Supelco, Bellefonte, PA) connected to a Waters 2690 solvent delivery system (Waters, Milford, MA). Metabolites were separated using a gradient system consisting of 0.05% acetic acid in water (solvent A) and methanol (solvent B) as follows: 35 to 70% B over 30 min, then 70 to 100% B over 5 min, followed by an isocratic hold at 100% B for another 5 min. The flow rate was 0.2 ml/min. The eluent from the column was introduced into a Micromass (Manchester, UK) Q-TOF-2 hybrid quadrupole/time-of-flight mass spectrometer (Micromass UK Ltd., Manchester, UK) equipped with negative ion electrospray. The resolving power was 5000 full width at half maximum. For exact mass measurements, raffinose ([M-H]- of m/z 503.1612) was introduced postcolumn as a lock mass. The mass accuracy obtained was better than 10 ppm unless noted otherwise. The ion source parameters were as follows: capillary 2.3 kV, cone voltage 30 V, source block temperature 120°C, and drying gas temperature 320°C. Tandem mass spectra were acquired at a quadrupole setting of 12, which corresponds to slightly less than unit resolution of the quadrupole. Tandem mass spectra were acquired at a collision energy of 25 eV using argon as the collision gas at a pressure of 2.0 x 10-5 mBar.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
) and Fabre et al. (2001). The major fragmentation pathway for the substrate 8-PN was retro Diels-Alder (RDA) reaction to produce 1,3A- and 1,3B- ions at m/z 219 and 119, respectively (Fig. 4). This pathway was very useful for the determination of whether metabolic transformation occurred on the A or B ring of the flavanone system. The ion of m/z 219 fragmented further by elimination of CO2 to form the ion of m/z 175. Based on this loss of CO2, Fabre et al. (2001) proposed that the structure of the RDA fragment ion is a lactone instead of the acyclic ketene structure that has been proposed for positive ion RDA fragment ions (Ma et al., 1997). Other minor fragmentation pathways were similar to those described for naringenin (Fabre et al., 2001) and included cleavage of the B ring (m/z 245) as well as loss of ketene and CO2 from the precursor ion. It should be noted that these assignments of fragment ion compositions are supported by exact mass measurements, which were within 20 ppm of the theoretical values.
|
|
|
|
|
|
Metabolites M1 and M2. Metabolites M1 and M2 were the most abundant oxidation products of 8-PN and eluted at the retention times of 15.6 and 16.0 min, respectively (see Fig. 2). Exact mass measurements indicated that M1 and M2 produced deprotonated molecules of m/z 355.1166, which corresponded to an elemental composition of C20H19O6 (-4.4 ppm). Therefore, M1 and M2 were identified as monooxidation products of 8-PN. The product ion mass spectra of these two metabolites were similar and are shown in Fig. 5. The most abundant fragment ion in these tandem mass spectra was detected at m/z 235, which suggested that oxidation took place on the prenyl moiety instead of the phenolic moiety (see proposed fragmentation scheme in Fig. 4). The most informative fragment ion was detected at m/z 325 and corresponded to a loss of formaldehyde (CH2O) from the precursor ion. This loss indicated the presence of a CH2OH group and strongly suggested that oxidation occurred on one of the two terminal methyl groups of the prenyl moiety to produce 8-(4''-hydroxyisopentenyl)naringenin (8-PN alcohol) (see structures of M1 and M2 in Fig. 5). Since the methyl groups were not oxidized to the same extent, we determined which group was preferentially oxidized by comparison with a synthetic standard of trans 8-PN alcohol. Synthetic trans 8-PN alcohol eluted at 16.0 min, which confirmed that the more abundant metabolite M2 was trans 8-PN alcohol and M1 was cis 8-PN alcohol.
Metabolite M3. Eluting at a retention time of 21.2 min (Fig. 2), metabolite M3 produced a deprotonated molecule at m/z 355.1160, which corresponded to a monooxidation product with an elemental composition of C20H19O6 (-6.0 ppm). The product ion tandem mass spectrum of the deprotonated molecule of M3 indicated that oxidation took place on the B ring of the flavanone system due to the formation of a fragment ion of m/z 135. Based on previous studies with naringenin (Nielsen et al., 1998; our unpublished data), hydroxylation occurred at the 3' carbon to form 8-prenyleriodictyiol. It should be noted that the amount of M3 relative to other oxidation products varied with the substrate concentration. At high substrate concentration (50 µM), M3 was almost as abundant as metabolites M1 and M2, but the relative yield of M3 decreased at lower initial substrate concentrations of 8-PN. For example, see the LC-MS chromatograms in Fig. 2 showing the metabolite profile of an incubation with 10 µM 8-PN.
Metabolite M4. Metabolite M4 produced a deprotonated molecule at m/z 353.1011 during negative ion electrospray, which was within 4.0 ppm of the elemental composition C20H17O6. This formula indicated that M4 was formed by introduction of one oxygen atom accompanied by a loss of two hydrogen atoms. Although there are several possible structures that are consistent with these data, the product ion tandem mass spectrum (see Table 1) suggested that this metabolite was formed by oxidation of a terminal methyl group to an aldehyde. This assignment was confirmed by comparison with an authentic synthetic standard of trans 8-PN aldehyde.
Metabolites M5 and M6. Two partially separated metabolites of 8-PN were detected at 
31 min as deprotonated molecules of m/z 337 (see M5 and M6 in Fig. 2). Exact mass measurements revealed that these metabolites were isomers with an elemental composition of C20H18O5 (-8.2 ppm) corresponding to a net loss of two hydrogen atoms from 8-PN. This metabolic transformation might occur by direct dehydrogenation or by oxidation followed by a loss of water. In addition, M5 and M6 eluted later than 8-PN, indicating that these metabolites were less polar. Attempts to obtain better chromatographic separation of M5 and M6 were unsuccessful, which complicated spectral interpretation due to the simultaneous presence of fragment ions from both species. Nevertheless, careful analysis of extracted ion currents for each particular fragment confirmed the presence of two separate metabolites. The negative ion electrospray product ion tandem mass spectrum of the deprotonated molecule M5 (Table 1) contained abundant product ions of m/z 217 (1,3A-) and m/z 119 (1,3B-), suggesting that the B ring was intact. The structure of M5 was determined to be 5,4'-dihydroxy-2'',2''-dimethylpyrano-[7,8: 6'',5'']flavanone based on comparison of retention times and tandem mass spectra with those of an authentic standard (Jang et al., 2002). This compound is analogous to a metabolite of xanthohumol described by Yilmazer et al. (2001b). In contrast to M5, fragment ions of m/z 219 and 117 were detected in the product ion tandem mass spectrum of the deprotonated molecule of M6 (see Table 1). The tandem mass spectrum of M6 indicated a highly stable aromatic molecule due to an abundant [M-H]- ion and the lack of abundant fragment ions. An important diagnostic fragment was detected at m/z 268, which corresponded to loss of a prenyl group that had not been altered during metabolism. Another interesting pair of fragment ions was recorded at m/z 281 and 282. We propose that the radical anion of m/z 282 originated from homolytic cleavage along the C1''-C2'' bond of the prenyl group to form a benzyl radical resonantly stabilized through conjugation to the extended aromatic system. This fragment ion might lose a hydrogen atom to form a stable quinone methide structure. Based on these considerations, we identified M6 as 8-prenylapigenin.
Minor metabolites. Several minor metabolites were detected in the incubation mixture during LC-MS. The abundance of these metabolites was generally very low, especially in incubations with low substrate concentration, but increased relative to other metabolites at higher 8-PN substrate concentrations.
Metabolites M7 through M11. The monooxidation products M7 through M11 were detected eluting between 17 and 20 min. The abundant M7 fragment ion of m/z 235 (see Table 1) indicated that oxidation took place on the prenyl group. Loss of water (m/z 337) from the precursor ion was relatively facile for M7 compared with other metabolites, suggesting the presence of a hydroxyl group near a double bond so that loss of water would form a stable conjugated structure. One such possibility is oxidation at the benzylic carbon of 8-PN. Another interesting feature of the product ion mass spectrum of M7 was a triplet of ions at m/z 283, 284, and 285, which corresponded to losses of 70 to 72 mass units. Since fragments weighing 70 to 72 mass units formed from this hydrocarbon must contain four carbon atoms and one oxygen atom, the additional oxygen added to M7 could not be located on the benzylic carbon. Instead, the most likely structure consistent with the formation of ions of m/z 337, 283, 284, and 285 is 5,7,4'-trihydroxy-8-(2-hydroxy-3-methylbut-3-enyl)flavanone (M7), in which oxygenation occurred at the 2'' carbon and the double bond of the prenyl side chain migrated to the terminal position. This assignment was confirmed by comparison of the HPLC retention time and tandem mass spectra with an authentic standard (Jang et al., 2002). Based on this structure, the fragment ion of m/z 284 may be explained by homolytic cleavage of the C1''-C2'' bond, and the fragment ion of m/z 285 is consistent with charge-initiated cleavage (negative charge located on the hydroxyl group) of the same bond with the elimination of a neutral aldehyde. Cleavage of the C1''-C2'' bond with a hydrogen transfer from the flavanone nucleus to the leaving group would form the fragment ion of m/z 283.
The negative ion electrospray CID product ion tandem mass spectrum of metabolite 3-hydroxy-8-prenylnaringenin (M8) (Table 1) is dominated by the unique fragment ion of m/z 193, which was not an abundant product ion in the tandem mass spectra of any other metabolite of 8-PN. Another pair of diagnostic fragment ions was detected at m/z 124 and 125. Oxidation at the C-3 carbon of the flavanone nucleus of 8-PN would be consistent with these data. For example, the most abundant fragment ion of m/z 193 represented a 1,4A- fragment, which is typically observed for C-3 oxidation products of catechin as noted by Miketova et al. (2000). We have also noted that this type of fragmentation is dominant for C-3-hydroxylated flavanones such as pinobanksin and aromadendrin (unpublished data). The fragment ions of m/z 124 and 125 represented elimination of the complete prenyl group (C5H9) and an isoprene unit (C5 H8), respectively, from the 1,4A- ion.
The product ion tandem mass spectrum of the deprotonated molecule of M9 (Fig. 6) shows an unusual loss of 110 units from the precursor ion to form a base peak at m/z 245. Exact mass measurements indicated that the molecular formula of the leaving group was C6H6O2. In addition, exact mass measurements showed that the related fragment ion of m/z 109 corresponded to an elemental composition of C6H5O2, and the loss of a hydrogen atom from this ion resulted in formation of a radical anion of m/z 108 (C6H4O2). The absence of an ion at m/z 235 indicates that oxidation did not take place on the prenyl group. Furthermore, the lack of an ion at m/z 135 showed that M9 did not contain a catechol moiety on the B ring. Oxygenation on the C ring could also be excluded, since loss of a group weighing 110 mass units would not be feasible from such a metabolite. Therefore, the most probable structure for M9 is a B-ring hydroxyl quinol (see structures in Fig. 6). This structure is consistent with all of the abundant fragment ions discussed above. In addition, these fragmentation pathways are essentially the same as those observed for quinol products of other 4'-hydroxylated flavanones (unpublished data).
The negative ion electrospray tandem mass spectra of metabolites M10 and M11, which eluted at 18.1 and 19.0 min, respectively, in the LC-MS chromatogram shown in Fig. 2 were dominated by 1,3A- and 1,3B- ions at m/z 235 and 119, respectively (data not shown). Although low in abundance, the ion of m/z 297 in the product ion mass spectrum of M10 was structurally significant and represented loss of an acetone moiety from the deprotonated molecule. This implied the presence of a tertiary alcohol. Based on tandem mass spectra, HPLC retention time, and comparison with an authentic standard, M10 was identified as 5,4'-dihydroxy-[2''-(1-hydroxy-1-methylethyl)dihydrofurano]-[5'',4'':7,8]flavanone. This metabolite is analogous to a xanthohumol metabolite described by Yilmazer et al., (2001b). The product ion mass spectrum of M11 was almost identical to that of M10. Although M11 might be an analog of M10 with the hydroxyl group in the 2'' position (M1 in Yilmazer et al., 2001b) or a diastereoisomer of M10, the structure of M11 could not be determined unambiguously.
Metabolite M12. Metabolite M12 eluted at a retention time of 28 min (Fig. 2) and produced a deprotonated molecule of m/z 245.0827, which was within 5.4 ppm of the molecular formula C14H13O4. This formula is consistent with the structure of 8-prenylchromone, which might be formed by elimination of the aromatic B ring during metabolism. This type of metabolite was also detected in our previous studies with other 4'-hydroxylated flavanones (unpublished data). The negative ion electrospray product ion tandem mass spectrum of the deprotonated molecule of M12 is shown in Fig. 6. The base peak in this tandem mass spectrum at m/z 190 was probably formed by homolytic cleavage of the C1''-C2'' bond of the prenyl group similar to that observed for M6 (see structures in Fig. 6). However, the stability of the ion of m/z 190 suggested that it was aromatic. The other significant but much less abundant ions of m/z 162 and 134 represented consecutive eliminations of molecules of carbon monoxide, which is a characteristic feature of chromone derivatives. Based on these considerations, metabolite M12 was tentatively identified as 8-prenylchromone. Structurally related to 8-prenylchromone, two peaks of low intensity were detected eluting at 9.1 and 9.2 min, which produced deprotonated molecules of m/z 261 (data not shown). Although the structures of these trace metabolites were not investigated further, they were probably monooxygenated derivatives of 8-prenychromone with the additional oxygen atom located on one of the other terminal methyls of the prenyl group.
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
|
The most abundant human liver microsomal metabolites of 8-PN were oxidation products of the prenyl group. Among these, hydroxylation of one of the terminal methyls of the prenyl group (M1 and M2) was the most abundant pathway. The formation of these metabolites followed Michaelis-Menten kinetics as shown in Table 2 and may be rationalized in terms of classical hydrogen abstraction/oxygen-rebound mechanism as shown in Fig. 7. Among the two possible sites of transformation, the trans product (M2) was more abundant and is probably the more stable isomer. The predominance of this metabolic pathway is not surprising given the favorable electronic and steric considerations for this site of attack. Hydrogen abstraction at a terminal methyl group is favorable due to formation of a stable allyl radical, and these groups are relatively exposed and thus accessible to enzymatic attack. Steric hindrance is probably responsible for the lack of formation of a benzylic oxidation product, whose resulting benzyl radical would be stabilized by delocalization of the unpaired electron into the phenyl ring. However, the benzylic protons were probably not readily accessible to the cytochrome P450 enzyme. This finding might be significant in terms of preventing toxicity, since benzylic oxidation has been shown to be an initial step in metabolic activation of various allyl-substituted aromatic molecules such as safrole.
|
Another interesting observation was the detection of a double-bond migration product (M7). We propose that this product was formed due to delocalization of the unpaired electron onto the other carbon of the allylic system. This type of transformation has also been observed for other allylic systems such as eugenol (Fischer et al., 1990).
The prenyl methyl group oxidation reaction was a predominant pathway at low substrate concentration and was proportionally less dominant at higher substrate concentration. This finding suggests that it is useful to investigate xenobiotic metabolism at several substrate concentrations to obtain a more complete assessment of the biological relevance of the particular metabolic pathway. In particular, many studies of the metabolism of natural products including the metabolism of xanthohumol by Yilmazer et al. (2001b) used relatively high substrate concentrations (100 µM), which might not be biologically relevant since this concentration is unlikely to be achieved in vivo. It is interesting to note that Yilmazer et al. (2001b) did not observe oxidation of prenyl methyl groups when investigating the metabolism of xanthohumol in the rat. Species differences cannot account for this discrepancy since we also observed these metabolites using rat liver microsomes (data not shown). However, it is conceivable that this pathway did not occur for the more rigid, open structure of xanthohumol, or that it was such a minor metabolite that it was undetectable using the analytical methodology used by the authors.
Biological activities of 8-PN alcohols are unknown at this time. However, it is very interesting to note that the isomeric product M7 has been found to be a moderately potent but highly selective inhibitor of human cyclooxygenase-2 (Jang et al., 2002). This finding illustrates an important aspect of careful study of metabolism of natural products; that is, a possibility that metabolic reaction can result not only in activation/deactivation but also in formation of completely new pharmacological activity, which can complicate in vivo extrapolation of in vitro findings.
In addition to the 8-PN alcohols, 8-PN aldehyde was detected and was probably formed by further oxidation of the primary alcohol. It is interesting to note that we observed only the trans isomer of the aldehyde. Perhaps the trans isomer is much more stable than the cis product so that interconversion occurred during sample preparation. Another explanation might be that the cis isomer is more prone to cyclization due to more favorable orientation of the reactive groups. Further oxidation of the aldehyde to a carboxylic acid was not observed, probably because the responsible enzyme(s) is(are) cytosolic instead of microsomal.
Another type of metabolic reaction of the prenyl group was epoxidation of the double bond. This metabolic pathway was found to be the major route for prenyl group metabolism in xanthohumol (Yilmazer et al., 2001b). Although no intact epoxide metabolite was detected, various cyclization products were formed by attack of the epoxide intermediate by a neighboring hydroxyl group (Fig. 7B). Hydroxyl group attack might occur at either of two positions to form a five- or a six-member ring. However, the six-member cyclic-hydroxylated product was not detected, probably due to the ease of dehydration to form stable conjugated structures such as M5. This dehydration pathway was also observed by Yilmazer et al. (2001b). These findings are also of potential toxicological significance since epoxide intermediates are often considered potentially toxic due to their reactivity toward biological nucleophiles. The presence of an ortho hydroxyl group, a very common structural feature in prenylated flavonoids, can deactivate this electrophilic epoxide intermediate through a rapid intramolecular reaction.
The second group of metabolites of 8-PN formed by human liver microsomes consisted of derivatives of the flavanone skeleton. For example, aromatic hydroxylation of the B ring produced metabolite M3. This type of transformation is a general route for 4'-hydroxylated flavonoids and has been reported by several groups (Nielsen et al., 1998; Yilmazer et al., 2001b; Nikolic and van Breemen, manuscript submitted for publication). B-ring hydroxylation was concentration-dependent since we found that at high substrate concentration, M3 was almost as abundant as M1 and M2. This is a potentially important finding since catechols are often considered potentially toxic intermediates due to their propensity to form reactive ortho quinones.
The unusual oxidation product M9 produced a fragmentation pattern that was similar to a similar metabolite of naringenin (unpublished data), which facilitated the identification of this metabolite as a hydroxy quinone (quinol) formed by ipso hydroxylation of the B ring. Ohe et al. (2000) reported this type of transformation for estrogens as a C10-hydroxylation product. We have also observed this kind of metabolite in our studies with other 4'-hydroxylated flavanones. M9 was probably an intermediate metabolite in the formation of the B-ring cleavage product, 8-prenylchromone (M12). Since we have also observed similar metabolites of naringenin (data not shown), it appears that the prenyl group is not essential for the formation of the unusual metabolites M9 and M12, and this metabolic pathway might be common for this class of flavonoids. The biological significance of these 8-PN metabolites is unclear at this time since the biological effects of these products have not been established.
A metabolite of 8-PN with the potential for enhanced estrogenicity was the C-ring desaturation product, 8-prenylapigenin (M6). The introduction of the new double bond into the C ring should increase the structural rigidity of the molecule. Interestingly, however, Kitaoka et al. (1998) found that the affinity of this compound for estrogen receptor is approximately the same as that of 8-PN.
Human exposure to 8-PN occurs primarily through the consumption of hop-derived products such as beer and an increasing number of dietary supplements containing hop extracts. Dietary exposure to hops through beer is relatively low and probably insignificant for most of the population (Tekel' et al., 1999). However, ingestion of dietary supplements containing hops or hops extracts is of more concern due to exposure to higher doses of the estrogenic 8-PN and its metabolites. Our study indicates that 8-PN is transformed into a wide array of metabolites, and some of these products might have different pharmacological activity than their precursors. It would be of interest to determine the biological activities and potential toxic effects of these metabolites.
|
Footnotes |
|---|
1 Abbreviations used are: 8-PN, 8-prenylnaringenin; HPLC, high-performance liquid chromatography; LC-MS, liquid chromatography-mass spectrometry; M5, 5,4'-dihydroxy-2'',2''-dimethylpyrano-[7,8:6'',5'']flavanone; M7, 5,7,4'-trihydroxy- 8-(2-hydroxy-3-methylbut-3-enyl)flavanone; M10, 5,4'-dihydroxy-[2''-(1-hydroxy- 1-methylethyl)dihydrofurano]-[5'',4'':7,8]flavanone; M2, 5,7-dihydroxy-8-[(E)-4-hydroxy-3-methyl-2-butenyl]-2-(4-hydroxyphenyl)-2,3-dihydro-4H-chromen-4-one; DMSO, dimethyl sulfoxide; M4, (E)-4-[5,7-dihydroxy-2-(4-hydroxyphenyl)-4-oxo- 3,4-dihydro-2H-chromen-8-yl]-2-methyl-2-butenal; CID, collision-induced dissociation; RDA, retro Diels-Alder; 8-PN alcohol, 8-(4''-hydroxyisopentenyl)naringenin; M6, 8-prenylapigenin; M12, 8-prenylchromone. 
Address correspondence to: Richard B. van Breemen, Department of Medicinal Chemistry and Pharmacognosy, College of Pharmacy, University of Illinois at Chicago, 833 S. Wood St., Chicago, IL 60612-7231. E-mail: breemen{at}uic.edu
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Minato, R. Suzuki, A. Asagarasu, T. Matsui, and M. Sato Biotransformation of 3-Amino-5,6,7,8-tetrahydro-2-{4-[4-(quinolin-2-yl)piperazin-1-yl]butyl}quinazolin-4(3H)-one (TZB-30878), a Novel 5-Hydroxytryptamine (5-HT)1A Agonist/5-HT3 Antagonist, in Human Hepatic Cytochrome P450 Enzymes Drug Metab. Dispos., May 1, 2008; 36(5): 831 - 840. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Guo, D. Nikolic, L. R. Chadwick, G. F. Pauli, and R. B. van Breemen IDENTIFICATION OF HUMAN HEPATIC CYTOCHROME P450 ENZYMES INVOLVED IN THE METABOLISM OF 8-PRENYLNARINGENIN AND ISOXANTHOHUMOL FROM HOPS (HUMULUS LUPULUS L.) Drug Metab. Dispos., July 1, 2006; 34(7): 1152 - 1159. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Nikolic and R. B. van Breemen NEW METABOLIC PATHWAYS FOR FLAVANONES CATALYZED BY RAT LIVER MICROSOMES Drug Metab. Dispos., April 1, 2004; 32(4): 387 - 397. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
