Inhibition of Peroxynitrite-Mediated LDL Oxidation by Prenylated

Chem. Res. Toxicol. 2003, 16, 1277-1286

1277

Inhibition of Peroxynitrite-Mediated LDL Oxidation by Prenylated Flavonoids: The r,β-Unsaturated Keto Functionality of 2′-Hydroxychalcones as a Novel Antioxidant Pharmacophore Jan F. Stevens,†,‡ Cristobal L. Miranda,§ Balz Frei,‡ and Donald R. Buhler*,§ Department of Chemistry, Department of Environmental and Molecular Toxicology, and the Linus Pauling Institute, Oregon State University, Corvallis, Oregon 97331-7301 Received October 24, 2002

Prenylated 2′-hydroxychalcones and flavanones from the inflorescences of the female hop plant (Humulus lupulus) were shown to inhibit peroxynitrite-mediated oxidation of low-density lipoproteins (LDL) at low micromolar concentrations. LDL oxidation was induced by the peroxynitrite generator, 3-morpholinosydnonimine (SIN-1), and measured by the formation of conjugated dienes and thiobarbituric reactive substances. Human intake of prenylated chalcones and flavanones is mainly through beer, which contains up to 4 mg/L of these polyphenols. The two main oxidation products obtained by SIN-1 and peroxynitrite treatment of xanthohumol (XN), the principal prenylflavonoid of hops, were the aurone, auroxanthohumol (AUXN), and an endoperoxy derivative of XN, named endoperoxyxanthohumol (EPOX). In addition, the reaction produced smaller amounts of the nitro and nitroso derivatives of XN and EPOX. The formation of these nitrated products was enhanced in the presence of sodium bicarbonate (25 mM). SIN-1-induced formation of AUXN is considered to be a superoxide-mediated reaction, while the structure of EPOX points to a two electron oxidation reaction involving a Michael type addition with peroxynitrite as the nucleophile, followed by cyclization yielding a (1,2)dioxepin-5-one ring structure. The flavanone isomer of XN, isoxanthohumol (IsoXN), unexpectedly showed a slight prooxidant effect instead of an inhibitory effect on LDL oxidation. Except for the formation of minor nitrated products, IsoXN remained largely unmodified upon treatment with SIN-1/peroxynitrite. Taken together, our results suggest that the R,βunsaturated keto functionality of chalcones is most reactive toward superoxide and peroxynitrite anions.

Introduction Reactive oxygen and nitrogen species are thought to play a role in the development of cancer, neurodegenerative diseases, and cardiovascular diseases such as atherosclerosis. These reactive species may cause oxidative modification of LDL,1 which is often associated with the initiation of atherosclerosis (1-3). Circulating plasma LDL can be oxidized by various oxidants, including peroxynitrite (1, 4), and it is protected against oxidation by water soluble antioxidants present in blood (5, 6). Peroxynitrite is a strong biological oxidant formed from the reaction of superoxide radical anions with nitric oxide (O2•- + NO• f ONOO-) (7). Endothelial cells and phagocytic cells (macrophages) are capable of simultaneously * To whom correspondence should be addressed. Tel: (541)737-1784. Fax: (541)737-0497. † Department of Chemistry. ‡ Linus Pauling Institute. § Department of Environmental and Molecular Toxicology. 1 Abbreviations: DAD, (photo)diode array detection; ESI-MS, electrospray ionization mass spectrometry; FTICR-MS, Fourier transform ion cyclotron resonance mass spectrometry; HMBC, heteronuclear multiple bond correlation, LDL, low-density lipoprotein; NOESY, nuclear Overhauser enhancement spectroscopy; SIN-1, 3-morpholinosydnonimine; TBARS, thiobarbituric acid reactive substances. General remarks: For compound acronyms and structures, see Tables 1 and 3 and Figure 1. Please note that atom numbering in chalcones and flavanones is different, cf. Figure 1. To avoid confusion, the chalcone atom numbering system was also used for AUXN and EPOX.

generating nitric oxide and superoxide anions, while neutrophils and macrophages from the lung (alveolar) and liver (Kupfer cells) can be stimulated by inflammation and infection to produce large amounts of peroxynitrite. Aside from oxidizing LDL, peroxynitrite can oxidize carbohydrates, proteins, nucleic acids, sulfhydryl compounds, and phenols. It can also mediate nitration and nitrosation reactions. The interaction of peroxynitrite with cellular components may lead to extensive cell damage, impaired function, and alterations in signaling pathways (8-10). Several epidemiological studies have suggested a relation between the consumption of foods rich in flavonoids and a reduced risk of coronary heart diseases (11-13), presumably due to the antioxidant and free radical scavenging properties of flavonoids. Flavonoids constitute the largest group of plant polyphenols with over 4000 representatives (14). They are ubiquitously distributed in green land plants, and estimates of their total daily intake through vegetable foods range from a few to several hundreds of milligrams, depending on dietary habits (15). While flavonols, flavones, anthocyanins, and catechins are the main flavonoid compounds in fruits and vegetables, little is known about the daily dietary intake of chalcones and flavanones, the key intermediates in the biosynthesis of flavonoids. Moreover, prenylation of flavonoids is relatively uncommon in the plant kingdom

10.1021/tx020100d CCC: $25.00 © 2003 American Chemical Society Published on Web 09/13/2003

1278 Chem. Res. Toxicol., Vol. 16, No. 10, 2003

with only three superorders (Urticales, Fabales, and Asterales) comprising over 80% of all prenylated flavonoids reported in the literature (16). In this study, we report on the inhibition of LDL oxidation by prenylated chalcones and flavanones from the inflorescences of the female hop plant (Humulus lupulus) that are used in the brewing industry to impart flavor and bitterness to beer. Human intake of prenylflavonoids is mainly through consumption of beer, which contains up to 4 mg/L of total prenylflavonoids, depending on hopping rate and brewery conditions (17, 18). In this study, we will show that prenylated 2′-hydroxychalcones are potent inhibitors of SIN-1-induced oxidation of LDL. We propose that the mechanism of peroxynitrite scavenging by 2′-hydroxychalcones does not involve radical intermediates as generally assumed for other classes of flavonoids (19, 20).

Materials and Methods Materials. Prenylated flavonoids (see Figure 1 for structures) were obtained by isolation from the inflorescences of female hop plants (H. lupulus) or by semisynthesis as described earlier (18, 21, 22). The purity of the flavonoids was 95+% by HPLC and 1H NMR analysis. LDL was prepared from the plasma of healthy volunteers using the method of Chung et al. (23) as modified by Sattler et al. (24). SIN-1 was purchased from Sigma. Peroxynitrite (ONOO-, 170-200 mM) was from Calbiochem (San Diego, CA) and was stored at -80 °C. In later experiments, peroxynitrite (108 mM) was synthesized chemically by the reaction of acidified NaNO2 (0.6 M) with H2O2 (0.6 M) in a sodium hydroxide-quenched flow setup consisting of two syringes connected to a reaction tube via a T-connection. The peroxynitrite concentration was determined spectrophotometrically (302 ) 1670 M-1 cm-1). HPLC Conditions. Analytical HPLC was carried out with a 5 µm C-18 LiChrospher column (250 mm × 4.6 mm; Merck, Darmstadt, Germany) using a linear solvent gradient from 40 to 100% MeCN in 1% aqueous formic acid over 30 min at 1.0 mL/min. The column was maintained at 30 °C using a column thermostat. On-line UV spectra of analytes were recorded with a DAD (Merck, Hitachi L-4500A) (Table 3). Preparative HPLC separations were achieved on a 10 µm Econosil RP-18 (250 mm × 22 mm) column (Alltech, Deerfield, IL). Linear gradient elution was used starting from 40 to 100% MeCN in 1% aqueous formic acid over 30 min at a flow rate of 10 mL/min. Semipreparative HPLC was carried out with an 10 µm LiChrospher RP-18 column (250 mm × 10 mm; Merck) using the same linear solvent gradient at a flow rate of 5 mL/min. The UV trace was recorded at 350, 370, or 395 nm. Peak fractions were collected manually, concentrated on a rotavapor, and then lyophilized. MS. Low-resolution ESI mass spectra were recorded in the positive ion mode on a Finnigan MAT TSQ 7000 triplequadrupole instrument using a spray voltage of 5 kV with the heated capillary set at 200 °C (Table 3). Samples were dissolved in MeCN-H2O (1:1) and introduced into the mass spectrometer by continuous flow injection at 5 µL/min using a syringe pump. MS-MS spectra were obtained by collision-induced dissociation using argon as the collision gas at a pressure of ca. 1.8 × 10-3 Torr. The collision energy was 25 eV. High-resolution ESI mass spectra were recorded on an APEX II FT-ICR mass spectrometer (Bruker Daltonics) equipped with a 7T superconducting magnet. Samples were introduced by continuous flow injection at a flow rate of 5 µL/min using a syringe pump. LC-MS/MS experiments were carried out on a Sciex APIIII+ triple-quadrupole mass spectrometer equipped with an atmospheric pressure chemical ionization source. Samples were introduced by HPLC via a heated nebulizer interface kept at 450 °C. Ionization of the analyte vapor mixture was initiated by a corona discharge needle at ca. 5 kV in the positive ion mode. The orifice plate voltage was 60 V. MS-MS experiments were

Stevens et al. performed with Ar-N2 (9:1) as the target gas at a thickness of ca. 1.9 × 1014 atoms/cm2. The collision energy was 20 V. LCMS/MS experiments using multiple reaction monitoring were carried out by detection of fragment ions formed by collisioninduced dissociation of the following molecular ions: m/z 353 (molecular ion) f 297 (fragment ion) for auroxanthohumol (AUXN); m/z 355 f 179 for xanthohumol (XN); m/z 369 f 313 for endoperoxyxanthohumol (EPOX); m/z 398 f 342 for 3-nitroso-endoperoxyxanthohumol (NOEPOX); m/z 400 f 179 for 3-nitro-xanthohumol (NO2XN); and m/z 414 f 358 for 3-nitroendoperoxyxanthohumol (NO2EPOX). HPLC separations were achieved on a Phenomenex C18 column (4 mm × 250 mm) using the same linear solvent gradient and flow rate as mentioned under HPLC Conditions. NMR Spectroscopy. NMR spectra were run on a Varian UNITY instrument at 500 MHz (1H) and 125 MHz (13C) in DMSO-d6 at room temperature. The solvent resonances (δH2.50 and δC39.51) were used as internal shift references. Twodimensional experiments (1H-1H COSY, 1H-13C HSQC and HMBC) were carried out using standard Varian pulse sequences. Lipid Peroxidation Induced by SIN-1 in LDL. Conjugated diene formation, which represents the initial stage of lipid peroxidation, and formation of TBARS, the end products of lipid peroxidation, were used as end points to determine the antioxidant and prooxidant activities of prenylated and nonprenylated flavonoids. LDL (final concentration of 0.1 mg/mL) was incubated at 25 °C with the individual flavonoids and SIN-1 (40 µM) in 50 mM phosphate buffer (pH 7.4) in a total volume of 200 µL/well on a 96 well plate. Conjugated diene formation was monitored by recording the absorbance at 250 nm in SpectraMax 250 every 30 min for 8 h and then at 18 and 24 h. After a 24 h incubation at 25 °C, TBARS were measured as follows. The contents of each well of the 96 well plate were transferred to 1.5 mL microcentrifuge tubes. Twenty microliters of ice-cold 50% trichloroacetic acid was added to the tubes. After the tubes were vortexed, 100 µL of 1% thiobarbituric acid in 0.28% NaOH was added to each tube, followed by the addition of 20 µL of 1 N HCl. The tubes were then heated at 90 °C for 20 min and centrifuged at 16 000g for 10 min, and aliquots (200 µL) of the supernatant were transferred to a 96 well plate. TBARS were then measured as the absorbance of the samples at 535 nm using a microplate reader (SpectraMax 250). Fate of XN in the LDL Incubations. Similar to the incubations described above, LDL (final concentration of 0.1 mg/ mL in 0.1 M sodium phosphate buffer, pH 7.4) was oxidized by peroxynitrite (50 µM) or SIN-1 (80 µM) in the presence of XN (25 µM). The open tubes were shaken in a Dubnoff metabolic shaker for 1 h at 37 °C. A 0.5 mL aliquot of the incubation mixture (0.5 mL) was mixed with 0.5 mL of methanol, and the tubes were placed on ice water. After 30 min, the incubation mixtures were centrifuged for 10 min at 14 000 rpm (4 °C). The supernatants (0.95 mL) were acidified with 50 µL of formic acid prior to LC-MS/MS analysis. Control experiments were carried out in the absence of LDL and the oxidant, peroxynitrite or SIN1. Reaction of XN and Isoxanthohumol (IsoXN) with SIN1. A solution of XN or IsoXN (200 µL, 10 mM in MeCN) was mixed with MeCN (200 µL) and sodium phosphate buffer (800 µL, 0.1 M, pH 7.4). To this solution was added 20 µL of SIN-1 (200 mM in water), and the reaction mixture was stirred for 1 h in an open microcentrifuge tube at 37 °C. After this period, 0.5 mL of the reaction mixture was mixed with 0.45 mL of water and 50 µL of formic acid. The reaction was carried out on a larger scale in order to obtain sufficient amounts of oxidation products for NMR analysis. To a solution of XN (60 mg, 0.17 mmol) in 30 mL of MeCN was added a solution of SIN-1 (35 mg, 0.17 mmol) in 30 mL of Na-phosphate buffer (0.1 M, pH 7.4). The reaction mixture was stirred under air for 4 h at 37 °C. After this period, MeCN was largely removed by rotary evaporation, and the aqueous layer was extracted with EtOAc (3 × 30 mL). The combined

Antioxidant Chemistry of Prenylated Flavonoids

Chem. Res. Toxicol., Vol. 16, No. 10, 2003 1279

Figure 1. (A) Structures of chalcones. For compound acronyms, compare with Table 1. (B) Structures of flavanones, genistein (GS), and quercetin (QC). For compound acronyms, compare with Table 1. EtOAc layers were evaporated, and the residue was subjected to preparative HPLC (UV 370 nm). Product peaks were collected and taken to dryness by rotary evaporation and lyophilization. The products obtained were AUXN (5.0 mg) and EPOX (0.5 mg). AUXN was shown to consist of two isomers by 1H NMR. Semipreparative HPLC of the mixture using the same condi-

tions yielded (Z)-auroxanthohumol ((Z)-AUXN; 3.7 mg) and (E)auroxanthohumol ((E)-AUXN; 1.2 mg). Reaction of XN and IsoXN with Peroxynitrite. A solution of XN or IsoXN (200 µL, 10 mM in MeCN) was mixed with MeCN (200 µL) and sodium phosphate buffer (800 µL, 0.1 M, pH 7.4). To this solution was added 37 µL of a solution of

1280 Chem. Res. Toxicol., Vol. 16, No. 10, 2003 peroxynitrite (108 mM). The reaction mixture was stirred for 1 h at 37 °C, and then, 0.5 mL of the reaction mixture was mixed with 0.45 mL of water and 50 µL of formic acid. Product formation was monitored by LC-MS. The reaction was carried out on a larger scale in order to obtain sufficient amounts of products for NMR analysis. A solution of XN (35 mg, 0.1 mmol) in MeCN (5 mL) was mixed with 5 mL of sodium phosphate buffer (0.1 M, pH 7.4) with stirring. Peroxynitrite (0.7 mL of a 200 mM solution) was added, and the reaction mixture was stirred for 24 h at 37 °C. After this period, 2 mL portions were acidified with 0.1 mL of formic acid and subjected to preparative HPLC (UV 350 nm). EPOX (10.4 mg) was isolated as the major product. Small amounts of 3-nitro-EPOX (0.4 mg) and nitroso-EPOX (0.4 mg) were also isolated. 3-Nitro-EPOX. 1H NMR: δH 7.79 (1H, br s, H-2), 7.46 (1H, d, J ) 8 Hz, H-6), 7.32 (1H, s, H-R), 7.05 (1H, d, J could not be determined, H-5), 6.21 (1H, s, H-5′), 5.16 (1H, t, J ) 6 Hz, H-2′′), 3.81 (3H, s, OMe), 3.3-3.1 (H-1′′ doublet obscured by the water signal), 1.74, and 1.64 (each 3H, s, H-4′′ and H-5′′). In the 1H spectrum of nitroso-EPOX, only signals belonging to the A-ring could be distinguished as follows: δH 6.62 (1H, s, H-5′), 5.30 (1H, t, J ) 7 Hz, H-2′′), 3.82 (3H, s, OMe), 3.3-3.1 (H-1′′ doublet obscured by the water peak), 1.76, and 1.65 (each 3H, s, H-4′′ and H-5′′). Reaction of XN and IsoXN with Peroxynitrite in the Presence of Sodium Bicarbonate (25 mM). A solution of XN or IsoXN (200 µL, 10 mM in MeCN) was mixed with MeCN (200 µL) and sodium phosphate buffer (680 µL, 0.1 M, pH 7.4). To this solution was added 120 µL of sodium bicarbonate (250 mM) and 37 µL of a solution of peroxynitrite (108 mM). The reaction mixture was stirred for 1 h at 37 °C in an open microcentrifuge tube, and then, 0.5 mL of the reaction mixture was mixed with 0.45 mL of water and 50 µL of formic acid. Control experiments were conducted in the absence of sodium bicarbonate as described above. Product formation was monitored by LC-MS, and the relative quantities were calculated from peak area determinations using BioMultiview 1.3. software (PE Sciex).

Results Exposure of LDL to NO• and O2•-, liberated from SIN-1 in solution (Scheme 1), leads to lipid peroxidation as demonstrated by the steady increase of conjugated diene formation up to 8 h of reaction (Figure 2). SIN-1-induced oxidation of LDL was inhibited by XN in a dose-dependent manner in the assay for conjugated diene formation (Figure 2A) and in the TBARS assay (Figure 2B), which measures the formation of electrophilic 2,4-dienone and epoxide moieties in oxidized unsaturated fatty acid residues by reaction with thiobarbituric acid as the nucleophile. When SIN-1-induced oxidation of LDL was repeated in the presence of a series of prenylated and nonprenylated flavonoids at 5 µM, strong inhibition was

Stevens et al.

Figure 2. Effects of various concentrations of XN on human LDL oxidation induced by SIN-1 (40 µM) as measured by the formation of conjugated dienes (A) or by the formation of TBARS (B). (A) Open squares, ethanol; closed squares, 2.5 µM XN; open circles, 5 µM; closed circles, 12.5 µM; and closed triangles, 25 µM XN. (B) Bars marked with an asterisk are significantly different from controls, p < 0.05, by a two-sided Student’s t-test. The control value for TBARS in panel B was 2.98 ( 0.15 µM.

observed for XN and related chalcones (4′,6′-dimethoxy2′,4-dihydroxychalcone (MCN), 2′,4′,6′,4-tetrahydroxy-3′C-prenylchalcone (TP), xanthogalenol (XG), 5′-prenylxanthohumol (PX), dehydrocycloxanthohumol (DX), and dehydrocycloxanthohumol hydrate (DH)) while the nonprenylated chalcone, chalconaringenin (CN), and the highly lipophilic prenylchalcones, 4′-O-5′-C-diprenylxanthohumol (DPX) and 2′,4′,6′,4-tetrahydroxy-3′-geranylchalcone (TG), were considerably less active as antioxidants (Table 1). Flavanones showed weak antioxidant activities as well, but within this group, the effect of

Scheme 1. Decomposition of SIN-1 in Solution: Liberation of Equimolar Amounts of O2•- and NO•



Table 1. Effect of Chalcones and Flavanones on Lipid Peroxidation (Measured as TBARS) Induced by SIN-1 in Human LDLa addition to reaction mixture

TBARS (% of control)

chalcones CN (chalconaringenin) MCN (4′,6′-dimethoxy-2′,4-dihydroxychalcone) TP (2′,4′,6′4-tetrahydroxy-3′-C-prenylchalcone) XG (xanthogalenol) XN (xanthohumol) PX (5′-prenylxanthohumol) DPX (4′-O-5′-C-diprenylxanthohumol) MXN (4′-O-methylxanthohumol) TX (tetrahydroxanthohumol) TG (2′,4′,6′4-tetrahydorxy-3′-geranylchalcone) DX (dehydrocycloxanthohumol) DH (dehydrocycloxanthohumol hydrate)

75 ( 2b 27 ( 1b 42 ( 5b 30 ( 3b 24 ( 5b 31 ( 2b 88 ( 2b 51 ( 2b 92 ( 2b 64 ( 1b 32 ( 1b 37 ( 2b

flavanones NG (naringenin) 6PN (6-prenylnaringenin) 8PN (8-prenylnaringenin) DPN (6,8-diprenylnaringenin) 6GN (6-geranylnaringenin) 8GN (8-geranylnaringenin) IsoXN (isoxanthohumol) isoflavone GS (genistein) flavonol QC (quercetin)

97 ( 7 78 ( 1b 67 ( 3b 85 ( 2b 84 ( 3b 68 ( 3b 107 ( 4b

Figure 3. HPLC analysis of oxidation products formed by treatment of XN with SIN-1 (A and B) and peroxynitrite (C) after 1 h of reaction. Key to peaks: 1, XN; 2, EPOX; 3Z, (Z)AUXN; 4, 3-nitrosoEPOX; 6, 3-nitroEPOX.

86 ( 2b 11 ( 3b

a GS (an isoflavone) and QC (a flavonol) are included for comparison. Results are expressed as mean values ( SE for four determinations. TBARS analysis was performed on human LDL exposed to 40 µM SIN-1 and 5 µM chalcone or flavanone for 24 h at 25 °C. b Significantly different from control, p < 0.05, by a twosided Student’s t-test.

Table 2. Effect of Combinations of XN and IsoXN on LDL Oxidation Induced by SIN-1a addition to reaction mixture

TBARS (% of control)b

5 µM XN 25 µM XN 5 µM IX 25 µM IX 5 µM XN + 5 µM IX 5 µM XN + 25 µM IX 25 µM XN + 5 µM IX 25 µM XN + 25 µM IX

26.3 ( 0.0a 14.7 ( 0.6bc 113 ( 1.3d 106 ( 1.3e 43.5 ( 0.7f 41.6 ( 1.7f 16.8 ( 1.0b 14.0 ( 0.3c

Results are expressed as mean values ( SE for four determinations. TBARS analysis was performed on human LDL exposed to 40 µM SIN-1 and 5 µM flavonoid for 24 h at 25 °C. b Values not sharing a common superscript are significantly different, p < 0.05. a

prenylation was not very pronounced (Table 1). IsoXN and tetrahydroxanthohumol (TX), both lacking the R,βunsaturated keto functionality, offered little or no protection against lipid peroxidation. On the contrary, IsoXN displayed a slight prooxidant effect by increasing the formation of TBARS (Tables 1 and 2) and the formation of conjugated dienes in LDL exposed to 40 µM SIN-1 (data not shown). However, the prooxidant activity of IsoXN was abolished by the addition of XN to the incubation mixture containing LDL and SIN-1 (Table 2). The observation that prenylated chalcones are more potent inhibitors of SIN-1-induced LDL oxidation than their nonprenylated or flavanone derivatives led us to examine the reaction of XN with SIN-1 and ONOO-. When XN (1.6 mM) was exposed to SIN-1 (3.2 mM) at pH 7.4, two major products were detected by HPLC-DAD, i.e., AUXN and EPOX (Figures 3 and 4). The large

Figure 4. On-line UV spectra of XN, (Z)-AUXN, and EPOX. For spectral data of other oxidation products, see Table 3.

bathochromic shift of AUXN (ca. 30 nm) and its molecular weight (Mr 353 by ESI-MS) indicated that an aurone type flavonoid was formed in the oxidation reaction (Table 3). The other product, EPOX, showed a molecular ion at m/z 369 [MH]+ in the ESI mass spectrum indicating that an oxygen atom was incorporated (see the following paragraph). Large-scale oxidation of XN with SIN-1 yielded (Z)-AUXN (8.2%) as the major oxidation product while a smaller amount of EPOX was also isolated by semipreparative HPLC (yield, 0.8%). Exposure of a solution the (Z)-aurone to light resulted in conversion of the Z isomer into the E isomer as shown by HPLC-DAD analysis and NMR spectroscopy, and this property was used to obtain a small amount of chromatographically pure (E)-AUXN. Light-induced isomerization of aurones is well-known (25). The identity of both AUXN isomers was confirmed by 1H NMR analysis (Table 4), while distinction between the E and the Z isomers was made by NOESY measurements. The major isomer showed interaction between the H-2/6 protons of the B-ring and the methyl protons of the prenyl substituent, and it was therefore identified as the Z isomer. The methine carbon of the Z isomer showed a resonance at δC 108.9, which is


Stevens et al.

Table 3. Chromatographic and Spectral Data of Products Obtained by Treatment of XN and IsoXN with SIN-1/ONOOHPLC-DAD dataa compd code

compd name

Rt (min)

XN EPOX (Z)-AUXN (E)-AUXN NO2XN NO2EPOX NOEPOX IsoXN NO2IsoXN

xanthohumol endoperoxyxanthohumol (Z)-auroxanthohumol (E)-auroxanthohumol 3-nitro-xanthohumol 3-nitro-endoperoxyxanthohumol 3-nitroso-endoperoxyxanthohumol isoxanthohumol 3′-nitro-isoxanthohumol

15.6 9.7 9.5 12.8 19.7 13.9 15.1 7.1 10.6

ESI-MSb (m/z)

UV λmax (nm) 368 363, 322 395, 336sh 404, 334sh 357, 291sh 382, 322sh 288 286

[MH]+

MS-MS of [MH]+

355 369 353 353 400 414 398 355 400

299 (17), 179 (100) 313 (100) 297 (100) 297 (100) 179 (100) 358 (100) 342 (100) 299 (31), 235 (13), 179 (100) 344 (51), 235 (11), 179 (100)

a HPLC separations were achieved on a 5 µm LiChrospher RP-18 column (4 mm × 250 mm) using a linear solvent gradient starting from 40% MeCN (B) in 1% aqueous formic acid (A) to 100% B over 30 min at 1.0 mL/min. b Electrospray mass spectrometry. MS-MS fragmentation was carried out with a collision energy of 25 eV.

Table 4. atom no. CdO R β 1 2/6 3/5 4 1′ 2′ 3′ 4′ 5′ 6′ 1′′ 2′′ 3′′ 4′′ 5′′ 2′-OH 4′-OH 4-OH 6′-OCH3 a

1H

XN δH 7.75 d J ) 16 Hz 7.65 d J ) 16 Hz 7.56 d J ) 8.5 Hz 6.82 d J ) 8.5 Hz

6.07 s 3.13 d J ) 7.0 Hz 5.12 t J ) 7.0 Hz 1.69 s 1.60 s 14.63 s 10.60 s 10.10 s 3.85 s

and

13C

NMR Data for XN, EPOX, and (Z)-AUXN (DMSO-d6, 500 MHz)a XN δC 191.4 123.6

EPOX δH

EPOX δC 179.1 131.08

7.18 s

142.4 125.9 130.4 115.8 159.7 104.4 164.4 107.2 162.2 90.8 160.3 21.1 122.9

136.0 7.10 d J ) 8.9 Hz 6.78 d J ) 8.9 Hz

6.22 s 3.23 d J ) 7.1 Hz 5.16 t J ) 7.1 Hz

129.8 17.7 25.5

1.74 s 1.64 s

55.7

10.86 s 9.38 s 3.80 s

(Z)-AUXN δH

149.0 118.0 116.0 154.2 103.7 164.8 103.8 164.4 93.9 156.5 20.7 121.8 131.07 17.5 25.3

55.5

(Z)-AUXN δC 178.6 146.3

6.54 s 7.78 d J ) 8.7 Hz 6.85 d J ) 8.7 Hz

6.19 s 3.32 d J ) 7.3 Hz 5.26 t J ) 7.3 Hz

108.9 123.5 132.7 115.9 158.7 103.2 165.2 103.9 166.2 94.3 157.1 21.2 122.2

1.80 s 1.65 s

130.7 17.7 25.5

3.79 s

NDa ND 55.5

For comparison, the same atom numbers were assigned to the same atoms in XN, EPOX, and AUXN; ND, not detected.

in line with δC-β values reported for other (Z)-aurones (range, 108.1-112.8 ppm) while (E)-aurones show δC-β values in the range of 119.8-122.2 ppm (26). When XN was treated with a 2-fold molar excess of peroxynitrite, EPOX appeared as the major product peak in the HPLC-UV370 chromatogram of the reaction mixture (Figure 3C). This reaction was therefore scaled-up in order to produce sufficient quantities of EPOX for preparative HPLC isolation and structural characterization. EPOX showed a molecular ion at m/z 369.1344 in the ESI-FTICR mass spectrum (C21H21O6+ calculates for 369.1332). This oxidation product thus resulted from the net addition of one oxygen atom to XN and loss of two hydrogen atoms. Collision-induced fragmentation of the molecular ion yielded an abundant prenyl cleavage fragment, m/z 313 [MH - C4H8]+. The 1H spectrum of EPOX showed an intact prenyl substituent, an unchanged B-ring, and all signals of XN’s A-ring with the exception of the low-field hydrogen-bonded 2′-OH group. The mutually trans-coupled H-R and H-β protons (J ) 16 Hz), characteristic of chalcones, were absent in EPOX, and these resonances were replaced with a new singlet

at δH 7.20, indicating that one of the two bridge carbons was oxygenated. The molecular weight of the compound (Mr 368) requires loss of two hydrogen atoms, which can be accounted for by the formation an additional ring involving the OH group at position 2′. The new quaternary carbon atom appeared at δC 136.2 in the 13C spectrum, which was considered too far downfield for a hemiketal carbon, and so, it was assumed that the newly formed heterocyclic ring contained an endoperoxide moiety. At this point, two possible structures emerged, one having a (1,2)-dioxinone ring with C-R as the bridge carbon and the other having a (1,2)-dioxepinone ring with the bridge at C-β. Distinction between the two possibilities was made by 1H-13C HMBC spectroscopy. In the HMBC spectrum of EPOX, the methine proton at δH 7.20 showed interactions with the keto carbon, C-β (δC 136.2) and C-1 but not with C-2/6 and C-1′. Because the R-proton of XN did not interact with C-1′ or C-2/6 either, the methine proton of EPOX was therefore assigned to H-R in favor of the second possibility. Moreover, the HMBC spectra of XN and (Z)-AUXN show that H-β interacts with the C-2/6 atoms in both compounds (3JH,C


interactions), which supports the absence of an H-β proton in EPOX. Although the NMR data provided no direct evidence for oxygenation at C-β in EPOX, the β-cyclization product was accepted as the correct structure of EPOX by comparison with the HMBC behavior of XN and (Z)-AUXN. Furthermore, EPOX was resistant to retro Diels-Alder fission of the OC-CR bond by collisional activation in MS-MS experiments, another argument against the presence of a six-membered heterocycle (in the MS-MS spectrum of IsoXN, the base peak at m/z 179 represents a retro Diels-Alder fragment) (21; cf. Table 3). Finally, EPOX showed no evidence of E,Z isomerism in solution, which also points to the absence of an exocyclic methylene functionality. In addition to EPOX, the reaction of XN with peroxynitrite yielded minor amounts of 3-nitro-EPOX (observed m/z 414.1195 in the ESI-FTICR mass spectrum, C21H20O8N+ calculates for 414.1183 Da) and 3-nitrosoEPOX (observed m/z 398.1243 in the ESI-FTICR mass spectrum, C21H20O7N+ calculates for 398.1234 Da). The MS-MS behavior of both EPOX products was very similar to that of EPOX itself, indicating that the EPOX skeleton remained unchanged in the nitration and nitrosation reactions. The position of the nitro substituent in nitroEPOX was determined by 1H NMR (data listed under Materials and Methods). The nitrated EPOX derivative showed A-ring signals and an H-R singlet that were similar to those of EPOX (cf. Table 4), while the B-ring gave rise to an AMX spin system. The downfield resonance at δH 7.79 was assigned to H-2, which places the nitro substituent at C-3 of the B-ring. The nitroso derivative of EPOX gave rise to A-ring signals similar to those observed for EPOX and 3-nitro-EPOX, indicating that the site of nitrosation was at the B-ring. Unfortunately, the B-ring signals of nitroso-EPOX could not be assigned due to peak broadening and weak intensity. Nitration and nitrosation of various substituted phenols with NO and peroxynitrite have been shown to occur preferentially at the para and ortho positions relative to the OH group when available (27); therefore, the site of nitrosation in EPOX is presumably ortho to the OH group of the B-ring (the para position is occupied). IsoXN was not oxidized to a significant extent when exposed to a 2-fold molar amount of SIN-1 or peroxynitrite. Trace amounts of two products were isolated, one of which was tentatively identified as the 3′-nitro derivative of IsoXN by ESI-MS-MS fragmentation. The retro Diels-Alder fragment appeared at m/z 179 as in IsoXN itself, indicating that the A-ring remained unchanged and that nitration took place at the B-ring. The nitro group was tentatively placed at the ortho position relative to the p-hydroxy group based on mechanistic grounds (2729). Carbon dioxide is a critical factor in the physiological chemistry of peroxynitrite; therefore, the XN oxidations were repeated in the presence of sodium bicarbonate at a physiologically relevant concentration (25 mM) to examine the effect of CO2 on the antioxidant chemistry of XN and IsoXN. The reactions were monitored by LCMS because this technique allows characterization of oxidation products in addition to relative quantification. The LC-MS analyses of the incubations containing bicarbonate revealed that the qualitative composition of the reaction mixtures was similar to that of the incubations without added bicarbonate. The quantitative composition of the reaction mixtures changed significantly as a result


Figure 5. Effect of bicarbonate addition on the profile of oxidation products obtained from the reaction of XN (1 mM, A) or IsoXN (1 mM, B) with peroxynitrite (2 mM) (conditions: 1 h of incubation at 37 °C, 0.1 M sodium phosphate buffer, pH 7.4). The reaction mixtures were analyzed by LC-MS using Q1 scanning in order to be able to detect any unexpected oxidation products other than those listed in Table 3 (no such products were found). Peak areas were determined from selected-ion chromatograms (m/z 353, 355, 369, 398, 400, and 414) using the BioMultiView 1.3 software and expressed as a percentage of the total. This way of presenting the results visualizes relative changes in product formation as a result of bicarbonate addition; no quantitative information can be extracted from the LC-MS data due to differences in response factors for individual products.

of bicarbonate addition in favor of NO2XN and 3′-nitroisoxanthohumol (NO2IsoXN) (Figure 5). The fate of XN in the LDL oxidations mediated by SIN-1 (80 µM) and peroxynitrite (50 µM) was examined by LC-MS/MS analysis of the LDL incubations in the presence of 25 µM XN. Comparison of the results with the LC-MS profile of a reaction mixture containing XN (1 mM) and peroxynitrite (2 mM) in the absence of LDL showed that the LDL-containing systems and the LDLfree system yielded similar sets of oxidation products with EPOX representing the major product in all incubations (cf. Figure 6). In the control incubations without SIN-1 or peroxynitrite, only XN and its isomerization product, IsoXN, could be detected by LC-MS/MS (not shown). These experiments suggest that the observed inhibition of LDL oxidation by XN is primarily due to scavenging of peroxynitrite.

Discussion Oxidation of LDL is considered to be involved in the early development of atherosclerosis and cardiovascular disease. Epidemiological studies have indicated that human intake of flavonoids by consumption of fruits and vegetables may offer protection against atherogenic processes. Little is known about the dietary intake of chalcone type flavonoids, but there is no doubt that beer constitutes the predominant dietary source of the prenylated chalcone, XN. Depending on hopping rate and brewery conditions, beer may contain up to 0.7 mg XN per liter and up to 4 mg/L of total prenylated flavonoids (17). Our present study shows that prenylated chalcones and to a lesser extent their nonprenylated and flavanone analogues inhibit the peroxynitrite-mediated oxidation of LDL, which can be explained at the molecular level. The oxidation of XN with SIN-1 yielded two major types of products, AUXN and EPOX, while reaction of XN with peroxynitrite yielded EPOX only. In the following paragraphs, we propose that formation of AUXN is a radical-


Figure 6. Qualitative composition of reaction mixtures obtained by incubation of XN (1 mM) with peroxynitrite (2 mM) in the absence of LDL, analyzed by LC-MS using Q1 scanning (A). (B and C) LC-MS/MS analyses of the reaction mixtures containing XN (25 µM) and peroxynitrite (50 µM) or SIN-1 (80 µM) in the presence of LDL (0.1 mg protein/mL), respectively. LC-MS/MS with multiple reaction monitoring (see Materials and Methods) was used for analysis of the LDL incubations in order to be able to detect small amounts of oxidation products. EPOX yielded the most intense absolute detector signal in all three incubations. Reaction conditions: sodium phosphate buffer, 0.1 M, pH 7.4; 1 h at 37 °C. Key to peaks: 1, XN; 2, EPOX; 3Z and 3E, Z and E isomers of AUXN; 4, NOEPOX; 5, NO2XN; 6, NO2EPOX; and 7, IsoXN (see also Table 3).

mediated reaction and that EPOX formation is assumed to be the result of two electron chemistry without involvement of radical intermediates. The observation that XN fails to form substantial amounts of AUXN on exposure to peroxynitrite indicates that peroxynitrite is not the oxidizing agent in the SIN1-mediated formation of AUXN. In solution, SIN-1 liberates NO• and O2•- radicals, of which the latter species is known to initiate radical reactions. From other studies on the antioxidant activities of flavonoids, it is known that donation of hydrogen atoms from phenolic hydroxy groups lies at the basis of the antioxidative properties of

Stevens et al.

flavonoids and other polyphenols (19, 20, 30). Thus, abstraction of a hydrogen atom from the 4-OH group of XN gives rise to an oxygen radical intermediate for which a resonance structure is possible with a carbon radical at C-R (Scheme 2). This intermediate can form a heterocycle after donation of a second hydrogen atom from the 2′-OH group of the A-ring. The oxidation product is an aurone type flavonoid whose (Z)-isomer is known to be more stable than the (E)-aurone (25), consistent with our observations. The proposed mechanism for the formation of AUXN (Scheme 2) differs from the epoxidation of the R,β-unsaturated keto functionality by treatment of 2′hydroxychalcones with alkaline H2O2 (Algar-FlynnOyamada or AFO reaction; 31), upon which both R- and β-cyclization takes place as a result of intramolecular nucleophilic attack by the 2′-OH oxygen. A mechanistic study of the AFO cyclization of 2′-hydroxy-6′-methoxychalcone epoxides revealed that R-attack is predominant, yielding a diasteroisomeric mixture of aurone hydrates that were stable for several days in aqueous media at neutral pH values (32). β-Attack yielded the corresponding 3-hydroxyflavanones (32). In our incubation experiments with XN/SIN-1, no formation of aurone hydrates or 3-hydroxyflavanones was observed; therefore, a two step hydrogen atom abstraction followed by selective R-cyclization as depicted in Scheme 2 seems to be a more plausible mechanism for the formation of AUXN. The formation of EPOX cannot be explained by a radical-mediated reaction because a carbon radical located at the β-position is not possible in terms of resonance stabilization. This led us to the hypothesis that oxidation of XN by peroxynitrite is not mediated by a radical species and that peroxynitrite reacts directly with the β-carbon of XN in a Michael type fashion with the ONOO- ion playing the role of Michael donor. Intramolecular nucleophilic attack at the newly introduced C-β oxygen of the ONOOXN adduct by the 2′-OH oxygen yields a seven-membered 3H-(1,2)-dioxepin-5-ol ring system whereby NO2- is released. Subsequent oxidation

Scheme 2. Proposed Superoxide Anion-Mediated Formation of AUXN Isomers from XN

Antioxidant Chemistry of Prenylated Flavonoids Scheme 3. Proposed Peroxynitrite-Mediated Formation of EPOX from XN

yields EPOX (Scheme 3). The mechanism we propose here for the formation of EPOX is similar to the mechanism by Bartberger et al. (33) who suggested on theoretical grounds that peroxynitrite reacts with acetone to yield dimethyldioxirane and nitrite by nucleophilic attack at the electrophilic keto carbon of acetone (in XN, the β-carbon is the electrophilic site for nucleophilic attack by peroxynitrite). From the pattern of reaction products obtained by treatment of XN with SIN-1 or peroxynitrite, it follows that the R,β-unsaturated keto moiety is the reactive part of XN. This finding contrasts with the pattern of oxidation products of XN obtained by Fenton type reactions, which yield primarily XN derivatives containing cyclized prenyl groups (Jurva, et al. Manuscript in preparation). The SIN-1/peroxynitrite-mediated oxidations investigated in this study allow us to rationalize the structureactivity relationships that emerge from the inhibitory effects of prenylated flavonoids on SIN-1-induced oxidation of LDL, because the LDL-free system and the LDL incubations all contained EPOX and other oxidation products listed in Table 3 (cf. Figure 6). Generally, chalcones are more potent inhibitors of LDL oxidation than flavanones or dihydrochalcones (e.g., TX), due to their R,β-unsaturated keto functionality that presumably acts as a Michael acceptor system for peroxynitrite. Monoprenylated chalcones are better inhibitors than their nonprenylated analogues because an electrondonating prenyl substituent at C-3′ enhances the nucleophilicity of the 2′-OH oxygen, which facilitates the formation of EPOX (Scheme 3). Introduction of additional prenyl groups further enhances the nucleopilicity of the 2′-OH group but also increases the compound’s lipophilicity and reduces its water solubility. We hypothesize that the more lipophilic prenylflavonoids form noncovalent complexes with the lipid residues of LDL and that they are therefore less effective as scavengers of reactive oxygen/nitrogen species from the aqueous medium. Cyclization of the prenyl substituent of XN (as in DH and DX) does not seem to cause a significant change in antioxidant activity, consistent with the proposed mechanism. In a previous study, it was shown that DH and

Chem. Res. Toxicol., Vol. 16, No. 10, 2003 1285 Scheme 4. Proposed Reaction Pathways for the Oxidation of XN Mediated by O2•-, Peroxynitrite, and ONOO-/CO2

DX are among the major metabolites of XN formed by incubation of XN with rat liver microsome preparations (34), suggesting that oxidation of XN by liver enzymes may not interfere with the scavenging of peroxynitrite. The same may be true for phase 2 metabolism of XN, yielding the 4- and 4′-glucuronides as the main conjugates (35) that still contain a reactive R,β-unsaturated keto functionality. An important aspect of peroxynitrate-mediated oxidations of biological molecules under physiological conditions relates to the fast reaction of peroxynitrite with CO2, yielding nitrosoperoxycarbonate (ONOO-CO2) that dissociates into nitrogen dioxide (•NO2) and carbonate radical (CO3•-) (36). Nitrogen dioxide may react with tyrosine phenoxy radicals to form 3-nitrotyrosine (29). The competition of CO2 with the reaction between peroxynitrite and XN was investigated by exposure of XN to peroxynitrite in the presence and absence of 25 mM sodium bicarbonate followed by LC-MS analysis of the reaction products (Figure 5). Addition of bicarbonate resulted in a dramatic increase in the formation of NO2XN and a relative decrease in the formation of EPOX and NO2EPOX. This indicates that peroxynitrite reacts faster with CO2 than with XN, shifting the fate of XN from EPOX formation to nitration of the 4-hydroxy phenyl ring. Furthermore, the bicarbonate experiment suggests that EPOX formation is not mediated by a decomposition product of peroxynitrite (one electron chemistry), which may be taken as support for the reactivity of peroxynitrite as a Michael donor (Scheme 3). A summary of the various reaction pathways is presented in Scheme 4.

Acknowledgment. This publication was made possible in part by Grant No. P30 ES00210 from the National Institute of Environmental Health Sciences, NIH, and by the Hop Research Council. We are grateful to Ms. Marilyn C. Henderson, Ms. Chelsea Johnston, Ms. Deborah J. Hobbs, and Mr. Alan W. Taylor for technical support. In addition, we acknowledge Mr. Jeff Morre´ of the Mass Spectrometry facility of the Environmental Health Sciences Center at Oregon State University. J.F.S. thanks Dr. Andrea Porzel for NMR measurements and Dr. Ju¨rgen Schmidt for recording FTICR mass spectra (Institute of Plant Biochemistry, Halle, Germany). We are grateful to Prof. Joseph S. Beckman (Environmental Health Sciences Center, Oregon State University) for valuable comments on the manuscript.

References (1) Darley-Usmar, V. M., Hogg, N., O., Leary, V. J., Wilson, M. T., and Moncada, S. (1992) The simultaneous generation of super-


(2)

(3)

(4)

(5) (6) (7) (8) (9)

(10)

(11)

(12)

(13) (14) (15)

(16) (17)

(18)

oxide and nitric oxide can initiate lipid peroxidation in human low-density lipoprotein. Free Radical Res. Commun. 1, 79-120. Esterbauer, H., Dieber-Rotheneder, M., Waeg, G., Striegl, G., and Jurgens, G. (1990) Biochemical, structural, and functional properties of oxidized low-density lipoprotein. Chem. Res. Toxicol. 3, 7792. Yla-Herttuala, S., Rosenfeld, M. E., Parthasarathy, S., Glass, C. K., Sigal, E., Witztum, J. L., and Steinberg, D. (1990) Colocalization of 15-lipoxygenase mRNA and protein with epitopes of oxidized low-density lipoprotein in macrophage-rich areas of atherosclerotic lesions. Proc. Natl. Acad. Sci. U.S.A. 87, 6959-6963. Hogg, N., Darley-Usmar, V. M., Wilson, M. T., and Moncada, S. (1993) The oxidation of alpha-tocopherol in human low-density lipoprotein by the simultaneous generation of superoxide and nitric oxide. FEBS Lett. 326, 199-203. Frei, B., England, L., and Ames, B. N. (1989) Ascorbate is an outstanding antioxidant in human blood plasma. Proc. Natl. Acad. Sci. U.S.A. 86, 6377-6381. Frei, B. (1995) Cardiovascular disease and nutrient antioxidants: Role of low-density lipoprotein oxidation. Crit. Rev. Food Sci. 35, 83-98. Beckman, J. S. (1996) Oxidative damage and tyrosine nitration from peroxynitrite. Chem. Res. Toxicol. 9, 836-844. Pryor, W. A., and Squadrito, G. L. (1995) The chemistry of peroxynitrite: a product from the reaction of nitric oxide with superoxide. Am. J. Physiol. 268 (5 Pt. 1), L699-L722. Christen, S., Woodall, A. A., Shigenaga, M. K., Southwell-Keely, P. T., Duncan, M. W., and Ames, B. N. (1997) γ-γ-tocopherol traps mutagenic electrophiles such as NO(X) and complements R-tocopherol: physiological implications. Proc. Natl. Acad. Sci. U.S.A. 94, 3217-3222. Go, Y. M., Patel, R. P., Maland, M. C., Park, H., Beckman, J. S., Darley-Usmar, V. M., and Jo, H. (1999) Evidence for peroxynitrite as a signaling molecule in flow-dependent activation of c-Jun NH(2)-terminal kinase. Am. J. Physiol. 277, H1647-H1653. Hertog, M. G. L., Feskens, E. J. M., Hollman, P. C. H., Katan, M. B., and Kromhout, D. (1993) Dietary antioxidant flavonoids and risk of coronary heart disease: the Zutphen elderly study. Lancet 342, 1007-1011. Hertog, M. G. L., Kromhout, D., Aravanis, C., Blackburn, H., Buzina, R., Fidanza, F., Giampaoli, S., Jansen, A., Menotti, A., Nedeljkovic, S., Pekkarinen, M., Simic, B. S., Toshima, H., Feskens, E. J. M., Hollman, P. C. H., and Katan, M. B. (1995) Flavonoid intake and long-term risk of coronary heart disease and cancer in the 7 countries studies. Arch. Intern. Med. 155, 381-386. Knekt, P., Jarvinen, R., Reunanen, A., and Maatela, J. (1996) Flavonoid intake and coronary mortality in Finland: a cohort study. Br. Med. J. 312, 478-481. Harborne, J. B. (1994) The Flavonoids: Advances in Research Since 1986, pp 1-676, Chapman and Hall, London. Kompulainen, J. T., Lehtonen, M., and Mattila, P. (1999) Trolox equivalent antioxidant capacity of average flavonoids intake in Finland. In Natural Antioxidants and Anticarcinogens in Nutrition, Health and Disease (Kumpulainen, J. T., and Salonen, J. T., Eds.) pp 141-150, The Royal Society of Chemistry, Cambridge. Barron, D., and Ibrahim, R. K. (1996) Isoprenylated flavonoidss a survey. Phytochemistry 43, 921-982. Stevens, J. F., Taylor, A. W., and Deinzer, M. L. (1999) Quantitative analysis of xanthohumol and related prenylflavonoids in hops and beer by liquid chromatography-tandem mass spectrometry. J. Chromatogr. A 832, 97-107. Stevens, J. F., Taylor, A. W., Clawson, J. E., and Deinzer, M. L. (1999) Fate of xanthohumol and related prenylflavonoids from hops to beer. J. Agric. Food Chem. 47, 2421-2428.

Stevens et al. (19) Rezk, B. M., Haenen, G. R. M. M., Van der Vijgh, W. J. F., and Bast, A. (2002) The antioxidant activity of phloretin: the disclosure of a new antioxidant pharmacophore in flavonoids. Biochem. Biophys. Res. Commun. 295, 9-13. (20) Choj, J. S., Chung, H. Y., Kang, S. S., Jung, M. J., Kim, J. W., No, J. K., and Jung, H. A. (2002) The structure-activity relationship of flavonoids as scavengers of peroxynitrite. Phytother. Res. 16, 232-235. (21) Stevens, J. F., Ivancic, M., Hsu, V. L., and Deinzer, M. L. (1997) Prenylflavonoids from Humulus lupulus. Phytochemistry 44, 1575-1585. (22) Stevens, J. F., Taylor, A. W., Nickerson, G. B., Ivancic, M., Henning, J., Haunold, M. L., and Deinzer, M. L. (2000) Prenylflavonoid variation in Humulus lupulus: distribution and taxonomic significance of xanthogalenol and 4′-O-methylxanthohumol. Phytochemistry 53, 759-775. (23) Chung, B. H., Segrest, J. P., Ray, M. J., Brunzell, J. D., Hokanson, J. E., Krauss, R. M., Beaudrie, K., and Cone, J. T. (1986) Single vertical spin density gradient ultracentrifugation. Methods Enzymol. 12, 181-209. (24) Sattler, W., Mohr, D., and Stocker, R. (1994) Rapid isolation of lipoproteins and assessment of their peroxidation by highperformance liquid chromatography postcolumn chemiluminescence. Methods Enzymol. 128, 469-489. (25) Markham, K. R., and Geiger, H. (1994) 1H nuclear magnetic resonance spectroscopy of flavonoids and their glycosides in hexadeuteriodimethyl sulfoxide. In The Flavonoids: Advances in Research Since 1986 (Harborne, J. B., Ed.) pp 441-497, Chapman and Hall, London. (26) Agrawal, P. K., and Bansal, M. C. (1989) In Carbon-13 NMR of Flavonoids (Agrawal, P. K., Ed.) pp 236-282, Elsevier, Amsterdam. (27) Yenes, S., and Messeguer, A. (1999) A study of the reaction of different phenol substrates with nitric oxide and peroxynitrite. Tetrahedron 55, 14111-14122. (28) Van Dalen, C. J., Winterbourn, C. C., Senthilmohan, R., and Kettle, A. J. (2000) Nitrite as a substrate and inhibitor of myeloperoxidase. J. Biol. Chem. 275, 11638-11644. (29) Zhang, H., Bhargava, K., Keszler, A., Feix, J., Hogg, N., Joseph, J., and Kalyanaraman, B. (2003) Transmembrane nitration of hydrophobic tyrosyl peptides. J. Biol. Chem. 278, 8969-8978. (30) Arora, A., Valcic, S., Cornejo, S., Nair, M. G., Timmermann, B. N., and Liebler, D. C. (2000) Reactions of genistein with alkylperoxyl radicals. Chem. Res. Toxicol. 13, 638-645. (31) Bennet, M., Burke, A. J., and O’Sullivan, W. I. (1996) Aspects of the Algar-Flynn-Oyamada (AFO) Reaction. Tetrahedron 52, 7163-7178. (32) Adams, C. J., and Main, L. (1991) Cyclisation and subsequent reactions of 2′-hydroxy-6′-methoxychalcone epoxide and related compounds. Tetrahedron 47, 4979-4990. (33) Bartberger, M. D., Olson, L. P., and Houk, K. N. (1998) Mechanisms of peroxynitrite oxidations and rearrangements: The theoretical perspective. Chem. Res. Toxicol. 11, 710-711. (34) Yilmazer, M., Stevens, J. F., Deinzer, M. L., and Buhler, D. R. (2001) In vitro biotransformation of xanthohumol, a flavonoid from hops (Humulus lupulus), by rat liver microsomes. Drug Metab. Dispos. 29, 223-231. (35) Yilmazer, M., Stevens, J. F., and Buhler, D. R. (2001) In vitro glucuronidation of xanthohumol, a flavonoid in hop and beer, by rat and human liver microsomes. FEBS Lett. 491, 252-256. (36) Squadrito, G. L., and Pryor, W. A. (2002) Mapping the Reaction of Peroxynitrite with CO2: Energetics, Reactive Species, and Biological Implications. Chem. Res. Toxicol. 15, 885-895.

TX020100D

Inhibition of Peroxynitrite-Mediated LDL Oxidation by Prenylated

Recommend Documents