The Acid-Promoted Reaction of the Green Tea Polyphenol

At lower EGCG concentrations (e.g., 25 μM) and at pH 3, reaction with .... solvent A; 3% formic acid in methanol, solvent B; 3% B, 0−5 min, from 3%...
1 downloads 0 Views 152KB Size
Download PDF
722

Chem. Res. Toxicol. 2005, 18, 722-729

The Acid-Promoted Reaction of the Green Tea Polyphenol Epigallocatechin Gallate with Nitrite Ions Lucia Panzella, Paola Manini, Alessandra Napolitano, and Marco d’Ischia* Department of Organic Chemistry and Biochemistry, University of Naples “Federico II”, Complesso Universitario Monte S. Angelo, Via Cinthia 4, I-80126 Naples, Italy Received December 20, 2004

Exposure of 400 µM (-)-epigallocatechin-3-O-gallate (EGCG), the main polyphenolic constituent of green tea, to equimolar concentrations of nitrite ions in 0.5 M HCl at 37 °C resulted in the formation of a distinct pattern of products that were trapped as phenazine derivatives by treatment with o-phenylenediamine. Repeated chromatographic fractionation eventually allowed isolation of four main species, which were identified by 2D NMR and MS analysis as 1b, derived from EGCG quinone 1a, the isomeric oximes 2b,c, arising from nitrosation of EGCG on the pyrogallol B-ring, and the dioxime 4b in which the A-ring was doubly substituted. At lower EGCG concentrations (e.g., 25 µM) and at pH 3, reaction with equimolar amounts of nitrite gave 1b as the first formed species, whereas nitrosation products 2b,c and 4b became detectable only with excess nitrite. Similar reaction of chemically prepared 1a with acidic nitrite led to the formation of 2b,c and 4b, suggesting that this quinone may be an intermediate in the nitrosation of EGCG. Exposure of green tea extracts to acidic nitrite ions resulted in the conversion of EGCG to 1a, detected as 1b. Overall these results substantiate literature reports suggesting that the protective effects of EGCG against nitrosation involve mainly an initial redox exchange process and hint at a hitherto unrecognized property of quinone 1a as a potential scavenger of nitrosating species.

Introduction The long history of use of green tea, the most popular beverage in Japan and China, has been accompanied by reports of decreased risk of a variety of disease states and illnesses, including cancers of the gastrointestinal tract (1-3). Green tea is obtained from freshly harvested leaves of the plant Camellia sinensis in which fermenting enzymes catalyzing oxidation of polyphenols are inactivated by steaming or roasting. This procedure ensures that the polyphenol composition is similar to that of fresh leaves, at variance with fermented black tea. Polyphenolic constituents of green tea comprise the epicatechin/gallate family of flavanols, namely, (-)-epicatechin, (-)-epigallocatechin (EGC),1 (-)-epicatechin-3-O-gallate, and (-)-epigallocatechin-3-O-gallate (EGCG) (4). This

latter is a potent antioxidant (5-7) and antiinflammatory agent (8) with anticarcinogenic properties in models of tumorigenesis (9-12). Besides its strong scavenging properties against reactive oxygen species, EGCG has * To whom correspondence should be addressed. E-mail: dischia@ cds.unina.it. Phone: +39-081674132. Fax: +39-081674393. 1 Abbreviations: EGC, (-)-epigallocatechin; EGCG, (-)-epigallocatechin-3-O-gallate; CAN, ceric ammonium nitrate; ESI+, positive ion electrospray ionization.

been shown to act as a most efficient inhibitor of potentially mutagenic and cell damaging reactions induced by nitrite ions in the acidic environment of the stomach (13, 14). Exposure to excess nitrite from the diet, for example, from polluted drinking water, spinaches and other vegetables, or cured/pickled meats, or arising from aberrant generation of endogenous nitric oxide at sites of chronic inflammation is implicated as a potential etiological factor in the development of stomach and colorectal cancers (15, 16). Main underlying mechanisms of cell damage would involve generation of a range of reactive nitrogen species such as nitrogen dioxide and the nitrosonium ion, which would cause N-nitrosation of secondary amines leading to carcinogenic nitrosamines, nitration of protein tyrosine residues and other aromatic biomolecules, and deamination of DNA bases (17-20). Recent studies showed that EGCG efficiently inhibits 3-nitrotyrosine formation by exposure of tyrosine (400 µM) to equimolar levels of nitrite ions in 0.5 M HCl at 37 °C and is more effective than the other catechins and ascorbic acid (21). Likewise, EGCG was found to be a most potent inhibitor of the deamination of adenine and guanine induced by nitrite at 37 °C and at acidic pH, causing conversion to hypoxanthine and xanthine, respectively. Spectrophotometric analysis showed that the interaction of the related polyphenol EGC, which lacks the galloyl moiety, with acidic nitrite is accompanied by marked changes in the absorption spectrum of the solution with a 4-fold increase in absorbance of peak at 280 nm and the appearance of a pronounced shoulder at 350 nm. On this basis, it was inferred that the mechanism of action of the catechin family of polyphenolics proceeds via oxidation of flavanols in scavenging reactive

10.1021/tx0496486 CCC: $30.25 © 2005 American Chemical Society Published on Web 03/08/2005

Reaction of Epigallocatechin Gallate with Acidic Nitrite

nitrogen species (21, 22). However, the exact nature of the products formed by reaction of EGCG with acidic nitrite was not investigated in detail. The present study was aimed at providing a chemical insight into the acid-promoted reaction of nitrite ions with EGCG by examining the process under the reported conditions (21), as well as other conditions aimed to model interactions that may occur in the gastric compartment following elevated nitrite intake (23). The specific goal was to isolate and structurally characterize the main products and to dissect the mechanisms underlying their formation.

Experimental Procedures Materials and Methods. o-Phenylenediamine (99.5%), sodium nitrite, ceric ammonium nitrate (CAN), sodium borohydride, pyrogallol, purpurogallin, tyrosine, and 3-nitrotyrosine were used as obtained. Griess reagent (1% sulfanilamide and 0.1% naphthylethylendiamine in 5% phosphoric acid) was used for product detection on TLC plates. UV spectra were performed using a diode array spectrophotometer (Hewlett-Packard model 8453E). IR spectra were recorded in diffuse reflectance mode using a FT-IR Nicolet Avatar 360 instrument equipped with a diffuse reflectance accessory (SpectraTech, model 0030-102). Samples were dissolved in methanol and placed on a steel capsule on a KBr layer, and solvent was left to evaporate; an empty capsule on the KBr layer was used as reference. 1H and 13C NMR spectra were recorded with a Bruker WM 400 at 400.1 and 100.6 MHz, respectively. The instrument was equipped with a 5 mm 1H/broadband gradient probe with inverse geometry. Gradient selected versions of inverse (1H detected) heteronuclear multiple quantum coherence and heteronuclear multiple bond correlation experiments were used. These latter used a 100 ms long-range coupling delay. 1H, 1H correlation spectroscopy experiments were run using standard pulse programs from the Bruker library. Chemical shifts are reported in δ values (ppm) downfield from TMS. Positive ion electrospray ionization (ESI+)/MS spectra were obtained in 2% formic acid-acetonitrile 1:1 v/v using a MicromassZQ Waters equipment. Main peaks are reported with their relative intensities (percent values are in parentheses). HPLC analyses were performed with a Gilson instrument equipped with a detector set at 280 and 350 nm. Analyses were run on an octadecylsilane coated column, 250 mm × 4.6 mm, 5 µm particle size (Sphereclone, Phenomenex) at a 1.0 mL/min flow rate. The following eluant systems were used: 50 mM phosphoric acid in water, solvent A; 50 mM phosphoric acid in acetonitrile, solvent B; from 2% to 20% B, 0-40 min; from 20% to 55% B, 40-55 min; 55% B, 55-65 min (eluant I); 50 mM phosphoric acid in water-acetonitrile (70:30 v/v) (eluant II); 3% formic acid in water, solvent A; 3% formic acid in methanol, solvent B; 3% B, 0-5 min, from 3% to 60% B, 5-45 min (eluant III). Quantum mechanical calculations were carried out using the Gaussian 03 program (24). The chemical shifts were determined at the GIAO/6-311+G(d,p) level on a structure optimized at the PBE0/6-31G(d) level. Isolation of EGCG from Green Tea Extracts. Dry green tea leaves (25 g) were suspended in 1.5 L of boiling water and allowed to stand under stirring for 10 min. After paper filtration, the supernatant was freeze-dried. The residue thus obtained (5 g) was dissolved in 95% ethanol and fractionated on a Sephadex LH-20 column (170 mm × 25 mm) using 95% ethanol as the eluant. Purification was followed by HPLC (eluant I). EGCG (0.73 g) >98% pure as determined by HPLC and NMR analysis was obtained. Reaction of EGCG with Nitrite. To a solution of EGCG (3 mg, 6.5 µmol) in methanol (100 µL), 0.5 M HCl (16.4 mL) was added followed by 1 molar equivalent of sodium nitrite, and

Chem. Res. Toxicol., Vol. 18, No. 4, 2005 723 the mixture was taken under stirring at 37 °C. The reaction course was followed by HPLC (eluant I). In other experiments the reaction was run (i) as above, but without added nitrite, (ii) as in the general procedure, but in the presence of tyrosine (1 molar equivalent with respect to EGCG), (iii) with the substrate at 25 µM concentration, in the presence of 250 µM nitrite, and with periodic diode array spectrophotometric analysis of the reaction mixture, (iv) using 0.1 M phosphate buffer (pH 3) as the reaction medium, with EGCG and nitrite at 25 µM concentration, or (v) as in (iv), with 25 or 250 µM EGCG, with addition of sodium nitrite at 15 min intervals for 2 h up to a 200 µM final concentration. When required, the reaction mixture was treated with o-phenylenediamine (3 mg, 28 µmol) at 10 min reaction time and, after 1 h, analyzed by HPLC (eluant I). In control experiments, the reaction mixture of EGCG with nitrite was extracted with ethyl acetate (3 × 10 mL) at 10 min reaction time, and the treatment with o-phenylenediamine was carried out on the combined organic layers. Reaction of EGCG with CAN. To a solution of EGCG (3 mg, 6.5 µmol) in methanol (100 µL), 1% H2SO4 (262 mL) was added, followed by CAN (11 mg, 20 µmol), and the mixture was stirred at room temperature. The reaction course was followed by HPLC (eluant I). In other experiments, the reaction was run (i) in 0.5 M HCl with the substrate at 400 µM and 4 molar equivalents of CAN with addition of sodium nitrite (200 µM) after 10 min or (ii) as in the general procedure but with ethyl acetate extraction of the mixture after 10 min followed by treatment of the organic layer with NaBH4. When required, the reaction mixture was treated with o-phenylenediamine (3 mg, 28 µmol) at 10 min reaction time and, after 1 h, analyzed by HPLC (eluant I). Isolation of the Phenazine Derivatives 1b, 2b, 2c, and 4b. For preparative purposes, the reaction of EGCG (400 µM) with nitrite (400 µM) was run in 0.5 M HCl using 480 mg of the starting material, and after 10 min the reaction mixture was treated with o-phenylenediamine (569 mg). After 1 h, the mixture was extracted with ethyl acetate, and the combined organic layers were dried over sodium sulfate and taken to dryness. The residue obtained (400 mg) was dissolved in 95% ethanol and fractionated on a Sephadex LH-20 column (170 mm × 25 mm) using 95% ethanol as the eluant. Fractions were collected based on HPLC analysis (eluant I) and further purified by preparative HPLC (eluant II) to give 1b (tR ) 53.5 min eluant I, 17 mg, 3% yield) (25), 2b (tR ) 52.3 min eluant I, 12 mg, 2% yield), 2c (tR ) 54.9 min eluant I, 12 mg, 2% yield), and 4b (tR ) 51.7 min eluant I, 12 mg, 2% yield). Compound 1b (3 mg, 1% yield) was also isolated from the reaction mixture of EGCG (270 mg) with CAN in 1% H2SO4, following treatment with o-phenylenediamine (321 mg) and fractionation on Sephadex LH-20 (eluant 95% ethanol). 2b. UV λmax (CH3OH) 283, 373 nm; IR νmax 3361, 1701, 1612, 1520, 1448, 1360, 1147, 1032, 988 cm-1; 1H and 13C NMR (CD3OD) see Table 1; ESI+/MS m/z 558 ([M + H]+, 44), 580 ([M + Na]+, 43). 2c. UV λmax (CH3OH) 282 nm; IR νmax 3352, 1683, 1614, 1519, 1454, 1365, 1344, 1235, 1148, 1034, 968 cm-1; 1H and 13C NMR (CD3OD) see Table 1; ESI+/MS m/z 558 ([M + H]+, 43), 580 ([M + Na]+, 100), 596 ([M + K]+, 16). 4b. UV λmax (CH3OH) 268, 323, 370 nm; IR νmax 3313, 1706, 1607, 1530, 1450, 1368, 1230, 1038, 954 cm-1; 1H NMR (acetoned6) (303 K) δ (ppm) 3.11 (m, 2H), 6.03 (br s, 2H), 6.98 (s, 2H), 7.58 (s, 1H), 7.93 (m, 2H), 8.05 (s, 1H), 8.21 (m, 2H); 1H NMR (acetone-d6) (203 K) δ (ppm) 3.50 (m, 2H), 5.84 (br s, 1H), 6.11 (br s, 1H), 6.95 (s, 2H), 7.62 (s, 1H), 8.01 (s, 1H), 8.05 (m, 2H), 8.24 (s, 2H), 9.08 (br s, 1H), 9.84 (br s, 2H), 11.17 (br s, 1H), 15.0 (br s, 1H), 17.0 (br s, 1H);13C NMR (acetone-d6) δ (ppm) 26.3 (CH2), 66.6 (CH), 80.4 (CH), 108.5 (CH), 109.6 (2 × CH), 112.1 (C), 117.5 (CH), 120.3 (C), 129.9 (2 × CH), 131.1 (CH), 131.6 (CH), 135.5 (C), 138.8 (C), 141.0 (C), 141.7 (C), 142.1 (C), 143.9 (C), 144.4 (C), 145.6 (C), 145.8 (2 × C), 153.5 (C), 161.9

724

Chem. Res. Toxicol., Vol. 18, No. 4, 2005 Table 1. 1H and

13C

NMR Data of 2b and 2ca

2b Cb 2 3 4 4a 5 6 7 8 8a 1′ 2′ 3′ 4′ 5′ 6′ 7′ 8′ 9′ 10′ 11′ 12′ 1′′ 2′′ 3′′ 4′′ COO

1H

(J, Hz)

5.48 (br s) 5.87 (s) 2.96 (d, 17.2) 3.03 (dd, 17.2, 4.0) 6.01 (d, 2.4)e 6.06 (d, 2.4)e

7.49 (br s) 8.01 (m)i 7.85 (d, 8.4)k 7.92 (d, 8.4)k 8.20 (m)i 6.79 (s)

Panzella et al.

2c 13C

74.1 68.0 26.8 99.4 157.9c 96.0 157.9c 97.0 158.0c 151.2 145.2g 180.5 143.5g 139.6 126.4 143.4g 145.1g 129.8 132.5 134.8 131.4 121.0 110.0 146.3 139.8 167.8

1H

(J, Hz)

5.67 (s) 5.95 (s) 2.95 (d, 17.6) 3.08 (dd, 17.6, 4.4) 5.95 (d, 2.0)f 6.01 (d, 2.0)f 7.09 (s)

8.31 (m) 8.10 (m) 8.10 (m) 8.31 (m) 6.79 (s)

13C

74.5 68.9 27.4 99.9 156.8d 96.4 158.6d 97.6 158.6d 152.4 128.5 183.0 142.0 141.5h 143.6 144.6h 144.8h 129.8j 132.2l 134.8l 136.1j 121.3 110.5 146.9 140.5 168.3

a Spectra run in CD OD; chemical shift values given in ppm. 3 Numbering as shown in structural formula 2b. For the sake of simplicity, the same numbering is retained for 2c. c-l Interchangeable.

b

(C), 165.4 (C), 183.4 (2 × C); ESI+/MS m/z 587 ([M + H]+, 49), 609 ([M + Na]+, 4). Reaction of Green Tea Extracts with Nitrite. To green tea extracts (15 mg) prepared under the conditions described above and previously dissolved in methanol (500 µL), 0.1 M phosphate buffer (pH 3) (262 mL) was added followed by sodium nitrite up to 25 µM final concentration, and the mixture was taken under stirring at 37 °C. The reaction course was followed by HPLC (eluant I). In other experiments, the reaction was run (i) in 0.5 M HCl (16.4 mL) with tea extracts at 0.9 mg/mL concentration and 400 µM nitrite, (ii) as in the general procedure, but with addition of nitrite at 15 min intervals for 2 h up to a 200 µM final concentration, (iii) as in the general procedure without added nitrite. When required, after 10 min the reaction mixture was treated with o-phenylenediamine (2.1 mg) and after 1 h analyzed by HPLC (eluant I). Reaction of Pyrogallol with Nitrite. To pyrogallol (3 mg, 24 µmol) previously dissolved in methanol (100 µL), 0.1 M phosphate buffer (459 mL) (pH 3) was added followed by sodium nitrite (24 µmol), and the mixture was taken under stirring at 37 °C. The reaction course was followed by HPLC (eluant III). For preparative purposes, the reaction was run using 15 mg of pyrogallol; after 1 h, the reaction mixture was extracted with ethyl acetate, and the combined organic layers were dried over sodium sulfate and taken to dryness. The residue obtained was directly analyzed by NMR.

Results and Discussion Reaction of EGCG with Nitrite Ions at High Concentration and at pH < 1. In an initial set of experiments, the reaction of EGCG with nitrite ions was examined spectrophotometrically by monitoring the changes in the absorbance spectrum after the addition of nitrite ions, as previously reported (21) (Figure 1). The results indicated the appearance of two shoulders around 235 and 330 nm after a reaction time of 10 min (trace

Figure 1. Spectrophotometric analysis of the reaction course of EGCG (25 µM) with nitrite (250 µM) in 0.5 M HCl at 0 (trace A), 10 (trace B), and 270 min (trace C).

B); after 4 h 30 min, the former disappeared, and a 2-fold increase in absorbance around 270 nm was observed (trace C). Based on this experiment, the reaction of EGCG with nitrite ions was carried out with the substrate and nitrite at 400 µM concentration in 0.5 M HCl, that is, under the same conditions reported in the literature for the inhibition of tyrosine nitration (21). Consistent with previous observations, HPLC analysis of the reaction mixture under the reported conditions showed complete loss of EGCG after 10 min with an increase of the solvent front probably due to the presence of unretained species. However, careful optimization of chromatographic conditions allowed the separation of a series of products, four of which (eluted at tR ) 22.5, 23.7, 24.4, and 24.8 min and designated A-D, in that order) were relatively more abundant (Figure 2, plot A). Peaks C and D were poorly resolved under a variety of conditions and could not be separated despite several efforts. Overall formation yield of products A-D was estimated at about 85% based on EGCG UV response at 280 nm. To inquire whether such products were derived from oxidation of EGCG, in a separate experiment EGCG was oxidized with CAN in acid, and the reaction was followed by HPLC. Formation of a single major species identical in all respects (retention time, UV spectrum) to that eluted under peak B was observed. These preliminary observations indicated that the reaction of EGCG with acidic nitrite leads to a main oxidation product, as inferred by the previous authors (21), but this is not the only component of the mixture, implying alternate routes of catechin consumption, possibily via phenolic nitrosation/nitration. Unfortunately all attempts to isolate the main reaction products met with failure due to extensive degradation/polymerization during work up, despite careful control of the temperature and the rigorous exclusion of oxygen. However, the problem was successfully tackled by making recourse to a widely used methodology for the isolation of catechin quinones (25-27), which is based on the treatment of the oxidation mixture with ophenylenediamine leading to stable phenazine derivatives. By this method, quite reproducible chromatograms were obtained (Figure 2, plot B), and a correlation between the native products and their phenazine derivatives could be established by collecting the fractions corresponding to the peaks and re-injecting them after treatment with o-phenylenediamine for 1 h. In particular, A gave the peaks eluted at tR ) 52.3 min (III) and 54.9 min (V), B was related to the peak at tR ) 53.5 min (IV), and C and D gave the couple of peaks eluted at

Reaction of Epigallocatechin Gallate with Acidic Nitrite

Chem. Res. Toxicol., Vol. 18, No. 4, 2005 725

arising evidently from the o-quinone of EGCG on the B-ring (1a). It should be noted that product 1b was previously described as an artifact arising by degradation of a dimeric oxidation product of EGCG following treatment with o-phenylenediamine at 80 °C (25). In the present study, artifactual formation of 1b from species other than the quinone 1a was ruled out by separate experiments in which the product eluted under peak B was carefully treated with sodium borohydride or with cysteine ethyl ester (25) and was readily and quantitatively converted to EGCG, thus suggesting that it was indeed quinone 1a. Products III and V were negative to the Griess reagent for nitroso compounds or nitrite-releasing species (28). They gave the same pseudomolecular ion peaks [M + H]+ and [M + Na]+ at m/z 558 and 580, respectively, in the ESI+/MS spectrum, suggesting two isomeric phenazine derivatives of EGCG of the type 1b substituted with a NO group. Consistent with this view, the 1H NMR spectra indicated in both cases the lack of one aromatic proton on the pyrogallol B ring, with the residual ones resonating at δ 7.49 (III) and 7.09 (V), whereas scrutiny of the 13 C NMR spectra, displaying a carbonyl-type carbon signal at δ 180.5 (III) and 183.0 (V), suggested quinone oxime derivatives. On this basis, the products III and V were formulated as the ortho and para quinone oximes 2b and 2c, respectively. This conclusion was supported

Figure 2. HPLC elution profile of the reaction mixture of EGCG (400 µM) with nitrite (400 µM) in 0.5 M HCl at 37 °C prior to (plot A, 10 min reaction time) and following (plot B) exposure to o-phenylenediamine (1.2 mM) for1 h.

tR ) 50.0 and 51.7 min, designated I and II. On this basis, the reaction was repeated with larger amounts of EGCG, and the ethyl acetate-extractable fraction was chromatographed on a Sephadex LH-20 column, and four main products eluted under peaks II-V could eventually be obtained by preparative HPLC. Product eluting under peak IV, which was also obtained by CAN-induced oxidation of EGCG after treatment with o-phenylenediamine, gave a pseudomolecular [M + H]+ ion peak in the ESI+/MS spectrum at m/z 529 and displayed in the 1H NMR spectrum the resonances expected for a phenazine derivative of EGCG oxidized on the pyrogallol B-ring. On this basis, and considering literature data (25), the product was formulated as 1b,

by a computational investigation of the model structures 3b and 3c by the DFT approach at the PBE0/6-31G(d) level in vacuo (24). Calculated proton and carbon resonances for these structures gave values in satisfactory agreement with experimental data for 2b and 2c. Particularly diagnostic were the chemical shifts differences between the residual protons of the B ring (∆δ ) 0.40 for 2b/2c and 0.52 for 3b/3c) on which basis the structure of the ortho quinone oxime 2b was assigned to the product with the most deshielded residual proton of the B ring (III). NMR data for 2b and 2c are reported in Table 1. Attempts to separate product eluting under peak I from other minor products were unsuccessful because of its

726

Chem. Res. Toxicol., Vol. 18, No. 4, 2005

Panzella et al.

unfavorable chromatographic properties accounting for markedly tailed peaks under a variety of elutographic conditions. Product II, on the other hand, could be isolated in pure form as a Griess negative yellow oil giving a pseudomolecular ion peak [M + H]+ at m/z 587 in the ESI+/MS spectrum consistent with a phenazine derivative containing two NO groups. Noticeable features of the 1H NMR spectrum were the lack of both protons on the resorcin A ring, whereas the pyrogallol B ring was unsubstituted. On this basis, the product was readily assigned the remarkable structure 4b in which the A ring

was present in the dioxime diketo form. Notably, the H-3 proton at δ 6.03 gave a distinct cross-peak in the 1H,13C HMBC spectrum with a carbon resonance at δ 112.1, due reasonably to the C-4a carbon. The apparent downfield shift of this carbon in 4b compared to 1b (δ 98.9) indicated loss of the shielding OH group on the adjacent C-5 position, and the presence of a signal at δ 183.4 in the 13C NMR spectrum supported its replacement by a carbonyl group. Interestingly, when the 1H NMR spectrum was run at -70 °C the signals for the OH oxime protons at δ 15.0 and 17.0 became distinguishable, in addition to the phenolic ones (δ 9-11). It must be noted that 2b, 2c, and 4b can exist as E and Z isomers at the oxime functionalities; however, 1 H and 13C NMR spectra analysis did not reveal a mixture of compounds. It can be argued that the isolated products correspond to the E isomers, on the basis of the reported higher stability of E keto oximes with respect to the Z isomers (29). Also, analysis of scalar coupling constants between H-2 and H-3 protons (30) in all the reaction products allowed us to rule out possible epimerization at the C-2 carbon. In further experiments, the reaction of EGCG with acidic nitrite was carried out as previously described but in the presence of 400 µM tyrosine (21) to determine whether the presence of the amino acid has any effect on the reaction course. HPLC analysis showed that under these conditions product formation and distribution was virtually identical to that obtained in the absence of tyrosine, indicating that the observed oxidation/nitrosation pathways are operative in the presence of biological constituents and other potential targets of reactive nitrogen species. Reactions of EGCG with Nitrite Ions at Lower Concentration and at pH 3. In further experiments, the reaction of EGCG with acidic nitrite was investigated at lower substrate and reagent concentrations, at 37 °C, and at pH 3, that is, under conditions of closer relevance to those occurring within the stomach during digestion (23). Incubation of EGCG at 25 µM concentration with 25 µM nitrite resulted in the initial formation of 1b as the main product after condensation with o-phenylenediamine. Additional treatment with 25 µM aliquots of nitrite at 15 min intervals of time led however to the gradual formation of the nitrosation products identified

Figure 3. HPLC elution profiles of green tea extracts prior to (plot A) and following (plot B) exposure to 400 µM nitrite in 0.5 M HCl for 1 h at 37 °C.

as the phenazine derivatives 2b,c and 4b. A time course study showed that EGCG conversion was complete after 2 h. A similar pattern of products was obtained running the reaction at higher concentration of EGCG (250 µM), but in this case, a minor substrate consumption was observed (58% after1 h, 83% after 2 h). Structural Modifications of ECGC in Green Tea Extracts Induced by Nitrite Ions at Acidic pH. To assess whether the chemistry observed with pure EGCG had any bearing on the actual behavior of this epicatechin in green tea extracts, the effect of acidic nitrite on a green tea extract was investigated. As shown in Figure 3 (plot A), the main detectable species in the green tea extract were EGCG, caffeine, and EGC; after reaction with 400 µM nitrite in 0.5 M HCl, the main product from EGCG was 1a (Figure 3, plot B), identified as 1b after the usual treatment with o-phenylenediamine, whereas nitrosation products were below detection limits. The same result was obtained when the reaction was run at pH 3 with 25 or 250 µM nitrite, but in this case, a minor consumption of EGCG was observed (10% after 1 h in the case of 25 µM nitrite). The finding that even in green tea extracts EGCG is readily converted to 1a would substantiate the role of this catechin in the protective effects of green tea against nitrosation (13, 14). The lack of detectable nitrosation products can be explained considering that the nitrite is consumed by all of the catechins and other polyphenolic constituents of green tea, which therefore sidetrack the nitrosating agent from further interaction with 1a. Mechanistic Remarks. The identification of products 1, 2, and 4 is of interest from the chemical viewpoint and

Reaction of Epigallocatechin Gallate with Acidic Nitrite

Chem. Res. Toxicol., Vol. 18, No. 4, 2005 727

Scheme 1

reveals novel aspects of the reactivity of EGCG with acidic nitrite under biomimetic conditions. At acidic pH, nitrite ions are converted to nitrous acid (pKa ) 3.25) (31), which decomposes to give a range of species according to the following equations:

NO2- + H+ a HNO2 HNO2 + H+ a NO+ + H2O NO+ + NO2- a N2O3 N2O3 a NO + NO2 Nitrous acid may act as an oxidant, since its standard reduction potential (E°) according to the equation HNO2 + H+ + e- ) NO + H2O is 0.983 V (31). An alternative, perhaps more reasonable path, would envision oneelectron oxidation of EGCG by the NO2 generated by decomposition of nitrous acid, as detailed above. Disproportionation or further oxidation of the resulting semiquinone might then lead to the formation of the quinone. In any case, whatever the actual species involved in the process, the important point here is that nitrite ions in acidic media can bring about oxidation of reactive polyphenolic compounds to the corresponding quinones with concomitant conversion to nitric oxide (NO) (32, 33). In light of these arguments, the observed oxidation of EGCG at the B ring leading to quinone 1a is not unexpected and reflects the higher oxidizability of the pyrogallol unit compared to the resorcine and gallate moieties. This is consistent with literature reports showing that the pyrogallol moiety confers the higher susceptibility to oxidation to catechin compounds (34, 35). To substantiate this oxidation mechanism, in a separate experiment the parent triphenol (pyrogallol) was reacted with acidic nitrite ions under conditions identical to those used for EGCG. Product analysis showed the formation of purpurogallin as the sole detectable product with no evidence for nitrosation or nitration products (NMR, HPLC). Our data indicated that quinone 1a is the main product at low concentrations, while nitrosation products become detectable on exposure to higher nitrite levels. EGCG oxidation has been investigated in considerable detail

(7, 25, 36-40), but to the best of our knowledge, this is the first report dealing with the generation and direct entrapment of the quinone 1a. It is possible that this is due to the acidic pH of the medium, decreasing ionization of the enol group and hindering dimerization via 1,4addition of the electron-rich 2′-position to the quinonoid system of another molecule of 1a (25). Products 2b and 2c reflect the two possible modes of condensation of the parent oxime 2a with o-phenylenediamine. Oxime 2a may exist in two tautomeric forms, shown in Scheme 1. Usually, nitrosation products of phenols tend to stay in the nitrosophenol rather than quinone monooxime form, unless they are framed within condensed ring systems (29, 41), so it is difficult to predict which of these forms prevails. This uncertainty however would not detract from the present mechanistic analysis. Oxime 2a arises from nitrosation of EGCG on the pyrogallol B ring; yet, the species that actually undergoes nitrosation is probably the quinone 1a and not EGCG. Consistent with this view, in another experiment pure quinone 1a prepared by CAN oxidation of EGCG was readily converted to 2b,c as well as 4b following reaction with acidic nitrite and subsequent derivatization with o-phenylenediamine. This apparently anomalous reaction, namely, the electrophilic nitrosation of an o-quinone, can be explained by considering that quinone 1a contains an enol-like functionality that accounts for a certain degree of nucleophilic character and may thus explain the observed reactivity toward nitrosating agents, for example, nitrosonium ions. That pyrogallol quinones may also behave like nucleophiles is well-known and is exemplified in the classic mechanism of formation of purpurogallin (42). Moreover, the R-nitrosation of carbonyl compounds via their enols is an established method in synthetic organic chemistry to introduce an oximino functionality (43). Dioxime 4a, the precursor to 4b, has also been shown to arise from nitrosation of 1a. The key mechanistic factor in this pathway, which would drive the second rapid nitrosation step, is probably represented by the generation of mononitroso intermediates of the type 5, in which the hydroxylated o-quinone oxime moiety would be more prone to nitrosation than the native A-ring in 1a because of the nucleophilic enol functionality. Consistent with this view is our failure to detect mononitrosation products on

728

Chem. Res. Toxicol., Vol. 18, No. 4, 2005

the A-ring despite several attempts. A plausible mechanistic scheme for formation of 2a and 4a summarizing the above-discussed pathways is outlined in Scheme 1. The incomplete mass balance, due to the presence of other ill-defined species that escaped isolation and characterization, prevents a more detailed mechanistic analysis, so it is possible that other reaction pathways of EGCG are operative. However, the above scheme establishes the central role of quinone 1a in most nitrosation pathways of EGCG, which is a most remarkable outcome of this study. Conclusions and Biological Relevance. In conclusion, the present results yield the first chemical insight into the products formed by reaction of EGCG with acidic nitrite under conditions of relevance to those occurring in the stomach after ingestion of high levels of nitrite. Nitrite concentrations in saliva are in the range of 50-200 µM (33), so the 400 µM concentration used in this study may seem to be rather high. However, it may be relevant to conditions supposed to be operative in the gastric juice derived from humans consuming 100 g of spinach (44, 45). The observed generation of EGCG quinone 1a with acidic nitrite substantiates previous conclusions in the literature (21) and would entail that EGCG, like other green tea catechins, can inhibit nitrite-induced nitrosation/nitration processes via a redox exchange process leading to the formation of NO. The consequences of an increase in the concentration of NO in the stomach in relation to the regulation of the mucosal blood flow, mucus formation, gastric motility, and acid secretion have already been addressed (46-48) and are out of the scope of the present discussion. Here, it is worth noting that quinone 1a, when exposed to excess nitrite ions in an acidic environment, is susceptible to nitrosation through competing pathways involving nucleophilic sites on the A and B rings. These results provide novel insights into the factors governing competition between oxidation and nitrosation pathways in plant flavonoids (32) and highlight the potential role of epigallocatechin quinones in the protective effects of tea polyphenols against nitriteinduced gastric cancer.

Panzella et al.

(4) (5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

(16)

(17) (18)

Acknowledgment. This study was carried out in the frame of the MIUR project “Sostanze naturali ed analoghi sintetici con attivita` antitumorale” (PRIN 2003). We thank the “Centro Interdipartimentale di Metodologie Chimico-Fisiche” (CIMCF, University of Naples Federico II) for NMR, mass, and computational facilities. We thank Professor O. Crescenzi for computational analysis and helpful discussion and Mrs. Silvana Corsani for technical assistance.

(19)

Supporting Information Available: 1H NMR, 13C NMR, HMQC, and 1H,13C HMBC spectra of compounds 2b, 2c, and 4b. This material is available free of charge via the Internet at http://pubs.acs.org.

(23)

(20) (21)

(22)

1H,13C

References (1) Yang, C. S., Chung, J. Y., Yang, G., Chhabra, S. K., and Lee, M. J. (2000) Tea and tea polyphenols in cancer prevention. J. Nutr. 130, 472S-478S. (2) Brown, M. D. (1999) Green tea (Camellia sinensis) extract and its possible role in the prevention of cancer. Altern. Med. Rev. 4, 360-370. (3) Katiyar, S. K., Agarwal, R., Wood, G. S., and Mukhtar, H. (1992) Inhibition of 12-O-tetradecanoylphorbol-13-acetate-caused tumor promotion in 7,12-dimethylbenz[a]anthracene-initiated SEN-

(24)

CAR mouse skin by a polyphenolic fraction isolated from green tea. Cancer Res. 52, 6890-6897. Graham, H. N. (1992) Green tea composition, consumption, and polyphenol chemistry. Prev. Med. 21, 334-350. Salah, N., Miller, N. J., Paganga, G., Tijburg, L., Paul Bolwell, G., and Rice-Evans, C. (1995) Polyphenolic flavanols as scavengers of aqueous phase radicals and as chain-breaking antioxidants. Arch. Biochem. Biophys. 322, 339-346. Rice-Evans, C. A., Miller, N. J., and Paganga, G. (1996) Structureantioxidant activity relationships of flavonoids and phenolic acids. Free Radical Biol. Med. 20, 933-956. Valcic, S., Burr, J. A., Timmermann, B. N., and Liebler, D. C. (2000) Antioxidant chemistry of green tea catechins. New oxidation products of (-)-epigallocatechin gallate and (-)-epigallocatechin from their reactions with peroxyl radicals. Chem. Res. Toxicol. 13, 801-810. Tedeschi, E., Suzuki, H., and Menegazzi, M. (2002) Antiinflammatory action of EGCG, the main component of green tea, through STAT-1 inhibition. Ann. N. Y. Acad. Sci. 973, 435-437. Valcic, S., Timmermann, B. N., Alberts, D. S., Wachter, G. A., Krutzsch, M., Wyner, J., and Guillen, J. M. (1996) Inhibitory effect of six green tea catechins and caffeine on the growth of four selected human tumor cell lines. Anticancer Drugs 7, 461-468. Lea, M. A., Xiao, Q., Sadhukhan, A. K., Cottle, S., Wang, Z. Y., and Yang, C. S. (1993) Inhibitory effects of tea extracts and (-)epigallocatechin gallate on DNA synthesis and proliferation of hepatoma and erythroleukemia cells. Cancer Lett. 68, 231-236. Lin, Y.-L., Juan, I.-M., Chen, Y.-L., Liang, Y.-C., and Lin, J.-K. (1996) Composition of polyphenols in fresh tea leaves and associations of their oxygen-radical-absorbing capacity with antiproliferative actions in fibroblast cells. J. Agric. Food Chem. 44, 1387-1394. Yoshizawa, S., Horiuchi, T., Fujiki, H., Yoshida, T., Okuda, T., and Sugimura, T. (1987) Antitumor promoting activity of (-)epigallocatechin gallate, the main constituent of ‘tannin’ in green tea. Phytother. Res. 1, 44-47. Wu, Y. N., Wang, H. Z., Li, J. S., and Han, C. (1993) The inhibitory effect of Chinese tea and its polyphenols on in vitro and in vivo N-nitrosation. Biomed. Environ. Sci. 6, 237-258. Tanaka, K., Hayatsu, T., Negishi, T., and Hayatsu, H. (1998) Inhibition of N-nitrosation of secondary amines in vitro by tea extracts and catechins. Mutat. Res. 412, 91-98. Bartsch, H., Ohshima, H., Pignatelli, B., and Calmels, S. (1992) Endogenously formed N-nitroso compounds and nitrosating agents in human cancer etiology. Pharmacogenetics 2, 272-277. Coss, A., Cantor, K. P., Reif, J. S., Lynch, C. F., and Ward, M. H. (2004) Pancreatic cancer and drinking water and dietary sources of nitrate and nitrite. Am. J. Epidemiol. 159, 693-701. Leaf, C. D., Wishnok, J. S., and Tannenbaum, S. R. (1989) Mechanism of endogenous nitrosation. Cancer Surv. 8, 323-334. Nyugen, T., Brunson, D., Crespi, C. L., Penman, B. W., Wishnok, J. S., and Tannenbaum, S. R. (1992) DNA damage and mutation in human cells exposed to nitric oxide in vitro. Proc. Natl. Acad. Sci. U.S.A. 89, 3030-3034. Bartsch, H., Ohshima, H., and Pignatelli, B. (1988) Inhibition of endogenous nitrosation. Mechanisms and implications in human cancer prevention. Mutat. Res. 202, 307-324. Felly-Bosco, E. (1998) Role of nitric oxide in genotoxicity; implications for carcinogenesis. Cancer Metab. Rev. 17, 25-37. Oldreive, C., Zhao, K., Paganga, G., Halliwell, B., and Rice-Evans, C. (1998) Inhibition of nitrous acid-dependent tyrosine nitration and DNA base deamination by flavonoids and other phenolic compounds. Chem. Res. Toxicol. 11, 1574-1579. Pannala, A., Rice-Evans, C. A., Halliwell, B., and Singh, S. (1997) Inhibition of peroxynitrite-mediated tyrosine nitration by catechin polyohenols. Biochem. Biophys. Res. Commun. 232, 164-168. Tricker, A. R., Haubner, R., Spiegelhalder, B., and Preussmann, R. (1988) The occurrence of tobacco-specific nitrosamines in oral tobacco products and their potential formation under simulated gastric conditions. Food Chem. Toxicol. 26, 861-865. Frisch, M. J., Trucks, G. W., Schlegel, H. B., Scuseria, G. E., Robb, M. A., Cheeseman, J. R., Montgomery, J. A., Jr., Vreven, T., Kudin, K. N., Burant, J. C., Millam, J. M., Iyengar, S. S., Tomasi, J., Barone, V., Mennucci, B., Cossi, M., Scalmani, G., Rega, N., Petersson, G. A., Nakatsuji, H., Hada, M., Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima, T., Honda, Y., Kitao, O., Nakai, H., Klene, M., Li, X., Knox, J. E., Hratchian, H. P., Cross, J. B., Adamo, C., Jaramillo, J., Gomperts, R., Stratmann, R. E., Yazyev, O., Austin, A. J., Cammi, R., Pomelli, C., Ochterski, J. W., Ayala, P. Y., Morokuma, K., Voth, G. A., Salvador, P., Dannenberg, J. J., Zakrzewski, V. G., Dapprich, S., Daniels, A. D., Strain, M. C., Farkas, O., Malick, D. K., Rabuck, A. D., Raghavachari, K., Foresman, J. B., Ortiz, J. V., Cui, Q.,

Reaction of Epigallocatechin Gallate with Acidic Nitrite

(25)

(26)

(27) (28) (29) (30) (31) (32)

(33)

(34) (35)

(36)

Baboul, A. G., Clifford, S., Cioslowski, J., Stefanov, B. B., Liu, G., Liashenko, A., Piskorz, P., Komaromi, I., Martin, R. L., Fox, D. J., Keith, T., Al-Laham, M. A., Peng, C. Y., Nanayakkara, A., Challacombe, M., Gill, P. M. W., Johnson, B., Chen, W., Wong, M. W., Gonzalez, C., and Pople, J. A. (2003) Gaussian 03, revision B.01, Gaussian, Inc., Pittsburgh, PA. Tanaka, T., Watarumi, S., Matsuo, Y., Kamei, M., and Kouno, I. (2003) Production of theasinesins A and D, epigallocatechin gallate dimers of black tea, by oxidation-reduction dismutation of dehydrotheasinensin A. Tetrahedron 59, 7939-7947. Tanaka, T., Mine, C.; Watarumi, S.; Fujioka, T.; Mihashi, K., Zhang, Y.-J, and Kouno, I. (2002) Accumulation of epigallocatechin quinone dimers during tea fermentation and formation of theasinesins. J. Nat. Prod. 65, 1582-1587. Tanaka, T., Kondou, K., and Kouno, I. (2000) Oxidation and epimerizzation of epigallocatechin in banana fruits. Phytochemistry 53, 311-316. Klebanoff, S. J. (1993) Reactive nitrogen intermediates and antimicrobial activity: role of nitrite. Free Radical Biol. Med. 14, 351-360. Ivanova, G., and Enchev, V. (2001) Does tautomeric equilibrium exist in ortho-nitrosonaphthols? Chem. Phys. 264, 235-244. Davis, A. L., Cai, Y., Davies, A. P., and Lewis J. R. (1996) 1H and 13C NMR assignments of some green tea polyphenols. Magn. Reson. Chem. 34, 887-890. Lide, D. R., Ed. (1995) CRC Handbook of Chemistry and Physics, 75th ed., CRC Press, Boca Raton, FL. Takahama, U., Oniki, T., and Hirota, S. (2002) Oxidation of quercetin by salivary components. Quercetin-dependent reduction of salivary nitrite under acidic conditions producing nitric oxide. J. Agric. Food Chem. 50, 4317-4322. Takahama, U., Yamamoto, A., Hirota, S., and Oniki, T. (2003) Quercetin-dependent reduction of salivary nitrite to nitric oxide under acidic conditions and interaction between quercetin and ascorbic acid during the reduction. J. Agric. Food Chem. 51, 6014-6020. Furuno, K., Akasako, T., and Sugihara, N. (2002) The contribution of the pyrogallol moiety to the superoxide radical scavenging activity of flavonoids. Biol. Pharm. Bull. 25, 19-23. Hagerman, A. E., Riedl, K. M., Jones, G. A., Sovik, K. N., Ritchard, N. T., Hartzfeld, P. W., and Riechel, T. L. (1998) High molecular weight plant polyphenolics (tannins) as biological antioxidants. J. Agric. Food Chem. 46, 1887-1892. Zhu, N., Huang, T.-C., Yu, Y., LaVoie, E. J., Yang, C. S., and Ho, C.-T. (2000) Identification of oxidation products of (-)-epigallocatechin gallate and (-)-epigallocatechin with H2O2. J. Agric. Food Chem. 48, 979-981.

Chem. Res. Toxicol., Vol. 18, No. 4, 2005 729 (37) Mizooku, Y., Yoshikawa, M., Tsuneyoshi, T., and Arakawa, R. (2003) Analysis of oxidized epigallocatechin gallate by liquid chromatography/mass spectrometry. Rapid Commun. Mass Spectrom. 17, 1915-1918. (38) Hatano, T., Hori, M., Kusuda, M., Ohyabu, T., Ito, H., and Yoshida, T. (2004) Characterization of the oxidation products of (-)-epigallocatechin gallate, a bioactive tea polyphenol, on incubation in neutral solution. Heterocycles 63, 1547-1554. (39) Tanaka, T., Mine, C., and Kouno, I. (2002) Structures of two new oxidation products of green tea polyphenols generated by model tea fermentation. Tetrahedron 58, 8851-8856. (40) Valcic, S., Muders, A., Jacobsen, N. E., Liebler, D. C., and Timmermann, B. N. (1999) Antioxidant chemistry of green tea catechins. Identification of products of the reaction of (-)epigallocatechin gallate with peroxyl radicals. Chem. Res. Toxicol. 12, 382-386. (41) Fischer, A., Golding, R. M., and Tennant, W. C. (1965) Nuclear magnetic resonance studies of some aromatic nitroso-compounds. J. Chem. Soc. 6032-6035. (42) Critchlow, A., Haslam, E., Haworth, R. D., Tinker, P. B., and Waldron, N. M. (1967) The oxidation of some pyrogallol and purpurogallin derivatives. Tetrahedron 23, 2829-2847. (43) Touster, O. (1953) Nitrosation of aliphatic carbon atoms. In Organic Reactions (Adams, R., Ed) Vol. VII, Chapter 6, pp 327377, J. Wiley and Sons, New York. (44) McKnight, G. M., Smith, L. M., Drummond, R. S., Duncan, C. W., Golden, M. and Benjamin, N. (1997) Chemical synthesis of nitric oxide in the stomach from dietary nitrate in humans. Gut 40, 211-214. (45) Van Maanen, J. M. S., Pachen, D. M. F. A., Dallinga, J. W., and Kleinjans, J. C. S. (1998) Formation of nitrosamines during consumption of nitrate- and amine-rich foods, and the influence of the use of mouthwashes. Cancer Detect. Prev. 22, 204-212. (46) Pique, J. M., Whittle, B. J. R., and Esplugues, J. V. (1989) The vasodilator role of endogenous nitric oxide in the rat gastric microcirculation. Eur. J. Pharmacol. 174, 293-296. (47) Brown, J. F., Hanson, P. J., and Whittle, B. J. (1992) Nitric oxide donors increase mucus gel thickness in rat stomach. Eur. J. Pharmacol. 223, 103-104. (48) Desai, K. M., Sessa, W. C., and Vane, J. R. (1991) Involvement of nitric oxide in the reflex relaxation of the stomach to accommodation of food or fluid. Nature 351, 477-479.

TX0496486