Effects of Polyphenols from Grape Seeds on Oxidative Damage to Cellular DNA

______http://www.paper.edu.cn

Molecular and Cellular Biochemistry 267: 67–74, 2004. c 2004 Kluwer Academic Publishers. Printed in the Netherlands.

Effects of polyphenols from grape seeds on oxidative damage to cellular DNA

Peihong Fan and Hongxiang Lou School of Pharmaceutical Sciences, Shandong University, Jinan, Shandong, P.R. China

Received 23 December 2003; accepted 26 May 2004

Abstract

Grape seed polyphenols have been reported to exhibit a broad spectrum of biological properties. In this study, eleven phenolic phytochemicals from grape seeds were puriﬁed by gel chromatography and high performance liquid chromatography (HPLC). The antioxidant activities of ﬁve representative compounds with different structure type were assessed by the free radical- scavenging tests and the effects of the more potent phytochemicals on oxidative damage to DNA in mice spleen cells were investigated. Procyanidin B4, catechin, epicatechin and gallic acid reduced ferricyanide ion and scavenged the stable free radical, α, α-diphenyl-β-picrylhydrazyl (DPPH) much more effectively than the known antioxidant vitamin ascorbic acid, while epicatechin lactone A, an oxidative derivative of epicatechin, did not reduce ferricyanide ion appreciably at concentrations used and was only about half as effective on free radical-scavenging as epicatechin. Mice spleen cells, when pre-incubated with relatively low concentration of procyanidin B4, catechin or gallic acid, were less susceptible to DNA damage induced by hydrogen peroxide (H2O2), as evaluated by the comet assay. In contrast, noticeable DNA damage was induced in mice spleen cells by incubating with higher concentration (150 µM) of catechin. Collectively, these data suggest that procyanidin B4, catechin, gallic acid were good antioxidants, at low concentration they could prevent oxidative damage to cellular DNA. But at higher concentration, these compounds may induce cellular DNA damage, taking catechin for example, which explained the irregularity of dose-effect relationship. (Mol Cell Biochem 267: 67–74, 2004)

Key words: antioxidants, comet assay, DNA damage, free radicals, grape seed polyphenols

Introduction and inﬂuencing gene expression [8] and antiulcer effect [9]. They have also exhibited properties against the oxidation of Grape seed extracts (GSE) have shown vascular effects low-density lipoproteins [10]. The last effect has explained through affecting vascular tone [1], collagen metabolism [2] the ‘French paradox’ [11–13]. In most cases, GSEs activi- and vascular permeability [3]; cytotoxic effects according to ties are related to their antioxidant properties and are mainly its cytotoxicity towards MCF-7, A-427 and CRL 1739 cells attributed to the phenolic compounds. [4]; chemopreventive effects by inhibiting adenomas polypo- Excessive reactive oxygen species (e.g, hydrogen peroxide sis coli (Apc) gene mutation-associated intestinal adenoma or H2O2) may cause irreparable DNA damage, leading to formation [5] and decreasing 12-O-tetradecanoylphorbol-13- mutagenesis and perhaps cancer [14]. Investigation into the acetate (TPA) induced production of reactive oxygen species, nature of DNA damage and repair have provided valuable DNA fragmentation in hepatic and brain tissues and lipid per- insight into aging, human genetics and cancer [15, 16]. Now, oxidation [6]; cytoprotective effects by ameliorating the toxic there is deep interest in identifying free radical scavengers effects of chemotherapeutic agents [7], protecting spermato- or antioxidants that inhibit oxidative DNA damage. We have gonial cells against radiation damage, apoptopic cell death restricted ourselves to the study of cellular DNA damage

Address for offprints:H.Lou, School of Pharmaceutical Sciences, Shandong University, Jinan, Shandong, 250012, P.R. China (E-mail: [email protected]) 中国科技论文在线______http://www.paper.edu.cn

68 preventing activities of phenolic compounds from grape concentrated until no acetone was left using a rotary evap- seeds using the single-cell gel electrophoresis (comet) as- orator under reduced pressure and a water bath temperature say, in which the cells with damaged DNA display increased <40 ◦. The concentrated solution was extracted four times migration of DNA from the nucleus towards the anode and with 1000 ml of ethyl acetate each time. The ethyl acetate the cell with broken DNA then resembles a “comet” with a extracts were combined, evaporated to remove ethyl acetate brightly fluorescent head and a tail region which was used to and were lyophilized. GSE was obtained. evaluate DNA damage [17–24]. In this paper, eleven compounds were obtained and iden- Gel chromatography fractionation tified from grape seeds adopting modified methods from literature [25–33], including three novel derivative structures, GSE was dissolved in water and was fractionated on a which were not referred in the literature at hand. With rapid Toyopearl TSK HW-40(F) column (70 × 8 cm), eluted screening procedures [34, 35], we determined the antioxi- with methanol in water gradiently. Fraction was detected dant and radical-scavenging activities of five representative by polyamide TLC [CHCl -MeOH-H Oasmobile phase, phenolic phytochemicals with different structure types. The 3 2 stained by 1% K Fe(CN) -FeCl ] and analytic RP-HPLC results were compared with the known antioxidant vitamin 3 6 3 [using a Phenomenex ODS column (5 µm, 4.6 × 250 mm), ascorbic acid. Then using mice spleen cells, we determined 0.2% V/V formate in 20% methanol as mobile phase, flow whether the potent antioxidants could inhibit oxidative dam- rate 1 ml/min, monitored at 270 nm UV absorbance]. age to DNA induced by H2O2. Because phenolic phytochemicals have pro-oxidant effects in non-cellular systems [36, 37], we also determined if phenolic phytochemicals by them- HPLC separation and purification of individual selves actually induce oxidative DNA damage in mice spleen polyphenols present in fractions from gel chromatography cells, taking catechin as representative. The fraction was separated by semi-preparative HPLC, Material and methods using a Waters ODS semi-preparative column (10 µm, 25 × 100 mm). The HPLC mobile phase contained solvent A(water), solvent B (5% V/V THF in 10% methanol) and Chemicals and equipment solvent C (methanol). The linear gradient system employed, at room temperature, was: 0–60 min 40% solvent A and 60% DPPH, low and normal melting point agarose products were solvent B to 35% solvent A, 55% solvent B and 10% solvent purchased from Sigma (ST. Louis, MO). All other chemi- C, 60–90 min 35% solvent A, 55% solvent B and 10% sol- cals and reagents used in the studies were obtained in the vent C to 100% solvent C. The solvent flow rate throughout purest form available commercially. HPLC was performed the run was 3 ml/min. The column eluate was monitored at using Waters 600–900 Semi-preparative liquid chromatogra- 270 nm UV absorbance in each case and individual polyphe- phy equipped with DAD detector. Lyophilization was done nol peaks were collected, the purity was identified by analytic by an Eyela FD-1000 Freeze-drying meter. Proton nuclear HPLC. If necessary, re-separation was performed using semi- magnetic resonance (1H-NMR) and carbon nuclear mag- preparative HPLC. netic resonance (13C-NMR) were recorded on an Avance 500 Brucker spectrometer. Mass spectra were recorded in an API 4000 mass spectrometer. Infrared spectra were recorded Identification of the individual polyphenols obtained with a Nicolet-Nexus 470 Ft-IR spectrometer. X-ray crystal structure data were collected on a Bruker-P4 Kappa CCD IR, MS, 1H-NMR, 13C-NMR spectra were analyzed and con- diffractometer. UV-VIS spectra were measured with an Agi- trasted with related literature, to identify the structure of the lent 8450 spectrometer. In comet assay, DNA was visualized individual phenols obtained. For certain compound 4, X-ray using Olympus IX-70a inverted fluorescence microscope, the crystal diffraction was used. resulting images were taken with Kodak image system. Initial screening of five representative polyphenols Preparation of grape seed extracts (GSE) from GSE for antioxidant activities

Grape seeds were collected from a brewery and milled to In view of the variety and amount of the compounds obtained, powder. The powder (5 kg) was defatted with petroleum catechin, epicatechin, epicatechin lactone A, gallic acid, pro- ether ﬁrstly, and then was macerated for 12 h at room tem- cyanidin B4 were selected as representatives for antioxidant perature three times with 2000 ml of water/acetone (30:70, activity assessment. Vit C was used as control. Sample so- V/V) each time. The three macerates were combined and lutions (0–3 mM, dissolved in methanol) were prepared, the 中国科技论文在线______http://www.paper.edu.cn

69 reducing power and free radical-scavenging activity of sam- Following centrifugation, the cells were resuspended in 1ml ple solutions at different concentration were evaluated. The of HBSS. Cells were then challenged at 4 ◦C with 50 µM reducing power of sample solutions were tested by measuring H2O2 for 20 min and immediately analyzed for extent of the absorbance at 700 nm of the resulting solution, after sam- DNA damage. In other experiments, cells were incubated ple solutions were reacted with potassium ferricyanide and with higher concentration (50 µM, 150 µM) of catechin to ferric chloride solutions, the higher the absorbance represents determine any pro-oxidant effects on DNA. the stronger the reducing power [34]. Free radical-scavenging activities of the samples were evaluated by determining their Assessment of cellular DNA damage abilities to chemically reduce the stable free radical, α,α- diphenyl-β-picrylhydrazyl (DPPH) [34, 35]. After DPPH· To determine the extent of DNA damage in cells, single cell methanolic solution reacted with samples, the absorbance at gel-electrophoresis or the “comet assay” [20, 21] was per- 525 nm of the resulting solutions were measured and com- formed under dim light. Briefly, following treatment and pared with the absorbance of DPPH in the absence of sample washing, cells were suspended in 0.8% low melting point solution. The lower the absorbance at 525 nm represents the agarose dissolved in HBSS (37 ◦C). Then 75 µlofthe mix- higher DPPH scavenging activity. The percentage of DPPH ture was pipetted onto a frosted microscope slide that had scavenging activity is expressed by the following Equation: been pre-coated with 75 µlof0.8% normal melting point agarose. Without delay, a glass cover slip was placed on top Atest (1 − ) × 100%. of the slide, and the agarose/cell mixture was allowed to com- A blank pletely congeal by putting the slides on a cold metal tray for 10 min. After removing the cover slip, the slide was immersed Isolation of mice spleen cells in ice-cold lysis solution [2.5 M NaCl, 100 mM disodium EDTA, 10 mM Tris, NaOH to PH 10, supplemented with 1% Grown mice were killed by dislocation, the spleen was (v/v) Triton X-100, 10% DMSO before used] for 2 h. Next, quickly removed and connective tissue attached was shelled slide was placed in a submarine gel electrophoresis unit con- in oxygenated Hanks’ balanced salt solution Ca-Mg-free taining 300 mM NaOH and 1 mM EDTA, PH 13 for 40 min (HBSS), the spleen then was crushed on stainless steel sieve to unwind before being electrophoresed at 18V (160 mA) for (100 meshes) with little HBSS to prepare single cell sus- 40 min. Following electrophoresis, the slides were twice im- pension. The suspension was collected in 10 ml centrifuge mersed in neutralizing buffer (0.4M Tris-HCl, PH 7.5, 4 ◦C), tube, centrifuged for 5 min at 1500 r.p.m (all at this condi- each time for 5 min, before finally applying 50 µlof5µg/ml tion if not referred specially). The supernatant was discarded ethidium bromide on them and leaving the slides in dark for and the pellet was resuspended in 4.5 ml distilled water for 20 min to stain the DNA. 30 sec to splinter red cells before 0.5 ml 0.9% NaCl was Slides were viewed using an inverted fluorescence mi- added to recovery. The cells were pelleted again and resus- croscope. Fifty nucleoids per slide were scored visually for pended in foetal calf serum, stored no more than 4 h in ice comet tail size based on an arbitrary scale of 0–4, i.e., rang- ◦ (−4 C) before treatment. The cells were diluted down to a ing from no damage to extensive damage of DNA. Individual 6 concentration of 10 cells / ml prior to use. The vitality was scores were added to give the final comet score reported. This evaluated by means of the Trypan Blue exclusion test. The basic procedure of scoring comets compares acceptably with investigation conforms with the Guide for the Care and Use compute imaging analysis [38]. The tail length of twenty-five of Laboratory Animals and approved by the Central Animal nucleoids per slide were measured and averaged, additionally. Care Committee of the Shandong University. Statistical treatment of data Cell treatment Data were analyzed with a standard SAS program using anal- The compounds of interest were dissolved in dimethyl ysis of variance (ANOVA) and Duncan’s new multiple range sulfoxide and diluted in FCS to the required concentration. test to determine any significant difference (P < 0.05). At the treatment stage the final dimethyl sulfoxide concentration was never higher than 0.05%. Under these conditions, Results dimethyl sulfoxide was neither toxic nor DNA-damaging. Cells (106/ml in a 24-well microplate) were pre-incubated Isolation and purification of polyphenol compounds without or with 10 µM, 25 µMofcompounds inside the ◦ humidified incubator (95%air/5%CO2,37 C) for 30 min to Eleven compounds with phenolic hydroxyl groups were allow for cellular uptake of the pheolic phytochemicals. Af- isolated from GSE and identified by spectra analysis (Fig. 1), terwards, cells were washed with ice-cold D-hanks (HBSS). including gallic acid (430 mg) from the 5%MeOH elute of 中国科技论文在线______http://www.paper.edu.cn

Fig. 1. Chemical structure of 11 individual polyphenols identiﬁed in GSE.

HW-40 column, catechin (7.07g) and epicatechin (3.4 g) Reducing power and free radical-scavenging effect from the 20% MeOH fraction, procyanidin B1 ( 40 mg ), B2 (600 mg), B3 (38 mg), B4 (1.67 g) and B5 (31 mg) The antioxidant activities of the five representative com- through reseperating the 30–50% fractions by HPLC, and pounds were first evaluated spectrophoto-metrically by as- three novel isomeric derivatives of flavon-3-ol with original sessing the capacity of each compound to reduce ferricyanide B-ring substituted by lactone ring, catechin lactone (40 mg), ion and scavenge DPPH, a stable free radical, and vitamin epicatechin lactone A (64 mg) and B (22 mg) obtained by ascorbic acid, a known commercial antioxidant was chosen as recrystalizing after rechromatographing the 10% MeOH control. In reducing power test, all compounds used showed fraction from HW-40. Catechin and epicatechin, as the con- concentration-dependant activity (Fig. 2). Procyanidin B4 stitutional units of procyanidins, were noticeable for their reduced the ferricyanide ion most quickly and thoroughly, good yield, while procyanidin B4 was the maximum among while epicatechin, gallic acid, catechin weakened gradation- the dimers, epicatechin lactone A among the derivatives. ally. Any of them did better than Vit C. In free radical- So, within the compounds we obtained, procyanidin B4, scavenging test, the concentration-dependant effect was also catechin, epicatechin, epicatechin lactone A and gallic acid observed and their IC50 were calculated (Fig. 3). Procyanidin were selected as representatives of different type of structure B4 was the strongest with IC50 0.109 mM, and gallic acid for bioactivity assessment. also exhibit rather strong effect (IC50, 0.112 mM). The IC50 中国科技论文在线______http://www.paper.edu.cn

screening, procyanidin B4, gallic acid and catechin were kept as representatives of potent antioxidants and assessed in further experiments. In antioxidant experiments, the solvents (i.e. methanol, PBS) never showed appreciable effect (data not shown).

Comet assay

Comet assay was used to determine whether the selected compounds could inhibit oxidative damage to cellular DNA. Comet image was analyzed (Fig. 4). Control cells had a comet score of 15 ± 3, whereas cells challenged with 50 µM H2O2 had a comet score of 180 ± 8 that indicates extensive DNA damage. However, 10 µMofcatechin, procyanidin B4 and gallic acid significantly inhibited H2O2-induced damage Fig. 2. Spectrophotometric analysis of the phenolic phytochemicals for re- to cellular DNA (Fig. 5), as indicated by the lower comet ducing power. Catechin(cat), epicatechin(epi), epicatechin lactone A (lac), scores of 120 ± 5, 87 ± 16 and 157 ± 11, respectively. The gallic acid(gal), proanthocyanidin B4 (pc B4), as well as the control, Vit effect of 25 µMofcompounds were also measured, comet C, were reacted with ferricyanide ion, the reduced product then complexed with ferric chloride. The absorbance of the resulting solution was measured scores of catechin, procyanidin B4 and gallic acid were at 706 nm. The increase of the absorbance reflected more ferricyanide ion 93 ± 12, 106 ± 18 and 114 ± 7 respectively, and as another was reduced. Data represent the mean of three experimental determinations. common evaluating indicator of DNA damage, the average The S.E.M was very small and not shown in figure to keep it distinct enough. tail length (DNA migration) were also measured (320 ± 15, 397 ± 15 and 410 ± 16 respectively) (Table 1). Comparing the effects of the samples at the concentration of 10 and 25 µM, we found that the dose-effect relationship was not so regular as shown in reducing power and free radical- scavenging tests, especially for procyanidin B4. For procyanidin B4, the protecting effect decreased with increasing concentration, instead. Finally, to explain the irregular trends of the phytochemicals and determine if higher concentration of them by themselves induce damage to DNA in mice spleen cells, 50 and 150 µM catechin were used to duplicate the experiments (Fig. 6). Compared with control cells (comet score 15 ± 2), 50 and 150 µM catechin had comet scores of 79 ± 16 and 183 ± 11. However, cells treated with low concentration of catechin (10 and 25 µM) had comet scores of only 30 ± 4 and 39 ± 6. Fig. 3. Spectraphotometric analysis of phenolic phytochemicals for free radical-scavenging activity. Catechin (cat), epicatechin (epi), epicatechin lactone A (lac), gallic acid (gal), proanthocyanidin B4 (pc B4), as well as the Discussion control, Vit C, were reacted with α,α-diphenyl-β-picrylhydrazyl (DPPH). The decrease in absorbance, reflecting chemical reduction of DPPH, was The central finding of the present study is that several phe- measured at 525 nm. The IC50 was calculated after linear regression was performed. Data represent the mean ± S.E.M. of three experimental deter- nolic phytochemicals isolated from grape seeds show strong minations. Values not sharing the same letter are significantly different from antioxidant activity in vitro and, in low concentration, afford one another (P < 0.05). significant protection against oxidative damage in the DNA of mice spleen cells. DNA damage can cause mutagenesis, of catechin and epicatechin were close (0.127, 0.133 mM, carcinogenesis and other chronic degenerative diseases, so respectively). Still, the four did better than Vit C (IC50, 0.39 our results may be an aid to explain the nutrition of GSE. mM). In both tests, epicatechin lactone A did not show strong Grape seed polyphenols mainly included procyanidins activities and was eliminated from our experiments. Epicat- [39]. Using gel chromatography and semi-preparative HPLC, echin was removed, too, because of its similarity to cate- 11 pure individual phenolic compounds were separated from chin in structure and antioxidant activity. Thus, after initial GSE. Epicatechin lactone A, B and catechin lactone are 中国科技论文在线______http://www.paper.edu.cn

Fig. 4. Representative photomicrographs of comet image in experiments (A. Control; B. H2O2 treatment; C. Protection against H2O2-induced damage by 10 µM catechin).

scavenging tests were selected to screening the antioxidant activity of phenolic compounds, quickly. The whole trends demonstrated that antioxidant activity is associated with the number of phenolic hydroxy groups, especially in reducing power test, and procyanidin B4 was proved to be the most ex- cellent antioxidant among the samples in experiments. How- ever, gallic acid with three phenolic hydroxy groups, abnor- mally, showed much stronger power than catechin or epicatechin in free radical-scavenging test. So the number of phenolic hydroxy groups is not the unique inﬂuence factor. In both tests, epicatechin lactone A was weak, just consistent with the higher oxidation state of its structure. At last, procyanidin B4, catechin and gallic acid were discussed in DNA damage preventing experimemt, which all did better than Vit Cinantioxidant tests in vitro. Fig. 5. Inhibitory effect of 10 µM catechin (cat), procyanidin B4 (pc B4) and Hydrogen peroxide is believed to cause DNA strand break- gallic acid (gal) on hydrogen peroxide-induced oxidative damage to DNA age by generation of the hydroxyl radicals close to the DNA in mice spleen cells. Cells were preincubated with 10 µMofcompound molecule, via the fenton reaction [40]. Thus, antioxidants ◦ µ ◦ at 37 C for 30 min before being challenged with 50 MH2O2 at 4 C for should be potential agents to protect DNA against oxidative 20 min. The extent of oxidative damage to DNA was assessed by comet assay. damage. Here, mice spleen cells were used as experimental Data represent the mean ± S.E.M. of three experimental determinations. Values not sharing the same letter are signiﬁcantly different from one another model. The viability of cells was measured by Trypan blue ex- (P < 0.05). clusion. Hydrogen peroxide induced damage to mice spleen cells, aggravating by higher concentration and longer actua- beyond the usual varieties of compounds from GSE, which tion duration, as measured by comet assay (data not shown). are presumed oxidative derivatives of epicatechin and In our experiments, low concentrations (i.e. 10 µMor catechin after analyzing the homology of their structures. 25 µM) of procyanidin B4, catechin, gallic acid inhibited Taking into account the variety and yield of the compounds, H2O2-induced damage to DNA in mice spleen cells partly. procyanidin B4, catechin, epicatechin, epicatechin lactone In agreement with its better antioxidant activity in screening A and gallic acid were brought into activity assessment, as experiments, procyanidin B4, at the concentration of 10 µM, representatives. was superior to either catechin or gallic acid, and was almost The antioxidant activity refers to the capacity of trans- twice stronger than catechin, just consistent with its double ferring electrons. The reducing power and free radical- number of phenolic hydroxy groups. But gallic acid was

Table 1. Inhibitory effect of catechin, procyanidin B4 and gallic acid on hydrogen peroxide-induced oxidative damage to DNA in mice spleen cells. The extent of oxidative damage to DNA was assessed by comet assay and was expressed by comet score and averaged tail length (migration of DNA from the head of nuclei). Data represent the mean ± S.E.M

Catechin Procyanidin Gallic acid H2O2 Control 50 µM10µM25µM10µM25µM10µM25µM

Comet score 15 ± 3 180 ± 8 120 ± 25 93 ± 12 80 ± 16 106 ± 18 114 ± 7 157 ± 11 Tail length ( µm) 32 ± 5 903 ± 47 691 ± 40 320 ± 15 315 ± 24 397 ± 15 707 ± 34 410 ± 16 中国科技论文在线______http://www.paper.edu.cn

may be due to their reaction with metal ions. For example, they may reduced Fe3+ and catalyze hydroxyl radical formation [41, 42]. While this study was in progress, a report was noticed that studied the effect of complex polyphenols and tannins from red wine (WCPT) on DNA oxidative damage of rat colon mucosa in vivo.Itwas concluded that cellular susceptibility to in vitro oxidative stress induced by different concentrations of H2O2 was not modified by WCPT treatment [43]. Though the constitutions of WCPT are similar to GSE, we cannot anticipate if the protecting effects of procyanidin B4, catechin or gallic acid still exist in vivo,orjust is similar to WCPT, because the metabolism of phenols is another troublesome problem. There are still many problems in front and further research is needed. In summary, procyanidin B4, catechin and gallic acid Fig. 6. Damage effects of different concentrations of catechin on DNA in mice spleen cells. Cells were incubated with 10, 25, 50, 150 µM catechin had substantial antioxidant activity. Using a cell model, low at 37 ◦C for 30 min. The extent of oxidative damage to DNA was assessed concentrations of them inhibited DNA damage induced by by comet assay. Data represent the mean ± S.E.M. of three experimental reactive oxygen. In contrast, high concentration of them by determinations. Values not sharing the same letter are significantly different themselves can cause cellular DNA damage. Their contrari- < . from one another (P 0 05). ant actions should be paid attention to. Since the discussed compounds were presentatives of those 11 compounds from beyond anticipation again, since it did not show better effect grape seeds according to their structure types, the conclusion than catechin. The protecting effect in cells should relate may be generalized to the other phenolic phytochemicals we positively to antioxidant activity in non-cellular environment obtained, even to the more with similar structure from grape in theory, but it is not plausible to take it for granted. After all, seeds. there are many other factors, which can influence the effect in vivo such as the actual concentrations of the compounds in cells. Acknowledgment It is noteworthy that the effect of each compound does not This work was supported by grants from National Natural increase inevitable with its concentration. 25 µM procyani- Science Foundation of China (subject number: 30070885). din B4 inhibited DNA strand breaks by 40% while 10 µM We thank Mr Zhiwei Deng of Beijing Normal Univesity, inhibited by 56%, even weaker than 25 µM catechin. Such Wentao Yu of Institute of Crystal Materials, Shandong Uni- contradiction may be explained by taking the pro-oxidant versity for structure identifying. We are very much obliged effect of phenolic compounds into account. In non-cellular to Miss Xiaoning Hao of School of Public Health and Miss models, phenolic phytochemicals can have both antioxidant Ting Liang of Institution of Experimental Nuclear Medicine, and pro-oxidant effects depending on the experimental con- Shandong University for technical assistance. dition [37]. It also was reported that dietary flavonoids at high concentrations inhibit growth, deplete glutathione, decrease viability and induce DNA breakage in normal human References lymphocytes and transformed cells [22]. In our experiments, we also found that, at relatitively higher concentrations of 1. Karim M, Kappagoda T, German B: Endothelium dependent vasorelax- catechin, DNA damage was induced by itself. The comet im- ing activity of polymeric phenolics (flavonoids) present in grape seed age of cells incubated with 50 µM catechin was almost visual extract. FASEB J 12(4): A382, 1998 and 150 µM catechin induced almost equal damage to 50 µM 2. Tixier JM, Godeau G, Robert AM, Hornebeck W: Evidence by in vivo µ and in vitro studies that binding of pycnogenols to elastin affects its rate H2O2, though 10 and 25 M catechin did not induce damage of degradation by elastases. Biochem pharmacol 33(24): 3933–3939, at all. Thus, it was not possible to totally inhibit H2O2-induced 1984 DNA damage by increasing catechin’s concentration because 3. Masquelier J, Dumon MC, Dumas J: Stabilization of collagen by pro- it assumed the role of pro-oxidant at high concentration. In a cyanidolic oligomers. Acta Ther 7: 101–105, 1981 similar way, 25 µM procyanidin B4 may exhibit pro-oxidant 4. Ye X, Krohn RL, Liu W, Joshi SS, Kuszynski CA, McGinn TR, Bagchi M, Preuss HG, Stohs SJ, Bagchi D: The cytotoxic ef- effect and antioxidant effect at the same time, so that the final fects of a novel IH636 grape seed proanthocyanidin extract on cul- protecting ability was weakened. Herein, the dose-effect re- tured human cancer cells. Mol Cell Biochem 196(1–2): 99–108, lationship is complicated. The pro-oxidant effect of phenols 1999 中国科技论文在线______http://www.paper.edu.cn

5. Arri M, Miki M, Hosoyama R, Aria H, Yamaji N, kataoka S: 24. Johnson MK, Loo G: Effects of epigallocatechin gallate and quercetin Chemopreventive effect of grape seed extract on intestinal carcino- on oxidative damage to cellular DNA. Mutat Res 459: 211–218, 2000 genesis in the apc mouse. Proc Am Assoc Cancer Res 39: 20, 25. Ricardo da Silva JM, Rigaud J, Cheynier V, Cheminat A, Moutounet M: 1998 Procyanidin dimers and trimers from grape seeds. Phytochemistry 30: 6. Bagchi D, Garg A, Krohn RL, Bagchi M, Bagchi DJ, Balmoori J, Stohs 1259–1264, 1991 SJ: Protection effect of grape seed proanthocyanidins and selected an- 26. Escribano-Bailon T, Gutierrez-Fernandez Y,Rivas-Gonzalo JC, Santos- tioxidants against TPA-induced hepatic and brain lipid peroxidation and Buelga C: Characterization of procyanidins of Vitis vinifera variety DNA fragmentation, and peritoneal macrophage activation in mice. Gen Tinta del Pais Grape seeds. J Agric Food Chem 40: 1794–1799, 1992 Pharmacol 30(5): 771–776, 1998 27. Sun B, Leandro C, Ricardo-da-Silva JM, Spranger I: Separation of grape 7. Joshi SS, Kuszynski CA, Benner EJ, Bagchi M, Bagchi D: Amelio- and wine proanthocyanidin according to their degree of polymerization. ration of cytotoxic effects of chemotherapeutic agents by grape seed J Agric Food Chem 46: 1390–1396, 1998 proanthocyanidin extract. Antioxid Redox Signal 1(4): 563–570, 1999 28. Sun B, Belchior GP, Ricardo-da-Silva JM, Spranger MI: Isolation and 8. Ray SD, Kumar MA, Bagchi D: In vivo abrogation of acetaminophen- purification of dimeric and trimeric procyanidins from grape seeds. J induced hepatic genomic DNA fragmentation anf apoptopic cell death Chromatogr A 841: 115–121, 1999 by a novel grape seed extract (GSPE). FASEB J 12(5): A779, 1998 29. Wang J, Hano Y, Nomura T, Chen Y: Procyanidins from the seeds of 9. Saito M, Hosoyama H, Ariga T, Kataoka S, Yamaji N: Antiulcer activity Vitis amurensis. Phytochemistry 53:1097–1102, 2000 of grape seed extract and procyanidins. J Agric Food Chem 46: 1460– 30. Lou HX, Yamazaki Y, Oka S: Proanthocyanidins: Chemistry and their 1464, 1998 biological significance. Curr Top Phytochemistry 4: 79–93, 2000 10. Meyer AS, Yi OS, Pearson DA, Waterhouse AL, Frankel EN: Inhibition 31. Geiss F, Heinrich M, Hunkler D, Rimpler H: Procyanidins with (+)- of human low-density lipoprotein oxidation in relation to composition epicatechin units from Byrsonima carassifolia bark. Phytochemistry of phenolic antioxidants in grapes (Vitis vinifera). J Agric Food Chem 39(3): 635–643, 1995 45: 1638–1643, 1997 32. De Bruyne T, Pieters LAC, Dommosse RA, Kolodziej H, Wray V, 11. Frankel EN, Kanner J, German JB, Parks E, Kinsella JE: Inhibition of Domke T, Vlietinck AJ: Unambiguous assignments for free dimeric oxidation of human low-density lipoprotein by phenolic substances in proanthocyanidin phenols from 2D NMR. Phytochemistry 43(1): 265– red wine. Lancet 341: 454–457, 1993 272, 1996 12. Kinsella JE, Frankel E, German B, Kanner J: Possible mechanisms for 33. De Freitas VAP, Glories Y, Bourgeois G, Vitry C: Characterisation the protective role of antioxidants in wine and plant foods. Food Technol of oligomeric and polymeric procyanidins from grape seeds by liquid 47: 85–89, 1993 secondary ion mass spectrometry. Phytochemistry 49(5): 1435–1441, 13. Renaud S, de Lorgeril M: Wine, alcohol,platelets, and the French para- 1998 dox for coronary heart disease. Lancet 339: 1523–1526, 1992 34. Guo JT, Lee HL, Chiang SH, Lin F, Chang CY: Antioxidant properties 14. Simic MG: Mechanisms of inhibition of free radical processes in muta- of the extracts from different parts of Broccoli in Taiwan. J Food Drug genesis and carcinogenesis. Mutat Res 202: 377–386, 1988 Anal 9: 96–101, 2001 15. Ames BN: Dietary carcinogens and anticarcinogens. Science 221: 1256– 35. Saint-cricq de Gaulejac N, Provost C, Vivas N: Comparative study 1264, 1983 of polyphenol scavenging activities assessed by different methods. J 16. Cerutti PA: Prooxidant states and tumor promotion. Science 227: 375– Agric Food Chem 47: 425–431, 1999 381, 1985 36. Hanasaki Y, Ogawa S, Fukui S: The correlation between active oxygen 17. Singh NP, Mccoy MT, Tice RR, Schneider EL: A simple techniquer for scavenging and antioxidative effects of flavonoids. Free radical Biol. quantitation of low levels of DNA damage in individual cells. Exp cell Med 16: 845–850, 1994 Res175: 184–191, 1988 37. Oshima H, Yoshie Y, Auriol S, Gilibert I: Antioxidant and pro-oxidant 18. Olive PL, Banath JP, Durand RE: Heterogeneity in radition-induced actions of flavonoids: Effects on DNA damage induced by nitric DNA damage and repair in tumor and normal cells measured using the oxide, peroxynitrite and nitroxyl anion. Free Radical Biol Med 25: comet assay. Radiat Res 112: 86–94, 1990 1057–1065, 1998 19. Narendra P, Raymond RT, Ralph ES, Edward LS: A microgel elec- 38. Mccarthy PJ, Sweetman SF, Mckenna PG, Mckelvey-Martin VJ: trophoresis technique for the direct quantitation of DNA damage and Evaluation of manual and image analysis quantitation of DNA damage repair in individual fibroblasts cultured on microscope slides. Mutat Res in the alkaline comet assay. Mutagenesis 12: 209–214, 1997 252: 289–296, 1991 39. Pieur C, Rigaud J, Cheynier V,Moutounet M: Oligomeric and polymeric 20. Ross GM, Mcmillan JJ, Wilcox P, Collins AR: The single cell microgel procyanidins from grape seeds. Phytochemistry 36: 781–784, 1994 electrophoresis assay (comet assay): Technical aspects and applications. 40. Halliwell B, Aruoma OI: DNA damage by oxygen-derived species: Its Mutat Res 337: 57–60, 1995 mechanisms and measurement in mammalian systems. FEBS Lett 281: 21. Duthie SJ, Collins AR, Duthie GG, Dobson VL: Queercetin and 9–19, 1991 myricetin protect against hydrogenperoxide-induced DNA damage 41. Yen GC, Chen HY, Peng HH. Antioxidant and pro-oxidant effects of (strand breaks and oxidized pyrimidines) in human lymphocytes. Mutat various tea extracts. J Agric Food Chem 45: 30–34, 1997 Res 393: 223–231, 1997 42. Sahu SC, Washington MC: Effects of antioxidants on quercerin-induced 22. Duthie SJ, Johnson W, Dobson VL: The effect of dietary flavonoids on nuclear DNA damage and lipid peroxidation. Cancer Lett 60: 259–264, DNA damage (strand breaks and oxidized pyrimidings) and growth in 1991 human cells. Mutat Res 390:141–151, 1997 43. Giovannelli L, Testa G, De Filippo C, Cheynier V, Clifford MN, Dolara 23. Duthie SJ, Dobson VL: Dietary flavonoids protect human coloncyte P. Effect of complex polyphenols and tannins from red wine on DNA ox- DNA from oxidative attack in vitro. Eur J Nutr 38:28–34, 1999 idative damage of rat colon mucosa in vivo. Eur J Nutr 39: 207–212, 2000