Mutagenesis, Vol. 17, No. 1, 45-53,
January 2002
© 2002 UK Environmental Mutagen Society/Oxford University Press
Flavonoids inhibit genetic toxicity produced by carcinogens in cells expressing CYP1A2 and CYP1A1
School of Biosciences, University of Birmingham, Birmingham B15 2TT, UK and 1 GenPharmTox Biotech AG, Fraunhoferstrasse 9, D-82152 Martinsreid/Planegg, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The effects of the flavonoids quercetin, apigenin and chrysin (10 µM) on the genetic toxicity of 2-amino-3-methylimidazo[4,5-f]quinoline (IQ) and benzo[a]pyrene (BaP) was investigated at sub-cytotoxic concentrations in Chinese hamster V79 cells expressing human or rat cytochromes P450. In V79 r1A2-NH and V79 h1A1-MZ cells, none of the flavonoids increased DNA strand breaks (SB) (measured by the Comet assay) or produced detectable DNA adducts (measured by 32P-post-labelling). Neither IQ nor BaP produced DNA damage in the absence of expressed CYP1A2 or CYP1A1, respectively. DNA damage measured as SB and DNA adducts was detectable in V79 r1A2-NH cells expressing rat CYP1A2 when treated with IQ (2.5–50 µM) and this was inhibited by quercetin. Likewise, DNA damage (SB and DNA adducts) was elevated in V79 h1A1-MZ cells expressing human CYP1A1 when treated with BaP (0.1–0.5 µM) and this was inhibited by chrysin and apigenin, but not by quercetin. The specificity of CYP1A1 inhibition by chrysin and apigenin and CYP1A2 inhibition by quercetin was confirmed by ethylresorufin O-deethylase assay.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Diet has a major influence on cancer (Wynder and Gori, 1977
) and it has been estimated that it may be possible to reduce cancer death rates through dietary modulation by as much as 35% (Doll and Peto, 1981). Both epidemiological and experimental data indicate protective effects of dietary constituents in vegetables against cancer (Hayatsu et al., 1988; Steinmetz and Potter, 1991; Doll, 1992; Steinmetz et al., 1993; Ferguson, 1994; Stavric, 1994; Wattenberg, 1985). Minor dietary non-nutrients are often found in foods of plant origin and a number of these have been considered as potential cancer chemopreventors (Wattenberg, 1985).
Flavonoids are plant derivatives of flavone with various degrees of hydroxylation and glycosidic substitutions (Figure 1; Kuhnau, 1976). These compounds are distributed ubiquitously in plants and are present in edible vegetables and fruits (Jongen and Dorgelo, 1986; Pierpoint, 1986; Rimm et al., 1996), although intake can vary widely (Hertog et al., 1995). Flavonoids have been reported to possess a wide range of biochemical and pharmacological activities, both potentially detrimental and protective (Bjeldanes and Chang, 1977; Rueff et al., 1986; NTP, 1991; Hertog and Holman, 1996; Hollman and Kata, 1999). One of the effects of flavonoids is the ability to modulate xenobiotic metabolism (Huang et al., 1983; Sousa and Marletta, 1985; Seiss et al., 1989; Chae et al., 1992; Lee et al., 1994; Obermeier et al., 1995; Kanazawa et al., 1998; Moon et al., 1998; Yannai et al., 1998; Zhai et al., 1998). These various studies indicate that a potential basis for protection is interference with enzymes such as specific cytochrome P450 (CYP450) forms that play an important role in the metabolic activation of a wide range of known carcinogens (Gonzalez and Gelboin, 1994). The activity of the specific form(s) of CYP450 involved in activation is likely to be a major determinant of susceptibility to carcinogens and will vary between individuals due to dietary habits as well as genetic characteristics. One of the problems associated with studies on the inhibition of CYP450 forms in rodent tissues is that the metabolism of probe substrates is often mediated by multiple enzyme systems, the predominant form depending on the substrate concentration. Inducers and inhibitors are rarely specific, thus it is difficult to establish the specific form of enzyme affected as well as difficult to determine the mechanism of enzyme inhibition and consequently the effect of low concentrations. The latter is important to make predictions about the relevance of human exposure to the concentrations of agents which are bioavailable from the diet. Furthermore, the substrate specificity of rodent CYP450 forms can differ markedly from that of human forms. These points indicate that findings in model rodent systems or even in human tissues are not easily extrapolated to predict the dietary influence of chemopreventors in man.
|
A recent approach to overcome some of these problems has been the use of cell lines expressing specific rodent or human CYP450 genes (Ellard et al., 1991
; Gonzalez et al., 1991; Langenbach et al., 1992; Doehmer, 1993). Using genetically engineered V79 cell lines expressing rat or human CYP1A1 or CYP1A2, the effects of the three flavonoids quercetin (flavonol), apigenin and chrysin (flavones) on carcinogen activation have been investigated. A polycyclic aromatic hydrocarbon, benzo[a]pyrene (BaP), was chosen as a CYP1A1 substrate and an aromatic amine, 2-amino-3-methylimidazo[4,5-f]quinoline (IQ), was used as a CYP1A2 substrate. Both compounds are known as procarcinogens which need CYP450-mediated activation to produce DNA adducts and thus genetic toxicity (Eisenbrand and Tang, 1993; Guengerich, 1994). The single cell gel electrophoresis (Comet) assay was chosen as a preliminary non-specific measure of DNA damage. Further studies of the effect of flavonoids on DNA adduct formation were conducted using the 32P-post-labelling technique. This approach, coupled with enzyme kinetic analyses, has allowed us to determine the influence of relatively low concentrations of the three flavonoids on activation of the carcinogens considered in the human and rat. |
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Cell cultures
The genetically engineered cell lines V79 h1A1-MZ, expressing human CYP1A1, V79 r1A1-MZ, expressing rat CYP1A1, V79 h1A2-NH, expressing human CYP1A2, V79 r1A2-NH, expressing rat CYP1A2, and the parental cell lines V79-NH, expressing an endogenous acetyltransferase activity, and V79-MZ, lacking acetyltransferase activity, were previously developed (Ellard et al., 1991
; Doehmer, 1993). These cell lines were tested for Mycoplasma contamination using a Mycotect kit (Gibco BRL).
Cell lines were cultured in Dulbecco's modified Eagle's medium (DMEM) (Sigma) with glucose supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Sigma) and 2 mM L-glutamine (Sigma). For selection of plasmid-containing cells, lines V79 h1A1-MZ, r1A1-MZ, h1A2-NH and r1A2-NH were maintained in DMEM supplemented with 0.4 mg/ml geneticin disulfate salt (G418) (Sigma). Cells were grown at 37°C in a humidified atmosphere (5% CO2 in 95% air). Prior to chemical treatment, the culture medium was discarded and replaced with serum-free DMEM. Cell viability was measured as release of lactate dehydrogenase (LDH) (Moldeus et al., 1978) or by the MTT assay (Lobner, 2000).
Chemical treatments
All chemicals were purchased from Sigma except IQ, which was purchased from Toronto Research Chemicals. IQ, apigenin and chrysin stock solutions were stored in dimethylsulfoxide (DMSO) (Sigma) at –20°C. All other chemicals were prepared on the day of use.
BaP, quercetin, apigenin and chrysin were dissolved in DMSO; cytosine ß-D-arabinofuranoside hydrochloride (araC) was dissolved in sterile phosphate-buffered saline (PBS).
Cells used for treatments were previously cultured for 48 (Comet assay) or 72 h (DNA adducts) initiated from 104 cells/ml in 25 (Comet assay) or 75 cm2 (DNA adducts) flasks. Cultures were pre-treated with 10 mg/ml araC for 1 h prior to carcinogen treatment (Comet assay) to inhibit religation of strand breaks, thus maximizing sensitivity.
Cultures were incubated for 1 (Comet assay) or 6 h (DNA adducts) with the test agents in the presence or absence of a flavonoid. The DMSO concentration used was 0.3%, which was also used in controls. H2O2 (100 mM) was used as a positive control. Concentrations of BaP tested ranged from 0.01 to 0.5 µM and concentrations of IQ tested ranged from 2.5 to 50 µM. Each of the flavonoids was tested at a concentration of 10 µM based on enzyme inhibition kinetics. Each treatment was performed in duplicate. After treatment, culture medium was discarded. The cells were rinsed with sterile PBS before being harvested by trypsinization and resuspended in supplemented DMEM.
Single cell gel electrophoresis (Comet assay)
All the following procedures were conducted in a dark room under yellow light to prevent the occurrence of additional DNA damage.
Slide preparation
Treated cells, harvested as previously described, were pelleted by centrifugation at 250 g for 5 min and resuspended in 100 µl of PBS. The basic alkali technique of Singh et al. (1988), as modified by Anderson et al. (1994), was followed. Each treated cell suspension was used to prepare two slides, i.e. 4 slides/treatment.
Electrophoresis
The slides were removed from the lysing solution and incubated for 20 min in an electrophoresis buffer consisting of 300 mM NaOH (Fisher) and 1 mM disodium EDTA in water. Electrophoresis was then conducted at room temperature for 20 min at 25 V/320 mA. At the end of electrophoresis the slides were drained and the excess alkali neutralized by covering each slide with 0.4 M Tris buffer, pH 7.5, for 5 min. This neutralizing procedure was repeated three times.
Staining
Each slide was stained by adding 50 µl of a 20 µg/ml ethidium bromide (2,7-diamino-10-ethyl-9-phenyl phenanthridinium bromide; Sigma) solution in water and covered with a coverslip. The slides were placed on humidified tissues in an air-tight box and kept in the dark at 4°C before image analysis.
Image analysis
Image analysis was completed within 48 h following the Comet assay. The slides were examined at 20x magnification with an Axiovert 10 fluorescence microscope, an excitation filter of 546 nm and a barrier filter of 590 nm (Zeiss). The microscope was attached to a Pulnix TM-765 AGC CCD video camera connected to a computer-based image analysis system. For each slide 50 `comets' were randomly captured (100–150 images analysed per treatment). Parameters and comets were analysed using Komet 3.0 software (Kinetic Imaging). The degree of DNA damage was evaluated either as the mean percentage of tail DNA or according to the classification of Anderson et al. (1994), dividing the cells into four categories depending on the percentage of tail DNA. Only the mean values are given in Results.
Statistical analysis
The comet data have been described as not following a normal distribution (Bauer et al., 1998; Pohl and Fouts, 1980). Therefore, the data were analysed with the non-parametric Mann–Whitney test using Minitab software with n = 4, 6 or 9 and total cell population (n = 100). Enzyme kinetic data were analysed by ANOVA.
Ethylresorufin O-deethylase (EROD) assay
Metabolism of O-ethylresorufin to resorufin was used as a probe to reflect CYP1A1 and CYP1A2 activities (Pohl and Fouts, 1980). V79 h1A1-MZ or V79 r1A2-NH cell cultures were initiated by seeding 104 cells/ml supplemented DMEM in 25 cm2 flasks. After 48 h the cells were incubated for 1 h at 37°C, 5% CO2 with concentrations of ethoxyresorufin (resorufin ethyl ether; Sigma) ranging from 0.2 to 16 µM in the absence or presence of either 10 µM quercetin, 10 µM apigenin or 10 µM chrysin. The cells were then harvested by trypsinization and total protein was measured using the Bradford assay (Bradford, 1976).
The EROD assay was conducted using cell culture supernatants according to the method of Paterson et al. (1984). Supernatant (4 ml) was mixed with 1 ml of 0.2 M acetate buffer, pH 4.5, and 2 ml of water. Two millilitres of this mixture were added to 0.5 ml of 1 mg/ml ß-glucuronidase in acetate buffer and incubated for 2 h at 37°C. Resorufin was extracted by adding 4 ml of diethyl ether containing 1.5% (v/w) isoamyl alcohol and samples were further incubated for 10 min at 37°C. The organic layer was back-extracted in 4 ml of 0.05 M glycine buffer, pH 10.4, at 4°C overnight. The resulting organic layer was used to measure resorufin by fluorimetry using excitation and emission wavelengths of 530 and 590 nm, respectively, and by comparison with a standard curve of resorufin (7-hydroxy-3H-phenoxazin-3-one; Sigma) for concentrations ranging from 0 to 0.64 µg/ml.
DNA adducts
DNA isolation from cell cultures
Treated cells harvested as previously described were washed twice in PBS. The final pellet was resuspended in 500 µl of 10 mM Tris–HCl, 1 mM EDTA and 1% SDS, pH 8.0, and 3 µl of 50 µg/ml (7.5 µU) RNase A (EC 3.1.27.5; Sigma) were then added before incubation for 1 h at 37°C on a shaker. An aliquot of 12.5 µl of 20 mg/ml (3 U) proteinase K (EC 3.4.21.64; Sigma) was added and the mixture was further incubated for 1 h at 37°C on a shaker. Purification of DNA was achieved by two extractions of this mixture with an equal volume of 25:24:1 (v/v) phenol/chloroform/isoamyl alcohol, pH 8.0 (Gibco BRL), followed by extraction with an equal volume of chloroform/TE, pH 8.0. DNA was allowed to precipitate by addition of 2.5 vol of ice-cold pure ethanol and 0.1 vol of 0.3 M sodium acetate. The DNA was finally dissolved in water and the absence of RNA was confirmed by gel electrophoresis. DNA concentration and purity were determined spectrophotometrically by measuring absorbance at 260 and 280 nm. DNA yields were between 100 and 200 µg. Samples were stored at –20°C.
DNA isolation from rat liver
Positive control DNA was prepared from treated rat liver. A 240 g male Wistar rat was injected i.p. with a single dose of 50 mg/kg 2-acetylaminofluorene (2-AAF) (Sigma) and killed after 24 h. The rat liver was collected and frozen at –70°C. Negative control DNA was obtained from the liver of an untreated 240 g male Wistar rat and from untreated V79 cells. Both rat livers were defrosted and homogenized in 8 vol of ice-cold buffer comprising 0.15 M NaCl and 0.015 M trisodium citrate, pH 7.0, then centrifuged at 400 g for 15 min. The supernatant was discarded and the pellet containing nuclei dispersed in 8 vol of 0.01 M Tris–HCl, 0.001 M EDTA and 1 M NaCl, pH 7.0. An equal volume of proteinase K buffer containing 5 mM EDTA and 0.5% sarcosyl, pH 8.0, and 100 µg/ml (24.4 U) proteinase K was added. This mixture was incubated at 50°C for 2 h on a shaker. The DNA was then extracted with phenol/chloroform/isoamyl alcohol, pH 8.0, as previously described for the cell cultures and the resulting DNA pellet was resuspended in 600 µl of RNase A buffer comprising 50 mM Tris–HCl and 10 mM NaCl, pH 8.0, and 3 µl of 50 µg/ml (7.5 µU) RNase A were added before incubation at 37°C for 30 min on a shaker. DNA from both rat livers was then extracted, precipitated and their purities and concentrations determined as described above.
32P-post-labelling of DNA adducts
The original method of 32P-post-labelling of DNA adducts (Gupta et al., 1982) with the butanol extraction enhancement protocol (Gupta, 1985) was used. DNA samples of 10 µg were digested to deoxynucleoside 3'-monophosphates. An aliquot of 1 µl of the digest was stored at –20°C for quantification of normal (undamaged) total nucleotides. The remaining 29 µl of the digest was diluted to the equivalent of 0.02 µg DNA/µl with aqueous 100 mM ammonium formate, pH 3.5, and 10 mM tetrabutylammonium chloride. DNA adducts were extracted twice by adding an equal volume of water-saturated butan-1-ol and back-extracted three times with an equal volume of water. The pH was adjusted by adding 3 µl of 200 mM Tris–HCl, pH 9.5, and the butanol extract was evaporated to dryness in a vacuum centrifuge prior to labelling with [
-32P]ATP (>5000 Ci/mmol, 10 µCi/µl; Amersham). The 5'-phosphorylation of adducted deoxynucleoside 3'-monophosphates with 32P was obtained by incubation at 37°C for 2 h.
32P-post-labelling of normal total nucleotides
The remaining 1 µl aliquot of the DNA digest kept as previously described was diluted to the equivalent of 0.1 ng DNA/µl with water and labelled with [-32P]ATP. One microlitre of 40 mU/µl potato apyrase (EC 3.6.1.5; Sigma) was then added to eliminate the excess ATP by incubation for 30 min at 37°C. The mixture was then diluted with a buffer comprising 10 mM Tris–HCl and 5 mM EDTA, pH 9.5, to the equivalent of 0.01 ng DNA before proceeding to thin layer chromatography.
Thin layer chromatography of 32P-labelled normal total nucleotides and adducts
32P-labelled normal and adducted nucleotides were separated on 20x10 cm plastic-backed PEI–cellulose TLC plates (Macherey-Nagel). For the normal total nucleotides the equivalent of 0.01 ng DNA diluted as previously described was applied 2 cm from the bottom of the plate and developed in D1 (to the top of the plate) for 2 h in 0.12 M sodium phosphate, pH 6.8.
The plates were developed in D1 with 1 M sodium phosphate, pH 6.0, overnight and the upper halves of the plates including the wicks were discarded. The remaining half-plates were washed in water for 10 min, then transferred to 0.1 M ammonium formate, pH 3.5, for 5 min to adjust the pH before being washed in water for 5 min. The TLC plates were dried and developed in D3 (rotation of 180° compared to D1) in 3 M lithium formate and 7 or 8.5 M urea, pH 3.5, for 16 h. The urea concentration used for the putative BaP–DNA adducts was 8.5 M, whereas 7 M was used to detect IQ–DNA adducts. The TLC plates were washed twice in 13 mM Tris base for 10 min, then washed in water for 5 min before being dried. The plates were developed in D4 (rotation of 90° compared to D3) in 0.8 M lithium chloride, 0.5 M Tris–HCl and 7 or 8.5 M urea, pH 8.0, for 6 h before being washed in water for 10 min and dried. Finally, a new wick was attached to the TLC plates which were developed in D5 (same direction as D4) in 0.35 M MgCl2 for 3 h (3–4 cm onto the paper wick).
The TLC plates were exposed to a storage phosphor screen for 3 h before being scanned by a phosphorimager (Molecular Dynamics). The results were processed using ImageQuant software (Molecular Dynamics).
Calculation of adduct levels
Relative adduct labelling (RAL) was calculated as previously described by Gupta (1985). Briefly, since adducts are evaluated from 10 µg DNA and total nucleotides are calculated from 0.01 ng DNA, RAL = c.p.m. adductsx 10–6/c.p.m. total nucleotides. The RAL values can be translated into fmol adducts/µg DNA by multiplying RALx107x0.3, assuming that 1 µg DNA = 0.3x107 fmol nucleotides. Therefore, adduct level in fmol adducts/µg DNA = c.p.m. adductsx3/c.p.m. total nucleotides.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Comet assay
Effects of flavonoids on the V79 cell lines The mean percentage tail DNA of V79 r1A2-NH and V79 h1A1-MZ cells was not significantly changed from control values (19.2 ± 3.3 and 16.7 ± 0.7% tail DNA, respectively) in the presence of 10 µM quercetin, 10 µM apigenin and 10 µM chrysin (24.4 ± 2.1, 21.2 ± 3.1 and 17.5 ± 0.4% tail DNA, respectively), although in one experiment there was evidence of a small inhibitory effect of quercetin (see below).
Effects of IQ and BaP in the V79 parent cell lines
Concentrations of IQ ranging from 2.5 to 50 µM did not significantly affect the mean per cent tail DNA values for V79-NH cells. Neither did concentrations of BaP ranging from 0.01 to 0.5 µM significantly affect the tail DNA in the V79-MZ cell line (maximum per cent tail DNA throughout was 18.5 ± 0.1% compared with a control value of 17.4 ± 0.7%). This indicates a requirement for heterologous expression of the respective CYP450 for metabolic activation.
CYP1A2: effect of quercetin on DNA damage induced by IQ
IQ induced DNA damage in the V79 r1A2-NH cell line, with an increasing response with increasing concentration (2.5–50 µM; Figure 2). The mean per cent tail DNA values for 25 (25.8 ± 1.4) and 50 µM IQ (25.3 ± 1.9) were significantly increased (P < 0.01) above controls (18.5 ± 0.1% tail DNA). DNA damage induced by 100 mM H2O2 as a positive control was 43.7 ± 3.0% tail DNA.
|
0.05, ***P In the presence of IQ plus 10 µM quercetin the mean per cent tail DNA was significantly reduced compared with IQ alone (17.3 ± 1.6 versus 25.8 ± 1.4 at 25 µM, P < 0.01; 15.20 ± 1.17 versus 25.3 ± 1.9 at 50 µM, P < 0.001). There was, indeed, some evidence for a reduced background level, which may relate to its known antioxidant capacity (Huang et al., 1983
; Musonda et al., 1997), although this was not investigated further. In the V79 h1A2-NH cell line IQ did not induce a significant dose-related response (data not shown) and thus the sensitivity of assays with this cell line was insufficient to study the inhibitory effects of flavonoids. Concentrations of IQ and quercetin ranging from 0 to 100 µM were found to be non-cytotoxic in both the V79 r1A2-NH and V79 h1A2-NH cell lines using the MTT assay (>90% cell viability; data not shown).
CYP1A1: effect of quercetin, apigenin and chrysin on DNA damage induced by BaP
Concentrations of 0.1 and 0.5 µM BaP increased the percentage of affected V79 h1A1-MZ cells and this was dependent on the concentration of BaP. BaP significantly increased (P < 0.001) the mean per cent tail DNA (to 27.9 ± 3.2) compared with the control value (18.7 ± 2.7) and approached the response given by the positive control (100 mM H2O2, 37.2 ± 4.9% tail DNA). However, 10 µM quercetin had no significant inhibitory effect on the mean percent tail DNA produced by BaP (Figure 3).
|
Similar results were observed for V79 r1A1-MZ cells exposed to the same range of BaP concentrations in the presence or absence of 10 µM quercetin (data not shown). For both cell lines concentrations of BaP ranging from 0 to 0.5 µM were found to be non-cytotoxic using the LDH assay (>90% cell viability; data not shown). Concentrations of quercetin ranging from 0 to 100 µM were also found to be non-cytotoxic (>90% of cell viability) for both cell lines using the MTT assay (data not shown).
The effects of chrysin on BaP- and H2O2-induced DNA damage in the V79h1A1-MZ cell line are shown in Figure 4. The mean per cent tail DNA for V79 h1A1-MZ cells exposed to 0.5 µM BaP (29.9 ± 3.0) was significantly (P < 0.01) reduced, to 17.4 ± 2.9, by 10 µM chrysin. In contrast, no significant effect of 10 µM chrysin on the mean per cent tail DNA was observed in cells treated with 100 mM H2O2.
|
In the presence of 10 µM apigenin the mean per cent tail DNA was also significantly decreased (19.2 ± 1.0, P < 0.005) compared with cells exposed to 0.5 µM BaP alone (35.3 ± 3.2) (Figure 5
). As with chrysin, no significant effect of apigenin on the mean per cent tail DNA was observed for V79 h1A1-MZ cells treated with 100 mM H2O2 (37.2 ± 4.9) compared with cells exposed to H2O2 alone (36.0 ± 7.5).
|
Concentrations of chrysin and apigenin ranging from 0 to 20 µM were found to be non-cytotoxic to the V79 h1A1-MZ cell line using the MTT assay (>90% of cell viability; data not shown).
EROD assay
Effect of quercetin on rat CYP1A2 activity
In the absence of quercetin kinetic parameters for EROD were calculated as Km = 0.26 ± 0.06 µM and Vmax = 83 ± 10 pmol/min/mg protein. In the presence of 10 µM quercetin these values were Km = 1.00 ± 0.34 µM and Vmax = 68 ± 8 pmol/min/mg protein (Figure 6A). The significant (P < 0.01) increase in Km value in the presence of quercetin coupled with the unchanged Vmax value suggests that this flavonoid competitively inhibits rCYP1A2 activity. A separate, more detailed study was also completed and the data plotted as Eadie–Hofstee plots (not shown). This gave similar values of Km = 0.45 ± 0.08 µM and Vmax = 70.1 pmol/min/mg. Vmax (71.0 pmol/min/mg) was not affected by 10 µM quercetin but Km (2.1 µM) was again significantly (P < 0.01) elevated. We also noted that hCYP1A2 was affected in a similar way: Km increased from 0.73 to 2.7 µM whereas Vmax was unaltered (109 and 96 pmol/min/mg for control and treated, respectively).
|
Effect of apigenin and chrysin on human CYP1A1 activity
In the absence of chrysin kinetic parameters for EROD were calculated as Km = 0.25 ± 0.10 µM and Vmax = 1165 ± 291 pmol/min/mg protein. In the presence of 10 µM chrysin these values were Km = 1.18 ± 0.16 µM and Vmax = 144 ± 38 pmol/min/mg protein (Figure 6B
). The significantly (P < 0.01) increased Km value and significantly (P < 0.01) reduced Vmax suggest a mixed type inhibition of hCYP1A1 by chrysin.
In a study to determine the effect of apigenin the Km and Vmax values in the absence of apigenin were Km = 0.35 ± 0.10 µM and Vmax = 1167 ± 303 pmol/min/mg protein. In the presence of 10 µM apigenin these values were Km = 8.10 ± 2.12 µM and Vmax = 806 ± 403 pmol/min/mg protein (Figure 6C). Km was significantly (P < 0.01) increased while the lowering of Vmax was not statistically significant, thus suggesting a predominantly competitive inhibition exerted by apigenin on hCYP1A1. As with quercetin, Eadie–Hofstee plots (not shown) supported these conclusions.
DNA adducts
The effects of apigenin and chrysin on BaP-induced DNA adducts are presented in Figure 7A. No DNA adducts were found in V79 h1A1-MZ cells treated with DMSO (Figure 7A). One major adduct spot along with some minor regions were observed after exposure of cells to 0.5 µM BaP (Figure 7B). The RAL value, calculated as previously described (Gupta, 1985), was 5.78 fmol/µg DNA for the major spot. Cells treated with BaP and 10 µM apigenin (Figure 7C) or 10 µM chrysin (Figure 7D) showed no detectable adducts. The absence of detectable adducts was also observed for cells treated with apigenin or chrysin alone (Figure 7E and F).
|
Figure 7B
also shows the adduct patterns for the V79 r1A2-NH cell line. Controls with DMSO indicated the absence of adducts (Figure 7G). One adduct (spot 1) was detected in the cells treated with 50 µM IQ with a RAL of 3.13 fmol/µg DNA (Figure 7H). This spot was not observed in the presence of IQ and 10 µM quercetin (Figure 7I). When treated with quercetin alone cells also did not exhibit detectable levels of adduct spots (Figure 7J).
A positive control (Figure 7K) was provided by the adduct pattern of liver DNA from a male Wistar rat pretreated with 50 mg/kg 2-AAF i.p. for 24 h. Two major spots were detected, namely spots 1 and 2, with RAL of 3.52 and 1.38 fmol/µg DNA, respectively.
Figure 7L depicts separation of normal nucleotides and unused ATP. This confirmed that the butanol-extracted adducts mixed with residual normal nucleotides were 32P-labelled in the presence of excess [-32P]ATP. Three radioactive spots were observed for normal nucleotides: guanosine (G), adenosine + thymidine (A+T), cytosine + methylated cytosine (C+m5C) and the inorganic phosphate spot (Pi).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
We have previously demonstrated an inhibitory effect of quercetin on drug metabolizing enzymes and oxygen radicals in a HepG2 cell line (Musonda et al., 1997
). Quercetin was shown to exert an inhibitory effect on human liver CYP450 activity towards ethoxycoumarin and ethylresorufin at micromolar concentrations. Yet, because of multiple enzyme involvement for these substrates, the mechanism of inhibition and isoenzyme specificity has not been determined. CYP1A2 is known to be expressed constitutively in the human liver, whereas CYP1A1 is essentially extrahepatic. Therefore, CYP1A2 was proposed as a likely candidate for the inhibitory effects of quercetin. Siess et al. (1995) have also reported inhibition of human EROD by quercetin (IC50 12 µM) and Bear and Teal (2000) found an association between the ability of flavonoids to inhibit rat CYP1A2 and protection against the bacterial mutagenicity of heterocyclic amines.
The results of the present study indicate that quercetin inhibits DNA damage mediated by rat CYP1A2 (IQ) but not human CYP1A1 (BaP). In contrast, both chrysin and apigenin exhibited inhibiting effects specific for human CYP1A1-mediated DNA damage, which was confirmed by enzymatic studies. Enzyme kinetic studies confirmed that the inhibitory effects of quercetin on DNA damage induced by BaP are associated with competitive inhibition of rat CYP1A2, and a similar enzyme inhibition was seen with the human form. However, the possibility of differences between the rodent and human forms is recognized and additional mechanisms of modulation of carcinogen activating enzymes may exist. Particularly relevant is the inhibition of CYP1A1 gene expression by quercetin in human HepG2 cells (Kang et al., 1999), with a consequent reduction in BaP-derived DNA adducts.
It is recognized that the C2,C3 double bond and the 5-hydroxyl groups on the B ring of the flavonoid molecule (Figure 1) are important for flavonoid inhibition of aryl hydroxylation (Lee et al., 1994). It has been proposed that quercetin may act by uncoupling CYP450 reactions and alter the binding site (Sousa and Marletta, 1985; Smith and Gupta, 1999). Moreover, quercetin and other hydroxylated flavonoids have been shown to inhibit the reduction of CYP450 (Buening et al., 1981).
A study of the effects of a wide range of potential chemopreventative agents on rat microsomes exposed to BaP and aflatoxin showed inhibition of DNA adduct formation at low levels (Buening et al., 1981; Shah and Bhattacharya, 1986; Bhattacharya and Firozzi, 1988; Smith and Gupta, 1999). Quercetin was found to be active at 100 µM (Shah and Bhattacharya, 1986). The unsubstituted flavone showed moderate inhibition of adduct formation while flavanones with a saturated C2,C3 double bond were inactive, as were most isoflavonoids and methylether derivatives of polyhydroxylated flavonoids. Structural features which contribute to the inhibitory activity of flavonoids towards CYP1A2 and IQ mutagenicity appear to be hydroxyl groups at positions C3', C4' and C5' on the B ring and C5 and C7 on the A ring (Figure 1) (Shah and Bhattacharya, 1986; Lee et al., 1994). Comparing different flavonoids with the same substituents, the order of antimutagenicity of these compounds towards IQ was as follows: flavones (apigenin) > flavonols (kaempferol) > flavanone (naringenin) > isoflavone (genistein) (Lee et al., 1994). The presence of a double bond at C2,C3 confers a co-planar conformation on the phenyl ring (ring B) with respect to the benzopyran rings (rings A–C). Planar flavonoid molecules preferentially interact with CYP1A1 and CYP1A2 and (through the latter) reduce metabolic activation of IQ. Flavones and flavonols were noted to be particularly potent in inhibition of 7-methoxy- and 7-ethoxyresorufin O-deethylase (Edenharder et al., 1997). Increased inhibitory effects on IQ mutagenicity was dependent on the number and position of hydroxyl groups. In our study, in contrast to the inhibitory effect of quercetin, apigenin did not inhibit IQ activation but preferentially inhibited CYP1A1 and activation of BaP. Methylation or glycosylation of hydroxyl groups rendered the flavonoids less active as inhibitors (Shah and Bhattacharya, 1986).
The influence on metabolic activation of BaP to the proximate carcinogen (±)-trans-7,8-dihydroxy-7,8-dihydrobenzo[a]pyrene has also been studied (Shah and Bhattacharya, 1986). The inhibition of metabolism by the flavonoids tested failed to correlate with their ability to inhibit adduct formation. These authors concluded that inactivation of benzo[a]pyrene metabolites was a major mechanism of inhibition of adduct formation. We now provide evidence that inhibition of CYP1A1 and CYP1A2 activities by chrysin, apigenin and quercetin, respectively, is correlated with the inhibition of DNA adduct formation.
Apigenin and chrysin were found to inhibit the hydroxylation of BaP in human liver microsomes with an IC50 of 50 µM (Smith and Gupta, 1999). Some naturally occurring flavonoids have been shown to inhibit CYP450 activities whereas other flavonoids exhibit a stimulatory activity towards CYP (Canivenc-Lanvier et al., 1996a,b). Buening et al. (1981) have proposed that flavonoids possessing hydroxyl groups tend to fall into the inhibitor category, whereas less polar flavonoid molecules lacking hydroxyl groups constitute CYP activators.
The modulating activities of naturally occurring compounds provide a system that can both activate and deactivate a wide range of xenobiotics and thus play a key role in chemoprevention (Wattenberg, 1985). Our results indicate that in this respect quercetin, chrysin and apigenin can be considered as potential inhibitors of chemical carcinogenesis. The precise mode of enzymatic inhibition will require further studies, in particular to determine the lowest CYP1A1 and CYP1A2 inhibiting concentrations of each of these flavonoids, in comparison to achievable hepatic concentrations in vivo, and to establish whether the inhibitory effects are direct or result from flavonoid products. Finally, it must be recognized that structural characteristics also need to be considered in relation to potential genotoxicity and co-genotoxicity of flavonoids. It has been reported that 3-hydroxylation of the C ring, 7-hydroxylation of the A ring and a catechol or pyrogallol group on the B ring favour genotoxicity of flavonols via auto-oxidation (Silva et al., 2000). Furthermore, the 3-hydroxyl group and a 2,3 double bond also appear important in the co-mutagenicity of flavonoids with 2-AAF (Ogawa et al., 1987).
|
Acknowledgments |
|---|
We thank the World Cancer Research Fund for financial support.
|
Notes |
|---|
2 To whom correspondence should be addressed. Email: j.k.chipman{at}bham.ac.uk
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
-
Anderson,D., Yu,T.W., Phillips,B.J. and Schmezer,P. (1994) The effect of various antioxidants and other modifying agents in oxygen-radical-generated DNA damage in human lymphocytes in the comet assay. Mutat. Res., 307, 261–271.[Web of Science][Medline]
Bauer,E., Recknagel,R.D., Fiedler,U., Wollweber,L., Bock,C. and Greulich,K.O. (1998) The distribution of the tail moments in single cell gel electrophoresis (comet assay) obeys a chi-square not a gaussian distribution. Mutat. Res., 398, 101–110.[Web of Science][Medline]
Bear,W.L. and Teal,R.W. (2000) Effects of citrus flavonoids on mutagenicity of heterocyclic amines and on cytochrome P450 1A2 activity. Anticancer Res., 5B, 3609–3614.
Bhattacharya,R.K. and Firozzi,P.F. (1988) Effect of plant flavonoids on microsome catalysed reactions of aflatoxin B1 leading to activation and DNA adduct formation. Cancer Lett., 39, 85–91.[Web of Science][Medline]
Bjeldanes,L.F. and Chang,G.W. (1977) Mutagenic activity of quercetin and related compounds. Science, 197, 577–578.
Bradford,M.M. (1976) A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein dye binding. Anal. Biochem., 72, 248–250.[Web of Science][Medline]
Buening,M.K., Chang,R.L., Huang,M.T., Fortner,J.G., Wood,A.W. and Conney,A.H. (1981) Activation and inhibition of benzo(a)pyrene and aflatoxin B1 metabolism in human liver microsomes by naturally occuring flavonoids. Cancer Res., 41, 67–7?.
Canivenc-Lanvier,MC, Bentejac,M., Miller,M.L., Leclerc,J., Siess,M.H., Latruffe,N. and Suschetet,M. (1996a) Differential effects of nonhydroxylated flavonoids as inducers of cytochrome P450 1A and 2B isozymes in rat liver. Toxicol. Appl. Pharm., 136, 348–353.[Web of Science][Medline]
Canivenc-Lanvier,M.C., Vernevaut,M.F., Totis,M., Siess,M.H., Magdalou,J. and Suschetet,M. (1996b) Comparative effects of flavonoids and model inducers on drug-metabolizing enzymes in rat liver. Toxicology, 114, 19–27.[Web of Science][Medline]
Chae,Y.H., Ho,D.K., Cassady,J.M., Cook,V.M., Marcus,C.B. and Baird,W.M. (1992) Effects of synthetic and naturally occurring flavonoids on metabolic activation of benzo(a)pyrene in hamster embryo cell cultures. Chem. Biol. Interact., 82, 181–193.[Web of Science][Medline]
Doehmer,J. (1993) V79 chinese hamster cells genetically engineered for cytochrome P450 and their use in mutagenicity and metabolism studies. Toxicology, 82, 105–118.[Web of Science][Medline]
Doll,R. (1992) The lessons of life: keynote address to the nutrition and cancer conference. Cancer Res., 52 (suppl.), 2024s–2029s.
Doll,R. and Peto,R. (1981) The causes of cancer: quantitative estimates of avoidable risks of cancer in the United States today. J. Natl Cancer Inst., 66, 1192–1265.
Edenharder,E., Rauscher,R. and Platt,K.L. (1997) The inhibition by flavonoids of 2-amino-3-methylimidazo[4,5-f]quinoline metabolic activation to a mutagen: a structure–activity relationship study. Mutat. Res., 379, 21–32.[Web of Science][Medline]
Eisenbrand,G. and Tang,W. (1993) Food-borne heterocyclic amines. Chemistry, formation, occurrence and biological activities. A literature review. Toxicology, 84, 1–82.[Web of Science][Medline]
Ellard,S., Yaseen,M., Dogra,S., Wolfel,C., Doehmer,J. and Parry,J.M. (1991) The use of genetically engineered V79 chinese hamster cultures expressing rat liver CYP1A1, 1A2 and 2B1 cDNAs in micronucleus assays. Mutagenesis, 6, 461–470.
Ferguson,L.R. (1994) Antimutagens as cancer chemopreventive agents in the diet. Mutat. Res., 307, 395–410.[Web of Science][Medline]
Gonzalez,F.J. and Gelboin,H.V. (1994) Role of human cytochromes P450 in the metabolic activation of chemical carcinogens and toxins. Drug Metab. Rev., 26, 165–183.[Web of Science][Medline]
Gonzalez,F.J., Crespi,C.L., Charles,L. and Gelboin,H.V. (1991) cDNA-expressed human cytochrome P450s: a new age of molecular toxicology and human risk assessment. Mutat. Res., 247, 113–127.[Web of Science][Medline]
Guengerich,F.P. (1994) Mechanisms of formation of DNA adducts from ethylene dihalides, vinyl halides and arylamines. Drug Metab. Rev., 26, 47–66.[Web of Science][Medline]
Gupta,R.C. (1985) Enhanced sensitivity of 32P-postlabeling analysis of aromatic carcinogen:DNA adducts. Cancer Res., 45, 5656–5662.
Gupta,R.C., Reddy,M.V. and Randerath,K. (1982) 32P-postlabeling analysis of non-radioactive aromatic carcinogen–DNA adducts. Carcinogenesis, 3, 1081–1092.
Gupta,R.C., Earley,K. and Sharma,S. (1988) Use of human peripheral blood lymphocytes to measure DNA binding capacity of chemical carcinogens. Proc. Natl Acad. Sci. USA, 85, 3513–3517.
Hayatsu,H., Arimoto,S. and Negishi,T. (1988) Dietary inhibitors of mutagenesis and carcinogenesis. Mutat. Res., 202, 429–446.[Web of Science][Medline]
Hertog,M.G.L. and Hollman,P.C.H. (1996) Potential health effects of the dietary flavonol quercetin. Eur. J. Clin. Nutr., 50, 63–71.[Web of Science][Medline]
Hertog,M.G., 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 seven countries study. Arch. Intern. Med., 155, 381–386.
Hollman,P.C.H. and Kata,K.B. (1999) Dietary flavonoids: intake, health effects and bioavaiability. Food Chem. Toxicol., 37, 937–942.[Web of Science][Medline]
Huang,M.T., Wood,A.W., Newmark,H.L., Sayer,J.M., Yagi,H., Jerina,D.M. and Conney,A.H. (1983) Inhibition of the mutagenicity of bay-region diol-epoxides of polycyclic aromatic hydrocarbons by phenolic plant flavonoids. Carcinogenesis, 4, 1631–1637.
Jongen,W.M.F. and Dorgelo,F.O. (1986) Naturally occuring carcinogens and modulating factors in food of plant origin. Neth. J. Agric. Sci., 34, 395–404.
Kanazawa,K., Yamashita,T., Ashida,H. and Danno,G.I. (1998) Antimutagenicity of flavones and flavonols to heterocyclic amines by specific and strong inhibition of the cytochrome P450 1A family. Biosci. Biotechnol. Biochem., 62, 970–977.[Medline]
Kang,Z.C., Tsai,S.-J. and Lee,H. (1999) Quercetin inhibits benzo(a)pyrene-induced adducts in human HepG2 cells by altering cytochrome P-450 1A1 gene expression. Nutr. Cancer, 35, 175–179.[Web of Science][Medline]
Kuhnau,J. (1976) The flavonoids. A class of semi-essential food components: their role in human nutrition. World Rev. Nutr. Diet., 24, 117–191.[Medline]
Langenbach,R., Smith,P.B. and Crespi,C. (1992) Recombinant DNA approaches for the development of metabolic systems used in in vitro toxicology. Mutat. Res., 277, 251–275.[Web of Science][Medline]
Lee,H., Wang,H.W., Su,H.Y. and Hao,N.J. (1994) The structure–activity relationships of flavonoids as inhibitors of cytochrome P-450 enzymes in rat liver microsomes and the mutagenicity of 2-amino-3-methyl-imidazo[4,5-f]quinoline. Mutagenesis, 9, 101–106.
Lobner,D. (2000) Comparison of the LDH and MTT assays for quantifying cell death: validity for neuronal apoptosis? J. Neurosci. Methods, 96, 147–152.[Web of Science][Medline]
Moldeus,P., Hogberg,J. and Orrenius,S. (1978) Isolation and use of liver cells. Methods Enzymol., 52, 60–71.[Medline]
Moon,J.Y., Lee,D.W. and Park,K.H. (1998) Inhibition of 7-ethoxycoumarin O-deethylase activity in rat liver microsomes by naturally occurring flavonoids: structure-activity relationships. Xenobiotica, 28, 117–126.[Web of Science][Medline]
Musonda,C.A., Helsby,N. and Chipman,J.K. (1997) Effects of quercetin on drug metabolizing enzymes and oxidation of 2',7-dichlorofluorescein in HepG2 cells. Hum. Exp. Toxicol., 16, 700–708.
NTP (1991) Technical Report on the Toxicology and Carcinogenesis Studies of Quercetin in F344/N Rats, NIH publication no. 91-3140. US Department of Health and Human Services, Public Health Service, Research Triangle Park, NC.
Obermeier,M.T., White,R.E. and Yang,C.S. (1995) Effects of bioflavonoids on hepatic P450 activities. Xenobiotica, 25, 575–584.[Web of Science][Medline]
Ogawa,S., Hirayama,T., Sumida,Y., Tokuda,M., Hirai,K. and Fuku,S. (1987) Enhancement of the mutagenicity of 2-acetylaminofluorene by flavonoids and the structural requirements. Mutat. Res., 190, 107–112.[Web of Science][Medline]
Paterson,P., Fry,J.R. and Horner,S. (1984) Influence of cytochrome P450 type on the pattern of conjugation of 7-hydroxycoumarin generated from 7-alkoxycoumarin. Xenobiotica, 14, 849–859.[Web of Science][Medline]
Pierpoint,W.S. (1986) Flavonoids in the human diet. Prog. Clin. Biol. Res., 213, 125–140.[Medline]
Pohl,R.J. and Fouts,J.R. (1980) A rapid method for assaying and metabolism of 7-ethoxyresorufin by microsomal subcellular fractions. Anal. Biochem., 107, 150–155.[Web of Science][Medline]
Rimm,E.B., Martijn,B.K., Asherio,A., Stampfer,M.J. and Willet,W.C. (1996) Relation between intake of flavonoids and risk for coronary heart disease in male health professionals. Ann. Intern. Med., 125, 384–389.
Rueff,J., Laires,A., Borda,H., Chaveca,T., Gomes,M.I. and Halpern,M. (1986) Genetic toxicology of flavonoids: the role of metabolic conditions in the induction of reverse mutation, SOS functions and sister-chromatid exchanges. Mutagenesis, 1, 179–183.
Shah,G.M. and Bhattacharya,R.K. (1986) Modulation of plant flavonoids and related phenolics of microsome catalysed adduct formation between benzo(a)pyrene and DNA. Chem. Biol. Interact., 59, 1–15.[Web of Science][Medline]
Siess,M.H., Leclerc,J., Canivenc-Lavier,M.C., Rat,P. and Suschetet,M. (1995) Heterogenous effects of natural flavonoids on monooxygenase activities in human and rat liver microsomes. Toxicology, 130, 73–78.[Web of Science]
Siess,M.H., Pennec,A. and Gaydou,E. (1989) Inhibition of ethoxy- and pentoxy-resorufin dealkylases of rat liver by flavones and flavonols: structure-activity relationship. Eur. J. Drug Metab. Pharmacokinet., 3, 235–239.
Silva,I.D., Gaspar,J., da Costa,G.G., Rodrigues,A.S., Laires,A. and Rueff,J. (2000) Chemical features of flavonols affecting their genotoxicity. Potential implications in their use as therapeutical agents. Chem. Biol. Interact., 124, 29–51[Web of Science][Medline]
Singh,N.P., McCoy,M.T., Tice,R.R. and Schneidel,E.L. (1988) A simple technique for quantitation of low-levels of DNA damage in individual cells. Exp. Cell, Res., 175, 184–191.[Web of Science][Medline]
Smith,W.A. and Gupta,R.C. (1999) Determining efficacy of cancer chemopreventative agents using a cell-free system concomitant with DNA adduction. Mutat. Res., 425, 143–152.[Web of Science][Medline]
Sousa,R.L. and Marletta,M.A. (1985) Inhibition of cytochrome P-450 activity in rat liver microsomes by the naturally occurring flavonoid, quercetin. Arch. Biochem. Biophys., 240, 345–357.[Web of Science][Medline]
Steinmetz,K.A. and Potter,J.D. (1991) Vegetables, fruit and cancer. I. Epidemiology. Cancer Causes Control, 2, 325–357.[Web of Science][Medline]
Steinmetz,K.A., Potter,J.D. and Folsom,A.R. (1993) Vegetables, fruit and lung cancer in the Iowa women's health study. Cancer Res., 53, 536–543.
Stravric,B. (1994) Antimutagens and anticarcinogens in foods. Food Chem. Toxicol., 32, 79–90.[Web of Science][Medline]
Wattenberg,L.W. (1985) Chemoprevention of cancer. Cancer Res., 45, 1–8.
Wynder,E.L. and Gori,G. (1977) Contribution of the environment to cancer incidence: an epidemiologic exercise. J. Natl Cancer Inst., 58, 825–832.
Yannai,S., Day,A.J., Williamson,G. and Rhodes,M.J.C. (1998) Characterization of flavonoids as monofunctional or bifunctional inducers of quinone reductase in murine hepatoma cell lines. Food Chem. Toxicol., 36, 623–630.[Web of Science][Medline]
Zhai,S., Dai,R., Wei,X., Friedman,F.K. and Vestal,R.E. (1998) Inhibition of methoxyresorufin demethylase activity by flavonoids in human liver microsomes. Life Sci., 63, 119–123.
Received on June 5, 2001; accepted on August 15, 2001.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  L. C. Wilms, A. W. Boots, V. C.J. de Boer, L. M. Maas, D. M.F.A. Pachen, R. W.H. Gottschalk, H. B. Ketelslegers, R. W.L. Godschalk, G. R.M.M. Haenen, F. J. van Schooten, et al. Impact of multiple genetic polymorphisms on effects of a 4-week blueberry juice intervention on ex vivo induced lymphocytic DNA damage in human volunteers Carcinogenesis, August 1, 2007; 28(8): 1800 - 1806. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. I. Haza, A. Zabala, N. Arranz, and P. Morales Protective Effect of a Lactobacillus salivarius Strain of Human Origin Food Science and Technology International, August 1, 2005; 11(4): 251 - 259. [Abstract] [PDF]
|
|
|
|
|
|
R. M. Lamuela-Raventos, A. I. Romero-Perez, C. Andres-Lacueva, and A. Tornero Review: Health Effects of Cocoa Flavonoids Food Science and Technology International, June 1, 2005; 11(3): 159 - 176. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||