xenobiotica, september 2003, vol. 33, no. 9, 913–925
Antioxidant and free radical scavenging activity of isoflavone metabolites G. RIMBACHy*, S. DE PASCUAL-TERESAy, B. A. EWINSy, S. MATSUGO§, Y. UCHIDA§, A. M. MINIHANEy, R. TURNERy, K. VAFEIADOUy and P. D. WEINBERGz yHugh Sinclar Human Nutrition Unit, School of Food Biosciences, University of Reading RG6 6AP, UK zSchool of Animal and Microbial Sciences, University of Reading, Reading RG6 6AP, UK §Department of Applied Chemistry and Biotechnology, Faculty of Engineering, Yamanashi University, Takeda, Kofu, Japan Received 20 March 2003 1. Soy isoflavones have been extensively studied because of their possible healthpromoting effects. Genistein and daidzein, the major isoflavone aglycones, have received most attention; however, they undergo extensive metabolism in the gut and liver, which might affect their biological properties. 2. The antioxidant activity, free radical-scavenging properties and selected cellular effects of the isoflavone metabolites equol, 8-hydroxydaidzein, O-desmethylangiolensin, and 1,3,5 trihydroxybenzene were investigated in comparison with their parent aglycones, genistein and daidzein. 3. Electron spin resonance spectroscopy indicated that 8-hydroxydaidzein was the most potent scavenger of hydroxyl and superoxide anion radicals. Isoflavone metabolites also exhibited higher antioxidant activity than parent compounds in standard antioxidant (FRAP and TEAC) assays. However, for the suppression of nitric oxide production by activated macrophages, genistein showed the highest potency, followed by equol and daidzein. 4. The metabolism of isoflavones affects their free radical scavenging and antioxidant properties, and their cellular activity, but the effects are complex.
Introduction Soy isoflavones have been extensively studied because of their potential beneficial effects in a number of disease processes (Bingham et al. 2003). Genistein and daidzein, the aglycones of the major soy isoflavones, have received considerably more attention than their metabolites, yet metabolism might be the key to understanding the beneficial effects of isoflavones since it is the metabolites, rather than the parent compounds, to which cells are predominantly exposed (Hendrich 2002). Isoflavones are ingested mainly in the conjugated form and undergo extensive hydrolysis by intestinal (jejunum) and bacterial (distal gut) b-glucosidases that release the principal aglycones, genistein and daidzein (Hur et al. 2002, Setchell et al. 2002). The aglycones can be absorbed directly from the small intestine by passive diffusion or can first undergo further biotransformation to a range of metabolites. Genistein is transformed to dihydrogenistein, which is *Author for correspondence; e-mail:
[email protected] Xenobiotica ISSN 0049–8254 print/ISSN 1366–5928 online # 2003 Taylor & Francis Ltd http://www.tandf.co.uk/journals DOI: 10.1080/0049825031000150444
914
G. Rimbach et al.
further metabolized to 60 -hydroxy-O-desmethylangolensin. Daidzein is metabolized to dihydrodaidzein and then transformed to both O-demethylangolensin (O-DMA) and equol (Joannou et al. 1995, Setchell et al. 1998). The formation of equol from daidzein is thought to be a particularly significant step, as the oestrogenic potency of equol is substantially higher than its precursor (Shutt and Cox 1972, Setchell 1998). Specific enzymes produced by a limited number of as yet largely unknown bacterial species in the colon appear to account for the above biotransformations (Bingham et al. 2003). Lampe et al. (2001) reported that faecal inoculums from equol-producing individuals caused a significant conversion of daidzein to equol and dihydrodaidzein in vitro. Antibiotics could inhibit equol and dihydrodaidzein production, clearly indicating the significance of the gut microflora. Coldham et al. (2002) reported the intermediate products of genistein incubated with faeces as dihydrogenistein and O-DMA, and the end products as 2-(4-hydroxyphenyl)propionic acid and 1,3,5-trihydroxybenzene (1,3,5-THB). Biotransformations occur in the liver as well as in the gut. The oxidative metabolism of genistein and daidzein was recently investigated in rat liver microsomes by Kulling et al. (2000). Both genistein and daidzein were extensively metabolized to monohydroxylated and dihydroxlylated products by cytochrome P450-dependent pathways. A major metabolite of daidzein was identified as 8-hydroxydaidzein (8-OH-daidzein). In both isoflavones, the additional hydroxyl groups were exclusively introduced into the ortho position of the existing phenolic hydroxyl groups.
HO
8
7
A 6
O C 4
3
2' 3'
5
B
O 6'
5'
4'
OH
O
OH
OH
Daidzein
Genistein
OH HO
O
HO
O
HO
2
O
O
OH
OH Equol
8-Hydroxydaidzein
HO
OH
OH 1,3,5-Trihydroxybenzene
HO
OH
O OH O-Demethylangolesin
Figure 1. Structure of the isoflavone test compounds. (Adapted from Bingham et al. 2003.)
Isoflavone metabolites
915
Gut and liver isoflavone metabolites are found in plasma and urine after soy consumption (Kulling et al. 2001). However, data about their chemical and biological properties are scarce. Although it is believed that many of the beneficial effects of isoflavones might be related, at least partially, to their antioxidant properties (Guo et al. 2002), there has been no comprehensive evaluation of the free-radical scavenging properties of isoflavone metabolites towards reactive oxygen species. Furthermore, little is known about the cellular activity of isoflavone metabolites. The aims of the present study were twofold: (1) to analyse the free-radical scavenging and antioxidant properties of isoflavone metabolites, compared with their parent isoflavones, using electron spin resonance spectroscopy (ESR), as well as the ferric reducing ability of plasma (FRAP) and trolox equivalent antioxidant capacity (TEAC) assays; and (2) to determine if isoflavones affect the production of nitritic oxide (NO) by activated RAW 264.7 macrophages. This murine cell line was chosen since it was an established model for examining determinants of NO production and other inflammatory processes (Guo et al. 2001).
Materials and methods Chemicals and reagents Potassium persulfate and neutral red were obtained from Fisher Scientific (Loughborough, UK). Daidzein, genistein, 1,3,5 THB, equol, 8-hydroxydaidzein and O-DMA were obtained from Plantech (Reading, UK). Purities of the isoflavone test components were 97–98%. 5,5-Dimethyl-1-pyrroline-Noxide was purchased from Labotec Ltd (Japan) and used without further purification. Hydrogen peroxide (H2O2, 31% w/v) was purchased from Mitasubishi Gas (Japan). Ferrous chloride, diethylenetriaminepentaacetic acid (DTPA) and hypoxanthine were purchased from Kanto Chemicals (Japan). Xanthine oxidase from buttermilk was purchased from Nacalai Tesque (Japan). All other chemical reagents used for ESR studies were purchased from Wako Chemicals (Japan). Rabbit polyclonal antiiNOS type II was purchased from BD Transduction Laboratories (San Diego, CA, USA). All other chemicals were obtained from Sigma-Aldrich (Poole, UK). Electron spin resonance spectroscopy Hydroxyl and superoxide anion radicals were detected by electron spin resonance spectroscopy (ESR). ESR experiments were carried out on a JEOL JES-TE 200 ESR spectrometer (X-Band microwave unit). A total of 50 ml of the reaction mixtures described below were introduced into a disposal capillary tube, and ESR spectra were recorded 3 min after starting the reaction. All ESR experiments were done in triplicate at room temperature. Assay for hydroxyl radical in the Fenton system For the ESR spin trapping study, a Fenton reaction was carried out in 100 ml aqueous solution in a micro test tube (1 ml) containing 10 ml 5 mM H2O2, 10 ml 100 mM DMPO, 20 ml re-distilled water, 25 ml 50 mM phosphate buffer, 25 ml 0.4–4.0 mM of the test compounds (final concentration 0.1–1 mM) and 10 ml 0.5 mM freshly prepared ferrous chloride solution. Metal free water was added instead of the test compounds for the control experiment. The Fenton reaction was initiated by adding an aliquot of FeCl2 solution finally. After stirring for 5 s, 50 ml of the reaction mixture were transferred into a 100-ml disposal capillary tube. The bottom of the capillary was sealed by a small amount of clay and ESR spectra were measured 3 min after the addition of FeCl2. ESR spectrometer settings were: central field 327.5 5 mT, microwave frequency 9.18 GHz; modulation amplitude 0.1 mT; microwave power 8 mW; time constant 0.1 s, gain 8.0 105, sweep time 2 min, scan width 100 G. Assay for superoxide radical generated in the xanthine/xanthine oxidase system For the ESR spin-trapping study, superoxide scavenging activity of antioxidants was carried out in 300 ml aqueous solution in a micro test tube (3 ml), containing 80 ml 2 mM hypoxanthine solution in 50 mM phosphate buffer solution, 10 ml 10.5 M DMPO, 80 ml 0.4 U ml1 xanthine oxidase solution in 50 mM phosphate buffer, 100 ml 0.3–3.0 mM antioxidants (final concentration 0.1–1 mM) and 30 ml
916
G. Rimbach et al.
10 mM DTPA solution. The reaction was initiated by adding an xanthine oxidase solution finally. After stirring for 5 s, 50 ml of the reaction mixture were transferred into a 100-ml disposal capillary tube. ESR spectra were measured 3 min after the addition of xanthine oxidase solution. ESR spectrometer settings were same as for the hydroxyl radical assay except gain, which was 5.0 105. Ferric reducing ability of plasma (FRAP) The FRAP assay was performed as described by Benzie and Strain (1996). FRAP reagent was freshly prepared each day by mixing together 10 mM 2,4,6-tripridyl-s-triazine (TPTZ) and 20 mM iron(III) chloride in 0.25 M acetate buffer, pH 3.6. The absorbance of the test components as compared with ascorbic acid was read at 593 nm (Perkin Elmer UV/Vis Lambda Bio 20) 6 min after incubation at room temperature against a blank of FRAP reagent and distilled water. Quercetin dihydrate, dissolved in ethanol was used as a positive control for the TEAC and FRAP assay. Trolox equivalent antioxidant capacity (TEAC) Sixteen hours before the assay being performed, the ABTS radical was prepared by adding 5 ml 4.9 mM potassium persulfate solution to 5 ml 14 mM ABTS solution (Re et al. 1999). This solution was diluted in distilled water to yield an absorbance of 0.70 at 734 nm (Perkin Elmer UV/Vis Lambda Bio 20). The final reaction mixture contained 10 ml standard or test compound in 1 ml ABTS solution. The samples were vortexed for 10 s, and 6 min after addition the absorbance was recorded and compared with ABTS radical solution plus distilled water. A standard curve was obtained by using Trolox as an internal standard (range 0–100 mM). TEAC values expressed the mmols of Trolox having the antioxidant capacity corresponding to 1.0 mmol of the test substance (Betancor-Fernandez et al. 2002). Cell culture RAW 264.7 cells, a murine cell line of monocyte macrophages (obtained from the European Collection of Cell Culture (ECACC), Salisbury, UK), were maintained at 37 C in 5% CO2 according to standard protocols (Rimbach et al. 2000a, b). The medium consisted of DMEM containing 10% heatinactivated foetal bovine serum, 2 mM L-glutamine, 1 mM sodium pyruvate, 100 U ml1 penicillin and 100 mg ml1 streptomycin. Cells were stimulated with 500 ng ml1 lipopolysaccharide (LPS) and 100 U ml1 interferon gamma (IFN-g). For all experiments, isoflavones were dissolved in dimethyl sulfoxide (DMSO) and stored at 80 C until required. When added to medium, the final DMSO concentration was 0.1% (v/v). Cell viability The uptake of the neutral red dye was used to measure cell viability as described by Valacchi et al. (2001). Briefly, after treatment of cells with isoflavones, the culture medium was replaced with fresh medium containing 60 mg ml1 neutral red. Following incubation for 3 h at 37 C, the medium was removed and the cells extracted using a solution comprising 50 : 49 : 1 (v/v/v) ethanol, water and glacial acetic acid. Absorbance at 540 nm was recorded using a microplate reader. NO production NO production was assessed by measurement of nitrite ðNO 2 Þ concentration in the medium using the Griess reaction. Supernatants from cultured macrophages were deproteinized with 0.3 M NaOH and 0.3 M ZnSO4. Equal volumes of Griess reagent (1% sulphanilamide/0.1% N-(1-naphthyl)ethylenediamine dihydrochloride/2.5% H3PO4) and deproteinized samples were incubated for 10 min at room temperature in the dark. Nitrite concentration was determined from absorbance at 548 nm using a standard curve for sodium nitrite (Park et al. 2000). Western blotting Protein from 106 RAW 264.7 cells was extracted in ice-cold lysis buffer (7 M urea, 2 M thiourea, 1% dithiothreitol (DTT), 2% 3-[(3-deoxycholic acid (cholamido-propyl) dimethylammonio]-1propanesulfonate (CHAPS), 0.8% pharmalyte (pH 3–10) and protease inhibitor cocktail from Sigma-Aldrich (according to the manufacturer’s instructions). The supernatant was used for protein determination using the Bradford assay. Solubilized extracts (70 mg) were electrophoresed on 4–20% Tris-glycine SDS gel and transferred electrophoretically to nitrocellulose membranes. After blocking for 2 h with 5% non-fat dried milk in Tris-buffered saline containing 0.1% Tween-20 (TBS-T), the membranes were incubated overnight at 4 C with rabbit polyclonal anti-iNOS type II antibodies (1 : 10 000) in 1% non-fat dried milk in TBS-T. The membranes were washed, incubated for 2 h with peroxidase-labelled secondary antibodies (1 : 3000). Bands were visualized by an Amplified Opti-4CN Substrate Kit from Bio-Rad (UK).
917
Isoflavone metabolites
Statistical analyses Data are presented as mean standard error of the mean (SEM) of between three and five independent experiments performed in triplicate. The data were checked for normality and log transformed where appropriate. Statistical analyses were carried out using ANOVA and a Student’s t-test, and the Kruskal–Wallis and Mann–Whitney U-test (SPSS for Windows version 10.0). Differences were considered significant if p
