Comparison of Food Antioxidants and Iron ... - ACS Publications

Article pubs.acs.org/JAFC

Comparison of Food Antioxidants and Iron Chelators in Two Cellular Free Radical Assays: Strong Protection by Luteolin Tim Hofer,*,† Trond Ø. Jørgensen,‡ and Ragnar L. Olsen# †

Department of Chemicals and Radiation, Division of Environmental Medicine, The Norwegian Institute of Public Health, N-0403 Oslo, Norway ‡ MabCent-SFI and #Norwegian College of Fishery Science, University of Tromsø, N-9037 Tromsø, Norway ABSTRACT: Liver (HepG2) cells were incubated with 21 edible flavonoids, carotenoids, polyunsaturated fatty acid (PUFA) chromones, and metal chelators for 1 h, washed in PBS, and challenged in the cellular antioxidant activity (CAA) and the cellular lipid peroxidation antioxidant activity (CLPAA) assays. These microplate format assays assess the compounds’ ability to protect against cytosolic peroxyl radicals (CAA) and induced membrane lipid peroxidation (CLPAA), respectively. Incubation encompassing a broad compound concentration range determined half-maximal inhibitory concentrations (IC50) by using sigmoidal curve fits. Overall, considering both assays, luteolin offered the greatest protection. The carotenoid astaxanthin offered only modest protection, whereas β-carotene was ineffective. Subtle structural differences between flavonoids were found to have amplified effects on protective abilities, and mechanisms of flavonoid antioxidant action are discussed. Membrane-permeable iron chelators (deferasirox and SIH) offered strong protective effects in CLPAA, but not in CAA, suggesting that CLPAA is dependent on membrane-associated free iron ions. KEYWORDS: catechin, curcumin, hydroxytyrosol, resveratrol, acetylsalicylic acid (ASA)

■

antioxidant activity (CAA) assay.16−18 Both assays are described below. Loosely chelated iron from the labile iron pool (LIP) can redox cycle (Fe3+ ↔ Fe2+) and mediate biomolecular oxidation reactions by ROS (Fenton chemistry) as well as accelerate lipid peroxidation reactions; however, the role of iron in the CLPAA and CAA assays has been less clear. The objective of this study was to test different classes of edible (low toxicity) hydrophilic and lipophilic antioxidants and investigate the role of LIP in the CAA and CLPAA assays using metabolically active human hepatocarcinoma cells (HepG2). The compounds studied are all much in focus as antioxidants and often studied using chemical antioxidant assays.

INTRODUCTION Several human diseases and conditions involve free radicals and/or metal dyshomeostasis.1−3 Iron levels rise in multiple tissues during disease states and aging,1,4 increasing the likelihood of oxidative damages from reactive oxygen species (ROS; O2•−, H2O2, ROOH, HOCl, etc.) during events such as inflammation and ischemia. ROS can induce lipid peroxidation,5,6 a process generating other toxic reactive species, for example, aldehydes. Antioxidant mechanisms are commonly related to free radical scavenging and metal chelating. Low-molecular-weight antioxidants and metal chelators have received increased interest both as protectors from disease onset and as therapeutics.7,8 A regular intake of fruits and vegetables containing antioxidants can counteract undesired health conditions, but it is not clear which dietary compounds are the most beneficial in terms of antioxidant effects. These effects are best evaluated in living cells and organisms. Antioxidants need to be constantly replenished because they are eventually oxidized (some are recycled, e.g., vitamin E and certain flavonoids)9 or are metabolized and excreted.10,11 Intake of flavonoid-rich foods (dark chocolate, red wine, green tea, etc.) have shown promising effects in humans, and flavonoids can scavenge radicals, chelate catalytic metal ions, and pass cellular membranes.6,12,13 Clinically administered iron chelators treat iron overload conditions, but will cause iron deprivation and other toxicities in normal subjects. Numerous papers describe the effects of antioxidants in chemical assays, but few demonstrate effects in living cells, which requires cell membrane passage and compound stability. Compound testing in cell cultures can provide valuable insights into which compounds are likely to have an impact in higher organisms. We recently established the cellular lipid peroxidation antioxidant activity (CLPAA) assay in the microplate format.14,15 Another commonly used free radical assay is the cellular © 2014 American Chemical Society

■

MATERIALS AND METHODS

Chemicals. All chemicals used were of the highest quality, and purity is indicated when relevant. The compounds tested for bioactivities are shown in Figure 1. Acetylsalicylic acid (ASA; Sigma A7356; 99.8%), apigenin (Fluka 10798; 95.4%), L-ascorbic acid (Sigma A5960; 99.3%), astaxanthin (Sigma A9335; synthetic; 95.6%), β-carotene (Fluka 22040; 98.0%), cumene hydroperoxide (cumOOH; Aldrich 513296, 88% (cautionary note: handle concentrated cumOOH inside a hood as vapors are toxic)), deferoxamine mesylate (DFOM; Sigma D9533; 95.0%), 2′,7′-dichlorofluorescin diacetate (DCFH-DA; Fluka 35847), resveratrol (Sigma R5010; 99.0%), and quercetin (Sigma Q0125; >99.9%) were purchased from Sigma-Aldrich Corp., St. Louis, MO, USA. 2,2′-Azobis(2-amidinopropane (abbreviated ABAP or AAPH; 82235), (+)-catechin (70940; >99.9%), epigallocatechin gallate (EGCG, 70935; >99.9%), 3,4-dihydroxyphenylethanol (hydroxytyrosol; 70604; 98.0%), and luteolin (10004161; 99.0%) were purchased from Cayman Europe, Received: Revised: Accepted: Published: 8402

May 16, 2014 July 28, 2014 July 29, 2014 July 29, 2014 dx.doi.org/10.1021/jf5022779 | J. Agric. Food Chem. 2014, 62, 8402−8410

Journal of Agricultural and Food Chemistry

Article

Figure 1. Categories and chemical structures of compounds tested. Numbering of the basic flavonoid structure is illustrated for apigenin. Apigenin and luteolin are flavones; kaempferol and quercetin are flavon-3-ols; and (+)-catechin, (−)-epicatechin, and EGCG are flavan-3-ols. Resveratrol is a stilbenoid, and DL-α-tocopherol is a lipophilic methylated phenol (a member of the vitamin E family). For metal chelators the number of ligands in the resulting major Fe(III) complex and the net charge at pH 7.4 are indicated. DFOM dissolves as DFO (charge +1) at pH 7.4 and forms a ferrioxamine− Fe3+ hexadentate ligand 1:1 complex with net charge +1. Deferasirox chelates Fe3+ as a tridentate ligand 2:1 complex with net charge −3.44 Also, SIH chelates Fe3+ as a tridentate ligand 2:1 complex but with net charge +1.43 Acetylsalicylic acid (ASA), the active ingredient in aspirin, is a small synthetic prodrug of salicylic acid, a phenol. L-Ascorbic acid (vitamin C) is a hydrophilic acidic sugar present as the ascorbate anion (AscH−) at pH 7.4. L-Tryptophan is an amino acid containing an indole group. (108374; 99.5%) was from Merck Chemicals, both located in Darmstadt, Germany. DL-α-Tocopherol (A17039; synthetic; 98.8%)

Tallinn, Estonia. (−)-Epicatechin (A3894; >99.9%) and kaempferol (A3288; 98.9%) were from AppliChem GmbH, and L-tryptophan 8403

dx.doi.org/10.1021/jf5022779 | J. Agric. Food Chem. 2014, 62, 8402−8410


Article

peroxidation antioxidant activity (CLPAA) assay in microplate format,14,15 where after cellular uptake of the lipophilic oxidation sensitive probe C11-BODIPY (fluorescent red) and subsequent antioxidant uptake, lipid peroxidation is induced by adding cumene hydroperoxide (cumOOH; lipophilic), capable of generating oxidizing (and initiating lipid membrane peroxidation) hydroxyl radicals (HO•) in the presence of redox-cycling metals such as ferrous iron (Fe2+).

was from Alfa Aesar, Ward Hill, MA, USA. Curcumin was from Alexis Biochemicals (350-028; 99.0%) purchased through Enzo Life Sciences Inc., Farmingdale, NY, USA. 4,4-Difluoro-5-(4-phenyl-1,3-butadienyl)4-bora-3a,4a-diaza-s-indacene-3-undecanoic acid (C 11 -BODIPY, D3861) was from Invitrogen, Eugene, OR, USA. Deferasirox (ICL670; 99.9%) was kindly provided by Novartis Pharma AG, Basel, Switzerland, and salicylaldehyde isonicotinoyl hydrazone (SIH; 99.0%) by Dr. Katherine J. Franz, Duke University, Durham, NC, USA. PUFA chromone 1 (all-(Z)-5,7-dihydroxy-2-(4Z,7Z,10Z,13Z,16Znonadecapentaenyl) and PUFA chromone 2 (2-((5-hydroxy-2((4Z,7Z,10Z,13Z,16Z)-nonadeca-4,7,10,13,16-pentaen-1-yl)-4-oxo4H-chromen-7-yl)oxy)acetic acid) were synthesized as described19 and donated by Dr. Trond V. Hansen, Oslo University, Norway. PUFA chromone 2 has a carboxylic acid group ether-linked to the chromone moiety, making it more hydrophilic than PUFA chromone 1 (see Figure 1). Screening a small synthetic library of natural compounds identified PUFA chromone 1 and PUFA chromone 2 as powerful antioxidants, particularly in CLPAA, but also in CAA.19 Compounds to be tested were weighed out in rubber-sealed cryo vials and dissolved in appropriate media using a vortex or, when necessary, an ultrasonicator combined with heating (60 °C) for the shortest possible time (minutes). Ascorbic acid, EGCG, (−)-epicatechin, DFOM, and L-tryptophan were dissolved in aqueous cell culture media (without fetal bovine serum, FBS). Quercetin was dissolved in 1 M NaOH. Catechin, hydroxytyrosol, kaempferol, luteolin, and resveratrol were dissolved in ethanol. Apigenin, astaxanthin, β-carotene, the chromones, curcumin, deferasirox, and SIH were dissolved in DMSO. Three stocks were prepared for each compound, which thereafter were stored at −80 °C. On assay days, two stocks containing different compounds were thawed and serially diluted 1:1 using cell culture media. Tested previously using HepG2 cells, short-term (1−2 h) incubations with up to 2% v/v dimethyl sulfoxide (DMSO) or ethanol did not significantly alter reaction rates or induce cell necrosis in any of the assays.15 For compounds dissolved in ethanol, cells were maximally exposed to 2% v/v ethanol at the highest compound concentration investigated, and the ethanol concentration was negligible at lower compound concentrations. Lipophilic compounds dissolved in DMSO, however, were diluted in culture media maintaining a minimum 5% v/v DMSO so as not to stick to plastics and were eventually diluted 1:5 (20 + 80 μL) into each well, giving a final DMSO concentration of 1% v/v DMSO on cells also in control wells. DL-α-Tocopherol, a highly sticky viscous liquid, was diluted in 100% DMSO but also incubated on cells at 1% v/v DMSO (final dilution 1 + 99 μL) along with controls. Aliquots of DCFH-DA (20 mM in ethanol), ABAP (200 mM in water), and C11-BODIPY (6 mM in DMSO) were kept at −80 °C and thawed immediately prior to use. Cell Culture. Human hepatocellular carcinoma (HepG2) cells were purchased from the American Type Culture Collection (Manassas, VA, USA) and routinely cultivated as monolayers in minimum essential medium Earle’s (MEM Earle’s; F0325 from Biochrom AG, Berlin, Germany) supplemented with gentamycin (10 μg/mL), L-alanyl-Lglutamine (2 mM), nonessential amino acids (1%), sodium pyruvate (1 mM), and 10% FBS in 175 cm2 flasks in a humidified 5% CO2 atmosphere at 37 °C. After washing cells with phosphate-buffered saline (PBS, Mg2+- and Ca2+-free), cells were detached using trypsinization and resuspended in cell culture medium. The cell concentration was determined by using Bürker chamber counting, and 80000 cells (in 100 μL) were seeded into wells (0.32 cm2 cell growth surface area) of black microplates with transparent bottoms (product 3603) obtained from Corning Inc. Life Sciences, Tewksbury, MA, USA. Cells were allowed to settle at room temperature to minimize plate edge effects20 before incubation at 37 °C. After 24 h, experiments were conducted when microplate wells had ∼50% monolayer cell confluence. Positive and negative controls were included on each microplate. IC50 values are influenced by the cell density and other experimental conditions; hence, maintaining constant conditions is necessary for reproducible results. Background levels of antioxidants (vitamin E, selenium) in cultured cells are commonly low and dependent on culture media additives. CLPAA Assay. To measure effects of antioxidants on cellular lipid membrane peroxidation, we recently established the cellular lipid

cumOOH + Fe2 + → cumO− + HO• + Fe3 +

(1)

R − H + HO• → R• + H 2O

(2)

R• + O2 → ROO•

(3)

•

•

ROO + RH → ROOH + R and so on

(4)

Free radical attack of C11-BODIPY generates two similar fluorescent green products (both having lost a phenyl moiety), correlative to the degree of lipid peroxidation that has taken place.21,22 C11‐BODIPY(red) (+ oxidation) → C11‐BODIPY oxidation products (green)

(5)

Usage of the C11-BODIPY probe provides an indirect measure of lipid peroxidation. The process generates reactive secondary oxidation products such as malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE) that are more challenging to measure. HepG2 cells grown in serum-supplemented media sustain a non-ferritin-bound redoxactive labile iron pool (LIP) (in isolated primary hepatocytes it is about 5−10 μM),23,24 so the addition of extra Fe3+ was found not to be necessary.14 After cells had been washed in PBS, 100 μL of cell culture medium (without FBS) containing 5 μM C11-BODIPY was added (at 5 μM, C11-BODIPY is not toxic to cells)25 to the microplate and placed inside a 37 °C water-saturated 5% CO2 chamber for 30 min. After medium withdrawal, 100 μL of cell culture medium (without FBS) containing diluted antioxidant was added before the microplate was placed back into the 5% CO2 incubator at 37 °C. After 60 min, the cells were washed in PBS and 100 μL of clear Hanks’ salt solution supplemented with 50 μM cumOOH was added. The microplate was immediately placed inside a Wallac VICTOR3 1420 multilabel counter (PerkinElmer, Waltham, MA, USA) preheated at 37 °C and set on bottom reading mode. The green (485/14 and 520/10 nm filters; CW-lamp setting: 20000)14 fluorescence was monitored over 60 min. The rate (slope) of fluorescently green C11-BODIPY oxidation product increase (obtained from linear regressions of slopes over 0−40 min using the inbuilt plate reader software Workout 2.0 from Dazdaq Solutions Ltd.) is related to the rate of cellular membrane lipid peroxidation.14 As diluted peroxides are unstable, a 0.1 M cumOOH stock was prepared immediately prior to use on a daily basis. CAA Assay. The CAA assay was used as described16 with minor modifications.15 CAA is based on simultaneous cellular uptake of antioxidants and the ROS-sensitive nonfluorescent probe 2′,7′dichlorofluorescin diacetate (DCFH-DA), which is hydrolyzed into cytosolically trapped DCFH that upon oxidation by peroxyl radicals forms 2′,7′-dichlorofluorescein (DCF; fluorescent green), which is monitored.15,16 The formation of peroxyl radicals from thermal decomposition of the azo compound ABAP is illustrated by the equations below:

− RNNR−(+ heat) → 2 −R• + N2 •

•

− R + O2 → − ROO

(6) (7)

After cells had been washed in PBS, 100 μL of cell culture medium (without FBS) containing diluted antioxidants and DCFH-DA (25 μM) was added and the microplate placed inside a 37 °C water-saturated 5% CO2 chamber. After 60 min of preincubation, the cells were washed in PBS, and 100 μL of clear Hanks’ salt solution supplemented with glucose (L2035 from Biochrom) containing ABAP (600 μM) was added before starting (t = 0 min) to monitor green fluorescence (485/14 nm excitation and 520/10 nm emission filters; CW lamp setting: 15000 stabilized 8404



Article

energy, 0.1 sec) inside a Wallac VICTOR3 1420 multilabel counter set at bottom reading mode. Control wells having no oxidant and a known potent antioxidant were included in all plates. The microplate was placed upon a specially designed alumina block in direct contact with all wells intended to provide uniform heating inside a water-saturated 5% CO2 chamber held at 37 °C. Finally, the green fluorescence was monitored again (t = 60 min). CAA data for each well were baseline corrected by subtracting the starting value from final value by using the inbuilt plate reader software Workout 2.0. This corrects for the antioxidant’s own light absorption as well as probe background fluorescence and ensures evaluation of only experimentally generated DCF. Curve Fitting. Log transformation of data and curve fitting was performed using Prism 5.0 (GraphPad Software, San Diego, CA, USA) by employing the four-parameter logistic equation (also called the Hill slope model) that yields half-maximal inhibitory concentrations (IC50, a measure of the effectiveness of a compound) and Hill slopes (examples are shown in Figure 2). A strong concentration dependence was seen for potent antioxidants in the micromolar−millimolar range (−6 to −3 on a log scale, see Figure 2). In cases when full protection could not be achieved, approximate IC50 values were obtained by extrapolating the curve to the lower plateau.

■

RESULTS AND DISCUSSION

In this study, cells were preincubated with the compound of interest for 1 h prior to washing and thereafter exposed to a radical inducer. This preincubation should allow sufficient time for cell membrane passage of most compounds and antioxidant defense establishment prior to adding the inducer, which may otherwise pass cellular membranes more quickly than the antioxidant investigated. Preincubation also ensures that protective effects occur within cells, and are not due to antioxidants reacting with the radical inducer outside cells. Nevertheless, some compounds (e.g., DFO, being taken up by endocytosis)26 may require longer incubations for effective cellular uptake, and rapidly membrane passing compounds could, to some extent, also leak from the cells and pass back into the medium. The HepG2 cell line was chosen due to its human origin, preservation of cellular functions,27,28 metabolic activity,28−30 ease of cultivation, and frequent use for similar tests.16,31 However, HepG2 cells have been reported to express low basal levels of phase I cytochrome P450 metabolizing enzymes and may (for certain compounds) be less responsive than, for example, the HepaRG cell line30,32 and human primary liver cells, in particular.32,33 Human primary liver cells, however, can show profound interindividual differences30 and are today unrealistically expensive for a study requiring large numbers of cells such as this one. Cellular uptake and eventual time-dependent compound metabolite formation may differ between cell types, and LIP levels may be influenced by the culturing conditions. These factors should be considered when results from different studies are compared. Testing two compounds simultaneously over broad concentration ranges for each microplate and then performing dose−response curve fitting by using the fourparameter logistic equation worked fine for both assays; see representative examples in Figure 2. The obtained IC50 values (given in both mg/L and μM units) are shown in Table 1 (CLPAA) and Table 2 (CAA), respectively. These results can be useful for selecting candidate compounds for testing in higher organisms and also provide insight into how the CLPAA and CAA assays function similarly and differ from one another. No obvious correlations between size (Mw) or lipophilicity (XLogP3; a calculated octanol−water partition coefficient) with antioxidant effect (IC50) among the tested compounds were found in either assay.

Figure 2. Dose−response testing compounds in the CLPAA and CAA assays. Representative examples are shown comparing two compounds in parallel over the range 0.2−800 μM using HepG2 cells. See Tables 1 and 2 for all obtained IC50 values. (A) Parallel dose−response testing of catechin and EGCG in the CLPAA assay. For this particular run, catechin and EGCG produced IC50 values of 458 and 4.82 μM, respectively. (B) Parallel dose−response testing of luteolin and resveratrol in the CAA assay. For this particular run, luteolin and resveratrol gave IC50 values of 13.3 and 482 μM, respectively. Experiments were performed in a 96-well microplate, and each concentration was tested in triplicate (N = 6 for each control (+added oxidant (▲); no added oxidant (▼)) selectively placed at concentrations distant from the tested range of antioxidants). For IC50 value calculations a “theoretical zero level” is used as the lower plateau (y value constrained to 0, reflecting a “perfect antioxidant”) for extrapolation of dose−response curves. The top plateau is constrained to the mean of the +added oxidant (▲) value, a measure of both added and endogenous oxidants. a.u., arbitrary units.

Flavonoids. With the exception of apigenin, which displayed no activity in either assay and also not at realistically obtainable concentrations in vivo (protection was noticeable first at several mM), most flavonoids were effective antioxidants in both assays. 8405



Article

Table 1. CLPAA Generated IC50 Values Using HepG2 Cellsa compound

Mw

XLogP3b

class

IC50 (mg/L)

IC50 (μM)

CV (%)

incubation in

SIH luteolin deferasirox EGCG quercetin hydroxytyrosolc PUFA chromone 1c curcumin DL-α-tocopherol PUFA chromone 2c kaempferol resveratrol astaxanthin (+)-catechin (−)-epicatechin apigenin DFOM ASA β-carotene L-tryptophan L-ascorbic acid

241.3 286.2 373.4 458.4 302.2 154.2 434.0 368.4 430.7 492.0 286.2 228.2 596.9 290.3 290.3 270.2 656.8 180.2 536.9 204.2 176.1

1.9 1.4 3.8 1.2 1.5 −0.7 8.4d 3.2 10.7 550 >550 >550 >550 >550 >550

0.833 ± 0.65 2.80 ± 1.35 2.82 ± 1.18 5.20 ± 0.33 9.59 ± 3.53 34.2 ± 13.8 14.13 ± 9.07 24.2 ± 1.82 24.5 ± 11.6 28.9 ± 3.26 173.05 ± 186.4 332.1 ± 56.9 221.1 ± 20.1 457.7 ± 66.0 528.5 ± 164.4 >4000 >4000 >4000 >4000 >4000 >4000

77.7 48.3 41.6 6.38 36.8 40.5 64.2 7.53 47.3 11.3 103 17.1 9.09 14.4 31.1

MEME + 1% DMSO MEME MEME + 1% DMSO MEME MEME MEME MEME + 1% DMSO MEME + 1% DMSO MEME + 1% DMSO MEME + 1% DMSO MEME MEME MEME + 1% DMSO MEME MEME MEME +1% DMSO MEME MEME MEME + 1% DMSO MEME MEME

Data are shown as averages (mean ± SD) from three separate experiments where separate preparations were tested on separate days. The PUFA chromenone data have been published,19 but are included for comparison. bLipophilicity data were obtained from NCBI Pubchem. cFor hydroxytyrosol and the PUFA chromones, only one preparation was available, which was tested on three different days. dClogPa value from ref 19. e Kaempferol gave a biphasic effect curve with CLPAA, and only the protective range (up to 100 μM) was used for curve-fits. High kaempferol concentrations (≥200 μM) offered no protection and increased probe oxidation. MEME = minimum essential medium Earle’s. a

Table 2. CAA Generated IC50 Values Using HepG2 Cellsa compound

Mw

XLogP3b

type

IC50 (mg/L)

IC50 as (μM)

CV (%)

incubation in

luteolin kaempferol EGCG curcumin quercetin L-ascorbic acid PUFA chromone 1c resveratrol astaxanthin hydroxytyrosolc PUFA chromone 2c deferasirox (+)-catechin (−)-epicatechin DFOM β-carotene L-tryptophan ASA DL-α-tocopherol apigenin SIH

286.2 286.2 458.4 368.4 302.2 176.1 434.0 228.2 596.9 154.2 492.0 373.4 290.3 290.3 656.8 536.9 204.2 180.2 430.7 270.2 241.3

1.4 1.9 1.2 3.2 1.5 −1.6 8.4d 3.1 10.3 −0.7 550 >550 >550 >550 >550 >550 >550e

11.90 ± 1.40 13.27 ± 2.86 18.57 ± 3.23 39.47 ± 3.04 71.23 ± 38.6 367.4 ± 66.1 159.5 ± 24.7 519.2 ± 41.9 225.7 ± 24.8 874.0 ± 242 341.5 ± 122 648.4 ± 308.9 1218.7 ± 594.1 1437.7 ± 227.4 1676.3 ± 507.8 >4000 >4000 >4000 >4000 >4000 >4000

11.8 21.5 17.4 7.7 54.1 18.0 15.5 8.1 11.0 27.7 35.8 47.6 48.8 15.8 30.3

MEME MEME MEME MEME + 1% DMSO MEME MEME MEME + 1% DMSO MEME MEME + 1% DMSO MEME MEME + 1% DMSO MEME + 1% DMSO MEME MEME MEME MEME + 1% DMSO MEME MEME MEME + 1% DMSO MEME + 1% DMSO MEME + 1% DMSO

Data are shown as averages (mean ± SD) from three separate experiments where separate preparations were tested on separate days. The PUFA chromenone data have been published,19 but are included for comparison. bLipophilicity data were obtained from NCBI Pubchem. cFor the PUFA chromones and hydroxytyrosol only one preparation was available, which was tested on three separate days. dClogPa value from ref 19. eSIH provided no protection and even increased the fluorescence over that of the positive control at ≥200 μM. a

Thus, luteolin’s dual ortho-positioned 3′,4′-dihydroxy substituents (catechol-type) on its extruding aromatic B-ring appear to have provided the extra edge. Luteolin was found to have a half-peak oxidation potential (Ep/2) of +180 mV, whereas apigenin could not be oxidized up to the maximum tested 1 V.12

Interestingly, luteolin, which displayed the highest antioxidant activity of all compounds tested considering both assays (see Tables 1 and 2), differs structurally from apigenin by just one extra hydroxyl group on the B-ring (Figure 1); both are flavones and have similar molecular weight and lipophilicity (Tables 1 and 2). 8406



Article

mixture of α- and γ-tocopherol.21 Curcumin and resveratrol offered protection in both assays, and at higher curcumin concentrations (≥100 μM) necrotic (cell detachment, etc.) effects were visually observed, reflecting cytotoxicity on the HepG2 cells; however, toxicity in cancer cells (i.e., HepG2) may be considered beneficial. Carotenoids. Carotenoids are highly lipophilic and partition into cellular membranes. Carotenoids contain multiple carbon−carbon double bonds that absorb UV light2 and can react with singlet oxygen (1O2) produced from UV light-induced photosensitization reactions;37 however, 1O2 is of relatively low physiological importance. Carotenoids may also react with peroxyl radicals38 and are oxidized during lipid peroxidation reactions but seem incapable of stalling this process, becoming degraded themselves.39 In humans, carotenoids are well absorbed, contrary to many experimental animals, but it has been suggested that their antioxidant role in humans is not very strong.40 In this study, astaxanthin and β-carotene stained the cells intensely red and orange, respectively, reflecting high cellular uptake. However, astaxanthin’s low and β-carotene’s absent effects in CLPAA suggest poor stalling of lipid peroxidation. The suggestion that astaxanthin would be a “super vitamin E”41 could not be supported. It should, however, be noted that oxidative damage indicative probes such as C11-BODIPY and DCFH could potentially oxidize more easily than biomolecules such as proteins or certain lipid constituents, and cellular membranes consist of a whole range of different lipids. For example, in a cell-free system measuring oxidation of 1-stearoyl-2-arachidonyl-sn-glycero-3phosphocholine (SAPC), C11-BODIPY was more sensitive to oxidative damage than the SAPC lipid.42 PUFA Chromones. The PUFA chromones tested are novel unexplored compounds that may offer promising activities in vivo. However, further toxicokinetic and toxicity testing is required. The more lipophilic PUFA chromone 1 resulted in a somewhat stronger response than PUFA chromone 2 (the PUFA chromone results were recently published19 but are included for comparison as the testing conditions were identical). Metal Chelators. The protective effects offered by the lipophilic cell membrane permeable Fe3+ chelators SIH43 and deferasirox44 in CLPAA (Table 1) suggest that this assay involves membrane-associated Fe ions that catalyze membrane peroxidation reactions upon cumOOH addition and eventually cause C11-BODIPY oxidation with the formation of measurable green fluorescent C11-BODIPY products. Thus, the CLPAA assay should be suitable for screening new Fe chelators. The low, or lacking, protective effect by deferasirox and SIH in CAA (Table 2), however, demonstrates that ABAP-induced oxidation of intracellular DCFH is largely independent of labile Fe ions. In CAA, SIH was the only compound that at higher (≥200 μM) concentrations increased the fluorescence compared to the positive control in a dose−response manner. Noticeably, deferasirox offered a relatively limited protection in CAA compared to its strong protection in CLPAA. The hydrophilic membrane impermeable Fe3+ chelator DFO offered no protection at reasonable concentrations in either assay (noticeable effects appeared first at mM concentrations; IC50 was 1.7 mM in CAA and even higher in CLPAA). DFO is mainly taken up intracellularly into lysosomes by endocytosis (a slow process)26 and requires subcutaneous infusion for optimal effect in patients. Whereas deferasirox and DFO are clinically used, SIH is not, presumably due to lower stability in vivo. Other Compounds. ASA has anti-inflammatory and antiplatelet (inactivates cyclooxygenase enzymes COX-1 and -2) physiological effects but is also a free radical scavenger,

Still, both apigenin and luteolin display iron chelating abilities,12 although not nearly as strong as the metal chelators tested in this study (pFe3+ values of deferasirox, DFO, and SIH are 22.5, 26.6, and 50, respectively;34 pFe3+ values for most compounds investigated in this study could not be found but are considerably lower). Thus, it seems that a potent flavonoid lipid peroxidation inhibitory activity requires a combination of metal chelating ability and low oxidation potential. The large difference between apigenin and luteolin in redox potential likely explains how the two compounds differed in performances in the CAA assay as well. Luteolin exhibits a high degree of electrochemical reversibility that is substantially higher than that of quercetin, for example, which is an even better electron donor (Ep/2 = +30 mV)12 that also can redox cycle its 3′,4′-dihydroxy substituents on its B-ring in a 2e− −2H+ step,9 implying that the C-ring structure affects the compound’s electrochemical properties. Higher concentrations of luteolin stained the cells yellow. Both flavanols, kaempferol and quercetin, displayed protective effects in both assays. However, kaempferol produced a biphasic effect in CLPAA (an increased dose−response protection was observed up to around 200 μM, but at higher concentrations, the protective effect disappeared), so only the protective range (low concentration) was used for curve fits. The reason for the biphasic effect curve in CLPAA is unclear; however, concentrations >200 μM are not biologically realistic in any case. Kaempferol (Ep/2 = +120 mV)12 has a single hydroxyl group on the B-ring as apigenin, but differs in that it is substituted with an oxidizable hydroxyl group at C-3 on its C-ring, which allows it to also oxidize as a conjugated system in a 2e− −2H+ step (its B- and C-ring −OH substituents → O).35 The electrochemical reversibility of luteolin and kaempferol could potentially be a disadvantage12 as redox cycling could produce ROS (as in the case of redox cycling quinones),36 but only unrealistically high (≥200 μM) kaempferol (not luteolin) concentrations had a pro-oxidant effect in this study (in the CLPAA assay only). The large flavonoid EGCG offered protection in both assays. EGCG is relatively lipophilic and has two 3′,4′,5′-trihydroxy (pyrogallol-type) phenolic rings that are also easily oxidized.9 (+)-Catechin and (−)-epicatechin (isomers) offered similar protection in both assays. The reason why (+)-catechin and (−)-epicatechin offered much less protection in comparison to luteolin (all three have ortho-positioned 3′,4′-dihydroxy substituents on the B-ring, and catechin’s Ep/2 of +160 mV12 is nearly the same as that of luteolin) and EGCG is unclear but may be due to the catechins’ different C-ring structure, which likely influences the compound’s reversibility or their lower lipophilicity. Possibly their metal chelatability or a cellular metabolic effect (HepG2 cells are metabolically active) may have affected the outcome. Several flavonoids were found to reach and protect the nuclei in HepG2 cells with maximum incorporation after 30 min, and quercetin and luteolin were found to be effective in suppressing ABAP-induced 8-oxodG formation in DNA.31 Other Phenols. It has been suggested that in particular the o-diphenolic substitution gives high antioxidant ability, as opposed to m-diphenolic or single hydroxyl substitution.8 This observation in addition to the obtained results for the flavonoids is further supported by the performance of hydroxytyrosol. Hydroxytyrosol is a small but major olive oil o-diphenolic derived from hydrolysis of oleuropein,8 which offered protection in both assays. DL-α-Tocopherol displayed antioxidant activity in cellular membranes (CLPAA), but not in the aqueous cytosol (CAA). It was previously observed that the oxidation of C11-BODIPY in rat-1 fibroblast was prevented by preincubation of the cells with a 8407



■

particularly against free hydroxyl radicals (•OH).3,45 Here, however, ASA offered no protection in either assay, indicating that free •OH has no major role in the CLPAA and CAA assays. Thus, the initiating CLPAA assay reactions 1 and 2 above appear to be “caged” events (no freely diffusing •OH).46 L-Ascorbic acid (vitamin C) offered protection in the CAA assay, but not in the CLPAA assay. Passive plasma membrane passage of ascorbate (AscH−) is expected to be very low due to low lipophilicity and charge, but active ascorbate import is mediated in most cell types (also hepatocytes) by sodium vitamin C transporters SVCT1 and SVTC2.47 The amino acid L-tryptophan (contains an indole group) offered no protection in either assay and was merely included as a negative control. Comparisons between CLPAA and CAA and Mechanistic Assay Insights. The two biologically relevant assays measure different parameters. CLPAA measures lipid peroxidation inhibition, and CAA measures intracellular peroxyl radical scavenging; hence, these are very complementing assays.14−16 For CLPAA, we previously concluded that the rate of green fluorescence appearance (C11-BODIPY decomposition products) was more sensitive than the disappearance of red fluorescence (C11 -BODIPY) 14 and that the background fluorescence from the plastic microplate interfered in the red light range.14 As CLPAA is based on analyzing the slope of green BODIPY product formation, it is not affected by individual absorption from tested compounds. With CAA, however, lightabsorbing effects by the compound can be an issue for particularly strong light (excitation and/or emitted light) absorbing compounds such as carotenoids by producing a false effect (increased effect at increased compound concentration). This can be checked and possibly normalized for by using a blank control with only the addition of test compound to cell-free medium containing a known concentration of the fluorescent probe. Moreover, the CAA assay is also more sensitive toward plate heating effects than CLPAA, and the variation among positive control values in CAA also increases (more than for CLPAA) with incubation time. Random placement of samples in the microplate can circumvent edge and corner effects (outer wells generally warm up more rapidly). Overall, our results together with those by others48,49 suggest that peroxides (cumOOH, H2O2, etc.) depend upon catalytic transition metal ions to oxidize organic compounds, whereas both production of azo compound (e.g., ABAP) oxidizing peroxyl radicals (ROO• as in the CAA assay) and their oxidation of organic compounds can take place independent of transition metals. Thus, the choice of oxidant in an assay relates to which antioxidant characteristics are being assessed. Implications. This study encourages further in vivo studies with administration of the better performing compounds (Tables 1 and 2), and an increased intake of foods rich in these compounds could be beneficial for our health. Luteolin, the most promising compound in this study, is present in multiple plants to various degrees; for example, high luteolin contents are present in peanut shells (here extractable luteolin exists in the studied aglycone form)50 and in certain edible herbs such as peppermint and sage,51 where luteolin mainly exists chemically attached to sugars (i.e., rutinose or glucose), bonds that hydrolyze during gastrointestinal passage.10,52 Interestingly, chrysanthemum flower tea extract is rich in luteolin constituents and has been used in folk medicine for centuries.52

Article

AUTHOR INFORMATION

Corresponding Author

*(T.H.) E-mail: [email protected]. Phone: +47-21076671. Fax: +47-21076686. Funding

We are grateful to the Norwegian Research Council (Project 208463) for financial support. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Novartis Pharma AG, Dr. Katherine J. Franz, and Dr. Trond V. Hansen for kindly supplying deferasirox, SIH, and the PUFA chromones, respectively.

■

ABBREVIATIONS USED ABAP (or AAPH), 2,2′-azobis(2-amidinopropane); ASA, acetylsalicylic acid; CAA, cellular antioxidant activity; C11BODIPY, 4,4-difluoro-5-(4-phenyl-1,3-butadienyl)-4-bora3a,4a-diaza-s-indacene-3-undecanoic acid; CLPAA, cellular lipid peroxidation antioxidant activity; cumOOH, cumene hydroperoxide; DCF, 2′,7′-dichlorofluorescein; DCFH, 2′,7′-dichlorofluorescin; DCFH-DA, 2′,7′-dichlorofluorescin diacetate; DFOM, deferoxamine mesylate; DMSO, dimethyl sulfoxide; EGCG, epigallocatechin gallate; FBS, fetal bovine serum; FRAP, ferric reducing antioxidant power; HepG2, human hepatocellular liver carcinoma cell line; IC50, half-maximal inhibitory concentration; PUFA chromone 1, all-(Z)-5,7-dihydroxy-2(4Z,7Z,10Z,13Z,16Z)-nonadecapentaenyl; PUFA chromone 2, 2-((5-hydroxy-2-((4Z,7Z,10Z,13Z,16Z)-nonadeca-4,7,10,13,16pentaen-1-yl)-4-oxo-4H-chromen-7-yl)oxy)acetic acid; SIH, salicylaldehyde isonicotinoyl hydrazone

■

REFERENCES

(1) Myhre, O.; Utkilen, H.; Duale, N.; Brunborg, G.; Hofer, T. Metal dyshomeostasis and inflammation in Alzheimer’s and Parkinson’s diseases: possible impact of environmental exposures. Oxid. Med. Cell. Longev. 2013, 2013, 1−19. (2) Snodderly, D. M. Evidence for protection against age-related macular degeneration by carotenoids and antioxidant vitamins. Am. J. Clin. Nutr. 1995, 62, 1448S−1461S. (3) Thome, J.; Zhang, J. J.; Davids, E.; Foley, P.; Weijers, H. G.; Wiesbeck, G. A.; Boning, J.; Riederer, P.; Gerlach, M. Evidence for increased oxidative stress in alcohol-dependent patients provided by quantification of in vivo salicylate hydroxylation products. Alcohol.: Clin. Exp. Res. 1997, 21, 82−85. (4) Hofer, T.; Marzetti, E.; Xu, J. Z.; Seo, A. Y.; Gulec, S.; Knutson, M. D.; Leeuwenburgh, C.; Dupont-Versteegden, E. E. Increased iron content and RNA oxidative damage in skeletal muscle with aging and disuse atrophy. Exp. Gerontol. 2008, 43, 563−570. (5) Zhang, R.; Brennan, M. L.; Shen, Z.; MacPherson, J. C.; Schmitt, D.; Molenda, C. E.; Hazen, S. L. Myeloperoxidase functions as a major enzymatic catalyst for initiation of lipid peroxidation at sites of inflammation. J. Biol. Chem. 2002, 277, 46116−46122. (6) Laguerre, M.; Lecomte, J.; Villeneuve, P. Evaluation of the ability of antioxidants to counteract lipid oxidation: existing methods, new trends and challenges. Prog. Lipid Res. 2007, 46, 244−282. (7) Rossi, L.; Mazzitelli, S.; Arciello, M.; Capo, C. R.; Rotilio, G. Benefits from dietary polyphenols for brain aging and Alzheimer’s disease. Neurochem. Res. 2008, 33, 2390−2400. (8) Tripoli, E.; Giammanco, M.; Tabacchi, G.; Di Majo, D.; Giammanco, S.; La Guardia, M. The phenolic compounds of olive oil: structure, biological activity and beneficial effects on human health. Nutr. Res. Rev. 2005, 18, 98−112.

8408



Article

human hepatoma (HepG2) cells. Chem. Res. Toxicol. 2014, 27, 852− 863. (30) Jetten, M. J. A.; Kleinjans, J. C. S.; Claessen, S. M.; Chesne, C.; van Delft, J. H. M. Baseline and genotoxic compound induced gene expression profiles in HepG2 and HepaRG compared to primary human hepatocytes. Toxicol. in Vitro 2013, 27, 2031−2040. (31) Kanazawa, K.; Uehara, M.; Yanagitani, H.; Hashimoto, T. Bioavailable flavonoids to suppress the formation of 8-OHdG in HepG2 cells. Arch. Biochem. Biophys. 2006, 455, 197−203. (32) Gerets, H. H. J.; Tilmant, K.; Gerin, B.; Chanteux, H.; Depelchin, B. O.; Dhalluin, S.; Atienzar, F. A. Characterization of primary human hepatocytes, HepG2 cells, and HepaRG cells at the mRNA level and CYP activity in response to inducers and their predictivity for the detection of human hepatotoxins. Cell Biol. Toxicol. 2012, 28, 69−87. (33) Wilkening, S.; Stahl, F.; Bader, A. Comparison of primary human hepatocytes and hepatoma cell line HEPG2 with regard to their biotransformation properties. Drug. Metab. Dispos. 2003, 31, 1035− 1042. (34) Fakih, S.; Podinovskaia, M.; Kong, X.; Collins, H. L.; Schaible, U. E.; Hider, R. C. Targeting the lysosome: fluorescent iron(III) chelators to selectively monitor endosomal/lysosomal labile iron pools. J. Med. Chem. 2008, 51, 4539−4552. (35) Brown, J. E.; Khodr, H.; Hider, R. C.; Rice-Evans, C. A. Structural dependence of flavonoid interactions with Cu2+ ions: implications for their antioxidant properties. Biochem. J. 1998, 330, 1173−1178. (36) Taguchi, K.; Shimada, M.; Fujii, S.; Sumi, D.; Pan, X.; Yamano, S.; Nishiyama, T.; Hiratsuka, A.; Yamamoto, M.; Cho, A. K.; Froines, J. R.; Kumagai, Y. Redox cycling of 9,10-phenanthraquinone to cause oxidative stress is terminated through its monoglucuronide conjugation in human pulmonary epithelial A549 cells. Free Radical Biol. Med. 2008, 44, 1645−1655. (37) Di Mascio, P.; Devasagayam, T. P.; Kaiser, S.; Sies, H. Carotenoids, tocopherols and thiols as biological singlet molecular oxygen quenchers. Biochem. Soc. Trans. 1990, 18, 1054−1056. (38) El-Agamey, A.; Lowe, G. M.; McGarvey, D. J.; Mortensen, A.; Phillip, D. M.; Truscott, T. G.; Young, A. J. Carotenoid radical chemistry and antioxidant/pro-oxidant properties. Arch. Biochem. Biophys. 2004, 430, 37−48. (39) Chen, G.; Djuric, Z. Carotenoids are degraded by free radicals but do not affect lipid peroxidation in unilamellar liposomes under different oxygen tensions. FEBS Lett. 2001, 505, 151−154. (40) Krinsky, N. I. Carotenoids as antioxidants. Nutrition 2001, 17, 815−817. (41) Miki, W. Biological functions and activities of animal carotenoids. Pure Appl. Chem. 1991, 63, 141−146. (42) MacDonald, M. L.; Murray, I. V. J.; Axelsen, P. H. Mass spectrometric analysis demonstrates that BODIPY 581/591 C11 overestimates and inhibits oxidative lipid damage. Free Radical Biol. Med. 2007, 42, 1392−1397. (43) Charkoudian, L. K.; Pham, D. M.; Franz, K. J. A pro-chelator triggered by hydrogen peroxide inhibits iron-promoted hydroxyl radical formation. J. Am. Chem. Soc. 2006, 128, 12424−12425. (44) Bruin, G. J. M.; Faller, T.; Wiegand, H.; Schweitzer, A.; Nick, H.; Schneider, J.; Boernsen, K. O.; Waldmeier, F. Pharmacokinetics, distribution, metabolism, and excretion of deferasirox and its iron complex in rats. Drug. Metab. Dispos. 2008, 36, 2523−2538. (45) Shi, X. L.; Ding, M.; Dong, Z. G.; Chen, F.; Ye, J. P.; Wang, S. W.; Leonard, S. S.; Castranova, V.; Vallyathan, V. Antioxidant properties of aspirin: Characterization of the ability of aspirin to inhibit silica-induced lipid peroxidation, DNA damage, NF-κB activation, and TNF-α production. Mol. Cell. Biochem. 1999, 199, 93−102. (46) Hofer, T. Oxidation of 2′-deoxyguanosine by H2O2-ascorbate: evidence against free OH• and thermodynamic support for twoelectron reduction of H2O2. J. Chem. Soc., Perkin Trans. 2 2001, 210− 213. (47) Liang, W. J.; Johnson, D.; Jarvis, S. M. Vitamin C transport systems of mammalian cells. Mol. Membr. Biol. 2001, 18, 87−95.

(9) Hendrickson, H. P.; Kaufman, A. D.; Lunte, C. E. Electrochemistry of catechol-containing flavonoids. J. Pharm. Biomed. 1994, 12, 325−334. (10) Manach, C.; Donovan, J. L. Pharmacokinetics and metabolism of dietary flavonoids in humans. Free Radical Res. 2004, 38, 771−785. (11) Crozier, A.; Jaganath, I. B.; Clifford, M. N. Dietary phenolics: chemistry, bioavailability and effects on health. Nat. Prod. Rep. 2009, 26, 1001−1043. (12) van Acker, S. A.; van den Berg, D. J.; Tromp, M. N.; Griffioen, D. H.; van Bennekom, W. P.; van der Vijgh, W. J.; Bast, A. Structural aspects of antioxidant activity of flavonoids. Free Radical Biol. Med. 1996, 20, 331−342. (13) Melidou, M.; Riganakos, K.; Galaris, D. Protection against nuclear DNA damage offered by flavonoids in cells exposed to hydrogen peroxide: the role of iron chelation. Free Radical Biol. Med. 2005, 39, 1591−1600. (14) Hofer, T.; Olsen, R. L. Cellebasert metode for måling av lipid peroksidasjon og antioksidantaktivitet. Bioingeniøren (Norwegian; archived at www.bioingenioren.no) 2010, 10, 6−12. (15) Hofer, T.; Eriksen, T. E.; Hansen, E.; Varmedal, I.; Jensen, I. J.; Andersen, J. H.; Olsen, R. L. Cellular and chemical assays for discovery of novel antioxidants in marine organisms. In Studies on Experimental Models; Basu, S., Wiklund, L., Eds.; Springer Science, Humana Press: New York, 2011; p 680. (16) Wolfe, K. L.; Liu, R. H. Cellular antioxidant activity (CAA) assay for assessing antioxidants, foods, and dietary supplements. J. Agric. Food Chem. 2007, 55, 8896−8907. (17) Song, W.; Derito, C. M.; Liu, M. K.; He, X. J.; Dong, M.; Liu, R. H. Cellular antioxidant activity of common vegetables. J. Agric. Food Chem. 2010, 58, 6621−6629. (18) Wolfe, K. L.; Kang, X. M.; He, X. J.; Dong, M.; Zhang, Q. Y.; Liu, R. H. Cellular antioxidant activity of common fruits. J. Agric. Food Chem. 2008, 56, 8418−8426. (19) Mohamed, Y. M. A.; Vik, A.; Hofer, T.; Andersen, J. H.; Hansen, T. V. Polyunsaturated fatty acid-derived chromones exhibiting potent antioxidant activity. Chem. Phys. Lipids 2013, 170−171, 41−45. (20) Lundholt, B. K.; Scudder, K. M.; Pagliaro, L. A simple technique for reducing edge effect in cell-based assays. J. Biomol. Screen. 2003, 8, 566−570. (21) Pap, E. H.; Drummen, G. P.; Winter, V. J.; Kooij, T. W.; Rijken, P.; Wirtz, K. W.; Op den Kamp, J. A.; Hage, W. J.; Post, J. A. Ratiofluorescence microscopy of lipid oxidation in living cells using C11BODIPY(581/591). FEBS Lett. 1999, 453, 278−282. (22) Drummen, G. P.; Gadella, B. M.; Post, J. A.; Brouwers, J. F. Mass spectrometric characterization of the oxidation of the fluorescent lipid peroxidation reporter molecule C11-BODIPY(581/591). Free Radical Biol. Med. 2004, 36, 1635−1644. (23) Ma, Y.; de Groot, H.; Liu, Z.; Hider, R. C.; Petrat, F. Chelation and determination of labile iron in primary hepatocytes by pyridinone fluorescent probes. Biochem. J. 2006, 395, 49−55. (24) Petrat, F.; Rauen, U.; de Groot, H. Determination of the chelatable iron pool of isolated rat hepatocytes by digital fluorescence microscopy using the fluorescent probe, phen green SK. Hepatology 1999, 29, 1171−1179. (25) Drummen, G. P.; van Liebergen, L. C.; Op den Kamp, J. A.; Post, J. A. C11-BODIPY(581/591), an oxidation-sensitive fluorescent lipid peroxidation probe: (micro)spectroscopic characterization and validation of methodology. Free Radical Biol. Med. 2002, 33, 473−490. (26) Persson, H. L.; Yu, Z. Q.; Tirosh, O.; Eaton, J. W.; Brunk, U. T. Prevention of oxidant-induced cell death by lysosomotropic iron chelators. Free Radical Biol. Med. 2003, 34, 1295−1305. (27) Javitt, N. B. Hep-G2 cells as a resource for metabolic studies − lipoprotein, cholesterol, and bile-acids. FASEB J. 1990, 4, 161−168. (28) Roe, A. L.; Snawder, J. E.; Benson, R. W.; Roberts, D. W.; Casciano, D. A. HepG2 cells − an in vitro model for P450-dependent metabolism of acetaminophen. Biochem. Biophys. Res. Commun. 1993, 190, 15−19. (29) Huang, M.; Zhang, L.; Mesaros, C.; Zhang, S. H.; Blaha, M. A.; Blair, I. A.; Penning, T. M. Metabolism of a representative oxygenated polycyclic aromatic hydrocarbon (PAH) phenanthrene-9,10-quinone in 8409



Article

(48) Balcerczyk, A.; Kruszewski, M.; Bartosz, G. Does the cellular labile iron pool participate in the oxidation of 2′,7′-dichlorodihydrofluorescein? Free Radical Res. 2007, 41, 563−570. (49) Lebel, C. P.; Ischiropoulos, H.; Bondy, S. C. Evaluation of the probe 2′,7′-dichlorofluorescin as an indicator of reactive oxygen species formation and oxidative stress. Chem. Res. Toxicol. 1992, 5, 227−231. (50) Zhou, P.; Li, L. P.; Luo, S. Q.; Jiang, H. D.; Zeng, S. Intestinal absorption of luteolin from peanut hull extract is more efficient than that from individual pure luteolin. J. Agric. Food Chem. 2008, 56, 296−300. (51) Fecka, I.; Turek, S. Determination of water-soluble polyphenolic compounds in commercial herbal teas from Lamiaceae: peppermint, melissa, and sage. J. Agric. Food Chem. 2007, 55, 10908−10917. (52) Lu, X. Y.; Sun, D. L.; Chen, Z. J.; Chen, T.; Li, L. P.; Xu, Z. H.; Jiang, H. D.; Zeng, S. Relative contribution of small and large intestine to deglycosylation and absorption of flavonoids from Chrysanthemun morifolium extract. J. Agric. Food Chem. 2010, 58, 10661−10667.

8410
