Biochimie 94 (2012) 403e415 Contents lists available at SciVerse ScienceDirect Biochimie journal homepage: www.elsevier.com/locate/biochi Research paper Antiradical and antioxidant activities of new bio-antioxidants V.D. Kanchevaa, *, L. Sasob, S.E. Angelovaa, M.C. Fotic, A. Slavova-Kasakovaa, C. Daquinoc, V. Encheva, O. Firuzid, J. Necheva a Institute of Organic Chemistry with Centre of Phytochemistry, Bulgarian Academy of Sciences, Sofia, Bulgaria Sapienza University of Rome, Rome, Italy c CNR- Institute of Biomolecular Chemistry, Catania, Italy d Medicinal and Natural Products Chemistry Research Center, Shiraz University of Medical Sciences, Shiraz, Iran b a r t i c l e i n f o a b s t r a c t Article history: Received 7 June 2011 Accepted 11 August 2011 Available online 25 August 2011 Antioxidants could be promising agents for management of oxidative stress-related diseases. New biologically active compounds, belonging to a rare class of natural lignans with antiangiogenic, antitumoral and DNA intercalating properties, have been recently synthesized. These compounds are benzo[kl] xanthene lignans (1,2) and dihydrobenzofuran neolignans (3,4). The radical scavenging and chainbreaking antioxidant activities of compounds 1e4 were studied by applying different methods: radical scavenging activity by DPPH rapid test, chain-breaking antioxidant activity and quantum chemical calculations. All studied compounds were found to be active as DPPH scavengers but reaction time with DPPH and compounds’ concentrations influenced deeply the evaluation. The highest values of radical scavenging activity (%RSAmax) and largest rate constants for reaction with DPPH were obtained for compounds 2 and 3. Comparison of %RSAmax with that of standard antioxidants DL-a-tocopherol (TOH), caffeic acid (CA) and butylated hydroxyl toluene (BHT) give the following new order of %RSA max: TOH (61.1%) > CA (58.6%) > 3 (36.3%) > 2 (28.1%) > 4 (6.7%) > 1 (3.6%) ¼ BHT (3.6%). Chain-breaking antioxidant activities of individual compounds (0.1e1.0 mM) and of their equimolar binary mixtures (0.1 mM) with TOH were determined from the kinetic curves of lipid autoxidation at 80  C. On the basis of a comparable kinetic analysis with standard antioxidants a new order of the antioxidant efficiency (i.e., protection factor, PF) of compounds 1e4 were obtained: 2 (7.2)  TOH (7.0)  CA (6.7) > 1 (3.1) > 3 (2.2) > ferulic acid FA (1.5) > 4 (0.6); and of the antioxidant reactivity (i.e. inhibition degree, ID): 2 (44.0) >> TOH (18.7) >> CA (9.3) >> 1 (8.4) > 3 (2.8) > FA (1.0) > 4 (0.9). The important role of the catecholic structure in these compounds, which is responsible for the high chain-breaking antioxidant activity, is discussed and a reaction mechanism is proposed. Higher oxidation stability of the lipid substrate was found in the presence of equimolar binary mixtures 2 þ TOH, 3 þ TOH and 4 þ TOH. However, an actual synergism was only obtained for the binary mixtures with compounds 3 and 4. The geometries of compounds and all possible phenoxyl radicals were optimized using density functional theory. For description of the scavenging activity bond dissociation enthalpies (BDE), HOMO energies and spin densities were employed. The best correlation between theoretical and experimental data was obtained for compound 2, with the highest activity, and for compound 4 with the lowest activity. The BDE is the most important theoretical descriptor, which correlates with the experimentally obtained antioxidant activity of the studied benzo[kl]xanthene lignans and dihydrobenzofuran neolignans. Ó 2011 Elsevier Masson SAS. All rights reserved. Keywords: Antiradical activity Antioxidant activity Bio-antioxidants Quantum-chemical calculations 1. Introduction During the last years a new class of biologically active antioxidants, namely bio-antioxidants have been proposed [1,2]. The idea * Corresponding author. Institute of Organic Chemistry with Centre of Phytochemistry, Bulgarian Academy of Sciences, Lipid Chemistry Department, “Acad. Georgi Bonchev Str.”, Block 9, Sofia, Bulgaria. Tel.: þ359 2 9606187; fax: þ359 2 8700225. E-mail address: vedeka@abv.bg (V.D. Kancheva). 0300-9084/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.biochi.2011.08.008 is to combine in one molecule by a specific synthetic way various functional fragments, which have synergetic effect at different stages of the complex multistage process of lipid oxidation [3,4]. Polyphenolics are multifunctional antioxidants and act as: reducing agents (free radical terminators), metal chelators, and singlet oxygen quenchers. The activity of antioxidants depends on complex factors including: the nature of antioxidants, the condition of oxidation, the properties of the oxidizing substrate and the stage of oxidation. 404 V.D. Kancheva et al. / Biochimie 94 (2012) 403e415 Lignans and related compounds such as neolignans are widely distributed in many plants and possess various biological properties including antiangiogenic [5] and antioxidant [6] activities. Diverse biological properties of lignans and related compounds make them very attractive as novel biologically active agents. The biomimetic phenolic oxidative coupling of caffeic acid esters have been utilized to synthesize unusual lignans belonging to the rare class of natural lignans such as mongolicumin A, rufescidride and yunnaneic acid H [7]. These compounds that include benzo[kl]xanthene lignans (1 and 2) and dihydrobenzofuran neolignans (3 and 4) (Fig. 1) have shown antitumoral [8] and antiangiogenic properties [9]. The aim of this study is to shed some light on the mechanism of action of these new compounds as antioxidants. These compounds were tested experimentally for their capacity as radical scavengers at physiological temperature (37  C) and as chain-breaking antioxidants. The effects of the solvent (acetone and methanol), of concentration (1e10 mM) and reaction time were also studied and compared by using the TLC DPPH rapid test. The radical scavenging activity towards DPPH gives information about the H-atom donating capacity of the studied compounds and some preliminary information about the possibility of using these compounds as antioxidants. In order to evaluate the antioxidant activity, the individual compounds as well as their binary mixtures with a-tocopherol (TOH) were therefore tested in model systems undergoing autoxidation. Quantum-chemical calculations were carried out to get full geometry optimization of the above mentioned neutral molecules and all possible radicals formed, and to find different energetical and structural characteristics. The aim was to find a theoretical descriptor or descriptors which characterize the activity of antioxidants to scavenge free radicals and on this base to be able to improve the selection of new antioxidants. 2. Materials and methods 2.1. Synthesis of the compounds The compounds including benzo[kl]xanthene lignans 1 and 2 and dihydrobenzofuran neolignans 3 and 4 were synthesized by metal-mediated oxidative coupling of caffeic acid esters and their structures were determined by spectroscopic methods. These biomimetic syntheses yield compounds [7] unprecedented in literature with a basic “natural” skeleton and a bioactivity profile similar to, or better than, that of natural analogues. A limit of this synthetic approach is that the lack of stereocontrol both in metal and enzyme mediated phenolic radical coupling frequently results in complex mixtures and racemate compounds. Details of the synthesis and structural characterization of the compounds (1, 3, 4) can be found in Daquino et al. [7]. The synthesis of compound 2 (Scheme 1) is instead reported in the following paragraphs. 2.1.1. General All chemicals were of reagent grade and purchased from SigmaeAldrich. LiChroprep Si-60 (Merck) was used as stationary phase for column chromatography. NMR spectra were recorded at 298 K on a Varian Unity Inova spectrometer operating at 499.86 (1H) and 125.70 MHz (13C) and equipped with gradient-enhanced, reverse-detection probe. Elemental analyses were performed on a PerkineElmer 240B microanalyzer. UVeVis spectra were recorded using a double-ray PerkineElmer model Lambda 25 spectrophotometer, whereas the FT-IR spectra were obtained with a Perkin Elmer Spectrum BX FT-IR System spectrophotometer. Liquid chromatographyemass spectrometry (LCeESIeMS) was carried out on an instrument (Waters 1525) equipped with ESI MS (Waters Micromass ZQ) and UV-DAD (Waters 996) detector; column: PhenomenexÒ Luna, C18, 250  4.6 mm (5 mm) at 298 K using as eluent system H2O/CH3CN, with a flow rate 1 ml min 1. 2.1.2. Synthesis of chlorogenic acid methyl ester (I) To a solution of chlorogenic acid (250 mg, 0.71 mmol) in MeOH (50 ml), 0.5 ml of H2SO4 (98%) were added; the resulting mixture was stirred for 25 min under reflux. After a fast cooling at room temperature 250 ml of ethyl acetate were added; the organic phase was washed with a 2 M NaHCO3 solution until effervescence ceased and with saturated brine (3  50 ml), and then dried over anhydrous Na2SO4. The aqueous phases were mixed, the pH adjusted to Fig. 1. Structures of studied compounds. 405 V.D. Kancheva et al. / Biochimie 94 (2012) 403e415 OH OH OH HO HO O HO OH O O MeOH/H+ 2,2-dimethoxypropane p-toluenesulfonic acid Mn(AcO)3 reflux, 25 min. r.t., 1h r.t., CHCl3, 9h O O O HO2C OH OH Chlorogenic Acid O MeO2C O OH MeO2C OH I (99%) OH O CO2Me O HO O OH OH CO2Me O O II ( 98%) OH O O O OH O OH 2 (40%) Scheme 1. Synthesis of compound 2. 7 with HCl 2 M, and then were extracted with ethyl acetate (3  50 ml). The organic solution was dried over anhydrous Na2SO4 and the solvent evaporated to dryness to afford 257 mg of I as a white solid (yield 99%). 1 H NMR [500 MHz, CD3OD, 298 K]: d ¼ 7.52 (d, 3JH,H ¼ 15.9 Hz, 1H), 7.04 (d, 4JH,H ¼ 2.0 Hz, 1H), 6.95 (dd, 3JH,H ¼ 8.2 and 4 JH,H ¼ 2.0 Hz, 1H), 6.78 (d, 3JH,H ¼ 8.2 Hz, 1H), 6.21 (d, 3JH,H ¼ 15.9 Hz, 1H), 5.27 (m, 1H), 4.13 (m, 1H), 3.73 (m, 1H), 3.69 (s, 3H), 2.23e2.11 (m, 3H), 2.03e1.99 (m, 1H) . - ESI MS: m/z ¼ 367.4 [M H] . 2.1.3. Synthesis of chlorogenic acid methyl ester acetonide (II) 250 mg (0.68 mmol) of I, 30 mg of p-toluenesulfonic acid (0.17 mmol) and 10 ml of 2,2-dimethoxypropane were placed in a flask. The cloudy solution was shaken for 1 h after which it was filtered on paper. 200 ml of ethyl acetate were added to the solution which was extracted with 2M aqueous NaHCO3 (3  50 ml) and brine (3  50 ml). After removal of the solvent under reduced pressure 273 mg of II were recovered as a white powder (yield 98%). Rf (TLC) ¼ 0.5 (8% MeOHeCH2Cl2) e UV (MeOH): lmax (3 ) ¼ 220 e 3522, 3346, 2991, 1784, 1715, (12000), 281 (21000) nm e IR (KBr): n 1640, 1209, 1175 cm 1 e 1H NMR [500 MHz, (CD3)2CO, 298 K]: d ¼ 7.57 (d, 3JH,H ¼ 15.9 Hz, 1H), 7.18 (d, 4JH,H ¼ 2.0 Hz, 1H), 7.06 (dd, 3 JH,H ¼ 8.2 and 4JH,H ¼ 2.0 Hz, 1H), 6.90 (d, 3JH,H ¼ 8.2 Hz, 1H), 6.28 (d, 3JH,H ¼ 15.9 Hz, 1H), 5.46 (m, 1H), 4.53 (m, 1H), 4.19 (m, 1H), 3.72 (s, 3H), 2.36e2.32 (m, 1H), 2.17e2.08 (m, 2H), 2.00e1.96 (m, 1H), 1.48 (s, 3H), 1.31 (s, 3H). 13C NMR [125 MHz, (CD3)2CO, 298 K]: d ¼ 174.56, 165.69, 147.91, 145.41, 145.11, 126.72, 121.70, 115.53, 114.74, 114.36, 108.56, 76.65, 73.52, 73.48, 70.37, 51.81, 36.47, 35.22, 27.58, 25.33. e ESI MS: m/z ¼ 407.4 [M H] . e C20H24O9 (408.399): calcd. C, 58.82 and H, 5.92%; found C, 58.79 and H, 5.95%. 2.1.4. Synthesis of bis[(3aR,4R,6S,7aR)-6-hydroxy-6(methoxycarbonyl)-2,2-dimethylhexahydro-1,3-benzodioxol-4-yl] 6,9,10-trihydroxybenzo[kl]xanthene-1,2-dicarboxylate (2) 200 mg (0.49 mmol) of II were dissolved in 40 ml of chloroform contained in a 200 ml flask. To the stirring solution 525 mg of Mn(OAc)3 (1.96 mmol) were therefore added and the mixture was left to react at room temperature for 9 h. Then, the solvent was evaporated to dryness. To the green yellow solid 15 ml of saturated ascorbic acid methanolic solution and 250 ml of ethyl acetate were added; the organic phase was extracted with a 2 M of solution of NaHCO3 (3  50 ml) and brine (3  50 ml). The organic phase was then dried over anhydrous Na2SO4 and the solvent was evaporated to dryness. The crude product was purified by silica gel column chromatography, using a gradient of CH2Cl2/MeOH (from 1% to 5%) to afford 80 mg of 2 as a yellowegreen amorphous powder (yield 40%). Rf (TLC) ¼ 0.4 (8% MeOHeCH2Cl2). e UV (MeOH): lmax (3 ) 272 e 3548, 2820, 1710, 1676, 1592, (24164), 390 (11100) nm. e IR (KBr): n 1467, 1381, 1152, 1089 cme1 e 1H NMR [500 MHz, (CD3)2CO, 298 K]: d ¼ 8.28 (s, 1H), 7.67 (s, 1H), 7.51 (d, 3JH,H ¼ 8.5 Hz, 1H), 7.34 (d, 3 JH,H ¼ 8.5 Hz, 1H), 6.70 (s, 1H), 5.96 (m, 1H), 5.60 (m, 1H), 4.64 (m, 1H), 4.58 (m, 1H), 4.46 (m, 1H), 4.34 (m, 1H), 3.70 (s, 3H), 3.67 (s, 3H), 2.51 (m, 1H), 2.40e2.10 (br. m, 6H), 1.99 (m, 1H), 1.72 (s, 3H), 1.49 (s, 3H), 1.45 (s, 3H), 1.32 (s, 3H). e 13C NMR [125 MHz, (CD3)2CO, 298 K]: d ¼ 174.62, 174.58, 168.96, 164.47, 147.94, 146.75, 142.08, 141.51, 136.83, 129.24, 126.62, 124.49, 124.02, 123.30, 121.44, 121.16, 119.83, 111.76, 109.84, 109.52, 108.68, 104.00, 77.49, 76.71, 74.39, 73.72, 73.70, 73.66, 71.68, 71.64, 51.84, 51.80, 36.49, 36.01, 34.98, 34.89, 27.72, 27.55, 25.48, 25.43. e ESI MS: m/z ¼ 809.6 [M H] . C40H42O18 (810.75); calcd. C, 59.26% and H, 5.22%; found C, 59.23% and H, 5.24%. 2.2. Screening for free radical scavengers by DPPH test 2.2.1. TLC DPPH rapid test e qualitative determination and separation into two main groups-active and non-active Compounds 1e4 were dissolved in acetone and spotted onto silica gel 60 F254 plates (E. Merck, Germany). The plates were airdried and sprayed with 0.03% 1,1-diphenylpicrylhydrazyl radical (DPPH) (SigmaeAldrich Chemie GmbH, Stenheim, Germany) solutions in methanol and in acetone for detecting the compounds with rapid scavenging properties. Both DPPH solutions (at about 0.03%) have a purple color. Substances possessing radical scavenging activity react with it and as a result, the purple color disappears. The compounds that showed white or light yellow spots on a purple background were considered as active radical scavengers [10]. The room temperature during this test was 30  C. The time of reaction with the radical, and the effect of the initial concentration (1 mM and 10 mM) of the samples on their radical scavenging activity were also studied. 2.2.2. Kinetics of DPPH absorbance decrease e quantitative determination of radical scavenging activity (RSA) The decrease in the absorption at 516e517 nm due to reaction of DPPH with AH (i.e., the H-atom abstraction from AeH by DPPH with formation of DPPHeH and A) was measured by the Perkin Elmer Lambda 16 UVeVis spectrophotometer, equipped with a HAAKE FE 2 thermostat (precision 1  C) at physiological temperature (37  C). We used acetone as a solvent according to Yordanov [11] and Kancheva et al. [12]. Reaction time is one of the most important criteria in this reaction. DPPH radical absorbance decrease was monitored for 20 min after mixing an excess of DPPH with AH (ratios [AH]/[DPPH] ¼ 0.25 and 0.40 mol/mol). There are two ways to obtain RSAs from the kinetic curves: 1) RSAs of compounds 1e4 were based on the DPPH test of Nenadis and Tsimidou [13] and were calculated from the absorbance at the start (0) and after some reaction time (t) as 406  %RSA ¼ Absð0Þ V.D. Kancheva et al. / Biochimie 94 (2012) 403e415  AbsðtÞ =AbsðtÞ  100% (1) Absorbance values were corrected for DPPH absorption using blank solutions. All tests were performed in duplicate. %RSAmax was obtained after 20 min reaction time. The total stoichiometry ntot {ntot¼(Abs0 Abst)/3 [AH], i.e., the number of free radicals scavenged by one molecule of AH} for t ¼ 20 min was calculated according to Goupy et al. [14]. In our experiments 3 ¼ 12 000 M 1 cm 1 is the molar absorption coefficient of DPPH in acetone solution at 37  C, [AH] e concentration of compounds tested; [AH] ¼ 2.7  10 5 M and 3.9  10 5 M, resp. 2) The initial rate constants ki (i ¼ 1e4) for the reaction between DPPH and compounds 1e4 which represent the most reliable indication of RSA were measured within the fast kinetics (reaction time 0e2 min) from the linear dependence ln[DPPH]/ [DPPH]0 ¼ ki [AH] t. The stoichiometrics of the fast kinetics also was calculated as n2min. 2.3. Chain-breaking antioxidant activity 2.3.1. Lipid samples Triacylglycerols of commercially available sunflower oil (TGSO) were cleaned from pro- and anti-oxidants by adsorption chromatography [15] and stored under nitrogen at 20  C. Fatty acid composition of the lipid substrate was determined by GC analysis of the methyl esters of the total fatty acids obtained according to Christie [16] with a GCeFID HewlettePackard 5890 equipment (HewlettePackard GmbH, Austria) and a capillary column HP INNOWAX (polyethylene glycol mobile phase, Agilent Technologies, USA) 30 m  0.25 mm  0.25 mm. The temperature gradient started from 165  C increased to 230  C with 4  C/min and held at this temperature for 15 min; injection volume was 1 ml. Injector and detector temperatures were 260 and 280  C, respectively. Nitrogen was the carrier gas at a flow rate 0.8 ml/min. The analyses were performed in triplicate. Six different fatty acids were present in TGSO: 16:0e6.7%; 18:0e3.6%; 18:1e25.1%; 18:2e63.7%; 20:0e0.2%; 22:0e0.7%; the numbers “n:x” indicate, respectively, the number of carbon atoms and double bonds in the fatty acid. Lipid samples containing various inhibitors were prepared directly before use. Aliquots of the antioxidant solutions in purified acetone were added to the lipid sample. Solvents were removed under a nitrogen flow. 2.3.2. Lipid autoxidation The process was carried out at 80  C (0.2  C) by blowing air (2.0 ml, at a rate of 100 ml min 1) through the samples in special vessels. The process was monitored by withdrawing samples at measured time intervals and subjecting them to iodometric determination of the primary products (hydroxyperoxides, LOOH) concentration, i.e. the peroxide value (PV) [17]. All kinetic data are expressed as the average of two independent measurents which were processed using the computer programs Origin 6.1 and Microsoft Excel-97. 2.3.3. Determination of the main kinetic parameters of the studied compounds Protection factor (PF) which represents the prolongation of the lag-phase caused by the antioxidant and determined as the ratio between the induction period in the presence (IPA) and in the absence (IPC) of antioxidant, i.e., PF ¼ IPA/IPC. Inhibition degree (ID), is a measure of the antioxidant reactivity e.g. how many times the antioxidant shortens the oxidation chain length, i.e., ID ¼ RC/RA. The initial oxidation rates RC in the absence and RA e in the presence of antioxidant were found from the tangents at the initial phase of the kinetic curves of hydroperoxides accumulation. Main rate of antioxidant consumption (Rm) represents the consumption rate of the inhibitor during the induction period of the inhibited lipid autoxidation, i.e., Rm ¼ [AH]/IPA. Relative main rate of antioxidant consumption (RRm) is the ratio RRm ¼ Rm/RA. Detailed information of all kinetic parameters’ determination is presented in Refs. [4,18,19] Synergism, additivism, antagonism [18]. If two or more antioxidants are added to oxidizing substrates, their combined inhibitory effect can be additive, antagonistic, or synergistic. Synergism e the inhibiting effect of the binary mixtures (IP1þ2) is higher than the sum of the induction periods of the individual phenolic antioxidants IP1þ2 >IP1 þ IP2 (2) The percent of the synergism was calculated by the formulae created from Shahidi [20]: %Synergism ¼  IP1þ2    IP1 þ IP2 IP1 þ IP2  100; % (3) Additivism e the inhibiting effect of the binary mixtures (IP1þ2) is equal to the sum of the induction periods of the individual phenolic antioxidants IP1þ2 ¼ IP1 þ IP2 (4) Antagonism e the inhibiting effect of the binary mixtures (IP1þ2) is lower than the sum of the induction periods of the individual phenolic antioxidants IP1þ2 < IP1 þ IP2 (5) 2.3.4. Statistical analysis of IP determination Ten independent experiments were carried out in association with previous results on inhibited oxidation according to Doerffel [21]. The standard deviation (SD) for different mean values of IP was (in h): IP ¼ 2.0, SD ¼ 0.2; IP ¼ 5.0, SD ¼ 0.3; IP ¼ 15.0, SD ¼ 1.0; IP ¼ 25, SD ¼ 1.5; IP ¼ 50.0, SD ¼ 3.0. The SD of different peroxide values (PV) determination (in meq kg 1), according to the modified iodometric method for different mean values of PV, was: PV ¼ 12.0, SD ¼ 1.0; PV ¼ 30.0, SD ¼ 2.0; PV ¼ 70.0, SD ¼ 5.0; PV ¼ 150.0, SD ¼ 10; PV ¼ 250.0, SD ¼ 20. The RA and RC were quite constant varying by less than 2%. 2.4. Quantum chemical calculations DFT methods give quite accurate results for geometry optimization of closed and open-shell molecules [22]. The Becke’s three parameter hybrid exchange-correlation (B3LYP) functional was chosen to be used in this study because its good performance in geometry optimization, as well as its quite accurate prediction of OeH bond dissociation energies. It has been reported that the B3LYP predicted bond dissociation energies are in good agreement with the values obtained by employing more accurate and expensive MP2 and CCSD methods [23]. The molecular geometries of benzo[kl] xanthene lignans (1 and 2), dihydrobenzofuran neolignans (3 and 4), a-tocopherol (TOH), caffeic acid (CA), ferulic acid (FA) and their radicals are optimized using unrestricted open-shell approach (UB3LYP) and 6-31G(d,p) basis set [24] without symmetry constraints (C1 symmetry was assumed) with the default convergence criteria. No spin contamination is found for radicals. Frequency calculations at the same level of theory were carried out to confirm that the obtained structures correspond to energy minima. Unscaled thermal corrections to enthalpy were added to the total energy values. V.D. Kancheva et al. / Biochimie 94 (2012) 403e415 For description of the free radicals scavenging activity, bond dissociation enthalpy (BDE) is employed. The BDEs for the generation of the respective radicals from the parent compounds 1e4 are calculated by the formula BDE ¼ H298 ðA$Þ þ ET ðH$Þ (6) H298 ðAHÞ where H298(A) and H298(AH) are the B3LYP/6-31G(d,p) calculated enthalpies at 298 K for radical species A and neutral molecule AH, respectively, and ET(H) (calculated total energy of H) is 313.93 kcal mol 1 ( 0.5002728 a.u.). Other energetic and structural parameters like energy of high occupied molecular orbital (HOMO) and spin density are also considered for the scavenging activity evaluation. All quantum chemical calculations were carried out using GAUSSIAN 09 program package [25]. 3. Results and discussion 3.1. Radical scavenging activity toward DPPH radical 3.1.1. Rapid TLC DPPH test This approach is useful for the preliminary selection of new compounds as active and non-active to scavenge free radicals, i.e., DPPH. The yellowish e white spots on the purple background of DPPH radical solutions in methanol and in acetone on the plates were observed (Figure S1 is presented in Supplementary data). It was studied the effects of reaction time (1 min and 10 min), of compounds’ concentrations (1 mM and 10 mM) and of DPPH solvent (acetone and methanol). Table 1 presents the radical scavenging activity of studied compounds depending on the brightness of the spots. Effect of the reaction time. Reaction time is one of the most important criteria in this reaction. We studied the activity of the compounds after 1 min and 10 min from spraying with DPPH radical solution. Compounds with strong activity (þþþ) as radical scavengers react immediately with DPPH and whiteeyellowish spots appear on the TLC. Compounds with moderate (þþ) and weak (þ) activity need longer reaction times. Therefore, this approach is very useful for the preliminary selection of new compounds which can be classified into three main groups, i.e.,: i) with strong activity (their spots appear immediately after spraying), ii) with moderate activity (their spots appear after 10 min), and with weak or no activity (without whiteeyellowish spots after 10 min or the spots are with low brithness). Effect of the concentration. At high concentration (10 mM) all the tested compounds showed yellowishewhite spots immediately or 10 min after spraying the DPPH solution in both solvents, i.e., they all are active towards free radicals. At lower concentration (1 mM) compounds 2 and 3 showed yellowishewhite spots immediately after spraying the DPPH solutions. The brightness of the spots for Table 1 Intensity of spots, characterizing the radical scavenging activity of compounds 1e4 determined by TLC DPPH rapid test. DPPH in solution of Methanol Acetone Concentration 1.0 mM 10 mM Reaction time, min 1 1 Compound Compound Compound Compound þ þ þþþ þþþ þ þþ þþþ þþþ þþ þþ þþþ þþþ þþ þþþ þþþ þþþ þþ þþ þþþ þþþ þþ þþþ þþþ þþþ þ þ þþ þþþ þ þþ þþþ 1 2 3 4 10 10 1.0 mM 10 mM 1 1 10 10 Strong (þþþ), moderate (þþ), weak (þ), and no ( ) radical scavenging activity. 407 compounds 1 and 4 was instead much lower (Table 1). It can, thus, be concluded that the lower concentration e 1 mM of tested compounds gives more precise separation of active radical scavengers. Effect of the solvent. The solvent of DPPH is very important. We usually used acetone as a solvent for antioxidants although we tested the radical scavenging activity of studied compounds in methanol as well. We have recently reported [26] that in the presence of methanol some activity towards the DPPH radical could be found even with compounds lacking free OH groups most likely because some hydrolysis can occur. In our case, we observed (Table 1) that the studied compounds demonstrated almost the same activity in both solvents at higher concentration (10 mM), whereas at lower concentration (1 mM) the activity was higher in acetone than in methanol. We can conclude that with this approach we are able to get much more precise separation of the activity of compounds by verifying the reaction time, concentration and DPPH solvent. 3.1.2. Quantitative determination of radical scavenging activity (RSA) The RSA determinations reported in literature are based on the DPPH absorbance decrease very often measured at different reaction times: 30 min in Blois method [27], shorter times (20, 10 and/or 5 min) from other authors [28e30]. However, it is obvious that reaction time and concentration of the tested compounds are of significance for their radical scavenging activity. In this study we therefore monitored the reaction kinetics for about 20 min keeping the concentration of 1e4 constant and using the same solvent. Fig. 2(a,b) shows the DPPH absorbance decrease for both ratios [AH]/[DPPH]/ ¼ 0.25 (Fig. 2a) and 0.40 mol/mol (Fig. 2b) in the presence of compounds 1e4 presented as (%RSA) according to the equation (1) and the initial kinetics in the time interval 0e2 min (Fig. 2a,b). The data show that, depending of the reaction time used, a different activity of the compounds is obtained. For example: - at the maximal reaction time t ¼ 20 min: compound 3 demonstrated in both ratios a higher %RSAmax, than of compound 2, and much higher one than of compounds 1 and 4 (see Fig. 2a,b). It can be seen that %RSA of 1e4 increased in the higher ratio. Taking into account the structural peculiarities of the compounds under study, we compared the %RSAmax of compounds 1 and 2, and also of 3 and 4. It can be seen that compound 2 showed 4-fold higher % RSAmax than compound 1, and compound 3 demonstrated almost 3.6-fold higher %RSAmax than compound 4 for the ratio [AH]/[DPPH] ¼ 0.25 mol/mol. Total stoichiometry (ntot) values obtained for ratio [AH]/[DPPH] ¼ 0.4 mol/mol are in the order: 3 (0.8) > 2 (0.6) > 4 (0.2)  1 (0.1). Comparison of %RSAmax data obtained with that of standard antioxidants TOH, CA and butylated hydroxyl toluene (BHT) at the same experimental conditions [31] give the following new order of %RSA max: TOH (61.1%) > CA (58.6%) > 3 (36.3%) > 2 (28.1%) > 4 (6.7%) > 1 (3.6%) ¼ BHT (3.6%). According to Ref. [31] we separated compounds into the same main groups as radical scavengers: Group A e compounds with strong activity (%RSA > 40%), i.e., TOH and CA; Group B e compounds with moderate activity (15 < %RSA < 40%), i.e., compounds 2 and 3; and Group C e compounds with weak activity (%RSA < 15%), i.e., compounds 1,4 and BHT. - for the fast kinetics (time interval 0e2 min): it can be seen that in this case compound 2 showed the highest RSA2min: 2-fold higher than that of compound 3; 6-fold higher that compound 4 and 7-fold higher that compound 1 (see Fig. 2c,d). Comparison with TOH, CA and BHT showed the following order (at ratio [AH]/[DPPH] ¼ 0.4) of %RSA2min: CA (43.7%) > TOH (38.8%) > 2 (23%) > 3 (11.8%) > 4 (3.7%) > 1 (3.1%) > BHT (2.5%). 408 V.D. Kancheva et al. / Biochimie 94 (2012) 403e415 Fig. 2. Kinetics of DPPH absorbance decrease at 517 nm at 37  C at different ratios of [AH]/[DPPH] ¼ 0.25 mol/mol (a, c) and [AH]/[DPPH] ¼ 0.40 mol/mol (b,d): radical scavenging activity, %RSA (a,b); the fast kinetics, during the first 2 min (c,d). The rate constants (23, 207, 50, 22 M 1 s 1 for 1e4 resp.) for the ratio [AH]/[DPPH] ¼ 0.4 mol/mol and the rate constants (13, 53, 32 and 18 M 1 s 1 for 1e4 respectively) for the ratio [AH]/[DPPH] ¼ 0.25 mol/mol were calculated and indicate the order 2 > 3 > 4 > 1 in agreement with the RSAs calculated within 2 min. New orders of relative stoichiometry ntot/n2min: 3 (20) >> 4 (7.9) > 1 (4.2) > 2 (3.5) and of relative radical scavenging activity % RSAmax/%RSA2min: 3 (3.1) > 2 (1.2) ¼ 1 (1.2) > 4 (0.6). As it can seen the relative stoichiometry of compound 3 is 20. This fact can be explained with the additional reactions of A with DPPH and with A leading to different products formation which may be also active as radical scavengers. 3.2. Chain-breaking antioxidant activity Abbreviations in the kinetic scheme: d e is the yield of lipid peroxide radicals (LO2) formation during the chain branching reaction; n e is the stoichiometry of the reaction between antioxidant and LO2, i.e., how many radicals were trapped by one molecule of antioxidant AH e monophenolic, QH2 e biphenolic, depends on antioxidant’s structure and mechanism. P1, P2, PA e reaction inactive products, Q e quinone, AeA and QHeQH are the corresponding dimers. 3.2.1. Basic kinetic scheme of lipid autoxidation [4,18,19] Non-inhibited lipid (LH) autoxidation (in absence of an antioxidant). Chain generation LH þ O2 ðYÞ/LO2 $ þ HO2 $ LO2 $ þ LHðþO2 Þ/LOOH þ LO2 $ Chain termination LO2 $ þ LO2 $/P1 þ ð1 LOOH þ O2 ðþLH; LOOHÞ/dLO2 $ dÞP2 ðkb Þ It has been proven, that during lipid autoxidation the key chain branching reaction is the pseudo-monomolecular one [32,33]. The rate of non-inhibited oxidation (R0) is presented by the following eq. R0 ¼ kp ½LHŠðRIN =kt Þ0:5 (7) Chain length of non-inhibited oxidation (n0) is presented as n0 ¼ R0/RIN In presence of an antioxidant (biphenolic, QH2 or monophenolic, AH) the following reactions are possible: Inhibited autoxidation (in presence of a biphenolic antioxidant QH2). Inhibition 1 LO2 $ þ QH2 /LOOH þ QH$ Inhibition 2 LO2 $ þ QH$/PA Inhibition 3 ðrecombinationÞ  kp  ðkt Þ The RIN during lipid autoxidation depends significantly from the chain branching reactions by different mechanisms: as monomolecular (LOOH); pseudo-monomolecular (LOOH þ LH) and bimolecular (LOOH þ LOOH) reactions: ðkA Þ   kA0 2QH$/QHeQH Inhibition 4ðdisproportionationÞ Rate of inhibited oxidationðRA Þ ðRIN Þ Chain propagation Chain branching ðkR Þ 2QH$/QH2 þ Q ðkD Þ RA ¼ kp ½LHŠRIN =nkA ½QH2 Š0 Inhibited autoxidation (in presence of a monophenolic antioxidant AH). Inhibition 1 LO2 $ þ AH/LOOH þ A$ Inhibition 2 LO2 $ þ A$/PA Inhibition 3ðrecombinationÞ ðkA Þ   kA0 2A$/AeA ðkR Þ V.D. Kancheva et al. / Biochimie 94 (2012) 403e415 Rate of inhibited oxidationðRA Þ 409 RA ¼ kp ½LHŠRIN =nkA ½AHŠ0 Chain length of inhibited oxidationðnA Þ nA ¼ RA =RIN One of the most important kinetic parameter the inhibition degree (ID) can be determined as the ratio between the oxidation chain length of non-inhibited and inhibited oxidation, i.e., by eq. ID ¼ R0 =RA ¼ n0 =nA (8) 3.2.2. Chain-breaking antioxidant activity of individual compounds Fig. 3(a,b) presents the kinetics of lipid (TGSO) autoxidation in the absence and in the presence of 0.1 mM (Fig. 3a) and 1.0 mM (Fig. 3b) of compounds 1e4. All main kinetic parameters characterizing the lipid oxidation process are shown in Table 2. These parameters show the following: a) at low concentration (0.1 mM) PF: 2 (7.2) > 1 (3.1) > 3 (2.2) >> 4 (0.6); ID: 2 (44.0) >> 1 (8.4) > 3 (2.8) >> 4 (0.9); Rm/10 8 M/s: 2 (0.3) < 1 (0.7) < 3 (1.0) < 4 (3.5) Compound 2 is the best chain-breaking antioxidant followed in order by compounds 1 and 3. Compound 4 not only did not show any antioxidant activity but it even proved to be a prooxidant. b) at high concentration (1.0 mM) PF: 2 (73) >> 1 (28)  3 (20) >> 4 (3.0) ID: 2 (88) >> 1 (44) > 3 (17.6) >> 4 (3.1) Rm/10 8 M/s: 2 (0.3) < 1 (1.0) < 3 (1.4) < 4 (9.3) The higher concentration of 1.0 mM increased by 10-fold the PF value of compounds 2 and 3 and by 9-fold that of compound 3 compared to those at 0.1 mM. For compound 4 at the higher concentration of 1.0 mM showed a week antioxidant activity. It can be concluded that compounds 1e3 were more efficient antioxidants at a 10-fold higher concentration (1.0 mM) since the protection factor (PF) and inhibition degree (ID) grew accordingly. 3.2.3. Comparable kinetic analysis with standard and known antioxidants The considerable differences in the molecular structures of 1e4 prevented us from drawing any conclusions on some structure/ activity relationships. Therefore, we preferred to make a comparable analysis with known and standard antioxidants. We selected caffeic acid (CA) taking into account not only the fact that it is one of the best biphenolic antioxidants with a catecholic structure but also that compounds 1e4 were obtained by caffeic acid phenethyl or methyl esters [7e9]. Ferulic acid (FA) was also chosen because it has the same ortho-methoxy-phenolic moiety like compound 4. On the other hand, a-tocopherol (TOH) was selected because it is one of the strongest natural phenolic antioxidant. All main kinetic parameters found with the same kinetic model (TGSO, 80  C) during bulk lipid autoxidation as antioxidant efficiency (PF), antioxidant reactivity (ID) and antioxidant capacity (Rm) were compared and analysed. The following orders regarding the main kinetic parameters PF, ID and Rm at the concentration of 0.1 mM were obtained: PF: 2 (7.2)  TOH (7.0)  CA (6.7) > 1 (3.1) > 3 (2.2) > FA (1.5) > 4 (0.6); ID: 2 (44.0) >> TOH (18.7) >> CA (9.3) >> 1 (8.4) > 3 (2.8) > FA (1.0) > 4 (0.9); Rm/10 8 M/s: CA (0.3) ¼ TOH (0.3) ¼ 2 (0.3) < 1 (0.7) < 3 (1.0) < FA (1.3) < 4 (3.5) Compound 2 demonstrated the highest antioxidant activity in terms of PF, whose value is close to that of one (TOH) of the most Fig. 3. Kinetics of TGSO autoxidation at 80  C in absence (C) and in presence of 0.1 mM (a) and in 1.0 mM (b) of compounds 1e4; and in presence of equimolar binary mixtures (1:1) of studied compounds (0.1 mM) with a-tocopherol (c): 1 (1 þ TOH), 2 (2 þ TOH), 3 (3 þ TOH), 4 (4 þ TOH). efficient antioxidants and higher than that of CA. Furthermore, compound 2 was also much more efficient to shorten the lipid oxidation chain length (see ID values) compared to TOH and CA. At higher concentration (1.0 mM) the following orders were obtained: PF: 2 (73) >> CA (43) >> 1 (28)  3 (20) ¼ TOH (20) >>> FA (4.0) > 4 (3.0) 410 V.D. Kancheva et al. / Biochimie 94 (2012) 403e415 Table 2 The main kinetic parameters, characterizing TGSO autoxidation at 80  C in presence of 0.1 and 1.0mM of the studied compounds: A) Individual compounds; B) Standard antioxidants; C) Binary mixtures with TOH (1:1). For control sample without additives: IPc ¼ (1.3  0.2) h, Rc¼ (8.8  0.5)  10 6, M/s. Compound Abbr. Concentrations [mM] Induction period, IPA [h] A) Individual compounds 1 0.1 4.0 1.0 30 2 0.1 9.3 1.0 93 3 0.1 2.8 1.0 18.4 4 0.1 0.8 1.0 3.3 B) Standard antioxidants TOH 0.1 10.5 1.0 20 FA 0.1 2.0 1.0 4.0 CA 0.1 10.0 1.0 43 C) Binary mixtures with a-tocopherol 1 þ TOH 0.1 10.0 2 þ TOH 0.1 15.0 3 þ TOH 0.1 14.0 4 þ TOH 0.1 13.8 Protection factor, PF [-] Initial rate RA, 10 6 [M/s] Inhibition degree, ID [-] Mean consump. rate, Rm [M/s]         0.3 3 0.8 5 0.3 2.1 0.1 0.3 3.1 21.5 7.2 73.1 2.2 14.1 0.6 2.5 1.1 0.2 0.2 0.1 3.2 0.5 9.6 2.8         0.3 0.1 0.1 0.1 0.3 0.1 0.5 0.2 8.0 44 44 88 2.8 17.6 0.9 3.1 6.9 9.9 3.0 2.9 9.9 1.4 3.5 9.3         10 10 10 10 10 10 10 10 9       0.8 2 0.2 0.3 0.8 3 7.0 13.3 1.5 2.7 6.7 33.1 0.3 0.4 5.6 1.8 0.6 0.2       0.1 0.1 0.5 0.3 0.2 0.2 18.7 14.0 1.0 3.1 9.3 28.0 0.3 1.4 1.3 6.9 0.3 0.6       10 10 10 10 10 10 8     0.8 0.8 0.8 0.8 8.0 12.0 11.2 11.0 0.5 0.4 0.4 0.5     0.3 0.2 0.2 0.3 18.3 22.2 22.2 19.0 0.3 0.2 0.2 0.2     10 10 10 10 8 9 9 9 9 8 8 8 8 8 8 8 8 8 8 8 Relative mean consump. rate, RRm [-] 6.6 2.0 15.0 2.9 3.1 2.8 3.6 3.3         10 3 10 2 10 2 10 2 10 3 10 2 10 3 10 2 1.0 3.5 2.3 3.8 0.5 3.0       10 10 10 10 10 10 2 0.6 0.5 0.5 0.4     10 10 10 10 2 2 3 2 2 2 2 2 2 Activity synergism, %, antagonism, % Refs. Moderate Moderate Strongest Strongest Weak Moderate No activity Weak tw tw tw tw tw tw tw tw Strongest Moderate Weak Weak Strongest Strongest 4 tw 4 14 4 4 Antagonism, 31% Antagonism, 24% Synergism, 5.3% Synergism, 22% tw tw tw tw tw- this work. ID: 2 (88) >> 1 (44) >> CA (28) > 3 (17.6) > TOH (14) >> 4 (3.1) ¼ FA (3.1) Rm/10 8 M/s: 2 (0.3) < CA (0.6) < 1 (1.0) < 3 (1.4) ¼ TOH (1.4) < FA (6.9) < 4 (9.3) At higher concentration (1.0 mM) compounds 1e3 are better antioxidants than TOH (see and compare PF and ID). The presence of catecholic structures in the antioxidant molecules is one of the most important factors that influence their chain-breaking antioxidant activity (see Fig. 3a,b and Table 2). The large antioxidant activity of catechols has been attributed to the reinforcement of the intramolecular hydrogen bond in the aryloxyl radical (QH) [34]. This phenomenon determines a weakening of the reactive OeH and boosts the antioxidant activity of the parent compounds. Compound 2 is the best antioxidant compared to all other compounds both at low and high concentrations. Its ID at 0.1 mM is 2-, 4-, 5.5- and 14-fold higher than that of TOH, CA, compound 1 and compound 4, respectively. At higher concentration (1.0 mM) compounds 1 and 3 which have catecholic structures showed an antioxidant activity similar to that of CA and TOH. These antioxidant activities are also controlled by the steric hindrance around the reactive OeH and by different levels of side reactions at which the aryloxyl radicals may participate. The most important factor in determining the antioxidant activity of a phenol remains however the BDE of its reactive OeH group. Generally, the lower the BDE of AH the faster is the reaction of H-atom abstraction by LO2 radicals (see below). However, solvent effects play an important role in these reactions [35]. The much lower chain-breaking antioxidant activity of compound 4, which is monophenolic compound, is in fact due essentially to its relatively high OeH BDE (see Table 3). The stoichiometric factor n of this compound is also significantly lower (n4 ¼ 0.15) than the other ones. This implies that the radicals A can recombine forming an inactive dimer (AeA). As a result, we observed with this compound little retardation effects at high concentration but some prooxidant activity at lower concentration, a result compatible with the following side reactions: Reaction of the phenoxyl antioxidant: AH þ LOOH / d (LO2) þ Product (acceleration of LOOH decomposition to active peroxide radicals) Reaction of the phenoxyl radical: A þ O2 / AO2 (oxidation) A þ LOOH (þO2) / LO2 þ AH k-A (reversed key reaction of inhibited oxidation)-this reaction is of significance for phenolic compounds without bulky substituent in ortho-position A þ LH (O2) / AH þ LO2 kLH (generation of LO2)-this reaction is significant for LH with high concentration of polyunsaturated fatty acids. As a result of this side reaction the activity of TOH decreases in higher concentrations especially in presence of polyunsaturated fatty acids of oxidizing lipid substrate. 3.2.4. Effect of equimolar binary mixtures of studied compounds with DL-a-tocopherol [36,37] In order to study the possible synergism between two phenolic antioxidants, the antioxidant efficiency (as PF) and reactivity (as ID) of the equimolar binary mixtures of studied compounds and TOH were tested and compared. Fig. 3c and Table 2 show all results obtained. A higher oxidation stability of the lipid substrate was found in the presence of binary mixtures 2 þ TOH, 3 þ TOH and 4 þ TOH. However, an actual synergism was only obtained for the binary mixtures with compounds 3 and 4. IPS (14.0) > IP3 (2.8) þ IPTOH (10.5) for the binary mixture 3 þ TOH e Synergism IPS (13.8) > IP4 (0.8) þ IPTOH (10.5) for the binary mixture 4 þ TOH e Synergism On the contrary, an antagonism was observed for the binary mixtures of (1 þ TOH) and (2 þ TOH): IPS (10.0) < IP1 (4.0) þ IPTOH (10.5) for the binary mixture 1 þ TOH e Antagonism IPS (15.0) < IP2 (9.3) þ IPTOH (10.5) for the binary mixture 2 þ TOH e Antagonism The compounds with strong (2) and moderate (1) antioxidant activities demonstrated antagonism in their binary mixtures with TOH. Addition of 1 to the substrate containing TOH didn’t lead to an increase in the oxidation stability. In fact, it was observed that the IP for the binary mixture was the same as for the substrate with TOH only. On the other hand, the mixture 2 þ TOH gave a better protection against oxidation than the single antioxidants, though there was no synergism between these two strong antioxidants. 411 V.D. Kancheva et al. / Biochimie 94 (2012) 403e415 Table 3 B3LYP/6-31G(d,p) calculated enthalpies (H298) at 298 K (in Hartree), enthalpy differences (∆H298) between rotamers and enantiomers of compounds 1-4 and CA, FA, TOH, and their radicals (in kcal/mol), bond dissociation enthalpy (BDE) (in kcal/mol), spin density (in e ) and energy of high occupied molecule orbital (HOMO) (in a.u.). Compound (eV) H298 1 1-6O 1-9O 1-10O 10 10 -6O 10 -9O 10 -10O 2 2-6O 2-9O 2-10O 20 20 -6O 20 -9O 20 -10O 3 (2R,3R) 3-30 O 3-40 O 3-7O 30 (2R,3R) 30 -30 O 30 -40 O 30 -7O 3 (2S,3S) 3-30 O 3-40 O 3-7O 30 (2S,3S) 30 -30 O 30 -40 O 30 -7O 4 (2R,3R) 40 (2R,3R) 4-40 O(2R,3R) 4 (2S,3S) 40 (2S,3S) 4-40 O(2S,3S) CA CA-3O CA-4O CA0 CA0 -3O CA0 -4O FA FA0 FA-4O TOH TOH0 TOH-O a b c d Value Value Value Value calculated calculated calculated calculated 1912.584824 1911.965566 1911.959253 1911.973574 1912.581943 1911.964069 1911.974189 1911.957797 2902.835741 2902.217355 2902.210381 2902.222884 2902.831042 2902.214428 2902.223843 2902.206411 1914.926273 1914.311355 1914.297075 1914.297343 1914.926051 1914.296093 1914.312057 1914.296453 1914.923209 1914.307427 1914.293477 1914.294945 1914.922627 1914.292024 1914.308278 1914.293800 1452.947914 1452.940784 1452.319119 1452.942030 1452.935184 1452.315149 648.511541 647.895731 647.884563 648.511018 647.880391 647.899250 687.787452 687.780274 687.160923 1284.995152 1284.994722 1284.383557 for for for for ∆H298 (molecules) BDE Spin density at O 5.41 9.37 0.39 74.7 78.6 69.6 0.276 0.287 0.325 6.35 0.00 10.29 73.8 67.4 77.7 0.276 0.236 0.367 4.07 8.45 0.60 74.2 78.6 70.8 0.269 0.289 0.335 5.91 0.00 10.94 73.1 67.2 78.2 0.267 0.308 0.388 0.44 9.40 9.23 71.9 80.9 80.7 0.353 0.399 0.412 10.02 0.00 9.79 81.4 71.4 81.2 0.400 0.355 0.413 0.53 9.29 8.37 72.5 81.2 80.3 0.355 0.400 0.411 10.20 0.00 9.09 81.8 71.6 80.7 0.401 0.357 0.413 0.00 80.6 0.386 2.49 2.21 9.22 79.4 0.386 72.5 79.5 0.358 0.328 11.83 0.00 81.8 70.0 0.407 0.301 79.2a/74.7b 0.314 69.9c/69.6d 0.370 ∆H298 (radicals) 0.00 1.81 0.00 2.95 0.00 0.14 1.92 2.29 0.00 4.47 3.69 7.99 0.00 0.33 0.00 4.50 0.00 0.27 HOMO 5.06 5.00 5.26 5.24 5.05 4.99 5.29 5.24 4.94 4.86 5.10 5.06 4.91 4.83 5.11 4.99 5.73 5.86 5.75 5.80 5.64 5.80 5.85 5.72 5.71 5.85 5.83 5.77 5.63 5.80 5.88 5.72 5.68 5.63 5.58 5.77 5.73 5.69 5.80 6.04 5.83 5.84 5.97 5.92 5.71 5.68 5.71 4.88 4.95 4.88 FA. FA0 . TOH. TOH0 . The latter behavior is similar to other cases of strong phenolic antioxidants in mixture with TOH previously reported by some of us [4,18,26]. The synergism manifested by some of the above phenols in admixture with TOH is due to reactions which regenerate the most active antioxidant: a) For monophenolic antioxidants, AH, (compound 4) with TOH regeneration of the strongest antioxidant (in our case TOH) is achieved by the reaction: TO þ AH / TOH þ A (regeneration of TOH) b) For active biphenolic compounds, QH2 (compound 3) regeneration of both antioxidants can be achieved through the crossedisproportionation reaction: QH þ TO / QH2 þ T ¼ O regeneration of QH2 TO þ QH / TOH þ Q regeneration of TOH. where: T ¼ O is tocopheryl quinone, formed by abstraction of a hydrogen atom from the tocopheryl radical TO and/or with reactions of the phenoxyl radical from one antioxidant with the second antioxidant: QH þ TOH / QH2 þ TO regeneration of QH2 TO þ QH2 / TOH þ QH regeneration of TOH. 412 V.D. Kancheva et al. / Biochimie 94 (2012) 403e415 3.3. Quantum chemical calculations The knowledge of the conformational, electronic and geometrical features of studied compounds is of crucial importance to understand the relationship between the molecular structure and antioxidant activity. In order to explain the differences in radical scavenging activity of benzo[kl]xanthene lignans 1 and 2 and dihydrobenzofuran neolignans 3 and 4, we have optimized the geometries of the compounds and all possible phenoxyl radicals of the parent compounds. The optimized structures of compounds 1e4 and their radicals are presented on Fig. 4(aed). It is proved [7] that enantiomers of compound 4 differ in their biological activity, e.g., (2R,3R) enantiomer shows promising antitumor properties against breast cancer and is more potent than (2S,3S) enantiomer. For that reason (2R,3R) and (2S,3S) enantiomers of compounds 3 and 4 are considered in the theoretical study. For the compounds 1e3 only rotamers with intramolecular hydrogen bonds are studied. These rotamers differ in the position of the hydrogen atoms from the hydroxyl groups in the catechol moiety of compounds 1e3. The denoted with and without “prime” rotamers are shown on Fig. 4aed. Two rotamers of compound 4, with and without intramolecular hydrogen bond, are considered. For the purpose of comparison rotameric forms and the possible radicals of CA, FA and TOH are also examined (Fig. 4e). The bond lengths of the investigated compounds are presented on Figure S2 in Supplementary data. Comparing bond lengths in the common fragment of 1 and 2, and of 3 and 4, no significant differences between them are found. Fig. 4. B3LYP/6-31G(d,p) optimized structures of different isomers and possible radicals of studied compounds: 1 (a), 2 (b), 3 (c), 4 (d) and of standard antioxidants CA, FA, TOH (e). V.D. Kancheva et al. / Biochimie 94 (2012) 403e415 413 Fig. 4. (continued). Quantum chemical calculated parameters characterizing the neutral molecules 1e4, all possible radicals formed (Fig. 4aed) and standard antioxidants TOH, CA and FA (Fig. 4e) are collected in Table 3. In all cases the rotamer of the neutral compound denoted without “prime” is more stable: 1 < 10 (1.81 kcal mol 1); 2 < 20 (2.95 kcal mol 1); 3 (2R,3R) < 30 (2R,3R) (0.14 kcal mol 1) and 3 (2S,3S) < 30 (2S,3S) (0.37 kcal mol 1) i.e. the rotamers of both enantiomers are almost isoenergetic; 4 (2R,3R) < 40 (2R,3R) (4.47 kcal mol 1) and 4 (2S,3S) < 40 (2S,3S) (4.30 kcal mol 1). The enantiomer (2R,3R) is more stable than (2S,3S) one in both compounds 3 and 4 by 1.92 and 3.69 kcal mol 1, respectively. The enthalpy difference (∆H) between the rotamers of CA is 0.33 kcal mol 1, while ∆H for the rotamers of FA is higher (4.50 kcal mol 1) due to the formation of the intramolecular hydrogen bond in the more stable form. Two rotameric forms are considered for TOH, the difference ∆H is 0.27 kcal mol 1. Formation of three radicals is possible from each rotamer of compounds 1e3, two radicals of CA and one radical of compound 4, TOH and FA, respectively (Table 3). The optimized geometries of the radical species are presented on Fig. 4aee and the relative energies of the radicals of compounds 1e4 and TOH, CA and FA are listed in Table 3. The complete geometrical parameters of all investigated systems are available on request. The rotamers 10, 30 (both enantiomers) and CA0 yield most stable radicals while for compound 2 the more stable rotamer yields most stable radical. However, in all cases the most stable radicals show lowest BDE. BDE values of OeH groups in position 9 of compounds 10 and 20 (10 -9O and 20 -9O) are much lower than in positions 6 and 10. For compounds 1 and 2 the lowest value of BDE is calculated for 1-10O and 2-10O. These results could be explained by the formation of intramolecular hydrogen bonds from the catecholic moiety and subsequent stabilization of corresponding radicals. BDE values for rotamers compound 3 (both enantiomers) are higher than those of 1 and 2. The values for 30 -40 O radicals of both enantiomers are lowest (71.4 and 71.6 kcal mol 1 for (2R,3R) and (2S,3S) enantiomer, respectively). These results are consistent with the results for compounds 1 and 2 e the 30 -40 O radicals are stabilized by intramolecular hydrogen bond. For compound 4, where such stabilization is impossible, the BDE values are much higher e 80.6 and 79.4 kcal mol 1 for 4-40 O (2R,3R) and 4-40 O (2S,3S) respectively. The results for compounds 1e4 are compared with these for standard antioxidants CA, FA and TOH. BDE (10 -9O) and BDE (20 -9O) values are lower than the value of CA0 -4O (70.0 kcal mol 1) and TOHeO (69.9 kcal mol 1). BDE (1-10O) and BDE (2e10O) values related to more stable rotameric forms are also close to the standard compound BDEs. BDE values of compounds 1 and 2 (and to some extent of 3) are close to those of the strongest antioxidants CA and TOH. The conclusion can be drawn that compounds 1 and 2 (and to some extent 3) will manifest antioxidant activity similar to that of CA and TOH. 414 V.D. Kancheva et al. / Biochimie 94 (2012) 403e415 Fig. 4. (continued). The spin density values of the atoms constituting the radicals studied show that the spin delocalization spreads over all atoms participating in the conjugation system of the molecule. The values of spin density at oxygen atom (O) from the hydroxyl group (listed also in Table 3) are not characteristic for the antioxidant activity. Calculated HOMO values (in eV) for all neutral molecules and radicals are listed in Table 3 and the visualization of HOMO is presented on Figure S3 in Supplementary data. The values in the table show that compound 2 is the one with the lowest HOMO energy. HOMO energies calculated for compound 2 are close to those calculated for TOH. The values for compound 1 are little bit higher than that of 2. HOMO values for 3 and 4 are higher and comparable to values calculated for CA and FA. According to Zhang and Chen [38] the higher the HOMO energy, the more active is the antioxidant; HOMO energy is very good to characterize the activity of some tocopherols. In our study HOMO energy and BDE of TOH are characterized with the lowest values (HOMO (TOH) ¼ 4.88 eV, BDE (TOH) < 70 kcal mol 1) in comparison to the values for other standard antioxidants (CA and FA). The results for compounds 1e4 are also inconsistent with the statement that active antioxidants have high HOMO energies e although compound 2 is predicted as most active one (BDE(20 -9O) ¼ 67.2 kcal mol 1) its HOMO energy is low. Consequently, HOMO energy cannot be used to predict the scavenging activity of investigated compounds. The invalidity of HOMO to characterize antioxidant activity of compounds 1e4 (and of standard antioxidants CA, FA and TOH) can be explained by the suggested weakness of this parameter in predicting of scavenging activity for antioxidants possessing intramolecular hydrogen bonds [38]. It can be concluded that the calculated BDE can be used as descriptor for prediction of RSA and/or antioxidant activity. 3.4. Correlation between experimental and theoretical data Correlation between all experimental and theoretical data obtained demonstrated that compound 2 has strongest activity in all methods applied; compounds 1 and 3 showed moderate/strong activity and compound 4 e weakest activity. There is an excellent correlation between the main kinetic parameters of the chainbreaking antioxidant activity of compound 2 with the highest activity and compound 4 with the weakest activity (PF and ID) and the theoretical descriptor BDE. Furthermore, it can be concluded that a synergism was found for the binary mixtures of TOH with compounds which BDE is higher than of TOH (see Table 3). Binary mixtures of TOH with compounds which BDE is lower than of TOH show an antagonism between them. For that reason we recommended BDE to be used as reliable theoretical descriptor for better selection of new bio-antioxidants. 4. Conclusions We report for the first time on the capacity of new biologically active compounds to scavenge free radicals (antiradical activity) and to inhibit lipid oxidation processes (chain-breaking antioxidant activity), i.e., their ability to react as bio-antioxidants. Compound 2 demonstrated strongest activity in all methods applied. Compounds 1 and 3 showed moderate/strong activity while compound 4 e the weakest activity. Compound 2 is the best antioxidant compared to all other compounds both at low and high concentrations. Its ID at 0.1 mM is 2-, 4-, 5.5- and 14-fold higher than that of TOH, CA, 1 and 4, respectively. At higher concentration (1.0 mM) compounds 1 and 3 which have catecholic structures showed an antioxidant activity similar to that of CA and TOH. An excellent correlation between the V.D. Kancheva et al. / Biochimie 94 (2012) 403e415 main kinetic parameters (PF and ID) of the chain-breaking antioxidants and their OeH BDE was found for compound 2 with the highest activity and compound 4 with the weakest activity. There is no clear correlation between the experimental and theoretical parameters for compounds 1 and 3, showing moderate/strong activities. We cannot explain the much lower RSA of compound 1 in comparison with compound 2. Additional studies of reaction products formed of compound 1 must be done, aiming to shed some light on the mechanism of action of this compound. In our kinetic investigations, we also observed synergism of compounds 3 and 4 with TOH which can be interpreted in terms of the higher OeH BDEs of these phenols compared to TOH. These phenols regenerate (the stronger antioxidant) TOH from TO radical and for this determine an increase in the IP of the mixture. On the other hand, compounds 1 and 2 which have lower OeH BDEs showed antagonism with TOH. Acknowledgements The authors appreciate the possibility to run DPPH absorbance kinetics on UVeVis spectrophotometer at 37  C and the helpful discussion with Prof. DSc L. Antonov (IOCCP -BAS). S.A and V.E. acknowledge the financial support of the Bulgarian Fund for Scientific Research under Grant DO 02-217/2008. The calculations were performed on the computer system installed at IOCCP-BAS with the financial support of the Bulgarian Scientific Research Fund under Project “MADARA” (RNF01/0110, contract N DO02-52/2008). The financial support from the Italian Ministry for Education, University and Research, General Management for the internationalization of scientific research is gratefully acknowledged. Appendix. Supplementary data Supplementary data associated with this article can be found in the online version, at doi:10.1016/j.biochi.2011.08.008. References [1] E.B. Burlakova, Russ. Chem. J. 51 (2007) 3e12. [2] M.C. Foti, S.K. Sharma, G. Shakya, A.K. Prasad, G. Nicolosi, P. Bovicelli, B. Ghosh, H.G. Raj, R.C. Rastogi, V.S. Parmar, Pure Appl. Chem. 77 (2005) 91e101. [3] V. Kancheva, New bio-antioxidants strategy, IL-50, in: B.o. Abstracts (Ed.) T3D-2010 International Symposium on Trends in Drug Discovery and Development, Delhi, India, 5e8 January 2010. [4] V. Kancheva, Phenolic antioxidants of natural origin e structure activity relationship and their beneficial effect on human health Chapter 1. in: A.A. Farooqui (Ed.), Phytochemicals and Human Health: Pharmacological and Molecular Aspects. Nova Science Publishers Inc., USA, 2010. [5] S. Apers, D. Paper, J. Bürgermeister, S. Baronikova, S.V. Dyck, G. Lemière, A. Vlietinck, L. Pieters, J. Nat. Prod. 65 (2002) 718e720. [6] Y.W. Choi, S. Takamatsu, S.I. Khan, P.V. Srinivas, D. Ferreira, J. Zhao, I.A. Khan, J. Nat. Prod. 69 (2006) 356e359. [7] C. Daquino, A. Rescifina, C. Spatafora, C. Tringali, Eur. J. Org. Chem. 36 (2009) 6289e6300. 415 [8] S. Di Micco, F. Mazué, C. Daquino, C. Spatafora, D. Delmas, N. Latruffe, C. Tringali, R. Riccio, G. Bifulco, Org. Biomol. Chem. 9 (2011) 701e710. [9] G. Basini, L. Baioni, S. Bussolati, F. Grasselli, C. Daquino, C. Spatafora, C. Tringali, Invest. New Drugs, 2010 Oct 6. [Epub ahead of print]. [10] K.R. Prabhakar, V.P. Veerapur, P. Bansal, V.K. Parihar, K.M. Reddy, A. Barik, B. Kumar, D. Reddy, P. Reddanna, K.I. Priyadarsini, M.K. Unnikrishnan, Bioorg. Med. Chem. 14 (2006) 7113e7120. [11] N.D. Yordanov, Appl. Magn. Reson. 10 (1996) 339e359. [12] V.D. Kancheva, P.V. Boranova, J.T. Nechev, I.I. Manolov, Biochimie 92 (2010) 1138e1146. [13] N. Nenadis, M. Tsimidou, J. Am. Oil Chem. Soc. 79 (2002) 1191e1195. [14] P. Goupy, C. Dufour, M. Loonis, Q. Dangles, J. Agric. Food Chem. 51 (2003) 615e622. [15] N. Yanishlieva, A. Popov, Die Nahrung 15 (1971) 671e681. [16] W.W. Christie, Lipid Analysis, third ed. The Oily Press, Dundee, Scotland, 2003. [17] A. Popov, N. Yanishlieva, J. Slavcheva, Compt. Rend. Acad. Bulg. Sci. 21 (1968) 443e446. [18] E.T. Denisov, T.G. Denisova, Handbook of Antioxidants. Bond Dissociation Energies, Rate Constants, Activation Energies and Enthalpies of Reactions, second ed. CRC Press, New York, 2001. [19] V.D. Kancheva, Eur. J. Lipid Sci. Technol. 111 (2009) 1072e1089. [20] F. Shahidi, P.K.J.P.D. Wanasundara, Crit. Rev. Food Sci. Nutr. 32 (1992) 67e103. [21] K. Doerffel, Statistics in the Analytical Chemistry. Mir, Moscow, 1994. [22] M. Leopoldini, N. Russo, M. Toscano, Food Chem. 125 (2011) 288e306. [23] G.A. DiLabio, J. Phys. Chem. A 103 (1999) 11414e11424. [24] M.J. Frisch, J.A. Pople, J.S. Binkley, J. Chem. Phys. 80 (1984) 3265e3269. [25] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery, J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.01. Gaussian Inc., Wallingford CT, 2009. [26] V.D. Kancheva, L. Saso, P.V. Boranova, M.K. Pandey, S. Malhorta, J.T. Nechev, A.K. Prasad, M.B. Georgieva, A.L.D. Pass, V.S. Parmar, Biochimie 92 (2010) 1089e1100. [27] M.S. Blois, Nature 181 (1958) 1199e1200. [28] V. Bondet, W. Brand-Williams, C. Berset, Lebensmittel-Wissenschaf und Tecnologie/Food Sci. Technol. 30 (1997) 609e615. [29] P. Molyneux, Songklanakarin J. Sci. Technol. 26 (2004) 211e219. [30] J.-K. Kim, J.H. Noh, S. Lee, J.S. Choi, H.Y.C.H. Suh, Y.-O.W.C. Choi, Bull. Korean Chem. Soc. 23 (2002) 661e662. [31] V. Kancheva, A. Slavova-Kazakova, P. Boranova, A. Prasad, V.S. Parmar, Structureeactivity Relationship of Some New Biophenolics. Correlation between Their Radical Scavenging and Chain-breaking Antioxidant Activities, 8th Euro Fed Lipid Congress, Oils, Fats and Lipids: Health & Nutrition, Chemistry & Energy, 21e24 November 2010, Munich, Germany, LAMI-001. [32] V.A. Belyakov, V.D. Kortenska, V.S. Rafikova, N.V. Yanishlieva, Kinet. Catal. 33 (1992) 617e622. [33] V.D. Kortenska-Kancheva, V.A. Belyakov, Riv. Ital. Delle Sost. Grasse 82 (2005) 177e185. [34] M.C. Foti, E.R. Johnson, M.R. Vinqvist, J.S. Wright, L. Ross, C. Barclay, K.U. Ingold, J. Org. Chem. 67 (2002) 5190e5196. [35] M.C. Foti, J. Pharm. Pharmacol. 59 (2007) 1673e1685. [36] V.D. Kortenska, N.V. Yanishlieva, J. Sci. Food Agric. 68 (1995) 117e126. [37] O.T. Kasaikina, V.D. Kortenska, E.M. Marinova, I.F. Rusina, N.V. Yanishlieva, Russ. Chem. Bull. 46 (1997) 1070e1073. [38] H.Y. Zhang, D.Z. Chen, Chinese Chem. Lett. 11 (2000) 727e730.