Journal of the Chilean Chemical Society
versão On-line ISSN 0717-9707
J. Chil. Chem. Soc. v.53 n.4 Concepción dez. 2008
J. Chil. Chem. Soc, 53, N° 4 (2008) págs: 1658-1662
ANTIOXIDANT ACTIVITYIN SOUTHERN BRAZIL HYPERICUM SPECIES
BERNARDI, A.P.M.,ALÓPEZ-ALARCÓN, C.B, ASPÉE, A, RECH, S.B.A, VON POSER, G.L,A* BRIDI, R.,AD DUTRAFILHO C. S.,DLISSI, EC
aPrograma de Pós Graduação em Ciências Farmacêuticas, UFRGS. Av. Ipiranga, 2752, 90610-000 Porto Alegre, RS, Brazil.
bDepartamento de Farmacia, Facultad de Química, Pontificia Universidad Católica de Chile. Av. Vicuña Mackenna n° 4860, Santiago, Chile.
cFacultad de Química y Biología, Universidad de Santiago de Chile, 307 Casilla 33, Correo 40, Santiago, Chile
dDepartamento de Bioquímica, Instituto de Ciências Básicas da Saúde, UFRGS. Rúa Ramiro Barcelos, 2600, 90035-003 Porto Alegre, RS, Brazil.
The present study was conducted to assess the antioxidant activity of Hypericum species endemic to South Brazil, H. caprifoliatum, H. carinatum, H. myrianthum and H. polyanthemum. The free radical scavenging properties of plant extracts were evaluated employing different methodologies, including the bleaching of a stable free radical (2,2-diphenyl-l-picrylhydrazyl, DPPH) and the peroxyl reactivity indexes TRAP (Total Reactive Antioxidant Potential) and ORAC-pyrogallol red (Oxygen Radicáis Absorbance Capacity). A fair correlation was found between total phenol content determined by Folin-Ciocalteau and DPPH consumption, both in crude methanol and n-hexane extracts. In particular, H. myrianthum and H. caprifoliatum showed the highest TRAP and ORAC-pyrogallol red values, respectively. This would imply that H. myrianthum contains a larger amount of antioxidants of lower reactivity.
Key Words: Hypericum; polyphenols, antioxidant activity; DPPH; TRAP; ORAC
Oxidative stress has been implicated in the pathogenesis of several neurodegenerative disorders, such as Parkinson's and Alzheimer's diseases.1,2 The lack of effective therapeutic agents for the treatment of these disorders has led to an intensive search of therapeutic alternatives based on the control of oxidative stress-induced neuronal damage. There is evidence that abnormalities in lipids may cause overproduction of reactive oxygen species (ROS) and/or decrease of the antioxidant activities. These phenomena may be related to the pathophysiology of major depression.3 In the last years, several studies have been devoted to evalúate the effect of antioxidant supplementation on the prevention of oxidative stress conditions. Inthis context, the identification and development of new compounds, able to prevent and/or diminish ROS-induced damage, is of great interest.
Recent studies have shown that polyphenols and benzophenones, as well as plant extracts rich in these compounds, can exert antioxidant actions.4' 7 Polyphenols, and in particular flavonoids, are usually recognized as the compounds which are responsible for most of the plant extracts antioxidant activity.8"12 The antidepressant activity of Hypericum perforatum (Guttiferae)13 and evidence on the antioxidant activity of Hypericum species in vi tro,14'22 has encouraged us to further survey the chemical composition of several Hypericum species native to southern Brazil. Recently, some species were studied in our laboratory, affording a variety of phenolic compounds, such as benzopyrans,23 benzophenones,24 flavonoids25 and phloroglucinol derivatives.26,27 Some of the species were evaluated for antidepressant,28 antiproliferative,29 antimicrobial,25,30 and analgesic activity.31 Recently, it has been also shown a significant monoamine oxidases MAO-A and MAO-B inhibitory activity for 5-hydroxy-6-isobutyryl-7-methoxy-2,2-dimethyl-benzopyran (HP3), a compound present in Hypericum polyanthemum non polar extracts.32 Furthermore, the antioxidant capacity of several compounds present in Hypericum such as uliginosin B, cariphenone A , cariphenone B, and flavonoids have been evaluated.24
In the present work, we have studied the free radical scavenging properties of four Brazil native Hypericum species. The interaction of the crude methanol extracts with peroxyl radicáis was estimated using both TRAP (Total Reactive Antioxidant Potential) and ORAC (Oxygen Radical Antioxidant Capacity) methodologies. The crude methanol extract and fractions (methanol, dichloromethane and n-hexane) were evaluated in terms of DPPH radical scavenging ability in ethanol. Also, the total phenol content in the samples was estimated by Folin's method.
Aerial parts oí Hypericum caprifoliatum Cham. & Schltdl. were collected in Viamão; H. myrianthum Cham. & Schltdl. and H.polyanthemum Klotzsch ex
Reichard were collected in Paraíso do Sul and Cacapava do Sul, respectively; H. carinatum Griseb. was collected in Glorinha. The vouchers specimens were deposited in the herbarium of the Universidade Federal do Rio Grande do Sul (ICN).
The plant material was dried at room temperature and powdered. The extracts were obtained by static maceration employing methanol at room temperature. The procedure involved three methanol extractions of 24 hrs, using new solvent each time in order to improve the extraction efficiency. Fractions of different polarity were obtained from the dried and powdered plant materials by consecutive extractions in a Soxhlet apparatus, using n-hexane, dichloromethane and methanol, successively. The Soxhlet extractions in each solvent were performed during 12h. After extraction, the solvent was evaporated to dryness under reduced pressure and the initial volume was reconstituted with ethanol or methanol.
Total phenol content in the extracts was determined according to the Folin-Ciocalteau colorimetric method,33 using quercetin as standard.34 Briefly, appropriate dilutions of the samples were oxidized with 0.2 N Folin-Ciocalteau reagent (Merck Darmstadt, Germany - 2N, diluted ten-fold). After 5 min, sodium carbonate (75 g/L) was added. The mixtures were incubated for 30 min and the absorbance of the resulting blue color measured at 765 nm using an ultraviolet-visible Hewlett Packard 8453 spectrophotometer. Results were expressed as milligrams of quercetin equivalents per gram of dry plant (QE) lml or per gram of dry extract (QE)extract.
Determination of DPPH radical scavenging activity
Samples were evaluated in terms of radical scavenging ability by kinetic assays measuring spectrophotometrically the bleaching of the stable free radical DPPH.35 Ethanol solutions of the crude methanol extracts and fractions (25 uL) were poured in 60 uM DPPH solution (final volume of 3 mL). DPPH and extracts solutions were daily prepared. The assays were carried out in an ultraviolet-visible Hewlett Packard 8453 (Palo Alto, CA, USA). The absorbance was measured at 517 nm (eJ17nm = 11500 M^cm"1). Temperature was controlled at 25.0 ± 0.2 °C. Measurements started immediately after mixing the solutions and the absorbance decrease was evaluated up to 600 seconds.
Total Reactive Antioxidant Potential (TRAP) assay
TRAP, total reactive antioxidant potential, was measure by luminol-enhanced chemiluminescence, according to the method of Lissi et al.36 This method is based on that antioxidant addition to a solution containing luminol and a free radical source (2,2'-azo-bis(2-amininopropane)dihydrochloride, AAPH) quenches the solution chemiluminescence for a period that is proportional to the amount of antioxidants inthe sample.36 Chemiluminescence was measured in the out-of-coincidence mode in a liquid scintillation counter (Wallac 1409) as counts per minute (CPM). The sample background was obtained by adding 3 mL of 10 mM AAPH in phosphate buffer 50 mM, pH 7.4, to a glass scintillation vial. Afterwards, 10 uL of luminol (4 mM) was added to each vial and the measured chemiluminescence was considered as the initial valué. Ten |iL of Trolox (6-hydroxy-2,5,8 tetramethylchroman-2-carboxylic acid, 320 or 160 |xM), quercetin (320, 160, 120 or 80 uM in ethanol) or Hypericum dried crude methanolic extracts dissolved in ethanol (1.0; 0.75; 0.5; 0.25 and 0.125 mg/mL) were added and the chemiluminescence was measured as a function of the elapsed time. The time at which the chemiluminescence intensity recovers to 20 % of the initial valué (induction time, IT) was measured.
Oxygen Radicáis Absorbance Capacity (ORAC) assay
Oxygen Radicáis Absorbance Capacity (ORAC) valúes of the samples were determined employing the procedure proposed by Lopez-Alarcón et al.37 This methodology measures the ability of the antioxidants present inthe tested sample to inhibit the bleaching of pyrogallol red (PGR) absorbance elicited by AAPH.
Stock solutions of dried crude methanolic extracts (Hypericum caprifoliatum 2.5 mg/mL; H. carinatum, H. myrianthum and H. polyanthemum 5 mg/mL) and Trolox were prepared in methanol immediately before their use. Stock solutions of AAPH (0.6M) and PGR (lxl0""M) were prepared daily in phosphate buffer 75 mM pH 7.4:methanol (60:40). Areaction mixture, with final volume 3 mL in phosphate buffer:methanol, containing AAPH (10 mM), PGR (5 uM) and the sample (25 uL) was incubated at 37°C in the thermostatized cuvette of a UV-visible spectrophotometer Hewlett Packard 8453 (Palo Alto, CA, USA). The consumption of PGR, associated to its incubation in the presence of AAPH, was evaluated from the progressive absorbance decrease measured at 540 nm. Valúes of (A/A0) were plotted as a function of time. Integration of the área under the curve (AUC) was performed up to a time such that (A/A0) reached avalué of 0.2. ORAC valúes, expressed as micromoles of Trolox equivalents (TE) per gram of dry extract, were calculated employing equation :
where AUC is the área under curve in presence of the extract, integrated between time zero and 80 % of probé consumption; AUC is the área under curve for the control; AUCTrolox , is the área under curve for Trolox, and f is a factor that takes into account the dilution of the extract. It is important to point out that ORAC valúes are independent of the additive concentration considered, because there is a linear relation between the AUC and the amount of the tested sample employed in the assay (data not shown). All experiments were carried out in triplícate.
RESULTS AND DISCUSSION
Determination of total phenols content
The amount of dry extract obtained from Hypericum species depend both on the employed solvent and the species considered. These data (Table 1) show that the amount of compounds in the crude extract are similar in the four species considered (between 27 to 36 grams of dry extract per 100 grams of dry plants), and that only a small fraction of the material is extracted with the non-polar solvents. This indicates that most of the material present in the crude extract corresponds to compounds of relatively high polarity.
Plant phenols constitute one of the major groups of compounds acting as primary antioxidants or free radical scavengers.38,39The Folin-Ciocalteaureagent was employed to evalúate the total amount of phenolic groups in the extracts.33 The total amount of phenolic compounds in the extracts was calculated with a linear equation based on a standard curve obtained employing quercetin:
where y is the absorbance and x is quercetin concentration (uM). The results for the extracts from Hypericum species, expressed as quercetin microequivalents (QE) per gram of dry extract, are given in Table 2.
Plant polyphenols are widely distributed in the plant kingdom and they are frequently present in large amounts.40 From the Hypericum species evaluated, the highest proportion was found in H. carinatum crude extract, where phenols (expressed as quercetin) amounts to ca. 23 % of the weight of the dry extract. This amount to ca. 7.7 g of phenols per 100 grams of dry plant.
A relevant conclusión of the data collected in Table 2 is that the fraction of phenols present in the extracts obtained in non polar solvents is considerable smaller than in the crude extract. The low extraction efficiency of non-polar solvents is further emphasized if the small amount of extracts in these solvents is taken into account (Table 1).
A peculiar feature of the data given in Table 2 is the fact that the fraction of phenols titrated in the crude extract is smaller than that detected in the methanolic fraction in H. caprifoliatum and H. polyanthemum. This could be due to the presence in the crude extract of non-phenolic compounds of low polarity (such as terpenoids) that are not present in the methanolic extraction. This proposal is compatible with the rather small fraction of phenolic compounds present in the non-polar solvents. Furthermore, it is interesting to note that if the results are expressed in amount of phenols per 100 grams of dry plants, always the larger amount is present in the crude extract.
The lowest percentage of phenolic compounds is present in the n-hexane extracts of H. polyanthemum and H. carinatum. This can be related to an important contribution of benzopyrans, such as HP3, whose structures are devoid of phenolic groups or present only one of these moieties).23
The data collected in Table 2 can be compared with that reported for Hypericum perforatum extracts. Zheng and Wang38 have determined the presence of 0.28 g of phenols (taking gallic acid as reference) per 100 grams of fresh plant. Furthermore, in the methanolic extract of this plant, Skerget et al.7 have reported the presence of 0.19 grams of phenols (expressed as gallic acid equivalents) per grams of extract and these valúes are similar to those reported in the present work.
Folin's valúes can be considered as a measure of the total amount of phenolic groups since the response of the method is almost directly proportional to the number of phenolic groups present in the sample. However, it has been reported that different phenols can give different responses.33 In particular, the presence of two vicinal hydroxyl groups can lead to an increase in the response of the method.41 In order to evalúate parameters to the free radical removal capacity of the different extracts we have performed a series of assays aimed to evalúate the antioxidant capacity of the samples.
DPPH radical bleaching.
All extracts, when added to DPPH ethanolic solutions bleached the visible absorbance of the sample, indicating capacity to scavenge free radicáis. The extent of DPPH bleaching elicited per a given amount of sample (i.e. an extract) has been considered as a measure of the amount of free radicáis scavengers present in the sample.42 Typical results are summarized in Table 3.
Comparison of these data with the phenols contents present in the extracts (Table 2), allows to conclude that the amount of DPPH bleached by the extracts correlates with the amount of phenols. In particular, H. carinatum extract presents the largest amount of phenols and the highest capacity to bleach DPPH radicáis. This is evidenced when the amount of DPPH bleaching is plotted againstthe amount of titrated phenols (Figure 1). A fair correlation is observed between both measurements that includes data obtained in n-hexane and crude extracts (r= 0.93, p = 5 x 10"4). Furthermore, the slope of the plot is near two, the expected theoretical valué if it is considered that each quercetin molecule is able to bleach two DPPH radicáis.
In H. carinatum and H. polyanthemum extracts, (Crude/n-hexane)DppH »(Crude/n-hexane)Folin
Total Reactive Antioxidant Potential (TRAP) of the crude methanol extracts were evaluated by a procedure based on the quenching of luminol chemiluminescence.36 Antioxidants can inhibit this chemiluminescence, giving an induction time which is directly proportional to the total antioxidant potential. The effect of the four crude methanol extracts, quercetin and Trolox is shown in Figure 2.
The chemiluminescence CL following the plant extracts or quercetin addition is qualitatively different to that obtained when Trolox is used. In particular, the rise of the CL intensity after the induction time is considerably faster in Trolox assays. This difference can be ascribed to the presence of relatively low reactivity, both in quercetin and the plant extraxts.43,44
The data given in Fig.2 allow an evaluation of the concentration of Trolox equivalents present in the extracts of the four species analyced.
where ti and tTrolox are the induction times measured for the sample and Trolox, respectively, and f is a factor that takes into account the extract dilution in the measuring vial. The data obtained for this procedure are given in Table 4
Comparison of these data with that given in Table 2 for the methanolic extracts allows to conclude that:
i) TRAP valúes are of the same order of magnitude than total phenols detected by Folin's. This would indícate that most phenols present in the extract are able to trap peroxyl radicáis.
ii) There is not correlation between TRAP and Folin's valúes (r = 0.44; p = 0.55). This indicates that differences in stoichiometric factors, included in TRAP valúes, are more important than the differences between the amounts of phenols present in the extracts.
Oxygen Radicáis Absorbance Capacity (ORAC) assay
Methodologies aimed to obtain Oxygen Radicáis Absorbance Capacities (ORAC) index are frequently employed to characterize the radical trapping capacity of puré compounds and their complex mixtures.37 In the present investigation we used pyrogallol red (PGR) as target molecule in the evaluation of the scavenging capabilities of the crude methanolic extracts through an ORAC-like methodology (ORAC-PGR), by following the decrease of PGR absorbance at 540 nm. Fig. 4 shows the protection afforded by H. caprifoliatum to the bleaching of PGR elicited by its incubation in presence of AAPH. The linearity of AUC vs antioxidant concentration, obtained both for Trolox and the tested extracts (data not shown), implies that ORAC valúes are independent of the additive concentration considered. Valúes of ORAC obtained with crude methanolic extracts from Hypericum species employing PGR under the present conditions are included in Table 4. The ORAC valúes vary from 240 to 820 umol of Trolox equivalents (TE)/g. Among the four Hypericum species evaluated, the highest ORAC valué was found in H. caprifoliatum crude extract. This can be, at least partially, associated to the exceeding amounts of phenols present in this extract (Table 2).
The ORAC methodology is influenced both by the reactivity of the tested compound and the number of radicáis that each molecule of the tested compound can remove. The relative importance of these factors depends upon the target molecule employed (phycoerythrin, fluorescein, c-phycocyanin and pyrogallol red), rendering ORAC valúes that are strongly dependent on the employed methodology. The efficiency of the tested compounds using PGR as a target molecule is considerably lower in the protection of PGR than in the protection of fluorescein as a target molecule. The competition for the peroxyl radicáis under these conditions is more difficult and the tested compounds only reduce, in a concentration dependent way, the rate of the target molecule consumption. The use of PGR as a target molecule for peroxyl radicáis provides ORAC indexes that are strongly influenced by the reactivity of the tested compounds.37 Therefore, this approach it is a good choice for an estimation of the average reactivity of the antioxidants present in a complex mixture. In particular, a comparison of TRAP valúes (determined by the amount of antioxidants) and PGR-ORAC valúes (strongly conditioned by reactivity) can be employed to estímate the average reactivity of the antioxidants present in the tested sample. Employing this criterion, the data of Table 4 would indícate that the extract bearingthe more reactive antioxidants is H. caprifoliatum, while H myrianthum and H. polyanthemum extracts comprise a large amount of relatively inefficient compounds.
The difference in the target molecule can explain the differences in ORAC valúes obtained in the present work with that reported in the literature. Zheng and Wang38 have reported, for the aqueous crude extract of H. perforatum and ORAC index of 16.8 uM per gram of fresh plant employing R-phycoerithrin as target molecule. This valué is considerably larger than those obtained in the present work for the crude methanolic extracts, ranging between 1.2 (H. myrianthum) and 6.8 (H. carinatum) umols/g of dry extract.
In summary, this study have demonstrated that the amount of dry extract obtained from H. caprifoliatum, H. carinatum, H. myrianthum and H. polyanthemum, depend both on the employed solvent and the species considered. A fair correlation was found between total phenol content determined by Folin-Ciocalteau and DPPH consumption, both in crude methanol and n-hexane extracts. In particular, H. myrianthum and//. caprifoliatum showed the highest TRAP and ORAC-pyrogallol red valúes, respectively. This would imply that H. myrianthum contains a larger amount of antioxidants of lower reactivity.
This work was supported by CAPES, CNPq, FAPERGS, PROPESQ-UFRGS, LARC-IBRO and FONDECYT (n° 11060323). Also, the support of Vicerrectoría Adjunta de Investigación y Doctorado (VRAID), Pontificia Universidad Católica de Chile (DIPUC n°2006/28) is acknowledged.
1. N. A. Simonian, J.T. Coyle, Annu. Rev. Pharmacol. Toxicol, 36, 83 (1996). [ Links ]
2. Y. Gilgun-Sherki, E. Melamed, D. Offen, Neuropharmacology, 40, 959 (2001). [ Links ]
3. M. Bilici, H. Efe, M.A. Koroglu, H.A. Uydu, M. Bekaroglu, O. Deger, J Affect Disord, 64, 43 (2001). [ Links ]
4. Z. Liu, X. Tao, C. Zhang, Y. Lu, D. Wei, Biomed. Pharmacother., 59, 481 (2005). [ Links ]
5. R. BridiR, F.P. Crossetti, V.M. Steffen, A.T. Henriques, Phytother. Res. 15, 449 (2001). [ Links ]
6. S. Sang, X. Cheng, R.E. Stark, R.T. Rosen, C.S. Yang, C.T. Ho. Bioorg. Med. Chem., 10, 2233 (2002). [ Links ]
7. M. Skerget, P. Kotnik, M. Hadolin, A. Rizner-Hras, M. Simonic, Z. Knez, FoodChem., 89, 191 (2005). [ Links ]
9. T.B. Ng, F. Liu, Z.T. Wang, Ufe Sci., 66, 709 (2000). [ Links ]
10. C. A. Rice-Evans, N.J. Miller, G. Paganga, Free Radie. Biol. Med., 20, 933 (1996). [ Links ]
11. C. A. Rice-Evans, N.J. Miller, G. Paganga, Trenas Plant Sci., 2, 152 (1997). [ Links ]
12. D. Amic, D. Davidovic-Amic, D. Beslo, N. Trinajstic, Croatica Chem. Acta, 76, 55 (2003). [ Links ]
13. A. R. Bilia, S. Gallori, F. Vincieri, Ufe Sci., 70, 3077 (2002). [ Links ]
14. E. J. Hunt, CE. Lester, E.A. Lester, R.L. Tackett, Ufe Sci., 69, 181 (2001). [ Links ]
15. M. Couladis, P. Baziou, E. Verykokidou, A. Loukis, Phytother. Res., 16, 769 (2002). [ Links ]
16. M. Couladis, R.B. Badisa, P. Baziou, S.K. Chaudhuri, E. Pilarinou, E. Verykokidou, C. Harvala, Phytother. Res., 16, 719 (2002). [ Links ]
17. F. Conforti, G.A. Statti, R. Tundis, F. Menichini, P. Houghton, Fitoterapia, 73, 479 (2002). [ Links ]
18. P. Valentao, E. Fernandes, F. Carvalho, P.B. Andrade, R.M. Seabra, M.L. Bastos, Biol. Pharm. Bull., 25, 1320 (2002). [ Links ]
19. P. Valentao, M. Carvalho, E. Fernandes, F. Carvalho, P.B. Andrade, R.M. Seabra, M.L. Bastos, J. Ethnopharmacol, 92, 79 (2004). [ Links ]
20. K. Athanasas, P. Magiatis, N. Fokialakis, A. Skaltsounis, H. Pratsinis, D. Kletsas, J. Nat. Prod., 67, 973 (2004). [ Links ]
21. D. Gamiotea-Turro, O. Cuesta-Rubio, S. Prieto-González, F. De Simone, S. Passi, L. Rastrelli, J. Nat. Prod, 67, 869 (2004). [ Links ]
22. B. A. Silva, F. Ferreres, J.O. Malva, A. Dias, Food Chem., 90, 157 (2005). [ Links ]
23. A. Ferraz, S. Bordignon, C. Staats, J. Schripsema, G.L. von Poser, Phytochemistry, 57, 1227 (2001). [ Links ]
24. A. Bernardi, A. Ferraz, D. Albring, S. Bordignon, J. Schripsema, R. Bridi, C. Dutra Filho, A. Henriques, G. von Poser, J. Nat. Prod., 68, 784 (2005). [ Links ]
25. R. Dalí' Agnol, A. Ferraz, A. Bernardi, D. Albring, C. Ñor, L. Sarmentó, L. Lamb, M. Hass, G. von Poser, E. Schapoval, Phytomedicine, 10, 511 (2003). [ Links ]
26. A. Ferraz, J. Schripsema, A. Pohlmann, G. Von Poser, Biochem. Syst. Ecol, 30, 989 (2002). [ Links ]
27. C. Ñor, D. Albring, A. Ferraz, J. Schripsema, V. Pires, P. Sonnet, D. Guillaume, G. von Poser, Biochem. Syst. Ecol, 32, 517 (2004). [ Links ]
28. R. Daudt, G. von Poser, G. Neves, S. Rates, Phytother. Res. 14, 344 (2000). [ Links ]
29. A. Ferraz, D. Faria, M. Benneti, A. da Rocha, G. Schwartsmann, A. Henriques, G. von Poser, Phytomedicine, 12, 112 (2005). [ Links ]
30. R. Fenner, M. Sortino, S. Rates, R. Dall'Agnol, A. Ferraz, C. Ñor, A. Bernardi, D. Albring, E. Schapoval, G. von Poser, S. Zacchino, Phytomedicine, 12, 236 (2005). [ Links ]
31. A. Viana, A. Heckler, R. Fenner, S. Rates, Braz. J. Med. Biol. Res., 36, 631(2003). [ Links ]
32. C. Gnerre, G. von Poser, A. Ferraz, V. Viana, B. Testa, S. Rates, J. Pharm. Pharmacol, 53, 1273 (2001). [ Links ]
33. V. Singleton, J. Rossi, Am. J. Enol. Vitic, 16, 144 (1965). [ Links ]
34. D. Ivanova, D. Gerova, T. Chervenkov, T. Yankova, J. Ethnopharmacol, 96, 145 (2005). [ Links ]
35. W. Brand-Williams, M. Cuvelier, C. Berset, Lebenson. Wiss. Technol, 28, 25 (1995). [ Links ]
36. E. Lissi, C. Pascual, M. del Castillo, Free Radie. Res. Comm., 17, 299 (1992). [ Links ]
37. C. López-Alarcón, E. Lissi, Free Radie. Res. 40, 979 (2006). [ Links ]
38. W. Zheng, S. Wang, J. Agrie. Food Chem., 49, 5165 (2001). [ Links ]
39. V. Katalinic, M. Milos, T. Kulisic, M. Jukic, Food Chem., 94, 550 (2006). [ Links ]
40. J. Harborne, New naturally occurring plant polyphenols. In: Polyphenolie phenomena. Paris: Scalbert, INRA, 1993. [ Links ]
41. E. Frankel, A. Waterhouse, P.L. Teissedre, J. Agrie. Food Chem., 43, 890 (1995). [ Links ]
42. J. Espín, C. Soler-Rivas, H. Wichers, J. Agrie. Food Chem., 48, 648 (2000). [ Links ]
43. C. Desmarchelier, M. Repetto, J. Coussio, S. Llesuy, G. Ciccia, Int. J. Pharmacog., 35, 288 (1997). [ Links ]
44. C. Desmarchelier, R. Lisboa, J. Coussio, G. Ciccia, J. Ethnopharmacol, 67, 69 (1999). [ Links ]
(Received: May 27, 2007 - Accepted: March 4, 2008)