View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Food Research International 45 (2012) 427–433

Contents lists available at SciVerse ScienceDirect

Food Research International

journal homepage: www.elsevier.com/locate/foodres

Consumption of baru [ alata Vog.], a Brazilian savanna , prevents -induced oxidative stress in rats

Egle Machado de Almeida Siqueira a,⁎, Alinne Martins Ferreira Marin b, Marcela de Sá Barreto da Cunha c, Adriana Medeiros Fustinoni c, Lívia Pimentel de Sant'Ana c, Sandra Fernandes Arruda d

a Biological Sciences Institute, Department of Cell Biology, Campus Universitário Darcy Ribeiro, Universidade de Brasília, Brasília, DF, PO Box 70.910.900, b Health Sciences Graduation Program, Health Sciences Faculty, Campus Universitário Darcy Ribeiro, Universidade de Brasília, Brasília, DF, PO Box 70.910.900, Brazil c Post-Graduation Program in Human Nutrition, Faculty of Health Sciences, Campus Universitário Darcy Ribeiro, Universidade de Brasília, Brasília, DF, PO Box 70.910.900, Brazil d Department of Nutrition, Faculty of Health Sciences, Campus Universitário Darcy Ribeiro, Universidade de Brasília, Brasília, DF, Brazil

article info abstract

Article history: The protective effect of -based foods in human health has been attributed to the presence of bioactive Received 14 July 2011 compounds in all parts of the . A previous study found a high level of minerals, tannins and phytic Accepted 12 November 2011 acid in the baru nut ( Vog.), which is a native of the Brazilian savanna. This study investi- gated the activity (AA) of the aqueous and ethyl acetate extracts of the baru nut and the effect of Keywords: the consumption of this nut on the oxidative status of rats supplemented orally with iron. The AA was eval- Antioxidant potential uated in vitro using the ferric reducing antioxidant power (FRAP), β-carotene/linoleic acid system and free- Malondialdehyde Carbonyl radical scavenging (DPPH) assays. The total polyphenol concentration was determined spectrophotometri- – Brazilian savanna cally using the Folin Ciocalteu reagent. The in vivo study was conducted in male Wistar rats that were fed Baru nut an AIN-93M diet with or without 10% baru nut or 1% phytic acid and supplemented daily with iron or saline by gavages for 17 days. The liver, heart and spleen were collected for the determination of the malondialde- hyde (MDA), carbonyl protein and iron concentrations. The specific activities of catalase, glutathione reduc- tase, glutathione peroxidase and glutathione S-transferase were also determined in these tissues. A T test was used to compare the results among the rats groups and between the different baru nut extracts (pb0.05). The aqueous extract of the baru nut contained a higher level of phenolic compounds and a higher antioxidant ac- tivity, as measured by FRAP and the β-carotene/linoleic system, relative to the EtOAc extract. The iron supple- mentation reduced the body weight gain, increased the levels of iron and MDA in the liver and the spleen and increased the carbonyl levels in all three tissues. Consumption of the baru nut reduced the carbonyl levels in the liver, heart and spleen of the iron-supplemented rats (p=0.002, 0.012 and 0.036, respectively) relative to the heart carbonyl level of rats that were fed the control diet (p=0.000); it also marginally reduced the iron- induced lipid oxidation in the liver (p=0.117) and the spleen (p=0.074). Phytic acid reduced the carbonyl level in the spleen (p=0.020) and marginally reduced the carbonyl level in the liver (p=0.098) of iron- supplemented rats. These results demonstrated that the consumption of the baru nut protects tissues against iron-induced oxidative stress, and the phytic acid from the baru nut may be partially responsible for this pro- tective effect; however, other compounds such as phenols may also be involved. © 2011 Elsevier Ltd. Open access under the Elsevier OA license.

1. Introduction and neurodegenerative disorders (Phung, Makanji, White, & Coleman, 2009). This protective effect of plant-based foods has been Epidemiological trials and experimental studies have shown a attributed to the presence of bioactive compounds that are spread negative association between the consumption of /vegetables across all parts of the plants, which, in turn, constitute the molecular and the incidence of several chronic human diseases such as cardio- defense system of plants in response to abiotic stress or the action of vascular diseases, diabetes, cancer, genetic diseases, and metabolic pathogens (Arruda, Siqueira, & Souza, 2004). The baru nut is a of the Baruzeiro plant (Dipteryx alata Vog.), a of belonging to the Leguminosae family that is native to the , ⁎ Corresponding author at: Laboratório de Bioquímica da Nutrição, Bloco J, 1˚ andar. blooms from November to May and produces fruit from July to Octo- Departamento de Biologia Celular, Instituto de Ciências Biológicas, Universidade de ber (Siqueira et al., 1986). The Cerrado, the second largest biome in Brasília. Campus Universitário Darcy Ribeiro, Asa Norte. 70.910.900, Brasília, DF, PO South America after the Amazon rainforest, is characterized by a typ- Box 70.910.900, Brazil. Tel.: +55 61 3107 3099; fax: +55 61 3273 3676. E-mail addresses: [email protected] (E.M.A. Siqueira), [email protected] ical hot climate, semi-humid and notably seasonal with rainy sum- (S.F. Arruda). mers and dry winters, similar to the savannas (Mendonça et al.,

0963-9969 © 2011 Elsevier Ltd. Open access under the Elsevier OA license. doi:10.1016/j.foodres.2011.11.005 428 E.M.A. Siqueira et al. / Food Research International 45 (2012) 427–433

1998). Despite these extreme climatic conditions, baru nuts are rich 7.5% sodium carbonate solution. After the mixture had been allowed in high-quality (23.9 to 29.9 g/100 g) and in lipids (38.2 to to stand for 30 min at room temperature, the absorbance was measured 41.9 g/100 g) that are predominantly unsaturated fatty acids (ap- at 765 nm using a UV–visible spectrophotometer (Shimadzu, model proximately 81.2%) (Sousa, Fernandes, Alves, Freitas, & Naves, 2011; UV-1800, Japan). The estimation of the phenolic compounds in the ex- Takemoto, Okada, Garbelotti, Tavares, & Aued-Pimentel, 2001). More- tracts was performed in triplicate, and the results were expressed as mg over, the baru nut contains high concentrations of calcium, iron and tannic acid equivalents (TAE)·100/g of peeled and roasted baru nuts. (141.4±7.0, 4.2±0.4 and 4.7±0.3 mg/100 g of fresh nut, re- spectively), as well as phytate and tannins (1073.6±114.9 and 2.5. Free radical-scavenging assay (DPPH) 472.2±12.5 mg/100 g, respectively) (Marin, Arruda, & Siqueira, 2009). The benefits of tannins, phytic acid and polyphenols in The antioxidant capacity was determined according to the Brand- human health as antioxidant compounds have been widely reported Williams, Cuvelier, and Berset (1995) and modified Sánchez-Moreno (Paiva-Martins, Rodrigues, Calheiros, & Marques, 2011; Brisdelli, et al. (1998) methods, which are based on the quantification of free D'Andrea, & Bozzi, 2009; Faure, Lissi, Torres, & Videla, 1990; radical scavenging. A methanol solution containing 2.5 mg DPPH•/ Benavente-García, Castillo, Marin, Ortuno, & Del Río, 1997; Foti, Piat- 100 mL was prepared. An aliquot of 100 μL of each extract of baru telli, Baratta, & Ruberto, 1996; Sakurai et al., 2010; Tempfer et al., nut (five different concentrations ranging from 300 to 3000 ppm 2007; Van Horn et al., 2008). Considering the wide distribution of were used) was added to 2.9 mL of the DPPH• solution. The decrease baruzeiro in the Cerrado region (Siqueira et al., 1986), the study of in the absorbance at 515 nm of each sample was measured in a Spec- the nutritional and bioactive potential of baru nuts becomes important tronic Genesys 2 instrument (Milton Roy) at 30 s intervals until the for the socioeconomic development and sustainability of this biome DPPH absorbance was stabilized. The antioxidant capacity was and the regions that have similar climatic and soil conditions, such as expressed in μmol of trolox/g of peeled and roasted baru nuts. The as- the savannas of Africa. Thus, the purpose of this study was to evaluate says were conducted in triplicate, and the results were expressed as the effect of the baru nut on the oxidative stress induced by oral iron sup- the mean values±standard deviation (SD). plementation in rats, with the objective of contributing additional knowledge about the nutritional and economic potential of this . 2.6. Ferric reducing antioxidant power (FRAP) assay

2. Material and methods The antioxidant capacity of each extract was estimated via a FRAP assay according to the Benzie and Strain (1999) method with modifi- 2.1. Chemicals cations. Briefly, the FRAP reagent was freshly prepared by mixing so- lutions of 0.3 mol/L acetate buffer (pH 3.6), 10 mmol/L 2,4,6-triazine- 2,6-Dichloroindophenol (DFI), 2,2-diphenyl-1-picryl-hydrazyl tripyridyl in 40 mmol/L hydrochloric acid, and an aqueous solution of (DPPH), 6-hydroxy-2,5,7,8-tetramethyl chroman-2-carboxylic acid 20 mmol/L ferric chloride in a 10:1:1 ratio and then incubating the (trolox), 2,4,6-tris(2-pyridyl)-s-triazine (TPTZ), ferrous sulfate and mixture at 37 °C for 30 min. Subsequently, 900 μL of FRAP reagent the Folin–Ciocalteu reagent were purchased from Sigma Chemical was mixed with 30 μL of nut extract and 90 μL of deionized water. A Co., São Paulo, Brazil. β-Carotene was purchased from Merck, Rio de tube containing the FRAP reagent was used as a blank solution, and Janeiro, Brazil. All solvents/chemicals used were of analytical or the absorbance at 595 nm was measured after 4 min of reaction. high-performance liquid chromatographic grade and were obtained Aqueous solutions of known Fe(II) concentrations in the range of from Sigma Chemical Co., São Paulo, Brazil, or Merck, Rio de Janeiro, 100–2000 μmol/L (FeSO4) were used to generate a calibration curve. Brazil. Deionized and distilled water was used in the experiments. The total extracts were assayed in triplicate, and the mean values were expressed in μmol Fe/g of peeled and roasted baru nut. The re- 2.2. Samples sults were expressed as the mean values±standard deviation.

The baru nuts (Dipteryx alata Vog.) that were obtained in trade 2.7. β-Carotene bleaching method from Pirenópolis-GO, Brazil, were washed with distilled water, dried at room temperature and roasted in a conventional oven at 150 °C The antioxidant capacity of each extract was estimated spectropho- for 40 min. After manually removing the peels, the nuts were ground tometrically based on the β-carotene discoloring induced by the oxida- and added to the rats' diets. The roasting process was used to inacti- tive degradation of linoleic acid described by Jayaprakasha, Singh, and vate the trypsin inhibitor present in the baru nut (DiPietro & Liener, Sakariah (2001) and modified by Singh et al. (2002). All of the determi- 1989; Kalume, Sousa, & Morhy, 1995). nations were conducted in triplicate. The decrease in the absorbance of each extract was measured at 15 min intervals until 180 min. The anti- 2.3. Extraction oxidant activity (AA) of the extracts was estimated by the formula

AA=100[1−(A0 −At)/(A°0 −A°t)], where A0 and A°0 are the absor- The antioxidant compounds were extracted from the ground bance values at 470 nm measured at zero time of the incubation for roasted nuts with 100% ethyl acetate and water (0.25 g/mL) at 30 °C the test sample and control, respectively, and At and A°t are the absor- for 1 h in a shaker at 125 rpm. The extracts were filtered through bance values measured for the test sample and control, respectively, JP41 filter paper for the removal of particles, and the residue was after incubation for 180 min, according to Singh et al. (2002). The re- re-extracted with the same solvent. The extracts were pooled, con- sults were expressed as the mean values±standard deviation (SD). centrated under vacuum at 40 °C and stored at −70 °C (Singh, Murthy, & Jayaprakasha, 2002). This entire procedure was conducted 2.8. Animals and diets in the dark. Thirty adult male Wistar rats, obtained from the Animal Facility at 2.4. Determination of total phenols the University of Brasilia, Brazil, with an average body weight of 278.9±20 g, were divided into five groups (six rats per group) and The total polyphenols in the ethyl acetate and aqueous extracts were housed individually in stainless steel cages at a temperature of 23± determined according to the method of Singh et al. (2002). The baru ex- 2 °C under a 12-h light/dark cycle. The rats had free access to water tracts were dissolved in methanol. The samples (0.2 mL) were mixed and food during the dark cycle. After three days of acclimatization, with 1.0 mL of 10-fold-diluted Folin–Ciocalteu reagent and 0.8 mL of a the rats were treated with one of the following diets: the control E.M.A. Siqueira et al. / Food Research International 45 (2012) 427–433 429 group received an AIN-93G rodent diet (Reeves, Nielsen, & Fahey, 2.11. Tissue protein oxidation 1993); the Ba group received the AIN-93G diet containing ground, roasted and peeled baru nuts (100 g/kg diet); and the iron- The oxidation of the tissue proteins was measured as the carbonyl supplemented groups (rats that received 10.5 mg of iron daily by ga- content of the liver, heart and spleen homogenates as described by vages in the form of FeSO4) were fed an AIN-93G diet (the Fe group), Richert, Wehr, Stadtman, and Levine (2002). The absorbance of the an AIN-93G diet containing baru nut (the Ba/Fe group) or an AIN-93G samples was measured at 398 nm, and the carbonyl content was diet containing phytic acid, inositol hexaphosphate (InsP6, 1 g/kg expressed as nmol carbonyl groups/mg total protein using an extinc- diet) (the PA/Fe group). The control and Ba groups received a saline tion coefficient of 22,000 mM−1 cm− 1. The total protein concentra- solution by gavages instead of iron sulfate. The AIN-93G composition tion of the homogenates was determined by the method of Hartree was modified to adjusted the baru diet compounds to the rats' re- (1972). quirements according to Table 1. The animals were weighed weekly, and the food intake was recorded daily. After 17 days of treatment, 2.12. Tissue lipid oxidation the animals were killed by cervical dislocation, and the liver, spleen and heart were excised, washed in a cold 0.9% NaCl solution, dried The oxidation of the tissue lipids was measured as the malondial- with paper towels to remove excess saline, and rapidly frozen in liq- dehyde (MDA) content of the liver, heart and spleen homogenates in uid nitrogen and stored at −70 °C for further analysis. The animal a high-performance liquid chromatographic (HPLC) system as de- care and use committee of the University of Brasília approved the ex- scribed by Candan and Tuzmen (2008). Briefly, 1 mL of 1% sulfuric perimental protocol. acid (H2SO4) was added to 0.1 g of tissue and homogenized in an Ultra-Turrax homogenizer (Ultra-Turrax T8, IKA-Werke, Staufen, Ale- manha). The homogenate was centrifuged at 12,000×g for 15 min at 2.9. Iron tissues determination 4 °C. An aliquot of 500 μL of the supernatant was mixed with 750 μLof

440 mmol/L phosphoric acid (H3PO4) and 250 μL of 42 mmol/L 2- The iron concentration was determined using the method de- thiobarbituric acid (TBA) and heated at 100 °C for 1 h. After cooling, scribed by Baranowska, Czernicki, and Aleksandrowicz (1995) and 500 μL of this mixture was mixed with 500 μL of a 91:9 mixture of modified by us. Briefly, 0.15 g of spleen and heart and 0.30 g of liver MeOH and 1 mol/L NaOH. The resulting precipitation solution was samples were digested with 5 mL concentrated HNO3 and 2.5 mL centrifuged at 12,000×g for 5 min at 4 °C, and the supernatant was H2SO4, in a Provecto Analítica Microwave System (DGT 100 Plus). filtered using a 47 mm, 0.45 μm, Tuffryn membrane filter (Gelman After the digestion, the samples were resuspended in 0.1 mol/L nitric Sciences). Aliquots of 20–30 μL of the tissues samples were injected acid to a final volume of 25 mL. The concentration of iron in the sam- onto a high-performance liquid chromatography system (Shimadzu ples was determined by inductively coupled plasma atomic emission VP system) using a Shim-pack CLC-ODS (M) column spectrometry (ICP-AES/Spectro, Kleve, Germany). An iron six-point (4.6 mmID×15 cm, 5 μm particle diameter 100 Å pore diameter). standard curve was generated using standard minerals solutions The samples were eluted with a mobile phase consisting of a 60:40

(Titrisol from Merck) in a concentration range of 0–10 ppm (Titrisol (v/v) mixture methanol and 50 mmol/L KH2PO4 (pH 6.8) and a flow from Merck, Germany; y=28552x+3396.6, r2 =0.9993). rate of 0.6 mL/min. The chromatograms were acquired at 532 nm (ex- citation) and 553 nm (emission) using a spectrofluorometric detector (RF-551, Shimadzu). The quantification was performed using a stan- 2.10. Hemoglobin concentration (Hb) dard curve prepared by diluting a 4.04 mmol/mL 1,1,3,3-tetraethoxy-

propane standard solution (TEP; C11H24O4, Sigma St. Louis, MO, USA) The basal hemoglobin concentration (Hb) was measured in a tail in 1% sulfuric acid. The acid hydrolysis of TEP yielded MDA. After dilu- vein blood sample using a HemoCue spectrophotometer. After the tion, the standard samples were treated as described for the tissue treatment, Hb was determined using a cell counter (Coulter T 890) samples, and a four-points standard curve was generated in the con- in blood samples containing anticoagulant that were collected by car- centration range of 0.27 to 2.69 nmol/mL (y=2.63×10−7x+0.0185, diac puncture immediately after sacrificing the animals. r2 =0.999). The MDA concentrations in the tissues were expressed as nmol of MDA/mg of tissue protein. The total protein concentration of the homogenate was determined by the method of Hartree (1972). Table 1 Diet compositions. 2.13. Tissue homogenates for enzyme assays

Components Diet 1 (g/kg) The liver, heart and spleen tissues were extracted 1:20 (w/v) in an Control, Fe and PA/Fe Ba and Ba/Fe Ultraturrax homogenizer using 0.5 mol/L phosphate buffer Cornstarch 397.5 381.7 (pH 7.2) containing 50 mmol/L EDTA and 1 mmol/L phenylmethylsul- – Starch (baru) 15.8 fonyl fluoride (PMSF) at 4 °C and centrifuged at 15,000×g for 20 min. Protein (casein) 200 176.1 Protein (baru) – 23,9 Dextrinized starch 132 132 2.14. Catalase (EC 1.11.1.6) assay Sucrose 100 100 Soybean oil 70 31.8 The activity of catalase was quantified by the consumption of Baru – 38.2 10 mmol/L H O , which was measured at 240 nm in buffer containing Fiber 50 36.6 2 2 – μ Fiber (baru) – 13.4 10 50 L of the tissue homogenates (Aebi, 1984). One unit of catalase mix 10 10 was defined as the amount of enzyme required to decompose 1 μmol mix 35 35 H2O2/min. L-cystine 3 3 Choline bitartrate 2.5 2.5 2.15. Glutathione reductase (GR, EC 1.6.4.2) assay 1Control: AIN-93G Rodent diet; Group Ba: AIN-93G diet containing baru nut ground, roasted and peeled (100 g/kg diet); iron-supplemented groups (rats that received The GR activity was determined by detecting the consumption of 10.5 mg of iron daily by gavages in the form of FeSO4): AIN-93G diet (Group +Fe); or the AIN-93G diet containing baru nut (Group Ba/Fe); or AIN-93G diet containing NADPH at 340 nm (Joanisse & Storey, 1996). The reaction assay con- phytic acid. sisted of 0.5 mmol/L EDTA, 1 mmol/L GSSG and 20–100 μL of the tissue 430 E.M.A. Siqueira et al. / Food Research International 45 (2012) 427–433

Table 2 Brazil Ltda. Significance was defined as pb0.05, and the variables Total phenol concentration and antioxidant activities (AA) of aqueous and ethyl acetate are presented as the mean±SD. extracts of roasted baru nut estimated by the free-radical scavenging (DPPH), ferric re- ducing antioxidant power (FRAP) and inhibition of β-carotene/linoleate oxidation assays.a 3. Results

Extract Antioxidant activities (AA)a The total phenol concentration and antioxidant activity (AA) of β Total phenolics DPPH FRAP -Carotene/linoleate the aqueous and EtOAc extracts of peeled and roasted baru nuts, esti- (mg TAE/100 g) (μmol TE/g) (FeSO4 μmol/g) inhibition (%/g) mated by free radical scavenging (DPPH), ferric reducing antioxidant β ⁎ ⁎ ⁎ power (FRAP) and inhibition of -carotene/linoleate oxidation assays Aqueous 154.6±5.5 0.8±0.1 8.3±0.2 6.0±0.5 are presented in Table 2. The aqueous extract of the baru nut exhib- Ethyl 3.1±1.6 0.6±0.1 1.2±0.2 0.6±0.1 fi fi acetate ited a signi cantly higher level of phenolic compounds and a signi - cantly higher antioxidant activity measured by FRAP and the β- Tannic acid equivalents, TAE; Trolox equivalent, (TE). a The data are expressed as the means of triplicate experiments on a dry weight carotene/linoleic system relative to the EtOAc extract. There was no basis. T test for independent samples. significant difference in the ability of the DPPH radical scavenging be- ⁎ pb0.05, significant differences in columns. tween the two extracts. The Pearson test indicated a high and signif- icant correlation between the total phenol concentration and the FRAP value (r=0.999; p=0.000) and the β-carotene/linoleic value homogenate in 50 mmol/L potassium phosphate buffer (pH 7.2). The (r=0.99; p=0.000). A correlation was also observed between the blanks were measured in the absence of GSSG. One unit (U) of glutathi- total phenol concentration and the DPPH value; however, it was one reductase was defined as the amount of enzyme required to oxidize only marginally significant (r=0.864; p=0.056). 1 nmol NADPH/min. The in vivo evaluation of the capacity of toasted baru nuts to pro- tect tissues against oxidative stress was conducted in rats fed a diet −1 2.16. Glutathione peroxidase (GPx, EC 1.11.1.9) assay with peeled and toasted baru nuts (10 g kg ), in the presence or ab- sence of an oral iron supplement introduced to the rats by gavages as an oxidant agent. Table 3 presents the food and iron intake, the body The GPx activity was quantified using H2O2 as a substrate in an assay coupled with the GR-catalyzed oxidation of NADPH, which was mea- weight gain and the variation in the hemoglobin concentration of the fi sured at 340 nm. First, the basal consumption of 0.15 mmol/L NADPH rats at the end of 17 days. There were no signi cant differences in the fi was measured in buffer containing 2 mmol/L azide, 5 mmol/L GSH, food intake; however, the Fe and Ba/Fe groups experienced a signi - 1.5 U glutathione reductase, and 10–50 μL of the tissue homogenates. cant reduction in their weight gain relative to that of the control group (p=0.027 and 0.009, respectively), whereas the PA/Fe group Then, 20 μLofH2O2 was added to a final concentration of 0.2 mmol/L. fi The blanks were measured in the absence of the tissue homogenate experienced a non-signi cant reduction (p=0.130). fi (Hermes-Lima & Storey, 1996). One unit (U) of glutathione peroxidase The oral iron supplementation resulted in a signi cant increase in was defined as the amount of the enzyme required to oxidize 1 nmol the liver and spleen iron levels but not in the iron levels of the heart NADPH/min. relative to the levels observed in the control group. The rats fed with the baru diet did not show significant differences in the levels of iron in tissues relative to the control group. The phytic acid and baru sup- 2.17. Glutathione-S-transferase (GST, EC 2.5.1.18) assay plementation did not result in a significant reduction of the iron level in the liver, heart or spleen (Fig. 1). The GST activity was measured by following the conjugation of The rats from the Fe group showed increased MDA levels in the 1 mmol/L GSH with 1 mmol/L 1-chloro-2,4-dinitrobenzene (CDNB) liver and spleen (p=0.004 and 0.028, respectively) relative to the at 340 nm in 50 mmol/L potassium phosphate buffer (pH 7.2) con- control group. The liver MDA levels observed in the Ba/Fe and PA/Fe taining 10–50 μL of the tissue homogenate (Hagib & Jakoby, 1981). groups were not different when compared to the Fe group; however, One unit (U) of glutathione-S-transferase was defined as the amount these values were similar to the values observed in the control and Ba of enzyme required to yield 1 nmol of conjugate/min. groups. The Ba/Fe group showed a marginal reduction of the spleen MDA level relative to that of the Fe group (p=0.074), and this 2.18. Statistical analysis value did not differ from that of the control group. The level of spleen MDA observed in the PA/Fe group was similar to that of the Fe group Comparisons of the mean values of data from the rat groups stud- but not different from that of the control group. There was no signif- ied in vivo and the mean values of the data from the two extracts of icant difference in the heart MDA levels among the rats from all baru nut studied in vitro were analyzed by the T test for independent groups (Fig. 2). samples. The correlation between the values of the total phenols and In contrast to the results observed for the lipid oxidative damage the antioxidant capacity in vitro was analyzed by the Pearson test. The to tissues, the effect of the treatments on the protein oxidative dam- analysis was performed using the SPSS Statistics 17.0 program, SPSS age, measured as the levels of carbonyl in tissues, was also observed

Table 3 Food and iron intake, iron by gavages, body weight gain and initial and final hemoglobin concentration1 of rats fed with the Control, Ba, Fe, Ba/Fe or PA/Fe diets, for 17 days.

Control Ba Fe Ba/Fe PA/Fe

Food intake (g) 450.6±24.9 478.3±59.3 415.2±61.9 410.0±39.3 431.3±27.5 Food iron intake (mg) 15.8±0.9 16.8±2.4 14.6±2.2 14.4±1.4 15.1±1.0 Gavages iron (mg) ––168.0 168.0 168.0 Body weight gain (g) 88.3±10. 93.7±2.4 60.9±23.2* 66.3±18.1* 75.4±18.0 ΔHb (g/dL)2 0.3±1.1 −1.6±1.9 0.5±3.1 −1.4±2.0 −0.7±1.2 ⁎ T test for independent samples. pb0.05 differences relative to the control group. 1 Mean±SD, n=6. 2 ΔHb=hemoglobin (Final−Initial). E.M.A. Siqueira et al. / Food Research International 45 (2012) 427–433 431

Fig. 1. Iron concentration in the liver, heart and spleen of rats fed with the Control, Ba, Fe, Fig. 3. Carbonyl concentration in the liver, spleen and heart of rats fed with the Control, Ba, Ba/Fe or PA/Fe diets for 17 days. Control: AIN-93G Rodent diet; Ba: AIN-93G diet+100 g Fe, Ba/Fe or PA/Fe diets for 17 days. Control: AIN-93G Rodent diet; Ba: AIN-93G diet+100 g of baru/kg diet; Fe: control diet plus 10.5 mg of oral iron supplementation daily; Ba/Fe: of baru/kg diet; Fe: control diet plus 10.5 mg of oral iron supplementation daily; Ba/Fe: Ba Ba diet plus oral iron supplementation; PA/Fe: AIN-93G diet plus 1 g phytic acid/kg diet diet plus oral iron supplementation; PA/Fe: AIN-93G diet plus 1 g phytic acid/kg diet and and oral iron supplementation. The data are expressed as the mean±SD (n=6). T test oral iron supplementation. The data are expressed as the mean±SD (n=6). *pb0.05; b b for independent samples. **p 0.01 and ***p 0.001 versus the control group. **pb0.01 and ***pb0.001 versus control group, T test for independent samples. &pb0.05 and &&pb0.01 versus Fe group, T test for independent samples. in the hearts of the rats. The rats fed with the baru diet (Ba group) exhibited a significant reduction in the heart carbonyl level relative to the control group (p=0.000), and the iron-supplemented rats 4. Discussion (Fe group) exhibited higher carbonyl levels than the control group in the liver, heart and spleen (p=0.025; 0.011 and 0.010, respective- The crucial role of iron in the maintenance of biological processes ly). The rats from the Ba/Fe group exhibited a lower carbonyl level in such as cellular respiration and DNA synthesis in addition to its direct the liver, heart and spleen than the rats from the Fe group (p=0.002; involvement in the endogenous production of the hydroxyl radical, a 0.012 and 0,036, respectively). The level of carbonyls in the spleen potential initiator of oxidative stress and lipid peroxidation in tissue in- was also lower in the PA/Fe group than in the Fe group (p=0.020). jury (Hsu, Chien, Chen, & Liu, 2007; Anderson, Frazer, & McLaren, 2009; The carbonyl level observed in the livers of rats from the PA/Fe Pacher, Nivorozhkin, & Szabo, 2006; Beckman, Beckman, Chen, group did not differ from that of rats in the control group, but it Marshall, & Freeman, 1990), drove biological systems to develop an ef- was marginally lower than that of the rats in the Fe group ficient molecular mechanism for the regulation of iron levels in the (p=0.098) (Fig. 3). body (Matsuura, 1983). This mechanism is capable of providing the physiological requirements of iron while avoiding the catalysis of the molecular oxidation reactions mediated by free iron. Despite this effi- cient mechanism for regulating endogenous levels of iron, acute iron toxicity has been widely reported and primarily associated with liver and kidney damage (Ebina et al., 1986; Takemoto et al., 2001; Bonvicino, Lemos, & Weksler, 2005); however, studies of the effect of the prescribed dose of oral iron supplementation, currently used in the treatment of anemia, on the oxidative damage of tissues is scarce. In the present study, an iron dose, proportional to that prescribed for human supplementation, was administrated to rats by gavages over the course of seventeen days as an oxidative stress promoter to investi- gate the potential antioxidant effect of the dietary baru nut. In addition to its nutritional values as a source of iron, zinc, calci- um (Marin et al., 2009), proteins and unsaturated fatty acids (Sousa et al., 2011; Caramori, Lima, & Fernandes, 2004), the toasted baru nut has an attractive taste. In addition, the presence of bioactive com- pounds such as , phytic acid and tannins (Marin et al., 2009; Caramori et al., 2004) suggest a high antioxidant potential for this nut. The aim of the present study was to investigate the antioxi- dant potential of the baru nut in vitro and in vivo. The reduction of the body weight gain observed in rats from the iron-supplemented groups (Fe and Ba/Fe) relative to the control group suggested a negative effect of the iron overload on the rats' de- Fig. 2. MDA concentration in the liver, spleen and heart of rats fed with the Control, Ba, velopment. The phytate added to the AIN-93 diet of the rats also sup- Fe or Ba/Fe diets for 17 days. Control: AIN-93G Rodent diet; Ba: AIN-93G diet+100 g of plemented with iron (PA/Fe group) seemed to minimize this negative baru/kg diet; Fe: control diet plus 10.5 mg of oral iron supplementation daily; Ba/Fe: Ba diet plus oral iron supplementation; PA/Fe: AIN-93G diet plus 1 g phytic acid/kg diet effect because this group experienced a weight gain similar to that of and oral iron supplementation. The data are expressed as the mean±SD (n=6). the control group. The iron overload found in the liver and spleen of *pb0.05 versus control group, T test for independent samples. the iron-supplemented rats, the organs that are responsible for the 432 E.M.A. Siqueira et al. / Food Research International 45 (2012) 427–433 iron storage and the iron recycling from senescent red cells, respec- the group that received iron supplementation although the level of tively, is in agreement with the fine control of body iron homeostasis iron and MDA in the heart did not change. These results suggest that is mediated by hepcidin, a hormone up-regulated by iron over- that heart muscle cells have a stringent mechanism for regulating load that induces ferroportin internalization and degradation, de- iron uptake, which protects the heart from lipid oxidation, mediated creasing the iron mobilization that results in iron accumulation in by iron overload; however, this mechanism cannot protect the heart tissues (Matsuura, 1983; Ganz, 2008). against protein oxidation, which may be mediated by other processes In addition to the iron accumulation, the oral iron supplementation resulting from changes in systemic iron homeostasis in the body. increased the oxidation of lipids in the liver and spleen and, despite the Similar to the results obtained in the present study, Legssyer, fact that the Ba/Fe and PA/Fe groups did not show a significant reduc- McArdle, Crichton, and War (2003) investigated the effect of four dif- tion of this oxidative damage relative to the Fe group, the lipid oxidation ferent injectable iron complexes on the iron loading of tissues and ob- levels in the liver and spleen of these two groups was similar to the served that only the iron dextran and the polymaltose, two high value observed in the control and Ba groups. These results suggest molecular weight complexes, caused a slight effect (2-fold relative that the iron overload in tissues induces the oxidative lipid damage to the control) in the iron loading of the heart, whereas iron seques- and that the baru nut at a dose of 100 g kg−1 in a rodent's diet may pro- tering tissues, such as spleen and liver, exhibited 10-fold and 20-fold tect the liver and spleen against oxidative lipid damage induced by iron higher iron content, respectively, relative to the control. supplementation. The similar responses obtained for the MDA levels The dietary iron supplementation should first increase the labile and iron accumulation in the liver and spleen of Ba/Fe and PA/Fe groups iron pool in the plasma, which, in turn, may increase the levels of re- suggest that the antioxidant potential of the baru nut against lipid per- active oxygen species (ROS), such as the hydroxyl radical (•OH), via a oxidation could be partly related to the phytic acid chelating property. Haber-Weiss reaction. This serum ROS could attack tissues, resulting The absence of a significant change in the iron and MDA levels in the in the enhancement of carbonyl proteins; however, the presence of hearts of the rats from all groups reinforces the suggestion that the tis- a free radical scavenger in the plasma may protect tissues against sue iron content was directly involved in the lipid damage. the ROS attack. Some authors suggest that proteins constitute the ini- The baru nut also seems to protect tissues against protein damage; tial targets of oxidation, whereas lipids and DNA are secondary dam- however, the results suggest that, in the heart, this protective mech- age targets for ROS (Wijeratne, Abou-Za, & Shahidi, 2006); therefore, anism is iron independent. Fig. 3 demonstrates that the dietary baru the short treatment time associated with the fact that the heart is not nut reduced the carbonyl level in the heart, although the levels of an organ directly involved in iron metabolism may explain the higher iron in this tissue did not vary; additionally, the iron-supplemented level of protein damage but not of lipid damage observed in the rats fed a diet containing phytate exhibited no reduction in the car- hearts of iron-supplemented rats relative to the control rats. bonyl levels in the heart. These results suggest that the baru nut The protective effect of the baru nut against the oxidative stress in may also contain bioactive compounds that act as free radical scaven- tissues induced by oral iron supplementation in rats that was ob- gers instead of iron chelators because phytic acid, an excellent iron served in this study suggests the presence of antioxidant compounds chelator, did not inhibit protein oxidation. in the composition of this nut. In a previous study, we found a high The low level of carbonyl protein found in the liver and spleen of the level of phytate and tannins (1073.6±114.9 and 472.2±12.5 mg/ iron-supplemented rats fed a diet containing baru nuts relative to that 100 g, respectively) in the baru nut compared with the levels found of the iron-supplemented rats fed with a control diet strengthens the in seventeen other fruits native to Cerrado (Marin et al., 2009), and antioxidant potential of the baru nut against protein oxidation; howev- in the present study, we also found higher total levels of phenolic er, in these tissues, phytic acid also exerted a protective role against the compounds in the peeled and toasted baru nut than in others nuts, oxidative damage of proteins. In an early study conducted by Grases and such as Brazil nuts, , , pine nuts, and macadamia collaborators to determine the phytate levels (InsP6) in the tissues of nuts (Sun, Chu, Wu, & Liu, 2002). The antioxidant effect of phenolic rats fed diets with different contents of phytate, it was observed that compounds as free radical scavengers and of phytate as a bivalent the majority of the InsP6 found in the organs and tissues was of a dietary metal chelator has been widely reported (Gary & Dao, 2003; origin, and the InsP6 exhibited a different distribution among the vari- Wijeratne et al., 2006). ous tissues studied. The authors found 10-fold more InsP6 in the brain These findings together suggest that phytic acid and phenolic than in the liver, kidney and bone. Thus, considering the InsP6 chelator compounds may be responsible for the baru nut's antioxidant activity properties, InsP6 could reduce the oxidative stress induced by iron with that was observed in this study. The absence of changes in the specific a different intensity in each tissue depending on the concentration of activities of catalase, GR, GPx and GST (Table 4) in the rats treated inositol in the tissues. with baru nut or phytate compared to the control and iron- In addition, similar to the results found in the previous study supplemented rats preclude the involvement of the baru nut com- (Arruda, Siqueira, & Valência, 2009), in the present study, the mech- pounds in the regulation of these enzymes as a mechanism of the an- anisms of oxidation of lipids and proteins seem to be different be- tioxidant action against the oxidative stress induced by oral iron cause there was an increase in the heart protein oxidation levels in during these 17 days of treatment.

Table 4 Specific activities of catalase, GR, GPx and GST in the liver, spleen and heart of rats fed with the Control, Ba, Fe, Ba/Fe or PA/Fe diets for 17 days.

Catalase Glutathione reductase Glutathione peroxidase Glutathione S-transferase

Specific activitya

Liver Heart Spleen Liver Heart Spleen Liver Heart Spleen Liver Heart Spleen

Control 69.2±20.5 23.2±8.6 30.9±4.3 23.6±5.6 6.0±1.4 17.1±4.3 983.5 ±273.6 412.3 ±108.3 515.9±93.7 340.9±56.7 12.9±6.3 17.2±3.3 Ba 67.3±20.8 16.2±5.7 33.6±3.3 23.9±6.0 6.2±0.6 16.7±2.0 799.7 ±147.1 444.7 ±122.5 517.4±42.8 325.8±56.8 14.6±5.0 13.8±4.0 Fe 61.7±4.4 20.3±5.3 30.1±4.4 20.8±5.6 5.5±1.8 17.7±2.9 837.7 ±173.5 436.2 ±72.4 500.8±42.0 298.7±40.1 13.3±2.3 16.2±4.5 Ba/Fe 67.7±10.3 16.3±1.3 33.3±4.3 24.0±6.8 6.4±1.7 16.7±1.4 622.7 ±288.3 491.2 ±45.8 532.0±36.3 298.0±95.5 9.1±4.0 15.5±0.5 PA/Fe 67.7±14.9 16.6±4.2 33.9±5.2 20.0±12.6 6.0±0.4 17.6±3.6 817.3 ±177.1 423.4 ±58.5 557.6±56.0 322.9±71.5 12.9±4.0 15.6±3.1

The data are expressed as the mean±SD (n=6). a Specific activity of enzymes was defined as follow: Catalase: 1 U=1 μmol H2O2/min; Glutathione reductase and glutathione peroxidase: 1 U=1 nmol NAPDH/min; Glutathione S-transferase: 1 U=1 nmol conjugate/min. E.M.A. Siqueira et al. / Food Research International 45 (2012) 427–433 433

5. Conclusions Hagib, W., & Jakoby, W. B. (1981). Glutathione S-transferases (rat and human). Methods in Enzymology, 27, 218–231. Hartree, E. (1972). Determination of protein: A modification of the Lowry method that This is the first time that a protective effect of the consumption of gives a linear photometric response. Analitical Biochemistry, 48, 422–427. a nut from Cerrado was observed in vivo to combat the oxidative Hermes-Lima, M., & Storey, K. B. (1996). Relationship between anoxia exposure and antioxidant status of the frog Rana pipens. American Journal of Physiology, 271, stress induced by oral iron supplementation in rats. The high levels R545–R553. of phenols and phytic acid in the baru nut composition may be re- Hsu, D. Z., Chien, S. P., Chen, K. T., & Liu, M. Y. (2007). The effect of sesamol on systemic sponsible by the antioxidant effect observed in this study. Studies oxidative stress and hepatic dysfunction in acutely iron-intoxicated mice. Shock, – aimed at isolating and identifying the compounds with antioxidant 28, 596 601. Jayaprakasha, G. K., Singh, R. P., & Sakariah, K. K. (2001). Antioxidant activity of grape activity that are present in the baru nut are now underway in our seed (Vitis Vinefera) extracts on peroxidation models in vitro. Food Chemistry, 73, laboratory. 285–290. Joanisse, D. R., & Storey, K. B. (1996). Oxidative damage and in Rana sylvatica, the freeze-tolerant wood frog. American Journal of Physiology-Heart and Circulatory Acknowledgments Physiology, 271, R545–R553. Legssyer, R., McArdle, H., Crichton, R., & War, R. (2003). Comparison of injectable iron complexes in their ability to iron load tissues and to induce oxidatative stress. Bio- We thank Gustavo Antonio Teixeira Chaves for his assistance in the Metals, 16, 425–433. treatment of the rats. We acknowledge Fundação de Amparo à Pesquisa Kalume, D. E., Sousa, M. V., & Morhy, L. (1995). Purification, characterization, sequence do distrito Federal (FAPDF) for the financial support and Conselho determination, and mass spectrometric analysis of a trypsin inhibitor from seeds of the brazilian tree Dipteryx alata (leguminosae). Journal of Protein Chemistry, 14, Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and 685–693. Coordenação de Aperfeiçoamento de Pessoal de Nível Superior Marin, A. M., Arruda, S. F., & Siqueira, E. M. A. (2009). Minerals, phytic acid and tannin (CAPES) for the scholarship support. contents of 18 fruits from the Brazilian savanna. International Journal of Food Sci- ence, 60, 180–190. Matsuura, R. (1983). Uptake of iron and nitrilotriacetate (NTA) in rat liver and the toxic – References effect of Fe-NTA. Acta Medica Okayama, 37, 393 400. Mendonça, R. C., Felfili, J. M., Walter, B. M. T., Silva, M. C., Rezende, A. R., Filgueiras, T. S., et al. (1998). Flora vascular do Cerrado. In S. M. Sano, & S. P. Almeida (Eds.), Cerrado: Aebi, H. (1984). Catalase in vitro. Methods in Enzymology, 105, 121–126. Ambiente e Flora Brasil (pp. 286–556). Brasil: Empresa Brasileira de Pesquisa Agrope- Anderson, G., Frazer, D., & McLaren, G. (2009). Iron absorption and metabolism. Current cuária do Brasil (EMBRAPA-CPAC). Opinion in Gastroenterology, 25, 129–135. Paiva-Martins, F., Rodrigues, V., Calheiros, R., & Marques, M. P. (2011). Characterization Arruda, S. F., Siqueira, E. M. A., & Souza, E. M. T. (2004). Malanga (Xanthosoma sagitti- of antioxidant olive oil biophenols by spectroscopic methods. Journal of the Science folium) and Purslane (Portulaca oleracea) reduce oxidative stress in vitamin of Food and Agriculture, 30(91), 309–314. Adeficient rats. Annals of Nutrition and Metabolism, 48, 288–295. Pacher, P., Nivorozhkin, A., & Szabo, C. (2006). Therapeutic effects of xanthine oxidase Arruda, S. F., Siqueira, E. M. A., & Valência, F. (2009). Vitamin A deficiency increases inhibitors: Renaissance half a century after the discovery of allopurinol. Pharmaco- hepcidin expression and oxidative stress in rats. Nutrition, 25, 472–478. logical Reviews, 58,87–114. Baranowska, I., Czernicki, K., & Aleksandrowicz, R. (1995). The analysis of lead, cadmi- Phung, O. J., Makanji, S. S., White, C. M., & Coleman, C. I. (2009). Almonds have a neutral um, zinc, copper and nickel content in human bones from the upper Silesian indus- effect on serum lipid profiles: A meta-analysis of randomized trials. Journal of the trial district. Science of the Total Environmental, 159, 155–162. American Dietetic Association, 109, 865–873. Beckman, J. S., Beckman, T. W., Chen, J., Marshall, P. A., & Freeman, B. A. (1990). Appar- Reeves, P., Nielsen, F., & Fahey, G. J. (1993). AIN-93 purified diets for laboratory ro- ent hydroxyl radical production by peroxynitrite: Implications for endothelial inju- dents: Final report of the American Institute of Nutrition Ad Hoc Writing Commit- ry from nitric oxide and superoxide. Proceedings of the National Academy of tee on the reformulation of the AIN-76A rodent diet. Journal of Nutrition, 123, Sciences, 87, 1620–1624. 1939–1951. Benavente-García, O., Castillo, J., Marin, F. R., Ortuno, A., & Del Río, J. A. (1997). Uses and Richert, S., Wehr, N., Stadtman, E., & Levine, R. (2002). Assessment of skin carbonyl properties of Citrus flavonoids. Review. Journal of Agricultural and Food Chemistry, content as a noninvasive measure of biological age. Archives of Biochemistry and 45, 4505–4515. Biophysics, 397, 430–432. Benzie, I. F. F., & Strain, J. J. (1999). Ferric reducing/antioxidant power assay: Direct Sakurai, T., Kitadate, K., Nishioka, H., Fujii, H., Kizaki, T., Kondoh, Y., et al. (2010). Oligo- measure of total antioxidant activity of biological fluids and modified version for merized grape seed polyphenols attenuate inflammatory changes due to antioxi- simultaneous measurement of total antioxidant power and ascorbic acid concen- dative properties in coculture of adipocytes and macrophages. The Journal of tration. Methods in Enzymology, 299,15–27. Nutritional Biochemistry, 21,47–54. Bonvicino, C., Lemos, B., & Weksler, M. (2005). Small mammals of Chapada dos Veadeiros Sánchez-Moreno, C., Larrauri, J. A., & Sauro-Calixto, A. (1998). Procedure to Measure National Park (Cerrado of Central Brazil): Ecologic, karyologic, and taxonomic consid- the Antiradical Efficiency of Polyphenols. Journal of the Science of Food and Agri- erations. Brazilian Journal of Biology, 65,395–406. culture, 76, 270–276. Brand-Williams, W., Cuvelier, M. E., & Berset, C. (1995). Use of free radical method to Siqueira,A.C.M.F.,Nogueira,J.C.B.,Morais,E.,Kageyama,P.Y.,Murgel,J.M.T.,& evaluate antioxidant activity. Lebensm-Wiss Technology, 28,25–30. Zandarin, M. A. (1986). O cumbaru — Dipteryx alata Vog, Estudo de diferentes Brisdelli, F., D'Andrea, G., & Bozzi, A. (2009). Resveratrol: a natural polyphenol with procedências e progênies. Boletim Técnico do Instituto Florestal, 40, 281–290. multiple chemopreventive properties. Current Drug Metabolism, 10, 530–546. Singh, R. P., Murthy, K. N. C., & Jayaprakasha, G. K. (2002). Studies on the antioxidant Candan, N., & Tuzmen, N. (2008). Very rapid quantification of malondialdehyde (MDA) activity of pomegranate (Punica granatum) peel and seed extracts using in vitro in rat brain exposed to lead, aluminium and phenolic antioxidants by high- models. Journal of Agricultural and Food Chemistry, 50,81–86. performance liquid chromatography-fluorescence detection. Neurotoxicology, 29, Sousa, A. G. O., Fernandes, D. C., Alves, A. M., Freitas, J. B., & Naves, M. M. V. (2011). Nutri- 708–713. tional quality and protein value of exotic almonds and nut from the Brazilian Savanna Caramori, S., Lima, C., & Fernandes, K. (2004). Biochemical characterization of selected compared to peanut. Food Research International, 44,2319–2325. plant species from Brazilian savannas. Brazilian Archives of Biology and Technology, Sun, J., Chu, Y. F., Wu, X., & Liu, R. H. (2002). Antioxidant and antiproliferative activities 47, 253–259. of common fruits. Journal of Agricultural and Food Chemistry, 50, 7449–7454. DiPietro, C. M., & Liener, I. E. (1989). Heat inactivation of the Kunitz and Bowman-Birk Takemoto, E., Okada, I. A., Garbelotti, M. L., Tavares, M., & Aued-Pimentel, S. (2001). soybean protease inhibitors. Journal of Agricultural and Food Chemistry, 37,39–44. Composição química da semente e do óleo de baru (Dipteryx alata Vog.) nativo Ebina, Y., Okada, S., Hamazaki, S., Ogino, F., Li, J., & Midorikawa, O. (1986). Nephrotox- do Município de Pirenópolis, Estado de Goiás. Revista do Instituto Adolfo Lutz, 60, icity and renal cell carcinoma after use of iron- and aluminum-nitrilotriacetate 113–117. complexes in rats. Journal of the National Cancer Institute, 76, 107–113. Tempfer, C. B., Bentz, E. K., Leodolter, S., Tscherne, G., Reuss, F., Cross, H. S., et al. (2007). Faure, M., Lissi, E., Torres, R., & Videla, L. A. (1990). Antioxidant activities of lignans and Phytoestrogens in clinical practice: a review of the literature. Fertility and Sterility, flavonoids. Phytochemistry, 29, 3773–3775. 87, 1243–1249. Foti, M., Piattelli, M., Baratta, M. T., & Ruberto, G. (1996). Flavonoids, coumarins, and Van Horn, L., McCoin, M., Kris-Etherton, P., Burke, F., Carson, J., & Champagne, C. (2008). cinnamic acids as antioxidants in a micellar system, structure–activity relationship. The evidence for dietary prevention and treatment of cardiovascular disease. Jour- Journal of Agricultural and Food Chemistry, 44, 497–501. nal of the American Dietetic Association, 108, 287–331. Ganz, T. (2008). Iron homeostasis: Fitting the puzzle pieces together. Cell Metabolism, 7, Wijeratne, S., Abou-Za, M., & Shahidi, F. (2006). Antioxidant polyphenols in 288–290. and its coproducts. Journal of Agricultural and Food Chemistry, 54, 312–318. Gary, R., & Dao, L. (2003). Antioxidant constituents of almond [Prunus dulcis (Mill.) D.A. Webb] hulls. Journal of Agricultural and Food Chemistry, 51, 496–501.