- Email: [email protected]
Bioactive compounds in different hazelnut varieties and their skins
Journal of Food Composition and Analysis 43 (2015) 203–208
Contents lists available at ScienceDirect
Journal of Food Composition and Analysis journal homepage: www.elsevier.com/locate/jfca
Original Research Article
Bioactive compounds in different hazelnut varieties and their skins Neslihan Go¨ncu¨og˘lu Tas¸, Vural Go¨kmen * Department of Food Engineering, Hacettepe University, 06800 Beytepe, Ankara, Turkey
A R T I C L E I N F O
A B S T R A C T
Article history: Received 4 April 2015 Received in revised form 12 June 2015 Accepted 13 July 2015 Available online 17 July 2015
Bioactive profiles of hazelnut skins belonging to fourteen hazelnut varieties were identified. Concentration of phenolic compounds, flavonoids and phenolic acids in soluble free, conjugated soluble and insoluble bound fractions together with their total concentrations were presented. In addition, tocopherol content and total antioxidant capacity of hazelnuts and their skins were revealed. Concentration of total phenolic compounds ranged between 51.9 and 203.1 mg gallic acid equivalent/g of skin among varieties, which is in accordance with the total antioxidant capacity. Total flavonoid content was almost 60% of the total phenolic compounds. Flavonoids and phenolic acids were found to be concentrated mostly in the conjugated soluble fraction. Tocopherol contents of hazelnut skins ranged from 226 to 593 mg/g, and a-tocopherol was the most abundant. Total antioxidant capacity was between 309 and 1375 mmol Trolox equivalent/g of hazelnut skins, which is more than 100 times higher than for hazelnuts without the skins. ß 2015 Elsevier Inc. All rights reserved.
Keywords: Hazelnut Hazelnut skin Hazelnut varieties Corylus avellana Phenolic compounds Antioxidant capacity Tocopherols Food analysis Food composition
1. Introduction Hazelnut (Corylus avellana L.), which belongs to the family of Betulaceae, is widely consumed (Pelvan et al., 2012). Turkey is the main hazelnut producer, with annual production of 549 000 tonnes, which is almost 65% of the total world production (FAO, 2013). There are 18 hazelnut varieties: Acı, Cavcava, C¸akıldak, Fos¸a, Ham, I˙ncekara, Kalınkara, Kan, Karafındık, Kargalak, Kus¸, Mincane, Palaz, Sivri, Tombul, Uzun Musa, Yassı Badem, Yuvarlak Badem, grown in Turkey (Pelvan et al., 2012). Among these varieties, Tombul contributes to 25–30% of Turkey’s total production (Alasalvar et al., 2003). The skin of hazelnut may be consumed directly intact with kernel or it could be easily removed after roasting. Hazelnut skin constitutes 2.5% of the total weight of the hazelnut (Alasalvar et al., 2009) and the skin may be considered an important constituent because of its distinctive phenolic profile and high antioxidant ¨ zdemir et al., activity (Contini et al., 2008; Pelvan et al., 2012; O 2014). Improvement in colon metabolism, decreasing in total and LDL-cholesterol, triacylglycerol and non-esterified fatty acids in blood serum could be attributed to the high phenolic content, phytosterols and dietary fibre in hazelnut skins (Caimari et al.,
* Corresponding author. Tel.: +90 312 2977108; fax: +90 312 2992123. E-mail address: [email protected] (V. Go¨kmen). http://dx.doi.org/10.1016/j.jfca.2015.07.003 0889-1575/ß 2015 Elsevier Inc. All rights reserved.
2015). Therefore, consumption of hazelnut with its brown skin is better from the point view of health. On the other hand, because brown skin removed after roasting is a major by-product of the hazelnut industry, its utilization is considered to be necessary. For example, defatted fine hazelnut skin particles were added to coffee before and after brewing to improve the physiologically positive effects of coffee phytochemicals (Contini et al., 2012). Five percent (w/w) of hazelnut skin was also used for fortification of fresh egg pasta (Zeppa et al., 2015) and bread (Anil, 2007) to increase antioxidant activity and dietary fibre content. Additionally, hazelnut skin showed prebiotic activity, allowing the growth of Lactobacillus crispatus P17613 and Lactobacillus plantarum P17630. Even 0.01% (w/v) of hazelnut skin acted as a cryoprotective ingredient during lyophilization, making it promising for application in probiotic-containing foods and nutraceuticals (Montella et al., 2013). Defatted hazelnut skin powder obtained after high shear homogenization was not only a good source of dietary fibre, antioxidant compounds and phenolic compounds but was also rich in colouring agents. Previous works investigated the total phenolic content, antioxidant activity and condensed tannins (Locatelli et al., 2010; Monagas et al., 2009; Alasalvar et al., 2009), dimeric B type procyanidins (Monagas et al., 2009; Esatbeyoglu et al., 2014), monomeric and oligomeric flavan-3-ol contents (Monagas et al., 2009; Del Rio et al., 2011) of roasted hazelnut skins. Shahidi et al. (2007) compared individual phenolic acids, radical-scavenging
204
N.G. Tas¸, V. Go¨kmen / Journal of Food Composition and Analysis 43 (2015) 203–208
activity and inhibition of human LDL cholesterol in extracts of whole hazelnut, roasted hazelnut skin and other by-products. Contents of free, bound and total phenolic acids, total phenolic content and antioxidant activity of natural and roasted commercial hazelnut varieties were reported (Pelvan et al., 2012). Despite the fact that hazelnut skins may be a good dietary source, an overall screening of bioactive compounds and their distribution in hazelnut skin is lacking in the literature. The objective of this study is to reveal the bioactive profile of natural hazelnut skins that belong to fourteen varieties. For that reason, analyses of phenolics and flavonoids (free soluble, conjugated soluble, bound), and phenolic acids (free soluble, conjugated soluble, bound) have been performed. In addition to that, total antioxidant activity and tocopherol profile of both hazelnuts and their skins have been measured. 2. Materials and methods 2.1. Chemicals and consumables Methanol, ethanol, acetonitrile, acetone, n-hexane, ethyl acetate, isopropanol, hexane and water were all HPLC-grade and were obtained from Sigma Aldrich (Steinheim, Germany). Standards of gallic acid, ferulic acid, catechin, potassium peroxy disulphate, ABTS (2,20 -azino-bis(3-ethyl-benzothiazoline-6-sulfonic acid)), Folin–Ciocalteu reagent, a-tocopherol (96%), b-tocopherol (96%), g-tocopherol (96%), d-tocopherol (90%) were also obtained from Sigma Aldrich (Steinheim, Germany). Formic acid (98%), hydrochloric acid (37%), sodium hydroxide, sodium carbonate, sodium nitrite, aluminium chloride and diethyl ether were purchased from Merck Co. (Darmstadt, Germany). Trolox [6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid] was purchased from Fluka Chemie AG (Buchs, Switzerland). Syringe filters (nylon, 0.45 mm) were supplied from Waters (Milford, MA). 2.2. Hazelnut and hazelnut skin Fourteen Turkish hazelnut (C. avellana L.) varieties (Acı, C¸akıldak, Fos¸a, I˙ncekara, Kalınkara, Kan, Kargalak, Kus¸, Palaz, Sivri, Tombul, Uzun Musa, Yassı Badem, Yuvarlak Badem), grown in the orchard of the Hazelnut Research Station in Giresun on the northeast coast of Turkey, were used for the analysis. Hazelnuts were left on the trees until the first week of August 2013; they were collected by hand when their green leafy covers became pale and moisture content decreased to 30%. They were then left in the sun until the moisture content became 6% for about 3–5 days. Pale leafy covers of hazelnuts were removed and three separate bags from each variety, each of them containing 2 kg hazelnuts, were immediately sent for analysis. Until analysis, hazelnuts were stored in their shells at –20 8C. Before analysis, hazelnut shells were cracked and hazelnut skin was removed manually by using a blade. About 2.5 g of hazelnut skin were obtained by scraping 100 g of hazelnut from each bag. Then, skin samples were finely ground using a coffee mill. Three independent hazelnut and hazelnut skin samples (n = 3) were analyzed with two analytical measurements. 2.3. Separation of phenolic fractions with alkaline hydrolysis Soluble free, soluble conjugated and insoluble bound phenolic compounds were extracted as described by Moore et al. (2005). First, 0.25 g of hazelnut skin were extracted with 4 mL of methanol/acetone/water (7:7:6, v/v/v) and centrifuged at 6000 g for 3 min. Extraction with 4 mL of methanol/acetone/ water (7:7:6, v/v/v) was repeated 5 times and supernatants of each step were collected and combined in a test tube (extract A). Pellet
obtained after this extraction was kept for the analysis of insoluble bound phenolic compounds. A quantity of 10 mL of the extract A was transferred to another tube and 10 mL of 4 N NaOH were added to it, in order to release soluble conjugated phenolic compounds (extract B). A quantity of 7 mL of 4 N NaOH was added to the pellet simultaneously to release insoluble bound phenolic compounds (extract C). Then, both tubes were left for 4 h in a shaker at the room temperature. After then, 5 mL of extract from A, B and C were transferred to test tubes and the pH was adjusted to 2 in all tubes, by using 6 N HCl. A quantity of 5 mL of diethyl ether:ethyl acetate (1:1, v/v) was added into these tubes, vortexed for 2 min and centrifuged at 6000 g for 2 min. The same extraction procedure with 5 mL of diethyl ether:ethyl acetate (1:1, v/v) was repeated four times. Supernatants were collected at each step and extracts were combined. Then, 5 mL of the combined extract were evaporated under an N2 stream at 30 8C to complete dryness. After then, phenolic compounds were redissolved in 1.5 mL of methanol and kept at 20 8C until the analyses were performed. As extract B contained both soluble free and soluble conjugated phenolic compounds, soluble conjugated phenolic compounds were calculated by subtracting the amount of soluble free phenolic compounds obtained from the analysis of extract A. 2.4. Analysis of total phenolic content Total phenolic content analyses were performed according to the Folin–Ciocalteu method (Hoff and Singleton, 1977). First, 25 mL of appropriate extracts were transferred into a test tube and 0.8 mL of 0.2 N Folin–Ciocalteu reagent were added to it. After 5 min of reaction time, the solution was neutralized with 0.8 mL of 20% Na2CO3 (w/v). The mixture was kept in the dark for 2 h until the characteristic blue colour was observed. After centrifugation at 6000 g for 4 min, absorbance of the supernatant was measured against methanol at 765 nm. Calibration curve was built with different concentrations of gallic acid dissolved in methanol and results were given as mg gallic acid equivalent (GAE) per g of hazelnut skin. 2.5. Analysis of individual phenolic acids Chromatographic analyses were performed on an Agilent 1200 HPLC system consisting of a diode array detector, quaternary pump, autosampler, and column oven (Agilent Technologies, Waldbronn, Germany). Before analysis, extracts were filtered through a 0.45-mm nylon syringe filters and taken into vials. Phenolic acids were separated on a Waters Atlantis C18 column (250 mm 4.6 mm id., 5 mm; Waters Corp., Milford, MA) by using 1% formic acid in water (A) and 1% formic acid in acetonitrile (B) at a flow rate of 1 mL/min with the following gradient programme: linear gradient elution from 10 to 20% B, 0–10 min; linear gradient elution from 20 to 40% B, 10–20 min; linear gradient elution from 40 to 10% B, 20–25 min and isocratic elution of 10% B, 25–30 min. The column temperature was 30 8C and injection volume was 10 mL. Analyses were performed by using diode array detector at 280 nm. Calibration curves were built for each of the compounds identified in the samples and phenolic acids were expressed as mg per g of hazelnut skin. 2.6. Analysis of total flavonoid content Total flavonoid content was determined according to the method described by Zhishen et al. (1999). First, 100 mL of appropriate extract were mixed with 50 mL 5% NaNO2. After waiting for 6 min of reaction time, 500 mL of 10% AlCl3 were added to the mixture to form aluminium-flavonoid complex. After 7 min,
N.G. Tas¸, V. Go¨kmen / Journal of Food Composition and Analysis 43 (2015) 203–208
250 mL of 1 N NaOH were added and the mixture was kept in the dark for 10 min. Then, the mixture was centrifuged at 6000 g for 5 min and the clear supernatant was measured against methanol at 510 nm. A calibration curve was built with different concentrations of catechin dissolved in methanol and results were given as mg catechin equivalent (CE) per g of hazelnut skin. 2.7. Analysis of total antioxidant capacity Total antioxidant capacity measurements were performed according to the QUENCHER method described by Serpen et al. (2009) by using 7 mM of ABTS aqueous solution with 2.45 mM potassium peroxy disulfate. The solution was kept in the dark for 12–16 h to obtain ABTS radical solution. ABTS working solution was prepared by diluting ABTS radical solution with ethanol–water (50:50, v/v) until the absorbance was 0.75–0.80 at 734 nm. Hazelnut skins were diluted 100 times with cellulose. A quantity of 20 mg of hazelnut and 10 mg of diluted hazelnut skin were weighed into test tubes and 10 mL of ABTS working solution was added. Tubes were placed in a shaker and left shaking in the dark. After centrifugation at 6000 g for 2 min, clear supernatants were measured at 734 nm exactly at the 30th minute of ABTS working solution addition onto samples. Trolox standard solutions were prepared in methanol between 0 and 600 mg/mL. Then, 100 mL of Trolox standard solutions and 10 mL ABTS working solutions were mixed and left 30 min in the dark. Absorbance of standard solutions was measured at 734 nm and a calibration curve was built. Antioxidant capacity of hazelnut and hazelnut skins were determined by using calibration curve and the results were expressed as mmol Trolox equivalent antioxidant capacity per g of hazelnut or hazelnut skin. 2.8. Analysis of tocopherols Hazelnut oil was extracted by hexane with Soxhlet extraction for 12 h after removal of their skins. Tocopherol analyses were performed both in hazelnut oils and skins. First, 0.5 g of hazelnut oil or hazelnut skin was weighed into test tubes and tocopherols were extracted with 2.5 mL of isopropanol–methanol (50:50, v/v) and centrifuged at 6000 g for 3 min. The procedure was repeated 3 times and supernatants were collected and combined. After passing through 0.45-mm filters, the extracts were collected in vials. Tocopherol analysis was performed using an Agilent LC/MS system containing a quaternary pump, temperature-controlled oven, Agilent 6130 MS detector and Agilent 1200 HPLC system (Agilent Technologies, Waldbronn, Germany). Separation of tocopherols was performed with an Agilent ZORBAX Rapid Resolution SB-C18 column (50 4.6 mm id., 3.5 mm; Waldbronn,
205
Germany) using mobile phases 1% formic acid in water (A) and methanol (B). An isocratic mixture of A and B (5:95, v/v) was used and the flow rate was 1 mL/min. Injection volume was 5 mL. Analyses were carried out in positive ionization mode. Monitored ions for a-tocopherol, b- and g-tocopherol, and d-tocopherol were m/z 431.7, 417.7, and 403.7, respectively. Standard solutions of each tocopherol were prepared at concentrations of 1, 5, 10, 20, 50 mg/L. Tocopherol contents of hazelnuts were calculated by using tocopherol contents of hazelnut oils and total fat contents of each variety. The data were expressed as mg tocopherol per g hazelnut or hazelnut skin (Yılmaz and Go¨kmen, 2013). 2.9. Statistical analysis All experiments were performed in triplicate for each hazelnut variety and all samples were analyzed with duplicate measurements. Experimental data were expressed as mean standard deviation. Significance of difference between varieties was analyzed by using one-way ANOVA and Duncan’s test (p < 0.05) by using SPSS Version 17.0. 3. Results and discussion 3.1. Soluble free, conjugated soluble, insoluble bound and total phenolic contents of hazelnut skins Soluble free, conjugated soluble, insoluble bound and total phenolic contents of hazelnut skins belong to fourteen hazelnut varieties are shown in Table 1. The average value for the concentration of soluble free phenolic compounds in the skins of fourteen hazelnut varieties was 17.3 mg GAE/g, 15% of the total phenolic compounds. Among hazelnut varieties, Yassı Badem had the lowest soluble free phenolic compounds with 7.1 mg GAE/g skin, while C¸akıldak had the highest with 30.6 mg GAE/g skin. Conjugated soluble phenolic compounds, comprised 74% of total phenolics with an average value of 84 mg GAE/g skin. Concentration of conjugated soluble phenolic compounds was lowest in Fos¸a with 30.1 mg GAE/g skin and highest in C¸akıldak with 155.2 mg GAE/g skin. Insoluble bound phenolics comprised 11% of the total phenolic compounds with an average value of 11.8 mg GAE/g skin. Insoluble bound phenolics were lowest in Acı with 7.6 mg GAE/g and highest in Uzun Musa with 29.4 mg GAE/g. Total phenolic compounds ranged between 51.9 and 203.1 mg GAE/g skin among the varieties. Previous works revealed the total phenolic content of roasted hazelnut skins as 233 mg CE/g skin of Tombul variety ¨ zdemir et al., 2014), or ranging between 41 and 127 mg (O polyphenols/g in roasted hazelnut skins of varieties from Turkey, Italy and Chile (Del Rio et al., 2011), and 182 mg CE/g in
Table 1 Concentration of soluble free, conjugated soluble, insoluble bound and total phenolic compounds of hazelnut skins (mg GAE/g)*. Soluble free phenolics Kargalak Palaz I˙ncekara Sivri Yassı Badem Fos¸a Kalınkara Yuvarlak Badem Kus¸ C¸akıldak Kan Uzun Musa Acı Tombul
d
18.8 4.9 14.4 1.5b,c,d 17.0 2.9d 15.2 5.0c,d 7.1 1.2a 11.1 0.1a,b,c 24.4 1.5e 9.8 0.7a,b 17.1 1.4d 30.6 0.6f 27.6 1.9e,f 16.8 0.4d 16.1 0.8c,d 15.7 1.3c,d
Conjugated soluble phenolics d,e
122.9 18.2 103.9 11.3d 73.2 11.2c 57.0 1.1b,c 45.8 0.4a,b 30.1 3.8a 67.0 6.5b,c 74.2 9.4c 107.3 7.0d,e 155.2 3.3f 44.7 8.2a,b 127.3 17.5e 54.2 8.3b,c 111.0 8.1d,e
Insoluble bound phenolics a
9.5 2.0 9.4 0.1a 11.8 2.5a,b 8.2 1.4a 8.0 0.4a 10.8 2.0a 9.1 1.6a 9.3 0.9a 7.7 0.8a 17.3 2.4c 10.9 0.2a 29.4 0.8d 7.6 1.7a 15.4 2.3b,c
Total phenolics 151.2 21.1f 127.7 9.7e 102.0 5.8d 80.4 2.4b,c,d 60.9 0.5a,b 51.9 2.0a 100.6 4.9c,d 93.4 9.2c,d 132.0 7.7e,f 203.1 1.5h 83.2 6.1b,c,d 173.5 17.1g 77.9 9.2b,c 142.2 11.7e,f
* Superscript letters in each column indicate statistically significant difference (p < 0.05) according to Duncan’s test. Three independent hazelnut skin sample (n = 3) were analyzed with two analytical measurements.
N.G. Tas¸, V. Go¨kmen / Journal of Food Composition and Analysis 43 (2015) 203–208
206
Table 2 Concentration of soluble free, conjugated soluble, insoluble bound and total flavonoid compounds of hazelnut skins (mg CE/g)*. Soluble free flavonoids Kargalak Palaz I˙ncekara Sivri Yassı Badem Fos¸a Kalınkara Yuvarlak Badem Kus¸ C¸akıldak Kan Uzun Musa Acı Tombul
c
3.4 0.0 2.6 0.8a,b,c 2.4 0.1a,b,c 2.0 0.2a,b 1.7 0.1a 2.0 0.5a,b 3.3 0.4c 2.6 0.5a,b,c 2.6 0.1a,b,c 4.5 0.2d 4.3 0.1d 2.8 0.6b,c 2.9 0.5b,c 2.3 0.5a,b
Conjugated soluble flavonoids c
Insoluble bound flavonoids a,b
52.4 4.1 48.1 2.8b,c 42.9 1.4a,b,c 35.5 0.4a,b,c 30.4 1.3a,b 25.5 1.0a 44.0 0.5b,c 41.4 2.4a,b,c 50.9 3.7c 82.4 2.3d 40.1 3.7a,b,c 81.7 3.4d 38.3 2.5a,b,c 49.4 1.3c
3.4 0.4 3.5 0.10a,b 4.2 0.40a,b 2.7 0.71a 2.7 0.50a 3.4 0.27a,b 3.3 0.27a,b 3.0 0.10a,b 2.7 0.41a 6.7 0.95b 5.6 0.10a,b 27.9 5.68c 2.9 0.18a,b 5.3 0.60a,b
Total flavonoids 59.2 3.73d 54.1 1.88c,d 49.5 1.25b,c,d 40.2 0.15a,b,c 34.8 1.91a,b 30.9 0.39a 50.7 1.25b,c,d 47.0 2.87a,b,c,d 56.2 4.08c,d 93.6 2.41e 50.0 3.75b,c,d 112.4 2.86f 44.0 3.14a,b,c,d 57.0 1.31c,d
* Superscript letters in each column indicate statistically significant difference (p < 0.05) according to Duncan’s test. Three independent hazelnut skin sample (n = 3) were analyzed with two analytical measurements.
medium-roasted hazelnut skin and 191 mg CE/g in high-roasted hazelnut skin (Locatelli et al., 2010). Pelvan et al. (2012) found 1.78–2.46 mg GAE/g total phenolic content in natural hazelnut varieties and suggested to consume hazelnuts with their skins as there was a significant loss in total phenolics, condensed tannins and free and bound phenolic acids after removal of skin. 3.2. Soluble free, conjugated soluble, insoluble bound and total flavonoid content of hazelnut skins Soluble free, conjugated soluble, insoluble bound and total flavonoid content of hazelnut skins is given in Table 2. Most of the flavonoids (85% of the total) were found in conjugated soluble form in hazelnut skin. In parallel with the results of conjugated phenolic compounds, Fos¸a was the variety containing least conjugated flavonoid with 25.5 mg CE/g and C¸akıldak was the highest with 82.4 mg CE/g. Insoluble bound flavonoids comprised 10% of the total flavonoids with an average 5.5 mg GAE/g skin. The least flavonoid-containing fraction was the soluble free fraction, with an average of 2.8 mg CE/g skin. Total flavonoid content of hazelnut skins ranged between 30.9 and 112.4 mg CE/g skin, which were almost 60% of the total phenolic compounds. Catechin, epicatechin, epicatechin gallate, gallocatechin, epigal¨ zdemir et al., 2014), procyanidin and locatechin gallate (O procyanidin dimers and trimers, and procyanidin dimer gallate were the flavan-3-ols identified in aqueous extracts of hazelnut skins (Del Rio et al., 2011). In addition, quercetin, myricetin,
quercetin-3-O-rhamnoside, myricetin rhamnoside, and kaempherol rhamnoside were the flavonols and phloretin 20 -O-glucoside was the dihydrochalcone detected in methanolic extracts of hazelnut skin (Del Rio et al., 2011). In this study, a combination of methanol/ acetone/water was used and alkaline hydrolysis performed. 3.3. Characterization of soluble free, conjugated soluble and insoluble bound phenolic acids of hazelnut skins Gallic acid and ferulic acid were the phenolic acids characterized in natural hazelnut skins (Table 3). Gallic acid was found in both soluble free and conjugated soluble fractions while ferulic acid was detected in conjugated soluble and insoluble fractions (Figure S1). Both phenolic acids were found to be dominant in the conjugated soluble fraction. Soluble free phenolic acids found in skin of Kargalak were lowest with 0.18 mg/g while Uzun Musa was highest with 0.5 mg/g skin although there was no significant difference between varieties (p > 0.05). Gallic acid found in conjugated form ranged between 0.5 and 3.7 mg/g skin. Ferulic acid in conjugated soluble form was approximately 8-fold higher than insoluble bound form. Ferulic acid ranged from 1.2 to 2.1 mg/g skin in conjugated form and 0.11 to 0.39 mg/g skin in insoluble bound form. Total phenolic acids determined in hazelnut skins were minimum of 3.0 and maximum of 8.3 mg/g skin. Pelvan et al. (2012) reported the total phenolic acid contents of natural hazelnuts ranged from 62.1 mg/g to 143.1 mg/g. ¨ zdemir et al. (2014) found 0.79 mg gallic acid/g skin in soluble free O fraction of roasted hazelnut skin.
Table 3 Concentration of soluble free, conjugated soluble, insoluble bound and total phenolic acids of hazelnut skins (mg/g)*.
Kargalak Palaz Incekara Sivri Yassı Badem Fos¸a Kalınkara Yuvarlak Badem Kus¸ C¸akıldak Kan Uzun Musa Acı Tombul
Soluble free phenolic acids
Conjugated soluble phenolic acids
Insoluble bound phenolic acids
Gallic acid
Ferulic acid
Gallic acid
Ferulic acid
0.18 0.03a 0.30 0.13a 0.33 0.08a 0.41 0.12a 0.28 0.15a 0.49 0.13a 0.50 0.13a 0.35 0.13a 0.34 0.10a 0.48 0.10a 0.41 0.03a 0.51 0.03a 0.22 0.03a 0.25 0.08a
2.1 0.01a 1.7 0.10a 1.8 0.12a 1.6 0.01a 1.8 0.10a 1.9 0.3a 1.5 0.17a 2.0 0.47a 2.0 0.25a 1.5 0.25a 2.2 0.23a 1.7 0.09a 2.0 0.39a 1.2 0.1a
3.7 0.25c,d 3.5 0.92c,d 1.7 0.05a,b,c 1.1 0.01a,b,c 0.7 0.04a,b 0.5 0.03a 1.6 0.07a,b,c 1.9 0.16a,b,c 2.7 0.40a,b,c 3.3 0.34b,c,d 0.6 0.18a,b 5.8 1.91d 1.5 0.10a,b,c 3.10 0.44a,b,c
0.20 0.03a,b 0.27 0.01a,b,c 0.22 0.09a,b,c 0.16 0.02a,b 0.22 0.01a,b,c 0.23 0.03a,b,c 0.18 0.01a,b 0.26 0.09a,b,c 0.19 0.01a,b 0.39 0.09c 0.14 0.03a,b 0.31 0.09b,c 0.11 0.02a 0.27 0.09a,b,c
Total phenolic acids
6.2 0.3a,b 5.8 1.2a,b 4.1 0.5a 3.3 0.2a 3.0 0.3a 3.1 0.5a 3.7 0.4a 4.5 0.9a 5.2 0.4a,b 5.6 0.4a,b 3.4 0.3a 8.3 1.8b 3.8 0.9a 4.9 0.4a
* Superscript letters in each column indicate statistically significant difference (p<0.05) according to Duncan’s test. Three independent hazelnut skin sample (n=3) were analyzed with two analytical measurements.
N.G. Tas¸, V. Go¨kmen / Journal of Food Composition and Analysis 43 (2015) 203–208
207
Table 4 Concentration of tocopherols in hazelnuts and hazelnut skins (mg/g)*. Hazelnut skin
Hazelnut
Kargalak Palaz I˙ncekara Sivri Yassı Badem Fos¸a Kalınkara Yuvarlak Badem Kus¸ C¸akıldak Kan Uzun Musa Acı Tombul
b + g Tocopherol
a-Tocopherol
Total tocopherols
b + g Tocopherol
a-Tocopherol
Total tocopherols
6.3 1.5a 21.6 3.1c,d,e,f,g 26.4 1.4e,f,g 25.1 0.1c,d,e,f,g 13.2 0.1a,b,c,d 12.0 0.7a,b,c 7.9 1.1a,b,c 12.9 5.3a,b 47.2 1.9a,b,c,d 29.5 1.5h 19.5 4.0b,c,d,e,f 25.9 2.3d,e,f,g 13.7 0.1a,b,c,d,e 33.5 0.3g
168.5 3.4c,d 129.4 10.0a,b 134.5 4.5a,b,c 145.8 0.4a,b,c,d 147.4 12.4a,b,c,d 155.4 7.5b,c,d 180.8 12.3d 158.0 4.0b,c,d 143.5 10.2a,b,c 136.1 0.6a,b,c 111.2 3.6a 155.9 14.9b,c,d 223.8 7.1f 248.3 1.7f
174.8 4.9a,b 151.0 13.1a,b 161.0 5.9a,b 170.9 0.5a.b 160.6 12.5a,b 167.4 8.2a,b 188.7 11.1b 170.8 9.3a,b 190.7 12.1b 165.5 2.1a,b 130.7 7.6a 181.8 17.2b 237.5 7.2c 281.7 2.0d
149.6 2.3d 150.8 1.1d 146.7 3.2c,d 197.7 0.9e 143.4 12.9c,d 75.9 5.8a,b 73.8 3.4a,b 79.9 2.4b 126.8 2.6c 146.5 11.4c,d 78.4 15.5a,b 57.9 0.5a 79.4 7.1a,b 89.1 0.9b
443.9 8.7h 250.0 0.4c,d,e 295.6 1.1d,e,f 357.5 19.1f,g 293.7 11.0d,e,f 372.0 9.4g 346.3 7.3f,g 236.7 22.5b,c,d 200.7 3.8a,b,c 256.6 39.2c,d,e 172.6 27.1a,b 168.2 3.8a 312.9 9.5e,f,g 207.0 0.6a,b,c
593.5 11.0f 400.7 0.8d,e 442.3 4.3e 555.2 18.2f 437.1 23.9e 447.9 15.2e 420.0 3.9e 316.6 24.9b,c 327.5 1.2b,c,d 403.1 50.6d,e 251.0 42.6a,b 226.1 3.2a 392.3 16.6c,d,e 296.1 1.6a,b
* Superscript letters in each column indicate statistically significant difference (p < 0.05) according to Duncan’s test. Three independent hazelnut skin and hazelnut sample (n = 3) were analyzed with two analytical measurements.
3.4. Tocopherol contents of hazelnuts and hazelnut skins Concentrations of (b + g)-tocopherol, a-tocopherol and total tocopherols found in both hazelnut varieties and their skins are given in Table 4. A total ion chromatogram of hazelnut skin and extracted ion chromatograms indicating a-tocopherol and (b + g)tocopherol are presented in Figure S2 in Supplementary Figures. a-Tocopherol was the most abundant tocopherol in both hazelnuts and hazelnut skin, ranging from 111.2 to 248.3 mg/g in hazelnuts. (b + g)-Tocopherol concentration of hazelnuts showed a large range between 6.3 and 47.2 mg/g. Total tocopherol content of hazelnuts was lowest in Kan with 130.7 mg/g and highest in Tombul with 281.7 mg/g. Ko¨ksal et al. (2006) identified a-, g- and d-tocopherols in natural hazelnut varieties and reported total tocopherol contents ranging between 196 and 414 mg/g. Total tocopherol content of Tombul was 513.1 mg/g and expectedly a-tocopherol constituted the highest with 404.0 mg/g (Alasalvar et al., 2006). Compared to the results of this study, higher tocopherol contents in hazelnuts were detected, as the skin was not removed in the other studies. In hazelnut skin, a-tocopherol was found to be ranging from 168.2 to 443.8 mg/g. (b + g)Tocopherol concentration of hazelnut skins was almost five times higher than hazelnuts and ranged from 57.9 to 197.7 mg/g skin. Total tocopherol content of hazelnut skins was lowest in Uzun Musa with 226.1 mg/g and highest in Kargalak with 593.5 mg/g; that was approximately two fold higher than hazelnuts. Tocopherol content of hazelnut skin oil was found to be 2770 mg/g ¨ zdemir et al., 2014) that was almost ten times higher than (O tocopherol contents of hazelnut skins detected in this study. 3.5. Total antioxidant capacity of hazelnuts and hazelnut skins Total antioxidant activity of hazelnut and hazelnut skins are expressed as Trolox equivalent antioxidant capacity in Table 5. Total antioxidant capacity of hazelnut varieties ranged from 5.4 to 8.8 mmol Trolox equivalent/g hazelnut. Total antioxidant capacity of hazelnut skins, with an average value of 878 mmol Trolox equivalent/g, was more than 100 times higher than hazelnuts. Among varieties, skin of Fos¸a was found to contain lowest amount of antioxidants with 309 mmol Trolox equivalent/g while skin of C¸akıldak had highest antioxidant capacity with 1375 mmol Trolox equivalent/g. Total antioxidant capacity of hazelnut skins was found to be in accordance with total phenolic compounds. Locatelli et al. (2010) found total antioxidant capacity of medium-roasted defatted skins as 1.10 mmol Trolox equivalent/g and 0.94 mmol Trolox equivalent/g for high-roasted defatted hazelnut skins.
Table 5 Total antioxidant capacity of hazelnuts and hazelnut skins (mmol Trolox equivalent/g)*.
Kargalak Palaz I˙ncekara Sivri Yassı Badem Fos¸a Kalınkara Yuvarlak Badem Kus¸ C¸akıldak Kan Uzun Musa Acı Tombul
Hazelnut
Hazelnut skin
7.8 0.5a,b,c 7.1 0.2a,b,c 7.5 1.5a,b,c 8.0 1.4b,c 6.5 1.1a,b,c 6.6 0.5a,b,c 7.0 0.3a,b,c 6.2 0.2a,b 8.3 1.1b,c 6.8 0.8a,b,c 5.4 0.5a 5.9 0.3a,b 7.7 1.5a,b,c 8.8 1.5c
1153 120g 887 3d,e,f 875 37d,e,f 775 139c,d,e 462 50a,b 309 14a 934 14e,f 706 17c,d 613 56b,c 1375 171h 952 117e,f 1343 233h 881 24d,e,f 1027 171f,g
* Superscript letters in each column indicate statistically significant difference (p < 0.05) according to Duncan’s test. Three independent hazelnut skin and hazelnut sample (n = 3) were analyzed with two analytical measurements.
Antioxidant capacity of hazelnut skin was 10 times higher than even from the most antioxidant-rich cereal, buckwheat (Serpen et al., 2008). In addition to that, antioxidant rich foods like cinnamon, dark chocolate (70% cocoa), blueberry, black filtered coffee and green tea were reported to have antioxidant capacities of 984, 134, 82, 28 and 25 mmol Trolox equivalent/g or mL, respectively (Blomhoff et al., 2006). Moreover, 1 g of the most antioxidant rich hazelnut skin, belonging to variety C¸akıldak, was equivalent to 1.4 g of cinnamon, 10 g of dark chocolate, 16.7 g of blueberry, 49.1 mL of black filtered coffee or 55 mL of green tea. However, these values could change among varieties as 1 g of the least antioxidant containing skin variety, Fos¸a, compensated 0.3 g of cinnamon, 2.3 g of dark chocolate, 3.8 g of blueberry, 11.0 mL of black filtered coffee or 12.4 mL green coffee. 4. Conclusion Hazelnut skin could be considered as a good source of bioactive compounds in comparison to its kernel. Bioactive profile and distribution of bioactive compounds in skin of hazelnut varieties was measured in this study. Phenolic compounds, flavonoids and phenolic acids were found to be rich in conjugated soluble fraction. Total phenolic compounds were lowest in skin of Fos¸a and highest in skin of C¸akıldak. Total flavonoid content of hazelnuts was 60% of total phenolic compounds. Moreover, hazelnut skin was found to
208
N.G. Tas¸, V. Go¨kmen / Journal of Food Composition and Analysis 43 (2015) 203–208
contain two times higher amount of tocopherols, especially a-tocopherol, than hazelnut itself. Total antioxidant capacity of hazelnut skin is much more higher (>100 times) than hazelnut and many other foods previously reported. Total antioxidant capacity was in accordance with total phenolic compounds; the skin of Fos¸a was the least antioxidant-containing variety and C¸akıldak was the highest. Although bioactive profile of hazelnut varieties covered a wide range, hazelnut skins of all varieties could have positive effects on health. Therefore, their consumption while intact with hazelnut kernel or as an ingredient of other foods should be taken into consideration. Acknowledgements This study was supported by The Scientific and Technological Research Council of Turkey (TUBITAK) in the frame of Intensified Cooperation (IntenC) Programme (Project no: 113O178). Authors are grateful to Hazelnut Research Station in Giresun for providing hazelnut varieties. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jfca.2015.07.003. References Alasalvar, C., Shahidi, F., Liyanapathirana, C.M., Oshima, T., 2003. Turkish Tombul hazelnut (Corylus avellana L.). 1. Compositional characteristics. J. Agric. Food Chem. 51, 3790–3796. Alasalvar, C., Amaral, J.S., Shahidi, F., 2006. Functional lipid characteristics of Turkish Tombul hazelnut (Corylus avellana L.). J. Agric. Food Chem. 54, 10177–10183. Alasalvar, C., Karamac, M., Kosinska, A., Rybarczyk, A., Shahidi, F., Amarowicz, R., 2009. Antioxidant activity of hazelnut skin phenolics. J. Agric. Food Chem. 57, 4645–4650. Anil, M., 2007. Using of hazelnut testa as a source of dietary fiber in breadmaking. J. Food Eng. 80, 61–67. Blomhoff, R., Carlsen, M.H., Andersen, L.F., Jacobs Jr., D.R., 2006. Health benefits of nuts: potential role of antioxidants. Brit. J. Nutr. 96, S52–S60. Caimari, A., Puiggros, F., Suarez, M., Crescenti, A., Laos, S., Ruiz, J.A., Alonso, V., Moragas, J., del Bas, J.M., Arola, L., 2015. The intake of a hazelnut skin extract improves the plasma lipid profile and reduces the lithocholic/ deoxycholic bile acid faecal ratio, a risk factor for colon cancer, in hamsters fed a high-fat diet. Food Chem. 167, 138–144. Contini, M., Baccelloni, S., Massantini, R., Anelli, G., 2008. Extraction of natural antioxidants from hazelnut (Corylus avellana L.) shell and skin wastes by long maceration at room temperature. Food Chem. 110, 659–669.
Contini, M., Bacelloni, S., Frangipane, M.T., Merendino, N., Massantini, R., 2012. Increasing espresso coffee brew antioxidant capacity using phenolic extract recovered from hazelnut skin waste. J. Functional Foods 4, 137–146. Del Rio, D., Calani, L., Dall’Asta, M., 2011. Polyphenolic composition of hazelnut skin. J. Agric. Food Chem. 59, 9935–9941. Esatbeyoglu, T., Juadjur, A., Wray, V., Winterhalter, P., 2014. Semisynthetic preparation and isolation of dimeric procyanidins B1B8 from roasted hazelnut skins (Corylus avellana L.) on a large scale using countercurrent chromatography. J. Agric. Food Chem. 62, 7101–7110. Food and Agriculture Organization of the United Nations (FAO). (2013). FAO Statistics Division. Retrieved April 3, 2015 from http://faostat3.fao.org/ download/Q/QC/E. Hoff, J.F., Singleton, K.I., 1977. A method for determination of tannin in foods by means of immobilized enzymes. J. Food Sci. 42, 1566–1569. Ko¨ksal, A.I., Artık, N., S¸ims¸ek, A., Gu¨nes¸, N., 2006. Nutrient composition of hazelnut (Corylus avellana L.) varieties cultivated in Turkey. Food Chem. 99, 509–515. Locatelli, M., Travaglia, F., Coisson, J.D., Martelli, A., Stevigny, C., Arlorio, M., 2010. Total antioxidant activity of hazelnut skin (Nocciola Piemonte PGI): impact of different roasting conditions. Food Chem. 119, 1647–1655. Monagas, M., Garrido, I., Lebron-Aguilar, R., Gomez-Cordoves, M.C., Raybarczyk, A., Amarowicz, R., Bartolome, B., 2009. Comparative flavan-3-ol profile and antioxidant capacity of roasted peanut, hazelnut, and almond skins. J. Agric. Food Chem. 57, 10590–10599. Montella, R., Coisson, J.D., Travaglia, F., Locatelli, M., Malfa, P., Martelli, A., Arlorio, M., 2013. Bioactive compounds from hazelnut skin (Corylus avellana L.): effects on Lactobacillus plantarum P17630 and Lactobacillus crispatus P17631. J. Functional Foods 5, 306–315. Moore, J., Hao, Z., Zhou, K., Luther, M., Costa, J., Yu, L.L., 2005. Carotenoid, tocopherol, phenolic acid, and antioxidant properties of Maryland-grown soft wheat. J. Agric. Food Chem. 53, 6649–6657. ¨ zdemir, K.S., Yılmaz, C., Durmaz, G., Go¨kmen, V., 2014. Hazelnut skin O powder: a new brown colored functional ingredient. Food Res. Int. 65, 291–297. Pelvan, E., Alasalvar, C., Uzman, S., 2012. Effects of roasting on the antioxidant status and phenolic profiles of commercial Turkish hazelnut varieties (Corylus avellana L.). J. Agric. Food Chem. 60, 1218–1223. Serpen, A., Go¨kmen, V., Pellegrini, N., Fogliano, V., 2008. Direct measurement of the total antioxidant capacity of cereal products. J. Cereal Sci. 48, 816–820. Serpen, A., Go¨kmen, V., Fogliano, V., 2009. Direct measurement of the total antioxidant capacity of foods: the ‘QUENCHER’ approach. Trends Food Sci. Technol. 20, 278–288. Shahidi, F., Alasalvar, C., Liyanapathirana, C.M., 2007. Antioxidant phytochemicals in hazelnut kernel (Corylus avellana L.) and hazelnut byproducts. J. Agric. Food Chem. 55, 1212–1220. Yılmaz, C., Go¨kmen, V., 2013. Compositional characteristics of sour cherry kernel and its oil as influenced by different extraction and roasting conditions. Ind. Crops Prod. 49, 130–135. Zeppa, G., Belviso, S., Bertolino, M., Cavallero, M.C., Dal Bello, B., Ghirardello, D., Giordano, M., Giorgis, M., Grosso, A., Rolle, L., Gerbi, V., 2015. The effect of hazelnut roasted skin from different cultivars on the quality attributes, polyphenol content and texture of fresh egg pasta. J. Sci. Food Agric. 95, 1678–1688. Zhishen, J., Mengcheng, T., Jianming, W., 1999. The determination of flavanoid contents in mulberry and their scavenging effects on superoxide radicals. Food Chem. 64, 555–559.
Contents lists available at ScienceDirect
Journal of Food Composition and Analysis journal homepage: www.elsevier.com/locate/jfca
Original Research Article
Bioactive compounds in different hazelnut varieties and their skins Neslihan Go¨ncu¨og˘lu Tas¸, Vural Go¨kmen * Department of Food Engineering, Hacettepe University, 06800 Beytepe, Ankara, Turkey
A R T I C L E I N F O
A B S T R A C T
Article history: Received 4 April 2015 Received in revised form 12 June 2015 Accepted 13 July 2015 Available online 17 July 2015
Bioactive profiles of hazelnut skins belonging to fourteen hazelnut varieties were identified. Concentration of phenolic compounds, flavonoids and phenolic acids in soluble free, conjugated soluble and insoluble bound fractions together with their total concentrations were presented. In addition, tocopherol content and total antioxidant capacity of hazelnuts and their skins were revealed. Concentration of total phenolic compounds ranged between 51.9 and 203.1 mg gallic acid equivalent/g of skin among varieties, which is in accordance with the total antioxidant capacity. Total flavonoid content was almost 60% of the total phenolic compounds. Flavonoids and phenolic acids were found to be concentrated mostly in the conjugated soluble fraction. Tocopherol contents of hazelnut skins ranged from 226 to 593 mg/g, and a-tocopherol was the most abundant. Total antioxidant capacity was between 309 and 1375 mmol Trolox equivalent/g of hazelnut skins, which is more than 100 times higher than for hazelnuts without the skins. ß 2015 Elsevier Inc. All rights reserved.
Keywords: Hazelnut Hazelnut skin Hazelnut varieties Corylus avellana Phenolic compounds Antioxidant capacity Tocopherols Food analysis Food composition
1. Introduction Hazelnut (Corylus avellana L.), which belongs to the family of Betulaceae, is widely consumed (Pelvan et al., 2012). Turkey is the main hazelnut producer, with annual production of 549 000 tonnes, which is almost 65% of the total world production (FAO, 2013). There are 18 hazelnut varieties: Acı, Cavcava, C¸akıldak, Fos¸a, Ham, I˙ncekara, Kalınkara, Kan, Karafındık, Kargalak, Kus¸, Mincane, Palaz, Sivri, Tombul, Uzun Musa, Yassı Badem, Yuvarlak Badem, grown in Turkey (Pelvan et al., 2012). Among these varieties, Tombul contributes to 25–30% of Turkey’s total production (Alasalvar et al., 2003). The skin of hazelnut may be consumed directly intact with kernel or it could be easily removed after roasting. Hazelnut skin constitutes 2.5% of the total weight of the hazelnut (Alasalvar et al., 2009) and the skin may be considered an important constituent because of its distinctive phenolic profile and high antioxidant ¨ zdemir et al., activity (Contini et al., 2008; Pelvan et al., 2012; O 2014). Improvement in colon metabolism, decreasing in total and LDL-cholesterol, triacylglycerol and non-esterified fatty acids in blood serum could be attributed to the high phenolic content, phytosterols and dietary fibre in hazelnut skins (Caimari et al.,
* Corresponding author. Tel.: +90 312 2977108; fax: +90 312 2992123. E-mail address: [email protected] (V. Go¨kmen). http://dx.doi.org/10.1016/j.jfca.2015.07.003 0889-1575/ß 2015 Elsevier Inc. All rights reserved.
2015). Therefore, consumption of hazelnut with its brown skin is better from the point view of health. On the other hand, because brown skin removed after roasting is a major by-product of the hazelnut industry, its utilization is considered to be necessary. For example, defatted fine hazelnut skin particles were added to coffee before and after brewing to improve the physiologically positive effects of coffee phytochemicals (Contini et al., 2012). Five percent (w/w) of hazelnut skin was also used for fortification of fresh egg pasta (Zeppa et al., 2015) and bread (Anil, 2007) to increase antioxidant activity and dietary fibre content. Additionally, hazelnut skin showed prebiotic activity, allowing the growth of Lactobacillus crispatus P17613 and Lactobacillus plantarum P17630. Even 0.01% (w/v) of hazelnut skin acted as a cryoprotective ingredient during lyophilization, making it promising for application in probiotic-containing foods and nutraceuticals (Montella et al., 2013). Defatted hazelnut skin powder obtained after high shear homogenization was not only a good source of dietary fibre, antioxidant compounds and phenolic compounds but was also rich in colouring agents. Previous works investigated the total phenolic content, antioxidant activity and condensed tannins (Locatelli et al., 2010; Monagas et al., 2009; Alasalvar et al., 2009), dimeric B type procyanidins (Monagas et al., 2009; Esatbeyoglu et al., 2014), monomeric and oligomeric flavan-3-ol contents (Monagas et al., 2009; Del Rio et al., 2011) of roasted hazelnut skins. Shahidi et al. (2007) compared individual phenolic acids, radical-scavenging
204
N.G. Tas¸, V. Go¨kmen / Journal of Food Composition and Analysis 43 (2015) 203–208
activity and inhibition of human LDL cholesterol in extracts of whole hazelnut, roasted hazelnut skin and other by-products. Contents of free, bound and total phenolic acids, total phenolic content and antioxidant activity of natural and roasted commercial hazelnut varieties were reported (Pelvan et al., 2012). Despite the fact that hazelnut skins may be a good dietary source, an overall screening of bioactive compounds and their distribution in hazelnut skin is lacking in the literature. The objective of this study is to reveal the bioactive profile of natural hazelnut skins that belong to fourteen varieties. For that reason, analyses of phenolics and flavonoids (free soluble, conjugated soluble, bound), and phenolic acids (free soluble, conjugated soluble, bound) have been performed. In addition to that, total antioxidant activity and tocopherol profile of both hazelnuts and their skins have been measured. 2. Materials and methods 2.1. Chemicals and consumables Methanol, ethanol, acetonitrile, acetone, n-hexane, ethyl acetate, isopropanol, hexane and water were all HPLC-grade and were obtained from Sigma Aldrich (Steinheim, Germany). Standards of gallic acid, ferulic acid, catechin, potassium peroxy disulphate, ABTS (2,20 -azino-bis(3-ethyl-benzothiazoline-6-sulfonic acid)), Folin–Ciocalteu reagent, a-tocopherol (96%), b-tocopherol (96%), g-tocopherol (96%), d-tocopherol (90%) were also obtained from Sigma Aldrich (Steinheim, Germany). Formic acid (98%), hydrochloric acid (37%), sodium hydroxide, sodium carbonate, sodium nitrite, aluminium chloride and diethyl ether were purchased from Merck Co. (Darmstadt, Germany). Trolox [6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid] was purchased from Fluka Chemie AG (Buchs, Switzerland). Syringe filters (nylon, 0.45 mm) were supplied from Waters (Milford, MA). 2.2. Hazelnut and hazelnut skin Fourteen Turkish hazelnut (C. avellana L.) varieties (Acı, C¸akıldak, Fos¸a, I˙ncekara, Kalınkara, Kan, Kargalak, Kus¸, Palaz, Sivri, Tombul, Uzun Musa, Yassı Badem, Yuvarlak Badem), grown in the orchard of the Hazelnut Research Station in Giresun on the northeast coast of Turkey, were used for the analysis. Hazelnuts were left on the trees until the first week of August 2013; they were collected by hand when their green leafy covers became pale and moisture content decreased to 30%. They were then left in the sun until the moisture content became 6% for about 3–5 days. Pale leafy covers of hazelnuts were removed and three separate bags from each variety, each of them containing 2 kg hazelnuts, were immediately sent for analysis. Until analysis, hazelnuts were stored in their shells at –20 8C. Before analysis, hazelnut shells were cracked and hazelnut skin was removed manually by using a blade. About 2.5 g of hazelnut skin were obtained by scraping 100 g of hazelnut from each bag. Then, skin samples were finely ground using a coffee mill. Three independent hazelnut and hazelnut skin samples (n = 3) were analyzed with two analytical measurements. 2.3. Separation of phenolic fractions with alkaline hydrolysis Soluble free, soluble conjugated and insoluble bound phenolic compounds were extracted as described by Moore et al. (2005). First, 0.25 g of hazelnut skin were extracted with 4 mL of methanol/acetone/water (7:7:6, v/v/v) and centrifuged at 6000 g for 3 min. Extraction with 4 mL of methanol/acetone/ water (7:7:6, v/v/v) was repeated 5 times and supernatants of each step were collected and combined in a test tube (extract A). Pellet
obtained after this extraction was kept for the analysis of insoluble bound phenolic compounds. A quantity of 10 mL of the extract A was transferred to another tube and 10 mL of 4 N NaOH were added to it, in order to release soluble conjugated phenolic compounds (extract B). A quantity of 7 mL of 4 N NaOH was added to the pellet simultaneously to release insoluble bound phenolic compounds (extract C). Then, both tubes were left for 4 h in a shaker at the room temperature. After then, 5 mL of extract from A, B and C were transferred to test tubes and the pH was adjusted to 2 in all tubes, by using 6 N HCl. A quantity of 5 mL of diethyl ether:ethyl acetate (1:1, v/v) was added into these tubes, vortexed for 2 min and centrifuged at 6000 g for 2 min. The same extraction procedure with 5 mL of diethyl ether:ethyl acetate (1:1, v/v) was repeated four times. Supernatants were collected at each step and extracts were combined. Then, 5 mL of the combined extract were evaporated under an N2 stream at 30 8C to complete dryness. After then, phenolic compounds were redissolved in 1.5 mL of methanol and kept at 20 8C until the analyses were performed. As extract B contained both soluble free and soluble conjugated phenolic compounds, soluble conjugated phenolic compounds were calculated by subtracting the amount of soluble free phenolic compounds obtained from the analysis of extract A. 2.4. Analysis of total phenolic content Total phenolic content analyses were performed according to the Folin–Ciocalteu method (Hoff and Singleton, 1977). First, 25 mL of appropriate extracts were transferred into a test tube and 0.8 mL of 0.2 N Folin–Ciocalteu reagent were added to it. After 5 min of reaction time, the solution was neutralized with 0.8 mL of 20% Na2CO3 (w/v). The mixture was kept in the dark for 2 h until the characteristic blue colour was observed. After centrifugation at 6000 g for 4 min, absorbance of the supernatant was measured against methanol at 765 nm. Calibration curve was built with different concentrations of gallic acid dissolved in methanol and results were given as mg gallic acid equivalent (GAE) per g of hazelnut skin. 2.5. Analysis of individual phenolic acids Chromatographic analyses were performed on an Agilent 1200 HPLC system consisting of a diode array detector, quaternary pump, autosampler, and column oven (Agilent Technologies, Waldbronn, Germany). Before analysis, extracts were filtered through a 0.45-mm nylon syringe filters and taken into vials. Phenolic acids were separated on a Waters Atlantis C18 column (250 mm 4.6 mm id., 5 mm; Waters Corp., Milford, MA) by using 1% formic acid in water (A) and 1% formic acid in acetonitrile (B) at a flow rate of 1 mL/min with the following gradient programme: linear gradient elution from 10 to 20% B, 0–10 min; linear gradient elution from 20 to 40% B, 10–20 min; linear gradient elution from 40 to 10% B, 20–25 min and isocratic elution of 10% B, 25–30 min. The column temperature was 30 8C and injection volume was 10 mL. Analyses were performed by using diode array detector at 280 nm. Calibration curves were built for each of the compounds identified in the samples and phenolic acids were expressed as mg per g of hazelnut skin. 2.6. Analysis of total flavonoid content Total flavonoid content was determined according to the method described by Zhishen et al. (1999). First, 100 mL of appropriate extract were mixed with 50 mL 5% NaNO2. After waiting for 6 min of reaction time, 500 mL of 10% AlCl3 were added to the mixture to form aluminium-flavonoid complex. After 7 min,
N.G. Tas¸, V. Go¨kmen / Journal of Food Composition and Analysis 43 (2015) 203–208
250 mL of 1 N NaOH were added and the mixture was kept in the dark for 10 min. Then, the mixture was centrifuged at 6000 g for 5 min and the clear supernatant was measured against methanol at 510 nm. A calibration curve was built with different concentrations of catechin dissolved in methanol and results were given as mg catechin equivalent (CE) per g of hazelnut skin. 2.7. Analysis of total antioxidant capacity Total antioxidant capacity measurements were performed according to the QUENCHER method described by Serpen et al. (2009) by using 7 mM of ABTS aqueous solution with 2.45 mM potassium peroxy disulfate. The solution was kept in the dark for 12–16 h to obtain ABTS radical solution. ABTS working solution was prepared by diluting ABTS radical solution with ethanol–water (50:50, v/v) until the absorbance was 0.75–0.80 at 734 nm. Hazelnut skins were diluted 100 times with cellulose. A quantity of 20 mg of hazelnut and 10 mg of diluted hazelnut skin were weighed into test tubes and 10 mL of ABTS working solution was added. Tubes were placed in a shaker and left shaking in the dark. After centrifugation at 6000 g for 2 min, clear supernatants were measured at 734 nm exactly at the 30th minute of ABTS working solution addition onto samples. Trolox standard solutions were prepared in methanol between 0 and 600 mg/mL. Then, 100 mL of Trolox standard solutions and 10 mL ABTS working solutions were mixed and left 30 min in the dark. Absorbance of standard solutions was measured at 734 nm and a calibration curve was built. Antioxidant capacity of hazelnut and hazelnut skins were determined by using calibration curve and the results were expressed as mmol Trolox equivalent antioxidant capacity per g of hazelnut or hazelnut skin. 2.8. Analysis of tocopherols Hazelnut oil was extracted by hexane with Soxhlet extraction for 12 h after removal of their skins. Tocopherol analyses were performed both in hazelnut oils and skins. First, 0.5 g of hazelnut oil or hazelnut skin was weighed into test tubes and tocopherols were extracted with 2.5 mL of isopropanol–methanol (50:50, v/v) and centrifuged at 6000 g for 3 min. The procedure was repeated 3 times and supernatants were collected and combined. After passing through 0.45-mm filters, the extracts were collected in vials. Tocopherol analysis was performed using an Agilent LC/MS system containing a quaternary pump, temperature-controlled oven, Agilent 6130 MS detector and Agilent 1200 HPLC system (Agilent Technologies, Waldbronn, Germany). Separation of tocopherols was performed with an Agilent ZORBAX Rapid Resolution SB-C18 column (50 4.6 mm id., 3.5 mm; Waldbronn,
205
Germany) using mobile phases 1% formic acid in water (A) and methanol (B). An isocratic mixture of A and B (5:95, v/v) was used and the flow rate was 1 mL/min. Injection volume was 5 mL. Analyses were carried out in positive ionization mode. Monitored ions for a-tocopherol, b- and g-tocopherol, and d-tocopherol were m/z 431.7, 417.7, and 403.7, respectively. Standard solutions of each tocopherol were prepared at concentrations of 1, 5, 10, 20, 50 mg/L. Tocopherol contents of hazelnuts were calculated by using tocopherol contents of hazelnut oils and total fat contents of each variety. The data were expressed as mg tocopherol per g hazelnut or hazelnut skin (Yılmaz and Go¨kmen, 2013). 2.9. Statistical analysis All experiments were performed in triplicate for each hazelnut variety and all samples were analyzed with duplicate measurements. Experimental data were expressed as mean standard deviation. Significance of difference between varieties was analyzed by using one-way ANOVA and Duncan’s test (p < 0.05) by using SPSS Version 17.0. 3. Results and discussion 3.1. Soluble free, conjugated soluble, insoluble bound and total phenolic contents of hazelnut skins Soluble free, conjugated soluble, insoluble bound and total phenolic contents of hazelnut skins belong to fourteen hazelnut varieties are shown in Table 1. The average value for the concentration of soluble free phenolic compounds in the skins of fourteen hazelnut varieties was 17.3 mg GAE/g, 15% of the total phenolic compounds. Among hazelnut varieties, Yassı Badem had the lowest soluble free phenolic compounds with 7.1 mg GAE/g skin, while C¸akıldak had the highest with 30.6 mg GAE/g skin. Conjugated soluble phenolic compounds, comprised 74% of total phenolics with an average value of 84 mg GAE/g skin. Concentration of conjugated soluble phenolic compounds was lowest in Fos¸a with 30.1 mg GAE/g skin and highest in C¸akıldak with 155.2 mg GAE/g skin. Insoluble bound phenolics comprised 11% of the total phenolic compounds with an average value of 11.8 mg GAE/g skin. Insoluble bound phenolics were lowest in Acı with 7.6 mg GAE/g and highest in Uzun Musa with 29.4 mg GAE/g. Total phenolic compounds ranged between 51.9 and 203.1 mg GAE/g skin among the varieties. Previous works revealed the total phenolic content of roasted hazelnut skins as 233 mg CE/g skin of Tombul variety ¨ zdemir et al., 2014), or ranging between 41 and 127 mg (O polyphenols/g in roasted hazelnut skins of varieties from Turkey, Italy and Chile (Del Rio et al., 2011), and 182 mg CE/g in
Table 1 Concentration of soluble free, conjugated soluble, insoluble bound and total phenolic compounds of hazelnut skins (mg GAE/g)*. Soluble free phenolics Kargalak Palaz I˙ncekara Sivri Yassı Badem Fos¸a Kalınkara Yuvarlak Badem Kus¸ C¸akıldak Kan Uzun Musa Acı Tombul
d
18.8 4.9 14.4 1.5b,c,d 17.0 2.9d 15.2 5.0c,d 7.1 1.2a 11.1 0.1a,b,c 24.4 1.5e 9.8 0.7a,b 17.1 1.4d 30.6 0.6f 27.6 1.9e,f 16.8 0.4d 16.1 0.8c,d 15.7 1.3c,d
Conjugated soluble phenolics d,e
122.9 18.2 103.9 11.3d 73.2 11.2c 57.0 1.1b,c 45.8 0.4a,b 30.1 3.8a 67.0 6.5b,c 74.2 9.4c 107.3 7.0d,e 155.2 3.3f 44.7 8.2a,b 127.3 17.5e 54.2 8.3b,c 111.0 8.1d,e
Insoluble bound phenolics a
9.5 2.0 9.4 0.1a 11.8 2.5a,b 8.2 1.4a 8.0 0.4a 10.8 2.0a 9.1 1.6a 9.3 0.9a 7.7 0.8a 17.3 2.4c 10.9 0.2a 29.4 0.8d 7.6 1.7a 15.4 2.3b,c
Total phenolics 151.2 21.1f 127.7 9.7e 102.0 5.8d 80.4 2.4b,c,d 60.9 0.5a,b 51.9 2.0a 100.6 4.9c,d 93.4 9.2c,d 132.0 7.7e,f 203.1 1.5h 83.2 6.1b,c,d 173.5 17.1g 77.9 9.2b,c 142.2 11.7e,f
* Superscript letters in each column indicate statistically significant difference (p < 0.05) according to Duncan’s test. Three independent hazelnut skin sample (n = 3) were analyzed with two analytical measurements.
N.G. Tas¸, V. Go¨kmen / Journal of Food Composition and Analysis 43 (2015) 203–208
206
Table 2 Concentration of soluble free, conjugated soluble, insoluble bound and total flavonoid compounds of hazelnut skins (mg CE/g)*. Soluble free flavonoids Kargalak Palaz I˙ncekara Sivri Yassı Badem Fos¸a Kalınkara Yuvarlak Badem Kus¸ C¸akıldak Kan Uzun Musa Acı Tombul
c
3.4 0.0 2.6 0.8a,b,c 2.4 0.1a,b,c 2.0 0.2a,b 1.7 0.1a 2.0 0.5a,b 3.3 0.4c 2.6 0.5a,b,c 2.6 0.1a,b,c 4.5 0.2d 4.3 0.1d 2.8 0.6b,c 2.9 0.5b,c 2.3 0.5a,b
Conjugated soluble flavonoids c
Insoluble bound flavonoids a,b
52.4 4.1 48.1 2.8b,c 42.9 1.4a,b,c 35.5 0.4a,b,c 30.4 1.3a,b 25.5 1.0a 44.0 0.5b,c 41.4 2.4a,b,c 50.9 3.7c 82.4 2.3d 40.1 3.7a,b,c 81.7 3.4d 38.3 2.5a,b,c 49.4 1.3c
3.4 0.4 3.5 0.10a,b 4.2 0.40a,b 2.7 0.71a 2.7 0.50a 3.4 0.27a,b 3.3 0.27a,b 3.0 0.10a,b 2.7 0.41a 6.7 0.95b 5.6 0.10a,b 27.9 5.68c 2.9 0.18a,b 5.3 0.60a,b
Total flavonoids 59.2 3.73d 54.1 1.88c,d 49.5 1.25b,c,d 40.2 0.15a,b,c 34.8 1.91a,b 30.9 0.39a 50.7 1.25b,c,d 47.0 2.87a,b,c,d 56.2 4.08c,d 93.6 2.41e 50.0 3.75b,c,d 112.4 2.86f 44.0 3.14a,b,c,d 57.0 1.31c,d
* Superscript letters in each column indicate statistically significant difference (p < 0.05) according to Duncan’s test. Three independent hazelnut skin sample (n = 3) were analyzed with two analytical measurements.
medium-roasted hazelnut skin and 191 mg CE/g in high-roasted hazelnut skin (Locatelli et al., 2010). Pelvan et al. (2012) found 1.78–2.46 mg GAE/g total phenolic content in natural hazelnut varieties and suggested to consume hazelnuts with their skins as there was a significant loss in total phenolics, condensed tannins and free and bound phenolic acids after removal of skin. 3.2. Soluble free, conjugated soluble, insoluble bound and total flavonoid content of hazelnut skins Soluble free, conjugated soluble, insoluble bound and total flavonoid content of hazelnut skins is given in Table 2. Most of the flavonoids (85% of the total) were found in conjugated soluble form in hazelnut skin. In parallel with the results of conjugated phenolic compounds, Fos¸a was the variety containing least conjugated flavonoid with 25.5 mg CE/g and C¸akıldak was the highest with 82.4 mg CE/g. Insoluble bound flavonoids comprised 10% of the total flavonoids with an average 5.5 mg GAE/g skin. The least flavonoid-containing fraction was the soluble free fraction, with an average of 2.8 mg CE/g skin. Total flavonoid content of hazelnut skins ranged between 30.9 and 112.4 mg CE/g skin, which were almost 60% of the total phenolic compounds. Catechin, epicatechin, epicatechin gallate, gallocatechin, epigal¨ zdemir et al., 2014), procyanidin and locatechin gallate (O procyanidin dimers and trimers, and procyanidin dimer gallate were the flavan-3-ols identified in aqueous extracts of hazelnut skins (Del Rio et al., 2011). In addition, quercetin, myricetin,
quercetin-3-O-rhamnoside, myricetin rhamnoside, and kaempherol rhamnoside were the flavonols and phloretin 20 -O-glucoside was the dihydrochalcone detected in methanolic extracts of hazelnut skin (Del Rio et al., 2011). In this study, a combination of methanol/ acetone/water was used and alkaline hydrolysis performed. 3.3. Characterization of soluble free, conjugated soluble and insoluble bound phenolic acids of hazelnut skins Gallic acid and ferulic acid were the phenolic acids characterized in natural hazelnut skins (Table 3). Gallic acid was found in both soluble free and conjugated soluble fractions while ferulic acid was detected in conjugated soluble and insoluble fractions (Figure S1). Both phenolic acids were found to be dominant in the conjugated soluble fraction. Soluble free phenolic acids found in skin of Kargalak were lowest with 0.18 mg/g while Uzun Musa was highest with 0.5 mg/g skin although there was no significant difference between varieties (p > 0.05). Gallic acid found in conjugated form ranged between 0.5 and 3.7 mg/g skin. Ferulic acid in conjugated soluble form was approximately 8-fold higher than insoluble bound form. Ferulic acid ranged from 1.2 to 2.1 mg/g skin in conjugated form and 0.11 to 0.39 mg/g skin in insoluble bound form. Total phenolic acids determined in hazelnut skins were minimum of 3.0 and maximum of 8.3 mg/g skin. Pelvan et al. (2012) reported the total phenolic acid contents of natural hazelnuts ranged from 62.1 mg/g to 143.1 mg/g. ¨ zdemir et al. (2014) found 0.79 mg gallic acid/g skin in soluble free O fraction of roasted hazelnut skin.
Table 3 Concentration of soluble free, conjugated soluble, insoluble bound and total phenolic acids of hazelnut skins (mg/g)*.
Kargalak Palaz Incekara Sivri Yassı Badem Fos¸a Kalınkara Yuvarlak Badem Kus¸ C¸akıldak Kan Uzun Musa Acı Tombul
Soluble free phenolic acids
Conjugated soluble phenolic acids
Insoluble bound phenolic acids
Gallic acid
Ferulic acid
Gallic acid
Ferulic acid
0.18 0.03a 0.30 0.13a 0.33 0.08a 0.41 0.12a 0.28 0.15a 0.49 0.13a 0.50 0.13a 0.35 0.13a 0.34 0.10a 0.48 0.10a 0.41 0.03a 0.51 0.03a 0.22 0.03a 0.25 0.08a
2.1 0.01a 1.7 0.10a 1.8 0.12a 1.6 0.01a 1.8 0.10a 1.9 0.3a 1.5 0.17a 2.0 0.47a 2.0 0.25a 1.5 0.25a 2.2 0.23a 1.7 0.09a 2.0 0.39a 1.2 0.1a
3.7 0.25c,d 3.5 0.92c,d 1.7 0.05a,b,c 1.1 0.01a,b,c 0.7 0.04a,b 0.5 0.03a 1.6 0.07a,b,c 1.9 0.16a,b,c 2.7 0.40a,b,c 3.3 0.34b,c,d 0.6 0.18a,b 5.8 1.91d 1.5 0.10a,b,c 3.10 0.44a,b,c
0.20 0.03a,b 0.27 0.01a,b,c 0.22 0.09a,b,c 0.16 0.02a,b 0.22 0.01a,b,c 0.23 0.03a,b,c 0.18 0.01a,b 0.26 0.09a,b,c 0.19 0.01a,b 0.39 0.09c 0.14 0.03a,b 0.31 0.09b,c 0.11 0.02a 0.27 0.09a,b,c
Total phenolic acids
6.2 0.3a,b 5.8 1.2a,b 4.1 0.5a 3.3 0.2a 3.0 0.3a 3.1 0.5a 3.7 0.4a 4.5 0.9a 5.2 0.4a,b 5.6 0.4a,b 3.4 0.3a 8.3 1.8b 3.8 0.9a 4.9 0.4a
* Superscript letters in each column indicate statistically significant difference (p<0.05) according to Duncan’s test. Three independent hazelnut skin sample (n=3) were analyzed with two analytical measurements.
N.G. Tas¸, V. Go¨kmen / Journal of Food Composition and Analysis 43 (2015) 203–208
207
Table 4 Concentration of tocopherols in hazelnuts and hazelnut skins (mg/g)*. Hazelnut skin
Hazelnut
Kargalak Palaz I˙ncekara Sivri Yassı Badem Fos¸a Kalınkara Yuvarlak Badem Kus¸ C¸akıldak Kan Uzun Musa Acı Tombul
b + g Tocopherol
a-Tocopherol
Total tocopherols
b + g Tocopherol
a-Tocopherol
Total tocopherols
6.3 1.5a 21.6 3.1c,d,e,f,g 26.4 1.4e,f,g 25.1 0.1c,d,e,f,g 13.2 0.1a,b,c,d 12.0 0.7a,b,c 7.9 1.1a,b,c 12.9 5.3a,b 47.2 1.9a,b,c,d 29.5 1.5h 19.5 4.0b,c,d,e,f 25.9 2.3d,e,f,g 13.7 0.1a,b,c,d,e 33.5 0.3g
168.5 3.4c,d 129.4 10.0a,b 134.5 4.5a,b,c 145.8 0.4a,b,c,d 147.4 12.4a,b,c,d 155.4 7.5b,c,d 180.8 12.3d 158.0 4.0b,c,d 143.5 10.2a,b,c 136.1 0.6a,b,c 111.2 3.6a 155.9 14.9b,c,d 223.8 7.1f 248.3 1.7f
174.8 4.9a,b 151.0 13.1a,b 161.0 5.9a,b 170.9 0.5a.b 160.6 12.5a,b 167.4 8.2a,b 188.7 11.1b 170.8 9.3a,b 190.7 12.1b 165.5 2.1a,b 130.7 7.6a 181.8 17.2b 237.5 7.2c 281.7 2.0d
149.6 2.3d 150.8 1.1d 146.7 3.2c,d 197.7 0.9e 143.4 12.9c,d 75.9 5.8a,b 73.8 3.4a,b 79.9 2.4b 126.8 2.6c 146.5 11.4c,d 78.4 15.5a,b 57.9 0.5a 79.4 7.1a,b 89.1 0.9b
443.9 8.7h 250.0 0.4c,d,e 295.6 1.1d,e,f 357.5 19.1f,g 293.7 11.0d,e,f 372.0 9.4g 346.3 7.3f,g 236.7 22.5b,c,d 200.7 3.8a,b,c 256.6 39.2c,d,e 172.6 27.1a,b 168.2 3.8a 312.9 9.5e,f,g 207.0 0.6a,b,c
593.5 11.0f 400.7 0.8d,e 442.3 4.3e 555.2 18.2f 437.1 23.9e 447.9 15.2e 420.0 3.9e 316.6 24.9b,c 327.5 1.2b,c,d 403.1 50.6d,e 251.0 42.6a,b 226.1 3.2a 392.3 16.6c,d,e 296.1 1.6a,b
* Superscript letters in each column indicate statistically significant difference (p < 0.05) according to Duncan’s test. Three independent hazelnut skin and hazelnut sample (n = 3) were analyzed with two analytical measurements.
3.4. Tocopherol contents of hazelnuts and hazelnut skins Concentrations of (b + g)-tocopherol, a-tocopherol and total tocopherols found in both hazelnut varieties and their skins are given in Table 4. A total ion chromatogram of hazelnut skin and extracted ion chromatograms indicating a-tocopherol and (b + g)tocopherol are presented in Figure S2 in Supplementary Figures. a-Tocopherol was the most abundant tocopherol in both hazelnuts and hazelnut skin, ranging from 111.2 to 248.3 mg/g in hazelnuts. (b + g)-Tocopherol concentration of hazelnuts showed a large range between 6.3 and 47.2 mg/g. Total tocopherol content of hazelnuts was lowest in Kan with 130.7 mg/g and highest in Tombul with 281.7 mg/g. Ko¨ksal et al. (2006) identified a-, g- and d-tocopherols in natural hazelnut varieties and reported total tocopherol contents ranging between 196 and 414 mg/g. Total tocopherol content of Tombul was 513.1 mg/g and expectedly a-tocopherol constituted the highest with 404.0 mg/g (Alasalvar et al., 2006). Compared to the results of this study, higher tocopherol contents in hazelnuts were detected, as the skin was not removed in the other studies. In hazelnut skin, a-tocopherol was found to be ranging from 168.2 to 443.8 mg/g. (b + g)Tocopherol concentration of hazelnut skins was almost five times higher than hazelnuts and ranged from 57.9 to 197.7 mg/g skin. Total tocopherol content of hazelnut skins was lowest in Uzun Musa with 226.1 mg/g and highest in Kargalak with 593.5 mg/g; that was approximately two fold higher than hazelnuts. Tocopherol content of hazelnut skin oil was found to be 2770 mg/g ¨ zdemir et al., 2014) that was almost ten times higher than (O tocopherol contents of hazelnut skins detected in this study. 3.5. Total antioxidant capacity of hazelnuts and hazelnut skins Total antioxidant activity of hazelnut and hazelnut skins are expressed as Trolox equivalent antioxidant capacity in Table 5. Total antioxidant capacity of hazelnut varieties ranged from 5.4 to 8.8 mmol Trolox equivalent/g hazelnut. Total antioxidant capacity of hazelnut skins, with an average value of 878 mmol Trolox equivalent/g, was more than 100 times higher than hazelnuts. Among varieties, skin of Fos¸a was found to contain lowest amount of antioxidants with 309 mmol Trolox equivalent/g while skin of C¸akıldak had highest antioxidant capacity with 1375 mmol Trolox equivalent/g. Total antioxidant capacity of hazelnut skins was found to be in accordance with total phenolic compounds. Locatelli et al. (2010) found total antioxidant capacity of medium-roasted defatted skins as 1.10 mmol Trolox equivalent/g and 0.94 mmol Trolox equivalent/g for high-roasted defatted hazelnut skins.
Table 5 Total antioxidant capacity of hazelnuts and hazelnut skins (mmol Trolox equivalent/g)*.
Kargalak Palaz I˙ncekara Sivri Yassı Badem Fos¸a Kalınkara Yuvarlak Badem Kus¸ C¸akıldak Kan Uzun Musa Acı Tombul
Hazelnut
Hazelnut skin
7.8 0.5a,b,c 7.1 0.2a,b,c 7.5 1.5a,b,c 8.0 1.4b,c 6.5 1.1a,b,c 6.6 0.5a,b,c 7.0 0.3a,b,c 6.2 0.2a,b 8.3 1.1b,c 6.8 0.8a,b,c 5.4 0.5a 5.9 0.3a,b 7.7 1.5a,b,c 8.8 1.5c
1153 120g 887 3d,e,f 875 37d,e,f 775 139c,d,e 462 50a,b 309 14a 934 14e,f 706 17c,d 613 56b,c 1375 171h 952 117e,f 1343 233h 881 24d,e,f 1027 171f,g
* Superscript letters in each column indicate statistically significant difference (p < 0.05) according to Duncan’s test. Three independent hazelnut skin and hazelnut sample (n = 3) were analyzed with two analytical measurements.
Antioxidant capacity of hazelnut skin was 10 times higher than even from the most antioxidant-rich cereal, buckwheat (Serpen et al., 2008). In addition to that, antioxidant rich foods like cinnamon, dark chocolate (70% cocoa), blueberry, black filtered coffee and green tea were reported to have antioxidant capacities of 984, 134, 82, 28 and 25 mmol Trolox equivalent/g or mL, respectively (Blomhoff et al., 2006). Moreover, 1 g of the most antioxidant rich hazelnut skin, belonging to variety C¸akıldak, was equivalent to 1.4 g of cinnamon, 10 g of dark chocolate, 16.7 g of blueberry, 49.1 mL of black filtered coffee or 55 mL of green tea. However, these values could change among varieties as 1 g of the least antioxidant containing skin variety, Fos¸a, compensated 0.3 g of cinnamon, 2.3 g of dark chocolate, 3.8 g of blueberry, 11.0 mL of black filtered coffee or 12.4 mL green coffee. 4. Conclusion Hazelnut skin could be considered as a good source of bioactive compounds in comparison to its kernel. Bioactive profile and distribution of bioactive compounds in skin of hazelnut varieties was measured in this study. Phenolic compounds, flavonoids and phenolic acids were found to be rich in conjugated soluble fraction. Total phenolic compounds were lowest in skin of Fos¸a and highest in skin of C¸akıldak. Total flavonoid content of hazelnuts was 60% of total phenolic compounds. Moreover, hazelnut skin was found to
208
N.G. Tas¸, V. Go¨kmen / Journal of Food Composition and Analysis 43 (2015) 203–208
contain two times higher amount of tocopherols, especially a-tocopherol, than hazelnut itself. Total antioxidant capacity of hazelnut skin is much more higher (>100 times) than hazelnut and many other foods previously reported. Total antioxidant capacity was in accordance with total phenolic compounds; the skin of Fos¸a was the least antioxidant-containing variety and C¸akıldak was the highest. Although bioactive profile of hazelnut varieties covered a wide range, hazelnut skins of all varieties could have positive effects on health. Therefore, their consumption while intact with hazelnut kernel or as an ingredient of other foods should be taken into consideration. Acknowledgements This study was supported by The Scientific and Technological Research Council of Turkey (TUBITAK) in the frame of Intensified Cooperation (IntenC) Programme (Project no: 113O178). Authors are grateful to Hazelnut Research Station in Giresun for providing hazelnut varieties. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jfca.2015.07.003. References Alasalvar, C., Shahidi, F., Liyanapathirana, C.M., Oshima, T., 2003. Turkish Tombul hazelnut (Corylus avellana L.). 1. Compositional characteristics. J. Agric. Food Chem. 51, 3790–3796. Alasalvar, C., Amaral, J.S., Shahidi, F., 2006. Functional lipid characteristics of Turkish Tombul hazelnut (Corylus avellana L.). J. Agric. Food Chem. 54, 10177–10183. Alasalvar, C., Karamac, M., Kosinska, A., Rybarczyk, A., Shahidi, F., Amarowicz, R., 2009. Antioxidant activity of hazelnut skin phenolics. J. Agric. Food Chem. 57, 4645–4650. Anil, M., 2007. Using of hazelnut testa as a source of dietary fiber in breadmaking. J. Food Eng. 80, 61–67. Blomhoff, R., Carlsen, M.H., Andersen, L.F., Jacobs Jr., D.R., 2006. Health benefits of nuts: potential role of antioxidants. Brit. J. Nutr. 96, S52–S60. Caimari, A., Puiggros, F., Suarez, M., Crescenti, A., Laos, S., Ruiz, J.A., Alonso, V., Moragas, J., del Bas, J.M., Arola, L., 2015. The intake of a hazelnut skin extract improves the plasma lipid profile and reduces the lithocholic/ deoxycholic bile acid faecal ratio, a risk factor for colon cancer, in hamsters fed a high-fat diet. Food Chem. 167, 138–144. Contini, M., Baccelloni, S., Massantini, R., Anelli, G., 2008. Extraction of natural antioxidants from hazelnut (Corylus avellana L.) shell and skin wastes by long maceration at room temperature. Food Chem. 110, 659–669.
Contini, M., Bacelloni, S., Frangipane, M.T., Merendino, N., Massantini, R., 2012. Increasing espresso coffee brew antioxidant capacity using phenolic extract recovered from hazelnut skin waste. J. Functional Foods 4, 137–146. Del Rio, D., Calani, L., Dall’Asta, M., 2011. Polyphenolic composition of hazelnut skin. J. Agric. Food Chem. 59, 9935–9941. Esatbeyoglu, T., Juadjur, A., Wray, V., Winterhalter, P., 2014. Semisynthetic preparation and isolation of dimeric procyanidins B1B8 from roasted hazelnut skins (Corylus avellana L.) on a large scale using countercurrent chromatography. J. Agric. Food Chem. 62, 7101–7110. Food and Agriculture Organization of the United Nations (FAO). (2013). FAO Statistics Division. Retrieved April 3, 2015 from http://faostat3.fao.org/ download/Q/QC/E. Hoff, J.F., Singleton, K.I., 1977. A method for determination of tannin in foods by means of immobilized enzymes. J. Food Sci. 42, 1566–1569. Ko¨ksal, A.I., Artık, N., S¸ims¸ek, A., Gu¨nes¸, N., 2006. Nutrient composition of hazelnut (Corylus avellana L.) varieties cultivated in Turkey. Food Chem. 99, 509–515. Locatelli, M., Travaglia, F., Coisson, J.D., Martelli, A., Stevigny, C., Arlorio, M., 2010. Total antioxidant activity of hazelnut skin (Nocciola Piemonte PGI): impact of different roasting conditions. Food Chem. 119, 1647–1655. Monagas, M., Garrido, I., Lebron-Aguilar, R., Gomez-Cordoves, M.C., Raybarczyk, A., Amarowicz, R., Bartolome, B., 2009. Comparative flavan-3-ol profile and antioxidant capacity of roasted peanut, hazelnut, and almond skins. J. Agric. Food Chem. 57, 10590–10599. Montella, R., Coisson, J.D., Travaglia, F., Locatelli, M., Malfa, P., Martelli, A., Arlorio, M., 2013. Bioactive compounds from hazelnut skin (Corylus avellana L.): effects on Lactobacillus plantarum P17630 and Lactobacillus crispatus P17631. J. Functional Foods 5, 306–315. Moore, J., Hao, Z., Zhou, K., Luther, M., Costa, J., Yu, L.L., 2005. Carotenoid, tocopherol, phenolic acid, and antioxidant properties of Maryland-grown soft wheat. J. Agric. Food Chem. 53, 6649–6657. ¨ zdemir, K.S., Yılmaz, C., Durmaz, G., Go¨kmen, V., 2014. Hazelnut skin O powder: a new brown colored functional ingredient. Food Res. Int. 65, 291–297. Pelvan, E., Alasalvar, C., Uzman, S., 2012. Effects of roasting on the antioxidant status and phenolic profiles of commercial Turkish hazelnut varieties (Corylus avellana L.). J. Agric. Food Chem. 60, 1218–1223. Serpen, A., Go¨kmen, V., Pellegrini, N., Fogliano, V., 2008. Direct measurement of the total antioxidant capacity of cereal products. J. Cereal Sci. 48, 816–820. Serpen, A., Go¨kmen, V., Fogliano, V., 2009. Direct measurement of the total antioxidant capacity of foods: the ‘QUENCHER’ approach. Trends Food Sci. Technol. 20, 278–288. Shahidi, F., Alasalvar, C., Liyanapathirana, C.M., 2007. Antioxidant phytochemicals in hazelnut kernel (Corylus avellana L.) and hazelnut byproducts. J. Agric. Food Chem. 55, 1212–1220. Yılmaz, C., Go¨kmen, V., 2013. Compositional characteristics of sour cherry kernel and its oil as influenced by different extraction and roasting conditions. Ind. Crops Prod. 49, 130–135. Zeppa, G., Belviso, S., Bertolino, M., Cavallero, M.C., Dal Bello, B., Ghirardello, D., Giordano, M., Giorgis, M., Grosso, A., Rolle, L., Gerbi, V., 2015. The effect of hazelnut roasted skin from different cultivars on the quality attributes, polyphenol content and texture of fresh egg pasta. J. Sci. Food Agric. 95, 1678–1688. Zhishen, J., Mengcheng, T., Jianming, W., 1999. The determination of flavanoid contents in mulberry and their scavenging effects on superoxide radicals. Food Chem. 64, 555–559.