Food Chemistry 145 (2014) 8–14
Contents lists available at ScienceDirect
Food Chemistry journal homepage: www.elsevier.com/locate/foodchem
Phytochemical and physical–chemical analysis of Polish willow (Salix spp.) honey: Identification of the marker compounds Igor Jerkovic´ a,⇑, Piotr Marek Kus´ b, Carlo Ignazio Giovanni Tuberoso c, Mladenka Šarolic´ d a
Department of Organic Chemistry, Faculty of Chemistry and Technology, University of Split, N. Tesle 10/V, 21000 Split, Croatia Department of Pharmacognosy, Wrocław Medical University, ul. Borowska 211a, 50-556 Wrocław, Poland c Department of Life and Environmental Sciences, University of Cagliari, via Ospedale 72, 09124 Cagliari, Italy d Department of Food Techology, Marko Marulic´ Polytechnic in Knin, Petra Krešimira IV, 30, 22300 Knin, Croatia b
a r t i c l e
i n f o
Article history: Received 10 April 2013 Received in revised form 22 July 2013 Accepted 1 August 2013 Available online 9 August 2013 Keywords: Salix spp. honey GC-FID/MS HPLC-DAD Volatiles Monoterpenes C13-norisoprenoids Abscisic acid Antioxidant activity CIE L⁄a⁄b⁄C⁄h colour coordinates
a b s t r a c t The case study of Polish Salix spp. honey was compared with published data on willow honey from other regions. GC-FID/MS (after HS-SPME and ultrasonic solvent extraction) and targeted HPLC-DAD were applied. Phenolic content, FRAP/DPPH assays and the colour coordinates were determined spectrophotometrically. Beside ubiquitous linalool derivatives, borneol (up to 10.9%), bicyclic monoterpenes with pinane skeleton (pinocarvone up to 10.6%, myrtenal up to 4.8% and verbenone up to 3.4%) and trans-bdamascenone (up to 13.0%) dominated in the headspace. The main compounds of the extractives were vomifoliol (up to 39.6%) and methyl syringate (up to 16.5%) along with not common 4-hydroxy-3-(1methylethyl)benzaldehyde (up to 11.1%). Abscisic acid (ABA) was found (up to 53.7 mg/kg) with the isomeric ratio (Z,E)-ABA:(E,E)-ABA = 1:2. The honey exhibited low antioxidant potential with pale yellow colour. The composition of Polish willow honey is similar to Mediterranean willow honeys with several relevant differences. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Salix genus, commonly occurring in Europe, comprises a large number of species and includes trees and shrubs (Mabberley, 1997). Willow species flower from March until May apart from Salixamygdalina L. that may bloom even in summer (Farkas & Zajácz, 2007). They are very valuable as an early bee pasture, since they serve as abundant nectar and pollen source supporting spring development of bee colonies (Farkas & Zajácz, 2007). The willow honey is rare in Poland, where the climate is relatively cold and it may be collected only from Salix species blooming in late April and May. Unifloral Salix spp. honey has been analysed in several countries, however more detailed phytochemical composition of Polish willow honey has not yet been investigated. Tan, Wilkins, Holland, and McGhie (1990) reported GC–MS phytochemical profile of methylated extract from willow honey sample from New Zealand, identifying (E,Z)- and (E,E)-abscisic acid (ABA) as the marker compounds. De la Fuente, Sanz, Martínez-Castro, Sanz, and Ruiz-Matute (2007) investigated the headspace and sugar composition of
⇑ Corresponding author. Tel.: +385 21 329434; fax: +385 21 329461. E-mail address: [email protected] (I. Jerkovic´). 0308-8146/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodchem.2013.08.004
willow honey sample from Spain, and phenylacetaldehyde, benzaldehyde, methyl salicylate, 2,6,6-trimethylcyclohepta-2,4-dien-1one as well as benzyl alcohol were found to be the main volatile compounds. Salix spp. honey from Croatia was analysed, in our previous studies (Jerkovic´ & Marijanovic´, 2010; Tuberoso, Jerkovic´, Bifulco, & Marijanovic´, 2011), by GC-FID/MS and HPLC-DAD. HPLC analysis confirmed (Z,E)-/(E,E)-ABA as the marker compounds of this honey type. Phenylacetic acid, pinocembrin, 8-hydroxy-4,7dimethylcoumarin and 3-hydroxy-trans-b-damascone were relevant volatile compounds in Croatian Salix spp. honey extracts, while safranal and lilac alcohols were relevant in the headspace. Baltrušaityte˙, Venskutonis, and Cˇeksteryte˙ (2007) investigated antioxidant activity (DPPH and ABTS assays) along with the content of several phenolics in willow and other honeys from Lithuania. The phytochemical composition of willow honey from different regions may vary, due to occurrence of different Salix species as well as the availability of other accompanying species for the bees. Additionally, it was already suggested (Kaškoniene˙ & Venskutonis, 2010) that the honey volatiles, even of the same floral origin, may vary in different regions. Therefore, as a continuation of previous investigations, the aims of present case-study were: (i) to investigate phytochemical composition of Polish Salix spp. honey applying GC-FID/MS and targeted HPLC-DAD, (ii) to determine its
I. Jerkovic´ et al. / Food Chemistry 145 (2014) 8–14
physical–chemical parameters (total phenols, antioxidant potential and colour) and (iii) to compare the obtained result with published characteristics of willow honey from other regions. Special emphasis was put on identifying marker compounds characteristic for Polish Salix spp. honey in comparison with already found chemical markers (e.g. ABA) of willow honey from other regions.
2. Experimental 2.1. Reagents and honey samples The solvents (diethyl ether, dichloromethane and pentane (Kemika, Zagreb, Croatia)) were analytically pure and redistilled before usage. The manual SPME fibre with the layer of divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/PDMS) was obtained from Supelco Co. (Bellefonte, PA, USA). Acetonitrile, anhydrous natrium sulphate, 1,1-diphenyl-2-picrylhydrazyl radical (DPPH), Folin–Ciocalteau’s reactive, gallic acid, benzoic acid, p-coumaric acid, 4-hydroxybenzoic acid, kynurenic acid, methylbenzaldehyde, methyl syringate, vanillic acid (±)-(2Z,4E)-abscisic acid, (±)-6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), methanol, phosphoric acid 85% w/w, and 2,4,6-tris(2-pyridyl)-1,3,5-triazine (TPTZ) were purchased from Sigma–Aldrich, Fluka (Milan, Italy). Standard of (±)-(2E,4E)-abscisic acid was purchased from A. G. Scientific Inc. (San Diego, CA). Ultrapure water (18 mX) was obtained using Milli-Q Advantage A10 System (Millipore, Milan, Italy). This study was carried out on selected three Salix spp. honey samples obtained from professional beekeepers from different parts of Poland during spring (Meret, Brat, Mertz, Lebrun, & Günata, 2011). All the samples, after acquisition from beekeepers, were stored in hermetically closed glass jars in dark at 4 °C. The samples were subjected to melissopalynological analysis according to the methodology recommended by the International Commission of Bee Botany and the International Honey Commission (Louveaux, Maurizio, & Vorwohl, 1978). 2.2. Ultrasonic solvent extraction (USE) Ultrasonic solvent extraction (USE) was performed in an ultrasound bath (Elmasonic Typ S 30 H, Germany) and indirect sonication mode at a frequency of 37 kHz at 25 ± 3 °C was applied. The extraction was performed as previously described (Jerkovic´, Marijanovic´, & Malenica-Staver, 2011). Briefly, forty grams of each sample were used for triplicate extraction with a mixture of pentane with diethyl ether (1:2, v/v, solvent A) and similar independent extraction was performed with dichloromethane (solvent B). The obtained extracts were concentrated to 0.2 mL through distillation with use of Kuderna-Danish concentrator and 1 lL was used for GC-FID and GC–MS analyses. 2.3. Headspace solid-phase microextraction (HS SPME) HS-SPME was carried out using a SPME fibre coated with layer of divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/ PDMS). Before use, the fibre was conditioned according to the manufacturer instructions. The isolation of headspace volatile compounds were carried out from honey/saturated water solution (5 mL, 1:1 v/v; saturated with NaCl) placed in 15 mL glass vial with PTFE/silicone septa and a stirring bar. The vials and septa were previously heated, at 150 °C for 24 h, to remove possible contaminants that may interfere. To allow equilibration, the samples were maintained at 60 °C and magnetically stirred for 1 h. The SPME fibre was immersed through the septum and exposed to the headspace for
9
40 min. The fibre was then transferred to the GC injector and desorbed for 6 min. 2.4. GC-FID/MS analysis GC-FID analyses were carried out on an Agilent gas chromatograph model 7890A (Agilent Technologies, Palo Alto, CA, USA) equipped with flame ionisation detector. Chromatographic separations were performed on 30 m x 0.25 mm i.d. coating thickness 0.25 lm capillary column HP-5MS ((5%-phenyl)-methylpolysiloxane, Agilent J & W GC column). The oven was temperature-programmed isothermal from 70 °C for 2 min, then increased to 200 °C, at a rate of 3 °C/min and held isothermal for 15 min. Helium at 1 mL/min was used as carrier gas. Injector temperature was set at 250 °C and detector temperature was 300 °C. The analyses by gas chromatography–mass spectrometry were carried out with an Agilent gas chromatograph model 7890A fitted with a mass selective detector model 5975C (Agilent Technologies, Palo Alto, CA, USA). Mass detector worked in the electron impact ionisation mode at 70 eV, the mass range was m/z 30–300 and ion source temperature 280 °C. Volatile compounds separation was obtained using the same column and oven temperature program as described above for GC-FID. The individual peaks were identified by comparison of their retention indices (relative to C9–C25 n-alkanes) to those of available authentic samples and literature (El-Sayed, 2012), as well as by comparing their mass spectra with Wiley 275 MS library (Wiley, New York, USA) and NIST98 (Gaithersburg, Germany) mass spectral databases. The percentage composition of the samples was calculated from the GC peak areas using the normalisation method (without correction factors). 2.5. HPLC-DAD analysis The analyses were performed according to Tuberoso et al. (2011) using HPLC-DAD Varian system ProStar fitted with a pump module 230, an autosampler module 410, and a ThermoSeparation diode array detector SpectroSystem UV 6000lp (ThermoSeparation, San Jose, CA, USA) set at 280 nm. The separation was obtained with a Phenomenex Gemini C18 110A column (150 mm 4.60 mm, 3 lm, Chemtek Analitica, Anzola Emilia, Bologna, Italy) using 0.2 M phosphoric acid (solvent A) and acetonitrile (solvent B) as mobile phase at a constant flow rate of 1.0 mL/min. The gradient (v/v) was generated, keeping 90% of solvent A for 5 min, then decreasing to 65% in 15 min, to 10% in 20 min, and remaining at this concentration for 10 min. The system was stabilized for 10 min with the initial A/B ratio (90:10, v/v) before each injection. The injection volume was 10 lL. The obtained chromatograms and spectra were elaborated with a ChromQuest V. 4.0 software (ThermoQuest, Rodano, Milan, Italy). Abscisic acid isomers along with other seven compounds were identified by both co-chromatography with standard compound and UV–Vis spectra evaluation. Standard solutions were prepared in methanol, and the working standard solutions in ultrapure water. Calibration curves were built with the method of external standard, correlating the area of the peaks vs. the concentration. The correlation values were 0.9989–0.9999 in the range of 0.2–20 mg/L. The honey samples were carefully diluted with ultrapure water (1:5, w/v), filtered through Econofilter RC membrane (0.45 lm, Ø 25 mm, Agilent Technologies, Milan, Italy), and injected in HPLC without any further purification. 2.6. CIE L*a*b*C*h chromatic coordinates The chromatic coordinates were measured with an UV/VIS spectrophotometer Varian series Cary 50 Scan (Varian, Leinì, TO, Italy), and data were processed using Cary Win UV Colour
10
I. Jerkovic´ et al. / Food Chemistry 145 (2014) 8–14
Application V. 2.00. Transmittances within 380–780 nm wavelength interval were measured using D65 illuminant with a 10° observation angle. The perfectly fluid and transparent honey samples liquefied in ultrasonic bath (t < 40 °C) were analysed in 10 mm optical polystyrene cuvettes without any dilution. 2.7. Total phenols Total phenols content was determined spectrophotometrically according to modified Folin–Ciocalteu method (Tuberoso et al., 2011). 100 lL of the honey diluted with ultrapure water 1:5 (w/ v)was added to 0.5 mL of Folin–Ciocalteu’s reactive. After 5 min, 3 mL of 10% Na2CO3 (w/v) was added, and the mixture was shaken and diluted with H2O to a final volume of 10 mL. After 90 min of incubation at room temperature, the absorbance was read at 725 nm in 10 mm quartz cuvette with Varian Cary 50 spectrophotometer against a blank. The total polyphenols content was obtained using a calibration curve prepared using fresh gallic acid standard solutions (10–500 mg/L) and expressed as mg/kg of gallic acid equivalent (GAE). 2.8. Antiradical activity (DPPH test) The spectrophotometric analysis was performed (Tuberoso et al., 2011). 50 lL of the honey diluted with ultrapure water (1:5 w/v) was dissolved in 2 mL of DPPH solution (0.04 mmol/L in MeOH). The calibration curve was prepared for Trolox, and data were expressed as Trolox equivalent antioxidant capacity (TEAC mmol/kg). The spectrophotometric measurments were carried out with Varian Cary 50 spectrophotometer at 517 nm using a 10 mm quartz cuvette. 2.9. Total antioxidant activity (FRAP test) Ferric reducing antioxidant assay (FRAP) was performed as previously (Tuberoso et al., 2011). A ferric complex of TPTZ and Fe3+ was prepared (0.3123 g TPTZ, 0.5406 g FeCl36H2O in 100 mL acetate buffer pH 3.6). 20 mL of the honey diluted with ultrapure water (1:5 w/v) were dissolved in 2 lL of ferric complex. Quantitative analysis was performed according to the external standard method (FeSO4, 0.1–2 mmol) correlating the absorbance (k = 593 nm) with the concentration. The results were expressed as mmol/kg of Fe2+. 2.10. Water content The water content of honey was determined with a portable refractometer that enables reading of the percentage of water within range of 12–26% (ATAGO Hand Refractometer Honey, Atago Co. Ltd., Tokyo, Japan). 3. Results and discussion The season of Salix spp. honey collection in Poland is mainly characterised by S. alba L., S. fragilis L., S. purpurea L. and S. amygdalina L. blooming in temperate climate in middle and late spring (Farkas & Zajácz, 2007). The melissopalynological analyses of the samples resulted with characteristic Salix spp. pollen content up to 66%. Other secondary pollen types, found with minor contribution, were Brassica napus L., Prunus spp., Acer spp. and Taraxacum spp. The phytochemical study of the headspace and extractives of Salix spp. honey revealed 26 and 47 compounds respectively. Targeted HPLC was focused on 9 compounds, especially on ABA isomers that were previously indicated as markers of willow honey (Tan et al., 1990; Tuberoso et al., 2011). Other general
physical–chemical characteristics were determined as well. Geographical origin and applied isolation methodologies may impact phytochemical composition and physical–chemical properties of the honey (Kaškoniene˙ & Venskutonis, 2010). Therefore, observed differences/similarities with published data on willow honey from other regions were further discussed. The comparison of Polish Salix spp. honey extracts with Croatian willow honey extracts (Jerkovic´ & Marijanovic´, 2010) is more reliable since the same experimental methodology was applied. 3.1. Phytochemicals extracted by HS-SPME The major constituents dominating the headspace of Polish willow honey were monoterpenes, mainly linalool derivatives such as lilac aldehyde isomers (8.0–36.9%), hotrienol (5.4–16.6%), cis-/ trans-linalool oxides (6.3–11.5%) along with terpinene-4-ol (2.4–3.1%) and several similar compounds, Table 1. Monoterpenes with bornane skeleton (borneol (5.0–10.9%)) and bicyclic pinane skeleton (pinocarvone (3.6–10.6%), myrtenal (1.7–4.8%) as well as verbenone (2.3–3.4%)) were other relevant compounds, not very common in other honey types. The headspace also contained relevant percentage of C13-norisoprenoid trans-b-damascenone (8.3–13.0%). Due to predominant monoterpenes abundance in the headspace, their biosynthetic relationships were further discussed taking into consideration uncommon skeletons in the honeys and the corresponding pathways as well as favored oxidative conditions within the hive (about 30 °C and acidic pH; Alissandrakis, Tarantilis, Harizanis, & Polissiou, 2007). Terpenes in general derive from geranyl pyrophosphate (GPP) that can rearrange to lynalyl pyrophosphate (LPP) and neryl pyrophosphate (NPP). After the elimination of pyrophosphate group (OPP ) from NPP, 1,6-cyclisation (Degenhardt, Köllner, & Gershenzon, 2009; Maffei, Gertsch, &
Table 1 Volatile compounds identified by GC-FID/GC-MS in headspace of Salix spp. honey samples after headspace solid-phase microextraction (HS-SPME). No.
Compounds
RI
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Dimethyl sulfide 2-Methylpropanenitrile 3-Methylbutanal Isoamyl alcohol 3-Methylpentanal 3-Methylpentanoic acid Benzaldehyde 1,3-Dimethyl-2-ethylbenzenea cis-Linalool oxide trans-Linalool oxide Linalool Hotrienol a-Isophorone Benzyl alcohol 4-Ketoisophorone Lilac aldehyde (isomer I) Lilac aldehyde (isomer I) Pinocarvone Lilac aldehyde (isomer II) Borneol Terpinene-4-ol a-Terpineol Myrtenal Lilac alcohola Verbenone trans-b-Damascenone
Contents lists available at ScienceDirect
Food Chemistry journal homepage: www.elsevier.com/locate/foodchem
Phytochemical and physical–chemical analysis of Polish willow (Salix spp.) honey: Identification of the marker compounds Igor Jerkovic´ a,⇑, Piotr Marek Kus´ b, Carlo Ignazio Giovanni Tuberoso c, Mladenka Šarolic´ d a
Department of Organic Chemistry, Faculty of Chemistry and Technology, University of Split, N. Tesle 10/V, 21000 Split, Croatia Department of Pharmacognosy, Wrocław Medical University, ul. Borowska 211a, 50-556 Wrocław, Poland c Department of Life and Environmental Sciences, University of Cagliari, via Ospedale 72, 09124 Cagliari, Italy d Department of Food Techology, Marko Marulic´ Polytechnic in Knin, Petra Krešimira IV, 30, 22300 Knin, Croatia b
a r t i c l e
i n f o
Article history: Received 10 April 2013 Received in revised form 22 July 2013 Accepted 1 August 2013 Available online 9 August 2013 Keywords: Salix spp. honey GC-FID/MS HPLC-DAD Volatiles Monoterpenes C13-norisoprenoids Abscisic acid Antioxidant activity CIE L⁄a⁄b⁄C⁄h colour coordinates
a b s t r a c t The case study of Polish Salix spp. honey was compared with published data on willow honey from other regions. GC-FID/MS (after HS-SPME and ultrasonic solvent extraction) and targeted HPLC-DAD were applied. Phenolic content, FRAP/DPPH assays and the colour coordinates were determined spectrophotometrically. Beside ubiquitous linalool derivatives, borneol (up to 10.9%), bicyclic monoterpenes with pinane skeleton (pinocarvone up to 10.6%, myrtenal up to 4.8% and verbenone up to 3.4%) and trans-bdamascenone (up to 13.0%) dominated in the headspace. The main compounds of the extractives were vomifoliol (up to 39.6%) and methyl syringate (up to 16.5%) along with not common 4-hydroxy-3-(1methylethyl)benzaldehyde (up to 11.1%). Abscisic acid (ABA) was found (up to 53.7 mg/kg) with the isomeric ratio (Z,E)-ABA:(E,E)-ABA = 1:2. The honey exhibited low antioxidant potential with pale yellow colour. The composition of Polish willow honey is similar to Mediterranean willow honeys with several relevant differences. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Salix genus, commonly occurring in Europe, comprises a large number of species and includes trees and shrubs (Mabberley, 1997). Willow species flower from March until May apart from Salixamygdalina L. that may bloom even in summer (Farkas & Zajácz, 2007). They are very valuable as an early bee pasture, since they serve as abundant nectar and pollen source supporting spring development of bee colonies (Farkas & Zajácz, 2007). The willow honey is rare in Poland, where the climate is relatively cold and it may be collected only from Salix species blooming in late April and May. Unifloral Salix spp. honey has been analysed in several countries, however more detailed phytochemical composition of Polish willow honey has not yet been investigated. Tan, Wilkins, Holland, and McGhie (1990) reported GC–MS phytochemical profile of methylated extract from willow honey sample from New Zealand, identifying (E,Z)- and (E,E)-abscisic acid (ABA) as the marker compounds. De la Fuente, Sanz, Martínez-Castro, Sanz, and Ruiz-Matute (2007) investigated the headspace and sugar composition of
⇑ Corresponding author. Tel.: +385 21 329434; fax: +385 21 329461. E-mail address: [email protected] (I. Jerkovic´). 0308-8146/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodchem.2013.08.004
willow honey sample from Spain, and phenylacetaldehyde, benzaldehyde, methyl salicylate, 2,6,6-trimethylcyclohepta-2,4-dien-1one as well as benzyl alcohol were found to be the main volatile compounds. Salix spp. honey from Croatia was analysed, in our previous studies (Jerkovic´ & Marijanovic´, 2010; Tuberoso, Jerkovic´, Bifulco, & Marijanovic´, 2011), by GC-FID/MS and HPLC-DAD. HPLC analysis confirmed (Z,E)-/(E,E)-ABA as the marker compounds of this honey type. Phenylacetic acid, pinocembrin, 8-hydroxy-4,7dimethylcoumarin and 3-hydroxy-trans-b-damascone were relevant volatile compounds in Croatian Salix spp. honey extracts, while safranal and lilac alcohols were relevant in the headspace. Baltrušaityte˙, Venskutonis, and Cˇeksteryte˙ (2007) investigated antioxidant activity (DPPH and ABTS assays) along with the content of several phenolics in willow and other honeys from Lithuania. The phytochemical composition of willow honey from different regions may vary, due to occurrence of different Salix species as well as the availability of other accompanying species for the bees. Additionally, it was already suggested (Kaškoniene˙ & Venskutonis, 2010) that the honey volatiles, even of the same floral origin, may vary in different regions. Therefore, as a continuation of previous investigations, the aims of present case-study were: (i) to investigate phytochemical composition of Polish Salix spp. honey applying GC-FID/MS and targeted HPLC-DAD, (ii) to determine its
I. Jerkovic´ et al. / Food Chemistry 145 (2014) 8–14
physical–chemical parameters (total phenols, antioxidant potential and colour) and (iii) to compare the obtained result with published characteristics of willow honey from other regions. Special emphasis was put on identifying marker compounds characteristic for Polish Salix spp. honey in comparison with already found chemical markers (e.g. ABA) of willow honey from other regions.
2. Experimental 2.1. Reagents and honey samples The solvents (diethyl ether, dichloromethane and pentane (Kemika, Zagreb, Croatia)) were analytically pure and redistilled before usage. The manual SPME fibre with the layer of divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/PDMS) was obtained from Supelco Co. (Bellefonte, PA, USA). Acetonitrile, anhydrous natrium sulphate, 1,1-diphenyl-2-picrylhydrazyl radical (DPPH), Folin–Ciocalteau’s reactive, gallic acid, benzoic acid, p-coumaric acid, 4-hydroxybenzoic acid, kynurenic acid, methylbenzaldehyde, methyl syringate, vanillic acid (±)-(2Z,4E)-abscisic acid, (±)-6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), methanol, phosphoric acid 85% w/w, and 2,4,6-tris(2-pyridyl)-1,3,5-triazine (TPTZ) were purchased from Sigma–Aldrich, Fluka (Milan, Italy). Standard of (±)-(2E,4E)-abscisic acid was purchased from A. G. Scientific Inc. (San Diego, CA). Ultrapure water (18 mX) was obtained using Milli-Q Advantage A10 System (Millipore, Milan, Italy). This study was carried out on selected three Salix spp. honey samples obtained from professional beekeepers from different parts of Poland during spring (Meret, Brat, Mertz, Lebrun, & Günata, 2011). All the samples, after acquisition from beekeepers, were stored in hermetically closed glass jars in dark at 4 °C. The samples were subjected to melissopalynological analysis according to the methodology recommended by the International Commission of Bee Botany and the International Honey Commission (Louveaux, Maurizio, & Vorwohl, 1978). 2.2. Ultrasonic solvent extraction (USE) Ultrasonic solvent extraction (USE) was performed in an ultrasound bath (Elmasonic Typ S 30 H, Germany) and indirect sonication mode at a frequency of 37 kHz at 25 ± 3 °C was applied. The extraction was performed as previously described (Jerkovic´, Marijanovic´, & Malenica-Staver, 2011). Briefly, forty grams of each sample were used for triplicate extraction with a mixture of pentane with diethyl ether (1:2, v/v, solvent A) and similar independent extraction was performed with dichloromethane (solvent B). The obtained extracts were concentrated to 0.2 mL through distillation with use of Kuderna-Danish concentrator and 1 lL was used for GC-FID and GC–MS analyses. 2.3. Headspace solid-phase microextraction (HS SPME) HS-SPME was carried out using a SPME fibre coated with layer of divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/ PDMS). Before use, the fibre was conditioned according to the manufacturer instructions. The isolation of headspace volatile compounds were carried out from honey/saturated water solution (5 mL, 1:1 v/v; saturated with NaCl) placed in 15 mL glass vial with PTFE/silicone septa and a stirring bar. The vials and septa were previously heated, at 150 °C for 24 h, to remove possible contaminants that may interfere. To allow equilibration, the samples were maintained at 60 °C and magnetically stirred for 1 h. The SPME fibre was immersed through the septum and exposed to the headspace for
9
40 min. The fibre was then transferred to the GC injector and desorbed for 6 min. 2.4. GC-FID/MS analysis GC-FID analyses were carried out on an Agilent gas chromatograph model 7890A (Agilent Technologies, Palo Alto, CA, USA) equipped with flame ionisation detector. Chromatographic separations were performed on 30 m x 0.25 mm i.d. coating thickness 0.25 lm capillary column HP-5MS ((5%-phenyl)-methylpolysiloxane, Agilent J & W GC column). The oven was temperature-programmed isothermal from 70 °C for 2 min, then increased to 200 °C, at a rate of 3 °C/min and held isothermal for 15 min. Helium at 1 mL/min was used as carrier gas. Injector temperature was set at 250 °C and detector temperature was 300 °C. The analyses by gas chromatography–mass spectrometry were carried out with an Agilent gas chromatograph model 7890A fitted with a mass selective detector model 5975C (Agilent Technologies, Palo Alto, CA, USA). Mass detector worked in the electron impact ionisation mode at 70 eV, the mass range was m/z 30–300 and ion source temperature 280 °C. Volatile compounds separation was obtained using the same column and oven temperature program as described above for GC-FID. The individual peaks were identified by comparison of their retention indices (relative to C9–C25 n-alkanes) to those of available authentic samples and literature (El-Sayed, 2012), as well as by comparing their mass spectra with Wiley 275 MS library (Wiley, New York, USA) and NIST98 (Gaithersburg, Germany) mass spectral databases. The percentage composition of the samples was calculated from the GC peak areas using the normalisation method (without correction factors). 2.5. HPLC-DAD analysis The analyses were performed according to Tuberoso et al. (2011) using HPLC-DAD Varian system ProStar fitted with a pump module 230, an autosampler module 410, and a ThermoSeparation diode array detector SpectroSystem UV 6000lp (ThermoSeparation, San Jose, CA, USA) set at 280 nm. The separation was obtained with a Phenomenex Gemini C18 110A column (150 mm 4.60 mm, 3 lm, Chemtek Analitica, Anzola Emilia, Bologna, Italy) using 0.2 M phosphoric acid (solvent A) and acetonitrile (solvent B) as mobile phase at a constant flow rate of 1.0 mL/min. The gradient (v/v) was generated, keeping 90% of solvent A for 5 min, then decreasing to 65% in 15 min, to 10% in 20 min, and remaining at this concentration for 10 min. The system was stabilized for 10 min with the initial A/B ratio (90:10, v/v) before each injection. The injection volume was 10 lL. The obtained chromatograms and spectra were elaborated with a ChromQuest V. 4.0 software (ThermoQuest, Rodano, Milan, Italy). Abscisic acid isomers along with other seven compounds were identified by both co-chromatography with standard compound and UV–Vis spectra evaluation. Standard solutions were prepared in methanol, and the working standard solutions in ultrapure water. Calibration curves were built with the method of external standard, correlating the area of the peaks vs. the concentration. The correlation values were 0.9989–0.9999 in the range of 0.2–20 mg/L. The honey samples were carefully diluted with ultrapure water (1:5, w/v), filtered through Econofilter RC membrane (0.45 lm, Ø 25 mm, Agilent Technologies, Milan, Italy), and injected in HPLC without any further purification. 2.6. CIE L*a*b*C*h chromatic coordinates The chromatic coordinates were measured with an UV/VIS spectrophotometer Varian series Cary 50 Scan (Varian, Leinì, TO, Italy), and data were processed using Cary Win UV Colour
10
I. Jerkovic´ et al. / Food Chemistry 145 (2014) 8–14
Application V. 2.00. Transmittances within 380–780 nm wavelength interval were measured using D65 illuminant with a 10° observation angle. The perfectly fluid and transparent honey samples liquefied in ultrasonic bath (t < 40 °C) were analysed in 10 mm optical polystyrene cuvettes without any dilution. 2.7. Total phenols Total phenols content was determined spectrophotometrically according to modified Folin–Ciocalteu method (Tuberoso et al., 2011). 100 lL of the honey diluted with ultrapure water 1:5 (w/ v)was added to 0.5 mL of Folin–Ciocalteu’s reactive. After 5 min, 3 mL of 10% Na2CO3 (w/v) was added, and the mixture was shaken and diluted with H2O to a final volume of 10 mL. After 90 min of incubation at room temperature, the absorbance was read at 725 nm in 10 mm quartz cuvette with Varian Cary 50 spectrophotometer against a blank. The total polyphenols content was obtained using a calibration curve prepared using fresh gallic acid standard solutions (10–500 mg/L) and expressed as mg/kg of gallic acid equivalent (GAE). 2.8. Antiradical activity (DPPH test) The spectrophotometric analysis was performed (Tuberoso et al., 2011). 50 lL of the honey diluted with ultrapure water (1:5 w/v) was dissolved in 2 mL of DPPH solution (0.04 mmol/L in MeOH). The calibration curve was prepared for Trolox, and data were expressed as Trolox equivalent antioxidant capacity (TEAC mmol/kg). The spectrophotometric measurments were carried out with Varian Cary 50 spectrophotometer at 517 nm using a 10 mm quartz cuvette. 2.9. Total antioxidant activity (FRAP test) Ferric reducing antioxidant assay (FRAP) was performed as previously (Tuberoso et al., 2011). A ferric complex of TPTZ and Fe3+ was prepared (0.3123 g TPTZ, 0.5406 g FeCl36H2O in 100 mL acetate buffer pH 3.6). 20 mL of the honey diluted with ultrapure water (1:5 w/v) were dissolved in 2 lL of ferric complex. Quantitative analysis was performed according to the external standard method (FeSO4, 0.1–2 mmol) correlating the absorbance (k = 593 nm) with the concentration. The results were expressed as mmol/kg of Fe2+. 2.10. Water content The water content of honey was determined with a portable refractometer that enables reading of the percentage of water within range of 12–26% (ATAGO Hand Refractometer Honey, Atago Co. Ltd., Tokyo, Japan). 3. Results and discussion The season of Salix spp. honey collection in Poland is mainly characterised by S. alba L., S. fragilis L., S. purpurea L. and S. amygdalina L. blooming in temperate climate in middle and late spring (Farkas & Zajácz, 2007). The melissopalynological analyses of the samples resulted with characteristic Salix spp. pollen content up to 66%. Other secondary pollen types, found with minor contribution, were Brassica napus L., Prunus spp., Acer spp. and Taraxacum spp. The phytochemical study of the headspace and extractives of Salix spp. honey revealed 26 and 47 compounds respectively. Targeted HPLC was focused on 9 compounds, especially on ABA isomers that were previously indicated as markers of willow honey (Tan et al., 1990; Tuberoso et al., 2011). Other general
physical–chemical characteristics were determined as well. Geographical origin and applied isolation methodologies may impact phytochemical composition and physical–chemical properties of the honey (Kaškoniene˙ & Venskutonis, 2010). Therefore, observed differences/similarities with published data on willow honey from other regions were further discussed. The comparison of Polish Salix spp. honey extracts with Croatian willow honey extracts (Jerkovic´ & Marijanovic´, 2010) is more reliable since the same experimental methodology was applied. 3.1. Phytochemicals extracted by HS-SPME The major constituents dominating the headspace of Polish willow honey were monoterpenes, mainly linalool derivatives such as lilac aldehyde isomers (8.0–36.9%), hotrienol (5.4–16.6%), cis-/ trans-linalool oxides (6.3–11.5%) along with terpinene-4-ol (2.4–3.1%) and several similar compounds, Table 1. Monoterpenes with bornane skeleton (borneol (5.0–10.9%)) and bicyclic pinane skeleton (pinocarvone (3.6–10.6%), myrtenal (1.7–4.8%) as well as verbenone (2.3–3.4%)) were other relevant compounds, not very common in other honey types. The headspace also contained relevant percentage of C13-norisoprenoid trans-b-damascenone (8.3–13.0%). Due to predominant monoterpenes abundance in the headspace, their biosynthetic relationships were further discussed taking into consideration uncommon skeletons in the honeys and the corresponding pathways as well as favored oxidative conditions within the hive (about 30 °C and acidic pH; Alissandrakis, Tarantilis, Harizanis, & Polissiou, 2007). Terpenes in general derive from geranyl pyrophosphate (GPP) that can rearrange to lynalyl pyrophosphate (LPP) and neryl pyrophosphate (NPP). After the elimination of pyrophosphate group (OPP ) from NPP, 1,6-cyclisation (Degenhardt, Köllner, & Gershenzon, 2009; Maffei, Gertsch, &
Table 1 Volatile compounds identified by GC-FID/GC-MS in headspace of Salix spp. honey samples after headspace solid-phase microextraction (HS-SPME). No.
Compounds
RI
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Dimethyl sulfide 2-Methylpropanenitrile 3-Methylbutanal Isoamyl alcohol 3-Methylpentanal 3-Methylpentanoic acid Benzaldehyde 1,3-Dimethyl-2-ethylbenzenea cis-Linalool oxide trans-Linalool oxide Linalool Hotrienol a-Isophorone Benzyl alcohol 4-Ketoisophorone Lilac aldehyde (isomer I) Lilac aldehyde (isomer I) Pinocarvone Lilac aldehyde (isomer II) Borneol Terpinene-4-ol a-Terpineol Myrtenal Lilac alcohola Verbenone trans-b-Damascenone
