Food Chemistry 124 (2011) 401–410

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Analytical Methods

Contribution to the characterisation of honey-based Sardinian product abbamele:

Volatile aroma composition, honey marker compounds and antioxidant activity
I. Jerković a,*, A. Kasum a, Z. Marijanović b, C.I.G. Tuberoso c
a
Department of Organic Chemistry, Faculty of Chemistry and Technology, University of Split, N. Tesle 10/V, 21000 Split, Croatia
b
Department of Food Technology, Marko Marulić Polytechnic in Knin, Petra Krešimira IV 30, 22300 Knin, Croatia
c
Department of Toxicology, University of Cagliari, via Ospedale 72, 09124 Cagliari, Italy

a r t i c l e i n f o a b s t r a c t

Article history: Sardinian abbamele is a typical product obtained from the honey recuperation from combs (traditional
Received 17 December 2009 procedure) or by concentration of the honey diluted in water (industrial procedure). Seven abbamele sam-
Received in revised form 11 May 2010 ples were obtained to study the volatiles’ composition, the presence of honey marker compounds and
Accepted 12 June 2010
their relationship with the production procedures. The long thermal treatment applied in abbamele pro-
duction caused very high (1007.0–4405.8 mg/kg) HMF content (HPLC-DAD), while glucose and fructose
amounts were quite similar to the honey ones (HPLC-RI). Total antioxidant activity (FRAP assay) of the
Keywords:
samples ranged between 13.3 and 71.2 mmol Fe2+/kg, while antiradical activity (DPPH assay) ranged
Abbamele
Headspace solid-phase microextraction
between 3.8 and 23.3 mmol TEAC/kg. Such high antioxidant values were linearly correlated with total
(HS-SPME) phenol amount (1297.8–4469.5 mg GAE/kg) determined by Folin–Ciocalteau method. Thermally derived
Ultrasonic solvent extraction (USE) furan derivatives and terpenes were abundant among the headspace volatiles (HS-SPME), particularly
GC and GC–MS limonene (0.5–76.0%) that probably originated from citrus rinds’ addition during abbamele production.
DPPH and FRAP assay GC and GC–MS analyses of USE isolates revealed HMF predominance as well as the honey marker com-
HPLC-DAD pounds (if/when existing) such as methyl syringate (up to 49.2%), marker of Asphodelus microcarpus
honey. High isophorone percentage (up to 30.9%) determined by HS-SPME followed by minor percentage
of 4-ketoisophorone and norisoprenoids in one sample indicated Arbutus unedo L. honey use in the pro-
duction. HPLC-DAD analysis confirmed the presence of specific honey markers: two samples showed high
methyl syringate concentrations (150.4–120.1 mg/kg) while homogentisic acid and other specific mark-
ers of A. unedo honey were found in one sample. The compared GC–MS and HPLC-DAD data proved to be
useful to obtain information about the use of specific honey in the production and to verify citrus
addition.
Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction honey diluted in water. In both ways of preparation it is common

to add peels or pieces of citrus fruits. A previous paper (Spano
Globalisation of food-stuff market led to a parallel develop- et al., 2008) reported a first chemical investigation of abbamele
ment of re-discovery and protection of traditional foods. Sardi- on typical parameters studied for honey (water content, electrical
nian abbamele is a typical product originally obtained from the conductivity, pH, free acidity, invertase activity, 5-hydroxy-
recuperation of honey from the combs. In ancient times and poor methyl-2-furaldehyde (HMF), total polyphenols and free amino
economies, no food was to be wasted and abbamele was perfect acids). As expected for a product submitted to heat treatment,
filler for sweets (Spiggia, 1997). According to abbamele traditional invertase activity was very low (less than 1.02 U/kg) while
production, honey is extracted from combs and the latter are HMF values ranged between 881 and 4776 mg/kg. The studied
crumbled and dipped into warm water (40 °C). Then, the emerg- parameters, although interesting for a preliminary characterisa-
ing wax separates and the remaining liquid (water, some honey tion of this product, were not useful to investigate the produc-
and pollen) is heated (up to 100 °C) until a brown, honey-like tion cycle of abbamele. In addition, it is stated (Spano et al.,
product is obtained (Spano et al., 2008). Today, abbamele is 2008) that the value of abbamele is usually much higher (up to
sometimes still prepared in this traditional way, but more and 10 times) than that of honey, but no information on its useful
more often it is prepared in industrial way by concentration of properties were reported. For instance, honey-based abbamele
could exhibit antioxidant properties contributing to its high va-
* Corresponding author. Tel.: +385 21 329434; fax: +385 21 329461.
lue. It is well known that antioxidant activity is one of the ben-
E-mail address: [email protected] (I. Jerković). eficial effects of honey that is greatly influenced by its botanical

0308-8146/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodchem.2010.06.047
402 I. Jerković et al. / Food Chemistry 124 (2011) 401–410

origin (Frankel, Robinson, & Berenbaum, 1998; Schramm et al., where the extracted volatiles were thermally desorbed directly to
2003) as well as heat treatment (Antony, Han, Rieck, & Dawson, the GC column.
2000).
The first goal of this work was to isolate volatiles from abbamele 2.3. Ultrasonic solvent extraction (USE)
and to obtain very representative chemical composition of more
and less volatile compounds applying solid-phase microextraction Ultrasound-assisted extraction (USE) was performed in an
(HS-SPME) and ultrasonic solvent extraction (USE). Isolated vola- ultrasound cleaning bath (Transsonic Typ 310/H, Germany) by
tiles were analysed by gas chromatography and mass spectrometry the mode of indirect sonication, at the frequency of 35 kHz at
(GC, GC–MS), in order to (i) investigate the volatile aroma com- 25 ± 3 °C. Forty grams of abbamele were dissolved with 22 mL of
pounds of abbamele and (ii) find potential compounds useful to ob- distiled water in a 100 mL flask. Magnesium sulphate (1.5 g) was
tain information on used honey type. Finding a marker compounds added and each sample was extensively vortexed. Different extrac-
in honey-based product is a powerful tool in the determination of tion solvents for USE were separately used for the representative
the honey used in the production so that this product can be sample: (1) pentane, (2) diethyl ether (on the same batch of
marked according to the honey origin. In this regard, HPLC-DAD abbamele after sonication with pentane and removing the pentane
was also applied to analyse the presence of nonvolatile honey mar- extract), (3) a mixture pentane: diethyl ether 1:2 (v/v) and (4)
ker compounds. The last aim of this research is determination of dichloromethane. Sonication was held for 30 min. After sonication,
abbamele total phenols’ amount (Folin–Ciocalteau assay), antiradi- the organic layer was separated by centrifugation and filtered over
cal (DPPH assay) and total antioxidant activities (FRAP assay). anhydrous MgSO4. Aqueous layer was returned to the flask and an-
other batch of the same extraction solvent (20 mL) was added and
extracted by ultrasound for 30 min. Organic layer was separated as
2. Materials and methods
the previous one, filtered over anhydrous MgSO4 and aqueous layer
was sonicated third time for 30 min with another batch (20 mL) of
2.1. Reagents and honey samples
the extraction solvent. Joined organic extracts were concentrated
up to 0.2 mL by fractional distillation and 1 lL was used for GC
Diethyl ether, pentane and dichloromethane were purchased
and GC–MS analyses. After determination of the most suitable
from Kemika (HR-Zagreb) and were distilled before usage.
extraction solvents for the representative sample, all the samples
Anhydrous MgSO4 and NaCl were obtained from Fluka Chemie
of abbamele were extracted using selected solvents as previously
(CH-Buchs). Methanol, acetonitrile, 5-hydrohymethylfurfural,
described.
methyl syringate, homogentisic acid, gallic acid, ferrous sulphate,
1,1-diphenyl-2-picrylhydrazyl radical (DPPH), (±)-6-hydroxy-
2.4. Gas Chromatography (GC) and gas chromatography–mass
2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), 2,4,6-tris-
spectrometry (GC–MS)
(2-pyridyl)-1,3,5-triazine (TPTZ), Folin–Ciocalteau reagens were
obtained from Sigma–Aldrich, Fluka (Milan, Italy). Standard of
Gas chromatography analyses were carried out on an Agilent
(±)-2-trans, 4-trans abscisic acid was purchased from A.G. Scien-
Technologies (Palo Alto, CA, USA) gas chromatograph model
tific, Inc. (San Diego, CA). Fructose, glucose, sucrose, sodium
7890A equipped with flame ionisation detector. Chromatographic
carbonate, ferric chloride and CuSO4
5H2O were supplied by Carlo
separations were performed on 30 m  0.25 mm i.d capillary col-
Erba (Milan, Italy). All other standards were purchased from
umn HP-5MS (5%-phenyl)-methylpolysiloxane, Agilent J & W GC
Sigma–Aldrich (Biovit d.o.o., Varaždin, Croatia). Ultrapure water
column) with coating thickness 0.25 lm. The oven was tempera-
(18 mX) was distilled and then purified with a Milli-Q Advantage
ture-programmed isothermal from 70 °C for 2 min, then increased
A10 System apparatus (Millipore, Milan, Italy).
to 200 °C, at a rate of 3 °C/min and held isothermal for 15 min. He-
This study was carried out on seven abbamele samples collected
lium at 1 mL/min was used as the carrier gas. Injector temperature
from professional producers in different areas of Sardinia (Italy)
was 250 °C and detector temperature was 300 °C. The injected vol-
during the years 2008–2009. Three samples (1–3) were produced
ume was 1 lL and split ratio was 1:50.
in the traditional way, while other four (4–7) were produced in
Analyses of volatile compounds by gas chromatography–mass
industrial way (Table 1). All the samples were stored in hermeti-
spectrometry were carried out with the Agilent gas chromatograph
cally closed glass bottles at 4 °C until the analysis.
model 7890A fitted with a mass selective detector model 5975C
(Agilent Technologies, Palo Alto, CA, USA). Mass detector worked
2.2. Headspace solid-phase microextraction (HS-SPME) in the electron impact ionisation mode at 70 eV, the mass range
was m/z 30–300 and ion source temperature was 280 °C. Volatile
The isolation of headspace volatiles was performed using man- compound separation was obtained using the same column and
ual SPME fibre with the layer of polydimethylsiloxane/divinylben- oven temperature programme as previously described for GC.
zene (PDMS/DVB) obtained from Supelco Co. (Bellefonte, PA, USA). The individual peaks were identified by comparison of their
The coating was 1 cm long. The fibre was conditioned prior to use retention indices (relative to C9–C25 n-alkanes for HP-5MS column)
according to the manufacturer’s instructions by inserting into the to those of authentic samples and literature (El-Sayed, 2007 and
GC injector port. references therein), as well as by comparing their mass spectra
For HS-SPME extraction 5 mL of abbamele/water solution (1:1 v/ with the Wiley 275 MS library (Wiley, New York, USA) and NIST02
v; the ionic strength was increased using saturated NaCl water (Gaithersburg, Germany) mass spectral database. The percentage
solution) was placed in 15 mL amber glass vial (volume ratio head- composition of the samples was computed from the GC peak areas
space:solution was 1:1 v/v) and hermetically sealed with PTFE/sil- using the normalisation method (without correction factors). The
icone septa. The vial was maintained in a water bath at 60 °C component percentages (Table 2) were calculated as mean values
during equilibration (15 min) and extraction (40 min) and was par- from duplicate GC and GC–MS analyses.
tially submerged so that the liquid phase of the sample was in the
water. All the experiments were performed under constant stirring 2.5. HPLC-RI analysis
velocity (1000 rpm) by magnetic stirrer. After sampling, the SPME
fibre was withdrawn into the needle, removed from the vial and in- Fructose, glucose and sucrose were detected using a Waters LC
serted into the injector (250 °C) of the GC and GC–MS for 6 min (Waters S.p.A., Vimodrone, Milan, Italy) fitted with a multisolvent
 I. Jerković et al. / Food Chemistry 124 (2011) 401–410 403

Table 1
Characteristics of the abbamele samples.

Sample*
1 2 3 4 5 6 7
Type of production traditional traditional traditional industrial industrial industrial industrial
Year of production 2007 2007 2007 2007 2006 2008 2008
Moisture (g/100 g) 19 ± 1.2 26.8 ± 0.1 23.8 ± 0.1 27.7 ± 0.2 16.2 ± 0.3 15.3 ± 0.6 18.3 ± 1.4
Fructose (g/100 g) 38.87 ± 0.79 33.26 ± 0.83 27.89 ± 0.05 31.85 ± 0.98 37.65 ± 0.25 36.38 ± 1.53 30.36 ± 0.15
Glucose (g/100 g) 35.21 ± 0.78 27.17 ± 1.16 27.32 ± 1.99 27.11 ± 0.37 35.44 ± 0.26 35.17 ± 3.19 30.35 ± 2.28
Sucrose (g/100 g) 1.26 ± 0.08 n.d. n.d. n.d. 1.11 ± 0.09 n.d. n.d.
Fructose/ Glucose 1.10 1.22 1.02 1.17 1.06 1.03 1.00
HMF (mg/kg) 1237.2 ± 87.5 1115.2 ± 76.6 1823.6 ± 54.6 1892.9 ± 123 2971.5 ± 13 1007 ± 56.3 4405.8 ± 133.4
Total polyphenols (mg GAEa /kg) 1297.8 ± 56.5 1959.8 ± 5.5 1377.6 ± 54.1 2039.0 ± 86.9 4469.5 ± 143.6 1491.2 ± 148.9 3468.2 ± 172.7
Methyl syringate (mg/kg) 150.4 ± 3.6 n.d 120.1 ± 4.1 n.d n.d n.d n.d
Homogentisic acid (mg/kg) n.d n.d n.d n.d 85.2 ± 2.9 n.d n.d
Unedone b(mg/kg) n.d n.d n.d n.d 112.9 ± 3.5 n.d n.d
trans,trans-abscisic acid (mg/kg) n.d n.d n.d n.d 125.0 ± 6.2 n.d n.d
cis,trans-abscisic acid (mg/kg) n.d n.d n.d n.d 127.1 ± 4.3 n.d n.d
FRAP c(mmol Fe2+/kg) 16.2 ± 0.1 19.4 ± 0.2 13.3 ± 0.1 24.6 ± 0.5 71.2 ± 0.6 19.9 ± 0.9 36.8 ± 0.7
DPPH d(mmol TEAC/kg) 3.8 ± 0.3 5.7 ± 0.1 3.8 ± 0.2 7.6 ± 0.2 23.3 ± 0.1 5.1 ± 0.3 12.2 ± 1.2
*
Values are means ± SD of triplicate determinations.
a
GAE: gallic acid equivalent.
b
dosed using c,t-ABA calibration curve.
c
FRAP value is expressed as Fe2+ millimolar concentration, obtained from a FeSO4 solution having an antioxidant capacity equivalent to that of the dilution of the abbamele.
d
DPPH value is expressed as TEAC millimolar concentration, obtained from a Trolox solution having an antiradical capacity equivalent to that of the dilution of the
abbamele.

delivery system 600, a column heater set at 35 °C, an autosampler 3. Results and discussion
717 plus with a 50-ll loop and a refractive index detector Varian
356-LC (Varian, Leinì, TO, Italy). Separation was obtained with a Honey-based product abbamele is produced from the honey
Spherisorb NH2 column (250  4.6 mm, 5 lm, Waters) using water heated up to 100 °C. Table 1 shows that final product vary signifi-
and acetonitrile 20:80 (v/v) as mobile phase at a constant flow rate cantly according to the preparation way. Moisture ranges from
of 0.8 mL/min. Standard and working solutions were prepared in 15.3 to 27.7 g/100 g, with a variation of 55% and total sugar
ultrapure water. Calibration curves were built with the method amount (determined by HPLC-RI) range from 55.2 g/100 g of the
of external standard, correlating the area of the peaks with the con- sample 3 to 75.3 g/100 g of sample 1. Sucrose was detected in
centration with correlation values ranging from 0.9991 to 0.9998. the samples 1 and 5 and the ratio fructose/glucose is similar to that
Abbamele samples were homogenised, diluted with ultrapure of honey showing a slight predominance of fructose. Differences
water, filtered through cellulose acetate GD/X septa (0.45 lm, among abbamele samples seem more connected with procedure
25 mm Ø, Whatman, Milan, Italy) and injected in HPLC without of preparation, rather than with different types of honey used.
any further purification. The long thermal treatment involved in the production of abbamele
caused very high content of 5-hydroxymethyl-furfural (HMF) in
2.6. HPLC-DAD analysis the samples. HPLC-DAD analysis showed the amount of HMF up
to 4405.8 mg/kg in sample 5 (Table 1), values comparable with
5-Hydrohymethyl-furfural (HMF) and specific markers of typi- those found by (Spano et al. (2008)). The HMF values detected in
cal Sardinian honeys were detected and quantified using an abbamele ought to be compared with those of other heated food
HPLC-DAD method as described in Tuberoso et al. (2010). Briefly, products, not the honey (Spano et al., 2008).
a Varian system ProStar HPLC fitted with a ThermoSeparation The two complementary isolation techniques used for the vola-
diode array detector SpectroSystem UV 6000lp (ThermoSeparation, tiles’ isolation (HS-SPME and USE) followed by GC and GC–MS
San Jose, CA, USA) set at 280 nm was employed. Separation was ob- analyses allowed to obtain very representative chemical composi-
tained with a Gemini C18 column (150  4.60 mm, 3 lm, Phenom- tion of abbamele more and less volatile compounds without the
enex, Casalecchio di Reno, BO, Italy) using 0.2 M phosphoric acid formation of artefacts. This approach is common among our re-
(solvent A) and acetonitrile (solvent B) as mobile phase. The gradi- search efforts for adequate fingerprinting of the honey volatiles
ent (v/v) was generated keeping 90% of solvent A for 5 min, than in research of typical volatile marker compounds of unifloral
decreasing to 65% in 15 min; and to 10% in 20 min and remaining botanical origin (Jerković, Marijanović, Kezić, & Gugić, 2009; Jerk-
at this concentration for 10 min. Before each injection the system ović, Tuberoso, Marijanović, Jelić, & Kasum, 2009). Some single
was stabilised for 10 min with the initial A/B ratio (90:10, v/v). compounds or groups of compounds have been reported by differ-
Chromatograms and spectra were elaborated with a ChromQuest ent researchers as indicative of the honey floral type, e.g. nonanol,
V. 2.51 data system (ThermoQuest, Rodano, Milan, Italy). nonanal, nonanoic acid and acetoin as being characteristic of
eucalyptus honey (Perez, Sanchez-Brunete, Calvo, & Tadeo, 2002);
2.7. Moisture content, total polyphenols and antioxidant activity acetophenone, 1-phenylethanol and 2-acetophenone being charac-
assays teristic of chestnut honey (Guyot, Bouseta, Scheirman, & Collin,
1998).
Moisture content was assessed by drying 1 g of sample for 4 h in a Heat treatment in food is related to the transformations (mainly
thermostatic oven at 105 ± 1 °C and weighing after it reached a con- Maillard reactions) in flavour, aroma, taste and colour closely re-
stant weight. The total phenol content, antiradical (DPPH test) and lated with temperature, time, pH, the nature of reactants (i.e., the
total antioxidant (FRAP test) activities were measured through spec- type of carbohydrates and amino acids or proteins in the honey),
trophotometric determinations as described by Tuberoso et al. etc. (Martins, Jongen, & Van Boeckel, 2001). Heating the honey at
(2009). temperatures as low as 50 °C leads to the formation of new volatile
Table 2

404
Compounds isolated from abbamele samples 1–7 by ultrasonic solvent extraction (USE) and headspace solid-phase microextraction (HS-SPME) analysed by GC and GC–MS.

No. Compound RI 1 2 3 4 5 6 7
A B C A B C A B C A B C A B C A B C A B C
Area percentage (%)
1 3-Methylbutanal* >900 0.1 – – 0.1 – – – – – – – – 0.1 – – 0.9 – – – – –
2 2-Methylbutanal* >900 0.4 – – 0.3 – – – – – – – – 0.4 – – 0.7 – – – – –
3 2,5-Dimethylfuran* >900 – – – – – – – – – – – – 0.2 – – – – – – – –
4 Dimethyl disulphide* >900 0.1 – – 0.1 – – – – – – – – 0.3 – – 0.7 – – 0.1 – –
5 Octane >900 – – – 0.1 – – – – – – – – – – – 0.4 – – – – –
6 2-Methyl-3-oxo-tetrahydrofuran* >900 – – – 0.1 – – 0.5 – – – – – 0.2 – – 0.1 – – – – –
7 3-Methylbutanoic acid (Isovaleric acid) >900 – – – – – – – – – – – – – – – 0.4 – – – – –
8 2-Furancarboxaldehyde (Furfural) >900 2.9 – – 3.6 – – 13.6 – – 1.3 – – 7.3 – – 7.3 – – 20.4 – –
9 2-Furanmethanol >900 – 0.1 – 0.3 0.1 0.1 0.6 0.1 – 0.1 0.1 – 0.1 0.4 – 0.2 0.1 – 0.1 0.1 –
10 4-Methyloctane* >900 – 0.1 – – – – – – – – – – – – – – – – – – –
11 Ethylbenzene >900 – 0.1 – – 0.1 – – 0.1 – – – – – – – – 0.1 – – – –
12 1,4-Dimethylbenzene** >900 – 0.4 – – 0.3 – – 0.6 – – 0.2 – – 0.2 – – 0.7 – – 0.2 –
13 Ethenylbenzene* >900 – 0.1 – – – – – – – – – – – – – – – – – – –
14 1,3-Dimethylbenzene** >900 – 0.1 – – 0.1 – – 0.1 – – – – – – – – 0.1 – – – –
15 Phenylacetylene* >900 0.5 – – – – – 4.5 – – – – – – – – 1 – – – – –

I. Jerković et al. / Food Chemistry 124 (2011) 401–410

16 Nonane 900 – – – – – – – – – 0.1 – – – – – 0.3 – – – – –
17 2-Buthoxyethanol* 907 – 0.1 – – – – – – – – – – – – – – – – – – –
18 1-(2-Furanyl)-ethanone 914 0.8 0.2 0.1 1 0.1 – 4.2 0.1 – 0.3 – 0.1 1 0.1 – 2 0.2 0.1 3.7 0.4 0.1
19 a-Pinene 939 – – – 0.2 – – – – – 0.4 – – – – – – – – – – –
20 2-Methylpropyl-2-methyl butanoate* 945 – – – – – – – – – – – – – – – 0.6 – – – – –
21 5-Methyl-2-furfural 965 1.1 0.1 – 1.3 0.1 – 6.1 0.1 – 0.5 – 0.1 2.6 0.2 – 1.5 – 0.1 7.5 0.4 0.1
22 Hexanoic acid 974 – – – 1.7 0.3 – – – – – – – – – – 0.1 – – – – –
23 Dimethyl trisulphide* 975 – – – – – – – – – – – – – – – 2.3 – – – – –
24 b-Pinene 981 – – – – – – – – – 0.2 – – – – – – – – – – –
25 b-Myrcene 992 – – – 0.8 – – – – – 1.1 – – – – – – – – – – –
26 a-Phellandrene 1007 – – – – – – – – – 0.2 – – – – – – – – – – –
27 d-3-Carene 1014 – – – – – – – – – 0.2 – – – – – – – – – – –
28 a-Terpinene 1020 0.3 – – – – – – – – 0.6 – – – – – – – – – – –
29 p-Cymene 1028 0.4 – – – – – – – – 0.5 – – 0.2 – – 3.2 – – 0.1 – –
30 2-Hydroxy-3-methyl-cyclopent-2-en-1-one* 1031 – 0.1 – – – – – – – – – – – – – – – – – – –
31 Limonene 1032 4.1 – – 70.8 0.4 0.5 0.4 – – 76 0.2 0.5 0.5 – – 1.6 – – 15.8 – –
32 Benzyl alcohol 1037 – – – – – – 0.6 – – – – – 0.6 0.1 – 0.7 – – 0.1 – –
33 Phenylacetaldehyde 1048 4.6 0.4 0.2 0.3 – – 3.2 0.1 – 0.2 – 0.1 0.7 – – 2.9 – 0.1 0.8 – –
34 trans-b-Ocimene* 1051 – – – – – – – – – 0.2 – – – – – – – – – – –
35 4,7-Dimethylundecane** 1061 – 0.1 – – – – – – – – – – – – – – – – – – –
36 c-Terpinene 1062 0.2 – – – – – – – – 9.5 – – – – – – – – – – –
37 2-Acetylpyrrole* 1063 – – – 0.4 0.1 – – 0.5 – – 0.6 – – – – – – – – – –
38 trans-Linalool oxide 1076 1.3 0.1 – 0.3 – – 0.8 – – – – – 1 – – 5.1 – – 0.2 – –
39 2-Furancarboxylic acid 1080 – 0.1 – – – – – – – – – – – – – – – – – – –
40 Methyl 2-furoate* 1084 0.8 – – 0.8 – – 2.5 – – 0.4 – – 2 – – 3.1 – – 3.3 – –
41 1-(2-Furanyl)-2-hydroxyethanone* 1087 – 2.5 3 – 0.3 0.4 – – 4.2 – – 3.1 – 2.4 1.4 – 2.3 5.7 – 3.7 0.8
42 a-Terpinolene 1090 – – – – – – – – – 1.2 – – – – – – – – – – –
43 p-Cymenene 1092 – – – – – – – – – – – – – – – 7.8 – – – – –
44 Fenchone 1093 9.3 0.1 – – – – – – – – – – – – – – – – 1.7 – –
45 Linalool 1101 0.3 – – 1.7 – – – – – 0.2 – – – – – – – – – – –
46 Nonanal 1105 0.2 – – – – – – – – – – – – – – 3.7 – – – – –
47 2-Methylbenzofuran 1110 – – – – – – – 2.4 2.3 – – – 0.3 – – 0.4 – 0.1 1.5 – –
48 2-Phenylethanol 1116 – – – – – – 51.5 – – 0.4 – – – – – 0.7 – – – – –
49 3-Hydroxy-2-methyl-4H-Pyran-4-one* 1119 – 0.1 – – 0.1 – – – – – – – – – – – – – – – –
50 2-Ethylhexanoic acid* 1121 0.9 – – – – – – – – – – – – – – – – – – – –
51 3,5,5-Trimethyl-cyclohex-2-en-1-one (Isophorone) 1124 2.5 0.1 – – – – – – – – – – 30.9 0.4 – 1.5 – – – – –
52 2,3-Dihydro-3,5-dihydroxy-6-methyl-4H-Pyran-4-one* 1144 – 0.6 0.5 0.2 0.3 – – 0.1 – – 0.2 0.4 0.8 1.7 0.9 1 1 2.6 – 0.5 0.6
53 3,5,5-Trimethylcyclohex-3-ene-1,4-dione (4- 1147 – – – – – – – – – – – – 1.9 – – 2.5 – – – – –
Ketoisophorone)
54 Camphor 1148 0.4 – – 0.3 – – – – – – – – – – – – – – – – –
55 2-Hydroxy-3,5,5-trimethyl-cyclohex-2-en-1-one* (2- 1151 – – – – – – – – – – – – 1.5 – – – – – – – –
Hydroxyisophorone)
56 Benzoic acid 1162 – 0.1 – – 0.2 – – 0.3 – – 0.4 0.6 – 0.1 – – – 0.1 – 0.2 –
57 Octanoic acid 1174 – – – 0.7 0.1 – – – – – – – – – – – – – – – –
58 Terpinen-4-ol 1180 0.2 – – 0.3 – – – – – 0.3 – – – – – – – – – – –
59 a-Terpineol 1192 1.2 – – – – – – – – 0.9 – – – – – – – – 15.4 – –
60 Methyl salicylate 1195 – – – – – – – – – – – – – – – 0.7 – – – – –
61 Methyl chavicol (Estragole) 1199 52.5 – – – – – – – – – – – – – – – – – – – –
62 1-Methoxy-4-(prop-2-enyl)-benzene* 1202 – 0.1 – – – – – – – – – – – – – – – – – – –
63 Decanal 1207 – – – – – – – – – – – – – – – 2 – – – – –
64 a-Ionene* 1213 – – – – – – – – – – – – – – – 0.9 – – – – –
65 3-Phenylfuran* 1223 0.6 – – – – – – – – – – – – – – 1.8 – – – – –
66 5-Hydroxymethylfurfural 1230 2.9 27.7 51.3 3.1 17.3 49.6 1.9 6.9 21 0.5 19.3 76.5 16.9 57.6 77.3 6.1 17 65.3 19.5 78.7 91.2
67 4-(1-Methylethyl)-benzaldehyde (Cuminal) 1243 – – – – – – – – – – – – – – – 1 – – – – –
68 Carvotanacetone* 1250 – – – – – – – – – – – – – – – 2.8 – – – – –
69 Nonanoic acid 1273 – – – – – – – – – – – – – – – 3.4 – – – – –
70 2,4,6-Trimethyphenol** 1274 – – – – – – – – – – – – 3.7 – – – – – – – –
71 2-Phenylbut-2-enal* 1275 0.7 – – – – – – – – – – – – – – – – – – – –
72 Phellandral* 1276 – – – – – – – – – – – – – – – 4.9 – – – – –

I. Jerković et al. / Food Chemistry 124 (2011) 401–410

73 Phenylacetic acid 1269 – 0.1 – – 0.3 – – 0.4 – – – – – – – – – – – 0.6 –
74 Anethole 1289 0.4 – – – – – – – – – – – – – – – – – – – –
75 4-(1-Methylethyl)-benzenethanol* (p-Cymen-7-ol) 1292 – – – – – – – – – – – – – – – 2.8 – – – – –
76 Carvacrol 1294 – – – 0.7 – – 1.3 – – – – – – – – – – – – – –
77 5-Acetyl-2-furanmethanol* 1301 – – 0.1 – 0.1 – – – – – – 0.2 – – – – – 0.1 – – –
78 2-Methoxy-benzene-1,4-diol* 1302 – 0.1 – – – – – – – – – – – – – – – – – – –
79 Thymol 1303 – – – – – – – 0.1 – – – – – – – 0.5 – – – – –
80 3-Methoxyacetophenone* 1306 – – – – – – – – – – – – 0.3 – – – – – – – –
81 3-Phenylprop-2-en-1-ol** (Cinnamyl alcohol) 1313 – – – – – – – 0.1 – – – – – – – – – – – – –
82 d-Elemene* 1314 – – – 1.6 – – – – – – – – – – – – – – – – –
83 3,4,5-Trimethylphenol** 1317 0.4 0.1 – – – – – – – – – – 16.1 1.2 0.5 3.5 – 0.1 – – –
84 1,2-Dihydro-1,1,6-trimethyl-naphtalene* 1354 – – – – – – – – – – – – – – – 0.9 – – – – –
85 Decanoic acid 1370 0.1 – – 0.6 0.6 0.4 – – – – – – – – – – – – – – –
86 a-Copaene* 1377 – – – – – – – – – 0.3 – – – – – – – – – – –
87 Ethyl decanoate (Ethyl caprate) 1397 – – – – – – 0.7 – – – – – – – – – – – – – –
88 4-Hydroxybenzyl alcohol 1426 – – – – – – – 0.1 – – – – – – – – – – – – –
89 3-Phenylprop-2-enoic acid (Cinnamic acid) 1434 – – – – 0.8 – – – – – – – – – – – – – – – –
90 Methyl-4-hydroxybenzoate 1471 – 0.1 – – – – – 0.1 – – – – – – – – – – – – –
91 5-Hydroxydec-2-enoicacid lactone* 1478 – – – – – – – – – – – – – – – 1.6 – – – – –
92 5-Methyl-2-phenylhex-2-enal* 1490 – – – – – – – – – – – – – – – 0.8 – – – – –
93 d-Selinene* 1491 – – – 0.8 – – – – – – – – – – – – – – – – –
94 Pentadecane 1500 – 0.1 – – 0.1 – – 0.1 – – – – – – – – – – – – –
95 4-Methyl-2,6-bis-(1,1-dimethylethyl)-phenol 1514 – 0.4 – – 0.4 0.6 – 0.3 – – 0.2 0.1 – 0.4 – 0.4 0.6 0.1 – 0.3 –
96 4-Hydroxybenzoic acid 1522 – 0.2 – – 1.1 – – 0.8 – – 15.5 1 – 3.3 0.1 – – – – – –
97 Methyl 4-hydroxy-3-methoxybenzoate (Methyl 1524 – 0.2 0.1 – – – – 0.1 – – – – – – – – – – – – –
vanillate)
98 4-Phenylbut-3-enoic acid* 1531 – 0.1 – – – – – – – – – – – – – – – – – – –
99 4-Hydroxy-3-methoxy-benzoic acid (Vanillic acid) 1566 – – – – – – – – – – 0.4 – – – – – – – – – –
100 Dodecanoic acid 1578 – – – – 0.2 0.3 – – – – – – – – – – – – – – –
101 4-Keto-a-ionone* 1655 – – – – – – – – – – – – – – – 1.8 1.8 – – – –
102 a-Ionol* 1656 – – – – – – – – – – 0.4 0.2 – 0.3 0.2 – 0.7 0.3 – – –
103 Menthofuran 1662 – – – – – – – – – – – – – 0.3 – – – 0.5 – – –
104 3-Hydroxy-4-phenyl-2(5H)-furanone* 1697 – 0.2 – – – – – – – – – – – – – – – – – – –
105 2,4,5-Trimethoxy-3-methylphenol** 1728 – 0.1 – – – – – – – – – – – – – – – – – – –
106 Methyl 4-hydroxy-3-methoxyphenylacetic acid* (Methyl 1752 – 0.1 – – – – – 0.2 – – 1.5 0.8 – 0.8 0.1 – 0.9 0.6 – – –
homovanillate)
107 Tetradecanoic acid 1768 – – – – 0.1 – – – – – – – – – – – – – – – –

405
(continued on next page)
 406
Table 2 (continued)

No. Compound RI 1 2 3 4 5 6 7
A B C A B C A B C A B C A B C A B C A B C
108 Methyl 3,5-dimethoxy-4-hydroxybenzoate (Methyl 1774 – 49.2 29.9 – 4 1.4 – 29.3 35.4 – 2.5 0.4 – 1.3 1.1 – 1.5 0.8 – 0.2 –
syringate)
109 4-Hydroxy-3,5,6-trimethyl-4-(3-oxobut-1-enyl)- 1795 – 0.9 0.6 – – – – – – – 0.9 1.1 – 7.2 6.1 – 6.7 6 – – –
cyclohex-2-en-1-one*
110 3-(4-Hydroxyphenyl)-prop-2-enoic acid* (4-Coumaric 1796 – – – – 2.2 – – – – – – – – – – – – – – – –
acid)
111 1H-Indene* 1834 – 0.1 – – – – – – – – – – – – – – – – – – –

I. Jerković et al. / Food Chemistry 124 (2011) 401–410

112 3-(4-Hydroxy-3-methoxyphenyl)-prop-2-enoic acid 1867 – – – – – – – – – – 0.1 – – – – – – – – – –
(Ferulic acid)
113 Diisobutyl phthalate 1869 – 0.1 – – – – – – – – – – – – – – – 0.2 – – –
114 Hexadecan-1-ol 1882 – – – – – – – 0.1 – – – – – – 0.2 – 0.1 0.2 – – –
115 Nonadecane 1900 – – – – 0.1 – – 0.1 – – – – – 1.6 – – 43.8 0.5 – – –
116 1,2,3-Trimethoxy-5-(prop-2-enyl)-benzene* (Elemicin) 1923 – 0.1 – – – – – – – – – – – – – – – – – – –
117 Methyl hexadecanoate (Methyl palimitate) 1934 – – – – 0.3 0.7 – 0.2 0.6 – – 0.2 – – – – – 0.5 – – –
118 1-Phenoxypropan-2-ol* 1957 – – – – – – – – – – 0.8 – – – 1 – – – – – –
119 Hexadecanoic acid (Palmitic acid) 1963 – 0.2 0.3 – 17.9 22.3 – 15.8 11.4 – 11.6 3.4 – 2 0.3 – – – – 0.5 0.6
120 Eicosane 2000 – – – – – – – – – – – – – 2.9 – – 13.5 9.9 – – –
121 Ethyl hexadecanoate (Ethyl palmitate) 2002 – – – – 0.9 – – 0.7 – – – – – – – – – – – – –
122 (Z)-octadec-9-enoic acid (Oleic acid) 2147 – – – – – – – 0.5 – – – – – – – – – – – – –
123 2-Ethyl-dibenzothiophene* 2061 – 0.2 – – – – – – – – – – – – – – – – – – –
124 (Z)-octadec-9-en-1-ol 2060 – – 0.4 – – – – – – – – 0.2 – – – – – – – – –
125 Octadecan-1-ol 2084 – 0.1 – – – – – – – – – – – – – – – 0.1 – – –
126 Methyl (Z,Z)-octadeca-9,12-dienoate (Methyl linoleate) 2101 – – – – 1.4 0.5 – – – – – – – – – – – – – – –
127 Methyl (Z,Z,Z)-octadeca-9,12,15-trienoate (Methyl 2109 – – – – 1.1 1.5 – – – – – – – – – – – – – – –
linolenate)
128 Heneicosane 2100 – 0.8 5.1 – – – – 6.4 1.2 – 5.3 0.5 – 0.2 0.1 – 2.2 0.3 – 0.3 –
129 (Z,Z,Z)-octadeca-9,12,15-trien-1-ol* 2175 – – – – – – – – – – 6.9 – – – – – – – – – –
130 (Z,Z)-Octadeca-9,12-dienoic acid (Linolenic acid) 2178 – – – – 31.5 5.2 – 27 11.1 – 8.4 2.7 – 1 0.8 – 1.1 – – – 0.5
131 Octadecanoic acid 2181 – – – – – – – – – – 1.5 2.8 – – – – – – – – –
132 Ethyl (Z,Z,Z)-octadeca-9,12,5-trienoate (Ethyl linolenate) 2197 – – – – 9.1 6.5 – 1.8 1.2 – – – – – – – – – – – –
133 Bis(2-ethylhexyl) phtalate* 2276 – 0.1 0.2 – – – – – – – – – – – – – – – – – –
134 Tetracosane 2400 – 1 – – 2.5 2.4 – 3.1 2 – 11.6 1.5 – 0.5 – – 1 0.1 – 0.2 –
Total identified (%) 90.2 88.0 91.8 92.2 94.6 92.4 92.4 98.7 90.4 95.6 88.6 96.5 89.6 86.2 90.1 88.6 95.4 94.4 90.2 86.3 93.9

1–7-number of abbamele sample; A-HS-SPME; B-USE with the mixture of pentane and diethyl ether (1:2 v/v); C-USE with dichloromethane; RI-retention indices on HP-5MS column.
*
tentatively identified (no reference compound was available).
**
correct isomer not identified.
 I. Jerković et al. / Food Chemistry 124 (2011) 401–410 407

compounds (generation of artefacts) and the GC peak areas of 3.2. Ultrasonic solvent extraction (USE)
many compounds varied significantly as a result of different heat-
ing conditions (Visser, Allen, & Shaw, 1988) either due to the oxi- Four solvents with different polarities were tested on represen-
dation or through Maillard reactions (Alissandrakis, Tarantilis, tative abbamele sample to determine the most suitable ones for the
Harizanis, & Polissiou, 2005). The effect of heating on HMF extraction of all samples with respect to the overall number of ex-
content of honey is greatly influenced by the honey botanical ori- tracted compounds (data not shown). From this preliminary re-
gin (Fallico, Zappala, Arena, & Verzera, 2004). search, two different solvents (mixture of diethyl ether and
pentane 1:2 v/v (solvent B) and pure dichloromethane (solvent
3.1. Headspace solid-phase microextraction (HS-SPME) C) were selected for USE in order to obtain more complete abbam-
ele profile of more and less polar volatile and semi-volatile
Seventy-six compounds were detected with the headspace compounds. 5-Hydroxymethylfurfural was the predominant com-
solid-phase microextraction (Table 2, column A). Thermally de- ponent in all USE extracts with the following distribution: solvent
rived furan derivatives were abundant: 2-furancarboxaldehyde B (6.9–78.7%), solvent C (21.0–91.2%). Other furan and pyran
(1.3–13.6%), 5-methyl-2-furancarboxaldehyde (0.5–6.1%), methyl derivatives were also present in USE extracts with solvent A such
furancarboxylate (0.4–3.1%), 5-hydrohymethylfurfural (0.5– as 2-furanmethanol (0.1–0.4%), 5-methyl-2-furfural (0.1–0.2%),
16.9%), dihydro-2-methyl-3(2H)-furanone (0.0–0.5%) and 2,3-dihy- 3-hydroxy-2-methyl-4H-pyran-4-one (0.0–0.1%) or 2,3-dihydro-
dro-3,5-dihydroxy-6-methyl-4H-pyran-4-one (0.0–1.0%). Their 3,5-dihydroxy-6-methyl-4H-pyran-4-one (0.1–1.7%). Several of
headspace percentages are not reliable, due to high water solubil- these compounds were present in USE extracts with solvent C,
ity and low volatility. On the other hand lower aliphatic aldehydes Table 2. In general, the percentages of furan and pyran derivatives
and acids (3-methylbutanal (0.0–0.9%), 2-methylbutanal (0.0– obtained by USE can be considered more reliable in comparison
0.7%) and 3-methylbutanoic acid (0.0–0.4%)) were only identified with HS-SPME due to higher polarity and less volatility.
by HS-SPME. They are also indicators of heat treatment and oxida- Higher aliphatic acids (such as (Z)-octadec-9-enoic acid,
tion reactions. Thermally derived compounds are abbamele charac- octadecanoic acid and linolenic acid), alcohols (such as (Z)-octa-
teristic, but not indicating the honey type used in the production. dec-9-en-1-ol and octadecan-1-ol) and esters (such as methyl
Terpenes were abundant in abbamele headspace, particularly hexadecanoate, ethyl hexadecanoate, methyl linoleate, ethyl linol-
limonene (0.5–76.0%) that probably originated from citrus rinds enate and methyl linoleate) as well as higher hydrocarbons (such
addition during production process. Limonene predominated in as tetracosane, heneicosane, eicosane and nonadecane) were abun-
the samples 4 (76%) and 2 (70.8%) and it can be pointed out as spe- dant in the samples 2 and 3 probably originated from beewax
cific aromatic headspace compound of abbamele, not found in such (Jerković, Marijanović, Ljubičić, & Gugić, 2010) during abbamele
high concentrations in different honeys. Therefore, limonene head- traditional processing from honeycombs.
space percentage is a key characteristic for detection of added cit- The most important marker of honey botanical origin found in
rus rinds during traditional and industrial processing of abbamele. USE extracts was methyl syringate (0.5–49.2%) with the highest
Other monoterpenes were also present (not in all the samples) percentages in the samples 1 and 3 (Table 2). Such high amounts
with minor percentages, such as a- and b-pinene (0.0–0.4%), of methyl syringate (Fig. 1B) can suggest that asphodel honey con-
b-myrcene (0.0–1.1%), a-phellandrene (0.0–0.2%), d-3-carene (0.0– tributed in samples 1 and 3 productions. In fact, it is reported
0.2%), a-terpinene (0.0–0.6%), p-cymene (0.0–0.5%), trans-b-ocim- (Tuberoso et al., 2009) that Sardinian Asphodelus microcarpus hon-
ene (0.0–0.2%), c-terpinene (0.0–9.5%), a-terpinolene (0.0–1.2%) ey is so far the honey with the highest amount of methyl syringate.
and others, Table 2. Very few sesquiterpenes such as a-copaene Very few terpenes in distinction from HS-SPME were identified
(0.0–0.3%) and c-selinene (0.0–0.8%) were found. Other honey and limonene was identified only in the samples 1 and 3 (Table 2).
ubiquitous terpenes such as linalool (0.0–1.7%) and trans-linalool Isophorones were not found as in HS-SPME, but oxygenated norisop-
oxide (0.0–5.1%) were also present. It is interesting to observe that renoides (such as 4-hydroxy-3,5,6-trimethyl-4-(3-oxo-1-butenyl)-
samples 3 and 5 show just a very little amount of limonene and no cyclohex-2-en-1-one) were identified in the samples 4, 5 and 6.
other terpenes were detected. It can be supposed that in such
samples only small amount of citrus peels was added. HS-SPME 3.3. HPLC-DAD
enabled detection of terpenes that are moderately visible in USE
extracts probably due to higher volatility and consequently more Besides HMF, the used chromatographic method allowed to de-
abundance in the headspace. tect the specific markers of unifloral Sardinian honeys. Fig. 2 re-
Ubiquitous honey volatile compounds–benzene derivatives ports the HPLC-DAD fingerprinting of samples 1 and 5. Methyl
were found such as benzyl alcohol (0.0–0.7%), phenylacetaldehyde syringate, marker of asphodel honey (Tuberoso et al., 2009), was
(0.2–4.6%) or 2-phenylethanol (0.0–51.5%). Highest content of found in the samples 1 and 3 (150.4 and 120.1 mg/kg, respec-
2-phenyethanol was found in sample 3, but it is not a specific mar- tively). Homogentisic acid (85.2 mg/kg), unedone (112.9 mg/kg),
ker since it has been reported in most honeys from a wide range of trans,trans-abscisic acid (125.0 mg/kg) and cis,trans-abscisic acid
floral sources. However, high percentage of 2-phenyethanol was (127.1 mg/kg), markers of strawberry-tree honey (Tuberoso et al.,
characteristic for several European honeys such as Amorpha frutico- 2010), were detected only in the sample 5. These results confirm
sa honey (Jerković, Marijanović et al., 2009; Jerković, Tuberoso et al., the information gathered from the volatile analyses: the samples
2009) or Calluna vulgaris honey (Guyot, Scheirman, & Collin, 1999). 1 and 3 were prepared with high amount of asphodel honey, while
High content (30.9%) of isophorone in sample 6 (Fig. 1A) fol- Sardinian straw-berry tree honey was used only in abbamele 5.
lowed by minor percentage of 4-ketoisophorone (1.9%), 2-hydrox- Heat treatment during abbamele production seems that it did not
yisophorone (0.0–1.5%) and norisoprenoides in some samples greatly influence the markers presence, although significant forma-
(Table 2) could be connected with Sardinian strawberry-tree honey tion of thermal artefacts occurred.
(Arbutus unedo L.). In fact, analysis of volatile (Bianchi, Cereri, &
Musci, 2005; de la Fuente, Sanz, Martínez-Castro, Sanz, & Ruiz- 3.4. Antioxidant activity and total phenol content
Matute, 2007) and semi-volatile fractions (Dalla Serra et al.,
1999) of strawberry-tree honey showed the presence of norisopr- A last aspect that deserves attention is the antioxidant activities
enoid compounds that are useful to characterise such a honey of abbamele. Total antioxidant activity measured with the FRAP as-
and were proposed as specific markers. say ranged from 13.3 to 71.2 mmol Fe2+/kg, while antiradical activity
408 I. Jerković et al. / Food Chemistry 124 (2011) 401–410

Fig. 1. Representative CG-MS chromatograms of abbamele samples 5 (A) and 1 (B) on HP-5MS column. Chromatogram (A) shows the marker compound isophorone obtained
by HS-SPME, while chromatogram (B) shows the marker compound methyl syringate obtained by ultrasonic extraction with the mixture pentane diethyl ether (1:2 v/v).

measured with the DPPH assay ranged from 3.8 to 23.3 mmol TEAC/ 3.7 and 4.4 mmol Fe2+/kg (Pichichero, Canuti, & Canini, 2009). Also
kg. Total phenolic amount ranged from 1297.8 to 4469.5 mg GAE /kg comparison with other foodstuff is very interesting because total
and it is linearly correlated with antioxidant and antiradical activi- phenol amount is similar or even higher than products such as red
ties (R2total phenols/FRAP = 0. 9109 and R2total phenols/DPPH = 0. 9395). Val- wines (Heinonen, Lehtonen, & Hopia, 1998), fruits and vegetables
ues were very high if compared to those published for the honeys, (Brat et al., 2006). Total phenolic values found in the samples of
although a direct comparison is very hard due to different types of abbamele are higher than those found by Spano et al. (2008), but dif-
antioxidant assay and way of quantification (Alvarez-Suarez, ferences can be due to different sample preparations and data
Tulipani, Romandini, Vidal, & Battino, 2009). However, dark and expression. Nevertheless, evaluation of these data needs some
honeydew honeys that are known to have the highest levels of total considerations. As HPLC-DAD analysis did not show high amount
phenolic compounds, usually do not exceed the level of 1250 mg of single phenolic compounds in abbamele samples, purification of
GAE/kg (Al et al., 2009; Ferreira, Aires, Barreira, & Estevinho, 2009). abbamele samples on SPE Oasis HLB column according to
FRAP values for honeys rich in phenolic compounds, such as Michalkiewicz, Biesaga, & Pyrzynska (2008) was performed. In this
chestnut, Satureja hortensis and honeydew honeys, ranged between way it was expected to separate phenolic compounds from other
 I. Jerković et al. / Food Chemistry 124 (2011) 401–410 409

Fig. 2. Representative HPLC-DAD chromatograms of abbamele samples 1 (A) and 5 (B) at 280 nm. Chromatographic conditions are described in the text. 1) 5-(hydroxymethyl)
furfural (HMF); 2) methyl syringate; 3) homogentisic acid; 4) furfural; 5) unedone; 6) trans,trans-abscisic acid; 7) cis,trans-abscisic acid.

eventually interfering compounds. Results showed that total phenol pounds to amino reductones or reductones and the formation of
amounts in these extracts were not statistically different from the polymers with antioxidant activity (Bailey & Um, 1992). Heterocy-
ones reported in Table 1 (data not shown). Interestingly, solutions clic compounds, such as furan and pyran derivatives, proved to inhi-
containing extracted phenolic compounds were more or less brown, bit oxidation (Osada & Shibamoto, 2006; Yong-Xin, Li, Qian, Kim, &
suggesting that the products of the Maillard reaction were not sepa- Kim, 2009). These products differ in molecular size and chemical
rated. However, because of the strong correlation between total structure with a common single antioxidative functional group,
polyphenols content and abbamele antioxidant activity found in this though the presence of entirely different antioxidants with different
research, the total phenol amount is an interesting aspect, even modes of action cannot be excluded (Lingnert, Eriksson, & Waller,
affected from the contribution of Maillard reaction products. It is 1983).
known that antioxidant activity of the honey is greatly influenced
by its botanical origin (Frankel et al., 1998; Schramm et al., 2003), 4. Conclusion
as well as heat treatment (Antony et al., 2000; Turkmen, Sari, Poy-
razoglu, & Velioglu, 2006). Antioxidants are formed at several stages Seven abbamele samples, from different industrial and tradi-
during the Maillard reaction, including degradation of Amadori com- tional producers, were obtained to study the relationship between
410 I. Jerković et al. / Food Chemistry 124 (2011) 401–410

the samples’ volatile aroma composition, honey marker com- Ferreira, I. C. F. R., Aires, E., Barreira, J. C. M., & Estevinho, L. M. (2009). Antioxidant
activity of Portuguese honey samples: Different contributions of the entire
pounds and the production procedures as well to determine their
honey and phenolic extract. Food Chemistry., 114, 1438–1443.
antioxidant activity. The investigation of the volatile fraction of Frankel, S., Robinson, G. E., & Berenbaum, M. R. (1998). Antioxidant capacity and
abbamele with different extraction methods allowed obtaining dif- correlated characteristics of 14 unifloral honeys. Journal of Apicultural Research,
ferent information about the production technology of this tradi- 37, 27–31.
Guyot, C., Bouseta, A., Scheirman, V., & Collin, S. (1998). Floral origin markers of
tional product. The presence of typical markers of Sardinian chestnut and lime tree honeys. Journal of Agricultural and Food Chemistry, 46,
unifloral honeys represents a powerful tool to connect abbamele 625–633.
with the territory of production. Terpenes are the key element to Guyot, C., Scheirman, V., & Collin, S. (1999). Floral origin markers of heather honeys:
Calluna vulgaris and Erica arborea. Food Chemistry, 64, 3–11.
determine adding of citrus fruit. Total phenol amount and antiox- Heinonen, I. M., Lehtonen, P. J., & Hopia, A. I. (1998). Antioxidant activity of berry
idant activities showed to be very interesting because they are and fruit wines and liquors. Journal of Agricultural and Food Chemistry, 46, 25–31.
much higher than those of honeys and comparable with that of Jerković, I., Marijanović, Z., Ljubičić, I., & Gugić, M. (2010). Contribution of the bees
and combs to honey volatiles: Blank-trail probe for honey biodiversity chemical
well-known products such as red wines and vegetables. The ap- profiling. Chemistry and biodiversity, 7, 1217–1230.
proach used in this study can be a model for the investigation of Jerković, I., Marijanović, Z., Kezić, J., & Gugić, M. (2009). Headspace, volatile and
high-quality traditional foodstuff, because the presence of specific semi-volatile organic compounds diversity and radical scavenging activity of
ultrasonic solvent extracts from Amorpha fruticosa honey samples. Molecules, 14,
markers connected with natural ingredients or technological pro- 2717–2728.
cesses can support products’ traceability and help to distinguish Jerković, I., Tuberoso, C. I. G., Marijanović, Z., Jelić, M., & Kasum, A. (2009).
the products clearly from other similar ones. Headspace, volatile and semi-volatile patterns of Paliurus spina-christi unifloral
honey as markers of botanical origin. Food Chemistry, 112, 239–245.
Lingnert, H., Eriksson, C. E., & Waller, G. R. (1983). Characterization of antioxidative
Acknowledgements Maillard reaction products from histidine and glucose. In The Maillard Reaction
in Foods and Nutrition. ACS Symposium Series (Vol. 215, pp. 335–345).
This research was supported by UKF project 25/08 (Evaluation Washington, DC: American Chemical Society.
Martins, S. I. F. S., Jongen, W. M. F., & Van Boeckel, M. A. J. S. (2001). A review of
of Unifloral Honeys – Chemical Fingerprinting and Nutritional Maillard reactions in food and implications to kinetic modelling. Trends in Food
Properties), Fondazione Banco di Sardegna (funding 436/2007. Science & Technology, 11, 364–373.
0065) and PIP d.o.o., KONCEPT-MEDIA d.o.o. and GODAX-PRO d.o.o. Michalkiewicz, A., Biesaga, M., & Pyrzynska, K. (2008). Solid-phase extraction
procedure for determination of phenolic acids and some flavonols in honey.
Journal of chromatography A, 1187, 18–24.
References Osada, Y., & Shibamoto, T. (2006). Antioxidative activity of volatile extracts from
Maillard model systems. Food Chemistry, 98, 522–528.
Al, M. L., Daniel, D., Moise, A., Bobis, O., Laslo, L., & Bogdanov, S. (2009). Physico- Perez, R. A., Sanchez-Brunete, C., Calvo, R. M., & Tadeo, J. L. (2002). Analysis of
chemical and bioactive properties of different floral origin honeys from volatiles from Spanish honeys by solid phase microextraction and gas
Romania. Food Chemistry., 112, 863–867. chromatography-mass spectrometry. Journal of Agricultural and Food
Alissandrakis, E., Tarantilis, P. A., Harizanis, P. C., & Polissiou, M. (2005). Evaluation Chemistry, 50, 2633–2637.
of four isolation techniques for honey aroma compounds. Journal of the Science Pichichero, E., Canuti, L., & Canini, A. (2009). Characterisation of the phenolic
of Food and Agriculture, 85, 91–97. and flavonoid fractions and antioxidant power of Italian honeys of
Alvarez-Suarez, J. M., Tulipani, S., Romandini, S., Vidal, A., & Battino, M. (2009). different botanical origin. Journal of the Science of Food and Agriculture,
Methodological aspects about determination of phenolic compounds and 89, 609–616.
in vitro evaluation of antioxidant capacity in the honey: A review. Current Schramm, D. D., Karim, M., Schrader, H. R., Holt, R. R., Cardetti, M., & Keen, C. L.
Analytical Chemistry., 5, 293–302. (2003). Honey with high levels of antioxidants can provide protection to
Antony, S. M., Han, I. Y., Rieck, J. R., & Dawson, P. L. (2000). Antioxidative effect of healthy human subjects. Journal of Agricultural and Food Chemistry, 51,
maillard reaction products formed from honey at different reaction times. 1732–1735.
Journal of Agricultural and Food Chemistry, 48, 3985–3989. Spano, N., Ciulu, M., Floris, I., Panzanelli, A., Pilo, M. I., Piu, P. C., et al. (2008).
Bailey, M. E., & Um, K. W. (1992). Maillard reaction products and lipid oxidation. In Chemical characterization of a traditional honey-based Sardinian product:
Lipid Oxidation in Food. ACS symposium series (pp. 122–130). Washington, DC: Abbamele. Food Chemistry, 108, 81–85.
American Chemical Society. Spiggia, S. (1997). Le api nella tradizione popolare della Sardegna. Sassari, Italy: Carlo
Bianchi, F., Cereri, M., & Musci, M. (2005). Volatile norisoprenoids as markers of Delfino Editore. pp. 43–44.
botanical origin of Sardinian strawberry-tree (Arbutus unedo L,) honey: Tuberoso, C. I. G., Bifulco, E., Caboni, P., Cottiglia, F., Cabras, P., & Floris, I. (2010).
Characterisation of aroma compounds by dynamic headspace extraction and Floral markers of strawberry-tree (Arbutus unedo L.) honey. Journal of
gas chromatography–mass spectrometry. Food Chemistry, 89, 527–532. Agricultural and Food Chemistry, 58, 384–389.
Brat, P., Georgé, S., Bellamy, A., Du Chaffaut, L., Scalbert, A., Mennen, L., et al. (2006). Tuberoso, C. I. G., Bifulco, E., Jerković, I., Caboni, P., Cabras, P., & Floris, I. (2009).
Daily polyphenol intake in France from fruit and vegetables. The Journal of Methyl syringate: A chemical marker of asphodel (Asphodelus microcarpus
Nutrition, 136, 2368–2373. Salzm. Et Viv.) monofloral honey. Journal of Agricultural and Food Chemistry, 57,
Dalla Serra, A., Franco, M. A., Mattivi, F., Ramponi, M., Vacca, V., & Versini, G. (1999). 3895–3900.
Aroma characterization of Sardinian strawberry tree (Arbutus unedo L,) honey. Turkmen, N., Sari, F., Poyrazoglu, E. S., & Velioglu, Y. S. (2006). Effects of prolonged
Italian Journal of Food Science, 11, 47–56. heating on antioxidant activity and colour of honey. Food Chemistry, 95,
de la Fuente, E., Sanz, M. L., Martínez-Castro, I., Sanz, J., & Ruiz-Matute, A. I. (2007). 653–657.
Volatile and carbohydrate composition of rare unifloral honeys from Spain. Food Visser, F. R., Allen, J. M., & Shaw, G. J. (1988). The effect of heat on the volatile
Chemistry, 105, 84–93. fraction from a unifloral honey. Journal Apicultural Research, 27, 175–181.
El-Sayed, A.M. (2007). The pherobase: Database of insect pheromones and Yong-Xin, L., Li, Y., Qian, Z.-J., Kim, M.-M., & Kim, S.-K. (2009). In vitro antioxidant
semiochemicals. Available from: http://www.pherobase.com. activity of 5-HMF isolated from marine red alga Laurencia undulata in free
Fallico, B., Zappala, M., Arena, E., & Verzera, A. (2004). Effects of conditioning on radical mediated oxidative systems. Journal of Microbiology and Biotechnology,
HMF content in unifloral honeys. Food Chemistry, 85, 305–313. 19, 1319–1327.