Food Chemistry 140 (2013) 686–691

Contents lists available at SciVerse ScienceDirect

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

Antioxidant capacity and vasodilatory properties of Mediterranean food: The case of

Cannonau wine, myrtle berries liqueur and strawberry-tree honey
Carlo Ignazio Giovanni Tuberoso a,⇑, Mladen Boban b, Ersilia Bifulco c, Danijela Budimir b,
Filippo Maria Pirisi a
a
Department of Life and Environmental Sciences, University of Cagliari, Via Ospedale 72, 09124 Cagliari, Italy
b
Department of Pharmacology, University of Split, School of Medicine, Soltanska 2, 21000 Split, Croatia
c
Institutt for Indremedisin, Universitetet i Bergen, Laboratoriebygget, 8.etg. Rom 8390, Jonas Lies vei, 5021 Bergen, Norway

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

Article history: The aim of this work was to use different assays to evaluate the antioxidant and vasodilatory properties
Available online 28 September 2012 of three typical food products from the Mediterranean area and to correlate these activities with their
phenolic content. For this purpose, red wines Cannonau, liqueurs obtained by cold maceration of myrtle
Keywords: (Myrtus communis L.) berries and bitter honeys obtained from strawberry-tree flowers (Arbutus unedo L.)
Antioxidant were analysed. The total phenolic (TP) content was measured spectrophotometrically with a modified
Direct vasodilatory effects Folin–Ciocalteau method and phenolic compounds were identified and dosed by HPLC-DAD and LC-
Phenolic compounds
MS/MS. Antioxidant activities were evaluated with DPPH, FRAP and ABTS assays and the in vitro vasod-
Cannonau wine
Myrtle liqueur
ilatory effects were assessed using norepinephrine precontracted rat aortic rings. Cannonau wines and
Strawberry-tree honey myrtle liqueurs showed high levels of TP (1978 ± 279 and 1741 ± 150 mg GAE/L, respectively), linearly
correlated to the results of FRAP, ABTS, and DPPH assays. Their maximal vasodilatory activity was
61.7 ± 4.1% and 53.0 ± 3.0%, respectively. Although strawberry-tree honey contained relatively high levels
of phenolic compounds (922 ± 38 mg GAE/kg), it did not induce vasodilation, even at the highest dose
tested (0.206 g/L). These results indicate that foods with high levels of phenolic compounds should be
studied using several different biological assays before being recommended to the general public as func-
tional foods.
Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction
Nutrition has become a central topic in government strategies
after scientific research clearly demonstrated that typical diseases
of the most industrialized countries, such as cardiovascular
diseases, cancers, and other major causes of mortality can be
prevented by an appropriate intake of nutrients and biologically
active compounds (Willett, 2002). Food supplies major nutrients
(carbohydrates, lipids and proteins) and minor components such
as vitamins or minerals that can exert peculiar activities. Antioxi-
dant activity is one of the properties where the scientific research
focused because a diet rich in antioxidant compounds is funda-
mental in preventing oxidative damages in humans (Nadtochiy &
Redman, 2011; Nisha & Deshwal, 2011). The so-called ‘‘antioxi-
dants’’ found in food is a heterogeneous category of molecules
and among them phenolic compounds are a group of molecules
with proved antioxidant properties. Plant polyphenols are the most
widespread group of secondary metabolites and comprise such di-
⇑ Corresponding author. Tel.: +39 0706758644; fax: +39 0706758612.
E-mail address: tuberoso@unica.it (C.I.G. Tuberoso).
verse classes as simple phenols and phenolic acids (cinnamic and
benzoic acid derivatives), depsides, phenylpropanoid and phenyl-
ethanoid glycosides, flavonoids (flavanones, flavones, flavonols,
flavanols, anthocyanins, isoflavones), stilbenes, and xanthones.
Numerous studies demonstrated their antioxidant activity, in-
creased nitric oxide production and protective effects against LDL
oxidation, and platelets aggregation (Choi, Lee, Hong, & Lee,
2012; Duthie & Crosier, 2000; Mudnic et al., 2009).
The Mediterranean diet, inscribed by the UNESCO in 2010 on
the Representative List of the Intangible Cultural Heritage of
Humanity, is famous worldwide for being rich in antioxidants
and especially in polyphenols. Despite the abundant data on anti-
oxidant activity of several food products, often these data are frag-
mentary or controversial (Kontou, Psaltopoulou, Panagiotakos,
Dimopoulos, & Linos, 2011; Valls-Pedret et al., 2012). The choice
of the antioxidant assay is important because the in vitro assays,
such as DPPH (using 2,2-diphenyl-1-picrylhydrazyl free radical),
deoxyribose, TEAC (Trolox equivalent antioxidant capacity), ORAC
(oxygen absorbing antioxidant capacity) or lipid peroxidation,
may only have limited relevance to in vivo conditions (Tang &
Halliwell, 2010). On the contrary, studies on living models and hu-
mans can give more reliable evidence, but systems are much more

0308-8146/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodchem.2012.09.071
 C.I.G. Tuberoso et al. / Food Chemistry 140 (2013) 686–691
complicated to evaluate and are subjected to physiological inter-
ferences (Boban et al., 2006).
In the present work, the antioxidant properties and in vitro
vasodilatory activity of three representative food products from
the Mediterranean area were investigated and correlated to their
phenolic content. For this purpose, the antioxidant capacity of
the red wine Cannonau, the liqueur obtained by cold maceration
of myrtle (Myrtus communis L.) berries and the bitter honey ob-
tained from strawberry-tree flowers (Arbutus unedo L.) were eval-
uated with different assays (DPPH, FRAP and ABTS assays). The
direct vasodilatory effects were examined on the isolated norepi-
nephrine precontracted rat aortic rings. The total phenols (TP) con-
tent was measured by spectrophotometric determination with a
modified Folin–Ciocalteau method and phenolic compounds were
identified and dosed by HPLC-DAD and LC-MS/MS.
2. Material and methods
2.1. Chemicals
Methanol, acetonitrile, gallic acid, syringic acid, homogentisic
acid, t-resveratrol, p-coumaric acid, caffeic acid, c,t-abscisic acid,
ferrous sulphate, 1,1-diphenyl-2-picrylhydrazyl radical (DPPH),
(±)-6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trol-
ox), 2,4,6-tris(2-pyridyl)-1,3,5-triazine (TPTZ), Folin–Ciocalteau’s
phenol reagent were obtained from Sigma-Aldrich, Fluka (Milan,
Italy). Phosphoric acid 85% w/w, sodium carbonate, ferric chloride
and CuSO4
5H2O were supplied by Carlo Erba (Milan, Italy). Stan-
dard procyanidin B1, procyanidin B2, (+)catechin, ( )epicatechin,
anthocianidin-3-O-glucosides, myricetin, myricetin-3-O-glucoside,
myricetin-3-O-galactoside, myricetin-3-O-rhamnoside, quercetin,
quercetin-3-O-glucoside, quercetin-3-O-rhamnoside and kaempf-
erol-3-O-glucoside were purchased from Extrasynthese (Genay,
France). Norepinephrine and acethilcoline were obtained from Sig-
ma Chemical Co. (St Louis, MO, USA). All the chemicals used in this
study were of analytical grade. Ultrapure water (18 mX) was ob-
tained with a Milli-Q Advantage A10 System apparatus (Millipore,
Milan, Italy).
2.2. Samples
All samples were commercial products from Sardinia (Italy), ob-
tained directly from producer and with known industrial pro-
cesses. Sensory analysis and physical–chemical characteristic
were evaluated to assess typicality of the products. Myrtle berries
liqueurs (n = 5) and Cannonau wines (n = 8) were produced by
2010–2011 harvesting. Wines were stored in stainless steel tanks
without contact with wood (barrels or barrique). Straw-berry three
honeys (n = 7) were produced in 2011 and selected according to
melissopalynological analysis and marker compound detection
(Tuberoso, Bifulco et al., 2010).
2.3. Antioxidant activity assays
2.3.1. In vitro vasodilatory effect
The animal study was approved by the Ethics Committee of the
University of Split School of Medicine. Ten male Wistar rats,
weighting 300 ± 20 g were used. At the beginning of the protocol,
animals received a 1.2 g/kg body weight intraperitoneal injection
of urethane and became unresponsive to noxious stimuli, what
was followed by decapitation and chirurgical procedure. The
descending thoracic aorta was dissected from connective tissue
and cut into small (3–4 mm) rings which were placed into
Krebs–Henseleit solution. After stabilization, rings were precon-
tracted with a test dose of norepinephrine (10 7 mol/L) and
687
functionality of endothelium was confirmed with acetylcholine in-
duced relaxation (10 6 mol/L). The relaxation was expressed as the
percentage decrease of the norepinephrine-induced vasoconstric-
tion. The whole procedure was published previously (Music
et al., 2005; Brizic et al., 2009). Triple washout and tension stabil-
ization of the rings was followed by next protocol: after precon-
traction with norepinephrine, rings were exposed to cumulative
concentrations of the samples (1–8:1000 final dilutions in organ
baths) to determine dose-dependent vasodilatory effect.
2.3.2. DPPH assay
A spectrophotometric analysis using DPPH, and a comparison
with the Trolox calibration curve was performed (Tuberoso, Rosa
et al., 2010). This assay is based on the ability of the antioxidant
to scavenge the radical cation 1,1-diphenyl-2-picrylhydrazyl radi-
cal (DPPH). Fifty microlitres of diluted sample (1:5–1:10, v/v or
w/v, with water) was dissolved in 2 mL of 0.04 mmol/L DPPH in
methanol. A calibration curve in the range 0.05–1.0 mmol/L was
used for the Trolox, and data were expressed as Trolox equivalent
antioxidant capacity (TEAC, mmol/L or mmol/kg). Spectrophoto-
metric readings were carried out with a Cary 50 spectrophotome-
ter (Varian, Leinì, TO, Italy) at 517 nm, using a 10 mm plastic
cuvette after an incubation period of 60 min in the dark.
2.3.3. ABTS assay
The ABTS assay is based on the inhibition of the absorbance of
the radical cation 2,2’-azino-bis(3-ethylbenzothiazoline-6-sulfo-
nate) (ABTS+) solution when it is exposed to an antioxidant (Re
et al., 1999). The ABTS+ cation radical was produced by the reac-
tion between 10 mL of 2 mM ABTS in H2O and 100 lL of 70 mM
potassium persulphate, stored in the dark at room temperature
for 24 h. The ABTS+ solution was then diluted with methanol to ob-
tain an absorbance of 0.70 at k = 734 nm and equilibrated at 30 °C.
Samples were prepared in triplicate by diluting 50 lL of extracts
(same dilution of DPPH assay) in 2 mL of the ABTS+ solution di-
luted with methanol and let to react for 1 min. Absorbances were
recorded on a Cary 50 spectrophotometer at 734 nm. The antioxi-
dant activities samples are expressed as TEAC values, defined as
the concentration of standard Trolox with the same antioxidant
capacity of the extract under investigation.
2.3.4. FRAP assay
The FRAP (ferric reducing antioxidant power) assay evaluate
antioxidants as reductants of Fe3+ to Fe2+, which is chelated by
2,4,6-tris(pyridin-2-yl)-1,3,5-triazine (TPTZ) to form a Fe2+–TPTZ
complex absorbing at 593 nm (Benzie & Strain, 1996, Tuberoso,
Rosa et al., 2010). The FRAP assay was performed preparing a ferric
complex TPTZ and Fe3+ (0.3123 g TPTZ, 0.5406 g FeCl3
6H2O in
100 mL acetate buffer pH 3.6). Twenty lL of sample (same dilution
for DPPH assay), was dissolved in 2 mL of ferric complex and, after
an incubation period of 4 min in the dark, absorbance at 593 nm
was measured with a Cary 50 spectrophotometer. Quantitative
analysis was performed according to the external standard method
(FeSO4, 0.1–2 mmol/L), correlating the absorbance with the con-
centration and results were expressed as mmol/L or mmol/kg of
Fe2+.
2.4. Polyphenols analysis
2.4.1. Total polyphenols
The total phenol content was measured spectrophotometrically
with modified Folin–Ciocalteau’s method (Tuberoso, Rosa et al.,
2010). One hundred microlitres of the sample diluted with water
(same dilution of DPPH assay) was added to 0.5 mL of Folin–
Ciocalteau’s phenol reagent. After 5 min, 3 mL of 10% Na2CO3
(w/v) was added, the mixture was shaken, and then diluted with
688 C.I.G. Tuberoso et al. / Food Chemistry 140 (2013) 686–691
water to a final volume of 10 mL. After a 90 min incubation period
at room temperature, the absorbance was read at 725 nm on a
10 mm quartz cuvette using a Varian Cary 50 Scan spectrophotom-
eter, against a blank. The total polyphenol content results,
expressed as mg/L or mg/kg of gallic acid equivalent (GAE), were
obtained using a calibration curve of a freshly prepared gallic acid
standard solution (10–200 mg/L).
2.4.2. HPLC-DAD
Detection and quantitative analyses of phenolic compounds
were carried out using an HPLC-DAD method as described by
Tuberoso, Rosa et al. (2010). An HPLC Varian system ProStar was
employed, fitted with a pump module 230, an autosampler module
410, and a ThermoSeparation diode array detector SpectroSystem
UV 6000lp (ThermoSeparation, San Jose, CA), set at 280, 360 and
520 nm. Separation was obtained with a Gemini C18 column
(150  4.60 mm, 3 lm, Phenomenex, Casalecchio di Reno, BO,
Italy) using 0.2 M phosphoric acid (solvent A), and acetonitrile (sol-
vent B) at a constant flow rate of 1.0 mL/min. The gradient (v/v)
was generated decreasing from 90% of solvent A to 85% in 2 min;
to 65% in 18 min; to 50% in 30 min; and to 10% in 40 min. Before
each injection, the system was stabilized for 10 min with the initial
A/B ratio (90:10, v/v). The injection volume was 10 lL. Chromato-
grams and spectra were elaborated with a ChromQuest V. 2.51 data
system (ThermoQuest, Rodano, Milan, Italy). Anthocyanins were
detected and dosed at 520 nm, flavonols at 360 nm, and all the
other compounds at 280 nm.
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. Gallic acid derivatives
were quantified as gallic acid. Quercetin-3-O-glucuronide was
dosed as quercetin-3-O-glucoside and myricetin-3-O-arabinoside
as myricetin-3-O-glucoside. The caffeic and p-coumaric tartaric es-
ters (caftaric and coutaric acids) were quantified using the calibra-
tion curves of caffeic and p-coumaric acid, respectively. The
samples were diluted with ultrapure water (1:20 v/v for wine
and myrtle liqueur, and 1:10 for honey, w/v), filtered through
Econofilter RC membrane (0.45 lm, Ø 25 mm, Agilent Technolo-
gies, Milan, Italy) and injected in HPLC without any further
purification.
2.4.3. LC–MS/MS
An HPLC–MS/MS Varian (Varian Palo Alto, CA, USA) system fit-
ted with two ProStar 210 pumps, an autosampler 410 ProStar, a
ProStar 330 PDA detector and a 1200L triple quadrupole mass
spectrometer with electrospray ionization source (ESI), was em-
ployed. Varian MS workstation version 6.6 software was used for
data acquisition and processing. The solvents used were water (sol-
vent A) and methanol (solvent B) at a constant flow rate of 0.3 mL/
min. The gradient (v/v) was generated decreasing from 90% of sol-
vent A to 65% in 15 min; to 50% in 30 min; and to 10% in 40 min.
Before each injection, the system was stabilized for 10 min with
the initial A/B ratio (90:10, v/v). Separation was obtained with
the Gemini C18 column and the injection volume was 10 lL. The
system was optimized in the positive mode for anthocyanins and
in negative for the other phenolic compounds. For the negative
mode, the electrospray capillary potential was set to 55 V. Air,
as desolvatation solvent gas, was used at 300 °C, while the housing
API temperature was kept at 41 °C. Deprotonated analyte mole-
cules of the parent compounds were subjected to collision-induced
dissociation using argon at 2.18 mTorr in the multiple reaction
monitoring (MRM) mode. The scanning time was 0.2 s, and the
voltage detector was set to 1500 V. For the anthocyanin molecules
the system was optimized to work in positive mode, and the
electrospray capillary potential was set to 30 V. Nitrogen at
365 °C was used as a desolvatation solvent gas, while the housing
API temperature was kept at 54 °C. Protonated analyte molecules
of the parent compounds were subjected to collision-induced dis-
sociation using argon at 2.18 mTorr in the multiple reaction mon-
itoring (MRM) mode. The scanning time was 0.2 s, and the voltage
detector was set to 1400 V. Pattern fragmentation was optimized
for [M]+, [M+H]+, and [M glycoside]+ detection in the positive
mode, and for [M H] and [M glycoside] detection in the nega-
tive mode. Also adducts with Na and K, and dimers were useful to
confirm peaks’ attribution.
2.5. Statistical analyses
The Graph Pad INSTAT software (GraphPad software, San Diego,
CA, USA) was used to calculate the means and standard deviations
of two or three independent experiments, involving triplicate or
duplicate analyses for each sample/condition. Evaluation of statis-
tical significance of differences was performed by a one-way anal-
ysis of variance (one-way ANOVA). Statistical analysis of
vasodilation responses was done using a 2-way analysis of vari-
ance, followed by Bonferroni post hoc test. All data are expressed
as means ± standard error of the means. P < 0.05 was considered
statistically significant. Because the tested samples differed signif-
icantly in total phenolic content, dilution and logarithm of dilution,
instead of concentration, were used to calculate EC50. EC50 was cal-
culated using nonlinear regression analysis.
3. Results and discussion
The sensory characteristics and the basic chemical–physical
analyses (pH, total acidity and, only for wine and liqueur, the alco-
hol percentage; data not shown) confirmed the typicity of the sam-
pled food. Cannonau wine is a typical red wine that, according to
ampelographical evidence, is closely related to the most wide-
spread and known Grenache or Garnacha. In Sardinia documented
historical traces since the XVI century can be found (Vodret, 2003).
Cannonau wine is protected with the controlled origin denomina-
tion (DOC) of ‘‘Cannonau di Sardegna’’ since 1972 (Gazzetta
Ufficiale no. 272, 1992) and in the last years producers are working
to upgrading it to the ‘‘controlled and guarantee origin denomina-
tion’’ (DOCG). Myrtle liqueur is traditionally obtained by 4–
5 weeks maceration of the violet-coloured mature berries with
ethanol at room temperature. Next, the extract is diluted in water
and sugar until the alcoholic grade is adjusted to 28–32% v/v and
the sugar content ranges between 20% and 30% w/v. A minimum
of 150 g of berries should be used to make a litre of liqueur and
no aromatic or preservative substances are added. The liqueur
‘‘Mirto di Sardegna’’ has been legally recognized by the European
Union as ‘‘spirit drinks with geographical designations’’. The anti-
oxidant capacity of industrial myrtle liqueurs measured with the
DPPH assay, showed values ranging from 11.67 to 12.89 mM Trol-
ox (Vacca, Piga, Del Caro, Fenu, & Agabbio, 2003), and from 43.2% to
76.5% inhibition (Alamanni & Cossu, 2004). Strawberry-tree ( A.
unedo L.) honey is known as ‘‘bitter honey’’ and is a typical product
of some Mediterranean regions like Sardinia, Corsica and restricted
areas of Spain, and Croatia. Homogentisic acid (2,5-dihydroxy-
phenylacetic acid) was reported to be the most abundant phenolic
compound in this honey, with an average amount of
414.1 ± 69.8 mg/kg (Tuberoso, Bifulco et al., 2010). This unifloral
honey emerged as the most active honey in the DPPH and FRAP
tests (4.8 ± 0.8 mmol TEAC/kg and 11.7 ± 1.7 mmol Fe2+/kg, respec-
tively) and the richest in total phenols (972 mg/kg GAE) (Rosa
et al., 2011).
 C.I.G. Tuberoso et al. / Food Chemistry 140 (2013) 686–691 689

Antioxidant activities of strawberry-tree honey found in this 100

experimentation were 12.0 ± 2.2 mmol Fe2+/kg (FRAP), 90 Cannonau wine
5.9 ± 1.5 mmol TEAC/kg (ABTS), and 4.5 ± 1.1 mmol TEAC/kg
80
(DPPH). These values are about half of those for Cannonau wine Myrtle liqueur
70

and myrtle liqueur which did not statistically differ between each
other (Table 1). Also, the vasodilatory activities of Cannonau wine
and myrtle liqueur were dose-dependent and similar. Maximal
relaxation (Emax) for the wine and liqueur samples was 62 ± 4%
and 53 ± 3%, respectively (P > 0.05). Cannonau wine and myrtle li-
queur proved to be vasodilators of similar potency, based on their
effective concentrations 50 (EC50) (4.69‰ vs. 4.97‰, P > 0.05). In
contrast to the Cannonau wine and Myrtle liqueur, the straw-
berry-tree honey showed almost no vasodilatory activity (Emax
1.1 ± 0.1%, P < 0.001) (Fig. 1). This was rather surprising since the
strawberry-tree honey contained relatively high levels of phenolic
compounds. It did not induce significant vasodilation, even at the
highest dose tested (0.206 g/L) (data not shown). This could be ex-
plained by the significant differences in the phenolic composition
of strawberry-tree honey relative to Cannonau wine and myrtle li-
queur. Namely, homogentisic acid as the most abundant phenolic
compound in this honey might not be an effective vasodilator. This
would be in line with our previous study in which we determined
and correlated antioxidative and vasodilatory activities of nine
phenolic acids from wine. We found that no correlation exists be-
tween antioxidative capacity and maximal vasodilatory effect of
the acids, and that determining antioxidative capacity of phenolic
compounds in vitro has no relevance for the prediction of their
other biological effects (Mudnic et al., 2010).
Antioxidant capacity measured with FRAP, ABTS and DPPH
assays was linearly correlated to total phenols measured with the
Folin–Ciocalteau’s assay, ranging from a minimum value of
rTP/ABTS+ = 0.8913 for Cannnau wine to a maximum of rTP/ABTS+ =
0.9529 for strawberry-tree honey (data not shown). Cannonau
wines and myrtle liqueurs showed comparable values of total
phenols (1978 ± 279 and 1741 ± 150 mg GAE/L, respectively), and
strawberry-tree honey contained almost half the quantity
(922 ± 38 mg GAE/kg). HPLC-DAD analysis was used to investigate
the phenolic fraction and the LC–MS/MS analysis was used to
confirm peaks’ attribution (Table 2). Acquired chromatograms at
280 nm were used to quantify hydroxybenzoic acids, hydroxycin-
namic acids, flavanols and t-resveratrol. Chromatograms at 360
and 520 nm were used to quantify flavonols and anthocyanins,
respectively. Cannonau wine and myrtle liqueur show several
common compounds although peculiarities can be noticed. For
instance, both Cannonau wine and myrtle liqueur are rich in
anthocyanins, with malvidin-3-O-glucoside the most representa-
tive, followed by petunidin-3-O-glucoside and delphinidin-3-O-
glucoside. Relatively high contents of anthocyanins in the Cannonau
relaxation (%)
Strawberry tree honey
60
50
40
30
20
10
* * * *
0
0 1 2 3 4 5 6 7 8
sample ‰
Fig. 1. Relaxation in the norepinephrine precontracted rat aortic rings induced by
test Cannonau wine, myrtle liqueur and strawberry-tree honey samples, n = 10.
Data are shown as mean ± S.E.M. ⁄P < 0.05 honey vs. liqueur and wine.
wine and the myrtle liqueur in comparison with the strawberry-tree
honey might be an additional argument for the observed differences
in vasodilatory activity. Namely, the anthocyanins have been shown
by several authors as the crucial phenolic fraction associated with
the direct vasodilatory activity (Burns et al., 2000; Mudnic et al.,
2012). Myrtle liqueur showed arabinoside derivatives (11 ± 3 mg/
L) that were absent in Cannonau wine, while no traces of acetate
and p-coumarate derivatives were found. These compounds were
detectable in Cannonau. Gallic acid and t-resveratrol were also de-
tected in Cannonau wines, that present a detectable quantity of
hydroxycinnamic acids and t-caftaric acid is the most represented
(2 ± 0 mg/L). It is interesting to note that Cannonau wine and myrtle
liqueur contain similar flavonols, but they differ in the type of glyco-
sides. Myrtle liqueur is characterized by myricetin-3-O-arabinoside
(not detected in Cannonau wine), and quercetin-3-O-glucuronide
and kaempferol-3-O-glucoside were detected only in Cannonau
wine. HPLC-DAD analysis of strawberry-tree honey showed homo-
gentisic acid (425 ± 76 mg/kg) as the most prominent phenolic
compound. These differences in the type of phenolic compound
can explain the different response of Cannonau wine, myrtle liqueur
and strawberry-tree honey to FRAP, ABTS and DPPH assays and the
vasodilatory activities. FRAP, ABTS and DPPH assays are typical
methods that involve redox reaction similar to that of Folin–
Ciocalteau’s assay: for this reason a good correlation between these
data are expected. These findings imply that biochemical in vitro
data cannot be directly transferred to the effects in more complex

Table 1
Antioxidant capacity and vasodilatory properties of Cannonau wine, myrtle liqueur and strawberry-tree honey.

Antioxidant and vasodilatory properties Food product

Cannonau wine (n = 8) Myrtle liqueur (n = 5) Strawberry-tree honey (n = 7)
In vitro antioxidant assays
FRAP (mmol Fe2+/L⁄)a 27.2 ± 5.2 26.7 ± 4.7 12.0 ± 2.2
DPPH (mmol TEAC/L⁄)b 7.9 ± 1.3 9.3 ± 0.6 4.5 ± 1.1
ABTS+ (mmol TEAC/L⁄)b 9.3 ± 1.4 11.5 ± 0.9 5.9 ± 1.5
In vitro vasodilation
Maximal vasodilatory activity, Emax (%) 61.7 ± 4.1 53.0 ± 3.0 1.1 ± 0.1
Effective concentration 50, EC50 (‰) 4.69 4.97 0.01
⁄
Strawberry-tree honey data are referred to kg of product.
a
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 food
product.
b
DPPH and ABTS values are expressed as TEAC millimolar concentration, obtained from a Trolox solution having an antiradical capacity equivalent to that of the dilution of
the food product.
690 C.I.G. Tuberoso et al. / Food Chemistry 140 (2013) 686–691

Table 2
Phenolic composition of Cannonau wine, myrtle liqueur and strawberry-tree honey (mg/L⁄).

Compound Identification Food product

Cannonau wine (n = 8) Myrtle liqueur (n = 5) Strawberry-tree honey (n = 7)
Total phenols (mg GAEa/L⁄) 1978 ± 279 1741 ± 150 922 ± 38
Phenols by HPLC (mg/L⁄) 170 ± 52 277 ± 61 428 ± 77
Hydroxybenzoic acids 28 ± 14 18 ± 8 428 ± 77
Gallic acid rt, UV–Vis, MS 24 ± 12 12 ± 7 3±2
Gallic acid derivatives UV–Vis, MS 2±1 6±3 nd
Syringic acid rt, UV–Vis, MS 3±3 nd nd
Homogentisic acid rt, UV–Vis, MS nd nd 425 ± 76
Hydroxycinnamic acids 4±1 nd nd
c-Caftaric UV–Vis, MS 1±0 nd nd
t-Caftaric UV–Vis, MS 2±0 nd nd
c-Coutaric UV–Vis, MS 0±0 nd nd
t-Coutaric UV–Vis, MS 1±1 nd nd
STILBENES 1±0 nd nd
t-Resveratrol rt, UV–Vis, MS 1±0 nd nd
Flavanols 48 ± 17 25 ± 9 nd
Procyanidin B1 rt, UV–Vis, MS 23 ± 10 nd nd
(+)Catechin rt, UV–Vis, MS 10 ± 3 25 ± 9 nd
Procyanidin B2 rt, UV–Vis, MS 13 ± 7 nd nd
(-)Epicatechin rt, UV–Vis, MS 2±1 nd nd
Flavonols 19 ± 6 124 ± 25 nd
Myricetin-3-O-arabinoside UV–Vis, MS nd 51 ± 11 nd
Myricetin-3-O-glucoside rt, UV–Vis, MS 1±0 nd nd
Myricetin-3-O-galactoside rt, UV–Vis, MS nd 34 ± 8 nd
Myricetin-3-O-rhamnoside rt, UV–Vis, MS nd 3±1 nd
Quercetin-3-O-glucuronide UV–Vis, MS 3±2 nd nd
Quercetin-3-O-glucoside rt, UV–Vis, MS 10 ± 3 7±2 nd
Quercetin-3-O-rhamnoside rt, UV–Vis, MS nd 6±3 nd
Kaempferol-3-O-glucoside rt, UV–Vis, MS 1±0 nd nd
Myricetin rt, UV–Vis, MS 2±1 20 ± 9 nd
Quercetin rt, UV–Vis, MS 3±3 3±1 nd
Anthocyanins 69 ± 21 110 ± 40 nd
Delphinidin-3-O-glucoside rt, UV–Vis, MS 2±1 20 ± 11 nd
Cyanidin-3-O-glucoside rt, UV–Vis, MS 0±0 5±3 nd
Petunidin-3-O-glucoside rt, UV–Vis, MS 3±2 22 ± 10 nd
Peonidin-3-O-glucoside rt, UV–Vis, MS 2±2 5±2 nd
Malvidin-3-O-glucoside rt, UV–Vis, MS 55 ± 16 57 ± 18 nd
Malvidin-3-O-acetylglucoside UV–Vis, MS 3±2 nd nd
Malvidin-3-O-t-p-coumarylglucoside UV–Vis, MS 3±1 nd nd
Anthocyanins arabinoside UV–Vis, MS nd 11 ± 3 nd

nd: not detected (below the limit of detection).

rt: comparison with retention time of pure standard.
UV–Vis: comparison with typical UV–Vis spectra of pure compound or similar pure standards.
MS: comparison with MS spectra of pure compound or literature data.
⁄
Strawberry-tree honey data are referred to kg of product.
a
GAE: gallic acid equivalent.

biological systems, especially not to the actions in the whole
organism (Halliwell, Aeschbach, Loliger, & Aruoma, 1995).
4. Conclusions
These results indicate that foods with high levels of phenolic
compounds should be studied using several different biological
assays, before being recommended to the general public as func-
tional foods. Given the current data on in-vitro and ex-vivo allevia-
tion of oxidative stress and in vitro vasodilatory activity, Cannonau
wine and myrtle liqueur are very promising nutraceutical foods.
Future studies should focus on extensive mechanistic investigation
into various types of bioactivity, complete pharmacokinetic stud-
ies, as well as on clinical evaluation of both pure phenolic com-
pounds, and complex extracts or fractions from the studied foods
to be used as supplements with antioxidant activities.
References
Alamanni, M. C., & Cossu, M. (2004). Radical scavenging activity and antioxidant
activity of liquors of myrtle (Myrtus communis L.) berries and leaves. Italian
Journal of Food Science, 16, 197–208.
Benzie, I. F., & Strain, J. J. (1996). The ferric reducing ability of plasma (FRAP) as a measure
of ‘‘antioxidant power’’: The FRAP assay. Analytical Biochemistry, 239, 70–76.
Brizic, I., Modun, D., Vukovic, J., Budmir, D., Katalinic, V., & Boban, M. (2009).
Differences in vasodilatory response to red wine in rat and guinea pig aorta.
Journal of Cardiovascular Pharmacology, 53, 116–120.
Boban, M., Modun, D., Music, I., Vukovic, J., Brizic, I., Salamunic, I., Obad, A., Palada, I.,
& Dujic, Z. (2006). Red wine induced modulation of vascular function:
Separating the role of polyphenols, ethanol, and urates. Journal of
Cardiovascular Pharmacology, 47, 695–701.
Burns, J., Gardner, P. T., O’Neil, J., Crawford, S., Morecroft, I., McPhail, D. B., Lister, C.,
Matthews, D., MacLean, M. R., Lean, M. E., Duthie, G. G., & Crozier, A. (2000).
Relationship among antioxidant activity, vasodilation capacity, and phenolic
content of red wines. Journal of Agricultural and Food Chemistry, 48, 220–230.
Choi, D.-Y., Lee, Y.-J., Hong, J. T., & Lee, H.-J. (2012). Antioxidant properties of natural
polyphenols and their therapeutic potentials for Alzheimer’s disease. Brain
Research Bulletin, 87. 144-15.
Duthie, G., & Crosier, A. (2000). Plant-derived phenolic antioxidants. Current Opinion
in Clinical Nutrition and Metabolic Care, 3, 447–451.
Gazzetta Ufficiale – Serie Generale no. 272, 18-11-1992. Modificazione al
disciplinare di produzione della denominazione di origine controllata dei vini
‘‘Cannonau di Sardegna’’.
Halliwell, B., Aeschbach, R., Loliger, J., & Aruoma, O. I. (1995). The characterization of
antioxidants. Food and Chemical Toxicology, 33, 601–617.
Kontou, N., Psaltopoulou, T., Panagiotakos, D., Dimopoulos, M. A., & Linos, A. (2011).
The Mediterranean diet in cancer prevention: A review. Journal of Medicinal
Food, 14, 1065–1078.
Mudnic, I., Modun, D., Brizic, I., Vukovic, J., Generalic, I., Katalinic, V., Bilusic, T.,
Ljubenkov, I., & Boban, M. (2009). Cardiovascular effects in vitro of aqueous
extract of wild strawberry (Fragaria vesca, L.) leaves. Phytomedicine, 16, 462–469.
 C.I.G. Tuberoso et al. / Food Chemistry 140 (2013) 686–691
Mudnic, I., Modun, D., Rastija, V., Vukovic, J., Brizic, I., Katalinic, V., Kozina, B., Medic-
Saric, M., & Boban, M. (2010). Antioxidative and vasodilatory effects of phenolic
acids in wine. Food Chemistry, 119, 1205–1210.
Mudnic, I., Budimir, D., Modun, D., Gunjaca, G., Generalic, I., Skroza, D., Katalinic, V.,
Ljubenkov, I., & Boban, M. (2012). Antioxidant and vasodilatory effects of
blackberry and grape wines. Journal of Medicinal Food, 15, 315–321.
Musić, I., Modun, D., Katalinić, V., Salamunić, I., Kozina, B., & Boban, M. (2005).
Effects of four-week moderate drinking of red wine and ethanol on the rat
isolated heart and aortic rings reactivity during ischemia and hypoxia.
Periodicum Biologorum, 107, 165–173.
Nadtochiy, S. M., & Redman, E. K. (2011). Mediterranean diet and cardioprotection:
The role of nitrite, polyunsaturated fatty acids, and polyphenols. Nutrition, 27,
733–744.
Nisha, K., & Deshwal, R. K. (2011). Antioxidants and their protective action against
DNA damage. International Journal of Pharmacy and Pharmaceutical Sciences, 3,
28–32.
Re, R., Pellegrini, N., Proteggente, A., Pannala, A., Yang, M., & Rice-Evans, C. (1999).
Antioxidant activity applying an improved ABTS radical cation decolorization
assay. Free Radical Biology and Medicine, 26, 1231–1237.
Rosa, A., Tuberoso, C. I. G., Atzeri, A., Melis, M. P., Bifulco, E., & Dessì, M. A. (2011).
Antioxidant profile of strawberry tree honey and its marker homogentisic acid
in several models of oxidative stress. Food Chemistry, 129, 1045–1053.
691
Tang, S. Y., & Halliwell, B. (2010). Medicinal plants and antioxidants: What do we
learn from cell culture and Caenorhabditis elegans studies? Biochemical and
Biophysical Research Communications, 394, 1–5.
Tuberoso, C. I. G., Bifulco, E., Caboni, P., Cottiglia, F., Cabras, P., & Floris, I. (2010).
Floral markers of strawberry tree (Arbutus unedo L.) honey. Journal of
Agricultural and Food Chemistry, 58, 384–389.
Tuberoso, C. I. G., Rosa, A., Bifulco, E., Melis, M. P., Atzeri, A., Pirisi, F. M., & Dessì, M.
A. (2010). Chemical composition and antioxidant activities of Myrtus communis
L. berries extracts. Food Chemistry, 123, 1242–1251.
Vacca, V., Piga, A., Del Caro, A., Fenu, P. A. M., & Agabbio, M. (2003). Changes in
phenolic compounds, colour and antioxidant activity in industrial red
myrtle liqueurs during storage. Zeitschrift fur Naturforschung C, 47, 442–
447.
Valls-Pedret, C., Lamuela-Raventós, R. M., Medina-Remón, A., Quintana, M., Corella,
D., Pintó, X., Martínez-González, M. Á., Estruch, R., & Ros, E. (2012). Polyphenol-
rich foods in the Mediterranean diet are associated with better cognitive
function in elderly subjects at high cardiovascular risk. Journal of Alzheimer’s
Disease, 29, 773–782.
Vodret, A. (2003) Storia regionale della vite e del vino in Sardegna. Ed. Unione
Italiana Vini, Milan, Italy.
Willett, W. C. (2002). Balancing life-style and genomics research for disease
prevention. Science, 296, 695–698.