Phenolic compounds and antioxidant capacity of virgin olive oil

Abstract

The characterisation of virgin olive oil from Arbequina, Carrasqueña, Corniche, Manzanilla Cacereña, Morisca, Picual, and Verdial de Badajoz varieties according to the individual phenolic compounds at different ripening stage was carried out. In all olive oil varieties studied, secoiridoid derivatives were most abundant, followed by phenolic alcohols, flavonoids and phenolic acids. The secoiridoid derivatives of hydroxytyrosol were the most important complex phenols for Picual and Carrasqueña, whereas the tyrosol derivatives were the major ones found in Manzanilla Cacereña, and Verdial de Badajoz. For secoiridoid derivatives of hydroxytyrosol and tyrosol, Arbequina was the oil variety showing the lowest concentration. Tyrosol, hydroxytyrosol, vanillic acid, p-cumaric acid, luteolin, and apigenin levels were greater in early harvested samples in almost all oils analysed. Antioxidant activity measurements (antiradical, lipid peroxide inhibition, H2O2 and NO scavenging) were also accomplished for the seven varieties in the first ripening stage.

Full text

This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/authorsrights

Author's personal copy
Phenolic compounds and antioxidant capacity of virgin olive oil
M
a
Nieves Franco
a
, Teresa Galeano-Díaz
b
, Óscar López
c
, José G. Fernández-Bolaños
c
, Jacinto Sánchez
a,
⇑
,
Concepción De Miguel
d
,M
a
Victoria Gil
e
, Daniel Martín-Vertedor
a
a
Technological Agri-Food Institute (INTAEX), Avda. Adolfo Suárez s/n, 06071 Badajoz, Spain
b
Analytical Chemistry Department, University of Extremadura, 06006 Badajoz, Spain
c
Organic Chemistry Department, University of Seville, 41012 Seville, Spain
d
Vegetal Biology, Ecology and Soil Sciences Department, University of Extremadura, 06007 Badajoz, Spain
e
Organic and Inorganic Chemistry Department, University of Extremadura, 06006 Badajoz, Spain
article info
Article history:
Received 16 December 2013
Received in revised form 8 April 2014
Accepted 24 April 2014
Available online 10 May 2014
Keywords:
Monovarietal virgin olive oil
Stage of maturation
Phenolic compounds
Antioxidant activity
abstract
The characterisation of virgin olive oil from Arbequina, Carrasqueña, Corniche, Manzanilla Cacereña,
Morisca, Picual, and Verdial de Badajoz varieties according to the individual phenolic compounds at
different ripening stage was carried out. In all olive oil varieties studied, secoiridoid derivatives were
most abundant, followed by phenolic alcohols, flavonoids and phenolic acids. The secoiridoid derivatives
of hydroxytyrosol were the most important complex phenols for Picual and Carrasqueña, whereas the
tyrosol derivatives were the major ones found in Manzanilla Cacereña, and Verdial de Badajoz. For
secoiridoid derivatives of hydroxytyrosol and tyrosol, Arbequina was the oil variety showing the lowest
concentration. Tyrosol, hydroxytyrosol, vanillic acid, p-cumaric acid, luteolin, and apigenin levels were
greater in early harvested samples in almost all oils analysed. Antioxidant activity measurements
(antiradical, lipid peroxide inhibition, H
2
O
2
and NO scavenging) were also accomplished for the seven
varieties in the first ripening stage.
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
A large number of phenolic compounds are present in virgin
olive oil (VOO). Although they comprise a minor family (found at
levels of mg/kg), they are responsible for the nutritional and
organoleptic properties, together with the higher oxidative
stability of this oil, in comparison with the rest of edible vegetable
oils (Velasco & Dobarganes, 2002). Furthermore, it has been
suggested that high concentrations of phenolic compounds in olive
oil may contribute to the healthy action of the Mediterranean diet
(Paiva-Martins et al., 2010) because they exhibit protective effects
against neuro-degenerative and cardiovascular diseases and even
show antiproliferative effects (Owen et al., 2000), contributing at
the same time to protect the organism against oxidative damage
(Valavanidis et al., 2004). Remarkably, the European Food Safety
Authority (EFSA) has recently claimed (http://www.efsa.europa.
eu/en/efsajournal/doc/2033.pdf) that ‘‘the consumption of olive
oil rich in polyphenols (hydroxytyrosol, 5 mg/day) contributes to
the protection of oxidative damage to lipids in blood’’.
A wide range of phenolic compounds have been identified in
virgin olive oil, including phenolic alcohols, secoiridoid deriva-
tives, phenolic acids, lignans, and flavonoids (Artajo, Romero,
Suárez, & Motilva, 2007). Traditionally, separation and quantifica-
tion of phenolic compounds in the extract isolated from VOO by
liquid–liquid extraction or by solid phase extraction have been
carried out by HPLC analysis coupled mostly with UV, electro-
chemical and mass spectrometry detection systems (Ouni et al.,
2011). Using mainly HPLC and phenolic compound profiles, sev-
eral studies have reported the differentiation of VOOs according
to their cultivar or geographical origin (Allalout et al., 2009) sug-
gesting that the oil antioxidant content is not constant; it depends
on the cultivar, fruit ripening stage, agroclimatic conditions and
olive growing techniques (Tovar, Romero, Alegre, Girona, &
Motilva, 2002; Uceda & Hermoso, 2001). Therefore, depending
on the minor components that are present and their ratio, a wide
variety of VOOs can be considered, with different biologic value.
See Figs. 1 and 2.
Olive fruit undergoes a series of changes during ripening.
Accordingly, it is necessary to develop a methodology for deter-
mining the best harvesting time for each variety, in order to opti-
mise the productivity of groves and to obtain high-quality olive
oil with an antioxidant profile leading to health benefits. Several
http://dx.doi.org/10.1016/j.foodchem.2014.04.091
0308-8146/Ó 2014 Elsevier Ltd. All rights reserved.
⇑
Corresponding author. Tel.: +34 924 012664; fax: +34 924 012674.
E-mail address: jacintojesus.sanchez@gobex.es (J. Sánchez).
Food Chemistry 163 (2014) 289–298
Contents lists available at ScienceDirect
Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Author's personal copy
authors have reported the influence of harvesting time in VOO
quality (Ouni et al., 2011). This illustrates the necessity to deter-
mine the best ripening stage for the picking and processing of
each olive cultivar. For that reason, the aim of this work is to
evaluate the individual phenolic profiles and the analyze of the
antioxidant activity profile of virgin olive oils obtained from the
seven olive varieties most representative of the southwest of
Spain (Arbequina, Carrasqueña, Corniche, Manzanilla Cacereña,
Morisca, Picual, and Verdial de Badajoz) at different stages of
ripening. Due to the relevant properties associated with polyphe-
nols, a correlation between the content of such minor compounds
and the antioxidant profile with the stage of the fruit ripening for
the seven selected varieties can provide valuable information to
the olive oil sector. Such information would indicate the best
ripening stage in each variety for which an oil with optimum
properties regarding nutritional, organoleptic and health aspects
can be produced. This is, therefore, the main goal of this
manuscript.
2. Material and methods
2.1. Samples
The study was carried out in an experimental olive (Olea
europaea L.) cultivar maintained by the Researcher Centre ‘‘Finca
La Orden’’ (Badajoz, Spain) within the limits of the olive-growing
area ‘‘Tierra de Barros’’ during the olive seasons 2011/2012 and
2012/2013. The UTM coordinates of cultivar were x = 702192.29,
y = 4303406.71. The climate of the area is Mediterranean and the
average annual rainfall was found to be 404 and 501 mm in
2011/2012 and 2012/2013 crop seasons, respectively. The olive
orchard was composed of fifteen-year-old olive trees (plantation
frame 6  6m
2
) of seven varieties that predominate in the
southwest of Spain: Arbequina, Carrasqueña, Corniche, Manzanilla
Cacereña, Manzanilla Sevillana, Picual, and Verdial de Badajoz. The
soil at the experimental orchard was a sandy loam (depth 2 m). The
orchard was managed under drip irrigation with linear irrigation
min5 10 15 20 25 30 35 40
mAU
0
50
100
150
200
250
300
350
400
min5 10 15 20 25 30 35 40
mAU
0
20
40
60
80
1
2
I.S. (1)
3
4
5
6
I.S. (2)
7
8
9
11
12
13
14
15
16
I.S. (2)
335 nm
280 nm
10
Fig. 1. HPLC chromatogram of the phenolic extract olive oil. Signal was recorded at f 280 and 335 nm. Peak assignments:1, hydroxytyrosol (3,4 DHPEA); 2, tyrosol (p-HPEA);
IS(1), p-hydroxyphenylacetic acid (internal standard); 3, vanillic acid; 4, vanillin; 5, p-coumaric acid; 6, 3,4-DHPEA-AC [hydroxytyrosol acetate, 4 (acetoxyethyl)-1,2-
dihydroxybenzene]; IS(2), o-cumaric acid; 7, 3,4-DHPEA EDA (dialdehydic form of elenoic acid linked to hydroxytyrosol); 8, pinoresinol; 9, 1-acetoxypinoresinol; 10, p-HPEA-
EDA (dialdehydic form of elenoic acid linked to tyrosol); 11, oleuropein aglycone (3,4-DHPEA-EA); 12, ligstroside aglycone (p-HPEA-EA); 13, caffeic acid; 14, ferulic acid; 15,
luteolin; 16, lpigenin.
290 M
a
Nieves Franco et al. / Food Chemistry 163 (2014) 289–298

Author's personal copy
scheduling, of 3582 cm
3
water/ha (15th May to 18th November)
and no tillage conditions; weeds were controlled with post-
emergence herbicides. It is noticeable that the fact of getting the
entire test samples from the same experimental olive grow
allowed to study the effect of the variables ‘‘variety’’ and ‘‘maturity
stage’’ on the minor components of fruits developed in the same
geographical area, under the same agronomic and pedoclimatic
conditions.
In the selection of the samples in our experiments we used a
statistical design which is based in randomised blocks with three
replicates for each variety, each elementary block consisting of
three olive trees. We studied the effect of different olive varieties
in analytical parameters, and also the evolution of these parame-
ters along the stage of maturation. A total of 63 samples of olives
were handpicked, in perfect sanitary conditions. This samples were
collected from November to January in three stages of maturation,
ripeness index of (0–2), (2–3) and (3–7), (green, spotted and ripe,
respectively), using the subjective evaluation of colour of the skin
and flesh, as proposed by Uceda and Frías (1975). The olive sam-
pling was carried out in the morning, taking samples randomly
in different parts of the central area of the olive tree, assuming a
total of 6 kg per variety and maturity. After harvesting, the olive
fruit samples were immediately transported to the laboratory in
ventilated storage trays to avoid compositional changes; the oil
was extracted within 24 h.
Oil extraction was carried out within 24 h from harvest in a
similar way as industrial extraction conditions using an Abencor
analyser (MC2 Ingeniery systems, Seville, Spain) according to the
procedure reported by Martínez, Muñoz, Alba, and Lanzón
(1975). Olives were crushed with a hammer mill and were
slowly mixed for 30 min at 25 °C. Then, the obtained paste
was centrifuged at 1438 g over 3 min. All the oil samples were
separated by decantation and were stored away from the light
in amber-coloured glass bottles at 4 °C until analysis (within
1 month).
2.2. Analytical quantifications of the phenolic content
Determination of Phenols by Solid Phase Extraction Reversed
Phase High-Performance Liquid Chromatography (SPE RP-HPLC).
A sample of filtered virgin olive oil was weighed (2.5 g), and
250
l
l of a methanolic solution of the internal standards (50 ppm
of p-hydroxyphenylacetic acid (PHFA) and 10 ppm of o-cumaric
acid were added. The solvent was evaporated under reduced pres-
sure at 35 °C, and then the remaining oil was dissolved in hexane
(6 ml).
A diol-bonded phase cartridge (Isolute, Biotage, PA, USA)
(Gómez-Caravaca, Carrasco-Pancorbo, Cañabate-Díaz, Segura-
Carretero, & y Fernández-Gutiérrez, 2005) was used to extract
the phenolic fraction. The cartridge was conditioned with metha-
nol (6 ml) and hexane (6 ml), and the oil solution was then applied
to the SPE column. The column was washed through the cartridge
with hexane (2  3 ml) and with an 85:15 (v/v) hexane-EtOAc mix-
ture (4 ml). Finally, the phenols were eluted with methanol
(15 ml), and the solvent was removed under vacuum at 35 °C.
The phenolic residue was dissolved in 1:1 (v/v) methanol-water
mixture (250
l
l) for HPLC analysis.
HPLC analysis was performed using an Agilent Technologies
series 1200 (Agilent Technologies, S.L., Madrid, Spain) system
equipped with an automatic injector, a column oven, and a diode
array UV detector. A Zorbax column SB-C18 (250  4.6 mm ID,
5
l
m particle size, (Agilent Technologies, Waldbronn, Germany)
was used, maintained at 30 °C with an injection volume of 20
l
l
and with a flow rate of 1.0 ml/min. The mobile phase was a 19:1
(v/v) mixture of water-acetic acid (solvent A), methanol (solvent
B), and acetonitrile (solvent C). The elution gradient was from
95% (A)-2.5% (B)-2.5% (C) to 34% (A)-33% (B)-33% (C) in 50 min, fol-
lowing by 100% (B) for 15 min to clean up the column. Chromato-
grams were taken at 240, 280 and 335 nm.
The polyphenolic content of the olive oils was analysed using
standards and their UV characteristic spectra. They were quantified
by a seven-point regression curve on the basis of the standards
obtained from commercial suppliers. To build the calibration
curves, standards methanolic solutions containing all phenolic
compounds in variable concentrations (Table 1) were prepared in
triplicate, 20
l
l of the solution were injected in the chromato-
graphic system and separation and detection were carried out under
the optimised conditions. Quantification of flavones, ferulic and caf-
feic acids were carried at 335 nm using o-cumaric acid as internal
standard; quantification of secoiridoid derivates and hydroxytyro-
sol acetate was accomplished at 280 nm using o-cumaric acid as
internal standard; the quantification of the rest of the phenolic
compounds was carried out at 280 nm using PHFA as internal stan-
dard. Once the chromatograms were obtained, the retention time
and the peak area were measured using the ChemStation software
package.
Fig. 2. Antioxidant activity comparison between the seven olive varieties.
M
a
Nieves Franco et al. / Food Chemistry 163 (2014) 289–298
291

Author's personal copy
2.3. Antioxidant assays
2.3.1. DPPH method
The antiradical activity of the oils of the seven olive varieties
(IM < 2, 2012/2013 crop season) were measured using 1,1-diphe-
nyl-2-picrylhydrazyl radical (DPPH) following the procedure
reported by Prior, Wu, and Schaich (2005). The assays were
performed in a Hitachi U-2900 spectrophotometer (Tokio, Japan),
using PS cuvettes.
To a 60
l
M methanolic solution of DPPH (HPLC-grade, 1.17 ml)
was added the methanolic solution of the extract (30
l
l, 7 different
concentrations), or pure methanol as the control. The correspond-
ing mixtures were kept in the darkness at rt for 30 min and then,
the absorbance was measured at 515 nm against a blank (MeOH).
Plotting the values of DPPH remaining vs. extract concentration
allows a straight line from which the EC
50
(the concentration of
the antioxidant required to reduce the concentration of the DPPH
to 50% of its initial value) is calculated. All measurements were
carried out in triplicate. The remaining DPPH concentration was
calculated using the expression:
%DPPH remaining ¼
A
sample
A
control
 100
A
sample
and A
control
refer to the absorbances at 515 nm of DPPH
in the sample and control solutions, respectively.
2.3.2. H
2
O
2
scavenging
H
2
O
2
-scavenging activity of the olive oils (IM < 2, 2012/2013
crop season) was measured using the procedure reported by
Bahorun et al. (1996). To a solution of 3  10
3
%H
2
O
2
(100
l
l),
0.1 M NaCl (100
l
l) and 0.1 M phosphate buffer (pH 7.4, 700
l
l)
was added a methanolic solution of the olive oil (4.0 mg/ml,
100
l
l), or pure MeOH (100
l
l) instead of the sample, for the con-
trol. The corresponding mixture was incubated in the darkness at
37 °C for 20 min. Then, a solution containing phenol red (0.2 mg/
ml) and horseradish peroxidase (0.1 mg/ml) in 0.1 M phosphate
buffer (pH 7.4, 1.0 ml) was added. The corresponding solution
was kept at 37 °C during 15 min. Then, 1 M NaOH (100
l
l) was
added, and the mixture was kept at room temperature for
10 min; after that, the absorbance was read at 610 nm against a
blank. All the measurements were carried out in triplicate. The
results are expressed as percentages of reduction of H
2
O
2
:
%reduction H
2
O
2
¼
A
H
2
O
2
 A
sample
A
H
2
O
2
 100
A
H2O2
refers to the control solution with MeOH instead of the
sample, and A
sample
refers to the absorbance of the oil-containing
solutions.
2.3.3. Lipid peroxidation assay (Ferric thiocyanate method, FTC)
Inhibition of the peroxidation-mediated degradation of a lipid
matrix (linoleic acid as the model compound and 2,2’-azobis
(2-methylpropionamidine) dihydrochloride (AAPH), as the free
radical initiator) was measured using the ferric thiocyanate
method (FTC), following the procedure reported by Olszewska,
Presler, and Michel (2012), olive oils with IM < 2, for the 2012/
2013 crop season were used.
To a methanolic solution of the olive oil (5.6 mg/ml, 37
l
l) and
1.3% (w/v) methanolic linoleic acid (175
l
l) in H
2
O (88
l
l) and
0.2 M phosphate buffer (pH 7.0, 175
l
l) in a screw-cap vial was
added AAPH in buffer (55.3 mM , 25
l
l). The control solution was
prepared by adding pure MeOH (37
l
l), instead of the sample.
The vial was incubated at 50 ± 0.1 °C for 24 h in the darkness. After
that, an aliquot (30
l
l) of the reaction mixture was dissolved in a
3:1 (v/v) H
2
O–MeOH solution (2.91 ml), and a 10% aqueous
solution of NH
4
SCN (30
l
l) and 20 mM FeCl
2
in 3.5% HCl (30
l
l)
were added; after 3 min of incubation at rt, the absorbance was
measured at 546 nm against the corresponding blank.
The results are expressed as the percentage of lipid peroxida-
tion inhibition:
%Inhibition ¼
A
control
 A
sample
A
control
 100
A
control
refers to the solution containing pure MeOH instead of
the sample, and A
sample
refers to the absorbance of oil-containing
solutions.
2.3.4. Nitric oxide scavenging activity
Nitric oxide scavenging activity was measured using the Griess
Illosvoy reaction (Garratt, 1964); olive oils with IM < 2, for the
2012/2013 crop season were used. To a solution of 40 mM phos-
phate buffer saline (pH 7.0, 0.15 ml) and MeOH (0.10 ml) were
added a methanolic solution of the olive oil (5.6 mg/ml, 50
l
l)
and a 10 mM solution of sodium nitroprusside in buffer (0.4 ml).
The control solution was prepared using pure MeOH (50
l
l)
instead of the sample. After incubation at rt for 125 min, a 2%
(w/v) solution of sulfanilamide in 4% (v/v) H
3
PO
4
(0.3 ml) was
added, and the mixture was incubated at rt for further 5 min. After
that, a 0.2% (w/v) solution of N-(1-naphthy)lethylenediamine dihy-
drochloride in water (0.3 ml) was added, and the mixture was
incubated at rt for additional 30 min. Finally, the absorbance was
measured at 540 nm against the corresponding blank.
The results are expressed as the percentage of lipid peroxida-
tion inhibition:
%Inhibition ¼
A
control
 A
sample
A
control
 100
Table 1
Analytical figures of merit of HPLC method for phenolic compounds.
Phenol Concentration range
(
l
g/ml)
Linear regression Determination
coefficient (r
2
)
LOD
a
(nig/
kg)
LOD
b
(nig/
kg)
LOQ
c
(nig/
kg)
Linearity
(%)
Analytical
sensitivity (
c
1
)
Hydroxytyrosol 0.1–40 y = 0.0441x  0.0219 0.999 0.04917 0.12665 0.1637361 99.0352 0.5710
Tyrosol 0.5–40 y = 0.0268x  0.0157 0.999 0.04774 0.12262 0.1589742 99.0635 0.5525
Caffeic acid 0.05–30 y = 0.2107x  0.0150 0.999 0.0603 0.1502 0.200799 98.5488 0.6713
Vanillic acid 0.5–40 y = 0.0719x  0.0210 0.997 0.06793 0.17455 0.2262069 98.6669 0.7864
Vanillin 0.5–40 y = 0.1808x  0.0873 0.999 0.04928 0.1266 0.1641024 99.0330 0.5704
p-cumaric acid 0.5–40 y = 0.2174x  0.0744 0.999 0.03517 0.09036 0.1171161 99.3098 0.4071
Ferulic acid 0.05–30 y = 0.1839x  0.0533 0.998 0.04402 0.10962 0.1465866 98.9409 0.4899
Oleuropein 2–1500 y = 0.0040x + 0.0458 0.999 2.03804 4.26565 6.7866732 99.1893 1.76819
Luteolin 0.05–30 y = 0.1495x  0.0044 0.998 0.05089 0.12677 0.1694637 98.7753 0.5665
Apigenin 0.05–30 y = 0.1849x  0.0278 0.998 0.0482 0.12005 0.160506 98.8402 0.5365
a
Limit of detection, Long and Winefordner method (Long & Winefordner, 1983).
b
Limit of detection, Clayton et al. method (
a
= b = 0.05) (Clayton et al., 1987).
c
Limit of quantification, from LOD (Long and Winefordner method)  3.33.
292 M
a
Nieves Franco et al. / Food Chemistry 163 (2014) 289–298

Author's personal copy
A
control
refers to the solution containing pure MeOH instead of
sample, and A
sample
refers to the absorbance of oil-containing
solutions.
2.4. Statistical analysis
For the statistical analysis of the phenolic content, a factorial
design was accomplished in order to considerer the different factor
effects; such factors involved in the experimentation were variety
and years in each stage of maturation separately (green, spotted
and ripe). The model included interactions between factors for each
of the dependent variables. The data were statistically analysed by
ANOVA and Duncan’s multiple range tests to determine at which
level the factors influence on the dependent variables considered.
The adequacy of the model was assessed through standardised
remainder study, to check the normality of the data and homogene-
ity of the variances. Statistical significance was accepted at a level of
p < 0.05. For variance analysis (ANOVA) the SPSS 18.0 software
(SPSS Inc., Chicago, IL, USA) was used. If the interaction was not sig-
nificant, the results were expressed as the mean values between
two crop years. The standard deviation (SD) was calculated.
3. Results
The individual phenolic compounds were identified and quanti-
fied by HPLC with using their corresponding external standards
calibration curves, except in the case of the secoiridoid derivatives,
which were quantified using standard calibration curves of oleu-
ropein due to the absence of pure standards. Table 1 shows the
analytical parameters of calibration used in the HPLC characterisa-
tion. Limits of detection (LODs) and Limits of quantification (LOQs)
ranged from 0.04 to 2.04 and 0.12 to 6.79 mg/kg in oil, respectively.
The linearity range of the analytical procedure was from 98.5 to
99.3% for the studied compounds. We used seven different concen-
trations for the phenolic standards and such standard solutions
were injected three times. All the calibration curves, which were
obtained as a function of the integrated peak area, were linear over
the studied range, with determination coefficients (r
2
) > 0.99 for all
the components.
The analysis of phenolic profiles allowed the separation and
identification of 24 phenolic compounds, all of them were identi-
fied by HPLC-ESI-MS, no qualitative differences were found in the
phenolic fraction profile between virgin olive oils from different
varieties.
However, significant quantitative differences were observed in
a wide number of phenolic compounds. In all olive oils studied,
secoiridoids derivatives were the most abundant, followed by phe-
nolic alcohols, flavonoids and phenolic acids. Oleuropein deriva-
tives: 3,4-DHPEA-EDA (an ester between hydroxytyrosol and to
the dialdehydic form of elenolic acid) and 3,4-DHPEA-EA (hydroxy-
tyrosol linked to the aldehydic form of elenolic acid), ligstroside
derivatives: p-HPEA-EDA (tyrosol linked to the dialdehydic form
of elenolic acid), p-HPEA-EA (tyrosol linked to the aldehydic form
of elenolic acid) and 3,4-DHPEA-AC (hydroxytyrosyl acetate) were
the major compounds in the seven varieties. Their concentration
were at least 28 times that of the minor compounds (3,4-DHPEA
(hydroxytyrosol), p-HPEA (tyrosol), vanillic acid, vanillin, p-cuma-
ric acid, caffeic acid, ferulic acid, luteolin, and apigenin).
On the other hand, gallic, 3,4-dihydroxyphenylacetic, gentisic,
4-hydroxybenzoic, syringic, m-coumaric, o-coumaric, and cin-
namic acids were not detected in oils from Extremadura varieties
by HPLC-ESI-MS.
A multivariate analysis was computed to assess the effect of
variety and crop season variables on each stage of maturation.
On the other hand, the major compounds showed a significant
interaction between crop seasons and varieties, so data from two
crop seasons are shown independently (Table 2).
On the other hand, there was not a clear tendency in 3,
4-DHPEA-EDA concentration throughout the ripening process in
both crop seasons. Carrasqueña variety presented the highest
values of 3,4-DHPEA-EDA concentration in the green stage of mat-
uration, while Corniche variety presented the highest values in the
spotted and ripe stages in both crop years.
The highest value of the p-HPEA-EDA was observed in the
second stage of maturation in Verdial de Badajoz variety and, the
lowest value was observed in Carrasqueña variety in ripe stage of
maturation in both crop seasons (36.11 and 14.52 mg.kg
-1
respec-
tively). In general, values found in spotted stage of maturation
were the highest ones.
Remarkably, 3,4-DHPEA-AC could not be detected in some of
the analysed varieties. Moreover, the highest values of the 3,4-
DHPEA-EA were observed in the spotted stage of maturation for
Arbequina, Corniche, and Morisca (Table 2). The general tendency
was to increase from green to spotted stage of maturation and to
decrease from spotted to ripe in both crop seasons. The same ten-
dency was observed for p-HPEA-EA. The highest values for this
compound were in Morisca, Corniche and Verdial de Badajoz
(151.16, 133.31 and 127.13 mg.kg
1
respectively) in 2011/2012
crop season and for Verdial de Badajoz and Corniche (139.56 and
117.92 mg.kg
1
respectively) in 2012/2013 crop season (Table 2).
Among the phenolic compounds present at lower concentra-
tions, phenolic acids were the first family to be identified in virgin
olive oil. Vanillic, p-cumaric and ferulic acids, and vanillin were
identified in different concentrations considering varieties and
maturation stages. None of the phenolic acids analysed, except
for ferulic acid, showed a correlation between crop seasons and
varieties in each stage of maturation. Therefore, in these cases,
we depict the mean values of the two crop seasons for these
variables (Table 3).
Vanillic acid was present at 0.34–2.00, 0.22–0.60 and
0.09–0.44 mg.kg
1
in green, spotted and ripe maturation stages,
respectively. Arbequina and Picual varieties showed the highest
values of vanillic acid in green and spotted stages of maturation.
Besides, this phenolic compound was not quantified in Corniche
and Arbequina varieties in ripe stage of maturation under limits
of quantification shown in Table 1. There was a decreasing
tendency in the concentration of these compounds upon ripening,
being statistically significant from green to spotted stage of matu-
ration. From spotted to ripe the concentration of vanillic acid
remained constant.
Vanillin concentration ranged from 0.17 to 0.47, 0.18 to 0.41
and 0.17 to 0.19 mg.kg
1
in green, spotted and ripe stages, respec-
tively, with no significant differences between varieties in ripe
stage of maturation. Arbequina and Picual varieties showed the
highest concentration in green stage of maturation. Thus, there
was a decrease of the concentration of these compounds with
the ripening of the fruit, being statistically insignificant in some
cases. Verdial de Badajoz variety decreased by 50% from green to
ripe stage of maturation while Corniche and Manzanilla Cacereña
did not show a significant variation throughout the ripening pro-
cess.p-Cumaric acid ranged from 0.29 to 1.37, 0.18 to 0.41 and
0.19 to 0.69 mg.kg
1
in green, spotted and ripe, respectively, with
significant differences between varieties. It is noticeable that
Morisca and Carrasqueña had the highest values significantly and
Corniche and Manzanilla Cacereña had the lowest values that the
rest of studied varieties in the three stage of maturation. Besides,
there was a decrease of the concentration of this compound with
the ripening of the fruit except for Carrasqueña and Corniche.
Ferulic acid could not be quantified in the most of the analysed
oils, being the values obtained in the first crop season lower than in
the second one.
M
a
Nieves Franco et al. / Food Chemistry 163 (2014) 289–298
293

Author's personal copy
Table 2
Composition of the major phenolic compounds obtained from seven varieties of VOO. Results are expressed as mean ± SD of three sample replicates. If the interaction was not significant, the results were expressed as two crop seasons
mean values ± SD. Different small letters in the same row indicate significant statistical differences (Duncan’s Test, p < 0.05) among varieties. Different capital letters in the same column indicate significant statistical differences
(Duncan’s Test, p < 0.05) during the stage of maturation.
Ripening index Variety Phenolic compounds (mg.Kg
1
)
3,4-DHPEA-AC 3,4-DHPEA-EDA p-HPEA-EDA 3,4-DHPEA-EA p-HPEA-EA
2011–2012 2012–2013 2011–2012 2012–2013 2011–2012 2012–2013 2011–2012 2012–2013 2011–2012 2012–2013
Green IM < 2 Arbequina 42.29 ± 3.55
b
A
nq 75.47 ± 23.91
aB
35.47 ± 7.70
aB
59.91 ± 6.95
aB
18.12 ± 9.05
aA
33.71 ± 1.88
aA
18.30 ± 4.30
aA
16.47 ± 0.98
aB
9.85 ± 0.23
aA
Carrasqueña 7.33 ± 0.28
b
NS
nq 141.11 ± 1.13
bB
59.31 ± 4.10
bB
73.47 ± 18.96
ab B
38.77 ± 5.04
bB
68.36 ± 14.93
cA
48.01 ± 6.64
eB
82.91 ± 6.88
cC
34.28 ± 3.99
bc B
Corniche nd nd 69.04 ± 14.72
aA
49.18 ± 6.13
bB
101.67 ± 21.52
bA
58.89 ± 7.51
cB
61.17 ± 16.54
bB
44.06 ± 8.68
de A
12.44 ± 2.30
aA
43.66 ± 8.66
cA
Manzanilla
Cacereña
nq nq 56.04 ± 9.61
aB
21.77 ± 4.20
aB
58.20 ± 7.29
aA
36.97 ± 7.95
bB
32.75 ± 9.75
aA
19.54 ± 4.29
ab NS
73.38 ± 0.93
cB
26.41 ± 4.98
bB
Morisca 6.20 ± 10.01
c
B
14.06 ± 1.70
A
70.56 ± 7.19
aB
52.14 ± 9.83
bB
41.43 ± 31.58
aA
15.24 ± 4.07
aA
30.39 ± 4.13
aB
28.91 ± 5.70
bc A
51.01 ± 1.18
bc B
27.97 ± 4.02
bA
Picual nq nd 89.25 ± 15.58
aB
23.62 ± 5.88
aA
44.20 ± 3.49
aA
19.10 ± 2.62
aA
73.31 ± 16.94
cB
16.51 ± 5.72
aA
39.80 ± 7.43
bA
12.90 ± 3.43
aA
Verdial de
Badajoz
nd nd 143.75 ± 34.23
bC
21.85 ± 8.83
aA
166.44 ± 15.02
cB
44.42 ± 7.48
bA
33.70 ± 5.74
aB
36.80 ± 1.27
cd AB
41.59 ± 11.75
bB
94.70 ± 0.04
dAB
Interactions crop
season x variety
⁄⁄⁄ ⁄ ⁄ ⁄ ⁄ ⁄ ⁄ ⁄
Spotted IM 2–3 Arbequina 98.17 ± 21.67
cB
151.09 ± 7.47
dB
68.18 ± 17.09
aB
23.50 ± 8.21
aA
86.90 ± 8.06
aC
38.05 ± 0.55
aB
122.24 ± 1.62
cC
125.64 ± 12.86
dC
14.59 ± 3.89
aAB
63.10 ± 6.85
bC
Carrasqueña 6.89 ± 2.65
a
27.84 ± 2.73
b
B
148.36 ± 26.53
cB
75.68 ± 4.62
cC
84.09 ± 8.36
aB
49.03 ± 6.21
aB
75.65 ± 11.12
bAB
85.98 ± 3.93
cC
40.00 ± 12.86
ab B
47.19 ± 4.54
bC
Corniche nd nd 156.08 ± 35.98
cB
95.03 ± 8.84
dC
172.42 ± 19.55
bB
118.38 ± 9.84
cC
80.65 ± 9.20
bC
139.54 ± 12.82
dB
133.31 ± 2.30
cB
117.92 ± 18.91
cd C
Manzanilla
Cacereña
nq 9.39 ± 0.94
a
59.97 ± 21.11
aB
14.74 ± 1.63
aA
68.67 ± 15.35
aB
36.52 ± 0.90
aB
49.28 ± 14.31
aB
14.35 ± 3.88
aNS
16.68 ± 3.69
aA
13.48 ± 4.06
aA
Morisca 33.45 ± 4.23
b
A
63.80 ± 7.45
c
C
171.03 ± 6.71
cC
77.17 ± 10.36
cC
91.50 ± 29.14
aB
40.27 ± 7.37
aB
133.36 ± 3.03
cC
117.06 ± 11.79
dB
151.16 ± 45.08
cC
94.57 ± 11.79
cB
Picual nq 14.59 ± 3.52
a
105.58 ± 13.02
bC
48.54 ± 7.23
bB
60.25 ± 3.96
aB
39.98 ± 6.54
aB
64.14 ± 7.26
bAB
41.96 ± 3.14
bc C
66.88 ± 13.10
bB
18.87 ± 1.07
aB
Verdial de
Badajoz
nq nd 111.71 ± 29.28
bB
44.84 ± 2.09
bB
291.44 ± 63.26
cC
132.70 ± 5.23
dB
71.62 ± 7.60
bC
67.67 ± 26.50
cB
127.13 ± 1.45
cC
139.56 ± 42.70
dB
Interactions crop
season x variety
⁄⁄⁄ ⁄ ⁄ ⁄ ⁄ ⁄ ⁄ ⁄
Ripe IM > 3 Arbequina 36.79 ± 1.47
c
A
70.80 ± 18.47
cA
44.70 ± 1.10
aA
23.45 ± 3.35
aA
36.82 ± 0.24
aA
17.29 ± 0.65
aA
104.35 ± 1.15
cB
97.32 ± 4.80
cB
9.68 ± 1.77
bA
36.89 ± 6.95
cd B
Carrasqueña 6.99 ± 1.37
a
14.80 ± 0.28
a
A
88.67 ± 10.47
bA
22.68 ± 1.59
aA
36.11 ± 3.06
aA
14.52 ± 2.63
aA
95.17 ± 33.36
cB
33.32 ± 6.74
aA
14.30 ± 1.68
bc A
11.99 ± 0.76
aA
Corniche 20.94 ± 2.51
b
B
nd 150.54 ± 12.60
cB
45.78 ± 8.33
bA
116.96 ± 32.40
cA
37.52 ± 5.26
bA
37.01 ± 9.33
ab A
56.45 ± 11.52
bA
10.54 ± 3.37
bA
22.74 ± 4.94
bA
Manzanilla
Cacereña
nd nq 40.56 ± 4.48
aA
nd 59.37 ± 8.13
bA
16.93 ± 0.33
aA
48.31 ± 5.99
bB
nq 10.90 ± 3.07
bA
nq
Morisca 43.26 ± 8.25
c
A
41.21 ± 2.59
b
B
42.52 ± 5.26
aA
22.54 ± 1.13
aA
40.32 ± 3.58
bA
14.75 ± 2.28
aA
18.99 ± 0.83
aA
33.72 ± 10.71
aA
23.89 ± 5.01
cA
27.35 ± 8.49
bA
Picual nq nq 71.73 ± 3.94
bA
27.79 ± 1.22
aA
39.68 ± 7.50
aA
22.75 ± 2.62
ab A
46.94 ± 0.81
bA
26.24 ± 3.10
aB
68.80 ± 9.62
dB
11.75 ± 1.39
aA
Verdial de
Badajoz
nd nd 71.64 ± 18.80
bA
18.00 ± 2.26
aA
123.49 ± 19.14
cA
52.17 ± 7.22
cA
16.47 ± 1.66
aA
25.81 ± 9.43
aA
6.53 ± 0.53
aA
41.72 ± 9.71
dA
Interactions crop
season x variety
⁄⁄⁄ ⁄ ⁄ ⁄ ⁄ ⁄ ⁄ ⁄
ns/NS: no significant and no significant interaction respectively; and ⁄: significant interaction.
294 M
a
Nieves Franco et al. / Food Chemistry 163 (2014) 289–298

Author's personal copy
Table 3
Composition of the minor phenolic compounds obtained from seven varieties of olive oils. Results are expressed as mean ± SD of three sample replicates. If the interaction was not significant, the results were expressed as two crop
seasons mean values ± SD. Different small letters in the same row indicate significant statistical differences (Duncan’s Test, p < 0.05) among varieties. Different capital letters in the same column indicate significant statistical differences
(Duncan’s Test, p < 0.05) during the stage of maturation.
Ripening
index
Variety Phenolic compounds (mg.Kg
1
)
3,4-DHPEA p-HPEA Vanillic acid Vanillin p-cumaric acid Ferulic acid Luteolin Apigenin
2011–2012 2012–2013 2011–2012 2012–2013 2011–2012 2012–2013
Green IM < 2 Arbequina 1.34 ± 0.41
aB
1.95 ± 0.29
aC
2.00 ± 0.33
cB
0.47 ± 0.15
bB
0.81 ± 0.40
bC
nq nq 2.68 ± 0.25
dA
0.68 ± 0.04
bc A
0.91 ± 0.08
eA
0.17 ± 0.00
aA
Carrasqueña 2.84 ± 0.84
bc B
3.49 ± 0.89
aA
1.16 ± 0.03
bB
0.36 ± 0.06
ab B
1.37 ± 0.20
cB
nq nq 0.91 ± 0.26
aA
0.57 ± 0.08
ab A
0.34 ± 0.08
bc NS
0.24 ± 0.04
bc A
Corniche 2.00 ± 0.84
ab B
3.05 ± 1.90
aB
0.34 ± 0.15
aB
0.17 ± 0.08
aNS
0.29 ± 0.14
aB
nd nq 0.74 ± 0.32
aA
0.49 ± 0.04
aA
0.17 ± 0.00
aA
nq
Manzanilla
Cacereña
1.46 ± 0.57
aB
5.92 ± 1.40
bA
0.72 ± 0.31
ab NS
0.25 ± 0.05
ab NS
0.48 ± 0.12
aC
nd nq 1.92 ± 0.10
cA
0.81 ± 0.01
cd A
0.56 ± 0.04
dA
0.34 ± 0.02
dA
Morisca 1.28 ± 0.27
aAB
3.00 ± 0.85
aB
0.49 ± 0.20
aB
0.32 ± 0.10
ab B
0.84 ± 0.22
dC
nd nq 2.96 ± 0.22
dB
1.19 ± 0.13
eA
0.67 ± 0.02
dB
0.27 ± 0.03
cA
Picual 3.31 ± 1.19 c
B
5.16 ± 1.27
bNS
1.91 ± 0.52
cB
0.43 ± 0.20
bB
1.25 ± 0.29
cC
nd nq 2.26 ± 0.18
cA
0.94 ± 0.15
dA
0.39 ± 0.13
cNS
0.21 ± 0.02
bA
Verdial de
Badajoz
1.40 ± 0.22
aNS
6.25 ± 0.58
bB
0.80 ± 0.30
ab NS
0.37 ± 0.22
ab C
1.07 ± 0.33
bc B
0.15 ± 0.02
A
nq 1.36 ± 0.09
bA
0.57 ± 0.05
ab A
0.22 ± 0.03
ab NS
nq
Interactions
crop
season x
variety
NS NS NS NS NS ⁄⁄ ⁄
Spotted IM
2–3
Arbequina 0.75 ± 0.19
aA
1.30 ± 0.33
aB
0.43 ± 0.20
bc A
0.41 ± 0.06
bB
0.22 ± 0.08
ab B
nd nq 6.21 ± 0.31
eB
2.64 ± 0.20
ab B
1.87 ± 0.31
cB
0.45 ± 0.05
aB
Carrasqueña 3.77 ± 0.70
dB
6.43 ± 1.08
de B
0.47 ± 0.22
cd A
0.23 ± 0.07
aA
0.18 ± 0.05
cd A
nq 0.31 ± 0.02
bB
1.55 ± 0.15
ab B
1.98 ± 0.31
aC
0.47 ± 0.04
a
0.92 ± 0.15
bC
Corniche 1.81 ± 0.77
bc B
3.22 ± 1.00
bc B
0.22 ± 0.05
aAB
0.19 ± 0.07
a
0.22 ± 0.00
aB
nd nq 1.26 ± 0.11
aA
2.54 ± 0.59
ab B
0.35 ± 0.03
aAB
0.55 ± 0.19
aB
Manzanilla
Cacereña
0.77 ± 0.16
aA
7.26 ± 1.00
eAB
0.45 ± 0.05
cd
0.24 ± 0.09
a
0.24 ± 0.10
aB
nd nq 3.21 ± 0.46
dB
3.01 ± 0.29
bC
1.01 ± 0.20
bB
1.51 ± 0.16
cC
Morisca 1.04 ± 0.22
ab A
1.97 ± 0.60
ab A
0.32 ± 0.12
abc A
0.19 ± 0.08
aA
0.41 ± 0.06
eB
nq 0.19 ± 0.02
a
1.80 ± 0.14
bc A
4.53 ± 0.19
cC
0.44 ± 0.04
aA
1.02 ± 0.12
bC
Picual 2.55 ± 1.13
cAB
3.79 ± 1.06
c
0.60 ± 0.26
dA
0.18 ± 0.04
aA
0.23 ± 0.08
bc B
nq nq 2.24 ± 0.24
cA
4.34 ± 0.60
cC
0.49 ± 0.08
a
0.92 ± 0.04
bC
Verdial de
Badajoz
1.60 ± 0.93
ab
5.84 ± 1.39
dAB
0.28 ± 0.04
ab
0.22 ± 0.01
aB
0.19 ± 0.07
dAB
0.24 ± 0.03
B
0.46 ± 0.03
cB
1.71 ± 0.24
ab B
2.00 ± 0.37
aB
0.37 ± 0.05
a
0.40 ± 0.06
aNS
Interactions
crop
season x
variety
NS NS NS NS NS ⁄⁄ ⁄
Ripe IM > 3 Arbequina 0.39 ± 0.02
aA
0.64 ± 0.02
aA
0.16 ± 0.01
ab A
0.17 ± 0.00
ns A
0.24 ± 0.01
aA
nq nq 2.76 ± 0.11
c
0.98 ± 0.03
e
Carrasqueña 0.65 ± 0.03
bA
3.44 ± 0.02
bc A
0.44 ± 0.03
bA
0.17 ± 0.02
A
0.53 ± 0.04
A
0.16 ± 0.03
ns
0.25 ± 0.02
aA
1.60 ± 0.02
a
0.65 ± 0.04
cd
Corniche 0.93 ± 0.28
cA
2.05 ± 1.02
ab A
0.09 ± 0.04
aA
0.17 ± 0.06 0.17 ± 0.02
A
nq nq 1.60 ± 0.83
a
0.31 ± 0.20
a
Manzanilla
Cacereña
0.74 ± 0.34
bc A
7.82 ± 1.52
dB
0.44 ± 0.07
b
0.18 ± 0.04 0.17 ± 0.02
A
nd nq 3.38 ± 0.93
cd
1.13 ± 0.15
e
Morisca 1.38 ± 0.70
dB
3.17 ± 1.42
bc B
0.26 ± 0.07
ab A
0.19 ± 0.04
A
0.17 ± 0.02
A
0.15 ± 0.01 nq 3.78 ± 0.57
d
0.75 ± 0.09
d
Picual 1.60 ± 0.82
dA
3.54 ± 2.66
bc
0.31 ± 0.08
ab A
0.17 ± 0.02
A
0.17 ± 0.02
A
nd nq 2.64 ± 0.67
bc
0.52 ± 0.13
bc
Verdial de
Badajoz
0.83 ± 0.41
bc
4.21 ± 1.07
cA
0.35 ± 0.34
ab
0.17 ± 0.00
A
0.17 ± 0.02
A
0.15 ± 0.02 0.31 ± 0.05
bA
1.87 ± 0.62
ab
0.39 ± 0.10
ab
Interactions
crop
season x
variety
NS NS NS NS NS ⁄ NS NS
ns/NS: no significant and no significant interaction respectively; and ⁄: significant interaction.
M
a
Nieves Franco et al. / Food Chemistry 163 (2014) 289–298
295

Author's personal copy
The main phenolic alcohols found in the examined varieties of
virgin olive oils were 3,4-DHPEA (hydroxytyrosol) and p-HPEA
(tyrosol). In this way, independently of the crop season, we could
find significant differences among varieties with regard to phenolic
alcohols. In general, the concentration of tyrosol was higher than
that of hydroxytyrosol. The highest concentration of hydroxytyro-
sol was observed in virgin olive oil from Picual variety in all stages
of maturation, followed by Carrasqueña in green and ripe stage of
maturation and by Morisca in ripe stage. On the contrary, Arbequ-
ina showed the lowest values of 3,4-DHPEA in the three stages of
maturation. The general trend of this compound is to decrease its
concentration along the ripening process, with the exception of
the Morisca variety that showed an increase from spotted to ripe
and Verdial de Badajoz, which does not suffer significant changes
in any stage of maturation. p-HPEA did not show a clear tendency
for any of the seven studied varieties along the ripening process. In
the case of p-HPEA, Manzanilla Cacereña variety presented highest
concentrations in the three stages of maturation with values higher
to 5 mg.kg
1
, followed by Verdial de Badajoz. On the other hand,
Arbequina showed values lower than 2 mg.kg
1
.
Luteolin and apigenin showed a significant interaction with
crop season and varieties factors in green and spotted stages of
maturation (Table 3). Furthermore, significant differences between
varieties can be observed. Morisca variety presented the highest
values of luteolin, while Manzanilla Cacereña presented the high-
est values of apigenin for stage of maturation at ripe.
Concerning the antioxidant profile of the seven oil varieties
considered herein, the results are depicted in Table 4. Four differ-
ent antioxidant tests have been conducted: DPPH and FTC method
(for quantifying the antiradical activity), H
2
O
2
and NO scavenging,
the latter being an example of a reactive nitrogen specie (RNS).
Butylated hydroxytoluene (BHT), a widely-used synthetic antioxi-
dant, is included as a reference compound. It is remarkable men-
tioning that, whereas DPPH considers antiradical activity against
a commercially-available free radical, FTC measures such activity
by thermal-inducement of alkyl peroxide-type free radicals using
a lipid matrix as model (linoleic acid) and AAPH as the free radical
initiator. Moreover, NO is also generated in situ by spontaneous
decomposition of sodium nitroprusside in a buffered aqueous
medium.
Preliminary results suggested that the strongest antioxidant
activity was found in the first maturity stage (IM < 2), so this is
only the stage considered for all the cases. Data show almost no
variation in the capacity of scavenging H
2
O
2
(69.9–76.8%), at
0.4 mg/ml final concentration.
Regarding DPPH, all the analysed fractions exhibited strong
anti-radical inhibition, but some differences existed; varieties
showing more potent anti-radical activity (the lowest EC
50
values)
were Carrasqueña, Arbequina and Corniche (EC
50
= 14.8, 16.9, and
17.0
l
g/ml, respectively), whereas those with a lower activity
compared with the former varieties were Verdial de Badajoz, Mor-
isca and Manzanilla Cacereña (26.6, 25.7, and 25.2
l
g/ml).
Although free radicals are also involved in the lipid peroxidation
assay, results are not the same as those found for DPPH; thus, for
example, the strongest inhibition of linoleic acid degradation was
found for Picual, Manzanilla Cacereña and Carrasqueña (45.1,
40.1, and 39.4%, respectively). Such results suggest the importance
of the medium and nature of the free radical employed for consid-
ering the antioxidant test (Table 4).
NO scavenging activity was the method which showed the most
pronounced differences in activity, ranging from 29.8% (Morisca) to
40.7% inhibition (Arbequina, Carrasqueña) at 0.4 mg/ml final con-
centration of the extract.
Therefore, in general, Arbequina and Carrasqueña are the varie-
ties with a better antioxidant profile, Morisca and verdial de Bad-
ajoz being the ones with the poorest antioxidant activity.
4. Discussion
The most important phenolic compounds that have been iden-
tified in VOO may be divided into different groups: phenolic acids,
phenolic alcohols, secoiridoids, lignans, and flavones (Bendini et al.,
2007; Brenes et al., 2000). HPLC profiles for the minor polar pheno-
lic compounds are similar, but their amounts present notable vari-
ations between varieties. In addition, qualitative and quantitative
profile of these compounds has been used, in recent years, to clas-
sify olive oils in terms of their stage of maturation, varietal and
geographical origin (Manai-Djabali et al., 2012).
A series of metabolic processes – chemical and enzymatic reac-
tions – take place during fruit ripening and processing, resulting in
the production of free phenols and inducing variations in the phe-
nolic profile of several compounds.
However, the secoiridoids derivatives of tyrosol and hydroxyty-
rosol are the main phenolic compounds in fresh olive oils
(Oliveras-López et al., 2007). 3,4-DHPEA-EDA was found to be the
most significant example of complex phenols for Picual for the
green maturation stage, whereas p-HPEA-EDA were the major ones
found in Manzanilla Cacereña (spotted and ripe maturation stages)
and Corniche (green maturation stage), for both crop seasons; for
both secoiridoid derivatives, the lowest concentration was found
in Arbequina. In this context, Manai-Djabali et al. (2012) showed
that the Betsijina fresh oil presented the highest values of secoirid-
oids (up to 32.3 and 213.3 mg/kg, respectively), values within
those found in our study.
Furthermore, some researchers have previously found that 3,
4-DHPEA-EA and p-HPEA-EA content was low in Arbequina olive
oil (Bakhouche et al., 2013; Lozano-Sánchez et al., 2010). On the
other hand, a decrease of the total phenolic content during ripen-
ing was generally observed, in all the varieties analysed. Tyrosol,
Table 4
Antioxidant activity of the oils from the seven olive varieties (DPPH, lipid peroxidation inhibition, H
2
O
2
and NO scavenging). Results are expressed as mean ± SD of three sample
replicates. If the interaction was not significant, the results were expressed as two crop seasons mean values ± SD. Different small letters in the same row indicate significant
statistical differences (Duncan’s Test, p < 0.05) among varieties.
Variety DPPH method (EC
50
,
l
g/ml) % Inhibition of lipid peroxidation % H
2
O
2
scavenging % NO scavenging
Arbequina 16.9 ± 1.8
ab
33.0 ± 2.0
ab
75.6 ± 1.3
cd
40.7 ± 7.7
ns
Carrasqueña 14.8 ± 2.1
a
39.4 ± 5.4
b
76.8 ± 1.8
d
40.7 ± 9.9
Corniche 17,0 ± 0.5
ab
37.1 ± 2.4
b
74.3 ± 1.7
bcd
24.1 ± 4.2
Manzanilla Cacereña 25.2 ± 4.8
bc
40.1 ± 8.1
b
72.4 ± 2.9
bcd
38.5 ± 15.8
Morisca 25.7 ± 6.0
bc
32.9 ± 11.4
ab
71.1 ± 3.8
bc
29.8 ± 10.1
Picual 18.4 ± 4.4
abc
45.1 ± 1.3
bc
75.3 ± 1.9
cd
42.2 ± 7.3
Verdial de Badajoz 26.6 ± 9.2
c
23.1 ± 10.8
a
69.9 ± 4.3
b
35.4 ± 6.1
BHT 15.4 ± 2.0
a
56.8 ± 2.5
c
10.0 ± 0.8
a
ns: no significant.
296 M
a
Nieves Franco et al. / Food Chemistry 163 (2014) 289–298

Author's personal copy
hydroxytyrosol, vanillic acid, p-cumaric acid, luteolin, and apigenin
levels were higher in early harvest samples in almost all the oils
analysed, these results being in agreement with Jiménez,
Sánchez-Ortiz, Lorenzo, and Rivas (2013) for the Picudo variety.
In this sense, the concentration of the simple phenol vanillin pro-
gressively decreased in a 54.5% as ripening progressed (Jiménez
et al., 2013). Besides, they reported that hydroxytyrosol levels
decreased by 50.40% in late harvest samples. These results are in
agreement with those obtained by Martinez, Hodaifa, and Lozano
Peña (2010), who reported that tyrosol and hydroxytyrosol con-
centrations decreased with increasing olive ripeness in Picual and
Arbequina varieties. The reason for the wide content distribution
observed for these two phenols may be that hydroxytyrosol and
tyrosol can be produced by the partial hydrolysis of their deriva-
tives (Montedoro, Servili, Baldioli, & Miniati, 1992). In fact,
Godoy-Caballero, Acedo-Valenzuela, and Galeano-Díaz (2012) ana-
lysed commercial VOO of several Spanish varieties and they found
higher concentration of these phenolic alcohols than those found
in our study.
With regard to the crop seasons studied, the major compounds
showed a significant interaction between crop seasons and varie-
ties. In this sense, we observed lower values in the first crop season
that could be mainly attributed to the higher rainfall occurred dur-
ing this crop year (97 mm more). It is well documented that the
total phenol content depends on the zones of cultivation, growing
conditions and, mainly, of the climatic characteristics (Aguilera
et al., 2005). Fuentes et al. (Unpublished results) found that the
lowest values of total phenolic compounds and oxidative stability
corresponded to crop year 2007/2008, featured with adverse
weather conditions. Moreover, the higher temperatures experi-
enced in southern Lebanon in 2010 compared to 2011 might be
responsible for the lower phenolic content of Baladi olive oil for
the former year (Inglese et al., 2011).
On the other hand, we have found no significant interaction for
the minor phenolic compounds, which indicate that these com-
pounds were not hypothetically affected by ambient conditions
in each year of study.
Low flavonoid levels represented by luteolin and apigenin were
observed in all olive oils analyzed. These results are in agreement
with several authors (Carrasco-Pancorbo et al., 2005). In the same
way, Ouni et al. (2011) observed concentrations of these com-
pounds that varied from 0.74 to 6.54 and 0.086 to 1.5 mg.kg
1
,
respectively. Besides, these researchers reported that, in spite of
their low concentrations, luteolin and apigenin showed significant
differences between oils from different varieties. Low flavonoid
levels represented by luteolin and apigenin were observed in all
the olive oils analyzed, with concentrations that varied from 0.49
to 6.21 mg.kg
1
and 0.17 to 1.87 mg.kg
1
, respectively.
Taking into consideration the complexity of oxidation processes
in living systems, where there is not only one single reactive inter-
mediate involved, different methods were chosen to quantity the
overall antioxidant profile of the samples. Each assay measures
the activity against a certain reactive oxygen/nitrogen specie, so
the combination of all these techniques reflects better the antiox-
idant capacity of the sample (Pérez, Vargas, Martínez, García, &
Hernández, 2003). Carrasqueña was found to be the variety with
the lowest EC
50
for the DPPH antiradical assay, and to have the
second best activity against NO and lipid peroxidation inhibition;
furthermore, it exhibited the best profile in H
2
O
2
scavenging. All
these results could be explained considering that Carrasqueña is
a variety featured, not only with a higher hydroxytyrosol content
than the rest of varieties, but also with a high concentration of
hydroxytyrosol derivatives, like DHPEA-AC, 3,4-DHPEA-EDA, and
3,4-DHPEA-EA, the two latter, being significant examples of the
major family of phenolic compounds contained in VOO. These
results could also explain the good oxidative stability exhibited
by Carrasqueña (Franco, Galeano-Díaz, Sánchez, De Miguel, &
Martín-Vertedor, in press). It is also the variety with the highest
content of p-HPEA-EA in 2011–2012 crop season and also exhibits
high concentration of p-HPEA-EDA; both compounds derive from
Ty, another important phenolic compound with relevant antioxi-
dant activity, only surpassed by HTy. As already indicated within
the manuscript, phenolic compounds contribute prominently to
the nutritional properties of oil which lead to health benefits.
Moreover, Arbequina variety, although exhibiting a poor
oxidative stability (Franco et al., in press), showed an excellent
antiradical activity in the DPPH assay, only surpassed by
Carrasqueña; it also presents the second best scavenging activity
against H
2
O
2
and NO. Maybe the fact that it is one of the few
varieties where 3,4-DHPEA-AC was detected (only quantified in
2011–2012 crop season) can explain such good antioxidant profile;
furthermore, 3,4-DHPEA-EDA is also present in high concentra-
tions, only surpassed by Carrasqueña, and statistically equal to
Verdial de Badajoz. Although this variety presents the overall
lowest polyphenolic content at the same maturation stage as com-
pared with the others, this observation is not contradictory with
the above results, as Arbequina, exhibits a high concentration of
hydroxytyrosol derivatives, which are known to exhibit strong
antioxidant properties.
As an example of variety with a high content of Ty and its deriv-
atives, Manzanilla Cacereña is remarkable. This variety is found in
the group of varieties with higher content in Ty, together with Ver-
dial of Badajoz and Picual. Furthermore, it is in the second group of
varieties with higher content in p-HPEA-EDA in 2012–2013, and it
is the second variety with the highest concentration in p-HPEA-EA
in 2011–2012 season. All this could justify their high antioxidant
power against lipid peroxidation, only surpassed by Picual; never-
theless, it exhibited only moderate oxidative stability (Franco et al.,
in press). Therefore, those oil varieties featured with strong lipid
peroxidation inhibition, together with strong free radical, NO and
H
2
O
2
scavenging properties are better preserved in terms of oxida-
tive degradation, and at the same time lead to relevant beneficial
health effects.
5. Conclusions
Phenolic compounds quantified by HPLC showed no qualitative
differences between the seven varieties studied, however they did
present quantitative differences. The phenolic compounds,
detected and quantified, mainly belong to the family of secoiridoid
derivatives of HTy (3,4-DHPEA-EA and 3,4-DHPEA-EDA) and of
Ty (p-HPEA-EA and p-HPEA-EDA); hydroxytyrosol acetate (3,4-
DHPEA-AC), phenolic alcohols (HTy and Ty), flavonoids (Luteolin
and Apigenin) and phenolic acids (vanillic acid, vanillin, p-cumaric
acid, cafeic acid, and ferulic acid) were also detected, the latter
ones as minor compounds. Most of the compounds identified
in the seven varieties studied were derivatives of oleuropein:
3,4-DHPEA-EDA and 3,4-DHPEA-EA; the ligustrosid derivatives:
p-HPEA-EDA, p-HPEA-EA and 3,4-DHPEA-AC, presenting a concen-
tration at least 28 times higher than that of the minor compounds.
Generally, an increase of these compounds from green to spotted
stage of maturation, and a decrease up to mature stage. Therefore,
the phenolic compound profile would allow their use as varietal
markers. This fact is a key point for mills due to the increase of
the competitiveness and quality protection of the companies in
the oleic sector.
The antioxidant capacity was measured for the first mature
stage in the 2012/2013 season, in order to establish differences
between the seven varieties; anti-radical capacity (against a model
free radical (DPPH method) and in the lipid peroxidation), and
H
2
O
2
and NO scavenging were tested. Considering the DPPH
M
a
Nieves Franco et al. / Food Chemistry 163 (2014) 289–298
297

Author's personal copy
method, the lowest EC
50
values (best antiradical activity) was
found for Carrasqueña variety (EC
50
<15
l
g/ml), in the same order
of magnitude as the widely-used synthetic antioxidant BHT. On the
other hand, for the lipid peroxidation technique, Picual variety was
the one exhibiting the best inhibition (45.1%), although slightly
below to that showed by BHT (56.8%) at the same concentration.
Carrasqueña variety also presented the best scavenging activity
of H
2
O
2
(76.8%), remarkably better than BHT (10%). Finally, regard-
ing NO scavenging assay, Morisca and Corniche variety presented
the lowest activity (29.8 and 24.1% inhibition, respectively); for
the rest of the cases, the scavenging properties were considerably
better (35.4–40.7%).
Acknowledgements
This study has been carried out with financial support from the
Government of Extremadura (Project 6/12). Mª N. Franco thanks to
the FSE and the Government of Extremadura for the Grant
(TEC09072). The authors wish to thank to the Research Centre
‘‘finca la Orden’’ by the field experiment varieties and to M. García
Serrano, P, Godoy-Caballero, J. Hernández Carretero, J.M. García
Ballesteros, E, Torrescusa, A, Fernández, and J. Barahona Nogales
for their help in the performance of this study. J. G. Fernández-
Bolaños and Ó. López wish also to thank the Dirección General
de Investigación of Spain and the Junta de Andalucía (grants
numbers P08-AGR-03751 and FQM134) for financial support.
References
Aguilera, M. P., Beltrán, G., Ortega, D., Fernández, A., Jiménez, A., & Uceda, M. (2005).
Characterisation of virgin olive oil of Italian olive cultivars: ‘‘Frantoio’’ and
‘‘Leccino’’, grown in Andalusia. Food Chemistry, 89, 387–391.
Allalout, A., Krichène, D., Methenni, K., Taamalli, A., Oueslati, I., Daoud, D., et al.
(2009). Characterization of virgin olive oil from super intensive Spanish and
Greek varieties grown in Northern Tunisia. Scientia Horticulturae, 120, 77–83.
Artajo, L. S., Romero, M. P., Suárez, M., & Motilva, M. J. (2007). Partition of phenolic
compounds during the virgin olive oil industrial extraction process. European
Food Research and Technology A, 225, 617–625.
Bahorun, T., Gressier, B., Trotin, F., Brunet, C., Dine, T., Luyckx, M., et al. (1996).
Oxygen species scavenging activity of phenolic extracts from hawthorn fresh
plant organs and pharmaceutical preparations. Arzneimittel Forschung, 46,
1086–1089
.
Bakhouche, A., Lozano-Sánchez, J., Beltrán-Debón, R., Joven, J., Segura-Carretero, A.,
& Fernández-Gutiérrez, A. (2013). Phenolic characterization and geographical
classification of comercial Arbequina extra-virgin olive oils produced in
southern Catalonia. Food Research International, 50, 401–408.
Bendini, A., Cerretani, L., Carrasco-Pancorbo, A., Gómez-Caravaca, A. M., Segura-
Carretero, A., Fernández-Gutiérrez, A., et al. (2007). Phenolic molecules in virgin
olive oils: A survey of their sensory properties, health effects, antioxidant
activity and analytical methods. An overview of the last decade. Molecules, 12,
1679–1719
.
Brenes, M., Hidalgo, F. J., García, A., Ríos, J. J., García, P., Zamora, R., et al. (2000).
Pinoresinol and 1-acetoxypinoresinol, two new phenolic compounds identified
in olive oil. Journal of the American Oil Chemist’s Society, 77, 715–720.
Carrasco-Pancorbo, A., Cerretani, L., Bendini, A., Segura-Carretero, A., Del Carlo, M.,
Gallina-Toschi, T., et al. (2005). Evaluation of the antioxidant capacity of
individual phenolic compounds in virgin olive oil. Journal of Agricultural and
Food Chemistry, 53, 8918–8925.
Clayton, C. A., Hines, J. W., & Elkins, P. D. (1987). Detection limits with specified
assurance probabilities. Analytical Chemistry, 59, 2506–2514.
Franco, M. N., Galeano-Díaz, T., Sánchez, J., De Miguel, C., & Martín-Vertedor, D. (in
press). Total phenolic compounds and tocopherols profiles of seven olive oil
varieties grown in the south-west of Spain. Journal of Oleo Science.
Garratt, D. C. (1964). The quantitative analysis of drugs (Vol. 3). Japan: Chapman and
Hall Ltd., pp. 456–458
.
Godoy-Caballero, M. P., Acedo-Valenzuela, M. I., & Galeano-Díaz, T. (2012). Simple
quantification of phenolic compounds present in the minor fraction of virgin
olive oil by LC-DAD-FLG. Talanta, 101, 479–487.
Gómez-Caravaca, A. M., Carrasco-Pancorbo, A., Cañabate-Díaz, B., Segura- Carretero,
A., & y Fernández-Gutiérrez, A. (2005). Electrophoretic identification and
quantification of compounds in the polyphenolic fraction of extra-virgin olive
oil. Electrophoresis, 26(18), 3538–3551.
Inglese, G., Famiani, F., Galvano, F., Servili, M., Esposto, S., & Urbani, S. (2011).
Factors affecting extra-virgin olive oil composition. Horticultural Reviews, 38,
83–147
.
Jiménez, B., Sánchez-Ortiz, A., Lorenzo, M. L., & Rivas, A. (2013). Influence of fruti
ripening on agronomic parameters, quality indices, sensory attributes and
phenolic compounds of Picudo olive oils. Food Research International. http://
dx.doi.org/10.1016/j.foodres.2013.08.016.
Long, G. L., & Winefordner, J. D. (1983). Limits of detection: A closer look at the
IUPAC definition. Analytical Chemistry, 55, 712–724.
Lozano-Sánchez, J., Segura-Carretero, A., Menéndez, J. A., Oliveras-Ferraros, C.,
Cerretani, L., & Fernandez-Gutierrez, A. (2010). Prediction of extra virgin olive
oil varieties through their phenolic profile. Potential cytotoxic activity against
human breast cancer cells. Journal of Agricultural and Food Chemistry, 58,
9942–9955
.
Manai-Djabali, H., Krichène, D., Ouni, Y., Gallardo, L., Sánchez, J., Osorio, E., et al.
(2012). Chemical profiles of five minor olive oil varieties grown in central
Tunisia. Journal of Food Composition and Analysis, 27 , 109–119.
Martinez, L., Hodaifa, G., & Lozano Peña, J. L. (2010). Changes in phenolic
compounds and Rancimat stability of olive oils from varieties of olives at
different stages of ripeness. Journal of the Science of Food and Agriculture, 90,
2393–2398
.
Martínez, J. M., Muñoz, E., Alba, J., & Lanzón, A. (1975). Informe sobre utilización del
Analizador de Rendimientos ‘‘Abencor’’. Grasas Aceites, 26, 379–385.
Montedoro, G., Servili, M., Baldioli, M., & Miniati, E. (1992). Simple and hydrolizable
phenolic compounds in virgin olive oil. 1. Their extraction, separation and
quantitative and semiquantitative evaluation by HPLC. Journal of Agricultural
and Food Chemistry, 40, 1571–1576.
Oliveras-López, M. J., Innocenti, M., Giaccherini, C., Ieri, F., Romani, A., & Mulinacci,
N. (2007). Study of the phenolic composition of spanish and italian
monocultivar extra virgin olive oils: Distribution of lignans, secoiridoidc,
simple phenols and flavonoids. Talanta., 73, 726–732.
Olszewska, M. A., Presler, A., & Michel, P. (2012). Profiling of phenolic compounds
and antioxidant acitivity of dry extracts from the selected Sorbus species.
Molecules, 17, 3093–3113.
Ouni, Y., Taamalli, A., Gómez-Caravaca, A. M., Segura-Carretero, A., Fenández-
Gutiérrez, A., & Zarrouk, M. (2011). Characterisation and quantification of
phenolic compounds of extra-virgin olive oils according to their geographical
origin by a rapid and resolutive LC-ESI-TOF MS method. Food Chemistry, 127,
1263–1267
.
Owen, R. W., Giacosa, A., Hull, W. E., Haubner, B., Spiegelhalder, B., & Bartsch, H.
(2000). The antioxidant/anticancer potential of phenolic compounds isolated
from olive oil. European Journal of Cancer, 36, 1235–1247.
Paiva-Martins, F., Fernandes, J., Santos, V., Silva, L., Borges, F., Rocha, S., et al. (2010).
Powerful protective role of 3,4-dihydroxyphenylethanol-elenolic acid
dialdehyde against erythrocyte oxidative-induced hemolysis. Journal of
Agricultural and Food Chemistry, 58, 135–140.
Pérez, R. M., Vargas, R., Martínez, F. J., García, E. V., & Hernández, B. (2003). Actividad
antioxidante de los alcaloides de Bocona arborea. Estudio sobre seis métodos de
análisis. Ars Pharmaceutica, 44(1), 5–21.
Prior, R. L., Wu, X., & Schaich, K. (2005). Standardized methods for the
determination of antioxidant capacity and phenolics in foods and dietary
supplements. Journal of Agricultural and Food Chemistry, 53, 4290–4302.
Tovar, M. J., Romero, M. P., Alegre, S., Girona, J., & Motilva, M. J. (2002). Composition
and organoleptic characteristics of oil from Arbequina olive (Olea europaea L.)
trees under deficit irrigation. Journal of the Science of Food and Agriculture, 82,
1755–1763
.
Uceda, M., & Frías, L. (1975). Harvest dates: Evolution of the fruits oil content, oil
composition and oil quality. Proceedings II. Seminario Oleícola Internacional
(pp. 125–128). COI: Córdoba.
Uceda, M., & Hermoso, M. (2001). La calidad del aceite de oliva. In El: ‘‘cultivo del
olivo’’ (pp. 588–614). Madrid: Mundi-Prensa.
Valavanidis, A., Nisiotou, C., Papageorgiou, Y., Kremli, I., Satravelas, N., Zinieris, N.,
et al. (2004). Comparison of the radical scavenging potential of polar and lipidic
fractions of olive oil and other vegetable oils under normal conditions and after
thermal treatment. Journal of Agricultural and Food Chemistry, 52, 2358–2365.
Velasco, J., & Dobarganes, C. (2002). Oxidative stability of virgin olive oil. European
Journal of Lipid Science and Technology, 104, 661–676.
298 M
a
Nieves Franco et al. / Food Chemistry 163 (2014) 289–298