Chapter II

____________________________________________________________________

II.1. Oxidative stress

The importance of oxidation in the body and in foodstuffs has been
widely recognized. Oxidative metabolism is essential for the survival of
cells. A side effect of this dependence is the production of free radicals and
other Reactive Oxygen Species (ROS) that cause oxidative changes
(Antolovich et al., 2002).
Oxygen is the primary oxidant in metabolic reactions designed to
obtain energy from the oxidation of a variety of organic molecules.
Oxidative stress results from the metabolic reactions that use oxygen, and it
has been defined as a disturbance in the equilibrium status of prooxidant/anti-oxidant systems in intact cells (Sies, 1991; Thomas, 1999). This
definition of oxidative stress implies that cells have intact pro-oxidant/antioxidant systems that continuously generate and detoxify oxidants during
normal aerobic metabolism (Thomas, 1999). All forms of life maintain a
reducing environment within their cells. The cellular redox environment is
preserved by enzymes that maintain the reduced state through a constant
input of metabolic energy. When disturbances in this normal redox state are
produced, toxic effects through the excessive production of free radicals or
ROS may occur. Consequently, the pro-oxidant systems outbalance the antioxidants, producing oxidative damage to lipids, proteins, carbohydrates and
nucleic acids. This may ultimately lead to cell death in case of severe
oxidative stress (Halliwell, 1994; Wood et al., 2006).
ROS generation is essential for many important biological processes.
Pincemail et al. (2002) and Havsteen (2002) indicated that ROS, have an
important physiological role, even at very low concentrations, acting as
secondary messengers capable of (1) regulating apoptosis; (2) activating
transcription factors (e.g. Nuclear factor kappa-beta) able to activate genes
implicated in immune and inflammatory responses; (3) modulating gene
expressions in order to activate antioxidant enzymes. However, when there
is an over-production of these ROS that exceeds the capacity of defense,

17

Bibliographic review
____________________________________________________________________

damage to valuable biomolecules may occur. The following important

cellular damages can be produced: (1) ruptures and mutations at the DNA
level; (2) denaturation of proteins; (3) oxidation of sugars; (4) induction of
lipid peroxidation processes (Rice-Evans et al., 1996).
A number of chemical and physical events can initiate oxidation,
which proceeds then continuously in the presence of a suitable substrate until
a blocking defense mechanism occurs. In vivo, many biochemical systems
can lead to high ROS production (inflammation processes, hemoglobin
oxidation, among others) (Pincemail et al., 2002). Besides, the environment
where we live and our life style are also sources oxidative stress increase in
our organism. Exercise can increase the levels of free radicals as can
environmental stimuli such as ionizing radiation (from industry, sun
exposure, cosmic rays, and medical X-rays), environmental toxins, altered
environmental conditions (e.g. hypoxia and hyperoxia; ozone and nitrous
oxide primarily from automobile exhaust), infectious agents, inadequate
eating. Lifestyle stressors such as cigarette smoking and excessive alcohol
consumption are also known to affect oxidative levels in our organism
(Aruoma, 1998; Pincemail et al, 2002).
Oxidation can also affect foods. It is in fact one of the major causes
of chemical spoilage (Colbert and Decaer, 1991), resulting in rancidity
and/or deterioration of the nutritional quality, colour, flavour, texture and
safety of foods (Shahidi et al., 1992). Also, photo-oxidation (Hamilton et al.,
1997; Robards et al., 1999) and the temperature increase (Halliwell, 1994;
Aruoma et al., 1997) are processes inducing radical formation that causes
oxidative deterioration in foods.
As an example, it is known that lipid oxidation is considered one of
the main causes of food deterioration. Lipids (e.g. triacylglycerols,
phospholipids) are naturally found in most biological materials consumed as
foods and added as functional ingredients in many processed foods
contributing to their texture, structure, mouth-feel and flavor. However,

18

Chapter II
____________________________________________________________________

lipids are also one of the most chemically unstable food components and will
readily undergo free-radical chain reactions (German, 1999).
Polyunsaturated fatty acids (PUFA) oxidation can cause unpleasant odors,
flavors and colors in foods, thereby reducing food shelf-life (Richards et al.,
2005). Thus, because of the abundance of unsaturated lipids in muscle foods,
such as meat and fish products and vegetable and fish oils, they are
especially susceptible to the undesirable effects of lipid oxidation (Frankel,
1996; Richards et al., 2005). Moreover, a special attention must be given to
the growing number of food products enriched in PUFA (omega-6 and
omega-3). In addition, the lipids are dispersed in an aqueous phase in most
of these foods. This dispersed state of lipids increases susceptibility to
oxidation by favouring their contact with oxygen and pro-oxidant species
(e.g. metal ions) dissolved in the aqueous phase. This increases the risk of
development of oxidation during technological processes or storage (Villire
et al., 2005; Villire et al., 2006).
Montero et al. (2005) mentioned that protein oxidation may occur
more rapidly than lipid oxidation in systems such as muscle, because protein
is within the aqueous phase where many free radicals are formed.
Furthermore, protein radicals can react with susceptible lipids to enhance the
rate of lipid oxidation.
Taking into account the previous considerations, it appears of utmost
importance to produce and characterize components that may protect the
food matrices from oxidative damage. On the other hand, mechanistic
studies on these protective properties may be helpful for the design of
optimized antioxidant agents.

II.1.1.

Reactive oxygen or nitrogen species

A free radical can be defined as any chemical species having one or

more unpaired electrons (Hamilton et al., 1997). Free radicals are thus

19

Bibliographic review
____________________________________________________________________

highly unstable, reactive and energized molecules that have electrons

available to react with various organic substrates (Rice-Evans and
Gopinathan, 1995; Schmidl and Labuza, 2000; Lee et al., 2004). One of
characteristics of free radical reactions is that frequently they are not
selective, but random.
There are numerous types of free radicals that can be formed within
the body. Most free radicals in biological systems are derivatives from
oxygen, but there are also derivatives of nitrogen. One of the most reactive
and damaging free radical is the hydroxyl radical (OH). Hydroxyl radical is
considered to be a principal actor in the toxicity of partially reduced oxygen
species since it is very reactive with all kinds of biological macromolecules.
ROS is a collective term describing radicals and other non-radical reactive
oxygen derivatives. These are responsible for the majority of radical
degradations (Aruoma et al., 1997). Important ROS in living organisms are
presented in Table 1.
Reactive Nitrogen Species (RNS) are nitrogen-based molecules that
can act to facilitate nitrosylation reactions. RNS include among others: nitric
oxide (NO), peroxynitrite (ONOO), peroxynitrous acid (ONOOH),
nitroxyl anion (NO), nitrogen dioxide (NO2), nitrous acid (HNO2)
(Schmidl and Labuza, 2000).
Both ROS and RNS may participate in reactions giving rise to free
radicals or damage to organic substrates.
Most free radicals are produced by mitochondria where 0.4 to 5% of
the oxygen to generate energy results in the formation of ROS, such as
superoxide and hydrogen peroxide. It is thus no wonder that most of the free
radical damage is to mitochondrial membranes and mitochondrial DNA
(Davies 1995; Lee et al., 2004).

20

Chapter II
____________________________________________________________________

Table 1. Some important radical and non-radical reactive oxygen species in

living organisms
Radicals

Non-Radicals

Hydroxyl

(OH)

Hypochloric acid

(HOCl)

Superoxide

(O2)

Hydrogen
Peroxide

(H2O2)

Peroxyl

(ROO)

Singlet Oxygen

(1O2)

Lipid peroxyl

(LOO)

Ozone

(O3)

Lipid peroxide

(LOOH)

Source: Adapted from ArUoma et al. (1997) and Schmidl and Labuza (2000).

Almost all cell types are capable of producing ROS during the
mitochondrial oxidation. In addition, several cell types do specifically
generate ROS or RNS for dedicated purposes. For example, ROS such as
superoxide and hydrogen peroxide are produced by phagocytes
(macrophages and neutrophils) in order to kill some types of bacterias. The
NO is produced within cells by the actions of a group of enzymes called
nitric oxide synthases. NO has the ability to dilate blood vessels and relax
smooth muscle tissue (Drew and Leeuwenburgh, 2002).
Free radicals react with target organic substrates such as lipids,
proteins, and DNA. Lipids, particularly the PUFA located in the membrane
lipophilic section are subject to free radicals attack, which may lead to the
production of lipid peroxides. This process is initiated by the action of an
oxidizing radical, such as hydroxyl radical or superoxide anion (R)
(Aruoma, 1998):

21

Bibliographic review
____________________________________________________________________

LH (substrate) + R  L (lipidic radical) + RH

L + O2  LOO (lipid peroxyl radical)
LOO + LH  LOOH (lipid peroxide) + L

II.1.2.

Antioxidant defenses and antioxidants

The aerobic organisms are protected from ROS and RNS by means
of defensive systems. These include various antioxidants, which have
different functions. An antioxidant may be defined as any substance that
when present at low concentration, compared with those of the oxidisable
substrate, significantly delays or inhibits oxidation of that substrate
(Halliwell, 1990; Gutteridge, 1994). The antioxidants comprise enzymes
(e.g. superoxide dismutase, glutathione peroxidase, and catalase), proteins
(albumin, ferritin) and small molecules (ascorbic acid, glutathione, uric acid,
tocopherol, carotenoids, polyphenols) (Prior et al., 2005). Foods are an
essential source of small-size antioxidants, as well as of trace elements,
which play an important role as enzyme or protein cofactors (Shi and
Noguchi, 2001). Table 2 shows some antioxidants, which constitute the
defense system in vivo.
As shown in Table 2, there are several lines of defense. The first line
of defense is to inhibit the formation of ROS. This defense system is
composed by preventive antioxidants that retard the rate of oxidation. This
may be achieved in a number of ways including removal of substances,
chelating transition metals or singlet oxygen quenching (Frankel and Meyer,
2000).

22

Chapter II
____________________________________________________________________

Table 2. Defense systems in vivo against oxidative damage

1. Preventive antioxidants: suppress the formation of free radicals.
(a) Non-radical decomposition of hydroperoxides and hydrogen peroxide:
Decomposition of hydrogen peroxide
2 H2O2  2 H2O + O2

Catalase

Glutathione
(cellular)

peroxidase

Decomposition of hydrogen peroxide and free fatty

acid hydroperoxides
H2O2 + 2 GSH  2 H2O + GSSG
LOOH + 2GSH  LOH + H2O + GSSG

Glutathione
(plasma)

peroxidase

Decomposition of hydrogen peroxide

phospholipid hydroperoxides
PLOOH + 2 GSH  PLOH + H2O + GSSG

Glutathione-S-transferase

and

Decomposition of lipid hydroperoxides

Peroxidase

Decomposition of hydrogen peroxide and lipid

hydroperoxides
LOOH + AH2  LOH + 2 H2O + A
H2O2+ AH2  2 H2O + A
(b) Sequestration of metal by chelation:
Transferrin, lactoferrin
Sequestration of iron
Haemopexin
Stabilisation of haem
Haptoglobin
Sequestration of haemoglobin
Ceruloplasmin, albumin
Sequestration of copper
(c) Quenching of superoxide and singlet oxygen:
Superoxide
dismutase Disproportionation of superoxide
(SOD)
2O2  + 2H+ H2O2 + O2
Carotenoids, vitamin E
Quenching of singlet oxygen
2. Radical-scavenging antioxidants: scavenge radicals to inhibit chain initiation
and break chain propagation.
Hydrophilic: vitamin C, uric acid, bilirubin, albumin
Lipophilic: vitamin E, ubiquinol, carotenoids, flavonoids
3. Repair and de novo enzymes: repair the damage and reconstitute membranes.
Source: Adapted from Shi and Noguchi (2001)

23

Bibliographic review
____________________________________________________________________

The radical-scavenging antioxidants are the second line of defense.

When these antioxidants are present in the system, they either delay or
inhibit the initiation step by reacting with the free radical (L) or inhibit the
propagation step by reacting with peroxyl (LOO) or alkoxyl (LO) radicals:

L + AH  LH + A

LOO + AH  LOOH + A

LO + AH  LOH + A

The antioxidant free radical (A) may further interfere with chainpropagation reactions by combining with other radicals (termination
reactions):

A + LOO  LOOA

A + LO  LOA
The third-line of defense is the repair. Various enzymes such as
lipases, proteases and DNA repair enzymes are responsibles for such defence
(Schmild and Labuza, 2000).
Gutteridge (1994) indicated that other mechanisms of action for
antioxidants include: enhancing endogenous antioxidant defenses by upregulating the expression of the genes encoding the antioxidant enzymes,
increasing elimination of damaged molecules and not repairing excessively
damaged molecules in order to minimize introduction of mutations.
Inhibition of oxidative enzymes (e.g. cyclooxygenase) is another mechanism
of action of antioxidants (Huang et al., 2005).
The most suitable antioxidants are those that perform one or more of
the above functions, without generating any toxic or reactive end-products.
As long as adequate amounts of antioxidants are present to provide sufficient

24

Chapter II
____________________________________________________________________

protection, the antioxidant-oxidant balance is maintained (Gutteridge, 1994;

Halliwell, 1996).
When the classical antioxidant defenses or biological antioxidants
are not sufficient to avoid the oxidative damage in our organism, the
consumption of external antioxidants must be increased through the diet or
through food supplements (e.g. vitamins C and E, carotenoids and
polyphenols) (Niki et al., 1995; Vansant et al., 2004). Compelling
epidemiological evidence is available on the benefits of consuming diets rich
in plant-based foods. This has led to a current advice consisting in the daily
ingestion of several portions of fruits and vegetables for optimal health
(Hooper and Cassidy, 2006). The consumption of fruits and vegetables
containing high amounts of natural antioxidant molecules has indeed been
associated with an improvement of the balance between free radicals and
antioxidants, which helps to minimize the oxidative stress in the body and to
reduce the risks of degenerative diseases (Prior, 2003; Lee et al., 2004).
Natural antioxidant compounds derived from plants, such as ascorbate,
tocopherol, carotenoids, bioactive plant phenols, are of the considerable
interest from the viewpoint of dietary antioxidants supplementation, as well
as of food preservation (Halliwell et al., 1995). As a result result, a lot of
research has been conducted to identify and evaluate antioxidants from
agricultural by-products, ethnic and traditional products, herbal teas, cold
pressed seed oils, exudate resins, hydrolysis products, not evaluated fruits
and edible leaves and other raw materials rich in antioxidants that have
nutritional importance and/or the potential for applications in the promotion
of health and prevention against damages caused by radicals (Dimitrios,
2006).

II.1.3.

Methods for testing antioxidant capacity

There are a great variety of chemical and biological methods to

evaluate the efficacy of antioxidants against oxidation. In oxidative
processes, multiple reactions and mechanisms as well as different steps are

25

Bibliographic review
____________________________________________________________________

usually involved. Therefore, no single assay will accurately reflect all the
radical sources or all the effects of antioxidants in a mixed or complex
system (Frankel and Meyer, 2000; Awika et al., 2003; Prior et al., 2005).
The features involved in oxidation are a substrate, an oxidant and an
initiator, intermediates and final products. The measurement of any one of
these can be used to assess antioxidant capacity (Clarkson, 1995). Thus, the
antioxidant capacity of a sample can vary as a function of the choice for a
particular substrate, oxidant, initiator or product formed (that indicates the
end-point of reaction), as well as with the particular combination of them in
the reaction medium. The medium conditions (e.g. pH, hydrophilicity,
temperature, polarity) where oxidation occurs are another factor that may
affect the antioxidant capacity of molecules (Moure et al., 2001). Taking all
these constraints into account, it is recommended to evaluate the efficiency
of antioxidant molecules on the basis of the results obtained by several
methods, taking into account the different oxidative pathways, the analytical
methods used to determine the extent and end-point of oxidation and
different physico-chemical properties of the oxidisable substrates (Frankel,
1993a; Arnao et al., 1999, Frankel and Meyer, 2000; Sanchez-Moreno,
2002).
It is very important that methods to assess the antioxidant status
reflect the oxidative processes associated with both the food and
physiological systems (Silva, 2006). In foods, it is necessary to determine
the efficacy of natural antioxidants for food protection against oxidative
damage, to avoid deleterious changes and loss of commercial and nutritional
value (Halliwell et al., 1995; Halliwell, 1997). On the other hand, Wood et
al. (2006) mentioned that is important to evaluate the intake of dietary
antioxidants and their real contribution to the antioxidant status of the human
organism.
Prior et al. (2005) and Roginsky and Lissy (2005) listed the major
requirements neccesary for a good method of antioxidant capacity
evaluation: (1) its mechanism of action must be based on a well-developed

26

Chapter II
____________________________________________________________________

theory and be described by a definitive kinetic scheme; (2) it must make use
of a biologically relevant radical source; (3) it has to be simple to allow the
determination to be carried out in any laboratory; (4) the instrumentation
must be readily available; (5) the repeatability of determinations should be
sufficient within each scientific work and the reproducibility of the assay
must be acceptable among all different laboratories; (6) it should be
adaptable for both hydrophilic and lipophilic antioxidants and for different
radical sources; (7) the mode of quantification must be clearly defined; (8) it
should be adaptable to high-throughput analysis for routine quality control
analyses.
In a review over the methodologies for the determination of
biological antioxidant capacity in vitro, MacDonalds-Wicks et al. (2006)
mentioned that ideally, the antioxidant capacity should be tested using both
in vitro and in vivo techniques but due to the high cost of conducting animal
and human feeding trials, many products undergo only in vitro testing.
Different in vitro assays have been described by the same authors such as:
methods that measure the inhibition of induced lipid autoxidation (possible
susbtrates could be: low density lipoproteins (LDL) or unsaturated fatty
acids) and others as oxygen radical absorbance capacity (ORAC) assay, total
radical-trapping antioxidant parameter (TRAP) assay, trolox equivalent
antioxidant capacity (TEAC) assay (or ABTS assay), ferric iron reducing
antioxidant power (FRAP) assay, 2,2-diphenyl-1-picrylhydrazyl radical
(DPPH) scavenging capacity assay. These same assays have been proposed
by Sanchez-Moreno (2002) to estimate the oxidation in foods and biological
systems. On the other hand, Antolovich et al. (2002) examined various
methods of measuring antioxidant capacity related to lipid oxidation such as:
accelerated stability tests, peroxide value, diene conjugation, thiobarbituric
acid reactive substances (TBARS) and measurement of hexanal.
An excellent compilation of methods currently used to determine the
in vitro antioxidant capacity has been performed by Prior et al. (2005) and
Souza (2007) and is presented in Table 3. The following sections describe

27

Bibliographic review
____________________________________________________________________

the in vitro methodologies that have been used in this thesis to determine the
antioxidant capacity.

II.1.3.1.

ORAC assay

The oxygen radical absorbance capacity (ORAC) assay has been

widely used in measuring the net resultant antioxidant capacity of botanical
and other biological samples (MacDonald-Wicks et al., 2006). Initially, this
method was developed by Cao et al., (1993) and is based on the
measurement of the free radical damage to a fluoresence probe through the
change in its fluorescence intensity. The change in fluorescent intensity is an
index of the degree of free radical damage. The inhibition of free radical
damage by an antioxidant, visualized throug the protection against the
change of probe fluorescence, is a mesure of its antioxidant capacity against
the free radical (Cao et al., 1993; Ou et al., 2001; Huang et al., 2002a).
In the ORAC assay, different generators can be used to produce
different radicals such as the peroxyl radical (ROO), the hydroxyl radical
(OH), or the transition metal Cu+2. However, the method has adopted the
ROO as standard radical, since it is the most common in biological systems.
Initially, the target protein was -phycoerythrin (-PE), whose loss of
fluorescence was an indication of the extent of damage from its reaction with
the peroxyl radical (Cao et al., 1993). Ou et al. (2001) adopted a new
fluorescent substance (fluorescein, FL) to replace -PE as a probe, because
-PE showed large inter-batch differences and interacted with polyphenols,
through non-specific protein binding, resulting thus in a loss of fluorescence
even without the added radical generator (Ou et al., 2001).
In general, either the sample, or a control or a Trolox standard (a
water soluble analogue of vitamin E, at four or five different concentrations,
used to construct a standard curve), are mixed with the FL solution and

28

Chapter II
____________________________________________________________________

Table 3. Comparison of methods for assessing in vitro antioxidant capacity

based upon mechanism, ROS generator, medium of reaction, endpoint,
quantitation method and the adaptability of assays to measure lipophilic and
hydrophilic antioxidants.
Antioxidant
assay

Mechanism

ROS
generator

Medium
of
reaction

ORAC
TRAP
FRAP

HAT
HAT
SET

AAPH
AAPH
TPTZ/Fe+3

pH 7.4
pH 7.4
pH 3.6

TEAC

SET/HAT

ABTS

FC
DPPH
CUPRAC
TOSC

SET
HAT
SET
HAT

TBARSLDL
DinesLDL
Hmolyse

HAT

FC reactive
DPPH
Cu+2
KMBAABAP
AAPH

pH 7.0/
EtOH
pH > 10
MeOH
H2O
pH 7.4

SET
HAT

Endpoint

Quantitation

Fixed time
AUC
Lag phase IC50, lag time
Times,
'DO fixed time
varies
Time
'DO fixed time

Lipophilic
and
hydrophilic
AOC
LP/HP
HP
HP
LP/HP

Fixed time 'DO fixed time

IC50
'DO fixed time
Time
'DO fixed time
IC50
AUC

HP
HP
HP
HP

pH 7.4

Lag phase

Lag time

HP

Cu+2

pH 7.4

Lag phase

Lag time

HP

AAPH

pH 7.4

Lag phase 'DO fixed time

HP

AAPH:
2,2-azobis(2-amidinopropane)
dihydrochloride ;
ABAP:
2,2-Azobis(2methylpropionamidine)
dichloride ;
ABTS:
2,2-azinobis(3-ethylbenzothiazoline)-6sulfonate; AOC: Antioxidant capacity; AUC: Area under the courve; CUPRAC: Cupricreducing antioxidant capacity; DO: absorbance; DPPH: 2,2-diphenyl-1-picrylhydrazyl;
EtOH :Ethanol; FC: Folin-Ciocalteu; FRAP: Ferric reducing ability of plasma; KMBA: Dketo--methiolbutyric acid; LDL: Low density lipoproteins; LP: Lipophilic; HP: Hydrophilic;
ORAC: Oxygen radical absorbance capacity; HAT: Hidrogen atom transfer; TBARS:
Thiobarbituric acid reactive substances; SET: Single electron transfer; TEAC: Trolox
equivalent antioxidant capacity; TOSC: Total oxidant scavenging capacity; TPTZ: 2,4,6Tripyridyl-1,3,5-Triazine; TRAP: Total radical-trapping antioxidant parameter (Source: Prior
et al., 2005; Souza, 2007).

incubated at 37C, before 2,2-azobis(2-amidinopropane) dihydrochloride

(AAPH) initiates the reaction. The reaction is measured at excitation: 485 nm
and emission: 525 nm for changes in fluorescence every minute for 35 min. As
the reaction progresses, FL is consumed and the fluorescence diminishes. In
the presence of an antioxidant, the decay of FL is retarded. The data from the

29

Bibliographic review
____________________________________________________________________

assay are obtained by calculating the area under curve (AUC) and net AUC
(AUCsample  AUCblank (without sample)) (Figure 1). A standard curve is built up
by plotting AUC against Trolox concentrations. The ORAC value of a
sample is expressed in Trolox equivalents thanks to the use of the standard
curve (Ou et al., 2001; Huang et al., 2002a).

Figure 1. Calculation for ORAC assay: Antioxidant capacity of tested sample

expressed as the net area under the curve (AUC) (Source: Prior et al., 2005).

ORAC can be automated and can test hydrophilic and lipophilic

antioxidant capacity (Huang et al., 2002b). The ORAC method is reported to
mimic the antioxidant capacity of phenols in biological systems better than
other methods since it uses biologically relevant free radicals and integrates
both time and degree of activity of antioxidants (Cao et al., 1993; Ou et al.,
2001).

II.1.3.2.

ABTS assay

The ABTS assay was developed by Miller et al. (1993) and Re et al.
(1999) and has been used widely for testing antioxidant capacity in food
samples. The ABTS assay measures the relative ability of an antioxidant to
scavenge the ABTS+ radical (compound intensely colored). The ABTS+
radical is generated by the reaction of a strong oxidizing agent (e.g.,
potassium permanganate or potassium persulfate) with the ABTS salt (Re et

30

Chapter II
____________________________________________________________________

al., 1999; Cano et al., 2002). The principle of the method is to monitor the
decay of the radical-cation ABTS+ (disseapareance of blue-green color)
caused by the addition of an antioxidant (Miller and Rice-Evans, 1997)
(Figure 2).

Figure 2. Reaction between ABTS+ and antioxidant (Source: Miller and RiceEvans, 1997; Huang et al., 2005).

ABTS+ has a strong absorption at the wavelength of 734 nm and can

be easily determined spectrophotometrically. In the absence of antioxidant,
ABTS+ is rather stable, but it reacts energetically in the presence of a
scavenger antioxidant, such as a phenolic compound. The extent of color
loss of ABTS+ is determined as a function of the concentration of an
antioxidant-containing sample and time and calculated relative to the Trolox

31

Bibliographic review
____________________________________________________________________

standard determined under the same conditions (Arnao et al., 2001; Awika et
al., 2003).
The ABTS assay is rapid and can be used in both aqueous and
organic solvent systems (Arnao et al., 1999; Lemanska et al., 2001). It has
also good repeatability and is simple to perform; hence, it is widely reported
(Arnao et al., 2001). Several limitations can be listed: 1) ABTS+ is not
found in biological systems and is not similar to radicals found in those
systems. 2) The reaction with ABTS+ occurs rather slowly with many
phenolics and samples of natural products (Campos and Lissi, 1997; Lissi et
al., 1999). Thus, the result of the determination of the ABTS assay is
expected to be dependent on the time of incubation as well as on the ratio of
sample quantity to ABTS+ concentration. 3) ABTS+ reacts with any
hydroxylated aromatic compound independently of its real antioxidative
potential (Arts et al., 2003).

II.1.3.3.

Methods based on lipid peroxidation

Several assays have been designed to evaluate lipid peroxidation. In

these assays a lipid (e.g. unsaturated fatty acid) or lipoprotein (LDL)
substrate is used under standard conditions and the degree of inhibition of
oxidation given by an antioxidant is measured (Snchez-Moreno and
Larrauri, 1998).
Conjugated dienes assay. In this method, the autoxidation of unsaturated
fatty acids (e.g linoleic acid) or LDL is induced by Cu+2/Fe+3 or an azo
initiator as AAPH. Conjugated dienes are intermediate compounds of lipid
oxidation (Figure 3). These compounds present a conjugated double-bond
and their resonance can be measured at 234 nm (Puhl et al., 1994). Thus, the
progress of oxidation in this reaction is monitored by UV absorbance at 234
nm (Pryor et al., 1993).

32

Chapter II
____________________________________________________________________

Tipically, the assay solution contains a free radical initiator, a

substrate and an antioxidant. A characteristic kinetics of conjugated diene
formation is represented by three phases: a latence phase (lag time), a
propagation phase (conjugated diene oxides accumulate rapidly) and a
decline phase (degradation phase). Thus, when the antioxidant inhibits the
oxidation process, longer lag times are obtained with respect to the reaction
without antioxidant. The duration of the lag time is dependent on the
concentration and capacity of the antioxidant molecule to inhibit oxidation
(Puhl et al., 1994; Huang et al., 2005). The quantification of the conjugated
dienes may be achieved by calculating the increase in absorbence per mass
of sample at a fixed time (Frankel et al., 1996), lag phase measurements or
as inhibition percentage (Antolovich et al., 2002).

Figure 3. Mechanism of lipid peroxidation (Source: Horton, 1987)

The measurement of the formation of conjugated dienes has the

advantage that it measures an early stage in the oxidation process.

33

Bibliographic review
____________________________________________________________________

Conjugated diene measurements often cannot be performed directly on

tissues and body fluids because many other interfering substances are
present, such as haem proteins, purines, pyrimidines that strongly absorb in
the same UV region (Maldhavi et al., 1996).
Thiobarbituric acid reactive substances (TBARS) assay. This method
is widely used to detect lipid oxidation (Puhl et al., 1994). This procedure
measures the malondialdehyde (MDA) formed as the split product of an
endoperoxide of unsaturated fatty acids resulting from oxidation of a lipid
substrate (Figure 3). MDA is a final compound of lipid oxidation
(degradation step). To evaluate the antioxidative efficacy of the compounds
against lipid oxidation, a substrate (e.g. linoleic or other fatty acids or LDL),
an initiator (Cu+2/Fe+3, or AAPH), an antioxidant and thiobarbituric acid
(TBA) are required. The MDA reacts with TBA to form a pink pigment
(Figure 4) that is measured spectrophotometrically at its maximum
absorption at 532 535 nm (Fernndez et al., 1997). Measurements can be
made also by fluorescence at a  excitation of 515 nm and  emission of 555 nm
(Janssens et al., 2002; Walter et al., 2004; Warnier, 2006; Souza et al.,
2008). The procedure involves two steps: the substrate is oxidized with
addition of the initiator and then the extent of oxidation is determined by the
addition of TBA and a spectrophometric measurement of the product
(TBARS) (Esterbauer and Cheeseman, 1990; Puhl et al., 1994). Oxidation is
inhibited by addition of an antioxidant and therefore a reduction in the
absorbance is seen.
Results are typically quantified against a calibration curve for
malondialdehyde
bis-(dimethylacetal)
or
malondialdehyde
bis(diethylacetal), which acts as a source of MDA. Results may also be
described in terms of inhibition percentage of the oxidation. The reaction in
the TBARS assay is not very specific because TBA forms complexes with
other compounds (e.g. aldehydes, oxidized lipids, etc) and also the reaction
conditions have a significant effect on colour development (Esterbauer and
Cheeseman, 1990).

34

Chapter II
____________________________________________________________________

Figure 4. Chromophore formed by condensation of MDA with TBA (Source:

Antolovich et al., 2002)

Erythrocyte oxidation assay. This assay was proposed by

Yamamoto et al. (1985) as a mechanism to evaluate oxidation of biological
membranes. Erythrocytes are vulnerable to lipid peroxidation due to their
high content of polyunsaturated lipids, their rich oxygen supply, and the
presence of transition metals (Miki et al., 1987; Niki, 1990; Tedesco et al.,
2000). ROS generated in the aqueous or lipid phases can attack erythrocyte
membranes and can induce the oxidation of lipids and proteins, triggering
disruptions of the membrane and hemolysis (Miki et al., 1987; Niki et al.,
1988, Tedesco et al., 2000). Numerous investigations have used erythrocytes
as model systems for studying biomembrane oxidative damage (Miki et al.,
1987; Niki et al., 1988; Janssens et al., 2002; Dai et al., 2006). In many of
these studies, free radical initiators such as AAPH have been used to
generate free radicals in the aqueous phase in order to attack the erythrocyte

35

Bibliographic review
____________________________________________________________________

membrane and propagate lipid peroxidation, leading to hemolysis (Miki et

al., 1987; Niki et al., 1988).
The assay consists in measuring the resistance of red blood cells in
the presence of an initiator (e.g. AAPH) and antioxidant compounds (e.g.
polyphenols, vitamin E or C) to the lipid peroxidation of their membranes.
The quantity of disrupted red cells (hemolysis) can be measured at 540 nm
(absorbance of oxyhemoglobin) (Mc Donald, 1994). This measurement
describes the erythrocyte degradation in function of time (Zhu et al., 2002;
Janssen et al., 2002; Dai et al., 2006). Protection of erythrocytes by an
antioxidant is deduced from the time required for half-hemolysis, as
compared to a control (without antioxidant). The inhibition of hemolysis of
erythrocytes is calculated as % inhibition (Miki et al., 1987; Zhu et al.,
2002).
Niki et al. (1988) mentioned that in the erythrocyte oxidation assay,
the water-soluble, chain-breaking antioxidants such as ascorbic acid was
able to scavenge the oxygen radicals present in the aqueous phase, while
lipid-soluble, chain-breaking antioxidants such as D-tocopherol scavenged
predominantly the radicals within the lipid region of the membranes. In
addition, Liao and Ying (2000) indicated that the interaction of flavonoids
with bio-membranes was an important factor in determining their structureactivity relationship. The interaction of an agent with membranes, or the
uptake of an agent into the membranes is strongly related to its partition
coefficient. Flavonols such as myricetin, quercetin and rutin which possess
an ortho-dihydroxyl functional group have shown much more effective antihemolysis activity than other flavonols (morin and kaempherol) bearing no
such functional group (Dai et al., 2006).

36

Chapter II
____________________________________________________________________

II.2. Phenolic compounds

In the last years, researchers and food manufacturers have become
increasingly interested in phenolic compounds. The main reason for this
interest is the recognition of the antioxidant properties of phenolics (Yang et
al, 2001), their great abundance in our diet, and their probable role in the
prevention of various diseases associated with oxidative stress, such as
cancer and cardiovascular and neurodegenerative diseases (Manach et al.,
2004).
Many properties of plant and plant products are associated with the
presence, type and content of their phenolic compounds (Robards et al.,
1999; Cheynier, 2005). In plants, phenolics play among others the role of
attractants for polynators, contributors to plant pigmentation, antioxidants,
protective agents against UV light (Robards and Antolovich, 1997; Harborne
and Williams, 2000), signal substances for establishment of symbiosis with
rhizobia, insulating materials to make cell walls impermeable to gas and
water and structural materials to give plant stability (Shahidi and Nackz,
2004). Besides, polyphenols are responsible for the major organoleptic
characteristics of plant-derived food and beverages, particularly in terms of
bitterness, astringency, colour, flavour and odour of products. They are also
reported to contribute to the health benefits associated with the consumption
of diets high in fruits and vegetables or plant-derived beverages (such as tea
and wine) (Cheynier, 2005).

II.2.1.

Classification and chemical structures

Phenolic compounds form one of the main classes of secondary

metabolites in plants with a large range of structures and functions. They are
found in fruits, vegetables, grains, barks, roots, stems, flowers, as well as in
their respective derived products.

37

Bibliographic review
____________________________________________________________________

Phenolics are not uniformly distributed in plants at the tissue,

cellular and subcellular levels. Insoluble phenolics are the components of
cell walls while soluble phenolics are compartmentalized within the plant
cell vacuoles (Yamaki, 1984; Beckman, 2000). At the tissue level, the outer
layers of plants contain higher levels of phenolics than those located in their
inner parts (Fernndez de Simn et al., 1992; Bengochea et al., 1997). Cell
wall phenolics, such as lignins (the polymer of monolignol units) and
hydroxycinnamic acids are linked to various cell components. These
compounds contribute to the mechanical strength of cell walls and play a
regulatory role in plant growth and morphogenesis and in the cell response
to stress and pathogens (Lewis and Yamamoto, 1990; Nackz and Shahidi,
2004).
The definition of phenolic compounds in terms of its metabolic
origin is: substances derived from the shikimate pathway and
phenylpropanoid metabolism (Robards et al., 1999; Shahidi and Nackz,
2004) (Figure 5).
Plant phenolics comprise a great diversity of compounds. These
compounds may be classified into different groups as a function of the
number of phenol rings that they contain and of the structural element that
binds these rings to one another. Distinctions are thus made between the
phenolic acids, stilbenes, lignins (nonflavonoids) and flavonoids (Figure 6)
(Robards et al., 1999; Manach et al., 2004). In addition to this diversity,
phenolic compounds may be associated with various carbohydrates and
organic acids and with one another (Cheynier, 2005). Next, a brief
description of some phenolic compounds found in the vegetal kingdom is
presented.

38

Chapter II
____________________________________________________________________

Figure 5. Production of phenylpropanoids, stilbenes, lignans, lignins, suberins,

cutins, flavonoids and tannins from phenylalanine. PAL denotes phenylalanine
ammonia lyase (Source: Shahidi and Nackz, 2004).

39

Bibliographic review
____________________________________________________________________

Phenolic compounds

Simple phenols
(C6)

Phenolic acids

Hydroxybenzoic
acids
(C1-C6)

Chalcones

Flavones

Isoflavones

Flavonoids
(C6-C3-C6)

Stilbenes
(C6-C2-C6)

Lignins
(C6-C3)n

Hydroxycinnamic
acids
(C3-C6)

Flavanones

Flavonols

Flavanonols

Proanthocyanidins

Anthocyanins

Flavanols

(Polymerization)

Figure 6. Classification of the major phenolic compounds (Source: Robards et

al., 1999).

II.2.1.1.

Phenolic acids

Phenolic acids have recently received considerable attention for their

protective antioxidant behavior and potential health benefits (Robbins, 2003;
Biroova et al., 2007). It has been reported that some phenolic acids posses
antitumor activity against colon carcinogenesis (Olthof et al., 2001), besides
inhibiting the AP-1 transcriptional activity implicated in the control of
inflammation, cell differentiation, and proliferation (Maggi-Capeyron et al.,
2001) and are potentially considered as protective compounds against cancer
and heart diseases (Wen et al., 2005).

40

Chapter II
____________________________________________________________________

The name phenolic acids, in general, describes phenols that possess

one carboxylic acid functionality. Phenolic acids are subdivided into
derivatives of hydroxycinnamic acids and derivatives of hydroxybenzoic
acids (Figure 7), on the bases of C3-C6 and C1-C6 skeletons, respectively
(Cheynier, 2005).
As seen in Figure 7(a), cinnamic acids present hydroxylations and
methylations in their structures. These compounds possess a phenyl ring (C6)
and a C3 side chain and are thus collectively termed phenylpropanoids. They
serve as precursors for the synthesis of lignins and many other compounds
(Shahidi and Nackz, 2004).
The most common hydroxycinnamic acid derivatives are pcoumaric, caffeic, sinapic and ferulic acids. These acids are rarely found in
the free form, except in processed food that has undergone freezing,
sterilization, or fermentation. The bound forms are glycosylated (glucose)
derivatives or esters of quinic acid, shikimic acid and tartaric acid. Likely,
the most familiar of these is chlorogenic acid (caffeic acid and quinic acid
esterified) (Herrmann, 1989; Manach et al., 2004).

The hydroxybenzoic acid content of edible plants is generally

very low, with the exception of certain red fruits, black radish and onions,
which can have concentrations of several tens of milligrams per kilogram
of fresh weight (Shahidi and Nackz, 1995). Similar to cinnamic acids,
hydroxylation and methylation of hydroxybenzoic acid leads to the
formation of dihydroxybenzoic acid (protocatechuic acid), vanillic acid,
syringic acid, p-hydroxybenzoic acid and gallic acid. Hydroxybenzoic
acids are commonly present in the bound form in foods and are often the
component of a complex structure like lignins and hydrolyzable tannins
(e.g. gallotannins, ellagitannins) (Khanbabaee and van Ree, 2001).

41

Bibliographic review
____________________________________________________________________

(a)

(b)

Figure 7. Phenolic acids: members of the cinnamic (a) and benzoic acid (b)
families found in food and nutraceuticals (Source: Nackz and Shahidi, 2004).

II.2.1.2.

Flavonoids

The flavonoids are typical phenolic compounds that act as potent

metal chelators and free radical scavengers (Harborne and Williams, 2000;
Heim et al., 2002) and they are powerful chain-breaking antioxidants.
Besides, some flavonoids have long been recognized to possess antiinflammatory, antiallergic, hepatoprotective, antithrombotic, antiviral and
anticarcinogenic activities (Middleton et al., 2000).

42

Chapter II
____________________________________________________________________

Flavonoids are built upon a C6-C3-C6 flavone skeleton in which the

three-carbon bridge between the phenyl groups is commonly cycled with
oxygen (Figure 8). Several classes of flavonoids are differentiated on the
basis of the degree of unsaturation and degree of oxidation of the threecarbon segment. Within the various classes, further differentiation is possible
on the basis of the number and nature of substituent groups attached to the
rings (Robards et al., 1999). Thus there are numerous substitution patterns in
which primary substituents (e.g. hydroxyl, methoxyl, aromatic or glycosyl
groups) can themselves be substituted (e.g. additionally glycosylated or
acylated), sometimes yielding highly complex structures (Cheynier, 2005).

Figure 8. Nuclear structure of flavonoids (Source: Heim et al., 2002).

Flavonoids are subdivided in several groups. Flavanols, flavonols

and anthocyanins are the quantitatively dominant groups in plants (Robards
and Antolovich, 1997).
Flavanols. Flavanols are also known as flavan-3-ols, flavans or
catechins. These compounds exist in the monomeric form as well as in the
form of oligomers and polymers, which are both referred to as condensed
tannins or proanthocyanidins. The main flavanols are catechins and they are
found in many types of fruits (e.g. apricots, apples), as well as in green tea,
red wine and chocolate (Yang et al., 2001). Catechin and epicatechin are the
main flavanols in fruits, while gallocatechin, epigallocatechin and
epigallocatechin gallate are found in certain seeds of leguminous plants,
grapes and more importantly in tea (Arts et al, 2000a, Arts et al., 2000b).
The flavan-3-ols (flavanol monomer) such as: catechin, epicatechin,

43

Bibliographic review
____________________________________________________________________

gallocatechin and epigallocatechin are found in fruits, generally in the free

form rather than in the glycosylated forms (Robards et al., 1999; Manach et
al., 2004). The chemical structure of different flavan-3-ol units is shown in
Figure 9.

Figure 9. Structure of flavan-3-ol units (Source: Gu et al., 2003).

Structurally, the proanthocyanidins are composed of mixtures of

oligomers and polymers containing flavan-3-ol units, linked mainly through
C4oC8 and/or C4oC6 bounds (both are called B-type, Figure 10). The
flavan-3-ol units can also be doubly linked by an additional ether bond
between C2oO7 (A-type) (Porter et al., 1991). In addition the
proanthocyanidins family also comprises forms with ester bonds to gallic
acid (Hammerstone et al., 2000; Nuez et al., 2006). The proanthocyanidins
consisting exclusively of (epi)catechin are designated as procyanidins,
whereas the proanthocyanidins containing (epi)gallocatechin as subunits are
named prodelphinidins (Lazarus et al., 1999; Xie and Dixon, 2005; Gu et al.,
2006), this last form being less common in nature.

44

Chapter II
____________________________________________________________________

(a)

(b)

Figure 10. Representative structures of procyanidin dimers (B-type

procyanidins). (a) Procyanidin B2 (b) Procyanidin B5. Where: R1 = OH, R2 = H
(Source: Hammerstone et al., 2000).

The molecular weight of proanthocyanidins expressed as degree of

polymerization (DP) is one of the most important characteristics of these
compounds. Gu et al. (2002) defined proanthocyanidins with DP = 2-10 and
DP > 10 as oligomers and polymers, respectively.
Flavonols. These are the most ubiquitous flavonoids in foods and
the main representatives are quercetin and kaempherol. The richest sources
are onions, leek, brocoli, tea and blueberries (Manach et al., 2004). These
compounds posses a double bound between C-2 and C-3 and also possess a
hydroxyl group in the 3-position (Figure 11). Flavonols may vary in the
number and distribution of hydroxyl groups as well as in their degree of
alkylation or glycosylation. The formation of flavonol glycosides depends on
the action of light; therefore, they are found mainly in leaves and fruit skins
with only trace amounts in parts of plants below the soil surface (Hermann,
1976).

45

Bibliographic review
____________________________________________________________________

Flavonol

Kaempferol
Quercetin
Myricetin
Morin

3,5,7,4 = OH
3,5,7,3,4 = OH
3,5,7,3,4,5 = OH
3,5,7,2,4= OH

Figure 11. Structures of the main flavonol compounds (Source: Shahidi and
Nackz, 2004).

Flavonols are present mainly as mono-, di- and triglycosides. The

monoglycosides occur mainly as 3-O-glycosides. The 3-O-diglycosides and
3,7-di-O-glycosides are also found frequently. Rutin, which present a
rutinose molecule (6-O--rhamnosyl-D-glucopyranose) in its structure, is an
example of a diglycoside quercetin (Shahidi and Nackz, 2004). The sugar
moieties of glycosides are usually composed of D-glucose, D-galactose, Lrhamnose, L-arabinose and their combinations. A number of flavonol
glycosides acylated with phenolic acids such as p-coumaric, ferulic, caffeic,
p-hydroxybenzoic and gallic acids have been found in fruits (Harborne and
Williams, 2000).
Anthocyanins. After the flavanols and flavonols, the anthocyanins
are the next most abundant and widely distributed group of flavonoids. They
are water-soluble pigments responsable for the bright red, blue and violet
colors of fruits and other plant organs (Mazza and Miniati, 1993).
Anthocyanins are present in many fruits, vegetables, flowers, leaves, roots
and other storage organisms of plants. Six anthocyanidins are widespread
and commonly contribute to the pigmentation of fruits. Cyanidin is the most
common and, in terms of frequency of occurrence, is followed in decreasing
order by delphinidin, peonidin, pelargonidin, petunidin and malvidin (Figure
12).

46

Chapter II
____________________________________________________________________

Figure 12. Structures of the anthocynidins (Source: Shahidi and Nackz, 2004).

Chemically, anthocyanins are glycosylated derivatives of the

3,5,7,3 -tetrahydroxy-flavylium cation. The aglycon moeities are collectively
named anthocyanidins. These are highly reactive and they do not occur
naturally in the free aglycon form. A wide range of anthocyanins exist. They
differ in the nature and number of sugars attached to the aglycon, as well as
in the position of the attachment and in the nature and number of aliphatic or
cinamic acids attached to the sugar residues. The presence of several
hydroxyl groups on the anthocyanidins as well as one or several sugar
molecules, make these compounds quite soluble in water, ethanol and
methanol (Escribano-Bailn et al., 2004).
The most commonly found sugars are xylose, arabinose and
rhamnose among pentoses and galactose and glucose among hexoses. They
are linked to the aglycon by a  or  linkage of the free hydroxyl. Di- and
trisaccharide functional groups are also common, the most usual ones being
rutinose (6-O--rhamnosyl-D-glucopyranose) and sophorose (2-O--

47

Bibliographic review
____________________________________________________________________

glucopyranosyl-D-glucopyranose). Sugar moieties are always found attached

to the hydroxyl group at position 3 of the anthocyanidin. When additional
sugar groups exist, they occupy positions 5 and/or 7, and less frequently 3
and 5 . Sugars may be substituted by cinnamic acids (e.g. caffeic, pcoumaric, ferulic, U- hydroxybenzoic or sinapic acid) and/or aliphatic acids
(e.g. acetic, malic, malonic, oxalic or succinic acid) (Delgado-Vargas et al.,
2000). These acyl substituents are commonly bound to the C-3 sugar through
an ester bond with the 6-OH or less frequently with the 4-OH group of the
sugar (Clifford, 2000).

II.2.2.

Phenolic compounds as antioxidants

A polyphenol substance can be defined as antioxidant only if it

fulfils two conditions: firstly, when present at low concentration relative to
the substrate to be oxidized, it can delay or prevent its oxidation; secondly,
the resulting radicals formed after scavenging must be stable (Kaur and
Kappor, 2001).
Andersen and Markhan (2006) mentioned that there are
approximately 10000 known plant phenolics and model studies have
demonstrated that many have antioxidant capacity (Robards et al., 1999).
However, there is a wide degree of variation among different phenolic
compounds in their efectiveness as antioxidants. Furthermore, there are a
number of different mechanisms by which phenolics may act as
antioxidants: via free radical scavenging, singlet oxygen quenching, metalion chelation (Robak and Gryglewski, 1988; Hamilton et al., 1997; Harborne
and Williams, 2000), or inhibition of oxidative enzymes (Francis, 1989).
Additional mechanisms may be involved in vivo where phenolics may
protect the -tocopherol from oxidation by preferentially oxidise themselves;
alternatively, they may regenerate -tocopherol by donating an hydrogen
atom to the -tocopherol radical (Miura et al., 1994).

48

Chapter II
____________________________________________________________________

Cook and Samman (1996) stated that flavonoids could act at any of
the three stages of the radical-mediated oxidation. Flavonoids could block
initiation by scavenging primary radicals such as superoxide. Flavonoids
could also react with peroxy radicals to slow down propagation. In addition,
the flavonoid radical intermediates, formed after reacting with peroxy
radicals, can react with the other radicals formed during propagation,
accelerating the termination process. Robards et al. (1999) indicated that
phenolic antioxidants function primarily as terminators of the free radical
reactions, through their ability to interfere with the chain propagation
reactions by rapid donation of hydrogen or electrons to lipid radicals.
Rice-Evans et al. (1997) and Fukumoto and Mazza (2000)
mentioned that the reactivity of phenolic compounds in relation to the radical
species follows a mechanism of exchange of reducing equivalents. Besides,
the reactivity of phenolic compounds to face radicals depends on the
configuration as well as the position and number of hydroxyl groups in their
molecules. Thus, the radical scavenging capability of flavonoids has been
related to their structural groups: the o-dihydroxyl structure of the B ring
(catechol ring), the 2,3-double bond in conjunction with the 4-oxo function,
and the additional presence of both 3-and 5-hydroxyl groups (Figure 13)
(Bors et al., 1990; Rice-Evans et al., 1996; Rice-Evans et al., 1997; Lien et
al., 1999; Heim et al., 2002). For this reason flavanoids containing a
catechol moiety (3 - and 4 -OH) in B-ring or an AC-ring with three OH
groups (3-, 5-, and 7-OH) are potent scavengers.

49

Bibliographic review
____________________________________________________________________

Figure 13. Representation of three important structures for neutralizing free

radicals by flavonoids (Source: Adapted of Rice-Evans et al., 1996; Rice-Evans
et al., 1997).

Heim et al. (2002) found that multiple hydroxyl groups conferred

substantial antioxidant, chelating and, in some cases, pro-oxidant activity to
the molecule. Methoxy groups introduce unfavorable steric effects, but
presence of a double bond and a carbonyl functionality in the C ring
increases the activity by affording a more stable flavanoid radical through
conjugation and electron delocalization (Heim et al., 2002). The degree of
glycosylation with monosaccharides or disaccharides significantly affects the
antioxidant properties of the compound: for example, aglycones of quercetin
and myricetin were found more active than their glycosides (Shahidi et al.,
1992). In vitro assays have demonstrated that the increase of the
polymerization degree of flavan-3-ols enhances their effectiveness against a
variety of radical species (Heim et al., 2002; Counet and Collin, 2003).
Extensive conjugation between 3-OH and B-ring catechol groups, together
with the abundant  48 linkages, endow a polymer with significantly
higher radical scavenging properties by increasing the stability of its radical
(Castillo et al., 2000). The presence of hydroxyl groups at C-3 and C-3
positions is essential for flavonoids to show a strong superoxide-scavenging
activity (Shi and Noguchi, 2001).
The antioxidant capacity of phenolic acids and their esters depends
on the number of hydroxy groups in the molecule. Hydroxycinnamic acids

50

Chapter II
____________________________________________________________________

have been found to be more effective than their benzoic acid counterparts.
Hydroxycinnamic acids are known as primary antioxidants and act as free
radical acceptors and chain breakers (Robbins, 2003; Shahidi and Nackz,
2004).
Another way, flavonoids may prevent radical formation is by
chelation of transition metals. These metals are essential to certain
physiological functions, such as the transfer of oxygen in the human body,
the stabilization of the three-dimensional structure of proteins, or the
catalytic activity of several types of enzymes. They are however also
involved in the formation of free radicals. By consequence, metals, such as
iron and copper (Fe+2, Cu+) play an important role in the initiation and
propagation steps of lipid oxidation. On one hand, the presence of a
transition metal can accelerate the initiation step through the removal of an
hydrogen from an unsaturated lipid to form a lipid radical.

RH + M+n  Rx + H+ + M+(n-1)
On another hand, metals can also decompose hydroperoxides to form
alkoxyl and peroxyl radicals, accelerating the lipid oxidation (Lee et al.,
2004).

Fe3+(Cu2+) + ROOH  Fe2+ (Cu+) + ROOx + H+

Fe2+ (Cu+) + ROOH  Fe3+ (Cu2+) + ROx + OHMetals are also responsible of hydroxyl radical formation, according to the
Fentons reaction (Favier, 2003):

Fe2+(Cu+) + H2O2  Fe+3(Cu+2) + OHx + OH

They are also implicated in the singlet oxygen formation:

Fe2+(Cu+) + 2O2  Fe+3(Cu+2) + 1O2 + O2 x 

51

Bibliographic review
____________________________________________________________________

Many flavonoid structures have the chemical properties to chelate these

metals in such a way that radical generation is inhibited. Chelation of metal
ions renders them catalytically inactive. The formation of stable and inert
metal complexes is thus a potential mechanism of antioxidant action (Lee et
al., 2004). Flavanoids are known to chelate metal ions at the 3-hydroxy-4keto group and/or 5-hydroxy-4-keto group (when the A-ring is hydroxylated
at the fifth position). An o-quinol group at the B-ring also bears a metal
chelating activity (Figure 14) (Brown et al., 1998).

Figure 14. Flavanoids and their proposed sites for chelation of metal ions
(Men+) (Source: Brown et al., 1998).

II.2.3.

Biological properties of phenolic compounds

Phenolic compounds have been reported to have a wide diversity of

beneficial effects on human health. Thus, phenolic compounds have received
attention as protective and preventive molecules against chronic diseases,
such as atherosclerosis and cardiovascular diseases, anti-inflammatory
process, cancer, osteoporosis (Middleton et al., 2000; Arts and Hollmanm
2005; Sarni-Manchado and Cheynier, 2006).

52

Chapter II
____________________________________________________________________

It is very important to indicate that in vivo properties of polyphenols

depend on their bioavailability (Scalbert and Williamson, 2000). Human and
animal studies on the absorption, distribution, metabolism and elimination of
dietary phenolics are key steps to know the conditions and quantity of
phenolic compounds that are effectively used in the body. These results are
relevant in order to further explain the systemic positive effects as well as
the nutritional quality that are attributed to polyphenols (Martin and
Andriantsitohaina, 2002), and also to correlate their consumption with health
beneficial properties.
Bioavailability varies considerably from one
polyphenol to the other (Yang et al., 2001).
In most countries, a high intake of saturated fats is strongly
correlated with high mortality from coronary heart disease (CHD). But
exceptions occur. The best example is the so-called French paradox, where
populations of the South of France, in spite of a high average body mass
index and a high fat intake are less exposed to cardiovascular diseases. This
behavior has been attributed to the regular intake of red wine in the diet
(Halliwell, 2000). Frankel et al. (1993b) have concluded that epicatechin and
quercetin are the most important wine constituents in reducing CHD. They
also supported a previous suggestion that it is the specific combination of
antioxidant phenolics in wine that protects against atherogenesis. Frankel et
al. (1993b) suggested that the beneficial effects of red wine may be
explained by the inhibition of oxidation of LDL by wine phenolics. The
oxidative damage to human LDL (particularly to the apolipoprotein B) is
considered to be an important step in the development of atherosclerosis. It
is a prerequisite for macrophage uptake and cellular accumulation of
cholesterol leading to the formation of the atheromal fatty streaks
(Esterbauer et al., 1992; Arts and Hollman, 2005).
A wide range of phenolic compounds have been reported to protect
human LDL against oxidative damage (Arts and Hollman, 2005; Kuriyama
et al., 2006). Satu-Gracia et al. (1999) reported that phenolics present in
Spanish sparkling wines (cavas) inhibit oxidation of LDL. This activity

53

Bibliographic review
____________________________________________________________________

correlates positively with the total phenolic content, as well as the content of
quercetin 3-glucuronide, trans-caffeic acid, protocatechuic acid, and
coumaric acid. Phenolic acids present in the human diet, such as caffeic and
chlorogenic acids, ellagic acid and protocatechuic acid, all protect isolated
LDL against oxidative damage (Laranjinha et al., 1994; Robbins, 2003; Wen
et al., 2005). The protective effects of anthocyanins against LDL oxidation
have been demonstrated in different studies (Ghiselli et al., 1998; Khknen
and Heinonen, 2003; Chang et al., 2006). Also, grape seed extracts rich in
procyanidins have been shown to have antiatherosclerotic activities
(Shrikhande, 2000; Corder et al., 2006). The protection of phenolic
compounds against LDL oxidation could be related to the capacity of
phenolic compounds to scavenge free radicals (Zhu et al, 2002), to chelate
metals and/or to interact with lipid-rich structures (Teissdre et al, 1996;
Liao and Ying, 2000; Dai et al., 2006). The partition coefficient of phenolic
compounds determines their interaction with these structures and influences
their antioxidant capacity performance at this level (Liao and Yin, 2000). As
an example, it is well known that the flavonoids and related polymers have
different types of interactions with lipid-rich structures. The more
hydrophobic flavonoids can partition in the hydrophobic core of these
structures, leading to a direct inhibition of lipid oxidation. The more
hydrophilic flavonoids interact by hydrogen bonding with the polar head
groups at the lipid-water interface of the lipid-rich structures. This type of
interaction may provide a direct or indirect protection from external and
internal aggressors (Oteiza et al., 2005).
In addition to their cardiovascular protective effect, phenolic
compounds appear to bear antitumor activities (Middleton et al., 2000; Yang
et al., 2001; Carnesechi et al., 2002; Ferguson et al., 2004). The
epigallocatechin-3-gallate, a polyphenolic component of green tea, has been
found to reduce the incidence of spontaneous and chemically-induced
tumors in experimental animals, as observed for tumors of liver, stomach,
skin, lung and esophagus (Huang et al., 1992). The ellagic, protocatechuic
and chlorogenic acids present in fruits and vegetables have been found to

54

Chapter II
____________________________________________________________________

serve as chemopreventive agents against several carcinogens (Nakamura et

al., 2001).
Phenolic compounds have also been studied for their antiinflammatory activities (Middleton et al., 2000; Crouvezier et al., 2007).
Cyclooxygenase and lipoxygenase play an important role as inflammatory
mediators. They catalyse the first steps of the production of proinflammatory eicosanoids from arachidonic acid (Nijveldt et al., 2001).
Selected phenolic compounds were shown to inhibit both cyclooxygenase
and 5-lipoxygenase pathways, through an inhibiton of the release of
arachidonic acid (Kim et al., 1998; Havsteen, 2002).
Isoflavones exert a broad spectrum of biological activities. Some
isoflavone derivatives have a chemical structure resembling that of
estrogens. They are thus able to bind to estrogen receptors. By consequence,
they bear hormonal agonistic or antagonistic properties depending on the
organ and on the age of the consumer. Other flavonoids such as flavones,
flavonols and flavanones have also shown affinity for the estrogen receptor
(Harborne and Williams, 2000). Besides estrogenic activities, isoflavones
protect against several chronic diseases. Results of epidemiological studies
indicate that consumption of soybean isoflavones lowers the incidence of
breast, prostate, urinary tract and colon cancers. These compounds also
provide protection against coronary heart diseases and osteoporosis (Moyad,
1999; Su et al., 2000).

II.2.4.
Methods
compounds

for

quantifying

phenolic

Numerous methods for quantifying phenolic compounds exist in the

literature, some being more used than others. The most common techniques
are based on spectrophotometric and chromatographic measurements.

55

Bibliographic review
____________________________________________________________________

II.2.4.1.

Spectrophotometric assays

Folin-Ciocalteu assay. This assay is commomly known as the total

phenolic assay (Singleton and Rossi, 1965). The basic mechanism is an
oxidation/reduction reaction and, as such, can be considered another method
to evaluate the antioxidant capacity (Prior et al., 2005). Reduction of
phosphomolybdic-phosphotungstic acid complexes (Folin-Ciocalteus
reagent, chromogen) to a blue-colored complex in an alkaline solution (pH ~
10) occurs in the presence of phenolic compounds. The color development is
due to the transfer of electrons to reduce the chromogens. Huang et al.
(2005) mentioned that, in essence, the molybdenum (Mo) is more easily
reduced in the complex and the electron-transfer reaction occurs between
reducing compound and Mo (VI) (Mo (VI) + e  Mo (V)). The reaction occurs
progressively and the absorbance increases during 30 minutes at
wavelengths between 725 and 765 nm.
The Folin-Ciocalteu reagent is not specific and detects all phenolic
groups found in extracts, including those found in the extractable proteins.
Another disadvantage of this assay is the interference of reducing substances
such as ascorbic acid and sugars. In general, the content in phenolics is
expressed as gallic acid equivalents, but others acids (e.g. chlorogenic acid)
or even flavonoids (e.g. catechin) can be used, depending on the major
phenolic compounds present in the sample.
DMACA (4-(dimethylamino)-cinnamaldehyde) assay. This
DMACA assay is used for the quantification of flavanoids that contain metaoriented di- or tri-hydroxy substituted benzene rings, with a single bond at
the 2,3 position of the C-ring, such as flavan-3-ols, flavan-4-ols, flavan-3,4diols, flavanones and derivatives. Delcour and Janssens de Varebeke (1985)
and Treutter (1989) demonstrated that DMACA does not react with a wide
range of flavonoids including dihydrochalcones, flavones and flavonols, and
phenolic acids.

56

Chapter II
____________________________________________________________________

In the DMACA assay, a condensation reaction occurs between the

flavanoid molecule and the DMACA in acidic medium, producing the
formation of a green chromophore. The DMACA colorimetric assay consists
in measuring the absorbance of the reaction medium at 635-640 nm, after
aproximately 15 min at room temperature. An appropriate blank must be
included. DMACA reacts only with the terminal groups of condensed
tannins (Figure 15). The results are usually expressed as mg of catechin
equivalents.
H3C

CH3

CH3
H3C

OH

OH

OH

CH3

OH

HO

O
HO

H2O

OH
OH

OH
OH

Figure 15. Reaction between flavan-3-ols and DMACA in an acidic medium

(Source: Mcmurrough and Baert, 1994).

Anthocyanins assay. Anthocyanins are colored pigments and can be

quantified by means of colorimetric measurements. The pH differential
method (Giusti and Worlstad, 2001) is the most widely used assay to
quantify monomeric anthocyanins. Anthocyanin pigments undergo
reversible structural transformation with a change in pH. These changes are
associated to strikingly different absorbance spectra. The colored flavylium
cation form predominates at pH 1.0 while the colorless hemiketal form is the
major form at pH 4.5 (Figure 16). For each pH value, the measurement is
performed at two wavelengths. To determine the total anthocyanin content,
the absorbance at pH 1.0 and 4.5 is measured at the wavelength of visiblemaximum and at 700 nm, which allows for haze correction (due to the
possible presence of sediments). When the predominant anthocyanin in the

57

Bibliographic review
____________________________________________________________________

Figure 16. Structural transformation of anthocyanins in function of pH (Source:

Malien-Aubert et al., 2001).

sample is not known, the wavelength maximum to choose is 520 nm. This
corresponds to the wavelength of cyanidin-3-glucoside, which is the most
widely found anthocyanin in colored vegetables. The difference in

58

Chapter II
____________________________________________________________________

absorbance is then calculated by subtracting the reading at pH 4.5 from that

at pH 1.0 ([Avis-max A700]pH 1.0 [A vis-max A700]pH 4.5), The result is divided
by the average extinction coefficients for the four major anthocyanins or by
the extinction coefficient of the principal anthocyanin of the sample in order
to yield the total anthocyanin content.

II.2.4.2.

Chromatographic assays

Various chromatographic techniques have been employed for the

separation, preparative isolation, purification, identification and
quantification of phenolic compounds such as the high performance liquid
chromatography
(HPLC)
technique,
high-speed
countercurrent
chromatography, thin layer chromatography (Merken and Beecher, 2000;
Shahidi and Nackz, 2004).
HPLC techniques are now the most widely used for the separation
and quantitation of phenolic compounds. The chromatographic conditions of
the HPLC methods include generally a reversed-phase C18 column, a UVVis diode array detector, and a binary solvent system containing acidified
water and a polar organic solvent (Tsao and Yang, 2003). Phenolic
compounds are identified and quantified by comparing their retention times
and UV-visible spectral data to known previously injected standards. The
advantages of this technique are 1) the possibility to quantify individual
phenolic compounds belonging to different phenolic families thanks to the
reverse phase conditions; 2) its versatility, high sensitivity, exactitude and
precision (Robards, 2003; Parejo et al., 2004). The main disavantage is its
high cost.
Prior to phenolic analysis by HPLC, several steps are usually
employed to favour a good performance during the run. These steps
comprise liquid-liquid extraction, solid-phase extraction (SPE) or the
fractionation of phenolics from crude extracts (Rodriguez-Saona and
Wrolstad, 2001; Robards, 2003; Prior et al., 2004). Several works have been

59

Bibliographic review
____________________________________________________________________

published on the application of the HPLC methodology for the analysis of

phenolics (Robards and Antolovitch, 1997; Merken and Beecher, 2000;
Mattila and Kumpulainen, 2002; Tsao and Yang, 2003; Wu and Prior, 2005).
On the other hand, normal phase conditions using a fluorometric detector
have been successfully applied for proanthocyanidin analysis (Gu et al.,
2002). Mass spectrometry (MS) detectors coupled to HPLC (HPLCMS)
have been widely employed for the structural characterization of phenolics
(Shahidi and Nackz, 2004). More information about this technique is given
in section II.3.4.2.

II.3. Extraction,
purification
and
processes for phenolic compounds

identification

Prior to the identification of phenolic compounds, several processes

including the extraction of plant materials and a purification step are needed.
Today, many methodologies, equipments and solvents are available to
achieve an adequate extraction and purification of phenolics. The adequate
choice and successfully sequence of operations can warranty an efficient
recovery of phenolics from different materials to be analyzed.
The following sections describe the extraction and purification
processes that have been used in the present thesis for analytical purposes to
identify phenolic compounds.

II.3.1.

Extraction processes

Extraction is the first step in the isolation and later identification of

phenolic compounds from botanicals. Due to the chemical nature of phenolic
compounds, no satisfactory solvent extraction system is suitable for the
isolation of all classes of phenolics. By contrast, different techniques of

60

Chapter II
____________________________________________________________________

phenolic extraction are used in different studies. Escribano-Bailn and

Santos-Buelga (2003) listed the three major techniques that may be used in
phenolics extraction are: a) extraction using solvents, b) solid-phase
extraction (SPE) and c) supercritical extraction. The extraction using
solvents can be divided in solid-liquid extraction and liquid-liquid
extraction. In the present section, more information about solid-liquid
extraction will be given because solid matrices are more frequently used as
source of phenolics than liquid matrices.
Gertenbach (2002) defined a solid-liquid extraction as the use of a
solvent to dissolve and remove a soluble fraction (called the solute, such as a
phenolic compound) from an insoluble, permeable solid matrix (plant
tissue). In this process the ground plant material is added to the solvent in
varying solvent to solid ratios. The mixture is then constantly stirred until an
uniform slurry is obtained. If a large contact area is obtained between the
two phases, the extraction occurs rapidly. After an appropriate amount of
time, the solvent is centrifugated or filtered out of the slurry and the final
extract is obtained (Shi et al., 2005).
The main factors that contribute to the efficiency of solvent
extraction are: type of solvent, pH, temperature and time of extraction,
number of extraction steps, solvent/material ratio, size of particle.
Type of solvent. The choice of solvent can significantly change both
the type and amount of phenolics that are extracted into the liquid, as well as
the rate at which the phenolics are extracted. The solubility of phenolic
compounds is governed by the type of solvent (polarity) used. In phenolic
extractions, highly polar solvents, such as water, or highly apolar solvents,
such as chloroform or hexane, do not usually give good recoveries (Liu et
al., 2000). Water solvents produce extracts with a lot of impurities (soluble
compounds with no phenolic nature), which make the later purification
process more difficult. When the extraction of many components from a
solid is required, mixtures of solvents are used, resulting in a solution with

61

Bibliographic review
____________________________________________________________________

medium polarity, allowing satisfactory constituent yields for compounds

with distinct polarities (Liu et al., 2000).
The most widely used solvent for extracting phenolic substances for
analytical purposes are methanol and methanol/water (Escribano-Bailn and
Santos-Buelga, 2003). Other solvents such as acetone, ethanol, ethyl acetate
and, to a lesser extent, propanol, and their combinations have also been
utilized for the extraction of phenolics (Nackz et al., 1992; Antolovich et al.,
2000; Ju and Howard, 2003).
Glycosylated phenolic compounds (more water soluble) are
generally extracted using combinations of water with methanol, ethanol or
acetone (Rice-Evans et al., 1997). In contrast, less polar phenolics
(aglycones) as isoflavones, flavanones and highly methoxylated flavones and
flavonols tend to be more soluble in rather hydrophobic solvents (Bradshaw
et al., 2001). Acetone and methanol are employed for extraction of flavan-3ol compounds but with distinct specificities in the extraction of these
substances. Thus, methanol is the best solvent for catechin extraction
(flavan-3-ol monomer), whereas acetone presents a better yield for
procyanidins (Escribano-Bailn and Santos-Buelga, 2003).
pH of the extraction medium. The addition of acid in the extraction
solvent determines the stability of phenolic compounds such as anthocyanins
(Rodriguez-Saona and Wrolstad, 2001), influences the possible
solubilization of the hydrolysable polymers, such as lignins,
hydroxycinnamic acids and procyanidins (Nackz and Shahidi, 2004; Tsao
and Deng, 2004) and finally may improve the desintegration of cell walls,
facilitating the solubilization of phenolic compounds.
Anthocyanins are usually extracted with an acidified organic solvent,
most commonly methanol and acetone (Rodriguez-Saona and Wrolstad,
2001). This solvent system destroys the cell membranes and simultaneously
dissolves the anthocyanins. The acid also provides favorable conditions for

62

Chapter II
____________________________________________________________________

the formation of the flavylium chloride salt, thus stabilizing the

anthocyanins. The most commonly used acids for anthocyanin extraction
are: HCl (stronger acid), tartaric and citric acids (weaker acids), or
trifluoroacetic acid (volatile stronger acid). Escribano-Bailn and SantosBuelga (2003) indicated that the use of methanol containing 0.1% HCl does
not cause a significant degradation of the most usual monoacylated
anthocyanins.
Tannins from sorghum and dry beans have been extracted using 1%
HCl in methanol (Shahidi and Nackz, 2004).
Temperature of extraction. Cacace and Mazza (2003) and LiyanaPathiirana and Shahidi (2005) mentioned that the polyphenol yield is
increased with increasing temperature. High temperatures improve the
efficiency of the extraction since heat renders the cell walls permeable,
increases the solubility and diffusion coefficients of the compounds to be
extracted and decreases the viscosity of the solvent, thus facilitating its
passage through the solid mass (Escribano-Bailn and Santos-Buelga, 2003).
Temperature is limited by the boiling point of the solvent. In
addition, many phytochemicals are quite unstable, and the extraction
temperature needs to be controlled to prevent thermal decomposition
(Gertenbach, 2002). In general, temperatures ranging from 50 to 60C are
recommended to extract polyphenols with good extraction recoveries (Pinelo
et al., 2004; Silva et al., 2007).
Liquid-solid ratio. The recovery of polyphenols from vegetable
products is also influenced by the ratio of sample to solvent (Shahidi and
Nackz, 2004; Silva et al., 2007). A higher liquid-to-solid ratio lowers the
concentration of dissolved phytochemicals at the surface of the solid particle,
thus providing a higher concentration gradient between the concentrations
inside and at the surface of the particles. This higher concentration gradient
gives a higher extraction rate. However, a more dilute extract requires a
more extensive downstream processing (Gertenbach, 2002). Liquid-solid

63

Bibliographic review
____________________________________________________________________

ratios varying from 1/10 to 1/60 are frequently reported in polyphenolic

extractions (Gertenbach, 2002; Silva et al., 2007).
Time of extraction. The extraction period is another factor that affects
the recovery of the polyphenolics. Extraction periods varying from 1 min to
24 h have been reported (Shahidi and Nackz, 2004). Longer extraction times
increase the chance of oxidation of phenolics unless reducing agents are
added to the solvent system (Naczk and Shahidi, 2004).
Particle size of the sample. The rate of extraction increases with
decreasing the particle size (Gertenbach, 2002). The rate-controlling step for
extraction is the migration of the solute through the pores of the particles to
the particle surface. A smaller particle will lead a shorter path for the solute
to reach the surface. A shorter diffusion path equates to a faster extraction
rate. Milling can enhance the extraction kinetics (Gertenbach, 2002). High
phenolic recoveries have been achieved with particle sizes that ranged from
0.6 to 0.8 mm (Pinelo et al., 2004; Silva et al., 2007).

II.3.2.

Purification processes

Extracts rich in phenolic compounds often require purification steps

to eliminate substances that would otherwise interfere with the identification
analysis of phenolics. The purification process leads to the removal of
interferences but also to the concentration of the analyte(s) of interest and to
the improvement of the analytical performances.
A first step in the purification process consists usually in the removal
of lipids from vaccum concentrated phenolic extracts through the use of
petroleum ether, hexane or chloroform. Next to this preliminary step,
purification methods are applied. The adoption of one or more of these
methods will depend on the nature of the phenolics to purify/isolate. These
methods can be divided into in solid-liquid and liquid-liquid phase
procedures (Tsao and Deng, 2004).

64

Chapter II
____________________________________________________________________

II.3.2.1.

Solid-liquid phase procedures

The solid-liquid phase procedures consist in the migration of a liquid

extract through a solid matrix. The solid matrix retains the compounds of
interest and the undesirable compounds are eliminated. The compounds of
interest are then recovered by means of an elution. The retention of desirable
compounds in the matrix depends on their affinity for the matrix. This
affinity is governed by specific characteristics such as polarity, ionic forces,
molecular weight. Besides, the recovery of the compounds of interest
depends on their affinity for the elution solvent. With regard to this
characteristic, a selective recovery can be performed. Due to this, in solidliquid procedures many systems that comprise matrix (adsorbent) and
solvent (eluent) differences have been proposed for the purification of
phenolic compounds.
One simple method of phenolic purification using SPE consists in
the passage of an aqueous solution rich in phenolics through a C18 cartridge
(matrix of cartridge is composed of C18 chains bound to silica), previously
conditioned with methanol and acidified water (pH ~ 2). The loaded
cartridge is washed with acidified water to eliminate interfering substances
and the retained phenolic compounds are eluted with a strong solvent such as
methanol (Vinha et al., 2005).
Rodrguez-Saona and Worlstad (2001) have proposed a purification
method to purify anthocyanins using SPE. The authors mentioned that this
method allows the removal of several interfering compounds present in the
crude extracts. After having been preconditioned with low acidic conditions,
these columns retain phenolic compounds (e.g., anthocyanins, flanovols,
flavan-3-ols), while allowing matrix interferences, such as sugars and acids,
to pass through and be discarded as waste. Then washing the retained
pigments with ethyl acetate will further remove phenolic compounds other
than anthocyanins, these last ones are recovered by means of a fast elution
with acidified methanol.

65

Bibliographic review
____________________________________________________________________

Column chromatography has also been employed for the

fractionation of phenolic extracts. Column chromatography using Sephadex
LH-20 gel has been used in various studies to separate the fraction of
proanthocyanins from others phenolics in different plant materials and
beverages (Kantz and Singleton, 1990; Prior et al., 2004; Sanchez-Moreno et
al., 2003). Sephadex LH-20 gel is made from hydroxypropylated dextran
beads that have been cross-linked to yield a polysaccharide network (Figure
17). The most usual solvents used in this system are ethanol, methanol and
acetone and their mixtures with water. The separation is based on the
establishment of hydrogen bonds between phenolic hydrogens or carboxylic
groups and H-bond acceptors in the gel. The strength of the adsorption
depends on the number of phenolic hydrogens per molecule, polymeric
polyphenols, such as condensed tannins, are adsorbed more readily than
monomers, such as cathechins. Acetone is a better desorbent, since carbonyl
oxygen acts as a strong acceptor regarding hydrogen bonding and it is
capable of displacing polymeric polyphenols (Kantz and Singleton, 1990).

Figure 17. Partial structure of Sephadex LH-20 (Amersham Bioscience, 2002).

Sanchez-Moreno et al. (2003) used SPE with Sephadex LH-20 to

remove phenolic acids, flavonoids and anthocyanins of red wines to obtain

66

Chapter II
____________________________________________________________________

procyanidins. The authors utilized 20-mL columns containing 5 g of

Sephadex LH-20. Wine samples were passed through the column previously
equilibrated with methanol. The column was then eluted with different
elution media: 25 mL of 20% (v/v) methanol/water to remove phenolic
acids, 40 mL of 60% (v/v) methanol/water to elute the flavonols and
anthocyanins and finally 90 mL of pure methanol for the elution of the
proanthocyanidins. A similar procedure was used by Prior et al., (2004) to
obtain anthocyanins and procyanidins from blueberries and cranberries.
Kantz and Singleton (1990) isolated non-polymeric and polymeric phenols
from grape extracts using a Sephadex LH-20 column (150 x 50 mm i.d.).
Non-polymeric compounds were eluted from the gel with 60% methanol and
polymeric phenols were recovered with 60% acetone. Goffman and
Bergman (2004) obtained low-molecular-weight phenolics and tannins of
rice kernels using small columns of Sephadex LH-20. Absolute ethanol was
first used to elute the low-molecular-weight phenolics fraction and then 70%
acetone was used to elute the high-molecular-weight phenolics or tannins of
rice kernels.

II.3.2.2.

Liquid-liquid phase procedures

Liquid-liquid procedures are used for the extraction of phenolics

from liquid samples (Krasteva et al., 2004), but also as a technique of
purification and fractionation of groups of phenolics from a sample. The
separation of compounds is based on their partition between two immiscible
liquids (organic and aqueous phase). The efficiency of recovery depends on
the solute affinity for extraction solvent, the ratio between the two
immiscible liquids and the number of extraction steps. This method is very
easy to apply and can be used with various solvents with different
solubilities and selectivities. But some disadvantages can be listed: 1) the
compounds with high affinity for water are partially extracted by the organic
solvent, resulting in a loss of analytes, 2) emulsions are formed, 3) large
volumes of aqueous phase are often to be manipulated, 4) the use of an
organic phase for the extraction requires its elimination during the

67

Bibliographic review
____________________________________________________________________

concentration process (Queiroz et al., 2001). Next, some liquid-liquid

procedures will be described.
Liquid-liquid extraction was applied by Balde (1997) to obtain
fractions of phenolics from red wine. Anthocyanins were mainly present in
the aqueous residual phase, whereas other phenolic compounds, such as
procyanidins, were present in the ethyl acetate phase. Zhang et al. (2000)
used liquid-liquid fractionation to identify the non-anthocyanin phenolics
present in litchi (Litchi chinensis Sonn.) peel. Aqueous extracts of litchi peel
were extracted three times with ethyl acetate (1:1, v/v). The phenolics
identified in the ethyl acetate fraction were mainly gallic acid, procyanidin
B2, B1 and B4, epicatechin, epigallocatechin, catechin and gallocatechin.
Kennedy (2002) indicated that applying a liquid-liquid partition with ethyl
acetate in proanthocyanidin-rich aqueous extracts from various products
allows the removal of monomeric flavan-3-ols, while oligomeric and
polymeric proanthocyanidins remained in the aqueous phase. Pedreschi and
Cisneros-Zevallos (2006) utilized a liquid-liquid purification to fractionate
phenolic compounds from Andean purple corn. In that work, the aqueous
phase, which contained phenolics of purple corn, was combined with ethyl
acetate in a separation funnel. The ethyl acetate phase was composed of
phenolic acids such as p-coumaric acid, vanillic acid, protocatechuic acid,
flavonoids, such as quercetin derivatives and a hesperitin derivative, while
the aqueous phase revealed the presence of mainly anthocyanins such as
cyaniding 3-glucoside, pelargonidin 3-glucoside and peonidin 3-glucoside.

II.3.3.
Separation and hydrolysis processes prior to
identification of phenolic compounds
II.3.3.1.

Separation processes

HPLC is perhaps the most popular and reliable system among all
chromatographic separation techniques for the separation of antioxidant
phytochemicals. Since quite a long time, HPLC is the method of reference

68

Chapter II
____________________________________________________________________

used for the analysis of the phenolic compounds. Several hundred papers
have been published on the analysis of phenolic compounds by HPLC. This
approach permits to detect one, two, or sometimes three subclasses in one
run. Thus, for example Merken and Beecher (2000) developed an HPLC
system for the separation and quantification of 17 flavonoids, which
represent all 5 subclasses and are expected to be prominent in commonly
consumed foods. Later, Sakakibara et al. (2003) proposed an HPLC method
to determine simultaneously different polyphenolics in vegetables, fruits,
and teas. Also, Tsao and Yang (2003) proposed an HPLC method for the
separation, identification and quantification of five major polyphenolic
groups found in fruits and related products, namely single ring phenolic
acids (hydroxybenzoic acid and hydroxycinnamic acid derivatives), flavan3-ols, flavonols, anthocyanins and dihydrochalcones. Additionally, these
authors reported on the identification of some proanthocyanidins.
The chromatographic conditions of the HPLC methods include, in
most of the cases, the use of a reverse phase (RP) C18 column ranging from
100 to 300 mm in length and from 3.0 to 4.6 mm in inside diameter, and a
binary solvent system containing an aqueous acidified solvent (solvent A),
such as aqueous acetic acid, perchloric acid, or formic acid, and an polar
organic solvent (Solvent B), such as methanol or acetonitrile, sometimes
acidified. Trifluoroacetic acid in both solvents enhances the resolution and
eliminates peak tailing of catechins (Sivan, 2002; Tsao and Yang, 2003).
The separation normally requires 1 h at a flow rate of 1.0 - 1.5 ml/min. In
general, the HPLC runs generally requiere 1 h maximum, with equilibration
between the runs. Columns are usually maintained at ambient temperature
and the injection volume generally ranges from 1 to 100 l (Tsao and Yang,
2003).

II.3.3.2.

Hydrolysis of phenolic compounds

Many of the phenolic compounds are often associated with sugar or

acyl moieties, which further diversifies the phenolic profiles of fruits,

69

Bibliographic review
____________________________________________________________________

vegetables and beverages and makes their identification more difficult.

Many researchers have thus proposed to apply hydrolysis techniques to the
phenolic extracts to facilitate their identification. The most common
techniques are acidic and basic hydrolysis. Basic hydrolysis releases the acyl
groups while the purpose of acidic hydrolysis is to release the sugar
molecules from the phenolic compounds (Sivan, 2002).
Anthocyanins are O-glycosides and acylglycosides of
anthocyanidins. Thus, anthocyanins are often hydrolyzed to anthocyanidins
by means of an acidic hydrolysis (Hong and Wrolstad, 1990; Durst and
Wrolstad, 2001). This reaction is typically carried out in MeOH2N HCl
(1:1, v/v) (Gao and Mazza, 1995) or in HCl solutions, such as 2N HCl (Durst
and Wrolstad, 2001), and conducted at high temperatures (100C) and
darkness. Alkaline hydrolysis has been used for acylated anthocyanins using
10% aqueous KOH under darkness for 8 min at room temperature. The
hydrolyzed mixture is then neutralized with 2N HCl. Similar conditions have
been used for flavonoids. Hertog et al. (1992) have proposed a protocol for
the acid hydrolysis of the main flavonoids in vegetables and fruits. In this
protocol, the sample is hydrolyzed using a 50% methanol acidified solution
(1.2M HCl) for 2 h at 90C. Besides, Llorach et al. (2003) used a basic
hydrolysis to release the acyl groups from highly acylated flavonoids from
cauliflower. The hydrolysis was performed using 4N NaOH under darkness
at room temperature for 16 h. In Figure 18, the application of the acidic (b)
and basic (c) hydrolysis to identify the flavonoid structures in a complex
flavonoid mixture are shown.
Hydrolysis of the ester of a carboxylic acid has been one strategy
used to simplify the analysis and gain a more specific picture of the phenolic
acid profiles in foods. The main procedures to cleave the ester bond reported
in the literature are basic hydrolysis (Robbins, 2003). The basic hydrolysis
involves the treatment of the sample with a solution of NaOH at
concentrations ranging from 1 to 4M. The time of reaction varies between 15

70

Chapter II
____________________________________________________________________

h and overnight, under darkness and involves the use of an inert atmosphere
(e.g. argon or nitrogen).

Figure 18. HPLC chromatograms of a flavonoid mixture before and after

hydrolysis. Detection was performed at 352 nm. (a) A flavonoid mixture
showing different groups of compounds: a kaempferol triglycoside (1), a set of
apigenin glycosides (2), a kaempferol diglycoside (3), a luteolin glycoside (4), a
set of acylated kaempferol glycosides (5), a chalcone (6), and luteolin (7). (b)
The alkaline hydrolysis products of the same mixture showing a large relative
increase in peak 3 and the loss of the acylated kaempferol glycoside peaks. (c)
The acid-hydrolyzed mixture showing luteolin and kaempferol (8) (Source:
Sivan, 2002).

71

Bibliographic review
____________________________________________________________________

Butanol-HCl is a solvent mixture used for the hydrolysis of

proanthocyanidins (condensed tannins). This hydrolysis is carried out in a
solution of butanol-concentrated HCl (95:5, v/v). In the presence of this hot
HCl solution, proanthocyanidins are converted to anthocyanidins (coloured
form), which are products of the autoxidation of the carbocations formed by
the cleavage of interflavanoid bonds (Porter et al., 1986). The hydrolysis is
carried out during 2 h in a boiling water bath, before cooling the mixture in
an ice-cold water bath. The anthocyanidins formed are then detected at 550
nm (Mole and Waterman, 1987).

II.3.4.

Identification processes
II.3.4.1.

HPLC

Food phenolics are commonly detected in HPLC using UVvis,

diode array detector (DAD), and UV-fluorescence detectors (Nackz and
Shahidi, 2004). Among these detections modes, UV-vis with DAD is the
most widely used. It is indeed seen as a convenient multiple-wavelength
detector used both to identify and quantify phenolic compounds on the basis
of their 1) UV-visible spectra and 2) retention time compared with known
phenolic standards previously characterized by HPLC. Retention time is
related to the hydrosolubility of the phenolic compounds. Under RP
conditions, more hydrophilic phenolics present a rapid elution (shorter
retention times) whereas the lately eluted phenolics have more lipophilic
characteristics (longer retention times).
Flavonoids present two characteristic UV absorption bands. Band II,
with a maximum in the 240 to 285 nm range, is believed to arise from the A
ring. Band I, with a maximum in the 300 to 550 nm range, presumably arises
from the B ring. In general, UV spectra of flavanones, isoflavones, and
catechins have an intense band II peak and a weak band I peak because of
little or no conjugation between A and B rings (Sivan, 2002). Phenolic acids

72

Chapter II
____________________________________________________________________

present main UV absorbance in the 217 to 326 nm range (Robbins, 2003). In

function to the maximum UV-Vis absorption, phenolics are identified and
quantified using four different wavelengths: 280 nm for the hydroxybenzoic
acid derivatives, flavan-3-ols (including their dimers), and dihydrochalcones,
320 nm for the hydroxycinnamic acid derivatives, 360 nm for the flavonols
and 520 nm for the anthocyanins (Tsao and Yang, 2003). Figure 19 shows
the representative UV-vis spectra of the major classes of flavonoid
aglycones and phenolic acids and Table 4 gives the maximum UV
absorption wavelengths.

Figure 19. UV-Vis absorbance patterns of the anthocyanidin delphinidin, the

flavan-3-ol epicatechin, the flavanone hesperitin, the flavone luteolin, the
flavonol quercetin, and the isoflavone genistein (Source: Sivan, 2002).

73

Bibliographic review
____________________________________________________________________

Tsao and Deng (2004) mentioned that although RP-HPLC has been
the primary separation means for procyanidins, more recent studies have
illustrated the difficulty in determining the degree of polymerization of these
antioxidants. Hammerstone et al. (1999) have developed a normal phase
(NP) HPLC method that utilized a series of linear gradients of methanol into
dichloromethane with constant amounts of acetic acid and water. This
method has been used in many studies for the analysis of proantocyanidins
(Lazarus et al., 1999; Gu et al., 2003; Gu et al., 2006).
Table 4. UV-vis absorption patterns of various phenolic compounds

Class of Compounds

Benzoic acids
Cinnamic acids
Anthocyanins
Flavonols
Flavanols
Flavones
Flavanones,
Flavanonols
Isoflavones
Chalcones

UV absorptiona
Band II
Omax
270-280
290-300
240-280
250-270
270-280
250-270
270-295

UV absorptiona
Band II
Omax

Visiblea
Omax

(315-325)*
(300) * 350-380

500-550

245-270
220-270

300-340
(300-320)*
340-390

330-350
(300-330)*

a
Methanol has been used as solvent with exception for anthocyanins where acidified
methanol was used. * Only when acylation is present. Source: Robards et al. (1999) and
Shahidi and Nackz (2004).

II.3.4.2.

HPLC- Mass Spectrometry

Over the past two decades, MS has proved to be one of the most
effective techniques in biomedical, pharmaceutical and food industry
research, among others, particularly for the analysis of complex mixtures in
samples. Its high sensitivity, specificity, and easy combination with
chromatographic techniques have placed it as the method of choice of many
analysts. The use of HPLC with mass spectrometry (HPLC-MS) detection

74

Chapter II
____________________________________________________________________

provides useful structural information and allows for tentative compound

identification when standard reference compounds are unavailable and when
peaks have similar retention time and similar UV-absorption spectra
(Hakkinen et al., 1999; Maatta-Riihinen et al., 2004). One of the most
common mass spectrometric methods used for the identification of phenolic
compounds is electrospray ionization (ESI)-MS.
ESI is a method of generating highly charged droplets from which
ions are ejected by an ion evaporation process. An electric field is generated
at the tip of a sprayer by applying a high voltage, with a close proximity of a
counter electrode. Ions of one polarity are preferentially drawn into the drops
by the electric field as they are separated from the bulk liquid (see Figure
20). This technique is typically performed either in the infusion mode or in
combination with HPLC or capillary electrophoresis. In the infusion mode,
the sample is introduced into a continuous liquid stream via an injection
valve. ESI interfaces (also referred to as ion-spray interfaces for certain
commercial variations) are most often used with quadrupole mass
spectrometers. Quadrupole instruments have limited mass-to-charge (m/z)
ranges (typically up to m/z 2000 or 4000) (Prasain et al., 2004).
ESI spectra of glycosidic compounds typically show a
pseudomolecular ion (e.g., [M+H]+), aglycone ion and ions associated with
the solvent. Acid (acetic or formic) is often added to mobile phases in
positive ion electrospray as a source of protons to assist ionisation (Ryan et
al., 1999). HPLC-ESI/MS may be applied in the identification or
quantification of a wide range of flavonoids. The advantages of this sytem
are that work in a high mass range present a multiple charge resolution ~
2000 and its sensitivity goes down to the femtomole or picomole. The
disadvantages are the following: it has a relatively low salt tolerance, the
multiple charge status can be confused in mixtures, and the analysis may be
difficult for non-ionizable compounds (Prasain et al., 2004). MS is
generally used to determine the molecular mass and to establish the
distribution of substituents on the phenolic ring(s). ESI/MS has been used

75

Bibliographic review
____________________________________________________________________

for the structural confirmation of phenolics in plums, peaches, nectarines,

grapeseeds, soyfoods, cocoa, olive oil, walnut leaves, among others (Nackz
and Shahidi, 2004).

Figure 20. Major components of the ESI source. The sample solution is passed
through an electrically charged needle and the liquid takes the shape of a Taylor
cone as it comes under the influence of the flow and the electrostatic field (the
force on the ions drags the liquid along while surface tension tries to pull it into
a sphere). There is rapid evaporation of the droplet and the capillary into which
the droplets fly is heated to aid solvent evaporation. In the case of the IonSpray
interface, the sample is dispersed by a nebulizing gas (Source: Prasain et al.,
2004).

II.4. Mashua
Mashua (Tropaeolum tuberosum Ruz & Pavn), also known as
isau, cubio, au, maswa, ysao, or puel, is a tuber crop indigenous to the
Andean highlands and is of economic value as a food and medicinal crop
(Travis, 1999). Other edible tuber crops from the same area comprise several
potato (Solanum spp.) species, ulluco (Ullucus tuberosa Caldas) and oca
(Oxalis tuberosus Molina) (Grau et al., 2003). According to The National

76