Content uploaded by Marica Orioli
Author content
All content in this area was uploaded by Marica Orioli on Feb 17, 2015
Content may be subject to copyright.
Introduction
There is an increasing need for a better definition of the thera-
peutic properties of honey, which are restricted to its dermatolo-
gical beneficial effects: honey reduces skin inflammation, edema
and exudation, promotes wound healing, diminishes scar size
and stimulates tissue regeneration [1]. However, it may also
give benefit in more severe pathologies, i.e., gastrointestinal ul-
cerative diseases, atherosclerosis and cancer where free radicals
play a key role [2]. Hamzaoglu reported that tumor implantation
in rats was markedly reduced by the application of honey pre-
and post-operatively, suggesting that the physico-chemical ac-
tion (decrease of oxygen availability in the tumor environment,
i.e., anti-angiogenic effect) and its antioxidants can prevent the
spread of metastatic cells [3]. Honey products (given per os)
have also proved successful in the prevention and therapy of
murine transplantable tumors [4], [5].
Until 1990, the chemopreventive action of honey was attributed
to its hydrogen peroxide-releasing properties, through induction
of cell apoptosis [6], [7], but recent findings point to a comple-
mentary role of the phytochemical antioxidants which can act
synergistically or independently from the release of H
2
O
2
[8].
Honey contains an array of chemicals endowed with antiradi-
Antioxidant and Radical Scavenging Activity of Honey
in Endothelial Cell Cultures (EA.hy926)
Giangiacomo Beretta
Marica Orioli
Roberto Maffei Facino
Affiliation
Istituto di Chimica Farmaceutica e Tossicologica ªPietro Pratesiº, Faculty of Pharmacy, University of Milan,
Milan, Italy
Correspondence
Dr. Giangiacomo Beretta ´ Istituto di Chimica Farmaceutica e Tossicologica ´ Faculty of Pharmacy ´ University of
Milan ´ viale Abruzzi 42 ´ 20131 Milan ´ Italy ´ Phone: +39-02-5031-7519 ´ Fax: +39-02-5031-7565 ´
E-mail: giangiacomo.beretta@unimi.it
Received April 6, 2007 ´ Revised July 9, 2007 ´ Accepted July 17, 2007
Bibliography
Planta Med 2007; 73: 1182±1189 Georg Thieme Verlag KG Stuttgart ´ New York
DOI 10.1055/s-2007-981598 ´ Published online September 7, 2007
ISSN 0032-0943
Abstract
The therapeutic properties of honey, once considered a form of
folk or preventive medicine, are acquiring importance for the
treatment of acute and chronic free radical-mediated diseases
(atherosclerosis, diabetes and cancer). The aim of this work was
to study the protective activity of a honey of multifloral origin,
standardized for total antioxidant power and analytically pro-
filed (HPLC-MS) in antioxidants, in a cultured endothelial cell
line (EA.hy926) subjected to oxidative stress. Cumene hydroper-
oxide (CuOOH) was used as free radical promoter. Native honey
(1% w/v pH 7.4, 10
6
cells) showed strong quenching activity
against lipophilic cumoxyl and cumoperoxyl radicals, with sig-
nificant suppression/prevention of cell damage, complete inhibi-
tion of cell membrane oxidation, of intracellular ROS production
and recovery of intracellular GSH. Experiments with endothelial
cells fortified with the isolated fraction from native honey enrich-
ed in antioxidants, exposed to peroxyl radicals from 1,1-diphe-
nyl-2-picrylhydrazyl (AAPH, 10 mM) and to hydrogen peroxide
(H
2
O
2
,50±100
m
M), indicated that phenolic acids and flavonoids
were the main causes of the protective effect. These results pro-
vide unequivocal evidence that, through the synergistic action of
its antioxidants, honey by reducing and removing ROS, may low-
er the risks and effects of acute and chronic free radical induced
pathologies in vivo.
Key words
Honey ´ endothelial cells ´ radical scavenging activity ´ antioxi-
dant fraction ´ GSH
Original Paper
1182
cal/anti-inflammatory activity, i.e., phenolic derivatives which
can play roles, alone or in combination, in its anti-tumor, anti-in-
flammatory effects [9]. The anti-tumoral effect of honey seems to
be due to a multifactorial process: 1) release of cytotoxic H
2
O
2
(and of HO radicals after Fenton reaction) [7]; 2) a direct inhibi-
tion of COX-2 by some specific constituent (chrysin and caffeic
acid phenyl ethyl ester, CAPE) [10]; and 3) scavenging action
against different reactive oxygen species (ROS) responsible for
induction of the inflammatory burst, which if not properly quench-
ed/contained can degenerate into cell malignancy [11]. Several
studies have focused on the antioxidant power of different types
of honey, primarily seeking to identify its floral and geographical
origin [12], [13], [14] and none of them have investigated the
protective effect of this complex matrix in living cells subjected
to oxidative stress. This was the aim of the present work.
We used: 1) a well established human cell culture line (endothe-
lial cells, EA.hy926); 2) a set of different free radical promoters
(CuOOH, H
2
O
2
, AAPH); 3) a prototype honey of multifloral origin,
standardized in antioxidant activity according to different meth-
ods [Folin-Ciocalteu, 1,1-diphenyl-2-picrylhydrazyl (DPPH), fer-
ric reducing antioxidant power (FRAP), and oxygen radical absor-
bance capacity, (ORAC)] [15]; and 4) an isolated fraction from
multiflora honey uniformly enriched in antioxidant constituents,
whose structure was analytically profiled by conventional tech-
niques (LC-UV/DAD-MS). Our aim was to find out whether the
cytoprotective activity of honey arose from a free radical scaven-
ging action of these constituents, and provide a mechanistic ex-
planation of the results.
Materials and Methods
Chemicals and cell lines
The organic solvents were all of analytical grade (Sigma-Aldrich;
Milan, Italy). 2,2
¢
-Azobis(2-amidinopropane) dihydrochloride
(AAPH) was from Wako Chemicals (Società Italiana Chimici;
Rome, Italy),
b
-phycoerythrin (
b
-PE) from Porphydium cruentum,
6-hydroxy-2,5,7,8-tetramethyl-2-carboxylic acid (Trolox), and N-
acetylcysteine (NAC) were purchased from Sigma±Aldrich. 1,1-Di-
phenyl-2-picrylhydrazyl (DPPH), cumene hydroperoxide (CuOOH),
Folin-Ciocalteu reagent, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphe-
nyltetrazolium bromide (MTT), o-phenyl-aldehyde (OPA) FeCl
3
´6-
H
2
OandFeSO
4
´7H
2
O and hydrogen peroxide 30% were from Fluka
(Buchs, Switzerland). 4,4-Difluoro-5-(4-phenyl-1,3-butadienyl)-4-
bora-3a,4a-diaza-S-indacene-3-undecanoyl acid (C
11
-BODIPY 581/
591, BODIPY) and 2
¢
,7
¢
-dichlorodihydrofluorescein diacetate
(DCFH-DA) were from Molecular Probes (Space Import±Export;
Milan, Italy). Chrysin, galangin, pinocembrin, luteolin, caffeic
acid, cinnamic acid, p-coumaric acid, syringic acid, benzoic acid,
Amberlite XAD-2, Dulbecco's phosphate buffered saline (PBS),
Dulbecco's modified Eagle's medium (DMEM), penicillin and
streptomycin, HAT media supplement, and
L-glutamine were all
from Sigma. Human endothelial cells (EA.hy926) were from Flow
Laboratories (Irvine, UK).
Spectrophotometric apparatus
Spectrophotometric measurements were conducted with a com-
puter-aided Perkin-Elmer UV-VIS spectrophotometer Lambda 16
(Perkin-Elmer; Monza, Italy). Fluorimetric determinations were
carried out with a Wallac Victor
2
Multiwell instrument (Perkin-
Elmer) with fluorescent filters (
l
ex
= 485 nm;
l
em
= 535 nm).
Honey
Commercial multiflora honey was purchased in a supermarket
(Milan, Italy), stored at 4 8C in the dark until tested by the usual
physicochemical tests (pH, electrical conductivity, titrable acid-
ity, ash content) and by qualitative tests [Lugol test and diastases
index for authenticity, hydroxymethylfurfural test (HMF) for
quality] and quantitative tests (reducing and non-reducing su-
gars) [16]. A sugar analogue consisting of 40% fructose, 30% glu-
cose, 10% maltose and 20 % water was used to check whether the
main sugar components of honey interfere in the assays.
Total phenolics and antioxidant activity in honey
The total phenol content of whole honey was determined by a
modification of the Folin-Ciocalteu method, and Trolox
was
used as standard to plot the calibration curve. The radical
scavenging activity of each XAD-2 honey fraction was evaluated
using different approaches (FRAP, DPPH and ORAC) according to
the methods previously described by us [15]. ORAC values are
expressed as Trolox equivalent (TE).
Cell cultures
Endothelial cells were cultured in DMEM supplemented with
10 % fetal calf serum, 2 mM
L-glutamine, antibiotics (100 U/mL
penicillin and 0.1 mg/mL streptomycin) and HAT supplement.
The cells were maintained in a humidified 5% CO
2
atmosphere
and sub-cultured 1:3 twice a week. All experiments were con-
ducted at 378C.
Model 1: protective effect of honey against intra-cellular and
cell membrane lipid oxidation
The cell monolayer was washed twice with the pre-warmed PBS
(pH 7.4) and then exposed to CuOOH (100
m
M) for 10 minutes.
The cells were then washed twice to remove the oxidant and
two types of fluorescent probe were incorporated: 1) BODIPY to
evaluate cell membrane oxidation, and 2) DCFH-DA for the intra-
cellular oxidative damage. Both the probes were incubated at 10
m
M (PBS) for 30 min. Then, the cell cultures were washed twice
with PBS and incubated with the following solutions:
± PBS (control)
± Honey solution (1% w/v in PBS)
± Trolox solution (30
m
M in PBS)
± Honey blank (1% w/v sugar analogue in PBS)
The membrane and intracellular oxidation kinetics were moni-
tored with a Victor
2
Wallac multi-well fluorescent plate reader
(excitation filter 500 10 nm; emission filter 530 12.5 nm)
using the following instrumental parameters: CW lamp energy
15,000 (arbitrary units), irradiation time 0.1 s, and cycle time 5
min for 46 cycles (time analysis = 60 min).
DCFH-DA assay
The DCFH-DA stock solution was prepared in ethanol (3.3 mM)
and stored under nitrogen at ±208C. Intracellular oxidative dam-
age was investigated by the DCFH-DA assay [17]. The endothelial
cells were washed twice with the pre-warmed PBS, incubated
with a DCFH-DA solution in PBS (10
m
M for 10 min), and the
probe incorporation was confirmed by fluorescence microscopy
Beretta G et al. Antioxidant and Radical ¼ Planta Med 2007; 73: 1182 ±1189
Original Paper
1183
(20 Ph1 ADL lens and a FITC filter -Eclipse TS 100; Nikon; Tokyo,
Japan).
The cell monolayer was washed twice with PBS, then incubated
with the following solutions:
± PBS
± Honey blank
± Honey solution (1% w/v in PBS)
± Trolox solution (30
m
M in PBS)
The fluorescence intensity of DCF, the DCFH-DA oxidation prod-
uct, was monitored as described at an excitation wavelength of
485 10 nm and an emission wavelength of 530 12.5 nm for 1
hour.
Model 2: preventive effect of honey against acute oxidative
insult
The endothelial cell monolayer was washed twice (PBS), then in-
cubated for 15 h with 1) D-MEM alone; 2) D-MEM supplemented
with 1 % honey (w/vol); 3) 50
m
M vitamin E, 4) 200
m
MNAC,or5)
with the honey blank (1% w/vol).
At the end of the pre-treatment, the cells were washed free of the
cell culture medium, the probe BODIPY and the CuOOH incorpo-
rated and the lipid oxidation kinetics were monitored as above
described.
Cell viability
Cell viability was determined by the MTT test as already reported
[18].
GSH analysis
Endothelial intracellular GSH was determined as already report-
ed [19].
Preparation of the antioxidant-enriched fraction from honey
Fifty g of the prototype multiflora honey were diluted with four
volumes of distilled water, stirred until completely dissolved, fil-
tered on cotton to remove solid particles and the pH adjusted to
2.5 with 6 N HCl. The filtrate was passed through a column
(25 cm2.5 cm) of Amberlite XAD-2 resin and after sequential
washing first with acidic (200 mL, pH 3) and then with neutral
water (200 mL) [20], the antioxidant-enriched fraction was re-
covered by elution with methanol (200 mL). The methanol frac-
tion was dried under vacuum at 40 8C, taken up in methanol/H
2
O,
2:1 (4 mg/mL) and analyzed by HPLC-UV-DAD-MS analysis.
All the fractions were tested for the total phenolic content (Folin-
Ciocalteu) and for radical scavenging/antioxidant activity (FRAP,
DPPH etc).
Characterization of honey antioxidant constituents (LC-DAD
and APCI-MS analyses)
HPLC-DAD experiments were carried out on a Thermoquest Sur-
veyor System (Thermoquest; Milan, Italy), equipped with a qua-
ternary pump, a Surveyor Model 6000 LP UV/VIS diode array pro-
grammable detector, operating at 280 nm, a Surveyor AS auto-
sampler, a vacuum degasser and a Xcalibur Software. Compo-
nents were separated with a Phenomenex Synergy RP8 column
(150 mm2 mm i.d.; particle size 4
m
m) protected with a Max-
RP guard column (4 mm 2 mm i. d.; particle size 4
m
m). Gradient
elution: 100% solvent A [95% ammonium formate (3.2 mM)/5%
CH
3
CN, pH 4.5 with HCOOH ] to 40% B [95% CH
3
CN, 5% ammo-
nium formate (3.2 mM), pH 4.5 with HCOOH] in 30 minutes,
then to 100% solvent B after 10 minutes; flow rate 0.2 mL/min.
APCI-HPLC-MS analysis was conducted with a Thermo Finnigan
LCQ Advantage ion trap mass spectrometer (Thermoquest). Va-
porizer temperature 4508C, ionization voltage 8 kV. Nebulizer
gas (N
2
) flow rate 0.5 L/min. Spectra were detected in positive
and negative ion mode (100±600 m/z, 0.5 scan/sec).
Cytoprotective activity of honey antioxidant-enriched
fraction
The cell monolayer was washed twice with pre-warmed PBS,
then incubated with: 1) BODIPY for cell oxidation, or 2) DCFH-
DA for intracellular oxidative damage. The cells were subjected
to oxidative stress by AAPH (10 mM in PBS), with or without the
honey antioxidant fraction and the oxidative damage was
checked after 1, 2, 3 h incubation times. The H
2
O
2
stress was de-
livered to cell cultures using increasing concentrations of the hy-
droperoxide (50± 100
m
M in PBS). After 1 h, the oxidative treat-
ment was stopped by washing the monolayers twice with PBS
and the increase of fluorescence of oxidized DCFH-DA was mon-
itored. In parallel cell morphology was evaluated. In both the oxi-
dative assays the antioxidant enriched fractions was incorpora-
ted in the endothelial cells (10 min, 378C) and these exposed to
the free radical inducers.
Statistical analysis
Statistical analyses were conducted with the Prism software for
Windows software package (GraphPad Inc.; San Diego, CA, USA).
Results are expressed as the mean (S.D). of four independent ex-
periments. Student's t-test was used; P values < 0.05 were con-
sidered to be significant.
Results and Discussion
Fig.1 reports the results relative to the cytoprotective activity of
honey and Trolox
in parallel with a honey blank in cell cultures
exposed to oxidative stress by the free radical promoter CuOOH
(100
m
M).
When endothelial cells were incubated with 100
m
M CuOOH,
even for a short time (10 min), there were marked alterations
to the original shape of the cell, as shown by phase-contrast mi-
croscopy (Fig.1A), which were paralleled by a significant loss of
cell viability: 733%vs1003%incontrolcells(p<0.001,
Fig.1B).
The dramatic change in morphology was caused by membrane
lipid oxidation (see the appearance and time-dependent in-
crease in BODIPY fluorescence, Fig.1C) due to Fe(II)-trace-cata-
lyzed decomposition of CuOOH to cumoxyl and cumoperoxyl
radicals, which diffuse from the membrane lipid bilayer into
the intracellular compartment. In parallel the cell begins to
show preliminary signs of a shrunken cytoplasm, a more rounded
shape, cell surface blebbing, and loss of inter-cellular adher-
ence.
Beretta G et al. Antioxidant and Radical ¼ Planta Med 2007; 73: 1182 ± 1189
Original Paper
1184
The addition of honey (1% w/v, equivalent to 0.85
m
g
TE
/10
6
cells)
reverted the progression of the damage: the BODIPY fluores-
cence, after a slight increase, remained almost constant within
the 1 h oxidation time. Similar inhibition, although less pro-
nounced, was observed for intracellular oxidation: after 1 h the
damage was 60 % less than in unprotected cells (Fig.1D). In the
presence of honey, cell viability slightly but significantly in-
creased, from 73 3% in unprotected cells to 87 4 % in cells
treated with honey (P < 0.05). Hence the antioxidant constitu-
ents of honey can inhibit the free radical species (i. e., cumoxyl
and cumoperoxyl radicals) both at membrane level and intracel-
lularly, limiting the propagation of the oxidative cascade. Alter-
natively these constituents may inactivate cytotoxic CuOOH by
metabolic reduction, through activation of GSH-peroxidase
(GSPx) [22]. In accordance with previous demonstrations in
homogeneous phase models [15], the honey blank did not give
any protective effect.
Then we evaluated the susceptibility of endothelial cells, incuba-
ted overnight with native honey, to CuOOH, checking membrane
lipid peroxidation, cell morphology and viability and intracellu-
lar redox status (GSH).
The cells showed 1) markedly less susceptibility to lipid peroxi-
dation (30 % lower than control cells, Fig. 2A); 2) more limited
morphological alterations (Fig. 2B); 3) a slight increase in cell
viability (15% gain, Fig. 2C).
The basal content of GSH in control cells shown in Fig. 3 (30.0
2.0 nmol/mg protein, column a), was not affected by incubation
with sugar blank (column b), honey (column c) or vitamin E (col-
umn d), but NAC induced a significant rise in intracellular GSH
(42% gain, column e).
Exposure to CuOOH, significantly reduced the level of GSH to 17.9
0.6 nmol/mg protein (column f). The GSH content (28.2 0.2
nmol/mg protein, column j) of cells incubated with NAC and ex-
posed to CuOOH was close to the control value, the smaller drop
being due to the additional GSH pool biosynthesized from the
precursor during the incubation. Most importantly, in cells
grown in DMEM enriched with honey (column h) then challeng-
ed with CuOOH, there was significant sparing of native GSH in
respect to control cells (GSH content 22.4 0.5 nmol/mg vs.
17.9 0.6 nmol/mg, column g, P < 0.05). Vitamin E, deeply and
firmly incorporated within the cell membrane, had no effect
(column i).
It therefore appears that some honey components, which are li-
pophilic, can cross the endothelial cell membrane and 1) spare
the endogenous antioxidant GSH from the oxidative attack of li-
Fig. 1 Protective effect in endothelial
cells (EA.hy926) of 1 % native honey in
PBS against oxidative stress. Free radical
promoter: CuOOH (100
m
M); control
(CTR): sugar blank (1 % w/v, PBS); Tro-
lox
(10
m
M). A Phase contrast micro-
scopy. B Cell viability (MTT test). C
Membrane lipid peroxidation (BODIPY).
D Intracellular ROS production (DCFH).
Results are expressed as the mean
S.D. of 4 independent experiments. * P
< 0.05, Student's t test, vs. CuooH + su-
gar blank.
Beretta G et al. Antioxidant and Radical ¼ Planta Med 2007; 73: 1182 ± 1189
Original Paper
1185
Fig. 2 Preventive effect of native hon-
ey (1 % PBS) against CuOOH (100
m
M)
promoted oxidative stress. Control
(CTR): sugar blank (1 % w/v, PBS). A
Membrane lipid peroxidation (BOPIDA).
B Phase contrast microscopy. C Cell via-
bility (MTT test). Results are expressed
as the mean S.D. of 4 independent ex-
periments.
Fig. 3 Intracellular GSH levels in endothelial cells pre-
treated overnight with: 1 % sugar blank, 1 % native honey,
50
m
M vitamin E and 200
m
M N-acetylcysteine (NAC). Unex-
posed cells (columns a ± e) and 100
m
M CuOOH exposed
cells (columns f ± l). Results are exprerssed as te mean
S.D. of 4 independent ewxperiments. * P < 0.05 Student's t
test, column h vs g.
Beretta G et al. Antioxidant and Radical ¼ Planta Med 2007; 73: 1182 ± 1189
Original Paper
1186
pophilic peroxyl radicals, and/or 2) regenerate GSH from GSSG by
a hydrogen-transfer mechanism, or both.
Finally, we evaluated the protective activity of the antioxidant
fraction isolated from native honey. Cells were exposed to a flux
of peroxyl-radicals generated by AAPH (10 mM) or to H
2
O
2
(50
and 100
m
M). Fig. 4 shows the HPLC profile of the antioxidant
fraction of multiflora honey, containing several constituents (Ta-
ble 1). These were identified on the basis of the UV-DAD profile,
MS fragmentation pattern, and by comparison of the chromato-
graphic, spectroscopic and MS properties of reference standards
(not shown), as caffeic acid, p-coumaric acid, eriodictiol, pino-
banksin, chrysin, pinocembrin, and galangin (typical of propolis
and beeswax), and as kaempferol and apigenin (from pollen).
There was a marked increase in membrane lipid-peroxidation in
unprotected endothelial cells exposed to AAPH (see the time-de-
pendent increase of BODIPY fluorescence, Fig. 5A), with no signs
of morphological alterations. Addition of the antioxidant fraction
(5, 10, 15
m
g
ext(TE)
/10
6
cells), induced a dose- and time-dependent
decrease in the membrane lipid peroxidation, which was sup-
pressed at 15
m
g
ext(TE)
/10
6
cells. Comparing this protective effect
with that given by 1% native honey (0.85
m
g
ext(TE)
/10
6
cells), the
latter is from 15 to 20-fold more active than its antioxidant frac-
tion. This suggests that by the chromatographic sorbtion method
adopted [12], [13], [14], [21], [23], we lose an appreciable amount
of antioxidant in the filtrate and in the first washings which can
synergistically concur with those eluted with methanol in the
cytoprotective activity of multiflora honey. Our assays of total
phenol content and total antioxidant activity gave direct evi-
dence for an overall loss in the washings of antioxidant material
of approximately 40% in respect to honey as such (not shown).
Very likely highly polar, hydrophilic antioxidant phenolics elute
in the first washings embedded in sugars. This is in accordance
with the data of Gheldof et al. and Weston et al. [21], [24], who
using the same method, found more than 40% of the total anti-
bacterial and antioxidant activity of honey in the first washings.
This phenol-positive material is not extractable with organic sol-
vents (unpublished data). However, it is also possible that some
other antioxidants in the matrix, structurally unrelated to poly-
phenols (vitamin E, carotenoids, small proteins, amino acids),
may not be detected by our LC-MS method. These too can play a
role in the cytoprotective activity of multiflora honey. Overall,
these considerations help to explain the contradictory reports
Table 1 Main components of the antioxidant enriched fraction from multiflora honey (RT retention time, MW molecular weight)
Peak number RT (min) MW (a. m. u.) Compound
1 6.43 467 convicine diglucoside*
2 10.50 163 4-N-aziridylbenzoic acid*
3 10.55 165 4-dimethylaminobenzoic acid*
4 17.92 180 caffeic acid
5 26.23 164 p-coumaric acid
6 39.55 288 eriodictiol
7 43.23 286 kaempferol
8 44.72 270 apigenin
9 45.14 272 pinobanksin
10 46.04 330 quercetin dimethyl ether
11 47.36 330 quercetin dimethyl ether (isomer)
12 48.03 254 chrysin
13 48.35 256 pinocembrin
14 48.52 270 galangin
* Nitrogen-containing compounds 1, 2 and 3 were only tentatively identified.
Fig. 4 HPLC-UV (280 nm) profile of the antioxidant enrich-
ed fraction from native multiflora honey. For structure as-
signments, see Table 1.
Beretta G et al. Antioxidant and Radical ¼ Planta Med 2007; 73: 1182 ± 1189
Original Paper
1187
on the biological activity of purified honey fractions compared to
native honey [24], and point to the need for studies and clinical
trials using native honey as such.
When we used the physiological oxidant H
2
O
2
, source of super-
oxide anion by Haber-Weiss reaction or of HO
.
radical by Fenton
reaction, free to permeate the cell compartments, there was a
strong increase of intracellular ROS (Fig. 5B), due to depletion of
GSH. Because of the weak oxidative conditions, close to in vivo
situations, phase contrast microscopy did not show up any mor-
phological alterations (not shown). After 1 h of incubation at the
concentration of honey fraction antioxidants (15
m
g
ext(TE)
/10
6
cells) which gave the highest degree of protection in the AAPH
experiment, oxidative damage was halved with 50
m
MH
2
O
2
and
there was a 35% inhibition with 100
m
MH
2
O
2
. This can be due to:
a) direct quenching of honey components against H
2
O
2
; b) the
ability of some antioxidant constituents to permeate the living
cell, drop into the intracellular space and spare thiol groups, or
c) activation of GSPx.
Whatever the mechanism, the results with a purified mixture of
honey antioxidants provide the first direct evidence that the pro-
tective effect of native honey in endothelial cells is basically due
to the phytochemical antioxidants in the matrix (Fig. 4 and Ta-
ble 1), acting against free radicals generated by different promo-
ters.
In addition, the presence of these phenolic acids and flavonoids
provides a reasonable explanation for the antioxidant power
we found in native multiflora honey, using different methods
[15].
The results of this study in living cells incubated with a reason-
ably low concentration of native honey and challenged with dif-
ferent free radical promoters provide unequivocal evidence that
honey possesses a wide spectrum of antioxidant activity, which
can give rise in vivo to preventive and curative effects in several
inflammatory pathologies and could play a role in cancer chemo-
prevention. A link has been demonstrated between inflamma-
tion and tumorigenesis in a mouse model of colitis-associated
cancer [11]. In addition in cancer cells honey might help prevent
proliferation by reducing the high oxidative state that cancer
cells need for proliferation [22].
The fact that native honey was an effective quencher of free radi-
cals when incorporated in endothelial cells at a 1% dilution ex-
cludes any intervention of the redox-active H
2
O
2
since its forma-
tion is negligible at this dilution. For the same reason, any protec-
tive effect due to the chemico-physical properties of the complex
matrix, or to the action of some trace mineral with antioxidant
properties, can be excluded.
One of the main findings reported in the present study is that the
honey components spare or regenerate endothelial GSH, which is
the first line of defence in the endothelial cells of the vascular
wall against the atherogenic action of peroxidized LDL and
against the strong oxidant peroxynitrite formed in those parts
of the arterial wall exposed to oscillatory shear stress (OSS),
where there is over-production of ROS due to up-regulation of
membrane NADPH oxidase [25]. Thus, honey components com-
ing into contact with a tissue rich in endothelial cells (such as the
arterial wall) can prevent the atherogenic action of oxidized LDL,
and boost the intracellular GSH pool which plays a key role in
Fig. 5 Protective effect of the antioxidant honey extract
from multiflora honey in endothelia cells against: (A) peroxyl
radicals (generator 10 mM AAPH, probe BODIPY) and (B)in-
tracellular ROS production from 50 and 100
m
MH
2
O
2
(probe
DCPH). The concentration was expressed as
m
gTE/10 cells.
Results are expressed as the mean S.D. of 4 independent
experiments.
Beretta G et al. Antioxidant and Radical ¼ Planta Med 2007; 73: 1182 ± 1189
Original Paper
1188
counteracting the action of circulating ROS and reactive nitrogen
species (RNS).
References
1
Molan PC. Potential of honey in the treatment of wounds and burns.
Am J Clin Dermatol 2001; 2: 13±9.
2
Mobarok Ali ATM, Al-Swayeh AO. Natural honey prevents ethanol-in-
duced increased vascular permeability changes in the rat stomach. J
Ethnopharmacol 1997; 55: 231± 8.
3
Hamzaoglu I, Saribeyoglu K, Durak H, Karahasanoglu T, Bayrak Altug I,
Sirin TF et al. Protective covering of surgical wounds with honey im-
pedes tumor implantation. Arch Surg 2000; 135: 1414±7.
4
Swellam T, Miyanaga N, Onozawa M, Hattori K, Kawai K, Shimazui T et
al. Antineoplastic activity of honey in an experimental bladder cancer
implantation model: in vivo and in vitro studies. Int J Urol 2003; 10:
213± 9.
5
Orsolic N, Terzic S, ver L, Bas
Ï
ic
Â
I. Honey-bee products in prevention
and/or therapy of murine transplantable tumours. J Sci Food Agric
2005; 85: 363± 70.
6
Bang LM, Buntting C, Molan P. The effect of dilution on the rate of hy-
drogen peroxide production in honey and its implications for wound
healing. J Altern Compl Med 2003; 9: 267±73.
7
Lopez-Lazaro M. Dual role of hydrogen peroxide in cancer: Possible re-
levance to cancer chemoprevention and therapy. Cancer Letters;
2006available online at http://www.sciencedirect.com
8
Weston RJ. The contribution of catalase and other natural products to
the antibacterial activity of honey: a review. Food Chem 2000; 71:
235±9.
9
Facino RM. Honey in tumor surgery. Arch Surg 2001; 136: 600.
10
Michaluart P, Masferrer JL, Carothers AM, Subbaramaiah K, Zweifel BS,
Koboldt C et al. Inhibitory effects of caffeic acid phenethyl ester on the
activity and expression of cyclooxygenase-2 in human oral epithelial
cells and in a rat model of inflammation. Cancer Res 1999; 59: 2347±
52.
11
Greten FR, Eckmann L, Greten TF, Park JM, Li Z-W, Egan LJ et al. IKK
b
links inflammation and tumorigenesis in a mouse model of colitis-
associated cancer. Cell 2004; 118: 285± 96.
12
Ferreres F, Tomàs-Barberàn FA, Gil MI, Tomàs-Lorente F. An HPLC tech-
nique for flavonoids analysis in honey. J Sci Food Agric 1991; 56: 49±
56.
13
Tomàs-Barberàn FA, Ferreres F, Amparo-Blazquez M, Garcia-Viguera C,
Tomas-Lorente F. High-performance liquid chromatography of honey
flavonoids. J Chromatogr 1993; 634: 41 ±6.
14
Ferreres F, Tomàs-Barberàn FA, Soler C, Garcia-Viguera C, Ortiz A, To-
màs-Lorente F. A simple extractive technique for honey flavonoid
HPLC analysis. Apidologie 1994; 25: 21±30.
15
Beretta G, Granata P, Ferrero M, Orioli M, Maffei Facino R. Standardiza-
tion of antioxidant properties of honey by a combination of spectro-
photometric/fluorimetric assays and chemometrics. Anal Chim Acta
2005; 533: 185 ±91.
16
Azeredo L da C, Azeredo MAA, de Souza SR, Dutra VML. Protein con-
tents and physicochemical properties in honey samples of Apis
mellifera of different floral origins. Food Chem 2003; 80: 249± 54.
17
Carini M, Aldini G, Piccone M, Maffei Facino R. Fluorescent probes as
markers of oxidative stress in keratinocyte cell lines following UVB ex-
posure. Farmaco 2000; 55: 526± 34.
18
Mosmann T. Rapid colorimetric assay for cellular growth and survival:
application to proliferation and cytotoxicity assays. J Immunol Meth-
ods 1983; 65: 55 ±63.
19
Aldini G, Granata P, Beretta G, Carini M, Maffei Facino R. Effects of UVB
radiation on 4-hydroxy-2-transnonenal metabolism and toxicity in
human keratinocytes. Chem Res Toxicol. 2007; 20: 416±23.
20
Markwell MA, Haas SM, Tolbert NE, Bieber LL. Protein determination
in membrane and lipoprotein samples: manual and automated proce-
dures. Methods Enzymol 1981; 72: 296±303.
21
Gheldof N, Wang XH, Engeseth NJ. Identification and quantification of
antioxidant components of honeys from various floral sources. J. Agric
Food Chem 2002; 50: 5870± 7.
22
Kluth D, Banning A, Paur I, Blomhoff R, Brigelius-Flohe R. Modulation
of pregnane X receptor- and electrophile responsive element-medi-
ated gene expression by dietary polyphenolic compounds. Free Radic
Biol Med 2007; 42: 315± 25.
23
Gómez-Caravaca AM, Gómez-Romero M, Arrµez-Romµn D, Segura-
Carretero A, Fernµndez-GutiØrrez A. Advances in the analysis of phe-
nolic compounds in products derived from bees. J Pharm Biomed
Anal 2006; 41: 1220±34.
24
Weston KR, Mitchell RJ, Allen KL. Antibacterial phenolic components
of New Zealand manuka honey. J Food Chem 1999; 64: 295 ±301.
25
Hsiai TK, Hwang J, Barr ML, Correa A, Hamilton R, Alavi M et al. Hemo-
dynamics influences vascular peroxynitrite formation: Implication for
low-density lipoprotein apo-B-100 nitration. Free Radic Biol Med
2007; 42: 519± 29.
Beretta G et al. Antioxidant and Radical ¼ Planta Med 2007; 73: 1182 ± 1189
Original Paper
1189
