Journal of Photochemistry and Photobiology B: Biology 85 (2006) 205–215

www.elsevier.com/locate/jphotobiol

Influence of astaxanthin, zeaxanthin and lutein on DNA damage

and repair in UVA-irradiated cells
Marcello Santocono a, Monica Zurria a, Marco Berrettini b, Donatella Fedeli b,
Giancarlo Falcioni b,*
a
Medical Department, SIFI SpA, Via E. Patti 36, Lavinaio (CT), Italy
b
Dipartimento di Biologia MCA, Università degli Studi di Camerino, Via Camerini, 2 I-62032 Camerino (MC), Italy

Received 16 January 2006; received in revised form 19 May 2006; accepted 18 July 2006
Available online 8 September 2006

Abstract

In order to gain more knowledge about the antioxidant role of the predominant carotenoids (lutein and zeaxanthin) of the human
retina, this study investigated their antioxidant activity and capacity. Astaxanthin was also studied, because its structure is very close
to that of lutein and zeaxanthin. The antioxidant activity of these molecules was evaluated using chemiluminescence techniques, with
lucigenin and luminol as chemiluminogenic probes for the superoxide radical and hydrogen peroxide, respectively. It was found that
all three carotenoids have similar superoxide-scavenging activity. The effect on the reduction of H2O2–luminol chemiluminescence
was present in the following order, zeaxanthin > astaxanthin P lutein.
Possible antioxidant capacity of these three compounds was sought using a biological system consisting of SK.N.SH human neuro-
blastoma and rat trachea epithelial cells subjected to oxidative stress from exposure to UVA radiation. In particular, we determined
whether these compounds were capable of minimizing DNA damage and influencing the kinetics of DNA repair.
DNA damage was assessed using the Comet assay, a rapid and sensitive single-cell gel electrophoresis technique used to detect pri-
mary DNA damage in individual cells. Neuroblastoma cells appeared more resistant to oxidative irradiation insult. The presence of
carotenoids reduced DNA damage when rat epithelial cells were exposed to UVA radiation for 2 min. A different result was obtained
in experiments performed on neuroblastoma cells; in this case, the presence of carotenoid during UVA exposition increased the damage.
The addition of carotenoids to epithelial cells after 2 min of UVA exposition did not seem to improve the kinetics of DNA repair; on
the contrary, zeaxanthin (after 60 0 incubation) and lutein (after 180 0 incubation) showed a genotoxic effect.
The addition of carotenoids to neuroblastoma cells after 30 0 UVA exposition positively influences the kinetics of DNA repair in the
first 15 min of incubation. At longer exposition times, while the behaviour measured was not constant, a genotoxic effect was not
observed. The data from this study provide additional information on the antioxidant and pro-oxidant activities of the predominant
macular pigment carotenoids of the human retina.
 2006 Elsevier B.V. All rights reserved.

Keywords: Carotenoids; ROS; DNA; Comet assay; UV radiation

1. Introduction Ultraviolet radiation induces oxidative stress because it

It is now well established that solar radiation is a geno-
toxic agent; damage is wrought by the UV region of the
spectrum (for a recent review [1]).
*
Corresponding author. Tel.: +39 0737 403211; fax: +39 0737 636216.
E-mail address: giancarlo.falcioni@unicam.it (G. Falcioni).
produces reactive oxygen species [2] and modulates the level
of antioxidants [3–6]. Oxidative stress leads to DNA, pro-
tein and lipid damage. In DNA, hydroxyl radicals (OH)
have been observed to cause strand breaks, and singlet oxy-
gen (1O2) has been observed to oxidize bases [2].
UV spectrum radiation encompasses UVA (315–
400 nm), UVB (280–315 nm) and UVC (100–280 nm).

1011-1344/$ - see front matter  2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jphotobiol.2006.07.009
206 M. Santocono et al. / Journal of Photochemistry and Photobiology B: Biology 85 (2006) 205–215
However, since UVC radiation is filtered out, for the most
part, by atmospheric ozone, both UVA and UVB radiation
play a more significant role in the initiation of photo-carci-
nogenesis [7]. At a molecular level, UVA and UVB differ in
the sites of action in the generation of pre-mutagenic
lesions. UVB radiation is site specific and is absorbed
directly by cellular DNA. The most frequent photolesions
resulting from UVB-induced DNA alterations are cyclobu-
tane pyrimidine dimers (CPDs) capable of interfering with
DNA replication. However, these can be removed by sev-
eral repair mechanisms, including excision repair.
Conversely, UVA radiation does not attack the DNA
directly, but is adsorbed by intracellular chromophores
such as riboflavin and membrane bound enzymes. This
results in an altered cellular redox potential via the genera-
tion of reactive oxygen species (ROS) and/or nitric oxide,
causing photosensitization.
Ultraviolet radiation provokes overproduction of free
radicals, which damages biological systems. This observa-
tion has stimulated research into the role of natural antioxi-
dants that can mitigate photobiologic damage. Our interest
in this field has focused on carotenoids, a group of natural
pigments that have been recently used in biological systems
for their antioxidant capacity [8,9]. It now appears that
important actions can be attributed to some carotenoids,
and evidence indicates that they may reduce development
rates of some human cancers [10]. However, there are
increasing concerns that carotenoids such as b-carotene
and lycopene may also exert pro-oxidant properties under
certain conditions (for recent reviews [11,12]).
In the present study, three different carotenoids – lutein,
zeaxanthin and astaxanthin – were used (Scheme 1). The
HO
O
ASTAXANTHIN
HO
ZEAXANTHIN
HO
LUTEIN
Scheme 1. Chemical structures of the compounds studied.
xantophylls lutein and zeaxanthin are the predominant
carotenoids of the macular pigment of the human retina
and the levels of their concentration in the retina have func-
tional and pathological consequences (i.e. age-related macu-
lar degeneration). Individuals suffering from age-related eye
disease have inferior xantophyll densities throughout their
retinas, and dietary zeaxanthin and lutein levels are inversely
linked to the risk of AMD as well as cataracts. Lutein and
zeaxanthin cannot be synthesized by humans and must be
obtained through diet via consumption of fruits and vegeta-
bles [13,14]. It has been reported that macular pigments
might prevent a wide range of human diseases, including
various cancers and other conditions [15]. Although asta-
xanthin has never been isolated from the human eye, it
was included in this study because its structure is very close
to that of lutein and zeaxanthin and because it affords pro-
tection from UV-light [16]. Like zeaxanthin and lutein, it
cannot be synthesized by mammals and must be acquired
through diet. Astaxanthin is found in salmon, trout, shrimp
and other aquatic animals, where it has several biological
functions.
Using chemiluminescence [17] with two different chemi-
luminogenic probes, lucigenin and luminol, we investigated
the antioxidant activity of these carotenoids versus super-
oxide anion ðO 2 Þ and hydrogen peroxide (H2O2). These
compounds were also studied to determine whether they
can minimize DNA damage inflicted on SK.N.SH human
neuroblastoma and rat trachea epithelial cells following
exposure to UVA radiation, and whether they can influ-
ence the kinetics of DNA repair. DNA damage inflicted
by UVA radiation was compared to H2O2-induced DNA
damage.
O
OH
OH
OH
 M. Santocono et al. / Journal of Photochemistry and Photobiology B: Biology 85 (2006) 205–215
DNA damage and repair were evaluated using comet
assay, a single-cell gel electrophoresis [18]. This test
involves embedding cells in agarose, followed by lysis, elec-
trophoresis and staining to visualize DNA damage using
fluorescent microscopy. Tail moment, the parameter used
as an index of DNA damage, is obtained by computerised
image analysis.
The present study evaluates SK.N.SH human neuro-
blastoma cells as an ‘‘in vitro’’ model of UVA induced pho-
toreceptor cell DNA damage and examines the effect of
zeaxanthin, lutein and astaxanthin on the process. Even
though significant and obvious differences between AMD
and CNS degenerations are present, they also have much
in common, suggesting that some of the principles emerg-
ing from research in field could potentially be relevant to
the other. Experiments involving rat trachea epithelial cells
were performed because these cells are less resistant to oxi-
dative irradiation insult.
2. Materials and methods
All reagents were of pure analytical grade. Astaxanthin,
Tempo, xanthine oxidase, lucigenin and luminol were pur-
chased from Sigma Chemical Co. (St. Louis, MO, USA),
zeaxanthin and lutein were provided by Extrasynthese,
Genay-France and by Kemin Foods, L.C, Iowa, USA,
respectively. Dulbecco’s Modified Eagle medium (MEM)
was purchased from Lifetech (Life Technologies Ltd., Pals-
ley, Scotland).
2.1. Chemiluminescence assays
The level of superoxide radical produced by the xan-
thine/xanthine oxidase system was measured using lucige-
nin as chemiluminogenic probe. Chemiluminescence was
measured in a reaction mixture containing 0.9 U/ml xan-
thine oxidase, 150 lM lucigenin in Tris 50 mM, pH 7.4 buf-
fer and different concentrations of the compounds ranging
from 0 to 90 lM dissolved in ethanol for lutein and methy-
lene chloride/methanol 1:1 for zeaxanthin and astaxanthin.
The reaction was started by injecting xanthine at a final
concentration of 50 lM and followed for 60 s as described
previously [17].
Luminol amplified chemiluminescence was performed
using a reaction mixture with 105 M luminol, and vari-
ous concentrations of carotenoids ranging from 0 to
25 lM (astaxanthin, zeaxanthin and lutein) dissolved in
methylene chloride/ethanol 1:1. The reaction was initi-
ated by injecting hydrogen peroxide into the solution at
a final concentration of 50 mM in methylene chloride/
ethanol 1:1.
Chemiluminescence (CL) was measured in an LB 953
Autolumat (Berthold) Co. Wildbad, Germany. The light
emission (photons) during the reactions was revealed by
the chemiluminescence detector as counts per second
(cps). Data refer to at least three individual experiments,
and appropriate controls were conducted throughout.
207
2.2. Preparation of cells
The epithelial cells used in this study were isolated from
the trachea of Wistar rats of both sexes (Charles River,
Calco, Italy). The rat was killed by exposure to CO2 for
20 s, and the trachea surgically removed and placed in
MEM (37 C) containing 400 U collagenase type IV/ml
for 10 min. Thereafter, the epithelium of the inner surface
of the trachea was carefully scraped off with a small lancet
and incubated for 30 min in the same MEM as above at
37 C. The resulting suspension was carefully filtered
through four layers of sterile gauze and centrifuged for
10 min at 3000 rpm. The cell pellet was then suspended in
phosphate buffer saline (PBS) and immediately used for
the experiments. The viability, measured with the trypan
blue exclusion test, was 90%.
Human neuroblastoma cells (SK.N.SH) were cultured in
a flask and maintained in minimum essential medium
(Eagle) with 2 mM L-glutamine and Earle’s BSS adjusted
to contain 1.5 g/L sodium bicarbonate, 0.1 mM non-essen-
tial amino acids, 1.0 mM sodium pyruvate, fetal bovine
serum 10% (Irvine Scientific, Santa Ana, CA), and 10%
fetal calf serum (Sigma, St. Louis, MO), and incubated at
37 C in a humidified incubator with 95% O2 and 5%
CO2. Trypsin was used to remove the cells from the flask;
finally, MEM was added to inactivate the trypsin. The cell
suspension was centrifuged 5 min at 3000 rpm, the surna-
tant was removed, and the cell pellet was suspended in
RPMI.
2.3. UVA irradiation
Rat epithelial cells (1 · 106) suspended in 1 ml of PBS,
or the same amount of human neuroblastoma cells sus-
pended in 1 ml of RPMI, isolated as described, were placed
in a 4 cm diameter Petri dish at room temperature; the
thickness of PBS or RPMI layer through which the cells
were exposed was 1 mm. The cells were exposed to UVA
irradiation (at 365 nm) by placing the dish 6 cm under a
Spectroline lamp (Spectronics Corporation Westbury,
New York, USA) at 2 W/m2 power. The dose of UVA irra-
diation was monitored with a Babuc/A (LSI) radiometer.
The UVA exposition time was chosen after preliminary
runs, so as to have a significant, notable degree of damage
that could be quantified using the Comet assay. The data
reported in the different figures are referred to different
UVA exposition time for the two cell types: 2 min for the
epithelial cells and 30 min for the neuroblastoma cells.
Hence the dose of UVA irradiation was 0.24 kJ/m2 for
the rat trachea epithelial cells and 3.6 kJ/m2 for the neuro-
blastoma cells.
2.4. Comet assay
The comet assay procedure used to measure the DNA
strand breaks in individual cells was essentially the same
as described previously [19]. Cells were suspended in
208 M. Santocono et al. / Journal of Photochemistry and Photobiology B: Biology 85 (2006) 205–215
0.7% low melting agarose in PBS and pipetted on micro-
scope slides pre-coated with a layer of 1% normal melting
agarose. The agarose with the cell suspension was allowed
to set on the pre-coated slides at 4 C for 10 min. Subse-
quently, another top layer of 0.7% low melting agarose
was added and allowed to set at 4 C for 10 min. The slides
were then immersed in lysis solution (1% sodium n-lauroyl-
sarcosinate, 2.5 M NaCl, 100 mM Na2EDTA, 10 mM Tris
HCl pH 10, 1% Triton X-100 and 10% DMSO) overnight
at 4 C in the dark in order to lyse the embedded cells
and to permit DNA unfolding. After incubation in lysis
solution, slides were exposed to alkaline buffer (1 mM
Na2EDTA, 300 mM NaOH buffer, pH > 13) for 20 min
to allow for DNA unwinding and then subjected to
20 min electrophoresis at 25 V (0.657 V/cm) in the same
alkaline buffer. After electrophoresis, the slides were
washed with 0.4 M Tris HCl buffer (pH 7.5) to neutralize
excess alkali and to remove detergents before staining with
ethidium bromide (2 lg/ml).
2.5. Evaluation of DNA damage
Cells were examined with an Axioskop 2 plus micro-
scope (Carl Zeiss, Germany) equipped with an excitation
filter of 515–560 nm and a magnification of ·20. Imaging
was performed using a specialised analysis system (‘‘Meta-
system’’ Altlussheim, Germany) to determine tail moment
(TM), a parameter correlated with the degree of DNA
damage in the single cell.
Data are presented as mean ± SEM. DNA single breaks
were analysed by one-way ANOVA (treatment) for the tail
moment parameter between group tests. The analyses were
done using Systat 10.0 [SPSS Inc. Chicago, IL].
Data refer to at least three individual experiments, three
slides/experiment, 450 scores/experiment. Differences were
considered significant when p values were <0.05 (*), <0.01
(**), <0.001 (***).
2.6. Evaluation of DNA damage and repair in the presence of
carotenoids
The comet assay experiments with epithelial cells were
conducted by irradiating 1 · 106 cells suspended in 1 ml
of PBS for 2 min with UVA, under three sets of
conditions:
(i) in the absence of carotenoids, observing DNA repair
for 3 h;
(ii) in the presence of 5, 50 and 100 lM of carotenoids;
(iii) by adding 50 lM of carotenoids immediately after
the irradiation and observing the DNA repair for
3 h. The kinetics of DNA repair was observed during
incubation of the cells in MEM + 20% FBS at 37 C.
After the UVA exposure, the epithelial cell viability was
checked using the trypan blue exclusion test, and found to
be 88.8%.
Human neuroblastoma cells (1 · 106) were suspended in
1 ml of RPMI and subjected to the following sets of
conditions:
(i) irradiated with UVA 15 or 30 min in the absence of
carotenoids;
(ii) irradiated with UVA 30 min and observing DNA
repair for 3 h;
(iii) irradiated with UVA 30 min in the presence of 40 lM
carotenoids;
(iv) irradiated with UVA 30 min adding 40 lM of carote-
noids immediately after the irradiation, and observ-
ing the DNA repair for 2 h.
After the exposure to UVA for 30 min, the viability of
the neuroblastoma cells was checked using the trypan blue
exclusion test, and found to be 98.5%.
The kinetics of DNA repair was observed during incu-
bation at 37 C of the cells resuspended in RPMI + 20%
fetal bovine serum (FBS).
2.7. Evaluation of DNA damage and repair in the presence of
hydrogen peroxide
Rat trachea epithelial cells and human neuroblastoma
cells at the concentration of 1 · 106 cells/ml were incubated
with H2O2 20 lM for 5 min in PBS. The viability of epithe-
lial cells and human neuroblastoma cells, checked by using
the trypan blue exclusion test after the treatment with
H2O2, was 98.5% and 89.8%, respectively.
After the treatment with H2O2 the epithelial cells were
washed with MEM + 20% fetal bovine serum (FBS) while
the neuroblastoma cells were washed with RPMI + 20%
fetal bovine serum (FBS). The DNA repair was observed
during incubation of the cells at 37 C in the same buffers.
3. Results
Chemiluminescence determinations were performed
using lucigenin or luminol as chemiluminogenic probes
for superoxide radical or hydrogen peroxide, respectively.
Fig. 1 shows the percentage of inhibition of lucigenin-
amplified CL signal produced by the reaction between this
probe and the superoxide anion generated by the xanthine/
xanthine oxidase system, in the presence of lutein,
astaxanthin and zeaxanthin. The same figure reports, as a
reference, the superoxide scavenging activity of the nitrox-
ide radical Tempo, a compound known to possess SOD
mimic activity [20]. All three compounds under study are
able to react with superoxide radicals; the effect of different
concentrations on the chemiluminescence measurements is
the same for all the compounds, and it is less pronounced
with respect to Tempo.
Tempo causes a significant inhibition in CL values: 50%
inhibition of CL peak was obtained using less than 20 lM
of Tempo. The presence of the same concentration of the
three carotenoids produces about 20% inhibition of the
 M. Santocono et al. / Journal of Photochemistry and Photobiology B: Biology 85 (2006) 205–215 209

100

90

80
inhibition of CL signal (%)
70

60

50

40

30

20

10

0
0 10 20 30 40 50 60 70 80 90 100
concentration (μM)

Fig. 1. Inhibition (%) of lucigenin-amplified chemiluminescence of the xanthine/xanthine oxidase system in the presence of (j) Tempo and the
carotenoids (r) lutein, (m) zeaxanthin and (d) astaxanthin. Chemiluminescence was measured in the presence of 0.9 U/ml xanthine oxidase and 150 lM
lucigenin; the reaction was initiated by injecting xanthine at a final concentration of 50 lM in Tris Æ HCl 50 mM pH 7.4 buffer. Inhibition values refer to
peak heights.

CL signal. Due to their poor solubility in the buffer used zeaxanthin and around 20 lM of other two compounds
for the assay, zeaxanthin and astaxanthin were tested only (Fig. 2).
at low concentrations. The comet assay was used to evaluate DNA oxidative
Chemiluminescence experiments performed using lumi- damage. The extent of DNA damage was quantified by
nol (a probe sensitive to hydrogen peroxide) demonstrated measuring the displacement of the genetic material between
the ability of all three compounds to reduce H2O2–luminol the cell nucleus (comet ‘‘head’’) and the resulting ‘‘tail’’.
chemiluminescence; this decreases in the following order: The alkaline comet assay revealed that UVA mediated
zeaxanthin > astaxanthin P lutein. Fifty percent inhibition DNA damage was present for rat trachea epithelial cells
of the CL signal was observed in presence of about 2 lM immediately after 2 min of exposure, as reported in Fig. 3

100

90

80
inhibition of CL signal (%)

70

60

50

40

30

20

10

0
0 5 10 15 20 25
concentration (μM)

Fig. 2. Inhibition (%) of the chemiluminescence signal generated by the luminol/H2O2 system. Chemiluminescence was measured in the presence of
105 M luminol and the desired amount of carotenoid dissolved in methylene chloride:ethanol 1:1. The reaction was initiated by injecting H2O2 at a final
concentration of 50 mM in methylene chloride:ethanol 1:1. Inhibition values refer to peak heights. (r) Lutein, (m) zeaxanthin and (d) astaxanthin.
210 M. Santocono et al. / Journal of Photochemistry and Photobiology B: Biology 85 (2006) 205–215

(0 min); tail moment, the parameter used as an index of
DNA damage and obtained by computerized image analy-
sis, increased significantly compared to the control. At
longer times of UVA irradiation (i.e. 30 0 ) it was not possi-
ble to evaluate the extent of DNA damage because the rate
of damage increased drastically. Besides we found that at a
constant dose, low intensity exposure with a long exposure
time does not change significantly the extent of DNA dam-
age (data not shown). Tail moment is a simple descriptor
calculated by computerized image analysis system consid-
ering both the length of DNA migration in the comet tail
and the percentage of nuclear material that migrated out
from the comet head to the comet tail. As a reference, we
reported in the same figure, the tail moment for cells
exposed for 5 min to 20 lM of hydrogen peroxide. As
can be observed, the exposure to H2O2 produces more
marked damage than 2 0 exposure to UVA. The DNA dam-
age in response to two different insults appeared to be a
rapid event. Fig. 3 also illustrates the kinetics of DNA
repair; after 3 h of incubation in an appropriate medium
consisting of MEM + 20% FBS, the level of DNA damage
for the cells exposed to UVA and to H2O2 was similar to
that of control cells.
Fig. 4 reports tail moments of epithelial cells exposed to
UVA (for 2 min) in presence of 0, 5, 50 and 100 lM of the
carotenoids under study. The protective effect of carote-
noids appears to be a function of their concentration,
and was quite similar for all three compounds used.
Fig. 5 shows the kinetics of DNA repair of epithelial
cells irradiated for 2 min with UVA and then incubated
in MEM + 20% FBS at 37 C in the presence of 50 lM
carotenoids. The addition of carotenoids after UVA expo-
sition does not seem to positively influence the kinetics of
DNA repair (a light protective effect was determined only
for lutein after 30 0 and for astaxanthin after 60 0 of incuba-
tion). On the contrary, the presence of zeaxanthin after 1 h
of incubation drastically increased the damage; this geno-
toxic effect was so extensive that DNA damage was not
measurable. After 180 0 of incubation, lutein also showed
a genotoxic effect, though less extensive than that of
zeaxanthin.
Parallel experiments were performed using human neu-
roblastoma cells (SK.N.SH). UVA-induced DNA damage
was produced at a later time after UVA exposure than in
rat epithelial cells (Fig. 6(a)). From Fig. 6(b) it is evident
that the induced damage was repaired following incubation
in RPMI + 20% FBS at 37 C. In addition, we report for
this kind of cells, as reference, the DNA damage induced
by treatment with 20 lM H2O2 for 5 min (Fig. 6(c)). The
same figure shows the capacity to repair the DNA damage
during incubation in RPMI + 20% FBS at 37 C.
Fig. 7(b) reports tail moments of SK.N.SH cells exposed
to UVA (for 30 min) in presence of 40 lM of the carote-
noids under study. As can be noted, the presence of carote-
noids not only fails to protect cells from DNA damage, but
it increases the damage. The presence of zeaxanthin pro-
duces greater damage than the other two carotenoids.
The same figure highlights the fact that, in absence of 30 0
UVA exposition, our compounds did not exert a genotoxic
effect.
Fig. 7(a) shows the kinetics of DNA repair of neuroblas-
toma cells irradiated for 30 min with UVA and then incu-
bated in RPMI + 20% FBS in absence and presence of
40 lM of the carotenoids. The presence of all three carote-
noids positively influences the repair of DNA damaged by
30 min UVA exposition only in the first incubation period

Fig. 3. Comet assay on rat trachea epithelial cells suspended in PBS and exposed for 2 min to UVA or 5 min to 20 lM H2O2 (0 0 ). After exposition, cells
were resuspended in appropriate medium consisting of MEM + 20% FBS and the kinetics of DNA repair was observed during incubation at 37 C at 30 0 ,
60 0 and 180 0 . *** p < 0.001, ** p < 0.01, * p < 0.05 compared with the control. h Control; cells + H2O2, P cells + UVA.
 M. Santocono et al. / Journal of Photochemistry and Photobiology B: Biology 85 (2006) 205–215 211

Fig. 4. Effect of different concentrations (5, 50, 100 lM) of zeaxanthin, lutein and astaxanthin on DNA of rat trachea epithelial cells. The ‘‘tail moment’’
parameter is reported using the Comet assay after 2 0 exposition to UVA in presence and in absence of carotenoids. *** p < 0.001 compared with UVA.
§§ p < 0.01, §§§ p < 0.001 compared with the control. h Control; j UVA; UVA + astaxanthin; P UVA + zeaxanthin; N UVA + lutein.

Fig. 5. Mean comet tail moments of rat trachea epithelial cells exposed for 2 min to UVA (0). After exposition, cells were resuspended in MEM + 20%
FBS and incubated at 37 C in the presence of 50 lM carotenoid. DNA repair was followed at 30 0 , 60 0 and 180 0 of incubation. § p < 0.05 compared with
the control. * p < 0.05, *** p < 0.001 compared with UVA at the same incubation time. h Control; j UVA; UVA + astaxanthin; P UVA + zeaxanthin;
N UVA + lutein.

(15 min). At longer incubation times, erratic behaviour was
observed, though not a genotoxic effect.
4. Discussion
Cellular phototoxicity induced by UVA radiation has
been reported in different cell lines [21–23]. UVA-
induced cellular alterations occur when ROS production
during photosensitization generates oxidative stress
[24].
Membrane leakage [25], DNA strand breaks [26], pyrim-
idine dimers and adducts [27] as well as damage to proteins
and enzymes [28] have been observed as consequences of
UVA exposure.
Researchers are devoting increased attention to the
molecular effects of UVA radiation on cellular DNA, par-
ticularly, the increased formation of reactive oxygen species
(ROS), which for the most part is mediated by yet to be
identified endogenous photosensitizers [29]. The source
and mode of exposure can influence the entity of oxidative
212 M. Santocono et al. / Journal of Photochemistry and Photobiology B: Biology 85 (2006) 205–215

a 12
***
10

Tail moment
6

4
**
2

0
basal contr15' UVA15' contr30' UVA30'

b 12

10
Tail moment

4
*** *** ***
2

0
0' 30' 60' 180'
incubation time (min)
c
12

10

8
***
Tail moment

§§§
6
***
4

2 *

0
Contr. 0' 30' 60' 180'
incubation time (min)

Fig. 6. (a) Mean comet tail moments of human neuroblastoma cells suspended in RPMI and exposed for 15 and 30 min to UVA. Control cells (without
exposition) were also included. *** p < 0.001 and ** p < 0.01 compared with the control. (b) Mean comet tail moments of human neuroblastoma cells
suspended in RPMI and exposed for 30 min to UVA (0 0 ) and during (30 0 , 60 0 and 180 0 ) incubation at 37 C in RPMI + 20% FBS. *** p < 0.001 compared
with 0 0 (cells after 30 min of exposition to UVA). (c) Mean comet tail moments of human neuroblastoma cells suspended in PBS and exposed 5 min to
20 lM H2O2 (0 0 ) and during (30 0 , 60 0 and 180 0 ) incubation at 37 C in RPMI + 20% FBS. *** p < 0.001; * p < 0.05 compared with control. §§§ p < 0.001
compared with 0 0 .

damage. Merwald et al. [30] found that human keratino-
cyte-derived cell lines exposed to fractional UVA exposure
at short intervals manifested increased degree of oxidative
damage as those exposed to prolonged radiation at
decreased intensity. For epithelial cells instead the oxida-
tive DNA damage, following exposition to the same dose
but reducing the UVA intensity, does not significantly
change. Longer intervals had the opposite effect. Because
UVA irradiation causes depletion of antioxidants and
oxidative stress, it is believed that supplementation with
non-enzymatic antioxidants may play a crucial role in ame-
liorating or indeed preventing photobiological damage
in vivo.
Only recently has the antioxidative activity of caroten-
oid molecules in biological systems been investigated and
exploited. Increased intake and serum levels of carotenoids
are associated with reduced risk of cardiovascular and ocu-
lar diseases [31,32]. Here, we discuss the superoxide and
hydrogen peroxide scavenging ability of lutein and zeaxan-
thin, carotenoids that are correlated with the function of
the human retina, and of astaxanthin, chosen because its
structure is very close to that of the other two molecules.
 M. Santocono et al. / Journal of Photochemistry and Photobiology B: Biology 85 (2006) 205–215 213

b
30
***
25
Tail Moment

20 *** ***

15
10

5
0
ea
ut
VA

a
l

t
ro

lu
st

ze
st

+l

+z
r+
nt

+a
a

VA

r+
r+
co

VA
nt
VA

nt
U
nt

co

U
co
co

Fig. 7. (a) Mean comet tail moments of human neuroblastoma cells suspended in RPMI and exposed for 30 min to UVA. After irradiation, the cells,
resuspended in RPMI + 20% FBS in the presence of 40 lM carotenoids, were incubated at 37 C. DNA repair was followed at 15 0 , 30 0 , 60 0 and 120 0 of
incubation. §§§ p < 0.001 compared with the control. ** p < 0.01, *** p < 0.001 compared with UVA at the same incubation time. h Control; j UVA;
UVA + astaxanthin; P UVA + zeaxanthin; N UVA + lutein. (b) Effect of 40 lM of astaxanthin, zeaxanthin and lutein on DNA of human
neuroblastoma cells suspended in RPMI exposed or not to UVA for 30 0 . *** p < 0.001 compared with UVA.

Information on antioxidant activities can be obtained
by using chemiluminogenic probes, which are now
increasingly used because they amplify the signal without
interfering with the reaction system. Different information
can be obtained, depending on the type of probe utilized
[33]. Lucigenin is a probe that is sensitive to the level of
the superoxide anion radical and permits determination
of the presence of O 2 in medium, while luminol is sensi-
tive to hydrogen peroxide. The chemiluminescence experi-
ments clearly indicate that the three compounds have
antioxidant activity versus superoxide anion and hydro-
gen peroxide. It is known that carotenoids are readily
and rapidly oxidized by a range of antioxidants. It has
been suggested that allylic hydrogen abstraction may be
a factor influencing carotenoid reactivity with free
radicals.
In the present study, we also examined the ability of
three different carotenoids to protect against UVA-induced
DNA damage in rat epithelial and in human neuroblas-
toma cells using the ‘‘comet assay’’. The comet assay has
been of particular advantage in assessing DNA damage
for various reasons: data are collected at the level of the
individual cells, only a few thousand cells are required,
and the assay is relatively sensitive and simple.
In both cell lines, the irradiation with UVA resulted in
time-dependent DNA damage. Exogenous H2O2 was also
able to induce DNA damage. The rate of DNA damage
induction was very different between the two cell lines [neu-
roblastoma cells (Fig. 6(a)) appear more resistant to oxida-
tive irradiation insult]. Although there is no experimental
evidence to explain this different behaviour, a possible
hypothesis is that the two cell types used contain different
concentrations of intracellular sensitizing substances or of
quenchers such as GSH and Vitamin E. Another factor
to be taken into consideration is the difference in efficacy
of the DNA repair systems.
214 M. Santocono et al. / Journal of Photochemistry and Photobiology B: Biology 85 (2006) 205–215
However, in general, UVA photons do not attack
DNA directly, but are adsorbed by intracellular compo-
nents such as riboflavin, porphyrins, nicotinamide and
certain membrane bound enzymes [34]. Protection of
UVA-mediated DNA damage by the three carotenoids
under study was present only for epithelial cells, and this
protection is nearly the same for all the compounds
(Fig. 4). On the other hand, a very different effect has
been observed for neuroblastoma cells (Fig. 7(b)), where
the presence of carotenoids during the 30 0 -irradiation
causes increased DNA damage.
A number of factors may influence the range of caroten-
oid behaviour from antioxidant to pro-oxidant. One could
be the interaction with ROS that forms a carotenoid per-
oxyl radical that is a potentially highly reactive specie.
There is increasing concern that carotenoids may also exert
pro-oxidant properties under certain conditions [11,12].
However, the role of antioxidants in UV-induced photo-
protection is complex, and the literature offers conflicting
evidence about a protective effect conferred by carotenoids,
in particular, against UVA-induced cellular alterations.
Maccarone et al. [35] recently reported that zeaxanthin
shows a pro-apoptotic effect in neuroblastoma cells,
although these cells are rather resistant to apoptosis, while
it is capable of preventing apoptosis in healthy cells. Fur-
thermore, we have shown that the UVA-induced DNA
damage is repaired upon incubation of both cell types in
appropriate buffer at 37 C. The addition of carotenoids
after UVA exposure influences the kinetics of DNA repair
in a very different manner; a genotoxic effect was also mea-
sured for epithelial cells. Since the chemical structure of
these carotenoids is extremely similar, their different behav-
iour is likely due to their physical interaction with the sys-
tem under study.
Data from previous reports on repair rates for UVA-
induced damage vary greatly [36–40].
In conclusion, it seems that the effectiveness of these
carotenoids as antioxidants depends upon a number of fac-
tors (e.g. concentration, cell type, cell status, timing of
insult exposure, location in the cell, interaction with other
antioxidants, etc.). This has been reported also for other
dietary compounds and several recent articles have
described the pro-oxidant activity of important phyto-
chemicals such as vitamin C, vitamin E and lipoic acid
[41–43].
5. Abbreviations
CL chemiluminescence
FBS fetal bovine serum
GSH glutathione reduced
MEM Dulbecco’s Modified Eagle medium
PBS phosphate buffer saline
ROS reactive oxygen species
SOD superoxide dismutase
AMD age-related macular degeneration
Acknowledgements
The authors are very grateful to SIFI SpA for its finan-
cial support.
References
[1] J. Cadet, E. Sage, T. Douki, Ultraviolet radiation-mediated damage
to cellular DNA, Mutat. Res. 571 (2005) 3–17.
[2] X. Zhang, B.S. Rosenstein, Y. Wang, M. Lebwohl, H. Wei,
Identification of possible reactive oxygen species involved in ultravi-
olet radiation-induced oxidative DNA damage, Free Radical Biol.
Med. 23 (1997) 980–985.
[3] J. Fuchs, M. Huflest, L. Rothfuss, D. Wilson, G. Caramo, L. Packer,
Acute effects of near ultraviolet light on the cutaneous antioxidant
defence system, Photochem. Photobiol. 50 (1989) 739–744.
[4] Y. Shindo, E. Witt, L. Pacher, Antioxidant defence mechanisms in
murine epidermis and dermis and their response to ultraviolet light, J.
Invest. Dermatol. 100 (1993) 260–265.
[5] A. Poswig, J. Wenk, P. Brenneisen, M. Wlaschek, C. Hommel, G.
Quel, K. Faisst, J. Dissemond, K. Briviba, T. Krieg, K. Scharffetter-
Kochanek, Adaptive antioxidant response of manganese-superoxide
dismutase following repetitive UVA irradiation, J. Invest. Dermatol.
112 (1999) 13–18.
[6] M.T. Leccia, M. Yaar, N. Allen, J.C. Beani, Solar simulated
irradiation modulates gene expression and activity of antioxidant
enzymes in cultured human dermal fibroblasts, Exp. Dermatol. 10
(2001) 272–279.
[7] B. Halliwell, J.M.C. Gutteridge, Insults on the skin, in: Free Radicals
in Biology and Medicine, Oxford University Press, New York, 1998,
pp. 530–537.
[8] A. Mortensen, L.H. Skibsted, T.G. Truscott, The interaction of
dietary carotenoids with radical species, Arch. Biochem. Biophys. 385
(1) (2001) 13–19.
[9] N.I. Krinsky, Actions of carotenoids in biological systems, Ann. Rev.
Nutr. 13 (1993) 561–587.
[10] R. Peto, R. Doll, J.D. Buckley, M.B. Sporn, Can dietary beta-
carotene materially reduce human cancer rates? Nature 290 (1981)
201–207.
[11] P. Palozza, Pro-oxidant actions of carotenoids in biological systems,
Nutr. Rev. 56 (1998) 257–265.
[12] A.J. Young, D. Philip, G.M. Lowe, Carotenoid antioxidant activity,
in: N. Krinsky, S.T. Mayne, H. Sies (Eds.), Carotenoids in Health
and Disease, Marcel Dekker Inc., New York, 2003.
[13] O. Sommerburg, J.E.E. Keunen, A.C. Bird, F.J.G.M. van Kiujk,
Fruits and vegetables that are sources for lutein and zeaxanthin; the
macular pigment in human eyes, Br. J. Ophtalmol. 82 (1998) 907–
910.
[14] Food and Nutrition Board, Institute of Medicine, Dietary Reference
Intakes for Vitamin C, Vitamin E, Selenium and Carotenoids,
National Academy Press, Washington, DC, 2000.
[15] J.A. Mares-Perlman, A.E. Millen, T.L. Ficek, S.E. Hankinson, The
body of evidence to support a protective role for lutein and
zeaxanthin in delaying chronic disease, Overview. J. Nutr. 132
(2002) 518S–524S.
[16] I. O’Connor, N.M. O’Brien, Modulation of UVA-induced oxidative
stress by b-carotene, lutein and astaxanthin in cultured fibroplasts, J.
Dermatol. Sci. 16 (1998) 226–230.
[17] A. Kantar, M. Oggiano, R. Gabbianelli, P.L. Giorgi, M. Biraghi,
Effect of imidazole salycilate on the respiratory burst of polymor-
phonuclear leucocytes, Curr. Ther. Res. 54 (1993) 1–7.
[18] N.P. Singh, M.T. McCoy, R.T. Tice, E.A. Schneider, Simple
technique for quantification of low levels of DNA damage in
individual cells, Exp. Cell. Res. 175 (1988) 184–191.
[19] M. Villarini, M. Moretti, E. Damiani, L. Greci, A.M. Santroni, D.
Fedeli, G. Falcioni, Detection of DNA damage in stressed trout
 M. Santocono et al. / Journal of Photochemistry and Photobiology B: Biology 85 (2006) 205–215
erythrocytes using the comet assay: protection by nitroxide radicals,
Free Radical Biol. Med. 24 (1998) 1310–1315.
[20] D.W. Laight, T.J. Andrews, A.I. Haj-Yehia, M.J. Carrier, E.E.
Anggard, Microassay of superoxide anion scavenging activity
in vitro, Environ. Pharmacol. Toxicol. 3 (1997) 65–68.
[21] S.M. De Leeuw, N.P.M. Smit, M. Van Veldhaven, E.M. Pennings, S.
Pavel, J.W.I.M. Simons, A.A. Chothorst, Melanin content of cultured
human melanocytes and UV-induced cytotoxicity, J. Photochem.
Photobiol. 61 (2001) 106–113.
[22] C. Alapetite, T. Wachter, E. Sage, E. Moustacchi, Use of the alkaline
comet assay to detect DNA repair deficiencies in human fibroblasts
exposed to UVC, UVB, UVA and gamma rays, Int. J. Radiat. Biol.
69 (1996) 359–369.
[23] L. Marrot, J.P. Belaidi, C. Chaubo, J.R. Meunier, P. Perez, C.
Agapakis-Causse, An in vitro strategy to evaluate the phototox-
icity of solar UV at the molecular and cellular level: application
to photo-protection assessment, Eur. J. Dermatol. 8 (1998) 403–
412.
[24] C. Didier, J.-P. Pouget, J. Cadet, A. Favier, J.C. Beani, M.-J.
Richard, Modulation of endogenous levels of thioredoxin in human
skin fibroblasts preventing damaging effect of ultraviolet A radiation,
Free Radical Biol. Med. 30 (2001) 537–546.
[25] B. Bose, S. Agarwal, S.N. Chatterjee, Membrane lipid peroxidation
by UV-A: mechanism and implications, Byotech. Appl. Biochem. 12
(2001) 557–561.
[26] N.S. Dissanayake, G.E. Greenoak, R.S. Mason, Effects of ultraviolet
irradiation on human skin- derived epidermal cells in vitro, J. Cell.
Physiol. 157 (1993) 119–127.
[27] C. Kielbassa, B. Epe, DNA damage induced by ultraviolet and visible
light and its wavelength dependence, Methods Enzymol. 319 (2000)
436–445.
[28] G.F. Vile, R.M. Tyrrel, UVA radiation-induced oxidative damage to
lipids and proteins in vitro and in human skinfibroblasts is dependent
on iron and singlet oxygen, Free Radical Biol. Med. 18 (1995) 721–
730.
[29] J. Cadet, M. Berger, T. Douki, B. Morin, S. Raoul, J.L. Ravanat, S.
Spinelli, Effects of UV and visible radiations on DNA – final base
damage, Biol. Chem. 378 (1997) 1275–1286.
[30] H. Merwald, G. Klosner, C. Kokesch, M. Der-Petrossian, H.
Honigsmann, F. Trautinger, UVA-induced oxidative damage and
cytotoxicity depend on the mode of exposure, J. Photochem.
Photobiol. B. Biol. 79 (2005) 197–207.
215
[31] K.F. Gey, Prospects for the prevention of free radical disease
regarding cancer and cardiovascular disease, Br. Med. Bull. 49 (1993)
679–699.
[32] P.F. Jacquesand, L.T. Chylack, Epidemiologic evidence of a role for
the antioxidant vitamins and carotenoids in cataract prevention, Am.
J. Clin. Nutr. 53 (1991) 352S–355S.
[33] M.E. Murphy, H. Sies, Visible-range low level chemiluminescence in
the biological system, Methods Enzymol. 186 (1990) 595–610.
[34] D.M. Carbonare, M.A. Pathak, Skin photosensitizing agents and the
role of reactive oxygen species in photoageing, J. Photochem.
Photobiol. B 14 (1992) 105–124.
[35] M. Maccarone, M. Bari, V. Gasperi, B. Demmig-Adams, The
photoreceptor zeaxanthin induces cell death in neuroblastoma cells,
Anticancer Res. 25 (2005) 3871–3876.
[36] M.E. Churchill, J.G. Peak, M.J. Peak, Correlation between cell
survival and DNA single-strand break repair proficiency in the
chinese hamster ovary cell lines AA8 and EM9 irradiated with 365 nm
ultraviolet-A radiation, Photochem. Photobiol. 53 (1991) 229–236.
[37] L. Roza, G.P. Van der Schans, P.H.M. Lohman, The induction and
repair of DNA damage and its influence on cell death in primary
human fibroblasts exposed to UVA or UVC radiation, Mutat. Res.
146 (1985) 89–98.
[38] J. Moan, H. Waksvik, T. Christiansen, DNA single-strand breaks and
sister chromatid exchanges induced by treatment with hematopor-
phyrin and light or by X-rays in human NHIK 3025 cells, Cancer Res.
40 (1980) 2915–2918.
[39] P.J. Mayer, M.O. Bradley, W.W. Nichols, Incomplete rejoining of
DNA double-strand breaks in unstimulated normal human lympho-
cytes, Mutat. Res. 166 (1986) 275–285.
[40] M.J. Pitout, W.K.A. Louw, H. Izatt, C.J. van Vuuren, N. Hugo, J.J.
van der Watt, A modified neutral sucrose centrifugation method for
rapid detection of DNA damaged by radiation and repair in human
lymphocytes, Int. J. Biochem. 14 (1982) 899–904.
[41] H. Moimi, l. Packer, N.E. Saris, Antioxidant and prooxidant
activities of a-lipoic acid and dihydrolipoic acid, Toxicol. Appl.
Pharmacol. 182 (2002) 229–234.
[42] P. Zhang, S.T. Omaye, DNA strand breaks and oxygen tension:
effects of b-carotene, a-tocopherol and ascorbic acid, Food Chem.
Toxicol. 39 (2001) 239–246.
[43] P. Zhang, S.T. Omaye, Antioxidant and prooxidant roles for b-
carotene, a-tocopherol and ascorbic acid in human lung cells,
Toxicol. In Vitro 15 (2001) 13–24.