Carcinogenesis vol.26 no.3 pp.657--664, 2005
doi:10.1093/carcin/bgi008
Modulation of light-induced skin tumors by N-acetylcysteine and/or
ascorbic acid in hairless mice
Francesco D’Agostini, Roumen M.Balansky1,
Anna Camoirano and Silvio De Flora2
Department of Health Sciences, University of Genoa, Via A.Pastore 1,
I-16132 Genoa, Italy
1
Permanent address: National Centre of Oncology, Str. Plovdivsko
pole 6, Sofia 1756, Bulgaria
2
To whom correspondence should be addressed
Email: [email protected]
The light emitted by halogen quartz bulbs contains a
broad spectrum of UV wavelengths, is strongly genotoxic
and is a potent inducer of skin tumors in hairless mice. By
using a UVC filter, this light mimics solar radiation and
induces a variety of genomic and transcriptional alterations in mouse skin. UV-related carcinogenesis involves
depletion of antioxidants and glutathione in skin cells.
On this basis, we evaluated modulation of carcinogenicity
of UVC-filtered halogen lamps in SKH-1 hairless mice by
the antioxidants N-acetyl-L-cysteine (NAC) and ascorbic
acid (AsA). Both agents were given in the drinking water,
either individually or in combination. The earliest skin
lesions were detected after 300 days’ exposure to light
and became confluent in a number of mice after 480
days. NAC administration prolonged the latency time by
90 days. Moreover, NAC considerably and significantly
decreased both incidence and multiplicity of lightinduced skin tumors, prevented the occurrence of malignant lesions (squamocellular carcinomas) and reduced
the tumor size. In contrast, AsA, which may behave as a
prooxidant rather than an antioxidant, increased the
multiplicity of total skin tumors, carcinomas in situ and
squamocellular carcinomas. Co-administration of NAC
with AsA significantly attenuated the negative effect of
AsA, presumably due to the ability of this thiol to maintain
a reduced environment. Therefore, in agreement with our
previous in vitro findings, oral NAC is able to attenuate the
detrimental effects of AsA.
Introduction
Solar radiation represents the most ubiquitous mutagen and
carcinogen in nature and has been estimated to account for
10% of human cancers (1). The carcinogenicity of sunlight is
due to several mechanisms, including genetic predisposition
factors, e.g. related to the phototype and DNA repair patterns,
formation of pyrimidine dimers and a variety of other DNA
photoproducts, generation of reactive oxygen species (ROS)
and UVB-related immunosuppression (2,3).
Abbreviations: AP-1, activator protein-1; AsA, ascorbic acid; COX-2,
cyclooxygenase 2; GSH, reduced glutathione; JNK, Jun N-terminal kinase;
NAC, N-acetyl-L-cysteine; ROS, reactive oxygen species.
Carcinogenesis vol.26 no.3 # Oxford University Press 2004; all rights reserved.
We found that the light emitted by halogen quartz bulbs, an
artificial illumination system to which millions of people are
exposed, is strongly genotoxic to both bacteria (4,5) and
human cultured cells (6) and is a potent inducer of skin tumors
in various strains of hairless mice (7,8). DNA damage and
carcinogenicity are due to the fact that these lamps emit a
broad spectrum of UV radiation, starting from the UVC region
(4--9). Neither genotoxic nor carcinogenic effects were produced when halogen quartz bulbs were covered with a common glass cover (4--8). This discovery prompted the
manufacturers to replace the traditional halogen lamps with
UV-block or UV-stop lamps, in which the quartz bulb is
‘doped’ in order to render it less permeable to UV wavelengths. In fact, the light emitted by these lamps failed to
produce either mutations in bacteria (10) or genotoxic effects
in human cultured cells (11).
Besides their possible impact on human health, halogen
lamps provide a suitable model for simulating solar radiation.
In fact, at variance with other experimental models using pulse
radiation, it is sufficient to expose the animals, typically hairless mice, every day for a predetermined number of hours fixed
with a timer, until the skin tumors become evident. Moreover,
by using a UVC filter, the radiation spectrum of halogen lamps
mimics the spectrum of solar radiation. Under these conditions
UVA- and UVB-related alterations could be detected in skin
cells of hairless mice, including biochemical and genomic
alterations (12), changes in multigene expression (13) and
skin tumors (our unpublished data).
Based on these premises, we designed a study aimed at
evaluating the efficacy of two chemopreventive agents having
antioxidant properties in modulating the yield of light-induced
skin tumors. We used, either individually or in combination,
the thiol N-acetylcysteine (NAC) and ascorbic acid (AsA,
vitamin C). NAC works extracellularly of itself and intracellularly as a precursor of L-cysteine and reduced glutathione
(GSH) through a variety of mechanisms, among which is
scavenging of ROS (14,15). AsA is a major water soluble
antioxidant present in cells and plasma and is a strong reducing
agent (16). These properties are particularly relevant to UVrelated skin carcinogenesis, due to the contribution of ROS
to this process and to the circumstances that cutaneous antioxidants are depleted in UV-exposed cells (17) and sensitivity
to UV is inversely related to the GSH content of cells (18). In
addition, combination of NAC with AsA has been proposed
due to the ability of NAC to counteract possible adverse effects
of AsA (19).
We show here that the two chemopreventive agents tested
display opposite modulations of UV carcinogenesis in
mouse skin. In fact, AsA further enhanced the multiplicity
of light-related tumors, while NAC efficiently decreased
their incidence, multiplicity and size and prevented the
occurrence of squamocellular carcinomas. In addition,
when given in combination, NAC attenuated the negative
effects of AsA.
657
F.D’Agostini et al.
Materials and methods
Mice
Female SKH-1 hairless mice (150), aged 4 weeks, were purchased from
Charles River (Calco, Italy). The mice were housed in MakrolonTM cages on
sawdust bedding, and maintained on standard rodent chow (MIL; Morini, San
Polo d’Enza, Italy) and tap water ad libitum. The housing conditions were as
follows: temperature 23 2 C, relative humidity 55%, 12 h day/night cycle.
The housing and treatment of mice were in accordance with our national and
institutional guidelines.
Exposure to light
One hundred and twenty mice were exposed to the light emitted by halogen
quartz bulbs (12 V, 50 W), incorporated into dichroic spotlights, which were
supplied by Leuci (File S.p.A., Lecco, Italy). This illumination system emits a
broad spectrum of visible light wavelengths and UVA, UVB and UVC radiation. Typical spectra and irradiance values are available in McKinlay et al.
(9). The bulbs, which were changed every 6 months, were covered with filters
cutting out UVC radiation (WG 280; Schott Optics Division, Mainz,
Germany). The distance from the back of the mice was regulated to yield an
illuminance level of 10 000 lux. Exposure was for 3 h/day, every day throughout the duration of the experiment.
Design of the study and treatment with chemopreventive agents
After acclimatization for 2 weeks, the mice were divided into seven groups.
Three groups, each of 10 animals, included mice that were not exposed to light
and were either untreated or treated with either NAC or AsA. The remaining
four groups, each of 30 animals, included light-exposed mice that were either
untreated or treated with either NAC or AsA or both agents. AsA was purchased from Sigma Chemical Co. (St Louis, MO) and NAC from Zambon
(Vicenza, Italy), in the form of a commercially available, lyophilized drug
preparation (Fluimucil). Both agents, either individually or in combination,
were dissolved in the drinking water to yield a calculated daily intake of 1 g/kg
body wt. The solutions were prepared every 2 days. Treatment with NAC and/
or AsA started 3 days before the first day of exposure to light and continued
throughout the duration of the experiment.
Survival, body weight gain, and skin tumor yield
The mice were inspected daily for survival and general appearance. At
monthly intervals they were individually weighed and scored for the presence,
morphology and size of skin lesions. The experiment was stopped after 480
days, when the lesions became confluent in a proportion of mice (see Results).
Representative skin lesions of various morphological appearance were excised,
fixed in neutral buffered formalin and embedded in paraffin. Tissue sections
were stained with hematoxylin and eosin. The histopathological analysis of
tumors followed the classification proposed by Gallagher et al. (20).
Statistical analysis
Data relative to survival and incidence of tumors were expressed in terms of
frequency and differences between groups were evaluated by x2 analysis. Data
relative to body weight and multiplicity were expressed in terms of means SE within the mice belonging to each experimental group. When the skin
lesions became confluent an arbitrary value of 15 lesions/mouse was given.
The overall differences between experimental groups regarding time-related
increases in body weight and multiplicity of skin lesions were evaluated by
ANOVA followed by the Bonferroni/Dunn post hoc test for evaluating differences at individual times. The differences between experimental groups
regarding the size of lesions were evaluated by Student’s t-test.
Results
Survival and body weights
Of the 150 mice under study, 138 (92.0%) were still alive
480 days after the start of exposure, when the experiment
was terminated. The survival rate ranged between a minimum
of 83.3% in light-exposed mice and a maximum of 96.7%
in light-exposed mice treated with both NAC and AsA, without
any significant difference related to treatment. The body
weights (mean SE) at the start of the experiment were
25.2 0.11 g in control mice, 24.9 0.18 g in light-exposed
mice, 25.6 0.25 g in light-exposed mice treated with NAC,
25.3 0.22 g in light-exposed mice treated with AsA and
25.5 0.22 g in light-exposed mice treated with both NAC
and AsA. At the end of the experiment the body weights were
658
36.4 0.15, 36.9 0.15, 36.8 0.13, 37.1 0.19 and 37.0 0.30 g, respectively. Thus, there was no significant difference
among the experimental groups in body weight gain.
Yield of skin tumors
Table I summarizes the data relative to the yield of skin
tumors, in terms of incidence and multiplicity, as related to
treatment of mice and to the time elapsed since the start of
exposure to light. No skin lesions were observed throughout
duration of the experiment in the 30 control mice that were not
exposed to light, 10 of which were untreated, 10 were treated
with NAC and 10 were treated with AsA. Therefore, these
mice were grouped together and served as negative controls.
The earliest light-related skin lesions were detectable, at a
monthly score, after 300 days’ exposure. In particular, lesions
were detected after 300 days in light-exposed mice, irrespective of treatment with AsA, after 330 days in light-exposed
mice treated with both NAC and AsA and after 390 days in
light-exposed mice treated with NAC. Thus, the oral administration of NAC delayed the appearance of light-induced skin
tumors by 3 months. The incidence and multiplicity of tumors
grew with time, until the lesions became confluent in a number
of mice. When the experiment was stopped, after 480 days, the
lesions were confluent in 7 of 28 surviving light-exposed mice,
22 of 28 light-exposed mice treated with AsA and 10 out of
29 light-exposed mice treated with both NAC and AsA. The
lesions did not become confluent in any of the light-exposed
mice treated with NAC.
As shown in Table I and Figure 1, administration of NAC
considerably and significantly decreased both incidence and
multiplicity of light-induced skin tumors. In contrast, administration of AsA increased the multiplicity of light-related
tumors, with significant overall differences between the two
time course--multiplicity curves (P 5 0.001) and at individual
times after 480 days (P 5 0.001). This negative effect was
attenuated by the combined administration of AsA and NAC.
In fact, there were significant differences between light þ
AsA and light þ NAC þ AsA regarding the overall time
course--multiplicity curves (P 5 0.001) and at individual
times after 390 days (P 5 0.05) and 480 days (P 5 0.001).
However, the combined treatment with NAC and AsA was
not as effective as NAC alone in decreasing the multiplicity
of light-induced tumors.
Histopathology and size of skin tumors
The light-induced skin lesions had various morphological
appearances, which were macroscopically categorized into
four main types (Figure 2). Intermediate lesions were also
detectable. Some of the type A lesions spontaneously
regressed, but in general there was a trend to a time-related
evolution into lesions of larger size and higher malignancy. At
the end of the experiment 10 lesions per morphological type
were subjected to histopathological analysis. As shown in
Figure 2, type A lesions had the appearance of small pink
warts and were microscopically classified as epidermal hyperplasias. Type B lesions were of a larger size, tended to be
reddish, were more or less protruding with a central excrescence and were microscopically classified as papillomas. Type
C lesions had the appearance of outgrowths, with a knobbly
surface, often hyperemic and with a keratinized surface. Most
of them were microscopically classified as carcinomas in situ,
but one keratoacanthoma-like tumor and an appendage/basal
tumor were also detected. Type D lesions were large,
0.17
0.33k
0.43h,j
0.56l
0.51e,l
0.78g,i,l
0
0
0.30
1.03
1.70
3.97
5.14
9.45
Mice not exposed to light, of which 10 were untreated, 10 were treated with NAC and 10 were treated with AsA.
Light emitted by halogen quartz bulbs (12 V, 50 W) covered with UVC filters, 3 h/day, at an illuminance level of 10 000 lux.
N-acetylcysteine in drinking water (1 g/kg body wt).
d
Ascorbic acid in drinking water (1 g/kg body wt).
e
P 5 0.05, fP 5 0.01 or gP 5 0.001 versus light, as assessed either by x2 analysis (incidence) or ANOVA followed by Bonferroni/Dunn post hoc test (multiplicity).
h
P 5 0.05 or iP 5 0.001 versus light þ AsA, as assessed by ANOVA followed by Bonferroni/Dunn post hoc test.
j
P 5 0.05, kP 5 0.01 or lP 5 0.001 versus light þ NAC, as assessed either by x2 analysis (incidence) or ANOVA followed by Bonferroni/Dunn post hoc test (multiplicity).
c
a
b
270
300
330
360
390
420
450
480
(0%)
(0%)
(0%)
(0%)
(0%)
(0%)
(0%)
(0%)
0/30
2/30
5/30
11/30
18/30
22/28
21/26
21/25
0
0
0
0
0
0
0
0
0/30
0/30
0/30
0/30
0/30
0/30
0/29
0/28
(0%)
(6.7%)
(16.7%)
(36.7%)
(60.0%)
(78.6%)
(80.8%)
(84.0%)
0
0.17
0.57
0.93
1.97
2.71
3.31
5.76
0.12
0.26
0.30
0.42
0.46
0.55
1.06
0/30
0/30
0/30
0/30
8/30
11/30
11/30
11/28
(0%)
(0%)
(0%)
(0%)
(26.7%)f
(36.7%)g
(36.7%)g
(39.3%)g
0
0
0
0
0.50
0.80
1.00
2.14
0.18
0.22e
0.26f
0.57g
(0%)
(10.0%)
(20.0%)
(56.7%)
(63.3%)
(76.7%)
(79.3%)
(89.3%)
0.23
0.42
0.40
0.57
0.57
0.63
0.92g
0/30
0/30
3/30
11/30
15/30
27/30
27/29
29/29
(0%)
(0%)
(10.0%)
(36.7%)l
(50.0%)j
(90.0%)l
(93.1%)l
(100.0%)l
0
0.40
1.10
1.83
3.30
4.50
6.34
11.57
0/30
3/30
6/30
17/30
19/30
23/30
23/29
25/28
Multiplicity
(mean SE)
Incidence
Multiplicity
(mean SE)
Incidence
Incidence
Multiplicity
(mean SE)
Incidence
Multiplicity
(mean SE)
Incidence
Multiplicity
(mean SE)
Lightb þ NACc þ AsAd
Lightb þ AsAd
Lightb þ NACc
Lightb
Controlsa
Time (days)
Table I. Yield of skin tumors in variously treated SKH-1 hairless mice at various times after the start of exposure to UV-containing light
Modulation of light-induced skin tumors
Fig. 1. Multiplicities (means SE) of skin tumors in variously treated mice,
as related to treatment and to the time elapsed since the start of
exposure to light.
irregularly shaped tumors and were often ulcerated. They were
microscopically classified as squamocellular carcinomas.
The morphology and histopathology of light-induced skin
lesions, when detectable, were not affected by treatment with
AsA and/or NAC, but treatment with the chemopreventive
agents affected the frequency of the different morphological
and histopathological types as well as their size. Table II shows
the yield of skin lesions of various types and their size 420
days after the start of exposure of mice to light, when the
lesions were not yet confluent. All four types of skin lesions
were detectable in light-exposed mice, including the occurrence of squamocellular carcinomas in 21.4% of mice. Treatment with NAC significantly decreased the incidence and
multiplicity of all types of light-induced lesions, and no squamocellular carcinoma was observed. Treatment with AsA
significantly increased the multiplicity of light-induced
carcinomas in situ and squamocellular carcinomas. As compared with AsA alone, treatment with both NAC and AsA
significantly increased the incidence and multiplicity of
light-induced epidermal hyperplasia but significantly
decreased the multiplicity of carcinomas in situ and the yield
of squamocellular carcinomas. Indeed, as in light-exposed
mice treated with NAC, the combined treatment with
NAC and AsA abolished the occurrence of squamocellular
carcinomas.
In parallel, both NAC and its combination with AsA significantly reduced the size of light-induced skin lesions, which
was not affected by AsA alone (Table II).
Discussion
The first result emerging from the present study is that exposure of hairless mice to the light of UVC-filtered halogen
lamps provides a suitable experimental model for assessing
the cutaneous carcinogenicity of UVA- and UVB-containing
659
F.D’Agostini et al.
Fig. 2. Macroscopic and histopathological appearance of the main skin lesions induced by light in hairless mice. (A) Epidermal hyperplasia; (B) papilloma;
(C) carcinoma in situ; (D) squamocellular carcinoma.
wavelengths mimicking sunlight exposure. A short daily
exposure (3 h) was chosen in order to allow a slower growth
of skin lesions and to better evaluate their modulation by
chemopreventive agents. Under these conditions the earliest
tumors appeared on the exposed mouse backs after 10 months
and progressively grew until they becam confluent in a proportion of mice after 16 months. The nature of light-induced
660
skin lesions ranged from preneoplastic lesions, such as epidermal hyperplasia, to benign tumors, such as papillomas, evolving towards keratoacanthoma-like tumors, appendage/basal
tumors, carcinomas in situ and squamocellular carcinomas.
This long-term effect parallels earlier alterations of intermediate biomarkers, detected after 4 weeks of exposure under
identical exposure conditions, excepting a longer daily
See Figure 2 for the morphological and histopathological appearance of tumors.
Mice not exposed to light, of which 10 were untreated, 10 were treated with NAC and 10 were treated with AsA.
c
Light emitted by halogen quartz bulbs (12 V, 50 W) covered with UVC filters, 3 h/day, at an illuminance level of 10 000 lux.
d
N-acetylcysteine in drinking water (1 g/kg body wt).
e
Ascorbic acid in drinking water (1 g/kg body wt).
f
Diameter in mm (mean SE).
g
P 5 0.05, hP 5 0.01 or iP 5 0.001 versus light, as assessed either by x2 analysis (incidence), ANOVA followed by Bonferroni/Dunn post hoc test (multiplicity) or Student’s t-test (size).
j
P 5 0.05, kP 5 0.01 or lP 5 0.001 versus light þ AsA, as assessed either by x2 analysis (incidence), ANOVA followed by Bonferroni/Dunn post hoc test (multiplicity) or Student’s t-test (size).
m
P 5 0.001 versus light þ NAC, as assessed by ANOVA followed by Bonferroni/Dunn post hoc test (multiplicity).
a
b
2.09 0.20i,l,m
1.29 0.27m
0.59 0.20k
0l
27/30 (90.0%)h,k,m
21/30 (70.0%)m
9/30 (30.0%)
0/30 (0%)h,j
2.23 0.14g,j
0.19
0.30g
0.84h
0.13h
0.95
1.77
1.23
0.55
12/30 (40.0%)
15/30 (50.0%)
12/30 (40.0%)
9/30 (30.0%)
4.73 0.69
0.33 0.09g
0.24 0.12h
0.22 0.07
0
(26.7%)g
(20.0%)g
(20.0%)g
(0%)h
0.42g
8/30
6/30
6/30
0/30
2.50
0.19
0.24
0.14
0.08
0.91
1.00
0.55
0.22
15/28 (53.6%)
15/28 (53.6%)
11/28 (39.3%)
6/28 (21.4%)
4.56 0.88
0
0
0
0
(0%)
(0%)
(0%)
(0%)
0/30
0/30
0/30
0/30
0
A
B
C
D
Size of lesionsf
Multiplicity
(mean SE)
Incidence
Incidence
Multiplicity
(mean SE)
Incidence
Multiplicity
(mean SE)
Incidence
Multiplicity
(mean SE)
Incidence
Multiplicity
(mean SE)
Lightc þ NACd þ AsAe
Lightc þ AsAe
Lightc þ NACd
Lightc
Controlsb
Type of lesiona
Table II. Yield of skin tumors of different morphology and histopathological type and size of lesions in variously treated SKH-1 hairless mice, 420 days after the start of exposure to UV-containing light
Modulation of light-induced skin tumors
exposure (9 h), in particular light-induced increases in lipid
peroxidation products, DNA photoproducts, 8-hydroxy-20 deoxyguanosine, P53 protein, proliferating cell nuclear antigen, cytogenetic damage and apoptosis (12). Moreover, 55 of
the 746 cancer-related genes (7.4%) analyzed were overexpressed in the skin of light-exposed mice, including oncogenes,
genes encoding for metalloproteinases and cytoskeleton
proteins and genes involved in xenobiotic metabolism, oxidative stress, the stress response, protein repair, DNA repair,
apoptosis, cell cycle regulation, signal transduction, inflammation and immune response (13).
The oral administration of NAC was quite efficient in inhibiting the formation of light-induced skin lesions and in totally
preventing the occurrence of squamocellular carcinomas. The
latency time of the earliest tumors was prolonged by 3 months
and both incidence and multiplicity of lesions as well as their
size were remarkably reduced. The dose tested (1 g/kg body wt)
has previously been shown to exert a variety of protective
effects in animal models (reviewed in 14,15). Although rather
high, this dose is not toxic, as shown not only by the unchanged
survival and body weight gains of the mice used in the present
study but also by the lack of alterations at the molecular level.
In fact, the same oral dose of NAC had a negligible influence
on the basal profiles of gene expression in organs of rodents,
including the lung of A/J mice (15), the liver of transplacentally exposed Swiss mouse fetuses (21) and both liver and lung
of Sprague--Dawley rats (A.Izzotti et al., unpublished data).
These experimental findings are consistent with the safety
of this molecule, as assessed in 40 years of extensive clinical
use (14,15).
NAC has previously been shown to inhibit the formation of
chemically induced skin tumors. In fact, administration of
NAC in the diet reduced the multiplicity of skin tumors resulting from the topical administration of benzo[a]pyrene in
cancer-prone p53 haploinsufficient Tg.AC (v-Ha-ras) transgenic mice (22). However, in this model NAC did not inhibit
the progression towards malignancy (22), while in the present
study no squamocellular carcinomas developed in lightexposed mice treated with NAC. The ability of NAC to inhibit
light-induced skin tumors is likely to be in part due to NAC
itself and in part to the generation of L-cysteine and GSH in
skin cells. In fact, GSH plays a central role in the endogenous
defense against UV radiation (23) and the formation of UVBinduced skin tumors in hairless mice has been shown to be
inhibited by administering GSH isopropyl ester prior to each
UV irradiation (24).
A major mechanism of NAC in UV-mediated skin carcinogenesis may be related to its ability to scavenge ROS, especially hydroxyl radicals generated in Fenton-type reactions
(25,26). Studies in cultured skin cells showed that NAC
inhibits the UVB-induced expression of cyclooxygenase-2
(COX-2) (27), which is up-regulated in the skin of lightexposed hairless mice (13). Moreover, NAC inhibited the
activation by UV radiation of factors involved in signal transduction, such as growth arrest and DNA damage-inducible
protein 45 (28), Jun N-terminal kinase (JNK), nuclear factor
kB and activator protein-1 (AP-1) (17). Note that activation of
AP-1 and overexpression of COX-2 are involved in UVBinduced promotion of skin tumors (29). Although in a number
of studies available in the literature NAC behaved as an
inhibitor of apoptosis, due to its ability to attenuate the DNA
damage triggering this process (14,15), in human C8161 melanoma cells NAC enhanced UV radiation-mediated apoptosis
661
F.D’Agostini et al.
(30). A further protective mechanism of NAC in UVB-related
skin carcinogenesis is the immunomodulating properties of
this thiol. In fact, independent of GSH synthesis, NAC inhibited UVB-induced immunosuppression in mice, consisting of
suppression of the T cell-mediated immune response, both
local and systemic (31). However, NAC had no effect on
UV-B-induced DNA damage, suggesting that NAC exerts its
photoprotective effect during the post-initiation phase of
photoimmunosuppression (32).
In contrast to the beneficial effect of NAC, in the present
study the oral administration of AsA, at the same dose, resulted
in a further increase in light-induced skin tumor multiplicity.
Previous studies had shown that the local application of AsA
protects animal skin against the harmful effects of UV radiation (33,34). Moreover, in cultured skin cells AsA mediated
cell damage and cell death by interfering at multiple levels
with the activity of the JNK/AP-1 pathway and by modulating
the expression of AP-1-related genes (35). However, our negative finding was not totally unexpected, since under certain
conditions AsA has already been shown to work as a prooxidant rather than as an antioxidant. Several studies provided
evidence that AsA may enhance both spontaneous and chemically induced mutations in prokaryotes and eukaryotes and
increase the yield of chemically induced tumors in rodents
(reviewed in 36,37). From a mechanistic point of view, AsA
has been proposed to act as a reducing agent for chelatecomplexed Fe3þ, thereby favoring production of hydroxyl
radicals from hydrogen peroxide through a Fenton-type reaction (38). Interestingly, UVA radiation mobilizes iron from
intracellular stores and labile iron acts as a catalyst to exacerbate the generation of secondary lipid messengers in skin cell
membranes (39). It has also been shown that AsA supplementation generates reactive aldehydes in vivo (40). In
cultured keratinocytes AsA 6-palmitate, a synthetic lipophilic
compound of AsA located at the cellular membrane, was
effective in scavenging intracellular ROS but strongly promoted UVB-induced lipid peroxidation, JNK activation and
cytotoxicity (41). Other studies showed that UV-induced AsA
oxidation in murine skin homogenates is caused by photoactivated reactions rather than by ROS reactions (42). Even in
humans, the dietary intake of AsA was inversely associated
with the UV-induced minimal erythema dose among 335
volunteers (43).
As to modulation by AsA of UV-induced skin tumors in
hairless mice, the results available in the literature are rather
controversial. When given topically on the dorsal surface 2 h
prior to each UVB irradiation, AsA (0.1 ml of a 5% solution)
slightly but significantly delayed tumor onset and decreased
the tumor yield. In parallel, topical AsA inhibited UVBinduced skin wrinkling. However, no inhibition of UVinduced skin wrinkling was afforded by oral AsA and topical
AsA failed to protect against UVA-induced skin sagging (33).
Linus Pauling’s papers (44,45) reported that supplementation
of the diet with AsA resulted in a pronounced decrease in the
incidence and delayed the onset of malignant lesions. Actually,
no multiplicity data were given and an analysis of the results
reported in these studies shows that an oral dose corresponding
to 0.55 g/kg body wt/day almost doubled the incidence of all
lesions 42 mm, while the incidence was decreased at the very
high doses of 8.3 and 16.7 g/kg body wt/day. A similar trend
was observed by Robinson et al. (46), who found that very
high doses of AsA in the diet, corresponding to 8, 16 and
32 g/kg body wt/day, decreased the multiplicity of
662
UV-induced skin tumors, while relatively lower doses (0.5, 1
and 2 g/kg body wt) increased it, with almost a doubling of
tumors at the lowest dose. Therefore, both the previous studies
(44--46) and the present study agree in the conclusion that at a
dose of 0.5--1.0 g/kg body wt/day, which would be quite high
in humans, AsA exacerbates UV-related skin carcinogenicity.
Only at exceedingly high doses (8--32 g/kg body wt/day) was
there a protective effect (44--46).
The combined administration of NAC and AsA attenuated
the adverse effect of the latter agent. In fact, compared with
treatment with AsA alone, co-administration of NAC delayed
the onset of tumors by 30 days, significantly attenuated skin
tumor multiplicity, abolished the occurrence of squamocellular
carcinomas and reduced the size of lesions. Previously we
showed that NAC antagonizes the enhancing effect of AsA
on ‘spontaneous’ mutagenicity in the Salmonella typhimurium
his strains TA102 and TA104, which are sensitive to oxidative mechanisms (19). Recycling of AsA from its oxidized form
is required to maintain intracellular stores of this vitamin (47)
and GSH has been shown, both in vitro and in vivo, to maintain
a reducing milieu in the cell which can reduce dehydroascorbic
acid (47--51). It is likely that NAC exerts similar functions both
of itself and as a precursor of intracellular L-cysteine and GSH.
There is evidence that NAC protects towards the adverse
effects not only of AsA (19, this study), but also of other
chemopreventive agents, such as green tea polyphenols (52)
and isothiocyanates (53), and attenuates the side-effects of
cytotoxic drugs, such as cyclophosphamide-induced urotoxicity (54) and doxorubicin-induced mutagenicity (55), clastogenicity (56), cardiotoxicity (57) and alopecia (56). NAC and
AsA have already been tested as a topical co-application on pig
skin and shown to protect against UVA-induced decomposition of the non-steroidal anti-inflammatory drug suprofen by
scavenging radicals and ROS (58). The results obtained in the
present study provide evidence that NAC can also attenuate the
detrimental effects of AsA when given by the oral route.
Acknowledgements
This study was supported by the Associazione Italiana per la Ricerca sul
Cancro (AIRC) and by the Bulgarian Ministry of Education and Science.
References
1. Higginson,J. and Muir,C.S. (1979) Environmental carcinogenesis: misconceptions and limitations to cancer control. J. Natl Cancer Inst., 63,
1291--1298.
2. IARC (1992) IARC Monographs on the Evaluation of the Carcinogenic
Risk to Humans, Vol 55, Solar and Ultraviolet Radiation. IARC, Lyon,
France.
3. Kripke,M.L. (1990) Effects of UV radiation on tumor immunity. J. Natl
Cancer Inst., 82, 1392--1396.
4. De Flora,S., Camoirano,A., Izzotti,A. and Bennicelli,C. (1990) Potent
genotoxicity of halogen lamps, compared to fluorescent light and sunlight.
Carcinogenesis, 11, 2171--2177.
5. De Flora,S., Camoirano,A., Izzotti,A. and Bennicelli,C. (1991) A bacterial
DNA-repair test evaluating the genotoxicity of light sources. Toxicol.
Methods, 1, 116--122.
6. D’Agostini,F., Izzotti,A. and De Flora,S. (1993) Induction of micronuclei
in cultured human lymphocytes exposed to quartz halogen lamps and its
prevention by glass covers. Mutagenesis, 8, 87--90.
7. De Flora,S. and D’Agostini,F. (1992) Halogen lamp carcinogenicity.
Nature, 356, 569.
8. D’Agostini,F. and De Flora,S. (1994) Potent carcinogenicity of uncovered
halogen lamps. Cancer Res., 54, 5081--5085.
9. McKinlay,A.F., Whillock,M.J. and Meulemans,C.C. (1989) Ultraviolet
radiation and blue-light emission from spotlights incorporating tungsten
Modulation of light-induced skin tumors
halogen lamps. In NRPB-R228. National Radiological Protection Board,
Chilton, UK, pp. 1--13.
10. Camoirano,A., Bennicelli,C., Bagnasco,M. and De Flora,S. (1999)
Genotoxic effects in bacteria of the light emitted by halogen tungsten
lamps having treated quartz bulbs. Mutat. Res., 441, 21--27.
11. D’Agostini,F., Caimo,A., De Filippi,S. and De Flora,S. (1999) Induction
and prevention of micronuclei and chromosomal aberrations in cultured
human lymphocytes exposed to the light of halogen lamps. Mutagenesis,
14, 433--436.
12. Balansky,R.M., Izzotti,A., D’Agostini,F., Camoirano,A., Bagnasco,M.,
Lubet,R.A. and De Flora,S. (2003) Systemic genotoxic effects produced
by light and synergism with cigarette smoke in the respiratory tract of
hairless mice. Carcinogenesis, 24, 1525--1532.
13. Izzotti,A., Cartiglia,C., Longobardi,M., Balansky,R.M., D’Agostini,F.,
Lubet,R.A. and De Flora,S. (2004) Alterations of gene expression in skin
and lung of mice exposed to light and cigarette smoke. FASEB J., 18,
1559--1561.
14. De Flora,S., Izzotti,A., D’Agostini,F. and Balansky,R.M. (2001)
Mechanisms of N-acetylcysteine in the prevention of DNA damage
and cancer, with special reference to smoking-related end-points.
Carcinogenesis, 22, 999--1013.
15. De Flora,S., Izzotti,A., Albini,A., D’Agostini,F., Bagnasco,M. and
Balansky,R.M. (2004) Antigenotoxic and cancer preventive mechanisms
of N-acetyl-L-cysteine. In Kelloff,G.J., Hawk,E.T. and Sigman,C.C. (eds)
Cancer Chemoprevention, Vol. 1, Promising Cancer Chemopreventive
Agents. Humana Press, Totowa, NJ, pp. 37--67.
16. Frei,B., England,L. and Ames,B.N. (1989) Ascorbate is an oustanding
antioxidant in human blood plasma. Proc. Natl Acad. Sci. USA, 86,
6377--6381.
17. Saliou,C., Kitazawa,M., McLaughlin,L. et al. (1999) Antioxidants
modulate acute solar ultraviolet radiation-induced NF-kappa-B
activation in a human keratinocyte cell line. Free Radic. Biol. Med., 26,
174--183.
18. Tyrrell,R.M. and Pidoux,M. (1988) Correlation between endogenous
glutathione content and sensitivity of cultured human skin cells to
radiation at defined wavelengths in the solar ultraviolet range. Photochem.
Photobiol., 47, 405--412.
19. D’Agostini,F., Balansky,R.M., Camoirano,A. and De Flora,S. (2000)
Interactions between N-acetylcysteine and ascorbic acid in modulating
mutagenesis and carcinogenesis. Int. J. Cancer, 88, 702--707.
20. Gallagher,C.H., Path,F.R.C., Canfield,P.J., Greenoak,G.E. and Reeve,V.E.
(1984) Characterization and histogenesis of tumors in the hairless mouse
produced by low-dosage incremental ultraviolet radiation. J. Invest.
Dermatol., 83, 169--174.
21. Izzotti,A., Balansky,R.M., Cartiglia,C., Camoirano,A., Longobardi,M. and
De Flora,S. (2003) Genomic and transcriptional alterations in mouse fetus
liver after transplacental exposure to cigarette smoke. FASEB J., 17,
1127--1129.
22. Martin,K.R., Trempus,C., Saulnier,M., Kari,F.W., Barrett,J.C. and
French,J.E. (2001) Dietary N-acetyl-L-cysteine modulates benzo[a]
pyrene-induced skin tumors in cancer-prone p53 haploinsufficient
Tg.AC (v-Ha-ras) mice. Carcinogenesis, 22, 1373--1378.
23. Steenvoorden,D.P.,
Hasselbaink,D.M.
and
Beijersbergen
van
Henegouwen,G.M. (1998) Protection against UV-induced reactive intermediates in human cells and mouse skin by glutathione precursors: a
comparison of N-acetylcysteine and glutathione ethylester. Photochem.
Photobiol., 67, 651--656.
24. Kobayashi,S., Takehana,M. and Tohyama,C. (1996) Glutathione isopropyl
ester reduces UVB-induced skin damage in hairless mice. Photochem.
Photobiol., 63, 106--110.
25. Aruoma,O.I., Halliwell,B., Hoey,B.M. and Butler,J. (1989) The antioxidant action of N-acetylcysteine. Free Radic. Biol. Med., 6, 593--597.
26. Izzotti,A., Orlando,M., Gasparini,L., Scatolini,L., Cartiglia,C.,
Tulimiero,L. and De Flora,S. (1998) In vitro inhibition by N-acetylcysteine of oxidative DNA modifications detected by 32P postlabeling. Free
Radic. Res., 28, 165--178.
27. Ashida,M., Bito,T., Budiyanto,A., Ichihashi,M. and Ueda,M. (2003)
Involvement of EGF receptor activation in the induction of
cyclooxygenase-2 in HaCaT keratinocytes after UVB. Exp. Dermatol.,
12, 445--452.
28. Wan,Y., Wang,Z., Shao,Y., Xu,Y., Voorhees,J. and Fisher,G. (2000) UVinduced expression of GADD45 is mediated by an oxidant sensitive
pathway in cultured human keratinocytes and human skin in vivo. Int. J.
Mol. Med., 6, 683--688.
29. Bowden,G.T. (2004) Prevention of non-melanoma skin cancer by
targeting ultraviolet-B-light signalling. Nature Rev. Cancer, 4, 23--35.
30. Rieber,M. and Rieber,M.S. (2003) N-acetylcysteine enhances UVmediated caspase-3 activation, fragmentation of E2F-4 and apoptosis in
human C8161 melanoma: inhibition by ectopic Bcl-2 expression.
Biochem. Pharmacol., 65, 1593--1601.
31. Steenvoorden,D.P. and Beijersbergen van Henegouwen,G.M. (1998)
Glutathione synthesis is not involved in protection by N-acetylcysteine
against UVB-induced immunosuppression in mice. Photochem.
Photobiol., 68, 97--100.
32. Nijmeijer,B.A.,
Steenvoorden,D.P.T.,
Beijersbergen
van
Henegouwen,G.M.J., Roza,L. and Vink,A.A. (1998) The mechanism of
N-acetylcysteine photoprotection is not related to dipyrimidine photoproducts. J. Photochem. Photobiol. B Biol., 44, 225--230.
33. Bissett,D.L., Chatterjee,R. and Hannon,D.P. (1990) Photoprotective effect
of superoxide-scavenging antioxidants against ultraviolet radiationinduced chronic skin damage in the hairless mouse. Photodermatol.
Photoimmunol. Photomed., 7, 56--62.
34. Darr,D., Combs,S., Dunston,S., Manning,T. and Pinnell,S. (1992) Topical
vitamin C protects porcine skin from ultraviolet radiation-induced damage.
Br. J. Dermatol., 127, 247--253.
35. Catani,M.V., Rossi,A., Costanzo,A., Sabatini,S., Levrero,M., Melino,G.
and Avigliano,L. (2001) Induction of gene expression via activator
protein-1 in the ascorbate protection against UV-induced damage.
Biochem. J., 15, 77--85.
36. Shamberger,R.J. (1984) Genetic toxicology of ascorbic acid. Mutat. Res.,
133, 135--159.
37. Lee,K.W., Lee,H.J., Suhr,Y.-J. and Lee,C.Y. (2003) Vitamin C and cancer
chemoprevention: reappraisal. Am. J. Clin. Nutr., 78, 1074--1078.
38. Benatti,U., Morelli,A., Guida,L. and De Flora,A. (1983) The production of
activated oxygen species by an interaction of methemoglobin with
ascorbate. Biochem. Biophys. Res. Commun., 11, 980--987.
39. Reelfs,O., Tyrrell,R.M. and Pourzand,C. (2004) Ultraviolet A radiationinduced immediate iron release is a key modulator of the activation
of NF-kappa B in human skin fibroblasts. J. Invest. Dermatol., 122,
1440--1447.
40. Cooke,M.S., Mistry,N., Ahmad,J. et al. (2003) Deoxycytidine glyoxal:
lesion induction and evidence of repair following vitamin C supplementation in vivo. Free Radic. Biol. Med., 34, 218--225.
41. Meves,A., Stock,S.N., Beyerle,A., Pittelkow,M.R. and Peus,D. (2002)
Vitamin C derivative ascorbyl palmitate promotes ultraviolet-B-induced
lipid peroxidation and cytotoxicity in keratocytes. J. Invest. Dermatol.,
119, 1103--1108.
42. Kitazawa,M., Podda,M., Thiele,J., Traber,M.G., Iwasaki,K., Sakamoto,K.
and Packer,L. (1997) Interactions between vitamin E homologues and
ascorbate free radicals in murine skin homogenates irradiated with
ultraviolet light. Photochem. Photobiol., 65, 355--365.
43. Broekmans,W.M., Vink,A.A., Boelsma,E., Klopping-Ketelaars,W.A.,
Tijburg,L.B., van’t Veer,P., van Poppel,G. and Kardinaal,A.F. (2003)
Determinants of skin sensitivity to solar irradiation. Eur. J. Clin. Nutr., 57,
1222--1229.
44. Dunham,W.B.,
Zuckerkandl,E.,
Reynolds,R.,
Willoughby,R.,
Marcuson,R., Barth,R. and Pauling,L. (1982) Effects of intake of
L-ascorbic acid on the incidence of dermal neoplasms induced in mice
by ultraviolet light. Proc. Natl Acad. Sci. USA, 79, 7532--7536.
45. Pauling,L., Willoughby,R., Reynolds,R., Blaisdell,B.E. and Lawson,S.
(1982) Incidence of sqamous cell carcinoma in hairless mice irradiated
with ultraviolet light in relation to intake of ascorbic acid (vitamin C) and
of D,L-alpha-tocopheryl acetate (vitamin E). Int. J. Vitam. Nutr. Res.
Suppl., 23, 53--82.
46. Robinson,A.B., Hunsberger,A. and Westall,F.C. (1994) Suppression of
squamous cell carcinoma in hairless mice by dietary nutrient variation.
Mech. Ageing Dev., 76, 201--214.
47. May,J.M., Cobb,C.E., Mendiratta,S., Hill,K.E. and Burk,R.F. (1998)
Reduction of the ascorbyl free radical to ascorbate by thioredoxin
reductase. J. Biol. Chem., 273, 23039--23045.
48. Meister,A. (1992) On the antioxidant effects of ascorbic acid and
glutathione (Commentary). Biochem. Pharmacol., 44, 1905--1915.
49. Meister,A. (1994) Glutathione, ascorbate and cellular protection. Cancer
Res., 54 (suppl.), 1969S--1975S.
50. Rose,R.C. and Bode,A.M. (1992) Tissue-mediated regeneration of
ascorbic acid: is the process enzymatic? Enzyme, 46, 196--203.
51. Winkler,B.S., Orselli,S.M. and Rex,T.S. (1994) The redox couple between
glutathione and ascorbic acid: a chemical and physiological perspective.
Free Radic. Biol. Med., 17, 339--349.
52. Kennedy,D.O., Matsumoto,M., Kojima,A. and Matsui-Yuasa,I. (1999)
Cellular thiols status and cell death in the effect of green tea polyphenols
in Ehrlich ascites tumor cells. Chem. Biol. Interact., 122, 59--71.
663
F.D’Agostini et al.
53. Jiao,D., Smith,T.J., Yang,C.S., Pittman,B., Deasi,D., Amin,S. and
Chung,F.-L. (1997) Chemopreventive activity of thiol conjugates
of isothiocyanates for lung tumorigenesis. Carcinogenesis, 18,
2143--2147.
54. Berrigan,M.J., Gurtoo,H.L., Sharma,S.D., Struck,R.F. and Martinello,A.J.
(1980) Protection by N-acetylcysteine of cyclophosphamide metabolism
related in vivo depression of mixed function oxygenase activity and
in vitro denaturation of cytochrome P-450. Biochem. Biophys. Res.
Commun., 93, 797--803.
55. De Flora,S., Camoirano,A., Izzotti,A., Zanacchi,P., Bagnasco,M. and
Cesarone,C.F. (1991) Antimutagenic and anticarcinogenic mechanisms
of aminothiols. In Nygaard,F. and Upton,A.C. (eds) Anticarcinogenesis
and Radiation Protection III. Plenum Press, New York, NY, pp. 275--285.
56. D’Agostini,F., Bagnasco,M., Giunciuglio,D., Albini,A. and De Flora,S.
(1998) Oral N-acetylcysteine inhibition of doxorubicin-induced
664
clastogenicity and alopecia: interaction between the two drugs in
preventing primary tumors and lung micrometastases in mice. Int. J.
Oncol., 13, 217--224.
57. Doroshow,J.H., Locker,J.Y., Ifrim,I. and Myers,C.E. (1981) Prevention
of doxorubicin cardiac toxicity in the mouse by N-acetylcysteine. J. Clin.
Invest., 68, 1053--1064.
58. Moison,R.M., Rijnkels,J.M., Podda,E., Righel,F., Tomasello,F., Caffieri,S.
and Beijersbergen van Henegouwen,G.M. (2003) Topically applied
vitamin C and cysteine derivatives protect against UVA-induced photodegradation of suprofen in ex vivo pig skin. Photochem. Photobiol., 77,
343--348.
Received October 19, 2004; revised December 14, 2004;
accepted December 15, 2004