Báo cáo khoa học: Ascorbic acid-pretreated quartz enhances cyclo-oxygenase-2 expression in RAW 264.7 murine macrophages pdf

Ascorbic acid-pretreated quartz enhances
cyclo-oxygenase-2 expression in RAW 264.7 murine
macrophages
Sonia Scarfı
`
1,2
, Umberto Benatti
2
, Marina Pozzolini
1,2
, Emanuela Clavarino
2
, Chiara Ferraris
1,2
,
Mirko Magnone
2
, Laura Valisano
3
and Marco Giovine
1,4
1 Advanced Biotechnology Center, Genoa, Italy
2 Department of Experimental Medicine, Section of Biochemistry, University of Genoa, Italy
3 Department for the Study of the Territory and its Resources, University of Genoa, Italy
4 Department of Biology, University of Genoa, Italy
Long-term exposure to quartz particles induces a
pathological process characterized by the development
of fibrotic nodules in the lung, due to the accumula-
tion of inflammatory cells, deposition of extracellular
matrix and cellular proliferation. The ensuing disease,
silicosis, is the result of a complex interaction of

molecular pathways, the molecular mechanisms of
which have been only partially elucidated [1].
The main difficulties arising from these studies are
due to the solid nature of quartz particles. Particulates
are intrinsically heterogeneous in dimension, shape and
composition. Furthermore, they never act as a constant
Keywords
ascorbic acid; inflammation; macrophages;
reactive oxygen species; silica
Correspondence
S. Scarfı
`
, Department of Experimental
Medicine, Section of Biochemistry,
University of Genoa, Viale Benedetto XV
n°1, 16132 Genoa, Italy
Fax: +39 010 354415
Tel: +39 010 3538151
E-mail: soniascarfi@unige.it
(Received 22 May 2006, revised 26 October
2006, accepted 30 October 2006)
doi:10.1111/j.1742-4658.2006.05564.x
Exposure to quartz particles induces a pathological process named silicosis.
Alveolar macrophages initiate the disease through their activation, which is
the origin of the later dysfunctions. Ascorbic acid is known to selectively
dissolve the quartz surface. During the reaction, ascorbic acid progressively
disappears and hydroxyl radicals are generated from the quartz surface.
These observations may be relevant to mammalian quartz toxicity, as sub-
stantial amounts of ascorbic acid are present in the lung epithelium. We
studied the inflammatory response of the murine macrophage cell line

RAW 264.7 incubated with ascorbic acid-treated quartz, through the
expression and activity of the enzyme cyclo-oxygenase-2 (COX-2). COX-2
expression and prostaglandin secretion were enhanced in cells incubated
with ascorbic acid-treated quartz. In contrast, no changes were observed in
cells incubated with Aerosil OX50, an amorphous form of silica. Quantifi-
cation of COX-2 mRNA showed a threefold increase in cells incubated
with ascorbic acid-treated quartz compared with controls. The transcription
factors, NF-jB, pCREB and AP-1, were all implicated in the increased
inflammatory response. Reactive oxygen species (H
2
O
2
and OH
•
) were
involved in COX-2 expression in this experimental model. Parallel experi-
ments performed on rat alveolar macrophages from bronchoalveolar lavage
confirmed the enhanced COX-2 expression and activity in the cells incuba-
ted with ascorbic acid-treated quartz compared with untreated quartz. In
conclusion, the selective interaction with, and modification of, quartz parti-
cles by ascorbic acid may be a crucial event determining the inflammatory
response of macrophages, which may subsequently develop into acute
inflammation, eventually leading to the chronic pulmonary disease silicosis.
Abbreviations
AA, ascorbic acid; BAL, bronchoalveolar lavage; COX-2, cyclo-oxygenase-2; EMSA, electrophoretic mobility-shift assay; GAPDH,
glyceraldehyde-3-phosphate dehydrogenase; IFN-c, interferon-c; PGE
2
, prostaglandin E
2
; ROS, reactive oxygen species.

60 FEBS Journal 274 (2007) 60–73 ª 2006 The Authors Journal compilation ª 2006 FEBS
entity, and their cytotoxic potential in a biological envi-
ronment depends on their mechanical, thermal and
chemical history as well as on the micromorphology at
the atomic level [2,3]. It is now generally accepted that
silicosis originates from inhalation of quartz particles,
which are subsequently incorporated by alveolar
macrophages. The cell’s inability to dissolve the crystal-
line particulate leads to the chronic inflammation
responsible for the development of the disease [4].
Ten years ago, Bavestrello et al . [5] reported that
ascorbic acid (AA) is able to partially dissolve the sur-
face of quartz, greatly increasing the concentration of
soluble silica in the surrounding medium. More
recently, Fenoglio et al. [6] demonstrated that during
this peculiar chemical reaction, while AA progressively
disappears, important modifications of the quartz sur-
face occur, leading to an increased production of free
hydroxyl radicals and H
2
O
2
. These findings are rele-
vant to mammalian quartz toxicity: by reacting with
AA, quartz could deprive the alveolar epithelium of
one of its most effective antioxidant defences, and the
surface modifications induced by AA increase the con-
centration of particle-derived reactive oxygen species
(ROS) in the alveolar space, which is one of the
mechanisms proposed for quartz fibrogenicity and

carcinogenicity [1]. Furthermore, AA-derived quartz
dissolution is specific for crystalline silica, as the amor-
phous silica particulate is not modified at its surface
by ascorbate treatment and does not produce hydroxyl
radicals [6].
These findings prompted us to further investigate the
cytotoxicity of AA-treated quartz particles in the murine
macrophage cell line RAW 264.7, a cell model widely
used for molecular studies on cell–particle interaction
[7,8]. In this study, we showed that AA-pretreated
quartz establishes a significantly higher cytotoxicity
compared with untreated quartz [9]. These results
suggested an active role for AA as a cofactor involved in
the early stages of quartz-induced pathology, and they
represent the basis of the present study on the effect of
AA on the quartz-induced inflammatory response in the
same cell model.
The inducible enzyme cyclo-oxygenase-2 (COX-2) is
one of the molecules principally involved in the imme-
diate cellular inflammatory response. It is encoded by
an immediate-early gene induced by various pro-
inflammatory agents, including endotoxin, cytokines,
mitogens and particulates. This enzyme has emerged as
primarily responsible for the synthesis of the prosta-
noids involved in acute and chronic inflammatory
states, and recently it has been documented that its
expression is also increased in cellular and animal
models after quartz exposure [7,10–12].
The aim of the present work was to evaluate the
production of COX-2 and prostaglandin E

2
(PGE
2
)in
RAW 264.7 cells incubated with AA-treated quartz
and to compare it with that induced by untreated
quartz particles and by a different polymorph of silica,
Aerosil OX50. This form of silica does not possess a
crystalline structure and has been demonstrated to be
unable to chemically interact with AA [6]. An assess-
ment of the major transcription factors (NF-jB,
CREB, AP-1) involved in COX-2 biosynthesis was per-
formed, and the relative amount of COX-2 mRNA
was quantified. The role of quartz-induced ROS in the
modulation of COX-2 expression was investigated.
Catalase, mannitol and desferrioxamine were used to
assess the importance of different ROS in triggering
the inflammatory response towards crystalline silica
pretreated or not with AA. Furthermore, experiments
performed on rat alveolar macrophages obtained from
bronchoalveolar lavage (BAL) confirmed the enhanced
COX-2 expression and activity in the cells incubated
with AA-pretreated quartz compared with cells chal-
lenged with untreated quartz.
These results suggest that the selective interaction of
AA with quartz could be a crucial event determining the
inflammatory response of macrophages, which is recog-
nized as the first necessary event eventually leading to
the quartz-induced chronic pulmonary disease silicosis.
Results

Quartz-induced COX-2 expression and PGE
2
production in RAW 264.7 cells
The inducible enzyme COX-2 is expressed in the early
stages of the inflammatory response and catalyses the
first step of the synthesis of PGE
2
, an important
inflammatory mediator. COX-2 and PGE
2
were quan-
tified in RAW 264.7 macrophages challenged with
quartz particles pretreated or not with AA.
RAW 264.7 cells were incubated with different con-
centrations of AA-treated Min-U-Sil 5 quartz or with
AA-treated Aerosil OX50 (a commercial amorphous
silica). COX-2 expression was evaluated, by western
blot analysis, after 6 h of treatment. Results were com-
pared with data obtained on cells incubated with
untreated particles and on untreated cells. Untreated
quartz suspensions, at concentrations of 15, 50 and
100 lgÆmL
)1
, induced COX-2 synthesis in RAW 264.7
cells (Fig. 1A, black bars) as well as AA-treated quartz
(Fig. 1A, striped bars). The statistical analysis of vari-
ance showed a significant difference in COX-2 synthe-
sis both between untreated and AA-treated quartz
particles (P<0.001) and among the different quartz
S. Scarfı

`
et al. Ascorbate-treated quartz enhances COX-2 expression
FEBS Journal 274 (2007) 60–73 ª 2006 The Authors Journal compilation ª 2006 FEBS 61
concentrations (P<0.001), and, in particular, cells
stimulated with AA-treated quartz showed higher
COX-2 expression than cells challenged with untreated
particles (Tukey test, P < 0.05).
Similar experiments were performed with Aerosil
OX50, which is not known to cause silicosis and does
not react with AA [6]. Exposure of RAW 264.7 cells to
the same concentrations of AA-treated (Fig. 1B, striped
bars) or untreated Aerosil OX50 particles (Fig. 1B,
black bars) stimulated the production of high amounts
of COX-2 with no significant difference between
AA-treated and untreated aerosil samples. However,
COX-2 expression in cells challenged with AA-treated
or untreated aerosil showed a significant, dose-dependent
COX-2 synthesis (analysis of variance, P < 0.001), with
values at 15 lgÆmL
)1
lower than those at 50 lgÆmL
)1
(Tukey test, P < 0.05) and values at 50 lgÆmL
)1
lower
than those at 100 lgÆmL
)1
(Tukey test, P < 0.05).
Aerosil particles have a surface area 10 times higher
than quartz particles. Thus, 100 lgÆmL

)1
quartz and
15 lgÆmL
)1
aerosil had a similar surface area (5.2 and
6.5 cm
2
, respectively; Fig. 1A,B) and induced a similar
increase in COX-2 expression. At 100 lgÆmL
)1
and a
surface area of 43 cm
2
ÆmL
)1
, aerosil particles induced
the highest increase in COX-2 expression in RAW cells
(Fig. 1B).
To quantify the enzymatic activity of COX-2 in our
experimental conditions, we also determined PGE
2
production in quartz-treated RAW 264.7 cells after 6
and 18 h of treatment. PGE
2
concentration in the
medium of cells incubated for 6 h (Fig. 2A, black bars)
or for 18 h (Fig. 2A, white bars) was significantly
increased by untreated quartz (analysis of variance,
P ¼ 0.000); furthermore, at 6 h, values measured at
100 lgÆmL

)1
were higher than values at 15 and
50 lgÆmL
)1
(Student–Newman–Keuls test, P < 0.05).
The relevant values were the following: at 6 h Q15
was 44.3 pgÆmL
)1
, Q50 was 54.1 pgÆmL
)1
and Q100
was 73.9 pgÆmL
)1
; at 18 h Q15 was 53.1 pgÆmL
)1
,
Q50 was 56.2 pgÆmL
)1
and Q100 was 75.6 pgÆmL
)1
.
Prostaglandin production in cell cultures challenged
with AA-treated particles (Fig. 2A, dotted bars) was
significantly different (1.3-fold to 1.65-fold higher)
from that measured in cultures stimulated with
untreated quartz at both 6 h (black, dotted bars) and
18 h (white, dotted bars) incubation (analysis of vari-
ance, P < 0.01). In this case, the PGE
2
concentration

in the medium of cells incubated with AA-treated
quartz increased in a significant, dose-dependent fash-
ion both at 6 and 18 h, with a significant difference
between increasing quartz concentrations (analysis
of variance, P ¼ 0.000, Student–Newman–Keuls test,
P < 0.05). The relevant values were: at 6 h, QA15 was
57.6 pgÆmL
)1
, QA50 was 79.7 pgÆmL
)1
and QA100
was 117.9 pgÆmL
)1
; at 18 h, QA15 was 72.9 pgÆmL
)1
,
QA50 was 92 pgÆmL
)1
and QA100 was 116.8 pgÆmL
)1
.
Murine macrophages were also challenged with AA-
pretreated or untreated quartz costimulated with
100 pgÆmL
)1
interferon-c (IFN-c) a cytokine released
by activated lymphocytes, mimicking in our model a
Fig. 1. COX-2 expression in quartz-treated and aerosil-treated cells.
(A) COX-2 expression in RAW 264.7 cells stimulated with AA-trea-
ted (striped bars) or untreated (black bars) Min-U-Sil quartz was

evaluated after 6 h of incubation by western blot analysis of total
cell lysates. Cells were challenged with untreated (q) or AA-treated
(qa) quartz particles at 15, 50 and 100 lgÆmL
)1
. Results are
expressed as density ratio between each COX-2 band and the cor-
responding b-actin band, relative to control. Values are the
mean ± SD from eight experiments. The asterisk indicates a statis-
tically significant difference between q and qa values (Tukey test,
P < 0.05). (B) COX-2 expression in RAW 264.7 cells stimulated
with AA-treated (striped bars) or untreated (black bars) Aerosil
OX50 silica particles was evaluated after 6 h of incubation by west-
ern blot analysis of total cell lysates. Cells were challenged with
untreated (a) or AA-treated (aa) aerosil particles at 15, 50 and
100 lgÆmL
)1
Results are expressed as density ratio between each
COX-2 band and the corresponding b-actin band, relative to control.
Values are the mean ± SD from eight experiments. The symbol #
indicates a significant increase in COX-2 at increasing aerosil
concentrations (15 versus 50 lgÆmL
)1
, P < 0,05; 50 versus
100 lgÆmL
)1
, P < 0,05, Tukey test).
Ascorbate-treated quartz enhances COX-2 expression S. Scarfı
`
et al.
62 FEBS Journal 274 (2007) 60–73 ª 2006 The Authors Journal compilation ª 2006 FEBS

later stage of inflammation, when these cells infiltrate
the lung tissue in large numbers and contribute to the
development of a chronic inflammatory state.
In the presence of IFN-c, PGE
2
release into the
medium was significantly increased by treatment of the
cells with quartz, AA-treated or untreated, by the par-
ticle concentration and by the incubation time
(Fig. 2B; analysis of variance, P < 0.001). Specifically,
costimulation of RAW 264.7 cells with untreated
quartz particles together with IFN-c for 6 h (black
bars) or 18 h (white bars) induced a greater increase in
PGE
2
production in all samples ranging between a 2.8-
fold and a 19.6-fold increase compared with the corres-
ponding values in the absence of IFN-c (Fig. 2A). The
relevant values of PGE
2
concentrations were the fol-
lowing: at 6 h, Q15 was 122.2 pgÆmL
)1
, Q50 was
149.6 pgÆmL
)1
and Q100 was 174.8 pgÆmL
)1
;at18h,
Q15 was 225.7 pgÆmL

)1
, Q50 was 598 pgÆmL
)1
and
Q100 was 1485 pgÆmL
)1
.
PGE
2
production by cells challenged with untreated
quartz was significantly increased at 18 h compared
with 6 h (Fig. 2B, analysis of variance, P < 0.001):
the increase in PGE
2
production at increasing concen-
trations (15–100 lgÆmL
)1
) was only significant (analy-
sis of variance, P ¼ 0.000) after 18 h of incubation
(Student–Newman–Keuls test, P < 0.05).
Furthermore, in the presence of IFN-c, PGE
2
release from cells stimulated with AA-treated Min-U-
Sil 5 quartz (Fig. 2B, dotted bars) was further signifi-
cantly increased, 1.4-fold to 4.8-fold, over that from
cells challenged with untreated quartz (analysis of vari-
ance, P < 0.001). The relevant values were: at 6 h,
QA15 was 256.5 pgÆmL
)1
, QA50 was 502.2 pgÆmL

)1
and QA100 was 836.3 pgÆmL
)1
; at 18 h, QA15 was
529.7 pgÆmL
)1
, QA50 was 1337 pgÆmL
)1
and QA100
was 2065 pgÆmL
)1
.
PGE
2
release from cells stimulated with AA-treated
quartz in the presence of IFN-c was affected by both
incubation time and particle concentration (Fig. 2B,
dotted bars, analysis of variance, P < 0.001); in parti-
cular, values recorded at 18 h were significantly higher
than those at 6 h (Tukey test, P < 0.05), and at both
time points a significant concentration-dependence of
the effect of AA-treated quartz was observed (Tukey
test, P < 0.05).
Summarizing, PGE
2
production and release by
RAW cells stimulated with both AA-treated and
untreated quartz occurred mainly during the first 6 h
and did not significantly increase during the subse-
quent 12 h (Fig. 2A). Conversely, in the presence of

IFN-c, both AA-treated and untreated quartz
particles induced a sustained PGE
2
release, leading
to a significantly higher PGE
2
concentration in the
Fig. 2. PGE
2
production in quartz-incubated cells. (A) PGE
2
con-
centration detected in RAW 264.7 cell supernatants after 6 h
(black bars) and 18 h (white bars) stimulation with quartz was
determined using a PGE
2
monoclonal EIA kit. Cells were chal-
lenged with untreated (q) or AA-treated (qa, dotted bars) quartz
particles at 15, 50 and 100 lgÆmL
)1
. Values are the mean ± SD
from eight experiments. The symbol § indicates a significant dif-
ference between the Q100 and both the Q15 and the Q50 val-
ues at 6 h (Student–Newman–Keuls test, P < 0.05). The asterisk
indicates a significant difference between Q and QA values, at
all particle concentrations and at both time points (analysis of
variance, P < 0.01). The symbol # indicates a significant differ-
ence between PGE
2
values at increasing QA concentrations, at

both time points (QA15 versus QA50 versus QA100, Student–
Newman–Keuls Test, P < 0.05). (B) PGE
2
concentration detected
in RAW 264.7 cell supernatants after 6 h (black bars) and 18 h
(white bars) stimulation with AA-treated (dotted bars) or
untreated quartz, in the presence of murine IFN-c (100 pgÆmL
)1
),
was determined using a PGE
2
monoclonal EIA kit. Values are the
mean ± SD from eight experiments. The symbol § indicates a
significant difference between Q15 versus Q50 versus Q100 val-
ues at 18 h (Student–Newman–Keuls test, P < 0.05). The aster-
isks (**) indicate a significant difference between Q values at
18 h versus Q values at 6 h (analysis of variance, P < 0.001).
The single asterisk indicates a significant difference between QA
and Q values at all particle concentrations and at both time
points (analysis of variance, P < 0.001). The symbol # indicates a
significant difference between the QA values at increasing con-
centrations, at both time points (Tukey test, P < 0.05).
S. Scarfı
`
et al. Ascorbate-treated quartz enhances COX-2 expression
FEBS Journal 274 (2007) 60–73 ª 2006 The Authors Journal compilation ª 2006 FEBS 63
medium at 18 h than at 6 h incubation (Fig. 2B).
Under the same conditions (particle concentration,
incubation time, presence or absence of IFN-c),
PGE

2
release from cells stimulated with AA-treated
quartz was always higher than that from cells chal-
lenged with untreated quartz.
COX-2 expression and activity were also evaluated
in primary cultures of rat alveolar macrophages iso-
lated from BAL of healthy animals and challenged for
6 h with or without 100 lgÆmL
)1
AA-treated or
untreated quartz. COX-2 expression (Fig. 3A) was sig-
nificantly increased by incubation of the cells with
untreated or AA-treated quartz (analysis of variance,
P ¼ 0.000), with increasing protein expression being
observed in control versus untreated versus AA-treated
quartz samples (Student–Newman–Keuls test, NT ver-
sus Q100 versus QA100, P < 0.05).
In line with this result, PGE
2
release (Fig. 3B) in the
culture medium of BAL macrophages was significantly
increased in AA-treated or untreated quartz samples
compared with control, untreated cells (Kruskal–Wallis
analysis of variance, 0.02 < P<0.05) with progres-
sively increasing PGE
2
concentrations being observed
in the media from control cells, cells stimulated with
untreated quartz, and cells challenged with AA-treated
quartz suspensions (multiple comparison test, NT ver-

sus Q100 versus QA100, P < 0.05). In particular,
PGE
2
release into the culture medium from cells incu-
bated with AA-treated quartz was 2.8-fold higher than
that measured in the supernatant from cells challenged
with untreated quartz (754.2 pgÆmL
)1
versus
272.9 pgÆmL
)1
).
Time-course of COX-2 mRNA synthesis in
RAW 264.7 macrophages
COX-2 mRNA synthesis was measured by quantitative
RT-PCR analysis in RAW 264.7 macrophages stimula-
ted with 100 lgÆmL
)1
AA-treated or untreated quartz.
COX-2 mRNA expression at the various incubation
times, normalized to the respective glyceraldehyde-3-
phosphate dehydrogenase (GAPDH) internal standard,
was compared with COX-2 expression at time zero
(Fig. 4). COX-2 transcription was significantly induced
30 and 60 min after cell exposure to both AA-treated
(striped bars) and untreated quartz (black bars), with
no significant difference observed for each stimulus
between the two time points. For both stimuli, COX-2
mRNA decreased below time-zero values 3 h after cell
exposure to the particles (Fig. 4). Thus, COX-2 mRNA

concentrations in cells challenged with untreated and
AA-treated quartz were statistically compared at 30
and 60 min of incubation. A significant difference
between samples challenged with untreated and
AA-treated quartz was observed (Scheirer–Ray–Hare
test, 0.001 < P<0.01), with AA-treated quartz
inducing a COX-2 transcription 2.5-fold higher than
that at time zero, and untreated quartz increased
COX-2 transcription 1.5-fold over time zero (mean of
values recorded at 30 and 60 min).
Fig. 3. COX-2 expression and PGE
2
production in quartz-incubated rat alveolar macrophages. (A) COX-2 expression in rat alveolar macro-
phages stimulated with AA-treated (striped bars) or untreated (black bars) Min-U-Sil 5 quartz was evaluated by western blot analysis of total
lysates after 6 h of incubation. The final concentration of quartz particles was 100 lgÆmL
)1
(Q100, QA100 untreated and treated). Results
are expressed as the density ratio between each COX-2 band and the corresponding b-actin band relative to control. Values are the
mean ± SD from three experiments. The asterisk indicates a significant increase in COX-2 expression in QA100 versus Q100 versus NT
samples (Student–Newman–Keuls test, P < 0.05). (B) PGE
2
concentration was detected in rat alveolar macrophage supernatants after 6 h of
stimulation with 100 lgÆmL
)1
untreated (black bars) or AA-treated quartz (striped bars) using a PGE
2
monoclonal EIA kit. Values are the
mean ± SD from six experiments. The symbol § indicates a significant increase in PGE
2
production in QA100 versus Q100 versus NT

samples (multiple comparison test, P < 0.05).
Ascorbate-treated quartz enhances COX-2 expression S. Scarfı
`
et al.
64 FEBS Journal 274 (2007) 60–73 ª 2006 The Authors Journal compilation ª 2006 FEBS
Nuclear translocation of NF-jB, AP-1 and pCREB
in RAW 264.7 macrophages
The nuclear translocation of the transcription factors
known to be responsible for COX-2 synthesis triggered
by inflammatory stimuli, i.e. NF-jB, AP-1 and
pCREB [13], was assessed by electrophoretic mobility-
shift assay (EMSA) on RAW 264.7 macrophages sti-
mulated with different concentrations of AA-treated or
untreated quartz for 30 min. The results of these
experiments are shown in Fig. 5.
Nuclear translocation of NF-jB (Fig. 5A) in
RAW 264.7 was significantly higher in cells incubated
with both AA-treated (striped bars) and untreated
(black bars) quartz at increasing particle concen-
tration (analysis of variance, P < 0.05). AA-treated
quartz suspensions at 15, 50 and 100 lgÆmL
)1
induced a concentration-dependent increase in NF-jB
translocation (analysis of variance, P < 0.001; Tukey
test, 15 versus 50 versus 100 lgÆmL
)1
, P < 0.05),
with 100 lgÆmL
)1
inducing a higher NF-jB trans-

location than untreated quartz (t-test, P < 0.01). In
untreated quartz suspensions, the dose-dependence of
the effect was significant only between 15 and
50 lgÆmL
)1
(analysis of variance, P < 0.05; Tukey test,
P < 0.05).
pCREB nuclear translocation (Fig. 5B) was also
significantly increased by incubation of the cells
with AA-treated and untreated quartz, at all particle
concentrations (analysis of variance, P < 0.001). In
both AA-treated and untreated quartz samples, pCREB
density values at 100 lgÆmL
)1
particle concentration
Fig. 5. NF-jB, pCREB and AP-1 nuclear translocation in quartz-trea-
ted RAW 264.7 cells. EMSA analyses of NF-jB (A), pCREB (B) and
AP-1 (C) were performed on nuclear extracts of RAW 264.7 cells
stimulated for 30 min with 15, 50 and 100 lgÆmL
)1
of AA-treated
(QA, striped bars) or untreated quartz (Q, black bars). Bars indicate
transcription factor band density relative to control, untreated cells
and are the mean ± SD from four experiments. (A) The asterisk
indicates a significant difference between Q and QA samples ver-
sus control (analysis of variance, P < 0.05). The symbol § indicates
a significant difference between QA100 and Q100 values (t-test,
P < 0.01). (B) The asterisk indicates a significant difference
between Q and QA samples versus control (analysis of variance,
P < 0.001). The symbol § indicates a significant difference between

QA100 and Q100 values (t-test, P < 0.01). (C) The asterisk indi-
cates a significant difference between Q and QA samples versus
control (analysis of variance, P < 0.05). The symbol § indicates a
significant difference between QA15 and Q15 values (t-test,
P < 0.01).
Fig. 4. RT-PCR of COX-2 mRNA in quartz-treated RAW 264.7 cells.
COX-2 mRNA transcription was monitored in RAW 264.7 macro-
phages by RT-PCR analysis from 30 min to 18 h after cell stimula-
tion with 100 lgÆmL
)1
AA-treated (striped bars) or untreated quartz
(black bars). Results are the mean of three independent experi-
ments performed in triplicate and expressed as COX-2 mRNA syn-
thesis normalized to the GAPDH transcription, relative to control
cells at time zero. At 30 and 60 min of incubation, a significant
difference (*) between samples challenged with untreated
and AA-treated quartz was observed (Scheirer–Ray–Hare test,
0.001 < P<0.01).
S. Scarfı
`
et al. Ascorbate-treated quartz enhances COX-2 expression
FEBS Journal 274 (2007) 60–73 ª 2006 The Authors Journal compilation ª 2006 FEBS 65
were significantly higher than at 15 and 50 lgÆmL
)1
,
but, in the case of untreated quartz, the difference
between the effect of 50 and 100 lgÆmL
)1
was lower
(analysis of variance, P < 0.05; Tukey test, P < 0.05)

than the one measured for the same concentrations of
AA-treated quartz (analysis of variance, P < 0.001;
Tukey test, P < 0.05). Besides, at 100 lgÆmL
)1
,
AA-treated quartz induced a higher pCREB trans-
location than untreated quartz (t-test, P < 0.01).
Finally, nuclear translocation of AP-1 (Fig. 5C) was
also significantly increased in cells incubated with AA-
treated or untreated quartz, at all particle concentra-
tions (analysis of variance, P < 0.05). For both the
AA-treated and untreated quartz suspensions, the
highest effect was observed at 15 lgÆmL
)1
(analysis of
variance, P < 0.05; Tukey test, P < 0.05, for both
series of data), with AA-treated quartz inducing a
higher AP-1 translocation than untreated quartz
(t-test, P < 0.01).
Summarizing, at 100 lgÆmL
)1
particle concentration,
translocation of NF-jB (Fig. 5A) and pCREB
(Fig. 5B) was significantly higher in cells stimulated
with AA-treated quartz (striped bars) than with
untreated quartz (black bars), whereas no significant
difference was observed at lower particle concentra-
tions; conversely, AP-1 (Fig. 5C) was more abundant
in the nuclei of cells stimulated with AA-treated quartz
compared with untreated quartz only at the lowest

quartz concentration (15 lgÆmL
)1
).
Role of oxygen radicals in quartz-induced COX-2
synthesis
In preliminary experiments, ROS production by
quartz-stimulated RAW cells was evaluated with a
ROS-specific fluorescent probe in the presence or
absence of the radical scavengers catalase, mannitol
and desferrioxamine, which remove hydrogen per-
oxide, hydroxyl radicals and iron, respectively (the
latter being required in the Fenton reaction, which
converts H
2
O
2
into OH
•
). AA-treated quartz at
100 lgÆmL
)1
induced a threefold increase in ROS
generation compared with the same concentrations of
untreated quartz, after 1 h incubation (not shown).
Desferrioxamine did not affect ROS generation by
either untreated or AA-treated quartz. Conversely,
mannitol completely quenched ROS production trig-
gered by untreated quartz, whereas it reduced the
probe’s fluorescence in the presence of AA-treated
quartz only by  24% (not shown). Unfortunately,

the effect of catalase could not be tested, because of
severe interference of the enzyme with the probe’s
fluorescence.
The increased ROS generation induced by AA-trea-
ted versus untreated quartz on RAW cells prompted
us to explore the effect of ROS scavengers on the
increased COX-2 expression triggered by AA-treated
quartz in RAW 264.7 macrophages. Cells were incuba-
ted for 6 h with AA-treated and untreated quartz in
the presence of excess of the radical scavengers, and
COX-2 expression was analyzed by western blot. Pre-
liminary experiments had demonstrated no effect of
the radical scavengers themselves on COX-2 synthesis
in RAW cells (not shown).
At all quartz concentrations tested, COX-2 expres-
sion in RAW 264.7 cells was increased by AA-treated
or untreated quartz and was affected by the presence
of the three scavengers (Fig. 6A, analysis of variance
P < 0.005; Fig. 6B, analysis of variance P < 0.001;
Fig. 6C, analysis of variance P < 0.01).
At 15 lgÆmL
)1
particle concentration (Fig. 6A), no
significant difference in the COX-2 synthesis triggered
by untreated quartz was observed in the absence (Q)
or presence of the various scavengers (Qc, Qm, Qd).
Conversely, the COX-2 expression stimulated by AA-
treated quartz was significantly inhibited by the pres-
ence of scavengers (analysis of variance, P < 0.001).
Indeed, the COX-2 density values in the presence of

catalase (QAc) and mannitol (QAm) were significantly
lower than the values in the absence of scavengers
(QA) or in the presence of desferrioxamine (QAd)
(Dunnett test, P < 0.05).
At 50 lgÆmL
)1
particle concentration (Fig. 6B), no
significant difference in the COX-2 synthesis stimulated
by AA-treated quartz was observed in the absence
(QA) or presence of the various scavengers (QAc,
QAm, QAd), whereas COX-2 expression triggered
by untreated quartz (Q) was significantly inhibited by
the presence of scavengers (analysis of variance,
P < 0.001). In fact, COX-2 density values in the pre-
sence of catalase (Qc) and mannitol (Qm) were signifi-
cantly lower than in the absence of scavengers (Q) or
in the presence of desferrioxamine (Qd) (Dunnett test,
P < 0.05).
Also at 100 lgÆmL
)1
particle concentration
(Fig. 6C), no significant difference in the COX-2 syn-
thesis stimulated by AA-treated quartz was observed
in the absence (QA) or presence of the various scav-
engers (QAc, QAm, QAd). Conversely, in untreated
quartz samples, a significant difference was observed
in the absence or presence of scavengers (Q, Qc, Qm,
Qd, analysis of variance, P < 0.005), although the
only significant reduction in COX-2 expression rela-
tive to the quartz-treated samples (Q) was observed

in the presence of catalase (Qc) (Dunnett test,
P < 0.05).
Ascorbate-treated quartz enhances COX-2 expression S. Scarfı
`
et al.
66 FEBS Journal 274 (2007) 60–73 ª 2006 The Authors Journal compilation ª 2006 FEBS
Summarizing, catalase reduced COX-2 expression in
cells stimulated with AA-treated quartz (QA, striped
bars) at the lowest particle concentration (15 lgÆmL
)1
,
Fig. 6A), while at higher particle concentrations (50–
100 lgÆmL
)1
) it reduced COX-2 expression only in
cells stimulated with untreated quartz (Fig. 6B,C).
Mannitol reduced COX-2 expression in cells incubated
with untreated quartz only at 50 lgÆmL
)1
particle
concentration (Fig. 6B), while in cells stimulated with
AA-treated quartz it was effective only at the lowest
particle concentration (15 lgÆmL
)1
, F ig. 6A). D esferri-
oxamine was always without effect. The failure of
desferrioxamine to inhibit quartz-induced COX-2
expression apparently rules out the involvement of a
Fenton reaction in mediating quartz effects on COX-2
expression.

These data confirm results obtained by others indi-
cating that oxygen radicals play an important role in
mediating quartz-induced COX-2 expression in macro-
phages, and indicate that their production is strictly
related to particle concentration [14]. Furthermore, the
above results indicate that ROS generation is higher in
cells challenged with AA-treated compared with
untreated quartz and that, at the lowest particle con-
centration, ROS scavengers are able to prevent COX-2
synthesis, whereas at higher particle concentrations
Fig. 6. COX-2 expression in quartz-stimula-
ted RAW 264.7 cells in the presence of
ROS scavengers. COX-2 was analyzed by
western blot in RAW 264.7 cells after 6 h
incubation with different concentrations of
AA-treated (QA, striped bars) or untreated
quartz (Q, black bars) in the presence of
4000 UÆmL
)1
catalase (c), 50 mM mannitol
(m) and 2 m
M desferrioxamine (d). Bars indi-
cate the COX-2 band density, normalized on
the corresponding b-actin bands, relative to
the density of the COX-2 band in cells sti-
mulated with untreated quartz (Q). Values
are the mean ± SD from four experiments.
(A) The symbol § indicates a significant dif-
ference between Qc, Qm, Qd, QA, QAm,
QAc, QAd values versus the Q value (analy-

sis of variance, P < 0.005). COX-2 expres-
sion in QA samples was significantly
inhibited (*) by the presence of catalase and
mannitol (QAc and QAm versus QA, analy-
sis of variance, P < 0.001, Dunnett test,
P < 0.05). (B) The symbol § indicates a sig-
nificant difference between Qc, Qm, Qd,
QA, QAm, QAc, QAd values versus the Q
value (analysis of variance, P < 0.001).
COX-2 expression in Q samples was signi-
ficantly inhibited (*) by the presence of
catalase and mannitol (Qc and Qm versus
Q, analysis of variance, P < 0.001, Dunnett
test, P < 0.05). (C) The symbol § indicates a
significant difference between Qc, Qm, Qd,
QA, QAm, QAc, QAd values versus the Q
value (analysis of variance, P < 0.01). COX-2
expression in Q samples was significantly
inhibited (*) by the presence of catalase (Qc
versus Q, analysis of variance, P < 0.005,
Dunnett test, P < 0.05).
S. Scarfı
`
et al. Ascorbate-treated quartz enhances COX-2 expression
FEBS Journal 274 (2007) 60–73 ª 2006 The Authors Journal compilation ª 2006 FEBS 67
protein production is apparently triggered by other
signals.
Discussion
Quartz toxicity towards mammalian cells is well
known in the field of occupational health research,

although the molecular mechanisms of quartz-induced
cell tissue damage are not yet completely understood.
Many experimental models have been investigated to
address the complex interactions between the heteroge-
neous lung tissue cell population and quartz particles.
One of the most consolidated pieces of evidence from
these studies is the strict dependence of cell reactivity
on the properties of particle surface.
The presence of AA in the lung fluids has been dem-
onstrated, and its possible involvement in the develop-
ment of quartz-induced lung injury had already been
hypothesized some years ago [15–17]. Ghio et al. [16]
observed a remarkable increase in the accumulation of
inflammatory cells in the lung fluids of guinea pigs fed
an AA-rich diet and exposed to Min-U-Sil 5. Con-
versely, control animals fed low doses of AA and
exposed to the same amount of quartz showed a
reduced proliferation of inflammatory cells.
In connection with this, we have recently demonstra-
ted that specific chemical modifications occurring on
the quartz surface after exposure to AA cause
increased quartz cytotoxicity in the murine macro-
phage cell line RAW 264.7 [9]. As cytotoxicity is only
one of the many aspects of macrophage reactivity to
quartz, we here investigated the inflammatory response
of RAW 264.7 macrophages, a cell model widely used
for in vitro studies of the biological response to quartz
particles and of primary rat alveolar macrophages to
AA-treated quartz [7,8,18–20]. Transcription of COX-
2, which catalyzes the first step of PGE biosynthesis,

occurs in the early stages of macrophage activation
and is known to be triggered by quartz [11].
Thus, we investigated COX-2 transcription, expres-
sion and enzymatic activity in terms of PGE
2
produc-
tion in the RAW cell line and BAL macrophages. The
transcription factors involved in COX-2 induction and
the possible role of specific ROS in triggering the
COX-2 synthetic pathway were also investigated.
In RAW 264.7 cells, AA-treated quartz particles
induced higher biosynthesis of COX-2 and greater pro-
duction of PGE
2
than untreated particles (Figs 1 and
2). Similar results were obtained with rat alveolar
macrophages freshly collected from healthy animals by
BAL (Fig. 3). These results are particularly relevant
because COX-2 over-expression seems to be strictly
related not only to inflammation development but also
cancer progression [21,22]. Interestingly, the enhancing
effect of AA-treated over untreated quartz on COX-2
synthesis was even more evident in the presence of
IFN-c (Fig. 2), a cytokine produced by activated
lymphocytes, which are believed to be recruited by
macrophages at a later stage of the lung inflammatory
process. The result of this experiment suggests that the
quartz surface modifications caused by AA are rele-
vant not only in the first steps of the lung inflam-
matory response, but also subsequently. The fact that

macrophages are unable to dissolve the internalized
quartz particles indeed prolongs macrophage activation
through multiple ingestion–re-ingestion cycles and
exposes the same cells to cytokines produced by activa-
ted lymphocytes [4]. The long-lasting presence of
quartz in the lung may expose the particles to AA pre-
sent in the bronchoalveolar fluid, inducing the chem-
ical modifications demonstrated by previous in vitro
experiments [6,9]; the resulting in vivo AA-modified
quartz could eventually favour an escalation of the
inflammatory response by macrophages, also stimula-
ted by cytokines released by other cell types during the
ongoing inflammation.
Results obtained here also demonstrate the specif-
icity of action of AA on crystalline (Min-U-Sil 5
quartz), as opposed to amorphous (Aerosil OX50),
silica. Indeed, no difference in COX-2 synthesis was
observed between cell cultures incubated with AA-
treated or untreated Aerosil OX50, although, rather
surprisingly, we observed significant COX-2 expres-
sion triggered by the amorphous silica. In contrast
with what was observed with quartz, however, cell
activation was transient, decreasing rapidly 18 h after
exposure to the particles (data not shown), suggest-
ing an acute cell response followed by a rapid recov-
ery. This time-course of COX-2 synthesis led us to
rule out a possible endotoxin contamination of the
amorphous silica particles, as it is well known that
lipopolysaccharide induces high production of prosta-
glandins, which lasts well after 18 h of stimulation

of the cells [23]. A possible explanation for the
inflammatory response elicited by amorphous silica
in RAW 264.7 cells comes from the dimensions of
the aerosil particles, which are smaller than the crys-
talline ones [24]. Particle dimension is critical during
macrophage phagocytosis, with smaller particles
being internalized better than larger ones and con-
sequently inducing greater cellular activation [25]. In
fact, the highest increase in COX-2 expression in
RAW cells was observed with an aerosil particle
concentration (100 lgÆmL
)1
) resulting in a surface
area 10 times higher than that of the same concen-
tration of quartz particles (Fig. 1).
Ascorbate-treated quartz enhances COX-2 expression S. Scarfı
`
et al.
68 FEBS Journal 274 (2007) 60–73 ª 2006 The Authors Journal compilation ª 2006 FEBS
In line with our in vitro results, both Johnston et al.
[26] and Warheit et al. [27] reported in in vivo experi-
ments a large, but transient, pulmonary inflammatory
response to amorphous silica, followed by a rapid
post-exposure decrease in the principal inflammatory
and cytotoxic biomarkers.
In RAW 264.7, activation of COX-2 by quartz is con-
trolled by NF-jb translocation to the nucleus and also
by AP-1 [11,14,28]. Indeed, nuclear translocation of
NF-jB and AP-1 in RAW 264.7 cells was stimulated to
a higher degree when macrophages were challenged with

AA-treated, as compared with untreated, quartz parti-
cles (Fig. 5). Moreover, we describe for the first time, to
our knowledge, involvement of pCREB in the inflam-
matory response triggered by quartz. As observed for
NF-jB and AP-1, AA-treated quartz was more effective
than untreated quartz in stimulating pCREB transloca-
tion. Interestingly, whereas NF-jB and pCREB were
activated by the highest quartz concentration
(100 lgÆmL
)1
), the AP-1 complex was activated by the
lowest (15 lgÆmL
)1
). These data indicate that quartz
can trigger different signal-transduction pathways in
macrophages depending on the number of particles
coming into contact with the phagocytic cells. In line
with its stimulation of transcription factor translocation,
AA-treated quartz induced a higher COX-2 mRNA syn-
thesis than untreated crystalline silica (Fig. 4).
It is generally accepted that free-radical production
by quartz is responsible for both quartz-induced toxicity
and NF-jB translocation leading to macrophage activa-
tion [29–32]. COX-2 production by RAW 264.7 cells sti-
mulated with 15 lgÆmL
)1
AA-treated quartz was indeed
significantly reduced in the presence of the ROS scaven-
gers catalase and mannitol (Fig. 6A), indicating that
AA treatment of quartz enhances COX-2 synthesis by

means of radicals derived from H
2
O
2
. A crucial role of
iron in the generation of hydroxyl free radicals triggered
by quartz has been reported [32]. However, the absence
of inhibition of COX-2 synthesis stimulated by AA-trea-
ted quartz by desferrioxamine (Fig. 6), together with
our previous results demonstrating hydroxyl radical
production by AA-treated quartz in the presence of
H
2
O
2
, suggests that the AA-modified quartz surface has
itself a specific reactivity towards H
2
O
2
, generating OH
•
without the need for iron [9]. If this hypothesis is cor-
rect, the iron content of quartz particles may play a
minor role in its radical-induced toxicity.
Summarizing the results obtained, our work demon-
strates that, indeed, COX-2 transcription, synthesis
and enzymatic activity, as assessed by PGE
2
produc-

tion, are significantly increased in murine and rat
macrophages challenged with AA-pretreated quartz
compared with untreated quartz particles. Further-
more, we confirm the recruitment of NF-jB and AP-1
transcription factors into the quartz-triggered macro-
phage activation pathway and provide evidence indi-
cating involvement of another transcription factor,
pCREB, which has already been implicated in COX-2
induction in macrophages but never associated with
quartz stimulation in these cells. Finally, a causal role
for H
2
O
2
-derived ROS in the mechanism by which
AA-modified quartz stimulates COX-2 transcription is
demonstrated.
In conclusion, the AA concentration in the lung epi-
thelium may play a pivotal role in enhancing the cyto-
toxic and pro-inflammatory properties of the quartz
particles, instead of preventing them by means of its
antioxidant properties, as currently believed.
Experimental procedures
Materials
All reagents were acquired from Sigma-Aldrich (Milan,
Italy), unless otherwise stated.
Cell cultures
The mouse macrophage cell line RAW 264.7 was obtained
from the American Type Culture Collection (Rockville,
MD, USA). Rat alveolar macrophages were obtained by

BAL from healthy animals (see below). Cells were cultured
at 37 °C in a humidified, 5% CO
2
atmosphere in Dul-
becco’s modified essential medium containing 4 mm gluta-
mine, supplemented with 10% defined fetal bovine serum
(HyClone, Logan, UT, USA) (complete medium). Cell
stimulation using different concentrations of both sterilized
quartz (Min-U-Sil 5; US Silica, Berkeley Spring Plant, spe-
cific surface area calculated by Brunaner, Emmett and
Teller, SSA
BET
¼ 5.2 m
2
Æg
)1
) and Aerosil OX50 (SSA
BET
¼
43.3 m
2
Æg
)1
, Degussa AG, Bitterfeld, Germany) was
obtained by adding 15, 50 or 100 lgÆmL
)1
particles treated
with distilled water or AA (prepared as described in [9]). In
detail, in terms of surface area ⁄ incubation volume, 15, 50
and 100 lgÆmL

)1
Min-U-Sil quartz particles corresponded
to 0.75, 2.6 and 5.2 cm
2
ÆmL
)1
, and for Aerosil OX50 they
corresponded to 6.5, 21.65 and 43.3 cm
2
ÆmL
)1
, respectively.
After 6 or 18 h culture, media were collected to detect
PGE
2
release, and cells were processed to obtain cell lysates
for western blot analyses, or nuclei were separated and
extracted for EMSA.
Collection of alveolar macrophages from BAL
samples
Male Sprague-Dawley rats (8–10 weeks) were purchased
from Harlan Italy (S. Pietro al Natisone, Italy) and housed
S. Scarfı
`
et al. Ascorbate-treated quartz enhances COX-2 expression
FEBS Journal 274 (2007) 60–73 ª 2006 The Authors Journal compilation ª 2006 FEBS 69
at the animal facility of the Biochemistry Section in the
Department of Experimental Medicine of the University of
Genoa. The program of animal use was approved by the
CBA ethics committee, and all procedures involving

animals were performed under protocols approved by the
European Community directives.
Three groups of four males were killed with sodium
pentobarbital (100 mgÆkg
)1
, intraperitoneally). Then, a tra-
cheal cannula was inserted, and BAL was performed using
ice-cold Ca
2+
⁄ Mg
+
-free Hanks’ medium. Lavages of 6–
8 mL were performed until a total of 50 mL lavage fluid
was collected from each rat. The samples were centrifuged
at 300 g for 10 min at 4 °C (Allegra X-22R, swinging
bucket rotor, Beckman Coulter SpA, Milan, Italy).
The supernatants were removed, and the cells from the
four rats were combined, resuspended in Hanks’ medium
(with Ca
2+
and Mg
+
) and centrifuged. Cells were then
resuspended in complete medium, then (1–1.5) · 10
6
cells ⁄
well (depending on the number of cells collected from each
group) were seeded on to 60 · 15 mm tissue culture dishes
(Falcon Becton Dickinson, Franklin Lakes, NJ, USA), and
cultured for 6 h at 37 ° C. Medium containing nonadherent

cells was discarded. Then 100 lgÆmL
)1
AA-treated or
untreated quartz in complete medium was added to the
cultures, which were further incubated for 6 h at 37 °C.
COX-2 expression and PGE
2
production were quantified as
described below.
Measurement of murine COX-2 expression and
PGE
2
production
Expression of murine COX-2 in RAW 264.7 macrophages
after 6 h incubation with 15, 50 or 100 lg ÆmL
)1
AA-treated
or untreated quartz or aerosil was measured by western blot
analysis of total cell lysates. PGE
2
production was quanti-
fied in the culture medium. Untreated cultures were used
as controls. Briefly, 3 · 10
6
cells were seeded on to
60 · 15 mm tissue culture dishes (Falcon BD) and cultured
as described above; after 18 h, the stimuli were added to the
culture medium, and cells were further incubated for 6 h at
37 °C. Thereafter, adherent cells were washed three times
with ice-cold NaCl ⁄ P

i
and lysed with 400 lL lysis buffer
(100 mm dithiothreitol, 2% SDS, 10% glycerol and 50 mm
Tris ⁄ HCl, adjusted to pH 6.8). The lysates were heated at
100 °C for 10 min, sonicated, and the protein concentration
was determined [33]. Identical amounts of lysate proteins
(40 lg per sample) were loaded on to SDS ⁄ 10% polyacryla-
mide gels, electrophoretically separated, and transferred to
Immun-Blot poly(vinylidene difluoride) membranes (Bio-
Rad, Milan, Italy). Membranes were blocked and cut at the
level of the 50-kDa precoloured marker. The upper part was
incubated with an anti-COX-2 mouse monoclonal IgG, and
the lower part was stained with an anti-(b-actin) goat poly-
clonal IgG, both at 1 lgÆmL
)1
(Santa Cruz Biotechnology,
Santa Cruz, CA, USA). Western blots were developed with
the ECL-PLUS kit (Amersham Pharmacia Biotech, Little
Chalfont, Bucks, UK), according to the manufacturer’s
instructions. Band detection and densitometry were per-
formed using the Chemi-Doc System and the quantity one
software package (Bio-Rad).
The PGE
2
concentration in the culture medium from
cells incubated for 6 or 18 h with quartz particles, in the
presence or absence of 100 pgÆmL
)1
murine IFN-c, was
quantified using the PGE

2
Monoclonal EIA Kit (Cayman
Chemical Company, Ann Arbor, MI, USA), according to
the manufacturer’s instructions.
Nuclear extracts
RAW 264.7 cells (3 · 10
6
per assay) were seeded on to
60 · 15 mm tissue culture dishes (Falcon BD) and cultured
as described above. After 18 h, the medium was discarded
and cells were incubated at 37 °C with 15, 50 and
100 lgÆmL
)1
AA-treated or untreated quartz for 30 min. At
the end of the incubation, cells were washed 3 times with ice-
cold NaCl ⁄ P
i
, recovered from the dishes with a cell scraper in
1 mL NaCl ⁄ P
i
, and pelleted in a microfuge at 14 000 g for
3 min at 4 °C (MICROcentrifugette 4212, fixed angle rotor,
ALC, Milan, Italy). The cell pellets were then resuspended in
400 lL ice-cold buffer A (20 mm Tris ⁄ HCl, pH 7.8; 50 mm
KCl; 10 lgÆmL
)1
leupeptin; 0.1 m dithiothreitol; 1 mm phe-
nylmethanesulfonyl fluoride), and 400 lL buffer B (buffer A
plus 1.2% Nonindet P40; Sigma) was then added. The sus-
pension was vortex-mixed for 10 s, centrifuged in a Micro-

fuge at 14 000 g for 30 s at 4 °C (MICROcentrifugette 4212,
fixed angle rotor, ALC), and the supernatant was discarded.
The pelleted nuclei were then washed with 400 lL buffer A
and centrifuged again at 14 000 g for 30 s at 4 °C (MICRO-
centrifugette 4212, fixed angle rotor, ALC). After removal of
the supernatant, the nuclear pellet was resuspended in
100 lL buffer B, mixed thoroughly in ice for 15 min to
disrupt nuclear membranes, sonicated for 10 s, and finally
centrifuged at 14 000 g for 20 min at 4 °C (MICROcentrifu-
gette 4212, fixed angle rotor, ALC). The supernatant contain-
ing the nuclear extracts was collected and the total protein
content was measured [33].
EMSA
EMSAs were performed by the method of Singh et al. [34],
slightly modified. Briefly, 67 ng of the specific oligonucleo-
tide (Santa Cruz) for NF-kB, AP-1 or pCREB was mixed
with 5 lL Forward Reaction Buffer (Invitrogen, Carlsbad,
CA, USA) and 0.5 lCi [c-
32
P]ATP (Amersham) in a final
volume of 15 lL and preincubated at 37 °C for 3 min.
Then 10 U T4 polynucleotide kinase (Invitrogen srl, Milan,
Italy) was added, and the mixture was incubated at 37 °C
for 20 min.
The radiolabelled oligonucleotide was then loaded and
purified on a Sephadex G25 mini-column (Amersham), and
Ascorbate-treated quartz enhances COX-2 expression S. Scarfı
`
et al.
70 FEBS Journal 274 (2007) 60–73 ª 2006 The Authors Journal compilation ª 2006 FEBS

the elution fraction from the column centrifuged in a
microfuge at 3800 g at room temperature (MICROcentri-
fugette 4212, fixed angle rotor, ALC). The radioactivity
of the eluted fraction was quantified with a b-Counter
(Beckman Coulter Inc., Fullerton, CA, USA).
The binding reaction was carried out at 25 °C for 20 min
in a final volume of 20 lL containing 10 lg nuclear extract,
0.25 lCi purified radiolabelled oligonucleotide and 0.5 lg
poly(dI)poly(dC) (Amersham) in binding buffer (25 mm
Hepes, pH 7.9; 0.5 mm EDTA; 0.5 mm dithiothreitol; 5%
glycerol; 50 mm NaCl; 0.5 mm phenylmethanesulfonyl
fluoride). Finally, the radiolabelled protein–DNA com-
plexes were electrophoretically resolved on a 4.5% nonde-
naturing polyacrylamide gel. The gel was dried on a Gel
Dryer 543 (Bio-Rad), and the radioactive complexes were
visualized and quantified with the Packard Cyclone Stor-
age Phosphor System and optiquant acquisition and
analysis software (PerkinElmer Inc., Boston, MA, USA).
Quantitative RT-PCR
Total RNA from RAW 264.7 macrophages was extracted
using Rnaeasy Mini Kit (Qiagen spa, Milan, Italy) and
RNase-Free DNase Set (Qiagen) according to the manufac-
turer’s instructions, from a starting material of 2 · 10
6
RAW 264.7 cells grown on 60 · 15 mm tissue culture
dishes (Falcon BD) in the presence of 100 lgÆmL
)1
quartz
pretreated with distilled water or AA for 30 min, 60 min,
3 h and 18 h.

Total cell cDNA was synthesized from 1 lg RNA in the
appropriate buffer containing 5 mm MgCl
2
, 40 U ribonuc-
lease inhibitor (RN
ASE
OUT; Invitrogen), 10 mm dithiothrei-
tol, and 200 U Superscript
TM
III (Invitrogen), at 50 °C for
50 min. Then complementary RNA was removed using 1 lL
Escherichia coli RNase H (Invitrogen) at 37 °C for 20 min.
The amount of COX-2 mRNA, normalized to the relative
GAPDH control, was determined by real-time quantitative
PCR using a Chromo 4 instrument (MJ Research, Bio-Rad).
PCR was performed in a 20-lL volume in nuclease-free
water containing 10 lL2· master mix iQ SYBR GreenÒ
(Bio-Rad), 0.2 lm each primer, and 0.5 lL cDNA or negat-
ive control. All samples were analysed in triplicate.
The following PCR conditions were used: 10 min initial
denaturation, followed by 40 cycles with denaturation at
95 °C for 15 s, annealing and elongation at 60 °C for 60 s.
The fluorescence was measured at the end of each elonga-
tion step. The next step was slow heating (1 °C per s) of
the amplified product from 55 °Cto92°C, in order to gen-
erate a melting temperature curve. This curve served as a
specificity control. The entire cycling process, including
data analysis, was monitored using the dna engine opti-
conÒ 2 real-time detection system Software program
(2.03 version).

The sequences of the GAPDH (M32599) primers were:
5¢-TCTCCCTCACAATTTCCATCCCAG-3¢ (forward pri-
mer) and 5¢-GGGTGCAGCGAACTTTATTGATGG-3¢
(reverse primer).
The sequences of the COX-2 (M64291) primers were: 5¢-
CCAGCAAAGCCTAGAGCAAC-3¢ (forward primer) and
(5¢-AGCACAAAACCAGGATCAGG-3¢) reverse primer.
To detect the PCR efficiency for each couple of primers,
an amplification curve was performed, using four different
dilutions of cDNA.
Data analysis to detect the relative gene expression of
COX-2, using the cDNA from untreated cells as calibra-
tor sample, was performed with the comparative thresh-
old Ct method [35] via gene expression analysis
software for the iCycler iQ Real Time Detection System
(Bio-Rad) [36].
Scavenger treatment
COX-2 expression in RAW 264.7 macrophages after stimu-
lation with 15, 50 and 100 lgÆmL
)1
AA-treated or
untreated quartz was also assessed in the presence of
4000 UÆmL
)1
catalase (Sigma), 50 mm mannitol (Sigma)
and 2 mm desferrioxamine (Sigma). Cells were cultured as
described above, and the scavengers were added together
with the quartz particles for a total incubation time of 6 h.
Subsequently, cells were processed for western blot analysis,
as described above.

Statistical analysis
COX-2 expression in RAW 264.7 cells (Fig. 1)
In both Fig. 1A and 1B, data were square-root-transformed
and analysed with analysis of variance, choosing P < 0.01
as basal level of significance as described by Underwood
[37]. Where significant F-ratios were obtained with analysis
of variance, multicomparison analyses were performed
using the Tukey test.
PGE
2
production in RAW 264.7 cells (Fig. 2)
In Fig. 2A, data were checked for normality and homosce-
dasticity of variance with the Cochran test and analysed
with analysis of variance. Where suitable, multicomparison
analyses were performed using the Tukey test or Student–
Newman–Keuls test; in Fig. 1B, data were log-transformed
and analysed with analysis of variance. Again, where suit-
able, multicomparison analyses were performed using the
Tukey test or Student–Newman–Keuls test.
COX-2 expression and PGE
2
production in rat alveolar
macrophages (Fig. 3)
In Fig. 3A, data were checked with the F-test and analysed
with analysis of variance. Multicomparison analyses were
performed using Student–Newman–Keuls test; in Fig. 3B,
S. Scarfı
`
et al. Ascorbate-treated quartz enhances COX-2 expression
FEBS Journal 274 (2007) 60–73 ª 2006 The Authors Journal compilation ª 2006 FEBS 71

data were analysed by Kruskal–Wallis analysis of variance
and multicomparison analysis [38].
Time-course of COX-2 mRNA synthesis in RAW 264.7
macrophages (Fig. 4)
Data were analysed with the Scheirer–Ray–Hare test and
multicomparison analysis.
NF-jB, pCREB and AP-1 nuclear translocation
in RAW 264.7 cells (Fig. 5)
In Fig. 5A,B, data were checked with the F-test and
analysed with analysis of variance and the Tukey test;
the t-test was used to check differences between untreated
and AA-treated quartz samples at the same concentra-
tion. In Fig. 5C, data were square-root-transformed and
analysed with analysis of variance and the Tukey test;
the t-test was used to check differences between untreated
and AA-treated quartz samples at the same concentra-
tion.
COX-2 expression in the presence of ROS scavengers
(Fig. 6)
In Fig. 6A–C, data were square-root-transformed and ana-
lysed with analysis of variance, choosing P < 0.01 as the
basal level of significance as described by Underwood [37].
Where significant F-ratios were obtained with analysis of
variance, multicomparison analyses were performed using
the Dunnett test.
Acknowledgements
This work was partially supported by 2004 CIPE Regi-
one Liguria and 2003 MURST-PRIN funds. We are
deeply indebted to Professor Antonio De Flora and
Professor Elena Zocchi for scientific discussion and

critical reading of the manuscript. We acknowledge the
Egenmann & Veronelli s.r.c. Company (Rho, Milan,
Italy) for kindly providing Aerosil OX50, and the US
Silica Company (Berkeley Springs, WV, USA) for pro-
viding Min-U-Sil 5.
References
1 Ding M, Chen F, Shi X, Yucesoy B, Mossman B &
Vallyathan V (2002) Diseases caused by silica: mecha-
nisms of injury and disease development. Int Immuno-
pharmacol 2, 173–182.
2 Fubini B (1998) Surface chemistry and quartz hazard.
Ann Occup Hyg 42, 521–530.
3 Donaldson K & Borm PJ (1998) The quartz hazard: a
variable entity. Ann Occup Hyg 42, 287–294.
4 Kreyling WG (1992) Intracellular particle dissolution in
alveolar macrophages. Environ Health Perspect 97, 121–
126.
5 Bavestrello A, Benatti U, Cerrano C, Cattaneo-Vietti R,
Cortesogno L, Gaggero L, Giovine M, Tonetti M &
Sara
`
M (1995) Quartz dissolution by the sponge Chon-
drosia reniformis (Porifera, Demospongiae). Nature 378,
374–376.
6 Fenoglio I, Martra G, Coluccia S & Fubini B (2000)
Possible role of ascorbic acid in the oxidative damage
induced by inhaled crystalline silica particles. Chem Res
Toxicol 13, 971–975.
7 Gozal E, Ortiz LA, Zou X, Burow ME, Lasky JA &
Friedman M (2002) Silica-induced apoptosis in murine

macrophage: involvement of tumor necrosis factor-alpha
and nuclear factor-kappaB activation. Am J Respir Cell
Mol Biol 27, 91–98.
8 Claudio E, Segade F, Wrobel K, Ramos S & Lazo PS
(1995) Activation of murine macrophages by silica parti-
cles in vitro is a process independent of silica-induced
cell death. Am J Respir Cell Mol Biol 13, 547–554.
9 Giovine M, Pozzolini M, Fenoglio I, Scarfı
`
S, Ghiazza
M, Benatti U & Fubini B (2003) Crystalline silica incu-
bated in ascorbic acid acquires a higher cytotoxic poten-
tial. Toxicol Ind Health 18, 249–255.
10 Cao Y & Prescott SM (2002) Many actions of cyclooxy-
genase-2 in cellular dynamics and in cancer. J Cell Phy-
siol 190, 279–286.
11 Chen F, Sun S, Kuhn DC, Gaydos LJ, Shi X, Lu Y &
Demers LM (1997) Involvement of NF-kappaB in
silica-induced cyclooxygenase 2 gene expression in rat
alveolar macrophages. Am J Physiol 272, L779–L786.
12 Zeidler PC, Roberts JR, Castranova V, Chen F, Butter-
worth L, Andrew ME, Robinson VA & Porter DW
(2003) Response of alveolar macrophages from indu-
cible nitric oxide synthase knockout or wild-type mice
to an in vitro lipopolysaccharide or silica exposure.
J Toxicol Environ Health 66, 995–1013.
13 Wadleigh DJ, Reddy ST, Kopp E, Ghosh S & Hersch-
man HR (2000) Transcriptional activation of the
cyclooxygenase-2 gene in endotoxin-treated RAW 264.7
macrophages. J Biol Chem 275, 6259–6266.

14 Castranova V (2004) Signaling pathways controlling the
production of inflammatory mediators in response to
crystalline silica exposure: role of reactive oxygen ⁄ nitro-
gen species. Free Radic Biol Med 37 , 916–925.
15 Skoza L, Snyder E & Kikkawa Y (1983) Ascorbic acid
in bronchoalveolar wash. Lung 161, 99–109.
16 Ghio AJ, Kennedy TP, Crissman KM, Richards JH &
Hatch GE (1998) Depletion of iron and ascorbate in
rodents diminishes lung injury after silica. Exp Lung
Res 24
, 219–232.
17 Titkova VA (1967) Effect of novocaine and vitamin C
on the fibrogenic action of quartz dust. Arch Pathol 29,
71–76.
Ascorbate-treated quartz enhances COX-2 expression S. Scarfı
`
et al.
72 FEBS Journal 274 (2007) 60–73 ª 2006 The Authors Journal compilation ª 2006 FEBS
18 Chen F, Sun SC, Kuh DC, Gaydos LJ & Demers LM
(1995) Essential role of NF-jb activation in silica-
induced inflammatory mediator production in macro-
phages. Biochem Biophys Res Commun 214, 985–992.
19 Chen F, Kuh DC, Sun SC, Gaydos LJ & Demers LM
(1995) Dependence and reversal nitric oxide production
on NF-jb in silica and lipopolysaccharide-induced
macrophages. Biochem Biophys Res Commun 214,
839–846.
20 Barrett EG, Johnston C, Oberdorster G & Finkelstein
JN (1999) Antioxidant treatment attenuates cytokine
and chemokine levels in murine macrophages following

silica exposure. Toxicol Appl Pharmacol 158, 211–220.
21 Hinz B & Brune K (2002) Cyclooxygenase-2 10 years
later. J Pharmacol Exp Ther 300, 367–375.
22 Brown JR & DuBois RN (2004) Cyclooxygenase as a
target in lung cancer. Clin Cancer Res 10, 4266s–4269s.
23 Riese J, Hoff T, Nordhoff A, DeWitt DL, Resch K &
Kaever V (1994) Transient expression of prostglandin
endoperoxide synthase-2 during mouse macrophage acti-
vation. J Leukoc Biol 55, 476–482.
24 Elias Z, Poirot O, Daniere MC, Terzetti F, Marande
AM, Dzwigaj S, Pezerat H, Fenoglio I & Fubini B
(2000) Cytotoxic and transforming effects of silica parti-
cles with different surface properties in Syrian hamster
embryo (SHE) cells. Toxicol In Vitro 14, 409–422.
25 Santini MT, Rainaldi G, Ferrante A, Romano R, Cle-
mente S, Motta A, De Berardis B, Balduzzi M, Paoletti
L & Indovina PL (2004) Environmental fine particulate
matter (PM 2.5) activates the RAW 264.7 macrophage
cell line even at very low concentrations as revealed by
1H NMR. Chem Res Toxicol 17, 63–74.
26 Johnston CJ, Driscoll KE, Finkelstein JN, Baggs R,
O’Reilly MA, Carter J, Gelein R & Oberdorster G
(2000) Pulmonary chemokine and mutagenic responses
in rats after subchronic inhalation of amorphous and
crystalline silica. Tox Sci 56, 405–413.
27 Warheit DB, McHugh TA & Hartsky MA (1995) Dif-
ferential pulmonary responses in rats inhaling crystal-
line, colloidal or amorphous silica dusts. Scand J Work
Environ Health 21, 19–21.
28 Ding M, Shi X, Dong Z, Chen F, Lu Y, Castranova

V & Vallyathan V (1999) Freshly fractured crystalline
silica induces activator protein-1 activation through
ERKs and p38 MAPK. J Biol Chem 274, 30611–
30616.
29 Shi XL, Dalal NS & Vallyathan V (1988) ESR evidence
for the hydroxyl radical formation in aqueous suspen-
sion of quartz particles and its possible significance to
lipid peroxidation in silicosis. J Toxicol Environ Health
25, 237–245.
30 Vallyathan V, Shi XL, Dalal NS, Irr W & Castranova
V (1988) Generation of free radicals from freshly frac-
tured silica dust. Potential role in acute silica-induced
lung injury. Am Rev Respir Dis 138, 1213–1219.
31 Shi X, Dong Z, Huang C, Ma W, Liu KYeJ, Chen F,
Leonard SS, Ding M, Castranova V & Vallyathan V
(1999) The role of hydroxyl radical as a messenger in
the activation of nuclear transcription factor NF-kap-
paB. Mol Cell Biochem 194, 63–70.
32 Rojanasakul YYeJ, Chen F, Wang L, Cheng N, Castra-
nova V, Vallyathan V & Shi X (1999) Dependence of
NF-kappaB activation and free radical generation on
silica-induced TNF-alpha production in macrophages.
Mol Cell Biochem 200, 119–125.
33 Bradford MM (1976) A rapid and sensitive method for
the quantitation of microgram quantities of protein util-
izing the principle of protein-dye binding. Anal Biochem
72, 248–254.
34 Singh IS, He JR, Calderwood S & Hasday JD (2002)
A high affinity HSF-1 binding site in the 5¢-untranslated
region of the murine tumor necrosis factor-a gene is

a transcriptional repressor. J Biol Chem 277,
4981–4988.
35 Aarskog NK & Vedeler CA (2000) Real-time quanti-
tative polymerase chain reaction. A new method that
detects both the peripheral myelin protein 22 duplica-
tion in Charcot-Marie-Tooth type 1A disease and the
peripheral myelin protein 22 deletion in hereditary
neuropathy with liability to pressure palsies. Hum Genet
107, 494–498.
36 Vandesompele J, De Preter K, Poppe B, Van Roy N,
De Paepe A & Speleman F (2002) Accurate normali-
zation of real-time quantitative RT-PCR data by
geometric averaging of multiple internal control genes.
Genome Biol 3, 0034.1–0034.12.
37 Underwood AJ (1981) Techniques of analysis of vari-
ance in experimental marine biology and ecology. Annu
Rev Oceanogr Mar Biology 19, 513–605.
38 Siegel S & Castellan NJ (1988) Nonparametric Statistics
for the Behavioural Sciences, 2nd edn. McGraw-Hill,
New York, NY.
S. Scarfı
`
et al. Ascorbate-treated quartz enhances COX-2 expression
FEBS Journal 274 (2007) 60–73 ª 2006 The Authors Journal compilation ª 2006 FEBS 73