Tumor sections from nonneoplastic (n=15), benign (n=28), and malignant ovarian tumors (n=20) were obtained from 63 women. Immunohistochemistry of the tumor sections demonstrated that inducible nitric oxide synthase (iNOS) expression was increased in ovarian cancer samples compared to nonneoplastic or benign tumor samples. Using the Griess method, nitric oxide (NO) metabolite levels were also found to be elevated in malignant tumor samples compared to benign tumor samples (P<.05). For stage I ovarian cancer, intracystic NO levels >80μM were more frequent than NO levels <80μM, and iNOS expression in well-differentiated carcinomas was greater than in moderately/poorly differentiated carcinomas (P<.05). These data suggest an important role for NO in ovarian carcinogenesis.
1. Introduction
Ovarian cancer is
the eighth most frequent malignant neoplasia and the fifth most common cause of
death from malignant tumor growth in women in the US
[1]. The frequency of ovarian
cancer increases in each decade of life, with the highest rate of diagnosis
occurring in women that are 75 years of age [2]. Early diagnosis of ovarian tumors is difficult since tumors
with diameters less than 5 cm cannot be recognized by
bimanual pelvic examination [3, 4].
However, pelvic examination, ultrasonography, detection of
tumor markers (such as CA 125, CA 15.3, CA 72.4, and CA 19.9),
as well as color Doppler imaging have been shown to be useful
in the diagnosis of ovarian cancer, despite the limitations of
these methods in differentiating between benign
and malignant tumors [5].
Nitric oxide (NO) is a biological messenger synthesized from
L-arginine by nitric oxide synthase (NOS). The endothelial and
neuronal isoforms of NOS (eNOS and nNOS, resp.) are
constitutively expressed in many cell types,
however, inducible NOS (iNOS) is only induced in leukocytes,
endothelial cells, and other specific cell types after
stimulation by bacterial endotoxins or
cytokines, resulting in higher concentrations of NO
[6].
NO can also exert both pro- and antitumor effects in the
tumor microenvironment. Production of NO is a mechanism by
which activated endothelium can lyse tumor cells
[7], however, it can also
regulate tumor growth and metastasis depending on its
concentration [8, 9]. Data from previous
studies also suggest that NO is both cytotoxic and cytostatic
against microorganisms and malignant cells
[10, 11] with
synthesis of NO by malignant cells causing NO-mediated
apoptosis [12]. As a free radical,
NO can react to produce peroxynitrites which can directly and
indirectly cause DNA damage [12].
If produced for a long period of time, excess NO production can
lead to mutations and ultimately to cancer
[13, 14].
In addition to increasing the metastatic potential of tumor cells via
mutations in the DNA, NO production by neoplastic cells
promotes angiogenesis, an essential process for the growth and
maintenance of tumors [15, 16].
Understanding the role of iNOS in ovarian cancer would
provide valuable insight into the development of additional
therapeutic options. The aim of the present study was to
identify differences between iNOS expression and the local
production of NO in patients with varying stages and grades of
ovarian cystic tumors. Therefore, levels of
intracystic NO metabolites and expression of iNOS were analyzed
in tumor sections from patients with nonneoplastic, benign, or
malignant ovarian tumors.
2. Materials and Methods2.1. Patients and Pathological Assessment
This study enrolled 63 randomly selected women who received
pelvic mass outpatient services from the Discipline of
Gynecology and Obstetrics of the Federal University of
Triângulo Mineiro (UFTM), Uberaba, Brazil. These patients
underwent surgery for an adnexal mass between February 1996 and
February 2007, and informed consent was obtained from patients
to allow their tissue to be used for examination and related
experiments. This study was approved by the UFTM Research
Ethics Committee.
Candidates for exploratory surgery were characterized by one
or more of the following criteria: cysts with ≥1 thick septum
(>3 mm) or ≥2 thin septa, a cyst diameter ≥7 cm,
persistence or increase in the cyst or ovarian volume over a
minimum of two follow-up periods, the presence of vegetation or
calcification, a solid or predominantly solid tumor, ascites,
elevated serum levels of tumor markers, or a resistance index
≤0.4 as detected by color Doppler imaging
[3, 5].
Inclusion criterion was the anatomicopathological finding of an
ovarian tumor (primary neoplastic or nonneoplastic tumor).
Exclusion criteria were adnexal torsion, cyst rupture,
metastasis of another primary tumor, or previous chemotherapy.
The anatomicopathological evaluation and staging of all cases
were performed according to guidelines published by the World
Health Organization (WHO) and the International Federation of
Gynecology and Obstetrics (FIGO) [17, 18]. Patients
were divided into 3 groups according to the classification of
tumor type: nonneoplastic (n=15), benign (n=28), or malignant (n=20) (which included cystadenomas of
borderline malignancy). Characterization of patient groups is presented in
Table 1.
Characteristics
of the three-patient groups compared in this study.
Nonneoplastic tumors (n=15)
Benign neoplasias (n=28)
Malignant neoplasias (n=20)
Age (years,
mean ± SD)
42.8±7.0
37.1±14.6
46.9±14.9
Parity child (mean ± SD)
2.3±1.5
2±1.8
3.1±2.8
Smokers
4 (26.7%)
7 (25.0%)
9 (45.0%)
Race:
Caucasian women
12 (80.0%)
17 (60.7%)
17 (85.0%)
Non-Caucasian women
3 (20.0%)
11 (39.3%)
3 (15.0%)
Use of hormonal
contraception
2 (13.3%)
5 (17.9%)
2 (10.0%)
Tubal ligation
3 (20.0%)
3 (10.7%)
4 (20.0%)
Hormonal status:
Reproductive age (years)
12 (80.0%)
23
(82.1%)
10 (50.0%)
In
menopause
3 (20.0%)
5 (17.9%)
10 (50.0%)
2.2. Collection of Intracystic Fluid
Cystic
fluid samples were aseptically collected by punction immediately following
resection of tissue. Bloody fluids caused by punction were excluded from analysis.
The collected fluids were immediately stored on ice until centrifugation (180 × g, 15 minutes) was performed. Cell
supernatants were transferred to fresh tubes, and the cell pellets were
resuspended in phosphate-buffered saline (PBS). Both samples were stored at −70°C until needed [19].
2.2. Determination of Nitrate Concentration
The
levels of NO metabolites (nitrite plus nitrate) in cystic samples were
determined by enzymatically reducing the nitrate present with nitrate reductase
as previously described [20]. Briefly, 50μL of nondiluted sample was incubated
with an equal volume of reductase buffer (0.1 M potassium phosphate (pH 7.5),
1 mM NADPH, 10 mM FAD, 4U of nitrate
reductase/mL) for 20 hours at 37°C.
A standard nitrate curve was obtained by incubating sodium nitrate (10 to 200μM) with reductase buffer. The total amount of nitrite was determined by the
Griess method [21]. Briefly, the samples were incubated with an equal volume of
freshly prepared Griess reagent (1% sulfanilamide, 0.1% naphthylenediamine
dihydrochloride in 5% phosphoric acid). Absorbance at 550 nm was determined
using a multi-well plate reader (Multiskan MCC/340MKII, Flow Laboratories) and
the results were reported as micromoles of NO3+NO2.
2.3. Immunohistochemistry
Specimens
obtained from surgical resection were fixed in 10% formalin before being
processed in paraffin. Sections stained with hematoxylin-eosin were reviewed by
a pathologist and a representative section for each case was selected for
immunohistochemical analysis.
Selected sections
were deparaffinized, rehydrated, and heated in a microwave oven in 0.01 M citrate buffer (pH
6.0) for 30 minutes. Endogenous peroxidase activity was blocked by 3% hydrogen
peroxide for 10 minutes, followed by a wash with PBS. The sections were
incubated overnight at 4°C with an anti-iNOS rabbit polyclonal antibody (Santa Cruz Biotechnology, 1:200).
The conjugate used was the avidin-biotin peroxidase detection solution (Dako
Cytomation LSAB and System-HRP). The signal was visualized using
diaminobenzidine (Dako Cytomation Liquid DAB and substrate Chromogen System, Dako).
Slides were counterstained with Harris’s haematoxylin, dehydrated, cleared, and
mounted. A skin sample with chronic granulomatous inflammation known to be
positive for iNOS was used as a positive control. Two independent observers
evaluated the sections and the intensity of staining was evaluated subjectively
using the following designations: 0 (no signal), 1 (weak), 2 (medium), 3
(strong) [22]. When scores of multiple tissue stainings were combined, scores
that were ≤1 were labeled “weak intensity”, and scores ≥2 were labeled
“strong intensity”.
2.4. Statistical Analysis
Data were analyzed
using GraphPad Instat software. For immunohistochemical staining, the
concordance between staining intensity scores was calculated according to the
following classifications: kappa <0.4: slight concordance; kappa ≥0.4 and <0.8: moderate concordance; kappa ≥0.8 and <1:
strong concordance; kappa = 1: perfect concordance. The first inter-rater
agreement was 90.7% (kappa = 0.94). All discordant cases were re-evaluated and
the result determined by consensus. The Fisher’s exact test was used to compare iNOS immunohistochemistry results and to assess the
relationship between iNOS expression and NO levels with the stage and grade of
the ovarian cancer samples analyzed. Data for nitrate levels were
expressed as the mean +/− standard deviation (SD) and values were compared by
ANOVA followed by the Tukey test for individual comparisons. The correlation
between intracystic nitrate levels and iNOS immunolabeling was
tested using Spearman’s rank correlation coefficient. The significance level
was set at less than 0.05.
3. Results3.1. Patients
Sixty-three
randomly selected women receiving pelvic mass outpatient services from the Discipline
of Gynecology and Obstetrics of the Federal University of Triângulo Mineiro
(UFTM) were enrolled in this study. For those diagnosed with nonneoplastic
tumors (n=15), 10 (66.7%) had serous ovarian cysts and 5 (33.3%) had functional
cysts (corpus luteum, follicular, and theca lutein cysts). The 28 patients
diagnosed with benign neoplasias included 11 (39.3%) with serous cystadenoma, 6
(21.4%) with mucinous cystadenoma, 6 (21.4%) with mature teratoma, 3 (10.7%)
with cystadenofibroma, as well as 1 (3.6%) with serous- and 1 (3.6%) with
mucinous-cystadenoma associated with Brenner’s tumor. For the 20 patients
diagnosed with malignant neoplasias, the cases included 7 (35%) serous
adenocarcinoma, 3 (15%) granulosa cell tumor, 3 (15%) mucinous cystadenoma of
borderline malignancy, 2 (10%) mucinous cystadenocarcinoma, 1 (5%) endometrioid
adenocarcinoma, 1 (5%) granulosa cell tumor associated with Brenner’s tumor, 1 (5%) anaplasic
adenocarcinoma, 1 (5%) immature teratoma with epidermoid carcinoma, and 1 (5%)
serous cystadenoma of borderline malignancy. In the ovarian cancer patient group, the number of
each tumor stage resected included 5 (25%) of I-A, 1 (5%) of I-B, 2 (10%) of I-C, 1 (5%) of
III-A, 10 (50%) of III-C, and 1 (5%) of IV. In the III-C and IV stages, five
complete surgeries, four satisfactory citorreductions and only 2 unsatisfactory
citorreductions were performed. Patients receiving adjuvant first-line
chemotherapy received either a combination of cisplatin, epirubicin, and
cyclophosphamide for epithelial tumors, or cisplatin, etoposide, and bleomycin for
granulosa cell tumors of stages ≥I-B. Of the 12 carcinomas, 5 (41.7%) were
well differentiated, 4 (33.3%) were moderately differentiated, and 3 (25%) were
poorly differentiated. Five patients were diagnosed with malignant ovarian
tumors had died before their follow-up.
3.2. iNOS Immunohistochemistry
There were
sufficient samples to perform iNOS immunohistochemistry on 15 nonneoplastic
tumors, 21 benign tumors, and 18 malignant tumors (Figures 1 and 2). The
results are summarized in Table 2. Samples with strong iNOS staining (scored ≥2) were more frequently found in ovarian cancer samples than in nonneoplastic (P=.0014) or benign neoplasia samples (P=.0003).
Immunohistochemical staining of
iNOS in nonneoplastic, benign, and malignant ovarian tissue samples.
Strong intensity
of iNOS
expression(a)
Weak intensity
of iNOS expression(b)
Nonneoplastic tumors (n=15)
5 (33.3%)
10 (66.7%)
Benign
neoplasia (n=21)
6 (28.6%)
15 (71.4%)
Malignant neoplasia (n=18)*+
16 (88.9%)
2 (11.1%)
(a) Received a
score of 2-3 for intensity of iNOS expression.
(b) Received a score of 0-1 for
intensity of iNOS expression.
P=.0014* compared to nonneoplastic tumor
samples.
P=.0003+ compared to benign neoplasia samples
(Fisher’s exact test).
Samples of malignant ovarian cancer
tissue were further classified into well-differentiated (n=11) or moderately/poorly
differentiated (n=7) tumors. For the well-differentiated tumors, 8 (72.7%) presented
strong staining of iNOS while the other 3 presented weak or medium staining of
iNOS. For the moderately/poorly differentiated tumor samples, all showed
weak/medium intensity of iNOS. Overall, well-differentiated tumors presented a
higher frequency of strong iNOS expression compared to moderately/poorly
differentiated carcinomas (P=.004; Fisher’s exact test). No
statistically significant correlation was found between the intensity of iNOS staining
and tumor stage.
Of the 5 (27.8%)
patients diagnosed with malignant ovarian tumors that had died by the time of
follow-up, only one of the samples previously collected from those five
individuals showed strong expression for iNOS.
3.3. Cystic Fluid Nitrate Concentration
Cystic fluid
samples were collected at the time of surgical resection and were subsequently
tested for NO metabolite levels. Cystic fluids from 1 nonneoplastic tumor, 1 benign neoplasia, and 2 malignant ovarian
tumors were not tested due to the viscous consistency of those fluids. The mean
levels of NO metabolites detected in the malignant tumor samples (75.7μM, n=18) were significantly higher (P=.045)
than the NO metabolite levels for benign ovarian tumors (38.5μM, n=27). However, statistically significant
differences were not detected between NO levels of malignant neoplasia samples versus
nonneoplastic samples (40.9μM, n=14) (Figure 3).
NO
metabolite levels determined from cystic fluid samples obtained from patients
with nonneoplastic tumors (n=14), benign tumors (n=27), and malignant neoplasias
(n=18). P=.045 (ANOVA); P*<.05 versus benign neoplasia
(Tukey).
To examine
whether intracystic NO metabolite levels could be predictive for tumor stage, patient
samples were divided into two groups: those with NO metabolite levels <80 μM and
those with NO metabolite levels >80 μM. The value of
80 μM was derived from the median value of NO detected
in malignant tumor samples (81.6 μM). No stage II
samples had detectable levels of NO metabolites. For stage I
samples, 6 (85.7%) had NO
metabolite levels >80 μM, and only 1 (14.3%) had NO
metabolite
levels <80 μM. In contrast, 3 of stage
III/IV tumor samples (27.3%) had NO metabolite levels >80 μM
and 8 (72.7%) had NO metabolite levels <80 μM.
These results indicate that intracystic NO
metabolite levels >80 μM were significantly more
frequent in stage I samples than in stage III/IV samples
(P=.0498;
Fisher exact test). However, there was no significant correlation between NO
metabolite levels and the grade of tumor differentiation.
3.4. iNOS Immunoreactivity and NO Metabolite Levels
Correlations between NO production and iNOS expression in the collected tumor tissues are
summarized in Table 3. Cystic levels of NO did not correlate with iNOS
expression in ovarian tissues.
Correlation between iNOS immunostaining and NO metabolite levels detected in tissue samples collected using Spearman’s
rank correlation coefficient.
Groups analyzed
iNOS immunostaining
× NO metabolite levels
Nonneoplastic tumors (n=14)
r=0.1595,
P=.586
Benign
neoplasia (n=20)
r=0.0942, P=.693
Malignant neoplasia (n=16)
r=0.1188, P=.661
4. Discussion
Differences in iNOS expression between
nonneoplastic, benign and malignant ovarian neoplasias suggest a role for NO in
ovarian carcinogenesis. Our experiments revealed greater expression of iNOS in
malignant ovarian neoplasias than in benign or nonneoplastic tumors. There were
two cases of ovarian cancer that did not show iNOS immunoreactivity, while
approximately one-third of benign and nonneoplastic tumor samples were positive
for iNOS expression. Our results are consistent with other studies that have
shown that a majority of ovarian malignant neoplasias present NOS activity, while
iNOS is detected at lower levels in patients without cancer [23]. Although we
hypothesize that iNOS may be a marker to detect malignant disease, it is not
considered by others to be a marker exclusive to malignant disease [24].
Expression
of NOS in malignant tissue derived from gynecological, breast, central nervous
system, gastric, and colorectal tumors has been reported, suggesting its role
in cancer progression [25–29]. A positive correlation between iNOS expression
and an increased density of tumor microvessels in human colorectal cancer was
shown by Cianchi et al. [28],
and iNOS expression has been associated with increased vascularization and
tumor invasion in endometrial malignant neoplasia [30]. Correspondingly,
patients with lung cancer, prostate cancer, or cervical cancer treated with NO
inhibitors showed antivascular activity [31]. These data support the hypothesis
that the inhibition of iNOS may provide a new therapeutic option for the treatment
of ovarian cancer [32].
Recently, it was demonstrated that iNOS
expression in serous and low-grade carcinomas was significantly higher than in nonserous
and high-grade carcinomas [22]. Advanced stage tumors expressed low levels of
iNOS and were associated with a shorter mean survival time although this was
not determined to be a statistically significant correlation. In our study,
patients with malignant tumors had significantly higher levels of intracystic
NO and iNOS expression in well-differentiated carcinomas compared to moderately/poorly
differentiated carcinomas. In addition, positive iNOS expression in ovarian
carcinoma had been identified as a positive disease-related survival indicator
[33], and was found to be consistent with the high levels of iNOS activity and
NO production of nonmetastatic cells versus metastatic cells [34].
In
contrast to our findings, a separate study found that patients with advanced ovarian serous
tumors express iNOS and COX-2 and experience a shorter disease-free interval
and survival rate [35]. In addition, patients negative for iNOS expression
presented a complete clinical response to a first-line treatment of
chemotherapy. A separate study found a positive correlation of NO synthesis
with tumor progression in a breast cancer model [25].
In
a study by Taveres-Murta et al.,
increased levels of NO metabolites in the tumor microenvironment were found in
patients with ovarian cancer, but not in patients with benign neoplasia [19].
Similarly, supernatants of cell cultures obtained from well-differentiated,
malignant ovarian tumors were found to contain higher levels of NO metabolites
compared to cell cultures from patients with poorly differentiated tumors [36].
However, to our knowledge, this is the first study to evaluate both intracystic
NO production and iNOS expression from the same tumor tissue. Our results
suggest that intracystic NO is not produced by tumor cells since no significant
correlation was found between the levels of NO metabolites detected and the
intensity of iNOS expression detected by immunohistochemical staining of the
tumor tissue. Instead, an analysis of NO production from effusions (ascitic,
cystic, or pleural fluids) of ovarian malignant tumors showed a significant
correlation between the percentage of macrophages present and detectable levels
of NO metabolites, suggesting that macrophages play a significant role in the
production of NO in the tumor microenvironment [36].
An
association between increased intracystic leukocyte infiltrates and NO
production has previously been associated with ovarian cancer [19]. NO produced
by ovarian carcinoma cell lines has also been shown to correlate with the
extent of tumor cell apoptosis observed [12]. These results, in combination
with our data, suggest that the level of iNOS expressed by tumor cells and
infiltrating leukocytes accounts for the intracystic NO production detected
although it cannot be ruled out that the iNOS associated with ovarian tumors
could also be expressed by immune cells [36, 37]. Studies using activated
macrophages treated with NOS-specific inhibitors have shown an inhibition of NO
production and induced cytotoxicity, suggesting that activated macrophages may
mediate NO-dependent cytotoxicity [13].
High
concentrations of NO can mediate cytotoxic activity against tumor cells, while
low concentrations have been associated with angiogenesis. High levels of NO have
also been shown to induce apoptosis. Correspondingly, high concentrations of NO
have not been found to be maintained in many malignant neoplasias [38]. These
opposing actions of NO have been attributed to factors such as differences in
the isoform of NOS expressed, the level of NOS expression, and the type of cell
involved in either in vitro or in vivo
systems [39].
In ovarian cancer
cell lines, high levels of NO donors or strong expression of the iNOS suppressed
survivin levels (a human gene that is part of the inhibitor of apoptosis family), and have been shown to induce apoptosis. It is
hypothesized that NO signaling could contribute to therapy resistance in
epithelial ovarian cancer by modulating survivin expression since low levels of
NO are associated with resistance to carboplatin- and paclitaxel-induced
apoptosis [40]. The ability to modulate NOS gene expression may
represent an opportunity to control the growth and metastasis of tumors in vivo [13].
5. Conclusions
In this study, increased expression of
iNOS and increased production of NO metabolites were detected in the tumor
microenvironment of ovarian cancer samples compared to benign and nonneoplastic
ovarian tissue samples. For ovarian cancer samples, iNOS expression correlated with
the extent of tumor differentiation and intracystic NO metabolite levels
correlated with the tumor stage. We
hypothesize that the absence of a correlation between NO production and iNOS
expression indicates that different cell types are involved in iNOS expression,
and that controlling NO production and inducing NOS may represent valuable strategies
in the prevention of benign tissue transitioning into well-differentiated
malignant ovarian tumors.
Acknowledgments
The authors wish to acknowledge
the funding received from the Conselho Nacional de Desenvolvimento Científico e
Tecnológico (CNPq), the Fundação de Amparo à Pesquisa do Estado de Minas Gerais
(FAPEMIG), and the Financiadora de Estudos e Projetos (FINEP).
American Cancer Society2006New York, NY, USAAmerican Cancer SocietyYancikR.RiesL. G.YatesJ. W.Ovarian cancer in the elderly: an analysis of surveillance, epidemiology, and end results program data19861543639647MurtaE. F. C.Da SilvaC. S.GomesR. A. S.Tavares-MurtaB. M.MeloA. L. K. O.Ultrasonographic criteria and tumor marker assay are good procedures for the diagnosis of ovarian neoplasia in preselected outpatients2004256707712PadillaL. A.padi0013@tc.umn.eduRadosevichD. M.MiladM. P.Limitations of the pelvic examination for evaluation of the female pelvic organs2005881848810.1016/j.ijgo.2004.09.015MurtaE. F. C.eddiemurta@mednet.com.brNomeliniR. S.Early diagnosis and predictors of malignancy of adnexal masses2006181141910.1097/01.gco.0000192967.67567.e9MoncadaS.PalmerR. M. J.HiggsE. A.Nitric oxide: physiology, pathophysiology, and pharmacology1991432109142LiL.KilbournR. G.AdamsJ.FidlerI. J.Role of nitric oxide in lysis of tumor cells by cytokine-activated endothelial cells1991511025312535FukumuraD.dai@steele.mgh.harvard.eduKashiwagiS.JainR. K.The role of nitric oxide in tumour progression20066752153410.1038/nrc1910RidnourL. A.ridnourl@mail.nih.govThomasD. D.DonzelliS.The biphasic nature of nitric oxide responses in tumor biology200687-81329133710.1089/ars.2006.8.1329ChhatwalV. J. S.MoochhalaS. M.ChanS. T. F.NgoiS. S.Nitric oxide and cancer1996461212410.1016/S0306-9877(96)90230-3AkaikeT.MaedaH.Nitric oxide and virus infection2000101330030810.1046/j.1365-2567.2000.00142.xRiederJ.MarthC.TotzkeG.SmolnyM.SeibelM.HoffmannG.Different patterns of inducible nitric oxide synthase gene expression in ovarian carcinoma cell lines200020532513258OhshimaH.BartschH.Chronic infections and inflammatory processes as cancer risk factors: possible role of nitric oxide in carcinogenesis1994305225326410.1016/0027-5107(94)90245-3ShiQ.XiongQ.WangB.LeX.KhanN. A.XieK.kepxie@mail.mdanderson.orgInfluence of nitric oxide synthase II gene disruption on tumor growth and metastasis2000601025792583AndradeS. P.HartI. R.PiperP. J.Inhibitors of nitric oxide synthase selectively reduce flow in tumour-associated neovasculature1992107410921095ChuS. C.Marks-KonczalikJ.WuH.-P.BanksT. C.MossJ.Analysis of the cytokine-stimulated human inducible nitric oxide synthase (iNOS) gene: characterization of differences between human and mouse iNOS promoters1998248387187810.1006/bbrc.1998.9062SerovS. F.ScullyR. E.SobinL. H.19739Geneva, SwitzerlandWorld Health OrganizationInternational Histological Classification of TumoursInternational Federation of Gynecology and Obstetrics (FIGO)Classification and staging of malignant tumors in the female pelvis19715017Tavares MurtaB. M.bmurtafarmaco@dcb.fmtm.brDe Queiróz CunhaF.MirandaR.AdadS. J.Candido MurtaE. F.Differential tumor microenvironment in human ovarian cystic tumors2004905491497SchmidtH. H. H. W.WilkeP.EversB.BöhmeE.Enzymatic formation of nitrogen oxides from L-arginine in bovine brain cytosol1989165128429110.1016/0006-291X(89)91067-XGreenL. C.Ruiz de LuzuriagaK.WagnerD. A.Nitrate biosynthesis in man1981781277647768ÖzelE.PeştereliH. E.ŞimşekT.ErdoğanG.KaraveliF. Ş.Expression of cyclooxygenase-2 and inducible nitric oxide synthase in ovarian surface epithelial carcinomas: is there any correlation with angiogenesis or clinicopathologic parameters?200616254955510.1111/j.1525-1438.2006.00567.xThomsenL. L.LawtonF. G.KnowlesR. G.BeesleyJ. E.Riveros-MorenoV.MoncadaS.Nitric oxide synthase activity in human gynecological cancer199454513521354KlimpA. H.HollemaH.KempingaC.van der ZeeA. G. J.de VriesE. G. E.DaemenT.Expression of cyclooxygenase-2 and inducible nitric oxide synthase in human ovarian tumors and tumor-associated macrophages2001611973057309ThomsenL. L.MilesD. W.HapperfieldL.BobrowL. G.KnowlesR. G.MoncadaS.Nitric oxide synthase activity in human breast cancer19957214144CobbsC. S.BrenmanJ. E.AldapeK. D.BredtD. S.IsraelM. A.Expression of nitric oxide synthase in human central nervous system tumors1995554727730JaiswalM.LarussoN. F.GoresG. J.Nitric oxide in gastrointestinal epithelial cell carcinogenesis: linking inflammation to oncogenesis20012813G626G634CianchiF.cianchif@mail.unifi.itCortesiniC.FantappièO.Inducible nitric oxide synthase expression in human colorectal cancer: correlation with tumor angiogenesis20031623793801IchinoeM.hanabi@med.kitasato-u.ac.jpMikamiT.ShiraishiH.OkayasuI.High microvascular density is correlated with high VEGF, iNOS and COX-2 expression in penetrating growth-type early gastric carcinomas200445661261810.1111/j.1365-2559.2004.02031.xLiW.wei5901482@163.netXuR.-J.JiangL.-H.ShiJ.LongX.FanB.Expression of cyclooxygenase-2 and inducible nitric oxide synthase correlates with tumor angiogenesis in endometrial carcinoma2005221637010.1385/MO:22:1:063NgQ.-S.GohV.MilnerJ.Effect of nitric-oxide synthesis on tumour blood volume and vascular activity: a phase I study20078211111810.1016/S1470-2045(07)70001-3MaloneJ. M.jmalone@dmc.orgSaedG. M.DiamondM. P.SokolR. J.MunkarahA. R.The effects of the inhibition of inducible nitric oxide synthase on angiogenesis of epithelial ovarian cancer200619441110111610.1016/j.ajog.2005.12.019AnttilaM. A.maarit.anttila@kuh.fiVoutilainenK.MerivaloS.SaarikoskiS.KosmaV.-M.Prognostic significance of iNOS in epithelial ovarian cancer200710519710310.1016/j.ygyno.2006.10.049XieK.DongZ.FidlerI. J.Activation of nitric oxide synthase gene for inhibition of cancer metastasis1996596797803RaspolliniM. R.AmunniG.VillanucciA.Expression of inducible nitric oxide synthase and cyclooxygenase-2 in ovarian cancer: correlation with clinical outcome200492380681210.1016/j.ygyno.2003.12.023ThomsenL. L.LLT36851@ggr.co.ukSargentJ. M.WilliamsonC. J.ElgieA. W.Nitric oxide synthase activity in fresh cells from ovarian tumour tissue: relationship of enzyme activity with clinical parameters of patients with ovarian cancer199856101365137010.1016/S0006-2952(98)00187-7ShermanP. A.LaubachV. E.ReepB. R.WoodE. R.Purification and cDNA sequence of an inducible nitric oxide synthase from a human tumor cell line19933243116001160510.1021/bi00094a017HellerA.heller@che.utexas.eduApoptosis-inducing high ⋅NO concentrations are not sustained either in nascent or in developed cancers20083101493149910.1002/cmdc.200800257XieK.FidlerI. J.Therapy of cancer metastasis by activation of the inducible nitric oxide synthase1998171557510.1023/A:1005956721457EngelsK.KnauerS. K.LoiblS.NO signaling confers cytoprotectivity through the survivin network in ovarian carcinomas200868135159516610.1158/0008-5472.CAN-08-0406