| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Tumor Biology |
Department of Pharmacology, University of Melbourne, Victoria 3010 [T. E. K., T. L. B., E. G., A. G. S.]; Bernard O’Brien Institute of Microsurgery, St. Vincent’s Hospital, Victoria 3065 [J. E. B.]; and Peter MacCallum Cancer Institute, Victoria 8006 [R. L. A.] Australia
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The promotion of tumor growth by NO in other models (1 , 4) may also involve the enhancement of angiogenesis. Angiogenesis is known to be essential for solid tumor growth (5) with a major angiogenic stimulus provided by VEGF. VEGF produced by tumor cells or administered exogenously requires a functional endothelial NO/cGMP pathway to promote angiogenesis (2) . Thus, selective inhibition of endothelial NO production would be predicted to reduce tumor angiogenesis. On the other hand, loss of the cytotoxic tumoricidal actions of NO would decrease tumor rejection and apoptosis (6) . The latter action would tend to oppose any effect of NOS inhibitors on tumor growth via suppression of angiogenesis.
The analysis of these discrepant results is also complicated by the enzyme isoform producing the NO. The NO required for VEGF-mediated endothelial mitogenesis is most probably provided by the constitutive endothelial isoform NOS I, whereas the NO involved in cytotoxicity and tumor inhibition is more likely to be provided by the inducible isoform, NOS II. This isoform has been found in tumors (2 , 7 , 8) , and its induction is a marker of inflammatory reactions (9) . Although the potential influence of NO derived from these two NOS isoforms could be investigated with inhibitors, those presently available lack the degree of isoform selectivity that would allow unequivocal interpretation in vivo, although good in vitro selectivity is shown (10) . The alternative approach of using animals lacking NOS II gene requires the availability of the gene deletion in a mouse strain syngeneic to the B16 melanoma but avoids the contentious issues arising from the use of inhibitors in vivo such as mode of treatment, dose given, duration of dosing, and selectivity. We have examined the growth of B16-F1 melanoma cells in an extensively phenotyped C57Bl/6 mouse strain lacking NOS II (11 , 12) .
NOS2-knockout (NOS II-/-) mice display a phenotype consistent with a loss of the cytotoxic actions of NO; they are susceptible to infection and show impaired macrophage cytotoxicity for parasites and tumor cells but otherwise appear to be normal in a pathogen-free environment (11) . We have demonstrated that angiogenesis-dependent skin flap survival (13) is reduced by NOS II-selective inhibitors in the rat and by NOS II gene provided by this isoform (14) . Therefore, in the current study, the relationship between NO derived from NOS II, tumor growth, and VEGF expression has been investigated using NOS II-/- mice. In this study, we report the novel finding that tumor growth and expression of VEGF165 mRNA levels in B16-F1 tumors are reduced in NOS II-/- mice.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MEM (2 mM L-glutamine) supplemented with heat-inactivated FCS (10% v/v). RBL-2H3 cells, which show mast cell-like characteristics (16)
, were maintained in culture in DMEM (2 mM L-glutamine) supplemented with FCS (5% v/v). The B16-F1 cell line was verified to be free from Mycoplasma contamination every 3 months using Mycoplasma T. C. Rapid detection system (Gene Probe).
Proliferation Studies.
B16-F1 cells were seeded into 24-well plates at a density of 5 x 103 cells/cm2 and grown to 70% confluency over 24 h. Cells were then rendered quiescent by incubation in FCS-free medium for 24 h before restimulation by FCS (1% v/v). L-NIL (0–100 µM) or deta NONOate (0–100 µM) were added 30 min before FCS and were left in contact with cells for 24 h before the [3H]thymidine pulse (1 µCi/ml) for 4 h. ONOO- (0–10 µM) was added 30 min before FCS and [3H]thymidine pulse and left in contact with the cells for 28 h. Details of the [3H]thymidine incorporation assay have been described elsewhere (17)
.
Tumor Growth.
Breeding pairs of mice that lacked the NOS II-/- and wild-type mice (NOS II+/+) were obtained from a colony established at the John Curtain School of Medical Research (Canberra, Australia). This colony was derived from breeding pairs generously donated by Prof. Carl Nathan (Cornell University, New York, NY). The animals have been subsequently bred under specific pathogen-free conditions at the Bernard O’Brien Institute of Microsurgery Animal House Facility (Melbourne, Australia). The NOS II-/- mice (background 129/SvEv x C57BL/6) were generated by gene targeting in embryonic stem cells as described previously (11)
. Homozygosity of NOS II-/- was repeatedly confirmed by PCR by the absence of NOS II mRNA and by the absence of NOS II immunostaining.5
NOS II+/+ control mice had the same strain background as NOS II-/- mice; i.e., 129/SvEv x C57BL/6. Mice were maintained in a pathogen-free environment with 12 h of darkness daily and an ambient air temperature of 22°C ± 2°C. The care and use of the mice followed guidelines set forth by The Australian Code of Practice and Animal Welfare with the approval of the St. Vincent’s Hospital Animal Ethics Committee. Male and female NOS II-/- and NOS II+/+ mice from separate homozygous colonies (12–19 weeks of age) were anesthetized by inhalation of penthrane and injected s.c. with 2 x 106 B16-F1 cells in a volume of 200 µl above the right hind limb. Tumors were harvested on day 14, and the wet weights were measured as an index of tumor growth. The tumors then either underwent RNA extraction using TRIzol (Life Technologies, Inc.) after homogenization in liquid nitrogen or were fixed in 4% paraformaldehyde and embedded in paraffin-wax for immunohistochemistry.
Detection of NOS II and VEGF.
The three layer alkaline phosphatase immunohistochemical method was used. Tissue sections (3 µm) were stained with monoclonal mouse antihuman VEGF (1:160 in 0.25% BSA/PBS; Genentech) or polyclonal rabbit antihuman NOS II (1:100 in 0.5% BSA/PBS; Transduction Laboratories). Biotin-labeled horse antimouse immunoglobulin secondary antibody (0.5% antibody in 1.5% normal horse serum/1% normal mouse serum/PBS; Vectastain) was used for VEGF, and biotin-labeled swine antirabbit immunoglobulin antibody (1:400 in PBS; Dako) was used for NOS II. Sections were further incubated with streptavidin conjugated to alkaline phosphatase (1:1500 in 0.25% BSA/PBS; Silenus, Australia), and subsequent specific staining was visualized with Fast Red TR/Napthol AS-MX, counterstained with Mayer’s hematoxylin and mounted in glycerol mount. Mast cells were identified using toluidine blue (0.1%) in morphologically related sections.
ONOO- Stimulation.
RBL 2H3 cells were grown to confluence in 6-well plates and washed twice in HEPES-buffered Tyrode salt solution at pH 7.2. ONOO- was synthesized and quantified spectrophotometrically (
302 1670 M-1 x cm-1) as described (18)
. ONOO- (1 µM or 10 µM), which is stable in the vehicle of 1 M NaOH, was added to cells bathed in HEPES-buffered Tyrode salt solution. The solution was rapidly mixed and left at room temperature for 5 min. Cells were then washed twice with HEPES-buffered Tyrode salt solution and incubated with 1 µM A23187 for 4 h in DMEM with 0.25% BSA. RNA was extracted from RBL-2H3 cells using TRIzol.
RBL-2H3 Cell Degranulation.
RBL-2H3 cells were incubated with [3H]5-HT (1 µCi/ml; 20–24 h) to label granule amine stores. At the end of this period, the medium was aspirated, and the cells were washed twice in PBS to remove nonincorporated label and then incubated in HEPES-buffered Tyrode salt solution. To elicit [3H]5-HT release, cells were exposed to A23187 (1 µM; 20 min; 37°C). Cells were exposed to ONOO- (1 µM-30 µM) for 5 min at room temperature 20–22°C in HEPES-buffered Tyrode salt solution at pH 7.2 10 min before A23187 or vehicle exposure. None of these treatments caused detectable cytotoxicity as evidenced by a lack (less than 5% total cell content) in lactate dehydrogenase released into cell supernatants.
Northern Analysis.
RNA (10 µg) was resolved on a 1.2% formaldehyde denaturing gel, transferred to an Immobilon-Ny+ nylon membrane (Millipore) and hybridized with a 32P-labeled full-length human VEGF A cDNA probe (a gift from Genentech). The membranes were hybridized overnight at 65°C, washed in 2 x SSC + 0.1% SDS at 55°C for 30 min, and 32P was detected in a Fuji Bas 1000 phosphorimager after 5 days of plate exposure. The membranes were stripped and hybridized with a GAPDH chicken cDNA probe (19)
and exposed overnight as described above. VEGF expression data were obtained by normalization to the signal for GAPDH.
Statistical Analysis.
Comparison between tumor weights was performed using the Mann-Whitney U test, because the data did not conform to a normal distribution. Tumor growth less than 2 mm3 was analyzed using a 
2 test. Data from in vitro experiments were analyzed using one-way ANOVA, and Dunn’s post hoc test was applied to identify individual differences when the ANOVA was significant. Comparisons were considered to be statistically significant when P < 0.05.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. In NOS II+/+ mice, the range of tumor weights was 0.02–3.13 g, and for tumors from NOS II-/- mice, the range was 0–2.33 g (P < 0.0001, Mann-Whitney U test). Generally, tumors require angiogenesis to grow beyond a size of 2 mm3 (5)
. In NOS II-/- mice, 10 of 44 had no detectable tumor mass, whereas in the NOS II+/+ mice, each of the 45 mice had detectable tumors, and only 2 of these were less than 2 mm3 (P < 0.05; 2 test).
|
. NOS II immunoreactivity was identified in tumor cells within the center and at the edge of tumors from NOS II+/+ mice but not in tumor cells growing in NOS II-/- mice (Fig. 2, C, D, E, and F)
.
|
and NOS II-/- mice (Fig. 2H)
and was not identified in any tumor cells. The number of immunoreactive VEGF positive cells within tumors from either NOS II+/+ or NOS II-/- mice was similar (data not shown). Because of the difficulty of accurate quantitation of high intensity granular immunohistochemical staining, no attempt was made to estimate whether the intensity of VEGF immunostaining was altered in the NOS II-/- strain. Northern blot analysis for VEGF mRNA was performed on RNA extracted from B16-F1 tumors removed after 14 days from 15 NOS II+/+ and 15 NOS II-/- mice. VEGF mRNA levels were significantly lower in tumors from NOS II-/- when compared with NOS II+/+ mice (P < 0.05; Fig. 3
).
|
. Therefore, we assessed if proliferation of these cells in culture could be directly stimulated by NO. DNA synthesis by B16-F1 cells was increased 4-fold after incubation with 1% FCS but was unaffected by the NOS II selective inhibitor, L-NIL at concentrations up to 100 µM (Table 1)
. Moreover, cultured B16-F1 melanoma cells did not produce detectable levels of the stable end products of NO oxidation, because NO3- levels were less than 1 µM after 48 h. The effects of exogenous NO and its by-product on DNA synthesis were assessed with authentic ONOO- and the NO donor deta NONOate. DNA synthesis was not affected by either 1–10 µM or 3–100 µM deta NONOate (Tables 1
and 2
).
|
|
).
|
, consistent with its localization in tumor-associated mast cells in vivo. This observation raised the possibility of VEGF release by exocytosis during degranulation. Thus, in the in vitro experiments, degranulation of RBL-2H3 cells was stimulated with the calcium ionophore A23187 and measured by release of preincorporated [3H]5-HT. Pretreatment of RBL-2H3 cells with ONOO- for 5 min before exposure to A23187 resulted in an enhancement of release of preincorporated [3H]5-HT (Fig. 4)
.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
However, not all of the data in the literature are consistent with NO-increasing tumor growth. Transfection of the K-1735 murine melanoma cells with the NOS II gene decreased the rate of growth of primary tumors by inducing apoptosis (20) . Pipili-Synetos et al. (3) demonstrated that the NO-releasing compounds isosorbide mononitrate or isosorbide dinitrate inhibited tumor growth and pulmonary metastasis in mice that received injections i.v. with Lewis lung carcinoma.
The proposition that NOS II activity promotes tumor growth is supported by several studies (1 , 2 , 4 , 21) demonstrating that ectopic expression of NOS II in tumor cells increases growth rates. However, these studies focused on NOS II expression in the tumor cells themselves, whereas we have shown that the host capacity for expression of NOS II is an important determinant of tumor growth. Shi et al. (22) found that B16-BL6 melanoma cells produced fewer and smaller metastases in NOS II-/- mice compared with the NOS II+/+ strain, a finding consistent with our results on primary tumor growth. Our results also demonstrated that it is the presence of NOS II in the host that allows tumor cells to express NOS II, because B16-F1 tumor cells in NOS II-/- mice did not express NOS II, although the tumor cells did so in NOS II+/+ mice. The simplest explanation for this unexpected finding is that the signal(s) for NOS II induction in B16-F1 tumor cells depend on NOS II expression by the host. Our attempts to induce NOS II in B16-F1 cells in culture by exposure of cells to a variety of cytokines were unsuccessful. Thus, the signals from NOS II+/+ mice for NOS II expression in the tumor are likely to be complex and specific.
Currently, the number of murine in vivo studies supporting pro-tumor effects of NO far outweigh the number against. Dong et al. (23) demonstrated how a combination of pro-inflammatory cytokines produced increased NO-mediated cytotoxicity against nonmetastatic cells while not affecting viability of metastatic cells. NO has been shown to stimulate p53 accumulation, which blocks cell proliferation causing a cytostatic effect. Reduced sensitivity to the cytostatic effects of NO in certain tumor cell lines appears to be explained by the expression of dysfunctional forms of p53 (23) . The B16-F1 melanoma cells were insensitive to NO-mediated cytotoxicity, but their p53 status is unknown. The enhancement of B16-F1 tumor growth by NOS II activity may be explained by the balance of influence of NO favoring angiogenesis (pro-growth) rather than cytostasis (anti-growth). Furthermore, the variance in the literature with respect to the impact of NO on tumor growth could potentially be explained by the p53 status of tumor cell lines.
Host-tumor signaling may also underlie the localization of NOS II in the tumor. NOS II immunoreactivity was differentially distributed within the B16-F1 tumor tissue of NOS II+/+ mice, being more frequently observed at the invasive edge, than in the tumor core. This pattern of NOS II distribution was also seen in tumors from head and neck cancer patients (2) . The increased NOS II activity correlated with tumor vascularization (2) , consistent with NO derived from the NOS II enzyme having a pro-angiogenic function.
The importance of angiogenesis in melanoma growth is controversial. Increased vascular density has been associated with a higher incidence of metastases and a worse prognosis for patients with different malignant neoplasms, including lung cancer, prostate cancer, and malignant melanoma (24) . However, there are also studies on cutaneous melanoma, uveal melanoma, and some other tumors in which no correlation between vascular density and overall survival was detected (25 , 26) . Folberg et al. (27) reported highly aggressive and metastatic melanoma cells are capable of forming highly patterned vascular channels in vitro that are composed of a basement membrane that stain with the periodic acid-Schiff reagent in the absence of endothelial cells and fibroblasts.
Recently, it was shown using skin window chambers in rodents that angiogenesis induced by tumor cells after implantation in the host begins at a very early stage, i.e., when the tumor mass contains roughly 100–300 cells (28) . Further work needs to be performed to monitor events at the initiation of angiogenesis at the outset of solid tumor growth in vivo. The influence of NOS II may be at these early angiogenic events because a proportion of tumors in NOS II-/- mice failed to grow to a macroscopic mass. Those tumors that did grow in NOS II-/- mice were of indistinguishable size to those growing in NOS II+/+ mice. Thus, it is difficult to predict whether the microvascular density would differ in those tumors in which it was measurable.
VEGF is known to be localized to macrophages and other nonendothelial cell types in sites of active angiogenesis (29) . In our study, immunoreactive VEGF was localized to mast cells in tumors from both NOS II+/+ and NOS II-/- mice. Mast cells have been implicated in the pathogenesis of various diseases that are accompanied by neovascularization (30) including tumors (31) . Only relatively recently, VEGF has been identified in human mast cells (30) , and its release after immunological stimulation has been demonstrated (32) . Although VEGF protein was detected in mast cells within tumors in either strain, VEGF mRNA expression was clearly diminished in the tumors in the NOS II-/- mice. This finding strongly suggests that NO regulates mast cell expression of VEGF. Similarly, Chin et al. (33) found NO released from a NO donor could increase VEGF mRNA levels in human carcinoma cells through a guanylate cyclase/cGMP-dependent mechanism. Thus, NO may function as an upstream regulator of expression as well as a downstream effector of the action of VEGF (34) . The RBL-2H3 cell line exhibits many of the characteristics of mast cells (16) , including the storage of VEGF in cytoplasmic granules. In contrast to the human carcinoma cell lines (33) , exogenous NO derived from the donor, deta NONOate, had no effect on mRNA for VEGF in cultures of RBL-2H3 cells.
In ischemic, inflammatory, or angiogenic environments such as those in the tumor, superoxide anion may be produced by mast cells (35) or by infiltrating macrophages (36 , 37) . Synthesis of NO in the presence of superoxide anion leads to production of the highly reactive cytotoxic agent ONOO-, which is known to be produced in increased amounts in inflammatory sites (38) . It was for this reason that we assessed VEGF expression (mRNA levels) and degranulation ([3H]5-HT release) after exposure of RBL-2H3 cells to ONOO-. When RBL-2H3 cells were exposed to ONOO- rather than NO, an increase in VEGF mRNA expression was observed, and A23187-stimulated 5-HT release was enhanced. Therefore, it appears that ONOO- and not NO could enhance VEGF availability in these tumors. The half-life of ONOO- at physiological pH is typically less than 2 s (39) . In view of this short half-life, exposure of the cells to a 10 µM bolus of ONOO- in this and in previous studies (40) is considered to be equivalent to a steady state exposure to very low concentrations of ONOO- (41) . Nevertheless, because the kinetics of ONOO- production in situ are not established, further investigation of different protocols for exposure of mast cells to ONOO- would provide additional insights. In our melanoma model, it is not possible to distinguish between host NOS II or tumor cell NOS II as the generator of ONOO- that may function to enhance VEGF release from mast cells. A study of NOS II-expressing tumors in NOS II-/- hosts could elucidate this point.
In summary, our experiments have revealed control of tumor growth exerted by the host ability to provide NO from NOS II. The other constitutive isoforms of NOS were not able to substitute for the inducible enzyme. This partial dependence of tumor growth on NOS II activity reinforces the importance of NOS II as a therapeutic target in cancer. The observed growth control may be mediated by increased VEGF availability through the production of ONOO- rather than NO itself.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported in part by NHMRC (Australia), Project 9936747. 
2 Present address: Leukocyte Biology Section, Imperial College, South Kensington, London, United Kingdom, SW72AZ.
3 To whom requests for reprints should be addressed, at Department of Pharmacology, University of Melbourne, Victoria 3010, Australia. Phone: (613) 8344-5675; Fax: (613) 8344-0241; E-mail: a.stew{at}clyde.its.unimelb.edu.au
4 The abbreviations used are: NO, nitric oxide; NOS, NO synthase; NOS II, inducible NOS; VEGF, vascular endothelial growth factor; ONOO-, peroxynitrite; 5-HT, serotonin; cGMP, cyclic GMP; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; L-NIL, L-N6-(1-iminoethyl) lysine.
Received 9/15/00. Accepted 1/30/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
receptor I expression. J. Exp. Med., 188: 1135-1145, 1998.This article has been cited by other articles:
|
|
 |
|
  J. Chen, Y. Yan, J. Li, Q. Ma, G. D. Stoner, J. Ye, and C. Huang Differential requirement of signal pathways for benzo[a]pyrene (B[a]P)-induced nitric oxide synthase (iNOS) in rat esophageal epithelial cells Carcinogenesis, June 1, 2005; 26(6): 1035 - 1043. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Mocellin, M. Provenzano, C. R. Rossi, P. Pilati, R. Scalerta, M. Lise, and D. Nitti Induction of Endothelial Nitric Oxide Synthase Expression by Melanoma Sensitizes Endothelial Cells to Tumor Necrosis Factor-Driven Cytotoxicity Clin. Cancer Res., October 15, 2004; 10(20): 6879 - 6886. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ostendorf, C. van Roeyen, R. Westenfeld, A. Gawlik, M. Kitahara, E. de Heer, D. Kerjaschki, J. Floege, and M. Ketteler Inducible Nitric Oxide Synthase-Derived Nitric Oxide Promotes Glomerular Angiogenesis via Upregulation of Vascular Endothelial Growth Factor Receptors J. Am. Soc. Nephrol., September 1, 2004; 15(9): 2307 - 2319. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D.-E. Hu, S. O. M. Dyke, A. M. Moore, L. L. Thomsen, and K. M. Brindle Tumor Cell-Derived Nitric Oxide Is Involved in the Immune-Rejection of an Immunogenic Murine Lymphoma Cancer Res., January 1, 2004; 64(1): 152 - 161. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. R. Kisley, B. S. Barrett, A. K. Bauer, L. D. Dwyer-Nield, B. Barthel, A. M. Meyer, D. C. Thompson, and A. M. Malkinson Genetic Ablation of Inducible Nitric Oxide Synthase Decreases Mouse Lung Tumorigenesis Cancer Res., December 1, 2002; 62(23): 6850 - 6856. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
) and NOS II-/- mice (n = 45, 
). Mice between 12–19 weeks of age received injections with 2 x 106 B16F1 cells. The median tumor weights were 0.74 g for NOS II+/+ and 0.15 g for NOS II-/- mice, as indicated by the box and whisker plots. *, P < 0.0001; Mann-Whitney U test.
