Hypoxia-inducible factor 1 alpha (HIF-1α) gene expression in human ovarian carcinoma

Cancer Letters 176 (2002) 215–223 www.elsevier.com/locate/canlet

Hypoxia-inducible factor 1 alpha (HIF-1a ) gene expression in human ovarian carcinoma Kentaro Nakayama a,b, Atsuko Kanzaki a, Kohkichi Hata b, Hidetaka Katabuchi c, Hitoshi Okamura c, Kohji Miyazaki b, Manabu Fukumoto a, Yuji Takebayashi a,* a

Department of Pathology, Institute of Development, Aging and Cancer, Tohoku University, 4-1 Seiryo-machi, Aoba-ku, Sendai 980-8570, Japan b Department of Obstetrics and Gynecology, Shimane Medical University, Izumo 693-8501, Japan c Department of Obstetrics and Gynecology, Kumamoto University School of Medicine, Kumamoto 860-8556, Japan Received 9 August 2001; received in revised form 31 August 2001; accepted 5 September 2001

Abstract Hypoxia-inducible factor 1a (HIF-1a) that regulates genes involved in response to hypoxia and promotes neo-angiogenesis, is a transcriptional factor for vascular endothelial cell growth factor (VEGF). The aim of this study was to examine the expression of HIF-1a and VEGF gene expressions and their relation to angiogenesis, clinicopathologic variables and survival in the patient with human ovarian carcinoma. We retrospectively analyzed HIF-1a and VEGF gene expression levels using reverse transcriptase polymerase chain reaction (RT-PCR) in 60 ovarian carcinomas. Intratumoral microvessel density (IMD) was assessed by immunostaining endothelial cells, using anti-CD 31 antibody in frozen sections. The relationships between the expression level of these genes, IMD and clinicopathologic variables were evaluated by Student’s t-test and chi-square tests. Survival analysis was performed by Kaplan–Meier curves. HIF-1a or VEGF gene expression level was independent of age, clinical stage and histological subtype besides grade of tumor. There was no relationship between HIF-1a or VEGF gene expression level and IMD in all carcinomas (R ¼ 0:118 and 0.224, respectively). In addition, a weak association between HIF-1a and VEGF gene expression level was observed (R ¼ 0:300, P ¼ 0:020). The association between VEGF gene expression and IMD was observed (R ¼ 0:501, P ¼ 0:016). However, no association between IMD and HIF-1a gene expression was observed. Further, both HIF-1a and VEGF gene expression levels had no effect on survival in the patient with ovarian carcinoma. These results suggest that VEGF upregulated by HIF-1a gene may be involved in angiogenesis of some type of ovarian carcinoma, but the expression levels of both genes have no effect on survival in the patients with ovarian carcinoma. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Ovarian carcinoma; Hypoxia-inducible factor 1 alpha; Angiogenesis; Intratumoral microvessel density; Vascular endothelial cell growth factor

1. Introduction Several basic and clinical studies have indicated that tumor growth is dependent on angiogenesis [1– * Correspondence author. Tel.: 181-22-717-8511; fax: 181-22717-8512. E-mail address: [email protected] (Y. Takebayashi). Abbreviations: IMD, intratumoral microvessel density

6]. Vascularization supplies nutrition and oxygen to proliferating cells. Apoptosis induced by nutrient deficiency counter balances cell proliferation, and limits tumor growth [7–9]. Clonal evolution of tumor cells in hypoxic microenvironment result(s) from selection of subpopulation(s) that not only resist apoptosis [10], but also promote the formation of new blood vessel [11]. In addition, further growth of the primary tumor,

0304-3835/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(01)00762-5

216

K. Nakayama et al. / Cancer Letters 176 (2002) 215–223

cellular adaptation to hypoxia and tumor neovascularization strongly correlate with the risk of invasion and metastasis [12,13]. Hypoxia-inducible factor 1 (HIF1) is a heterodimeric transcription factor that regulates O2 homeostasis and physiological response to O2 deprivation [14,15]. HIF-1consists of two subunits, HIF-1a and HIF-1b, that belongs to a subfamily of basic helix–loop–helix (bHLH) transcription factors containing a Per-ARNT-Sim (PAS) motif [16]. A decrease in cellular O2 tension leads to elevation of HIF-1 activity via stabilization of the HIF-1a protein, conversely, ubiquitin-mediated proteolysis of HIF-1a on reexposure to normoxic environment results in rapid decay of HIF-1 activity [17–20]. Further, HIF1a mediates the angiogenesis by its induction of vascular endothelial growth factor (VEGF) [21]. VEGF is a novel angiogenic factor, and its expression may be prognostic indicator in some types of human solid carcinomas [22–25]. Therefore, HIF-1a gene expression may be supposed to play an important role in VEGF-mediated angiogenesis in human solid carcinomas. Ovarian carcinoma is the leading cause of deaths from gynecologic malignancy. The incidence and mortality of ovarian carcinoma has not been improved in the past decade. This is due to late diagnosis, usually occurring during the late stages of this disease. The abundant molecular biological analysis of oncogenes and tumor suppressor genes as well as angiogenesis and angiogenic factors has been analyzed, in order to find the new target(s) for the improvement of prognosis of patients with ovarian carcinoma. Overexpresssion of HIF-1a protein has been demonstrated in a variety of human carcinoma [26]. Recently, Birner et al. have reported that HIF-1a expression was an independent prognostic marker in early stage cervical carcinoma [27]. The aim of this study was to investigate the clinicopathologic significance of HIF-1a expression, using reverse transcriptase polymerase chain reaction (RT-PCR) in human ovarian carcinoma.

2. Materials and methods 2.1. Patients and tissue specimens Tumors were obtained from 60 patients with ovarian carcinoma underwent surgery at the Department

of Obstetrics and Gynecology, Shimane Medical University Hospital and Kumamoto University Hospital between January 1991 and December 2000. Informed consent was obtained from each patient, prior to this study. All the samples were embedded in OCT compound (Sakura Finetechnical Co., Ltd, Tokyo) and stored at 2808C, immediately until processing. The patients ranged in age from 19 to 82 years with an average age of 54.5 years. None of the patients had undergone preoperative radiotherapy or chemotherapy. Patients received postoperative therapy; however, there was no difference in outcome among the various treatment modalities. The patients had no other form of malignancies. All patients were primarily treated with reduction surgery (Transabdominal hysterectomy, bilateral salpingoophorectomy or modified radical hysterectomy) and postoperative chemotherapy, which consisted of a platinum-based regimen (CAP; Cycrophosphamide, Adriamycin and Cisplatin). Clinical staging for primary ovarian carcinoma was performed according to the International Federation of Obstetrics and Gynecology [28], and tumors are histologicaly classified according to World Health Organization criteria [29]. Follow-up for all the patients included in the survival analysis was updated in April 25, 2001 (median follow-up was 40.6 months; range, 6–128 months). At that time, 15 patients had died of ovarian carcinoma and 45 were alive. The characteristics of the patients with ovarian carcinoma are summarized in Table 1. 2.2. RT-PCR Total RNA from human ovarian carcinoma was prepared by Trizol (Gibco, Life Tech, CA, USA). cDNA was synthesized with 3 mg of total RNA ,random hexadeoxynucleotide primer (Gibco, Life Tech, CA, USA) in 20 ml of a solution containing RT. Then cDNA was diluted 1:5 with water and stored at 2208C until use. PCR was performed with cDNA derived from 30 ng RNA. PCR reactions were carried out in a total volume of 25 ml containing cDNA and dNTPs at concentrations of 200 and 0.4 mM, respectively, of each primer and 0.25 unit of ExTaq polymerase (Takara Shuzo, Japan). The PCR condition consisted of 10 min at 948C followed by 35 cycles of 30 s at 948C, 30 s at 558C and 1 min at 728C, followed by 728C for 10 min. PCR products were

K. Nakayama et al. / Cancer Letters 176 (2002) 215–223

217

Table 1 Relationship of HIF-laVEGF gene expression and clinicopathological variables in patients with ovarian carcinoma a Variables

Total

All patients

HIF-la

VEGF

Low

High

53 22–81

55 19–87

23 37

12 (52%) 18 (49%)

11 (48%) 19 (51%)

Histology Serous Mucinous Others

29 17 14

14 (48%) 10 (59%) 6 (43%)

15 (52%) 7 (41%) 8 (57%)

Grade LMP Well Moderate Poor

5 17 16 22

2 (40%) 14 (82%) 5 (31%) 9 (41%)

3 (60%) 3 (18%) 11 (69%) 13 (59%)

c

High

53 22–81

55 19–87

14 (61%) 16 (43%)

9 (39%) 21 (57%)

18 (62%) 7 (41%) 5 (36%)

11 (38%) 10 (59%) 9 (64%)

2 (40%) 11 (65%) 10 (63%) 7 (32%)

3 6 6 15

NS

Stage (FIGO) I, II III, IV

a

Low

Significance

60

Age (years) Median Range

b

Significance

NS

NS

NS

NS

NS

NS 0.003 b 0.009 c

NS (60%) (35%) (37%) (58%)

0.041 b NS

NS, no significance. Well vs. moderate. Well vs. poor.

subjected to electrophoresis in 2% agarose gel. The PCR primer sequences of HIF-1a, VEGF and glyceraldehyde-3-A phosphate-dehydrogenase (GAPDH) which was used as an internal control were as follows: sense primer for HIF-1a, 5 0 -CTCAAAGTCGGACAGCCTCA-3 0 and antisense primer for HIF-1a, 5 0 -CCCTGCAGTAGGTTTCTGCT-3 0 corresponding to 45 4 bp (residues 2248–2702); sense primer for VEGF, 5 0 -CCTCCGAAACCATGAACTTT-3 0 and antisense primer for VEGF, 5 0 -AGAGATCTGGTTCCCGAAAC-3 0 corresponding to 637 bp (residues to 89–726); sense primer for GAPDH, 5 0 -CCCCTGGCCAAGGTCATCCATGACAACTTT-3 0 and antisense primer for GAPDH, 5 0 -GGCCATGAGGTCCACCACCCTGTTGCTGTA-3 0 corresponding to 513 bp (residues 515–1027). 2.3. Quantitative detection of mRNA and PCR In order to evaluate the amplified PCR products semi-quantitatively, the optimal conditions for detection and quantification of HIF-1a and GADPH genes were determined using cDNA derived from placenta.

At 40 cycles of PCR in each molecule, the relative yields of PCR products are similar, indicating that this number of cycles corresponds to plateau. At 25 cycles or less, each gene expression cannot be clearly distinguished (data not shown). We have also performed same experiments using several ovarian carcinoma tissues. The results were similar (data not shown). Therefore we used 35 PCR cycles for the detection of each gene with these findings. The amplified cDNA fragment was electrophoresed in 2% agarose gel containing 0.4 mg of ethidium bromide/ml, visualized by UV, and quantified using NIH Image 161. 2.4. Microvessel staining CD34 monoclonal antibody was used to immunostain vessels in ovarian carcinoma using the standard immunoperoxidase procedure (Vectastain Elite ABC kit, Vector, Burlingame, CA). After sectioning the tissues at 2.5 mm thickness recovered from O.C.T. compound, the sections were fixed in 10% neutral buffered formalin. Endogenous peroxidase in the sections was blocked by incubating them in 0.03%

218

K. Nakayama et al. / Cancer Letters 176 (2002) 215–223

H2O2 in absolute methanol for 30 min at room temperature. Each section was incubated with 3% of skim milk in phosphate buffered saline (PBS) for 30 min at room temperature. The sections were incubated with anti-CD34 monoclonal antibody (1:1000 diluted) overnight at 48C. The sections were equilibrated to room temperature, and then rinsed six times every 5 min in PBS. The sections were incubated in biotinylated horse anti-mouse IgG at 1:200 with 1.5% normal horse serum for 30 min at room temperature. After washing them for six times every 5 min in PBS, they were incubated for 30 min at room temperature in the avidin–biotin horseradish peroxidase macromolecular complex. After washing for six times every 5 min in PBS, the sections were incubated for 6 min in 0.05% diaminobenzidine in PBS with 0.03% H2O2. The slides were rinsed in deionized water and counterstained with hematoxylin, dehydrated and mounted. 2.5. Evaluation of microvessel density After screening the areas with intense neovascularization at low power ( £ 100 field), microvessels in the areas with the highest number of intratumoral microvessels (IMD) were assessed in a £200 field, because the microvessel density was more precise than in a £ 100 field. We carefully observed the shape of the stained objectives. Because CD34 occasionally stains tumor cell, stromal fibroblast and inflammatory cell, the cell without a visible lumen and with the round and bigger nucleus was excluded from the microvessel (Fig. 1B). Two investigators (K.N. and Y.T.) simultaneously assessed the microvessel density without knowledge of the clinicopathologic factors by the two-head light microscopy. When two investigators (K.N. and Y.T.) did not reach the same results, they observed them on the TV-captured images and counted them again together. 2.6. Statistical analysis Associations of clinical variables were evaluated using the Student’s t-test or chi-square test. Survival curves were determined by Kaplan–Meier method and the prognostic significance was evaluated by log–rank test. Two-sided P values were calculated and was considered significant when less than 0.05.

3. Results 3.1. Expression of HIF-1a or VEGF gene Fig. 1A is representative of PCR results of HIF-1a and VEGF genes in human ovarian carcinoma. The expression levels of HIF-1a gene arranged in order of magnitude of each case were given in Fig. 1C,D. Both HIF-1a and VEGF gene expression were detected in 71.7% (43/60) and distributed through low to high expression levels, respectively (Fig. 1C,D). After categorizing low and high expression of HIF-1a or VEGF gene by their median values, we examined the relationship between clinicopathologic variables and each gene expression level. HIF-1a and VEGF gene expression level were independent of age, clinical stage and histological subtype. However, the expression level of these genes in moderately or poorly differentiated carcinoma was significantly higher than that in well-differentiated carcinoma (P ¼ 0:027, 0.004 and 0.041, respectively; Table 1). 3.2. Relationships between expression of HIF-1a , VEGF gene and IMD To observe the relationship between the expression level of HIF-1a or VEGF gene and angiogenesis, we assessed the IMD in ovarian carcinoma. IMD arranged in the order of magnitude of that in 60 ovarian carcinoma were plotted in a graph (Fig. 1E). IMD varied from 1.3 to 64.1 and the average of IMD was 18:94 ^ 12:86 mm2 , (mean ^ SD) in ovarian carcinoma. There was no relationship between HIF-1a or VEGF gene expression level and IMD (R ¼ 0:118 and 0.224, respectively, Fig. 2A,B). In addition, a weak association between HIF-1a and VEGF gene expression level was observed (R ¼ 0:300, P ¼ 0:020; Fig. 2C). To clarify the angiogenic property of HIF-1a and VEGF genes in ovarian carcinoma, we further investigated the relationships between these gene expression levels and IMD in poorly differentiated carcinomas. The significant correlations between VEGF gene expression and IMD or HIF-1a gene expression were observed (R ¼ 0:501, 0.424, P ¼ 0:016, 0.048, respectively, Fig. 2D,E). However, no association between IMD and HIF-1a gene expression was observed (data not shown).

K. Nakayama et al. / Cancer Letters 176 (2002) 215–223

219

Fig. 1. HIF-1a gene expression in ovarian carcinoma. (A) Expression of HIF-1a gene in 60 patients with ovarian carcinoma, determined by RT-PCR, as described in Section 2. Lanes C, RNA of placenta was used as positive controls for the detection of HIF-1a gene. Lanes W, PCR was performed without each cDNA. Data were expressed relative to the expression of GAPDH gene in each ovarian carcinoma. Case nos. 1, 5, serous cystadenocarcinoma, poorly differentiated; Case no. 10, serous cystadenocarcinoma, moderately differentiated; Case nos. 6, 33, mucinous cystadenocarcinoma, well-differentiated. (B) Immunohistochemical features of endothelial cells within ovarian carcinoma. Black arrows indicate IMD in ovarian carcinoma ( £ 200 field). White arrows indicate stained cells excluded from microvessels. White scale bar, 25 mm. (C, D) Expression level of HIF-1a and VEGF gene is arranged in the order of magnitude of each in 60 ovarian carcinoma. (E) IMD is arranged in the order of magnitude of each in 60 ovarian carcinoma.

3.3. Postoperative survival and the expression of HIF1a and VEGF gene expression in ovarian carcinoma The postoperative survival of 60 patients who underwent curative resection of ovarian carcinoma was analyzed according to the expression of HIF-1a or VEGF gene (Fig. 3). Although we observed both disease free interval and overall survival, no significant differences could be detected (Fig. 3A–D). Further, after categorizing into four groups regarding status of HIF-1a or VEGF gene expression level

(VEGF high, HIF-1a high/VEGF high, HIF-1a low/VEGF low, HIF-1a high /VEGF low, HIF-1a low), we also examined survival influences. There were no significant differences in these four groups (P ¼ 0:485, 0.579, 0.9443 and 0.736, respectively).

4. Discussion The present study provides that: (i) HIF-1a or VEGF gene expression level was independent of

220

K. Nakayama et al. / Cancer Letters 176 (2002) 215–223

Fig. 2. (A, B) Association of HIF-1a and VEGF gene expression level and IMD in ovarian carcinoma. Expression level of HIF-1a gene in 60 patients with ovarian carcinoma, determined by RT-PCR, as described in Section 2. Each gene expression level is reported relative to GAPDH gene. IMD was plotted against HIF-1a and VEGF gene expression in a graph. (C) Correlation of HIF-1a and VEGF gene expression in ovarian carcinomas. (D) Correlation of IMD and VEGF gene expression in poorly differentiated ovarian carcinomas.

age, clinical stage and histological subtype besides grade of tumor (Table 1), (ii) VEGF upregulated by HIF-1a gene may be involved in angiogenesis of some type of ovarian carcinoma and (iii) both HIF1a and VEGF genes have no effect on survival in ovarian carcinoma. HIF-1a may stimulate angiogenesis via transactivation of the VEGF gene, thus supporting tumor growth [21]. The present study revealed that a weak association between HIF-1a and VEGF gene expression level was observed (Fig. 2C). Further, both HIF1a or VEGF gene did not have a strong angiogenic effect in ovarian carcinoma (Fig. 2A,B), although the association between VEGF gene expression and IMD was observed in poorly differentiated ovarian carci-

noma (Fig. 2D). The previous study reported a positive correlation between IMD and VEGF expression in ovarian carcinoma [30]. However, no correlation between them was found in other studies [31–40]. Although Shen et al. showed VEGF expression by RT-PCR analysis as an independent prognostic factor, the effects of VEGF expression levels on the patient’s survival in ovarian carcinoma was not mediated by angiogenesis [41]. Taken our results and these reports together, the angiogenesis by VEGF gene upregulation-mediated through HIF-1a gene may not play an important role in ovarian carcinoma. Folkman demonstrated that the growth of solid tumors and their metastasis are dependent on angiogenesis, and angiogenesis is now regarded as one of

K. Nakayama et al. / Cancer Letters 176 (2002) 215–223

221

Fig. 3. (A, B) Kaplan–Meier survival curve in 60 patients with ovarian carcinoma according to HIF-1a gene expression level. P value was conducted by log–rank test. (C, D) Kaplan–Meier survival curve in 60 patients with ovarian carcinoma according to VEGF gene expression level. (A, C) Disease free survival. (B, D) Ovarall survival.

the important events in processing of tumors. IMD has been reported to be a prognostic factor in numerous human solid tumors after the demonstration that IMD was an independent prognostic factor in human breast carcinoma by Weidner et al. [3]. Recently, several studies of angiogenesis in ovarian carcinoma have been reported [30–41]. However, the effect of angiogenesis on survival of patients with ovarian carcinoma is still argued. Recently, we compared the angiogenesis of ovarian carcinoma with that of breast carcinoma, and reported that the angiogenesis of ovarian carcinoma may be less than that of breast carcinoma [42]. This might also be the reason for less significance of HIF-1a and VEGF genes in angiogenesis

and prognostic significance of the patients with ovarian carcinoma. We found that HIF-1a gene expression in moderately or poorly differentiated ovarian carcinoma was higher than that in well-differentiated carcinoma. In the previous report, high-grade carcinoma has more genetic event rather than low-grade ovarian carcinoma [43]. In addition, the incidence of gene amplification was different between well-differentiated and poorly differentiated esophageal/gastric carcinoma [44,45]. High expression of HIF-1a in high-grade ovarian carcinoma might be involved in carcinogenesis limited to poorly differentiated carcinoma. In conclusion, both HIF-1a and VEGF gene

222

K. Nakayama et al. / Cancer Letters 176 (2002) 215–223

expression may not play an important role in the angiogenesis induced by hypoxia in ovarian carcinoma.

References [1] J. Folkman, K. Watson, D. Ingber, D. Hanahan, Induction of angiogenesis during the transition from hyperplasia to neoplasia, Nature 339 (1989) 58–61. [2] L.A. Liotta, W.G. Stetler-Stevenson, Tumor invasion and metastasis: an imbalance of positive and negative regulation, Cancer Res. 51 (1991) 5054s–5059s. [3] N. Weidner, J.P. Semple, W.R. Welch, J. Folkman, Tumor angiogenesis and metastasis – correlation in invasive breast carcinoma, N. Engl. J. Med. 324 (1991) 1–8. [4] E.R. Horak, R. Leek, N. Klenk, S. LeJeune, K. Smith, N. Stuart, M. Greenall, K. Stepniewska, A.L. Harris, Angiogenesis, assessed by platelet/endothelial cell adhesion molecule antibodies, as indicator of node metastases and survival in breast cancer, Lancet 340 (1992) 1120–1124. [5] M. Toi, J. Kashitani, T. Tominaga, Tumor angiogenesis is an independent prognostic indicator in primary breast carcinoma, Int. J. Cancer 55 (1993) 371–374. [6] N. Weidner, P.R. Carroll, J. Flax, W. Blumenfeld, J. Folkman, Tumor angiogenesis correlates with metastasis in invasive prostate carcinoma, Am. J. Pathol. 143 (1993) 401–409. [7] L. Holmgren, M.S. O’Reilly, J. Folkman, Dormancy of micrometastases: balanced proliferation and apoptosis in the presence of angiogenesis suppression, Nat. Med. 1 (1995) 149–153. [8] M.S. O’Reilly, L. Holmgren, C. Chen, J. Folkman, Angiostatin induces and sustains dormancy of human primary tumors in mice, Nat. Med. 2 (1996) 689–692. [9] S. Parangi, M. O’Reilly, G. Christofori, L. Holmgren, J. Grosfeld, J. Folkman, D. Hanahan, Antiangiogenic therapy of transgenic mice impairs de novo tumor growth, Proc. Natl Acad. Sci. USA 93 (1996) 2002–2007. [10] T.G. Graeber, C. Osmanian, T. Jacks, D.E. Housman, C.J. Koch, S.W. Lowe, A.J. Giaccia, Hypoxia-mediated selection of cells with diminished apoptotic potential in solid tumours, Nature 379 (1996) 88–91. [11] D. Hanahan, J. Folkman, Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis, Cell 86 (1996) 353–364. [12] J.M. Brown, A.J. Giaccia, The unique physiology of solid tumors: opportunities (and problems) for cancer therapy, Cancer Res. 58 (1998) 1408–1416. [13] C.V. Dang, G.L. Semenza, Oncogenic alterations of metabolism, Trends Biochem. Sci. 24 (1999) 68–72. [14] K. Guillemin, M.A. Krasnow, The hypoxic response: huffing and HIFing, Cell 89 (1997) 9–12. [15] G.L. Semenza, Regulation of mammalian O2 homeostasis by hypoxia-inducible factor 1, Annu. Rev. Cell Dev. Biol. 15 (1999) 551–578. [16] G.L. Wang, B.H. Jiang, E.A. Rue, G.L. Semenza, Hypoxia-

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25] [26]

[27]

[28]

[29]

[30]

inducible factor 1 is a basic-helix–loop–helix-PAS heterodimer regulated by cellular O2 tension, Proc. Natl Acad. Sci. USA 92 (1995) 5510–5514. G.L. Semenza, G.L. Wang, A nuclear factor induced by hypoxia via de novo protein synthesis binds to the human erythropoietin gene enhancer at a site required for transcriptional activation, Mol. Cell. Biol. 12 (1992) 5447–5454. S. Salceda, J. Caro, Hypoxia-inducible factor 1alpha (HIF1alpha) protein is rapidly degraded by the ubiquitin-proteasome system under normoxic conditions. Its stabilization by hypoxia depends on redox-induced changes, J. Biol. Chem. 272 (1997) 22642–22647. L.E. Huang, J. Gu, M. Schau, H.F. Bunn, Regulation of hypoxia-inducible factor 1alpha is mediated by an O2-dependent degradation domain via the ubiquitin-proteasome pathway, Proc. Natl Acad. Sci. USA 95 (1998) 7987–7992. P.J. Kallio, W.J. Wilson, S. O’Brien, Y. Makino, L. Poellinger, Regulation of the hypoxia-inducible transcription factor 1alpha by the ubiquitin-proteasome pathway, J. Biol. Chem. 274 (1999) 6519–6525. W.G. An, M. Kanekal, M.C. Simon, E. Maltepe, M.V. Blagosklonny, L.M. Neckers, Stabilization of wild-type p53 by hypoxia-inducible factor 1alpha, Nature 392 (1998) 405–408. M. Toi, S. Hoshina, T. Takayanagi, T. Tominaga, Association of vascular endothelial growth factor expression with tumor angiogenesis and with early relapse in primary breast cancer, Jpn. J. Cancer Res. 85 (1994) 1045–1049. K. Maeda, Y.S. Chung, Y. Ogawa, S. Takatsuka, S.M. Kang, M. Ogawa, T. Sawada, M. Sowa, Prognostic value of vascular endothelial growth factor expression in gastric carcinoma, Cancer 77 (1996) 858–863. L.M. Ellis, Y. Takahashi, W. Liu, R.M. Shaheen, Vascular endothelial growth factor in human colon cancer: biology and therapeutic implications, Oncologist 5 (2000) 11–15. G. Gasparini, Prognostic value of vascular endothelial growth factor in breast cancer, Oncologist 5 (2000) 37–44. H. Zhong, A.M. De Marzo, E. Laughner, M. Lim, D.A. Hilton, D. Zagzag, P. Buechler, W.B. Isaacs, G.L. Semenza, J.W. Simons, Overexpression of hypoxia-inducible factor 1alpha in common human cancers and their metastases, Cancer Res. 59 (1999) 5830–5835. P. Birner, M. Schindl, A. Obermair, C. Plank, G. Breitenecker, G. Oberhuber, Overexpression of hypoxia-inducible factor 1alpha is a marker for an unfavorable prognosis in earlystage invasive cervical cancer, Cancer Res. 60 (2000) 4693– 4696. Changes in definitions of clinical staging for carcinoma of the cervix and ovary: International Federation of Gynecology and Obstetrics, Am. J. Obstet. Gynecol. 156 (2000) 263–264. V.V. Serov, T.B. Zhuravleva, L.H. Vasilevskaia, G. Mel’nikov Iu, Morphogenesis of uterine myomas, Akush. Ginekol. (Mosc.) 49 (1973) 3–8. G. Gasparini, E. Bonoldi, G. Viale, P. Verderio, P. Boracchi, G.A. Panizzoni, U. Radaelli, A. Di Bacco, R.B. Guglielmi, P. Bevilacqua, Prognostic and predictive value of tumour angiogenesis in ovarian carcinomas, Int. J. Cancer 69 (1996) 205– 211.

K. Nakayama et al. / Cancer Letters 176 (2002) 215–223 [31] H.C. Hollingsworth, E.C. Kohn, S.M. Steinberg, M.L. Rothenberg, M.J. Merino, Tumor angiogenesis in advanced stage ovarian carcinoma, Am. J. Pathol. 147 (1995) 33–41. [32] M. Emoto, H. Iwasaki, K. Mimura, T. Kawarabayashi, M. Kikuchi, Differences in the angiogenesis of benign and malignant ovarian tumors, demonstrated by analyses of color Doppler ultrasound, immunohistochemistry, and microvessel density, Cancer 80 (1997) 899–907. [33] W.M. Schoell, D. Pieber, O. Reich, M. Lahousen, M. Janicek, F. Guecer, R. Winter, Tumor angiogenesis as a prognostic factor in ovarian carcinoma: quantification of endothelial immunoreactivity by image analysis, Cancer 80 (1997) 2257–2262. [34] M. Orre, M. Lotfi-Miri, P. Mamers, P.A. Rogers, Increased microvessel density in mucinous compared with malignant serous and benign tumours of the ovary, Br. J. Cancer 77 (1998) 2204–2209. [35] S. Heimburg, M.K. Oehler, T. Papadopoulos, H. Caffier, P. Kristen, J. Dietl, Prognostic relevance of the endothelial marker CD 34 in ovarian cancer, Anticancer Res. 19 (1999) 2527–2529. [36] O. Abulafia, J.E. Ruiz, K. Holcomb, T.M. Dimaio, Y.C. Lee, D.M. Sherer, Angiogenesis in early-invasive and low-malignant-potential epithelial ovarian carcinoma, Obstet. Gynecol. 95 (2000) 548–552. [37] K. Hata, R. Fujiwaki, Y. Maede, K. Nakayama, M. Fukumoto, K. Miyazaki, Expression of thymidine phosphorylase in epithelial ovarian cancer: correlation with angiogenesis, apoptosis, and ultrasound-derived peak systolic velocity, Gynecol. Oncol. 77 (2000) 26–34. [38] Y. Nakanishi, J. Kodama, M. Yoshinouchi, K. Tokumo, S. Kamimura, H. Okuda, T. Kudo, The expression of vascular endothelial growth factor and transforming growth factor-beta

[39]

[40]

[41]

[42]

[43]

[44]

[45]

223

associates with angiogenesis in epithelial ovarian cancer, Int. J. Gynecol. Pathol. 16 (1997) 256–262. E.M. Hartenbach, T.A. Olson, J.J. Goswitz, D. Mohanraj, L.B. Twiggs, L.F. Carson, S. Ramakrishnan, Vascular endothelial growth factor (VEGF) expression and survival in human epithelial ovarian carcinomas, Cancer Lett. 121 (1997) 169– 175. M. Orre, P.A. Rogers, VEGF, VEGFR-1, VEGFR-2, microvessel density and endothelial cell proliferation in tumours of the ovary, Int. J. Cancer 84 (1999) 101–108. G.H. Shen, M. Ghazizadeh, O. Kawanami, H. Shimizu, E. Jin, T. Araki, Y. Sugisaki, Prognostic significance of vascular endothelial growth factor expression in human ovarian carcinoma, Br. J. Cancer 83 (2000) 196–203. K. Nakayama, A. Kanzaki, Y. Takebayashi, M. Toi, H. Bando, T. Nabei, K. Miyazaki, M. Fukumoto, Different features of angiogenesis between ovarian and breast carcinoma, Cancer Lett. 170 (2) (2001) 161–167. P.C. Cheng, J.A. Gosewehr, T.M. Kim, M. Velicescu, M. Wan, J. Zheng, J.C. Felix, K.F. Cofer, P. Luo, B.H. Biela, G. Godorov, L. Dubeau, Potential role of the inactivated X chromosome in ovarian epithelial tumor development, J. Natl Cancer Inst. 88 (1996) 510–518. J. Yokota, T. Yamamoto, N. Miyajima, K. Toyoshima, N. Nomura, H. Sakamoto, T. Yoshida, M. Terada, T. Sugimura, Genetic alterations of the c-erbB-2 oncogene occur frequently in tubular adenocarcinoma of the stomach and are often accompanied by amplification of the v-erbA homologue, Oncogene 2 (1988) 283–287. Y. Kanda, Y. Nishiyama, Y. Shimada, M. Imamura, H. Nomura, H. Hiai, M. Fukumoto, Analysis of gene amplification and overexpression in human esophageal- carcinoma cell lines, Int. J. Cancer 58 (1994) 291–297.