Protein Kinase B Phosphorylation Correlates with Vascular Endothelial Growth Factor A and Microvessel Density in Gastric Adenocarcinoma

Abstract

Objective:

The signalling molecule protein kinase B (Akt) modulates many cellular processes. Phosphatidylinositol 3-kinase (PI3K)/Akt signalling pathways play important roles in tumour angiogenesis. The aim of this study was to determine the role of phosphorylated Akt (pAkt) in angiogenesis and its correlation with vascular endothelial growth factor (VEGF)-A in gastric adenocarcinoma.

Methods:

Tumour tissue and matched healthy gastric mucosa were obtained from patients undergoing surgical resection of gastric adenocarcinoma. Akt and pAkt were detected via Western blotting. VEGF-A, pAkt and CD34 were examined by immunohistochemistry.

Results:

Akt and pAkt protein levels were significantly higher in gastric cancer tissue than in normal tissue (n = 48 patients). Positive VEGF-A immuno -staining was significantly associated with pAkt immunostaining. Microvessel density was correlated with both VEGF-A and pAkt positivity.

Conclusions:

Phosphorylated Akt and VEGF-A are involved in angiogenesis of gastric adenocarcinoma, and Akt activation may contribute to angiogenesis via VEGF-A upregulation. The PI3K/Akt/VEGF signalling pathway may be involved in gastric adenocarcinoma.

Keywords

Angiogenesis Gastric adenocarcinoma Microvessel density Protein kinase B Vascular endothelial growth factor-A

Introduction

Gastric cancer is one of the most common neoplasms and is a leading cause of death worldwide.¹ The overall survival rate for patients with gastric carcinoma has increased, as a result of improved detection of early cancers and wider implementation of radical surgery, but many patients die due to primary treatment failure or late-stage presentation.²

Angiogenesis is an important aspect of tumour growth and progression, and is an essential step in the metastasis of cancer cells by vascular spread.³ Vascular endothelial growth factor (VEGF) regulates the growth of endothelial cells, with VEGF-A being the most potent growth factor for angiogenesis.³ The level of VEGF-A has been correlated with tumour progression and poor survival in gastric cancer,^4,5 and inhibition of angiogenesis is a potentially useful treatment approach.

The serine/threonine protein kinase B (Akt) is a downstream target of growth factor receptors via the phosphatidylinositol 3-kinase (PI3K) signalling pathway.⁶ The PI3K/Akt signalling pathway regulates various cellular processes such as cell proliferation and survival, glucose utilization and inhibition of apoptosis.^6,7 Akt was initially identified as an oncogene in mouse thymoma.⁸ Activated (phosphorylated) Akt (pAkt) is upregulated in various human tumours including colon, breast, pancreatic and gastric cancers.^{9 – 11} We have shown that Akt is involved in chemoresistance of gastric cancer cells against etoposide and doxorubicin induced cell death.¹² In addition, PI3K/Akt signalling pathways are known to be important in angiogenesis.^{13– 18} The relationship between Akt phosphorylation and VEGF-A and microvessel density in human gastric adenocarcinoma tissue is unclear, however.

The present study investigated the relationship between Akt phosphorylation and VEGF-A levels, in addition to microvessel density, in human gastric adenocarcinoma.

Materials and methods

Study Population

Consecutive patients undergoing surgical resection for gastric adenocarcinoma at the Department of Gastroenterology, First Affiliated Hospital, Nanchang University, Nanchang, China, between May 2010 and June 2011, were recruited to the study. Patients who had received any chemo-, radio- or immunotherapy before surgery were excluded from the study. Tumours were staged and classified according to the tumour–node–metastasis (TNM) staging system and Lauren classification.¹⁹

The study was carried out in accordance with the Declaration of Helsinki (2000) and was approved by the Ethics Committee of Nanchang University. Written informed consent was obtained from all participants.

Sample Collection

Samples of gastric adenocarcinoma and matched normal gastric mucosa (≥ 10 cm from the distal tumour margin) were collected from each patient and flash-frozen in liquid nitrogen. A portion of each sample was fixed in 10% buffered formalin, embedded in paraffin wax, and cut into 3-μm thick sections for histopathological analysis (hematoxylin and eosin staining) and immunohistochemical staining. A randomly selected subset of samples was used for Western blotting. All methods are described below.

Western Blotting

Frozen gastric adenocarcinoma samples and their matched normal mucosa tissues were homogenized and lysed in buffer containing 40 mM Tris-HCl (pH 8.0), 120 mM sodium chloride, 0.5% nonyl phenoxypoly ethoxyl -ethanol (NP-40), 1 mM phenylmethyl -sulphonyl fluoride, and 10 μg leupeptin per ml. Lysates were incubated on ice for 60 min then centrifuged at 9901 g for 20 min. For each sample, 50 g of protein was separated via 12% sodium dodecyl sulphate– polyacrylamide gel electrophoresis at 25 °C and 200 V, then electrotransferred onto nitrocellulose membranes (Santa Cruz Biotechnology, Santa Cruz, CA, USA) at 25 V for 2 h. Membranes were blocked with 5% nonfat milk in TBST (10 mM Tris [pH 7.4], 100 mM sodium chloride, 0.5% Tween 20) then washed three times with TBST (5 min each wash). Membranes were then incubated with mouse monoclonal antihuman pAkt (Ser⁴⁷³) (1 : 1000 dilution; Cell Signaling Technology, Beverly, MA, USA), rabbit monoclonal antihuman Akt (1 : 1000 dilution; Cell Signaling Technology) or mouse monoclonal anti-human β-actin (1 : 1000 dilution; Santa Cruz Biotechnology) for 1 h at 4 °C. After washing three times with TBST at 25°C (5 min each wash), membranes were incubated with peroxidase conjugated goat antimouse or goat antirabbit immunoglobulin (Ig)G (1 : 2000 dilution; Santa Cruz Biotechnology) for 1 h at 25°C. Protein signals were visualized using enhanced chemiluminescence (ECL Western blotting detection, Amersham Life Science, Amersham, UK) and exposure to medical X-radiographic film for 1 min. Films were scanned (Epson GT-9500; Seiko Corp, Nagano, Japan) and protein bands were quantitated by densitometry (Image PC alpha 9; National Institutes of Health, Bethesda, MD, USA) with reference to β-actin levels.

Immunohistochemistry

Tissue sections were deparaffinized, rehydrated, heated in antigen retrieval buffer (DakoCytomation, Glostrup, Denmark) at 99°C for 20 min, and incubated in 3% hydrogen peroxide for 15 min to quench endogenous peroxidase. Sections were incubated in mouse monoclonal antihuman pAkt (Ser⁴⁷³) (1 : 65 dilution; Cell Signaling Technology) or rabbit polyclonal antihuman VEGF-A (1 : 300 dilution; SC-152, Santa Cruz Biotechnology) at 4 °C overnight. Negative controls were incubated with phosphate buffered saline (PBS; 0.01 M, pH 7.3) in place of primary antibody. Sections were washed three times with PBS (2 min each wash), incubated with biotinylated goat antimouse or goat antirabbit IgG (1 : 200 dilution, Santa Cruz Biotechnology) for 1 h at room temperature, and washed three times with PBS (2 min each wash). Immunoreactivity was detected using a streptavidin–peroxidase detection system (Santa Cruz Biotechnology) and visualized using 3,3′-diaminobenzidine (DAB) with haematoxylin as counterstain. Staining was classified as weak (light yellow), moderate (deep yellow) or strong (brown), and graded based on the percentage of cells stained with moderate or strong intensity: negative, < 5 % cells; moderate, 5 – 50 %; high, > 50 %. Cases with moderate or high staining were considered positive. All immunostained slides were reviewed by the same pathologist, who was blinded to patient outcome.

Quantification of Microvessel Density

Tissue sections were deparaffinized, rehydrated and incubated in 3% hydrogen peroxide for 15 min to quench endogenous peroxidase. Slides were heated in a microwave oven for 10 min in 10 mM citrate buffer (pH 6.0), then incubated with rabbit monoclonal antihuman CD34 (1: 40 dilution; Maxim Biotech, San Francisco, CA, USA) overnight at 4 °C. Slides were washed three times with 0.01 M PBS (pH 7.35) for 3 min each wash, incubated with peroxidase-labelled goat antirabbit IgG (1 : 200 dilution, DakoCytomation) for 10 min at 37°C and washed three times with PBS (3 min each wash). Immunostaining was visualized with DAB and sections were counterstained with haematoxylin.

Intratumoral microvessel density was determined as described.²⁰ In brief, the area of most intense neovascularization (preferably on the tumour margin) was selected by scanning on low magnification (× 10 – 100). Only the vascularity of tumour areas considered to be viable (i.e. non-necrotic) was taken into account. Individual microvessels were counted at × 200 magnification (0.723 mm²/field). Any brown-stained, nucleus-containing endo -thelial cell that was clearly separate from adjacent microvessels, tumour cells and other connective tissue elements was considered a single, countable microvessel, without requirement for a lumen or the presence of erythrocytes. Each patient's microvessel count was the mean of two separate counts made by two different pathologists who remained blind to patient outcome.

Statistical Analyses

Correlations between microvessel density and the presence of pAkt and VEGF-A were determined using χ²-test or two-sided Fisher's exact test. The prevalence of Akt in adenocarcinoma and normal tissue was compared by two-tailed Student's t-test. Statistical analyses were performed using SPSS^® version 11.0 (SPSS Inc., Chicago, IL, USA) for Windows^®. A P-value of < 0.05 was considered statistically significant.

Results

The study included 48 patients with gastric adenocarcinoma (28 males/20 females; mean age 58.5 ± 23.7 years, range 29 – 80 years).

Western blotting for pAkt and Akt was performed on tissues from a selection of 20 patients. A representative Western blot is shown in Fig. 1. Both pAkt and Akt were present at significantly higher levels in adenocarcinoma tissue than in matched normal tissue (P < 0.001 for both comparisons; Table 1).

Figure 1:

Representative Western blot analyses for protein kinase B (Akt) and phosphorylated (p)Akt in tumour tissue (C) and matched normal gastric mucosa (N) from four patients with gastric adenocarcinoma. Levels of both Akt and pAkt were higher in tumour tissue than in normal tissue in all cases shown

Table 1:

Levels of protein kinase B (Akt) and phosphorylated (p)Akt determined by Western blotting in a selection of human gastric adenocarcinoma and matched normal gastric mucosa tissue samples (n = 20)

Protein	Normal mucosa	Adenocarcinoma
Akt	0.069 ± 0.009	0.186 ± 0.013***
pAkt	0.040 ± 0.006	0.164 ± 0.025***

Data presented as mean ± SD, normalized to β-actin levels.

***

P < 0.001 vs normal mucosa; Student's t-test.

Immunohistochemical staining for pAkt was located in the cytoplasm of tumour cells and normal gland epithelial cells (Figs 2A and 2B). Phosphorylated Akt staining was found in 67.7% (32/48) of gastric adenocarcinomas and 58.3% (28/48) of control tissue samples. In tumour tissue, positive pAkt staining was associated with poor differentiation (P = 0.016), late TNM stage (P = 0.014) and high microvessel density (P = 0.021) (Table 2).

Figure 2:

Representative photomicrographs of immunohistochemical staining for (A) phosphorylated protein kinase B (pAkt) in normal human gastric mucosa; (B) pAkt in gastric adenocarcinoma; (C) vascular endothelial growth factor A in gastric adenocarcinoma; and (D) identification of microvessels in gastric adenocarcinoma via CD34 staining (arrow)

Table 2:

Correlations between demographic, clinical and histopathological characteristics and the presence of immunostaining for phosphorylated protein kinase B (pAkt) or vascular endothelial growth factor (VECF)-A in tumour tissue from patients with gastric adenocarcinoma (n = 48)

Characteristic	Total	pAkt immunostaining				VECF-A immunostaining
Characteristic	Total	Negative n = 16	Moderate n = 20	High n = 12	Statistical significance^a	Negative n = 23	Moderate n = 15	High n = 10	Statistical significance^a
Gender
Male	28 (58.3)	10 (20.8)	12 (25.0)	6 (12.5)	NS	13 (27.1)	10 (20.8)	5 (10.4)	NS
Female	20 (41.7)	6 (12.5)	8 (16.7)	6 (12.5)		10 (20.8)	5 (10.4)	5 (10.4)
Age
< 50	23 (47.9)	7 (14.6)	9 (18.8)	7 (14.6)	NS	9 (18.8)	7 (14.6)	7 (14.6)	NS
≥ 50	25 (52.1)	9 (18.8)	11 (22.9)	5 (10.4)	14 (29.2)	8 (16.7)	3 (6.3)
Differentiation
Well/moderately	34 (70.8)	14 (29.2)	16 (33.3)	4 (8.3)	P = 0.016	17 (35.4)	10 (20.8)	7 (14.6)	NS
Poor/undifferentiated	14 (29.2)	2 (4.2)	4 (8.3)	8 (16.7)		6 (12.5)	5 (10.4)	3 (6.3)
Lauren classification¹⁹
Intestinal	34 (70.8)	10(20.8)	14(29.2)	10 (20.8)	NS	14 (29.2)	12 (25.0)	8 (16.7)	NS
Diffuse	14 (29.2)	6 (12.5)	6 (12.5)	2 (4.2)		9(18.8)	3 (6.3)	2 (4.2)
Tumour diameter, cm
< 5	26 (54.2)	8 (16.7)	13 (27.1)	5 (10.4)	NS	10(20.8)	9 (18.8)	7(14.6)	NS
≥ 5	22 (45.8)	8 (16.7)	7 (14.6)	7 (14.6)	13 (27.1)	6 (12.5)	3 (6.3)
TNM stage¹⁹
I - II	22 (45.8)	12 (25.0)	7 (14.6)	3 (6.3)	P = 0.014	14 (29.2)	6 (12.5)	2 (4.2)	P = 0.029
III - IV	26 (54.2)	4 (8.3)	13 (27.1)	9 (18.8)		9 (18.8)	9 (18.8)	8 (16.7)
Lymph node metastasis
0	19 (39.6)	9 (18.8)	5 (10.4)	5 (10.4)	NS	14 (29.2)	4 (8.3)	1 (2.1)	P = 0.011
≥ 1	29 (60.4)	7 (14.6)	15 (31.3)	7 (14.6)		9 (18.8)	11 (22.9)	9 (18.8)
MVD^b
Low (< 40.6 vessels/ 0.723 mm²)	23 (47.9)	12 (25.0)	8 (16.7)	3 (6.3)	P = 0.021	16 (33.3)	4 (8.3)	3 (6.3)	P = 0.015
High (≥ 40.6 vessels/ 0.723 mm²)	25 (52.1)	4 (8.3)	12 (25.0)	9 (18.8)		7 (14.6)	11 (22.9)	7 (14.6)

Data presented as n (%) of patients.

TNM, tumour-node-metastasis; MVD, microvessel density.

Staining was graded based on the percentage of cells stained with moderate or strong intensity: negative, < 5%; moderate, 5 - 50%; high, > 50%.

χ²-test.

Relative to mean MVD (40.6 vessels/0.723 mm²).

Immunohistochemical staining for VEGF-A was visible in the cytoplasm in 52.1% (25/48) of tumour tissues (Fig. 2C) and 18.8% (nine of 48) of normal gastric mucosa samples. Presence of VEGF-A was associated with late TNM stage (P = 0.029), lymph node metastasis (P = 0.011) and high microvessel density (P = 0.015) (Table 2). Positive VEGF-A staining was significantly associated with positive pAkt staining (P = 0.012; Table 3).

Table 3:

Association between the presence of phosphorylated protein kinase B (pAkt) and vascular endothelial growth factor (VEGF)-A in tumour tissue from patients with gastric adenocarcinoma (n = 48)

pAkt immunostaining	VEGF-A immunostaining			Statistical significance^a
pAkt immunostaining	Negative	Moderate	High	Statistical significance^a
Negative	13 (27.1)	2 (4.2)	1 (2.1)	P = 0.012
Moderate	7 (14.6)	9 (18.8)	4 (8.3)
High	3 (6.3)	4 (8.3)	5 (10.4)

Data presented as n (%) of patients.

Staining was graded based on the percentage of cells stained with moderate or strong intensity: negative, < 5%; moderate, 5 – 50%; high, > 50%.

χ²-test.

The mean intratumoral microvessel density was 40.6 ± 15.1 vessels/0.723 mm² (range 14 – 115 vessels/0.723 mm²). Intratumoral microvessels were found mainly at the periphery of the infiltrating tumour in nearly all instances (Fig. 2D).

Discussion

A detailed understanding of the mechanisms underlying angiogenesis may be of therapeutic value for cancers that require neovascularization. Angiogenesis, as quantitated via microvessel density, plays significant clinical and pathological roles in cancer,^21,22 and may be an independent, highly significant and accurate prognostic indicator.²⁰

Gastric cancer is a leading cause of death in Asia.¹ Tumour angiogenesis is considered to be the most important predictor of overall survival in gastric cancer,²³ but little is known about the molecular mechanism underlying gastric cancer angiogenesis. The angiogenic process has been shown to be triggered by several key tumour-derived growth factors,^24,25 including VEGF-A and platelet-derived growth factor B.^24,26,27 Studies have found VEGF-A to be overexpressed in 40 – 90% of gastric cancers,^{5,28– 36} which is consistent with the 52.1% found in the present study. VEGF-A positivity has been correlated with increased microvessel density, vascular invasion, lymph node and distant metastases, and advanced tumour stage, suggesting that VEGF-A might be a useful biomarker of tumour aggressiveness.^28,29 In the present study, VEGF-A positivity was correlated with lymph node metastasis, advanced tumour stage and increased microvessel density.

Upregulation of pAkt is closely related to poor prognosis in several tumour types, including colon, breast, pancreatic and gastric cancers.^{9 –11} Levels of both Akt and pAkt were significantly higher in tumour tissue than normal tissue, in the present study. There was no correlation between presence of pAkt and age, gender, Lauren classification, tumour size or lymph node metastasis in the current study. Immunostaining for pAkt was significantly stronger in poorly differentiated and late TNM stage tumours, however. We have previously shown that Akt is important in the chemoresistance of gastric cancer cells against etoposide and doxorubicin induced cell death,¹² but the role of Akt in the angiogenesis of gastric cancer remains unclear.

An important finding of the present study was the strong correlation between microvessel density and pAkt and VEGF-A in human gastric adenocarcinoma. The association between the PI3K/Akt signalling pathway and angiogenesis has been demonstrated in several in vitro and in vivo studies.^17,18,37,38 An inhibitor of mammalian target of rapamycin (mTOR) induced downregulation of VEGF-A secretion via the PI3K/Akt/mTOR signalling pathway, eventually inhibiting neuroblastoma growth in vitro and in vivo.³⁷ The PI3K/Akt signalling pathway was also found to be required for hypoxia inducible factor (HIF)-1α and VEGF-A expression in 4-hydroxy oestradiol-induced carcinogenesis.³⁸ HIF-1α activated VEGFA gene expression and influenced both angiogenesis and tumour growth. The PI3K/Akt signalling pathway was also found to regulate HIF1A expression.^17,18 An inhibitor of Akt phosphorylation significantly inhibited VEGF production in vitro, revealing the existence of a PI3K/Akt/VEGF signalling pathway.³⁹ Taken together, these studies suggest that the PI3K/Akt signalling pathway contributes to tumour angiogenesis via VEGF-A and HIF-1α.

In conclusion, the correlations between microvessel density and pAkt and VEGF-A positivity in the present study suggest that pAkt and VEGF-A are involved in angiogenesis of gastric adenocarcinoma. In addition, the relationship between pAkt and VEGF-A suggests that Akt activation contributes to angiogenesis via VEGF-A upregulation, indicating the presence of a PI3K/Akt/VEGF signalling pathway in gastric adenocarcinoma.

Footnotes

Acknowledgements

The study was supported by funding from National Science Foundation Grants of China (No. 81160307), the Jiangxi Science & Technology Pillar Program and the Science Foundation for Young Scholars of Jiangxi Province (No. 2007GQY1167), and Foundation of Jiangxi Educational Committee (No. GJJ10063).

Conflicts of interest: The authors had no conflicts of interest to declare in relation to this article.

References

Parkin

, Pisani

, Ferlay

: Estimates of the worldwide incidence of 25 major cancers in 1990. Int J Cancer 1999; 80: 827–841.

Hundahl

, Phillips

, Menck

: The National Cancer Data Base Report on poor survival of U.S. gastric carcinoma patients treated with gastrectomy: Fifth Edition American Joint Committee on Cancer staging, proximal disease, and the “different disease” hypothesis. Cancer 2000; 88: 921–932.

Leung

, Cachianes

, Kuang

: Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 1989; 246: 1306–1309.

Konno

, Ohta

, Baba

: The role of circulating IL-8 and VEGF protein in the progression of gastric cancer. Cancer Sci 2003; 94: 735–740.

Ding

, Li

, Lin

: Distinct roles of VEGF-A and VEGF-C in tumour metastasis of gastric carcinoma. Oncol Rep 2007; 17: 369–375.

Manning

, Cantley

: AKT/PKB signaling: navigating downstream. Cell 2007; 129: 1261–1274.

Jiang

, Liu

: PI3K/PTEN signaling in tumorigenesis and angiogenesis. Biochim Biophys Acta 2008; 1784: 150–158.

Bellacosa

, Testa

, Staal

: A retroviral oncogne, Akt, encoding a serine-threonine kinase containing as SH2-like region. Science 1991; 254: 274–277.

Carnero

: The PKB/AKT pathway in cancer. Curr Pharm Des 2010; 16: 34–44.

10.

Tsutsui

, Matsuyama

, Yamamoto

: The Akt expression correlates with the VEGF-A and -C expression as well as the microvessel and lymphatic vessel density in breast cancer. Oncol Rep 2010; 23: 621–630.

11.

Wen

, Wang

, Zhou

: Mutation analysis of tumor suppressor gene PTEN in patients with gastric carcinomas and its impact on PI3K/AKT pathway. Oncol Rep 2010; 24: 89–95.

12.

, Ai

, Yu

: Phosphoinositide 3-kinase/Akt pathway plays an important role in chemoresistance of gastric cancer cells against etoposide and doxorubicin induced cell death. Int J Cancer 2008; 122: 433–443.

13.

, Tan

, Zhang

: Activation of Akt-mTOR-p70S6K pathway in angiogenesis in hepatocellular carcinoma. Oncol Rep 2008; 20: 713–719.

14.

Fang

, Xia

, Cao

: Apigenin inhibits VEGF and HIF-1 expression via PI3K/AKT/p70S6K1 and HDM2/p53 pathways. FASEB J 2005; 19: 342–353.

15.

, Wu

, Lin

: Hispidulin, a small flavonoid molecule, suppresses the angiogenesis and growth of human pancreatic cancer by targeting vascular endothelial growth factor receptor 2-mediated PI3K/Akt/mTOR signaling pathway. Cancer Sci 2011; 102: 219–225.

16.

Sliva

, Jedinak

, Kawasaki

: Phellinus linteus suppresses growth, angiogenesis and invasive behaviour of breast cancer cells through the inhibition of AKT signalling. Br J Cancer 2008; 98: 1348–1356.

17.

Silva

, Bowen

, Rychahou

: VEGFR-2 expression in carcinoid cancer cells and its role in tumor growth and metastasis. Int J Cancer 2011; 128: 1045–1056.

18.

Heeg

, Hirt

, Queisser

: EGFR overexpression induces activation of telomerase via PI3K/AKT-mediated phosphorylation and transcriptional regulation through Hif1-alpha in a cellular model of oral-esophageal carcinogenesis. Cancer Sci 2011; 102: 351–360.

19.

Sobin

, Wittekind

(eds): TNM Classification of Malignant Tumors (UICC), 5th edn. Hoboken: Wiley-Blackwell, 1997.

20.

Weidner

, Semple

, Welch

: Tumor angiogenesis and metastases - correlation in invasive breast carcinoma. N Engl J Med 1991; 324: 1–8.

21.

Delli Carpini

, Karam

, Montgomery

: Vascular endothelial growth factor and its relationship to the prognosis and treatment of breast, ovarian, and cervical cancer. Angiogenesis 2010; 13: 43–58.

22.

Huang

, Huang

, Chang

: The expression and clinical significance of pSTAT3, VEGF and VEGF-C in pancreatic adenocarcinoma. Neoplasma 2012; 59: 52–61.

23.

Vidal

, Metges

, Elizalde

: High preoperative serum vascular endothelial growth factor levels predict poor clinical outcome after curative resection of gastric cancer. Br J Surg 2009; 96: 1443–1451.

24.

Carmeliet

: Angiogenesis in health and disease. Nat Med 2003; 9: 653–660.

25.

Yancopoulos

, Davis

, Gale

: Vascular-specific growth factors and blood vessel formation. Nature 2000; 407: 242–248.

26.

Nissen

, Cao

, Hedlund

: Angiogenic factors FGF2 and PDGF-BB synergistically promote murine tumor neovascularization and metastasis. J Clin Invest 2007; 117: 2766–2777.

27.

Cao

: Opinion: emerging mechanisms of tumour lymphangiogenesis and lymphatic metastasis. Nat Rev Cancer 2005; 5: 735–743.

28.

Kondo

, Kaneko

, Baba

: VEGF-C and VEGF-A synergistically enhance lymph node metastasis of gastric cancer. Biol Pharm Bull 2007; 30: 633–637.

29.

Miyamoto

, Yamamoto

, Taniguchi

: Differential expression of angiogenesis-related genes in human gastric cancers with and those without high-frequency microsatellite instability. Cancer Lett 2007; 254: 42–53.

30.

Lee

, Kim

, Sohn

: No association of vascular endothelial growth factor-A (VEGF-A) and VEGF-C expression with survival in patients with gastric cancer. Cancer Res Treat 2009; 41: 218–223.

31.

Zhao

, Qin

, Chen

: Microvessel density is a prognostic marker of human gastric cancer. World J Gastroenterol 2006; 12: 7598–7603.

32.

Saito

, Osaki

, Murakami

: Prediction of sites of recurrence in gastric carcinoma using immunohistochemical parameters. J Surg Oncol 2007; 95: 123–128.

33.

Kolev

, Uetake

, Iida

: Prognostic significance of VEGF expression in correlation with COX-2, microvessel density, and clinicopathological characteristics in human gastric carcinoma. Ann Surg Oncol 2007; 14: 2738–2747.

34.

Suzuki

, Dobashi

, Hatakeyama

: Clinicopathological significance of platelet-derived growth factor (PDGF)-B and vascular endothelial growth factor-A expression, PDGF receptor-ß phosphorylation, and microvessel density in gastric cancer. BMC Cancer 2010; 10: 659.

35.

Nikiteas

, Tzanakis

, Theodoropoulos

: Vascular endothelial growth factor and endoglin (CD-105) in gastric cancer. Gastric Cancer 2007; 10: 12–17.

36.

Yang

, Zhao

, Wu

: Expression and biological relationship of vascular endothelial growth factor-A and matrix metalloproteinase-9 in gastric carcinoma. J Int Med Res 2011; 39: 2076–2085.

37.

Johnsen

, Segerström

, Orrego

: Inhibitors of mammalian target of rapamycin downregulate MYCN protein expression and inhibit neuroblastoma growth in vitro and in vivo. Oncogene 2008; 27: 2910–2922.

38.

Gao

, Nester

, Sarkar

: 4-Hydroxy estradiol but not 2-hydroxy estradiol induces expression of hypoxia-inducible factor 1a and vascular endothelial growth factor A through phosphatidylinositol 3-kinase/Akt/FRAP pathway in OVCAR-3 and A2780-CP70 humanovarian carcinoma cells. Toxicol Appl Pharmacol 2004; 196: 124–135.

39.

Villaume

, Blanc

, Gouysse

: VEGF secretion by neuroendocrine tumor cells is inhibited by octreotide and by inhibitors of the PI3K/AKT/mTOR pathway. Neuroendocrinology 2010; 91: 268–278.