Ann Surg Oncol DOI 10.1245/s10434-014-3765-8
ORIGINAL ARTICLE – TRANSLATIONAL RESEARCH AND BIOMARKERS
Anti-Prokineticin1 (PROK1) Monoclonal Antibody Suppresses Angiogenesis and Tumor Growth in Colorectal Cancer Takanori Goi, MD, PhD, Toshiyuki Nakazawa, MD, Yasuo Hirono, MD, PhD, and Akio Yamaguchi, MD, PhD First Department of Surgery, University of Fukui, Eiheiji, Fukui, Japan
ABSTRACT Background. The prokineticin1 (PROK1) gene has been cloned as an angiogenic growth factor from endocrine gland cells. However, we have not known about potentials of anti-PROK1 monoclonal antibody in human cancers. Here we investigated how the anti-PROK1 monoclonal antibody (mAb; established by our department) would affect the high-PROK1-expressing colorectal cancer (CRC) cells in vitro and vivo. Methods. We confirmed PROK1 protein expression in the CRC cells by performing immunohistochemical staining and measured the amount of soluble PROK1 protein. Next, we mixed the CRC cell culture fluid with the antiPROK1mAb to examine angiogenic activity in vitro and in vivo. Additionally, we investigated whether the antiPROK1mAb would affect the tumor-forming capability of high PROK1-expressing CRC cells implanted into mice. Results. PROK1 protein expression was confirmed in 3 CRC cell lines, and soluble PROK1 protein was also confirmed in the CRC cell culture fluid. The culture fluid increased angiogenesis in vitro and vivo, whereas the antiPROK1mAb suppressed angiogenesis. Subcutaneous tumor formation and tumor angiogenesis in mice were suppressed by the anti-PROK1mAb treatment. The antiPROK1mAb significantly suppressed the number of CD31 stained cells in mice. Conclusions. The in vitro and vivo experimental system indicated that the anti-PROK1mAb could suppress angiogenesis and tumor growth in the CRC strains.
Ó Society of Surgical Oncology 2014 First Received: 12 March 2014 T. Goi, MD, PhD e-mail: [email protected]
In general, it is considered important to suppress lymph node metastasis, hematogenous metastasis, or peritoneal metastasis for improving the survival rate of cancer patients.1,2 In the case of colorectal cancer, hematogenous metastasis is particularly frequent, and establishment of a countermeasure is desired.2 Hematogenous metastasis of colorectal cancer is considered to occur as follows: colorectal cancer cells in the primary lesion break the basement membrane, enter the capillary vessel, pass through the portal vein, and infiltrate a distant organ, e.g., the liver.3 Angiogenic growth factors are thought to play an important role at several steps.4 Molecular target treatment drugs targeting angiogenic growth factors have recently been applied in clinical practice for advanced/recurrent colorectal cancer, and they have produced useful results in improving patients’ prognosis.5–9 The prokineticin1 (PROK1) factor, which is investigated in this study, was reported by Ferrara as an angiogenic growth factor.10 On the mRNA level, it is expressed in a number of tissues such as adrenal body, ovary, and testis, apparently sharing structural homology with the venom protein A in snake venom. Characteristically, it demonstrates proliferation and migration activities only in endocrine cells and promotes growth of vascular endothelial cells under hypoxic conditions. However, there is no homology with vascular endothelial growth factor (VEGF), and it is thought to be completely different from the existing VEGF family, based on the sequence of cysteine.10 We have recently studied PROK1 mRNA expression in primary lesions of patients who underwent colorectal cancer surgery. We found that the prognosis was significantly poorer in cases positive for PROK1 mRNA expression compared to negative cases, and that angiogenesis was more likely to occur, as the expression of PROK1 gene was increased, leading to liver metastasis.11,12 Furthermore, other reports on malignant tumors indicated that PROK1 was associated with metastasis of prostate cancer and neuroblastoma, and that the degree of
T. Goi et al.
malignance increased in pancreatic duct cancer.13–15 Accordingly, the significance of PROK1 appears to be very high in malignant tumors. In the present study, we used an in vitro and in vivo experimental system that included high PROK1 protein expression-type colorectal cancer cells to confirm that the anti-PROK1 monoclonal antibody (mAb) could suppress angiogenesis and tumor formation.
PROK1 ELISA Assay
METHODS
A 200 ll cell culture fluid plus the mAb-YURA21 (5 lg) or normal mouse IgG (Santa Cruz Biotechnology, USA) were added to wells of an angiogenesis kit (Kurabo, Japan). On day 5, cells were fixed in 70 % ethanol and stained with anti-CD31 mAb (Tubule staining kit, Kurabo). For the evaluation of capillary tube formation (the stained tubelike structures), each well was photographed and total tube length was analyzed by MacSCOPE program (Mitani Corp., Japan).11 The experiments were repeated 4 times.
Anti-PROK1 Monoclonal Antibody and Western Blot Analysis A mouse was immunized subcutaneously 3 times with a PROK1-GST fusion protein. The total RNA was extracted from HCT116 colon cancer cells. The 50 primer was PROK1-AX, GGATCCATGAGAGGTGCCACGCGAG TCTCAATC of the published human PROK1 sequence, and the 30 primer was PROK1-BX, GAATTCAAAATT GATGTTCTTCAAGTCCATGGAGCAGCGG.10 The PCR products were cloned into pGEX2T vector (Pharmacia, Sweden). Spleen cells were fused with the murine myeloma cell line: NS-1 cells.16 The hybridoma culture supernatants were assayed for reactivity with PROK1 protein (Shenandoah Biotechnology Inc., USA) using an enzyme-linked immunosorbent assay and immunoblotting. Subcloning and limiting dilution were repeated 3 times to obtain the antiPROK1mAb (mAb-YURA21). The Ig subclass was IgG1. PROK1 protein (5 lg) was run on a 10 % SDS-polyacryl-amide gel and transferred to PVDF membrane (BioRad Laboratories Inc., USA). The membrane was incubated with the mAb-YURA21. Density was visualized by enhanced chemiluminescence according to the manufacturer’s instructions (ImmunoStar LD, Wako Pure Chemical Industries, Ltd., Japan). Cell Culture The human colon cancer cell lines, DLD-1, HCT116, and LoVo (obtained from European Collection of Cell Cultures, UK) were cultured at 37 °C in 5 % CO2 in RPMI 1640 medium containing 10 % fetal bovine serum (FBS). Cell Culture Fluid After preparation of the cell lines as described previously, each cell line was passaged at 60 % confluence in a 60 mm culture plate and cultured in RPMI 1640 (containing 10 % FBS). The culture fluid was collected after culture of the cell lines for 3 days.
The cell culture fluid was used for quantifying PROK1 levels by applying a PROK1-specific ELISA kit from R&D Systems, Inc. (USA) and following the vendor’s protocol. The experiments were repeated 4 times. In Vitro Tube Formation Assay
Detection of Vascularization with Dorsal Air Sac Method A Millipore chamber (Millipore; diameter, 10 mm; filter pore size, 0.45 lm) was filled with 200 ll culture medium plus the mAb-YURA21, or normal mouse IgG was implanted subcutaneously into the dorsal side of 6-weekold female SHO nude mice (Charles River Laboratories, Japan). At 7 days after implantation, a rectangular incision was made in the skin on the dorsal side. The chambercontacting region was photographed. Tumor Formation and Microvessel Counting in Nude Mice The 6-week-old female SHO nude mice (Charles River Laboratories) were subcutaneously injected in the right armpit region with 1.0 9 106 cells in 0.1 mL of matrix gel (BD Biosciences, USA). Two groups of mice were tested. Group 1 was injected with nonstimulated colon cancer cells (DLD-1, HCT116, and LoVo) and normal mouse IgG. Group 2 was injected with colon cancer cells and the mAbYURA21 (5 lg). The tumor size was measured every 3 days with calipers. The tumor volume was calculated with the formula: (L 9 W2)/2, where L is the length and W is the width of the tumor.12 After 21 days, the tumor was resected, photographed, and weighted. Tumors and subcutaneous tissues for histological examination were embedded in OCT compound (Sakura Finetechnical, Tokyo, Japan). Then 4-lm-thick sections were analyzed for CD31 protein expression (antiCD31mAb DAKO, Denmark) by the ChemMate method using the EnVision system (DAKO).
Anti-PROK1mAb Inhibits Angiogenesis and Tumor Growth FIG. 1 a The anti-PROK1mAb (mAbYURA21) reacted with a PROK1 protein. Left: Coomassie blue, purified PROK1 protein. Right: Western blot, the anti-PROK1mAb detected a PROK1 protein. b PROK1 protein expression in colorectal cancer cell lines. The PROK1 expression was detected in colorectal cancer cell lines (DLD-1, HCT116, LoVo). 9200. c Amount of PROK1 protein in culture fluid after culturing colorectal cancer cell lines. The cell culture media were collected and used for quantifying PROK1 levels by using an PROK1specific ELISA kit
B
DLD-1
A
C
kD
(pg/ml) 400
30
350
HCT116
300 250
10 200 150 100 50
LoVo 0
DLD-1
HCT 116
LoVo
For vessel counting, 1 field magnified 200-fold in each of 5 vascularized areas was counted, and average counts were recorded.
Investigation of the Expression of PROK1 Protein in Culture Fluid After Culturing Colorectal Cancer Cell Lines
Statistical Considerations
PROK1 protein was observed in the culture fluids of all colorectal cancer cell lines: 271.1 pg/ml for DLD-1, 287.3 pg/ml for HCT116, and 235.3 pg/ml for LoVo (Fig. 1c).
Statistical analysis was performed by the v2 test or t test using Stat Mate IV (ATMS Co., Ltd., Japan). Data are given as mean ± SEM. Values of P \ 0.05 were considered as statistically significant. RESULTS Generation of the Anti-PROK1mAb Reactive with PROK1 Protein The anti-PROK1mAb (mAb-YURA21) was specifically reacted with PROK1 protein (11.7 kD) (Fig. 1a). Investigation of the Expression of PROK1 Protein in Colorectal Cancer Cell Lines Immunohistochemical staining was carried out with the anti-PROK1mAb (mAb-YURA21) in human colorectal cancer cell lines to examine the expression of PROK1 protein. The PROK1 expression was detected in colon cancer cell lines: DLD-1, HCT116, and LoVo (Fig. 1b).
Investigation of Tube Formation When AntiPROK1mAb was Added to the Culture Fluid After 3-Day Culture of Colorectal Cancer Cell Lines The length of tube formation in the control (culture fluid had not been used for culturing) was 680 lm, while it became significantly longer (970–1,200 lm) when culture fluid had been used for culturing of the cell lines. When the anti-PROK1mAb (mAb-YURA21) was added to the culture fluid, tube formation was significantly suppressed to 570–620 lm compared with cell lines cultured in the presence of culture fluid (Fig. 2). Investigation of Subcutaneous Angiogenesis in Mice When Anti-PROK1mAb was Added to the Cultures The blood vessels increased in length and size in response to culture fluids from all cultured colorectal cancer cell lines. However, when the anti-PROK1mAb
T. Goi et al. FIG. 2 Investigation of tube formation in culture fluids after culturing colorectal cancer cell lines. a Representative photographs of tube formation: left, DLD-1 culture fluid alone; right, DLD-1 culture fluid plus the anti-PROK1mAb. b Quantitative analyses of the tube length in culture fluids. Tube formation was evaluated by measurements of tube length after treatment with the antiPROK1mAb. Data represent mean ± SEM (n = 4). (*t test P \ 0.05)
A
B
(µm) 1400
Culture medium
+
+
1200
Anti-PROK1 mAb
-
+
1000
*
*
*
+ + - +
+ + - +
HCT116
LoVo
800 600 400 200 0 Culture + + medium Ant- + PROK1mAb DLD-1
A
B Culture medium
Culture medium
Culture medium +Anti-PROK1 mAb
Culture medium +Anti-PROK1 mAb
DLD1
HCT116
LoVo
*
counts 50
Microvessel density
C
*
*
40 30 20 10
0 Anti-PROK1 mAb b
–
+ DLD1
–
+
HCT116
–
+ LoVo
FIG. 3 Investigation of subcutaneous angiogenesis in mice in response to colorectal cancer cell fluid. a Representative photographs of blood vessels: left, culture fluid alone; right, anti-PROK1mAb plus culture fluid. b Representative photographs of CD31 immunostaining:
left, HCT116 culture fluid alone; right, anti-PROK1mAb plus HCT116 culture fluid. c The numbers of positively CD31 stained cells. Data represent mean ± SEM. (*t test P \ 0.05)
(mAb-YURA21) was added to the culture fluid, blood vessels were suppressed in length and size (Fig. 3a).
(mAb-YURA21). In mice treated with the LoVo fluid, 34 cells were positive per visual field, while only 16 cells were positive per visual field with the addition of the antiPROK1mAb (mAb-YURA21). The anti-PROK1mAb (mAb-YURA21) significantly suppressed the number of stained cells in response to all fluids (Fig. 3b, c).
Subcutaneous CD31 Expression in Mice in Response to Colorectal Cancer Cell Fluids In mice treated with the DLD-1 fluid, 42 cells were positive per visual field, while 27 were positive per visual field with the addition of the anti-PROK1mAb (mAbYURA21). In mice treated with the HCT116 fluid, 32 cells were positive per visual field, while 17 were positive per visual field with the addition of the anti-PROK1mAb
Suppression of Mouse Colorectal Cancer Cell Line Tumor Formation by the Anti-PROK1mAb After subcutaneously implanting high PROK1 proteinexpressing colorectal cancer cells into mice (Fig. 4a), the
Anti-PROK1mAb Inhibits Angiogenesis and Tumor Growth FIG. 4 Investigation of subcutaneous tumor formation by the anti-PROK1Ab. SHO nude mice were subcutaneously injected in the right armpit region with 1.0 9 106 cells and the anti-PROK1mAb in matrix gel. At 3 weeks later, the tumor was resected, photographed, and weighed. Representative photographs of tumor formation: left, HCT116 alone; right, anti-PROK1mAb plus HCT116. The measurement of subcutaneous tumor weight. Data represent mean ± SEM. (t test P \ 0.05)
A
Cancer cells
B
*
*
400
500
300
300 200 100
Weight (mg)
LoVo 350
Weight (mg)
HCT116 600
Weight (mg)
DLD-1 500
400 300 200 100
Anti-PROK1 mAb
(–)
(+)
*
250 200 150 100 50 0
0
0
FIG. 5 Investigation of CD31 expression of subcutaneous tumor formation. a Representative photographs of anti-CD31Ab stained cells by performing immunohistochemical staining: left, HCT116 alone; right, anti-PROK1mAb plus HCT116. b The numbers of positively CD31 stained cells in subcutaneous tumors. Left, cells alone; right, anti-PROK1mAb plus cells. Data represent mean ± SEM. (*t test P \ 0.05)
Cancer cells +Anti-PROK1 mAb
(–)
(+)
(–)
(+)
A Culture medium
B
Culture medium +Anti-PROK1 mAb
*
counts
*
*
40
Microvessel density
35 30 25 20 15 10 5 0
Anti-PROK1 mAb
–
+ DLD1
DLD-1 cell tumor was 410 mg in weight, but 245 mg with the addition of the anti-PROK1mAb. The HCT116 tumor was 570 mg in weight, but 250 mg with the addition of the
– HCT116
+
–
+ LoVo
anti-PROK1mAb (mAb-YURA21). The LoVo tumor was 310 mg in weight, but 145 mg with the addition of the antiPROK1mAb (mAb-YURA21). Tumor formation was
T. Goi et al.
significantly suppressed by adding the anti-PROK1mAb (Fig. 4b). CD31 Expression at the Tumor-Forming Site in Mice The number of positive cells at the DLD-1 tumor site was 28 per visual field; with the addition of the anti-PROK1mAb (mAb-YURA21), the number of positive cells at the tumor site was 13 per visual field. The number of positive cells at the HCT116 tumor site was 31 per visual field, and there were 14 per visual field when the anti-PROK1mAb (mAb-YURA21) was added. The number of positive cells at the LoVo tumor site was 30 per visual field, and there were 12 per visual field at the tumor site with the addition of the anti-PROK1mAb (mAb-YURA21). The number of stained cells was significantly suppressed by the addition of the anti-PROK1mAb (mAb-YURA21) (Fig. 5). DISCUSSION The mechanism of gastrointestinal cancer metastasis has recently been studied from the perspective of molecular biology, and various genes are considered to be involved in metastasis.17,18 VEGF is known as one of the important factors for hematogenous metastasis, in particular. Molecular target treatment drugs have been developed for targeting this factor, clearly contributing to the improvement of prognosis.5 PROK1 used in our study has been reported to be a growth factor that is different from VEGF.10 As no protein-level investigations have been conducted of normal human colon mucosa and colorectal cancer, our department prepared an anti-PROK1mAb (mAb-YURA21) for the present study. First, the immunohistochemical staining method did not reveal any expression in the normal colon mucosa, producing a similar result as for RNA expression. Next, the anti-PROK1mAb was used for examining human colorectal cancer resected from patients in our department. Expression was observed in about 40 % cases of malignant colorectal cancer. Of these cases, the frequency of expression was high in cases with hematogenous metastasis, suggesting the involvement of the PROK1. Also, the blood vessel area was significantly higher in PROK1 expression-positive cancers than in PROK1 expression-negative cancers. Recent reports indicated that cancer stem cells and the surrounding microenvironment (niche) are important for cancer growth.19–22 Growth of cancer stem cells is considered to require an optimum environment containing a number of closely related surrounding factors such as mesenchymal cells, paraneoplastic macrophages, fibroblasts, various cytokines, extracellular matrix, and angiogenic
factors.23–25 Thus, the cancer cell itself possibly needs to promote the expression of PROK1 protein, an angiogenic factor, attempting to create the surrounding microenvironment required for the growth and development of colorectal cancer. At present, antibody drugs have been clinically used to target VEGF as a therapy against colorectal cancer, and improvement has been observed particularly for the prognosis of unresectable advanced colorectal cancer.5 However, the present treatments alone cannot completely control colorectal cancer, and further improvement of treatment is desired. As the first step toward the clinical application of PROK1-targeting treatments, we used high PROK1 protein-expressing colorectal cancer cell lines and an antiPROK1mAb to study angiogenesis suppression. The PROK1 protein released from colorectal cancer cells was soluble at different concentrations in the extracellular supernatants. Using these culture supernatants, we examined subcutaneous angiogenesis in mice. Compared with the control medium, angiogenesis was intensified in response to the supernatants. Addition of the antiPROK1mAb suppressed this angiogenic activity. Thus, the anti-PROK1mAb might possibly suppress the activity of colorectal cancer-produced PROK1 protein and thus might suppress angiogenic activity. Furthermore, because the angiogenic growth factor reportedly plays an important role in cancer cell proliferation, we administered cancer cells and the anti-PROK1mAb to mice to investigate subcutaneous tumor formation. When the anti-PROK1mAb was added, tumor formation was significantly suppressed and the surrounding angiogenic activity was also reduced, suggesting the importance of PROK1 in colorectal cancer cells. In recent years, we studied the relationship between PROK1 and angiogenesis in clinical human colorectal cancer specimens.26 Significantly more blood vessels were found in the PROK1-positive specimens, compared with the PROK-1 negative samples, indicating the clinical significance of this factor. Also, Prokineticin receptors, when expressed on colorectal cancer cells, are likely to be involved in invasion via an autocrine mechanism, and PROK1 is considered to be a part of the important signal transduction network associated with metastasis.27 In our recent in vitro and in vivo experiment with colorectal cancer cell lines, we found that the antiPROK1mAb could suppress angiogenic activity and tumor formation, suggesting that our finding might be the first step toward a new therapy. ACKNOWLEDGMENT The technical assistance of Ms. Saitoh M. with this research was appreciated. This work was supported in part by a Grant-in-Aid for Science Research from the Ministry of Education, Sports, Science, and Technology of Japan (No. 25462047).
Anti-PROK1mAb Inhibits Angiogenesis and Tumor Growth DISCLOSURE interests.
The authors declare that they have no competing
REFERENCES 1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int. J Cancer. 2010;127:2893–917. 2. Watanabe T, Itabashi M, Shimada Y, Tanaka S, Ito Y, Ajioka Y, et al. Japanese society for cancer of the colon and rectum: Japanese society for cancer of the colon and rectum (JSCCR) guidelines 2010 for the treatment of colorectal cancer. Int J Clin Oncol. 2012;17:1–29. 3. Hanahan D, Folkman J. Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. Cell. 1996;86:353–64. 4. Ferrara N. Pathways mediating VEGF-independent tumor angiogenesis. Cytokine Growth Factor Rev. 2010;21:21–6. 5. Saltz LB, Clarke S, Dı´az-Rubio E, Scheithauer W, Figer A, Wong R, et al. Bevacizumab in combination with oxaliplatin-based chemotherapy as first-line therapy in metastatic colorectal cancer: a randomized phase III study. J Clin Oncol. 2008;26:2013–9. 6. Van Cutsem E, Ko¨hne CH, Hitre E, Zaluski J, Chang Chien CR, Makhson A, et al. Cetuximab and chemotherapy as initial treatment for metastatic colorectal cancer. N Engl J Med. 2009;360: 1408–17. 7. Amado RG, Wolf M, Peeters M, Van Cutsem E, Siena S, Freeman DJ, et al. Wild-type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer. J Clin Oncol. 2008;26:1626–34. 8. Grothey A, Van Cutsem E, Sobrero A, Siena S, Falcone A, Ychou M, et al. Regorafenib monotherapy for previously treated metastatic colorectal cancer (CORRECT): an international, multicenter, randomized, placebo-controlled, phase 3 trial. Lancet. 2013;381:303–12. 9. Tabernero J, Van Cutsem E, Lakomy R, Prausova´ J, Ruff P, van Hazel GA, et al. Aflibercept versus placebo in combination with fluorouracil, leucovorin and irinotecan in the treatment of previously treated metastatic colorectal cancer: prespecified subgroup analyses from the VELOUR trial. Eur J Cancer. 2014;50:320–31. 10. LeCouter J, Kowalski J, Foster J, Hass P, Zhang Z, Dillard-Telm L, et al. Identification of an angiogenic mitogen selective for endocrine gland endothelium. Nature. 2001;412:877–84. 11. Nagano H, Goi T, Koneri K, Hirono Y, Katayama K, Yamaguchi A. Endocrine gland-derived vascular endothelial growth factor (EG-VEGF) expression in colorectal cancer. J Surg Oncol. 2007;96:605–10. 12. Goi T, Fujioka M, Satoh Y, Tabata S, Koneri K, Nagano H, et al. Angiogenesis and tumor proliferation/metastasis of human colorectal cancer cell line SW620 transfected with endocrine glandderived-vascular endothelial growth factor, as a new angiogenic factor. Cancer Res. 2004;64:1906–10.
13. Ngan ES, Sit FY, Lee K, Miao X, Yuan Z, Wang W, et al. Implications of endocrine gland-derived vascular endothelial growth factor/prokineticin-1 signaling in human neuroblastoma progression. Clin Cancer Res. 2007;13:868–75. 14. Pasquali D, Rossi V, Staibano S, De Rosa G, Chieffi P, Prezioso D, et al. The endocrine-gland-derived vascular endothelial growth factor (EG-VEGF)/prokineticin 1 and 2 and receptor expression in human prostate: up-regulation of EG-VEGF/prokineticin 1 with malignancy. Endocrinology. 2006;147:4245–51. 15. Morales A, Vilchis F, Cha´vez B, Chan C, Robles-Dı´az G, Dı´azSa´nchez V. Expression and localization of endocrine glandderived vascular endothelial growth factor (EG-VEGF) in human pancreas and pancreatic adenocarcinoma. J Steroid Biochem Mol Biol. 2007;107:37–41. 16. Goi T, Yamaguchi A, Nakagawara G, Urano T, Shiku H, Furukawa K. Reduced expression of deleted colorectal carcinoma (DCC) protein in established colon cancers. Br J Cancer. 1998; 77:466–71. 17. Kowanetz M, Ferrara N. Vascular endothelial growth factor signaling pathways: therapeutic perspective. Clin Cancer Res. 2006;12:5018–22. 18. Avraamides CJ, Garmy-Susini B, Varner JA. Integrins in angiogenesis and lymphangiogenesis. Nat Rev Cancer. 2008;8:604–17. 19. Clarke MF, Dick JE, Dirks PB, Eaves CJ, Jamieson CH, Jones DL, et al. Cancer stem cells—perspectives on current status and future directions: AACR Workshop on cancer stem cells. Cancer Res. 2006;66:9339–44. 20. Robert GJ, Huch M, Clevers H. Stem cells and cancer of the stomach and intestine. Nat Rev Cancer. 2010;6:373–84. 21. Moumen M, Chiche A, Decraene C, Petit V, Gandarillas A, Deugnier MA, et al. Myc is required for b-catenin-mediated mammary stem cell amplification and tumorigenesis. Mol Cancer. 2013;12:132. 22. Olino K, Wada S, Edil BH, Pan X, Meckel K, Weber W, et al. Tumor-associated antigen expressing Listeria monocytogenes induces effective primary and memory T-cell responses against hepatic colorectal cancer metastases. Ann Surg Oncol. 2012;19: S597–607. 23. Kalluri R, Zeisberg M. Fibroblasts in cancer. Mol Oncol. 2006;4:392–401. 24. Kaplan RN, Riba RD, Zacharoulis S, Bramley AH, Vincent L, Costa C, et al. VEGFR1-positive haematopoietic bone marrow progenitors initiate the pre-metastatic niche. Nature. 2005;438:820–7. 25. Miller AR, McBride WH, Hunt K, Economou JS. Cytokine-mediated gene therapy for cancer. Ann Surg Oncol. 1994;1:436–50. 26. Goi T, Nakazawa T, Hirono Y, Yamaguchi A. Prokineticin 1 expression in gastrointestinal tumors. Anticancer Res. 2013;33: 5311–5. 27. Tabata S, Goi T, Nakazawa T, Kimura Y, Katayama K, Yamaguchi A. Endocrine gland-derived vascular endothelial growth factor strengthens cell invasion ability via prokineticin receptor 2 in colon cancer cell lines. Oncol. Rep. 2013;29:459–63.
ORIGINAL ARTICLE – TRANSLATIONAL RESEARCH AND BIOMARKERS
Anti-Prokineticin1 (PROK1) Monoclonal Antibody Suppresses Angiogenesis and Tumor Growth in Colorectal Cancer Takanori Goi, MD, PhD, Toshiyuki Nakazawa, MD, Yasuo Hirono, MD, PhD, and Akio Yamaguchi, MD, PhD First Department of Surgery, University of Fukui, Eiheiji, Fukui, Japan
ABSTRACT Background. The prokineticin1 (PROK1) gene has been cloned as an angiogenic growth factor from endocrine gland cells. However, we have not known about potentials of anti-PROK1 monoclonal antibody in human cancers. Here we investigated how the anti-PROK1 monoclonal antibody (mAb; established by our department) would affect the high-PROK1-expressing colorectal cancer (CRC) cells in vitro and vivo. Methods. We confirmed PROK1 protein expression in the CRC cells by performing immunohistochemical staining and measured the amount of soluble PROK1 protein. Next, we mixed the CRC cell culture fluid with the antiPROK1mAb to examine angiogenic activity in vitro and in vivo. Additionally, we investigated whether the antiPROK1mAb would affect the tumor-forming capability of high PROK1-expressing CRC cells implanted into mice. Results. PROK1 protein expression was confirmed in 3 CRC cell lines, and soluble PROK1 protein was also confirmed in the CRC cell culture fluid. The culture fluid increased angiogenesis in vitro and vivo, whereas the antiPROK1mAb suppressed angiogenesis. Subcutaneous tumor formation and tumor angiogenesis in mice were suppressed by the anti-PROK1mAb treatment. The antiPROK1mAb significantly suppressed the number of CD31 stained cells in mice. Conclusions. The in vitro and vivo experimental system indicated that the anti-PROK1mAb could suppress angiogenesis and tumor growth in the CRC strains.
Ó Society of Surgical Oncology 2014 First Received: 12 March 2014 T. Goi, MD, PhD e-mail: [email protected]
In general, it is considered important to suppress lymph node metastasis, hematogenous metastasis, or peritoneal metastasis for improving the survival rate of cancer patients.1,2 In the case of colorectal cancer, hematogenous metastasis is particularly frequent, and establishment of a countermeasure is desired.2 Hematogenous metastasis of colorectal cancer is considered to occur as follows: colorectal cancer cells in the primary lesion break the basement membrane, enter the capillary vessel, pass through the portal vein, and infiltrate a distant organ, e.g., the liver.3 Angiogenic growth factors are thought to play an important role at several steps.4 Molecular target treatment drugs targeting angiogenic growth factors have recently been applied in clinical practice for advanced/recurrent colorectal cancer, and they have produced useful results in improving patients’ prognosis.5–9 The prokineticin1 (PROK1) factor, which is investigated in this study, was reported by Ferrara as an angiogenic growth factor.10 On the mRNA level, it is expressed in a number of tissues such as adrenal body, ovary, and testis, apparently sharing structural homology with the venom protein A in snake venom. Characteristically, it demonstrates proliferation and migration activities only in endocrine cells and promotes growth of vascular endothelial cells under hypoxic conditions. However, there is no homology with vascular endothelial growth factor (VEGF), and it is thought to be completely different from the existing VEGF family, based on the sequence of cysteine.10 We have recently studied PROK1 mRNA expression in primary lesions of patients who underwent colorectal cancer surgery. We found that the prognosis was significantly poorer in cases positive for PROK1 mRNA expression compared to negative cases, and that angiogenesis was more likely to occur, as the expression of PROK1 gene was increased, leading to liver metastasis.11,12 Furthermore, other reports on malignant tumors indicated that PROK1 was associated with metastasis of prostate cancer and neuroblastoma, and that the degree of
T. Goi et al.
malignance increased in pancreatic duct cancer.13–15 Accordingly, the significance of PROK1 appears to be very high in malignant tumors. In the present study, we used an in vitro and in vivo experimental system that included high PROK1 protein expression-type colorectal cancer cells to confirm that the anti-PROK1 monoclonal antibody (mAb) could suppress angiogenesis and tumor formation.
PROK1 ELISA Assay
METHODS
A 200 ll cell culture fluid plus the mAb-YURA21 (5 lg) or normal mouse IgG (Santa Cruz Biotechnology, USA) were added to wells of an angiogenesis kit (Kurabo, Japan). On day 5, cells were fixed in 70 % ethanol and stained with anti-CD31 mAb (Tubule staining kit, Kurabo). For the evaluation of capillary tube formation (the stained tubelike structures), each well was photographed and total tube length was analyzed by MacSCOPE program (Mitani Corp., Japan).11 The experiments were repeated 4 times.
Anti-PROK1 Monoclonal Antibody and Western Blot Analysis A mouse was immunized subcutaneously 3 times with a PROK1-GST fusion protein. The total RNA was extracted from HCT116 colon cancer cells. The 50 primer was PROK1-AX, GGATCCATGAGAGGTGCCACGCGAG TCTCAATC of the published human PROK1 sequence, and the 30 primer was PROK1-BX, GAATTCAAAATT GATGTTCTTCAAGTCCATGGAGCAGCGG.10 The PCR products were cloned into pGEX2T vector (Pharmacia, Sweden). Spleen cells were fused with the murine myeloma cell line: NS-1 cells.16 The hybridoma culture supernatants were assayed for reactivity with PROK1 protein (Shenandoah Biotechnology Inc., USA) using an enzyme-linked immunosorbent assay and immunoblotting. Subcloning and limiting dilution were repeated 3 times to obtain the antiPROK1mAb (mAb-YURA21). The Ig subclass was IgG1. PROK1 protein (5 lg) was run on a 10 % SDS-polyacryl-amide gel and transferred to PVDF membrane (BioRad Laboratories Inc., USA). The membrane was incubated with the mAb-YURA21. Density was visualized by enhanced chemiluminescence according to the manufacturer’s instructions (ImmunoStar LD, Wako Pure Chemical Industries, Ltd., Japan). Cell Culture The human colon cancer cell lines, DLD-1, HCT116, and LoVo (obtained from European Collection of Cell Cultures, UK) were cultured at 37 °C in 5 % CO2 in RPMI 1640 medium containing 10 % fetal bovine serum (FBS). Cell Culture Fluid After preparation of the cell lines as described previously, each cell line was passaged at 60 % confluence in a 60 mm culture plate and cultured in RPMI 1640 (containing 10 % FBS). The culture fluid was collected after culture of the cell lines for 3 days.
The cell culture fluid was used for quantifying PROK1 levels by applying a PROK1-specific ELISA kit from R&D Systems, Inc. (USA) and following the vendor’s protocol. The experiments were repeated 4 times. In Vitro Tube Formation Assay
Detection of Vascularization with Dorsal Air Sac Method A Millipore chamber (Millipore; diameter, 10 mm; filter pore size, 0.45 lm) was filled with 200 ll culture medium plus the mAb-YURA21, or normal mouse IgG was implanted subcutaneously into the dorsal side of 6-weekold female SHO nude mice (Charles River Laboratories, Japan). At 7 days after implantation, a rectangular incision was made in the skin on the dorsal side. The chambercontacting region was photographed. Tumor Formation and Microvessel Counting in Nude Mice The 6-week-old female SHO nude mice (Charles River Laboratories) were subcutaneously injected in the right armpit region with 1.0 9 106 cells in 0.1 mL of matrix gel (BD Biosciences, USA). Two groups of mice were tested. Group 1 was injected with nonstimulated colon cancer cells (DLD-1, HCT116, and LoVo) and normal mouse IgG. Group 2 was injected with colon cancer cells and the mAbYURA21 (5 lg). The tumor size was measured every 3 days with calipers. The tumor volume was calculated with the formula: (L 9 W2)/2, where L is the length and W is the width of the tumor.12 After 21 days, the tumor was resected, photographed, and weighted. Tumors and subcutaneous tissues for histological examination were embedded in OCT compound (Sakura Finetechnical, Tokyo, Japan). Then 4-lm-thick sections were analyzed for CD31 protein expression (antiCD31mAb DAKO, Denmark) by the ChemMate method using the EnVision system (DAKO).
Anti-PROK1mAb Inhibits Angiogenesis and Tumor Growth FIG. 1 a The anti-PROK1mAb (mAbYURA21) reacted with a PROK1 protein. Left: Coomassie blue, purified PROK1 protein. Right: Western blot, the anti-PROK1mAb detected a PROK1 protein. b PROK1 protein expression in colorectal cancer cell lines. The PROK1 expression was detected in colorectal cancer cell lines (DLD-1, HCT116, LoVo). 9200. c Amount of PROK1 protein in culture fluid after culturing colorectal cancer cell lines. The cell culture media were collected and used for quantifying PROK1 levels by using an PROK1specific ELISA kit
B
DLD-1
A
C
kD
(pg/ml) 400
30
350
HCT116
300 250
10 200 150 100 50
LoVo 0
DLD-1
HCT 116
LoVo
For vessel counting, 1 field magnified 200-fold in each of 5 vascularized areas was counted, and average counts were recorded.
Investigation of the Expression of PROK1 Protein in Culture Fluid After Culturing Colorectal Cancer Cell Lines
Statistical Considerations
PROK1 protein was observed in the culture fluids of all colorectal cancer cell lines: 271.1 pg/ml for DLD-1, 287.3 pg/ml for HCT116, and 235.3 pg/ml for LoVo (Fig. 1c).
Statistical analysis was performed by the v2 test or t test using Stat Mate IV (ATMS Co., Ltd., Japan). Data are given as mean ± SEM. Values of P \ 0.05 were considered as statistically significant. RESULTS Generation of the Anti-PROK1mAb Reactive with PROK1 Protein The anti-PROK1mAb (mAb-YURA21) was specifically reacted with PROK1 protein (11.7 kD) (Fig. 1a). Investigation of the Expression of PROK1 Protein in Colorectal Cancer Cell Lines Immunohistochemical staining was carried out with the anti-PROK1mAb (mAb-YURA21) in human colorectal cancer cell lines to examine the expression of PROK1 protein. The PROK1 expression was detected in colon cancer cell lines: DLD-1, HCT116, and LoVo (Fig. 1b).
Investigation of Tube Formation When AntiPROK1mAb was Added to the Culture Fluid After 3-Day Culture of Colorectal Cancer Cell Lines The length of tube formation in the control (culture fluid had not been used for culturing) was 680 lm, while it became significantly longer (970–1,200 lm) when culture fluid had been used for culturing of the cell lines. When the anti-PROK1mAb (mAb-YURA21) was added to the culture fluid, tube formation was significantly suppressed to 570–620 lm compared with cell lines cultured in the presence of culture fluid (Fig. 2). Investigation of Subcutaneous Angiogenesis in Mice When Anti-PROK1mAb was Added to the Cultures The blood vessels increased in length and size in response to culture fluids from all cultured colorectal cancer cell lines. However, when the anti-PROK1mAb
T. Goi et al. FIG. 2 Investigation of tube formation in culture fluids after culturing colorectal cancer cell lines. a Representative photographs of tube formation: left, DLD-1 culture fluid alone; right, DLD-1 culture fluid plus the anti-PROK1mAb. b Quantitative analyses of the tube length in culture fluids. Tube formation was evaluated by measurements of tube length after treatment with the antiPROK1mAb. Data represent mean ± SEM (n = 4). (*t test P \ 0.05)
A
B
(µm) 1400
Culture medium
+
+
1200
Anti-PROK1 mAb
-
+
1000
*
*
*
+ + - +
+ + - +
HCT116
LoVo
800 600 400 200 0 Culture + + medium Ant- + PROK1mAb DLD-1
A
B Culture medium
Culture medium
Culture medium +Anti-PROK1 mAb
Culture medium +Anti-PROK1 mAb
DLD1
HCT116
LoVo
*
counts 50
Microvessel density
C
*
*
40 30 20 10
0 Anti-PROK1 mAb b
–
+ DLD1
–
+
HCT116
–
+ LoVo
FIG. 3 Investigation of subcutaneous angiogenesis in mice in response to colorectal cancer cell fluid. a Representative photographs of blood vessels: left, culture fluid alone; right, anti-PROK1mAb plus culture fluid. b Representative photographs of CD31 immunostaining:
left, HCT116 culture fluid alone; right, anti-PROK1mAb plus HCT116 culture fluid. c The numbers of positively CD31 stained cells. Data represent mean ± SEM. (*t test P \ 0.05)
(mAb-YURA21) was added to the culture fluid, blood vessels were suppressed in length and size (Fig. 3a).
(mAb-YURA21). In mice treated with the LoVo fluid, 34 cells were positive per visual field, while only 16 cells were positive per visual field with the addition of the antiPROK1mAb (mAb-YURA21). The anti-PROK1mAb (mAb-YURA21) significantly suppressed the number of stained cells in response to all fluids (Fig. 3b, c).
Subcutaneous CD31 Expression in Mice in Response to Colorectal Cancer Cell Fluids In mice treated with the DLD-1 fluid, 42 cells were positive per visual field, while 27 were positive per visual field with the addition of the anti-PROK1mAb (mAbYURA21). In mice treated with the HCT116 fluid, 32 cells were positive per visual field, while 17 were positive per visual field with the addition of the anti-PROK1mAb
Suppression of Mouse Colorectal Cancer Cell Line Tumor Formation by the Anti-PROK1mAb After subcutaneously implanting high PROK1 proteinexpressing colorectal cancer cells into mice (Fig. 4a), the
Anti-PROK1mAb Inhibits Angiogenesis and Tumor Growth FIG. 4 Investigation of subcutaneous tumor formation by the anti-PROK1Ab. SHO nude mice were subcutaneously injected in the right armpit region with 1.0 9 106 cells and the anti-PROK1mAb in matrix gel. At 3 weeks later, the tumor was resected, photographed, and weighed. Representative photographs of tumor formation: left, HCT116 alone; right, anti-PROK1mAb plus HCT116. The measurement of subcutaneous tumor weight. Data represent mean ± SEM. (t test P \ 0.05)
A
Cancer cells
B
*
*
400
500
300
300 200 100
Weight (mg)
LoVo 350
Weight (mg)
HCT116 600
Weight (mg)
DLD-1 500
400 300 200 100
Anti-PROK1 mAb
(–)
(+)
*
250 200 150 100 50 0
0
0
FIG. 5 Investigation of CD31 expression of subcutaneous tumor formation. a Representative photographs of anti-CD31Ab stained cells by performing immunohistochemical staining: left, HCT116 alone; right, anti-PROK1mAb plus HCT116. b The numbers of positively CD31 stained cells in subcutaneous tumors. Left, cells alone; right, anti-PROK1mAb plus cells. Data represent mean ± SEM. (*t test P \ 0.05)
Cancer cells +Anti-PROK1 mAb
(–)
(+)
(–)
(+)
A Culture medium
B
Culture medium +Anti-PROK1 mAb
*
counts
*
*
40
Microvessel density
35 30 25 20 15 10 5 0
Anti-PROK1 mAb
–
+ DLD1
DLD-1 cell tumor was 410 mg in weight, but 245 mg with the addition of the anti-PROK1mAb. The HCT116 tumor was 570 mg in weight, but 250 mg with the addition of the
– HCT116
+
–
+ LoVo
anti-PROK1mAb (mAb-YURA21). The LoVo tumor was 310 mg in weight, but 145 mg with the addition of the antiPROK1mAb (mAb-YURA21). Tumor formation was
T. Goi et al.
significantly suppressed by adding the anti-PROK1mAb (Fig. 4b). CD31 Expression at the Tumor-Forming Site in Mice The number of positive cells at the DLD-1 tumor site was 28 per visual field; with the addition of the anti-PROK1mAb (mAb-YURA21), the number of positive cells at the tumor site was 13 per visual field. The number of positive cells at the HCT116 tumor site was 31 per visual field, and there were 14 per visual field when the anti-PROK1mAb (mAb-YURA21) was added. The number of positive cells at the LoVo tumor site was 30 per visual field, and there were 12 per visual field at the tumor site with the addition of the anti-PROK1mAb (mAb-YURA21). The number of stained cells was significantly suppressed by the addition of the anti-PROK1mAb (mAb-YURA21) (Fig. 5). DISCUSSION The mechanism of gastrointestinal cancer metastasis has recently been studied from the perspective of molecular biology, and various genes are considered to be involved in metastasis.17,18 VEGF is known as one of the important factors for hematogenous metastasis, in particular. Molecular target treatment drugs have been developed for targeting this factor, clearly contributing to the improvement of prognosis.5 PROK1 used in our study has been reported to be a growth factor that is different from VEGF.10 As no protein-level investigations have been conducted of normal human colon mucosa and colorectal cancer, our department prepared an anti-PROK1mAb (mAb-YURA21) for the present study. First, the immunohistochemical staining method did not reveal any expression in the normal colon mucosa, producing a similar result as for RNA expression. Next, the anti-PROK1mAb was used for examining human colorectal cancer resected from patients in our department. Expression was observed in about 40 % cases of malignant colorectal cancer. Of these cases, the frequency of expression was high in cases with hematogenous metastasis, suggesting the involvement of the PROK1. Also, the blood vessel area was significantly higher in PROK1 expression-positive cancers than in PROK1 expression-negative cancers. Recent reports indicated that cancer stem cells and the surrounding microenvironment (niche) are important for cancer growth.19–22 Growth of cancer stem cells is considered to require an optimum environment containing a number of closely related surrounding factors such as mesenchymal cells, paraneoplastic macrophages, fibroblasts, various cytokines, extracellular matrix, and angiogenic
factors.23–25 Thus, the cancer cell itself possibly needs to promote the expression of PROK1 protein, an angiogenic factor, attempting to create the surrounding microenvironment required for the growth and development of colorectal cancer. At present, antibody drugs have been clinically used to target VEGF as a therapy against colorectal cancer, and improvement has been observed particularly for the prognosis of unresectable advanced colorectal cancer.5 However, the present treatments alone cannot completely control colorectal cancer, and further improvement of treatment is desired. As the first step toward the clinical application of PROK1-targeting treatments, we used high PROK1 protein-expressing colorectal cancer cell lines and an antiPROK1mAb to study angiogenesis suppression. The PROK1 protein released from colorectal cancer cells was soluble at different concentrations in the extracellular supernatants. Using these culture supernatants, we examined subcutaneous angiogenesis in mice. Compared with the control medium, angiogenesis was intensified in response to the supernatants. Addition of the antiPROK1mAb suppressed this angiogenic activity. Thus, the anti-PROK1mAb might possibly suppress the activity of colorectal cancer-produced PROK1 protein and thus might suppress angiogenic activity. Furthermore, because the angiogenic growth factor reportedly plays an important role in cancer cell proliferation, we administered cancer cells and the anti-PROK1mAb to mice to investigate subcutaneous tumor formation. When the anti-PROK1mAb was added, tumor formation was significantly suppressed and the surrounding angiogenic activity was also reduced, suggesting the importance of PROK1 in colorectal cancer cells. In recent years, we studied the relationship between PROK1 and angiogenesis in clinical human colorectal cancer specimens.26 Significantly more blood vessels were found in the PROK1-positive specimens, compared with the PROK-1 negative samples, indicating the clinical significance of this factor. Also, Prokineticin receptors, when expressed on colorectal cancer cells, are likely to be involved in invasion via an autocrine mechanism, and PROK1 is considered to be a part of the important signal transduction network associated with metastasis.27 In our recent in vitro and in vivo experiment with colorectal cancer cell lines, we found that the antiPROK1mAb could suppress angiogenic activity and tumor formation, suggesting that our finding might be the first step toward a new therapy. ACKNOWLEDGMENT The technical assistance of Ms. Saitoh M. with this research was appreciated. This work was supported in part by a Grant-in-Aid for Science Research from the Ministry of Education, Sports, Science, and Technology of Japan (No. 25462047).
Anti-PROK1mAb Inhibits Angiogenesis and Tumor Growth DISCLOSURE interests.
The authors declare that they have no competing
REFERENCES 1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int. J Cancer. 2010;127:2893–917. 2. Watanabe T, Itabashi M, Shimada Y, Tanaka S, Ito Y, Ajioka Y, et al. Japanese society for cancer of the colon and rectum: Japanese society for cancer of the colon and rectum (JSCCR) guidelines 2010 for the treatment of colorectal cancer. Int J Clin Oncol. 2012;17:1–29. 3. Hanahan D, Folkman J. Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. Cell. 1996;86:353–64. 4. Ferrara N. Pathways mediating VEGF-independent tumor angiogenesis. Cytokine Growth Factor Rev. 2010;21:21–6. 5. Saltz LB, Clarke S, Dı´az-Rubio E, Scheithauer W, Figer A, Wong R, et al. Bevacizumab in combination with oxaliplatin-based chemotherapy as first-line therapy in metastatic colorectal cancer: a randomized phase III study. J Clin Oncol. 2008;26:2013–9. 6. Van Cutsem E, Ko¨hne CH, Hitre E, Zaluski J, Chang Chien CR, Makhson A, et al. Cetuximab and chemotherapy as initial treatment for metastatic colorectal cancer. N Engl J Med. 2009;360: 1408–17. 7. Amado RG, Wolf M, Peeters M, Van Cutsem E, Siena S, Freeman DJ, et al. Wild-type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer. J Clin Oncol. 2008;26:1626–34. 8. Grothey A, Van Cutsem E, Sobrero A, Siena S, Falcone A, Ychou M, et al. Regorafenib monotherapy for previously treated metastatic colorectal cancer (CORRECT): an international, multicenter, randomized, placebo-controlled, phase 3 trial. Lancet. 2013;381:303–12. 9. Tabernero J, Van Cutsem E, Lakomy R, Prausova´ J, Ruff P, van Hazel GA, et al. Aflibercept versus placebo in combination with fluorouracil, leucovorin and irinotecan in the treatment of previously treated metastatic colorectal cancer: prespecified subgroup analyses from the VELOUR trial. Eur J Cancer. 2014;50:320–31. 10. LeCouter J, Kowalski J, Foster J, Hass P, Zhang Z, Dillard-Telm L, et al. Identification of an angiogenic mitogen selective for endocrine gland endothelium. Nature. 2001;412:877–84. 11. Nagano H, Goi T, Koneri K, Hirono Y, Katayama K, Yamaguchi A. Endocrine gland-derived vascular endothelial growth factor (EG-VEGF) expression in colorectal cancer. J Surg Oncol. 2007;96:605–10. 12. Goi T, Fujioka M, Satoh Y, Tabata S, Koneri K, Nagano H, et al. Angiogenesis and tumor proliferation/metastasis of human colorectal cancer cell line SW620 transfected with endocrine glandderived-vascular endothelial growth factor, as a new angiogenic factor. Cancer Res. 2004;64:1906–10.
13. Ngan ES, Sit FY, Lee K, Miao X, Yuan Z, Wang W, et al. Implications of endocrine gland-derived vascular endothelial growth factor/prokineticin-1 signaling in human neuroblastoma progression. Clin Cancer Res. 2007;13:868–75. 14. Pasquali D, Rossi V, Staibano S, De Rosa G, Chieffi P, Prezioso D, et al. The endocrine-gland-derived vascular endothelial growth factor (EG-VEGF)/prokineticin 1 and 2 and receptor expression in human prostate: up-regulation of EG-VEGF/prokineticin 1 with malignancy. Endocrinology. 2006;147:4245–51. 15. Morales A, Vilchis F, Cha´vez B, Chan C, Robles-Dı´az G, Dı´azSa´nchez V. Expression and localization of endocrine glandderived vascular endothelial growth factor (EG-VEGF) in human pancreas and pancreatic adenocarcinoma. J Steroid Biochem Mol Biol. 2007;107:37–41. 16. Goi T, Yamaguchi A, Nakagawara G, Urano T, Shiku H, Furukawa K. Reduced expression of deleted colorectal carcinoma (DCC) protein in established colon cancers. Br J Cancer. 1998; 77:466–71. 17. Kowanetz M, Ferrara N. Vascular endothelial growth factor signaling pathways: therapeutic perspective. Clin Cancer Res. 2006;12:5018–22. 18. Avraamides CJ, Garmy-Susini B, Varner JA. Integrins in angiogenesis and lymphangiogenesis. Nat Rev Cancer. 2008;8:604–17. 19. Clarke MF, Dick JE, Dirks PB, Eaves CJ, Jamieson CH, Jones DL, et al. Cancer stem cells—perspectives on current status and future directions: AACR Workshop on cancer stem cells. Cancer Res. 2006;66:9339–44. 20. Robert GJ, Huch M, Clevers H. Stem cells and cancer of the stomach and intestine. Nat Rev Cancer. 2010;6:373–84. 21. Moumen M, Chiche A, Decraene C, Petit V, Gandarillas A, Deugnier MA, et al. Myc is required for b-catenin-mediated mammary stem cell amplification and tumorigenesis. Mol Cancer. 2013;12:132. 22. Olino K, Wada S, Edil BH, Pan X, Meckel K, Weber W, et al. Tumor-associated antigen expressing Listeria monocytogenes induces effective primary and memory T-cell responses against hepatic colorectal cancer metastases. Ann Surg Oncol. 2012;19: S597–607. 23. Kalluri R, Zeisberg M. Fibroblasts in cancer. Mol Oncol. 2006;4:392–401. 24. Kaplan RN, Riba RD, Zacharoulis S, Bramley AH, Vincent L, Costa C, et al. VEGFR1-positive haematopoietic bone marrow progenitors initiate the pre-metastatic niche. Nature. 2005;438:820–7. 25. Miller AR, McBride WH, Hunt K, Economou JS. Cytokine-mediated gene therapy for cancer. Ann Surg Oncol. 1994;1:436–50. 26. Goi T, Nakazawa T, Hirono Y, Yamaguchi A. Prokineticin 1 expression in gastrointestinal tumors. Anticancer Res. 2013;33: 5311–5. 27. Tabata S, Goi T, Nakazawa T, Kimura Y, Katayama K, Yamaguchi A. Endocrine gland-derived vascular endothelial growth factor strengthens cell invasion ability via prokineticin receptor 2 in colon cancer cell lines. Oncol. Rep. 2013;29:459–63.
