Endocrine gland-derived vascular endothelial growth factor (EG-VEGF) and its receptor PROKR2 are associated to human colorectal cancer progression and peritoneal carcinomatosis

Abstract

BACKGROUND:

The highest risk factor for mortality among malignant tumors is metastasis. Endocrine gland-derived vascular endothelial growth factor (EG-VEGF) is an angiogenic factor which biological activity is mediated via two G protein-coupled receptors, prokineticin receptor1 (PROKR1) and PROKR2. Recent studies suggested that EG-VEGF expression is deregulated in multiple cancers including colorectal cancer (CRC).

METHODS:

Using distinctive CRC and peritoneal carcinomatosis (PC) cohorts and a corresponding control cohort, we determined the circulating levels of EG-VEGF and its in situ expression, and that of its related receptors.

RESULTS:

Circulating EG-VEGF levels were significantly increased in patients with metastatic PC compared to CRC and control patients ( $p<$ 0.05). Furthermore, according to clinicopathologic examinations, local EG-VEGF expression correlated with higher tumor and nodal stages ( $p<$ 0.001) of CRC. EG-VEGF and PROKR2 were highly expressed in colorectal primary lesions compared to positive controls. PROKR1 expression was lower and did not change in tumor specimens. Also, EG-VEGF and its receptor PROKR2 were differentially expressed in the colorectal primary lesions and in the control groups.

CONCLUSION:

Altogether these findings suggest that EG-VEGF/receptors system might be an important actor in the CRC progression into PC and might be involved in the ability of tumor cells to invade other organs. Circulating EG-VEGF could be proposed as a prognostic marker in human CRC and its progression into PC.

Keywords

Prokineticin colorectal cancer peritoneal carcinomatosis EG-VEGF

1. Introduction

Associated metastasis to gastrointestinal cancer, especially colorectal cancer (CRC) still constitutes a serious health problem, and is the major cause of death in CRC patients, with approximately 50% of survival in patients with multiple metastatic sites [1, 2]. Among processes that contribute to the course of metastasis in CRC is angiogenesis. This process is one of the most critical steps in tumorigenesis [3, 4]. Recently, attention has been directed towards anti-angiogenesis therapies, in which tumor proliferation is inhibited by blocking the formation of new tumor vessels, using angiogenesis inhibitors and drugs [5, 6].

Endocrine gland-derived vascular endothelial growth factor (EG-VEGF), also known as Prokineticin-1 (PROK-1) is the canonical member of the prokineticin family. This family accounts two members PROK-1 and PROK-2, also called Bv8 [7]. EG-VEGF was reported to selectively act on the endothelium of endocrine gland cells [8, 9]. Although EG-VEGF exhibits the same actions on the vascular system as the ones described for VEGF, these factors do not have any homologies, they constitute distinct proteins from two different families, with multiple regulatory functions [7]. Prokineticin receptor (PROKR) 1 and 2 were identified as EG-VEGF receptors, they belong to G-protein coupled receptors and mediate multiple functions upon their activation by EG-VEGF or its analog, the prokineticin 2 also called Bv8 [10, 11]. Human EG-VEGF cDNA comprises 1378-bp, encoding for a protein of 105 amino acids, including a classical signal sequence of 19 residues and a mature protein of 86 amino acids with a molecular weight of 8.6 kDa. Its biological effects have been associated with neuronal survival, pain sensation, contraction of gastrointestinal smooth muscle, placental development and hematopoiesis [12, 13].

Among the different types of cancer, CRC is characterized by its susceptibility to trigger hematogeneous metastasis [14]. It is one of the third most common malignancies worldwide, and the fourth leading cause of cancer-related mortality that affects both genders equally [15]. The incidence and survival rates of CRC are significantly different between different countries around the world [18]. Once a gastrointestinal malignancy has reached the stage of peritoneal carcinomatosis (PC), this is usually considered as the final stage of the metastatic cancer.

To date, 15 to 20% of patients with metastatic CRC show isolated PC with significant clinical relevance. For a long time, PC has been considered a consequence of systemic dissemination of the cancer via the bloodstream [19]. It is now well established that angiogenic processes contribute to CRC progression. These processes are orchestrated by multiple growth factors that control the angiogenesis balance.

PC arising from digestive cancers is mostly associated with a poor prognosis. Although its symptoms are non-specific, indirect signs such as bowel obstruction ascites can be indicative but are not disease specific [19]. For CRC, peritoneal seeding of cancer cells leading to PC is rather common [20]. At the time of the first diagnosis, it is estimated that the peritoneal surface is already effected in 8–15% of CRC cases [21, 22]. Importantly, in approximately 25% of patients with recurrent CRC, the peritoneal cavity seems to be the only metastatic site of the disease [23, 24]. To date, much is known about CRC development; nevertheless, mechanisms that are associated with its progression are still to be investigated.

Recent studies reported positive correlations between the prokineticin family of proteins and CRC development. PROK-2 has recently been qualified as an angiogenic factor of colorectal cancer [25]. The canonical member of the prokineticin family, EG-VEGF has been reported to be associated with CRC development and metastasis [26, 27]. Importantly, EG-VEGF mRNA expression, measured within CRC tumor specimens was correlated to the severity of the CRC disease [28]. Nevertheless, no studies have investigated EG-VEGF circulating levels and expression in CRC and PC patients. Also, few studies have investigated the status of the expression of EG-VEGF receptors, PROKR1 and PROKR2 in CRC.

The present study was conducted to determineEG-VEGF, PROKR1 and PROKR2 expression in normal and tumor colorectal tissues. Both circulating and local levels were compared between control, CRC and PC patients.

2. Materials and methods

2.1 Blood collection

Collection and processing of human samples was approved by the university ethics committee, and informed consent was obtained from each patient. The consent authorizes the use of the blood and tissues for research purposes.

The study was restricted to patients, with histologically confirmed CRC and with PC as the only site of metastasis, with the ability to perform informed consent, for whom enough clinical documentation is available, with sampling at the primary CRC excision and with no previous chemotherapy for metastatic or advanced CRC.

Blood was collected into EDTA vacutainers and centrifuged at 5000 rpm for 10 minutes. Plasma was removed from the erythrocytes and stored at $-$ 80 ${}^{\circ}$ C immediately until defrosted and used for ELISA. We collected 50 plasma samples of CRC patients aged between 16 and 74 years, 8 plasma samples from patients with PC aged between 21 and 65 years, and 41 plasma samples collected from control aged between 16 and 73 years, control sera were kindly obtained from healthy volunteers at the Regional Blood Transfusion Center of Casablanca. The collection was performed between September 2015 and April 2016 at the Ibn Rochd University Hospital of Casablanca.

2.2 Tumor biopsies collection

Colorectal specimens were obtained from 10 patients undergoing surgical resection of human advanced CRC. In addition 10 colorectal tissues from normal patients were used as positive control. The control specimens were obtained from patients who undergone regular colonoscopy and/or resection for benign colorectal diseases, such as hemorrhoidectomy. Shortly after collection, a fragment of each colorectal tissue, was fixed in 4% (vol/vol) paraformaldehyde for 24 h at room temperature, embedded in paraffin and cut in 5 $\mu$ m sections using a microtome (Leica Microsystems Nusloch, Germany), at the department of anatomo-pathology at the Ibn Rochd University Hospital (Casablanca Morocco). All blood samples and tumor biopsies were collected from the same group of patients.

2.3 EG-VEGF ELISA

EG-VEGF was measured by ELISA (Peprotech, Neuilly-sur-Seine, France) in the collected plasma. A total of 0.5 $\mu$ g/mL EG-VEGF capture antibody diluted in 1xPBS (pH 7.4) was adsorbed onto 96-well sterile plate (Nunc Maxisorp) overnight at room temperature (100 $\mu$ L/well). The plate was washed 4 times with 1xPBS supplemented with 0.05% Tween-20 (PBS-T) before and after blocking with 300 $\mu$ l/well of 1% BSA diluted in 1xPBS for at least 1 h at RT. Two separated standard curves (ranging from 16 pg/mL to 1 ng/mL and blank (wash buffer)) were constructed to allow accurate reading of samples at upper and lower ranges of the assay. After adding recombinant human EG-VEGF standard and samples (100 $\mu$ l/well), the plate was incubated at RT for at least 2 h, gently washed 4 times, then the detection antibody (100 $\mu$ l/well, biotinylated goat anti-human EG-VEGF) was added. It was diluted in PBS-T 0.1% BSA to [0.5 $\mu$ g/ml] and incubated 2 h at RT. After washing, 6 $\mu$ l of Avidin Peroxidase diluted in PBS-T 0.1% BSA [1:2000] was added to the plate (100 $\mu$ l/well), the plate was then incubated at 37 ${}^{\circ}$ C for 30 min. 100 $\mu$ l per well of ABTS (2,2’-Azinobis [3-ethylbenzothiazoline-6-sulfonic acid]) substrate solution was then added. After 4 washes, absorbance was measured immediately with an ELISA plate reader at 405 nm with wavelength correction set at 650 nm. All samples were in the linear range of the standard curves. The detection limit of the assay was 16 pg/ml.

Table 1
Demographic characteristics of all controls, colorectal (CRC) and peritoneal carcinomatosis (PC) patients

Characteristics	Controls $n$ (%)	CRC $n$ (%)	OR [95% CI]	P-value	PC $n$ (%)	OR [95% CI]	P-value
Age (years)				0.437			0.258
Mean $\pm$ SD	44.6 $\pm$ 14.7	49 $\pm$ 12.9			52.6 $\pm$ 13.4
Median	44	49.5			55.5
Range	16–73	16–74			21–65
Gender			1.00 [0.2–3.9]	0.960		0.75 [0.4–14.9]	0.850
Male	13 (31.7)	20 (40)			3 (37.5)
Female	28 (68.3)	30 (60)			5 (62.5)
Origin			0.73 [0.1–4.2]	0.724		0.75 [0.03–14.9]	0.850
Urban	30 (73.2)	38 (76)			5 (62.5)
Rural	11 (26.8)	12 (24)			3 (37.5)
Antecedents
Diabetic	4 (9.8)	7 (14)	0.86 [0.7–0.9]	0.433	2 (25)
Hypertensive	2 (4.9)	3 (6)	1.00 [0.9–1.0]	0.819	2 (25)
Surgery	17 (41.5)	17 (34)	0.76 [0.2–2.7]	0.680	4 (50)
Smoking	6 (14.6)	14 (28)	0.71 [0.5–0.9]	0.132	3 (37.5)	0.5 [0.2–1.1]	0.206
Alcoholic	3 (7.3)	6 (12)	0.84 [0.7–0.9]	0.456	–
Family	–	13 (26)			1 (12.5)
Total	41	50			8

CRC, colorectal cancer; PC, peritoneal carcinomatosis; OR, Odds Ratio; CI, confidence interval; SD, standard deviation.

Table 2

Clinicopathologic characteristics of colorectal cancer patients

Characteristics	CRC (N $=$ 50)
	$n$	%
Tumor
Colon	32	64
Rectum	18	36
Histology
Leiberkuhnian ADK	40	80
Mucinous ADK	2	4
Tubular ADK	2	4
Differentiation
Poor	7	14
Moderate	27	54
Well	6	12
Symptoms
Vomit	14	28
Pelvic pain	31	62
Diagnosis delay
$>$ 6 months	30	60
$<$ 6 months	20	40
Tumor Stage
pTis	7	14
pT1	1	2
pT2	16	32
pT3	12	24
pT4	8	16
Nodal stage
N0	25	50
N1	12	24
N2	8	16
Metastases
M0	45	90
M1	5	10

2.4 Immunohistochemistry

Immunohistochemical staining was performed using the avidin-biotin-peroxidase method. Paraffin-embedded tissue Sections 5 $\mu$ m thick) placed on poly-L-lysine-coated slides were deparaffinized in 100% Xylene and rehydrated through decreasing grades of alcohol (100%, 95%, 70% and 50%). Endogenous peroxidase activity was quenched by pretreatment with 3% hydrogen peroxide (H ${}_{2}$ O ${}_{2}$ ) (Sigma Aldrich, France) in methanol for 20 min. Nonspecific binding sites were blocked with 5% normal Goat serum (NGS) diluted in 1X phosphate buffered saline (PBS1X) containing 2% Bovine Serum Albumin (BSA) for 45 min. Sections were incubated overnight in a humid chamber at 4 ${}^{\circ}$ C with monoclonal rabbit anti-human EG-VEGF, PROKR1 or PROKR2 primary antibodies (Covalab, Lyon, France) diluted at 1:150 (0.45 $\mu$ g/ml), 1:50 (0.35 $\mu$ g/ml) and 1:50 (0.50 $\mu$ g/ml) respectively in the blocking buffer. The tissue sections were subsequently washed three times with PBS and incubated with monoclonal biotinylated goat anti-rabbit IgGs (1:400 dilution in blocking solution; Sigma-Aldrich, Saint-Quentin Fallavier, France) for 1 h in a humid chamber at room temperature. After three PBS washes, the slides were incubated with an avidin-biotin complex (Vectastain ABC kit; Vector Laboratories, Burlingame, CA) for 1 h. After a final PBS wash, the immunoreactive proteins were visualized after the addition of two drops of freshly prepared 3,3’-diaminobenzidine (DAB) to sections for 30 sec to 1 min to give a Browne color (liquid DAB and substrate chromogen system) (Dako). The reaction was stopped in deionized water. All staining steps were performed at room temperature. The sections were then washed, lightly counterstained with hematoxylin, dehydrated, and mounted with Merckoglass. In each case negative controls were performed using blocking buffer instead of the primary antibody. Photographs were taken using a Zeiss AxioCam HRc coupled to a Zeiss Axioplan microscope.

2.5 Statistical analysis

Quantitative and qualitative variables were summarized in terms of median (range), mean (standard deviation), and frequency (percentage). Statistical comparisons between groups were made using chi-square or one-way ANOVA test. Calculations were performed using SPSS statistics version 21 (SPSS Inc., Chicago, IL, USA) and SigmaPlot (Jandel scientific software, San Rafael, CA). The differences were considered significant at $p$ values lower than 0.05 or 0.001.

3. Results

3.1 Patient characteristics

The CRC was located in the colon in 32 patients (64%), and in 18 patients (36%) in the rectum. The mean ages ( $\pm$ SD) of patients from whom the CRC and PC samples were obtained were 49 ( $\pm$ 12.9) and 52.6 ( $\pm$ 13.5) years, respectively. The mean age ( $\pm$ SD) of the control individuals was 44.6 ( $\pm$ 14.7) years. All patients and controls baseline demographic and clinical characteristics, including age, gender, origin, personal antecedents, diagnostic delay, TNM (Tumor Node Metastasis) stage of the tumor, were collected, and are summarized in Tables 1 and 2.

The experimental groups consisted of 58 patients of both genders (CRC $n=$ 50, 60% female; PC $n=$ 8, 62.5% female), most of them lives in urban areas, 60% of CRC patients (Table 2) and 50% of PC patients had a diagnosis delay greater than 6 months from the time of the onset of the symptoms. 41 healthy volunteers of both genders constitute the control group (68.3% female); 73.2% of them live in urban areas. The group aging between 40–60 years represented the most affected group; this was observed both for the CRC (54%) and PC 62.5% groups (Fig. 1). According to our data, there were no statistically significant differences in the characteristics between all groups. However, three factors: diabetes (OR [95% CI]: 0.86 [0.7–0.9]), smoking (OR [95% CI]: 0.71 [0.5–0.9]) and alcohol (OR [95% CI]: 0.84 [0.7–0.9]) were correlated to CRC development. Interestingly, gender, smoking and origin are not considered as risk factors to PC development (OR [95% CI]: 0.75 [0.4–14.9], 0.5 [0.2–1.1], 0.75 [0.03–14.9] respectively). However no significant association or statistical significance could be drawn ( $p>$ 0.05) (Table 1).

Figure 1.

Distribution of all control colorectal cancer (CRC) and peritoneal carcinomatosis (PC) patients by age groups.

Figure 2.

EG-VEGF serum levels measured by ELISA in control (CTL), colorectal cancer (CRC) and peritoneal carcinomatosis (PC) patients. A total of 99 serum samples were analyzed (* $p<$ 0.05, *** $p<$ 0.001).

Considering the serum tumor Carcinoembryonic antigen (CEA) marker of our patients, we analyzed the relationship of its levels in correlation with the TNM stage of CRC. Positive correlation ( $p<$ 0.001) was found between CEA levels, tumor size and metastatic stages (Fig. 3A).

Of the 50 patients with CRC, 8 (16%) had liver metastasis as site of tumor dissemination. In the PC patients, no clinical or radiological evidence of other systemic spreading associated to PC was observed.

Figure 3.

A) CEA circulating levels in relation to TNM stage of CRC. (* $p<$ 0.001). B) Relationship between EG-VEGF circulating levels and the TNM stage in all CRC patients. pTis: primary carcinoma in situ, N: nodal stage, M: metastatic stage, ns: no significance (* $p<$ 0.001).

Figure 4.

EG-VEGF protein expression in healthy human colorectal mucosa and human primary colorectal cancer tissues. Expression was assessed by immunohistochemistry. Representative photographs show EG-VEGF immunoreactivity in normal human colorectal mucosa (A), in human primary colorectal cancer (B). Photograph in panel C reports a negative control. All sections were counterstained with heamatoxylin. TC: tumor cell, CL: crypts of lieberkùhn, MM: muscularis mucosae, LP: lamina propria, L: Lumen, GC: goblet cells, AE: absorptive erythrocytes. Magnification: A: x400, B and C: x200.

Figure 5.

Immunohistochemical staining of PROKR1 in human colorectal tissues. Photographs show representative PROKR1 immunoreactivity in normal human colorectal mucosa (A), in human primary colorectal cancer (B) and a negative control (C). All sections were counterstained with Heamatoxylin. TC: tumor cell, CL: crypts of lieberkùhn, LP: lamina propria, GC: goblet cells, AE: absorptive erythrocytes, MM: muscularis mucosae, L: Lumen. Magnification: A: x400, B and C: x200.

3.2 Circulating EG-VEGF in human colorectal cancer and peritoneal carcinomatosis

First, we compared EG-VEGF protein circulating levels in the control group and in the patient groups with CRC and PC, using ELISA test. Circulating EG-VEGF was detectable in all tested sera. There was a significant difference between control cases, CRC patients and PC patients ( $p<$ 0.05). CirculatingEG-VEGF levels were around 407.451 pg/ml in CRC patients, and around 309.392 pg/ml in control cases. This value increased 2.4-fold (754.143 pg/ml) in PC patients (Fig. 2) and the differences in the mean values between groups were statistically significant ( $p<$ 0.05, $p<$ 0.001).

3.3 EG-VEGF protein expression and clinicopathologic factors in CRC patients

Analysis of the relationship between EG-VEGF circulating levels in CRC patients and the clinicopathologic characteristics reported interesting correlations when considering the TNM classification of CRC. The expression of EG-VEGF was positively correlated with the tumor size (pT3, pT4) and nodal (N2) stage of the disease (Fig. 3B). These correlations were statistically significant ( $p<$ 0.001). However, a negative correlation was found between metastatic stage and the circulating EG-VEGF ( $p>$ 0.001).

3.4 Immunolocalization of EG-VEGF protein and its receptors PROKR1 and PROKR2 in colorectal cancer

3.4.1 Comparison of EG-VEGF protein expression in normal and colorectal cancer tissues

The EG-VEGF protein was strongly expressed in the primary lesion of CRC in comparison with normal colorectal tissue (Fig. 4A and B). Its expression in healthy human colorectal tissue was weak, heterogeneous and limited to the cytoplasm of tubular epithelial cells (crypts of lieberkùhn). The staining was detected in the cytoplasm of both the goblet and the absorptive cells. In the primary lesion of CRC, strong staining was observed in both the cell membrane and the cytoplasm, also we observed an expression in the nucleus of the lamina propria cells. The negative control did not show any staining. Representative photographs are reported in Fig. 4C.

3.4.2 Comparison of PROKR1 and PROKR2 protein expression in normal and colorectal cancer tissues

The expression of PROKR1 and PROKR2 receptors in normal and CRC tissues showed heterogeneous localizations in CRC cases compared to control tissues. PROKR1 was localized both in the nucleus and in the cytoplasm of tumor cells with weak expression in the lamina propria, in CRC cases (Fig. 5B). Overall, in positive controls, widespread cytoplasmic distribution of PROKR1 expression was observed in absorptive cells along with a weak expression in the cells of the lamina propria (Fig. 5A). Overall, the intensity and patterns of this distribution varied. No significant difference could be observed when comparing control and CRC sections (Fig. 5). In contrast, PROKR2 protein was localized predominantly, both at the cell membrane and in the cytoplasm of tumor cells in CRC lesions, and the intensity was stronger at the cell membrane compared to the cytoplasm (Fig. 6B). In positive control cases, PROKR2 expression was especially localized in the nuclei and the cytoplasm of lamina propria cells and erythrocytes (Fig. 6A). No staining was observed in all negative controls. PROKR2 immunoreactivity was significantly higher in the colorectal primary lesions, compared to the positive control sections (Fig. 6).

Figure 6.

Comparison of PROKR2 immunohistochemical staining in normal and pathological colorectal mucosa. PROKR2 immunoreactivity was detected in normal human colorectal mucosa (A) and in human primary colorectal cancer (B). (C) Represent a negative control of the staining. All sections were counterstained with Hematoxylin. TC: tumor cell, CL: crypts of lieberkùhn, LP: lamina propria, GC: goblet cells, AE: absorptive erythrocytes. Magnification: A and C: x400, B: x200.

4. Discussion

Prokineticins are a family of secreted peptides with diverse regulatory roles [29]. EG-VEGF, the canonical member of this family exhibits the same effects on endothelial cells as those described for VEGF, the classical angiogenic factor. Nevertheless, EG-VEGF is structurally different from VEGF and uses different types of receptors. According to previous studies, the expression of EG-VEGF in normal human tissue is limited to steroidogenic tissues hormone-producing cells; including, ovary, testis and placenta [30], and does not occur in normal gastrointestinal membranes such as the stomach and colon [7, 31]. This factor is strongly involved in normal and pathological angiogenesis [32] and plays an important role in the regulation of tumor cell growth and survival [4, 13, 33, 34].

In the present study, we searched for a potential new marker for the growth of CRC and its progression to PC. Recent studies suggested that EG-VEGF might be deregulated in this type of cancer without investigating its circulating levels in patient’s vs normal individuals. Our results confirmed EG-VEGF and PROKR2 deregulations in situ in the CRC and bring strong evidences of its increase in the blood stream of CRC and PC patients, suggesting its potential use as prognostic biomarker of the disease.

CRC is a major cause of cancer-related morbidity and mortality worldwide [31, 35]. Our results show that CRC typically occurs in middle-age to elderly patients. In addition, most affected individuals are aged between 40–60 years. Hence, we propose that circulatingEG-VEGF and its levels of expression and localization in CRC could be informative on its progression into PC.

In our study, EG-VEGF protein levels increased in CRC patients compared to the control group and were even higher in PC; this strongly supports a potential role of this factor in the processes that contributes to the progression and metastasis of CRC into PC. This results are concordant to previous finding of Nagano et al. who demonstrated that in situ EG-VEGF mRNA levels are correlated to the severity of CRC [28]. Importantly, EG-VEGF was also shown to increase the invasion of tumor cells, which is an indicator of cell malignancy and an important element for the development of hematogenous, lymph node and peritoneal metastasis [36].

We observed that the mean value of circulatingEG-VEGF was higher in patient’s age group of 41–50 years old, however among PC patients, the higher mean value of circulating EG-VEGF was observed in patients with age group 51–60 years. Tumor staging was by far the most important prognostic predictor of clinical outcome for patients with CRC. We found that the expression of EG-VEGF was increased in patients with high tumor size (pT3, pT4) and nodal (N2) stage of CRC, as well as in those with high CEA levels, which correlate with advanced tumor stage. In a recent study, the CEA levels were reported to correlate with the number of circulating tumor cells (CTCs) in CRC patients [37]. This relationship between the increase in the levels of EG-VEGF expression and the progression of the tumor, like the traditional tumor biomarker CEA, confirms and supports the idea that EG-VEGF could be considered to act as prognosis-determining factor in the progression of human CRC. In accordance with a previous study, tumor (T) stage, nodal stage and liver metastasis were considered independent predictors of metachronous colorectal PC [38]. Moreover, Nakazawa et al. [26], suggest that the recurrence of CRC was more frequent and the prognosis was worse for the stage T3 patients with positive lesions expressing EG-VEGF compared to stage T3 negative lesions. This indicates that EG-VEGF expression could be an important factor for the infiltration and metastasis of human CRC. In the same study, clinicopathologic examination confirmed that EG-VEGF expression frequently occurred in cases with vascular invasion and hematogenous metastasis, including liver metastasis [26].

Goi et al. defined EG-VEGF and its receptorPROKR2 as prognostic markers in colorectal carcinoma [39, 40]. Importantly, this group also demonstrated that synergistic anti-tumoral effects were obtained when both EG-VEGF and VEGF signaling pathways were targeted. These findings strongly suggest complementary actions between VEGF andEG-VEGF in CRC development and progression [41].

It was estimated that the peritoneal cavity was the main site of metastases in approximately 25% of patients with CRC [23]. Furthermore, multiple studies support the association between angiogenic factors and tumor malignancy [42]. Kim et al. [43] showed increased angiogenesis in tumors with rapid proliferation, whereas Folkman [3] reported an increase in the tumor volume due to the active angiogensis in tumor tissue, confirming that tumor growth is angiogenesis dependent.

Multiple studies reported the association ofEG-VEGF with malignant tumors. These studies reported that, i) increased EG-VEGF expression is associated with increased malignancy of prostate cancer, neuroblastoma, thyroid cancer and pancreatic duct cancer, ii) EG-VEGF mRNA expression in the resected primary lesion of CRC patients is associated with a worse prognosis, and that iii) increasedEG-VEGF expression is associated with angiogenesis involving hepatic metastasis [14, 36, 44]. Hence, these studies strongly suggest that EG-VEGF must be considered as a new central actor of cancer development and progression.

PC from CRC was considered as a terminal condition resulting from tumor with median survival not exceeding six months [38]. Due to the low sensitivity of preoperative imaging techniques, the exact incidence of PC is unknown [45, 46]. Jayne et al. [38] reported that 8% from a total of 3019 CRC patients presented a synchronous PC and 5% with metachronous PC. In addition, a recent study in Sweden by Segelman et al. [47] confirmed that 4.3% of 11124 CRC patients presented with synchronous PC, while 4% with metachronous PC.

The immunolocalization experiments performed in this study revealed interesting patterns of distribution of the EG-VEGF and its two receptors PROKR1 and PROKR2 in CRC sections. This particular aspect strongly suggests the involvement of EG-VEGF as a potential prognostic factor in CRC. EG-VEGF, PROKR1 and PROKR2 were slightly expressed in normal colorectal mucosa, and highly expressed in CRC cells, indicating that EG-VEGF/receptors systems could be considered as new targets to treat CRC. Previous study evaluating the expression of PROKR2, reported that the prevalence of PROKR2 expression increased as tumor stage advances. This study suggested that PROKR2 could be used as a prognostic factor in human CRC.

Metastasis is the strongest risk factor for mortality among tumors malignancies. Understanding the mechanism of such processes may lead to new therapies of cancer. Many discoveries suggest that prokineticins, expressed at higher levels in cancers compared to normal tissues, may be cancer specific and, possibly, tissue specific targets. These aspects could open new frames for cancer-specific therapies.

4.1 Conclusion

Our results support the use of EG-VEGF and its rec- eptor PROKR2 as predictor and prognosis-determining factors of the progression of the CRC into PC. These proteins could also, represent therapeutic targets in the pathogenesis of this cancer.

Limits of the study

Because no previous study has addressed the comparison of EG-VEGF circulating levels in colorectal cancer, the present study was conducted with the main objective to measure and compare circulating EG-VEGF levels in control vs CRC and PC patients. In situ comparison of the levels of EG-VEGF, PROKR1 and PROKR2 expression was also investigated. While, EG-VEGF measurement at the circulating level were conclusive and brought strong informative new message, in situ analyses will need more sampling to better quantify the local levels of expression of the prokineticin system. A large cohort including, sera, mRNA and protein specimens collected in CRC and PC patients is under construction to validate these results at larger scale.

Footnotes

Acknowledgments

We acknowledge the following sources of funding: Institut National de la Santé et de la Recherche Médicale (U1036), University Grenoble-Alpes, Commissariat à l’Energie Atomique (DSV/iRTSV/BCI), University Hassan II of Casablanca, Faculty of Medicine and Pharmacy of Casablanca, and the National Center for Scientific and Technical Research (CNRST) of Rabat.

Conflict of interest

The authors declare no potential conflicts of interest.

Abbreviations

EG-VEGF: Endocrine Gland-Derived Vascular Endothelial Growth Factor. PROK: Prokineticin. CRC: Colorectal Cancer. PC: Peritoneal Carcinomatosis.CEA: Carcinoembryonic Antigen.

References

Kanthan

Senger

J.-L.

and Kanthan

S.C.

, Molecular events in primary and metastatic colorectal carcinoma: A review, Pathol Res Int 2012 (2012). doi: 10.1155/2012/597497.

de Krijger

Mekenkamp

L.J.M.

Punt

C.J.A.

and Nagtegaal

I.D.

, MicroRNAs in colorectal cancer metastasis, J Pathol 224 (2011), 438–447. doi: 10.1002/path.2922.

Folkman

, Tumor angiogenesis: Therapeutic implications, N Engl J Med 285 (1971), 1182–1186. doi: 10.1056/NEJM197111182852108.

Yancopoulos

G.D.

Davis

Gale

N.W.

Rudge

J.S.

Wiegand

S.J.

and Holash

, Vascular-specific growth factors and blood vessel formation, Nature 407 (2000), 242–248. doi: 10.1038/35025215.

Hotz

H.G.

Reber

H.A.

Hotz

Sanghavi

P.C.

Foitzik

Buhr

H.J.

and Hines

O.J.

, Angiogenesis inhibitor TNP-470 reduces human pancreatic cancer growth, J Gastrointest Surg Off J Soc Surg Aliment Tract 5 (2001), 131–138.

O’Reilly

M.S.

Boehm

Shing

Fukai

Vasios

Lane

W.S.

Flynn

Birkhead

J.R.

Olsen

B.R.

and Folkman

, Endostatin: An endogenous inhibitor of angiogenesis and tumor growth, Cell 88 (1997), 277–285.

LeCouter

Kowalski

Foster

Hass

Zhang

Dillard-Telm

Frantz

Rangell

DeGuzman

Keller

G.A.

Peale

Gurney

Hillan

K.J.

and Ferrara

, Identification of an angiogenic mitogen selective for endocrine gland endothelium, Nature 412 (2001), 877–884. doi: 10.1038/35091000.

Kaser

Winklmayr

Lepperdinger

and Kreil

, The AVIT protein family, EMBO Rep 4 (2003), 469–473. doi: 10.1038/sj.embor.embor830.

Ferrara

LeCouter

Lin

and Peale

, EG-VEGF and Bv8: A novel family of tissue-restricted angiogenic factors, Biochim Biophys Acta 1654 (2004), 69–78. doi: 10.1016/j.bbcan.2003.07.001.

10.

Soga

ichiro Matsumoto

Oda

Saito

Hiyama

Takasaki

Kamohara

Ohishi

Matsushime

and Furuichi

, Molecular cloning and characterization of prokineticin receptors, Biochim Biophys Acta 1579 (2002), 173–179.

11.

Masuda

Takatsu

Terao

Kumano

Ishibashi

Suenaga

Abe

Fukusumi

Watanabe

Shintani

Yamada

Hinuma

Inatomi

Ohtaki

Onda

and Fujino

, Isolation and identification of EG-VEGF/prokin- eticins as cognate ligands for two orphan G-protein-coupled receptors, Biochem Biophys Res Commun 293 (2002), 396–402. doi: 10.1016/S0006-291X(02)00239-5.

12.

Bullock

C.M.

Knauer

D.J.

Ehlert

F.J.

and Zhou

Q.Y.

, Identification of two prokineticin cDNAs: Recombinant proteins potently contract gastrointestinal smooth muscle, Mol Pharmacol 59 (2001), 692–698.

13.

Lecouter

Lin

and Ferrara

, EG-VEGF: A novel mediator of endocrine-specific angiogenesis, endothelial phenotype, and function, Ann N Y Acad Sci 1014 (2004), 50–57.

14.

Ngan

E.S.W.

Sit

F.Y.L.

Lee

K.L.

Miao

Yuan

Wang

Nicholls

J.M.

Wong

K.K.Y.

Garcia-Barcelo

Lui

V.C.H.

and Tam

P.K.H.

, Implications of endocrine gland – derived vascular endothelial growth factor/prokineticin-1 signaling in human neuroblastoma progression, Clin Cancer Res 13 (2007), 868–875. doi: 10.1158/1078-0432.CCR-06-2176.

15.

Haggar

F.A.

and Boushey

R.P.

, Colorectal cancer epidemiology: Incidence, mortality, survival, and risk factors, Clin Colon Rectal Surg 22 (2009), 191–197. doi: 10.1055/s-0029-1242458.

16.

Cunningham

Atkin

Lenz

H.-J.

Lynch

H.T.

Minsky

Nordlinger

and Starling

, Colorectal cancer, Lancet Lond Engl 375 (2010), 1030–1047. doi: 10.1016/S0140-6736(10)60353-4.

17.

Siegel

Naishadham

and Jemal

, Cancer statistics, 2012, CA Cancer J Clin 62 (2012), 10–29. doi: 10.3322/caac.20138.

18.

Labianca

Beretta

G.D.

Kildani

Milesi

Merlin

Mosconi

Pessi

M.A.

Prochilo

Quadri

Gatta

de Braud

and Wils

, Colon cancer, Crit Rev Oncol Hematol 74 (2010), 106–133. doi: 10.1016/j.critrevonc.2010.01.010.

19.

Piso

and Arnold

, Multimodal treatment approaches for peritoneal carcinosis in colorectal cancer, Dtsch Ärztebl Int 108 (2011), 802–808. doi: 10.3238/arztebl.2011.0802.

20.

Verwaal

V.J.

van Ruth

Witkamp

Boot

van Slooten

and Zoetmulder

F.A.N.

, Long-term survival of peritoneal carcinomatosis of colorectal origin, Ann Surg Oncol 12 (2005), 65–71. doi: 10.1007/s10434-004-1167-z.

21.

Gómez Portilla

Cendoya

de Tejada

I. López

Olabarría

de Lecea

C. Martínez

Magrach

Gil

Echevarrí a

Valdovinos

and Larrabide

, Carcinomatosis peritoneal de origen Estado actual del tratamiento: Revisión y puesta al día, Rev Esp Enfermedades Dig 97 (2005), 716–737.

22.

Russell

A.H.

Pelton

Reheis

C.E.

Wisbeck

W.M.

Tong

D.Y.

and Dawson

L.E.

, Adenocarcinoma of the colon: An autopsy study with implications for new therapeutic strategies, Cancer 56 (1985), 1446–1451. doi: 10.1002/1097-0142(19850915)56:6<1446::AID-CNCR2820560636>3.0.CO;2-5.

23.

Koppe

M.J.

Boerman

O.C.

Oyen

W.J.G.

and Bleichrodt

R.P.

, Peritoneal carcinomatosis of colorectal origin, Ann Surg 243 (2006), 212–222. doi: 10.1097/01.sla.0000197702.46394.16.

24.

Carraro

P.G.S.

Segala

Cesana

B.M.

and Tiberio

, Obstructing colonic cancer: Failure and survival patterns over a ten-year follow-up after one-stage curative surgery, Dis Colon Rectum 44(2) (2001), 243–250. doi: 10.1007/BF02234300.

25.

Kurebayashi

Goi

Shimada

Tagai

Naruse

Nakazawa

Kimura

Hirono

and Yamaguchi

, Prokineticin 2 (PROK2) is an important factor for angiogenesis in colorectal cancer, Oncotarget 6 (2015), 26242–26251. doi: 10.18632/oncotarget.4385.

26.

Nakazawa

Goi

Hirono

and Yamaguchi

, Prokineticin 1 protein expression is a useful new prognostic factor for human sporadic colorectal cancer, Ann Surg Oncol 22 (2015), 1496–1503. doi: 10.1245/s10434-014-4150-3.

27.

Tabata

Goi

Nakazawa

Kimura

Katayama

and Yamaguchi

, Endocrine gland-derived vascular endothelial growth factor strengthens cell invasion ability via prokineticin receptor 2 in colon cancer cell lines, Oncol Rep 29 (2013), 459–463. doi: 10.3892/or.2012.2124.

28.

Nagano

Goi

Koneri

Hirono

Katayama

and Yamaguchi

, Endocrine gland-derived vascular endothelial growth factor (EG-VEGF) expression in colorectal cancer, J Surg Oncol 96 (2007), 605–610. doi: 10.1002/jso.20716.

29.

Monnier

and Samson

, Cytokine properties of prokineticins, FEBS J 275 (2008), 4014–4021. doi: 10.1111/j.1742-4658.2008.06559.x.

30.

Brouillet

Murthi

Hoffmann

Salomon

Sergent

Mazancourt

P.D.

Dakouane-Giudicelli

Dieudonné

M.N.

Rozenberg

Vaiman

Barbaux

Benharouga

Feige

J.-J.

and Alfaidy

, EG-VEGF controls placental growth and survival in normal and pathological pregnancies: Case of fetal growth restriction (FGR), Cell Mol Life Sci 70 (2013), 511–525. doi: 10.1007/s00018-012-1141-z.

31.

Pasquali

Rossi

Staibano

De Rosa

Chieffi

Prezioso

Mirone

Mascolo

Tramontano

Bellastella

and Sinisi

A.A.

, 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 147 (2006), 4245–4251. doi: 10.1210/en.2006-0614.

32.

Monnier

and Samson

, Prokineticins in angiogenesis and cancer, Cancer Lett 296 (2010), 144–149. doi: 10.1016/j.canlet.2010.06.011.

33.

Ren

L.-N.

Q.-F.

Xiao

F.-J.

Yan

Yang

Y.-F.

Wang

L.-S.

Guo

X.-Z.

and Wang

, Endocrine glands-derived vascular endothelial growth factor protects pancreatic cancer cells from apoptosis via upregulation of the myeloid cell leukemia-1 protein, Biochem Biophys Res Commun 386 (2009), 35–39. doi: 10.1016/j.bbrc.2009.05.149.

34.

Traboulsi

Sergent

Boufettal

Brouillet

Slim

Hoffmann

Benlahfid

Zhou

Q.Y.

Balboni

Onnis

Bolze

P.-A.

Salomon

Sauthier

Mallet

Aboussaouira

Feige

J.-J.

Benharouga

and Alfaidy

, Antagonism of EG-VEGF receptors as targeted therapy for choriocarcinoma progression in vitro and in vivo , Clin Cancer Res Off J Am Assoc Cancer Res (2017). doi: 10.1158/1078-0432.CCR-17-0811.

35.

Chen

Zheng

Baade

P.D.

Zhang

Zeng

Bray

Jemal

X.Q.

and He

, Cancer statistics in China, 2015, CA Cancer J Clin 66 (2016), 115–132. doi: 10.3322/caac.21338.

36.

Goi

, Endocrine gland-derived vascular endothelial growth factor strengthens cell invasion ability via prokineticin receptor 2 in colon cancer cell lines, Oncol Rep (2012). doi: 10.3892/or.2012.2124.

37.

Zhao

J.-X.

Liu

L.-R.

Yang

X.-Y.

Liu

and Zhang

Z.-G.

, Serum CA19-9 as a marker of circulating tumor cells in first reflux blood of colorectal cancer patients, Oncotarget 8 (2017), 67918–67932. doi: 10.18632/oncotarget.18912.

38.

Jayne

D.G.

Fook

Loi

and Seow-Choen

, Peritoneal carcinomatosis from colorectal cancer, Br J Surg 89 (2002), 1545–1550. doi: 10.1046/j.1365-2168.2002.02274.x.

39.

Goi

, Kurebayashi

, Ueda

, Naruse

, Nakazawa

, Koneri

, Hirono

, Katayama

, Yamaguchi

,Expression of prokineticin-receptor2(PK-R2) is a new prognostic factor in human colorectal cancer, Oncotarget 6 (2015), 31758–31766. doi: 10.18632/oncotarget.5565.

40.

Goi

Nakazawa

Hirono

and Yamaguchi

, The prognosis was poorer in colorectal cancers that expressed both VEGF and PROK1(No correlation coefficient between VEGF and PROK1), Oncotarget 6 (2015), 28790–28799.

41.

Goi

Nakazawa

Hirono

Yamaguchi

Goi

Nakazawa

Hirono

and Yamaguchi

, The anti-tumor effect is enhanced by simultaneously targeting VEGF and PROK1 in colorectal cancer, Oncotarget 6 (2015), 6053–6061. doi: 10.18632/oncotarget.3474.

42.

Risau

, Mechanisms of angiogenesis, Nature 386 (1997), 671–674. doi: 10.1038/386671a0.

43.

Kim

K.J.

Winer

Armanini

Gillett

Phillips

H.S.

, and Ferrara

, Inhibition of vascular endothelial growth factor-induced angiogenesis suppresses tumour growth in vivo , Nature 362 (1993), 841–844. doi: 10.1038/362841a0.

44.

Jiang

, Abiatari

, Kong

, Erkan

, Oliveira

T. De

, Giese

N.A.

, Michalski

C.W.

, Friess

, Kleeff

,Pancreatic islet and stellate cells are the main sources of endocrine gland-derived vascular endothelial growth factor/prokineticin-1 in pancreatic cancer, Pancreatol Off J Int Assoc Pancreatol IAP Al 9 (2009), 165–172. doi: 10.1159/000178888.

45.

Klaver

Y.L.

Lemmens

V.E.

Nienhuijs

S.W.

Luyer

M.D.

and de Hingh

I.H.

, Peritoneal carcinomatosis of colorectal origin: Incidence, prognosis and treatment options, World J Gastroenterol WJG 18 (2012), 5489–5494. doi: 10.3748/wjg.v18.i39.5489.

46.

Pasqual

E.M.

Bertozzi

Bacchetti

Londero

A.P.

Basso

S.M.M.

Santeufemia

D.A.

Lo Re

and Lumachi

, Preoperative assessment of peritoneal carcinomatosis in patients undergoing hyperthermic intraperitoneal chemotherapy following cytoreductive surgery, Anticancer Res 34 (2014), 2363–2368.

47.

Segelman

Granath

Holm

Machado

Mahteme

and Martling

, Incidence, prevalence and risk factors for peritoneal carcinomatosis from colorectal cancer, Br J Surg 99 (2012), 699–705. doi: 10.1002/bjs.8679.

Endocrine gland-derived vascular endothelial growth factor (EG-VEGF) and its receptor PROKR2 are associated to human colorectal cancer progression and peritoneal carcinomatosis

Abstract

BACKGROUND:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Materials and methods

2.1 Blood collection

2.2 Tumor biopsies collection

2.3 EG-VEGF ELISA

Table 1 Demographic characteristics of all controls, colorectal (CRC) and peritoneal carcinomatosis (PC) patients

2.5 Statistical analysis

3. Results

3.1 Patient characteristics

3.3 EG-VEGF protein expression and clinicopathologic factors in CRC patients

3.4 Immunolocalization of EG-VEGF protein and its receptors PROKR1 and PROKR2 in colorectal cancer

3.4.1 Comparison of EG-VEGF protein expression in normal and colorectal cancer tissues

3.4.2 Comparison of PROKR1 and PROKR2 protein expression in normal and colorectal cancer tissues

4.1 Conclusion

Limits of the study

Footnotes

Acknowledgments

Conflict of interest

Abbreviations

References

Table 1
Demographic characteristics of all controls, colorectal (CRC) and peritoneal carcinomatosis (PC) patients