Abstract
Objective
Angiogenesis depends on interaction between a variety of promoting and inhibiting factors, and is known to involve vascular endothelial growth factor (VEGF) A and the Notch signaling pathway. The present study investigated the expression of Notch ligand delta-like (DLL) 4 (Drosophila), and VEGFA in colon cancer and colorectal adenoma tissue, and the association with tumour angiogenesis.
Methods
Protein level DLL4, VEGFA and CD34 molecule (CD34) expression was detected immunohistochemically in tissue sections from patients with colon cancer and colorectal adenoma.
Results
Out of 80 cases (35 with colon cancer, 45 with colorectal adenoma) DLL4 and VEGFA expression was closely related to tumour diameter, clinical stage, histological grade and lymph node metastasis. DLL4 expression was significantly higher in colon cancer tissue than colorectal adenoma tissue.
Conclusion
High levels of DLL4 expression were closely related to metastasis and prognosis in patients with colon cancer. The results of the present study support the conclusion that prognosis of colon cancer is significantly correlated with angiogenesis.
Introduction
The growth and metastasis of colon cancer are angiogenesis-dependent, therefore, progressively more anti-angiogenic strategies are being applied in the treatment of colon cancer.1–3 The Notch signalling pathway is known to play an important role in angiogenesis, and the migration, proliferation and differentiation of endothelial cells are regulated by members of the fibroblast growth factor and vascular endothelial growth factor (VEGF) gene families.4–7 In mammalian cells, the Notch signalling pathway comprises five membrane-bound ligands (Jaggedl, Jagged2, Delta-like (DLL)1, DLL3 and DLL4) and four Notch transmembrane receptors (Notch l–4). 8 DLL4 is expressed during angiogenesis at the site of proliferating endothelial cells. 9 Increased expression of DLL4 has been shown in the vascular system of breast cancer, colon cancer and ovarian cancer.10–12 In human endothelial cells, DLL4 has an important role in regulating proliferation, migration, survival and the formation of the endothelial reticular system. 13
The present study investigated expression of DLL4 and VEGFA in tissue from patients with colorectal adenoma and colon cancer to explore the value of assessing DLL4 levels in determining the prognosis of colon cancer, and the feasibility of DLL4-targeted therapy. The correlation between DLL4 and VEGFA expression and microvessel density was investigated, together with differences in expression of related factors between colorectal adenoma and colon cancer.
Patients and methods
Study population
Consecutive patients with colorectal cancer or colorectal adenoma (confirmed histologically by the Department of Gastrointestinal Surgery, Renmin Hospital of Wuhan University, Wuhan, Hubei, China), who underwent radical surgery at the Department of Gastrointestinal Surgery, Renmin Hospital of Wuhan University between January 2009 and August 2011, were eligible for inclusion in this study. None of the patients received postoperative adjuvant therapy. Histological analysis was used to define all colorectal cancer tumours according to stage (I–IV), using the tumour–node–metastasis (TNM) classification system developed by the Union for International Cancer Control. Patients’ clinicopathological data were collected, including tumour growth pattern, tumour location, tumour differentiation, lymph node metastases and liver metastases. There were no additional inclusion or exclusion criteria.
The Ethics Committee of the Cancer Research Centre of Wuhan University approved the study and all patients provided written informed consent.
Immunohistochemistry
Colorectal cancer and adenoma tissue was obtained from patients who had undergone surgical treatment or polypectomy. Colorectal adenoma samples were limited to large adenomas (≥10 mm in diameter). Tissue samples were fixed in 10% formalin, embedded in paraffin wax, and cut into 4 µm-thick sections. Immunohistochemical staining was performed using an avidin–biotin–peroxidase (ABC) method. Briefly, after deparaffinizing in xylene and rehydrating in graded ethanol solutions, sections were heated in citrate buffer (0.01 M, pH 6.5) at 120℃ for 10 min. The antigen retrieval step was identical for each antibody used. Sections were then incubated with mouse anti-human DLL4 (1:500 dilution; Santa Cruz Biotechnology, Santa Cruz, CA, USA) or mouse anti-human VEGF (1:1000 dilution; Santa Cruz Biotechnology) overnight at 4–8℃. Sections were washed four times with 0.1 M phosphate buffered saline (PBS; pH 7.4), then incubated with biotinylated goat anti-mouse immunoglobulin G (1:200 dilution; Vector Laboratories, Burlingame, CA, USA) at room temperature for 90 min. Slides were washed three times with 0.1 M PBS (pH 7.4) and immunoreactive staining was visualized using 0.5 mg/ml 3,3′-diaminobenzidine tetrahydrochloride and 0.03% hydrogen peroxide. Bovine serum albumin was used in place of primary antibody, as a negative control.
Quantification of DLL4 and VEGF
The DLL4 and VEGF staining in colorectal cancer and colorectal adenoma tissue was blindly scored and quantified by three independent observers (YD, QW and SL). Each observer analysed one tissue section per patient, comprising 10 fields of view per section. The mean result was quantified as a product of the staining intensity score: 0, negative; 1, weak; 2, moderate; 3, strong × percentage of cells stained: 0%, 0; 1–24%, 1; 25–49%, 2; 50–74%, 3; 75–100%, 4. Out of a final immunoreactive score of 0–12, a tissue section (patient) score ≥1 was defined as positive. Immunoreactive staining was quantified under high magnification (×200), using an Aperio Image Scope and Aperio ImageScope Viewer software, version MAN-0001 revision H (Aperio Technologies, Inc., Vista, CA, USA).
Microvessel detection and quantification
Microvessel density was evaluated as previously described. 14 Briefly, one section per patient was analysed, and intratumour microvessels were highlighted using an anti-CD34 monoclonal antibody (Ventana Medical Systems, Tucson, AZ, USA) as follows: slides were incubated with primary anti-CD34 antibody (1:100 dilution) overnight at 4℃, then washed 3 times with 0.1 mol/l phosphate buffered saline (PBS, pH 7.4). Slides were then incubated with biotinylated goat anti-mouse immunoglobulin G (1:200 dilution; Vector Laboratories, Burlingame, CA, USA) for 90 min at room temperature, followed by incubation with horse radish peroxidase-conjugated streptavidin (1:100 dilution; Chemicon, Temecula, CA, USA) for 20 min at room temperature, and a final wash 3 times with 0.1 mol/l PBS (pH 7.4). Antibody binding was visualized with 3,3-diaminobenzidine tetrahydrochloride, and sections were lightly counter stained with Mayer's hematoxylin.
Each tissue section was then examined under low power (×40 magnification) to identify the three most highly microvascularized CD34-stained regions (hot spots) of the section. 14 One field per hot spot was then analysed at × 400 magnification by three independent observers (YD, QW and SL), who counted the number of CD34-stained intratumoural microvessels. A single microvessel was defined as any CD34-stained endothelial cell separated from adjacent microvessels, tumour cells, and other connective tissue elements. Large vessels with thick muscular walls were excluded in the counts. The lumen was not required to identify a microvessel. The mean of three hotspots (one field per hotspot) was calculated for each independent observer, then the number of microvessels per section was calculated as the mean of the three independent observer results.
Quantification of DLL4 or VEGFA positive microvessels
Any CD34-stained microvessels that were also stained positive for DLL4 or VEGFA were considered to be DLL4-positive or VEGFA-positive microvessels.
One tissue section per patient was analysed, and 10 fields of view were randomly selected per section (analysed at × 400 magnification). The number of DLL4-positive or VEGFA-positive microvessels observed in each field was calculated and scored as follows: DLL4 or VEGFA-positive score, ≥one positive stained microvessel observed in one field; negative score, no positive stained microvessels observed in one field. Thus, the maximum score per tissue section equates to 10 positive microvessels in 10 fields; the minimum score equates to no positive microvessels in 10 fields.
Quantification of microvessel density in concentrated regions of DLL4-positive microvessels
One tissue section per patient was analysed at × 400 magnification, and one region with the highest concentration of DLL4-positive microvessels was marked on each section. The number of CD34-stained microvessels was then counted.
Statistical analyses
Data are presented as mean ± SD, and was analysed using SPSS software, version 18.0 (SPSS, Inc., Chicago, IL, USA) for Windows®. Between-group differences were analysed using Student’s t-test. Correlation analysis was performed using Pearson’s correlation coefficient. A P-value < 0.05 was considered to be statistically significant.
Results
Tissue from a total of 80 patients (35 with colorectal adenoma and 45 with colon cancer) were included in this study. In colorectal adenoma and colon cancer tissue, DLL4 and VEGFA immunoreactive brown staining appeared mainly in the cell cytoplasm, but also in small amounts in the cell membranes (Figure 1).
Representative photomicrographs showing expression of delta-like (DLL) 4 (Drosophila), vascular endothelial growth factor (VEGF) A and CD34 molecule in colon cancer and colon adenoma tissue. (A1–4) colon adenoma tissue, DLL4 positive, VEGFA positive, CD34 positive, and DLL4 negative, respectively; (B1–4) colon cancer tissue, DLL4 positive, VEGFA positive, CD34 positive, and DLL4 negative, respectively. Black arrows, positively stained cells; 200 × magnification; Scale bar = 25 µm.
Expression of DLL4 and VEGFA in relation to clinicopathological characteristics
Relationship between tissue delta-like (DLL) 4 (Drosophila) and vascular endothelial growth factor (VEGF) A expression, and demographic/clinicopathological characteristics of patients with colon cancer and colorectal adenomas.
Data presented as n (%) of patient tissue samples.
TNM, tumour node metastasis classification system (stages I–IV) developed by the Union for International Cancer Control.
NS, no statistically significant between-group difference (P ≥ 0.05; Students t-test).
Microvessel DLL4 and VEGFA expression in colon cancer and colorectal adenoma tissue
One tissue section per patient was analysed, resulting in 10 fields randomly selected for each of 35 colon cancer sections and 45 colorectal adenoma sections.
In colorectal adenoma tissue, the mean ± SD DLL4-positive microvessel score was 3.7 ± 1.1 out of 10 fields per section, compared with 8.5 ± 2.4 out of 10 fields per section in colon cancer tissue (t = 11.328, P < 0.001). A similar result was observed in terms of VEGFA expression (mean ± SD number of VEGFA positive fields per section in colorectal adenomas was 4.6 ± 1.4 out of 10 versus 9.2 ± 2.2 out of 10 in colon cancer tissue; t = 13.685, P < 0.001).
Differences in microvessel density in concentrated regions of DLL4-positive vessels between colon cancer and colorectal adenoma tissue
In each tumour tissue section, a mark was made in the region with the highest concentration of DLL4-positive vessels and the number of CD34-stained microvessels was counted (Figure 1A3 and B3). In colorectal adenoma tissue sections, 1947 microvessels were counted in 45 concentrated regions of DLL4-positive microvessels (31.4 ± 6.2 positive microvessels in each region). In colon cancer tissue sections, 5402 microvessels were counted in 35 concentrated regions of DLL4-positive microvessels (59.3 ± 8.7 positive microvessels in each region). Microvessel density in the concentrated regions of DLL4-positive microvessels was significantly higher in colon cancer tissue compared with colorectal adenoma tissue (t = 12.475, P < 0.001).
Correlation between DLL4 and VEGFA expression in colon cancer
Correlation between expression of delta-like (DLL)4 (Drosophila) and vascular endothelial growth factor (VEGF)A in colon cancer (n = 35) and colorectal adenoma (n = 45) tissue (Pearson’s correlation coefficient).
Data presented as n patient incidence.
Discussion
Excessive angiogenesis and VEGFA overexpression in colon cancer are strongly indicative of a poor prognosis. 15 To gain a deeper understanding of the mechanisms involved, the present study evaluated DLL4, a ligand involved in the Notch signalling pathway in colon cancer angiogenesis, as a potential prognostic indicator. DLL4 and VEGFA are mainly expressed in the cytoplasm and cell membrane of colon cancer and colorectal adenoma cells. 11 In the present study, expression of DLL4 and VEGFA was higher in colon cancer tissue than colorectal adenoma tissue, and expression of DLL4 was related to higher microvessel density in colon cancer tissue compared with colorectal adenoma.
Transcript levels of VEGF mRNA have been shown to be up-regulated in a colon cancer model. 16 In the present study, DLL4 expression significantly correlated with expression of VEGFA, and there was a marked increase in CD34 expression in DLL4-positive regions in cancer tissue compared with colorectal adenoma tissue. These results are consistent with the results of other studies into bladder cancer and renal clear cell carcinoma, in which DLL4 expression significantly and positively correlated with VEGF expression. 8 In vitro experiments have shown DLL4 expression to be regulated by VEGF in endothelial cells.17–19
Expression of DLL4 has been shown to significantly correlate with CD34 expression in both normal and tumour blood vessels. 16 The expression of CD34 in DLL4-negative blood vessels was obviously lower than that in DLL4-positive blood vessels, suggesting that DLL4 regulates the density of blood vessels through interaction with Notch receptors of pericytes. 20 In bladder cancer, expression of DLL4 was shown to be closely related to vascular maturation and microvessel density, however, its mechanism of action requires further verification. 20 DLL4-related signal transduction involves Notchl and Notch4 receptors, and has a key role in endothelial cells,21–23 however, the influence of DLL4 from endothelial cells on the tunica adventitia remains unreported. The Notch signalling pathway is known to play a role in the maturation of blood vessels. Blocking DLL4 has been reported to cause an increase in immature and non-functional blood vessels in tumour tissues,24,25 and DLL4-deficient mice were found to develop abnormal blood vessels that were generally not covered with smooth muscle or outer membrane. 26 Such abnormal blood vessel development was similar to that observed in platelet-derived growth factor B-deficient or platelet-derived growth factor B receptor-deficient mice.27,28
Tumours with a high proportion of mature vessels may have a poor response to some antiangiogenic therapies such as anti-VEGF therapy, as vascular adventitia may have a protective effect on the lack of growth factors. 29 DLL4 overexpression in human endothelial cells has been found to decrease the expression of VEGF receptor 2, the main endothelial cell receptor involved with VEGFA-induced angiogenesis, and to decrease expression of the VEGFA coreceptor, neuropilin-1. Reduction in these receptors coincided with reduction in VEGFA-induced cell proliferation, suggesting a role for DLL4 in resistance to antiangiogenic treatment involving anti-VEGF therapy. 13
Expression of DLL4 occurs in the intermediate stage of the final differentiation of endothelial cells, promoting cell maturation and functional differentiation. 9 These results suggest that DLL4 could be a potential target for cancer therapy. The value of DLL4 as a tumour prognostic marker could be assessed on a preliminary basis through correlating DLL4 expression with clinical follow-up data. High levels of DLL4 may indicate a risk of increased tumour recurrence or increased mortality rate. The sample sizes in the present study are relatively small, however, the study showed that there was a high level of DLL4 expression in colon cancer tissue compared with colorectal adenoma tissue. DLL4 was also found to be preferentially expressed in the vascular system of colon cancer and DLL4 expression was related to increased microvessel density and positively correlated with VEGFA expression. These results suggest that DLL4 may have value as a prognostic marker in colon cancer. Appropriate levels of DLL4 are necessary for endothelial cell function.19,30 Further investigations into targeted therapy involving DLL4 blockade or overexpression against colon cancer are required, particularly against tumours which are insensitive to anti-VEGF therapy.
Footnotes
Declaration of conflicting interest
The authors declare that there is no conflict of interest.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.