MMP-11 and VEGF-C expression correlate with clinical features of colorectal adenocarcinoma

Abstract

Colorectal cancer is one of the most common types of cancer and has a high mortality rate, making accurate prediction of clinical prognosis an important concern. This study sought to investigate the expression of two proteins, MMP-11 and VEGF-C, in colorectal adenocarcinoma tissues to provide options for better tumor detection. The correlation of the expression of these proteins with the clinical and pathological features of colorectal adenocarcinoma and with the invasion and metastasis of colorectal adenocarcinoma were assessed. Immunohistochemistry with light microscopy was used to detect the expression level of MMP-11 and VEGF-C in 78 samples of colorectal adenocarcinoma tissues and in 20 samples of normal colorectal tissues. Expression of MMP-11 in colorectal adenocarcinoma tissues and normal colorectal tissues was detected in 55.1% and 30.0% of samples, respectively; VEGF-C was detected in 78.2% and 15.0% of samples, respectively. The expression of MMP-11 and VEGF-C in colorectal adenocarcinoma tissues were not related to patient age, sex, or tumor location. However, expression was related to Dukes’ staging and lymph node metastasis. Expression of MMP-11 was correlated with distant metastasis from tumors. Lastly, expression of MMP-11 in colorectal adenocarcinoma was positively correlated with VEGF-C. In brief, expression of MMP-11 and VEGF-C in colorectal adenocarcinoma tissues serves as a significant index for evaluating the degree of malignancy, clinical stages, lymph nodes, and distant metastasis of colorectal adenocarcinoma. The detection of both of these proteins may help predict the infiltration, metastasis, and prognosis of colorectal adenocarcinoma, and potentially offers new targets for clinical treatment.

Introduction

Colorectal cancer is one of the most common malignant tumors of the digestive tract, and its poor prognosis results from infiltration, metastasis, and postoperative recurrence. The extracellular matrix (ECM) acts as a barrier to tumor cell infiltration and metastasis [1]. ECM is degraded by matrix metalloproteinases (MMPs) like MMP-11 [2], which is synthesized and secreted by tumor and connective tissues. MMPs are mainly composed of five classes: stromelysins, membrane type metalloproteinases, gelatinases, collagenases, and other secreted type MMPs [2]. MMP-11 belongs to the stromelysins, which were first discovered in the interstitial cells of breast cancer [3,4]. MMP-11 participates in various pathophysiological processes involving matrix degradation and tumor metastasis, and is the most important proteolytic enzyme for ECM degradation [5]. It can degrade various fibrous connective tissues, destroying the tissue barrier and facilitating the invasion and infiltration of tumors; its expression is up-regulated in various tumors [6-8]. It has been proposed that inhibiting MMP-11 expression could inhibit tumor cell proliferation [9].

Human vascular endothelial growth factor-C (VEGF-C), also called a lymphangiogenic growth factor, belongs to the vascular endothelial growth factor family along with VEGF-A, VEGF-B, and VEGF-D. VEGF proteins control vascularization, vasculogenesis, vascular permeability changes, lymphangiogenesis lymphatic vessel density, and endothelial cell survival of tumors [10]. VEGF-C plays an important role in regulating lymphangiogenesis and is also closely related to tumor invasion and metastasis [11]. It also functions in determining the lymphatic vessel density of tumors [12]. Lymph node metastasis is one of the major routes for the invasion and metastasis of tumors and is closely dependent on the activity of VEGF-C and relevant factors. It has been proposed that the high expression of VEGF-C promoted lymphatic metastasis from tumors by promoting neo-lymphangiogenesis and activating existing lymphatic vessels [13,14].

The known and potential roles of MMP-11 and VEGF-C make them attractive candidates as predictors of cancer progression. In this study, immunohistochemistry was used to detect the expression of MMP-11 and VEGF-C in 78 cases of colorectal adenocarcinoma and in 20 normal colorectal mucosal tissues. We also investigated the relationship of MMP-11 and VEGF-C expression with various clinical and pathological factors, as well as the correlation between the expression of MMP-11 and the expression of VEGF-C. The results offer potential new prognostic and diagnostic factors.

Materials and methods

Human tissues

Tissue samples were obtained from 78 confirmed colorectal adenocarcinomas resected between January and December 2013 and assessed in The Department of Pathology, The First Affiliated Hospital of Zhengzhou University (Zhengzhou, China). None of the patients underwent chemotherapy or radiotherapy. The samples were obtained from 52 males and 26 females (mean age, 47.1 ± 3.3 years; range, 36 to 73 years). Of the 78 samples, 33 were obtained from recta and 45 were from colons. Samples were assessed histologically; 9 samples were of Grade I, 46 were Grade II, and 23 were Grade III. Samples were also assessed for Dukes’ staging: 3 samples were in Stage A, 39 were in Stage B, 27 were in Stage C, and 9 were in Stage D. Of the 78 samples, 28 underwent lymph node metastasis and 9 underwent distant metastasis. For a control group, 20 samples of normal colorectal mucosal tissues were obtained during the same time period.

Reagents

Rabbit anti-human MMP-11 monoclonal antibody and rabbit anti-human VEGF-C polyclonal antibody were purchased from Beijing Zhongshan Golden Bridge Biotechnology Co., Ltd (Beijing, China). An immunohistochemistry kit and a DAB chromogenic reagent were purchased from Fuzhou Maixin Biotechnology Development Co., Ltd (Fuzhou, China).

Immunohistochemistry

All tissue samples were fixed with 10% neutral formalin, embedded in paraffin, and cut into 4-μm serial sections using a microtome. Sections were collected onto glass slides, dewaxed with xylene, dehydrated in a graded alcohol series, and subjected to heat-mediated antigen retrieval. The samples were then incubated overnight in 0.3% H₂O₂ at room temperature with primary antibodies (MMP-11 or VEGF-C), washed with PBS, incubated with biotin-labeled secondary antibody, and washed with PBS again. The samples were then treated with horseradish peroxidase-labeled working solution of streptavidin and incubated at 37°C for 45 minutes. Next, the samples were developed with DAB color-producing reaction, where the chromogenic time was controlled through microscopic observation. Finally, the samples were stained with haematine, cleared with xylene, and sealed with neutral resins. The sections of known positive colorectal adenocarcinoma were used as positive controls and samples with PBS solution substituted for the primary antibody were used as negative controls.

Quantification of immunohistochemical staining results

Two experienced pathologists performed blind assessment of each section by examining 10 high-power fields of vision that were randomly selected under a microscope (Olympus, BX51-P). Samples were scored for the proportion of positive cells stained: a score of 0 for ≤ 1% positive, 1 for 2%-20%, 2 for 21%-50%, and 3 for ≥ 51%. Samples were also scored for staining intensity: a 0 score for no staining, a 1 for lightly staining, and a 2 for deeply staining. A semi-quantitative analysis was performed by combining the scores for staining and intensity by multiplying the two scores together. With this combined score, ≤ 1.9 indicated negative result (-), 2-2.9 indicated weakly positive result (+), 3-3.9 indicated positive result (++), and ≥ 4 indicated strongly positive result (+++).

Statistical analysis

Statistical analysis was performed using SPSS 18.0 statistical software (IBM, New York City, USA). A χ² test was performed to compare the expression of MMP-11 and VEGF-C between groups, and a Spearman rank correlation was used to analyze the correlation between MMP-11 and VEGF-C expression.

Results

Expression of MMP-11 and VEGF-C in colorectal adenocarcinoma tissues and normal colorectal mucosal tissues

MMP-11 and VEGF-C proteins were mainly localized within the cytoplasm and were observable as brown or tan granules. MMP-11 was expressed in 55.1% of colorectal adenocarcinoma tissues but in 30.0% of normal colorectal tissues (Table 1); this difference in the percentage of samples expressing MMP-11 was significantly different (p < 0.05). VEGF-C was detected in 78.2% of colorectal adenocarcinoma tissues and 15.0% of controls (Table 1); this difference was also significantly different.

Table 1.

The positive expression rates of MMP-11 and VEGF-C in colorectal adenocarcinoma and normal colorectal tissues

Tissue type	n	MMP-11a	VEGF-Cb
		Positive expression (%)	Positive expression (%)
Colorectal adenocarcinoma	78	43 (55.1)	61 (78.2)
control	20	6 (30.0)	3 (15.0)

^a p = 0.045, χ² = 4.021;

^b p = 0.000, χ² = 29.639.

MMP-11 and VEGF-C expression correlate with clinical and pathological factors of colorectal adenocarcinoma

The expression of MMP-11 was correlated with several known clinical and pathological factors of colorectal adenocarcinoma. The expression of MMP-11 in colorectal adenocarcinoma tissues was related to Dukes’ staging, lymph node metastasis, and distant metastasis (p < 0.05; Table 2). Alternatively, VEGF-C expression was related to histological grading, Dukes’ staging, and lymph node metastasis (p < 0.05; Table 3).

Table 2.

The correlation between MMP-11 expression and clinicopathological factors in colorectal adenocarcinoma

clinicopathological factors	n	MMP-11
		Positive samples	Negative samples	Positive expression (%)	χ2 value	P value
ages
≤ 50	47	29	18	61.7	2.066	0.151	P > 0.05
> 50	31	14	17	45.2
gender
male	52	31	21	59. 6	1.270	0.260	P > 0.05
female	26	12	14	46. 2
tumor location
rectal carcinoma	33	17	16	51.5	0.302	0.583	P > 0.05
colon carcinoma	45	26	19	57.8
histological grading
Grade I	9	3	6	33.3	3.828	0.147	P > 0.05
Grade II	46	24	22	52.2
Grade III	23	16	7	69. 6
Dukes’ staging
A-B stage	42	18	24	42.9	5.539	0.019	P < 0.05
C-D stage	36	25	11	69.4
lymphatic metastasis
exist	28	19	9	67.9	4.692	0.030	P < 0.05
none	50	24	26	48.0
distant metastasis
exist	9	8	1	88.9	4.688	0.030	P < 0.05
none	69	35	34	50.7

Table 3.

The correlation between VEGF-C expression and clinicopathological factors in colorectal adenocarcinoma

Clinicopathological factors	n	VEGF-C
		Positive cases	Negative cases	Positive expression (%)	χ2 value	p value
ages
≤ 50	47	39	8	83.0	1.518	0.209	P > 0.05
> 50	31	22	9	71.0
gender
male	52	44	8	84.6	3.761	0.052	P > 0.05
female	26	17	9	65.4
tumor location
rectal carcinoma	33	24	9	72.7	1.007	0.316	P > 0.05
colon carcinoma	45	37	8	82.2
histological grading
Grade I	9	4	5	44.4	7.186	0.028	P < 0.05
Grade II	46	37	9	73.9
Grade III	23	20	3	87.0
Dukes’ staging
A-B stage	42	28	14	66.7	7.108	0.008	P < 0.05
C-D stage	36	33	3	91.7
lymphatic metastasis
exist	28	27	1	96.4	8.510	0.004	P < 0.05
none	50	34	16	68.0
distant metastasis
exist	9	9	0	100.0	2.835	0.092	P > 0.05
none	69	52	17	75.4

Correlation between the expression of MMP-11 and VEGF-C in colorectal adenocarcinoma

The expression of MMP-11 was positively correlated with the expression of VEGF-C (r = 0.273, p = 0.016; Table 4).

Table 4.

Correlation of expression of MMP-11 and VEGF-C in colorectaladenocarcinoma tissues

MMP-11	VEGF-C		p value	r value
	Positive	Negative
positive	38	5	0.016	0.273
negative	23	12

Discussion

Colorectal cancer ranks third in the incidence of cancer and ranks fourth in mortality [15], so improving diagnostic and prognostic factors for this type of cancer is important. The results of this study indicate that both MMP-11 and VEGF-C may be useful as diagnostic or prognostic markers in colorectal cancer.

Both MMP-11 and VEGF-C expression were more common in colorectal adenocarcinoma tissues than in normal tissues. Additionally, the expression of MMP-11 was closely related to the Dukes’ staging, lymph node metastasis, and distant metastasis without a marked correlation with patient age or sex or tumor location. These results indicate that MMP-11 might participate in the process of invasion and infiltration of tumors, and play an important role in lymph node metastasis and distant metastasis from the tumors [16]. VEGF-C expression was related to histological grading, Dukes’ staging, and lymph node metastasis. These results suggest that the increased degree of malignancy of tumors could possibly promote the expression of VEGF-C, thus promoting tumor lymphangiogenesis [17]. The proliferated and expanded lymphatic vessels would then pave the way for the lymph node metastasis from the tumors and allow the tumor cells to pass the neo-lymphatic vessels and enter the lymph nodes. This would allow further expansion toward the whole body, and thereby could promote the further growth of the tumors.

In summary, the expression of MMP-11 and VEGF-C in colorectal adenocarcinoma is associated with infiltration, invasion, and metastasis of these tumors. Additionally, the MMP-11 and VEGF-C expression are positively correlated, indicated that a related pathway may be present between these two proteins. In further studies, MMP-11 and VEGF-C could serve as reference indexes for evaluating the biological behaviors and prognosis of colorectal adenocarcinoma and possibly become a new therapeutic target.

Disclosure of conflict of interest

None.