Expression of MMP-2, TIMP-2 protein and the ratio of MMP-2/TIMP-2 in gallbladder carcinoma and their significance

Abstract

AIM: To study the correlation between expression of MMP-2, TIMP-2 protein and the ratio of MMP-2/TIMP-2 and clinical-pathological parameters of patients with gallbladder carcinoma.

METHODS: Carcinomas (n = 45) and polypoid lesions (n = 15) of the gallbladder were studied for the expression of MMP-2 and TIMP-2 protein by immunohistochemical avidin-biotin-complex method and image analysis. Clinical-pathological data of patients with gallbladder carcinoma such as histological type, grade of differentiation, level of infiltration, liver invasion and lymph node involvement, etc, were recorded.

RESULTS: There was significant difference between the average level (1.123 ± 0.108 vs 1.030 ± 0.054, P = 0.002) of MMP-2, the ratio (1.050 ± 0.013 vs 0.937 ± 0.078, P = 0.003) of MMP-2/TIMP-2 in gallbladder carcinomas and in polypoid lesions of the gallbladder. Significant difference was found between the expression of MMP-2 in early stage and advanced tumors, but there was no correlation between MMP - 2 protein expression and histological type, differentiation degree, infiltration level, lymph node involvement or liver invasion. Although no difference was observed between TIMP-2 expression and histological type or differentiation degree, significant difference was found between TIMP-2 expression and different Nevin stage, infiltration level, local lymph node involvement or liver invasion (1.168 ± 0.067 vs 1.048 ± 0.075, 1.170 ± 0.062 vs 1.039 ± 0.069, 1.039 ± 0.076 vs 1.147 ± 0.083, 1.048 ± 0.074 vs 1.103 ± 0.095, P < 0.05). MMP-2/TIMP-2 ratio did not correlate with histological type, grade of differentiation and liver invasion, but significant differences were found between MMP-2/TIMP-2 ratio and different Nevin stage, infiltration level and lymph node involvement in patients with carcinoma of gallbladder.

CONCLUSION: TIMP-2 and MMP-2/TIMP-2 ratio could reflect more accurately biological characteristic of gallbladder carcinoma and MMP-2/TIMP-2 ratio might be a new significant marker in early diagnosis, in the judgment of invasion or metastasis and the estimate of prognosis in patients with gallbladder carcinomas.

Key Words: $[Keywords]

INTRODUCTION

Primary carcinoma of gallbladder represents a very lethal malignant tumor because of its early metastasis, strong invasion and poor prognosis[1-3]. It is very important to estimate the malignant degree and invasion tendency in order to guide clinical diagnosis and treatment of gallbladder carcinoma. Breakage or degradation of ECM (extracellular matrix) and BM (basement membrane) is necessary in the process of tumor invasion[4,5]. Matrix metalloproteinases (MMPs) and their tissue inhibitors of metalloproteinases (TIMPs), specially, MMP-2 and its tissue inhibitor (TIMP-2), take important roles in degradation of ECM and BM[6,7] and relating to tumor invasion[8-11]. So far, there are no reports on the expression of MMP-2, TIMP-2 and the ratio of MMP-2/TIMP-2 in gallbladder carcinoma. An effort is presently made to examine expression of MMP-2 and TIMP-2 protein and ratio of MMP-2/TIMP-2 in carcinomas and polypoid lesions of the gallbladder, to study correlation between expression of MMP-2, TIMP-2 and ratio of MMP-2/TIMP-2 and clinical-pathological parameters of patients with gallbladder carcinoma and to evaluate their clinical significance.

MATERIALS AND METHODS

Materials

45 carcinomas and 15 polypoid lesion of the gallbladder underwent operational resection and confirmed histopathologically at Tongji hospital of Tongji University from 1995 to 2000 were studied. All samples were fixed in 10% formalin and embedded in paraffin. In 45 patients with carcinoma of gallbladder, there were 6 males and 39 females with a mean age of 61.9 years (range 36-80 years). Of these, there were histologically 30 adenocarcinomas, 6 papillary carcinomas and 9 others ( squamous, mucinous, undifferentiated and clear cell carcinoma); there were 24 G1 (well differentiation), 11 G2 (moderate differentiation) and 10 G3 (poor differentiation), according to the criteria established by World Health Organization for histological type of tumors of the gallbladder and extrahepatic bile ducts. Clinical-pathological dates of each patient with carcinoma of gallbladder, such as histological type, grade of differentiation, level of infiltration, liver invasion and lymph node involvement, etc, were recorded. 15 polypoid lesions of the gallbladder were used as controls. Of them, there were 7 males, 8 females with a mean age of 45.5 years (range 27-74 years).

Methods

Immunohistochemistry Immunohistochemical staining was performed on sections from formalin-fixed paraffin-embedded blocks by the avidin-biotin-complex method (SABC kit, BOSTER). Monoclonal MMP-2 antibody (Neomarker’s) was used at a concentration of 1:100 and TIMP-2 antibody (Antibody Diagnostic) at 1:20. Goat serum, biotinylated secondary antibody (goat anti-mouse IgG) and DAB are all purchased from BOSTER. For negative control, the slides were treated with PBS in place of primary antibody.

Quantified analysis of stained intensity[12] Stained intensity was quantified with the software Analysis System of Clinicopathological Image (version 2, for windows 95/OSR). The methods were as follows: (1) estimating the percentage of positive tumor cells, no positive cells, regarded as 0%. (2) examining average gray value of positive cells of individual slide with image analysis system to be used as antigen concentration. (3) calculating antigen content: Content = gray value × percentage of positive cells. Control samples were examined with the same methods.

Data alternating and statistical analysis Because some samples were negative the examined data needed to be alternated. Regarding examined data as X, alternated data as Y. Alternating method was as following:

Y = e^x e = 2.71828

In this way MMP-2/TIMP-2 ratio was calculated.

All the statistical analyses were performed using SPSS 10.0 for windows. P < 0.05 or F < 0.05 was considered to be of statistical significance.

RESULTS

Expression of MMP-2 and TIMP-2 protein and the ratio of MMP-2/TIMP-2 in carcinomas and polypoid lesions of the gallbladder

Expression of MMP-2 and TIMP-2 protein was observed in tumor cells and epithelial cells of benign lesions, tumor stromal tissues, muscularis of gallbladder and vas wall. Some endangium of vas wall were stained. The protein of MMP-2 and TIMP-2 was mainly expressed in cytoplasm of positive cells (Figures 1, 2, 3 and 4). Although the expressing pattern of MMP-2 and TIMP-2 protein in carcinomas of gallbladder was identical to that in polypoid lesions of gallbladder, expressed value of MMP-2 in carcinoma was significantly higher than that in polypoid lesions while there was no difference in TIMP-2 protein expression between the two groups (Table 1). In addition, the ratio of MMP-2/TIMP-2 in carcinoma of gallbladder was significantly higher than that in polypoid lesions of gallbladder (P = 0.003, Table 1).

Table 1 Expression of MMP-2, TIMP-2 and ratio of MMP-2/TIMP-2 in carcinoma and polypoid lesions of gallbladder (^-x±s).

	n	MMP-2	TIMP-2	MMP-2/TIMP-2
PLG	15	1.030 ± 0.054	1.104 ± 0.072	0.937 ± 0.078
GBC	45	1.123 ± 0.108a	1.077 ± 0.090	1.050 ± 0.013b

PLG: polypoid lesions of the gallbladder; GBC: gallbladder carcinoma.

^aP = 0.002,

^bP = 0.003.

Figure 1 Expression of MMP-2 proteins in polypoid lesion cells of the gallbladder ( × 200).

Figure 2 Expression of MMP-2 proteins in adenocarcinoma cells of the gallbladder ( × 200).

Figure 3 Expression of TIMP-2 protein in polypoid lesion cells of the gallbladder ( × 200).

Figure 4 Expression of TIMP-2 protein in adenocarcinoma cells of the gallbladder ( × 200) .

Correlation between expression of MMP-2, TIMP-2 protein and ratio of MMP-2/TIMP-2 and clinical-pathological parameters in patients with carcinoma of gallbladder

Correlation between expression of MMP-2, TIMP-2 protein and ratio of MMP-2/TIMP-2 and clinical-pathological parameters in patients with carcinoma of gallbladder were shown in Table 2 and Figures 3, 4, 5, 6, 7, 8. There were no correlation between expression of MMP-2 or TIMP-2 protein and ratio of MMP-2/TIMP-2 and the patients’ histological type or degree of differentiation (F test, P > 0.05). Based on Nevin stage criteria, our files included early stage (S1, S2. n = 11) and advanced stage tumors (S3-S5, n = 34). Expression of MMP-2 protein was only correlated with Nevin stage (P < 0.05); while expression level of TIMP-2 protein was positively correlated with Nevin stage and infiltration level (P < 0.05), but was reversely correlated with lymph node metastasis and liver invasion (P < 0.05). There were significant differences between ratio of MMP-2/TIMP-2 and different Nevin stages, infiltration levels, lymph node metastasis and liver invasion in patients with gallbladder carcinoma.

Table 2 Expression and correlation of MMP-2 and TIMP-2 proteins and ratio of MMP-2/TIMP-2 and clinical-patho-logical parameters of patients with carcinoma of the gall-bladder (^-x±s).

	n	MMP-2	TIMP-2	MMP-2/TIMP-2
histological type
adenocarcinoma	30	1.140 ± 0.113	1.077 ± 0.089	1.064 ± 0.143
papillary carcinoma	6	1.101 ± 0.106	1.117 ± 0.094	0.990 ± 0.106
others	9	1.082 ± 0.086	1.042 ± 0.084	1.043 ± 0.101
differentiated degree
well	24	1.117 ± 0.099	1.089 ± 0.095	1.031 ± 0.114
moderate	11	1.041 ± 0.131	1.062 ± 0.083	1.081 ± 0.159
poor	10	1.120 ± 0.110	1.064 ± 0.086	1.059 ± 0.144
Nevin stage
S1,S2	11	1.063 ± 0.077	1.168 ± 0.067	0.914 ± 0.086
S3-S5	34	1.143 ± 0.110a	1.048 ± 0.075b	1.094 ± 0.112c
Infiltration level
muscular	13	1.088 ± 0.099	1.170 ± 0.062	0.932 ± 0.093
serosal	32	1.137 ± 0.109	1.039 ± 0.069b	1.097 ± 0.114c
Lymph node
LN (+)	29	1.131 ± 0.109	1.039 ± 0.076	1.092 ± 0.12
LN (-)	16	1.120 ± 1.107	1.147 ± 0.083b	0.973 ± 0.114c
Liver invasion
(+)	21	1.136 ± 0.107	1.048 ± 0.074	1.087 ± 0.107
(-)	24	1.112 ± 0.110	1.103 ± 0.095b	1.017 ± 0.143

^aP < 0.05,

^bP < 0.05, ^CP < 0.05.

Figure 5 Expression of MMP-2 protein in adenocarcinomas of the gallbladder with LN (-) (Expression level 1. 101, × 200).

Figure 6 Expression of TIMP-2 protein in adenocarcinomas of the gallbladder with LN (-) (Expression level 1. 268, × 200).

Figure 7 Expression of MMP-2 protein in adenocarcinomas of the gallbladder with LN (+) (Expression level 1. 270, × 200).

Figure 8 Expression of TIMP-2 protein in adenocarcinomas of the gallbladder with LN (+) (Expression level 1. 000, × 200).

DISCUSSION

One of the typical characteristics of malignant tumor is invasion and metastasis. Its ability of invasion and metastasis is mainly responsible for their lethality. So it is necessary to understand the molecular and cellular mechanism of tumor dissemination so as to develop novel therapies basing on this knowledge. Tumor invasion is considered to be a dynamic, complex and multi-step process[13], but the essential step is the degradation of extracellular matrix (ECM) and basement membrane (BM)[13,14]. It was reported that MMPs (matrix metalloproteinases) are important for the degradation of ECM. MMPs hydrolyze specifically type IV,V,VII,Xcollagens and fibronectin, elastin et al which are all important components of ECM and BM, and are closely associated with the invasiveness and metastasis of tumor[15-20]. TIMPs (tissue inhibitors of metalloproteinases), as the specific inhibitors of MMPs, have such ability to form tight binding, non-covalent inhibitory complexes with multiple members of the MMP family that they inhibit MMP activity of ECM degradation and have anti-metastasis function[21,22].

There have been many reports about correlation between expression of MMP-2 and TIMP-2 and tumor development[22-28]. However, no reports concerning correlation between expression of MMP-2 and TIMP-2 and clinicopathological parameters in gallbladder carcinoma have been found. In an immunohistochemical study of 27-pancreatic cancers conducted by Branhall[29], a strong correlation was found between over-immunoreactivity of MMP-2 and the aggressive phenotype of pancreatic carcinoma. Another study of 177 breast cancer consisting mainly of invasive ductal carcinoma showed that the activation rate of pro-MMP-2 is significantly higher in node-positive carcinomas than in nod-negative cancers or benign neoplasms. Patients with positive staining for MMP-2 were significantly associated with shortened survival and a 3.6 fold increase in the risk of death. MMP-2 was an independent prognostic indicator in multivariate analysis[30]. Transfection of TIMP-2 gene into HT-1080 cell line and C-H-ras transformed rat embryo fibroblast inhibited their invasion potential in a murine tumor model. Ara et al[4] have observed the expression of high level of TIMP-2 in early stage and the inverse correlation with the corresponding MMP-2 in cases of neuroblastoma, which may represent a mechanism by which tumor cells and stromal tissues control the proteolysis and remodeling of ECM that occurs during invasion and advancement of stages. Also, the correlation of TIMP-2 over-expression with better chance of survival seemed to be associated with the ability of TIMP-2 to inhibit MMP-2 activities and tumor invasion. But recently some studies have suggested that the expression level of TIMP-2 relates to poor prognosis. Ree et al[31] showed that TIMP-2 mRNA level correlated with the development of distant metastases. It was reported in another study that TIMP-1 and TIMP-2 mRNA levels were positively correlated with lymph node metastasis, reduced 5-year survival and Duke’s classification in primary colorectal carcinomas. A study on stomach cancer showed that TIMP-1 and TIMP-2 were identified in 41% and 57% of tumors, respectively, whereas normal gastric mucosa was negative. No correlation was observed between the presence of TIMP-2 and tumor stage, histological type, lymph node status or survival[32].

Primary carcinoma of the gallbladder represents a lethal malignant tumor because of its early metastasis, poor prognosis and great difficulties in management[1-3]. Focusing on the key step of ECM degradation in metastasis, studying on the expression of MMP-2 and TIMP-2 and evaluating the correlation between these expression levels and clinicopathological parameters would be important for early diagnosis, judgment of invasion or metastasis and prognosis in patients with carcinoma of gallbladder. The present study examined the expression of MMP-2 and TIMP-2 in primary carcinoma of gallbladder by immunohistochemical and image analysis methods in order to evaluate accurately the invasive potential of tumor cells. Results showed that several types of tissues expressed MMP-2 and TIMP-2 protein, such as tumor cells, muscularis of gallbladder and vas wall, stromal cells and epithelial cells of benign lesions, indicating that MMP-2 and TIMP-2 are from epithelial cells (normal or transformed) and other stromal cells. The staining pattern of tumor cells and epithelial cells of benign lesions was cytoplasmic type, identical to previous immunohistochemical studies[33]. Although the expression of MMP-2 and TIMP-2 could be observed in carcinomas and polypoid lesions of the gallbladder, the expression level of MMP-2 and the ratio of MMP-2/TIMP-2 in tumor was significantly lower than those in benign lesions. But there was no difference in TIMP-2 levels between the two groups, indicating that the relatively high expression of MMP-2 was the basis of tumor invasion and metastasis.

With regard to correlations between the expression of MMP-2 or TIMP-2 and clinicopathological parameters of patients with carcinoma of gallbladder, the present study showed that the expression of MMP-2 related to Nevin stage. Expression value in advanced stage was obviously higher than in early stage (1.143 ± 0.110 vs 1.063 ± 0.077, P < 0.05), but had no correlation with lymph node status and infiltration level. These seemed in contradiction with above reports. The reasons were, on one hand, difference of study method and antibody[14], on the other hand, double effects of TIMP-2. TIMP-2 not only inhibits MMP-2 activities but also takes part in activation of MMP-2 on the cell surface. If the relationship between MMP-2 and TIMP-2 was not considered, invasive characteristic of cancer cells would not be reflected completely by only examining the expression of MMP-2. In addition, there was no significant difference in the expression value of TIMP-2 between groups of different histological type and grade of differentiation. Expression of TIMP-2 in early stage was obviously higher than in advanced stage, and correlated with infiltration level, local lymph node metastasis and liver invasion. All these suggested that there was a correlation between TIMP-2 expression and the clinicopathological parameters standing for the patients’ prognosis. TIMP-2 may be one of indicators in the judgment of Nevin stage, invasion and lymph node metastasis in patients with carcinoma of gallbladder.

MMPs play a major role in ECM degradation while TIMPs are the specific regulators of MMPs. The balance of MMPs/ TIMPs is the decisive factor for the maintenance of ECM steadiness and integrity, and the roles of MMPs in tumor metastasis do not depend on the absolute concentration of MMPs in local area, but the MMPs/TIMPs ratio. The expression levels of MMPs and TIMPs would alter in varying degrees in many biological or pathological processes, such as blastocyst implantation, embryonic development, never growth, tissue remodeling, wound healing and the breakdown of blood brain barrier, ulceration, liver fibrosis et al[34-41], but the MMPs/TIMPs balance is controlled accurately. During tumor metastasis the balance of MMPs/TIMPs is broken, favoring ECM degradation without regulation. A study demonstrated recently that the increased adhesion of A2058 cells was associated with increased secretion of MMP-2, which was then inhibited by the increase of TIMP-2. The author believed the imbalance of MMP-2 and TIMP-2 was critical for tumor cells to have strong invasive potential. Thus examining the absolute concentration of MMPs in local area had no significance and could not reflect the invasive potential of tumor cells. MMPs/TIMPs might act as a prognostic factor indicating malignant invasion and metastasis.

Based on this point, the author evaluated for the first time the correlation between the ratio of MMP-2/TIMP-2 and the clinicopathological parameters of gallbladder carcinoma when examining the expression of MMP-2 and TIMP-2 protein. The results showed that the ratio of MMP-2/TIMP-2 was obviously lower in ploypoid lesions of the gallbladder than that in carcinoma of gallbladder (P = 0.003). The ratio of MMP-2/TIMP-2 related to Nevin stage and was significantly higher in advanced cases than in early cases. Significant difference of the ratio between lymph involvement (+) and (-) was also observed (1.092 ± 0.122 vs 0.973 ± 0.114, P < 0.05). Moreover, the ratio of MMP-2/TIMP-2 also correlated with infiltration level. Gohji and coworkers[42] have examined MMP-2/TIMP-2 of 53 advanced urothelial cancers, and reported that MMP-2/TIMP-2 ratio was obviously higher in 31 patients with recurrence than that in patients without recurrence. COX multiple regression analysis showed MMP-2/TIMP-2 as a new marker predicting recurrence in urothelial tumors. They[43] also found the higher the serum MMP-2/TIMP-2 ratio of advanced bladder cancer, the earlier recurrence and the poorer prognosis. Serum MMP-2/TIMP-2 ratio could exert a predictive role, identical to our conclusion. We propose that, compared with the expression of MMP-2 or TIMP-2, the ratio of MMP-2/ TIMP-2 could reflect more accurately the biological characteristics of invasion and metastasis in patients with carcinoma of the gallbladder. As a new prognostic marker, the ratio of MMP-2/TIMP-2 might be useful in early diagnosis and in estimating the invasion and metastasis or survival crisis in patients with carcinoma of the gallbladder.