Introduction
The metastasis of breast cancer involves several stages and lymph node metastasis is the main early metastatic route. The first stage in this progression involves that the tumor cells lose their local constraints, which emanated from neighboring normal cells and surrounding stroma. The second step is termed intravasation, which involves entering the circulatory system (blood and/or lymphatic). In the third step, the disseminated tumor cells survive within hostile ectopic sites. Finally, in the fourth step, organotropic colonization of tumor cells proliferates in the compatible microenvironment. A series reports have invoked epithelial-mesenchymal transition (EMT) as the mechanistic basis for breast cancer invasion and metastasis [
1,
2]. EMT involves the loss of “epithelial” characteristics and switching on of mesenchymal characteristics such as the change of cell-cell adhesion. The E-cadherin (E-Cad) to N-cadherin (N-Cad) switch is often a hallmark of tumor cells that have undergone EMT [
3]. EMT is supposed to be an initial change of breast cancer metastasis and is related to several factors that may promote the capability of degrading adjacent extracellular matrix (ECM) and increase the mobility and angiogenesis of tumor cells. However, there was still a controversy whether EMT was the main mechanism for breast cancer invasion and metastasis or not [
4]. Vascular endothelial cell growth factor (VEGF) is an apparently endothelial cell-specific mitogen that promotes the lymphangiogenesis and is associated with a primary tumor’s risk of metastasis to lymph nodes. Matrix metalloproteinase-9 (MMP-9) is an enzyme capable of degrading the ECM, altering cell-cell and cell-ECM interactions and is involved in the development of tumor invasion and metastasis. Cyclooxygenase-2 (COX-2) is the rate-limiting enzyme that catalyzes the formation of prostaglandins. Moreover, it is highly expressed in many aggressive metastatic cancers and may play a critical role in cancer progression, being involved in tumor proliferation, lymphangiogenesis, and metastasis. In the current study, immunohistochemical expressions of E-Cad, N-Cad, VEGF, MMP-9 and COX-2, and microvascular density (MVD) were investigated to elucidate the relationship between EMT and microenvironment variation involved many factors related with the lymph node metastasis in breast cancer.
Materials and methods
Patients and specimens
The pathologic samples tested were collected from The Second Affiliated Hospital of Dalian Medical University, including 74 cases (female, 39 cases with lymph node metastasis, 35 cases without lymph node metastasis) with ages ranging from 24 to 70 (median age: 53) years. All tissues were surgically resected. Cancerous tissues and metastatic lymph nodes were all fixed in 10% phosphate-buffered formalin immediately after resection, embedded in paraffin, and cut into 4-μm-thick sections for immunohistochemical and routine histological examinations. The histological types of patients were all infiltrating carcinoma, according to the World Health Organization (WHO) Criteria (2003), including 60 cases of infiltrating ductal carcinoma, 5 cases of infiltrating lobular carcinoma, and 9 of other types (3 of medullary carcinoma, 3 of mucinous carcinoma, 2 of clear cell carcinoma, and 1 of adenoid cystic carcinoma). The Union Internationale Centre le Cancer (UICC) breast cancer tumor-node-metastasis (TNM) system for pathological staging and the Bloom-Richardson system for tissue grading were employed. The follow-up rate was 64.9% (48/74), and the mean follow-up time was 72 (ranging from 72 to 121) months.
Immunohistochemistry
The sections were dewaxed and rehydrated by sequential immersion in xylene and graded ethanol, and water. Endogenous peroxidase activity was blocked by incubation with 3% hydrogen peroxide methanol for 20 min. Then, the sections were rinsed and subjected to water-bath heating for antigen retrieval. Nonspecific antibody binding sites were blocked by the incubation of the tissue section in normal goat serum for 30 min at room temperature. Anti-N-Cad and anti-VEGF (IBL Co., Japan), anti-COX-2 (Maixin Biotechnology Co., Fuzhou, China), anti-E-Cad, anti-MMP-9, and anti-CD34 (Zhongshan Golden Bridge Biotechnology Co., Beijing, China) were added and incubated overnight at 4°C. After being washed with phosphate buffered saline (PBS), a biotin-labeled secondary antibody (Zhongshan Golden Bridge Biotechnology Co., Beijing, China) was added and incubated for 30 min. The sections were then incubated with peroxidase-conjugated streptavidin for 30 min at room temperature and visualized with diaminobenzidine in darkness for 10 min. Finally, sections were counterstained with hematoxylin. In each experiment, negative controls without the primary antibody were included to check for nonspecific staining.
Evaluation of immunostaining and microvessel counting
To evaluate E-Cad, N-Cad, VEGF, MMP-9, and COX-2 expressions, scoring criteria were established based on the percentage of positive cells and staining intensity. There were two count criteria: (a) percentage of positive cells (0: no immunopositive cells, 1:<10% positive cells, 2: 11%-50% positive cells, 3: 51%-75% positive cells, and 4: >75% positive cells) and (b) staining intensity (0: negative, 1: weak, 2: moderate, 3: high). The result of (a) multiplied by (b) was regarded as a score. Scores of 0-2, 3-5, 6-9, and 10-12 corresponded to grades negative (-), positive (+), positive (++), and positive (+++), respectively. Scores less than 6 were regarded as low expressions, otherwise, were high expressions. Intratumoral microvessels were highlighted by immunostaining with anti-CD34 monoclonal antibody. Any single or clustering endothelial cells stained brown clearly separated from the adjacent microvessels, tumor cells, and other connective tissue elements were considered as vessels. Branching structures were counted as a single vessel unless there was a discontinuity in the structure. The stained sections were screened at 100-magnification under a light microscope to identify the 3 regions of the section with the highest vascular density. Vessels were counted in 5 regions at 200 magnification, and the average number of microvessels were recorded. Two observers did the counting separately, and the mean value was used for analysis.
Statistical analysis
The correlations between expressions of E-Cad, N-Cad, VEGF, MMP-9, COX-2, and clinical pathological parameters were assessed by the Mann-Whitney or Kruskal-Wallis rank test. Mean differences in microvessel counts were compared by using the Student’s t-test. The expressions between the primary tumors and metastasic lymph nodes were assessed by the Two-related-samples tests Wilcoxon test. Correlations between the expressions of every protein were evaluated by the Spearman rank correlation coefficient. The Cox proportional hazard model was used for multivariate analysis of prognostic factors. P<0.05 was considered statistically significant. All P values were represented as two sided. All statistical analyses were performed with SPSS 13.0.
Results
Correlation between E-Cad, N-Cad, VEGF, MMP-9, COX-2, MVD, and clinicopathologic features
Expressions of E-Cad and N-Cad were found on the membrane and in the cytoplasm of the tumor cells (Fig. 1a, b). The E-Cad expressing in tumor cells of partial cases in metastatic group mainly located in the cytoplasm; VEGF immunoreaction was also present in some vascular endothelial cells in addition to the cytoplasm of tumor cells (Fig. 1c). The expressions of MMP-9 and COX-2 were shown as a cytoplasmic staining pattern and primarily distributed within the periphery of tumor masses (Fig. 1d, e). Statistical analysis indicated that the expressions of N-Cad, VEGF, MMP-9, and COX-2 in the group with lymph node metastasis were significantly higher than those without lymph node metastasis (P<0.05), while the expression of E-Cad was correlated negatively to the lymph node metastasis (P<0.05) (Table 1). The metastasis rate of lymph node in the cases with EMT (lower E-Cad expression and higher N-Cad expression) was 78.3%, while that without EMT (higher E-Cad expression and lower N-Cad expression) was 11.1%. The expression of COX-2 protein was higher in cases with grade III and grade II than that with grade I, respectively (P<0.05). In the cases with higher grade, the expression of E-Cad decreased, while the expression of N-Cad increased. Table 2 showed that there were significant differences in the expressions of E-Cad and N-Cad between cases with histological grade II and those with grade I (P<0.05). There was no significant association between the expressions of these proteins and the other clinicopathological parameters including patient age, tumor size, histological type, and clinical stage. CD34 immunoreaction was present in vascular endothelial cells (Fig. 1f). MVD was higher within the periphery of the tumor masses. The correlation between MVD and lymph node metastasis was shown in Table 3. High MVD was significantly related to the cases with lymph node metastasis (P<0.05).
Correlation between expressions of E-Cad, N-Cad, VEGF, MMP-9, COX-2, and MVD
A significantly negative correlation was found between the expression levels of E-Cad and MMP-9, COX-2 or VEGF, respectively (P<0.05). A significantly positive correlation was found between the expression levels of N-Cad and MMP-9, COX-2 or VEGF, respectively (P<0.05). The correlation among the expression levels of MMP-9, COX-2, and VEGF were significantly positive (P<0.05). The correlation between the expressions of MMP-9, COX-2 or VEGF, and MVD were summarized in Table 3. The mean MVD values in the tumors with high expression of MMP-9, COX-2, or VEGF were significantly higher than those in the tumors with low expression of MMP-9, COX-2, or VEGF (P<0.05).
Multivariate survival analysis
Multivariate survival analysis showed that lymph node metastasis (RR=4.107, P<0.05), MVD value (RR=1.034, P<0.05), and patient age (RR=1.091, P<0.05) were independent prognostic factors. The survival time of those with old age, higher MVD value, and with lymph node metastasis was shorter.
Discussion
The first key step in the metastatic cascade of malignant tumor is that the tumor cells become loose and then invade the adjacent stroma. Subsequently, the invasive tumor cells may pass through (blood vessel and/or lymphatic) endothelium into the circulatory system, reside in a new environment, and then proliferate to form metastatic focus in local lymph nodes and distant organs. Recent studies indicated that EMT was the mechanistic basis for tumor migration. EMT is referred to as the process by which the intracellular adhesions of polarized epithelial cells are reduced and converted into individual and motile cells, that is, the carcinoma cells are stably or transiently lose their original epithelial polarities and acquire a mesenchymal phenotypes and thus may disseminate to lymph nodes or even other organs of the body [
2]. However, there was still a controversy whether EMT is the only mechanism for breast cancer invasion and metastasis or not [
4]. Genome-wide transcriptional profiling of increasingly larger numbers of human breast cancer cell lines have confirmed the existence of a subgroup of cell lines (termed Basal B/Mesenchymal) with enhanced invasive properties and a predominantly mesenchymal gene expression signature, being distinct from other subgroups with predominantly luminal (termed Luminal) or mixed basal/luminal (termed Basal A) phenotypes [
5]. Related transduction pathways and signal factors may involve the EMT mechanism, most of which transcriptionally repress E-Cad during cancerous development. The down-regulation of E-Cad is the initial change of EMT. Tumor cells maybe lost the epithelial properties due to the down-regulation of E-Cad in many diffuse-type solid cancers such as breast infiltrating lobular carcinoma. The individual invasive tumor cells located in the boundary between carcinoma nest and stroma may also be converted to a lower expression of E-Cad, so-called EMT, which may be a transient change, and its opposite converse, MET, may appear when the tumor cells reach and reform a cancerous lesion in target organs. The cells of the secondary tumor then regained the up-regulation of E-Cad and the epithelial characteristics.
N-Cad, as a mesenchymal cadherin, owns reverse function to E-Cad mainly expressed in the tissue that originates from neuroectoderm and mesoderm. Up-regulation of N-Cad may increase the adhesion between tumor cells and stoma and induce invasion and metastasis of tumor cells [
6]. In the current study, the expression of E-Cad in the group with lymph node metastasis was higher than that without lymph node metastasis. The metastatic rate of lymph node in the cases with EMT (lower E-Cad expression and higher N-Cad expression) was significantly higher than that without EMT (higher E-Cad expression and lower N-Cad expression). It suggested that the EMT played a crucial function in the cascade of lymph node metastasis in breast cancer.
The mechanisms involved in EMT are integrated in concert with master pathways regulating tumor growth, angiogenesis, and mobility [
7]. The promoting factors for lymphangiogenesis, produced by tumor cells, may increase the lymphatic microvessel density, which is a basal event to tumor metastasis. VEGF may promote the endothelium proliferation of lymphatic vessels and is a useful prognostic marker by correlating significantly with lymphovascular invasion [
8]. The signal axis of VEGF-C, VEGF-D and their receptors is the main signal pathway for the formation of lymphatic vessels and lymphatic metastasis. Previous studies suggested that VEGF-C promoted lymphangiogenesis, and tumor lymphangiogenesis in turn promoted metastasis to lymph node and lung [
9]. In this study, the expression of VEGF in the group with lymph node metastasis was higher than that without lymph node metastasis, which indicated that the VEGF produced by tumor cells was prone to lymphatic metastasis. Meanwhile, VEGF may also increase the MDV by inducing angiogenesis. The high permeability of newborn vascular wall makes tumor cells easily immigrate to blood. MVD value is usually used to evaluate the potential of tumor vascular metastasis. In this study, there was a significant correlation between the expression of VEGF and MVD value. MVD value in group with lymph node metastasis was higher than that without lymph node metastasis and related with the prognosis of breast cancer. VEGF-C not only induces tumor lymphangiogenesis but also promotes tumor metastasis to regional as well as distal lymph nodes and organs, via promotion of lymphatic network expansion of sentinel lymph nodes. This expansion of lymphangiogenesis in sentinel lymph node was correlated with increased tumor metastasis to distant lymph nodes and eventually to the lung [
10]. Kaplan [
11] and Hiratsuka [
12] suggested that the remote establishment by primary tumors produced growth factors/cytokines of special permissive microenvironments in target organs prior to metastasis (pre-metastatic niches) induced the formation of secondary tumors.
MMP-9 as one kind of gelatinase may degrade the extracellular matrix, which is a barrier for tumor invasion, and is involved in the metastasis of many tumors [
13,
14]. In this study, the expression of MMP-9 in group with lymph node metastasis was significantly higher than that without lymph node metastasis, which indicated that the case with overexpression of MMP-9 was prone to undergo lymph node metastasis. The significant positive association between the expression of MMP-9 and VEGF, and the co-upregulation of the expression of MMP-9 and the value of MVD demonstrated that MMP-9 may promote angiogenesis, which was consistent with previous studies [
15]. MMPs are up-regulated in nearly all tumor types, which have been shown to induce EMT in cultured cells [
16]. During EMT, E-Cad is down-regulated, which correlates with increased motility and invasion of cells. In this study, there was a negative correlation between E-Cad and MMP-9, which was consistent with the previous study, suggesting that the up-regulation of MMP-9 may represent a mechanism for down-modulation of E-Cad in ovarian cancer [
17].
COX-2 as an inducible enzyme is induced in inflammatory and cancerous tissues. An increased level of COX-2 expression in tumor tissue and cell lines compared with that in normal tissue has been reported in many types of human tumors, and it has been demonstrated that COX-2 promoted malignant behavior [
18]. In this study, the expression of COX-2 in group with lymph node metastasis was significantly higher than that without lymph node metastasis; the expression of COX-2 was significantly positively correlated with the expressions of MMP-9 and VEGF in breast cancer; the MVD values in the tumors with strong expression of COX-2 were significantly higher than those among tumors with weak expression of COX-2, which has shown to be related with the increased level of vascular endothelial growth factors and MMPs induced by COX-2. The mechanism of the tumor angiogenesis functions of COX-2 has been shown to be derived from stimulation of endothelial migration by generating prostaglandin E2 (PGE2) and up-regulation of VEGF, and inhibition of endothelial apoptosis by facilitating the activation of Bcl-2 or Akt [
19,
20]. COX-2 may stimulate the lymphangiogenesis by up-regulating VEGF through the EP1/SRC/HER-2/NEu pathway. COX-2 activity also modulates the expression of MMP-9, which may be a part of the molecular mechanism by which COX-2 promotes cell invasion and migration. The increased expression of MMP-9 can be inhibited by the specific inhibitor of COX-2 [
21]. Our data suggested that the angiogenesis and lymphangiogenesis induced by the overexpression of COX-2 may be stimulated by the up-regulation of VEGF and MMP-9. A panel of investigations have suggested the possible biological multifunction of COX-2 in tumorgenesis and metastasis correlated with factors as stimulation of cell proliferation, inhibition of tumor cell apoptosis, enhancement of adhesion between tumor cells, and ECM,
etc. Overexpression of COX-2 and loss of E-Cad were observed in several cancer tissues, and the COX-2 inhibitor can induce E-Cad expression in bladder cancer [
22], which suggested that overexpression of COX-2 may be related to EMT. In this study, the correlation of high expression of COX-2 with the poor differentiation of breast cancer was consistent with the previous report [
23], suggesting an important function of COX-2 in tumorigenesis and differentiation of breast cancer.
EMT is considered to be a crucial event in the invasive and metastatic process. The cells owned EMT and metastatic proclivity are postulated to be the migrated cancer stem cell (CSC) [
24], in the descendants, of which the decreased expression of E-Cad, and the increased expressions of N-Cad, MMP-9, COX-2, and VEGF were observed. The reverse transition from a mesenchymal to an epithelial phenotype (MET) would be observed in metastatic focus where the CSC migrated to. In this study, although the average level of E-Cad in metastatic lymph node was lower than that in primary tumor, without a statistically significant difference, it was also observed that the expression of E-Cad was higher in some metastatic lymph nodes than in primary cancer. Thus, it would be postulated that MET of cancer cells occurs under the influences of local microenvironment.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(219KB)
