Is the expression of matrix metalloproteinases (MMP-2, -9) and tissue inhibitors of metalloproteinases (TIMP-1, -2, and -3) associated with angiogenesis and clinicopathological features for breast cancer?

Introduction: Matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) are involved in the progression of several tumors, including breast cancer. Our aim was to investigate the association of immunohistochemical expression of protein MMP-2, and -9 and tissue inhibitors TIMP-1,2,-3 by tumoral cells in the process of angiogenesis and to define their relation with clinicopathological features for breast cancer. Methods: Immunohistochemical analysis of MMP-2,-9, TIMP-1,-2,-3, endoglin/CD105, estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) status was performed on 79 tissue samples of breast cancer with axillary lymph node dissection. Results: Statistically significant difference was found between mean age of patients and tissue inhibitors of metalloproteinase (TIMP-1) expression status (p=0.008), i.e., women with TIMP-1 negative tumors were on average younger (mean age 46.5) compared to women with TIMP-1 positive tumors (mean age 58.1); TIMP-2 expression status showed association with ER status (p=0.017), while TIMP-3 negative tumors were on average more frequently ER and PR negative (p=0.016; p=0.027). Status of protein expression of MMP-9 was associated with TIMP-1 protein expression status (p=0.033), i.e., breast cancers with overexpression of protein MMP-9 were more frequently TIMP-1 protein positive. Conclusion: Only TIMPs were associated with clinicopathological features for breast cancer. TIMP-2 expression was associated with worse (TIMP-2 positive tumors were frequently ER-negative), while TIMP-3 © 2017 Suada Kuskunović-Vlahovljak, et al.; licensee University of Sarajevo Faculty of Health Studies. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. *Corresponding author: Suada Kuskunović-Vlahovljak, Department of Pathology, School of Medicine, University of Sarajevo, Čekaluša 90, 71000 Sarajevo, Bosnia and Herzegovina. Tel./Fax.: +387 33 666 964. E-mail: suada.kuskunovic@mf.unsa.ba Submitted: 1 September 2017/Accepted: 18 September 2017 DOI: https://doi.org/10.17532/jhsci.2017.460 RESEARCH ARTICLE Open Access expression in tumoral cells was associated with better clinicopathological features for breast cancer (TIMP-3 positive tumors were frequently ERand PR-positive).


INTRODUCTION
Considering the fact that extracellular matrix (ECM) plays an important role in cancer progression and that matrix metalloproteinases (MMPs) are main molecules responsible for its remodeling, over the past years many studies have observed their influence to cancer progression. Since their discovery, there have been attempts to develop MMP inhibitor programs, especially for breast cancer, which as a multifactorial and heterogeneous disease is urging for more targeted and personalized therapy (1). MMPs are main regulators during several phases of the angiogenic process, from deposition and remodeling of ECM components to cell proliferation and migration (2). They contribute to angiogenesis either by degrading basement membranes or by promoting and maintaining the angiogenic phenotype (1,3). In breast cancer progression, particular importance was given to MMP-2 and MMP-9 due to their specificity for Collagen IV and possible implication in metastatic spread (4,5).
Although the results of a large number of studies showed a correlation between high levels of MMPs with a more aggressive form of disease and shorter overall survival period (4,6,7), some studies reported conflicting results (8,9).
The results are even more inconsistent when it is about the main tissue inhibitors of MMPs. TIMPs, as well as MMPs, are secreted by tumoral and stromal cells. Increased TIMP expression may also mean a stromal response to a tumor invasion, or may indicate a tissue response during the control of the activity of the MMP in maintaining the integrity of the ECM. TIMPs have an anti-MMP activity which are responsible for tumor suppression but can also stimulate growth or influence the apoptosis. The balance between anti-MMP and antiapoptotic effect on tumor growth may depend on the amount of bioavailable TIMP proteins in the tumor microcirculation. Due to their main biological role, it is expected that increased levels of TIMPs inhibit tumor invasion and formation of distant metastases and thus improve prognosis (10). However, elevated TIMP levels are reported in association with cancer progression and identified as poor prognostic indicators in several human tumor types, including breast cancer (11,12). Recent studies have found TIMP-1 serum levels enhanced in patients with triple-negative breast cancer (TNBC) and associated with poor prognosis (12), while others suggest TIMP-2 as a novel biological therapy for TNBC (13).
Having in mind all those facts, we aimed to investigate the possible association of immunohistochemical expression of protein MMP-2, MMP-9, TIMP-1, -2, and -3 by breast carcinoma cells with angiogenesis and to compare it with standard clinicopathological features for breast cancer in Bosnian women.

Clinicopathological data
The biopsy samples of 79 patients with invasive breast cancer (IBC) were diagnosed at the Department of Pathology, School of Medicine, Sarajevo, Bosnia and Herzegovina. All patients with IBC underwent partial or total mastectomy with axillary lymph node dissection. Patients who received neoadjuvant chemotherapy or radiotherapy, or had distant metastases at the time of diagnosis were excluded from the study.
Mean age of the patients at the time of diagnosis was 56.08 (range from 30 to 87).
All clinicopathological data are summarized in Table 1.

Immunohistochemical evaluation
The evaluation of immunohistochemical staining was carried out blind to the patient's data and pathological features.
Estimation of protein expression of MMP, TIMP, ER, PR, and HER-2 in breast carcinoma cells, as well as estimation of CD105 positive newly formed blood vessels in tumor tissue, was performed by the semi-quantitative method.
The staining intensity of protein MMP-2, MMP-9, TIMP-1, TIMP-2, and TIMP-3 and the number of stained tumor cells were both taken into consideration. Stains were scored on a scale of 0-3: Score 0, was assigned for no staining; Score 1, if cytoplasm and cell membrane weakly stained in <10% tumor cells; Score 2, for weak to moderate staining in 10-30% of tumor cells; and Score 3, if more than 30% of tumor cells strongly stained. The immunohistochemical expression was classified as positive or negative, considering 0 and 1 as negative and 2 or 3 as positive. Samples were considered to be MMP-2, MMP-9, TIMP-1, TIMP-2, and TIMP-3 positive when ≥10% of tumor cells were immunoreactive.

Quantification of angiogenesis
Sections stained with anti-CD105 antibody were used for the quantification of newly formed blood vessels in breast carcinoma tissue. The microscopic fields of breast carcinoma were observed by magnification ×200. Newly formed, CD105 positive blood vessels with visible vascular lumen were counted in 10 visual fields, separately. The total number of newly formed blood vessels was quantified as the mean value for each tumor sample and expressed as the total number of newly formed blood vessels by the high power field magnification. Isolated cells or vascular cells without visible lumen were not taken into account.

Statistical analysis
Statistical analysis was performed using SPSS version 21.0 (SPSS Inc., Chicago, IL).

Relationship between MMP protein expression in breast cancer tissue and clinicopathological features
Protein expression of MMP-2 and MMP-9 in breast cancer cells showed no relation to patients age, tumor size, tumor grade, lymphovascular invasion, lymph node status, nor to estrogen receptor (ER), progesteron receptor (PR) status and HER2 status ( Table 2).

Relationship between TIMP protein expression in breast cancer tissue and clinicopathological features
Statistically significant difference in mean age of patients was found to exist with respect to TIMP-1 expression status (p= 0.008), i.e., women with TIMP-1 negative expression were on average younger (mean age 46.5) compared to women with TIMP-1 positive expression (mean age 58.1). TIMP-1 expression status showed association with patients' age group (<50 or ≥50 years), i.e. patients aged ≥50 were on average more frequently TIMP-1 positive (p=0.000), while to other clinicopathological features showed no relation.
TIMP-3 expression status showed association with estrogen and progesterone receptor status; TIMP-3 negative tumors were on average more frequently ER-and PR-negative (p = 0.016 and p = 0.027, respectively) ( Table 2).

Relationship between mean number of CD105 newly formed blood vessels in breast cancer tissue and clinicopathological features
Statistically significant difference in a mean number of CD105 newly formed blood vessels, and tumor size (pT) was found to exist (between pT1 and pT2, p = 0.017) ( Table 2).
Regarding the tumor grade (G), statistically significant difference in a mean number of CD105 newly formed blood vessels were shown to exist between tumors of G1 and G2 (p = 0.002) and between tumors of G1 and G3 (p = 0.002) ( Table 2).
The mean number of newly formed blood vessels was higher in ER-negative and PR-negative tumors (compared to ER-and PR-positive tumors), (p = 0.002, and p = 0.006, respectively).

Relationship between MMP, TIMP protein expression, and CD105 newly formed blood vessels
An analysis of MMPs and TIMPs tissue expression revealed a significant association only between MMP-9 and TIMP-1. As shown in Table 3, the status of tissue expression of MMP-9 was associated with TIMP-1 tissue expression status (p = 0.033). We found that breast cancers with overexpression of protein MMP-9 were more frequently TIMP-1 protein positive. However, no statistically significant  difference in a mean number of CD105 newly formed blood vessels was found to exist between different groups with respect to MMPs and TIMPs expression status.

DISCUSSION
The development of metastases is a complex process involving angiogenesis and degradation of ECM, which are essential for the spread and proliferation of cancer cells. These events include the activity of MMPs and their tissue inhibitors (TIMPs) which seem to be involved in the propagation of various tumors, including breast cancer (3,15). The expression of MMPs is generally very low and mainly regulated by transcription. Other mechanisms of their regulation include post-translational modification, latent-zymogen forms of MMPs, and coexpression of TIMPs. Many members of MMP family are dysregulated in human cancers, especially MMP-1, -2, 7, -9, and 13 (1). In a variety of studies related to MMPs and TIMPs in breast cancer, different methods of assessment were used, including immunohistochemistry and gene expression, its activity in serum or plasma, or expression by mRNA in situ hybridization with the sole purpose to target them as diagnostic and therapeutic molecules in breast cancer (16).
In this study, we evaluated the significance of immunohistochemical expression of protein MMP-2, and -9 and their tissue inhibitors TIMP-1,-2, and -3 by tumoral cells in the process of angiogenesis, by comparing it with well-established clinicopathological features for breast carcinoma. The advantages of immunohistochemical evaluation are the direct readout of protein levels and the possibility to distinguish the expression of examined proteins in tumor versus stromal cells (16).
Immunohistochemical expression of the MMP-2 protein was positive in 64.56% of cancer cells, and MMP-9 protein in 92.41%. MMP-2 and MMP-9 status showed no association with clinicopathological features for breast cancer (Table 2).
Using immunohistochemistry, Vizoso et al. (17), Li et al. (18), and Talvensaari-Mattila et al. (19) reported opposite results in their studies, i.e., that the expression of metalloproteinases is related to 6±3.9 MMP-2: Matrix metalloproteinase-2, MMP-9: Matrix metalloproteinase-9, TIMP-1: Tissue inhibitor of metalloproteinase-1, TIMP-2: Tissue inhibitor of metalloproteinase-2, TIMP-3: Tissue inhibitor of metalloproteinase-3, N/A: Not applicable poor prognosis in breast cancer patients. Pellikainen et al. in the study of 421 patients found high MMP-9 expression in stromal cells to be prognostic for poor recurrence-free survival and breast cancer specific survival, while MMP-9 expression in tumoral cells was associated with smaller tumors and better survival outcomes in this cohort (20). Scorilas and colleges found that lower expression of MMP-9 by tumoral cells is associated with poor prognostic factors, i.e., with tumors of greater diameter (8).
No statistically significant difference in the mean number of CD105 newly formed blood vessels was found to exist between different groups with respect to MMP-2 and MMP-9 (Table 3), although the higher number of CD105 positive newly formed blood vessels was recorded in MMP-2 and MMP-9 positive tumors (p > 0.05).
In general, TIMPs are known for their ability to inhibit MMP activity, but recently many MMPindependent functions of TIMPs have been discovered, thus highlighting their dual role in cell-ECM interactions and tumor behavior (21).
TIMPs are able to inhibit all active MMPs, but not with the same efficacy. TIMP-1 mostly inhibits MMP-7, MMP-9, MMP-1, and MMP-3, while TIMP-2 is the most effective inhibitor of MMP-2. TIMP-3 can inhibit MMP-2 and MMP-9 (22). In our study, the status of tissue expression of MMP-9 was associated with TIMP-1 tissue expression status (p = 0.033) ( Table 3), which was expected due to previously mentioned facts. Jinga et al. pointed to the possibility that imbalance between MMP-9 and TIMP1 can be involved in the development of invasive breast carcinoma (4). By contrast, Thorsen et al., in a large study of 465 breast cancer patients examining the plasma concentration of MMP-9/TIMP-1 complex, found no correlation with disease-free survival (23). High serum and tumor levels of TIMP-1 have been reported in association with poor response to chemotherapy and decreased survival (24,25). Our results showed the statistically significant difference in mean age of patients with respect to TIMP1 expression status (i.e., between groups with positive and negative expression of TIMP-1) (p = 0.008). Women with TIMP-1 negative expression were on average younger (mean age 46.5) compared to women with TIMP-1 positive expression (mean age 58.1). TIMP-1 expression status showed association with patients' age group (<50 or ≥50 years), i.e., patients aged ≥50 were on average more frequently TIMP-1 positive (p = 0.000) ( Table 2). TIMP-2 expression status showed association with ER status (p = 0.017), i.e., TIMP-2 positive tumors were on average more frequently ER-negative, while TIMP-2 negative tumors were more frequently ER-positive (Table 2). Same results showed other studies (11,26) once again indicating to the relation between TIMP-2 and aggressive behavior of breast cancer.
Among all TIMPs, TIMP-3 has the broadest spectrum of inhibition, and unlike others, TIMP-3 is tightly bounded to ECM (27). TIMP-3 expression status was associated with estrogen and progesterone receptor status; TIMP-3 positive tumors were on average more frequently ER-and PR-positive (p = 0.016 and p = 0.027, respectively) ( Table 2). Measuring mRNA levels of TIMPs, Span et al. reported TIMP-3 as an only possible predictor for relapse-free survival in breast cancer patients, i.e., breast cancers with high TIMP-3 expression showed a better response to endocrine therapy (28). Likewise, Vizoso et al. have reported significantly higher TIMP-3 expression in ER-positive tumors. They also found that TIMP-3 expression by fibroblastic cells, but not by tumoral cells correlates positively with distant metastases (17).
We also investigated possible differences in angiogenesis, i.e., the mean number of CD105 positive newly formed blood vessels with regard to standard clinicopathological features for breast carcinoma. Statistically, significant difference was observed in a mean number of CD105 newly formed blood vessels between tumors of G1 and G2 (p = 0.002), and between tumors of G1 and G3 (p = 0.002) ( Table 2), i.e., tumors of Grade 2 had on average more CD105 positive newly formed blood vessels compared to Grade 1 tumors as well as tumors of Grade 3 compared to Grade 1 tumors. An increase in the number of newly formed blood vessels was observed with increasing of tumor size, although statistically significant difference was found to exist only between tumors of pT1 and pT2 (p = 0.017) ( Table 2), showing that tumors larger than 2 cm need more blood supply for progression. The mean number of newly formed blood vessels was higher in ER-negative and PR-negative tumors (compared to ER-and PR-positive tumors), (p = 0.002, and p = 0.006, respectively), which confirmed the findings of previous studies and also indicate that ER-and PR-negative tumors are associated with more aggressive clinical behavior, higher histological grade and negative clinicopathological features for breast carcinoma (29).

CONCLUSIONS
To the best of our knowledge, this is the first study involving MMPs and TIMPs in breast cancer of Bosnian women. When it comes to tissue inhibitors of MMPs our results especially highlight the complex role of TIMPs in cancer progression as well as the complexity of TIMP-MMP interaction. In support of that are findings that some of TIMPs were associated with worse (i.e., TIMP-2) and some with a better pathological prognostic-predictive factor for breast carcinoma (i.e., TIMP-3), as well as the existence of MMP-TIMP dependency (i.e., MMP-9 and TIMP-1). Results of our study do not indicate a significant association of MMPs with angiogenesis (mean number of CD105 newly formed blood vessels) nor with clinicopathological features for breast cancer. Possible reasons for that could be relatively small sample size, ethnic differences, the usage of different antibodies or laboratory procedures compared to previous studies and the fact that we assessed the protein expression only in tumoral cells. Further studies, especially prospective ones, with greater sample size and the usage of different methods to asses MMPs and TIMPs in breast cancer, will be necessary to determine the impact of MMPs and TIMPs to progression and outcome in Bosnian women.