Expression of matrix metalloproteinase 2 (MMP-2), E-cadherin and Ki-67 in metastatic and non-metastatic canine mammary carcinomas

In order to invade, the neoplastic cell released from a primary tumour has to overcome a dense extracellular matrix (ECM) of proteins comprised of collagen, laminin, fibronectin, vitronectin and many other particles, which together form the basement membrane (BM). This membrane forms a mechanical barrier between the epithelium or endothelium and the surrounding tissue [1]. The ability of tumour cells to degrade components of the ECM is directly associated with their metastatic potential [2]. Matrix metalloproteinases (MMPs) are enzymes that play an important role in the destruction of the ECM. Metalloproteinases are Zn²⁺ dependent proteases [3], which are mainly produced in fibroblasts, monocytes, leukocytes, macrophages, neutrophils, endothelial cells and are secreted by tumour cells [4, 5]. The vascular endothelial growth factor (VEGF), tumour necrosis factor alpha (TNF-α), interleukin-1 (IL-1) and prostaglandin (PG) were shown to stimulate the synthesis of MMPs [5]. The gelatinases, particularly MMP-2 and MMP-9, are thought to play an important role in carcinogenesis. Their increased expression and correlation with unfavourable prognostic factors have been found in human and canine breast cancer [6–12]. Those proteinases lead to a breakdown of type IV collagen, which constitutes the basement membrane scaffolding, including that of the vascular endothelium. This facilitates the intravasation of tumour cells, which is an important stage of metastasis.

The mechanism through which a cell metastasizes and adheres to the target site is not fully understood. However, this process always involves the release of a tumour cell from the primary malignancy. Cells require various chemical signals (cytokines, chemokines, hormones, neurotransmitters) to begin migration. Cellular adhesion molecules (CAM), located in cell membranes, play an important role at this stage of the metastatic cascade. The level of their expression influences the strength of connections between the neighbouring cells [13, 14]. There are four main classes of adhesion molecules: cadherins, integrins, selectins and immunoglobulin-like particles. Cadherin plays an important role in the invasion and metastasis of the tumour [15]. It is a calcium-dependent transmembrane protein [8]. Three major molecules are included in this group: E-cadherin (Epithelial-cadherin), N-cadherin (Neuronal cadherin) and P-cadherin (Placental cadherin). Cadherins connect to neighbouring cells by binding to other cadherins through an extracellular N-terminal amino acid sequence of His-Ala-Val. The cytoplasmic domain of E-cadherin binds to a group of interconnected proteins, called catenins (α, β and γ). β and γ-catenins compete with one another and directly bind to E-cadherin. α-catenin, on the other hand, connects E-cadherin to actin F and α-actinin, which constitute the cell cytoskeleton [14]. Any functional disturbances in the cadherin-catenin complex (dysfunction and/or a lack of cadherin or catenin) reduce cell adhesion and disturb cellular differentiation. This leads to an increase in the invasive potential of the tumour and may support metastasis [16]. It has been shown that transfection of a murine mammary tumour cell line with E-cadherin cDNA decreases its invasiveness [17]. This underlines the importance of E-cadherin in the acquisition of a malignant phenotype of tumour cells.

The aim of this study was to demonstrate the expression (tumour cell positivity as well as expression intensity) of proteins associated with the metastatic potential of mammary gland carcinomas, i.e., MMP-2 and E-cadherin. We also aimed to determine the enhancement of their expression when correlated with the expression of the Ki-67 antigen in canine mammary carcinomas. The expression of Ki-67 can be observed already during G1 phase of the cell cycle; it increases markedly during S and G2 phases, reaches its peak level during M phase, and is absent in G0 cells. Consequently, Ki-67 is detected mostly in proliferating cells [18, 19]. Furthermore, the expression of the studied markers in metastatic tumours was compared to the results obtained from the tumours that had not metastasized.

The material for the study was sampled during surgery from 32 female dogs of various breeds, 8 to 14 years old. The tumours were verified as carcinomas based on a histopathological examination according to the classification by Goldschmidt et al. [20] as simple tubular (n = 17) and simple tubulopapillary carcinomas (n = 15) (12 cases with lymph node metastasis – diagnosed based on a fine needle biopsy, and 20 cases without metastasis following a 2 year observation). In order to confirm or rule out metastasis, the female dogs were examined 2 months after the surgical tumour removal. Thoracic radiography and abdominal ultrasonography were carried out in each dog. In every case, samples from the superficial inguinal or/and axillary lymph nodes (which have not been removed during mastectomy) were obtained by using a fine needle biopsy. Those examinations were repeated every 6 months.

Formalin-fixed, paraffin-embedded tissues were freshly cut into 4-μm- thick sections and mounted on Superfrost Plus slides (Menzel Gläser, Braunschweig, Germany). The sections were then dewaxed with xylene and gradually hydrated in alcohol. The activity of the endogenous peroxidase was blocked by a 5 min exposure to 3 % H₂O₂. The sections were then boiled in a microwave oven for 15 min in an Antigen Retrieval Solution [21, 22] (DakoCytomation, Glostrup, Denmark). In order to measure the levels of the studied antigens, the following antibodies were applied for 1 hour at room temperature (RT) in the following concentrations: polyclonal rabbit anti-MMP-2 (1:100; Chemicon Millipore, Billerica, MA, USA), monoclonal mouse anti-E-cadherin antibody (clone NCH-38, 1:150; DakoCytomation), monoclonal mouse anti-Ki-67 (clone MIB-1, 1:100; DakoCytomation). All the utilized antibodies were diluted in the Background Reducing Antibody Diluent (DakoCytomation). Next, the samples were incubated (15 min, RT) with secondary biotinylated antibodies and a streptavidin-horseradish peroxidase complex (LSAB2, HRP, DakoCytomation). The 3,3'-diaminobenzidine (7 min, room temperature, DakoCytomation) served as a substrate for the reaction. All the sections were counterstained with Meyer’s haematoxylin. In all the cases, controls were assessed and the specific antibody was substituted with the Primary Negative Control (DakoCytomation). Microphotographs of all the studied tumours were subjected to a computer-aided image analysis via a computer coupled to a BX53 optical microscope (Olympus, Tokyo, Japan). The set was able to record images and to analyse them digitally. The measurements were carried out using the Cell^A software (Olympus Soft Imaging Solution GmbH, Germany).

The expression of MMP-2 and E-cadherin was appraised using the modified semiquantitative IRS scale according to Remmele and Stegner [12, 23]. The method takes into account both the proportion of positively stained tumour cells and the intensity of the reaction colour, while its final result represents the product of both parameters, with values ranging from 0 to 12 points (no reaction = 0 points (−); weak reaction = 1–2 points (+), moderate reaction = 3–4 points (++), intense reaction = 6–12 points (+++)). The expression of the Ki-67 antigen was evaluated quantitatively by estimating the percentage of positive tumour cells (0–5 % = no reaction (0 points), 6–25 % = weak reaction (1 point), 26–50 % = moderate reaction (2 points), above 50 % = intense reaction (3 points)). For purpose of the statistical analysis the final IRS scores of MMP-2 and E-cadherin, as well as Ki-67 score for particular cases were used for deriving the mean values. All the analyses were performed by two independent pathologists, and in case of discrepant results, the sections were re-evaluated until a consensus was achieved.

The results were subjected to a statistical analysis using Prism 5.0 (GraphPad, La Jolla, CA, USA). The differences between the two groups were compared using the non-parametric test of Mann-Whitney, whereas associations between the expression of the analysed markers were assessed using the Spearman correlation and Fisher exact test. The results were considered significant at p < 0.05 in all the analyses. For purpose of statistical analysis the expression of studied antigens was dichotomized into low (none, low and moderate expression of studied markers) and high (intense expression) expressing tumours.

We did not show significant differences in MMP-2, E-cadherin and Ki-67 expression between metastatic and non-metastatic tumours due to low number of cases studied, however further experiments are necessary to assess the role of these antigens in the process of canine mammary tumours metastasis. Furthermore, the obtained results may suggest that canine carcinomas may be used as an experimental model for future studies on the mechanism of carcinogenesis and metastasis in humans. Nevertheless, the results of this study should be regarded as preliminary, as the analysis was performed on a small number of cases.

BM, basement membrane; CAM, cellular adhesion molecules; E-cadherin, epithelial-cadherin; ECM, dense extracellular matrix; IL-1, interleukin-1; MMP-2, matrix metalloproteinase 2; N-cadherin, neuronal cadherin; P-cadherin, placental cadherin; PG, prostaglandin; RT, room temperature; TNF-α, tumour necrosis factor alpha; VEGF, vascular endothelial growth factor