Abstract
Background: A novel model system to study the cellular, molecular, and genetic characteristics of breast cancer stem cells is needed. Personalized prognostic models are indispensable for assessing the effect of so-called tailor-made adjuvant therapy. Materials and Methods: Surgically-extirpated tissues were dispersed with Dispase and cultured in commonly used medium. The expression of tumor markers was detected with immunohistochemistry, and gene expression profiles of tissues and cells were analyzed using the Agilent Human Microarray. Results: We established a primary culture, which exhibited mesenchymal morphology. The gene expression profile of the primary culture revealed that the cells underwent epithelial-mesenchymal transition and had cancer stem cell properties. Conclusion: A primary culture of breast cancer cells can be easily established and could be used for studying breast cancer stem cells and assessing treatment of a patient.
Metastasis control is an important issue for treating many types of cancer, including breast cancer. Establishment of a primary culture of cells from surgically-extirpated tumor tissue is a promising method for better designing suitable treatments for individual patients (1). Moreover, cancer stem cells (CSCs) are thought to be important targets for cancer therapy (2, 3). Previous researchers have established a primary culture of breast cancer cells with stem cell properties (4, 5). This system represents a suitable in vitro model for developing therapeutic strategies aimed at eradicating the CSCs in breast cancer.
In this study, we established primary cultured cells from a surgically-extirpated tumor. Comprehensive characterization of normal and malignant tissues, a normal breast epithelial cell line, and the primary cell culture with immunohistochemistry and gene chip analysis revealed that the established primary culture underwent epithelial-mesenchymal transition (EMT) and also had properties of CSCs. These characteristic features represent a good model system for studying breast cancer stem cells (BCSCs). Moreover, we used this method as a model system for evaluating the effects of hyperthermia as a possible therapeutic treatment. This system may help predict response-to-treatment.
Materials and Methods
Tissue samples. The normal and malignant tissues were extra portions of surgically-extirpated scirrhous carcinoma of the female breast and were obtained at the Kobe Urban Breast Clinic (Kobe, Japan). The patient was 55 years old and received neoadjuvant systemic chemotherapy with letrozole, tegafur/gimeracil/oteracil, and cyclophosphamide/docetaxel. The patient provided written informed consent, and this study was approved by the Human Ethics Committee of Konan University (11-02).
Cells and cell culture. The chemically-transformed normal mammary gland breast epithelial cell line CRL-8798 was obtained from the American Type Culture Collection (Manassas, VA, USA), and human dermal fibroblast cells (HDF) were purchased from Summit Pharmaceutical (Tokyo, Japan). The breast tissue was digested overnight at 37°C with 1,000 pU/ml Dispase II (Eidia, Tokyo, Japan) or plated on a Petri dish and cultured for transmigration. The obtained primary cells were cultured in two different media: MCF medium [Earle's Eagle's Minimum Essential Medium, 1 mM sodium pyruvate, 10 μg/ml bovine insulin, 1× non-essential amino acids (Sigma, St. Louis, MO, USA), 10% Fetal Bovine Serum (EuroClone, Milano, Italy)] and a commercially available medium [designated ME medium in this study: Normal Human Mammary Epithelial Cell System (CC-3150; Lonza, Walkersville, MD, USA)].
Expression profiles of characteristic genes in MCKH cells.
Histochemistry. The extirpated tissues were fixed with 4% formalin and embedded in paraffin. Cultured cells were grown on coverslips and fixed with 4% formalin followed by fixation in 100% methanol.
Antibodies. The antibodies for the selected tumor markers were as follows: anti-estrogen receptor (ER) (Ab-14; Thermo Scientific, Cheshire, UK), anti-cytokeratin-19 (CK19) (BioLegend, San Diego, CA, USA), anti-CD66/carcinoembryonic antigen (CEA) (CEACAM5; Santa Cruz Biotechnology, Santa Cruz, CA, USA), anti-mucin 1 (MUC1; Abcam, Cambridge, UK), anti-breast cancer antigen 225 (BCA225; Abcam), anti-tumor associated glycoprotein 72 (TAG72; Abcam), anti-cancer antigen 125 (CA125; Abcam), anti-beta-tubulin (T5293; Sigma), and fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit IgG (American Qualex, San Clemente, CA, USA).
Gene expression profiles. The total RNAs of extirpated tissues and cultured cells were extracted with RNeasy (Qiagen, Tokyo, Japan) according to the manufacturer's instructions. The gene expression profiles were analyzed with Hokkaido System Science (Sapporo, Japan) using the SurePrint G3 Human GE Microarray Kit 8X60k (Agilent, Santa Clara, CA, USA). These data were compared to data obtained from normal tissue adjacent to the tumor, extirpated from the same patient.
Heat-shock experiment and semi-quantitative reverse transcription-polymerase chain reaction (RT-PCR). Cells were plated onto 35-mm culture dishes (2.5×105 cells/dish), and heat shocked by incubation at 40°C for 1 h, 43°C for 10 min, or 43°C for 1 h. Heat shock was carried out three times. The cell survival rate was measured with the 3-(4, 5-dimethylthiazolyl-2)–2,5-diphenyltetrazolium bromide (MTT) method. Total RNAs were extracted 1 h after the heat shock, and cDNAs were obtained with reverse transcription. The PCR program was 94°C for 2 min, followed by 19 cycles of 94°C for 1 min, 56°C for 1 min, and 72°C for 1 min, followed by a final cycle of 72°C for 4 min. Under these conditions, amplification was not saturated. The amounts of mRNAs were normalized to the amount of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA, and the expression profiles are shown as the ratio to that of control cells, which were cultured at 37°C.
Expression profiles of tumor antigens.
Results
Establishment of primary cultures from the breast tumor. Primary cultures were established using two different methods: a tissue-fragment culture and a dispersed cell culture. Cells were grown in two different media, MCF and ME medium. In this study, we used Dispase II for cell dispersal and obtained primary cultures from all plated dishes. On the other hand, the tissue-fragment culture method was less effective for establishing a primary culture. We found that the MCF medium was more suitable for establishing a primary culture from the breast tumor.
Both the tissue-fragment culture and the dispersed cell culture methods produced similar populations of primary cultured cells. At the initial phase of cell spreading, morphological heterogeneity was observed in which both a long, flattened mesenchymal phenotype and a columnar epithelial phenotype were present. However, several passages later, all primary cultures exhibited a uniform mesenchymal phenotype. In this study, these cell lines were deemed equivalent and were designated as MCKH (primary culture of mammary carcinoma from KH) primary cultured cells.
The primary culture recapitulated some original tumor characteristics but also exhibited a characteristic expression pattern including stemness. To characterize the primary cultured cells, we carried out gene expression analyses and performed immunocytochemical detection of selected tumor markers in the normal and malignant tissues, and CRL-8798 and MCKH cells. The tumor of this patient was originally characterized in a clinical test as ER/progesterone receptor-positive, human epidermal growth factor receptor-2-negative. In addition, the levels of CK19 in the patient's peripheral blood (3.7 ng/ml) were slightly higher than the normal levels, whereas levels of B-type natriuretic peptide (BNP), CA125, CA19-9, CEA, and squamous cell carcinoma antigen (SCC) were normal.
Table I summarizes the gene expression analyses. The RNA expressions of ER, progesterone receptor, CK19, MUC1, and some other genes were dramatically lower in MCKH cells compared to the malignant tissue. Some tumor markers [breast cancer anti-estrogen resistance 1 (BCAR1) and breast carcinoma amplified sequence 4 (BCAS4)], some genes in the NOTCH and WNT signaling pathways [NOTCH1, delta-like (DLL) 3, and WNT1], and a keratin gene (KRT12) maintained a normal expression pattern.
Immunocytochemical analyses showed that ER, CK19, CD66/CEA, MUC1, BCA225, and TAG72 were clearly expressed in the extirpated malignant tissue, whereas their expression was lost in the primary culture (Table II, Figure 1). Only CK19 was expressed faintly in early passages of the primary culture (Figure 1J and Table II). These observations suggest down-regulation of these genes during establishment of the primary culture.
Recently, CSCs have become an intriguing cell population within the tumor (5). Markers for human BCSCs are CD44+ and CD24−/low (6). As shown in Table I, MCKH cells lost CD24 expression and had up-regulated CD44 expression. Moreover, p21 (CIP1) and musashi homolog 1 (MSI1) have also been suggested to be putative stem cell markers in the breast (7). In MCKH cells, p21 was up-regulated more than five-fold than in the malignant tissue, and MSI1 was up-regulated more than 7-fold. All of these changes in the expression of stemness marker genes clearly show the stem cell characteristics of MCKH cells.
Moreover, the expression of a metallopeptidase gene (matrix metallopeptidase 1, MMP1) was dramatically increased (more than 2,800-fold) in MCKH cells, suggesting invasive characteristics of MCKH cells. CD133 expression may be related to metastatic localization in the bone (8). Slight up-regulation of CD133 in MCKH cells suggests high invasiveness into bone. In fact, bone metastasis was found six months after surgery in the patient who donated the tissue. Loss or reduced retinoblastoma expression is seen in high-grade breast adenocarcinomas (9), but no significant change in retinoblastoma gene expression was observed in MCKH cells.
Immunocytochemical detection of some tumor markers. A-H: Histochemistry of malignant tissue (A-D) and normal tissue (E-H). Sections were stained with hematoxylin and eosin (A and E), or immunohistochemically stained for cytokeratin-19 (CK19; B and F), mucin-1 (MUC1; C and G), and breast cancer antigen 225 (BCA225; D and H). Immunostained tissues were double stained with DAPI (2 μg/ml). Mammary ducts were rich in nuclei (F-H) and negative for the proteins examined. I-L: Primary cultured MCKH cells were photographed under phase-contrast microscopy (I) or immunohistochemically stained for CK19 (J and K) and tubulin (L). CK19 staining was positive in early-passage cells (J: population doubling level=1.68) and negative in aged cells (K: population doubling level=9.43). Scale bar=100 μm.
Primary cell culture exhibited the EMT phenotype. EMT is characterized by down-regulation of epithelial proteins including E-cadherin, epithelial cytokeratins (KRT8, KRT18, and KRT19), junctional adhesion molecules including occludins (OCLN), claudins (CLDN), and desmoplakin (DSP); and by up-regulation of mesenchymal proteins including vimentin (VIM), fascin-1 (FSCN), α-smooth muscle actin (ACTA2), N-cadherin (CDH2), cadherin-11 (CDH11), TGFβ-1 (TGFB1), and transcription factors related to cadherin gene expression including forkhead box C1 and C2 (FOXC1 and -2), goosecoid homeobox (GSC), snail homolog 1 and 2 (SNAI1 and -2), E12/E47 (TCF3), twist homolog 1 and 2 (TWIST1 and -2), and Zinc finger E-box binding homeobox 1 and 2 (ZEB1 and -2) (10).
Survival rate and heat shock protein gene expression after heat shock treatment. A: Relative viability 24 h after heat-shock. After heat shock at 43°C for 1 h, the difference between CRL-8798 and MCKH cells was significant (*p=0.036). For the MCKH cells, the difference between heat shock at 43°C for 10 min and 43°C for 1 h was also significant (**p=0.039). Significant differences were calculated with Student's t-test. B-D: Relative expression of heat-shock protein genes, 1 h after heat shock in HDF (B), CRL-8798 (C), and MCKH (D) cells.
As shown in Table III, MCKH cells exhibited down-regulation of epithelial markers and up-regulation of mesenchymal markers. In particular, E-cadherin was down-regulated 400-fold, and N-cadherin almost 200-fold compared to those in malignant tissue. These changes were not seen with CRL-8798 cells. Moreover, both CLDN genes were dramatically reduced, and ACTA2 and several transcription factors (SNAI1, SNAI2, and TWIST2) were increased by more than 4 times. These data strongly suggested the presence of EMT in MCKH cells.
Primary culture as an assay system for adjuvant treatment. Because MCKH cells retained some original tumor characteristics, we heat-shocked the MCKH cells to examine the usefulness of this model system for testing possible treatments for the patient. As shown in Figure 2A, although more than 50% of CRL-8798 cells died after heat shock at 43°C for 1 h, about 80% of HDF and MCKH cells survived after the same heat shock. HDF cells highly-overexpressed heat-shock protein (HSP)70, and this overexpression was present even after the 40°C heat shock (Figure 2C). In MCKH cells, in addition to overexpression of HSP70, all HSP genes examined in this experiment were overexpressed (Figure 2D) compared to the CRL-8798 cells (Figure 2B). This overall overexpression of HSP genes suggests resistance by MCKH cells to heat shock. Thus, the MCKH model system shows that heat shock may only be minimally effective in treating this patient.
Expression profiles of the epithelial-mesenchymal transition (EMT)-related genes in MCKH cells.
Discussion
In this study, the most suitable method for establishing a primary culture from a surgically-extirpated breast tumor was the dispersed cell culture method using Dispase. MCKH cells exhibited EMT, which may represent the metastatic state of this tumor. As EMT is thought to be a crucial event for carcinoma progression, basic studies of EMT using model systems are still important for understanding how to control the metastatic progression of tumors (1).
The other important feature of MCKH cells is their stemness. CSCs exist within many types of cancer and are thought to play pivotal roles in both tumor initiation and maintenance of tumor growth (11). However, the origin of these CSCs remains unknown. Two hypotheses have been proposed suggesting that CSCs originate from normal tissue stem cells or arise from differentiated cancer cells through EMT (12). MCKH cells clearly exhibited both EMT and stemness, strongly suggesting a close relationship between EMT and CSCs.
BCSCs have been well-studied. For example, Ponti et al. established CD44+/CD24low BCSCs (4), which can be propagated in vitro as non-adherent mammospheres. Li et al. carried out proteomic analyses using mammospheres as a model system (13). Although we could not culture MCKH cells as mammospheres or demonstrate their tumorigenic activity, primary culture of breast cancer cells with a mesenchymal phenotype and stemness, as was seen with MCKH cells, is a unique in vitro culture system.
Some reports also established a cultured cell line from immortalized MDA-MB453 human breast cancer cells (14) and primary culture of breast cancer (1), which display CSC and EMT characteristics, respectively. These two reports analyzed the expression of only a limited number of genes. In our experiment, comprehensive analysis of gene expression was used to compare the four closely related types of tissues and cells (MCKH cells, CRL-8798 cells, malignant breast tumor, and normal breast tissue). The large amount of data generated in this study will provide accurate information about MCKH cells and will likely make these cells a good experimental model for analyzing the mechanisms of EMT and the maintenance and differentiation of CSCs. Understanding the cellular, molecular, and genetic mechanisms of EMT, and the cellular, molecular, and genetic characteristics of CSCs using a primary culture model system such as MCKH cells may provide potential therapeutic targets for treating the disease.
In this study, using MCKH cells as a model system, we assessed the effects of hyperthermia. MCKH cells were relatively resistant to heat shock, which may be due to overexpression of HSPs, but these data also suggest that prolonged hyperthermia may favorably affect cancer regression. On the other hand, resistance to high temperature supports the idea that MCKH cells have stem cell properties.
Acknowledgements
This work was partly supported by Grants-in-Aid for Scientific Research and the “Core Research” project (2009-2014) from the Ministry of Education, Culture, Sports, Science and Technology (Japan).
Footnotes
-
This article is freely accessible online.
- Received April 4, 2013.
- Revision received June 1, 2013.
- Accepted June 3, 2013.
- Copyright© 2013 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
