Background
MicroRNAs (miRNAs) are functional small nucleic acids that regulate the stability and translational efficiency of target messenger RNAs[
1]. Altered expression of mi-RNAs has been demonstrated in several human cancers where miRNA 'signatures' are found to be informative for tumour classification and clinical outcome[
2,
3]. Although several miRNAs are upregulated in specific tumour types[
4], a global reduction of miRNA abundance is the more common trait in human cancers. This results in considerable influence on the transformed phenotype[
5]. In human breast cancer the deregulation of miRNA expression was first demonstrated by Iorio et al.[
6], who suggested a possible role for miRNAs as robust biomarkers for breast cancer diagnosis and prognosis. Recent studies[
7] provide evidence that miRNAs are involved in many of the cellular regulatory processes, including activation of different signaling pathways and induction of apoptosis[
8,
9].
The heterogeneity of human cancers requires the use of multiple therapeutic modalities, including radiation therapy. However, the development of radioresistance presents a problem over prolonged courses of treatment[
7,
10]. Only a few studies have described the effect of radiation on miRNA expression profiles[
7,
11,
12]. There are indications that radiation sensitivity may be manipulated by influencing the expression of a single miRNA species[
11,
13,
14]. However, little is known of the underlying mechanisms. A more detailed knowledge about radiation influences on miRNA expression in tumour cells is important for improving the effectiveness and reducing the side effects of radiotherapy. Overexpression of miR-21 has been previously reported[
15‐
17]. Patients with high tumour miR-21 expression have a worse clinical outcome than those with low tumour miR-21 expression[
18]. A possible explanation was provided by a genome wide search for miR-21 targets. This suggested a functional link between miR-21 and the p53 tumour suppressor pathway[
17,
19], where p53-induced proteins provoke apoptosis in response to DNA damage after irradiation in cancer. One possibility to improve therapeutic strategy is the modulation of cell cycle progression. The fact that the radiation-induced G2-phase block is a universal event in tumour cells renders the G2/M checkpoint as target for improved efficacy of radiation therapy[
20]. Most of the cancer cells have mutations in genes involved in the G1 checkpoint such as p53, Rb, p16, MDM2 and cyclin D1[
21,
22]. Interestingly the G2 checkpoint is usually retained in the cancer cells with impaired G1 checkpoint. Therefore if the G2 checkpoint is selectively disrupted the cancer cells with impaired G1 checkpoint would become more sensitive to the DNA-damaging treatment compared with normal cells because normal cells still retain G1 checkpoint intact[
21].
In this study, we characterized expression of miR-21 in 86 invasive mammary carcinomas, supporting poor prognostic effects with high miR-21 expression. Additionally, we identify changes in miR-21 levels and cellular response regulation after irradiation in breast cancer cells. Furthermore, we give evidence that modulating the miR-21 expression level would be an important milestone in efficient breast cancer radiation therapy treatment.
Methods
Growth and maintenance of cell lines
The breast cancer cell line MDA–MB–361 was cultured in DMEM (Dulbecco Modified Eagles medium) with 20% FCS, (Invitrogen, Carlsbad, CA) and T47D was maintained in RPMI 1640 (Roswell Park Memorial Institute medium) supplemented with 10% FCS and human insulin (10 μg/ml). The cell cultures were maintained in a water humidified 37°C incubator with 5% CO2.
Ionizing radiation treatment
Irradiation of cell cultures containing 1 × 106 log phase cells was performed with a Cs-137 irradiator (HWM D-2000, Siemens, Germany) at a dose rate of 0.95 Gy/min. Doses of 2.5 Gy; 5.0 Gy or 7.5 Gy were administered at room temperature and control cells were sham irradiated. The exposed and sham irradiated cells were subsequently incubated at 37°C and harvested after indicated time points for RNA and protein isolation. The experiment was repeated for each dose in triplicate.
Lentivirus production and infection of breast cancer cell lines
Replication-defective lentiviral particles were produced by transient co-transfection of HEK293T cells in a 10 cm petri dish with 16 μg, 8 μg and 4 μg of packaging plasmids pMDLg/pRRE, pRSV. Rev and pMD2.G (a kind gift from D. Trono, École polytechnique fédérale de Lausanne) and 8 μg of lentiviral transduction vector pGreenPuro (pGP; System Biosciences, California) using Lipofectamine 2000 (Life Technologies, California) according to the manufacturer’s instructions. The pGP vector (named EV – empty virus in results section) was used as the backbone for miR-21 overexpression and miR-21 downregulation (anti-miR-21) by specific miRNA oligo cloning (pmiRZIP-21 - Cat. Nr. MZIP21-PA-1-GVO-SB; Biocat, Heidelberg, Germany).
The virus particles were harvested 48 hours after transfection, cleared and concentrated as previously described[
23]. According to virus titer determination virus productions ranged between 10
8 and 10
9 TU/ml (TU - Transduction Units). Viral infection of breast cancer cells was performed using protocols previously described[
24]. Briefly, 2 × 10
5 cells per well were infected with 4 × 10
5 TU/ml (defined as 2 MOI – multiplicity of infection) and three days after infection GFP expression was monitored. After infection 5 × 10
5 cells were irradiated for indicated time points. Microscopic analysis was done 48 hours post irradiation (HBO 50/AC and AxioCam MRC, Carl Zeiss AG, Germany).
RNA isolation for miRNA expression analysis
Paraffin-embedded tissue was microdissected with a sterile needle from 5 μm thick sections using a stereo microscope (Stemi 2000, Zeiss, Germany). A consecutive H&E-stained section was used for guidance. Tumour cell material (containing at least
> 80% tumour cells) was collected from all cases. Additionally, histologically normal ductal epithelium material was collected from five cases as control tissue. Total RNA was isolated from microdissected tissues as previously described[
25]. After digestion in lysis buffer and 500 μg/ml proteinase K the RNA was purified by phenol/chloroform extraction, ethanol precipitated, and dissolved in 20 μl RNase-free water. Five microlitres (100 ng) of RNA were reverse-transcribed using MultiScribeTM reverse transcriptase (Applied Biosystems; Foster City, CA, USA)[
26]. Further processing and evaluation of the results was performed according to the manufacturer’s instructions.
Total RNA was isolated from each of the breast cancer cell lines (MDA-MB-361 and T47D) after irradiation. Cells were pelleted by centrifugation at 1500 rpm for 5 min, and washed with 1 ml Dulbecco’s phosphate-buffered saline (PBS) without MgCl2 and CaCl2 (Invitrogen, Carlsbad, CA, USA). Small RNAs (<200 nucleotides) were isolated from the cells using the mirVana™ miRNA isolation kit (Applied Biosystems; Foster City, CA, USA) following the protocol for total RNA isolation. The quantity and quality of the total RNA and miRNA was measured with the Nanodrop spectrophotometer (PeqLab Biotechnology; Germany) and by running 2% agarose gels stained with ethidium bromide, respectively.
TaqMan-miRNA assays and data analysis
A specific single TaqMan – miRNA assay (Applied Biosystems, Forster City, CA, USA) was used for miR-21 expression analysis (Cat.Nr. 4427975; Assay ID 000397) in total RNA isolations from FFPE samples and from cells treated with irradiation. Quantitative PCR was performed on StepOnePlus Detection System (Applied Biosystems, Foster City, CA) according to the manufacturer’s instructions. The relative expression values of specific miRNA were calculated by using the 2
–ΔΔCT method[
27] normalized to the control miRNA (RNU43 and RNU44 - Cat.Nr. 4427975; Assay ID 001094 and 001095) and to the FFPE control or non-irradiated sample. All reactions were performed at least twice in duplicate.
Cell Proliferation and survival
Cell proliferation and viability was determined with a colorimetric cell proliferation WST1 kit (Roche, Manheim, Germany). Twenty-four hours before irradiation, 1000 to 2000 cells per well were seeded into 24-well plates. Three days after irradiation, 200 μl fresh growth medium and 20 μl WST1 labeling reagent were added and the cells were incubated for 2 hours in a 37°C incubator with 5% CO2. After incubation the absorbance was determined at 450 nm with reference length at 650 nm using a spectrophotometer plate reader (TECAN, Switzerland). For the measurement of clonogenic survival, cells were seeded in range of densities (500–2000 cells per plate) and 24 h later irradiation was performed. After 10–14 days, the colony formation capacity was assayed after ethanol fixation and Giemsa staining.
Cell cycle and subG1 fraction analysis
DNA staining of isolated nuclei for cell cycle analysis was performed using a modification of the method of Nüsse et al.,[
28]. At each indicated time, the treated cells were trypsinized and collected by centrifugation at 300 g for 5 min, and the supernatant was carefully removed. The cell pellet was gently resuspended in 500 μl of a solution containing 10 mM NaCl, 4 mM Na-citrate, 10 μg/ml RNase, 0.3% Nonidet P-40, and 50 μg/ml propidiumiodide (PI). The cell suspensions were incubated for 60 min at room temperature followed by the addition of 500 μl of solution containing 70 mM citric acid, 250 mM sucrose and 50 μg/ml PI. The cell suspensions were mixed and stored at 4°C before flow cytometry. Cell cycle distributions were analyzed on a FACScan LSR II (Becton- Dickinson) (excitation wavelength: 488 nm; emission wavelength: 610 nm, LSR II, Becton Dickinson/FACS DIVA Software). Cells with a DNA content less than that of cells in the G1 phase of the cell cycle (<2n) were assigned to the subG1 fraction and were considered to be apoptotic.
Patients and tumour samples
Formalin-fixed and paraffin-embedded (FFPE) archival material, obtained from 86 patients with invasive ductal breast carcinomas (IDC), was used for miRNA analysis. Forty-nine tumours were lymph node negative and 57 tumours were small in size (≤2 cm). Nine of the tumours were histological grade 1, 56 were grade 2, and 23 were grade 3[
29,
30]. The patients age ranged from 15 to 84 years (median 66 years). All patients were surgically treated, and no patient received preoperative adjuvant chemotherapy treatment. Postoperative 29 patients received radiation therapy treatment and 4 patients received Novaldex with radiation therapy. Detailed long-term clinical follow-up was available for all patients with a median follow-up period of 113 months (min. 5 months, max. 468 months). Forty patients relapsed with distant metastases within the total follow-up period. Ethical approval for the study was obtained from the Ethics Committee of the Medical Faculty of the Technical University of Munich.
Statistics
Correlation between histopathological markers and miRNA expression was examined by Spearman's rank correlation test. For univariate survival analysis Kaplan-Meier curves were calculated from 86 patients, and differences between strata were tested with the log-rank Chi-Square value. Results obtained in the in vitro experiments were tested using one- or two-way ANOVA and GraphPad Prism. In all analysis statistical significance was considered at the p <0.05 levels.
Discussion
Upregulation of miR-21 is a frequent miRNA alteration described in human cancers[
32]. The consequences of overexpression of miR-21 is that it acts as an “oncomir” blocking apoptosis[
33], promoting cell proliferation[
18,
34] and causing invasion and metastasis[
35,
36]. It appears that miR-21 targets multiple tumour-supressive pathways[
31] and recent studies showed convincing evidence that miR-21 negatively regulates Cdc25A and cell cycle progression in colon cancer[
37] and in human glioblastoma cells[
38]. According to miR-21 target analysis, Lu et al., demonstrated that miR-21 promotes cell transformation by targeting the programmed cell death 4 gene (PDCD4) in MCF7 breast cancer cells[
39]. There is some evidence that miRNAs are also involved in modulating radiation sensitivity in lymphoblastic cell lines[
7], endothelial cells[
14] and for resistance to cytotoxic anticancer therapy in lung cancer cells[
11]. Data published recently from Gwak et al.,[
40] demonstrate the importance of miR-21 knockdown in radiosensitation of glioblastomas. In correlation with our results they present importance of high miR-21 expression levels in conferring radiation resistance in glioblastomas. The roles of miR-21 expression in modulating response of breast tumour cells to irradiation have not been previously analyzed.
Therefore, we have investigated miR-21 expression in radiation resistant and radiation sensitive breast cancer cells after exposure to γ-irradiation. We observed that the expression of miR-21 was not significantly changed after 5 Gy exposure of the radiosensitive MDA-MB-361 cells, but was transiently increased in radiation resistant T47D cells. This data support hypothesis that miR-21 is not merely upregulated in association with oncogenesis, but rather can act as radioresistant miRNA when transiently overexpressed after radiation treatment[
36].
The G2/M checkpoint arrest is prominent after exposure to DNA damage reagents such as γ-irradiation[
21,
41]. Our cell cycle data analysis showed that the anti-apoptotic action of miR-21 is also evident after radiation exposure and correlates with radiation resistance. In addition, miR-21 influence cell cycle progression via the DNA damage-G2 checkpoint induction. In this matter miR-21 inhibition (anti-miR-21) is able to reduce the G2/M block and to enhance apoptosis induction 24 hours after radiation treatment (Figure
5A). All together these data suggest the importance of combination therapy such as radiotherapy with efficient G2/M check point inhibitor anti-miR-21. Supporting our results in the manuscript of Li et al.[
38], it is presented that miR-21 inhibitor reduces G2/M arrest what is inconsistent with recently published data from Gwak et al.,[
40] showing G2/M induction after miR-21 knockdown in glioblastoma cells. This highlights the importance of G2/M arrest after radiation treatment to be studied in different tumour cell types to further support a general conclusion about miR-21 function in radioresitance.
The data presented in Figure
6 confirm previously published data from Yan et al.[
18], demonstrating increased expression of miR-21 in breast cancer. We identify that patients with low expression levels of miR-21 have better clinical outcome. Previously it has been reported that high levels of miR-21 expression correlate with advanced clinical stage, lymph node metastasis and shortened survival of the patients[
18,
42]. This is confirmed by the association we observe between low miR-21 expression and distant metastasis free survival. The role of miR-21 in shaping the response to radiotherapy is suggested by the increased clinical survival seen for low miR-21 patient group after radiation therapy.
Conclusions
Taken together, our results show that miR-21 expression transiently increases in response to irradiation treatment in the T47D radiation resistant cell line. Furthermore, the miR-21 knockdown improved radiation induced apoptosis and growth arrest in radiation resistant cells almost to the same extent as in sensitive breast cancer cells (MDA-MB-361). These findings are important concerning the better clinical outcome for patients with low miR-21 expression levels and the use of miR-21 as potential target in breast cancer therapy.
Acknowledgements
The authors thank Katrin Lindner and Stefanie Winkler for excellent technical assistance. Partially supported by a grant from the Deutsche Wirtschaftsministerium (KF 2341803SB1) and by the European Commisson (Seventh Framework Programme) 'STORE' Project, no. 232628.
Competing interests
The authors declare no potential conflict of interest.
Authors’ contributions
NA, MA and MJA coordinated the study and drafted the manuscript. NA, IH, IGV, KR, and NL performed experiments and analyzed data. GA contributed important material for analysis. HB performed statistical analysis. All authors read and approved the final manuscript.