Abstract

Background

Previous studies have shown that altered forms of MRE11, a protein known to play a vital role in DNA double-strand break repair, DNA replication, and telomere maintenance are associated with cancer outcomes. We investigated the role of MRE11 in breast cancer in both clinical and in vitro settings.

Methods

We examined MRE11 expression in tumor tissues from invasive ductal carcinoma breast cancer patients (n = 254) by immunohistochemistry, and associations with clinicopathological characteristics and overall survival were assessed using Cox proportional hazards regression models and Kaplan–Meier survival curves. Effect of MRE11 overexpression and knockdown on cell proliferation, invasion, and radioresistance was assessed in vitro using breast cancer cell lines (MCF-7 and MDA-MB-231). We also investigated the mechanisms involved. Effect of MRE11 overexpression on tumor growth was assessed in an orthotopic xenograft model (n = 8 mice per group). All statistical tests were two-sided.

Results

Of the 254 tissue samples, 69.3% and 30.7% showed high and low MRE11 expression, respectively. High MRE11 expression was statistically significantly associated with malignant cancer behavior compared with low MRE11 expression (eg, stages III and IV vs stage I, P = .004; poor overall survival, P = .005). MRE11 overexpression in breast cancer cell lines promoted cell proliferation through STAT3, cell cycle entry, invasion and migration, and radioresistance via enhanced DNA repair activity and also inhibited apoptosis; knockdown of MRE11 had the opposite effect. In xenograft tumor–bearing mice (n = 8 per group), increased tumor growth was observed in the MRE11-overexpressing group compared with the control group (tumor volume at week 8, control vs MRE11-overexpressing tumor originating from MCF-7 cells, mean = 280.4mm 3 , 95% confidence interval [CI] = 62.4 to 498.4mm 3 vs mean = 631.0mm 3 , 95% CI = 296.9 to 965.0mm 3 , P = .043).

Conclusion

High MRE11 expression was associated with a more malignant behavior in breast cancer. MRE11 may be a novel oncoprotein and may therefore serve as a new therapeutic intervention against breast cancer.

Breast cancer is the leading malignancy in women worldwide and still continues to grow at a rapid pace. Various risk factors for breast cancer have been identified, including age, early menarche and late menopause, exogenous hormone usage, familial history, geographic location, increased breast density, lifestyle, nulliparity and older age at first delivery, obesity after menopause, and radiation exposure ( 1–4 ). However, the underlying molecular mechanisms for breast cancer are still poorly understood, and further investigation is merited in order to improve breast cancer prevention, treatment, and prognosis.

DNA repair process is essential for maintenance of genomic integrity, and it becomes activated when DNA damage occurs, thus making it an essential component for cell survival and functioning ( 5 , 6 ). DNA double-strand breaks (DSBs) are one of the major threats to genomic integrity—causing chromosome breaks, deletions, and translocations—events frequently observed in cancer cells ( 6–8 ). DSBs are repaired by either one of two mechanistically distinct pathways, namely homologous recombination and nonhomologous end-joining (NHEJ) pathways ( 9–12 ). MRE11/RAD50/NBS1 (MRN) nuclease complex is involved in both homologous recombination and NHEJ pathways ( 13 , 14 ), and upon ionizing radiation, MRE11 (meiotic recombination 11 homolog A; also known as MRE11A), RAD50, and NBS1 (also known as nibrin) colocalize to the DSB sites and form distinct repair foci ( 15 , 16 ). Other than its roles in DSB repair and coordinating checkpoint response, MRN complex also plays a vital role in DNA replication and maintenance of telomeres, which are linked to cellular immortalization and neoplastic formation ( 17 , 18 ).

MRE11, a core protein of the MRN repair complex ( 13 , 14 ), is also involved in the recruitment and activity of telomerase ( 19–21 ), and null mutation in MRE11 leads to murine embryonic stem cell death ( 22 ). Furthermore, MRE11 is highly expressed in proliferating tissues as well as proto-oncogene simian virus 40 large T antigen-immortalized cardiomyocytes ( 23 , 24 ). Therefore, it is presumably possible that other than its function in DSB repair, MRE11 may also facilitate the development of human cancers under certain circumstances.

In this study, we explored the involvement of MRE11 protein in breast cancer development in vitro and in a mouse model in vivo. We also investigated the underlying mechanisms.

Materials and Methods

Patient Samples

Surgically treated female breast cancer patients (n = 254) with confirmed pathology of invasive ductal carcinoma were included, and their clinicopathological characteristics are summarized in Table 1 . The breast cancer tissues were obtained from the patients when undergoing surgical treatment at the Department of Surgery, Kaohsiung Medical University Hospital, during the period from 2006 to 2010. All the patients received primary treatment by surgery followed by adjuvant radiotherapy, chemotherapy, or hormone therapy. The ethics committee approved the study, and informed consent was obtained from each patient. The histological types and grades of the primary tumors were determined according to a system modified from the WHO classification and the Modified Bloom–Richardson Grading Scheme, respectively ( 25 , 26 ), whereas the staging was according to the AJCC TNM staging system ( 1 ). We obtained data on estrogen receptor (ER) and progesterone receptor (PR) status, which was analyzed by immunohistochemical staining, and HER2 staining was evaluated by the standard Hercept Test procedure (Dako 5204). Chemotherapy consisted of six cycles of fluorouracil, epirubicin, and cyclophosphamide or six cycles of docetaxel, epirubicin, and cyclophosphamide. Hormonal therapy, usually tamoxifen and anastrozole, was administered to patients who were ER positive.

Table 1.

Clinicopathological characteristics of breast cancer patients and association with MRE11 expression*

    MRE11 Expression   
    Low  High   
Variable  Patient, no. (%)  No. (%)  No. (%)  P †  
No.  254 (100)  78 (30.7)  176 (69.3)  – 
Stage‡         
    I
    II
    III and IV  90 (35.4)
 93 (36.6)
 71 (28.0)  37 (47.4)
 29 (37.2)
 12 (15.4)  53 (30.1)
 64 (36.4)
 59 (33.5)  .004 
Grade§         
    1
    2
    3  17
173
64  10 (12.8)
52 (66.7)
16 (20.5)  7 (4.0)
121 (68.8)
48 (27.3)  .026 
Age, y         
    ≤50
    >50  122
132  43 (55.1)
35 (44.9)  79 (44.9)
97 (55.1)  .13 
BMI, kg/m 2         
    ≤24
    >24  141
113  39 (50.0)
39 (50.0)  102 (58.0)
74 (42.0)  .24 
Tumor size, cm         
    <2
    2–5
    >5  137
90
27  56 (71.8)
19 (24.4)
3 (3.8)  81 (46.0)
71 (40.3)
24 (13.6)  <.001 
LN metastasis         
    Negative
    Positive  163
91  58 (74.4)
20 (25.6)  105 (59.7)
71 (40.3)  .024 
Recurrence in patients with radiotherapy (n = 156)         
    No
    Yes  129
27  41 (95.3)
2 (4.7)  88 (77.9)
25 (22.1)  .010 
Recurrence in patients with chemotherapy (n = 219)         
    No
    Yes  190
29  63 (94.0)
4 (6.0)  127 (83.6)
25 (16.4)  .035 
Recurrence in patients with hormone therapy (n = 161)         
    No
    Yes  148
13  52 (96.3)
2 (3.7)  96 (89.7)
11 (10.3)  .15 
ER status         
    ER-
    ER+  83
171  21 (26.9)
57 (73.1)  62 (35.2)
114 (64.8)  .19 
PR status         
    PR−
    PR+  112
142  33 (42.3)
45 (57.7)  79 (44.9)
97 (55.1)  .70 
HER2 status         
    HER2−
    HER2+  164
90  53 (67.9)
25 (32.1)  111 (63.1)
65 (36.9)  .45 
    MRE11 Expression   
    Low  High   
Variable  Patient, no. (%)  No. (%)  No. (%)  P †  
No.  254 (100)  78 (30.7)  176 (69.3)  – 
Stage‡         
    I
    II
    III and IV  90 (35.4)
 93 (36.6)
 71 (28.0)  37 (47.4)
 29 (37.2)
 12 (15.4)  53 (30.1)
 64 (36.4)
 59 (33.5)  .004 
Grade§         
    1
    2
    3  17
173
64  10 (12.8)
52 (66.7)
16 (20.5)  7 (4.0)
121 (68.8)
48 (27.3)  .026 
Age, y         
    ≤50
    >50  122
132  43 (55.1)
35 (44.9)  79 (44.9)
97 (55.1)  .13 
BMI, kg/m 2         
    ≤24
    >24  141
113  39 (50.0)
39 (50.0)  102 (58.0)
74 (42.0)  .24 
Tumor size, cm         
    <2
    2–5
    >5  137
90
27  56 (71.8)
19 (24.4)
3 (3.8)  81 (46.0)
71 (40.3)
24 (13.6)  <.001 
LN metastasis         
    Negative
    Positive  163
91  58 (74.4)
20 (25.6)  105 (59.7)
71 (40.3)  .024 
Recurrence in patients with radiotherapy (n = 156)         
    No
    Yes  129
27  41 (95.3)
2 (4.7)  88 (77.9)
25 (22.1)  .010 
Recurrence in patients with chemotherapy (n = 219)         
    No
    Yes  190
29  63 (94.0)
4 (6.0)  127 (83.6)
25 (16.4)  .035 
Recurrence in patients with hormone therapy (n = 161)         
    No
    Yes  148
13  52 (96.3)
2 (3.7)  96 (89.7)
11 (10.3)  .15 
ER status         
    ER-
    ER+  83
171  21 (26.9)
57 (73.1)  62 (35.2)
114 (64.8)  .19 
PR status         
    PR−
    PR+  112
142  33 (42.3)
45 (57.7)  79 (44.9)
97 (55.1)  .70 
HER2 status         
    HER2−
    HER2+  164
90  53 (67.9)
25 (32.1)  111 (63.1)
65 (36.9)  .45 

P values were calculated using a two-sided χ 2 test.

‡ Staging was based on the AJCC TNM staging system (1).

§ Grading was based on the Modified Bloom–Richardson Grading Scheme (26).

Tissue Microarray and Immunohistochemical Analysis

All tissues for tissue microarray were obtained from formalin- fixed, paraffin-embedded tissue blocks. Slides from hematoxylin- eosin–stained sections were reviewed by a pathologist to select representative areas of tumor and normal regions for core punches within the primary paraffin block by the technicians. The construction of the tissue microarray was performed using Alphelys Arraying Device (Alphelys, Plaisir, France), as described before ( 27 ).

Five-micrometer sections were cut from the recipient tissue microarray block by using a microtome with an adhesive-coated tape sectioning system (Alphelys). The detailed protocol for immunohistochemical analysis has been described elsewhere ( 28 ). Goat anti-human MRE11 polyclonal antibody (dilution 1:100) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA), and the same antibody was also used for immunoblotting. Images of immunohistochemically stained sections were captured by the Nikon E600 microscope (Nikon, Tokyo, Japan) and then processed by Adobe Photoshop 6.0 (Adobe System Incorporated, San Jose, CA). Notably, only the nuclear staining of MRE11 in tumor cells (about 1000 cells in three to four high-power fields) was counted. The results for MRE11 staining were stratified into quartiles for statistical analysis on the basis of the percentage of positively stained cells (score 1, ≤25% positive cells; score 2, 26%–50% positive cells; score 3, 51%–75% positive cells; and score 4, ≥76% positive cells). For further statistical analysis, scores 1 and 2 were categorized as low expression, and scores 3 and 4 were categorized as high expression. The staining was determined separately for each specimen by two independent experts simultaneously under the same condition. In rare cases, discordant scores were reevaluated and scored on the basis of consensual opinion.

Cell Culture

Human breast cancer cell lines (MDA-MB-231, MCF-7, ZR75-1, and T47D) used in this study were cultured according to the instructions from The American Type Culture Collection (ATCC, Manassas, VA) and were grown respectively in Leibovitz’s L-15 medium for MDA-MB-231, Minimum Essential Medium (MEM) for MCF-7, or Dulbecco’s modified Eagle medium (DMEM) for ZR75-1 and T47D. The genotypes and phenotypes of the cell lines were authenticated by Bioresource Collection and Research Center, Taiwan ( www.bcrc.firdi.org.tw ). All media were purchased from Invitrogen (Carlsband, CA) and were supplemented with 10% (vol/vol) fetal bovine serum (FBS) (Hyclone, Logan, UT) and antibiotics (100 units/mL penicillin, 100 µg/mL streptomycin, and 2.5 µg/mL amphotericin B) (Biological Industries, Haemek, Israel). In addition, human renal epithelial cells (293T) were maintained in DMEM supplemented with 10% FBS and antibiotics (above concentrations). All cells were grown and maintained at 37°C in a 5% CO 2 incubator (Thermo Electron Corp, New Castle, DE).

Transient Transfection Using MRE11-Specific shRNA for MRE11 Knockdown

Three short hairpin RNA (shRNA) target sequences for human MRE11 (shRNA1, 5′-GCTGCGTATTAAAGGGAGGAA-3′; shRNA2, 5′-GCCCACATCTTTATTGAACTT-3′; and shRNA3, 5′-TAAGAGACAGACACTTGGTTA-3′) (National RNAi Core Facility, Academia Sinica, Taiwan) were tested in MCF-7, T47D, and MDA-MB-231 breast cancer cells for MRE11 silencing. The double-stranded shRNA oligonucleotides were cloned into the pLKO.1-hairpin lentiviral vector. A control shRNA unrelated to human sequences was used as a negative control (SHC002, Sigma, St Louis, MO). Lentiviruses were prepared by transfecting three plasmids (the packing plasmid pCMV8.91, the envelope plasmid pMD.G, and MRE11-shRNA plasmid pLKO.1) (National RNAi Core Facility) into 293T cells using Lipofectamine 2000 (Invitrogen, Carlsband, CA), according to the manufacturer’s protocol. Viral soup was collected continuously for 3 days after transfection. For the best effective MRE11 knockdown without cytotoxicity and off-targeting effect, MDA-MB-231 cells were transfected by the lentivirus at a range of 50–100 multiplicity of infection. The transfection efficiency of lentiviral MRE11-shRNA in MDA-MB-231 cells was high (mean [SD]% of cells transduced = 97.50 [0.51]%). MRE11-shRNA and control shRNA transfectants were examined for MRE11 expression by immunoblotting. Notably, we could not obtain any stable clones of MDA-MB-231 cells with MRE11 knockdown from three rounds of clonal selection experiments. For all subsequent experiments, MRE11 knockdown clones were used 3 days after transfection because optimal knockdown of MRE11 expression was observed at this time point.

Stable Transfection of a Plasmid Construct for MRE11 Overexpression

MRE11 gene was cloned into pReceiver-M03 (Genecopoeia, Rockville, MD) vector backbone for fusion with the GFP reporter gene. The vector also contained the neomycin resistance gene. MCF-7 cells (5.8×10 5 ) were seeded on a 6-cm dish (Corning Life Sciences, Corning, NY) and incubated at 37°C overnight. Lipofectamine LTX (Invitrogen, Carlsband, CA) was used for transfection. The cells were grown in G418 (A.G. Scientific, San Diego, CA) at 800 μg/mL for stable clone selection. Selected stable clones were maintained in the growth medium with G418 at 500 μg/mL. MRE11-GFP expression in the stable clones was determined by immunoblotting and densitometric analysis using a UVP BioSpectrum 500 imaging system (UVP LLC, Upland, CA). The stable clones were categorized into relatively high, moderate, or low expression of MRE11-GFP and used for further experiments. Also, two clones stably transfected with empty vector and confirmed by immunoblotting were used as control for further experiments.

Cell Viability Assay

For cell viability assay, MDA-MB-231 cells with MRE11 transient knockdown (1×10 5 cells per well in six-well culture plate) and MCF-7 cells with stable MRE11 overexpression (1×10 5 cells per well in six-well culture plate) were treated with trypsin-EDTA (Invitrogen, Carlsband, CA) and resuspended in 100 μL medium, mixed thoroughly with 100 μL of 0.4% trypan blue (Invitrogen, Carlsband, CA), and incubated at room temperature for 15 minutes. Total cell count and viability data were obtained using an automated cell counter, Countess (Invitrogen), according to the manufacturer’s instruction. Three independent experiments were performed in triplicate.

XTT Colorimetric Assay

Cell viability and proliferation of MDA-MB-231 cells with MRE11 transient knockdown and MCF-7 cells with stable MRE11 overexpression (5×10 3 cells per well in 96-well culture plate for each cell line) were determined by tetrazolium salt 2,3-bis[2-methyloxy-4- nitro-5-sulfophenyl]-2H-tetrazolium-5-carboxanilide (XTT) assay (Roche Applied Science, Indianapolis, IN). At different time points after seeding, the culture medium was removed and XTT assay was performed by measuring the optical density (OD) at 490nm and subtracting the nonspecific background at 650nm (OD 490–650 nm ); we have described the details in a previous article ( 29 ). Three independent experiments (five replicates in each) were performed.

To study the effect of STAT3 inhibition on cell viability, MCF-7 cells with stable MRE11 overexpression (5×10 3 cells per well) were treated with 10 µM Stattic (Tocris Bioscience, Bristol, UK), a specific inhibitor of STAT3 phosphorylation and dimerization, for 24 hours ( 29 ). Three independent experiments (five replicates in each) were performed.

Apoptosis Assays

Annexin V Staining.

Annexin V-FITC fluorescence microscopy kit (BD Biosciences, San Jose, CA) was used to detect early apoptotic cells during apoptotic progression according to the manufacturer’s user manual. Briefly, MDA-MB-231 cells with MRE11 transient knockdown or MCF-7 cells with stable MRE11 overexpression (2×10 3 cells per well) were seeded on the eight-well chamber Labtek II slide (Nunc, Naperville, IL). The cells were fixed with 4% of paraformaldehyde (Sigma, St Louis, MO) for 15 minutes and washed twice with ice-cold phosphate-buffered saline (PBS; 137mM NaCl, 2.7mM KCl, 10mM Na 2 HPO 4 , and 2mM KH 2 PO 4 [pH 7.4]). The cells were then stained with 1:10 diluted Annexin V-FITC in the 1X Annexin V binding buffer for 15 minutes at room temperature. Consequently, the stained cells were observed under a fluorescent microscope (Zeiss Axioskop; Carl Zeiss Poughkeepsie, NY) after washing with the binding buffer. Three independent experiments were performed, and 1000 cells were counted in each experiment.

TUNEL Staining.

DNA fragmentation was analyzed by terminal deoxynucleotidyl transferase–mediated dUTP nick end labeling (TUNEL) staining using DeadEnd Colorimetric TUNEL kit (Promega, Madison, WI) according to the manufacturer’s instruction. Briefly, MDA-MB-231 cells with MRE11 transient knockdown or MCF-7 cells with stable MRE11 overexpression (2×10 3 cells per well) were seeded onto the eight-well chamber Labtek II slide (Nunc) and were fixed in 4% paraformaldehyde (Sigma, St Louis, MO) for 15 minutes and permeabilized with 0.2% Triton X-100 (Sigma) in PBS for 5 minutes. Cells were washed with PBS and incubated with 200-μL TUNEL stain–specific buffer for 1 hour at 37°C. The TUNEL reactions were terminated by washing the cells with washing buffer for 5 minutes three times at room temperature. Apoptotic cells (dark brownish color) were counted under a light microscope (TE2000U, Nikon, Japan). Three independent experiments were performed, and 1000 cells were counted for TUNEL positivity in each experiment.

Cell Cycle Analysis

Measurements of cell cycle distribution were performed by flow cytometric analysis. Briefly, MDA-MB-231 cells with MRE11 transient knockdown or MCF-7 cells with stable MRE11 overexpression were plated on six-well plates (Corning Life Sciences, Corning, NY). At 80% to 90% confluence, the cells were treated with trypsin-EDTA, fixed in 75% ethanol (Sigma, St Louis, MO), and stored at −20 ° C for at least 30 minutes. Before analysis, cells were permeabilized with TritonX-100 and stained for 30 minutes at 37°C with PBS containing 50 μg/mL propidium iodide and 4 kU/mL RNase (BD Biosciences, San Jose, CA). The cellular DNA content was measured using the FACScan flow cytometer (BD Biosciences), and cell cycle distribution was calculated using the Winmdi 2.8 software (J. Trotter, Scripps Research Institute, La Jolla, CA). Three independent experiments were performed.

Immunoblotting Analysis

Immunoblotting was performed as described in a previous article ( 30 ). The chemiluminescent signal was captured by UVP BioSpectrum 500 imaging system (UVP LLC, Upland, CA). The following polyclonal antibodies were used: goat anti-human MRE11 (1:1000 dilution), rabbit anti-human RAD50 (mouse polyclonal, 1:1000), and pSTAT3 Ser-727 (rabbit polyclonal, 1:1000) were obtained from Santa Cruz Biotechnology (Santa Cruz, CA); rabbit anti-human CDK6 (1:1000 dilution), rabbit anti-GFP (1:1000 dilution), mouse anti-human STAT3 (1:1000 dilution), chicken anti-human c-MYC (1:1000 dilution), rabbit anti-human ACTB (1:2000 dilution), and HRP-conjugated goat anti-rabbit and anti-mouse antibodies were obtained from Genetex (Irvin, CA). The following monoclonal antibodies were used: mouse anti-human CDK4 (1:1000 dilution) and rabbit anti-human pSTAT3 Tyr-705 (1:500 dilution) were obtained from Cell Signaling Technology (Beverly, MA); rabbit anti-human BCL2L1 (1:1000 dilution), mouse anti-human NBS1 (1:1000 dilution), mouse anti-human RB1 (1:1000 dilution), and rabbit anti-human γ-H2AX (1:1000 dilution) were obtained from Genetex (Irvin, CA).

Anchorage-Independent Soft Agar Assay

Colony formation in soft agar was performed by the following protocol ( www.lbl.gov/lifesciences/BissellLab/labprotocols/softagar.htm ) with minor modification. Briefly, MEM mixed with 0.5% agar (Sigma, St Louis, MO) was paved onto 6-cm dishes (Corning Life Sciences) and allowed to solidify. The dishes were overlaid with MRE11-overexpressing MCF-7 cells (2×10 4 , 1×10 4 , or 5×10 3 cells) in MEM with 5% FBS and 0.3% agarose. Cultures were maintained for 15 days and replenished with 1mL of MEM supplemented with 5% FBS every 2–3 days. Consequently, colonies were fixed with 75% ethanol and then stained with 0.005% crystal violet (Sigma) for 1 hour. MRE11-overexpressing MCF-7 cell colonies (>1mm in diameter, about 50 cells) were counted using a light microscope (Nikon, Tokyo, Japan) ( 31 ). Three independent experiments were performed.

In Vivo Orthotopic Breast Tumor Growth in a Mouse Model

To study the effect of MRE11 overexpression on xenograft tumor growth, a 1.7-mg β-estradiol pellet with 60-day release (Innovative Research of America, Sarasota, FL) was implanted subcutaneously in each female nonobese diabetic/severe combined immunodeficient (NOD/SCID) mouse aged 6–8 weeks (National Laboratory Animal Center, Tainan, Taiwan) to support the MCF-7 tumor growth (n = 16 mice). A power calculation was not done to determine the sample size, and the primary objective was to compare tumor sizes between the two groups and determine the time course of tumor development. NOD/SCID mice (n = 8 per group) were randomly divided into two groups at 1 week after the implantation of β-estradiol pellet. Vector control or MRE11-overexpressing MCF-7 cells (8×10 6 ) were suspended in matrigel (BD Biosciences, San Jose, CA) and injected into the mammary fat pads of the mice. Tumor sizes were measured every week by one author (CYH), who was blinded to the groups and randomly chose mice from control or experimental groups. The tumor volumes were calculated according to a standard formula (width 2 × length/2). Mice were killed by cervical dislocation after week 8, following the terms of the original protocol.

All experiments using mice were performed according to the guidelines of the Animal Committee and with ethics approval from the institutional review board of E-Da Hospital/I-Shou University, Taiwan.

Wound-Healing Assays

MDA-MB-231 cells with MRE11 transient knockdown and MCF-7 cells with stable MRE11 overexpression (1.5×10 5 cells per well in 12-well culture plate for each cell line) were plated on the coverslips placed in 12-well culture plates in their respective complete culture medium and grown to confluence overnight. A wound was made by scraping with a sterilized 200-mL pipette tip in the middle of the cell monolayer, and the cells were allowed to migrate to the denuded area for 24 hours for MDA-MB-231 cells or 48 hours for MCF-7 cells. Cell migration was visualized by magnification and photographed using Nikon Eclipse 80i (Nikon, Tokyo, Japan) light microscope. The wound area was measured by the program Image J software (NIH, Bethesda, MD). The percentage of wound closure was estimated by the following equation: Wound closure % = [1 − (wound area at Tt /wound area at To )] × 100%, where Tt is the time after wounding and To is the time immediately after wounding. Three independent experiments were performed.

Transwell Invasion Assays

In vitro cancer cell invasion assay was performed using a 12-well transwell insert with 8-μm pore size and coated with matrigel (Corning Life Sciences, Corning, NY). The upper wells of the inserts were coated with 100-μL Matrigel (1mg/mL) (BD Bioscience, San Jose, CA) and polymerized at 37°C for 4 hours. The coating was then rinsed once with serum-free medium. MDA-MB-231 cells with MRE11 transient knockdown or MCF-7 cells with stable MRE11 overexpression (1×10 5 cells) in 0.2-mL serum-free medium were added to the upper well; the bottom chamber contained growth medium with 10% FBS. After incubation for 48 hours, cells in the top well were removed by peeling off the matrigel and swiping the top of the membrane with cotton swabs. Cells on the underside of the membrane were stained with 0.4g/L crystal violet (Sigma, St Louis, MO) and counted by light microscopy based on five-field digital images taken randomly at 200× magnification. Three independent experiments were performed.

Gelatin Zymography Analysis

Gelatin zymography is mainly used for the detection of two matrix metalloproteinases (MMPs; also known as gelatinases), MMP-2 and MMP-9 ( 32 ). MDA-MB-231 cells with MRE11 transient knockdown or MCF-7 cells with stable MRE11 overexpression were plated in 24-well plates (5×10 5 cells per well) in 0.5mL of growth medium and allowed to attach at 37°C. Conditioned media were collected at 48 hours after incubation, cleared by centrifugation at 13 000 g for 10 minutes at 4°C, and stored at −80 o C until gelatin zymography analysis was performed. Cell number was determined after staining with 0.4g/L crystal violet. Briefly, media samples were electrophoresed on a 7.5% SDS-PAGE containing 0.1% gelatin obtained from porcine skin (Sigma). The volume of each medium sample analyzed was normalized based on cell number. At the end of electrophoresis, gels were washed in 2.5% Triton X-100 (Sigma) for approximately 45 minutes with a change of solution and in a reaction buffer (50mM Tris-HCl [pH 8.0], 5mM CaCl 2 , 0.02% NaN 3 ) for 30 minutes. Gels were incubated in the reaction buffer at 37 o C for about 20 hours, then stained with 0.25% Coomassie brilliant blue R-250 (Sigma) in 10% acetic acid and 30% ethanol, and destained in the same solution without the dye. Quantification of gelatinases was achieved by computerized image analysis using UVP BioSpectrum 500 Imaging System (Upland, CA). Three independent experiments were performed.

Assays After Ionizing Radiation Treatment of Breast Cancer Cells

MCF-7 cells with stable MRE11 overexpression were grown on culture plates until they reached 70%–80% confluence and then irradiated 1–20 Gy (3 Gy per minute) at room temperature using a 6-MV linear accelerator (ELEKTA-Sli, Norcross, GA) at the Department of Radiation Oncology, E-Da Hospital. The cells were then incubated at 37°C for 3 hours before performing further experiments.

Phosphatase Treatment. Phosphatase treatment of protein lysates was conducted as previously described ( 33 ). The lysates were then placed on ice and were subjected to immunoblotting analysis immediately.

Immunofluorescence Analysis of MRE11-GFP and γ-H2AX. MCF-7 cells with stable MRE11 overexpression (2×10 3 cells per well) were grown on eight-well chamber Labtek II slide (Nunc, Rochester, NY), and after 10-Gy γ-irradiation (3 Gy per minute), the medium was removed and the cells were fixed with 3% paraformaldehyde at room temperature for 20 minutes. To block nonspecific binding, cells were incubated with the blocking buffer (1% bovine serum albumin and 0.5% of Triton-X 100 [Sigma, St Louis, MO] in PBS) for 1 hour at room temperature. After washing with PBS, cells were incubated with rabbit anti-γ-H2AX antibody (dilution 1:200) (Genetex, Irvine, CA) for 1.5 hours followed by Texas Red-conjugated secondary antibody (dilution 1:500) (Santa Cruz Biotech, Santa Cruz, CA) for 45 minutes. Nuclei were stained with 4,6-diamidine-2-phenylindole, dihydrochloride (DAPI; dilution 1:1000) (Invitrogen, Carlsband, CA) for 5 minutes at room temperature. Slides were mounted with immunofluorescence mounting medium (DakoCytomation, Carpinteria, CA), and nuclear foci for MRE11-GFP and γ-H2AX were visualized under an immunofluorescence scanning microscope system (Zeiss Axioskop; Carl Zeiss Poughkeepsie, NY). For each sample in this experiment, 50–200 cells were scored. Positive cells were defined as containing three or more nuclear foci. Quantification of staining intensity was done with Adobe Photoshop 7.0 (Adobe System Incorporated, San Jose, CA) and normalized to the number of analyzed cells. Three independent experiments were performed.

Neutral Comet Assay. Irradiation-induced DSBs in MCF-7 cells with stable MRE11 overexpression were determined by neutral comet single-cell gel electrophoresis ( 34–36 ), according to the manufacturer’s protocol (Trevigen, Gaithersburg, MD), and the procedure has been described in a previous article ( 37 ). Comet tail produced by DNA damage in the individual cells was visualized under a fluorescence microscope and automatically quantified using the Comet Assay Software Project (CASP) image analysis software (CASP-1.2.2; University of Wroclaw, Poland). Tail moments with the definition of [(tail mean − head mean) × (% tail DNA/100)] were automatically calculated. Three independent experiments were performed, and more than 50 consecutive cells were quantified for each data point.

Colony-forming Assay. Following 1–10 Gy irradiation (3 Gy per minute), MCF-7 cells (2 × 10 4 cells per 35-mm culture dish) with stable MRE11 overexpression were incubated at 37°C for 14 days. The cells were then washed twice with PBS and stained with 0.1% crystal violet for 15 minutes before counting. Clusters of at least 50 cells were scored as colonies, and all the colonies on the 35-mm culture dish were counted. Three independent experiments were performed.

Statistical Analysis

Comparisons between MRE11 high-expression and low-expression groups for tumor stage, tumor grade, age at diagnosis, tumor size, lymph nodes status, recurrence, ER status, PR status, and HER2 status were done by the two-sided χ 2 test. Survival curves were generated using the Kaplan–Meier estimates, and the statistical significance of difference between curves was evaluated by the two-sided log-rank test. Furthermore, hazard ratios (HRs) and 95% confidence intervals (CIs), which were computed from univariate and multivariable Cox proportional hazards regression models (approximate proportionality was verified by visual examination of the Kaplan–Meier estimates), were used to assess associations between overall survival and clinicopathological characteristics. Two-sided Student t test was used for in vitro and mouse xenograft studies.

All of the statistical analyses were performed using the SPSS 14.0 statistical package for PC (SPSS, Chicago, IL). All P values less than .05 were considered statistically significant.

Results

Association of MRE11 Expression in Breast Cancer Tissues With Clinicopathological Characteristics and Prognosis

MRE11 expression in cancer tissues (n = 254 patients) was examined by immunohistochemistry and stratified into four scores (quartiles) for statistical analysis (score 1, n = 24 patients; score 2, n = 54 patients; score 3, n = 76 patients; and score 4, n =100 patients). Based on the scores, tissues were further categorized into high (scores 3 and 4; 69.3%) and low (scores 1 and 2; 30.7%) expression groups ( Figure 1 , A and Table 1 ). When compared with the adjacent normal mammary gland tissues, cancer tissues had a low nuclear MRE11 expression ( Figure 1 , A ), which was consistent with previous studies ( 38–40 ).

Figure 1.

Association between MRE11 expression levels in breast cancer tissues and patient survival. Overall survival according to MRE11 expression levels and ER status were determined by Kaplan–Meier survival curves. The hazard ratios and 95% confidence intervals were calculated using a multivariable-adjusted Cox model, and the P values were calculated using a two-sided log-rank test. A ) Nuclear immunohistochemical staining of MRE11 in breast cancer tissues. Staining was scored according to the percentage of positively stained cells in four quantitative categories (score 1, ≤25% positive cells; score 2, 26%–50% positive cells; score 3, 51%–75% positive cells; and score 4, ≥76% positive cells). The expression of MRE11 in the matched adjacent normal breast tissues and noncancerous tissue (negative control) is also shown. Images are representative of triplicate samples. MRE11 staining scores 1 and 2 were categorized as low expression, and scores 3 and 4 were categorized as high expression. Scale bar = 50 μm. B ) Overall survival analysis in patients according to MRE11 staining scores. C ) Overall survival analysis in patients according to MRE11 high and low expression levels. D ) Overall survival analysis in patients according to ER status. E ) Overall survival analysis in patients according to combinations of MRE11 expression and ER status. HR = hazard ratio; CI = confidence interval; ER = estrogen receptor.

Figure 1.

Association between MRE11 expression levels in breast cancer tissues and patient survival. Overall survival according to MRE11 expression levels and ER status were determined by Kaplan–Meier survival curves. The hazard ratios and 95% confidence intervals were calculated using a multivariable-adjusted Cox model, and the P values were calculated using a two-sided log-rank test. A ) Nuclear immunohistochemical staining of MRE11 in breast cancer tissues. Staining was scored according to the percentage of positively stained cells in four quantitative categories (score 1, ≤25% positive cells; score 2, 26%–50% positive cells; score 3, 51%–75% positive cells; and score 4, ≥76% positive cells). The expression of MRE11 in the matched adjacent normal breast tissues and noncancerous tissue (negative control) is also shown. Images are representative of triplicate samples. MRE11 staining scores 1 and 2 were categorized as low expression, and scores 3 and 4 were categorized as high expression. Scale bar = 50 μm. B ) Overall survival analysis in patients according to MRE11 staining scores. C ) Overall survival analysis in patients according to MRE11 high and low expression levels. D ) Overall survival analysis in patients according to ER status. E ) Overall survival analysis in patients according to combinations of MRE11 expression and ER status. HR = hazard ratio; CI = confidence interval; ER = estrogen receptor.

Next we examined the associations of MRE11 expression with clinicopathological characteristics of malignant cancer behavior. Patients were followed for a median period of 40 months (range = 5–68 months). Increasing scores of MRE11 expression were associated with increasingly worse survival, and patients in the score 4 group had the worst survival compared with patients in the score 1 group (HR of death = 3.96, 95% CI = 0.48 to 32.7, P = .005) ( Figure 1 , B ). High MRE11 expression was statistically significantly associated with a number of clinicopathological variables including cancer stage ( P = .004), tumor grade ( P = .026), tumor size ( P < .001), lymph node metastasis ( P < .024), and higher recurrence rates after radiotherapy ( P < .010) and chemotherapy ( P < .035) ( Table 1 ).

Next we performed a univariate Cox regression analysis to assess associations of clinicopathological characteristics with overall survival and understand the associations of MRE11 expression levels with survival better. Tumor stage (stage III and IV vs stage I, HR of death = 17.96, 95% CI = 2.35 to 137.29, P = .005; stage II vs stage I, HR of death = 9.88, 95% CI = 1.25 to 78.08, P = .030), ER status (ER vs ER + , HR of death = 5.65, 95% CI = 2.23 to 14.33, P < .001), PR status (PR vs PR + , HR of death = 4.64, 95% CI = 1.72 to 12.50, P = .002), and MRE11 status (high vs low expression, HR of death = 4.02, 95% CI = 1.19 to 13.58, P = .025) were statistically significantly associated with overall survival (Table 2). Similar results were observed for disease-free survival by univariate analysis ( Supplementary Table 1 , available online).

In a multivariable Cox regression analysis (adjusted for stage, age, ER and PR status, and MRE11 expression), tumor stage (stages III and IV vs stage I, adjusted HR of death = 22.76, 95% CI = 2.90 to 178.55, P = .003; stage II, adjusted HR of death = 7.78, 95% CI = 0.98 to 62.06, P = .053), ER status (adjusted HR of death = 4.90, 95% CI = 1.46 to 16.45, P = .010), and MRE11 expression (adjusted HR of death = 4.43, 95% CI = 1.21 to 16.25, P = .025) were statistically significantly associated with overall survival (Table 2). Similar results were observed for the analysis of disease-free survival by multivariable analysis ( Supplementary Table 1 , available online).

We further analyzed the associations of overall survival with high vs low MRE11 expression and according to ER status by Kaplan–Meier survival curves. Patients in the high MRE11 expression group had a statistically significantly poorer overall survival than patients in the low MRE11 expression group (adjusted HR of death = 4.43, 95% CI = 1.21 to 16.25, P = .015) ( Figure 1 , C ), and ER group had a poorer overall survival than ER + group (HR of death = 4.90, 95% CI = 1.46 to 16.45, P <. 001), respectively ( Figure 1 , D ). We further analyzed the combined association of MRE11 expression and ER status with overall survival by Kaplan–Meier survival curves. Patients with both ER status and MRE11 high expression had the worst overall survival (MRE11 high and ER vs MRE11 low and ER + , HR of death = 6.91, 95% CI = 1.57 to 30.47, P < .001) ( Figure 1 , E ). Similar results for disease-free survival were observed ( Supplementary Figure 1 , available online).

Next we studied the effect of MRE11 expression on patient survival after adjuvant radiotherapy, chemotherapy, and hormone therapy by Kaplan–Meier survival curves. Although MRE11 high expression in cancer tissues was associated with worse survival rates after radiotherapy or chemotherapy ( P = .012 and P = .031, respectively), it had no effect on patient survival after hormone therapy ( P = .19) ( Figure 2 ). Similar outcomes for disease-free survival after adjuvant radiotherapy, chemotherapy, and hormone therapy were also observed ( Supplementary Figure 2 , available online). We also analyzed the combined association of radiotherapy and chemotherapy on disease-free survival and overall survival by Kaplan–Meier survival curves. The results showed that after combined radiotherapy and chemotherapy MRE11 high expression in cancer tissues was associated with worse disease-free and overall survival rates ( Supplementary Figure 3 , available online).

Figure 2.

Overall survival of breast cancer patients in the MRE11 low and high expression groups according to adjuvant treatment. We included 78 patients in the MRE11 low group and 176 patients in the MRE11 high group. Analysis was done using Kaplan–Meier method. The hazard ratios and 95% confidence intervals were calculated using a multivariable-adjusted Cox model, and the P values were calculated using a two-sided log-rank test. A ) Survival analysis according to administration of adjuvant radiotherapy. B ) Survival analysis according to administration of adjuvant chemotherapy. C ) Survival analysis according to administration of adjuvant hormone therapy. HR = hazard ratio; CI = confidence interval.

Figure 2.

Overall survival of breast cancer patients in the MRE11 low and high expression groups according to adjuvant treatment. We included 78 patients in the MRE11 low group and 176 patients in the MRE11 high group. Analysis was done using Kaplan–Meier method. The hazard ratios and 95% confidence intervals were calculated using a multivariable-adjusted Cox model, and the P values were calculated using a two-sided log-rank test. A ) Survival analysis according to administration of adjuvant radiotherapy. B ) Survival analysis according to administration of adjuvant chemotherapy. C ) Survival analysis according to administration of adjuvant hormone therapy. HR = hazard ratio; CI = confidence interval.

Figure 3.

Effect of MRE11 knockdown and overexpression on cell viability, cell proliferation, apoptosis, and STAT3 signaling. A ) Effect of MRE11 knockdown on cell viability. MDA-MB-231 cells with MRE11 transient knockdown and vector control were stained with trypan blue, and the number of viable cells was counted using an automated cell counter. An XTT colorimetric assay was also done using these cells to assess cell viability by measuring the optical density at 490 nm. The results were expressed as percentage of vector control. B ) Effect of MRE11 knockdown on apoptosis. MDA-MB-231 cells with MRE11 transient knockdown and vector control were stained with annexin V and TUNEL, and 1000 cells were counted for positivity for each stain. C ) Effect of MRE11 knockdown on cell cycle distribution. Flow cytometry analysis of MDA-MB-231 cells with MRE11 transient knockdown or vector control, 3 days after transfection, to determine the percentage of cells in G0/G1, S, and G2/M phases. In panels A, B, and C, the mean number of viable cells and 95% confidence intervals ( error bars ) from three independent experiments are shown. P values were calculated using a two-sided Student t test. D ) Effect of MRE11 knockdown on the expression of regulatory proteins involved in G1 to S transition. Immunoblot analysis of MDA-MB-231 cells with MRE11 transient knockdown or vector control, 3 days after transfection. The blot is representative of three independent experiments. E ) Effect of MRE11 overexpression on cell proliferation. MCF-7 cells were stably transfected with MRE11-GFP or vector control. Stable clones with relatively high, moderate, and low expression of MRE11-GFP were selected for the XTT assay. Mean OD values and 95% confidence intervals ( error bars ) from three independent experiments are shown. P values were calculated using a two-sided Student t test. F ) Effect of MRE11 overexpression on cell cycle distribution. Flow cytometry analysis of MCF-7 cells stably transfected with MRE11-GFP (clone with MRE11 overexpression) or vector control for distributions in G0/G1, S, and G2/M phases. Mean values and 95% confidence intervals ( error bars ) from three independent experiments are shown. P values were calculated using a two-sided Student t test. G ) Effect of MRE11 overexpression on the expression of regulatory proteins involved in G1 to S transition. Immunoblot analysis of MCF-7 cells with MRE11-GFP overexpression was done. Two stable clones transfected with empty vector are shown as vector I and vector II. The blot is representative of three independent experiments. H ) Effect of MRE11 overexpression on STAT3 signaling. MCF-7 cells with MRE11-GFP overexpression or vector control were treated with 10 µM Stattic, a small molecule inhibitor of STAT3 activation, for 24 hours, and immunoblot analysis was done to determine the level of total and phosphorylated STAT3 and its downstream effectors. ACTB was used as the protein loading control. I ) Effect of Stattic on cell viability. MCF-7 cells with MRE11-GFP overexpression or vector control were treated with 10 µM Stattic for 24 hours, and cell viability was compared with no Stattic treatment (as 100%) by XTT assay. Mean values and 95% confidence intervals ( error bars ) from three independent experiments are shown. P values were calculated using a two-sided Student t test. KD = knockdown.

Figure 3.

Effect of MRE11 knockdown and overexpression on cell viability, cell proliferation, apoptosis, and STAT3 signaling. A ) Effect of MRE11 knockdown on cell viability. MDA-MB-231 cells with MRE11 transient knockdown and vector control were stained with trypan blue, and the number of viable cells was counted using an automated cell counter. An XTT colorimetric assay was also done using these cells to assess cell viability by measuring the optical density at 490 nm. The results were expressed as percentage of vector control. B ) Effect of MRE11 knockdown on apoptosis. MDA-MB-231 cells with MRE11 transient knockdown and vector control were stained with annexin V and TUNEL, and 1000 cells were counted for positivity for each stain. C ) Effect of MRE11 knockdown on cell cycle distribution. Flow cytometry analysis of MDA-MB-231 cells with MRE11 transient knockdown or vector control, 3 days after transfection, to determine the percentage of cells in G0/G1, S, and G2/M phases. In panels A, B, and C, the mean number of viable cells and 95% confidence intervals ( error bars ) from three independent experiments are shown. P values were calculated using a two-sided Student t test. D ) Effect of MRE11 knockdown on the expression of regulatory proteins involved in G1 to S transition. Immunoblot analysis of MDA-MB-231 cells with MRE11 transient knockdown or vector control, 3 days after transfection. The blot is representative of three independent experiments. E ) Effect of MRE11 overexpression on cell proliferation. MCF-7 cells were stably transfected with MRE11-GFP or vector control. Stable clones with relatively high, moderate, and low expression of MRE11-GFP were selected for the XTT assay. Mean OD values and 95% confidence intervals ( error bars ) from three independent experiments are shown. P values were calculated using a two-sided Student t test. F ) Effect of MRE11 overexpression on cell cycle distribution. Flow cytometry analysis of MCF-7 cells stably transfected with MRE11-GFP (clone with MRE11 overexpression) or vector control for distributions in G0/G1, S, and G2/M phases. Mean values and 95% confidence intervals ( error bars ) from three independent experiments are shown. P values were calculated using a two-sided Student t test. G ) Effect of MRE11 overexpression on the expression of regulatory proteins involved in G1 to S transition. Immunoblot analysis of MCF-7 cells with MRE11-GFP overexpression was done. Two stable clones transfected with empty vector are shown as vector I and vector II. The blot is representative of three independent experiments. H ) Effect of MRE11 overexpression on STAT3 signaling. MCF-7 cells with MRE11-GFP overexpression or vector control were treated with 10 µM Stattic, a small molecule inhibitor of STAT3 activation, for 24 hours, and immunoblot analysis was done to determine the level of total and phosphorylated STAT3 and its downstream effectors. ACTB was used as the protein loading control. I ) Effect of Stattic on cell viability. MCF-7 cells with MRE11-GFP overexpression or vector control were treated with 10 µM Stattic for 24 hours, and cell viability was compared with no Stattic treatment (as 100%) by XTT assay. Mean values and 95% confidence intervals ( error bars ) from three independent experiments are shown. P values were calculated using a two-sided Student t test. KD = knockdown.

Effect of Knockdown and Overexpression of MRE11 Expression on Cell Viability, Cell Proliferation, Apoptosis, Cell Cycle, and Signaling In Vitro

Because high MRE11 expression was associated with clinically malignant breast cancer, we further explored the effect of MRE11 expression on breast cancer cell behavior in vitro. We analyzed the expression of MRE11 in different breast cancer cell lines (T47D, MCF-7, ZR75-1, and MDA-MB-231) by immunoblotting and found that MRE11 expression is generally expressed in different breast cell lines and did not show a substantial difference in expression levels among these cell lines ( Supplementary Figure 4, A , available online). Knockdown of MRE11 expression in MCF-7, MDA-MB-231, and T47D cells was conducted using three lentivirus-based shRNA constructs; MDA-MB-231 cells and shRNA1 were chosen for further study based on silencing efficiency (data not shown).

Figure 4.

Effect of MRE11 overexpression on anchorage-independent colony information in vitro and xenograft tumor growth in vivo. A ) Anchorage-independent colony formation of MCF-7 cells carrying MRE11-GFP (low, moderate, and high expression levels) was determined by soft agar assay after culture for 15 days. Colonies were stained with crystal violet. Representative images (2×10 4 cells seeded) from three independent experiments are shown. The bar graphs represent the mean number of colonies with 95% confidence intervals ( error bars ). P values were calculated using a two-sided Student t test. Scale bar = 30 μm. B ) An orthotopic breast tumor model was developed by injecting female nonobese diabetic/severe combined immunodeficient (NOD/SCID) mice with MCF-7 cells with MRE11-GFP overexpression or vector only (n = 8 mice per group). Tumor volumes were calculated using a standard formula. Mean tumor volumes with 95% confidence intervals ( error bars ) are shown. P values were calculated using a two-sided Student t test.

Figure 4.

Effect of MRE11 overexpression on anchorage-independent colony information in vitro and xenograft tumor growth in vivo. A ) Anchorage-independent colony formation of MCF-7 cells carrying MRE11-GFP (low, moderate, and high expression levels) was determined by soft agar assay after culture for 15 days. Colonies were stained with crystal violet. Representative images (2×10 4 cells seeded) from three independent experiments are shown. The bar graphs represent the mean number of colonies with 95% confidence intervals ( error bars ). P values were calculated using a two-sided Student t test. Scale bar = 30 μm. B ) An orthotopic breast tumor model was developed by injecting female nonobese diabetic/severe combined immunodeficient (NOD/SCID) mice with MCF-7 cells with MRE11-GFP overexpression or vector only (n = 8 mice per group). Tumor volumes were calculated using a standard formula. Mean tumor volumes with 95% confidence intervals ( error bars ) are shown. P values were calculated using a two-sided Student t test.

Next we examined the effect of MRE11 knockdown on cell viability using trypan blue staining and XTT assay. We observed a statistically significantly reduced number of viable MDA-MB-231 cells when MRE11 expression was inhibited compared with vector control (trypan blue stain, control vs MRE11 knockdown, mean [number of viable cells] = 2.0×10 5 , 95% CI = 1.7×10 5 to 2.4×10 5 vs mean = 5.4×10 4 , 95% CI = 2.9×10 4 to 8.0×10 4 , P = .010; XTT assay, control vs MRE11 knockdown, mean [% of control] = 100%, 95% CI = 94.0% to 105.9% vs mean = 45.3%, 95% CI = 38.7% to 52.1%, P <.001) ( Figure 3 , A ). Annexin-V and TUNEL assays showed a statistically significant increase in apoptosis in MDA-MB-231 cells with MRE11 knockdown compared with vector control (annexin-V stain, P = .001; TUNEL, P = .023) ( Figure 3 , B ).

The effect of MRE11 silencing on cell cycle distribution was examined by flow cytometry. Knockdown of MRE11 showed a statistically significant increase in cell cycle arrest at the G0/G1 phase ( P = .003) and decreased number of cells at the S phase ( P = .004) compared with vector control ( Figure 3 , C ). Therefore, we examined the expression of regulatory proteins required for G1 to S transition by immunoblotting. The levels of phosphorylated RB1 and total CCND1, CDK4, and CDK6 were reduced in MRE11-silenced MDA-MB-231 cells ( Figure 3 , D ).

Because high expression of MRE11 in patient tissues showed a statistically significant association with malignant cancer behavior, we generated stably overexpressed MRE11-GFP construct in MCF-7 cells, which is a less malignant breast cancer cell line than MDA-MB-231, to further study the biological effects of MRE11. Quantitative assessment of cell proliferation by XTT assay revealed a statistically significant increased proliferation at day 3 in cells with high, moderate, and low levels of MRE11-GFP expression compared with vector control (GFP only) (high MRE11 vs vector control, mean [OD 490–650 nm ] = 0.452, 95% CI = 0.445 to 0.458 vs mean [OD 490–650 nm ] = 0.368, 95% CI = 0.359 to 0.377, P < .001; moderate MRE11, P = .001; and low MRE11, P = .016) ( Figure 3 , E ). A statistically significant decrease in G0/G1 phase for high and moderate MRE11-GFP expression (high, P = .013; moderate, P = .040) and a statistically significant increase in S phase (high, P = .046; moderate, P = .049) and G2/M phase (high, P = .038) were observed compared with vector control ( Figure 3 , F ). We examined the expression levels of G1-S transition proteins by immunoblotting analysis and found increased levels of phosphorylated RB1, and total CDK4, CDK6, and CCND1 in high MRE11-GFP-expressing MCF-7 cells compared with vector controls ( Figure 3 , G ).

We also tested the effect of MRE11 overexpression on cell viability using MDA-MB-231 cells transiently transfected with MRE11 construct. Overexpression was confirmed by immunoblotting, and an XTT assay showed that cell viability after transfection for 3 days was statistically significantly increased compared with vector control ( P = .005) ( Supplementary Figure 4, B , available online). We also examined the effect of MRE11 knockdown on MCF-7 cell viability by transient transfection of lentivirus-based MRE11 shRNA construct. Knockdown of MRE11 expression was confirmed by immunoblotting, and an XTT assay showed that MCF-7 cell viability after transfection for 3 days was statistically significantly decreased compared with their respective control vector ( P = .016) ( Supplementary Figure 4, C , available online). The results indicate that MRE11 may have similar effects on other breast cancer cells.

Activated STAT3 signaling promotes cell proliferation and prevents apoptosis ( 41 ), and treatment with Stattic, a small-molecule inhibitor of STAT3 activation, induces apoptotic cell death ( 29 ). In this study, we analyzed the effect of MRE11 overexpression on STAT3 activity by immunoblotting analysis. MRE11-GFP overexpression in MCF-7 cells resulted in an increase in of the level of phosphorylated STAT3 at tyrosine-705 and serine-727 residues ( Figure 3 , H ). It is known that c-MYC, CCND1, and BCL2L1 are direct transcriptional targets of STAT3 ( 41 ). These downstream targets were overexpressed in cells expressing high and moderate levels of MRE11 compared with low levels of MRE 11 and vector control ( Figure 3 , H ). When treated with Stattic, the level of STAT3 phosphorylation at both tyrosine-705 and serine-727 residues and the total level of c-MYC, CCND1, and BCL2L1 were reduced ( Figure 3 , H ). In an XTT cell viability assay, treatment with Stattic statistically significantly suppressed cell viability in MCF-7 cells with high MRE11-GFP expression (cell viability [%] after Stattic treatment vs no treatment [as 100%], high MRE11 vs vector, mean = 27.4%, 95% CI = 22.8% to 32.0% vs mean = 46.6%, 95% CI = 40.8% to 52.4%, P = .025) and moderate MRE11-GFP expression ( P = .044) compared with vector control ( Figure 3 , I ).

Effect of MRE11 Overexpression on Anchorage-Independent Growth In Vitro and Tumor Growth in Vivo in Mice

In an anchorage-independent assay for colony formation in vitro, MRE11-overexpressing MCF-7 cells had an increased colony formation in soft agar compared with vector control (high MRE11 vs vector at day 15, 2×10 4 cells seeded, mean number of colonies = 1448, 95% CI = 1236 to 1659, vs mean number of colonies = 246, 95% CI = 184 to 309, P < .001) ( Figure 4 , A ).

To further confirm the association of MRE11 expression with tumor formation in vivo, we used a NOD/SCID mouse xenograft model where MCF-7 cells with high MRE11 expression or vector control were injected into the mammary fat pads of female mice (n = 8 per group) for growth of orthotopic breast tumors. Tumor growth was statistically significantly increased in the high MRE11 expression group compared with vector control (tumor volume at week 7, control vs MRE11-overexpressing tumor, mean = 147.7mm 3 , 95% CI = 23.5 to 271.9mm 3 vs mean = 434.9mm 3 , 95% CI = 265.1 to 604.8mm 3 , P = 0.018; tumor volume at week 8, control vs MRE11-overexpressing tumor, mean = 280.4mm 3 , 95% CI = 62.4 to 498.4mm 3 vs mean = 631.0mm 3 , 95% CI = 296.9 to 965.0mm 3 , P = .043) ( Figure 4 , B ).

Effect of MRE11 Knockdown on Cell Migration and Invasion In Vitro

Directed migration of cells in vitro, measured in a wound-healing assay as the percentage of wound closure 24 hours after the wound was made, showed a reduced motility in MDA-MB-231 cells with MRE11 knockdown compared with that of vector control cells (mean wound closure = 16.7%, 95% CI = 13.6% to 19.8% vs mean wound closure = 75.4%, 95% CI = 69.1% to 81.7%, P < .001) ( Figure 5 , A ). In matrigel-covered transwell assay, MDA-MB-231 cells with MRE11 knockdown showed attenuated cell invasion after 48 hours of incubation compared with vector control cells (mean number of cells = 1850, 95% CI = 1484 to 2216 vs mean number of cells = 5180, 95% CI = 4288 to 6072, P = .002) ( Figure 5 , B ).

Figure 5.

Effect of MRE11 knockdown or overexpression on cell migration and invasion in vitro. A ) Effect of MRE11 knockdown on wound healing. MDA-MB-231 cells with MRE11 transient knockdown or vector control were grown in a monolayer, a wound was made by scraping with a pipette tip in the middle of the cell monolayer, and cells were allowed to migrate to the denuded area for 24 hours. The mean percentage of wound closure with 95% confidence intervals ( error bars ) from three independent experiments is shown. P value was calculated using a two-sided χ 2 test. Scale bar = 200 μm. B ) Effect of MRE11 knockdown on cell invasion. Invasion of MDA-MB-231 cells with MRE11 transient knockdown through transwell inserts containing matrigel-coated membranes was assessed 48 hours after incubation. The mean number of invasive cells with 95% confidence intervals ( error bars ) from three independent experiments is shown. P value was calculated using a two-sided χ 2 test. Scale bar = 100 μm. Three independent experiments were performed. C ) Effect of MRE11 knockdown on MMP secretion. MDA-MB-231 cells with MRE11 transient knockdown were grown, and conditioned medium was collected after 48 hours to assess the level of secretion of MMP-2 and MMP-9 by gelatin zymography analysis. The secretion of MMP-2 and MMP-9 in vector cells was taken as 100%. The mean relative percentage of secretion with 95% confidence intervals ( error bars ) from three independent experiments is shown. P values were calculated using a two-sided χ 2 test. D ) Effect of MRE11 overexpression on wound healing. MCF-7 cells with MRE11-GFP overexpression or vector control were grown in a monolayer; a wound was made and cells were allowed to migrate to the denuded area for 24 hours. The mean percentage of wound closure with 95% confidence intervals ( error bars ) from three independent experiments is shown. P value was calculated using a two-sided χ 2 test. Scale bar = 200 μm. E ) Effect of MRE11 overexpression on cell invasion. Invasion of MCF-7 cells with MRE11-GFP overexpression through transwell inserts was assessed 48 hours after incubation. The mean number of invasive cells with 95% confidence intervals ( error bars ) from three independent experiments is shown. P value was calculated using a two-sided χ 2 test. Scale bar = 100 μm. F ) Effect of MRE11 overexpression on MMP secretion. MCF-7 cells with MRE11-GFP overexpression were grown, and conditioned medium was collected after 48 hours to assess the level of secretion of MMP-2 and MMP-9 by gelatin zymography analysis. The secretion of MMP-2 and MMP-9 in vector cells was taken as 100%. The mean relative percentage of secretion with 95% confidence intervals ( error bars ) from three independent experiments is shown. P values were calculated using a two-sided χ 2 test. KD = knockdown; OE = overexpression; MMP = matrix metalloproteinase.

Figure 5.

Effect of MRE11 knockdown or overexpression on cell migration and invasion in vitro. A ) Effect of MRE11 knockdown on wound healing. MDA-MB-231 cells with MRE11 transient knockdown or vector control were grown in a monolayer, a wound was made by scraping with a pipette tip in the middle of the cell monolayer, and cells were allowed to migrate to the denuded area for 24 hours. The mean percentage of wound closure with 95% confidence intervals ( error bars ) from three independent experiments is shown. P value was calculated using a two-sided χ 2 test. Scale bar = 200 μm. B ) Effect of MRE11 knockdown on cell invasion. Invasion of MDA-MB-231 cells with MRE11 transient knockdown through transwell inserts containing matrigel-coated membranes was assessed 48 hours after incubation. The mean number of invasive cells with 95% confidence intervals ( error bars ) from three independent experiments is shown. P value was calculated using a two-sided χ 2 test. Scale bar = 100 μm. Three independent experiments were performed. C ) Effect of MRE11 knockdown on MMP secretion. MDA-MB-231 cells with MRE11 transient knockdown were grown, and conditioned medium was collected after 48 hours to assess the level of secretion of MMP-2 and MMP-9 by gelatin zymography analysis. The secretion of MMP-2 and MMP-9 in vector cells was taken as 100%. The mean relative percentage of secretion with 95% confidence intervals ( error bars ) from three independent experiments is shown. P values were calculated using a two-sided χ 2 test. D ) Effect of MRE11 overexpression on wound healing. MCF-7 cells with MRE11-GFP overexpression or vector control were grown in a monolayer; a wound was made and cells were allowed to migrate to the denuded area for 24 hours. The mean percentage of wound closure with 95% confidence intervals ( error bars ) from three independent experiments is shown. P value was calculated using a two-sided χ 2 test. Scale bar = 200 μm. E ) Effect of MRE11 overexpression on cell invasion. Invasion of MCF-7 cells with MRE11-GFP overexpression through transwell inserts was assessed 48 hours after incubation. The mean number of invasive cells with 95% confidence intervals ( error bars ) from three independent experiments is shown. P value was calculated using a two-sided χ 2 test. Scale bar = 100 μm. F ) Effect of MRE11 overexpression on MMP secretion. MCF-7 cells with MRE11-GFP overexpression were grown, and conditioned medium was collected after 48 hours to assess the level of secretion of MMP-2 and MMP-9 by gelatin zymography analysis. The secretion of MMP-2 and MMP-9 in vector cells was taken as 100%. The mean relative percentage of secretion with 95% confidence intervals ( error bars ) from three independent experiments is shown. P values were calculated using a two-sided χ 2 test. KD = knockdown; OE = overexpression; MMP = matrix metalloproteinase.

A critical event in cancer cell migration and invasion is the degradation of extracellular matrix (ECM), and the expression of MMPs is necessary for ECM degradation and metastasis ( 42 , 43 ). We examined the levels of secreted metastasis-associated MMP proteins (MMP-2 and MMP-9) by gelatin zymography and found that the levels were markedly suppressed in the supernatant of MDA-MB-231 cells with MRE11 knockdown compared with the vector control ( P = .004 for MMP-2; P = .005 for MMP-9) ( Figure 5 , C ). In contrast, MRE11 overexpression promoted MCF-7 cell motility and wound closure at 48 hours after wound was made compared with vector control (mean wound closure = 68.2%, 95% CI = 52.1% to 84.4% vs mean wound closure = 38.9%, 95% CI = 28.7% to 49.1%, P = .002) ( Figure 5 , D ). MRE11-overexpressing MCF-7 cells also showed enhanced cell invasion compared with vector control cells after 48 hours of incubation (mean number of cells = 64.3, 95% CI = 46.7 to 82.0 vs mean number of cells = 41.3, 95% CI = 39.7 to 42.9, P = .047) ( Figure 5 , E ). The secretion of MMP-2 and MMP-9 after 48 hours of incubation was also increased in the supernatant of MRE11-overexpressing cells compared with vector control cells ( P = .015 for MMP-2; P = .001 for MMP-9) ( Figure 5 , F ).

Effect of MRE11 Overexpression on DNA Repair

When DSBs are generated, the MRN protein complex is recruited to the sites of the DNA lesions, which induces the activation of ataxia telangiectasia mutated, a kinase that phosphorylates several downstream targets such as histone H2AX (known as γH2AX) ( 44–46 ). Immunoblotting results showed that MDA-MB-231 cells with MRE11 knockdown had a reduced expression of RAD50 and NBS1, whereas MCF-7 cells overexpressing MRE11 had increased expression ( Supplementary Figure 5, A and B , available online). Upon γ-irradiation treatment, both MRE11-GFP and endogenous MRE11 were phosphorylated, and pretreatment with phosphatase reduced the level of phosphorylated protein, which was determined by immunoblotting ( Figure 6 , A ). We also did an immunofluorescence assay for γ-H2AX foci (an indicator of DSB severity) and MRE11-GFP repair foci after γ-irradiation treatment, and the result showed that γ-H2AX foci and MRE11-GFP foci colocalized in the nuclei after γ-irradiation ( Figure 6 , B , and Supplementary Figure 6 , available online). The absence of γ-H2AX foci at 24 hours after γ-irradiation is associated with the completeness of DSB repair ( Supplementary Figure 6 , available online). The time course of γ-H2AX foci formation in MRE11-overexpressing MCF-7 cells and vector control cells after γ-irradiation treatment was also analyzed. The γ-H2AX-positive cells increased dramatically at 15 minutes after irradiation in both MCF-7-MRE11 and control cells ( Figure 6 , C and Supplementary Figure 6 , available online). A statistically significant decrease in γ-H2AX-positive cells was observed in MRE11-GFP-overexpressing MCF-7 cells at 3, 6, and 24 hours after irradiation compared with vector control cells ( P = .001, P < .001, and P = .030, respectively) ( Figure 6 , C ). These results were consistent with γ-H2AX expression pattern by immunoblotting analysis ( Figure 6 , D ).

Figure 6.

Effect of MRE11 overexpression on DNA repair and cell survival upon ionizing radiation in vitro. A ) Effect of MRE11 overexpression on phosphorylation of MRE11. MCF-7 cells with MRE11-GFP overexpression and vector control were treated with 10 or 20 Gray (Gy) IR, and after 3 hours cells were lysed with 200 units of calf intestinal phosphatase for 2 hours. Immunoblotting analysis was done to determine the level of phosphorylated MRE11. B ) Effect of MRE11 overexpression on colocalization of IR-induced MRE11-GFP foci ( green ) and γ-H2AX foci ( red ). MCF-7 cells with MRE11-GFP overexpression were treated with 10 Gy IR and were immunostained for γ-H2AX. The foci for MRE11-GFP and γ-H2AX were observed under fluorescence microscope. The localization of MRE11-GFP and γ-H2AX in the nucleus of MRE11-GFP-overexpressing MCF-7 cells is shown. Scale bar = 10 μm. C ) Effect of MRE11 overexpression on the dynamic change of colocalization of MRE11-GFP foci and γ-H2AX foci. MCF-7 cells with MRE11-GFP overexpression and vector control were treated with 10 Gy IR and immunostained for γ-H2AX, and the foci for γ-H2AX were analyzed. Mean γ-H2AX foci positive cells with 95% confidence intervals ( error bars ) from three independent experiments are shown. P values were calculated using a two-sided Student t test. D ) Effect of MRE11 overexpression on the kinetic change of γ-H2AX level. MCF-7 cells with MRE11-GFP overexpression and vector control were treated with 10 Gy IR, and immunoblot analysis was done to determine the level of γ-H2AX. The blot is representative of three independent experiments. ACTB was used as the protein-loading control. E ) Effect of MRE11 overexpression on the kinetic change of DNA comet tail moments (DNA double-strand breaks). MCF-7 cells with MRE11-GFP overexpression and vector control were treated with 10 Gy IR. Comet tail moments were determined by neutral comet single-cell gel electrophoresis and observed under a fluorescence microscope. At least 50 consecutive cells were quantified per sample. Representative images of cells without or with comet tail are shown. Mean tail moments with 95% confidence intervals ( error bars ) from three independent experiments are shown. P values were calculated using a two-sided Student t test. Scale bar = 25 μm. F ) Effect of MRE11 overexpression on apoptosis. TUNEL staining of MCF-7 cells with MRE11 overexpression and vector control was analyzed at 72 hours after 10 Gy IR exposure. One-thousand cells were counted for positivity for each stain. The mean percentage of TUNEL-positive cells with 95% confidence intervals ( error bars ) from three independent experiments is shown. P values were calculated using a two-sided Student t test. G ) Effect of MRE11 overexpression on clonogenic survival. The colony-forming efficiency of MCF-7 cells with MRE11 overexpression and vector control was analyzed at 14 days after IR exposure. Clusters of at least 50 cells were scored as colonies, and all the colonies on the culture dish were counted. Representative images of MRE11 overexpression and vector control by colony-forming assay were shown. Mean survival time with 95% confidence intervals ( error bars ) from three independent experiments is shown. P values were calculated using a two-sided Student t test. OE = overexpression; p = phosphorylated; IR = ionizing irradiation.

Figure 6.

Effect of MRE11 overexpression on DNA repair and cell survival upon ionizing radiation in vitro. A ) Effect of MRE11 overexpression on phosphorylation of MRE11. MCF-7 cells with MRE11-GFP overexpression and vector control were treated with 10 or 20 Gray (Gy) IR, and after 3 hours cells were lysed with 200 units of calf intestinal phosphatase for 2 hours. Immunoblotting analysis was done to determine the level of phosphorylated MRE11. B ) Effect of MRE11 overexpression on colocalization of IR-induced MRE11-GFP foci ( green ) and γ-H2AX foci ( red ). MCF-7 cells with MRE11-GFP overexpression were treated with 10 Gy IR and were immunostained for γ-H2AX. The foci for MRE11-GFP and γ-H2AX were observed under fluorescence microscope. The localization of MRE11-GFP and γ-H2AX in the nucleus of MRE11-GFP-overexpressing MCF-7 cells is shown. Scale bar = 10 μm. C ) Effect of MRE11 overexpression on the dynamic change of colocalization of MRE11-GFP foci and γ-H2AX foci. MCF-7 cells with MRE11-GFP overexpression and vector control were treated with 10 Gy IR and immunostained for γ-H2AX, and the foci for γ-H2AX were analyzed. Mean γ-H2AX foci positive cells with 95% confidence intervals ( error bars ) from three independent experiments are shown. P values were calculated using a two-sided Student t test. D ) Effect of MRE11 overexpression on the kinetic change of γ-H2AX level. MCF-7 cells with MRE11-GFP overexpression and vector control were treated with 10 Gy IR, and immunoblot analysis was done to determine the level of γ-H2AX. The blot is representative of three independent experiments. ACTB was used as the protein-loading control. E ) Effect of MRE11 overexpression on the kinetic change of DNA comet tail moments (DNA double-strand breaks). MCF-7 cells with MRE11-GFP overexpression and vector control were treated with 10 Gy IR. Comet tail moments were determined by neutral comet single-cell gel electrophoresis and observed under a fluorescence microscope. At least 50 consecutive cells were quantified per sample. Representative images of cells without or with comet tail are shown. Mean tail moments with 95% confidence intervals ( error bars ) from three independent experiments are shown. P values were calculated using a two-sided Student t test. Scale bar = 25 μm. F ) Effect of MRE11 overexpression on apoptosis. TUNEL staining of MCF-7 cells with MRE11 overexpression and vector control was analyzed at 72 hours after 10 Gy IR exposure. One-thousand cells were counted for positivity for each stain. The mean percentage of TUNEL-positive cells with 95% confidence intervals ( error bars ) from three independent experiments is shown. P values were calculated using a two-sided Student t test. G ) Effect of MRE11 overexpression on clonogenic survival. The colony-forming efficiency of MCF-7 cells with MRE11 overexpression and vector control was analyzed at 14 days after IR exposure. Clusters of at least 50 cells were scored as colonies, and all the colonies on the culture dish were counted. Representative images of MRE11 overexpression and vector control by colony-forming assay were shown. Mean survival time with 95% confidence intervals ( error bars ) from three independent experiments is shown. P values were calculated using a two-sided Student t test. OE = overexpression; p = phosphorylated; IR = ionizing irradiation.

DSB repair activity was also assessed by the neutral comet assay; this assay detects DNA DSBs and fragmented DNA and presents a “comet” appearance of the nuclei when exposed to an electric field ( 35–37 ). The tail moment (ie, length) is proportional to the DNA damage. The mean tail moments of MRE11-GFP-overexpressing cells and vector alone cells were similar at 15 minutes after γ-irradiation ( Figure 6 , E ). The mean tail moments of MRE11-GFP-overexpressing cells were statistically significantly reduced compared with those of vector control cells at 3, 6, and 24 hours after irradiation ( P < .001, P = .002, and P < .001, respectively) ( Figure 6 , E ).

The dynamics and colocalization of γ-H2AX and MRE11-GFP in MRE11-GFP-overexpressing cells upon γ-irradiation is shown in Supplementary Figure 6 (available online). Weak punctate MRE11-GFP and γ-H2AX foci began to form and colocalize at 15 minutes, and they gradually increased at 1 and 3 hours after irradiation. Although MRE11-GFP foci continued to exist and their colocalization with γ-H2AX foci reached a maximum at 6 hours after irradiation, a decrease of γ-H2AX foci also started at this time point. At 24 hours after irradiation, homogenous MRE11-GFP distribution and complete absence of γ-H2AX were observed.

Next we analyzed apoptosis by TUNEL staining 3 days after 10-Gy γ-irradiation. We observed that apoptosis was statistically significantly lower in MRE11-GFP-overexpressing MCF-7 cells compared with vector control cells (mean TUNEL-positive cells = 27.4%, 95% CI = 22.2% to 32.7% vs mean TUNEL-positive cells = 41.8%, 95% CI = 36.4% to 47.3%, P = .001) ( Figure 6 , F ).

Furthermore, we assessed the effect of MRE11 overexpression on radiosensitivity or radioresistance of MCF-7 cells in vitro. The cell survival curves determined by colony-forming assay demonstrated that MRE11-GFP-overexpressing MCF-7 cells exhibited higher colony formation (indicating radioresistance) with increasing dose of radiation compared with vector control cells ( Figure 6 , G ). A colony survival analysis showed a statistically significantly increased survival of MRE11 overexpressing cells compared with control (at 1-Gy γ-irradiation: control vs MRE11 overexpressing, mean = 69.4%, 95% CI = 66.5% to 72.3% vs mean = 89.2%, 95% CI = 80.7% to 97.6%, P = .012; at 5-Gy γ-irradiation: control vs MRE11 overexpressing, mean = 7.2%, 95% CI = 5.0% to 9.4% vs mean = 29.7, 95% CI = 21.6% to 37.7%, P = .006; at 10-Gy γ-irradiation: control vs MRE11 overexpressing, mean = 1.7%, 95% CI = 0.8% to 2.5% vs mean = 4.9%, 95% CI = 3.0% to 6.8%, P = .004). The results suggest that MRE11 overexpression contributed to DNA damage repair in irradiated cells.

Discussion

Analysis of clinical samples in this study showed that high MRE11 expression in breast cancer cells was associated with malignant behavior; higher levels of MRE11 expression were associated with higher recurrence rates, lymph node metastasis, resistance to radiotherapy and chemotherapy, and decreased patient survival, compared with lower levels of expression. In vitro studies using breast cancer cell lines showed that MRE11 overexpression promoted cell proliferation, cell cycle entry (S and G2/M phases), invasion and migration, and resistance to ionizing radiation via enhanced DNA repair activity. We also found that activation of STAT3 was involved in MRE11-mediated increased cell proliferation. MRE11 knockdown in breast cancer cells suppressed cell proliferation, induced cell cycle arrest, and showed increased apoptosis. Increased tumor growth was observed in an orthotopic breast tumor model in mice.

Previous studies have documented the role of MRE11 in cell survival and proliferation, yet it is only recently that the detailed mechanism and genetic involvement of MRE11 in human cancers has begun to emerge ( 22–24 ). In human malignancies, genetic mutations in MRE11 are relatively rare, whereas aberrant expression of MRE11 is more frequent ( 47 , 48 ). In our current study, we aimed to clarify the phenotypic associations of aberrant MRE11 expression in breast cancer, first by corroborating levels of MRE11 expression and phenotype behavior in a clinical study and second by investigating the underlying molecular mechanisms in in vitro studies using breast cancer cells. This study demonstrated that MRE11 high-expression phenotypes in breast cancer tissues were statistically significantly associated with malignant cancer behavior. Three distinct roles of MRE11 were found in breast tumor cells—proliferation via activation of STAT3, DSB repair facilitating radioresistance, and tumor cell invasion and migration via MMP-2 and MMP-9. These experimental findings provide a possible molecular explanation for the observed MRE11 phenotype behaviors of breast cancers studied in the clinical setting. More importantly, they may have therapeutic implications in terms of prognostic markers, treatment stratification, and molecular targets for both radiosensitization and inhibition of tumor cell proliferation.

Because MRN complex plays an important role in forming DSB repair foci, high expression of MRE11 within tumor cells would be hypothesized to allow increased DSB repair and confer radioresistance, leading to increased local recurrence and diminished rates of patient survival ( 49 ). Our study supports this hypothesis and clearly indicates that breast cancer patients with MRE11 high expression in cancer tissues had increased recurrence rates and decreased overall and disease-free survival after radiotherapy and chemotherapy. The importance of MRE11 expression is further highlighted by the fact that after adjusting for stage, age, ER and PR status, and MRE11 expression in multivariable Cox regression analyses, only tumor stage, ER status, and MRE11 expression were statistically significantly associated with patient survival.

One other important finding from Kaplan–Meier survival curves was that the group with both MRE11 high expression and ER status had the lowest overall survival and disease-free survival. This subgroup selects patients who are both resistant to chemotherapy and radiotherapy and not amenable to treatment with hormonal therapy. Thus, MRE11 high expression and ER status may act as a profound marker of poor multimodal therapeutic response and consequently a predictor of decreased survival.

Conversely, some clinical studies on MRE11 phenotypes in human cancers have suggested an association between high MRE11 expression and improved survival rates in bladder cancer, colorectal cancer, and breast cancer ( 39 , 48 , 50 ). Because of differences in cell types, patient inclusion criteria, treatment modalities, definition for MRE11 expression as well as the comparison of MRE11 expression with normal or cancer tissues, it is difficult to validly compare these studies. Discrepancy in clinical studies may be explained instead by differing endpoints, such as local vs distant recurrence, but also in the methods of quantification of MRE11 and definition of low and high expression ( 39 , 50 ). At present there is no standard method or definition for MRE11 expression, and care must therefore be taken in interpreting clinical studies. Furthermore, the only other study investigating MRE11 in breast cancer failed to show any relationship of MRE11 with local recurrence but only with distant recurrence ( 39 ). This also differs from the observed therapeutic effect of radiotherapy in breast cancer, as locoregional rather than distant control is exerted.

The findings in this study demonstrate that overexpression of MRE11 in breast cancer cells leads to cell proliferation through activation of STAT3 signaling and the downstream effectors including CCND1, BCL2L1, c-MYC, CDK4, and CDK6. However, these effects can be prevented by Stattic, a specific inhibitor of STAT3. Furthermore, treatment of MRE11-overexpressing breast cancer cells with Stattic results in increased apoptosis. These results indicate that MRE11 is a potential molecular target for breast cancer therapy.

Second, this study demonstrates the role of high MRE11 expression in DSB repair and associated radioresistance. In the clinical part of this study, we identified that one of the significant poor prognostic factors in determining clinical outcomes after radiotherapy and chemotherapy was high levels of MRE11 expression. In the corresponding in vitro experiment, MRE11 high expression showed an improved DSB repair and cell survival upon γ-irradiation treatment. In agreement with our results, previous reports also demonstrated that inherited or acquired MRE11 deficiency leads to inhibition of DSB repair, followed by radio- and chemosensitivity in normal as well as cancer cells ( 51–56 ). Delineating the mechanisms behind tumor radioresistance may allow stratification of treatment options based on MRE11 expression levels, as well as potentially permitting molecular targeting to increase radiosensitivity ( 50 ).

Third, another important role that high MRE11 expression may play in breast tumor cells is the ability to invade and migrate and hence facilitate metastasis. In the clinical part of the study, high MRE11 expression in breast cancer tissues was statistically significantly associated with lymph node metastasis. Consistent with this finding, MRE11-overexpressing cells had an enhanced ability of migration and invasion through activation of MMP-2 and -9. Moreover, MMP-2 and MMP-9 promoters have been found to directly interact with STAT3, and activation of the STAT3-MMP axis has been shown to promote invasiveness and metastasis of cancer cells ( 57–59 ). Thus, in this study, it is likely that the activation of STAT3-MMP-2 or STAT3-MMP-9 axis is also responsible for MRE11-mediated enhanced migration and invasion. To the best of our knowledge, this is the first report showing the influence of MRE11 on cancer cell metastasis, and further studies are required to dissect the underlying mechanism.

In conclusion, high MRE11 expression in breast cancer cells was associated with a more malignant cancer cell behavior in vitro and in vivo, poor response to radiotherapy and chemotherapy, and a poor survival in breast cancer patients. In this context, MRE11 should be regarded as an oncoprotein for breast cancer. The underlying molecular mechanisms of MRE11 in breast tumors demonstrated in this study have a number of important clinical implications and offer a number of novel therapeutic targets. First, specific inhibitors of STAT3 such as Stattic may reduce proliferation of breast tumors, particularly in high MRE11 expression groups. Second, stratification of breast cancers into MRE11 low and high expression groups may allow identification of tumors that are particularly radiosensitive and suitable for adjuvant or neoadjuvant radiotherapy. Last, MRE11 may be a great prognostic marker in breast cancer, as we have shown that MRE11-high and ER tumors have a highly statistically significant decrease in overall and disease-free survival.

The present study has a number of limitations. Although the similarity of the clinical phenotype with the in vivo mouse xenograft model provides insights into the possible mechanisms underlying the role of MRE11 in malignant cell behavior, it may not fully represent the clinical mechanisms in humans. Perhaps more importantly, a standardized approach for clinical assessment of MRE11 expression in breast cancer is required in order to provide a more reproducible method of phenotype stratification and for reliable interpretation of clinical studies. We would recommend that future clinical studies adopt the same criteria and methods for MRE11 expression levels that we have suggested in this study.

Table 2.

Association between clinicopathological characteristics of breast cancer patients and overall survival *

Variable  Univariate  Multivariable 
 HR (95% CI)  P  HR (95% CI)  P§ 
Stage||         
    I
    II
    III and IV  1.00 (referent)
9.88 (1.25 to 78.08)
17.96 (2.35 to 137.29)  .030
.005  1.00 (referent)
7.78 (0.98 to 62.06)
22.76 (2.90 to 178.55)  .053
.003 
Grade¶         
    1
    2
    3  1.00 (referent)
1.48 (0.20 to 11.27)
2.95 (0.37 to 23.73)  .71
.31  —  — 
Age, y         
    ≤50
    >50  1.00 (referent)
2.33 (0.96 to 5.67)  .06  1.00 (referent)
1.53 (0.61 to 3.83)  .36 
BMI, kg/m 2         
    ≤24
    >24  1.00 (referent)
1.56 (0.69 to 5.57)  .29  —  — 
ER status         
    ER+
    ER−  1.00 (referent)
5.65 (2.23 to 14.33)  <.001  1.00 (referent)
4.90 (1.46 to 16.45)  .010 
PR status         
    PR+
    PR−  1.00 (referent)
4.64 (1.72 to 12.50)  .002  1.00 (referent)
2.34 (0.64 to 8.55)  .20 
HER2 status         
    HER2−
    HER2+  1.00 (referent)
1.43 (0.63 to 3.26)  .40  —  — 
Radiotherapy         
    No
    Yes  1.00 (referent)
2.04 (0.76 to 5.50)  .16  —  — 
Chemotherapy         
    No
    Yes  1.00 (referent)
1.31 (0.31 to 5.59)  .72  —  — 
MRE11         
    Low
    High  1.00 (referent)
4.02 (1.19 to 13.58)  .025  1.00 (referent)
4.43 (1.21 to 16.25)  .025 
Variable  Univariate  Multivariable 
 HR (95% CI)  P  HR (95% CI)  P§ 
Stage||         
    I
    II
    III and IV  1.00 (referent)
9.88 (1.25 to 78.08)
17.96 (2.35 to 137.29)  .030
.005  1.00 (referent)
7.78 (0.98 to 62.06)
22.76 (2.90 to 178.55)  .053
.003 
Grade¶         
    1
    2
    3  1.00 (referent)
1.48 (0.20 to 11.27)
2.95 (0.37 to 23.73)  .71
.31  —  — 
Age, y         
    ≤50
    >50  1.00 (referent)
2.33 (0.96 to 5.67)  .06  1.00 (referent)
1.53 (0.61 to 3.83)  .36 
BMI, kg/m 2         
    ≤24
    >24  1.00 (referent)
1.56 (0.69 to 5.57)  .29  —  — 
ER status         
    ER+
    ER−  1.00 (referent)
5.65 (2.23 to 14.33)  <.001  1.00 (referent)
4.90 (1.46 to 16.45)  .010 
PR status         
    PR+
    PR−  1.00 (referent)
4.64 (1.72 to 12.50)  .002  1.00 (referent)
2.34 (0.64 to 8.55)  .20 
HER2 status         
    HER2−
    HER2+  1.00 (referent)
1.43 (0.63 to 3.26)  .40  —  — 
Radiotherapy         
    No
    Yes  1.00 (referent)
2.04 (0.76 to 5.50)  .16  —  — 
Chemotherapy         
    No
    Yes  1.00 (referent)
1.31 (0.31 to 5.59)  .72  —  — 
MRE11         
    Low
    High  1.00 (referent)
4.02 (1.19 to 13.58)  .025  1.00 (referent)
4.43 (1.21 to 16.25)  .025 

* Univariate and multivariable Cox regression model was used for statistical analysis. Multivariable analysis was adjusted for stage, age, ER and PR status. BMI =body mass index; ER =estrogen receptor; PR =progesterone receptor; HR = hazard ratio; CI = confidence interval; — = not applicable.

† Variables with P values less than .10 in the univariate analysis were included in multivariable analysis.

‡ Two-sided P values were calculated using a univariate Cox proportional hazards regression model.

§ Two-sided P values were calculated using a multivariable Cox proportional hazards regression model. Staging was based on the AJCC TNM staging system (1).

¶ Grading was based on the Modified Bloom-Richardson Grading Scheme (26).

Funding

National Health Research Institutes, Taiwan (NHRI-EX93-9306BI to SSFY); National Science Council, Taiwan (NSC97-2314-B-214-002-MY2 to SSFY); Department of Health, Taiwan (DOH101-TD-C-111-002 to MFH); E-Da Hospital, Taiwan (EDPJ97012 to SSFY).

References

1.
Benson
JR
Jatoi
I
Keisch
M
Esteva
FJ
Makris
A
Jordan
VC
Early breast cancer.
Lancet
 
2009
;
373
(
9673
):
1463
1479
2.
Amir
E
Freedman
OC
Seruga
B
Evans
DG
Assessing women at high risk of breast cancer: a review of risk assessment models.
J Natl Cancer Inst
 
2010
;
102
(
10
):
680
691
3.
Ganz
PA
Assessing the quality and value of quality-of-life measurement in breast cancer clinical trials.
J Natl Cancer Inst
 
2011
;
103
(
3
):
196
199
4.
Boyd
NF
Martin
LJ
Bronskill
M
Yaffe
MJ
Duric
N
Minkin
S
Breast tissue composition and susceptibility to breast cancer.
J Natl Cancer Inst
 
2010
;
102
(
16
):
1224
1237
5.
Hakem
R
DNA-damage repair; the good, the bad, and the ugly.
EMBO J
 
2008
;
27
(
4
):
589
605
6.
Powell
SN
Bindra
RS
Targeting the DNA damage response for cancer therapy.
DNA Repair (Amst)
 
2009
;
8
(
9
):
1153
1165
7.
Scott
SP
Pandita
TK
The cellular control of DNA double-strand breaks.
J Cell Biochem
 
2006
;
99
(
6
):
1463
1475
8.
Schultz
J
Data from DNA repair studies questioned as leading researcher investigated.
J Natl Cancer Inst
 
2003
;
95
(
18
):
1358
1359
9.
Helleday
T
Lo
J
van Gent
DC
Engelward
BP
DNA double-strand break repair: from mechanistic understanding to cancer treatment.
DNA Repair (Amst)
 
2007
;
6
(
7
):
923
935
10.
Kobayashi
J
Iwabuchi
K
Miyagawa
K
et al
Current topics in DNA double-strand break repair.
J Radiat Res
 
2008
;
49
(
2
):
93
103
11.
Natarajan
AT
Palitti
F
DNA repair and chromosomal alterations.
Mutat Res
 
2008
;
657
(
1
):
3
7
12.
Hartlerode
AJ
Scully
R
Mechanisms of double-strand break repair in somatic mammalian cells.
Biochem J
 
2009
;
423
(
2
):
157
168
13.
Assenmacher
N
Hopfner
KP
MRE11/RAD50/NBS1: complex activities.
Chromosoma
 
2004
;
113
(
4
):
157
166
14.
Shrivastav
M
De Haro
LP
Nickoloff
JA
Regulation of DNA double- strand break repair pathway choice.
Cell Res
 
2008
;
18
(
1
):
134
147
15.
Williams
RS
Williams
JS
Tainer
JA
Mre11-Rad50-Nbs1 is a keystone complex connecting DNA repair machinery, double-strand break signaling, and the chromatin template.
Biochem Cell Biol
 
2007
;
85
(
4
):
509
520
16.
Hopfner
KP
DNA double-strand breaks come into focus.
Cell
 
2009
;
139
(
1
):
25
27
17.
Callén
E
Surrallés
J
Telomere dysfunction in genome instability syndromes.
Mutat Res
 
2004
;
567
(
1
):
85
104
18.
Stewart
SA
Weinberg
RA
Telomeres: cancer to human aging.
Annu Rev Cell Dev Biol
 
2006
;
22
:
531
557
19.
Chai
W
Sfeir
AJ
Hoshiyama
H
Shay
JW
Wright
WE
The involvement of the Mre11/Rad50/Nbs1 complex in the generation of G-overhangs at human telomeres.
EMBO Rep
 
2006
;
7
(
2
):
225
230
20.
Royle
NJ
Méndez-Bermúdez
A
Gravani
A
et al
The role of recombination in telomere length maintenance.
Biochem Soc Trans
 
2009
;
37
(p
t 3
):
589
595
21.
Ueno
M
Roles of DNA repair proteins in telomere maintenance.
Biosci Biotechnol Biochem
 
2010
;
74
(
1
):
1
6
22.
Xiao
Y
Weaver
DT
Conditional gene targeted deletion by Cre recombinase demonstrates the requirement for the double-strand break repair Mre11 protein in murine embryonic stem cells.
Nucleic Acids Res
 
1997
;
25
(
15
):
2985
2991
23.
Petrini
JH
Walsh
ME
DiMare
C
Chen
XN
Korenberg
JR
Weaver
DT
Isolation and characterization of the human MRE11 homologue.
Genomics
 
1995
;
29
(
1
):
80
86
24.
Lanson
NA
Jr,
Egeland
DB
Royals
BA
Claycomb
WC
The MRE11-NBS1-RAD50 pathway is perturbed in SV40 large T antigen-immortalized AT-1, AT-2 and HL-1 cardiomyocytes.
Nucleic Acids Res
 
2000
;
28
(
15
):
2882
2892
25.
Ermilova
VD
International histological classification of breast cancer by the WHO (1968) and its prognostic value.
Arkh Patol
 
1980
;
42
(
4
):
13
19
26.
Frierson
HF
Jr,
Wolber
RA
Berean
KW
et al
Interobserver reproducibility of the Nottingham modification of the Bloom and Richardson histologic grading scheme for infiltrating ductal carcinoma.
Am J Clin Pathol
 
1995
;
103
(
2
):
195
198
27.
Lee
YC
Yang
YH
Su
JH
Chang
HL
Hou
MF
Yuan
SS
High visfatin expression in breast cancer tissue is associated with poor survival.
Cancer Epidemiol Biomarkers Prev
 
2011
;
20
(
9
):
1892
1901
28.
Yeh
YT
Hou
MF
Chung
YF
et al
Decreased expression of phosphorylated JNK in breast infiltrating ductal carcinoma is associated with a better overall survival.
Int J Cancer
 
2006
;
118
(
11
):
2678
2684
29.
Schust
J
Sperl
B
Hollis
A
Mayer
TU
Berg
T
Stattic: a small-molecule inhibitor of STAT3 activation and dimerization.
Chem Biol
 
2006
;
13
(
11
):
1235
1242
30.
Chang
HL
Hsu
HK
Su
JH
et al
The fractionated Toona sinensis leaf extract induces apoptosis of human ovarian cancer cells and inhibits tumor growth in a murine xenograft model.
Gynecol Oncol
 
2006
;
102
(
2
):
309
314
31.
Jiang
Y
Rom
WN
Yie
TA
Chi
CX
Tchou-Wong
KM
Induction of tumor suppression and glandular differentiation of A549 lung carcinoma cells by dominant-negative IGF-I receptor.
Oncogene
 
1999
;
18
(
44
):
6071
6077
32.
Konttinen
YT
Ainola
M
Valleala
H
et al
Analysis of 16 different matrix metalloproteinases (MMP-1 to MMP-20) in the synovial membrane: different profiles in trauma and rheumatoid arthritis.
Ann Rheum Dis
 
1999
;
58
(
11
):
691
697
33.
Robison
JG
Elliott
J
Dixon
K
Oakley
GG
Replication protein A and the Mre11.Rad50.Nbs1 complex co-localize and interact at sites of stalled replication forks.
J Biol Chem
 
2004
;
279
(
33
):
34802
34810
34.
Olive
PL
Banáth
JP
Durand
RE
Heterogeneity in radiation-induced DNA damage and repair in tumor and normal cells measured using the “comet” assay.
Radiat Res
 
1990
;
122
(
1
):
86
94
35.
Hellman
B
Vaghef
H
Boström
B
The concepts of tail moment and tail inertia in the single cell gel electrophoresis assay.
Mutat Res
 
1995
;
336
(
2
):
123
131
36.
Bauer
E
Recknagel
RD
Fiedler
U
Wollweber
L
Bock
C
Greulich
KO
The distribution of the tail moments in single cell gel electrophoresis (comet assay) obeys a chi-square (chi2) not a gaussian distribution.
Mutat Res
 
1998
;
398
(
1–2
):
101
110
37.
Chen
HM
Chang
FR
Hsieh
YC
et al
A novel synthetic protoapigenone analogue, WYC02-9, induces DNA damage and apoptosis in DU145 prostate cancer cells through generation of reactive oxygen species.
Free Radic Biol Med
 
2011
;
50
(
9
):
1151
1162
38.
Angèle
S
Treilleux
I
Brémond
A
Tanière
P
Hall
J
Altered expression of DNA double-strand break detection and repair proteins in breast carcinomas.
Histopathology
 
2003
;
43
(
4
):
347
353
39.
Söderlund
K
Stål
O
Skoog
L
Rutqvist
LE
Nordenskjöld
B
Askmalm
MS
Intact Mre11/Rad50/Nbs1 complex predicts good response to radiotherapy in early breast cancer.
Int J Radiat Oncol Biol Phys
 
2007
;
68
(
1
):
50
58
40.
Bartkova
J
Tommiska
J
Oplustilova
L
et al
Aberrations of the MRE11-RAD50-NBS1 DNA damage sensor complex in human breast cancer: MRE11 as a candidate familial cancer-predisposing gene.
Mol Oncol
 
2008
;
2
(
4
):
296
316
41.
Bowman
T
Garcia
R
Turkson
J
Jove
R
STATs in oncogenesis.
Oncogene
 
2000
;
19
(
21
):
2474
2488
42.
Watanabe
H
Extracellular matrix–regulation of cancer invasion and metastasis.
Gan To Kagaku Ryoho
 
2010
;
37
(
11
):
2058
2061
43.
Bernhard
EJ
Gruber
SB
Muschel
RJ
Direct evidence linking expression of matrix metalloproteinase 9 (92-kDa gelatinase/collagenase) to the metastatic phenotype in transformed rat embryo cells.
Proc Natl Acad Sci USA
 
1994
;
91
(
10
):
4293
4297
44.
Kinner
A
Wu
W
Staudt
C
Iliakis
G
Gamma-H2AX in recognition and signaling of DNA double-strand breaks in the context of chromatin.
Nucleic Acids Res
 
2008
;
36
(
17
):
5678
5694
45.
Pilch
DR
Sedelnikova
OA
Redon
C
Celeste
A
Nussenzweig
A
Bonner
WM
Characteristics of gamma-H2AX foci at DNA double-strand breaks sites.
Biochem Cell Biol
 
2003
;
81
(
3
):
123
129
46.
Kuo
LJ
Yang
LX
Gamma-H2AX—a novel biomarker for DNA double-strand breaks.
In Vivo
 
2008
;
22
(
3
):
305
309
47.
Dzikiewicz-Krawczyk
A
The importance of making ends meet: mutations in genes and altered expression of proteins of the MRN complex and cancer.
Mutat Res
 
2008
;
659
(
3
):
262
273
48.
Gao
J
Zhang
H
Arbman
G
Sun
XF
RAD50/MRE11/NBS1 proteins in relation to tumour development and prognosis in patients with microsatellite stable colorectal cancer.
Histol Histopathol
 
2008
;
23
(
12
):
1495
1502
49.
Le Scodan
R
Cizeron-Clairac
G
Fourme
E
et al
DNA repair gene expression and risk of locoregional relapse in breast cancer patients.
Int J Radiat Oncol Biol Phys
 
2010
;
78
(
2
):
328
336
50.
Choudhury
A
Nelson
LD
Teo
MT
et al
MRE11 expression is predictive of cause-specific survival following radical radiotherapy for muscle-invasive bladder cancer.
Cancer Res
 
2010
;
70
(
18
):
7017
7026
51.
Gatti
RA
The inherited basis of human radiosensitivity.
Acta Oncol
 
2001
;
40
(
6
):
702
711
52.
Ewald
B
Sampath
D
Plunkett
W
ATM and the Mre11-Rad50-Nbs1 complex respond to nucleoside analogue-induced stalled replication forks and contribute to drug resistance.
Cancer Res
 
2008
;
68
(
19
):
7947
7955
53.
Kuroda
S
Fujiwara
T
Shirakawa
Y
et al
Telomerase-dependent oncolytic adenovirus sensitizes human cancer cells to ionizing radiation via inhibition of DNA repair machinery.
Cancer Res
 
2010
;
70
(
22
):
9339
9348
54.
Liikanen
I
Dias
JD
Nokisalmi
P
et al
Adenoviral E4orf3 and E4orf6 proteins, but not E1B55K, increase killing of cancer cells by radiotherapy in vivo.
Int J Radiat Oncol Biol Phys
 
2010
;
78
(
4
):
1201
1209
55.
Deng
R
Tang
J
Ma
JG
et al
PKB/Akt promotes DSB repair in cancer cells through upregulating Mre11 expression following ionizing radiation.
Oncogene
 
2011
;
30
(
8
):
944
955
56.
Zhang
J
Xin
X
Chen
Q
et al
Oligomannurarate sulfate sensitizes cancer cells to doxorubicin by inhibiting atypical activation of NF-kappaB via targeting of Mre11
Int J Cancer.
  2012;130(2):467–477.
57.
Huang
C
Cao
J
Huang
KJ
et al
Inhibition of STAT3 activity with AG490 decreases the invasion of human pancreatic cancer cells in vitro.
Cancer Sci
 
2006
;
97
(
12
):
1417
1423
58.
Xie
TX
Wei
D
Liu
M
et al
Stat3 activation regulates the expression of matrix metalloproteinase-2 and tumor invasion and metastasis.
Oncogene
 
2004
;
23
(
20
):
3550
3560
59.
Wu
X
Yan
Q
Zhang
Z
Du
G
Wan
X
Acrp30 inhibits leptin-induced metastasis by downregulating the JAK/STAT3 pathway via AMPK activation in aggressive SPEC-2 endometrial cancer cells.
Oncol Rep
 
2012
;
27
(
5
):
1488
1496

Notes

We thank Dr Ying-Hsien Kao at E-Da Hospital for critical reading of the manuscript and Dr Shyh-An Yeh at E-Da Hospital for technical support with ionizing radiation experiment, and we especially thank Dr Yi-Hsin Yang at Kaohsiung Medical University Hospital for statistical advice.

The authors are solely responsible for the study design, data collection, analysis and interpretation of the data, writing the manuscript, and decision to submit the manuscript for publication.