Triple-negative breast cancer (TNBC) is a highly aggressive tumor subtype lacking effective prognostic indicators or therapeutic targets. Mitochondrial function is dysregulated frequently in cancer cells to allow for adaptation to a harsh tumor microenvironment. Targeting mitochondrial biogenesis and bioenergetics is, therefore, an attractive therapeutic strategy. In this study, we performed quantitative proteomic analyses in human parental and metastatic breast cancer cell lines to identify mitochondrial proteins involved in TNBC metastasis. We found that single-strand DNA-binding protein 1 (SSBP1) was downregulated in highly metastatic breast cancer cells. Moreover, SSBP1 downregulation promoted TNBC cell metastasis in vitro and in vivo. Mechanistically, SSBP1 loss decreased mitochondrial DNA copy number, thereby potentiating calcineurin-mediated mitochondrial retrograde signaling that induced c-Rel/p50 nuclear localization, activated TGFβ promoter activity, and TGFβ-driven epithelial-to-mesenchymal transition. Low SSBP1 expression correlated with tumor progression and poor prognosis in patients. Collectively, our findings identified SSBP1 as a novel metastasis suppressor and elucidated the mechanisms by which dysregulated mitochondrial signaling contributes to metastatic potential, providing potential new prognostic indicators for patients with TNBC. Cancer Res; 76(4); 952–64. ©2015 AACR.

Triple-negative breast cancer (TNBC) is a highly aggressive tumor subtype associated with poor prognosis. TNBC tumors are generally larger in size and of a higher grade, often have lymph node involvement at the time of diagnosis and tend to more frequently affect younger patients (1–6). However, the absence of well-defined molecular targets for this tumor subtype makes TNBC treatment challenging (7–9). As the major energy and metabolite source in cells, mitochondrial function is often dysregulated in cancer, and there is growing interest in understanding how altered mitochondrial function may be targeted to inhibit tumor progression. The modified bioenergetic and biosynthetic states of mitochondria communicate with the nucleus through mitochondrial retrograde signaling to modulate signaling pathways and transcriptional circuits to meet the requirements of cancer cells. Thus, the discovery of novel biomarkers involved in this pathophysiologic process can provide broad new insights into the mechanisms of tumorigenesis and tumor progression.

In our previous study, the highly metastatic cell line MDA-MB-231HM was developed from parental MDA-MB-231 cells by in vivo selection in mice. Accordingly, MDA-MB-231HM cells exhibited increased invasiveness to the lungs compared with parental MDA-MB-231 cells. Another highly bone-metastatic cell line, MDA-MB-231Bo, which was previously isolated via in vivo selection and also had stronger lung metastasis capacity compared with MDA-MB-231 cells (Supplementary Fig. S1A), was obtained from Dr. Toshiyuki Yoneda (The University of Texas, Houston, TX; refs. 10, 11). Because of the similar genetic background of these cells, they provide a unique model for identifying candidate metastasis-associated biomarkers and potential therapeutic targets for TNBC. To discover mitochondrial biomarkers that may be associated with TNBC progression, we thoroughly investigated the mitochondrial proteome profile of the MDA-MB-231 cell line series. Using iTRAQ-labeling technology followed by nanoscale high-performance liquid chromatography (nano-HPLC)-MS/MS, we identified single-strand DNA-binding protein 1 (SSBP1) as a differentially expressed protein associated with the metastatic potential of TNBC.

SSBP1 (also known as mtSSB) exists in eukaryotic mitochondria (12). cDNA analysis revealed that SSBP1 is translated as a 148-amino acid polypeptide, and the first 16 amino acids are proposed to be removed after the import of this protein into the mitochondria, resulting in a mature protein consisting of 132 amino acids (13). SSBP1 is distinct from nuclear SSB and plays an important role in mtDNA replication, recombination, and repair by binding to single-stranded mtDNA. SSBP1 has been shown to stimulate the DNA polymerase activity of pol γ in humans (14, 15) and the unwinding activity of human mtDNA helicase (16). In addition, SSBP1 stimulates strand-displacement DNA synthesis promoted by the combined action of mtDNA helicase (17). Recently, it has been reported that SSBP1 is associated with human cancer. The level of SSBP1 protein correlates with the aggressiveness of human osteosarcoma cells, suggesting a link between mtDNA replication and cancer progression (18). SSBP1 has also been identified as a novel binding partner of the tumor suppressor p53 in mitochondria biophysical measurements (19). In this study, we show for the first time that SSBP1 is a suppressor of TNBC metastasis.

iTRAQ-nano-HPLC-MS/MS analyses

The cell lysates of parental MDA-MB-231 and highly metastatic MDA-MB-231 (MDA-MB-231HM and MDA-MB-231Bo) breast cancer cells were quantified using a Bradford assay, labeled with iTRAQ-labeling reagents (Applied Biosystems), and digested with trypsin. The resulting peptides were fractionated with a Waters ultra-performance liquid chromatography device, and the fractions were then separated by nano-HPLC (Eksigent Technologies) on a secondary reverse-phase (RP) analytical column. A Triple TOF 4600 mass spectrometer (MS) was operated in information-dependent data acquisition mode to automatically switch between MS and tandem MS (MS/MS) acquisition. MS/MS spectra were extracted and charge state deconvoluted using an MS Data Converter from AB Sciex.

Cell culture

All breast cancer cell lines, normal breast MCF10A cells and HEK 293T cells were obtained from the Shanghai Cell Bank, Type Culture Collection Committee of Chinese Academy of Science (Shanghai, China) in 2014, and they were authenticated using short tandem repeat profiling by the cell bank. These cell lines were conserved in our laboratory and subjected to routine cell line quality examinations (e.g., morphology, Mycoplasma) by HD Biosciences every 3 months. The cells for experiments were passaged for less than 6 months. The MDA-MB-231Bo cells, obtained from Dr. Toshiyuki Yoneda (The University of Texas, Houston, TX) were cultured in ATCC-formulated Leibovitz L-15 medium supplemented with 10% FBS, 100 IU/mL penicillin, and 100 μg/mL streptomycin. The remaining cell lines were maintained in complete growth medium as recommended by the manufacturers.

Study population

This study involved 250 breast cancer patients with a diagnosis of pathologically invasive ductal breast cancer with a follow-up period of at least 5 years. The diagnoses were verified by two independent pathologists in the Department of Pathology of the Fudan University Shanghai Cancer Center (FDUSCC). Samples were collected from these patients in the Department of Breast Surgery of FDUSCC from August 2001 to March 2006. This study was approved by the Ethics Committee of FDUSCC, and each participant signed an informed consent document.

Tissue microarrays, IHC staining, and IHC variable evaluation

Tissue microarrays (TMA; Supplementary Table S1) were constructed from archival formalin-fixed, paraffin wax-embedded samples of carcinomas obtained from the 250 breast cancer patients described above. Tissue cylinders with diameters of 2 mm were punched from a previously marked tumor area in each block (donor block) and inserted into a recipient paraffin wax block, resulting in a 10 × 10 array. Tissue samples from all 250 patients were punched twice into the microarray to compare staining patterns in different areas of the same tumor. The brief details of IHC staining, and IHC variable evaluation are described in the Supplementary Methods.

Quantitative real-time PCR

Total RNA was extracted with the TRIzol reagent (Invitrogen) and reverse transcribed using the PrimeScript RT Reagent Kit (Perfect Real-Time; TaKaRa Biotechnology). Subsequently, real-time PCR was performed with SYBR Premix Ex Taq (TaKaRa Bio) using an ABI Prism 7900 instrument (Applied Biosystems). mtDNA content was measured by real-time PCR using specific primers for the mtDNA-encoded cytochrome oxidase I and nuclear DNA-encoded GAPDH.

The primer sequences used in this study are as follows:

  • GAPDH: F 5′-GGTGGTCTCCTCTGACTTCAACA-3′,

  • GAPDH: R 5′-GTTGCTGTAGCCAAATTCGTTGT-3′,

  • SSBP1: F 5′-TGGAGTCGTGTGTTTTGGCT-3′,

  • SSBP1: R 5′-TAGGCTTTTCCTGAAAACCGAGG-3′,

  • TGF-β: F 5′-CCCTGGACACCAACTATTGC-3′,

  • TGF-β: R 5′-CTTCCAGCCGAGGTCCTT-3′,

  • TβR1: F 5′-CCCTGGACACCAACTATTGC-3′,

  • TβR1: R 5′-CTTCCAGCCGAGGTCCTT-3′,

  • SMAD3: F 5′-AGACCCCACCCCCTGGCTACCTG-3′,

  • SMAD3: R 5′-GGGGACATCGGATTCGGGGA-3′,

  • SMAD7: F 5′-TGGATGGCGTGTGGGTTTA-3′,

  • SMAD7: R 5′-TGGCGGACTTGATGAAGATG-3′,

  • IL-11: F 5′-ACTGCTGCTGCTGAAGACTC-3′,

  • IL-11: R 5′-CCACCCCTGCTCCTGAAATA-3′,

  • CXCR4: F 5′-CAGTGGCCGACCTCCTCTT-3′,

  • CXCR4: R 5′-CAGTTTGCCACGGCATCA-3′,

  • PTHrP: F 5′-TTCTTCCCAGGTGTCTTGAG-3′,

  • PTHrP: R 5′-TTTACGGCGACGATTCTTCC-3′,

  • Snail1: F 5′-CGAGTGGTTCTTCTGCGCTA-3′,

  • Snail1: R 5′-CTGCTGGAAGGTAAACTCTGGA-3′,

  • Snail2: F 5′-GGCTGGCCAAACATAAGCAG-3′,

  • Snail2: R 5′-GCTTCTCCCCCGTGTGAGTT-3′.

Plasmids and shRNA

Human SSBP1 cDNA was subcloned from the MDA-MB-231 breast cancer cell line into the pCDH-CMV-MCS-EF1-Puro lentiviral vector with a Flag tag. The cloned primer sequence is as follows:

  • Forward: 5′-CCGGAATTCGCCACCATGGACTACAAGGACGATGATGACAAG CTCGATGGAGGAATGTTTCGAAGACCTGTATT-3′

  • Reverse: 5′-CGCGGATCCCTACTCCTTCTCTTTCGTCT-3′

SSBP1 shRNAs and the negative control were purchased from GeneChem and expressed in the GV248 backbone. The target sequences are as follows:

  • shRNA-1: 5′-AGTTTGGTTCTTGAAAGAT-3′;

  • shRNA-2: 5′-TGGCACAGAATATCAGTAT-3′.

Lentivirus packaging and infection

Briefly, 293T cells were cotransfected with lentiviral vectors and the packaging vectors PCDH (or GV248), psPAX2 and pMD2G. Forty-eight hours after transfection, the viral supernatants were collected, filtered, and concentrated by ultracentrifugation. Polybrene (Sigma-Aldrich) was added at a working concentration of 8 μg/mL. The cells were incubated with virus for 12 hours, and then medium containing FBS was added. Twenty-four hours later, the infected cells were subjected to selection with 2 μg/mL puromycin for 1 week.

Western blot analysis

Whole-cell lysates were generated using Pierce T-PER (Tissue Protein Extraction Reagent; Thermo Fisher Scientific Inc.) containing protease inhibitor cocktail tablets (Roche) and phosphatase inhibitors (Roche). In total, 30 μg of the cell lysates were resolved by SDS-PAGE and transferred to polyvinylidene fluoride membranes (Pall). The membranes were blocked in 5% milk or 5% BSA, and then incubated with various primary antibodies followed by the appropriate horseradish peroxidase–conjugated secondary antibodies. Immunoreactive bands were identified using enhanced chemiluminescence, according to the manufacturer's instructions, and quantified by densitometry. The antibodies used in this study are listed in Supplementary Table S2.

Transwell assays

Cells (5 × 105 for the migration assay and 1 × 106 for the invasion assay) were plated in the top chamber of a non-coated membrane or Matrigel-coated Transwell chambers (BD Biosciences) in medium without FBS. Medium supplemented with serum was used as a chemoattractant in the bottom chamber. The cells were incubated for 15 to 20 hours, and cells that did not migrate through the pores were removed with a cotton swab. The cells on the lower surface of the membrane were stained with methanol and 0.1% crystal violet and then counted.

Calcineurin activity

Calcineurin activity was measured using a calcineurin activity assay kit (Millipore) according to the manufacturer's instructions. Briefly, total phosphatase activity in cellular extracts was measured by incubation with a calcineurin-specific RII phosphopeptide substrate in the presence or absence of EGTA. The detection of free phosphate release is based on the malachite green assay in the presence of EGTA, which was subtracted from the total activity to obtain calcineurin-specific phosphatase activity.

Measurement of mitochondrial membrane potential

Tumor cells plated in phenol red–free growth medium were treated with 1 mg/mL JC-1 dye for 30 minutes at 37°C and analyzed by FACS. Red or green JC-1 fluorescence signals were resolved by detection in the FL1 and FL3 channels, respectively. Loss of mitochondrial membrane potential was measured by a reduction in the red/green fluorescence intensity ratio.

Measurement of mitochondrial reactive oxygen species

Mitochondrial-specific superoxide production was measured by labeling live cells with MitoSOX-Red (Invitrogen). Cells were incubated with 5 μmol/L MitoSOX-red for 20 minutes at 37°C in serum-free DMEM, washed twice with Dulbecco PBS, and then immediately analyzed by FACS.

Phosphoprotein profiling with the Phospho Explorer antibody microarray

The Phospho Explorer antibody microarray, which was designed and manufactured by Full Moon Biosystems, Inc., contains 1,318 antibodies. Each antibody has two replicates that are printed on a coated glass microscope slide together with multiple positive and negative controls. The antibody array experiment was performed according to an established protocol.

Immunofluorescence

MCF10A-stable cell lines grown on coverslips were fixed with 4% paraformaldehyde for 30 minutes at room temperature, permeabilized with 0.5% Triton X-100 for 5 minutes at 4°C, and incubated with primary antibodies for 2 hours at 37°C. The slides were then incubated with Alexa 695–conjugated (red; Abcam) or Alexa 594–conjugated (red; Invitrogen) secondary antibodies for 40 minutes at room temperature. Images were captured with a confocal laser microscope (Leica TCS SP5 II). At least 100 cells were analyzed for each group.

Metastasis assays in nude mice

All animal work was performed in accordance with the guidelines of the Institutional Animal Care and Use Committee of Fudan University under approved protocols. In total, 5 × 105 cells were washed in PBS and injected intravenously into female BALB/c nude mice (n = 6) to study lung metastasis activity. Noninvasive bioluminescence imaging was performed to quantify the metastasis burden in target organs using an in vivo imaging system.

Luciferase assay

In the luciferase reporter assays, HEK 293T cells were seeded in 96-well plates at a density of 1 × 104 to 2 × 104 cells per well and cultured for 24 hours followed by cotransfection with c-Rel or the expression control vector at 150 μg/well; TGFβ luciferase constructs of different promoter lengths at 150 μg/well; or the promoter control vector pGL3-basic with 7.5 ng/well of the control Renilla luciferase reporter construct pRL-TK (Promega). After 48 hours, luciferase assays were performed as recommended by the Dual-Glo Luciferase Assay System (Promega) detected by a SynergyHT Multi-Mode Microplate Reader (BioTek) at 595 nm. Primers for plasmid construction are listed in Supplementary Table S3.

Chromatin immunoprecipitation

After incubating with 1% formaldehyde at room temperature for 10 minutes, the cells were pelleted and resuspended in 200 mL of lysis buffer (1% SDS; 10 mmol/L EDTA; 50 mmol/L Tris-HCl, pH 8.0). The cell lysates were then sonicated (Sonics & Materials, Inc.) until the DNA was cleaved into fragments of 200 to 500 bp. These extracts were immunoprecipitated using anti–c-Rel or normal mouse IgG (Santa Cruz Biotechnology, Inc.), and chromatin immunoprecipitation (ChIP) experiments were performed using the EZ–Magna G Chromatin Immunoprecipitation Kit (Millipore, Merck KGaA) according to the manufacturer's instructions. Subsequently, PCR for the TGFβ promoter was performed using specific primers, which are listed in Supplementary Table S4.

Statistical analyses

Results are reported as the mean ± SD or the mean ± SEM as indicated in the figure legends. The results were analyzed using the SPSS 16.0 software (SPSS) and PRISM 5.0 (GraphPad Software Inc.). Comparisons of quantitative data between two groups were analyzed with the Student t test (two-tailed; P < 0.05 was considered significant). The χ2 test was used to compare qualitative variables. Cumulative survival time was calculated using the Kaplan–Meier method, which was analyzed using the log-rank test. Univariate and multivariate analyses were based on the Cox proportional hazards regression model. P values <0.05 were considered significant.

Identification of SSBP1 as a downregulated gene in highly metastatic breast cancer cells

Using the iTRAQ-labeling method with our model system, we analyzed MDA-MB-231 (human breast cancer) cell and its derived highly metastatic sublines MDA-MB-231HM and MDA-MB-231Bo (Fig. 1A). Our analysis spanned two pairs of cell lines allowing for overlapped comparisons of mitochondrial proteins from matched parental and derivative cell lines. To generate mitochondrial proteins based on quantitative abundance data, we considered proteins upregulated or downregulated at least 1.5-fold in highly metastatic variants as compared with their matched, parental cell lines. This analysis yielded quantitative mitochondrial proteins consisting of 50 and 66 proteins for the 231HM/231 and 231Bo/231 comparisons, respectively (part of results showed in Supplementary Table S5). We focused our interesting candidate on one protein SSBP1, which is significantly downregulated in MDA-MB-231HM and MDA-MB-231Bo cells compared with the parental MDA-MB-231 cells (Fig. 1B and Supplementary Fig. S1B–S1E). In addition to these two metastatic breast cancer cell lines, a panel of breast cancer cell lines was also evaluated to observe the general trend of SSBP1 gene expression. In accordance with the iTRAQ results, SSBP1 mRNA expression was generally observed in luminal breast cancer cells (MCF-7, T47D, and ZR-75-30), which are considered weakly metastatic cell lines, but SSBP1 was downregulated in basal-like breast cancer cells (BT549, MDA-MB-468, and MDA-MB-231). In particular, the highly metastatic cell lines (MDA-MB-231HM and MDA-MB-231Bo) had greater decreases in SSBP1 expression levels (Fig. 1C). In addition, Oncomine expression analysis revealed SSBP1 downregulation in various human cancers (Fig. 1D). We also studied the prognostic value of SSBP1 using an online Kaplan–Meier plot for survival analysis. For the entire cohort of patients and patients with basal-subtype tumors, low SSBP1 expression had a negative impact on relapse-free survival (Fig. 1E). However, SSBP1 expression did not correlate with relapse-free survival for patients with ER-positive and Her-2–positive breast tumors (Fig. 1E). Furthermore, patients in The Cancer Genome Atlas (TCGA) cohort with stomach adenocarcinoma and SSBP1 gene alterations had notably shorter lifespans than those with stomach adenocarcinoma without an SSBP1 gene alteration (Supplementary Fig. S2A). Taken together, these data suggest that SSBP1 may play an oncogenic role in multiple cancer types, making it worthy of further investigation.

Figure 1.

SSBP1 is downregulated in highly metastatic breast cancer cells. A, schematic overview of quantitative analysis approaches. B, SSBP1 is downregulated in both 231HM and 231Bo cells compared with parental 231 cells. C, SSBP1 mRNA expression in breast cancer cell lines. D, Oncomine box plots of SSBP1 expression in multiple advanced human cancers. E, survival analysis of SSBP1 expression in patients of the indicated subtypes using an online Kaplan–Meier plotter.

Figure 1.

SSBP1 is downregulated in highly metastatic breast cancer cells. A, schematic overview of quantitative analysis approaches. B, SSBP1 is downregulated in both 231HM and 231Bo cells compared with parental 231 cells. C, SSBP1 mRNA expression in breast cancer cell lines. D, Oncomine box plots of SSBP1 expression in multiple advanced human cancers. E, survival analysis of SSBP1 expression in patients of the indicated subtypes using an online Kaplan–Meier plotter.

Close modal

Loss of SSBP1 promotes breast cancer cell migration and metastasis in vitro and in vivo

To further investigate the role of SSBP1 in tumor migration and metastasis, we used three breast cancer cell line models (MDA-MB-231, MDA-MB-468, and MDA-MB-231Bo) for in vitro Transwell migration and in vivo metastasis studies. First, we used two independent shRNAs targeting SSBP1 in MDA-MB-231 and MDA-MB-468 cells. An empty vector (shCon) served as control. Western blot analysis demonstrated that both shRNA constructs strongly reduced SSBP1 protein levels (Fig. 2A). Furthermore, we generated an SSBP1 stably overexpressing MDA-MB-231Bo cell line (Fig. 2A), and the results demonstrated that SSBP1 knockdown in MDA-MB-231 and MDA-MB-468 cells significantly promoted cell migration and invasion in Transwell assays (Fig. 2B–E). Using a complementary inverse approach, elevated SSBP1 expression suppressed the invasion ability of MDA-MB-231Bo cells in an invasion assay (Fig. 2F and G).

Figure 2.

Low SSBP1 expression promotes breast cancer metastasis in vitro and in vivo. A, SSBP1 was stably knocked down in MDA-MB-231 and MDA-MB-468 cells using two independent shRNA constructs, and it was constitutively overexpressed in the highly metastatic MDA-MB-231Bo cells. B–G, the migration and invasion abilities of each cell line were evaluated by Transwell assays in vitro. Top, photos of representative fields (magnification, ×100) of invasive cells; bottom, the histograms of the results. B and C, for MDA-MB-231 cells; D and E, for MDA-MB-468 cells; and F and G, for MDA-MB-231Bo cells. Statistical analysis was performed using the Student t test (n = 3). Error bars, SD; *, P < 0.05 compared with control. H, BLI of three representative mice in each group at week 6 after injection with indicated cells. I, BLI signals of each mouse.

Figure 2.

Low SSBP1 expression promotes breast cancer metastasis in vitro and in vivo. A, SSBP1 was stably knocked down in MDA-MB-231 and MDA-MB-468 cells using two independent shRNA constructs, and it was constitutively overexpressed in the highly metastatic MDA-MB-231Bo cells. B–G, the migration and invasion abilities of each cell line were evaluated by Transwell assays in vitro. Top, photos of representative fields (magnification, ×100) of invasive cells; bottom, the histograms of the results. B and C, for MDA-MB-231 cells; D and E, for MDA-MB-468 cells; and F and G, for MDA-MB-231Bo cells. Statistical analysis was performed using the Student t test (n = 3). Error bars, SD; *, P < 0.05 compared with control. H, BLI of three representative mice in each group at week 6 after injection with indicated cells. I, BLI signals of each mouse.

Close modal

Next, to examine whether SSBP1 affects invasion and metastasis in vivo, we constructed mouse models. We labeled SSBP1 knockdown MDA-MB-231 cells and SSBP1 stably overexpressing MDA-MB-231Bo cells with a retroviral construct expressing a GFP/luciferase fusion protein (20), and tumor cell colonization and outgrowth in the lungs were monitored by noninvasive bioluminescent imaging (BLI) 6 weeks after the intravenous delivery of cells into the nude mice. Our results demonstrated that low SSBP1 expression significantly increased lung metastasis in vivo, and the lung metastasis-promoting effect was reversed after SSBP1 overexpression in MDA-MB-231Bo cells (Fig. 2H and I and Supplementary Fig. S2B). These findings confirm our hypothesis that low SSBP1 expression promotes breast cancer cell metastasis in vitro and in vivo.

Downregulation of SSBP1 induces mitochondria retrograde signaling

As SSBP1 is essential for mtDNA replication (14, 15), we hypothesized that SSBP1 downregulation would reduce the mtDNA content in cells. First, we assessed the relative mtDNA copy number by specific primers. We found that compared with control cells, the mtDNA copy number in MDA-MB-231 shSSBP1#1 and MDA-MB-468 shSSBP1#1 cells was significantly decreased without obvious differences in the amplification of the GAPDH nuclear gene (Fig. 3B). Furthermore, MDA-MB-231Bo cells had a greater decrease in mtDNA copy number compared with the parental MDA-MB-231 cells (Fig. 3A), indicating that the mtDNA copy number is associated with tumor metastasis. Next, to determine whether the reduced mtDNA content potentiates mitochondrial reactive oxygen species (ROS) generation and loss of the mitochondrial membrane potential (ΔΨm), we examined the level of mitochondrial ROS generation and ΔΨm in MDA-MB-231 and MDA-MB-468 cells. As shown in Fig. 3C and D, there was a significant increase in mitochondrial ROS generation and ΔΨm loss in SSBP1-downregulated cells. In addition, it showed that all these downstream changes similarly generated when mtDNA copy number was reduced in MDA-MB-231 cells by treated with ethidium bromide (EB). We used 30 ng/mL of EB, which is the concentration required for partial depletion of mtDNA (Supplementary Fig. S3A). The mtROS levels and p-SMAD3 levels were increased markedly in “mtDNA-reduced” MDA-MB-231 cells (MDA-MB-231 EB; Supplementary Fig. S3B and S3C). Accordingly, the “mtDNA-reduced” cells acquired increased migratory capacity (Supplementary Fig. S3D).

Figure 3.

Downregulation of SSBP1 activates mitochondria retrograde signaling. A and B, the relative mtDNA content was analyzed by real-time PCR amplification of mtDNA-encoded COX I and nuclear-encoded GAPDH (*, P < 0.05 compared with control). C, FACS analysis of mitochondrial membrane potential. Left, representative graph of the indicated cells; right, the statistical FACS data (*, P < 0.05 compared with control). D, statistical data of mitochondrial ROS generation in the indicated cells (*, P < 0.05 compared with control). E, relative calcineurin (Cn) activity in the indicated cells (*, P < 0.05 compared with control).

Figure 3.

Downregulation of SSBP1 activates mitochondria retrograde signaling. A and B, the relative mtDNA content was analyzed by real-time PCR amplification of mtDNA-encoded COX I and nuclear-encoded GAPDH (*, P < 0.05 compared with control). C, FACS analysis of mitochondrial membrane potential. Left, representative graph of the indicated cells; right, the statistical FACS data (*, P < 0.05 compared with control). D, statistical data of mitochondrial ROS generation in the indicated cells (*, P < 0.05 compared with control). E, relative calcineurin (Cn) activity in the indicated cells (*, P < 0.05 compared with control).

Close modal

Dysfunctional mitochondria involved in increased ROS production and ΔΨm loss in mammalian cells can trigger changes in mitochondrial Ca2+ signaling, reducing the energetic drive for mitochondria Ca2+ import and increasing the cytosolic Ca2+ concentration, thus activating calcineurin (21). For these reasons, we assessed the role of SSBP1 in calcineurin activation and found that SSBP1 downregulation played a role in increased calcineurin activity in both MDA-MB-231 and MDA-MB-468 cells (Fig. 3E). Conversely, SSBP1 overexpression in MDA-MB-231Bo cells decreased calcineurin activity (Fig. 3E). Taken together, these results indicate that SSBP1 loss activates the calcineurin-dependent mitochondrial retrograde signaling pathway in breast cancer cells.

Knockdown of SSBP1 induces c-Rel/p50 nuclear localization and activates TGFβ promoter activity

Next, to identify the intracellular signaling network underlying the enhanced metastasis due to decreased SSBP1 expression, we surveyed potential signaling pathways using a phospho-antibody microarray technique. Analysis of the array revealed the induction of several key components of the NF-κB and TGFβ/SMAD pathways as a result of SSBP1 depletion, including NF-κB-p105/p50 (phospho-Ser893), NF-κB-p105/p50 (phospho-Ser337), SMAD2 (phospho-Ser467), and SMAD3 (phospho-Ser204; Fig. 4A). As mitochondrial retrograde signaling may serve to reprogram the nucleus, we first confirmed activation of the NF-κB pathway. The expression of NF-κB-p105/p50 (phospho-Ser893) was evaluated in the two SSBP1-downregulated breast cancer cell lines MDA-MB-231 and MDA-MB-468 (Fig. 4B). Activation of the canonical NF-κB pathway leads to nuclear translocation of the c-Rel/p50 dimer, which functions as a transcriptional activator (22). Compared with control shRNA-infected MDA-MB-231 cells, SSBP1 shRNA induced the inactivation of IκBβ (Supplementary Fig. S3E), the nuclear translocation of c-Rel/p50, and reduced cytoplasmic c-Rel. Moreover, compared with mock-infected MDA-MB-231Bo cells, SSBP1-overexpressing MDA-MB-231Bo cells reduced nuclear c-Rel and p50 expression as demonstrated by fractionation assays (Fig. 4C). Furthermore, we used MDA-MB-231 cells treated with CCCP- (the mitochondria-specific ionophore that can induce disrupted ΔΨm) and CnA (the catalytic subunit of calcineurin) overexpressing MDA-MB-231 cells as positive controls. Results showed that addition of CCCP and overexpression of CnA caused reduced cytoplasmic IκBβ levels and increased nuclear c-Rel/p50 levels, which was similar as transfection with SSBP1 shRNA. In addition, more results showed that all these changes could be reversed by FK506, a calcineurin inhibitor (23). FK506 treatment was used as a control to ascertain the role of shSSBP1-induced calcineurin activation in IκBβ reduction and c-Rel/p50 nuclear translocation. The addition of FK506 resulted in restoration of IκBβ and decreased c-Rel/p50 in both cells transfected with SSBP1 shRNA and CnA (Supplementary Fig. S3E and S3F).

Figure 4.

Knockdown of SSBP1 induces c-Rel/p50 nuclear localization and activates TGFβ promoter activity. A, phosphoproteome array data showing the fold change of indicated phosphoproteins upon SSBP1 depletion after normalization to total protein expression. The gray-colored area indicates defined induction/reduction boundaries (≤75%/≥150%). B, Western blot analysis of MDA-MB-231 and MDA-MB-468 cells confirming the phosphorylation of p105/50. C, Western blot analysis of c-Rel, p105, p50, p65, in cytoplasmic and nuclear fractions of SSBP1-knockdown MDA-MB-231 and SSBP1-overexpressing MDA-MB-231Bo cells. GAPDH and lamin A were used as cytoplasmic and nuclear markers, respectively. D, scheme showing primer design for the plasmid construction for luciferase assays. E, ChIP analysis for c-Rel binding to the TGFβ promoter in MDA-MB-231 cells transfected with vector carrying c-Rel or negative control. F, dual-luciferase reporter assay analyzing luciferase activity in 293T cells transiently cotransfected with c-Rel, Renilla luciferase, and various TGFβ promoter fragments in the pGL3-basic reporter vector as indicated (**, P < 0.01 compared with control).

Figure 4.

Knockdown of SSBP1 induces c-Rel/p50 nuclear localization and activates TGFβ promoter activity. A, phosphoproteome array data showing the fold change of indicated phosphoproteins upon SSBP1 depletion after normalization to total protein expression. The gray-colored area indicates defined induction/reduction boundaries (≤75%/≥150%). B, Western blot analysis of MDA-MB-231 and MDA-MB-468 cells confirming the phosphorylation of p105/50. C, Western blot analysis of c-Rel, p105, p50, p65, in cytoplasmic and nuclear fractions of SSBP1-knockdown MDA-MB-231 and SSBP1-overexpressing MDA-MB-231Bo cells. GAPDH and lamin A were used as cytoplasmic and nuclear markers, respectively. D, scheme showing primer design for the plasmid construction for luciferase assays. E, ChIP analysis for c-Rel binding to the TGFβ promoter in MDA-MB-231 cells transfected with vector carrying c-Rel or negative control. F, dual-luciferase reporter assay analyzing luciferase activity in 293T cells transiently cotransfected with c-Rel, Renilla luciferase, and various TGFβ promoter fragments in the pGL3-basic reporter vector as indicated (**, P < 0.01 compared with control).

Close modal

Moreover, as a result of SSBP1 downregulation, MDA-MB-231 and MDA-MB-468 cells produced higher amounts of TGFβ protein and mRNA. In contrast, SSBP1 overexpression inhibited TGFβ production in MDA-MB-231Bo cells. Thus, we hypothesized that the c-Rel–p50 complex could activate TGFβ transcription. To investigate this possibility, we performed ChIP assays of SSBP1-downregulated MDA-MB-231 cells followed by PCR of the conserved TGFβ promoter region encompassing the consensus c-Rel–binding motifs. Chromatin immunoprecipitates from the c-Rel antibody were enriched for the TGFβ promoter region compared with those precipitated with an IgG control, confirming that this region contains a c-Rel–binding site. To further identify the role of SSBP1 depletion in regulating TGFβ promoter transcription, a DNA fragment between −1,033 and +316 relative to TGFβ transcription start site (TSS) was cloned into a pGL3-Basic plasmid to yield pGL3–TGFβ recombinant vector delivering TGFβ promoter (Fig. 4D and E). Results showed that sequence constructs P1, P2, and P3 increased the luciferase activity in 293T cells, but sequence P4 remained the luciferase activity similar as the control, indicating that c-Rel was able to increase the transcriptional activity of the TGFβ promoter by binding to the region located between −500 and +316 upstream of the TSS (Fig. 4F). These findings confirm that the loss of SSBP1 induces c-Rel/p50 nuclear localization, leading to increased TGFβ transcription.

SSBP1 regulates TGFβ signaling and TGFβ-induced epithelial-to-mesenchymal transition

Analysis of the antibody array revealed the induction of several key proteins within the TGFβ–SAMD pathway as a result of SSBP1 depletion. We next investigated the effects of SSBP1 modulation on key components of the TGFβ–SMAD pathway and found the elevated production of TGFβ and increased phosphorylation of SMAD3 (S423+S425), a key mediator of TGFβ signaling, in both MDA-MB-231 and MDA-MB-468 SSBP1-deficient cells (Fig. 5A–C). In contrast, SSBP1 overexpression in MDA-MB-231Bo cells markedly reduced TGFβ secretion and downregulated p-SMAD3 protein levels (Fig. 5A–C). Moreover, we examined the expression of other components of the TGFβ pathway, including TGFβ receptor 1 (TβR1) and SMAD7, but no changes were observed (Fig. 5C).

Figure 5.

Downregulation of SSBP1 activates TGFβ signaling and TGFβ-induced EMT. A, relative TGFβ mRNA expression of the indicated cells (mean ± SEM, n = 3 independent experiments; *, P < 0.05 compared with control). B, TGFβ levels in the medium of the indicated cells (mean ± SEM, n = 3 independent experiments; *, P < 0.05 compared with control). C, Western blot analysis of p-SMAD3, SMAD3, TβRI, SMAD7, and fibronectin expression in SSBP1-knockdown and SSBP1-overexpressing cells. D, Transwell assay of the indicated cells treated with GW788388 (10 μmol/L) for 48 hours. E, Western blot analysis of MCF10A cells with stable knockdown of SSBP1 treated with GW788388 (10 μmol/L) for 48 hours. F, Western blot analysis of MCF10A cells with stable knockdown of SSBP1 treated with MitoTEMPO (10 mmol/L) for 48 hours. G, representative images showing that depletion of SSBP1 in MCF10A cells decreases the level of the epithelial marker E-cadherin and increases the level of the mesenchymal marker N-cadherin after treatment with or without for GW788388 (10 μmol/L) for 48 hours.

Figure 5.

Downregulation of SSBP1 activates TGFβ signaling and TGFβ-induced EMT. A, relative TGFβ mRNA expression of the indicated cells (mean ± SEM, n = 3 independent experiments; *, P < 0.05 compared with control). B, TGFβ levels in the medium of the indicated cells (mean ± SEM, n = 3 independent experiments; *, P < 0.05 compared with control). C, Western blot analysis of p-SMAD3, SMAD3, TβRI, SMAD7, and fibronectin expression in SSBP1-knockdown and SSBP1-overexpressing cells. D, Transwell assay of the indicated cells treated with GW788388 (10 μmol/L) for 48 hours. E, Western blot analysis of MCF10A cells with stable knockdown of SSBP1 treated with GW788388 (10 μmol/L) for 48 hours. F, Western blot analysis of MCF10A cells with stable knockdown of SSBP1 treated with MitoTEMPO (10 mmol/L) for 48 hours. G, representative images showing that depletion of SSBP1 in MCF10A cells decreases the level of the epithelial marker E-cadherin and increases the level of the mesenchymal marker N-cadherin after treatment with or without for GW788388 (10 μmol/L) for 48 hours.

Close modal

TGFβ stimulates the epithelial-to-mesenchymal transition (EMT), migration, invasion, and metastasis of breast cancer cells. MCF10A cells are a classical in vitro EMT experimental model (24, 25). MCF10A cells with stable SSBP1 depletion have typical EMT characteristics. Western blot and immunofluorescence assays demonstrated that SSBP1 deficiency resulted in lower E-cadherin expression and higher N-cadherin and fibronectin expression. Of note, treatment with GW788388 and MitoTEMPO, an inhibitor of TGFβ/SMAD3 signaling and mitochondrial ROS, respectively (26–28), mitigated the effects of SSBP1 deficiency on the cell invasive ability and EMT (Fig. 5D–G). In addition, it showed that EMT phenotype changes similarly generated when mtDNA copy number was reduced in MCF10A cells by treated with EB. And treatment with MitoTEMPO reversed the effects of EB on EMT in MCF10A-EB cells (Supplementary Fig. S3G). Furthermore, qRT-PCR analysis of MDA-MB-231 and MDA-MB-468 cells after shRNA-mediated SSBP1 knockdown indicated that SSBP1 depletion was required for the efficient induction of metastasis-related TGFβ target genes, such as IL11, PTHrP, Snail1, and Snail2. Together, these results suggest that SSBP1 depletion potentiates TGFβ/SMAD signaling and TGFβ-driven EMT.

Low SSBP1 expression correlates with SMAD3 activation and poor clinical outcome

To determine the clinical relevance of the above findings in advanced human cancers, we first analyzed the Oncomine database and found that SSBP1 is downregulated in various human cancers (Fig. 1D). On the basis of these data, downregulation or loss of SSBP1 may be an early event in tumorigenesis that contributes to metastasis. In addition, using the cBio database (29), patients with stomach adenocarcinoma and SSBP1 gene alterations have shorter lifespans than those with stomach adenocarcinoma without an SSBP1 alteration in the TCGA cohort.

In light of our observations, we sought to determine whether SSBP1 and p-SMAD3 levels correlated with breast cancer patient prognosis. We examined the expression of SSBP1 and p-SMAD3 in TMAs containing 250 breast tumor specimens by immunohistochemical analysis (Fig. 6A). Of note, patients with low SSBP1 expression were significantly associated with histologic grades (grades III and IV; P = 0.01, Pearson χ2 test; Fig. 6B) and lymph node metastasis (P = 0.009; Fig. 6B). Importantly, low SSBP1 expression and high p-SMAD3 expression, alone or combined, correlated with worse disease-free survival (Fig. 6C–E). Moreover, univariate analysis indicated that low SSBP1 expression at diagnosis was associated with a higher risk for disease relapse (Table 1). Additional multivariate COX analysis exhibited a similar trend as the univariate analysis (Table 1). Finally, we tested the correlation between SSBP1 and p-SMAD3 expression in the same breast tumor samples and observed a significant inverse correlation between their expressions (Spearman ρ = −0.282, P < 0.001; Fig. 6F and G). These data suggest the potential use of SSBP1 for prognostic stratification of patients with breast cancer and confirm our hypothesis that SSBP1 loss promotes human breast cancer metastasis by controlling SMAD-dependent TGFβ signaling.

Figure 6.

Low SSBP1 protein expression correlates with a high p-SMAD3 level and poor breast cancer patient survival. A, immunohistochemical images of the SSBP1 and p-SMAD3 proteins are shown in large (magnification, ×400) and small images (magnification, ×100). B, histogram showing the frequency of low SSBP1 protein expression in patients of different histologic grades of breast cancer and lymph node metastasis statuses (P values determined using the Pearson χ2 test). C–E, cumulative DFS curves with respect to SSBP1 and p-SMAD3 protein expression (P values determined using the log-rank test). F, representative images show that SSBP1 and p-SMAD3 expression is inversely correlated in the same patient. G, the percentage of specimens displaying low or high SSBP1 expression compared with the p-SMAD3 expression level. H, a novel model for the regulation of TGFβ-induced EMT by SSBP1.

Figure 6.

Low SSBP1 protein expression correlates with a high p-SMAD3 level and poor breast cancer patient survival. A, immunohistochemical images of the SSBP1 and p-SMAD3 proteins are shown in large (magnification, ×400) and small images (magnification, ×100). B, histogram showing the frequency of low SSBP1 protein expression in patients of different histologic grades of breast cancer and lymph node metastasis statuses (P values determined using the Pearson χ2 test). C–E, cumulative DFS curves with respect to SSBP1 and p-SMAD3 protein expression (P values determined using the log-rank test). F, representative images show that SSBP1 and p-SMAD3 expression is inversely correlated in the same patient. G, the percentage of specimens displaying low or high SSBP1 expression compared with the p-SMAD3 expression level. H, a novel model for the regulation of TGFβ-induced EMT by SSBP1.

Close modal
Table 1.

Univariate and multivariate survival analysis of SSBP1 expression in breast cancer patients

Univariate analysisMultivariate analysis
HR (95% CI)PHR (95% CI)P
Age 0.978 (0.573–1.668) 0.935 0.933 (0.523–1.665) 0.815 
Timor size 1.373 (0.795–2.371) 0.256 0.953 (0.532–1.707) 0.871 
Lymph node status 2.701 (1.563–4.668) 0.000 2.230 (1.247–3.988) 0.007 
Histologic grade 1.769 (1.031–3.035) 0.038 1.339 (0.744–2.410) 0.330 
ER status 0.815 (0.471–1.409) 0.464 1.379 (0.657–2.895) 0.395 
PR status 0.484 (0.228–1.026) 0.058 0.371 (0.142–0.967) 0.043 
Her-2 status 0.915 (0.530–1.582) 0.752 0.709 (0.378–1.329) 0.283 
SSBP1 0.348 (0.202–0.598) 0.000 0.385 (0.211–0.703) 0.002 
Univariate analysisMultivariate analysis
HR (95% CI)PHR (95% CI)P
Age 0.978 (0.573–1.668) 0.935 0.933 (0.523–1.665) 0.815 
Timor size 1.373 (0.795–2.371) 0.256 0.953 (0.532–1.707) 0.871 
Lymph node status 2.701 (1.563–4.668) 0.000 2.230 (1.247–3.988) 0.007 
Histologic grade 1.769 (1.031–3.035) 0.038 1.339 (0.744–2.410) 0.330 
ER status 0.815 (0.471–1.409) 0.464 1.379 (0.657–2.895) 0.395 
PR status 0.484 (0.228–1.026) 0.058 0.371 (0.142–0.967) 0.043 
Her-2 status 0.915 (0.530–1.582) 0.752 0.709 (0.378–1.329) 0.283 
SSBP1 0.348 (0.202–0.598) 0.000 0.385 (0.211–0.703) 0.002 

NOTE: Bold values are statistically significant (P < 0.05).

Abbreviations: ER, estrogen receptor; PR, progesterone receptor; CI, confidence interval.

In this report, we demonstrate that SSBP1 downregulation induces a significant decrease in the mtDNA copy number in breast cancer cells, which initiates the calcineurin-mediated mitochondrial retrograde signaling pathway to alter nuclear gene expression. The importance of retrograde signaling has been demonstrated for changes in mitochondrial Ca2+ signaling and mitochondrial ROS production (30). Decreases in mitochondrial membrane electrochemical potential reduce the energetic drive for mitochondria to import Ca2+, disrupting Ca2+ homeostasis in the cytoplasm (21, 31). Numerous studies have supported a role for mtDNA defects in tumorigenesis in a wide range of human cancers (32–36), but the upstream regulators of mtDNA alteration remain largely elusive. Here, our data demonstrate that SSBP1 downregulation regulated mtDNA content, activated mitochondrial retrograde signaling, and drove EMT in breast cancer cells. This finding led to the identification of SSBP1 as a novel mitochondrial biomarker for breast tumor progression.

The TGFβ–SMAD pathway has been shown to play a role in tumor progression and EMT. Tumor cells normally secrete abundant TGFβ, which is the most potent inducer of EMT in epithelial cancers and promotes cell invasion and metastasis (37, 38). TGFβ exerts its effects through the activation of its receptor and phosphorylated major effector SMADs (SMAD2/SMAD3) translocate to the nucleus to activate EMT (39). Although substantial data have clarified the downstream molecular networks that trigger EMT through TGFβ, the regulatory controls of the oncogenic functions of TGFβ have not been well elucidated. More importantly, elucidation of the regulatory controls governing the function of TGFβ is critical for the design of novel therapies targeting the oncogenic arm of the pathway. Our data show that SSBP1 loss results in calcineurin-mediated mitochondrial retrograde signaling, which plays a causal role in inducing c-Rel/p50 nuclear localization and activating TGFβ transcription. These results indicate that SSBP1 may serve as a novel negative regulator of TGFβ–SMAD3-driven EMT.

When activated, EMT is thought to be a driver of invasion and metastasis in different types of epithelial cancers and to potentiate cancer cell resistance to current chemotherapies (40). The results from MCF10A cells with SSBP1 downregulation are consistent with the induction of a mesenchymal transdifferentiation program. Loss of SSBP1 in MCF10A cells results in a partial EMT morphology change, including the upregulation of fibronectin and N-cadherin, well-known markers of early EMT induction. Thus, cumulative results indicate that SSBP1 loss may play an early role in the induction of EMT.

In addition, using Oncomine expression analysis, we showed that SSBP1 undergoes a copy-number reduction in many human cancers, including breast cancer, suggesting that the potential importance of SSBP1 in initiation and progression is not limited to breast cancer. On the basis of data from the TCGA cohort, patients with stomach adenocarcinoma and SSBP1 gene alterations have shorter lifespans than those with stomach adenocarcinoma without an SSBP1 alteration. Moreover, p-SMAD3 levels correlated with SSBP1 levels in our breast cancer TMA, further validating that SSBP1 plays an essential role in the control of TGFβ–SMAD pathways. Furthermore, low SSBP1 expression is closely associated with breast cancer metastasis in mouse cancer models. TMAs of samples from breast cancer patients revealed that low SSBP1 and high p-SMAD3 expression correlated with poor prognosis. Our results suggest that further studies should examine the potential clinical utility of SSBP1 and p-SMAD3 as prognostic biomarkers for aggressive disease.

In summary, using a combination of in vivo and in vitro functional metastasis assays and extensive clinical correlation analysis, we provide compelling evidence that SSBP1 is a tumor suppressor involved in breast cancer and provide a novel paradigm for SSBP1-mediated activation of mitochondrial retrograde signaling, increased TGFβ transcription, and subsequent TGFβ–SMAD3-driven cell invasion and cancer metastasis (Fig. 6H). In contrast, in the presence of SSBP1, TGFβ secretion decreases, and the oncogenic TGFβ–EMT signaling is restrained. In recent years, many studies have demonstrated that various TGFβ antagonists have partial success in clinical trials (41, 42). Our findings that SSBP1 deficiency promotes invasion and metastasis, whereas its overexpression restricts this process provide insights into mechanisms and strategies for the therapeutic intervention of the TGFβ pathway and cancer metastasis.

No potential conflicts of interest were disclosed.

Conception and design: H.-L. Jiang, H.-F. Sun, W. Jin

Development of methodology: H.-L. Jiang, H.-F. Sun

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): H.-L. Jiang, H.-F. Sun

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): H.-L. Jiang, H.-F. Sun, S.-P. Gao, L.-D. Li, S. Huang

Writing, review, and/or revision of the manuscript: H.-L. Jiang, H.-F. Sun

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): X. Hu, S. Liu, J. Wu, Z.-M. Shao

Study supervision: W. Jin

The authors thank Dr. Toshiyuki Yoneda (The University of Texas, Houston, TX) for the high-osteolytic metastasis (Bo) MDA-MB-231 Bo cells.

This work was supported by the grants from National Natural Science Foundation of China (31170778 and 81272923) and Program for New Century Excellent Talents in University (NCET-12-0127).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Haffty
BG
,
Yang
Q
,
Reiss
M
,
Kearney
T
,
Higgins
SA
,
Weidhaas
J
, et al
Locoregional relapse and distant metastasis in conservatively managed triple negative early-stage breast cancer
.
J Clin Oncol
2006
;
24
:
5652
7
.
2.
Thike
AA
,
Cheok
PY
,
Jara-Lazaro
AR
,
Tan
B
,
Tan
P
,
Tan
PH
. 
Triple-negative breast cancer: clinicopathological characteristics and relationship with basal-like breast cancer
.
Mod Pathol
2010
;
23
:
123
33
.
3.
Thike
AA
,
Iqbal
J
,
Cheok
PY
,
Chong
AP
,
Tse
GM
,
Tan
B
, et al
Triple negative breast cancer: outcome correlation with immunohistochemical detection of basal markers
.
Am J Surg Pathol
2010
;
34
:
956
64
.
4.
Dent
R
,
Trudeau
M
,
Pritchard
KI
,
Hanna
WM
,
Kahn
HK
,
Sawka
CA
, et al
Triple-negative breast cancer: clinical features and patterns of recurrence
.
Clin Cancer Res
2007
;
13
:
4429
34
.
5.
Bauer
KR
,
Brown
M
,
Cress
RD
,
Parise
CA
,
Caggiano
V
. 
Descriptive analysis of estrogen receptor (ER)-negative, progesterone receptor (PR)-negative, and HER2-negative invasive breast cancer, the so-called triple-negative phenotype: a population-based study from the California cancer Registry
.
Cancer
2007
;
109
:
1721
8
.
6.
Tischkowitz
M
,
Brunet
JS
,
Begin
LR
,
Huntsman
DG
,
Cheang
MC
,
Akslen
LA
, et al
Use of immunohistochemical markers can refine prognosis in triple negative breast cancer
.
BMC Cancer
2007
;
7
:
134
.
7.
Pegram
MD
,
Lipton
A
,
Hayes
DF
,
Weber
BL
,
Baselga
JM
,
Tripathy
D
, et al
Phase II study of receptor-enhanced chemosensitivity using recombinant humanized anti-p185HER2/neu monoclonal antibody plus cisplatin in patients with HER2/neu-overexpressing metastatic breast cancer refractory to chemotherapy treatment
.
J Clin Oncol
1998
;
16
:
2659
71
.
8.
Wiggans
RG
,
Woolley
PV
,
Smythe
T
,
Hoth
D
,
Macdonald
JS
,
Green
L
, et al
Phase-II trial of tamoxifen in advanced breat cancer
.
Cancer Chemother Pharmacol
1979
;
3
:
45
8
.
9.
Carey
LA
,
Dees
EC
,
Sawyer
L
,
Gatti
L
,
Moore
DT
,
Collichio
F
, et al
The triple negative paradox: primary tumor chemosensitivity of breast cancer subtypes
.
Clin Cancer Res
2007
;
13
:
2329
34
.
10.
Hiraga
T
,
Kizaka-Kondoh
S
,
Hirota
K
,
Hiraoka
M
,
Yoneda
T
. 
Hypoxia and hypoxia-inducible factor-1 expression enhance osteolytic bone metastases of breast cancer
.
Cancer Res
2007
;
67
:
4157
63
.
11.
Stark
AM
,
Anuszkiewicz
B
,
Mentlein
R
,
Yoneda
T
,
Mehdorn
HM
,
Held-Feindt
J
. 
Differential expression of matrix metalloproteinases in brain- and bone-seeking clones of metastatic MDA-MB-231 breast cancer cells
.
J Neurooncol
2007
;
81
:
39
48
.
12.
Kaguni
LS
. 
DNA polymerase gamma, the mitochondrial replicase
.
Annu Rev Biochem
2004
;
73
:
293
320
.
13.
Tiranti
V
,
Rocchi
M
,
DiDonato
S
,
Zeviani
M
. 
Cloning of human and rat cDNAs encoding the mitochondrial single-stranded DNA-binding protein (SSB)
.
Gene
1993
;
126
:
219
25
.
14.
Farr
CL
,
Wang
Y
,
Kaguni
LS
. 
Functional interactions of mitochondrial DNA polymerase and single-stranded DNA-binding protein. Template-primer DNA binding and initiation and elongation of DNA strand synthesis
.
J Biol Chem
1999
;
274
:
14779
85
.
15.
Oliveira
MT
,
Kaguni
LS
. 
Functional roles of the N- and C-terminal regions of the human mitochondrial single-stranded DNA-binding protein
.
PLoS ONE
2010
;
5
:
e15379
.
16.
Korhonen
JA
,
Gaspari
M
,
Falkenberg
M
. 
TWINKLE Has 5′ → 3′ DNA helicase activity and is specifically stimulated by mitochondrial single-stranded DNA-binding protein
.
J Biol Chem
2003
;
278
:
48627
32
.
17.
Korhonen
JA
,
Pham
XH
,
Pellegrini
M
,
Falkenberg
M
. 
Reconstitution of a minimal mtDNA replisome in vitro
.
EMBO J
2004
;
23
:
2423
9
.
18.
Shapovalov
Y
,
Hoffman
D
,
Zuch
D
,
de Mesy Bentley
KL
,
Eliseev
RA
. 
Mitochondrial dysfunction in cancer cells due to aberrant mitochondrial replication
.
J Biol Chem
2011
;
286
:
22331
8
.
19.
Wong
TS
,
Rajagopalan
S
,
Townsley
FM
,
Freund
SM
,
Petrovich
M
,
Loakes
D
, et al
Physical and functional interactions between human mitochondrial single-stranded DNA-binding protein and tumour suppressor p53
.
Nucleic Acids Res
2009
;
37
:
568
81
.
20.
Minn
AJ
,
Gupta
GP
,
Siegel
PM
,
Bos
PD
,
Shu
W
,
Giri
DD
, et al
Genes that mediate breast cancer metastasis to lung
.
Nature
2005
;
436
:
518
24
.
21.
Amuthan
G
,
Biswas
G
,
Ananadatheerthavarada
HK
,
Vijayasarathy
C
,
Shephard
HM
,
Avadhani
NG
. 
Mitochondrial stress-induced calcium signaling, phenotypic changes and invasive behavior in human lung carcinoma A549 cells
.
Oncogene
2002
;
21
:
7839
49
.
22.
Hwang
SY
,
Hwang
JS
,
Kim
SY
,
Han
IO
. 
O-GlcN. Acylation and p50/p105 binding of c-Rel are dynamically regulated by LPS and glucosamine in BV2 microglia cells
.
Br J Pharmacol
2013
;
169
:
1551
60
.
23.
O'Keefe
SJ
,
Tamura
J
,
Kincaid
RL
,
Tocci
MJ
,
O'Neill
EA
. 
FK-506 and CsA-sensitive activation of the interleukin-2 promoter by calcineurin
.
Nature
1992
;
357
:
692
4
.
24.
Tran
DD
,
Corsa
CA
,
Biswas
H
,
Aft
RL
,
Longmore
GD
. 
Temporal and spatial cooperation of Snail1 and Twist1 during epithelial–mesenchymal transition predicts for human breast cancer recurrence
.
Mol Cancer Res
2011
;
9
:
1644
57
.
25.
Sarrio
D
,
Rodriguez-Pinilla
SM
,
Hardisson
D
,
Cano
A
,
Moreno-Bueno
G
,
Palacios
J
. 
Epithelial-mesenchymal transition in breast cancer relates to the basal-like phenotype
.
Cancer Res
2008
;
68
:
989
97
.
26.
Petersen
M
,
Thorikay
M
,
Deckers
M
,
van Dinther
M
,
Grygielko
ET
,
Gellibert
F
, et al
Oral administration of GW788388, an inhibitor of TGF-beta type I and II receptor kinases, decreases renal fibrosis
.
Kidney Int
2008
;
73
:
705
15
.
27.
Liang
HL
,
Sedlic
F
,
Bosnjak
Z
,
Nilakantan
V
. 
SOD1 and MitoTEMPO partially prevent mitochondrial permeability transition pore opening, necrosis, and mitochondrial apoptosis after ATP depletion recovery
.
Free Radic Biol Med
2010
;
49
:
1550
60
.
28.
Liu
M
,
Liu
H
,
Dudley
SC
 Jr
. 
Reactive oxygen species originating from mitochondria regulate the cardiac sodium channel
.
Circ Res
2010
;
107
:
967
74
.
29.
Cerami
E
,
Gao
J
,
Dogrusoz
U
,
Gross
BE
,
Sumer
SO
,
Aksoy
BA
, et al
The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data
.
Cancer Discov
2012
;
2
:
401
4
.
30.
Wallace
DC
. 
Mitochondria and cancer
.
Nat Rev Cancer
2012
;
12
:
685
98
.
31.
Correia
RL
,
Oba-Shinjo
SM
,
Uno
M
,
Huang
N
,
Marie
SK
. 
Mitochondrial DNA depletion and its correlation with TFAM, TFB1M, TFB2M, and POLG in human diffusely infiltrating astrocytomas
.
Mitochondrion
2011
;
11
:
48
53
.
32.
Tseng
LM
,
Yin
PH
,
Chi
CW
,
Hsu
CY
,
Wu
CW
,
Lee
LM
, et al
Mitochondrial DNA mutations and mitochondrial DNA depletion in breast cancer
.
Genes Chromosomes Cancer
2006
;
45
:
629
38
.
33.
Xia
P
,
An
HX
,
Dang
CX
,
Radpour
R
,
Kohler
C
,
Fokas
E
, et al
Decreased mitochondrial DNA content in blood samples of patients with stage I breast cancer
.
BMC Cancer
2009
;
9
:
454
.
34.
Copeland
WC
,
Wachsman
JT
,
Johnson
FM
,
Penta
JS
. 
Mitochondrial DNA alterations in cancer
.
Cancer Invest
2002
;
20
:
557
69
.
35.
Chinnery
PF
,
Samuels
DC
,
Elson
J
,
Turnbull
DM
. 
Accumulation of mitochondrial DNA mutations in ageing, cancer, and mitochondrial disease: is there a common mechanism?
Lancet
2002
;
360
:
1323
5
.
36.
Morais
R
,
Zinkewich-Peotti
K
,
Parent
M
,
Wang
H
,
Babai
F
,
Zollinger
M
. 
Tumor-forming ability in athymic nude mice of human cell lines devoid of mitochondrial DNA
.
Cancer Res
1994
;
54
:
3889
96
.
37.
Han
G
,
Lu
SL
,
Li
AG
,
He
W
,
Corless
CL
,
Kulesz-Martin
M
, et al
Distinct mechanisms of TGF-beta1-mediated epithelial-to-mesenchymal transition and metastasis during skin carcinogenesis
.
J Clin Invest
2005
;
115
:
1714
23
.
38.
Kalluri
R
,
Weinberg
RA
. 
The basics of epithelial–mesenchymal transition
.
J Clin Invest
2009
;
119
:
1420
8
.
39.
Wodarz
A
,
Nathke
I
. 
Cell polarity in development and cancer
.
Nat Cell Biol
2007
;
9
:
1016
24
.
40.
Mani
SA
,
Guo
W
,
Liao
MJ
,
Eaton
EN
,
Ayyanan
A
,
Zhou
AY
, et al
The epithelial–mesenchymal transition generates cells with properties of stem cells
.
Cell
2008
;
133
:
704
15
.
41.
Ikushima
H
,
Miyazono
K
. 
TGFbeta signalling: a complex web in cancer progression
.
Nat Rev Cancer
2010
;
10
:
415
24
.
42.
Nagaraj
NS
,
Datta
PK
. 
Targeting the transforming growth factor-beta signaling pathway in human cancer
.
Expert Opin Investig Drugs
2010
;
19
:
77
91
.

Supplementary data