Abstract
Introduction
Breast cancer, the malignancy originated from epithelial tissue of mammary gland, has become a worldwide problem and one of the leading causes of cancer death among women. 1 The growing incidence and complicated pathogenesis pose austere challenges to breast cancer treatments in contemporary society.2,3 In this sense, excavating an innovative therapeutic concept has become the urgent affair for the prevention and treatment of breast cancer.
Autophagy, a primary catabolic process that takes place in all eukaryotic cells during metabolic stress or other adverse conditions, is applied to clean the waste protein complexes and damaged organelle by lysosome under the guidance of autophagosome.4,5 Owing to the autophagy, cells maintain homeostasis and accommodate the dynamic microenvironment, which may give rise to malignant transformation, such as survival advantages and drug resistance,6,7 in disease development. Mitophagy refers to mitochondria degradation via autophagy in a selective process, which is indispensable for the well-being of cells by getting rid of the dysfunctional mitochondria and the by-product of respiratory metabolism, such as reactive oxygen species (ROS).8,9 BCL2/adenovirus E1B 19 kDa interacting protein 3 (BNIP3) and BNIP3-like (BNIP3L, namely, NIX) are proteins with homology to Bcl-2 in the BH3 domain. Functionally, they are also known as the receptors that anchored on the surface of mitochondrial outer membrane, which are deeply integrated into each step of mitophagy. At the beginning of mitophagy, BNIP3 and BNIP3L/NIX compete with Beclin 1, the mammalian ortholog of yeast autophagy-related gene 6 (Atg6), which is engaged in the formation of class III phosphatidylinositol 3-kinase (PI3K) complexes, for the interaction of Bcl-2 using their BH3-binding pockets, thereby releasing Beclin 1 and allowing it to induce autophagy.10,11 BNIP3 and BNIP3L/NIX also can recruit LC3 (microtubule-associated protein 1 light chain 3) and gamma-aminobutyric acid receptor-associated protein (GABARAP) to mitochondria, which accelerate the formulation and maturity of autophagosome.12,13 However, some argued that silencing BNIP3 had no effect on the activation of LC3, which implies that BNIP3 may also function in the later stage of mitophagy. 14
Fructose-1,6-bisphosphatase (FBP1), a rate-limiting enzyme in gluconeogenesis, is involved in the catalytic reaction of the biotransformation from fructose-1,6-bisphosphate to fructose 6-phosphate and inorganic phosphate. 15 Recently, with an increasing attention focusing on energy metabolism of tumor cells, emerging evidences prove that FBP1, as a tumor suppressor, 16 makes a critical difference in the occurrence and progression of many malignant neoplasms, such as breast, 17 liver, 18 lung, 19 and pancreatic 20 carcinoma. However, the modulation effects of FBP1 on the biological behavior, especially the activity of mitophagy, and their potential mechanisms remain an enigma in breast cancer cells.
In this study, we first found out FBP1-induced apoptosis by inhibiting mitophagy in MCF-7 and MDA-MB-231 human breast cancer cells, along with its underlying molecular mechanisms focusing on the pairwise crosstalk with Bcl-2, Beclin 1, and hypoxia-inducible factor 1-alpha (HIF-1α)/BNIP3 signaling pathway, which may offer a novel perspective in the cellular and molecular mechanisms of mitophagy and provide a latent target gene for breast cancer treatment.
Materials and methods
Cell culture
Human breast cancer cells MCF-7 and MDA-MB-231 were obtained from the American Type Culture Collection (ATCC, Rockville, MD, USA) and cultivated in Dulbecco’s modified Eagle’s medium (DMEM; Gibco, Carlsbad, CA, USA). The culture mediums were supplemented with 10% fetal bovine serum (Gibco), penicillin (100 U/mL), and streptomycin (100 µg/mL). Both of the two cell lines were incubated at 37°C in the presence of 5% CO2.
Small interfering RNA, plasmid, and transient transfection
For FBP1 knockdown, specific small interfering RNA against FBP1 (siFBP1) or scrambled control siRNA (siNC; GenePharma, Shanghai, China) were transfected into MCF-7 cells. The sequences used were as follows: siFBP1 sense, 5′-AACAUGUUCAUAACCAGGUCG-3′ and anti-sense, 5′-CGACCUGGUUAUGAACAUGUU-3′; siNC sense, 5′-UUCUCCGAACGUGUCACGUTT-3′ and anti-sense, 5′-ACGUGACACGUUCGGAGAATT-3′. For FBP1 overexpression, pcDNA3.1(+) negative control plasmid or pcDNA3.1-Flag-FBP1 plasmid (Public Protein/Plasmid Library, Nanjing, China) was transfected into MDA-MB-231 cells. The transfection was carried out in accordance with the protocol offered by Lipofectamine 2000 (Invitrogen, Grand Island, NY, USA). For quantitative real-time polymerase chain reaction (qRT-PCR), western blot, immunofluorescence staining, flow cytometry, mitochondrial membrane potential (MMP), and transmission electron microscope (TEM) analysis, 2.5 µg of each plasmid or 100 pmol of each siRNA was added into in each well of six-well plates. For immunoprecipitation, 8.0 µg of each plasmid or 600 pmol of each siRNA was added into in each 10-cm dish. After 6 h of incubation, the complex was changed by fresh serum-containing media. The cells were extracted for further detection of messenger RNA (mRNA) and protein expression after 24 and 48 h transfection, respectively.
RNA extraction and quantitative real-time PCR
Total RNA was harvested with TRIzol reagent (TaKaRa, Kusatsu, Japan) and reverse transcribed into cDNA with PrimeScript RT reagent kit (TaKaRa, Kusatsu, Japan) according to the manufacturer’s instruction. The expression of FBP1, HIF1A, BNIP3, and BNIP3L/NIX was quantified by qRT-PCR. All the reactions were performed in a volume of 10 µL in triplicate with the Bio-Rad Miniopticom Real-time PCR system using SYBR Premix EX Taq™ II Kit (TaKaRa). Primers for FBP1, HIF1A, BNIP3, BNIP3L/NIX, and β-actin (house-keeping gene) were designed and synthesized by Invitrogen. The specific sequences of each primer were listed in Table 1. The precise reaction condition was followed by initial denaturation at 95°C for 30 s and 40 cycles of denaturation at 95°C for 5 s, annealing at 53.5°C for 20 s, and extension at 72°C for 30 s. The second derivative maximum method was used to determine the crossing point (Cp) for each sample (in triplicate). The relative expression of each gene was calculated by 2−ΔΔCp. ΔCp referred to Cp of the target gene minus Cp of β-actin gene.
Primer sequences for quantitative real-time PCR.
PCR: polymerase chain reaction.
Western blot
Total cellular proteins from MCF-7 and MDA-MB-231 were obtained by radioimmunoprecipitation assay (RIPA) lysis buffer (Beyotime, Haimen, China) containing 1 mM phenylmethylsulfonyl fluoride (PMSF; Beyotime) as protease inhibitor. The protein concentration was measured by bicinchoninic acid (BCA) protein assay reagent kit (Beyotime). After being mixed with sodium dodecyl sulfate (SDS)-loading buffer (Boster, Wuhan, China) and boiled for 5 min, the protein samples (30 µg each lane) were separated by 12% SDS-polyacrylamide gel electrophoresis (PAGE) gels and transferred to polyvinylidene difluoride (PVDF) membrane (Millipore Corporation, Billerica, MA, USA) at 210 mA for 1 h. Subsequently, the bands blocked by 5% fat-free milk powder were dissolved by Tris-buffered saline containing 0.1% Tween-20 (TBS-T) for 1.5 h at room temperature. Primary antibodies against FBP1 (1:2000; Abcam, Cambridge, MA, USA), BNIP3 (1:2500; Abcam), BNIP3L/NIX (1:2500; Abcam), p62/SQSTM1 (1:3500; Abcam), Bcl-2 (1:3000; Cell Signaling Technology, Danvers, MA, USA), Caspase 3 (1:2500; Cell Signaling Technology), LC3 (1:3500; Cell Signaling Technology), HIF-1α (1:2000; Santa Cruz Biotechnology, Santa Cruz, CA, USA), Beclin 1 (1:2000; Santa Cruz Biotechnology), Bax (1:1500; Bioworld, Nanjing, China), PARP (1:2500; Bioworld), β-actin (1:2000; Bioworld), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH; 1:2000; Bioworld) were incubated with corresponding membranes at 4°C overnight. After TBS-T washing, the membranes were incubated with horseradish peroxidase (HRP)-conjugated goat anti-mouse or rabbit IgG (secondary antibodies; 1:2500; Bioworld) at room temperature for 1 h. After TBS-T washing, the antigen–antibody reaction was visualized by enhanced chemiluminescence (ECL) assay (Millipore Corporation), and the blots were analyzed using Image Lab 3.0 (Bio-Rad, Hercules, CA, USA). Gray value of each protein sample was calculated by Image J 1.50i (National Institutes of Health, USA) and normalized by internal control β-actin or GAPDH.
Immunofluorescence staining
Cells were grown on glass cover slips under common condition. After 48 h of transfection, washed the glass cover slip with pre-cold phosphate-buffered saline (PBS) three times and then fixed with 4% formaldehyde (Solarbio, Beijing, China) for 20 min at room temperature. After that, cells were permeabilized with 0.15% Triton X-100 (Solarbio; diluted by PBS) for 10 min and blocked with 5% goat serum (Solarbio; diluted by PBS) for 1 h at room temperature. Primary antibodies against FBP1 (1:100; Abcam; diluted by 5% goat serum), LC3 (1:200; Cell Signaling Technology; diluted by 5% goat serum), or TOM20 (1:200; Abcam; diluted by 5% goat serum) were applied to stain the cell on glass cover slips at 4°C overnight. After PBS washing, tetraethylrhodamine isothiocyanate (TRITC)-conjugated goat anti-rabbit IgG and fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit IgG (fluorescent secondary antibodies; 1:250; diluted by 0.15% Triton X-100; ZSGB-BIO, Beijing, China) were incubated with primary antibody-labeled cells. At that moment, cells were required to keep in dark and reacted with the fluorescent secondary antibodies for 1 h at room temperature. The nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI; Boster) for 10 min at room temperature away from light. Finally, images were captured using a fluorescence microscope (ECLIPSE Ti-s; Nikon, Tokyo, Japan), and relevant images were taken with a SPOT Diagnostic (Sterling Heights, MI, USA) charge coupled device (CCD) camera.
JC-1 (5,5,6,6-tetrachloro-1,1,3,3-tetraethylbenzimidazolylcarbocyanine iodide) MMP assay
MMP was assessed by a fluorescent probe JC-1 (Beyotime). According to the manufacturer’s directions, cells in each group were stained with 10 µg/mL JC-1 at 37°C for 20 min in the dark. After incubation, cells were rinsed with ice-cold PBS and fixed in 2% paraformaldehyde at 4°C for 5 min. Finally, cells were examined under inverted fluorescence microscope (ECLIPSE Ti-s; Nikon, Tokyo, Japan) and images were captured with a SPOT Diagnostic (Sterling Heights, MI, USA) CCD camera.
Immunoprecipitation
The SureBeads™ Protein G Magnetic Beads (Bio-Rad) were washed with PBS-T (PBS + 0.1% Tween 20) buffer for three times. After magnetization and separation, 8 µg mouse anti-Bcl-2 (Cell Signaling Technology) or mouse IgG was added into final volume of 400 µL PBS and then mixed with 60 µL of the SureBeads™ to each sample. The mixture was resuspended and rotated at room temperature for 30 min. Meanwhile, cells were harvested from 10-cm dishes and lysed with Cell Lysis Buffer for western blotting and immunoprecipitation (IP; Beyotime) containing Protease Inhibitor Cocktail (Biotool, Houston, TX, USA). One-fifth of whole cell lysate was mixed with 5× SDS-loading buffer (Boster) proportionally as INPUT and the rest of lysate was incubated with anti-Bcl-2 pre-bound SureBeads™ at 4°C overnight for immunoprecipitation. After PBS-T washing, magnetization, and separation, the immunoprecipitates were eluted with 2× SDS-loading buffer (Boster) in boiling water bath for 5 min. Finally, the immunoprecipitated protein complexes were analyzed by western blotting.
Flow cytometry
Cells were detached by ethylenediaminetetraacetic acid (EDTA)-free trypsin and resuspended in growth medium. After centrifugation, the cell precipitates were washed with cold PBS for three times and resuspended at a concentration of 1 × 106 cells in 1 mL PBS. For the detection of cellular ROS level, cells were tagged by DCFH-DA probe from Reactive Oxygen Species Assay Kit (Beyotime) at 10 µM, which was diluted by serum-free medium, for 20 min in the common culture condition. After washing with serum-free medium for three times, the fluorescence intensity of each sample was detected by FACScan flow cytometer (BD Biosciences). For the quantification of apoptotic rate, cells were stained by Annexin V-fluorescein isothiocyanate (FITC; BD Pharmingen, San Diego, CA, USA) and propidium iodide (PI; BD Pharmingen) in dark successively according to the manufacturer’s instructions. According to the results presented from flow cytometry, Annexin V-FITC-positive cells were identified as apoptotic cells.
TEM analysis
Cells were collected and fixed in 2% paraformaldehyde and 0.1% glutaraldehyde in 0.1 M sodium cacodylate for 2 h, post-fixed with 1% osmium tetroxide for 1.5 h, washed and stained for 1 h in 3% aqueous uranyl acetate for 1 h. After washing, the samples were dehydrated with a graded alcohol with a series of increasing concentrations and embedded in Epon-Araldite resin (Canemco & Marivac, Lakefield, QC, Canada). Ultrathin sections were cut by ultramicrotome (Reichert-Jung, Inc., Cambridge, UK) and counterstained with 0.3% lead citrate. Finally, the samples were examined under TEM (H-7500; Hitachi, Tokyo, Japan).
Acquisition and analysis of The Cancer Genome Atlas data
Data were downloaded from The Cancer Genome Atlas (TCGA) Breast Cancer database in the UCSC Xena Public Data Hub (http://xena.ucsc.edu/public-hubs/) and visualized or analyzed by Microsoft Excel (Microsoft Corporation, Redmond, WA, USA) and/or the cBio visualization portal (http://www.cbioportal.org).
Statistical analysis
Student’s
Results
FBP1 expression is inversely correlated with autophagic level in breast cancer
To identify the relationship between FBP1 expression and autophagic level in breast cancer, we first detected the basic expression of FBP1 both in luminal (MCF-7, BT-474, T47D, and ZR-75-1) and basal-like (MDA-MB-231, BT-549, MDA-MB-435, and MDA-MB-468) breast cancer cells. According to the results from qRT-PCR (Figure 1(a)) and western blot (Figure 1(b)), we found both the mRNA and protein levels of FBP1 in luminal cells were significantly higher than those in basal-like cells. We also examined the expression of LC3 and p62 by western blot in the mentioned eight cell lines. Contrast to FBP1, the original level of autophagy in luminal cells was greatly lower than that in basal-like cells (Figure 1(c)). To further validate the clinical relevance of the above phenotypes, we first compared the differential expression of FBP1 mRNA in gene-expression array dataset (TCGA Nature 2012) and gene-expression RNAseq dataset (TCGA AWG). The results illustrated that FBP1 expression was distinctly higher in luminal subtype than that in basal-like subtype (Figure 1(d)), which was identical to our previous results from qRT-PCR (Figure 1(a)) and western blot (Figure 1(b)) analysis. Besides, we also took full advantage of TCGA Breast Cancer database to evaluate the gene-expression correlations between FBP1 and some autophagy markers including autophagy-promoting gene, such as microtubule-associated protein 1 light chain 3 beta (MAP1LC3B), BNIP3, and phosphatidylinositol 3-kinase catalytic subunit type 3 (PIK3C3), and autophagy-inhibiting gene, such as RUN and cysteine-rich domain containing Beclin 1 interacting protein (RUBCN). The results suggested that the expression of FBP1 gene was negatively correlated with the expression of MAP1LC3B, BNIP3, and PIK3C3 genes, but positively correlated with the expression of RUBCN gene (Figure 1(e)). Correspondingly, low expression of FBP1 and RUBCN together with high expression of BNIP3 were associated with poor clinical outcome in TCGA breast cancer patients (Supplementary Figure 1). Taken together, these results indicate that the expression of FBP1 and cellular autophagic level presented a negative relationship in breast cancer.
FBP1 expression inversely correlates with autophagic level in breast cancer. (a) The mRNA and (b) protein levels of FBP1 in luminal and basal-like breast cancer cells were detected by qRT-PCR and western blot, respectively. Quantitative analysis of FBP1 mRNA and protein levels after normalization with β-actin. (c) A representative image of western blot and densitometry analysis show the expression of p62 and LC3-II in luminal and basal-like breast cancer cells. GAPDH served as loading control. (d) The expression of FBP1 mRNA in luminal and basal-like breast cancer samples from TCGA Nature 2012 and TCGA AWG databases. Statistical analysis was performed using Student’s t-test. All data are presented as mean ± SD from three independent experiments (**p < 0.01). (e) Correlation between mRNA levels of FBP1 and autophagy markers in TCGA Breast Cancer gene-expression database. Scattering plots with value of sample number (N), Pearson’s r (r), and p value (p) are presented.
FBP1 inhibits autophagy
To investigate the causal relationship between FBP1 and autophagy, we established transient clones with FBP1 overexpression and knockdown in human breast cancer MCF-7 and MDA-MB-231 cells by plasmid and siRNA transfection, respectively. Not unexpectedly, the specific siRNAs targeting FBP1 strikingly reduced the levels of FBP1 in MCF-7 cells. Likewise, the expression of FBP1 was conspicuously upregulated after the transfection of recombinant FBP1-overexpression plasmids in MDA-MB-231 cells (Figure 2(a)). The results of immunofluorescence staining showed that the expression of LC3 was upregulated in FBP1-knockdown MCF-7 cells and downregulated in FBP1-overexpressing MDA-MB-231 cells (Figure 2(b)). More interestingly, we noticed that FBP1 expression significantly decreased the protein level of Beclin 1 (Figure 2(c)). In addition, the accelerative impacts of FBP1-knockdown on autophagy were also close to those of Rapamycin (RAPA, an allosteric mammalian target of rapamycin (mTOR) inhibitor that induces autophagy in the early stage) administration at 50 nM for 24 h. Correspondingly, the inhibitory effects of FBP1-overexpression on autophagy approached to those of 3-methyladenine (3-MA, a class III PI3K inhibitor that effectively blocks autophagy in the early stage) treatment at 8 mM for 24 h. Besides, the co-treatment of FBP1 and 3-MA could develop cumulative effects on autophagy inhibition in breast cancer cells (Figure 2(c)). Collectively, it is clear to see that FBP1 plays a negative role in the regulation of autophagy in breast cancer.
FBP1 represses autophagy. (a) A representative image of western blot and densitometry analysis show the silencing or overexpressing efficiency of FBP1. GAPDH served as loading control. (b) Cells in each group were co-stained with LC3 (green) and FBP1 (red) by corresponding fluorescent antibodies. Nuclei were visualized with DAPI (blue). Cellular morphology was captured in the phase-contrast images. Scale bar represents 20 µm. (c) A representative image of western blot and densitometry analysis shows the expressions of p62, LC3-II, and Beclin 1 in MCF-7 with or without 50 nM RAPA treatment for 24 h and MDA-MB-231 cells with or without 8 mM 3-MA administration for 24 h. GAPDH served as loading control. Statistical analysis was performed using Student’s t-test. All data are presented as mean ± SD from three independent experiments (**p < 0.01). siNC:MCF-7 cells were transfected with scrambled control siRNA by Lipofectamine 2000; siFBP1:MCF-7 cells were transfected with specific siRNA against FBP1 by Lipofectamine 2000; Ctrl:MDA-MB-231 cells were transfected with pcDNA3.1(+) negative control plasmid by Lipofectamine 2000; and FBP1:MDA-MB-231 cells were transfected with pcDNA3.1-Flag-FBP1 plasmid by Lipofectamine 2000.
FBP1 promotes ROS generation and cell apoptosis
To further examine the influences of FBP1-indeced autophagy inhibition on apoptosis in human breast cancer cells, we first measured intracellular ROS level using DCFH-DA probe. The mean fluorescence intensity of MCF-7 and MDA-MB-231 cells in each group was measured by flow cytometry, which certified that knockdown of FBP1 significantly reduced the levels of ROS in MCF-7 cells, whereas FBP1 overexpression induced a substantial increase in ROS level in MDA-MB-231 cells (Figure 3(a)). Meanwhile, on the basis of the dot plot given by flow cytometry, we noticed that the apoptotic rates of MCF-7 cells in FBP1-knockdown group (4.06% ± 0.95%) were much lower than those in control group (8.48% ± 0.94%). Similarly, the apoptotic rates of MDA-MB-231 cells in FBP1-overexpression group (17.59% ± 0.79%) was much higher than those in control group (8.45% ± 1.33%; Figure 3(b)). Consistently, FBP1-expressing MCF-7 and MDA-MB-231 cells had higher expression of pro-apoptotic proteins, such as cleaved PARP, cleaved Caspase 3, and Bax, and lower expression of anti-apoptotic proteins, such as PARP, Caspase 3, and Bcl-2 (Figure 3(c)), which suggested that FBP1 was capable of promoting apoptosis in breast cancer cells.
FBP1 facilitates cellular ROS production and apoptosis. (a) The cellular ROS levels were calculated by flow cytometry and presented as the percentage of control values in the form of mean fluorescence intensity. (b) A representative image of flow cytometric plots and bar graphs show the percentage of apoptosis. (c) Western blot analysis of the expression of apoptosis-related proteins. Quantitative analyses of protein levels are shown as fold changes in optical density compared with total levels and normalized to GAPDH. Statistical analysis was performed using Student’s t-test. All data are presented as mean ± SD from three independent experiments (*significant differences between the indicated groups versus the siNC or Ctrl groups, p < 0.05; **highly different between the indicated groups versus the siNC or Ctrl groups, p < 0.01). siNC:MCF-7 cells were transfected with scrambled control siRNA by Lipofectamine 2000; siFBP1:MCF-7 cells were transfected with specific siRNA against FBP1 by Lipofectamine 2000; Ctrl:MDA-MB-231 cells were transfected with pcDNA3.1(+) negative control plasmid by Lipofectamine 2000; FBP1:MDA-MB-231 cells were transfected with pcDNA3.1-Flag-FBP1 plasmid by Lipofectamine 2000.
FBP1 restrains the removal of disordered mitochondria
To determine the effects of FBP1 on mitophagy in breast cancer cells, we first utilized the JC-1 fluorescent dye to examine the MMP of MCF-7 and MDA-MB-231 cells, which had been treated with carbonyl cyanide m-chlorophenyl hydrazine (CCCP, an inhibitor of oxidative phosphorylation that impairs mitochondrial protein synthesis) at 50 µM for 6 h. The results showed that the ratios of red to green fluorescence intensity were shrunk in cells with high levels of FBP1 expression, which indicated that the FBP1 decreased the proportion of cells with a healthy MMP (Figure 4(a)). Then, we also used TEM to monitor the ultrastructural feature and autophagy of MCF-7 and MDA-MB 231 cells with CCCP administration at 50 µM for 6 h. Under the TEM observation, cells with low level of FBP1 expression displayed the malformed mitochondria completely engulfed by autophagic vacuole and partly separated by phagophore derived from rough endoplasmic reticulum, which illustrated a high proportion of mitophagy. However, cells with high level of FBP1 expression were characterized by severe swelling mitochondria with lacking cristae and destructing inner membranes but without mitophagy (Figure 4(b)). Additionally, we detected the behaviors of translocase of outer membrane 20 (TOM20), a well-established mitochondrial marker, by immunofluorescence staining. The results showed that after the treatment of CCCP at 50 µM for 6 h, the bulk of mitochondria aggregated around the perinuclear region in FBP1-expressing MDA-MB-231 cells but dispersed among the cytoplasm in FBP1-knockdown MCF-7 cells. The results of quantitative analysis also suggested that the mean fluorescence intensity of TOM20 was elevated in FBP1-expressing cells (Figure 4(c)). The above-mentioned results suggested that FBP1 acts as a suppressor during mitophagy.
FBP1 acts as a suppressor during mitophagy. (a) Cells in each group were stained with JC-1 fluorochrome to monitor mitochondrial membrane potentials. Scale bar represents 20 µm. Histograms show semi-quantitation of red/green fluorescence intensity ratios and controls are set to 100%. (b) Mitophagy of MCF-7 and MDA-MB-231 cells was observed by TEM. The disrupted mitochondria surrounded by mitophagosomes (black arrows) and phagophore (white arrows) were readily detected in siFBP1 and Ctrl groups. Pictures in siNC and FBP1 groups revealed the swelling mitochondria with fractured and fuzzy cristae (triangles). (M, normal mitochondria; N, nucleus of MCF-7 and MDA-MB-231 cells.) Scale bar represents 500 nm. (c) Cells in each group were immunostained with TOM20 (green). Nuclei were visualized with DAPI (blue). Scale bar represents 20 µm. Histograms display the mean fluorescence intensities of TOM20 as the percentage of control values. All data are presented as mean ± SD from three independent experiments (*significant differences between the indicated groups versus the siNC or Ctrl groups, p < 0.05; **highly different between the indicated groups versus the siNC or Ctrl groups, p < 0.01). siNC:MCF-7 cells were transfected with scrambled control siRNA by Lipofectamine 2000; siFBP1:MCF-7 cells were transfected with specific siRNA against FBP1 by Lipofectamine 2000; Ctrl:MDA-MB-231 cells were transfected with pcDNA3.1 (+) negative control plasmid by Lipofectamine 2000; and FBP1:MDA-MB-231 cells were transfected with pcDNA3.1-Flag-FBP1 plasmid by Lipofectamine 2000.
FBP1 blocks HIF-1α/BNIP3 pathway and promotes the interaction between Beclin 1 and Bcl-2 in the process of inhibiting mitophagy
To further explore the underlying molecular mechanism of FBP1-mediated mitophagy inhibition in breast cancer cells, we first accessed the mRNA and protein expression of HIF-1α, BNIP3L/NIX, and BNIP3, which are the members in HIF-1α/BNIP3 pathway. As expected, FBP1 gave rise to a remarkable downregulation of HIF-1α, BNIP3L/NIX, and BNIP3 both in mRNA (Figure 5(a)) and protein (Figure 5(b)) levels. Moreover, the results of immunoprecipitation assay pointed out that Bcl-2 dissociated from BNIP3L/NIX and BNIP3 and interplayed with Beclin 1 after FBP1 overexpression (Figure 5(c)). These data demonstrate that FBP1 cripples the activity of mitophagy by suppressing HIF-1α/BNIP3 pathway and enhancing mutual effect between Beclin 1 and Bcl-2 in breast cancer cells.
Roles of HIF-1α/BNIP3 pathway and interaction between Bcl-2 and Beclin 1 in FBP1-mediated mitophagy inhibition. (a) Relative mRNA levels of HIF-1α, BNIP3L/NIX, and BNIP3 were analyzed by qRT-PCR. Quantitative analysis after normalization with β-actin. (b) A representative image of Western blot and densitometry analysis show the relative protein levels of HIF-1α, BNIP3L/NIX, and BNIP3. Quantitative analysis of protein levels are shown as fold changes in optical density compared with total levels and normalized to GAPDH. (c) Whole cell lysates extracted from each group were divided into immunoprecipitation (IP) part and INPUT part by a ratio of four to one. Heavy chain IgG and GAPDH served as loading control in IP and INPUT, respectively. Quantitative analysis of protein levels is shown as fold changes in optical density compared with INPUT levels and presented as a percentage of vector control values. All data are presented as mean ± SD from three independent experiments. (*significant differences between the indicated groups versus the siNC or Ctrl groups, p < 0.05; **highly different between the indicated groups versus the siNC or Ctrl groups, p < 0.01). siNC:MCF-7 cells were transfected with scrambled control siRNA by Lipofectamine 2000; siFBP1:MCF-7 cells were transfected with specific siRNA against FBP1 by Lipofectamine 2000; Ctrl:MDA-MB-231 cells were transfected with pcDNA3.1 (+) negative control plasmid by Lipofectamine 2000; FBP1:MDA-MB-231 cells were transfected with pcDNA3.1-Flag-FBP1 plasmid by Lipofectamine 2000.
Discussion
To investigate the effects of FBP1 on mitophagy and apoptosis, as well as its regulatory mechanisms in breast cancer cells, results in this study suggest that FBP1 inhibits the transcriptional activity of HIF-1α and leads to the decline of BNIP3 and BNIP3L/NIX both in mRNA and protein levels, which disrupts the BNIP3 /NIX-Bcl-2 complex formed in normal condition (Figure 6(a)) but consolidates the combination between Bcl-2 and Beclin 1 (Figure 6(b)). Besides, the expression of Beclin 1 is remarkably downregulated by Bcl-2 after combination, which leads to the blockage of mitophagy and excitation of ROS-induced cell apoptosis (Figure 6(b)).
Hypothetical model depicting the effects of FBP1 on mitophagy and apoptosis in breast cancer cells. (a) In general, the activation of HIF-1α initiates the transcription of BNIP3 and BNIP3L/NIX. The corresponding protein products, the receptors located on mitochondrial outer membrane, compete with Beclin 1 for the binding of Bcl-2, thereby releasing Beclin 1 and allowing it to induce mitochondrial selective autophagy, which is necessary to prevent the accumulation of ROS levels. Ultimately, the timely removal of ROS helps cells survive. (b) The overexpression of FBP1 inhibits the transcriptional activity of HIF-1α. Accordingly, the mRNA and protein expressions of BNIP3 and BNIP3L/NIX are downregulated, which encourages Bcl-2 to dissociate from BNIP3 and BNIP3L/NIX and combine with Beclin 1. After the combination, the expression of Beclin 1 is dramatically suppressed by Bcl-2, which leads to the inhibition of mitophagy and aggregation of ROS. Accumulated poisonous ROS eventually causes cell apoptosis.
Since the concept of Warburg effect was proposed in 1920s, an increasing number of studies have demonstrated that abnormal energy metabolism of tumor cells plays an important role in occurrence and development of many types of cancers.21,22 Aberrant activation of nuclear factor-kappa B (NF-κB) signaling pathway can methylate FBP1 promoter, which impedes the transcription of FBP1 gene and advances the process of glycolysis, resulting in Warburg effect. 23 FBP1, a rate-limiting enzyme in gluconeogenesis, not only impairs the tumorigenicity and cancer stem cell (CSC)-like characteristics from the viewpoint of energy metabolism 24 but more interestingly can affect cell-cycle distribution by acting on cyclin in lung cancer.25,26 Moreover, clinical proteomics analysis in other research also indicated that the loss of FBP1 was strongly related to palindromia after tamoxifen treatment in patients with breast cancer. 27 The results of clinical tissue microarrays showed that autophagy level and LC3-II expression were higher in patients with basal-like breast cancer than in those with any other subtypes of breast cancer, 28 supporting the results shown in Figure 1(c) in our work. However, few studies have explored the relationship between FBP1 expression and autophagy. For the first time, we found that FBP1 expression was negatively correlated with autophagic level both in luminal and basal-like breast cancer cells (Figure 1(a) to (c)). This was further confirmed by statistical correlation analysis using TCGA breast cancer database, and the results were consistent with our conclusion (Figure 1(d) and (e)). Coincidentally, low expressions of FBP1 and high expression of autophagy-promoting marker, such as BNIP3, might be an adverse prognostic factor of breast cancer (Supplementary Figure 1). To further explore effects of FBP1 on mitophagy and apoptosis in breast cancer cells, we knocked down FBP1 in MCF-7 (luminal) cells and overexpressed FBP1 in MDA-MB-231 (basal-like) cells in the following experiments.
During the past decades, pioneering studies have corroborated that the prevalence of drug resistance and the evasion of programmed cell death are related to the upregulation of autophagy.29,30 Inhibiting autophagy is well known to increase the efficacy of therapy during cancer treatment.31–33 Sure enough, data showed in Figure 2 suggest that FBP1 weakened the activity of autophagy, and the negative effects of FBP1 on autophagy were nearly equivalent to those of 3-MA administration in MBA-MB-231 cells. More intriguingly, we also discovered that the co-treatment with FBP1 and 3-MA was more effective than either treatment used alone in inhibiting autophagy, which may be significant for clinical medication and chemoresistance reversal in breast cancer therapy.
Substantial evidences have demonstrated the overlaps in molecular regulators of apoptosis and mitophagy. BNIP3 and BNIP3L/NIX, for example, are two of them.34,35 As pro-apoptotic proteins, BNIP3 and BNIP3L/NIX induce opening of the mitochondrial permeability transition pore and early loss of plasma membrane integrity when cells are subjected to severe stress, which mediates mitochondrial dysfunction and ultimately results in apoptosis.36,37 However, previous studies argued that BNIP3 overexpression only led to a small amount of apoptosis (approximately 10% increase in 48 h) in MCF-7 and MEFs cells under normoxic conditions. Additionally, apoptosis occurred more slowly (more than 48 h) than that in the presence of potent inducers such as Bid and Bax (approximately 12 h). Moderate hypoxia (0.5% and above) increased the expressions of HIF-1α and BNIP3, but had no effect on the long-term survival of HeLa cells, which was also occurred following knockdown of BNIP3 or BNIP3L/NIX under severe hypoxia condition.38,39 Thus, BNIP3 and BNIP3L/NIX cause cell death in a weak and delayed manner, which may explain why downregulation of BNIP3 and BNIP3L/NIX (Figure 5(d) and (e)) and upregulation of Bax, cleaved-PARP, and cleaved Caspase 3 (Figure 3(c)) eventually result in cell apoptosis (Figure 3(b)) in FBP1-expressing cell lines. From another aspect, as reported in other research, 24 our results also showed that FBP1 contributes to increasing levels of cellular ROS (Figure 3(a)), one of the accessory substances of respiratory metabolism that mainly derived from mitochondria; thus, we examined whether the elevated ROS level is associated with a disorder of mitochondrial removal. To verify this hypothesis, we first treated both MCF-7 and MDA-MB-231 cells with CCCP after FBP1 knockdown or overexpression. Next, we demonstrated that cells with higher FBP1 expression failed to be rescued from injured MMP by CCCP administration (Figure 4(a)). We also observe the subcellular structure in MCF-7 and MDA-MB-231 cells by TEM and found that FBP1 protected mitochondria from sequestration by the autophagosome (Figure 4(b)). Besides, we noticed that mitochondria accumulated in FBP1-expressing MDA-MB-231 cells rather than in FBP1-knockdown MCF-7 cells from the results of TOM20-labeled immunofluorescence staining (Figure 4(c)). By this token, the accumulated ROS originated from FBP1-mediated mitophagy inhibition, which may also explain the increase in apoptosis following FBP1 overexpression (Figure 3(b)).
In addition to being the markers of mitophagy,40,41 BNIP3 and BNIP3L/NIX are also the downstream signal molecules in HIF-1α pathway. Under oxygen-deficit conditions, expression of BNIP3 and BNIP3L/NIX is elevated, which triggers mitophagy soon afterwards. 42 Here, we noticed that FBP1 reduces the mRNA (Figure 5(a)) and protein (Figure 5(b)) expression of HIF-1α, together with downstream BNIP3L/NIX and BNIP3 both in MCF-7 and MDA-MB-231 cells, supporting that FBP1 can suppress cell proliferation by inhibiting nuclear HIF function via direct interplay with the HIF inhibitory domain in clear cell renal cell carcinoma. 43 Besides, the results of immunoprecipitation assay also indicated that FBP1 collapsed the formation of BNIP3/NIX-Bcl-2 complex and solidified the interaction between Bcl-2 and Beclin 1 (Figure 5(c)), which implied that the inefficiency of mitophagy by FBP1-mediated repression was accomplished during the initial period. However, the data suggested that 3-MA and FBP1 share a common function phase in autophagy inhibition, which may account for the additive effect on autophagy inhibition by co-treatment (Figure 2(c)). However, we expounded that FBP1-induced Beclin 1 reduction was directly regulated by Bcl-2 repression.
Given this, the increase in cell death induced by FBP1-mediated mitophagy inhibition suggests that crosstalk occurs between autophagy and apoptosis in breast cancer.
As described above, in our present experiments, we disclose a negative correlation between FBP1 expression and autophagic level in breast cancer cells for the first time. FBP1 leads to the inactivation of HIF-1α/BNIP3 pathway and intensive interplay between Beclin 1 and Bcl-2, resulting in ever-increasing ROS level and ensuing apoptosis in breast cancer cells. With all this in mind, FBP1 may be considered to be the stumbling block and promising therapeutic target for breast cancer. One limitation of this study is that we did not evaluate the effects of FBP1 in vivo and in clinical specimens, which will be analyzed in future works.
Footnotes
Declaration of conflicting interests
Funding
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
