Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 May 2;8(5):569-85.
doi: 10.15252/emmm.201606255. Print 2016 May.

The mitochondrial calcium uniporter regulates breast cancer progression via HIF-1α

Anna Tosatto  1 , Roberta Sommaggio  2 , Carsten Kummerow  3 , Robert B Bentham  4 , Thomas S Blacker  4 , Tunde Berecz  4 , Michael R Duchen  4 , Antonio Rosato  5 , Ivan Bogeski  3 , Gyorgy Szabadkai  6 , Rosario Rizzuto  7 , Cristina Mammucari  8
Affiliations

The mitochondrial calcium uniporter regulates breast cancer progression via HIF-1α

Anna Tosatto et al. EMBO Mol Med. .

Abstract

Triple-negative breast cancer (TNBC) represents the most aggressive breast tumor subtype. However, the molecular determinants responsible for the metastatic TNBC phenotype are only partially understood. We here show that expression of the mitochondrial calcium uniporter (MCU), the selective channel responsible for mitochondrial Ca(2+) uptake, correlates with tumor size and lymph node infiltration, suggesting that mitochondrial Ca(2+) uptake might be instrumental for tumor growth and metastatic formation. Accordingly, MCU downregulation hampered cell motility and invasiveness and reduced tumor growth, lymph node infiltration, and lung metastasis in TNBC xenografts. In MCU-silenced cells, production of mitochondrial reactive oxygen species (mROS) is blunted and expression of the hypoxia-inducible factor-1α (HIF-1α) is reduced, suggesting a signaling role for mROS and HIF-1α, downstream of mitochondrial Ca(2+) Finally, in breast cancer mRNA samples, a positive correlation of MCU expression with HIF-1α signaling route is present. Our results indicate that MCU plays a central role in TNBC growth and metastasis formation and suggest that mitochondrial Ca(2+) uptake is a potential novel therapeutic target for clinical intervention.

Keywords: HIF‐1α; breast cancer; metastasis; mitochondrial Ca2+ uptake; reactive oxygen species.

PubMed Disclaimer

Figures

Figure 1
Figure 1. MCU expression correlates with breast tumor progression and TNBC cell migration
  1. A, B

    Correlation of MCU and MCUb expression levels with breast cancer clinical stages. Median‐centered log2 mRNA expression levels of MCU and MCUb were collected from the TCGA breast cancer dataset (http://tcga-data.nci.nih.gov/docs/publications/brca_2012/). Data were plotted and analyzed against tumor size (T1–T4) (A) and regional lymph node infiltration (N0–N3) (B), according to the AJCC Cancer Staging Manual (7th edition). Linear regression analysis with different stages was implemented. Parameters of linear regression are shown. Numbers of samples for each stage are shown in parentheses.

  2. C–E

    MCU silencing reduces [Ca2+]mit uptake in TNBC cells. Cells were transfected with siMCU or siControl. After 48 h, [Ca2+]mit uptake upon ATP (C, E) or histamine (D) stimulation was measured (n = 10). P‐values: ***P = 0.0008 (C), ***P < 0.0001 (D), ***P = 0.0001 (E), respectively.

  3. F–H

    MCU silencing impairs TNBC cell migration. Cells were transfected with siMCU or siControl. The day after transfection, a linear scratch was obtained on the cell monolayer through a vertically held P200 tip (time point 0 h). Cell migration into the scratched area was monitored 48 h later. The covered area was measured and expressed as a percentage relative to 0‐h time point (n = 12). P‐value: ***P < 0.0001.

  4. I–K

    Cell proliferation is mainly unaffected by MCU depletion. Cells were transfected with siMCU or siControl. Cell number was counted every 24 h for 3 days (the 72‐h time point corresponds to the 48‐h time point of wound healing assay). Results are expressed as ratio R/R0 where R0 is the number of cells at the time of transfection (0‐h time point) (n = 6). P‐value: *P = 0.05.

Data information: In each panel, data are presented as mean ± SD. A two‐tailed unpaired t‐test was performed. See also Appendix Figs S1, S2 and S3.
Figure 2
Figure 2. MCU silencing blunts cell invasiveness without affecting cell viability

  1. Re‐expression of mMCU rescues cell motility of MCU‐silenced cells. Cells were transfected with siMCU or siControl. Ad‐mMCU was used to re‐express MCU (Ad‐GFP was used as a control). MCU protein expression was verified by Western blot. The day after transduction, a linear scratch was made (0‐h time point). Cell migration into the wounded area was monitored at 48‐h time point, and the covered area was measured (n = 12). P‐value: ***P < 0.0001.

  2. MCU silencing blunts cell invasiveness. Stable shMCU‐ and shControl‐expressing spheroids were plated and let grow into collagen I (0‐h time point). Spheroid area was measured at 0 h and 48 h (n = 8). Scale bar: 300 μm. P‐value: ***P = 0.0003.

  3. MCU silencing reduces the clonogenic potential of MDA‐MB‐231 cells. Stable shMCU‐ and shControl‐expressing cells were plated at low confluence (2 × 103/well of a 6‐well plate). After 7 days, the number of colonies was counted (minimum 30 cells/colony, n = 8). P‐value: **P = 0.0027.

  4. MCU depletion does not induce cell death. Cells were transfected with siMCU or siControl. Seventy‐two hours later, cell apoptosis and necrosis were measured by FITC‐Annexin V and propidium iodide (PI) detection (Q1: PI positive, Q2: PI and FITC‐Annexin V positive, Q3: PI and FITC‐Annexin V negative, Q4: FITC‐Annexin V positive; n = 6).

  5. MCU depletion does not alter cell cycle. Cells were transfected with siMCU or siControl. Seventy‐two hours later, cell cycle distribution was monitored by propidium iodide (PI) detection (n = 6).

Data information: In each panel, data are presented as mean ± SD. A two‐tailed unpaired t‐test was performed. See also Appendix Fig S4.
Figure 3
Figure 3. MCU deletion hampers tumor growth and metastasis formation in MDA‐MB‐231 xenografts
Control MDA‐MB‐231 cells and MCU −/− clones 1 and 2 carrying the firefly luciferase reporter gene were injected into the fat pad of SCID mice.

  1. Tumor mass volume was measured at specific time points until the day of sacrifice (day 39 post‐injection for control, day 46 and 56 p.i. for MCU −/− cl.1 and cl.2, respectively). P‐values: (cl.1) ***P = 0.0001, (cl.2) ***P < 0.0001.

  2. Left: in vivo metastasis at the homolateral axillary area of three representative mice per group at the time of sacrifice. Right: total flux analysis. P‐values: **P = 0.01, *P = 0.02.

  3. Lymph nodes weight at the time of sacrifice. P‐values: ***P = 0.0010, **P = 0.0014.

  4. Human cytokeratin 7 (CK7) IHC staining of three representative lymph nodes per group. Scale bar: 500 μm.

  5. Left: images of three representative lungs per group collected ex vivo at the time of sacrifice. Right: total flux analysis. P‐values: **P = 0.0031, ***P = 0.0004.

Data information: In each panel, data are presented as mean ± SE (n = 9 for Control, n = 8 for MCU −/− cl.1, n = 10 for MCU −/− cl.2). A two‐tailed unpaired t‐test was performed. See also Appendix Fig S4.
Figure 4
Figure 4. MCU downregulation alters cellular redox state
  1. A–C

    FLIM analysis of cellular NADH/NADPH levels. Fluorescence lifetimes of NAD(P)H autofluorescence in stable shControl‐ and shMCU‐expressing cells were imaged. Representative images of the distribution of τbound on an intensity weighted pseudocolored scale (2.2–2.5 ns) are shown. Scale bars: 20 μm (A). Mean ± SE of τbound (B) and relative NADH and NADPH intensities (C) calculated from equation in Blacker et al (2014) are shown (n = 3). P‐values: **P = 0.01 (B), **P = 0.002 (C).

  2. D, E

    Measurement of the redox state of the NADH/NAD+ couple. Representative measurements of NADH intensity at steady state and at minimal and maximal reduced state (D). Percentage of the steady‐state redox state (E) (n = 3).

  3. F

    MCU depletion impairs the mitochondrial rate of ATP production. Cells were transfected with siMCU or siControl. Forty‐eight hours later, cells were treated with 5.5 mM 2‐deoxy‐D‐glucose for 1 h and cellular ATP levels were quantified (n = 6). P‐value: ***P = 0.0009.

Data information: In each panel, data are expressed as mean ± SE. A two‐tailed unpaired t‐test was performed.
Figure 5
Figure 5. MCU depletion reduces mitochondrial ROS production
  1. A

    Antioxidant treatments decrease cell migration. A linear scratch was obtained on cell monolayer through a vertically held P200 tip (0‐h time point). Cells were treated for 48 h with N‐acetylcysteine (NAC) or dithioerythritol (DTE). Cell migration into the wounded area was monitored at 48‐h time point, and the covered area was measured (n = 12). P‐values: (DTE 100 μM) **P = 0.008, ***P < 0.0001, (NAC 100 μM) **P = 0.005.

  2. B

    Scavenging of mitochondrial ROS decreases cell migration. A linear scratch was obtained on a cell monolayer through a vertically held P200 tip (0‐h time point). Cells were treated for 48 h with 50 μM MitoTEMPO. Cell migration into the wounded area was monitored at 48‐h time point, and the covered area was measured (n = 12). P‐value: ***P < 0.0001.

  3. C

    MCU silencing does not affect matrix pH. Cells were transfected with siMCU or siControl and SypHer2 probe. Forty‐eight hours later, matrix pH was measured (n = 22).

  4. D, E

    Mitochondrial H2O2 levels are critically blunted after MCU depletion. Cells were transfected with shMCU or shControl, together with the ratiometric YFP‐based biosensor pHyper‐dMito (D) or the mitochondrial H2O2‐sensitive HyPerRed probe (E). Forty‐eight hours later, H2O2 production was measured (n = 35). P‐values: *P = 0.02 (D), *P = 0.05 (E).

  5. F

    Mitochondrial superoxide levels are critically blunted after MCU silencing. Cells were transfected with siMCU or siControl. Forty‐eight hours later, cells were loaded with the red dye MitoSOX™ and superoxide anion levels were measured (n = 25). P‐value: *P = 0.04.

  6. G

    Mitochondrial GSSG/GSH ratio is critically reduced after MCU silencing. Cells were transfected with shMCU or shControl, together with the mitochondrial targeted mitGrx1‐roGFP2 probe. Ninety‐six hours later, the glutathione redox potential (EGSH) was measured (n = 46). P‐value: ***P < 0.0001.

Data information: In panels (A, B), data are expressed as mean ± SD. In panels (C–G), data are expressed as mean ± SE. A two‐tailed unpaired t‐test was performed. Scale bars: 10 μm.
Figure 6
Figure 6. MCU depletion critically affects HIF‐1α levels and signaling
  1. A

    MCU silencing reduces HIF‐1α protein levels. Cells were transfected with siMCU or siControl. HIF‐1α protein levels were detected 48 h later.

  2. B

    MCU silencing reduces MG132‐mediated HIF‐1α and hydroxylated HIF‐1α protein accumulation. Cells were transfected with siMCU or siControl. Forty‐eight hours later, cells were treated with 10 μM of the proteasome inhibitor MG132. Left: Protein levels were revealed by Western blot. Right: quantification by densitometry (n = 5).

  3. C

    ROS increase HIF1A transcription. Cells were treated overnight with 100 μM paraquat to induce ROS production. HIF‐1α mRNA levels were measured by real‐time PCR (n = 3). P‐value: **P = 0.002.

  4. D–J

    MCU silencing reduces mRNA levels of HIF1A, HIF2A, and HIF‐1α target genes. Cells were transfected with siMCU or siControl. mRNA expression was measured by real‐time PCR (n = 3). P‐values: for HIF‐1α **P = 0.0031 (20% O2), **P = 0.009 (1% O2); for HIF‐2α **P = 0.01; for LOX ***P = 0.001 (20% O2), ***P = 0.0005 (1% O2); for PDK1 *P = 0.02, **P = 0.009; for G6PI *P = 0.02, **P = 0.005; for CAIX **P = 0.0026 (20% O2), **P = 0.0022 (1% O2); for HK2 **P = 0.0024, *P = 0.03.

  5. K

    HIF‐1α overexpression rescues siMCU‐mediated migration impairment. Cells were transfected with siMCU or siControl. Wild‐type (wt) and constitutively active (ca) HIF‐1α were expressed by retroviral infection (pBABE was used as a control). The day after transduction, cells were scratched (0‐h time point). Cell migration into the wounded area was monitored at 48‐h time point, and the covered area was measured (n = 12). P‐values: ***P < 0.0001, *P = 0.04.

  6. L, M

    MCU expression levels correlate with HIF1A (L) and HIF‐1α‐regulated genes (M). A linear model (lm) to test the power of MCU expression levels predicting the expression of HIF1A and HIF‐1α‐regulated genes was calculated and plotted for each of the 532 samples of the TCGA database (see Fig 1). Equation and R 2 values of the linear regression and significance indicating deviation from 0 are shown. The area of 95% prediction limit is shaded below and above the linear regression line. The HIF‐1α‐regulated gene set was compiled from Broad Institute GSEA database (http://www.broadinstitute.org/gsea/msigdb/cards/V$HIF1_Q5.html, merged sets of V$HIF1_Q3 and V$HIF1_Q5).

Data information: In each panel, data are expressed as mean ± SD. A two‐tailed unpaired t‐test was performed. See also Appendix Fig S5.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Ainscow EK, Rutter GA (2001) Mitochondrial priming modifies Ca2+ oscillations and insulin secretion in pancreatic islets. Biochem J 353: 175–180 - PMC - PubMed
    1. Baughman JM, Perocchi F, Girgis HS, Plovanich M, Belcher‐Timme CA, Sancak Y, Bao XR, Strittmatter L, Goldberger O, Bogorad RL et al (2011) Integrative genomics identifies MCU as an essential component of the mitochondrial calcium uniporter. Nature 476: 341–345 - PMC - PubMed
    1. Belousov VV, Fradkov AF, Lukyanov KA, Staroverov DB, Shakhbazov KS, Terskikh AV, Lukyanov S (2006) Genetically encoded fluorescent indicator for intracellular hydrogen peroxide. Nat Methods 3: 281–286 - PubMed
    1. Blacker TS, Mann ZF, Gale JE, Ziegler M, Bain AJ, Szabadkai G, Duchen MR (2014) Separating NADH and NADPH fluorescence in live cells and tissues using FLIM. Nat Commun 5: 3936 - PMC - PubMed
    1. Bogeski I, Kappl R, Kummerow C, Gulaboski R, Hoth M, Niemeyer BA (2011) Redox regulation of calcium ion channels: chemical and physiological aspects. Cell Calcium 50: 407–423 - PubMed

Publication types

Substances