Research ArticleCELL STRESS

Mitochondrial redox sensing by the kinase ATM maintains cellular antioxidant capacity

See allHide authors and affiliations

Science Signaling  10 Jul 2018:
Vol. 11, Issue 538, eaaq0702
DOI: 10.1126/scisignal.aaq0702
  • Fig. 1 Mitochondrial ROS promote redox-dependent ATM dimerization.

    (A) MitoSOX staining by flow cytometry in HeLa cells treated with vehicle (−) or 50 μM MitoTEMPO (MT; +) for 24 hours before being treated with vehicle (−) or 20 μM menadione (MD; +) for 30 min. Data are mean fluorescence intensities ± SD of biological triplicates. *P < 0.05 by Student’s t test. (B) Western blot analysis of HeLa cells treated with vehicle (−) or 25 or 50 μM MitoTEMPO (+) for 2 hours before treatment with vehicle (−) or 20 μM MD (+) for 30 min. ATM monomers (ATM-M) and disulfide-linked dimers (ATM-D) were resolved by nonreducing SDS-PAGE. Mitochondrial HSP60 was probed as a loading control. (C) Primary MEFs from wild-type (WT) or mitochondrial catalase–overexpressing (mCAT) mice treated with vehicle or 20 μM MD for 30 min were analyzed for cellular ROS using 2′,7′-dichlorofluorescein diacetate (DCFDA) and flow cytometry as described in (A). (D) Western blots of ATM and HSP60 of MEFs treated as in (C) were performed as described in (B), except that catalase was also probed to demonstrate its overexpression. Blots are representative of three experiments.

  • Fig. 2 Conversion of mitochondrial superoxide to membrane-permeable hydrogen peroxide is necessary for ATM dimerization.

    (A) MitoSOX staining in HeLa cells treated with vehicle (veh), 20 or 40 μM MD, 2 μM antimycin A (AA), or 0.4 μM rotenone (ROT) for 30 min. Data are mean fluorescence intensities ± SD of biological triplicates. **P < 0.01 by Student’s t test. (B) MitoPY1 staining in HeLa cells treated with vehicle, 20 or 40 μM MD, 2 μM antimycin A, or 0.4 μM rotenone for 30 min. Plotted is the mean fluorescence intensity ± SD of biological triplicates. **P < 0.01 by Student’s t test. (C) Western blot analysis of HeLa cells treated with vehicle (−), 20 or 40 μM MD, 2 μM antimycin A, or 0.4 μM rotenone for 30 min. ATM monomers and dimers and PRDX1 monomers (PRDX1-M) and dimers (PRDX1-D) were resolved by nonreducing SDS-PAGE. Calnexin was probed as a loading control. Blots are representative of three experiments.

  • Fig. 3 Mitochondrial hydrogen peroxide–induced ATM dimers are in the nucleus but do not interfere with ATM chromatin association.

    (A) Western blot analysis of HeLa cells expressing short hairpin RNA (shRNA) targeting a scrambled sequence (shSCR) or ATM (shATM) treated with 20 μM MD (+) or vehicle (ethanol; −) for 30 min or 0.75 μM topotecan for 1 hour, as indicated. ATM dimer, ATM monomer, CHK2, phosphorylated CHK2 Thr68 (P-CHK2), and phosphorylated H2AX Ser139 (P-H2AX) were probed, along with HSP60 as a loading control. (B) HeLa cells treated with vehicle (ethanol; −) or 30 μM MD (+) for 1 hour were separated into nuclear (N) and postnuclear (PN) fractions, which were analyzed by Western blot along with a cell-equivalent amount of whole-cell (WC) lysate. ATM monomer, ATM dimer, histone H3, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and HSP60 were analyzed. (C) HeLa cells treated with vehicle (−) or 20 μM MD (+) for 30 min and/or 0.75 μM topotecan for 1 hour were subjected to salt extraction of chromatin-bound proteins. An equivalent volume of the supernatant fraction (S) and the pellet (P) fraction (containing chromatin-bound proteins) was loaded for Western blot. ATM, phosphorylated ATM Ser1981 (P-ATM), histone H3, GAPDH, and HSP60 were analyzed. Blots are representative of two experiments.

  • Fig. 4 TRX1 negatively regulates ATM redox dimerization.

    (A) Western blot analysis of HeLa cells expressing shRNA-targeting enhanced green fluorescent protein (shEGFP) as a negative control or TRX1 (shTRX1) treated with vehicle or the indicated concentration of MD for 30 min. Blot was probed for ATM monomer, ATM dimer, and TRX1, with calnexin as a loading control. Blots are representative of two experiments. (B) DCFDA staining by flow cytometry, in cells treated with vehicle, 5 or 20 μM MD as described in (A). Plot is representative of three experiments using biological duplicate samples in each. (C) Western blot of HeLa cell treated with 20 μM MD for 30 min after which the medium was changed and cells were allowed to recover for 0, 10, 20, 30, and 40 min, as indicated. Blot was probed for ATM monomer, ATM dimer, and TRX1, with HSP60 and GAPDH as loading controls. Blots are representative of three experiments. (D) Western blot of HeLa cells transfected with control (siCtrl) or siRNA-targeting PRDX1 and PRDX2 for 72 hours and treated with vehicle (−) or 20 μM MD (+) for 30 min. Blot was probed for ATM monomer, ATM dimer, PRDX1, and PRDX2, with calnexin as a loading control. Blots are representative of three experiments.

  • Fig. 5 Redox sensing by ATM up-regulates antioxidant capacity and G6PD expression.

    (A and B) Analysis of U2OS cells transfected with empty vector (vector) and control shRNA (shCtrl) or overexpressing WT or the C2991L (CL)–mutant ATM while having their endogenous ATM knocked down by shRNA-targeting ATM (shRNA1) were treated with vehicle or 20 μM MD for 30 min. Cellular and mitochondrial ROS were then measured by DCFDA (A) or MitoSOX (B) staining and flow cytometry. Plotted is the mean fluorescence intensity ± SD of biological triplicates. *P < 0.05 and **P < 0.01 by Student’s t tests. (C) G6PD activity. Plotted is the mean ± SD of technical quadruplicates from a representative of three experiments. **P < 0.01 by Student’s t tests. (D) G6PD mRNA abundance by quantitative polymerase chain reaction (qPCR). Plotted is the mean ± SD of biological triplicates. **P < 0.01 by Student’s t tests. (E) Western blot of G6PD protein. Here, endogenous ATM was knocked down by one of two shRNAs (1 and 2). Calnexin was probed as a loading control. Blots are representative of three experiments.

  • Fig. 6 Cells deficient in ATM redox signaling flux less glucose through the PPP and have increased total glutathione.

    (A) G6PD activity limits metabolic fluxes through the PPP, which generates pentose phosphate compounds. Generation of these compounds was greater in cells expressing the WT allele of ATM. (B) Amount of light (M + 0) and heavy (M + 2) pentose phosphate isobars 6 and 24 hours after cells were labeled with heavy glucose in U2OS cells expressing either WT or C2991L-mutant ATM after endogenous ATM knockdown by each of two shRNAs (1 and 2). Plotted is the mean ± SD of biological triplicates. (C) Ratio of the G6PD substrate [glucose-6-phosphate (G6P)] and its downstream product 6-phosphogluconate (6PG) were measured and plotted as described in (B). (D) Amount of heavy glutathione plotted as described in (B). *P < 0.05 and **P < 0.01 by Student’s t tests.

Supplementary Materials

  • www.sciencesignaling.org/cgi/content/full/11/538/eaaq0702/DC1

    Fig. S1. DCFDA staining of TRX1 knockdown cells and ATM dimerization in PRDX1 and PRDX2 knockout cells.

    Fig. S2. Additional characterization of the ATM U2OS cell system and replicate experiments with a second ATM shRNA.

    Fig. S3. ATM C2991L-mutant cells consume more glucose for glycolysis.

    Fig. S4. ATM C2991L-mutant cells flux more carbon from glucose to the TCA cycle.

    Fig. S5. ATM C2991L-mutant cells have more total glutathione than WT cells, but GSH/GSSG ratios are similar.

    Table S1. Sequences of sgRNAs used to knockout PRXD1 or PRDX2 in HeLa cells.

  • This PDF file includes:

    • Fig. S1. DCFDA staining of TRX1 knockdown cells and ATM dimerization in PRDX1 and PRDX2 knockout cells.
    • Fig. S2. Additional characterization of the ATM U2OS cell system and replicate experiments with a second ATM shRNA.
    • Fig. S3. ATM C2991L-mutant cells consume more glucose for glycolysis.
    • Fig. S4. ATM C2991L-mutant cells flux more carbon from glucose to the TCA cycle.
    • Fig. S5. ATM C2991L-mutant cells have more total glutathione than WT cells, but GSH/GSSG ratios are similar.
    • Table S1. Sequences of sgRNAs used to knockout PRXD1 or PRDX2 in HeLa cells.

    [Download PDF]

Stay Connected to Science Signaling

Navigate This Article