Figure 1.
Nuclear and cytoplasmic activation of DNA-PK by Dbait32Hc.
Immunostaining of DNA-PK (S2056-P) (green) and γ-H2AX (red) in 8H and Dbait32Hc-treated in (A) MRC-5 cells and (B) and MO59K (DNA-PK+/+) or DNA-PK-deficient M059J (DNA-PK-/-) cells. (C) Immunostaining of Hsp90α (Thr7-P) (green) and γ-H2AX (red) in in 8H and Dbait32Hc-treated MRC-5 cells. (A–C) Cells were fixed 1 h after the end of transfection with 8H or Dbait32Hc molecule. DNA was stained with DAPI (blue). Scale bar: 20 µm. (D) γ-H2AX activation in protein extracts 1h after the end of Dbait32Hc or 8H or γ-irradiated (5 Gy) treatment in M059K (DNA-PK+/+) and M059J (DNA-PK-/-) cells. The protein extracts were subjected to western blotting and the blots were probed with antibodies against γ-H2AX and β-actin. (E) Flow cytometric cell cycle analysis of 8H- or Dbait32Hc-transfected or irradiated (5 Gy) MRC-5 cells. Cells were fixed at the indicated times and DNA was stained with propidium iodide.
Figure 2.
Differential phosphorylation pattern and total protein status after Dbait32Hc treatment.
(A) SDS-polyacrylamide gel electrophoresis after isoelectric focusing (pH range 4.5–5.5) of MRC-5 lysates treated with Dbait32Hc or 8H. Total protein was detected by Sypro Ruby (red, SR) staining and phosphorylation was monitored by Pro-Q Diamond (green, Pro-Q) staining of the same gel. Spots displaying a marked increase in phosphorylation after treatment are highlighted (white arrows). (B) Higher magnification of selected spots from (A) showing higher levels of phosphorylation (arrows) of the indicated proteins after Dbait32Hc treatment than after transfection with the control, 8H. No difference in total protein levels was founds. Proteins displaying at least a 10-fold increase in Pro-Q Diamond staining that could be unambiguously assigned to proteins stained by Sypro Ruby were excised and analyzed by LC-MS/MS. (C) In vitro phosphorylation of vimentin by DNA-PK. Purified DNA-PK (DNA-PKcs and Ku) was incubated with [γ-32P]ATP and the indicated amounts of purified vimentin protein. Dbait32Hc was added where indicated, to activate DNA-PK, and the proteins were then denatured, separated by SDS-polyacrylamide gel electrophoresis and analyzed by autoradiography. (D) Peptides and phosphosites of in vitro DNA-PK-phosphorylated vimentin, as identified by LC-MS/MS with the LTQ-Orbitrap after trypsin digestion. (pT) and (pS) correspond to phosphorylated threonine and serine, respectively.
Table 1 Identified proteins phosphorylated upon DNA-PK activation.
Figure 3.
The Ser459 residue of vimentin is phosphorylated in response to Dbait32Hc.
(A) Double immunostaining of anti-γ-H2AX (red) and anti-vimentin (S459-P) (green) in MRC-5 cells transfected with Dbait32Hc or 8H or irradiated (10Gy). The arrow indicates a cell negative for both γ-H2AX and phosphorylated vimentin (S459). (B) MRC-5, SK28, M059K or DNA-PK-deficient M059J cells were transfected with Dbait32Hc or 8H or irradiated (IR) with 10 Gy, as indicated. Proteins were extracted 1 h after the end of treatment for western blot analysis and the lysates were probed for vimentin (Vim-S459-P), γ-H2AX and β-actin. (C) MRC-5 cells were treated 2 h before and during Dbait32Hc or 8H treatment with 10 µM Nu7026 (DNA-PK inhibitor), 20 µM wortmannin (PIKK inhibitor), 10 µM KU-55933 (ATM inhibitor) or vehicle (DMSO). The total cell extracts were processed and probed as in (B). (D) The indicated cell lines were transfected with Dbait32Hc or 8H, fixed 1 h after the end of transfection and immunostained for vimentin (Vim-S459-P) (green) and γ-H2AX (red). (E) Double immunostaining of 32Hc- or 8H-treated SK28 cells with antibodies against total vimentin (Vim, red) and phosphorylated vimentin (Vim-S459-P, green). DNA was stained with DAPI (blue). Scale bar: 20 µm.
Figure 4.
Diagram of the domain structure of vimentin and C-terminus conservation.
Top, vimentin consists of a central α-helical rod domain, flanked by distal head (N-terminal) and tail (C-terminal) regions. Bottom, alignment of the sequence of the C-terminus of human vimentin (residues 441-466) with those of vimentin from various species. The human serine 459 residue is highlighted by gray boxes; residues differing between species are shown in light gray.
Figure 5.
DNA-PK-independent phosphorylation of the Ser459 residue of vimentin in M-phase.
(A) Merged images of immunostaining for vimentin (S459-P) (green) and DNA-staining with DAPI (blue) in untreated MRC-5 cells at the indicated stages of mitosis. (B) Immunostaining as in (A) of MO59K (DNA-PK+/+) or DNA-PK-deficient M059J (DNA-PK-/-) cells and ATM-defective AT5BI cells. (C) Immunostaining as in (A) of MRC-5 cells treated for 5 h with 10 µM NU7026 (DNA-PK inhibitor), 20 µM wortmannin (PIKK and PI3K inhibitor), 10 µM KU-55933 (ATM inhibitor) or (D) 60 µM roscovitine (CDK inhibitor). (E) MOS depletion does not affect vimentin phosphorylation at Ser459 during mitosis. HeLa cells were transfected with a control (CTL) or Mos-silencing siRNAs for 72h, followed by fixation and immunostaining as in A. (A-E) Cells in mitosis (arrow) or interphase are shown. Scale bar: 20 µm. (F) MOS depletion verification. HeLa cells were transfected with a control (CTL) or Mos-silencing siRNAs for 72h, followed by protein extraction, which were subjected to western blotting and the blots were probed with antibodies against MOS and β-actin. (G) MOS expression level in MRC-5 cells and HeLa cells. (H) MOS depletion does not affect Dbait32Hc-induced vimentin phosphorylation on Ser459. HeLa cells were transfected as described previously (E), 72h after transfection cells were treated with Dbait32Hc or 8H followed by proteins extraction, then subjected to western blotting and the blots were probed with vimentin (Vim-S459-P), γ-H2AX and β-actin.
Figure 6.
Inhibition of cell migration and adhesion by DNA-PK activation.
(A) SK-28 cells were transfected with Dbait32Hc or control 8H, fixed 1 h after the end of transfection and immunostained for vimentin (red) and phospho-vimentin (S459-P) (green). A higher magnification of typical cell protrusions is shown. The bar chart provides a quantitative representation of the signal intensity ratio of vimentin (S459-P) and total vimentin for cell protrusions and cytoplasm, with standard deviations, from three analyses of 50 cells per set of conditions. DNA was stained with DAPI (blue). Scale bar: 20 µm. (B) The indicated cell lines were transfected with Dbait32Hc or 8H, wounded 1 h after the end of transfection and wound healing was observed by videomicroscopy for 16 h. The bar chart gives the percentage wound closure, with standard deviations, from three independent experiments. (C) The indicated cell lines were transfected with Dbait32Hc or 8H, then 1 h after the end of transfection cells were detached by trypsinization and let to adhere on fibronectin for 1h. The bar chart gives the relative adhesion to fibronectin measure by colorimetric assay, with standard deviations, from two independent experiments (Ab560). The significance of differences was assessed by the Kruskal-Wallis method. *, p<0.0005.