DNA-PK phosphorylation of RPA32 Ser4/Ser8 regulates replication stress checkpoint activation, fork restart, homologous recombination and mitotic catastrophe

Abstract

Genotoxins and other factors cause replication stress that activate the DNA damage response (DDR), comprising checkpoint and repair systems. The DDR suppresses cancer by promoting genome stability, and it regulates tumor resistance to chemo- and radiotherapy. Three members of the phosphatidylinositol 3-kinase-related kinase (PIKK) family, ATM, ATR, and DNA-PK, are important DDR proteins. A key PIKK target is replication protein A (RPA), which binds single-stranded DNA and functions in DNA replication, DNA repair, and checkpoint signaling. An early response to replication stress is ATR activation, which occurs when RPA accumulates on ssDNA. Activated ATR phosphorylates many targets, including the RPA32 subunit of RPA, leading to Chk1 activation and replication arrest. DNA-PK also phosphorylates RPA32 in response to replication stress, and we demonstrate that cells with DNA-PK defects, or lacking RPA32 Ser4/Ser8 targeted by DNA-PK, confer similar phenotypes, including defective replication checkpoint arrest, hyper-recombination, premature replication fork restart, failure to block late origin firing, and increased mitotic catastrophe. We present evidence that hyper-recombination in these mutants is ATM-dependent, but the other defects are ATM-independent. These results indicate that DNA-PK and ATR signaling through RPA32 plays a critical role in promoting genome stability and cell survival in response to replication stress.

Keywords: Checkpoint regulation; DNA repair; Homologous recombination; Replication stress.

Figure 1

Etoposide-induced RPA32 phosphorylation depends on ATR and DNA-PK but not ATM. (A) UM-SCC-38 cells stably expressing WT RPA32 were treated for 3 h with KU55933 (ATMi), NU7026 (DNA-PKi), or 48 h with siRNA targeting ATR, then treated for 2 h with 20 μM etoposide. Perhaps because of drug uptake or other factors, more typical NU7026 concentrations (10–20 μM) were ineffective (data not shown). Cells were harvested before etoposide treatment, and 0–24 h after release. Inhibitors remained in the medium during etoposide treatment and recovery. Protein from whole-cell lysates were separated by SDS-PAGE and immunoblotted with antibodies to RPA32. (B) Western blot of ATR from cells treated with ATR-siRNA and/or etoposide; nucleolin was used as a loading control. (C) RPA32 was analyzed as in panel A except cells were not treated with etoposide.

Figure 2

Etoposide-induced cell cycle arrest depends on RPA32 Ser4/Ser8 phosphorylation, ATR, and DNA-PK, but not ATM. UM-SCC-38 cells stably expressing WT or S4A/S8A RPA32 were treated with etoposide as described in Fig. 1. Cells were treated for 3 h with KU60019 (ATMi), VE-821 (ATRi) or NU7441 (DNA-PKi) prior to, during, and after etoposide treatment as indicated. Cells were harvested before etoposide treatment, and 0–24 h after release and cell cycle profiles of propidium iodide stained cells were determined by flow cytometry. (A) PIKK roles in etoposide-induced cell cycle arrest in cells expressing WT RPA32. For each treatment group, representative cell cycle profiles are presented above. G1 cells are the left-most peak of each profile; average percentages of G1 cells (±SEM) for 3 determinations are shown below. Dashed line indicates percent G1 cells 24 h after release. Statistics calculated using t tests are shown for etoposide treated vs etoposide + PIKK inhibitor treated cells at 18 and 24 h: * indicates P < 0.05 and ** indicates P < 0.02, and P values trending toward significance are shown. (B) Cell cycle profiles and percent G1 cells for etoposide-treated RPA32 S4A/S8A cells as in panel A. Statistics shown are for comparisons of WT vs S4A/S8A mutant cells in each treatment group at 24 hr; WT values (from panel A) are shown as hatched bars to illustrate these comparisons. * indicates P < 0.05 and ** indicates P < 0.02. At 24 h, there were no statistically significant differences in S4A/S8A cells treated vs not treated with any of the PIKK inhibitors.

Figure 3

DNA-PKcs suppresses replication stress-induced HR in an ATM-dependent manner. (A) Representative metaphase spreads of DNA-PKcs KD mutant cells treated with 5 mM HU or mock-treated. Arrows indicate SCEs. (B) Quantitation of SCEs in DNA-PKcs WT, null, and KD cells following HU or mock treatment. (C) Direct repeat HR frequencies in untreated DNA-PKcs WT, null, and KD cells, or treated with CPT or HU. Values are averages (±SEM) for 2–5 determinations; * and ** indicate P<0.05 or P<0.01, respectively (t tests).

Figure 4

DNA-PKcs and RPA32 regulate replication recovery after stress. (A) WT, DNA-PKcs null and KD cells were treated for 60 min with 5 mM HU, HU was removed, 10 μM BrdU was added, and BrdU incorporation was detected with immunofluorescence/flow cytometry. (B) DNA fiber analysis procedure is outlined above and representative images are shown. The plot includes data based on >250 fibers scored for each of WT, DNA-PKcs null and KD cells. Values are averages (±SEM) for 4–6 images. ** indicates P<0.01 (t tests). (C) BrdU incorporation in WT and RPA32 S4A/S8A cells procedures are outlined above for controls and replication recovery after etoposide. Histograms of flow cytometry data scoring BrdU incorporation and DNA content by PI staining. Points above diagonal line (green) represent BrdU positive cells. Both cell lines are replication competent prior to etoposide (control). (D) RPA32 S4A/S8A abrogates S/G2 arrest after release from cis-platin-induced replication stress. Procedure is outlined above, and cell cycle profiles of PI-stained cells were determined by flow cytometry. Percent G1 cells for each condition are shown in black boxes.

Figure 5

DNA-PKcs – RPA32 signaling suppresses mitotic catastrophe after replication stress. (A) Representative images of WT, DNA-PKcs null and KD cells after 16 h treatment with 10 mM HU and staining with DAPI (blue) and anti-tubulin antibodies (green). A higher HU concentration was used in this experiment to increase sensitivity to detect abnormal cells. Arrows indicate anaphase bridges (failed cytokinesis) and giant cells. (B) Average percentage (±SEM) of abnormal cells for 4 determinations. (C) Images of mitotic RPA32 S4A/S8A and WT cells stained with DAPI following etoposide or mock treatment. WT cells treated with etoposide are not shown because they arrest before M phase. In etoposide-treated cells, arrows indicate anaphase bridges (left and center) and nuclear blebs (left and right).