PIDD mediates the association of DNA-PKcs and ATR at stalled replication forks to facilitate the ATR signaling pathway

Abstract

The DNA-dependent protein kinase (DNA-PK), consisting of the DNA binding Ku70/80 heterodimer and the catalytic subunit DNA-PKcs, has been well characterized in the non-homologous end-joining mechanism for DNA double strand break (DSB) repair and radiation resistance. Besides playing a role in DSB repair, DNA-PKcs is required for the cellular response to replication stress and participates in the ATR-Chk1 signaling pathway. However, the mechanism through which DNA-PKcs is recruited to stalled replication forks is still unclear. Here, we report that the apoptosis mediator p53-induced protein with a death domain (PIDD) is required to promote DNA-PKcs activity in response to replication stress. PIDD is known to interact with PCNA upon UV-induced replication stress. Our results demonstrate that PIDD is required to recruit DNA-PKcs to stalled replication forks through direct binding to DNA-PKcs at the N' terminal region. Disruption of the interaction between DNA-PKcs and PIDD not only compromises the ATR association and regulation of DNA-PKcs, but also the ATR signaling pathway, intra-S-phase checkpoint and cellular resistance to replication stress. Taken together, our results indicate that PIDD, but not the Ku heterodimer, mediates the DNA-PKcs activity at stalled replication forks and facilitates the ATR signaling pathway in the cellular response to replication stress.

Figure 1.

The PL motif of DNA-PKcs is required for robust DNA replication. (A) Sequence comparison of human DNA-PKcs amino acid 337–350 with the PIP box motif of other PIPs (left panel) and homology with other vertebrates (right panel). Consensus sequence of the PIP box, QxxΨxxϑϑ. Ψ = M, L, I; ϑ = F, Y. The DNA-PKcs PL mutant (mPL) carries alanine substitutions at F345 and Y346. (B) Expression of DNA-PKcs in wild-type CHO AA8 cells, DNA-PKcs deficient V3 cells and V3 cells complemented with empty vector (V3), wild-type DNA-PKcs (WT) and mutants carrying alanine substitutions at the Thr2609 cluster (V3–6A) or the PL motif (V3-mPL). (C) Growth curve of V3 and derived cell lines. Cell numbers were analyzed by a cell counter (Beckman Coulter Z2). Data are presented as mean ± s.d., N = 2. (D) V3-WT and V3-mPL cells were pulse-labeled with iododeoxyuridine (IdU, 10 min) and chlorodeoxyuridine (CldU, 20 min) sequentially, and were analyzed by the DNA fiber assay. The lengths of IdU (red) and CldU (green) tracks were analyzed from >100 ongoing DNA replication tracks.

Figure 2.

The PL motif of DNA-PKcs is required for DNA replication stress response. (A–C) V3 derivative cell lines were analyzed for their colony forming ability against IR (A), UV (B) and HU (C). Data are presented as mean ± s.e.m., N = 3. (D) V3 derived cells were pulse-labeled with 50 μM EdU, treated with HU (2 mM) for 2 h, pre-extracted with 0.1% TX-100, and stained against anti-RPA2 antibody and EdU. Data are presented as RPA2 densities in EdU+ nuclei (N > 150). ****P < 0.0001. (E) V3 derived cells were labeled with BrdU (10 μM) for 30 h and then treated with HU (2 mM) for 2 h. BrdU-labeled single-stranded DNA was stained with anti-BrdU antibody under non-denaturing condition. Data are presented as BrdU densities in nuclei (N > 150). ****P < 0.0001; ***P < 0.001.

Figure 3.

DNA-PKcs PL motif participates in the ATR signaling pathway and intra-S checkpoint. (A) V3-WT and V3-mPL1 cells were exposed to UV (20 J/m²) and harvested at the indicated time points for WB using regular and phospho-specific antibodies against DNA-PKcs. (B) V3-WT and V3-mPL1 cells were pulse-labeled with 50 μM EdU, exposed to UV and then stained against EdU (red) and anti-pT2647 antibody (green). (C and D) Cells were treated with UV or HU and harvested at the indicated time points. Whole cell lysates were analyzed with indicated antibodies. (E) V3-WT and V3-mPL1 cells were sequentially labeled with IdU (100 μM, 10 min) and CldU (100 μM, 20 min) with or without UV exposure in between labels, and then analyzed by DNA fiber assay. The length of DNA tracks labeled with IdU (red) and CldU (green) were measured. The ratios of CldU to IdU in length were calculated from ongoing replication tracks (red-green, N ≥ 100).

Figure 4.

The DNA-PKcs PL motif mediates the DNA-PKcs and PIDD association in vitro and in vivo. (A) Recombinant His-tagged PIDD death domain (a.a. 778–873) was pulled down by the GST fusion protein carrying the wild-type DNA-PKcs N’ terminal fragment (a.a. 1–403) but not the PL mutant fragments (2A or 4A), as indicated. The loading of the GST fusions was demonstrated by Ponceau S staining. (B) V3-WT or V3-mPL1 cells were transfected with FLAG-tagged PIDD-C (a.a. 446–910) or PIDD-CC (a.a. 588–910) constructs. Whole cell lysates were subjected to co-IP with anti-DNA-PKcs antibody and western blotted against anti-DNA-PKcs or anti-FLAG antibodies. (C) V3-WT and V3-mPL1 cells transfected with the FLAG-PIDD construct were UV (20 J/m²) irradiated and harvested at 30 min for PLA (green) using anti-DNA-PKcs and anti-FLAG antibodies. The right panel shows the quantification of PLA spots per nucleus. N > 200. ****P < 0.0001. (D and E) HeLa cells transfected with an IRES-DsRed/FLAG-PIDD plasmid were exposed to UV followed by PLA analysis using anti-FLAG coupled with (D) anti-DNA-PKcs or (E) anti-RPA2 antibodies. Right, quantification. N > 200. ****P < 0.0001.

Figure 5.

PIDD bridges DNA-PKcs and ATR upon replication stress. (A) 293FT cells were transfected with control (siCon) and siRNA against PIDD (siPIDD), and analyzed for PIDD protein expression. (B) Control and siPIDD-transfected 293FT cells were exposed to UV (20 J/m²) and harvested at the indicated time points. Whole cell lysates were subjected to western blot. (C and D) Control and siPIDD-transfected 293FT cells were exposed to UV and harvested at 30 min for PLA (green) using anti-ATR in combination with (C) anti-DNA-PKcs or (D) anti-RPA2 antibodies. The right panel shows the quantification of the PLA spots per nucleus. N > 300. ****P < 0.0001. (E) HeLa cells transfected with the control and siPIDD were exposed to UV. Soluble nuclear protein fraction (S3) and chromatin nuclear matrix fraction (P3) were prepared for WB as indicated. (F) 293FT-transfected siCon and siPIDD were analyzed for the ATR signaling pathway upon UV. (G) HCT116 and derivative DNA-PKcs^−/− cells were treated with HU and analyzed for ATR autophosphorylation.

Figure 6.

Ku80 is dispensable for DNA-PKcs and ATR association in response to UV irradiation. HCT116 Ku86^Flox/− cells were sham treated or incubated with Ad-Cre adenovirus expressing Cre recombinase for 4 days. (A) Cells were exposed to UV (20 J/m²) and analyzed for DNA-PKcs phosphorylation by western blot. (B) Cells were pulse-labeled with EdU, exposed to UV, followed by immunostaining against EdU (red) and anti-pT2647 (green) antibodies. (C) Cells were exposed to UV and analyzed by PLA using anti-ATR and anti-DNA-PKcs antibodies. Right, quantification. Ku86^Flox/− sham, N > 300. ****P < 0.0001.

Figure 7.

PIDD is required for intra-S-phase checkpoint response. (A) HeLa cells were sham-treated or transfected with siPIDD followed by DNA fiber analysis. DNA tracks labeled with IdU (red) and CldU (green) were detected using monoclonal mouse and rat anti-BrdU antibodies, respectively. (B) The lengths of IdU and CldU tracks were analyzed from ongoing replication tracks (red-green, N > 100). (C) The ratios of CldU to IdU in length were calculated from ongoing replication tracks. N > 100.