Autophosphorylation of DNA-PKCS regulates its dynamics at DNA double-strand breaks

Abstract

The DNA-dependent protein kinase catalytic subunit (DNA-PK(CS)) plays an important role during the repair of DNA double-strand breaks (DSBs). It is recruited to DNA ends in the early stages of the nonhomologous end-joining (NHEJ) process, which mediates DSB repair. To study DNA-PK(CS) recruitment in vivo, we used a laser system to introduce DSBs in a specified region of the cell nucleus. We show that DNA-PK(CS) accumulates at DSB sites in a Ku80-dependent manner, and that neither the kinase activity nor the phosphorylation status of DNA-PK(CS) influences its initial accumulation. However, impairment of both of these functions results in deficient DSB repair and the maintained presence of DNA-PK(CS) at unrepaired DSBs. The use of photobleaching techniques allowed us to determine that the kinase activity and phosphorylation status of DNA-PK(CS) influence the stability of its binding to DNA ends. We suggest a model in which DNA-PK(CS) phosphorylation/autophosphorylation facilitates NHEJ by destabilizing the interaction of DNA-PK(CS) with the DNA ends.

Figure 1.

DNA-PK_CS accumulates at DSB sites induced by laser microirradiation and heavy charged particles. (A) TUNEL labeling (red) and γH2AX immunostaining (blue) of YFP-DNA-PK_CS–expressing V3 cells after microirradiation. (B, top) Time-lapse imaging of YFP-DNA-PK_CS–expressing V3 cells before and after microirradiation. (bottom) Kinetics of relative fluorescence of YFP-DNA-PK_CS accumulation at the DSB site after microirradiation. Each data point is the average of 10 independent measurements. Error bars represent the SD. (C) Time-lapse imaging of YFP-DNA-PK_CS– expressing V3 cells before and after irradiation with uranium-charged particles. (top) Living cells before and 30-s after irradiation. (bottom left) Merged γH2AX staining. (bottom right) Tracks of the uranium particles through the cell nucleus. Arrows point to DNA-PK_CS accumulation sites. (D) Kinetics of DNA-PK_CS accumulation at the DSB site after microirradiation (red) or uranium irradiation (blue). Each data point is the average of 10 independent measurements. Error bars represent the SD. (E) Coimmunostaining of microirradiated YFP-DNA-PK_CS–expressing V3 cells with γH2AX antibody (blue) and phosphospecific antibodies to the Ser-2056 or Thr-2609 amino acid residues of DNA-PK_CS (red). Bar, 10 μm.

Figure 2.

Ku80 is essential for the recruitment of DNA-PK_CS to DSB sites in vivo. (A, left) Time-lapse imaging of YFP-DNA-PK_CS–expressing Xrs6 cells before and after microirradiation. (right) γH2AX staining. Arrow points to the microirradiated site. (B) Time-lapse imaging of Ku80-complemented, YFP-DNA-PK_CS–expressing Xrs6 cells before and after microirradiation. (C) Kinetics of DNA-PK_CS and Ku80 accumulation at the DSB site after microirradiation. Red, YFP-DNA-PK_CS in V3 cells; blue, YFP-DNA-PK_CS in Ku80-complemented Xrs6 cells; green, YFP-Ku80 in Xrs6 cells. Each data point is the average of 10 independent measurements. Error bars represent the SD. (D) Time-lapse imaging of YFP-Ku80–expressing Xrs6 cells before and after microirradiation.

Figure 3.

Impairment of either kinase activity or clustered phosphorylation of DNA-PK_CS leads to reduced DSB repair and the maintained presence of DNA-PK_CS at DSB sites. (A) Summary of known in vitro and in vivo phosphorylation sites of DNA-PK_CS. Asterisks indicate in vivo sites of IR-induced phosphorylation. The modifications we made in our 7A DNA-PK_CS mutant are as follows: S2056A, T2609A, S2612A, T2620A, S2624A, T2638A, and T2647A. (B) Coimmunostaining of KD (left) and 7A cells (right), using phosphospecific antibodies against the S2056, T2609, or T2647 amino acid residues of DNA-PK_CS (red) and against γH2AX (blue). Bars, 10 μm. (C) Initial accumulation kinetics of WT, KD, 7A, and WT + wortmannin at the DSB site after microirradiation. Each data point is the average of 10 independent measurements. Error bars represent the SD. (D) 2-h time-course after microirradiation, showing the kinetics of WT, 7A, and KD in V3 cells and WT in XR1 cells. Each data point is the average of 10 independent measurements. Error bars represent the SD.

Figure 4.

Mutation of the phosphorylation cluster of DNA-PK_CS leads to longer maintained presence of DNA-PK_CS at DSB sites than mutation of single phosphorylation sites. (A) Coimmunostaining of microirradiated S2056A and T2609A cells with phosphospecific antibodies against the S2056, T2609, or T2647 amino acid residues of DNA-PK_CS (red) and against γH2AX (blue). Bars, 10 μm. (B) Initial accumulation kinetics of WT, S2056A, and T2609A at a DSB site after microirradiation. Each data point is the average of 10 independent measurements. Error bars represent the SD. (C) 2-h time-course after microirradiation, showing the kinetics of WT, 7A, S2056A, and T2609A. Each data point is the average of 10 independent measurements. Error bars represent the SD.

Figure 5.

Kinase activity and clustered phosphorylation of DNA-PK_CS influence the exchange rate between DNA-bound and free DNA-PK_CS at DSB sites. (A) Time-lapse imaging of WT cells after photobleaching of the DSB site. The first picture (t = 0 s) is the prebleach situation, the second picture (t = 3 s) was taken immediately after photobleaching. (B) FRAP curves of WT DNA-PK_CS, 7A, and KD at the DSB site. Each data point is the average of 15 independent, normalized measurements. Error bars represent the SD. Prebleach intensity levels were normalized to 1, postbleach intensity levels were normalized to 0. (C) FRAP curves of WT DNA-PK_CS, 7A, and KD in an area of the cell nucleus where no DNA damage is present. Each data point is the average of 10 independent, normalized measurements. Error bars represent the SD. Prebleach intensity levels were normalized to 1, postbleach intensity levels were normalized to 0.

Figure 6.

A model for the DNA-PK_CS autophosphorylation process. The model is based on data presented in this paper, and on previously reported experiments. After the onset of a DSB, the DNA ends are recognized by the Ku70/80 heterodimer, which attracts unphosphorylated DNA-PK_CS. The dynamic exchange between DNA-bound and free DNA-PK_CS takes place at a low rate as long as DNA-PK_CS remains unphosphorylated. This protects the DNA ends from premature processing or ligation. After tethering of the broken DNA ends, DNA-PK_CS autophosphorylation takes place. This effectively liberates the DNA ends, thereby enabling processing and ligation.