Cooperation of DNA-PKcs and WRN helicase in the maintenance of telomeric D-loops

Abstract

Werner syndrome is an inherited human progeriod syndrome caused by mutations in the gene encoding the Werner Syndrome protein, WRN. It has both 3'-5' DNA helicase and exonuclease activities, and is suggested to have roles in many aspects of DNA metabolism, including DNA repair and telomere maintenance. The DNA-PK complex also functions in both DNA double strand break repair and telomere maintenance. Interaction between WRN and the DNA-PK complex has been reported in DNA double strand break repair, but their possible cooperation at telomeres has not been reported. This study analyzes thein vitro and in vivo interaction at the telomere between WRN and DNA-PKcs, the catalytic subunit of DNA-PK. The results show that DNA-PKcs selectively stimulates WRN helicase but not WRN exonuclease in vitro, affecting that WRN helicase unwinds and promotes the release of the full-length invading strand of a telomere D-loop model substrate. In addition, the length of telomeric G-tails decreases in DNA-PKcs knockdown cells, and this phenotype is reversed by overexpression of WRN helicase. These results suggest that WRN and DNA-PKcs may cooperatively prevent G-tail shortening in vivo.

Figure 1.. D-loop unwinding by WRN in the absence and presence of DNA-PKcs.

(A) The D-loop substrate consisted with INV, BT and BB. 5'-end of INV was radiolabeled as indicated by asterisk. WRN (3.3 nM, lanes 3-6) and increasing amounts of DNA-PKcs (0.67 nM, lane 4; 3.3 nM, lanes 5 and 7; 16.7 nM, lane 6) were incubated in standard reaction buffer prior to addition of the telomeric D-loop substrate. Reaction products were analyzed by native (B) or denaturing gel electrophoresis (C). Lanes 1 in (B) and (C): A DNA ladder marker.

Figure 2.

Differential Effect of DNA-PKcs on WRN helicase and exonuclease activities. (A) WRN (3.3 nM, lanes 3-5) and DNA-PKcs (3.3 nM, lane 4; 16.7 nM, lanes 5 and 6) were incubated in standard reaction buffer lacking ATP prior to addition of the D-loop substrate. Reaction products were analyzed by denaturing gel electro-phoresis. Lanes 1 and 7: A DNA ladder marker. (B) WRN (E84A) (3.3 nM, lanes 3-5) was preincubated with either DNA-PKcs (16.7 nM, lane 4) or Ku (3.3 nM, lane 5) in standard reaction buffer prior to addition of the D-loop substrate. Reaction products were analyzed by native gel electrophoresis. Lane 1: heat-denatured D-loop substrate denoted by a filled triangle. Lane 6: A DNA ladder marker.

Figure 3.

Differential effect of DNA-PKcs on WRN and BLM helicase activities. BLM (3.3 nM, lanes 3-5) and either DNA-PKcs (16.7 nM, lane 4) or RPA (16.7 nM, lanes 5 and 6) were incubated in standard reaction buffer prior to addition of the D-loop substrate. Lane 1: A DNA marker, [³²P]-INV annealed with BB. Lane 7: heat-denatured D-loop substrate denoted by a filled triangle.

Figure 4.

Effect of DNA-PKcs on WRN helicase activity on telomeric and non-telomeric D-loops. WRN wild type (WT) (3.3 nM, lanes 5, 6, 12, and 13) or WRN (E84A) (3.3 nM, lanes 3, 4, 10, and 11) was preincubated with DNA-PKcs (16.7 nM, lanes 4, 6, 11, and 13). A telomeric (lanes 2-6) or a non-telomeric D-loop substrate (lanes 9-13) was added to the reaction. Lanes 1 and 8: A DNA ladder marker. Lanes 7 and 14: heat-denatured telomeric and non-telomeric D-loop substrates, respectively, denoted by filled triangles.

Figure 5.

DNA-PKcs fails to alter WRN helicase activity on forked duplex, Holliday junction and G-tailed telomeric DNA substrates. DNA helicase assays were carried out in the presence of the indicated proteins and DNA substrates. (A) WRN (1 nM, lanes 2, 3, 5, 8, 9, and 11) and either DNA-PKcs (5 nM, lanes 3, 4, 9, and 10) or RPA (5 nM, lanes 5, 6, 11, and 12) were incubated in standard reaction buffer prior to addition of a 34 bp forked duplex (0.5 nM, lanes 1-6) or a 22 bp forked duplex (0.5 nM, lanes 7-12). (B) WRN (4 nM, lanes 2-5) or BLM (2.5 nM, lanes 8-11), and DNA-PKcs (4 nM, lane 3; 8 nM, lane 4; 20 nM, lanes 5; 2.5 nM, lane 9; 5 nM, lane 10; 12.5 nM, lane 11) were incubated with in HJ reaction buffer prior to addition of Holliday junction (0.5 nM, lanes 1-11). Lane 6: DNA-PKcs (20 nM) alone. Lane 12: heat-denatured Holliday junction denoted with filled triangles. (C) G-tailed duplex (0.5 nM, lanes 1-5 and 7) was incubated with WRN (7.5 nM, lane 2-5) and DNA-PKcs (6.25 nM, lane 3; 12.5 nM, lane 4; 25 nM, lanes 5 and 7) in standard reaction buffer. Lane 6: heat-denatured G-tailed duplex denoted by a filled triangle.

Figure 6.

Quantification of telomere G-tail length by hybridization protection assay in DNA-PKcs knockdown U-2 OS cells. (A) A schematic of the HPA for telomere G-tail. Non-denatured genomic DNA was incubated with acridinium ester (AE)-labeled 29-mer telomere HPA probe. The AE of unhybridized and mis-hybridized probes was hydrolyzed, and chemilumines-cence from AE of hybridized probes was measured. (B and C) G-tail length of cells expressing an shRNA control or an shRNA against WRN was examined in panel B. G-tail length of cells transfected with siRNA against control (left), siRNA against DNA-PKcs (middle left), siRNA against DNA-PKcs with pEYFP-WRN (middle right), or siRNA against DNA-PKcs with pEYFP-WRN (E84A) (right) was examined in panel C. The G-tail length in the control cells was represented as 100%. Data are represented as mean +/- standard errors of two independent experiments.

Figure 7.

A model for protection of G-tails by DNA-PKcs. See text for detailed description of the model.