Essential role of Tip60-dependent recruitment of ribonucleotide reductase at DNA damage sites in DNA repair during G1 phase

Abstract

A balanced deoxyribonucleotide (dNTP) supply is essential for DNA repair. Here, we found that ribonucleotide reductase (RNR) subunits RRM1 and RRM2 accumulated very rapidly at damage sites. RRM1 bound physically to Tip60. Chromatin immunoprecipitation analyses of cells with an I-SceI cassette revealed that RRM1 bound to a damage site in a Tip60-dependent manner. Active RRM1 mutants lacking Tip60 binding failed to rescue an impaired DNA repair in RRM1-depleted G1-phase cells. Inhibition of RNR recruitment by an RRM1 C-terminal fragment sensitized cells to DNA damage. We propose that Tip60-dependent recruitment of RNR plays an essential role in dNTP supply for DNA repair.

Figure 1.

Tip60-dependent recruitment of RNR at DSB sites. (A) HeLa cells were exposed to IR at 1 Gy, subjected to in situ detergent extraction after 5 min, and immunostained with the indicated antibodies. Bars, 5 μm. (B) GM02063 cells were subjected to UVA microirradiation and immunostained with the indicated antibodies after 5 min. RRM1 or RRM2 and γH2AX signals are shown in green and red, respectively, in merged images. Bars, 10 μm. (C) IR-irradiated HeLa cell lysates treated with the indicated siRNAs were fractionated as described in the Materials and Methods. (Left panels) The fractions were subjected to immunoblotting using the indicated antibodies. (Right panel) The RRM1 bands were quantitated, and the results are presented as percentages of S1 fraction. Data are mean ± standard deviation (n = 3). (D) Sf9 lysates expressing RRM1-HA, RRM2, or GST-His-Tip60 were subjected to GST pull-down or HA pull-down assays using the indicated antibodies. (E) Chromatin fractions from IR- or mock-treated HeLa cells (after 5 min) were solubilized with micrococcal nuclease. The solubilized extracts were immunoprecipitated with anti-Tip60 antibodies or control IgG. The resulting precipitates and a 10% input (1/10 Input) were immunoblotted with the indicated antibodies. (F) The affinity-purified Tip60 complexes, as described in the Materials and Methods, were subjected to immunoblotting using the indicated antibodies. (G) GM02063 cells were treated with control, Tip60, or GFP siRNAs and then subjected to UVA microirradiation as in B.

Figure 2.

RRM1 is recruited at DSB sites in a Tip60-dependent manner. (A) Map of the I-SceI cassette construct containing the I-SceI site, the probe for Southern blotting, and a set of primers for the ChIP assay. (B) STEFKu70^−/−phprt-DR-GFP cells infected with I-SceI adenoviruses were subjected to ChIP analysis using the indicated antibodies as described in the supporting Materials and Methods. Data are shown as percentages of increases in PCR products from cells expressing I-SceI (I-SceI) relative to those from cells expressing Lac Z (LZ). Data are mean ± standard deviation (n = 3). (C) STEFKu70^−/−phprt-DR-GFP cells were transfected with two independent Tip60 siRNAs (Tip60-1 and Tip60-2) or control siRNA. ChIP analysis was performed as in B. (Bottom panels) Aliquots of cell lysates were subjected to immunoblotting using anti-Tip60 antibodies. (D) The constructs used are schematically represented, and the specific interaction between RRM1 mutants and Tip60 was assayed using yeast two-hybrid screening. (E) An in vitro RNR assay of complexes containing wild-type or various RRM1 mutants was performed as described in the Materials and Methods. (Black bars) −HU; (white bars) +HU (10 mM). Data are mean ± standard deviation (n = 3). (F) Sf9 lysates expressing GST-His-Tip60 and the indicated RRM1-HA were subjected to GST pull-down assay using the indicated antibodies. (G) Knockout–knock-in STEFKu70^−/−phprt-DR-GFP cells expressing wild-type or A776CΔ781-C RRM1-HA were generated by transfection with vectors for either wild-type or A776CΔ781-C RRM1 and then with RRM1 siRNA. Expression vectors of wild type and A776CΔ781-C contain mutations in a specific sequence targeted by siRNA. (Top panel) Cells were subjected to ChIP analysis using anti-HA antibodies as in B. (Bottom panels) Aliquots of cell lysates were subjected to immunoblotting using the indicated antibodies.

Figure 3.

Recruitment of active RNR at DNA damage sites is a prerequisite for effective DNA repair. (A) HeLa cells were treated with (open bars) or without (filled bars) 2.5 mM HU, exposed to IR (4 Gy), and subjected to a comet assay as described in the Materials and Methods. The results were obtained by counting at least 50 cells per sample in three independent experiments. (B) HeLa cells were transfected with a control (filled bars) or RRM1 or RRM2 siRNA (open bars), and DNA repair was evaluated as in A. Cell lysates were subjected to immunoblotting using the indicated antibodies. (C) HeLa cells were transfected with or without (filled bars) either wild-type (gray bars), C429S (open bars), A776CΔ781-C (dotted), or Δ761–786 (hatched) RRM1. RRM1-transfected cells were then transfected with RRM1 siRNA. Expression vectors of wild type and various RRM1 mutants contain mutations in a specific sequence targeted by siRNA. DNA repair activity and expression of RRM1 were examined as in B. (D) Knockout–knock-in HeLa cells expressing wild type or A776CΔ781-C RRM1-HA were exposed to IR, and cell lysates were subjected to immunoblotting as in C.

Figure 4.

Inhibition of recruitment of RNR at DSB sites by ectopic expression of NLS-RC1-HA abrogates DNA repair and sensitizes cells to DNA damage. (A) STEFKu70^−/−phprt-DR-GFP cells expressing NLS-RC1-HA (SV40 NLS-RC1 fragment, 701–792 amino acids) or NLS-GFP-HA (GFP fragment, 1–93 amino acids) were subjected to ChIP analysis as in Figure 2B. (Top panels) Cell lysates were subjected to immunoblotting using the indicated antibodies. (B, left panels) Tet-on HeLa cells expressing NLS-RC1-HA or NLS-GFP-HA were treated with or without tetracycline (1 μg/mL), exposed to IR (4 Gy), and subjected to immunostaining with the indicated antibodies and a comet assay as in Figure 3A. (Right panels) IR-untreated lysates were subjected to immunoblotting using the indicated antibodies. (C) Asterisk (*) represents nonspecific bands. (D) These cells were exposed to the indicated dose of IR, and a quantitative colony formation assay was performed 8 d after treatment. Data are mean ± standard deviation (n = 3). (E) Knockout–knock-in HeLa cells expressing either wild-type (filled bars) or A776CΔ781-C (open bars) RRM1-HA were synchronized as described in the Materials and Methods. Synchronized cells were then released into G1 phase or S phase (time −3) and exposed to IR (4 Gy) 3 h after release (time 0). (Right panels) DNA repair was evaluated as in A. (Left panels) Cell cycle distributions are presented.