A divalent FHA/BRCT-binding mechanism couples the MRE11-RAD50-NBS1 complex to damaged chromatin

Abstract

The MRE11-RAD50-NBS1 (MRN) complex accumulates at sites of DNA double-strand breaks in large chromatin domains flanking the lesion site. The mechanism of MRN accumulation involves direct binding of the Nijmegen breakage syndrome 1 (NBS1) subunit to phosphorylated mediator of the DNA damage checkpoint 1 (MDC1), a large nuclear adaptor protein that interacts directly with phosphorylated H2AX. NBS1 contains an FHA domain and two BRCT domains at its amino terminus. Here, we show that both of these domains participate in the interaction with phosphorylated MDC1. Point mutations in key amino acid residues of either the FHA or the BRCT domains compromise the interaction with MDC1 and lead to defects in MRN accumulation at sites of DNA damage. Surprisingly, only mutation in the FHA domain, but not in the BRCT domains, yields a G2/M checkpoint defect, indicating that MDC1-dependent chromatin accumulation of the MRN complex at sites of DNA breaks is not required for G2/M checkpoint activation.

Figure 1

Both the FHA domain and the tandem BRCT domain of NBS1 are required for the interaction with the phosphorylated SDT region of MDC1 in vitro. (A) Schematic representation of full-length human NBS1 and its domain composition. The enlarged area shows a sequence alignment of the FHA and BRCT domains of human, mouse and Xenopus NBS1. The putative secondary structure of the first (amino-terminal) BRCT domain is indicated by pale colours. The secondary structure of the second (carboxy-terminal) BRCT domain (indicated by bright colours) was derived from Xu et al (2008). Phospho-interacting amino acids are highlighted in yellow. (B) Purified MDC1 GST-SDT fragment was preincubated with CK2 and ATP. The fragment was then incubated with in vitro-translated ³⁵S-labelled NBS1 wild type or mutants for 1 h, washed and resolved by SDS–PAGE and autoradiography. (C) Purified MDC1 GST-SDT fragment was preincubated with CK2 and ATP. The fragment was then incubated with purified MRN complex where the NBS1 subunit was either wild type or contained a point mutation in the FHA domain (R28A) or in the BRCT tandem domain (K160M). Bound proteins were separated by SDS–PAGE followed by immunoblotting. The blots were probed with a polyclonal antibody against NBS1. (D) Human embryonic kidney 293T cells were transiently transfected with Flag-tagged MDC1(800) fragment and Myc-tagged NBS1 wild type and mutants, as indicated. Flag antibodies were used for co-immunoprecipitation and Myc antibodies for western blot analysis. CK2, casein kinase 2; GST, glutathione-S-transferase; IP, immunoprecipitation; MDC1, mediator of the DNA damage checkpoint 1; MRN, MRE11–RAD50–NBS1 complex; NBS1, Nijmegen breakage syndrome 1; SDS–PAGE, sodium dodecyl sulphate–polyacrylamide gel electrophoresis; wt, wild-type.

Figure 2

The BRCT tandem domain of NBS1 is required for focal accumulation of the MRN complex at sites of DSBs in vivo. (A) NBS-iLB1 fibroblasts and NBS-iLB1 fibroblasts stably transduced with wild-type, R28A or K160M mutant NBS1, respectively, were irradiated at 5 Gy. The irradiated cells were incubated for 1 h, fixed with methanol and probed with the indicated antibodies. Cells were then analysed by confocal microscopy and nuclear foci-positive cells were counted for statistical evaluation. (B) NBS-iLB1 fibroblasts stably transduced with wild-type, R28A or K160M mutant NBS1, respectively, were micro-irradiated as described in Methods. The irradiated cells were incubated for 1 h, fixed with methanol and probed with the indicated antibodies. Cells were then analysed by confocal microscopy. DAPI, 4′,6-diamidino-2-phenylindole; MRN, MRE11–RAD50–NBS1 complex; NBS1, Nijmegen breakage syndrome 1.

Figure 3

Mutation in the tandem BRCT domain of NBS1 does not yield a G2/M DNA-damage checkpoint defect. NBS-iLB1 fibroblasts and NBS-iLB1 fibroblasts stably transduced with wild-type, R28A or K160 mutant NBS1, respectively, were left untreated or irradiated at 1 and 10 Gy. Cells were harvested 1 h after irradiation, fixed with methanol and stained with an antibody against phosphorylated H3 (P-H3) and propidium iodide. The percentage of P-H3-positive cells was determined by fluorescence-activated cell sorting analysis. In this graph, three independent experiments (each performed in triplicate) are summarized. The error bars represent the standard deviation. NBS1, Nijmegen breakage syndrome 1.

Figure 4

Experimental uncoupling of the MRN complex from damaged chromatin does not trigger a G2/M checkpoint defect. (A) Nuclear foci formation of NBS1 in inducible U2OS YFP-BRCT-overexpressing cells after irradiation at 5 Gy. Non-induced cells (top) and YFP-BRCT-expressing cells (bottom). (B) Microlaser-induced DNA-damage recruitment analysis of NBS1 in inducible U2OS YFP-BRCT-overexpressing cells. Non-induced cells (top) and YFP-BRCT-expressing cells (bottom). (C) Overexpression of the MDC1 BRCT domains does not trigger a G2/M checkpoint defect. Expression of YFP-BRCT fusion protein was induced 8 h before irradiation (+DOX). Mock-induced cells acted as the control (−DOX). Depletion of endogenous MDC1 by siRNA (siM) partly abrogated the G2/M checkpoint regardless of whether or not MDC1 is proficient for γH2AX binding. The error bars represent the standard deviation. DAPI, 4′,6-diamidino-2-phenylindole; DOX, doxocyclin; MDC1, mediator of the DNA damage checkpoint 1; MRN, MRE11–RAD50–NBS1 complex; NBS1, Nijmegen breakage syndrome 1; siRNA, small interfering RNA; YFP, yellow fluorescent protein.