Nucleosome binding proteins potentiate ATM activation and DNA damage response by modifying chromatin

Abstract

In the nucleus of the eukaryotic cells, the genome is continuously damaged by multiple external and internal processes producing several types of DNA lesions. Correct repair of these lesions is crucial for the maintenance of genome integrity and for the orderly progression of developmental and differentiation programs. Several distinct DNA damage response (DDR) pathways have evolved to repair the various types of DNA lesions, including double strand breaks (DSB).1 These pathways involve an orderly assembly of the multiple components of the repair machinery to the damage site.2,3 Since the damaged DNA lesions are repaired in the context of chromatin the repair process involves changes in the local and global structure of the chromatin fiber. Indeed, changes in the organization of various chromatin components in the vicinity of the damaged site and the vital function of chromatin remodelers in the repair process are already documented.3,4 A major question which has not yet been fully explored is whether the global architecture of the chromatin fiber and the steady state nuclear organization of various DDR factors prior to the induction of DNA damage predetermine the subsequent cellular response to DNA damage. We obtained insights into this question by studying the role of the chromatin-binding protein HMGN1 in DSB repair.5 HMGN1 is a nucleosome binding protein that affects the architecture of chromatin and the levels of several posttranslational modifications in the tail of the nucleosomal histones6. The interaction of HMGN1 with chromatin is highly dynamic: the protein moves continuously among nucleosomes and its residence time at any particular site is short.7 HMGN1 interacts with chromatin within a network of chromatin-binding architectural proteins and competes with histone H1 for chromatin binding sites.8 This competition imparts structural and functional plasticity to the chromatin fiber since the two proteins have opposite effects on chromatin structure: HMGN1 destabilizes9 while histone H1 enhances chromatin compaction.10 Previous studies with Hmgn1-/- mice and cells revealed that loss of HMGN1 protein leads to alterations in histone modifications6 as well as to DSB hypersensitivity and to an increase in tumorigenicity.11 In our recent manuscript we demonstrate that the hyper sensitivity of Hmgn1-/- cells to DSB is due to faulty activation of the ATM kinase, the major cellular transducer of the DSB signal.5 Significantly, rescue experiments with wild type and mutant HMGN1 indicated that the interaction of HMGN1 with chromatin is important for the activation of ATM. Detailed analysis of the ATM activation pathway revealed that loss of HMGN1 affects the interaction of ATM with chromatin, a key step in its activation. Examination of the activation of DSB sensors which are upstream to ATM such as the MRN complex, 53BP1 and MDC1 did not reveal any malfunction in the Hmgn1-/- cells. Surprisingly, loss of HMGN1 enhances the interaction of ATM with chromatin not only after, but already prior to the induction of DNA damage. Although HMGN1 affects the activation of ATM and its chromatin interactions, the two proteins do not colocalize. Taken together, the results suggest that the interaction of HMGN1 with nucleosomes induces changes in chromatin that affect the global organization of ATM in the nucleus and predetermine the rate of DSB-induced ATM activation.5 Since the interaction of HMGN1 with chromatin affects the levels of histone modifications6 we tested directly whether these modifications affect ATM activation. We found that induction of DSB lead to global increase in the acetylation of H3K14 and that loss of HMGN1 reduces the levels of this acetylation. Significantly, treatment of cells with HDAC inhibitors, which do elevate the levels of histone acetylation, abolished the effects of HMGN1 on the interaction of ATM with chromatin and on its kinetics of activation.5 Thus, by modulating the levels of histone modifications, HMGN1 affects the chromatin organization and the activation kinetics of ATM. HMGN1 is not the only chromatin architectural protein that affect DNA repair. Depletion of H1 histone, a competitor of HMGN1,7,8 leads to the opposite effect on the DDR pathway; increased resistance to DSB formation along with increased activation of the ATM pathway.12 These results lead to several general conclusions. First, that the global organization of DDR factors in the nucleus already prior to the induction of DNA damage may affect the rate of DNA repair. Second, that chromatin structure and the level of histone modifications affect the organization of DDR factors and their rate of activation. Third, that by modifying the properties of chromatin, nucleosome-binding architectural protein may affect the rate of DNA repair. These results provide new insights into the molecular mechanisms whereby these structural proteins affect the rate of DNA repair and perhaps other biological activities in the context of chromatin.