How is Ataxia-Telangiectasia Mutated protein kinase activated?

Ataxia-Telangiectasia Mutated (ATM) is a key protein kinase in the cell's response to double-stranded breaks in DNA. This damage is detected by the Mre11-Rad50-Nbs1 (MRN) complex, which recruits ATM to the DNA. Upon activation, ATM phosphorylates a vast range of substrates, which triggers a multitude of downstream pathways such as DNA damage repair, cell cycle arrest, senescence and sometimes apoptosis. In addition to its role in the DNA damage response (DDR), ATM is reported to be a sensor of oxidative stress. The increased concentration of reactive oxygen species (ROS) under oxidative stress conditions was proposed to cause formation of a disulfide bond between two ATM monomers. ATM is then activated to phosphorylate numerous substrates involved in pathways such as autophagy, ROS regulation, and cell cycle checkpoint activation. The precise molecular details for how ATM is activated, either by the MRN complex at sites of DNA damage or by oxidative stress, are still not fully understood. In this thesis, I describe my work aimed at understanding the molecular basis of ATM activation in response to DNA damage and to oxidative stress. Firstly, I investigate the role of the Nbs1 C-terminus in ATM activation. Using pull-downs, kinase assays and negative stain electron microscopy, I determined that it does not directly cause changes in ATM to activate it. Instead, it acts indirectly by binding to ATM to localise it to the site of damage. Secondly, I detail attempts at elucidating the architecture of the complex that ATM forms with MRN and DNA using cryogenic electron microscopy (cryo-EM). Unfortunately, the cryo-EM samples did not yield particles that were sufficiently homogeneous to construct a high-resolution density map. These structural efforts were carried out in parallel with optimisation of in vitro ATM kinase assays to qualitatively and quantitatively assess ATM activity. From these, I determined that the catalytic turnover for substrate phosphorylation of ATM activated by MRN and DNA was 700-fold greater than ATM basal activity, and that this value for activated ATM was almost 50-fold higher than previously determined by another group. Moreover, assessment of ATP hydrolysis kinetic parameters for ATM basal activity compared to those of substrate phosphorylation suggest that substrate binding is a limiting step in the catalytic reaction, and not ATP binding. Studies into the MRN complex found that ATP hydrolysis by MRN might be playing an important role in ATM activation. Lastly, I focus on ATM activation by oxidative stress and how this compares to activation in DDR. I compare the kinetic parameters for ATM activation and demonstrate that ATM is far more active under DNA damaging conditions than under oxidative stress conditions, and that these activating conditions are distinct and not additive. These assays were also used to inform sample conditions for cryo-EM grids, aiming to visualise structural changes induced by oxidative stress. From these, I could determine high-resolution structures of the ATM closed inactive state and a novel ATM "H2O2-activated" state with removal of an autoinhibitory element (PRD), alongside a lower resolution map with an apparent disulfide bridge at the two-fold interface, which allowed formation of a hypothesis for ATM activation under oxidative stress conditions.