Restoration of ATM Expression in DNA-PKcs-Deficient Cells Inhibits Signal End Joining

Single-molecule imaging reveals the kinetics of non-homologous end-joining in living cells

Non-homologous end joining (NHEJ) is the predominant pathway that repairs DNA double-stranded breaks (DSBs) in vertebrates. However, due to challenges in detecting DSBs in living cells, the repair capacity of the NHEJ pathway is unknown. The DNA termini of many DSBs must be processed to allow ligation while minimizing genetic changes that result from break repair. Emerging models propose that DNA termini are first synapsed ~115Å apart in one of several long-range synaptic complexes before transitioning into a short-range synaptic complex that juxtaposes DNA ends to facilitate ligation. The transition from long-range to short-range synaptic complexes involves both conformational and compositional changes of the NHEJ factors bound to the DNA break. Importantly, it is unclear how NHEJ proceeds in vivo because of the challenges involved in analyzing recruitment of NHEJ factors to DSBs over time in living cells. Here, we develop a new approach to study the temporal and compositional dynamics of NHEJ complexes using live cell single-molecule imaging. Our results provide direct evidence for stepwise maturation of the NHEJ complex, pinpoint key regulatory steps in NHEJ progression, and define the overall repair capacity NHEJ in living cells.

Unravelling the complexities of DNA-PK activation by structure-based mutagenesis

It has been known for decades that the DNA-dependent protein kinase (DNA-PK) is only an active serine/threonine protein kinase when it is bound to a DNA double-stranded end; still, the molecular details of how this activation is achieved have remained elusive. The recent surge in structural information for DNA-PK complexes has provided valuable insights into the process of DNA end recognition by DNA-PK. A particularly intriguing feature of this kinase is a region of the protein that can transition from a seemingly structurally disordered state to a single alpha-helix that traverses down the DNA binding cradle. The DNA-PK bound DNA end of the DNA substrate engages with and appears to split around this helix which has been named the DNA End Blocking helix (DEB). Here a mutational approach is utilized to clarify the role of the DEB, and how DNA ends activate the enzyme. Our data suggest two distinct methods of kinase activation that is dependent on the DNA end chemistry. If the DNA end can split around the helix and stabilize the interaction between the DNA end and the DEB with a recently defined Helix-Hairpin-Helix (HHH) motif, the kinase forms an end-protection monomer that is active towards DNA-PK's many substrates. But if the DNA end cannot stably interact with the DEB [because of the DNA end structure, for instance hairpins, or because the DEB has been disrupted by mutation], the kinase is only partially activated, resulting in specific autophosphorylations of the DNA-PK monomer that allows nucleolytic end-processing. We posit that mutants that disrupt the capacity to stably generate the DEB/HHH DNA end-interaction are inefficient in generating the dimer complex that is requisite for NHEJ. In support of this idea, mutations that promote formation of this dimer partially rescue the severe cellular phenotypes associated with mutation of the DEB helix.

Efficient high-precision homology-directed repair-dependent genome editing by HDRobust

Homology-directed repair (HDR), a method for repair of DNA double-stranded breaks can be leveraged for the precise introduction of mutations supplied by synthetic DNA donors, but remains limited by low efficiency and off-target effects. In this study, we report HDRobust, a high-precision method that, via the combined transient inhibition of nonhomologous end joining and microhomology-mediated end joining, resulted in the induction of point mutations by HDR in up to 93% (median 60%, s.e.m. 3) of chromosomes in populations of cells. We found that, using this method, insertions, deletions and rearrangements at the target site, as well as unintended changes at other genomic sites, were largely abolished. We validated this approach for 58 different target sites and showed that it allows efficient correction of pathogenic mutations in cells derived from patients suffering from anemia, sickle cell disease and thrombophilia.

Two distinct long-range synaptic complexes promote different aspects of end processing prior to repair of DNA breaks by non-homologous end joining

Non-homologous end joining is the major double-strand break repair (DSBR) pathway in mammals. DNA-PK is the hub and organizer of multiple steps in non-homologous end joining (NHEJ). Recent high-resolution structures show how two distinct NHEJ complexes "synapse" two DNA ends. One complex includes a DNA-PK dimer mediated by XLF, whereas a distinct DNA-PK dimer forms via a domain-swap mechanism where the C terminus of Ku80 from one DNA-PK protomer interacts with another DNA-PK protomer in trans. Remarkably, the distance between the two synapsed DNA ends in both dimers is the same (∼115 Å), which matches the distance observed in the initial description of an NHEJ long-range synaptic complex. Here, a mutational strategy is used to demonstrate distinct cellular function(s) of the two dimers: one promoting fill-in end processing, while the other promotes DNA end resection. Thus, the specific DNA-PK dimer formed (which may be impacted by DNA end structure) dictates the mechanism by which ends will be made ligatable.

Catalytically inactive DNA ligase IV promotes DNA repair in living cells

DNA double strand breaks (DSBs) are induced by external genotoxic agents (ionizing radiation or genotoxins) or by internal processes (recombination intermediates in lymphocytes or by replication errors). The DNA ends induced by these genotoxic processes are often not ligatable, requiring potentially mutagenic end-processing to render ends compatible for ligation by non-homologous end-joining (NHEJ). Using single molecule approaches, Loparo et al. propose that NHEJ fidelity can be maintained by restricting end-processing to a ligation competent short-range NHEJ complex that 'maximizes the fidelity of DNA repair'. These in vitro studies show that although this short-range NHEJ complex requires DNA ligase IV (Lig4), its catalytic activity is dispensable. Here using cellular models, we show that inactive Lig4 robustly promotes DNA repair in living cells. Compared to repair products from wild-type cells, those isolated from cells with inactive Lig4 show a somewhat increased fraction that utilize micro-homology (MH) at the joining site consistent with alternative end-joining (a-EJ). But unlike a-EJ in the absence of NHEJ, a large percentage of joints isolated from cells with inactive Lig4 occur with no MH - thus, clearly distinct from a-EJ. Finally, biochemical assays demonstrate that the inactive Lig4 complex promotes the activity of DNA ligase III (Lig3).

Cryo-EM of NHEJ supercomplexes provides insights into DNA repair

Non-homologous end joining (NHEJ) is one of two critical mechanisms utilized in humans to repair DNA double-strand breaks (DSBs). Unrepaired or incorrect repair of DSBs can lead to apoptosis or cancer. NHEJ involves several proteins, including the Ku70/80 heterodimer, DNA-dependent protein kinase catalytic subunit (DNA-PKcs), X-ray cross-complementing protein 4 (XRCC4), XRCC4-like factor (XLF), and ligase IV. These core proteins bind DSBs and ligate the damaged DNA ends. However, details of the structural assembly of these proteins remain unclear. Here, we present cryo-EM structures of NHEJ supercomplexes that are composed of these core proteins and DNA, revealing the detailed structural architecture of this assembly. We describe monomeric and dimeric forms of this supercomplex and also propose the existence of alternate dimeric forms of long-range synaptic complexes. Finally, we show that mutational disruption of several structural features within these NHEJ complexes negatively affects DNA repair.

Autophosphorylation and Self-Activation of DNA-Dependent Protein Kinase

The DNA-dependent protein kinase catalytic subunit (DNA-PKcs), a member of the phosphatidylinositol 3-kinase-related kinase family, phosphorylates serine and threonine residues of substrate proteins in the presence of the Ku complex and double-stranded DNA. Although it has been established that DNA-PKcs is involved in non-homologous end-joining, a DNA double-strand break repair pathway, the mechanisms underlying DNA-PKcs activation are not fully understood. Nevertheless, the findings of numerous in vitro and in vivo studies have indicated that DNA-PKcs contains two autophosphorylation clusters, PQR and ABCDE, as well as several autophosphorylation sites and conformational changes associated with autophosphorylation of DNA-PKcs are important for self-activation. Consistent with these features, an analysis of transgenic mice has shown that the phenotypes of DNA-PKcs autophosphorylation mutations are significantly different from those of DNA-PKcs kinase-dead mutations, thereby indicating the importance of DNA-PKcs autophosphorylation in differentiation and development. Furthermore, there has been notable progress in the high-resolution analysis of the conformation of DNA-PKcs, which has enabled us to gain a visual insight into the steps leading to DNA-PKcs activation. This review summarizes the current progress in the activation of DNA-PKcs, focusing in particular on autophosphorylation of this kinase.

Structure of an activated DNA-PK and its implications for NHEJ

DNA-dependent protein kinase (DNA-PK), like all phosphatidylinositol 3-kinase-related kinases (PIKKs), is composed of conserved FAT and kinase domains (FATKINs) along with solenoid structures made of HEAT repeats. These kinases are activated in response to cellular stress signals, but the mechanisms governing activation and regulation remain unresolved. For DNA-PK, all existing structures represent inactive states with resolution limited to 4.3 Å at best. Here, we report the cryoelectron microscopy (cryo-EM) structures of DNA-PKcs (DNA-PK catalytic subunit) bound to a DNA end or complexed with Ku70/80 and DNA in both inactive and activated forms at resolutions of 3.7 Å overall and 3.2 Å for FATKINs. These structures reveal the sequential transition of DNA-PK from inactive to activated forms. Most notably, activation of the kinase involves previously unknown stretching and twisting within individual solenoid segments and loosens DNA-end binding. This unprecedented structural plasticity of helical repeats may be a general regulatory mechanism of HEAT-repeat proteins.

Temporally uncoupled signal and coding joint formation in human V(D)J recombination

In vertebrate antigen receptor gene rearrangement, V(D)J recombination events can occur by deletion or by inversion. For deletional events, the signal joint is deleted from the genome. Nearly half of the immunoglobulin light chain genes undergo V(D)J recombination in an inversional manner, and both signal and coding joint formation must occur to retain chromosomal integrity. But given the undetermined amount of pre-B and pre-T cell death that occurs during V(D)J recombination, the efficiency with which both joints are completed is not known, nor is the relative efficiency (balance) of signal versus coding joint formation. Signal joint formation only requires Ku and XRCC4:DNA ligase 4 of the nonhomologous DNA end joining repair pathway. Coding joint formation requires these proteins as well, but in addition requires Artemis and DNA-dependent protein kinase to open the hairpin DNA coding ends, which the RAG complex generated; and further processing is required because the hairpin opening generates incompatible 3' overhangs. Mutations in some of the end processing enzymes affect one, but only minimally the other joint. We have devised a precise cellular assay that does not have any cellular, enzymatic or biochemical selective bias to assess signal and coding joint formation independently, and it can detect intermediates for which one joint has formed but not the other. We find that intermediates with only one completed joint are more abundant than molecules with both joints completed. This indicates that either joint can form independent of the other and joint formation can be a relatively slow process.

Activation of DNA-PK by hairpinned DNA ends reveals a stepwise mechanism of kinase activation

As its name implies, the DNA dependent protein kinase (DNA-PK) requires DNA double-stranded ends for enzymatic activation. Here, I demonstrate that hairpinned DNA ends are ineffective for activating the kinase toward many of its well-studied substrates (p53, XRCC4, XLF, HSP90). However, hairpinned DNA ends robustly stimulate certain DNA-PK autophosphorylations. Specifically, autophosphorylation sites within the ABCDE cluster are robustly phosphorylated when DNA-PK is activated by hairpinned DNA ends. Of note, phosphorylation of the ABCDE sites is requisite for activation of the Artemis nuclease that associates with DNA-PK to mediate hairpin opening. This finding suggests a multi-step mechanism of kinase activation. Finally, I find that all non-homologous end joining (NHEJ) defective cells (whether deficient in components of the DNA-PK complex or components of the ligase complex) are similarly deficient in joining DNA double-stranded breaks (DSBs) with hairpinned termini.

Human DNA-PK activates a STING-independent DNA sensing pathway

Detection of intracellular DNA by the cGAS-STING pathway activates a type I interferon-mediated innate immune response that protects from virus infection. Whether there are additional DNA sensing pathways, and how such pathways might function, remains controversial. We show here that humans-but not laboratory mice-have a second, potent, STING-independent DNA sensing pathway (SIDSP). We identify human DNA-dependent protein kinase (DNA-PK) as the sensor of this pathway and demonstrate that DNA-PK activity drives a robust and broad antiviral response. We show that the E1A oncoprotein of human adenovirus 5 and the ICP0 protein of herpes simplex virus 1 block this response. We found heat shock protein HSPA8/HSC70 as a target for inducible phosphorylation in the DNA-PK antiviral pathway. Last, we demonstrate that DNA damage and detection of foreign DNA trigger distinct modalities of DNA-PK activity. These findings reveal the existence, sensor, a specific downstream target, and viral antagonists of a SIDSP in human cells.

Deciphering the role of distinct DNA-PK phosphorylations at collapsed replication forks

It has recently been established that the marked sensitivity of ATM deficient cells to topoisomerase poisons like camptothecin (Cpt) results from unrestrained end-joining of DNA ends at collapsed replication forks that is mediated by the non-homologous end joining [NHEJ] pathway and results in the induction of copious numbers of genomic alterations, termed "toxic NHEJ". Ablation of core components of the NHEJ pathway reverses the Cpt sensitivity of ATM deficient cells, but inhibition of DNA-PKcs does not. Here, we show that complete ablation of DNA-PKcs partially reverses the Cpt sensitivity of ATM deficient cells; thus, ATM deficient cells lacking DNA-PKcs are more resistant to Cpt than cells expressing DNA-PKcs. However, the relative sensitivity of DNA-PKcs proficient ATM deficient cells is inversely proportional to DNA-PKcs expression levels. These data suggest that DNA-PK may phosphorylate an ATM target (that contributes to Cpt resistance), explaining partial rescue of Cpt sensitivity in cells expressing high levels of DNA-PKcs. Although crippling NHEJ function by mutagenic blockade of the critical ABCDE autophosphorylation sites in DNA-PKcs also sensitizes cells to Cpt, this sensitization apparently occurs by a distinct mechanism from ATM ablation because blockade of these sites actually rescues ATM deficient cells from toxic NHEJ. These data are consistent with autophosphorylation of the ABCDE sites (and not ATM mediated phosphorylation) in response to Cpt-induced damage. In contrast, blockade of S3205 (an ATM dependent phosphorylation site in DNA-PKcs) that minimally impacts NHEJ, increases Cpt sensitivity. In sum, these data suggest that ATM and DNA-PK cooperate to facilitate Cpt-induced DNA damage, and that ATM phosphorylation of S3205 facilitates appropriate repair at collapsed replication forks.

ATM antagonizes NHEJ proteins assembly and DNA-ends synapsis at single-ended DNA double strand breaks

Two DNA repair pathways operate at DNA double strand breaks (DSBs): non-homologous end-joining (NHEJ), that requires two adjacent DNA ends for ligation, and homologous recombination (HR), that resects one DNA strand for invasion of a homologous duplex. Faithful repair of replicative single-ended DSBs (seDSBs) is mediated by HR, due to the lack of a second DNA end for end-joining. ATM stimulates resection at such breaks through multiple mechanisms including CtIP phosphorylation, which also promotes removal of the DNA-ends sensor and NHEJ protein Ku. Here, using a new method for imaging the recruitment of the Ku partner DNA-PKcs at DSBs, we uncover an unanticipated role of ATM in removing DNA-PKcs from seDSBs in human cells. Phosphorylation of DNA-PKcs on the ABCDE cluster is necessary not only for DNA-PKcs clearance but also for the subsequent MRE11/CtIP-dependent release of Ku from these breaks. We propose that at seDSBs, ATM activity is necessary for the release of both Ku and DNA-PKcs components of the NHEJ apparatus, and thereby prevents subsequent aberrant interactions between seDSBs accompanied by DNA-PKcs autophosphorylation and detrimental commitment to Lig4-dependent end-joining.

The recent advances in non-homologous end-joining through the lens of lymphocyte development

Lymphocyte development requires ordered assembly and subsequent modifications of the antigen receptor genes through V(D)J recombination and Immunoglobulin class switch recombination (CSR), respectively. While the programmed DNA cleavage events are initiated by lymphocyte-specific factors, the resulting DNA double-strand break (DSB) intermediates activate the ATM kinase-mediated DNA damage response (DDR) and rely on the ubiquitously expressed classical non-homologous end-joining (cNHEJ) pathway including the DNA-dependent protein kinase (DNA-PK), and, in the case of CSR, also the alternative end-joining (Alt-EJ) pathway, for repair. Correspondingly, patients and animal models with cNHEJ or DDR defects develop distinct types of immunodeficiency reflecting their specific DNA repair deficiency. The unique end-structure, sequence context, and cell cycle regulation of V(D)J recombination and CSR also provide a valuable platform to study the mechanisms of, and the interplay between, cNHEJ and DDR. Here, we compare and contrast the genetic consequences of DNA repair defects in V(D)J recombination and CSR with a focus on the newly discovered cNHEJ factors and the kinase-dependent structural roles of ATM and DNA-PK in animal models. Throughout, we try to highlight the pending questions and emerging differences that will extend our understanding of cNHEJ and DDR in the context of primary immunodeficiency and lymphoid malignancies.

Simultaneous precise editing of multiple genes in human cells

When double-strand breaks are introduced in a genome by CRISPR they are repaired either by non-homologous end joining (NHEJ), which often results in insertions or deletions (indels), or by homology-directed repair (HDR), which allows precise nucleotide substitutions to be introduced if a donor oligonucleotide is provided. Because NHEJ is more efficient than HDR, the frequency with which precise genome editing can be achieved is so low that simultaneous editing of more than one gene has hitherto not been possible. Here, we introduced a mutation in the human PRKDC gene that eliminates the kinase activity of the DNA-dependent protein kinase catalytic subunit (DNA-PKcs). This results in an increase in HDR irrespective of cell type and CRISPR enzyme used, sometimes allowing 87% of chromosomes in a population of cells to be precisely edited. It also allows for precise editing of up to four genes simultaneously (8 chromosomes) in the same cell. Transient inhibition of DNA-PKcs by the kinase inhibitor M3814 is similarly able to enhance precise genome editing.

mEAK-7 Forms an Alternative mTOR Complex with DNA-PKcs in Human Cancer

MTOR associated protein, eak-7 homolog (mEAK-7), activates mechanistic target of rapamycin (mTOR) signaling in human cells through an alternative mTOR complex to regulate S6K2 and 4E-BP1. However, the role of mEAK-7 in human cancer has not yet been identified. We demonstrate that mEAK-7 and mTOR signaling are strongly elevated in tumor and metastatic lymph nodes of patients with non-small-cell lung carcinoma compared with those of patients with normal lung or lymph tissue. Cancer stem cells, CD44+/CD90+ cells, yield elevated mEAK-7 and activated mTOR signaling. mEAK-7 is required for clonogenic potential and spheroid formation. mEAK-7 associates with DNA-dependent protein kinase catalytic subunit isoform 1 (DNA-PKcs), and this interaction is increased in response to X-ray irradiation to regulate S6K2 signaling. DNA-PKcs pharmacologic inhibition or genetic knockout reduced S6K2, mEAK-7, and mTOR binding with DNA-PKcs, resulting in loss of S6K2 activity and mTOR signaling. Therefore, mEAK-7 forms an alternative mTOR complex with DNA-PKcs to regulate S6K2 in human cancer cells.

Deciphering phenotypic variance in different models of DNA-PKcs deficiency

DNA-PKcs deficiency has been studied in numerous animal models and cell culture systems. In previous studies of kinase inactivating mutations in cell culture systems, ablation of DNA-PK's catalytic activity results in a cell phenotype that is virtually indistinguishable from that ascribed to complete loss of the enzyme. However, a recent compelling study demonstrates a remarkably more severe phenotype in mice harboring a targeted disruption of DNA-PK's ATP binding site as compared to DNA-PKcs deficient mice. Here we investigate the mechanism for these divergent results. We find that kinase inactivating DNA-PKcs mutants markedly radiosensitize immortalized DNA-PKcs deficient cells, but have no substantial effects on transformed DNA-PKcs deficient cells. Since the non-homologous end joining mechanism likely functions similarly in all of these cell strains, it seems unlikely that kinase inactive DNA-PK could impair the end joining mechanism in some cell types, but not in others. In fact, we observed no significant differences in either episomal or chromosomal end joining assays in cells expressing kinase inactivated DNA-PKcs versus no DNA-PKcs. Several potential explanations could explain these data including a non-catalytic role for DNA-PKcs in promoting cell death, or alteration of gene expression by loss of DNA-PKcs as opposed to inhibition of its catalytic activity. Finally, controversy exists as to whether DNA-PKcs autophosphorylates or is the target of other PIKKs; we present data demonstrating that DNA-PK primarily autophosphorylates.

PAXX Is an Accessory c-NHEJ Factor that Associates with Ku70 and Has Overlapping Functions with XLF

In mammalian cells, classical non-homologous end joining (c-NHEJ) is critical for DNA double-strand break repair induced by ionizing radiation and during V(D)J recombination in developing B and T lymphocytes. Recently, PAXX was identified as a c-NHEJ core component. We report here that PAXX-deficient cells exhibit a cellular phenotype uncharacteristic of a deficiency in c-NHEJ core components. PAXX-deficient cells display normal sensitivity to radiomimetic drugs, are proficient in transient V(D)J recombination assays, and do not shift toward higher micro-homology usage in plasmid repair assays. Although PAXX-deficient cells lack c-NHEJ phenotypes, PAXX forms a stable ternary complex with Ku bound to DNA. Formation of this complex involves an interaction with Ku70 and requires a bare DNA extension for stability. Moreover, the relatively weak Ku-dependent stimulation of LIG4/XRCC4 activity by PAXX is unmasked by XLF ablation. Thus, PAXX plays an accessory role during c-NHEJ that is largely overlapped by XLF's function.

Ataxia telangiectasia syndrome: moonlighting ATM

INTRODUCTION

Ataxia-telangiectasia (A-T) a multisystem disorder primarily characterized by cerebellar degeneration, telangiectasia, immunodeficiency, cancer susceptibility and radiation sensitivity. Identification of the gene defective in this syndrome, ataxia-telangiectasia mutated gene (ATM), and further characterization of the disorder together with a greater insight into the function of the ATM protein have expanded our knowledge about the molecular pathogenesis of this disease. Area covered: In this review, we have attempted to summarize the different roles of ATM signaling that have provided new insights into the diverse clinical phenotypes exhibited by A-T patients. Expert commentary: ATM, in addition to DNA repair response, is involved in many cytoplasmic roles that explain diverse phenotypes of A-T patients. It seems accumulation of DNA damage, persistent DNA damage response signaling, and chronic oxidative stress are the main players in the pathogenesis of this disease.

Mutational phospho-mimicry reveals a regulatory role for the XRCC4 and XLF C-terminal tails in modulating DNA bridging during classical non-homologous end joining

XRCC4 and DNA Ligase 4 (LIG4) form a tight complex that provides DNA ligase activity for classical non-homologous end joining (the predominant DNA double-strand break repair pathway in higher eukaryotes) and is stimulated by XLF. Independently of LIG4, XLF also associates with XRCC4 to form filaments that bridge DNA. These XRCC4/XLF complexes rapidly load and connect broken DNA, thereby stimulating intermolecular ligation. XRCC4 and XLF both include disordered C-terminal tails that are functionally dispensable in isolation but are phosphorylated in response to DNA damage by DNA-PK and/or ATM. Here we concomitantly modify the tails of XRCC4 and XLF by substituting fourteen previously identified phosphorylation sites with either alanine or aspartate residues. These phospho-blocking and -mimicking mutations impact both the stability and DNA bridging capacity of XRCC4/XLF complexes, but without affecting their ability to stimulate LIG4 activity. Implicit in this finding is that phosphorylation may regulate DNA bridging by XRCC4/XLF filaments.

DNA Breaks and End Resection Measured Genome-wide by End Sequencing

DNA double-strand breaks (DSBs) arise during physiological transcription, DNA replication, and antigen receptor diversification. Mistargeting or misprocessing of DSBs can result in pathological structural variation and mutation. Here we describe a sensitive method (END-seq) to monitor DNA end resection and DSBs genome-wide at base-pair resolution in vivo. We utilized END-seq to determine the frequency and spectrum of restriction-enzyme-, zinc-finger-nuclease-, and RAG-induced DSBs. Beyond sequence preference, chromatin features dictate the repertoire of these genome-modifying enzymes. END-seq can detect at least one DSB per cell among 10,000 cells not harboring DSBs, and we estimate that up to one out of 60 cells contains off-target RAG cleavage. In addition to site-specific cleavage, we detect DSBs distributed over extended regions during immunoglobulin class-switch recombination. Thus, END-seq provides a snapshot of DNA ends genome-wide, which can be utilized for understanding genome-editing specificities and the influence of chromatin on DSB pathway choice.

The ATM Kinase Restrains Joining of Both VDJ Signal and Coding Ends

The evidence that ATM affects resolution of RAG-induced DNA double-strand breaks is profuse and unequivocal; moreover, it is clear that the RAG complex itself cooperates (in an undetermined way) with ATM to facilitate repair of these double-strand breaks by the classical nonhomologous end-joining pathway. The mechanistic basis for the cooperation between ATM and the RAG complex has not been defined, although proposed models invoke ATM and RAG2's C terminus in maintaining the RAG postcleavage complex. In this study, we show that ATM reduces the rate of both coding and signal joining in a robust episomal assay; we suggest that this is the result of increased stability of the postcleavage complex. ATM's ability to inhibit VDJ joining requires its enzymatic activity. The noncore C termini of both RAG1 and RAG2 are also required for ATM's capacity to limit signal (but not coding) joining. Moreover, potential phosphorylation targets within the C terminus of RAG2 are also required for ATM's capacity to limit signal joining. These data suggest a model whereby the RAG signal end complex is stabilized by phosphorylation of RAG2 by ATM.