Learn more: PMC Disclaimer | PMC Copyright Notice
Ataxia telangiectasia mutated (Atm) and DNA-PKcs kinases have overlapping activities during chromosomal signal joint formation
Associated Data
Abstract
Lymphocyte antigen receptor gene assembly occurs through the process of V(D)J recombination, which is initiated when the RAG endonuclease introduces DNA DSBs at two recombining gene segments to form broken DNA coding end pairs and signal end pairs. These paired DNA ends are joined by proteins of the nonhomologous end-joining (NHEJ) pathway of DSB repair to form a coding joint and signal joint, respectively. RAG DSBs are generated in G1-phase developing lymphocytes, where they activate the ataxia telangiectasia mutated (Atm) and DNA-PKcs kinases to orchestrate diverse cellular DNA damage responses including DSB repair. Paradoxically, although Atm and DNA-PKcs both function during coding joint formation, Atm appears to be dispensible for signal joint formation; and although some studies have revealed an activity for DNA-PKcs during signal joint formation, others have not. Here we show that Atm and DNA-PKcs have overlapping catalytic activities that are required for chromosomal signal joint formation and for preventing the aberrant resolution of signal ends as potentially oncogenic chromosomal translocations.
Developing lymphocytes assemble antigen receptor genes through the process of V(D)J recombination, which is initiated by the recombinase activating gene (RAG)-1 and -2 proteins that together form an endonuclease, hereafter referred to as RAG (1). RAG introduces DNA double strand breaks (DSBs) at the border of recombining variable (V), diversity (D), and joining (J) gene segments and their flanking RAG recognition sequences, termed recombination signals (RSs). This pairwise DNA cleavage results in two hairpin-sealed coding ends and two blunt phosphorylated signal ends. The coding ends are joined imprecisely as a coding joint, and the signal ends are joined relatively precisely as a signal joint (2, 3).
Whereas coding joint formation is required to generate the second exon of all antigen receptor genes, signal joint formation is dispensable for this process. Indeed, when the two recombining gene segments are in the same transcriptional orientation, the signal end-flanked intervening DNA fragment is excised from the chromosome, and signal joints form on an extrachromosomal circle that is lost upon cell division. However, when gene segments are in the opposite transcriptional orientation, rearrangement leads to inversion of the intervening region, and the signal joint remains within the chromosomal context. Importantly, in this latter case, signal joint formation is essential for reestablishing the linear integrity of the chromosome. However, signal joint formation may be less efficient than coding joint formation, raising the question of how chromosome integrity is maintained in cells undergoing inversional rearrangements (4, 5).
Coding ends and signal ends are processed and joined by proteins of the nonhomologous end-joining (NHEJ) pathway of DNA DSB repair (2, 3). Elements of this pathway include DNA Ligase IV, XRCC4, Artemis, and the DNA-PK complex comprising Ku70, Ku80 and the DNA-PKcs catalytic subunit. XLF/Cernunnos also may function in repairing RAG DSBs, although its activity is not essential for V(D)J recombination in lymphoid cells (6). DNA Ligase IV, XRCC4, Ku70 and Ku80 are essential for both signal and coding joint formation. Artemis possesses a nuclease activity required to open hairpin-sealed coding ends that is essential for coding joint formation (2, 3).
DNA-PKcs is a member of the PI-3-like family of serine/threonine kinases that forms a complex with Ku70 and Ku80. During coding joint formation, DNA-PKcs regulates the hairpin-opening activity of Artemis (2, 3). Although DNA-PKcs promotes precise signal joint formation, the extent to which signal end joining per se depends on DNA-PKcs has been less clear, with variability in the findings potentially due to the different DNA-PKcs mutations and cell types analyzed (2, 7–12).
Like DNA-PKcs, the ataxia telangiectasia mutated (Atm) protein also is a member of the PI-3-like family of serine/threonine kinases that functions primarily during coding joint formation (13–15). However, Atm is not required to open hairpin-sealed coding ends (13). Rather, it stabilizes coding ends in postcleavage complexes until they are joined by NHEJ and suppresses the aberrant resolution of these DNA ends as chromosomal translocations (13). In contrast, Atm has no known activities in signal joint formation (13, 16).
In response to DNA DSBs, Atm and DNA-PKcs phosphorylate many proteins that act broadly in DNA damage responses (17, 18). Importantly, some of these substrates can be phosphorylated and regulated by either kinase (19, 20). The functional relevance of these overlapping activities is evidenced by the more profound defects in Ig class switch recombination and in activation of DNA damage responses in lymphocytes deficient in both Atm and DNA-PKcs compared with those with isolated deficiencies of either protein (19, 20). Moreover, whereas mice deficient in Atm or DNA-PKcs are viable and live well into adulthood, those with a combined deficiency of these proteins exhibit early embryonic lethality (21, 22).
Given these observations, we reasoned Atm and DNA-PKcs could have important overlapping functions in the repair of RAG-mediated DSBs, and more specifically during signal joint formation, in developing lymphocytes. To test this notion, we used several approaches to generate cells with isolated or combined deficiencies in Atm and DNA-PKcs and in which chromosomal V(D)J recombination can be induced and the processing and joining of signal ends assessed. Our findings reveal that Atm and DNA-PKcs have important overlapping activities during signal joint formation and in preventing the aberrant resolution of signal ends and, furthermore, that these activities depend on the catalytic activities of these two proteins.
Results
Chromosomal and Extrachromosomal Signal Joint Formation.
RAG-mediated DNA DSBs can be induced and their repair assayed in Abelson-transformed pre-B cells, hereafter referred to as abl pre-B cells (13, 23). Treating these cells with the abl kinase inhibitor STI571 (Imatanib or Gleevec) leads to G1 cell cycle arrest, induction of RAG expression, and robust rearrangement of chromosomally integrated retroviral recombination substrates (13, 23). The pMX-DELSJ retroviral recombination substrate contains a single RS pair that, upon rearrangement, forms a chromosomal signal joint (Fig. 1A) (13). The pMX-DELCJ retroviral recombination substrate is identical to pMX-DELSJ except that the RSs have been inverted such that the signal joint forms on the excised extrachromosomal DNA fragment (Fig. S1).
Chromosomal signal joint formation in Atm- and DNA-PKcs- deficient abl pre-B cells. (A) Schematic of unrearranged (UR) pMX-DELSJ and pMX-DELSJ with a signal end (SE) cleavage intermediate and a signal joint (SJ) product. Retroviral long terminal repeats (LTR), packaging sequence (ψ), GFP cDNA, IRES-hCD4 cDNA (i-hCD4), 5′ 12-RS (filled triangle), and 3′ 23-RS (open triangle) are shown. Approximate positions of the EcoRV (E) site and C4b probe are also shown. (B) EcoRV-digested genomic DNA samples from Lig IV−/−:DELSJ (Lig4 25.3.4sj1), WT:DELSJ (A70.2sj35), Atm−/−:DELSJ (ATM2Esj87), and SCID:DELSJ (SB1.1sj3) abl pre-B cells treated with STI571 for the indicated times were assayed for pMX-DELSJ rearrangement by Southern blotting using the C4b probe. TCRβ probe hybridization is shown as a DNA loading control.
Several WT (WT:DELSJ) and DNA Ligase IV−/− (Lig IV−/−:DELSJ) abl pre-B cells were generated with single pMX-DELSJ integrants (Table S1 and Fig. S2). Because DNA Ligase IV is required to join signal and coding ends, RAG induction in Lig IV−/−:DELSJ abl pre-B cells leads to the accumulation of unrepaired chromosomal signal ends without signal joint formation (Fig. 1B; 2.2 kb band, SE). In contrast, RAG induction in WT:DELSJ abl pre-B cells leads to robust pMX-DELSJ signal joint formation (Fig. 1B). Moreover, unrepaired pMX-DELSJ signal ends were not observed at any time after RAG induction in WT:DELSJ abl pre-B cells (Fig. 1B; 4-kb band, SJ). We conclude that in WT abl pre-B cells chromosomal signal ends generated by RAG cleavage are rapidly resolved as signal joints.
Previous studies have concluded that signal joining is less efficient than coding joint formation (4, 5). Because most of the signal ends analyzed in these studies exist on extrachromosomal DNA fragments excised during deletional rearrangements, we reasoned that the difference in these findings and ours may be due to unique requirements for the repair of chromosomal and extrachromosomal signal ends. In agreement with this notion, we readily detect unrepaired extrachromosomal pMX-DELCJ signal ends 48 h after RAG induction in WT:DELCJ abl pre-B cells (Fig. S1). Thus, in abl pre-B cells undergoing V(D)J recombination chromosomal (pMX-DELSJ) signal joints form more efficiently than extrachromosomal (pMX-DELCJ) signal joints.
Neither Atm nor DNA-PKcs Have Unique Activities During Chromosomal Signal Joint Formation.
Atm and DNA-PKcs have unique activities during coding joint formation. To assess the function of Atm and DNA-PKcs in chromosomal signal joint formation, several Atm−/− and SCID (DNA-PKcs deficient) abl pre-B cells were generated with single pMX-DELSJ integrants (Atm−/−:DELSJ and SCID:DELSJ) (Table S1). Both Atm−/−:DELSJ and SCID:DELSJ abl pre-B cells exhibit robust pMX-DELSJ signal joint formation with unrepaired chromosomal signal ends rarely detected at 48 or 96 h after RAG induction in these cells (Fig. 1B). The signal joints that form in Atm−/−:DELSJ abl pre-B cells are relatively precise, whereas those that form in SCID:DELSJ abl pre-B cells show increased imprecision as previously reported (Fig. S3) (8, 11, 12).
Studies implicating DNA-PKcs in signal joint formation have focused primarily on extrachromosomal signal ends (7, 9–12). In agreement, we find that unrepaired extrachromosomal pMX-DELCJ signal ends persist in SCID:DELCJ abl pre-B cells for up to 96 h after RAG induction, whereas these signal ends are largely resolved as signal joints by 96 h in both WT:DELSJ and Atm−/−:DELSJ abl pre-B cells (Fig. S1). Thus, DNA-PKcs promotes precise signal joining and efficient extrachromosomal signal joint formation. However, neither Atm nor DNA-PKcs have essential independent activities during chromosomal signal joint formation.
Atm and DNA-PKcs Have Redundant Functions During Chromosomal Signal Joint Formation.
To determine whether Atm and DNA-PKcs have overlapping functions during chromosomal signal joint formation, we generated lentiviral vectors to express Atm-specific shRNAs in Scid abl pre-B cells. That these shRNAs can achieve a functional knockdown of Atm was confirmed by analysis of WT abl pre-B cells harboring these shRNAs and the pMX-INV recombination substrate, which undergoes rearrangement by inversion (WT:INV; Fig. S4A) (13). Knockdown of Atm in WT:INV abl pre-B cells leads to defects in pMX-INV rearrangement similar to those observed in Atm−/−:INV abl pre-B cells, including increased pMX-INV hybrid joint formation and the accumulation of unrepaired pMX-INV coding ends (Fig. S4 B–D) (13). Thus, the knockdown of Atm achieved through the use of Atm-specific shRNAs recapitulates defects in V(D)J recombination observed in Atm−/− abl pre-B cells (13).
As expected, knockdown of Atm in WT:DELSJ abl pre-B cells had no effect on chromosomal signal joint formation (Fig. 2 A and B). In sharp contrast, knockdown of Atm in SCID:DELSJ abl pre-B cells led to a significant accumulation of unrepaired chromosomal signal ends (Fig. 2B). Moreover, inhibition of residual Atm activity in these cells with an Atm kinase inhibitor (KU55933) led to a further reduction in pMX-DELSJ signal joint formation and an increased accumulation of unrepaired signal ends (Fig. 2C). Expression of a nontargeting (nt) shRNA in SCID:DELSJ abl pre-B cells or the Atm shRNA in Atm−/−:DELSJ abl pre-B cells had no effect on signal joint formation (Fig. 2B). Expression of a different Atm-specific shRNA in an additional SCID:DELSJ abl pre-B cell line led to a similar defect in chromosomal signal joint formation (Fig. S5). We conclude that Atm and DNA-PKcs have overlapping functions in chromosomal signal joint formation.
Accumulation of unrepaired signal ends in abl pre-B cells deficient in both Atm and DNA-PKcs proteins. (A) Western blot analyses of Atm and DNA-PKcs in WT:DELSJ (A70.2sj35), Atm−/−:DELSJ (ATM2Esj87), and SCID:DELSJ (SB1.1sj10) cell lines expressing either nontargeting (nt) or Atm-specific lentiviral shRNA vectors. PLCγ2 is shown as a protein loading control. (B) Southern blot analysis of STI571-treated abl pre-B cells from A and Lig IV−/−:DELSJ (Lig4 25.3.4sj1) carried out as described in the legend of Fig. 1B. (C) Southern blot analysis of pMX-DELSJ rearrangement in Atm−/−:DELSJ and SCID:DELSJ lines from B, treated with KU55933 or DMSO vehicle. TCRβ probe hybridization is shown as a DNA loading control. Signal end accumulation in Lig IV−/−:DELSJ (Lig4 25.3.4sj31) cells is also shown.
At later times (96 h) after the induction of RAG, unrepaired signal ends that accumulate in cells with combined deficiencies in Atm and DNA-PKcs appear to be resected compared with those found in DNA Ligase IV-deficient cells (Fig. 2 B and C). Thus, Atm and DNA-PKcs have overlapping activities in preventing signal end resection that could conceivably be linked to their function in the joining of these DNA ends. In this regard, Atm and DNA-PKcs can both phosphorylate H2AX, forming γ-H2AX in chromatin flanking RAG DSBs, and γ-H2AX inhibits the resection of signal and coding ends in G1-phase lymphocytes (24–26).
Kinase Activities of Atm and DNA-PKcs Are Redundant During Chromosomal Signal Joint Formation.
To determine whether the overlapping functions of Atm and DNA-PKcs in signal joint formation rely on their kinase activities, we treated WT:DELSJ, Atm−/−:DELSJ, and SCID:DELSJ abl pre-B cells with kinase inhibitors specific for Atm (KU55933) or DNA-PKcs (NU7026 or NU7441). Chromosomal signal joint formation was unaffected in WT:DELSJ abl pre-B cells treated with any of these inhibitors (Fig. 3A and Fig. S6). In contrast, treatment of Atm−/−:DELSJ abl pre-B cells with either of the DNA-PKcs kinase inhibitors (Fig. 3 and Fig. S6) or treatment of SCID:DELSJ abl pre-B cells with the Atm kinase inhibitor (Fig. 3B) led to an accumulation of unrepaired signal ends. Treatment of Atm−/−:DELSJ abl pre-B cells with the Atm kinase inhibitor or SCID:DELSJ abl pre-B cells with the DNA-PKcs kinase inhibitors had no effect on signal joint formation, thereby confirming the specificity of these inhibitors (Fig. 3 and Fig. S6).
Atm or DNA-PKcs kinase activity is required for normal signal joint formation. (A) Southern blot analysis of pMX-DELSJ rearrangement in C4b-probed EcoRV-digested genomic DNA (Fig. 1A) from WT:DELSJ (A70.2sj35), Atm−/−:DELSJ (ATM2Esj87), and SCID DELSJ (SB1.1sj10) abl pre-B cell lines treated with STI571 and the DNA-PKcs kinase inhibitors NU7026 or NU7441 or DMSO vehicle (−) for the indicated times. Lig IV−/−:DELSJ (Lig4 25.3.4sj1) cells treated with STI571 also are shown. (B) Southern blot analysis of pMX-DELSJ rearrangement (as described in A) in Atm−/−:DELSJ (ATM2Esj80), and SCID:DELSJ (SB1.1sj18) cells treated with STI571 and NU7026, KU55933, or DMSO vehicle and Lig IV−/−:DELSJ (Lig4 25.3.4sj31) abl pre-B cells treated with STI571.
At endogenous antigen receptor loci, chromosomal signal joints are formed during rearrangements that occur by inversion. In agreement with our analysis of deletional pMX-DELSJ rearrangement, inhibition of DNA-PKcs kinase activity in Atm−/−:INV abl pre-B cells also led to an increase in unrepaired signal ends during inversional rearrangement of pMX-INV (Fig. S7). Together, these findings demonstrate that Atm and DNA-PKcs have overlapping catalytic activities that are required for the efficient formation of chromosomal signal joints.
Aberrant Signal End Resolution Is Suppressed by Atm and DNA-PKcs.
Although Atm suppresses the aberrant resolution of coding ends as potentially dangerous chromosomal translocations or deletions, Atm appears to be dispensable for suppressing aberrant signal end resolution (13). To determine whether Atm and DNA-PKcs have important overlapping activities in suppressing aberrant signal end resolution, we assessed pMX-DELSJ configuration in clones of abl pre-B cells after induction of V(D)J recombination. Specifically, abl pre-B cell lines were treated with STI571 for 96 h followed by removal of STI571 and subcloning. The configuration of pMX-DELSJ in individual subclones was determined by Southern blotting, which can distinguish clones with unrearranged pMX-DELSJ from those where rearrangement has formed either a normal chromosomal signal joint or an aberrant joint based on the size of hybridizing DNA fragments (Fig. 4 A and B).
Atm and DNA-PKcs suppress aberrant signal joint formation. (A) Schematic of the pMX-DELSJ configurations in abl pre-B cells recovered after treatment with STI571. The (UR) and (SJ) configurations are described in Fig. 1A. Most aberrant joints (*) will yield hybridizing bands greater that 2.2 kb, which is the distance from the EcoRV site in the LTR to the SE (open triangle). (B) Representative Southern blot analyses of pMX-DELSJ configurations in Atm−/−:DELSJ (ATM2Esj80) clones recovered after treatment with STI571 and NU7026 or DMSO vehicle for 96 h. Recovered clones with aberrant rearrangements are indicated (*). (C) Relative frequencies of aberrant pMX-DELSJ rearrangements in WT: DELSJ (A70.2sj35), Atm−/−:DELSJ (ATM2Esj87), and SCID:DELSJ (SB1.1sj10) clones recovered after treatment with STI571 and NU7026 or DMSO vehicle for 96 h. P value measures for the statistical significance of observed differences were determined using a one-tailed Fisher exact test. Numbers of clones analyzed with rearrangements are noted for each cell line. Aberrant rearrangements characterized in detail from Atm−/−:DELSJ clones recovered after 96 h of RAG induction and NU7026 treatment are in Fig. S8.
WT:DELSJ, Atm−/−:DELSJ and SCID:DELSJ abl pre-B cells clones isolated after the induction of V(D)J recombination exhibit low and approximately equivalent levels of aberrantly resolved pMX-DELSJ signal ends (Fig. 4C). In contrast, inhibition of DNA-PKcs kinase activity in Atm−/−:DELSJ abl pre-B cells led to a significant increase in cells with aberrantly resolved pMX-DELSJ signal ends (Fig. 4C). Notably, inhibition of DNA-PKcs kinase activity in WT:DELSJ abl pre-B cells did not lead to an increase in aberrant signal end resolution compared with WT:DELSJ treated with vehicle alone (Fig. 4C). The breakpoints of nine aberrantly resolved pMX-DELSJ signal ends from Atm−/−:DELSJ abl pre-B cells treated with the DNA-PKcs inhibitor were cloned and sequenced (Fig. S8). Eight of these breakpoints lie in the IgLκ locus within 20 bp of an RS and would form a chromosomal translocation between pMX-DELSJ on chromosome 7 and the IgLκ locus on chromosome 6 (Fig. S8). The other aberrant joint involved a breakpoint on chromosome 7 that generated a 68 kb deletion (Fig. S8). The bias for breakpoint targets at the IgLκ locus, which also undergoes inducible rearrangement in these cells, is reminiscent of what was observed for the aberrant resolution of coding ends in Atm-deficient abl pre-B cells (13, 27).
Discussion
We show that, after induction of V(D)J recombination in abl pre-B cells, chromosomal signal ends (pMX-DELSJ) are efficiently resolved into signal joints, whereas extrachromsomal signal ends (pMX-DELCJ) can persist unrepaired for a longer period. Moreover, in WT abl pre-B cells, chromosomal signal joints (pMX-DELSJ) are formed with kinetics similar to that observed for chromosomal coding joints (pMX-DELCJ) (13). These findings suggest that the presence of unrepaired pMX-DELCJ signal ends, but not coding ends, in WT lymphocytes may reflect differences in the ability of NHEJ to efficiently repair chromosomal vs. extrachromosomal DNA ends rather than differences in the efficiency of signal and coding joint formation per se. These findings could reflect mechanistic differences in the NHEJ-mediated repair of chromosomal vs. extrachromosomal DNA ends in general. Furthermore, chromosomal and extrachromosomal DNA ends may differ in their ability to activate DNA damage response pathways.
Although DNA-PKcs appears to have independent activities in extrachromosomal signal joint formation, neither DNA-PKcs nor Atm is uniquely required for efficient chromosomal signal end joining. Rather, these proteins have critical overlapping activities in chromosomal signal joint formation. After the induction of V(D)J recombination, a combined deficiency in Atm and DNA-PKcs leads to an increase in unrepaired chromosomal signal ends with these DNA ends frequently resolved aberrantly as potentially dangerous chromosomal translocations and deletions. The kinase activities of Atm and DNA-PKcs are critical for their function in chromosomal signal joint formation, suggesting that these proteins phosphorylate common substrates that participate in this process. Many NHEJ factors required for V(D)J recombination can be phosphorylated by DNA-PKcs and/or Atm either in vitro or in vivo, including Artemis, DNA Ligase IV, XRCC4, Ku70, Ku80, and XLF (3, 28–31). However, mutational analyses have yet to reveal a clear function for these phosphorylation events in either coding or signal joint formation.
Atm and DNA-PKcs kinase activities could be important for promoting the dissociation of proteins from DSB complexes that otherwise would inhibit signal end joining. In this regard, Rag-1 and Rag-2 avidly bind to signal ends in postcleavage complexes generated in vitro, and this association inhibits signal end ligation (32). Rag-1 and Rag-2 both have serine/threonine residues that could be targets for Atm or DNA-PKcs, and Rag-2 can be phosphorylated by DNA-PKcs in vitro (33, 34). Thus, Atm or DNA-PKcs could promote the removal of Rag-1 and Rag-2 from signal ends before their joining. DNA-PKcs binds to broken DNA ends and may itself have to be removed through a series of phosphorylation events mediated by DNA-PKcs auto-phosphorylation or by Atm (35–39). However, we find that DNA-PKcs-deficient cells require Atm kinase activity for efficient signal joint formation. Thus, during signal joint formation, Atm and DNA-PKcs must function in ways other than promoting the dissociation of DNA-PKcs from signal ends.
During deletional rearrangement, extrachromosomal signal joint formation may be important to preserve genomic stability by preventing excised signal end-flanked DNA segments from being reinserted ectopically into the genome (40–42). However, during rearrangements that occur by inversion, signal joints form within the chromosomal context and, as such, the joining of these signal ends is critical for preserving the linear integrity of the chromosome. In addition, RAG can also generate DNA DSBs at single chromosomal signal joints with the rejoining of these DNA ends also needed to preserve the linear integrity of the chromosome (43). Thus, during inversional rearrangements, efficient signal joint formation and the rejoining of DSBs generated by RAG cleavage at these joints are needed to maintain genomic stability. There are single V gene segments in the T-cell receptor β (Vβ14) and δ (Vδ5) loci that undergo rearrangement by inversion in developing T cells. Moreover, approximately half of the 160 Vκ gene segments in the IgLκ locus undergo rearrangement by inversion in developing B cells. Thus, the overlapping functions of Atm and DNA-PKcs in promoting chromosomal signal joint formation would be important for both developing B and T lymphocytes. Finally, RAG cleavage produces blunt signal ends with 5′ phosphates and 3′ hydroxyl groups that are likely similar in structure to DSBs generated under other circumstances. We speculate that the overlapping activity of DNA-PKcs and Atm during normal signal joint formation may be more broadly required for DSB repair and for suppressing the aberrant resolution of DNA breaks generated by genotoxins or as intermediates in other essential physiologic processes.
Methods
Mice.
Animals were housed in a specific pathogen-free animal facility at Washington University. Animal protocols were approved by the Washington University Institutional Animal Care and Use Committee.
Generation and Culture of abl pre-B Cell Lines.
V-abl–transformed pre-B cells were generated by culturing bone marrow from 3- to 5-wk-old mice with the pMSCV v-abl retrovirus as described previously (Table S1) (13). All cells were generated from mice harboring the Eμ-Bcl-2 transgene (13). DNA Ligase IV−/− (Lig IV−/−) abl pre-B cells were generated from DNA Ligase IVloxP/loxP abl pre-B cells by incubation of these cells with a Tat-Cre fusion protein (44). Briefly, abl pre-B cells were incubated for 1 h in media containing 50 μg/mL Tat-Cre, allowed to recover for 4 d, and then subcloned by limiting dilution. Lig IV−/− abl pre-B cells then were identified by PCR. Abl pre-B cell clones with single chromosomal integrants of the pMX-DELCJ, pMX-DELSJ and pMX-INV retroviral recombination substrate were made as previously described (Table S1) (13). Cells were treated with 3.0 μM STI571 (Novartis) for the indicated times at 106 cells/mL The Atm kinase inhibitor KU55933 (KuDos) was used at 15 μM. Cells were treated with DNA-PKcs kinase inhibitors NU7026 (Sigma) and NU7441 (KuDos) at 20 μM and 5 μM, respectively.
Construction and Use of Lentiviral shRNA Viral Vectors.
Atm and nt shRNAs were cloned into the pFLRU lentiviral vector (45). The sequences of the forward (F) and reverse (R) oligonucleotides used to generate the nt, Atm-specific shRNAs are in Table S2. To generate lentiviral vectors, 1 × 106 HEK293T cells plated in a 6 cm2 plate were cultured to 70–80% confluency before transfection. 4 μg pFLRU:shRNA vector was cotransfected with 4 μg pHR′Δ8.2R packaging plasmid and 0.5 μg pCMV-VSVg envelope plasmid using Lipofectamine 2000 (Invitrogen) reagent per the manufacturer's instructions. Culture media was changed 24 h after transfection, and virus-containing media was harvested 48 h after transfection and stored at −80 °C until used for infection. To transduce abl pre-B cells, 5 × 106 cells in 2.5 mL medium were combined with 2.5 mL virus-containing medium per well of a six-well tissue culture plate. Polybrene was added to a final concentration of 5 μg/mL. Cultures were centrifuged at 1,800 rpm (650 × g) for 90 min at room temperature. Cells expressing pFLR:shRNA vectors were identified by YFP expression 72 h after transduction and purified by flow-cytometric sorting.
Southern Blotting and PCR Analysis.
Southern blot analyses were carried out on genomic DNA from cells harboring pMX-INV, pMX-DELCJ, and pMX-DELSJ as previously described (13). Probe C4b is an 880bp PCR product amplified from pMX-delSJ using primers hCD4PR-1F and hCD4PR-2R (Table S2). Probe TCRβ is a 604-bp fragment between Vβ12 and Vβ13 amplified from genomic DNA using primers TCRβ-F and TCRβ-R (Table S2). G probe is a 780-bp fragment amplified from pMX- DELSJ using primers GFP-F and GFP-R (Table S2). Cloning and sequencing of SJs from the substrate pMX-DELSJ was performed as previously described.
Western Blot Analyses.
Western blotting was carried out as previously described except that 5% acrylamide gels were used (46). Primary antibodies were used at the following concentration dilutions: anti-Atm, 1:500 (Genetex, GTX70107), anti–DNA-PKcs, 1:500 (MS-423-P1; Thermo-Fisher), anti-PLCγ2, 1:500 (Q20 sc-407; Santa Cruz). Secondary antibodies were used at the following concentration dilutions: goat antimouse, 1:5,000 (62–6520; Invitrogen), donkey antirabbit F(ab')2 Fragment, 1:5,000 (45000-683; Fisher).
Acknowledgments
We thank Dr. Eugene Oltz for critical review of the manuscript and Drs. Greg Longmore and Yunfeng Feng for pFLRU, pHR’Δ8.2R, and pCMV-VSVg. This work is supported by National Institutes of Health Grants AI074953 (to B.P.S.), AI47829 (to B.P.S.), CA136470 (to B.P.S and C.H.B), CA125195 (to C.H.B.), and CA096832 (to P.J.M.). B.Y. and E.J.G. are supported by the Cancer Research Institute Pre-doctoral Emphasis Pathway in Tumor Immunology Training grant awarded to the University of Pennsylvania (to B.Y.) and Washington University (to E.J.G.). J.J.B. is supported by a National Institutes of Health Ruth L. Kirschstein National Research Service Award (T32 HD007499) and a Children's Discovery Institute Fellows Award. C.H.B. is a Pew Scholar in the Biomedical Sciences program.
Footnotes
The authors declare no conflict of interest.
*This Direct Submission article had a prearranged editor.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1013295108/-/DCSupplemental.