Introduction

Protein arginine methylation results in the addition of one or two methyl groups to the guanidino nitrogen atoms of arginine 1. There are three main forms of methylated arginine identified in eukaryotes: ω-NG-monomethylarginines, ω-NG,NG-asymmetric dimethylarginines and ω-NG,NG-symmetric dimethylarginines. There are nine protein arginine N-methylation enzymes (PRMTs) that catalyze the transfer of a methyl group from S-adenosylmethionine to a guanidino nitrogen of arginine 2. Arginines located within glycine-arginine-rich (GAR) motifs are preferred sites of methylation of several PRMTs including PRMT1 2, the predominant enzyme in mammalian cells, responsible for most of the asymmetrical arginine dimethylation reactions in human cells 3. PRMT1-null mice are embryonically lethal and contain numerous hypomethylated substrates 4. Using a conditional null PRMT1 allele in mice, it was shown that PRMT1-deficient mouse embryonic fibroblasts (MEFs) exhibit spontaneous DNA damage, chromosome instability, polyploidy and defective checkpoint activation following DNA damage 5. These findings highlight the importance of arginine methylation in the DNA damage response pathway and the maintenance of genomic stability. A large number of GAR motif-containing proteins were shown to be arginine methylated using candidate and proteomic approaches 6. These substrates are functionally involved in diverse intracellular processes from nuclear export and gene expression to DNA damage signaling 2.

The MRE11/Rad50/NBS1 (MRN) complex is the primary sensor rapidly recruited to DNA double-strand breaks (DSBs) 7, 8, 9. The MRN complex tethers the DNA ends 10, 11, leading to the recruitment and activation of the ATM kinase 12, 13. The MRN complex is also thought to participate in ATR-CHK1 kinase activation by facilitating DNA end resection 14, 15, 16, 17, 18. It participates in multiple downstream pathways for checkpoint signaling, DNA replication, telomere maintenance, non-homologous end joining and meiotic recombination 19, 20, 21, 22, 23, 24.

The in vivo function of the MRN complex has been revealed by hypomorphic alleles in human MRE11 and NBS1, leading to ataxia-telangiectasia-like disorder (ATLD) and Nijmegen breakage syndrome (NBS), respectively 25, 26. These disorders share characteristic features including hypersensitivity to ionizing radiation, immunodeficiency and an increased predisposition to the development of malignancies 27. The in vivo deletion of components of the MRN complex is embryonically lethal in mice 28, 29, 30, 31. However, animal models of ATLD (Mre11ATLD1/ATLD1) and NBS have been generated and share many common features of the human disorders 32, 33.

MRE11 is a conserved protein with an N-terminal nuclease domain 11, 34 and a DNA-binding region 10, 35 encompassing the GAR motif 36. MRE11 nuclease-defective mice have been generated (Mre11H129N/H129N), defining a physiological role of the MRE11 nuclease activity in homologous recombination and the maintenance of genomic stability 30. Mammalian MRE11 is also involved in the classical and alternative non-homologous end-joining (NHEJ) pathways 21, 23. The arginines within the MRE11 GAR motif are asymmetrically dimethylated by PRMT1 and have been shown to regulate its exonuclease activity in vitro 36, 37, but the physiological significance of arginine methylation remains to be elucidated. In this study, we generated an Mre11 allele (Mre11RK) in mice that substitutes the arginines with lysines within the GAR motif. We report the requirement for the MRE11 GAR motif in regulating the ATR activation during DNA damage signaling and the maintenance of genomic stability.

Results

Generation of Mre11RK knock-in mice

We generated a mouse knock-in allele at the Mre11 locus that substitutes the nine arginines within the GAR motif with lysines (Figure 1A). This Mre11RK allele was generated to assess the in vivo physiological role of the methylarginines within the MRE11 GAR motif. Lysine was chosen to maintain the positive charge of the residues. The Mre11RK allele was generated by homologous recombination targeting exon 14 that encodes the GAR motif (Figure 1B). Mouse genotypes were verified by PCR using genomic DNA (Figure 1C). Moreover, the genomic DNA and the mRNA expressed from this mutant allele were sequenced, verifying that the codons encoding the nine arginines within the GAR motif were replaced with lysine-encoding codons (data not shown). The Mre11RK allele was also engineered to introduce an EcoRI site within the middle of the exon 14 to distinguish the mRNAs encoding wild-type Mre11 and Mre11RK. As expected, the DNA fragment generated by RT-PCR spanning the exon 14 was undigested by EcoRI in wild-type cells, digested 50% and 100% in Mre11RK/+ and Mre11RK/RK cells, respectively, further confirming the genotypes (Figure 1D).

Figure 1
figure 1

Generation of Mre11RK/RK mice and MEF cells. (A) Schematic representation of the wild-type MRE11 with the glycine-arginine rich (GAR) motif and the sequence of MRE11RK substituting the arginine with lysine residues. (B) Schematic representation of the mouse Mre11 allele and Mre11RK allele. The initial strategy (MRE11RKNeo) shows the targeting position at the exon 14, with substitution of arginine with lysine residues as shown in A and the targeting plasmid structural elements. Shown in Mre11RK is the targeted locus with deletion of Neo. The exons are the black boxes and the line represents introns not drawn to scale. The white and blank triangles denote loxP sites and FRT sites, respectively, and the small arrows denote the primers used for PCR analysis. The expected size of the PCR DNA fragment amplified by the primer pair of primer 3 and primer 6 for the wild-type allele is 525 bp, while the size of the DNA fragment for the Mre11RK allele is 591 bp. (C) Genomic DNA isolated from primary MEFs was analyzed by PCR using the primer pair of primer 3 and primer 6 as indicated in B and the DNA fragments visualized on an ethidium-bromide-stained agarose gel. M denotes molecular mass markers of the 1 kb ladder (Invitrogen). (D) Total cellular RNA was isolated from primary MEFs and subjected to reverse transcription-PCR. The DNA fragment was purified and digested with EcoRI and then separated on an agarose gel. M denotes molecular mass markers of the 1 kb ladder (Invitrogen).

The Mre11RK/RK mice were born in the expected Mendelian ratio and did not display any overt phenotypes (data not shown), unlike the Mre11Δ/Δ and Mre11H129N/H129N nuclease-defective mice, which die during early embryogenesis 30. The Mre11RK/RK females and males were fertile and gave rise to normal litter sizes of six to eight pups (data not shown), unlike the Mre11ATLD1/ATLD1 mice where the females are subfertile 32.

Mre11RK/RK mice and MEFs are hypersensitive to γ-IR

We first decided to challenge the mice with γ-irradiation (IR), as a known phenotype of MRN hypomorphic alleles is their hypersensitivity to IR. Cohorts of Mre11+/+, Mre11RK/+ and Mre11RK/RK mice were irradiated with 10 Gy of IR and closely monitored for radiation toxicity. All of the Mre11RK/RK mice succumbed to 10 Gy of IR treatment within 2 weeks, while less than half of the wild-type and the Mre11RK/+ mice died within 35 days (Figure 2A).

Figure 2
figure 2

Mre11RK/RK mice and MEFs are hypersensitive to IR, but the B lymphocytes have no significant impact on class switching recombination. (A) Mre11+/+, Mre11+/RK and Mre11RK/RK mice were treated with 10 Gy IR and monitored for signs of radiation toxicity over 35 days. Mre11RK/RK mice showed a significant reduction in survival rates compared to Mre11+/+ and Mre11+/RK mice according to the log-rank test (P = 0.0066). The percentage survival was plotted as a function of days post IR. (B) Approximately 200-400 immortalized MEFs were seeded on a 10 cm tissue culture dish and treated with various doses of IR. The cells were then maintained in regular medium. Fourteen to twenty days later, the cells were fixed and cell colonies were stained with crystal violet and counted. The colony number was normalized to percentage of untreated cells and plotted as a function of IR dosage. The graphs shown represent the average and standard deviation (SD) from four independent immortalized MEF cell lines (clones) of each genotype. The asterisks denotes P < 0.01 using the Student's t-test. (C) Flow cytometric analysis of immunoglobulin class switching from IgM to IgG in B lymphocytes cultured with IL-4 and anti-CD40 for four days. Bar graph depicts comparisons of IgG1+ cell populations relative to Mre11+/cond control, from an average (+SD) of three mice per genotype. Mre11cond is Mre11Δ in B lymphocytes. (D) Flow cytometric analysis of immunoglobulin class switching from IgM to IgG in B lymphocytes cultured with IL-4 and anti-CD40 for 4 days shown in panel C.

The hypersensitivity to IR was also demonstrated in the cells isolated from Mre11RK/RK mice. Immortalized Mre11RK/RK and wild-type MEFs were generated and treated with varying IR doses. After IR treatment, the cell colony number was significantly reduced in Mre11RK/RK MEFs, demonstrating more than 100-fold hypersensitivity to IR compared to wild-type MEFs (Figure 2B).

MRE11 participates in NHEJ pathways 21, 23, which are required for class switch recombination (CSR). Complete absence of MRE11 in developing B lymphocytes causes a significant reduction in CSR, whereas MRE11 defective only in nuclease activities causes a mild CSR defect 23. To determine if the MRE11 GAR motif is required for its function in NHEJ, we compared the impact on CSR of Mre11RK/RK to that of Mre11Δ/Δ and Mre11Δ/H129N. Conversion of Mre11cond to Mre11Δ in mature IgM+ B lymphocytes was facilitated by the CD21-Cre transgene 23, 38. CSR was assessed in B lymphocytes isolated from spleens. Comparing Mre11RK/RK to Mre11+/+, no difference in spleen size or cellularity was evident (data not shown). We induced switching from IgM to IgG1 and noted defects in Mre11Δ/Δ and Mre11Δ/H129N cells similar to those reported previously (Figure 2C, 2D) 23. In contrast, Mre11RK/RK cells display a small statistically significant difference (P = 0.0409), which is unlikely to be biologically significant (Figure 2C, 2D). Taken together, these findings show that we have generated a novel Mre11 hypomorphic allele and that the GAR motif affects sensitivity to IR, but has no major effects on mouse viability, fertility or CSR.

Defective genomic stability and checkpoint control in Mre11RK/RK MEFs

Metaphase spreads were then prepared to determine chromosome damage with passage 2 primary Mre11RK/RK and wild-type MEFs treated with 1 Gy of IR. We observed a significant increase in chromosomal anomalies in these Mre11RK/RK MEFs (Figure 3A, left panel, and quantified in Figure 3B). Genomic instability was also observed in later stage Mre11RK/RK MEFs without DNA damage treatment (Figure 3A, right panel, and quantified in Figure 3C). For example, in passage 7 primary cells, 60% of Mre11RK/RK MEFs harbored at least one chromosome aberration per metaphase, compared to 22% in wild-type cells (Figure 3C). Interestingly, we observed a significant number of radial chromosomes in the passage 7 Mre11RK/RK MEFs but not in the wild-type MEFs (Figure 3A, right panel and Table 1). These findings demonstrate the requirement for the MRE11 GAR motif for the maintenance of chromosomal stability.

Figure 3
figure 3

Chromosomal instability and DNA damage checkpoint defect in Mre11RK/RK MEFs. (A) Representative metaphases from IR-treated passage 2 Mre11RK/RK MEF cells (left) and untreated passage 7 Mre11RK/RK MEF cells (right). Breaks were indicated by arrows and radial chromosomes were cycled. (B) Chromosome breakage analysis of IR-treated passage 2 wild-type and Mre11RK/RK MEFs. The metaphases were grouped into four categories according to the number of chromosome aberrations per metaphase observed and the percentage of the metaphases in each group was calculated (left). The data were also expressed by average aberrations per metaphase at both 2N phase and 4N phase, respectively (right). Statistical significance was assessed using Student's t-test. *(RK/RK versus +/+; 2N) P < 0.05 and **(RK/RK versus +/+; 4N) P < 0.001. (C) Chromosome breakage analysis of untreated passage 7 wild-type and Mre11RK/RK MEFs. The result was presented as that in B. Statistical significance was assessed using Student's t-test. **(RK/RK versus +/+; 2N or 4N) P < 0.001. (D) G2/M checkpoint analysis. Mre11+/+, Mre11+/RK and Mre11RK/RK MEFs were left untreated or treated with 5 or 10 Gy of IR. At 1.5 h after treatment, the cells were fixed and stained with propidium iodide and anti-pS10-histone H3 antibody to identify the cells in mitosis. The percentage of pS10-histone H3-positive cells was determined by flow cytometry and expressed as a ratio of IR treated to non-IR treated. The experiments were performed more than three times for each dosage. Statistical significance was assessed using Student's t-test. *(RK/RK versus +/+ or +/RK; 5 Gy) P < 0.01 and **(RK/RK versus +/+ or +/RK; 10 Gy) P < 0.001.

Table 1 Mre11RK/RK MEFs accumulate radial chromosomes

In response to DNA damage, cell cycle checkpoints are activated to arrest cell cycle progression, allowing time for repair 39. ATLD cells are known to have checkpoint defects in response to DNA damage 25, 32. To examine the activation of the G2/M checkpoint of Mre11RK/RK MEFs, we measured the abundance of cells entering mitosis 90 min after IR treatment with anti-histone H3pS10 antibody and expressed it as a mitotic ratio with untreated cells. After 10 Gy IR treatment, a mitotic ratio of 0.1 was observed for wild-type and Mre11+/RK MEFs, suggesting that 10% of the cells progressed through M phase (Figure 3D). By contrast, a mitotic ratio of 0.3 was observed for Mre11RK/RK MEFs, suggesting that 30% of the cells progressed through the M phase (Figure 3D). A similar significant difference was also observed with 5 Gy IR (Figure 3D), suggesting that a significant fraction of the Mre11RK/RK cells lost their G2/M checkpoint.

MRN complex formation, localization to sites of DNA damage and ATM activation are normal in Mre11RK/RK MEFs

To define the molecular defects observed in Mre11RK/RK cells, we first assessed the integrity of the MRKRN complex. Using control and MRE11 immunoprecipitations, we showed that the MRE11RK protein was slightly less abundant than endogenous MRE11, but it maintained its interaction with NBS1 and RAD50 (Figure 4A). We also investigated the arginine methylation status of MRE11RK with the anti-methylarginine-specific antibody ASYM25b. Indeed, MRE11RK was hypomethylated in Mre11RK/RK MEFs (Figure 4A), and the newly introduced lysines were not acetylated (Supplementary information, Figure S1). We next examined the ability of the MRKRN complex to localize to sites of DNA damage by indirect immunofluorescence. MRE11RK localized at sites of DNA damage in response to IR treatment similar to wild-type MRE11 (Figure 4B). These findings show that MRE11RK maintains interactions within the MRN complex.

Figure 4
figure 4

MRN complex formation, localization to sites of DNA damage and IR-induced ATM activation are normal in Mre11RK/RK MEFs. (A) Whole-cell lysates from wild-type and Mre11RK/RKMEFs, respectively, were immunoprecipitated with anti-MRE11 antibody. The bound proteins were separated by SDS-PAGE and immunoblotted with indicated antibodies. (B) Mre11+/+ and Mre11RK/RK MEFs were treated with 10 Gy of IR or left untreated. Two hours later, the cells were visualized by indirect immunofluorescence with anti-MRE11 antibody. The scale bar represents 10 μm. (C) Mre11+/+ and Mre11RK/RK MEFs were treated with 10 Gy of IR or left untreated (−). The cells were harvested at the indicated times after IR treatment. Total cellular proteins were subjected to immunoblotting with anti-ATMpS1981, anti-γH2AX and anti-CHK2 antibodies.

We next examined whether Mre11RK/RK MEFs harbor defects in ATM activation, since MRE11 is required for recruiting and activating the kinase and cells with MRE11 deficiencies are impaired in this pathway 12, 25, 30, 32. The dynamic of ATM activation was normal in Mre11RK/RK MEFs following IR, as assessed by ATM autophosphorylation using the anti-ATMpS1981 antibody 40 and by evaluating phosphorylation of the ATM substrate CHK2, visualized as a slower migrating species by SDS-PAGE (Figure 4C). Moreover, the IR-induced foci of γ-H2AX and 53BP1 were equally formed in Mre11RK/RK and the wild-type MEFs (Figure 5), consistent with the MRKRN complex properly localizing at DSBs, resulting in ATM activation.

Figure 5
figure 5

IR-induced nuclear foci of γH2AX and 53BP1 are normal in Mre11RK/RK MEFs. Wild-type (Mre11+/+) and Mre11RK/RK MEF cells, respectively, were left untreated or treated with 10 Gy of IR. After 2 h of recovery, the cells were visualized by indirect immunofluorescence with anti-γH2AX and anti-53BP1 antibodies, respectively. Cells with > 5 foci were counted and expressed as a percentage. The graphs show the average and standard error of the mean from two independent experiments performed in duplicates, where > 20 different fields were analyzed. In total more than 500 cells were counted for each sample in each experiment. The scale bar represents 10 μm.

Defective CHK1 activation in Mre11RK/RK MEFs in response to IR treatment

Cells from ATLD patients have both impaired ATM activation and G2/M checkpoint control 32, whereas the MRE11 nuclease-defective cells have both normal ATM activation and G2/M checkpoint control, although MRE11 nuclease activity is essential for cell survival and DNA damage repair 30. Therefore, the Mre11RK hypomorphic allele provides a unique genetic system to assay ATM-independent MRE11 contributions to G2/M checkpoint control in mammalian cells. It has been proposed that the dual action of ATM and ATR is required to initiate the G2/M checkpoint in response to IR 41. Moreover, ATM and MRE11 are required to enhance IR-induced ATR-dependent CHK1 phosphorylation 14, 42, but the role of MRE11 and its GAR motif in ATR activation remains unclear. To assess ATR function, we monitored the phosphorylation of CHK1 using anti-CHK1pS345 antibodies. We observed a dramatic defect in CHK1 activation in Mre11RK/RK MEFs compared to wild-type cells in response to IR treatment (Figure 6A). A CHK1 activation defect was also observed in Mre11RK/RK MEFs by immunoprecipitating CHK1 and assaying its activity using an in vitro kinase assay (Figure 6B). CDC25A is a CHK1 substrate and its phosphorylation is required for its ubiquitination and degradation in response to IR 43. We observed that CDC25A degradation was impaired in Mre11RK/RK MEFs, compared to wild-type MEFs, after IR treatment, consistent with the Mre11RK/RK MEFs exhibiting a CHK1 activation defect (Figure 6C). We also observed a mild defect in CHK1 phosphorylation in Mre11RK/RK MEFs compared to wild-type cells in response to UV and hydroxyurea treatment, respectively (Supplementary information, Figure S2).

Figure 6
figure 6

IR-induced CHK1 phosphorylation and activation are defective in Mre11RK/RK MEFs. (A) Mre11+/+ and Mre11RK/RK MEFs were treated with 10 Gy of IR or left untreated (−). The cells were harvested at the indicated times after IR treatment. Total cellular proteins were subjected to immunoblotting with the indicated antibodies. (B) Mre11+/+ and Mre11RK/RK MEFs were treated with 10 Gy of IR or left untreated (−). The cells were harvested at the indicated times after IR treatment, and whole-cell lysates were subjected to immunoprecipitation with the anti-CHK1 antibody. The bound proteins were used for CHK1 activity assay as described in Materials and Methods. (C) MEFs were treated as in A. The cells were harvested at the indicated times after IR treatment. Total cellular proteins were subjected to immunoblotting with the anti-CDC25A and anti-α-tubulin antibodies as a loading control. (D) MEFs were infected with the empty retroviral vector pMSCV-puro (Vector) or the vector which expresses wild-type human MRE11 (WT) or MRE11 RK mutant (RK). The infected cells were selected with 2 μg/ml of puromycin for 2 days after infection. The cells were then treated with 10 Gy of IR (+) or left untreated (−). The cells were harvested at 15 min after IR treatment. Total cellular proteins were subjected to immunoblotting with the indicated antibodies.

We next performed a rescue experiment with ectopic expression of wild-type MRE11 and MRE11RK in Mre11RK/RK MEFs. We observed that increasing the protein levels of wild-type MRE11 in the Mre11RK/RK MEFs rescued the CHK1 phosphorylation defect (Figure 6D, lanes 9 and 10), but increasing the protein levels of MRE11RK had little or no effect on CHK1 activation (Figure 6D, lanes 11 and 12). The MRE11RK protein level in the MRE11RK viral vector-infected Mre11RK/RK MEFs was comparable to that in the empty viral vector-infected wild-type MEFs (see MRE11 panel, Figure 6D, comparing lanes 11 and 12 with lanes 1 and 2). These findings show that Mre11RK/RK MEFs with restored MRE11RK to levels of endogenous wild-type MRE11 still display impaired ATR/CHK1 activation. These findings show that the lack of a functional MRE11 GAR motif directly contributes to the ATR/CHK1 activation defects observed in Mre11RK/RK MEFs.

Defective exonuclease activity and DNA binding of MRKRN complex

The presence of RPA-coated single-strand DNA (RPA-ssDNA) promotes ATR-ATRIP recruitment, which leads to CHK1 phosphorylation and activates the G2/M checkpoint 44. Hence, the CHK1 activation defect in Mre11RK/RK MEFs might result from abnormal DNA resection, leading to a reduction in RPA-ssDNA formation. To assess these molecular mechanisms, we first monitored whether purified MRKRN displayed DNA resection defects compared to the wild-type MRN complex (Figure 7A). We showed previously that recombinant MRE11RK had impaired 3′-5′ exonuclease activity in vitro 36, 37, and we now extend these findings to show that MRE11RK within the MRKRN complex also has impaired 3′-5′ exonuclease activity. Interestingly, three times more MRKRN was required to achieve the same level of DNA resection as MRN, suggesting that the MRKRN complex has impaired resecting activity (Figure 7B, compare MRN at 5 nM and MRKRN at 15 nM). In addition, we performed a DNA-binding analysis comparing the ability of the MRN and MRKRN complexes to bind ssDNA, dsDNA and splayed arm DNA. As expected, the MRN complex bound both dsDNA and ssDNA, as well as splayed arm DNA (Figure 7C, 7D). The MRKRN complex bound all three types of DNA with relative weaker affinity than the MRN complex (Figure 7C, 7D). For example, at the concentration of 5 nM, the wild-type MRN complex bound almost 80% splayed arm DNA, whereas the MRKRN complex bound only 30% splayed arm DNA (Figure 7D). These findings suggest that the DNA-binding ability of MRE11 may be required for the processivity of its intrinsic exonuclease activity.

Figure 7
figure 7

Defect in exonuclease activity and DNA binding of MRKRN complex. (A) SDS-PAGE of purified wild-type MRN (WT) and MRKRN (RK). (B) Exonuclease assays of MRN (2.5-15 nM) and MRKRN (2.5-15 nM) on dsDNA. The concentration of protein is calculated in function of MRE11. (C) Competition electrophoretic mobility shift assays were performed with MRN or MRKRN and ssDNA (SS), dsDNA (DS) and splayed arm (SA) substrates. The concentration of protein is calculated in function of MRE11. (D) Quantification of the DNA binding of panel C.

Defective IR-induced RPA2 and RAD51 foci formation in Mre11RK/RK MEFs

To assess whether RPA-ssDNA complexes were formed in Mre11RK/RK MEFs, we assayed RPA foci formation in response to DSBs using anti-RPA2 antibodies after IR treatment. We observed a more than 50% decrease of Mre11RK/RK MEFs containing >5 RPA2 foci compared to wild-type MEFs (Figure 8A, 8B). Since RPA-ssDNA complexes are subsequently replaced with RAD51 for homologous recombination to proceed 45, 46, we examined the formation of RAD51 foci. Indeed, we also observed a significant reduction in RAD51 foci formation in Mre11RK/RK MEFs (Figure 8A, 8B). Two hours after 10 Gy IR treatment, 40% of wild-type MEFs had > 5 RAD51 foci compared to 20% in Mre11RK/RK MEFs (Figure 8A, 8B). At 4 h, 40% of the wild-type MEFs retained > 5 RAD51 foci, while the Mre11RK/RK MEFs with >5 RAD51 foci increased to nearly 30%. Moreover, Mre11RK/RK MEFs had less RPA2 foci per cell than wild-type MEFs (Figure 8C). These results show that the Mre11RK/RK MEFs are defective in recruiting RPA and RAD51 to DSBs.

Figure 8
figure 8

Mre11RK/RK MEFs have defects in IR-induced RPA2 and RAD51 foci formation. Mre11+/+ and Mre11RK/RK MEFs, respectively, were left untreated or treated with 10 Gy of IR. After varying hours of recovery, the cells were visualized by indirect immunofluorescence with anti-RPA2 or anti-RAD51 antibody. (A) A typical image was shown for each sample. (B) The cells with > 5 foci were counted and expressed as a percentage. The graph shows the average and standard error of the mean (SEM) from two independent experiments performed in duplicates, where > 20 different fields were analyzed. In total, more than 500 cells were counted for each sample in each experiment. Statistical significance was assessed using Student's t-test. *P < 0.05 and **P < 0.001. (C) The number of foci in the cells with > 5 RPA2 foci was counted. The graph shows the average number of foci in each cell, which has > 5 RPA2 foci and SEM from two independent experiments performed in duplicates. In total, more than 40 cells were counted for each sample. Statistical significance was assessed using Student's t-test. *P < 0.01 and **P < 0.001.

Discussion

In the present study, we have generated a mouse knock-in allele of Mre11 that substitutes the arginines with lysines in the GAR motif. The Mre11RK/RK mice were sensitive to γ-IR and the MEFs isolated from the Mre11RK/RK mice accumulated increased number of aberrant chromosomes, defining a physiological role for the GAR motif in maintaining genomic stability. The Mre11RK/RK MEFs displayed defects of ATR/CHK1 signaling and G2/M checkpoint activation, and reduced recruitment of RPA and RAD51 proteins to the damaged sites in response to γ-IR. MRE11RK was assembled normally within the MRKRN complex, localized to sites of DNA damage and normally activated the ATM pathway. In vitro biochemical analysis suggested that the MRKRN complex exhibited defects in exonuclease processivity and DNA binding, which are likely to be responsible for the impaired DNA end resection and ATR/CHK1 activation observed in the Mre11RK/RK MEFs in response to IR. Our findings provide genetic evidence for the critical role of the MRE11 GAR motif in DSB repair, and define a mechanistic link between the MRE11 GAR motif, MRN exonuclease processivity, the processing of DSBs and cell cycle checkpoint activation (Figure 9).

Figure 9
figure 9

Model. The MRE11 GAR motif is required for IR-induced ssDNA resection and ATR activation but not for ATM activation. Defects in the MRE11 GAR motif lead to defects in RPA-ssDNA complexes, leading to defects in ATR and CHK1 activation. Subsequently, RAD51 cannot be properly recruited, leading to DNA repair defects and genomic instability.

The GAR motif is a characteristic signature of clusters of methylated arginines. The arginines located within the GAR motif have been shown to mediate protein-protein and protein-nucleic acid interactions 2. Moreover, the GAR motif was shown to regulate protein localization, but this may be indirect due to protein-protein and protein-nucleic acid interactions. Using the unique genetic system and several biochemical and cytological methods, we demonstrate the molecular mechanism by which GAR motif regulates MRE11 function for the activation of ATR signaling and recruitment of RPA and RAD51 proteins to the DNA damage sites, both of which rely on ssDNA resection, a relatively later event of MRE11 action in response to DSBs. In contrast, the GAR motif mutation did not affect the initial response of the MRN complex to DSBs, as it was normally recruited to sites of DNA damage and normally activated the ATM pathway. The Mre11RK/RK mice provide an important tool to study the roles of the MRN complex at the later stage in response to DNA damage.

The physiological function of the MRE11 nuclease domain was assessed by the generation of mice carrying the nuclease-defective Mre11H129N mutation 30. Homozygous Mre11H129N/H129N mice are embryonically lethal, like Mre11-null mice, confirming the requirement for the MRE11 nuclease domain during embryogenesis. Cells from Mre11H129N/H129N mice exhibited hypersensitivity to IR and genomic instability. The MH129NRN complex was stable, localized to sites of DNA damage and activated ATM. The cellular defect of Mre11H129N/H129N mice lies in their inability to properly repair DNA by homologous recombination 30. We demonstrated significant impairment of the 3′-5′ exonuclease processivity of the MRKRN complex (Figure 7) 36, 37. The intrinsic activity of nuclease activities of MRE11 may be sufficient to promote fork degradation 47. In conjunction with other factors, such as 5′-3′ exonucleases CtIP and EXO1, the MRN complex is functionally involved in the extensive degradation, leading to DNA end resection and homologous recombination 15, 16, 48, 49. Our findings suggest that the GAR motif may modulate the MRE11 exonuclease activity within the MRN complex in vivo, contributing to the defective ssDNA resection observed in Mre11RK/RK MEFs (Figure 8).

The impaired DNA-binding activity of MRE11RK that we observed may also contribute to the defect of DNA end processing observed with the MRKRN complex. The MRE11 GAR motif is likely to be required to bind selective types of DNA ends at the sites of DNA damage to modulate ATR activation during DSB repair. Indeed, we found that MRN bound splayed arm DNA, representative of a DNA replication intermediate, preferentially over ssDNA and dsDNA substrates. We infer that the MRE11 exonuclease activity is required to generate ssDNAs needed to activate the ATR pathway via ssDNA binding by ATRIP. The fact that Mre11RK/RK mice are viable, suggests that the exonuclease defect we that observed is not as severe as observed in Mre11H129N/H129N mice in which both exonuclease and endonuclease activities are defective 30. We observed that Mre11RK/RK MEFs display a G2/M checkpoint defect not observed in Mre11H129N/H129N MEFs 30. These findings suggest that some phenotypes of Mre11RK/RK MEFs cannot be solely explained by impaired MRE11 exonuclease activity (Table 2). As such, it is important to note that cells from ATR-deficient Seckel Syndrome 50, like Mre11RK/RK MEFs, also display a G2/M checkpoint defect. In addition, we observed the formation of radial chromosomes in Mre11RK/RK MEFs (Figure 3A and Table 1), a feature also observed in ATR and Fanconi-deficient cells. Our results suggest that defects in the methylation of mammalian MRE11 GAR motif are likely to manifest themselves as defects in the level of ATR activation, leading to genomic instability and cell death.

Table 2 Comparison of Mre11RK/RK with Mre11ATLD1/ATLD1 and Mre11H129/H129 mice and MEF cells

In conclusion, our findings provide genetic and biochemical evidence for the critical role of the MRE11 GAR motif in DSB repair. Our data define a mechanistic link between the MRE11 GAR motif, MRN exonuclease processivity, DSB end processing and cell cycle checkpoint activation. The fact that the GAR motif is widely found in proteins implies that this mechanism of regulation may also occur in other systems.

Materials and Methods

Reagents and antibodies

Rabbit anti-MRE11 antibodies were from Novus Biologicals (Littleton, CO, USA), Cell Signalling Technology (Danvers, MA, USA) and GeneTex Inc. (Irvine, CA, USA), respectively. Rabbit anti-53BP1, anti-RAD50 and anti-mouse NBS1 antibodies were from Novus Biologicals. Mouse anti-ATM-pS1981 (mouse Ser1987) antibody was from Rockland (Gilbertsville, PA, USA). Mouse anti-γH2AX monoclonal antibody, rabbit anti-phospho-histone H3-Ser10, ASYM25b, and anti-CHK2 antibodies were obtained from Millipore (Billerica, MA, USA). Rabbit anti-RAD51, anti-CHK1 and anti-CDC25A antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Rabbit anti-phospho-CHK1 ser345 antibody was from Cell Signalling Technology. Rabbit anti-acetyl-lysine antibody was from Abcam (Cambridge, MA, USA). Rabbit anti-RPA2 antibodies were generated with a peptide located at the C-terminal end of the mouse RPA2 using the following peptide as antigen, NEGHIYSTVDDDHFKSTDAEC. Propidium iodide (PI) and the Alexa Fluor 488-conjugated goat anti-rabbit antibodies and anti-mouse antibodies were from Invitrogen (Carlsbad, CA, USA). Protein A-sepharose and mouse anti-α-tubulin monoclonal antibody were from Sigma (St Louis, MO, USA). Protease inhibitor cocktail and protein phosphatase inhibitor cocktail were from Roche (Mississauga, ON, USA). CHK1 substrate peptide (KKKVSRSGLYRSPENLNRPR) derived from CDC25C was purchased from AnaSpec Inc. (San Jose, CA, USA). Immunoprecipitations and immunoblotting were performed as previously described 5.

Generation of the Mre11RK/RK mice

The mutant Mre11 knock-in allele (Mre11RK) was generated using a targeting construct 3lox-2frt flanking a neomycin resistance cassette for selection 5. The primers used to introduce arginine to lysine mutations were: 5′-CTGCCCTTTGGCTCCTTTCTTCCCTTTGCCTTTGCCTTTGCCTTTGCTCGGCACTGCTG-3′ and 5′-AATTCGGCACCTAAAGGAGGCTCTCAGAAAGGCCGAGGTCAGCACC-3′. For the 5′ arm of the construct, a 3.2 kb DNA fragment was amplified using the following oligonucleotides: 5′-TTTTCCGCGGATGGAATCTGAACACACACTGAGTGG-3′ and 5′-TTTTGCGGCCGCATGTTTGTAAAAGACAGTCATGAG-3′ and the fragment was subcloned into SacI and NotI unique sites of the pGK-neo vector. A 3.1 kb fragment for the 3′ arm was amplified by PCR with 5′-TTTTGTCGACTACTATCAGGAGATAAAGTACTTACATG-3′ and 5′-TTTTGGCCGGCCTGAAAGATGTAGTCCTGTCAC-3′ and subcloned into SalI and FseI. To insert the mutated exon 14 of Mre11, gDNA was amplified and engineered to have AscI and XhoI ends. The completed vector was then sequenced to verify the absence of any undesired mutations. In order to clearly identify the mutated allele, an EcoRI site was introduced, also changing serine at position 583 to an asparagine. An FseI site was introduced at the 3′ end of the 3.1 kb fragment and was used to linearize the plasmid for electroporation into embryonic stem (ES) cells. Potential homologous recombinant ES cells emerging from neomycin selection were screened, as determined through a long template PCR system (Roche, 11681842001). Primers for verification included: 5′ arm: ATTGGCACCTATTGTGCAGC and AGGTCGAGGGACCTAATAACTTCG, 3′ arm: GCGTGCAGAATGCCGGGCTTCCGGAG and GATCTGAAAGCTAGTATG, and the fragment between the insert and 5′ arm: AGTGCAGTCAGTGCTCTTTA and CTCTCATTCAGTCATATCAA. Approximately 500 ES colonies were screened and several clones were identified that contained the Mre11 mutant allele. Targeted ES cells were injected into 3.5-day-old blastocytes and were transferred into CD-1 foster mothers, and animals classified as chimeras by coat color were mated with C57BL/6 mice. Germline transmission was achieved and mice were maintained in C57BL/6 background. These mice were then crossed with a transgenic mouse containing FLP1 recombinase gene (Jackson Lab) under the direction of the Gt(ROSA)26Sor promoter to promote recombination and removal of the neomycin cassette. The resulting DNA sequence yielded an 600 bp band in the recombinant allele due to the insertion of 90 bp encompassing the loxP and FRT site of the vector.

Mouse genotyping and RT-PCR of the Mre11 transcript

All mouse procedures were performed in accordance with McGill University guidelines, which are set by the Canadian Council on Animal Care. Genomic DNA was isolated from ear biopsies and analyzed by PCR analysis. The Mre11 allele was identified using the following oligonucleotides: 5′-TGTAGTAGAACTTGGACAGT-3′ and 5′-TGAACCCAGGTCATCTAGAA-3′ yielding a 525 bp band in the wild-type allele and a 591 bp band in the Mre11RK allele. Total cellular RNA was prepared by TRIzol reagent according to the manufacturer's protocol (Invitrogen) to ensure proper splicing of the knock-in transcript. Total RNA (6 μg) was reverse transcribed, and cDNA samples were subjected to PCR analysis. The following 5′ and 3′ primers were used to evaluate the Mre11 transcript: 5′-GCGGTTTCTTAAGGAGCGCCATATT-3′ and 5′-TGTGCCCGACCACCTTTGATCAGCC-3′, yielding a 550 bp band.

Isolation, immortalization and culture of MEFs

Primary MEFs were isolated from E14.5 embryos. Spontaneously immortalized MEFs were created according to the standard 3T3 protocol. All cells were grown in DMEM containing 10% fetal bovine serum.

Ionizing radiation treatment of mice and MEFs

Mice and cells were irradiated at room temperature using a Theratron T-780 Cobalt Unit located in the Department of Radiation Oncology at the Jewish General Hospital (Montreal, Quebec, Canada). Doses ranging from 2 to 10 Gy were delivered at a dose rate of 0.66 Gy/min. The cells were returned to an incubator after the IRs and maintained at 37 °C for further analysis. For in vivo survival experiments, 8- to 10-week-old mice (n = 7) of each genotype were whole-body irradiated at a dose of 10 Gy and then monitored for radiation toxicity.

Chromosome breakage studies: scoring of chromosome aberrations

We have analyzed more than 50 Giemsa-stained metaphases for the passage 7 MEFs without and passage 2 MEFs with 1 Gy IR for each genotype, respectively. The number and type of structural chromosome abnormalities were scored. Chromatid and isochromatid gaps, chromatid and isochromatid breaks, deletions and fragments were scored as a single break. Structural rearrangements including translocations and radial figures were scored as two breakage events. The total number of chromosome aberrations and the mean number of chromosome aberrations per metaphase were scored. Diploid and tetraploid cells were analyzed separately.

Clonogenic assay

Approximately 200-400 cells were plated on 10 cm dishes and treated with varying dosage of IR in duplicate 24 h after plating. The cells were maintained in the regular medium and allowed to grow for 14-20 days. The colonies were fixed and stained with 0.1% crystal violet for 30 min. The stained colonies were counted. The surviving fraction was determined by dividing the average number of colonies for each treatment by the average number of colonies in the control plates.

Class switch recombination

Breeding and genotyping of mice harboring combinations of the Mre11cond, Mre11Δ, Mre11H129N and CD21-Cre alleles were as previously described 23, 30. Analyses of CSR was as previously described 23.

Cytometry analysis

For all flow cytometry experiments, both cells growing on the surface of the dishes and in the culture medium were harvested, fixed with 75% ethanol and stored at −20 °C for less than 1 week before staining and analysis. All flow cytometry measurements were performed using BD FACSCalibur flow cytometer (BD Biosciences). Data were analyzed using BD CellQuest Pro software. For measurement of phosphorylated histone H3, fixed cells were first stained with anti-H3 pS10 rabbit antibody for 1 h and then with FITC-conjugated goat anti-rabbit IgG (Invitrogen) for another hour after washing with dilution buffer (1% FBS and 0.1% Triton X-100 in PBS) as described previously 5. Cells were then counterstained with PI and subjected to flow cytometry analysis.

Immunofluorescence

Cells growing on glass coverslips were washed with PBS twice and fixed with 4% paraformaldehyde at room temperature for 10 min. The cells were then permeabilized (0.5% Triton X-100 in PBS) for 10-15 min. Following three washes with PBS, cells were blocked with 10% serum in PBS and stained with mouse anti-γH2AX (1:1 000), anti-MRE11 (1:200), rabbit anti-53BP1 (1:200), anti-RPA2 (1:100) or anti-RAD51 (1:20) antibodies diluted in PBS containing 5% serum and 0.1% Triton X-100. After three washes, the cells were then stained with Alexa Fluor 488-conjugated goat anti-rabbit or anti-mouse secondary antibodies. DNA was counterstained with 4,6-diamidino-2-phenylindole after three washes with PBS and coverslips were mounted with Immu-Mount purchased from Thermo Scientific. Images were taken using a Zeiss M1 fluorescence microscope.

CHK1 activity assays

MEFs were lysed with lysis buffer (10 mM Tris-HCl, pH 7.4, 150 mM NaCl supplemented with phosphatase inhibitor cocktails and protease inhibitor cocktails) and whole-cell lysates were incubated with anti-CHK1 antibody for 2 h and then protein A-sepharose beads for 1 h. The beads were then washed three times with lysis buffer and then twice with kinase reaction buffer (50 mM Tris-HCl, pH 7.0, 1 mM DTT, 5 mM MgCl2, 0.4 mM MnCl2). The washed beads were resuspended in 20 μl kinase reaction buffer containing 10 μg peptide and 10 μCi γ-32P-ATP and incubated at 30 °C for 20 min with rotation. After spinning for 10 s at 4 °C, the reaction tubes were placed on ice and immediately 10 μl of supernatant was spotted onto thick Fisherbrand Whatmann paper. The papers were then washed five times with 0.5% orthophosphoric acid and rinsed once with ethanol and dried. After drying, the papers were counted with a multi-purpose Scintillation Counter LS 6500, Beckman Coulter (Mississauga, ON, USA). The CHK1 activity was expressed as percentage of the activity in the non-treated samples.

Purification of MRN and MRKRN

MRN and MRKRN were purified as follows. Insect cells (800 ml, 106 cells/ml) were co-infected with MRE11-HIS (or MRE11RK-HIS) and FLAG-RAD50 dual baculovirus and a GST-NBS1 baculovirus for 48 h in a spinner flask. The cells were collected by centrifugation and the pellet was frozen on dry ice. Cells were lysed in PBS300 (1× PBS, 150 mM NaCl, 1 mM EDTA, 1 mM DTT, 0.05% Triton X-100, and protease inhibitors). The cell lysate was centrifuged in 35 000 rpm. for 40 min. Glutathione Sepharose (2 ml, GE Healthcare) was added to the supernatant and incubated for 1.5 h at 4 °C. The beads were washed three times with PBS300, twice with PBS500 (1× PBS, 350 mM NaCl, 1 mM EDTA, 1 mM DTT, 0.05% Triton X-100 and protease inhibitors) and once with P5 buffer (20 mM NaPO4, pH 7.0, 500 mM NaCl, 10% glycerol, 0.02% Triton X-100, 5 mM imidazole). The MRN complex was eluted by cleavage with PreScission protease (60 U/ml, GE Healthcare) for 5 h at 4 °C. The supernatant was added to 400 μl of Talon resin (Clontech) and incubated for 1.5 h at 4 °C. The resin was washed with 10 ml of P30 Talon washing buffer (20 mM NaHPO4, pH 7.4, 1 M NaCl, 10% glycerol, 0.02 % Triton X-100, 30 mM imidazole). MRN complexes were eluted with buffer containing 500 mM imidazole and dialyzed in storage buffer (20 mM Tris-Cl, pH 7.5, 200 mM NaCl, 10% glycerol, 1 mM DTT).

Exonuclease and DNA-binding assays

Exonuclease assays were performed as described previously 36. DNA-binding reactions (10 μl) contained 32P-labelled DNA oligonucleotides (25 nM in nucleotides of each substrate) and MRN or MRKRN, at the indicated concentrations, in binding buffer (25 mM MOPS (morpholinepropanesulfonic acid), pH 7.0, 60 mM KCl, 0.2% Tween, 2 mM DTT, 1 mM Mg(CH3COO)2). Reaction mixtures were incubated at 37 °C for 15 min, followed by 15 min of fixation in 0.2% glutaraldehyde. Reactions were loaded on an 8% TBE acrylamide gel.