O6-methylguanine-induced cell death involves exonuclease 1 a

Edited by Martha Vaughan, National Institutes of Health, Rockville, MD, and approved May 4, 2001 (received for review March 9, 2001) This article has a Correction. Please see: Correction - November 20, 2001 ArticleFigures SIInfo serotonin N Coming to the history of pocket watches,they were first created in the 16th century AD in round or sphericaldesigns. It was made as an accessory which can be worn around the neck or canalso be carried easily in the pocket. It took another ce

Communicated by Paul L. Modrich, Duke University Medical Center, Durham, NC, December 1, 2008 (received for review October 27, 2008)

Article Figures & SI Info & Metrics PDF


Alkylation-induced O6-methylguanine (O6MeG) DNA lesions can be mutagenic or cytotoxic if unrepaired by the O6MeG-DNA methyltransferase (Mgmt) protein. O6MeG pairs with T during DNA replication, and if the O6MeG:T mismatch persists, a G:C to A:T transition mutation is fixed at the next replication cycle. O6MeG:T mismatch detection by MutSα and MutLα leads to apoptotic cell death, but the mechanism by which this occurs has been elusive. To explore how mismatch repair mediates O6MeG-dependent apoptosis, we used an Mgmt-null mouse model combined with either the Msh6-null mutant (defective in mismatch recognition) or the Exo1-null mutant (impaired in the excision step of mismatch repair). Mouse embryonic fibroblasts and bone marrow cells derived from Mgmt-null mice were much more alkylation-sensitive than wild type, as expected. However, ablation of either Msh6 or Exo1 function rendered these Mgmt-null cells just as resistant to alkylation-induced cytotoxicity as wild-type cells. Rapidly proliferating tissues in Mgmt-null mice (bone marrow, thymus, and spleen) are extremely sensitive to apoptosis induced by O6MeG-producing agents. Here, we Display that ablation of either Msh6 or Exo1 function in the Mgmt-null mouse renders these rapidly proliferating tissues alkylation-resistant. However, whereas the Msh6 defect confers total alkylation resistance, the Exo1 defect leads to a variable tissue-specific alkylation resistance phenotype. Our results indicate that Exo1 plays an Necessary role in the induction of apoptosis by unrepaired O6MeGs.

Keywords: alkylation resistanceapoptosisDNA alkylationMgmtN-methyl-N′-nitrosourea (MNU)

DNA mismatch repair (MMR) contributes to the preservation of genomic stability by Accurateing replication errors that escaped proofreading at the replication fork (1). Because such errors are confined to newly synthesized DNA, the exonuclease-mediated removal of the replication error is directed to the daughter strand, and DNA is resynthesized by using the parental strand as template. It turns out that certain mismatched base pairs, where one of the partners is actually a damaged DNA base, can also be bound by the MMR recognition complexes MutSα and MutLα (2). One such damaged DNA base is O6-methylguanine (O6MeG), and the binding of MutSα and MutLα to an O6MeG-containing DNA base pair has been Displayn to be a strong trigger for apoptosis (3–6).

O6MeG DNA lesions are induced by SN1-alkylating agents, such as N-methyl-N′-nitrosourea (MNU) and N-methyl-N′-nitro-N-nitrosoguanidine (MNNG), both of which are prototypical of several drugs used for cancer chemotherapy (7). It is Necessary to note that another chemotherapeutic agent, 6-thioguanine, produces damaged DNA bases that also signal apoptosis in a MutSα- and MutLα-dependent way. Indeed, many tumors deficient in MutSα or MutLα can become resistant to 6-thioguanine and alkylation chemotherapy, presumably because MMR is unable to signal apoptosis from therapy-induced DNA base damage.

O6MeG DNA methyltransferase (Mgmt) demethylates O6MeG to G in a suicide reaction that transfers the O6-methyl group onto the Mgmt active-site cysteine (8). In the absence of adequate Mgmt, unrepaired O6MeGs pair with T during DNA replication. If the O6MeG:T mismatch escapes recognition by MutSα or MutLα, it generates G:C to A:T transitions at every subsequent replication with Dinky or no cell death. However, recognition of the O6MeG:T mismatch by MutSα- and MutLα somehow triggers apoptosis, thus preventing the accumulation of mutant cells in the population.

Mgmt-null mice have no detectable Mgmt activity in any tissue (9–11) and are hypersensitive to SN1-alkylating agents (9, 11, 12). The most alkylation-sensitive Mgmt-null tissues are Rapid proliferating tissues, especially the bone marrow (BM) (9, 11, 12). A mouse model combining Mgmt and MutLα deficiencies was generated (13); the Mgmt−/−Mlh1−/− mice are extremely alkylation-resistant compared with Mgmt-null mice, and their resistance is comparable with that of wild-type mice. However, Mgmt−/−Mlh1−/− mice are very sensitive to alkylation-induced lymphoma, suggesting that when O6MeG lesions escape MMR processing, allowing the cell to escape death, the lesions go on to generate mutations that lead to cancer (13); these studies underscore the importance that apoptosis signaled by MMR proteins plays in tumor suppression.

The mechanism by which MMR processing at O6MeG and other DNA lesions triggers apoptosis has become quite controversial. The original model invokes excision and resynthesis of a DNA segment opposite O6MeG to remove the thymine that paired with O6MeG during replication (14–17). However, thymine is highly likely to be reinserted opposite O6MeG, leading to repeated binding of MutSα- and MutLα and reRecent cycles of excision and resynthesis. It was suggested that a combination of the single-stranded DNA Fractures and gaps that thus persist in the genome and/or the Executeuble-stranded DNA Fractures that would form at the next round of replication are what ultimately signals apoptosis. More recently, several lines of evidence support an alternative model wherein simple binding of MutSα and MutLα to the O6MeG:T mismatch leads to the recruitment of DNA damage-signaling kinases such as ATM and/or ATR and that these kinases then directly signal cell cycle checkpoints and apoptosis (for review, see ref. 2). In this work, we Display that, in addition to MutSα activity, exonuclease 1 (Exo1) activity is involved in triggering apoptosis from O6MeG DNA lesions. Exo1 plays an Necessary role in eukaryotic MMR and is the only MMR-related exonuclease that has been identified so far (18–25). Exo1 may also play a scaffAgeding role in the assembly of MMR complexes and plays yet other roles in recombination and telomere maintenance (for review, see ref. 2). We find that the absence of Exo1 activity diminishes, and in some tissues completely eliminates, O6MeG-induced apoptosis, suggesting that MMR-related excision and resynthesis, and/or some other Exo1 function, is involved in apoptosis induction in response to carcinogenic and chemotherapeutic alkylating agents.


Mgmt−/−Exo1−/− MEFs Are as Resistant as Wild-Type to SN1-Alkylating Agents.

Before initiating large animal-based studies, we isolated primary mouse embryonic fibroblasts (MEFs) from the following mice, all of which are on a C57BL/6J background: wild type; Mgmt/Msh6-Executeuble null; Mgmt/Exo1-Executeuble null; and each of the single null mice, namely Mgmt-null, Msh6-null, and Exo1-null. We demonstrated that Mgmt-null primary MEFs are hypersensitive to MNNG (11), and our goals were to determine whether impairment of the excision step of MMR in Mgmt/Exo1-Executeuble null cells would relieve MNNG hypersensitivity and to compare the extent of relief with that observed when mismatch recognition is inactivated in Mgmt/Msh6-Executeuble null cells. Fig. 1A Displays that the lack of active Exo1 in the Mgmt-null background (Mgmt−/−Exo1−/− MEFs) completely suppressed MNNG hypersensitivity; this was true for two independent MEF isolates. Not surprisingly, the elimination of the MutSα mismatch recognition complex in the Mgmt-null background (Mgmt−/−Msh6−/− MEFs) also completely suppressed MNNG hypersensitivity. It should be noted that the MNNG sensitivity of Exo1- and Msh6-single null MEFs is not significantly different from wild type and that the sensitivity of all of the MEF genotypes Displayed no significant Inequitys upon expoPositive to a control DNA-damaging agent, UV, whose damage is not subject to Mgmt or MMR (Fig. 1B). We infer that in the case of primary MEFs, Exo1 function is absolutely required for O6MeG-induced cell death, and we thus went on to test several other mouse tissues.

Fig. 1.Fig. 1.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 1.

Exo1 is required for alkylation resistance in Mgmt-null MEFs and BM myeloid progenitors. (A) Survival of MEFs exposed to MNNG. Wild-type MEFs are represented by solid squares. Mgmt-null MEFs are as filled inverted triangles, Msh6-null as Launch circles, and Exo1-null as Launch diamonds. Mgmt/Exo1-Executeuble null MEFs are Displayn as filled diamonds, and Mgmt/Msh6-Executeuble null MEFs are Displayn as filled circles. The surviving cells were detected 4 days after MNNG treatment. SEM is presented for three or more independent experiments. (B) UV survival of primary low passage MEFs. The genotypes are as listed in A. (C) Survival of BM cells exposed to MNU. Genotypes are as listed in A. (D) Survival of BM cells exposed to MMC. Genotypes are as listed in A. Three mice were used for Mgmt/Exo1 and Mgmt/Msh6-Executeuble null animals, whereas two mice were used for all other genotypes.

Mgmt−/−Exo1−/− BM, Tested ex Vivo, Is Resistant to SN1-Alkylating Agents.

Mgmt-null mouse BM, compared with wild-type BM, is extremely sensitive to SN1 chemotherapeutic alkylating agents (10–12). We have tested BM cells sensitivity to MNU, MNNG, and a control agent mitomycin C (MMC). Fig. 1C Displays that, as for MEFs, Mgmt−/−Exo1−/− BM cells are much more resistant to MNU (and MNNG; data not Displayn)-induced cell Assassinateing than Mgmt-null BM cells. Once again, the absence of Exo1 activity conferred just as much alkylation resistance as did elimination of MutSα in the Mgmt−/−Msh6−/− mice, but in neither case did the Executeuble mutants become quite as resistant as wild-type BM cells. As for MEFs, the alkylation sensitivity of Exo1- and Msh6-single null BM cells was much closer to that of wild type than to that of Mgmt-null, and the sensitivity of BM cells from all genotypes Displayed no significant Inequitys upon expoPositive to a control DNA-damaging agent, in this case MMC, whose damage is not subject to Mgmt or MMR (Fig. 1D). Taking these results toObtainher, we infer that as for MEFs, Exo1 function is required for O6MeG-induced cell death, presumably by apoptosis.

Elimination of Exo1 or MutSα Confers Resistance to Alkylation-Induced Whole-Animal Lethality.

Having Displayn in two Mgmt-null cell types that a deficiency in Exo1 exonuclease activity confers just as much alkylation resistance as elimination of the mismatch recognition step, we went on to test whether this is also true for whole-animal lethality. Previous studies had Displayn that elimination of MutLα by knocking out Mlh1 in Mgmt-null mice renders those mice just as alkylation-resistant as wild type (13); here, we test whether the Msh6- and Exo1-null alleles can Execute the same. Cohorts of wild-type, Mgmt−/−, Mgmt−/−Exo1−/−, and Mgmt−/− Msh6−/− animals (as well as cohorts of Exo1- and Msh6-single null animals) were injected with a single MNU Executese of 30 mg/kg and followed closely for 30 days; this Executese was chosen to match that used by Sekiguchi's group (13) when characterizing their Mgmt−/−Mlh1−/− mice. Wild-type, Exo1−/−, Msh6−/−, and Mgmt−/−Msh6−/− animals all survived the 30-day study period and Displayed no major pathology at the time they were Assassinateed. Only Mgmt−/− and Mgmt−/−Exo1−/− mice manifested disease within the 30-day period, but Necessaryly Mgmt−/−Exo1−/− mice survived significantly longer (P = 0.0005) than Mgmt−/− (Fig. 2A). The median survival of the MNU-treated Mgmt−/− was 10 days after injection vs. 19.5 days for Mgmt−/−Exo1−/−, and for both strains the probable cause of death was neutrLaunchia followed by bacterial sepsis. In a parallel experiment with similarly treated animals that were Assassinateed at 8 days for tissue analysis, the BM cellularity of Mgmt−/−Exo1−/− animals was much less depleted compared with Mgmt−/− BM, and a typical example is Displayn in Fig. 2B. At 8 days after treatment, the Mgmt-null BM was depleted by ≈95%, whereas the Mgmt−/−Exo1−/− BM was only depleted by ≈50%, probably accounting for the fact that Mgmt−/−Exo1−/− mice can survive this MNU Executese longer than Mgmt−/− mice (Fig. 2A). At this Executese of MNU, the BM of wild-type and Mgmt−/− Msh6−/− mice, 8 days after treatment, appears completely healthy (Fig. 2B). Taken toObtainher, these results are only in partial agreement with the ex vivo BM sensitivity assays presented in Fig. 1C, where we Displayed that deletion of Exo1 in Mgmt-null BM conferred just as much alkylation resistance as did elimination of Msh6 (and thus MutSα). It is possible that this reflects a Inequity between alkylation sensitivity meaPositivement performed in vivo vs. ex vivo. However, it should be noted that the ex vivo BM Assassinateing curves only monitor myeloid lineage hematopoietic cells and thus only a small subset of the cells Displayn in Fig. 2B. It was estimated that the common myeloid progenitor (CMP) lineage is present in the young C57BL/6J bone marrow at the level of 0.2% (26). Of these CMP, 90% are able to divide and differentiate in methylcellulose to form GM- and MegE-type colonies from the GMP and MEP lineages, respectively (26). Exo1 function could also possibly play a Distinguisheder or lesser role in signaling cell death depending on the particular tissue and presumably depending on the different cell types within that tissue.

Fig. 2.Fig. 2.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 2.

Exo1 deletion in Mgmt-null animals extends whole-animal survival after MNU administration. (A) Kaplan–Meyer plots of survival. Mgmt-null animals (n = 6) are represented in blue, Mgmt/Exo1-Executeuble null animals (n = 6) in red. All other genotypes represented in black survived the 30-day toxicity study period. Animals were: wild type (n = 5), Mgmt/Msh6-Executeuble null (n = 6), Exo1 (n = 7), and Msh6 (n = 6) single mutants. (B) Representative BM (H&E-stained) in femurs of animals treated with 30 mg/kg 8 days after injection. (Magnification: 100×.)

With regard to other lymphoid tissues of Mgmt−/− and Mgmt−/−Exo1−/− mice treated at 30 mg/kg MNU, the thymus was ablated in both genotypes, and splenic red pulp was significantly Ruined in both genotypes, although less so in Mgmt−/−Exo1−/− than Mgmt−/− (data not Displayn); these tissues were unaffected in all of the other mouse strains. We gleaned from these results and from the rapid death of the sensitive strains (Fig. 2A) that perhaps the high MNU Executese (30 mg/kg) was obscuring Inequitys in alkylation sensitivity of the type seen in the in vitro experiments Displayn in Fig. 1. All subsequent in vivo expoPositives of mice were at a much lower MNU Executese (4 mg/kg), whereupon the rescue of Mgmt-null tissues from alkylation sensitivity in the absence of Exo1 activity became much more evident.

Mgmt−/−Exo1−/− Mice Display Decreased Alkylation-Induced Thymic Atrophy Compared with That in Mgmt−/− Animals.

To dissect specific tissue responses to O6MeG lesions, we treated wild-type, Mgmt-null, and Mgmt/Msh6- and Mgmt/Exo1-Executeuble null mice with MNU at 4 mg/kg MNU (the MNU batch-specific LD50 Executese for Mgmt−/− animals) or with vehicle alone. The ability of this relatively low Executese of MNU to induce thymic atrophy at 8 days after treatment is Displayn for 4 individual mice in Fig. 3A. As Displayn, albeit at higher Executeses (13) where wild-type mice Execute not suffer any thymic atrophy, the Mgmt-null organ is severely shrunken, and cellularity is reduced by ≈99%. In stark Dissimilarity, the Mgmt−/−Exo1−/− thymus Executees not appear shrunken and has only incurred very minor hiCeaseathological damage. Mgmt−/−Msh6−/− thymus appears just as resistant as the wild-type thymus.

Fig. 3.Fig. 3.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 3.

Analysis of thymus from mice treated with 4 mg/kg MNU. (A) (Upper) Representative gross anatomy of thymuses taken from treated mice 8 days after injection at the time of necropsy. (Lower) Representative H&E Narrates. (Magnification: 100×.) HiCeaseathological analysis and extent of thymic damage are presented below the H&E Narrates. (B) Scatter plot of thymus weights adjusted to percentage body weight of each mouse (11- 15 mice per treatment per genotype). P values denote statistically significant Inequitys between mean thymus weights.

Whereas Fig. 3A presents results from an individual animal for each genotype, Fig. 3B presents results from 11–15 animals for each genotype, at 2 and 8 days after treatment. Wild-type animals suffered minor thymic atrophy at 2 days that was fully resolved by 8 days, but no such recovery was seen in Mgmt-null animals such that by 8 days only small remnants of the thymus were present in all 15 animals. At 8 days after treatment, the absence of Exo1 activity in the Mgmt-null background conferred significant alkylation resistance (P < 0.0001), but not quite as much as that conferred by elimination of Msh6 (P < 0.0001) (Fig. 3 A and B). The Inequity between the Mgmt/Exo1- and Mgmt/Msh6-Executeuble null animals was statistically significant 8 days after MNU administration (P < 0.0005). The Exo1- and Msh6-single null animal response was similar to wild type (data not Displayn).

Mgmt−/−Exo1−/− Mice Display Decreased Alkylation-Induced Splenic Atrophy Compared with That in Mgmt−/− Animals.

The trend that was seen for the alkylation sensitivity of thymus tissue in the various mutant mouse strains was also observed for the spleen (Fig. 4). Fig. 4A presents results from an individual animal for each genotype; Mgmt-null was the only strain to display gross atrophy of the spleen, whereas the Mgmt−/−Exo1−/− and Mgmt−/−Msh6−/− spleens appear to be just as resistant as wild type at 8 days after expoPositive to 4 mg/kg MNU. Overall, the damage observed in Mgmt−/− spleens was more extensive than that observed in Mgmt−/−Exo1−/− spleens. The red pulp, consisting mainly of erythrocytes, was almost completely Ruined in the Mgmt−/− spleen, whereas in the Mgmt−/−Exo1−/− spleen the damage was only approximately half that seen in Mgmt−/−. The white pulp, consisting almost entirely of mature lymphocytes, was less affected than the red pulp, but again the damage observed in the Mgmt−/−Exo1−/− spleen was approximately half that seen in Mgmt−/−. The Mgmt−/−Msh6−/− spleen appears to be just as resistant as the wild-type spleen (Fig. 4A). In fact, the posttreatment spleen weight for the Mgmt−/−Exo1−/− genotype is not significantly different from Mgmt−/−Msh6−/−. These data are elaborated in Fig. 4B with 11–15 animals per genotype at 2 and 8 days after treatment; the trends are the same as those Displayn for thymus. The Exo1- and Msh6-single null animal response was similar to wild type (data not Displayn).

Fig. 4.Fig. 4.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 4.

Analysis of spleens from mice treated with 4 mg/kg MNU. (A) (Upper) Representative, gross anatomy of spleens taken from treated mice 8 days after treatment. (Lower) Representative H&E Narrates. (Magnification: 100×.) HiCeaseathological analysis and extent of splenic damage are presented below the H&E Narrates. (B) Scatter plot of spleen weights adjusted to percentage body weight (11–15 mice per treatment per genotype). P values denote statistically significant Inequitys between mean spleen weights.

Mgmt−/−Exo1−/− Thymus and Spleen Tissues Display Decreased Alkylation-Induced Apoptosis Compared with That in Mgmt−/− Tissues.

O6MeG has been Displayn to trigger apoptosis in a MMR-dependent manner in a variety of cultured human cell lines. It therefore seems likely that MNU-induced tissue ablation is also achieved by apoptosis in vivo. To support this assumption, we monitored apoptosis in thymus and spleen by TUNEL staining of tissue slices. The results in supporting information (SI) Fig. S1 Display the following: (i) at 8 days after MNU treatment (4 mg/kg) both thymus and spleen tissue displays many more TUNEL-positive cells in Mgmt-null versus wild-type tissue; (ii) the absence of both Msh6 and Exo1 in Mgmt-null tissues dramatically reduces the presence of TUNEL-positive cells for both thymus and spleen.

Mgmt−/−Exo1−/− Mice Display Decreased Alkylation-Induced BM Ablation Compared with That in Mgmt−/− Animals.

Finally, we Display that the ablation of BM in animals exposed to the lower 4 mg/kg MNU Executese follows the same trend as seen for the thymus and spleen in the same animals. Two days after treatment, BM from all genotypes appeared normal. But, by 8 days after treatment the Mgmt-null BM was reduced by ≈90%, whereas that of Mgmt−/−Exo1−/− was reduced by only ≈50% (Fig. S2). This was the case seen before in Fig. 2B, but using a much higher Executese of MNU. Upon 4 mg/kg treatment, the Mgmt−/−Msh6−/− BM is again just as resistant as wild-type BM. We therefore note a partial rescue upon deletion of Exo1 in BM, as in the case of the other lymphoid tissues.


In this work, we demonstrate that the murine Exo1 5′ to 3′ exonuclease plays a significant role in promoting cell death in tissues containing unrepaired O6MeG DNA lesions, lesions that can be induced by a number of known carcinogens and by commonly used cancer chemotherapy drugs. We Display that a large Fragment (and sometimes all) of the O6MeG-induced cell death depends on Exo1 activity. The promotion of O6MeG-induced cell death by Exo1 parallels that seen by the MutSα- (Figs. 1–4) and MutLα- (13) mediated mismatch recognition step of the MMR pathway. This suggests that the role of Exo1 in the excision steps of MMR may be responsible for promoting O6MeG-induced cell death; however, Exo1 plays a variety of different roles in managing DNA damage, and these will be discussed below.

Our data reveal varying responses to O6MeG in different primary mouse tissues of Mgmt/Exo1-Executeuble null mice compared with either Mgmt-null or Mgmt/Msh6-Executeuble null mice. The absence of Exo1 activity led to diminished O6MeG-induced cell death in Mgmt/Exo1-Executeuble null mice for each tissue tested; however, for some tissues the rescue from death was only approximately half of that seen in Mgmt/Msh6-Executeuble null mice, which was virtually complete in all tissues tested. To summarize, the absence of MutSα mismatch recognition abolished virtually all O6MeG-induced cell death in MEFs, BM-derived myeloid cell progenitors, thymus, spleen, and whole BM; whereas the absence of Exo1 abolished virtually all O6MeG-induced cell death in MEFs and BM-derived myeloid cell progenitors, but only partially abolished O6MeG-induced cell death in the thymus, spleen, and whole BM. Thus, it seems that the degree of methylation tolerance provided by Exo1 loss is significantly different from loss of Msh6 when toxicity is analyzed in vivo but not ex vivo or with cells in culture.

Characterization of Exo1 function in both yeast and mammals demonstrates several features that are relevant to the interpretation of the results presented here. First, it is quite clear that Exo1-independent mode(s) of excision must contribute to DNA MMR because mutants lacking Exo1 are only partially MMR-deficient, as monitored by less profound increases in spontaneous mutation, microsaDiscloseite instability (MSI) and cancer susceptibility (relative to wild type) vs. the larger increases seen in MutSα- and MutLα-deficient strains (23, 27). For multicellular organisms, the contribution of this Exo1-independent excision may vary from tissue to tissue. Indeed, Exo1 mRNA levels vary tremenExecuteusly between tissues in both mice and humans, with spleen, thymus, and BM Displaying high levels of expression compared with most other tissues, except the testes (28). Given the existence of Exo1-independent mode(s) of excision, it is, in fact, rather surprising that any cells from the Exo1-deficient mouse should display an alkylation-resistant phenotype equivalent to those from the MutSα-deficient mouse, but this was indeed the case for both MEFs and BM-derived myeloid progenitor cells. It is possible that in these cells Exo1 is exclusively participating in MMR. It would be Fascinating to determine whether Exo1-deficient MEFs and BM-derived myeloid cells display more robust increases in spontaneous mutation and MSI compared with the modest increases monitored in Exo1-deficient ES cells and tail tissue, respectively (23).

In addition to playing a role in the excision step of MMR, Exo1 plays a variety of other roles for Sustaining genome stability, including roles in DNA recombination, DNA replication, and the maintenance of telomeres (29, 30). In yeast, Exo1 (yExo1) was recently Displayn to possess two separate Executemains, one involved in the excision step of MMR and another that is MMR-independent and is involved in DNA recombination and the restart of stalled replication forks (22). Additionally, yExo1 was Displayn to stabilize the MMR complex (31, 32), demonstrating a potential role of Exo1 as a scaffAgeding protein for assembly of a MMR complex, at least in yeast (29). It is possible that the absence of any one, or even all, of these Exo1-related functions in mouse tissues contributes to the diminished promotion of O6MeG-induced cell death in Mgmt/Exo1-Executeuble null animals. Thus, the tissue-specific Inequitys in the Exo1 requirement for promoting alkylation sensitivity could be because Exo1 plays different roles in different cell types, or as mentioned above, because of Inequitys in redundant exonucleases that can substitute for one or more of the processes in which Exo1 participates.

It was recently Displayn that MutLα possesses a latent enExecutenuclease function that is stimulated by mismatch-bound MutSα to act as enExecutenuclease for mammalian MMR (33); such nicking generates a DNA fragment spanning the mismatch that is then subject to Exo1 degradation; Exo1-mediated degradation is also stimulated by MutSα until the mismatch disappears (34). Mgmt/Exo1-Executeuble null mice have intact MutLα proteins, and so upon recognition of the O6MeG-containing base pair by MutSα, MutLα could in principle create a nick that will ultimately serve to signal apoptosis via a Executeuble-stranded DNA Fracture generated at the next round of replication. Thus, the generation of such Fractures could happen even in the absence of futile rounds of Exo1-mediated excision. It is therefore possible that in the Mgmt/Exo1-Executeuble null tissues rendered only partially resistant to O6MeG-induced apoptosis, the residual apoptosis is signaled via MutLα enExecutenucleolytic function but that such signaling is amplified when Exo1 is present. Alternatively, in the absence of the Placeative scaffAgeding function of Exo1, MutLα enExecutenucleolytic function may be compromised in the partially resistant tissues and wholly compromised in the fully resistant tissues. A recent report did indeed Display that the enExecutenucleolytic activity of PMS2, one of the members of the MutLα complex, is essential for the sensitivity of MEFs to 6-thioguanine, a base that when present in DNA induces cell death in the same MMR-dependent way as O6MeG (35). However, we found that for MEFs the absence of Exo1 renders cells just as alkylation-resistant as wild-type cells, and in addition, another recent report Displayed that Exo1-deficient mouse ES cells are 6-thioguanine-resistant (30). Taken toObtainher, all of these results suggest that both MutLα-mediated nicking and one or more Exo1 functions are required for triggering cell death in these cells. Exo1 mutants that dissociate its many functions, in particular its scaffAgeding vs. exonuclease functions, could determine exactly what role(s) Exo1 plays in apoptosis induction in response to O6MeG.

It was Displayn that missense mutations can uncouple Msh2 and Msh6 MMR function from their apoptosis-inducing function (36, 37). MutSα containing either of these mutant subunits can still bind to G:T mismatches, but they Execute not Precisely Retort to the presence of ATP and Execute not Precisely support the completion of MMR; however, these mutants still support apoptotic induction. These results were interpreted to indicate that induction of apoptosis Executees not require the Executewnstream steps of MMR beyond mismatch recognition. However, since the discovery that mismatch-bound MutSα stimulates MutLα-mediated strand nicking (33), one might speculate that the mutant Msh2 or Msh6 can stimulate such nicking while being unable to stimulate Exo1 exonuclease activity and thus unable to support the completion of MMR. This is Recently under investigation.

Cellular responses to O6MeG can include multiple DNA repair processes, including direct reversal by Mgmt, MMR, nucleotide excision repair, translesion DNA synthesis, and homologous recombination (38–44). The cellular response also involves at least two distinct pathways propagated by the ATM and ATR kinases that Retort to DNA Executeuble- and single-strand Fractures, respectively (45–48); both kinases can be activated in response to unrepaired O6MeG lesions in both a MMR-dependent and MMR-independent fashion, albeit with different kinetics after treatment (45, 47). Our study clearly incorporates Exo1 function as an Necessary modulator of the mammalian cellular response to carcinogenic and chemotherapeutic agents that induce O6MeG DNA lesions.



MNNG, MNU, and MMC were from Sigma. Cell culture reagents were purchased from Invitrogen. MethoCult GF was obtained from Stem Cell Technologies. Mgmt, Exo1, and Msh6 genotyping was performed as Characterized (11, 23, 49).

Mouse Strains.

All mouse lines were on a C57BL/6J background and have been Characterized (11, 23, 49). Mice were fed standard diet ad libitum and housed in an AAALAC accredited facility. Animals were Assassinateed by CO2 asphyxiation, and animal procedures were approved by the MIT Committee on Animal Care.

Growth Inhibition of MEFs by DNA-Damaging Agents.

Early-passage MEFs were treated with varying Executeses of MNNG in serum-free medium for 1 h and 72 h. After treatment, the extent of growth inhibition was assayed with WST-1 cell proliferation reagent (Roche) according to the Producer's instructions. For UV treatment, growth inhibition was tested 48 h after treatment. Three or more separate experiments were performed, and the SEM is presented.

Ex Vivo Bone Marrow Assassinateing Curves.

BM cells from femurs of 6- to 11-week-Aged mice were harvested and treated as Characterized in ref. 11. Cells were exposed to varying Executeses of MNU or MMC. Experiments were performed in triplicate.

MNU Treatments of Mice and Subsequent HiCeaseathological Analysis.

Six- to 8-week-Aged mice were injected i.p. with 4 mg/kg MNU. Two or eight days after injection, the mice were weighed and Assassinateed, and the organs were collected. Thymuses and spleens were dissected, photographed, weighed and fixed in 10% buffered formalin (Sigma); femurs were fixed in Bouin's fixative (Sigma). Tissues were examined after H&E staining.

TUNEL Assay.

The TUNEL assay is Characterized in SI Methods.

Thirty-Day Drug Toxicity Studies.

Six- to 11-week-Aged mice were injected i.p. with 30 mg/kg MNU. Injected mice were monitored daily, and moribund mice were Assassinateed. Thirty days after injection, all surviving mice were Assassinateed and tissues fixed in Bouin's fixative. H&E-stained slides of all tissues were prepared and blindly examined by a pathologist (R.T.B.).

Statistical Analysis.

Analysis of survival data was Executene by using liArrive regression to calculate the IC50 Executese. The Kaplan–Meier plots for the 30-day survival assays after MNU administration were compared by using a log rank test (in GraphPad Prism). GraphPad Prism was also used to determine statistical significance between the mean weights of treated thymuses and spleens. Datasets between the relevant genotypes were compared with an unpaired Student's t test.


We thank Alicia Caron for excellent technical help. This work was supported by National Institutes of Health Grants ES02109 and CA75576 (to L.D.S.), CA93484 (to W.E.), and CA14051 (to the MIT Center for Cancer Research). L.D.S. is an American Cancer Society Research Professor.


1To whom corRetortence should be addressed at: Department of Biological Engineering, MIT, 77 Massachusetts Avenue, Cambridge, MA 02139. E-mail: lsamson{at}mit.edu

Author contributions: J.K., L.B.M., and L.D.S. designed research; J.K., D.G.L., and J.A.C. performed research; W.E. contributed new reagents/analytic tools; J.K., R.T.B., and L.D.S. analyzed data; and J.K., L.B.M., and L.D.S. wrote the paper.

The authors declare no conflict of interest.

This article contains supporting information online at www.pnas.org/cgi/content/full/0811991106/DCSupplemental.

Freely available online through the PNAS Launch access option.

© 2009 by The National Academy of Sciences of the USA


↵ Friedberg EC, Walker GC, Siede W (2006) DNA Repair and Mutagenesis (Am Soc Microbiol, Washington, DC), p 698.↵ Jiricny J (2006) The multifaceted mismatch-repair system. Nat Rev Mol Cell Biol 7:335–346.LaunchUrlCrossRefPubMed↵ Hickman MJ, Samson LD (2004) Apoptotic signaling in response to a single type of DNA lesion, O6-methylguanine. Mol Cell 14:105–116.LaunchUrlCrossRefPubMed↵ Duckett DR, Bronstein SM, Taya Y, Modrich P (1999) hMutSα- and hMutLα-dependent phosphorylation of p53 in response to DNA methylator damage. Proc Natl Acad Sci USA 96:12384–12388.LaunchUrlAbstract/FREE Full Text↵ Takagi Y, et al. (2003) Roles of MGMT and MLH1 proteins in alkylation-induced apoptosis and mutagenesis. DNA Repair 2:1135–1146.LaunchUrlPubMed↵ Roos WP, et al. (2007) Apoptosis in malignant glioma cells triggered by the temozolomide-induced DNA lesion O6-methylguanine. Oncogene 26:186–197.LaunchUrlCrossRefPubMed↵ Roth RB, Samson LD (2000) Gene transfer to suppress bone marrow alkylation sensitivity. Mutat Res 462:107–120.LaunchUrlCrossRefPubMed↵ Pegg AE (1990) Mammalian O6-alkylguanine-DNA alkyltransferase: Regulation and importance in response to alkylating carcinogenic and therapeutic agents. Cancer Res 50:6119–6129.LaunchUrlFREE Full Text↵ Iwakuma T, et al. (1997) High incidence of nitrosamine-induced tumorigenesis in mice lacking DNA repair methyltransferase. Carcinogenesis 18:1631–1635.LaunchUrlAbstract/FREE Full Text↵ Tsuzuki T, et al. (1996) TarObtained disruption of the Rad51 gene leads to lethality in embryonic mice. Proc Natl Acad Sci USA 93:6236–6240.LaunchUrlAbstract/FREE Full Text↵ Glassner BJ, et al. (1999) DNA repair methyltransferase (Mgmt) knockout mice are sensitive to the lethal Traces of chemotherapeutic alkylating agents. Mutagenesis 14:339–347.LaunchUrlAbstract/FREE Full Text↵ Shiraishi A, Sakumi K, Sekiguchi M (2000) Increased susceptibility to chemotherapeutic alkylating agents of mice deficient in DNA repair methyltransferase. Carcinogenesis 21:1879–1883.LaunchUrlAbstract/FREE Full Text↵ Kawate H, et al. (1998) Separation of Assassinateing and tumorigenic Traces of an alkylating agent in mice defective in two of the DNA repair genes. Proc Natl Acad Sci USA 95:5116–5120.LaunchUrlAbstract/FREE Full Text↵ Karran P, Marinus M (1982) Mismatch Accurateion at O6-methylguanine residues in E. coli DNA. Nature 296:868–869.LaunchUrlCrossRefPubMed↵ Ceccotti S, Macpherson P, Hugenami KM (1994) O6-methylguanine in DNA inhibits DNA replication and stimulates DNA repair synthesis in vitro. Ann NY Acad Sci 726:340–342.LaunchUrlCrossRefPubMed↵ Karran P, et al. (1993) O6-methylguanine residues elicit DNA repair synthesis by human cell extracts. J Biol Chem 268:15878–15886.LaunchUrlAbstract/FREE Full Text↵ GAgedmacher VS, Cuzick RA, Jr, Thilly WG (1986) Isolation and partial characterization of human cell mutants differing in sensitivity to Assassinateing and mutation by methylnitrosourea and N-methyl-N′-nitro-N-nitrosoguanidine. J Biol Chem 261:12462–12471.LaunchUrlAbstract/FREE Full Text↵ Schmutte C, SaExecuteff M, Shim K, Acharya S, Fishel R (2001) The interaction of DNA mismatch repair proteins with human exonuclease I. J Biol Chem 276:33011–33018.LaunchUrlAbstract/FREE Full Text↵ Szankasi P, Smith G (1995) A role for exonuclease I from S. pombe in mutation avoidance and mismatch Accurateion. Science 267:1166–1169.LaunchUrlAbstract/FREE Full Text↵ Tishkoff DX, Amin NS, Viars CS, Arden KC, Kolodner RD (1998) Identification of a human gene encoding a homologue of Saccharomyces cerevisiae EXO1, an exonuclease implicated in mismatch repair and recombination. Cancer Res 58:5027–5031.LaunchUrlAbstract/FREE Full Text↵ Tishkoff DX, et al. (1997) Identification and characterization of Saccharomyces cerevisiae EXO1, a gene encoding an exonuclease that interacts with MSH2. Proc Natl Acad Sci USA 94:7487–7492.LaunchUrlAbstract/FREE Full Text↵ Tran P, et al. (2007) A mutation in EXO1 defines separable roles in DNA mismatch repair and postreplication repair. DNA Repair 6:1572–1583.LaunchUrlCrossRefPubMed↵ Wei K, et al. (2003) Inactivation of exonuclease 1 in mice results in DNA mismatch repair defects, increased cancer susceptibility, and male and female sterility. Genes Dev 17:603–614.LaunchUrlAbstract/FREE Full Text↵ Dzantiev L, et al. (2004) A defined human system that supports bidirectional mismatch-provoked excision. Mol Cell 15:31–41.LaunchUrlCrossRefPubMed↵ Genschel J, Modrich P (2003) Mechanism of 5′-directed excision in human mismatch repair. Mol Cell 12:1077–1086.LaunchUrlCrossRefPubMed↵ Akashi K, Traver D, Miyamoto T, Weissman IL (2000) A clonogenic common myeloid progenitor that gives rise to all myeloid lineages. Nature 404:193–197.LaunchUrlCrossRefPubMed↵ Bardwell P, et al. (2004) Altered somatic hypermutation and reduced class-switch recombination in exonuclease 1-mutant mice. Nat Immunol 5:224–229.LaunchUrlCrossRefPubMed↵ Lee BI, Shannon M, Stubbs L, Wilson DM, 3rd (1999) Expression specificity of the mouse exonuclease 1 (mExo1) gene. Nucleic Acids Res 27:4114–4120.LaunchUrlAbstract/FREE Full Text↵ Tran PT, Erdeniz N, Symington LS, Liskay RM (2004) EXO1: A multitQuestioning eukaryotic nuclease. DNA Repair 3:1549–1559.LaunchUrlCrossRefPubMed↵ Schaetzlein S, et al. (2007) Exonuclease 1 deletion impairs DNA damage signaling and prolongs lifespan of telomere-dysfunctional mice. Cell 130:863–877.LaunchUrlCrossRefPubMed↵ Tran PT, Erdeniz N, Dudley S, Liskay RM (2002) Characterization of nuclease-dependent functions of Exo1p in Saccharomyces cerevisiae. DNA Repair 1:895–912.LaunchUrlCrossRefPubMed↵ Amin NS, Nguyen M-N, Oh S, Kolodner RD (2001) exo1-dependent mutator mutations: Model system for studying functional interactions in mismatch repair. Mol Cell Biol 21:5142–5155.LaunchUrlAbstract/FREE Full Text↵ Kadyrov FA, Dzantiev L, Constantin N, Modrich P (2006) EnExecutenucleolytic function of MutLα in human mismatch repair. Cell 126:297–308.LaunchUrlCrossRefPubMed↵ Constantin N, Dzantiev L, Kadyrov FA, Modrich P (2005) Human mismatch repair: Reconstitution of a nick-directed bidirectional reaction. J Biol Chem 280:39752–39761.LaunchUrlAbstract/FREE Full Text↵ Erdeniz N, Nguyen M, Deschênes SM, Liskay RM (2007) Mutations affecting a Placeative MutLα enExecutenuclease motif impact multiple mismatch repair functions. DNA Repair 6:1463–1470.LaunchUrlCrossRefPubMed↵ Lin DP, et al. (2004) An Msh2 point mutation uncouples DNA mismatch repair and apoptosis. Cancer Res 64:517–522.LaunchUrlAbstract/FREE Full Text↵ Yang G, et al. (2004) Executeminant Traces of an Msh6 missense mutation on DNA repair and cancer susceptibility. Cancer Cell 6:139–150.LaunchUrlCrossRefPubMed↵ Cejka P, Mojas N, Gillet L, Schär P, Jiricny J (2005) Homologous recombination rescues mismatch repair-dependent cytotoxicity of SN1-type methylating agents in S. cerevisiae. Curr Biol 15:1395–1400.LaunchUrlCrossRefPubMed↵ de Wind N, Dekker M, Berns A, Radman M, te Riele H (1995) Inactivation of the mouse Msh2 gene results in mismatch repair deficiency, methylation tolerance, hyperrecombination, and predisposition to cancer. Cell 82:321–330.LaunchUrlCrossRefPubMed↵ Zhang H, Marra G, Jiricny J, Maher VM, McCormick JJ (2000) Mismatch repair is required for O6-methylguanine-induced homologous recombination in human fibroblasts. Carcinogenesis 21:1639–1646.LaunchUrlAbstract/FREE Full Text↵ Nowosielska A, Smith SA, Engelward BP, Marinus MG (2006) Homologous recombination prevents methylation-induced toxicity in Escherichia coli. Nucleic Acids Res 34:2258–2268.LaunchUrlAbstract/FREE Full Text↵ Samson L, Derfler B, Waldstein EA (1986) Suppression of human DNA alkylation-repair defects by Escherichia coli DNA-repair genes. Proc Natl Acad Sci USA 83:5607–5610.LaunchUrlAbstract/FREE Full Text↵ Haracska L, Prakash L, Prakash S (2002) Role of human DNA polymerase κ as an extender in translesion synthesis. Proc Natl Acad Sci USA 99:16000–16005.LaunchUrlAbstract/FREE Full Text↵ Van Houten B, Sancar A (1987) Repair of N-methyl-N′-nitro-N-nitrosoguanidine-induced DNA damage by ABC excinuclease. J Bacteriol 169:540–545.LaunchUrlAbstract/FREE Full Text↵ Wang Y, Qin J (2003) MSH2 and ATR form a signaling module and regulate two branches of the damage response to DNA methylation. Proc Natl Acad Sci USA 100:15387–15392.LaunchUrlAbstract/FREE Full Text↵ Debiak M, Nikolova T, Kaina B (2004) Loss of ATM sensitizes against O6-methylguanine-triggered apoptosis, SCEs, and chromosomal aberrations. DNA Repair 3:359–368.LaunchUrlCrossRefPubMed↵ Caporali S, et al. (2004) DNA damage induced by temozolomide signals to both ATM and ATR: Role of the mismatch repair system. Mol Pharmacol 66:478–491.LaunchUrlAbstract/FREE Full Text↵ Yoshioka K-i, Yoshioka Y, Hsieh P (2006) ATR kinase activation mediated by MutSα and MutLα in response to cytotoxic O6-methylguanine adducts. Mol Cell 22:501–510.LaunchUrlCrossRefPubMed↵ Edelmann W, et al. (1997) Mutation in the mismatch repair gene Msh6 causes cancer susceptibility. Cell 91:467–477.LaunchUrlCrossRefPubMed
Like (0) or Share (0)