Potential Role of MLH1 in the Induction of P53 and Apoptosis by Blocking Transcription on Damaged DNA Templates

Vol. 1, 747–754, August 2003 Molecular Cancer Research 747

Potential Role of MLH1 in the Induction of p53 and Apoptosis by Blocking Transcription on Damaged DNA Templates

Sunitha Yanamadala1 and Mats Ljungman1,2

1Department of Radiation Oncology, Division of Radiation and Cancer Biology, University of Michigan Comprehensive Cancer Center, University of Michigan Medical School and 2Department of Environmental Health Sciences, School of Public Health, University of Michigan, Ann Arbor, MI

Abstract that MMR proteins participate in the regulation of both Defects in DNA mismatch repair (MMR) are common mitotic and meiotic recombination (1, 2) and in transcription- in human cancers, confer tolerance to certain types of coupled repair (3), although their roles in transcription- chemotherapeutic agents, and lead to genomic instabil- coupled repair has been disputed (4, 5). Furthermore, it has ity. In addition to their mismatch-correcting roles during been shown that MMR proteins can bind to certain DNA 6 DNA replication, MMR proteins can bind to certain DNA lesions such as O -methylguanine, 8-hydroxyguanine, and lesions and signal p53 and apoptosis by an unknown N-acetyoxy-N-acetyl-2-aminofluorene and that this binding is mechanism. To further study the mechanism by which linked to the induction of p53 and apoptosis (6–9). The loss of the MMR protein MLH1 is involved in the induction of p53 MMR in tumor cells is thought to give these cells a selective and apoptosis, we exposed the colon carcinoma cell line advantage (10) and subsequently would make the tumor more HCT116 (MLH1-deficient) and mlh1-corrected HCT116 resistant to the induction of apoptosis by anticancer therapy. In sublines to alkylating agents or hydrogen peroxide fact, MMR-deficient cells have been shown to be tolerant to

(H2O2). It was found that while alkylating agents induced many types of anticancer agents such as cisplatin (11), alkylating both apoptosis and phosphorylation of the Ser-15 site of agents (12), methotroxate (13), DNA topoisomerase II inhibitors p53 in a MLH1-dependent manner, induction of apopto- (14), oxidative stress (15, 16) and 5-fluorouridine (17). sis, but not p53 phosphorylation, was MLH1 dependent The role of MMR proteins in the induction of p53 and following treatment with H2O2. The MLH1-dependent apoptosis following induction of DNA damage is not yet induction of p53 phosphorylation by alkylating agents understood, but at least three different mechanisms have been did not appear to be cell cycle dependent, arguing against proposed (18–20). First, replication past a damaged nucleotide a futile repair mechanism operating during S phase as may activate MMR proteins to remove the newly synthesized the sole mechanism for the MLH1-dependent DNA strand opposite the damaged nucleotide. This would lead to a damage signaling. Importantly, we found that both futile cycle of nicking and rejoining, which could subsequently alkylating agents and H2O2 caused significant inhibi- lead to the formation of double-strand breaks and the induction tion of mRNA synthesis in MLH1-expressing but not of p53 and apoptosis. Second, MMR proteins may be involved in MLH1-deficient cells. These findings suggest a novel in preventing recombinational repair of double-strand breaks mechanism of MLH1 in the induction p53 and apop- induced by replication blockage at sites of DNA lesions (19). A tosis by inhibiting RNA polymerase II-dependent third mechanism, i.e., not necessarily replication dependent, transcription on damaged DNA templates. suggests that MMR proteins may bind DNA lesions and form DNA damage signaling complexes capable of inducing p53 and Introduction apoptosis (20). DNA mismatch repair (MMR) proteins play an important In this study, we investigated whether the MMR protein role in repairing mismatches caused by DNA polymerase MLH1 may trigger the induction of p53 and apoptosis errors (1). In addition to mismatch correction, it is thought following treatment with alkylating or oxidative stress-inducing agents by blocking transcription. The rationale of our study was that we have previously shown that there is a strong correlation between the blockage of transcription and the induction of p53 and apoptosis (21–28). Using the MMR-deficient colorectal Received 4/9/03; revised 6/4/03; accepted 6/24/03. The costs of publication of this article were defrayed in part by the payment of cancer cell line HCT116 and MMR-corrected sublines contain- page charges. This article must therefore be hereby marked advertisement in ing either an extra copy of the human chromosome 3 on which accordance with 18 U.S.C. Section 1734 solely to indicate this fact. the mlh1 gene resides (12) or a vector containing the mlh1 Grant support: NIH grant CA82376, a grant from the Gastrointestinal Oncology Program Fund of the University of Michigan Comprehensive Cancer Center, and gene (29), we show that transcription is inhibited in a a grant from the Biomedical Research Council at the University of Michigan MLH1-dependent manner in cells treated with the alkylating Medical School. agents N-methyl-N-nitrosourea (MNU) and N-methyl-NV- Requests for reprints: Mats Ljungman, Department of Radiation Oncology, Division of Radiation and Cancer Biology, University of Michigan Medical nitro-N-nitrosoguanidine (MNNG). Importantly, this inhibition School, 4306 CCGC, 1500 East Medical Center Drive, Ann Arbor, MI 48109- correlated with the induction of p53 phosphorylation and 0936. Phone: (734) 764-3330; Fax: (734) 647-9654. E-mail: [email protected] apoptosis. Furthermore, induction of apoptosis following Copyright D 2003 American Association for Cancer Research. exposure to the oxygen radical-inducing agent hydrogen

peroxide (H2O2) correlated to inhibition of mRNA synthesis. To investigate the potential mechanism of the MLH1- Based on these results, we propose that MLH1 triggers the dependent induction of p53 and apoptosis by alkylating agents, induction of p53 and apoptosis by blocking transcription on we used the colorectal cancer cell line HCT116 (MLH1- damaged DNA templates. deficient) and MMR-corrected HCT116 cell clones transfected with either a copy of the human chromosome 3 (HCT116.ch3) (12) or the human mlh1 gene (29). To verify that alkylating Results agents induce apoptosis in a MLH1-dependent manner in this MLH1-Dependent Induction of Apoptosis and cell system, we treated the MLH1-deficient and MLH1- Phosphorylation of p53 by Alkylating Agents expressing HCT116 cells with either MNNG or MNU for 1 h The alkylating agents MNNG and MNU induce, among in the presence of O6-BG and then scored apoptosis 72 h later. 6 other lesions, O -methylguanine in cellular DNA (30). This At the doses used, these treatments resulted in significant type of lesion is mutagenic, toxic, and normally repaired by the induction of apoptosis in MLH1-proficient but not in MLH1- 6 O -methylguanine methyl transferase, an enzyme that can be deficient cells (Fig. 1A). These results are in accordance with 6 6 inhibited by O -benzylguanine (O -BG) (31, 32). MMR previous reports (6–9). Furthermore, the induction of apoptosis 6 proteins have been shown to bind to O -methylguanine adducts was mediated by O6-methylguanine lesions because it was only (33), and this binding is thought to trigger the induction of p53 observed in cells in which O6-methylguanine methyl transfer- and apoptosis (6–9). However, the mechanism by which MMR ase activity had been inhibited by O6-BG. The MLH1- proteins trigger p53 and apoptosis following exposure to proficient and MLH1-deficient cells were similarly sensitive alkylating agents is unknown. to the induction of apoptosis by actinomycin D, suggesting that the MLH1-proficient cells were not hypersensitive to apoptotic activators in general. It was recently shown that the induction of phosphorylation of the Ser-15 site of p53 following treatments with MNU or MNNG, but not UV light or etoposide, is dependent on MSH2 and MLH1 (7). In that study, the induction of Ser-15 phosphorylation was observed at 18 h and persisted for at least 72 h. In this study, we examined the early kinetics of the induction of Ser-15 phosphorylation in the MLH1-proficient and MLH1-deficient HCT116 cells and found that Ser-15 phosphorylation following MNU treatment was MLH1 dependent as previously reported (7). Furthermore, Ser-15 phosphorylation of p53 could be detected as early as 4 h after treatment (Fig. 1B). No induction of Ser-15 phosphorylation was observed in the MLH1-deficient cells.

Induction of Ser-15 Phosphorylation by MNU Do Not Appear to be Cell Cycle Dependent It has been proposed that the signal that triggers the induction of p53 following exposure to alkylating agents originates from DNA strand breaks generated through futile attempts by MMR proteins to remove the newly synthesized strand opposite O6-methylguanine adducts. If this model is correct, then cells must traverse the S phase following the induction of DNA damage in order for p53 to get induced. To explore whether the p53 phosphorylation in MLH1-proficient cells following exposure to MNU is cell cycle dependent, we used dual-labeling flow cytometry to measure Ser-15 phospho- FIGURE 1. MLH1-dependent induction of apoptosis and phosphoryla- rylation as a function of the cell cycle phase in unsynchro- tion of p53 by alkylating agents. A. HCT116 and HCT116 + vector (MLH1- deficient) and HCT116.ch3 and HCT-mlh1 (MLH1-proficient) cells were nized cells. Although the Ser-15 phosphospecific anti-p53 exposed to 10-AM MNNG, 2-mM MNU, or 200-nM actinomycin D (act D), antibodies gave a very distinct and strong band when used and the amount of apoptosis was assessed by flow cytometry 72 h later. for Western blotting, this antibody gave a weaker signal Legend: light gray bars, MLH1-deficient cells; black bars, MLH1-proficient cells; dark gray bars, MLH1-proficient cells treated with MNU in the when using flow cytometry. We observed a shift representing absence of O6-BG. Columns, average of three to five experiments; bars, a 20–40% increase in the mean antibody FITC signal for the SEM. B. HCT116 and HCT116 + vector (MLH1-deficient) and treated sample compared with untreated controls. Importantly, HCT116.ch3 and HCT116 + mlh1 (MLH1-proficient) cells were exposed 6 for 1 h to 2-mM MNU (in the presence of O -BG), and the phosphorylation this shift was not confined to a certain phase of the cell of p53 at the Ser-15 site was assessed at 0, 4, or 8 h following treatment. cycle; rather, we observed that phosphorylation of p53 at Phosphorylation of the Ser-15 site of p53 was assessed by immunoblotting with a phosphospecific anti-p53 antibody. The Coomassie blue staining Ser-15 increased in all phases of the cell cycle following pattern for total proteins on the membranes showed equal loading. MNU treatment (Fig. 2A). Although slightly more Ser-15

FIGURE 2. MLH1-dependent phosphorylation of p53 is not cell cycle dependent. A. HCT116.ch3 cells were treated with 2-mM MNU as described in Fig. 1, and cells were collected and fixed 8 h later. The amount of Ser-15 phosphorylation as a function of the cell cycle was determined by dual-labeling flow cytometry of the fixed cells stained with phosphospecific anti-p53 antibodies and propidium iodide. Columns, increased amount of Ser-15 antibody reactivity in MNU-treated cells above untreated controls and average of three different experiments; bars, SEM. B. HCT116.ch3 cells were treated with 2-mM MNU and incubated with 33.3-AM BrdUrd during and for 16 h following MNU treatment to specifically label cells that progressed through S phase during the experiment. The data were gated as BrdUrdÀ and BrdUrd+ cells and plotted as number of cells against level of FITC signal (Ser-15-P signal). White plots, untreated cells; gray plots, MNU-treated cells. This experiment was performed nine times, and the mean of the average Ser-15 phosphorylation signal increased 38% and 37% for BrdUrdÀ and BrdUrd+, respectively. These increases were statistically significant with P < 0.005 (BrdUrdÀ) and P < 0.001 (BrdUrd+).

phosphorylation was found in the S and G2/M phases of the cell linked to blockage of transcription (21–24, 28, 35–38). It is cycle, the fact that cells in the G1 phase also triggered Ser-15 possible that RNA polymerase II complexes may serve as phosphorylation suggests that this function of MLH1 is not sensors of DNA damage that when blocked at DNA lesions exclusively S phase dependent. signal p53 and apoptosis as well as repair proteins involved in Although it cannot be excluded that the cells found in the G1 transcription-coupled repair (25). While UV light-induced phase of the cell cycle at the time of fixation had traversed the S cyclobutane pyrimidine dimers and 6-4 photoproducts and G2/M phases during the 8-h postincubation period, this efficiently block the elongation of RNA polymerase II in possibility is unlikely because alkylating agents induce a vitro (39, 40), the premutagenic DNA lesion O6-methylguanine pronounced G2 arrest in these cells (34). However, to further have been shown to be efficiently bypassed by RNA investigate whether the MLH1-dependent induction of Ser-15 polymerases in vitro (41). phosphorylation of p53 is S phase dependent, we performed To test the hypothesis that the MLH1 protein may play a role flow cytometry experiments with cells that had been incubated in DNA damage signaling by interacting with DNA lesions and in the presence of bromodeoxyuridine (BrdUrd) during and blocking transcription, we investigated what effect alkylating after the treatment with MNU. BrdUrd specifically labels cells agents had on the synthesis of mRNA in the MLH1-proficient that are traversing the S phase, and by using anti-BrdUrd and MLH1-deficient HCT116 cells. Our results show that antibodies, the cell samples could be divided into a population mRNA synthesis was not significantly affected in the MMR- of cells that traversed S phase (BrdUrd+) and a population that deficient HCT116 cells following treatment with MNNG did not traverse S phase (BrdUrdÀ) during the 16-h post- or MNU in the presence of O6-BG (Fig. 3). This suggests incubation period. Using this approach and flow cytometry, it that O6-methylguanine lesions do not inhibit transcription was found that following treatment with MNU, the shift in Ser- directly in the absence of MLH1. In the MLH1-corrected cells, 15 phosphorylation of p53 occurred in both BrdUrdÀ and mRNA synthesis was significantly inhibited 4 h following BrdUrd+ cells (Fig. 2B). The shift in Ser-15 phosphorylation treatment. We used the transcription inhibitor actinomycin D as following treatment was found to be significant in both a control and found equal inhibition of transcription in both cell BrdUrdÀ and BrdUrd+ populations. Taken together, these lines (data not shown). In the absence of O6-BG, no inhibition findings suggest that the involvement of MLH1 in signaling of transcription was observed in the MLH1-proficient cells p53 phosphorylation is not restricted to the S phase of the cell treated with MNU, strongly suggesting that O6-methylguanine cycle and argues against an exclusive role of a MMR-directed lesions were responsible for the reduced mRNA synthesis in the futile repair mechanism during the S phase in the induction of MLH1-corrected cells. Time course experiments revealed that Ser-15 phosphorylation of p53. inhibition of transcription was transient and maximal at 4 h after treatment (data not shown). This suggests that transcription MLH1-Dependent Suppression of Messenger RNA blocking may result from a time-dependent MMR complex Synthesis Following Treatment With Alkylating Agents formation at the sites of O6-methylguanine lesions in the To further explore possible mechanisms for the MLH1- transcribed strand of active genes and that the blockage dependent induction of p53 and apoptosis, we tested whether is eventually resolved. Taken together, these results show that alkylating agents may inhibit general mRNA synthesis in a O 6-methylguanine lesions do not by themselves inhibit MLH1-dependent manner. It has previously been shown that transcription, but in the presence of a functional MMR system, the induction of p53 and apoptosis following UV irradiation is mRNA synthesis is significantly reduced.

studies using our cell system, we treated cells with the hydroxyl radical-inducing agent H2O2 and found that apoptosis was induced to a much greater extent in the MLH1-proficient cells than in the MLH1-deficient cells (Fig. 4A). We next investigated the role of MLH1 in the induction of Ser-15 phosphorylation following treatment with H2O2 and found that Ser-15 phosphorylation occurred in both MLH1-proficient and MLH1-deficient cells (Fig. 4B). Furthermore, the phosphorylation was rapid and evident directly after the 1-h H2O2 treatment. This finding may be unexpected because the MLH1-proficient cells were clearly more prone to undergo apoptosis following H2O2 treatment than were the MLH1-deficient cells (Fig. 4A). However, the MLH1- independent induction of Ser-15 phosphorylation could be explained by the significant number of DNA strand breaks induced by H2O2, resulting in the phosphorylation of the Ser-15 site via a MLH1-independent pathway involving the ATM kinase (43–45). Thymine glycols and 8-hydroxyguanine are two biologi- cally important types of lesions induced by ionizing radiation

FIGURE 3. MLH1-dependent inhibition of mRNA synthesis following or treatment with H2O2 (46). Studies using transcription exposure to alkylating agents. HCT116 and HCT116 + vector (MLH1- assays with DNA templates containing site-specific lesions deficient) and HCT116.ch3 and HCT116 + mlh1 (MLH1-proficient) cells suggest that RNA polymerases can bypass both of these types were exposed to 10-AM MNNG or 2-mM MNU, and the amount of nascent mRNA synthesis was assessed 4 h later. Legend as in Fig. 1A. Columns, of lesions in vitro (41, 47, 48). However, 8-hydroxyguanine average of 10 – 12 experiments for the HCT116 and HCT116.ch3 cells and adducts have been shown to inhibit transcription from 6 experiments for the HCT116 + vector and HCT116 + mlh1 cells; bars, plasmids in human cells (49). Thus, there may be specific SEM. Statistically significant differences between the two cell lines are indicated (*P < 0.005, **P < 0.001). factors in cells that bind these lesions and block transcription (47). To directly examine whether oxidative stress may inhibit endogenous mRNA synthesis from genomic DNA and whether a functional MMR status is required for such Induction of Apoptosis, but not p53, Correlates to inhibition, we treated the MMR-deficient and proficient MLH1-Dependent Inhibition of Messenger RNA HCT116 cells for 1 h with 400-AM H2O2 and measured Synthesis Following Treatment With H2O2 mRNA synthesis 4 h later. It was found that H2O2 treatment It has been previously shown that, in addition to alkylating had no significant effect on mRNA synthesis in the MMR- agents, MMR-deficient cells are more tolerant than cor- deficient HCT116 cells, suggesting that H2O2-induced responding MMR-proficient cells to the oxidative stress- DNA lesions, such as thymine glycols and 8-hydroxygua- generating agents ionizing radiation (low-dose rate) (16), nines, do not by themselves block transcription (Fig. 4C). In H2O2, and tert-butyl hydroperoxide. (15, 42). To confirm these contrast, the same treatment of the MMR-corrected HCT116

FIGURE 4. MLH1-dependent induction of apoptosis, but not Ser-15 phosphorylation of p53, following exposure to H2O2 correlates with inhibition of mRNA synthesis. A. HCT116 and HCT116 + vector (MLH1-deficient) and HCT116.ch3 and HCT116 + mlh1 (MLH1-proficient) cells were treated with 400-AM H2O2 for1hat37jC, and apoptosis was analyzed 72 h later as sub-G1 DNA content using flow cytometry. B. MLH1-deficient and corrected cell lines were treated as in A, and the time course of phosphorylation at the Ser-15 site of p53 was assessed using Western blot. The zero time point is from cells harvested 3 directly following the H2O2 treatment. C. The cells were treated as in A, and 4 h later, nascent RNA synthesis was labeled with [ H]uridine. Columns, average of 10 – 12 experiments for the HCT116 and HCT116.ch3 cells and 6 experiments for the HCT116 + vector and HCT116 + mlh1 cells; bars, SEM. Statistically significant differences between the two cell lines are indicated (*P < 0.005, **P < 0.001).

cells resulted in significant inhibition of mRNA synthesis both MLH1-proficient and MLH1-deficient cells. However, in (40–50%). These results suggest that, similarly to O6- agreement with a previous study (7), we found a clear methylguanine lesions, H2O2-induced DNA lesions result in dependence of MLH1 for the induction of Ser-15 phospho- significant inhibition of transcription in a MLH1-dependent rylation of p53 following treatment of cells with MNU or manner. Thus, these findings may explain the earlier findings MNNG in the presence of O6-BG. A recent study found an showing that transcription of plasmids containing site-specific ATM-dependent induction of Ser-15 phosphorylation of p53 8-hydroxyguanine lesions is blocked in cells (49) but not in following MNNG treatment that correlated to induction of vitro (41, 47, 48). DNA strand breaks (51). However, that study used higher doses of MNNG and did not include O6-BG during treatment or the postincubation period. Thus, their results probably reflect base Discussion excision repair-induced DNA strand breaks leading to the In this study, we explored the mechanism by which MMR activation of the ATM kinase. proteins may trigger DNA damage signaling following Our results that no inhibition of transcription was observed exposure to the alkylating agents MNNG and MNU and the in MLH1-deficient HCT116 cells following treatments with 6 oxygen radical-inducing agent H2O2. We show that both the MNNG, MNU, or H2O2 suggest that O -methylguanine induction of apoptosis and the phosphorylation of the Ser-15 adducts or H2O2-induced lesions do not by themselves block site of p53 following exposure to alkylating agents were transcription, which is in accordance with results obtained using dependent on functional MLH1 (Fig. 1) as previously reported in vitro transcription assays (41). However, mRNA synthesis (6–9). Importantly, we observed that the MLH1-dependent was significantly reduced in the MLH1-proficient cells phosphorylation of the Ser-15 site of p53 following treatment following exposures to either alkylating agents or H2O2. It has with MNU occurred within 4 h and in all phases of the cell been shown that both O6-methylguanine adducts (52, 53) and cycle (Fig. 2). These results lend support to a model in which oxidative DNA adducts (49, 54) are subject to transcription- the MLH1-mediated induction of p53 by MLH1 is not coupled base excision repair (tcBER). Furthermore, the MMR dependent on replication. Thus, the ‘‘futile repair’’ model protein MSH2 has been suggested to be required in the tcBER of suggesting that induction of p53 and apoptosis is triggered by oxidative lesions (55). It is possible that the transcription- the attempted mismatch correction of the newly synthesized blocking activity of MLH1 at sites of O6-methylguanine and strand opposite damaged bases cannot fully explain this cell oxidative DNA adducts may in addition to signaling p53 and cycle-independent induction of p53. apoptosis be responsible for the recruitment of base excision We have previously shown that there is a connection repair proteins for tcBER. between blocked transcription and induction of p53 and What is the mechanism by which MLH1 inhibits mRNA apoptosis (21–28). To test the hypothesis that O6-methylguanine synthesis in cells exposed to alkylating agents or oxidative and hydroxyl radical-induced DNA lesions may trigger the stress? One mechanism for the MLH1-dependent reduction in MLH1-dependent p53 and apoptosis by inhibiting transcription, mRNA synthesis could be that MLH1 and other MMR proteins we measured mRNA synthesis in MLH1-deficient and MLH1- bind to DNA lesions and signal the down-regulation of general corrected HCT116 cells. The results show that the MLH1- transcription initiation. However, this is unlikely because we mediated induction of p53 and apoptosis following treatment found that p53 was phosphorylated at the Ser-15 site following with MNNG or MNU correlated to MLH1-dependent inhibition MNU treatment, which we have previously shown is associated of transcription (Fig. 3). This MLH1-dependent inhibition was with inhibition of transcription elongation (28, 56). A more mediated by O6-methylguanine adducts because no inhibition likely mechanism involves the binding of MLH1 and other of mRNA synthesis was detected in MNU-treated cells in MMR proteins to the lesions on DNA, which converts them the absence of the O6-methylguanine transferase inhibitor into transcription-blocking lesions (Fig. 5). Alternatively, 6 O -BG. Furthermore, treatment with H2O2 resulted in MLH1 proteins may travel with the RNA polymerase, and significant inhibition of mRNA synthesis in the MLH1- when encountering lesions in the transcribed strand, the MLH1 corrected but not in the MLH1-deficient HCT116 cells (Fig. proteins may halt transcription leading to the phosphorylation 4C). However, phosphorylation of the Ser-15 site of p53 of the Ser-15 site of p53. following H2O2 treatment was induced regardless of MLH1 Mismatch repair plays an important role in suppressing the status most likely due to the presence of DNA strand breaks formation of human tumors (57, 58). It is well established that inducing phosphorylation of p53 via the ATM kinase. Thus, the mismatch-correcting function of MMR proteins in the the MLH1-dependent induction of apoptosis following wake of DNA replication suppresses mutations and ensures treatment with H2O2 correlated with the inhibition of mRNA genetic stability (18). However, the novel function of MLH1 synthesis but not with the induction of p53. This finding is as a transcription-blocking factor on damaged DNA templates similar to what we have observed in human fibroblasts described in this study may contribute to tumor suppression as exposed to UV light or cisplatin where apoptosis closely well by causing the induction of p53, apoptosis, and perhaps correlated to failure to recover mRNA synthesis rather than tcBER. These results extend our model of RNA polymerase to the induction of p53 (24, 26, 27). II as a sensor for DNA damage to also include lesions that It has been reported that ATM activation/activity following do not directly block transcription but through the interaction exposure to ionizing radiation is not dependent on the MMR with additional factors, such as MMR proteins, cause status of cells (50). Our results using H2O2 would be consistent transcription blockage and subsequent induction of DNA with this finding because Ser-15 phosphorylation occurred in damage signaling. This hypothesis provides a framework for

MNNG or 2-mM MNU in serum-free medium containing 6 25-AM O -BG. Cells were washed and supplied with fresh 6 medium containing 25-AM O -BG and incubated at 37jCfor different periods of time. For treatments with H2O2, the cells were exposed for 1 h at 37jC to 400-AM H2O2 in serum-free medium. The cells were then washed and supplied with fresh media and incubated at 37jC.

Measurement of Messenger RNA Synthesis Cells were seeded in growth media containing 185-Bq/ml [14C]thymidine to uniformly label cellular DNA for 24 h. Cells were then treated with MNNG, MNU, or H2O2 for 1 h as described above. Following a 4-h incubation in fresh media containing O6-BG, nascent RNA was labeled for 1 h by adding [3H]uridine (7.5 Â 106 Bq/ml) and measured as previously FIGURE 5. Model for MLH1-dependent induction of p53 and apoptosis described (22, 56). Briefly, the labeled cells were rinsed twice by blocking transcription elongation. In MLH1-deficient cells (top), the elongating RNA polymerases readily bypass the lesions, and as a result, in ice-cold PBS, detached by scraping, collected by centrifu- the cells do not induce p53 or apoptosis. This scenario would lead to gation, and lysed using a lysis buffer from the straight A’s increased mutation rates and enhanced tolerance toward alkylating agents. In MLH1-proficient cells (bottom), transcription is blocked at sites mRNA isolation system (Novagen, Madison, WI). Polyadenylic of alkylation damage in a MLH1-dependent manner leading to the acid RNA was isolated using polydeoxythymidylic acid- induction of p53 and/or apoptosis. The mechanism of transcription conjugated magnetic beads (Novagen), and total nascent RNA inhibition by MLH1 may involve the formation of transcription-blocking complexes together with other MMR components such as MSH2 at sites of synthesis was measured by precipitating cell lysates with an DNA lesions. Alternatively, MLH1 may associate with the elongating equal volume of 10% ice-cold TCA. The [3H] and [14C] counts transcription machinery and scan the template for certain DNA lesions present on the beads (mRNA) or on the filters (total RNA) were such as O6-methylguanine, thymine glycols, or 8-hydroxyguanine. counted in a scintillation counter using a dual-counting program. Relative total RNA and polyadenylic acid RNA synthesis was then determined by calculating the ratio of [3H]/ future studies searching for additional cellular proteins that may [14C] for each sample and comparing it with the ratio from an signal p53 and apoptosis by converting potentially mutagenic untreated control sample. The data are presented as the DNA adducts into apoptosis-inducing events through the percentage of the [3H]/[14C] ratio for each treatment compared blockage of transcription. Furthermore, these studies may lead with the value determined from untreated control cells. to the development of new cancer therapeutic agents that either target transcription elongation directly or indirectly through the Western Blot interaction with chemotherapy-induced DNA adducts. Cell lysates (30 Ag) were subjected to SDS-PAGE and Western blot as previously described (28). The membranes were Materials and Methods immunoblotted with Ser-15 phosphospecific anti-p53 antibody Cell Lines (Cell Signaling Technology, Beverly, MA). Equal loading and The human colorectal adenocarcinoma cell line HCT116 transfer of total proteins was confirmed by Coomassie blue (MLH1 deficient) and a HCT116 hybrid clone containing an staining of the membranes following Western blot. extra copy of chromosome 3 on which mlh1 resides (HCT116.ch3) were kindly provided to us by Dr. R. Boland Measurement of Apoptosis (University of California at San Diego) (12). The HCT116 cell Attached and floating cells were collected and analyzed for lines harboring an empty vector (HCT116 + vector) or a sub-G1 DNA content using flow cytometry as previously vector expressing the human mlh1 gene (HCT116 + mlh1) described (59). were kindly provided by Dr. F. Praz (Institut Gustave Roussy, Villejuif, France) (29). The cell lines were grown as Measurement of Ser-15 Phosphorylation of p53 in the monolayers in RPMI 1640 medium supplemented with 10% Cell Cycle fetal bovine serum. To the media in which the HCT116.ch3 6 Cells were pretreated for 2 h with 25-AM O -BG followed by cells were growing, 400-Ag/ml geneticin disulfate was added treatment with 2-mM MNU for 1 h at 37jC in serum-free to select for cells carrying this extra chromosome. The 6 medium containing 25-AM O -BG. The cells were then washed expression of MLH1 in HCT116.ch3 and HCT116 + mlh1,or 6 and supplied with fresh medium containing 25-AM O -BG and absence thereof in HCT116 cells, was verified by Western blot incubated at 37jC. In experiments involving BrdUrd, 33.3-AM (data not shown). BrdUrd was added concomitantly with the 2-mM MNU and was included throughout the experiments. Cells were then collected Drug Treatments and fixed in 70% ethanol overnight. The fixed cells were 6 Cells were preincubated in media containing 25-AM O -BG pelleted and rinsed with HBT (5% fetal bovine serum and 0.5% for 2 h at 37jC to inhibit O6-methylguanine transferase [v/v] Tween 20 in PBS) followed by incubation with Ser-15 activity. The cells were then exposed for 1 h at 37jC to 10-AM phosphospecific anti-p53 antibodies with or without Alexa

Fluor-conjugated anti-BrdUrd antibody (Molecular Probes, 16. DeWeese, T. L., Shipman, J. M., Larrier, N. A., Buckley, N. M., Kidd, L. R., Groopman, J. D., Cutler, R. G., te Riele, H., and Nelson, W. G. Mouse embryonic Eugene, OR) for 30 min. Cells were washed in HBT and stem cells carrying one or two defective Msh2 alleles respond abnormally to resuspended and incubated with FITC-conjugated secondary oxidative stress inflicted by low-level radiation. Proc. Natl. Acad. Sci. USA, 95: antibodies. The cells were stained in a 0.5-ml ice-cold HBT 11915 – 11920, 1998. solution containing propidium iodide (180 Ag/ml) and RNase A 17. Meyers, M., Wagner, M. W., Hwang, H. S., Kinsella, T. J., and Boothman, D. A. Role of the hMLH1 DNA mismatch repair protein in fluoropyrimidine- (400 Ag/ml) for 30 min at 4jC. The amount of propidium iodide mediated cell death and cell cycle responses. Cancer Res., 61: 5193 – 5201, staining and Alexa Fluor (BrdUrd) and FITC (Ser-15 2001. phosphorylation) signal were analyzed on a single-cell basis 18. Bellacosa, A. Functional interactions and signaling properties of mammalian using fluorescence-activated cell sorting Calibur two-parameter DNA mismatch repair proteins. Cell Death Differ., 8: 1076 – 1092, 2001. 19. Karran, P. Mechanisms of tolerance to DNA damaging therapeutic drugs. flow cytometer (Becton Dickinson, Franklin Lakes, NJ) and Carcinogenesis, 22: 1931 – 1937, 2001. Cell Quest software. 20. Fishel, R. The selection for mismatch repair defects in hereditary non- polyposis colorectal cancer: revising the mutator hypothesis. Cancer Res., 61: 7369 – 7374, 2001. Acknowledgments We thank Dr. Richard Boland for the generous gift of the HCT116 and 21. Ljungman, M. and Zhang, F. Blockage of RNA polymerase as a possible HCT116.ch3 cells lines and Dr. Franc¸oise Praz for the gift of the HCT116 + trigger for UV light-induced apoptosis. Oncogene, 13: 823 – 831, 1996. vector and HCT116 + mlh1 cells. We thank the Flow Cytometry Core at the 22. Ljungman, M., Zhang, F. F., Chen, F., Rainbow, A. J., and McKay, B. C. University of Michigan Comprehensive Cancer Center for technical assistance, Inhibition of RNA polymerase II as a trigger for the p53 response. Oncogene, 18: Steve Kronenberg for help with some graphics, and Dr. Philip Hanawalt and the 583 – 592, 1999. members of the Ljungman Laboratory for valuable discussions. 23. McKay, B. C., Ljungman, M., and Rainbow, A. J. Persistent DNA damage induced by ultraviolet light inhibits p21(waf1) and bax expression: implications for DNA repair, UV sensitivity and the induction of apoptosis. Oncogene, 17: References 545 – 555, 1998. 1. Modrich, P. and Lahue, R. Mismatch repair in replication fidelity, genetic recombination, and cancer biology. Annu. Rev. Biochem., 65: 101 – 133, 1996. 24. McKay, B. and Ljungman, M. Role for p53 in the recovery of transcription and protection against apoptosis induced by ultraviolet light. 2. Kolodner, R. D. and Marsischky, G. T. Eukaryotic DNA mismatch repair. Neoplasia, 1: 276 – 284, 1999. Curr. Opin. Genet. Dev., 9: 89 – 96, 1999. 25. Ljungman, M. Dial 9-1-1 for p53: mechanisms of p53 activation by cellular 3. Mellon, I., Rajpal, D. K., Koi, M., Boland, C. R., and Champe, G. N. stress. Neoplasia, 2: 208 – 225, 2000. Transcription-coupled repair deficiency and mutations in human mismatch repair genes. Science, 272: 557 – 560, 1996. 26. McKay, B. C., Chen, F., Perumalswami, C. R., Zhang, F. F., and Ljungman, M. The tumor suppressor p53 can both stimulate and inhibit ultraviolet light- 4. Sonneveld, E., Vrieling, H., Mullenders, L. H., and van Hoffen, A. Mouse induced apoptosis. Mol. Biol. Cell, 11: 2543 – 2551, 2000. mismatch repair gene Msh2 is not essential for transcription-coupled repair of UV-induced cyclobutane pyrimidine dimers. Oncogene, 20: 538 – 541, 2001. 27. McKay, B. C., Becerril, C., and Ljungman, M. P53 plays a protective role against UV- and cisplatin-induced apoptosis in transcription-coupled repair 5. Rochette, P. J., Bastien, N., McKay, B. C., Therrien, J. P., Drobetsky, E. A., proficient fibroblasts. Oncogene, 20: 6805 – 6808, 2001. and Drouin, R. Human cells bearing homozygous mutations in the DNA mismatch repair genes hMLH1 or hMSH2 are fully proficient in transcription- 28. Ljungman, M., O’Hagan, H. M., and Paulsen, M. T. Induction of ser15 and coupled nucleotide excision repair. Oncogene, 21: 5743 – 5752, 2002. lys382 modifications of p53 by blockage of transcription elongation. Oncogene, 20: 5964 – 5971, 2001. 6. D’Atri, S., Tentori, L., Lacal, P. M., Graziani, G., Pagani, E., Benincasa, E., Zambruno, G., Bonmassar, E., and Jiricny, J. Involvement of the mismatch repair 29. Jacob, S., Aguado, M., Fallik, D., and Praz, F. The role of the DNA mismatch system in temozolomide-induced apoptosis. Mol. Pharmacol., 54: 334 – 341, 1998. repair system in the cytotoxicity of the topoisomerase inhibitors camptothecin and etoposide to human colorectal cancer cells. Cancer Res., 61: 6555 – 6562, 2001. 7. Duckett, D. R., Bronstein, S. M., Taya, Y., and Modrich, P. hMutS a- and 30. Margison, G. P., Koref, M. F. S., and Povey, A. C. Mechanisms of hMutL a-dependent phosphorylation of p53 in response to DNA methylator 6 damage. Proc. Natl. Acad. Sci. USA, 96: 12384 – 12388, 1999. carcinogenicity/chemotherapy by O -methylguanine. Mutagenesis, 17: 483 – 487, 2002. 8. Hickman, M. J. and Samson, L. D. Role of DNA mismatch repair and p53 in 6 signaling induction of apoptosis by alkylating agents. Proc. Natl. Acad. Sci. USA, 31. Pegg, A. E. Repair of O -alkylguanine by alkyltransferases. Mutat. Res., 462: 96: 10764 – 10769, 1999. 462: 83 – 100, 2000. 32. Friedman, H. S., Keir, S., Pegg, A. E., Houghton, P. J., Colvin, O. M., 9. Wu, J. X., Gu, L. Y., Wang, H. X., Geacintov, N. E., and Li, G. M. Mismatch 6 repair processing of carcinogen-DNA adducts triggers apoptosis. Mol. Cell. Biol., Moschel, R. C., Bigner, D. D., and Dolan, M. E. O -benzylguanine-mediated 19: 8292 – 8301, 1999. enhancement of chemotherapy. Mol. Cancer Ther., 1: 943 – 948, 2002. 10. Bardelli, A., Cahill, D. P., Lederer, G., Speicher, M. R., Kinzler, K. W., 33. Duckett, D. R., Drummond, J. T., Murchie, A. I. H., Reardon, J. T., Sancar, A., Lilley, D. M., and Modrich, P. Human MutS a recognizes damaged DNA base Vogelstein, B., and Lengauer, C. Carcinogen-specific induction of genetic 4 instability. Proc. Natl. Acad. Sci. USA, 98: 5770 – 5775, 2001. pairs containing O-6-methylguanine, O -methylthymine, or the cisplatin-d(GpG) adduct. Proc. Natl. Acad. Sci. USA, 93: 6443 – 6447, 1996. 11. Aebi, S., Kurdi-Haidar, B., Gordon, R., Cenni, B., Zheng, H., Fink, D., Christen, R. D., Boland, C. R., Koi, M., Fishel, R., and Howell, S. B. Loss of 34. Hawn, M., Umar, A., Carethers, J., Marra, G., Kunkel, T., Boland, C., and DNA mismatch repair in acquired resistance to cisplatin. Cancer Res., 56: 3087 – Koi, M. Evidence for a connection between the mismatch repair system and the 3090, 1996. G2 cell cycle checkpoint. Cancer Res., 55: 3721 – 3725, 1995. 12. Koi, M., Umar, A., Chauhan, D., Cherian, S., Carethers, J., Kunkel, T., and 35. Yamaizumi, M. and Sugano, T. UV-induced nuclear accumulation of p53 is Boland, C. Human chromosome 3 corrects mismatch repair deficiency and evoked through DNA damage of actively transcribed genes independent of the microsatellite instability and reduces N-methyl-NV-nitro-N-nitrosoguanidine cell cycle. Oncogene, 9: 2775 – 2784, 1994. tolerance in colon tumor cells with homozygous hMLH1 mutation. Cancer 36. Dumaz, N., Duthu, A., Ehrhart, J. C., Drougard, C., Appella, E., Anderson, Res., 54: 4308 – 4312, 1994. C. W., May, P., Sarasin, A., and DayaGrosjean, L. Prolonged p53 protein 13. Frouin, I., Prosperi, E., Denegri, M., Negri, C., Donzelli, M., Rossi, L., accumulation in trichothiodystrophy fibroblasts dependent on unrepaired Riva, F., Stefanini, M., and Scovassi, A. I. Different effects of methotrexate on pyrimidine dimers on the transcribed strands of cellular genes. Mol. Carcinog., DNA mismatch repair proficient and deficient cells. Eur. J. Cancer, 37: 1173 – 20: 340 – 347, 1997. 1180, 2001. 37. Brash, D. E., Wikonkal, N. M., Remenyik, E., van der Horst, G. T. J., 14. Fedier, A., Schwarz, V. A., Walt, H., Carpini, R. D., Haller, U., and Fink, D. Friedberg, E. C., Cheo, D. L., van Steeg, H., Westerman, A., and van Kranen, H. J. Resistance to topoisomerase poisons due to loss of DNA mismatch repair. Int. J. The DNA damage signal for Mdm2 regulation, Trp53 induction, and sunburn cell Cancer, 93: 571 – 576, 2001. formation in vivo originates from actively transcribed genes. J. Invest. Dermatol., 117: 1234 – 1240, 2001. 15. Hardman, R. A., Afshari, C. A., and Barrett, J. C. Involvement of mammalian MLH1 in the apoptotic response to peroxide-induced oxidative stress. Cancer 38. Blattner, C., Bender, K., Herrlich, P., and Rahmsdorf, H. Photoproducts Res., 61: 1392 – 1397, 2001. in transcriptionally active DNA induce signal transduction to delay UV-

responsive genes for collagenase and metallothionein. Oncogene, 16: 49. Le Page, F., Kwoh, E. E., Avrutskaya, A., Gentil, A., Leadon, S. A., Sarasin, 2827 – 2834, 1998. A., and Cooper, P. K. Transcription-coupled repair of 8-oxoguanine: requirement 39. Donahue, B. A., Yin, S., Taylor, J. S., Reines, D., and Hanawalt, P. C. for XPG, TFIIH, and CSB and implications for Cockayne syndrome. Cell, 101: Transcript cleavage by RNA polymerase II arrested by a cyclobutane pyrimidine 159 – 171, 2000. dimer in the DNA template. Proc. Natl. Acad. Sci. USA, 91: 8502 – 8506, 1994. 50. Brown, K. D., Rathi, A., Kamath, R., Beardsley, D. I., Zhan, Q. M., 40. Tornaletti, S., Reines, D., and Hanawalt, P. C. Structural characterization of Mannino, J. L., and Baskaran, R. The mismatch repair system is required for RNA polymerase II complexes arrested by a cyclobutane pyrimidine dimer in the S-phase checkpoint activation. Nat. Genet., 33: 80 – 84, 2003. transcribed strand of template DNA. J. Biol. Chem., 274: 24124 – 24130, 1999. 51. Adamson, A. W., Kim, W. J., Shangary, S., Baskaran, R., and Brown, K. D. 41. Viswanathan, A. and Doetsch, P. W. Effects of nonbulky DNA base damages ATM is activated in response to N-methyl-NV-nitro-N-nitrosoguanidine-induced on Escherichia coli RNA polymerase-mediated elongation and promoter DNA alkylation. J. Biol. Chem., 277: 38222 – 38229, 2002. clearance. J. Biol. Chem., 273: 21276 – 21281, 1998. 52. Thomale, J., Hochleitner, K., and Rajewsky, M. F. Differential formation and 6 42. Glaab, W. E., Hill, R. B., and Skopek, T. R. Suppression of spontaneous and repair of the mutagenic DNA alkylation product O -ethylguanine in transcribed hydrogen peroxide-induced mutagenesis by the antioxidant ascorbate in mismatch and nontranscribed genes of the rat. J. Biol. Chem., 269: 1681 – 1686, 1994. repair-deficient human colon cancer cells. Carcinogenesis, 22: 1709 – 1713, 2001. 53. Ali, R. B., Teo, A. K., Oh, H. K., Chuang, L. S., Ayi, T. C., and Li, 6 43. Canman, C. E., Lim, D. S., Cimprich, K. A., Taya, Y., Tamai, K., B. F. Implication of localization of human DNA repair enzyme O - Sakaguchi, K., Appella, E., Kastan, M. B., and Siliciano, J. D. Activation of the methylguanine-DNA methyltransferase at active transcription sites in transcrip- 6 ATM kinase by ionizing radiation and phosphorylation of p53. Science, 281: tion-repair coupling of the mutagenic O -methylguanine lesion. Mol. Cell. 1677 – 1679, 1998. Biol., 18: 1660 – 1669, 1998. 44. Banin, S., Moyal, L., Shieh, S. Y., Taya, Y., Anderson, C. W., Chessa, L., 54. Leadon, S. A. and Cooper, P. K. Preferential repair of ionizing radiation- Smorodinsky, N. I., Prives, C., Reiss, Y., Shiloh, Y., and Ziv, Y. Enhanced induced damage in the transcribed strand of an active human gene is phosphorylation of p53 by ATN in response to DNA damage. Science, 281: defective in Cockayne syndrome. Proc. Natl. Acad. Sci. USA, 90: 10499 – 10503, 1674 – 1677, 1998. 1993. 45. Xie, S. Q., Wang, Q., Wu, H. Y., Cogswell, J., Lu, L., Jhanwar-Uniyal, M., 55. Le Page, F., Randrianarison, V., Marot, D., Cabannes, J., Perricaudet, M., and Dai, W. Reactive oxygen species-induced phosphorylation of p53 on Feunteun, J., and Sarasin, A. BRCA1 and BRCA2 are necessary for the serine 20 is mediated in part by polo-like kinase-3. J. Biol. Chem., 276: transcription-coupled repair of the oxidative 8-oxoguanine lesion in human cells. 36194 – 36199, 2001. Cancer Res., 60: 5548 – 5552, 2000. 46. Friedberg, E., Walker, G., and Siede, W. DNA Repair and Mutagenesis. 56. Ljungman, M. and Paulsen, M. T. The cyclin-dependent kinase inhibitor Washington, DC: ASM Press, 1995. roscovitine inhibits RNA synthesis and triggers nuclear accumulation of p53 that is unmodified at Ser15 and Lys382. Mol. Pharmacol., 60: 785 – 789, 2001. 47. Tornaletti, S., Maeda, L. S., Lloyd, D. R., Reines, D., and Hanawalt, P. C. Effect of thymine glycol on transcription elongation by T7 RNA polymerase and 57. Kolodner, R. D. Mismatch repair: mechanisms and relationship to cancer mammalian RNA polymerase II. J. Biol. Chem., 276: 45367 – 45371, 2001. susceptibility. Trends Biochem. Sci., 20: 397 – 401, 1995. 48. Kuraoka, I., Endou, M., Yamaguchi, Y., Wada, T., Handa, H., and 58. Jiricny, J. and Nystrom-Lahti, M. Mismatch repair defects in cancer. Curr. Tanaka, K. Effects of endogenous DNA base lesions on transcription Opin. Genet. Dev., 10: 157 – 161, 2000. elongation by mammalian RNA polymerase II—implications for transcription- 59. Chen, F., Chang, D., Goh, M., Klibanov, S. A., and Ljungman, M. Role of coupled DNA repair and transcriptional mutagenesis. J. Biol. Chem., 278: p53 in cell cycle regulation and apoptosis following exposure to proteasome 7294 – 7299, 2003. inhibitors. Cell Growth & Differ., 11: 239 – 246, 2000.

Downloaded from mcr.aacrjournals.org on September 24, 2021. © 2003 American Association for Cancer Research. Potential Role of MLH1 in the Induction of p53 and Apoptosis by Blocking Transcription on Damaged DNA Templates 1 1 NIH grant CA82376, a grant from the Gastrointestinal Oncology Program Fund of the University of Michigan Comprehensive Cancer Center, and a grant from the Biomedical Research Council at the University of Michigan Medical School.

Sunitha Yanamadala and Mats Ljungman

Mol Cancer Res 2003;1:747-754.

Updated version Access the most recent version of this article at: http://mcr.aacrjournals.org/content/1/10/747

Cited articles This article cites 57 articles, 32 of which you can access for free at: http://mcr.aacrjournals.org/content/1/10/747.full#ref-list-1

Citing articles This article has been cited by 10 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/1/10/747.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/1/10/747. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.