ATR regulates hexavlent chromium-induced S-phase checkpoint through phosphorylation of SMC1

Abstract

Hexavalent chromium (Cr[VI]) is an industrial waste product known to cause nasal and lung cancer in exposed workers. Intracellularly, Cr[VI] undergoes a series of enzymatic reductions resulting in the formation of reactive chromate intermediates and oxygen free radicals. These metabolites react with DNA to cause numerous types of genomic lesions, but the cellular response to these genotoxic insults is poorly understood. Recently, we demonstrated that in response to DNA damage induced by Cr[VI], an ataxia-telangiectasia mutated (ATM) and structural maintenance of chromosomal protein 1 (SMC1)-dependent S-phase checkpoint is activated. Interestingly, this checkpoint response was only ATM-dependent in cells exposed to low doses of Cr[VI], we demonstrate that the ATM and Rad3 related kinase, ATR, is required to activate the S-phase checkpoint. In response to all doses of Cr[VI], ATR is activated and phosphorylates SMC1 to facilitate the checkpoint. Further, chromatin binding ability of Rad17 is required for this process. Taken together, these results indicate that the Rad17-ATR-SMC1 pathway is essential for Cr[VI]-induced S-phase checkpoint activation.

1. Introduction

In nature, mammalian cells are exposed to a host of physical and chemical agents capable of causing DNA damage. If left unchecked, this damage can result in mutations, which are passed on to daughter cells and can ultimately lead to tumorigenesis. In order to cope with DNA damage and ensure that genomic fidelity is maintained, cells have evolved a DNA damage response consisting of a series of cell cycle checkpoints that become activated in the face of DNA damage and function to halt progression at all phases of the cell cycle [1,2].

Controlling the DNA damage response is a concisely organized signal transduction network that contains a number of gene products. Central to this network are the phospho-inositide kinase (PIK)-related proteins ataxia telangiectasia mutated (ATM) and ATM-Rad3-related (ATR) [3]. ATM and ATR are high molecular weight protein kinases that are signal transducers in the DNA damage response. ATM and ATR function in parallel, many times phosphorylating the same downstream targets to initiate arrest at all phases of the cell cycle. These proteins do, however, exhibit functional differences. For example, ATM acts mainly in response to DNA double strand breaks induced by ionizing radiation, while ATRs role is mainly to cope with DNA damage induced by UV radiation and hydroxurea (HU) [4].

In contrast to the wealth of data available concerning DNA damage response transducers and effectors, very little is known about the proteins that sense DNA damage. DNA damage sensors should display an essential genetic role in the activation of the DNA damage response and possess the ability to bind to DNA, and four proteins initially characterized in yeast exhibit these abilities. Three of these proteins, Rad9, Rad1, and Hus1 are PCNA like proteins that form a doughnut-like heteromer and could theoretically be loaded onto damaged DNA in a manner similar to PCNA loading onto primed DNA [5–7]. The fourth putative sensor is Rad17. It shares homology with the large subunit of replication factor C (RFC) and has been proposed to interact with the small RFC subunits to tether the Rad9-Rad1-Hus1 (9-1-1) complex to sites of DNA damage and to maintain the damage signal until repair is completed [8]. In addition to these functions, Rad17 interacts with and is phosphorylated by ATR and co-localizes with this protein at sites of DNA damage [6,9]. The significance of these interactions remains in question, although human cells containing a mutant Rad17 incapable of being phosphorylated by ATR display a loss of G2 arrest in response to DNA damage.

Hexavalent chromium (Cr[VI]) is an industrial waste product and environmental pollutant known to cause nasal and lung cancer in humans. Cr[VI] readily enters cells where it is reductively metabolized to Cr[III], in the process generating oxygen free radicals and reactive chromate intermediates. DNA damage induced by chromium metabolism includes single- and double-strand breaks, Cr–DNA crosslinks, protein–DNA crosslinks, and DNA inter- and intra-strand crosslinks. Much data is available concerning the types of DNA damage induced by Cr[VI], but little is known about the mechanisms that cells use to sense and cope with this damage. We recently observed that an ATM-SMC1 dependent S-phase arrest is triggered in response to low doses of Cr[VI] [10], but have since observed an ATM independent S-phase arrest in response to high chromium doses, suggesting that ATM may not be as vital to the Cr[VI] induced DNA damage response as previously believed. In the present study we report that ATR is the critical element in regulating cellular damage response to Cr[VI], and that it, more so than ATM, functions to phosphorylate SMC1 and initiate the S-phase checkpoint. Furthermore, we find that Rad17 chromatin binding ability is required for ATR-mediated S-phase checkpoint activation.

2. Materials and methods

2.1. Cell culture, adenovirus, and Cr[VI] treatment

Simian virus 40-transformed human fibroblast lines from a healthy A-T heterozygote or from a patient with A-T (GM0637 and GM9607, respectively) were obtained from the NIGMS Human Mutant Cell Repository (Camden, NJ). HeLa and HCT 116 cells were purchased from the American Type Culture Collection (Manassas, Va.). The ATR flox cells, which allow for cre-lox mediated removal of ATR, were obtained from Dr. Stephen J. Elledge [11] (Baylor College of Medicine, Houston, TX) and maintained in MEMα medium + 10% FBS + 100 μg/ml G418. Expression of Cre recombinase in these cells was accomplished through infection with adenovirus AdCre1, which was obtained from the LSUHSC vector core laboratory. These cell lines were grown at 37 °C in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum in a humidified 5% CO₂ atmosphere. Potassium chromate (K2CrO₄) was obtained from Sigma (St. Louis, MO) and was dissolved in sterile PBS.

2.2. Electrophoresis and immunoblotting

At indicated times after Cr[VI] treatment, cells were harvested and subsequently lysed in TGN buffer (50 mM Tris–HCl, pH 7.5/50mM glycerophosphate/150mM NaCl/10% glycerol/1% Tween 20/1 mM NaF/1 mM MaVO₄/1 mM phenylmethylsulfonyl fluoride/2 mg/ml pepstatin A/5mg/ml leupeptin/10mg/ml aprotinin/1 mM dithiothreitol) for 15 min on ice. The cell lysates were clarified by microcentrifugation and the supernatant fraction removed and saved. Proteins were resolved by electrophoresis on 4–15% gradient SDS-PAGE gels and transferred to nitrocellulose membranes. Membranes were then probed for 2 h with a polyclonal SMC1 antibody (Bethyl Laboratories) or polyclonal phospho-specific SMC1 Ser966 antibody (Bethyl Laboratories). The membranes were then probed with HRP-conjugated secondary antibody for 1 h and the blots visualized by ECL Western blotting detection reagents (Amersham Biosciences, Piscataway, NJ). Equal protein loading was confirmed by immunoblotting with an antibody to β-tubulin (Sigma).

2.3. Cell transfection

Transfections with empty pcDNA3.1 expression plasmid, constructs encoding, wild-type SMC1 or the Ser957Ala/Ser966Ala SMC1 mutant (provided by Dr. Michael Kastan, St. Jude Children's Research Hospital), or constructs containing wild-type RAD17 or K132E RAD17 were performed on cells in log-phase growth using Lipofectamine (Life Technologies). The efficiency of transfection was assessed by transfection with a green fluorescent protein (GFP) reporter plasmid. Cells were analyzed for GFP expression by flow cytometry 36 h after transfection.

2.4. S-phase checkpoint assay

Activation of the S-phase checkpoint was determined using the method described previously [12–14]. Following transfection cells were prelabeled for ∼24 h by culture in complete growth media containing 10 nCi/ml [¹⁴C]thymidine (NEN). Following prelabeling, the medium was replaced with normal DMEM, and the cells incubated for 6 h. The cells were then treated with Cr[VI] followed by pulse-labeling with 2.5 μCi/ml [³H]thymidine for 15 min (NEN). Following pulse-labeling, cells were harvested, washed twice with PBS, and fixed in cold (−4 °C) 70% methanol for a minimum of 30 min. Afterwards, the cells were applied to Whatman GF/A filters which were then rinsed sequentially in 70% and 95% methanol, air dried, and then placed in glass vials containing 3 ml of scintillation fluid. The filters were subsequently analyzed on a Beckman scintillation counter with windows set to record both ¹⁴C and ³Hdpm. The measure of DNA synthesis was derived from resulting ratios of ³H cpm to ¹⁴C cpm and corrected for counts resulting from channel crossover.

3. Results

3.1. An ATM-independent S-phase checkpoint is activated by high doses of Cr[VI] exposure

ATM is a high molecular weight protein kinase best known for its critical role in the DNA damage response to DNA double strand breaks [15]. ATM regulates cell cycle checkpoints by phosphorylation of a number of downstream targets. Previously, our lab has shown ATM is required for proper S-phase arrest in response to low doses of Cr[VI] (less than 10 μM) [10]. While examining the S-phase checkpoint in response to Cr exposure in ATM-proficient (GM0637) and -deficient (GM9607) cells, we tested a range of Cr concentrations. Consistent with observations from previous experiments using HeLa or 293T cells (unpublished data), we found that the ATM-proficient GM0637 cells exhibited a dose-dependent S-phase arrest in response to 4 h Cr[VI] exposure. Interestingly, we observed that at 20 μM and above, Cr[VI] exposure also induced S-phase arrest in ATM-deficient cells (Fig. 1A). This suggests that an ATM-independent S-phase arrest is activated by high dose Cr exposure.

Fig. 1.

ATM-dependent and -independent S-phase checkpoints are activated in response to Cr[VI] exposure. (A) Replicative DNA synthesis was assessed 4 h after various doses of Cr[VI] exposure in control cells (GM0637) and cells defective in ATM (GM9607). Error bars represent the standard deviation (S.D.) of at least triplicate samples. (B) Cr[VI]-induced SMC1 phosphorylation was assessed by immunoblot analyses using anti-phospho-Ser 966 of SMC1 and an anti-SMC1 antibody in normal (GM0637), ATM deficient (GM9607) cells after treatment of 10 or 40 μM doses of Cr[VI] for 4 h.

Previously, we reported that the Cr[VI]-induced S-phase checkpoint is governed by ATM phosphorylation of SMC1. Since we observed an ATM-independent S-phase checkpoint, we further investigated whether there was ATM-independent SMC1 phosphorylation in response to high dose Cr[VI] exposure. Using a phospho-specific antibody recognizing SMC1 phosphorylation, we performed Western blot analysis in whole cell lysates obtained from HeLa cells treated with 10 or 40 μM of Cr[VI] for 4 h. We found that SMC1 phosphorylation signals are detectable at both 10 and 40 μM of Cr[VI] exposure. As previously reported, we observed that Cr[VI] induced SMC1 phosphorylation is defective in the ATM-deficient cells at the lower dose (10 μM) (Fig. 1B). However, in response to the higher dose of Cr[VI] exposure (40 μM), we observed SMC1 phosphorylation in ATM-deficient cells, indicating that a protein kinase other than ATM is involved in the damage response to high doses of Cr[VI].

3.2. ATR is required for the S-phase checkpoint in response to Cr[VI]

We previously reported that Cr[VI]-induced S-phase checkpoint is a caffeine-sensitive process in the whole does range of Cr[VI] exposure. Caffeine is considered an inhibitor of ATM and its related kinase ATR. Since ATM is not required for S-phase checkpoint activation caused by high doses of Cr[VI], it is quite likely that ATR will play a role in the process. To test this hypothesis, we utilized an established cell system that had been engineered to contain a conditional allele of ATR [11]. ATR can be deleted by introducing Cre recombinase after AdCre virus infection, resulting in ATR-null cells (ATR−/−) (Fig. 2A). HCT 116 (ATR+/+), which is the parental cell line of this system, was used as a positive control. We found that ATR−/− cells display an impaired S-phase arrest in response to the entire dose range of Cr[VI] exposure, as we see little or no decrease in DNA synthesis as compared to ATR(+/+) cells. This suggests that ATR plays a more important role in Cr[VI]-induced S-phase arrest than ATM. Interestingly, we found that ATR(−/−) cells appeared to be synthesizing even more DNA when exposed to low doses of Cr[VI]. At doses higher than 80 μM, the S-phase arrest becomes ATR-independent (data not shown) and is most likely due to physical blocking of the replication machinery and not to the activation of a true DNA damage checkpoint.

Fig. 2.

Cr[VI]-induced S-phase checkpoint requires functional ATR. (A) Manipulation of the ATR gene is assessed by immunoblot analyses using an anti-ATR antibody in cell lysates resulted from HCT116 (ATR +/+) and ATR flox cells exposed to adenovirus containing CRE recombinase. Tubulin levels are shown as a loading control. (B) Replicative DNA synthesis was assessed 4 h after various doses of Cr[VI] exposure in control cells (ATR+/+) and cells defective in ATR (ATR−/−). Error bars represent the standard deviation (S.D.) of at least triplicate samples.

3.3. ATR phosphorylates SMC1 in response to Cr exposure

SMC1 is a member of the structural maintenance of chromosomes protein family, which functions in chromatin condensation and sister chromatid cohesion [16]. We observed an ATM-independent SMC1 phosphorylation after high dose Cr[VI] exposure. To determine if ATR is responsible for SMC1 phosphorylation, we performed Western blot analysis on cell lysates formed from the conditional ATR knock-out cells 30 min after treatment with 40 μM of Cr[VI]. A phospho-specific antibody recognizing Ser966 phosphorylation of SMC1 was used to study SMC1 phosphorylation. We found that depletion of ATR results in defective SMC1 phosphorylation (Fig. 3A), demonstrating that in addition to ATM, ATR is required for SMC1 phosphorylation after Cr[VI] exposure.

Fig. 3.

ATR-dependent SMC1 phosphorylation is required for the Cr[VI] induced S-phase checkpoint. (A) ATR proficient and deficient cells were treated with 40 μM Cr[VI] for 4 h and then lysates were obtained. Cr[VI]-induced SMC1 phosphorylation was assessed by immunoblot analyses using anti-phospho-Ser 966 of SMC1 and an anti-SMC1 antibody. (B) Replicative DNA synthesis was assessed in 293T cells exposed to 40 μM Cr[VI] for 4 h after transfection with empty vector, wildtype SMC1 (wtSMC1), or phosphorylation site mutant SMC1 (2S/A). Error bars represent the standard deviation (S.D.) of at least triplicate samples.

3.4. ATR-mediated SMC1 phosphorylation is required for the Cr[VI] induced S-phase checkpoint

Since ATM-mediated SMC1 phosphorylation is required for low dose Cr[VI]-induced S-phase checkpoint and ATR is required for SMC1 phosphorylation in response to Cr[VI], we further hypothesize that this ATR-mediated SMC1 phosphorylation is needed to activate the checkpoint at high dose Cr[VI] exposure. To test this hypothesis, we over-expressed vector-only, wild-type or serine to alanine mutant SMC1 constructs into 293T cells. We found that cells transiently transfected either with an empty vector or with wild-type Smc1 exhibited a normal S-phase arrest in response to 40 μM Cr[VI] exposure (Fig. 3B). In contrast, cells transiently transfected with vectors expressing SMC1 containing serine to alanine mutation at Ser957 and Ser966 exhibited substantially impaired S-phase arrest (Fig. 3B). These observations demonstrate that ATR-dependent SMC1 phosphorylation in response to Cr[VI] is involved in activation of Cr[VI]-induced S-phase checkpoint. It is noted that over-expression of the exogenous SMC1 protein does not affect the basal level of DNA synthesis (data not shown).

3.5. Rad17 chromatin binding ability is essential for the ATR-SMC1 dependent S-phase arrest after Cr[VI] exposure

Rad17 is believed to serve as a DNA damage sensor due to its ability to bind chromatin and the similarity to the five replication factor C (RFC) subunits [5–8]. ATPase activity is required for Rad17 to bind to chromatin and this chromatin binding ability is required for ATR-mediated cellular response to low dose UV radiation. To study whether Rad17 is involved in Cr[VI] response we investigated the effects of loss of Rad17 chromatin-binding ability on Cr[VI]-induced SMC1 phosphorylation and S-phase checkpoint activation. 293T cells were exposed to 40 μM Cr[VI] for 4 h after being transiently transfected with wild type or K132E mutant Rad17, or empty vector. To assess the transfection efficiency in each experiment, we transfected with a green fluorescent protein (GFP) reporter vector and analyzed for GFP expression by flow cytometric analysis 36 h after transfection. The efficiency of transfection in multiple assessments was always between 90% and 97% (data not shown). Our data shows that cells expressing empty vector or wild-type Rad17 have an optimal SMC1 phosphorylation in response to Cr[VI] exposure (Fig. 4A). However, cells expressing the K132E mutant form of Rad17 have diminished SMC1 phosphorylation, suggesting that the ATPase and chromatin binding activity of Rad17 is required for ATR-dependent SMC1 phosphorylation in response to Cr[VI] exposure. By looking at the Cr[VI]-induced S-phase checkpoint, we found that over-expression of K132E abolished the S-phase arrest, resulting in the radio-resistant DNA synthesis phenotype, while wild-type Rad17 and vector transfected cells showed a robust decrease in DNA synthesis after Cr[VI] exposure (Fig. 4B). These data taken together suggest that chromatin binding ability of Rad17 is essential for the Cr[VI]-induced DNA damage response and that it is required for ATR-mediated SMC1 phosphorylation and S-phase checkpoint activation.

Fig. 4.

Chromatin binding activity of Rad17 is required for ATR-mediated SMC1-phosphorylation and S-phase checkpoint after Cr[VI] exposure. 293T cells were transfected with empty vector, wild-type Rad17, or K132E Rad17. These cells were then exposed to 40 μM Cr[VI] for 4 h and were assessed for (A) SMC1 phosphorylation by immunoblot analyses using anti-phospho-Ser 966 of SMC1 and an anti-SMC1 antibody. (B) Inhibition of DNA synthesis. Error bars represent the standard deviation (S.D.) of at least triplicate samples.

4. Discussion

DNA damage-induced S-phase checkpoint is one of the active cellular responses that may enhance cell survival and limit heritable genetic abnormalities [2]. Recently, we reported that in response Cr[VI] exposure cells employ an optimal S-phase response by down regulation of DNA synthesis [10]. This S-phase checkpoint response is particularly important following cellular exposure to Cr[VI] containing compounds because of the ability of these compounds to not only induce DNA double strand breaks but to also generate numerous replication blocking lesions. Since a major function of the S-phase checkpoint is coordinate DNA replication with DNA repair processes, failure of this checkpoint response could lead to replication fork collapse, loss of genomic fidelity, and ultimately tumorigenesis [2].

The S-phase checkpoint is in part governed by the ATM kinase and its phosphorylation of the downstream target SMC1 [10,17]. In the present study, we extend these insights by demonstrating that the ATM homolog ATR plays a critical role in Cr-induced S-phase arrest. This ATR-mediated S-phase checkpoint is regulated by the Rad17 chromatin binding ability and ATR-mediated phosphorylation of SMC1.

ATM and ATR are protein kinases that play numerous important roles in the DNA damage response. The major difference between ATM and ATR is the type of damage to which each responds. For example, it is well established that ATM plays a critical role in DNA double strand breaks (DSBs) induced by ionizing radiation (IR). This explains why A-T patients are hypersensitive to IR. However, ATM deficient cells respond to UV radiation normally. In contrast to ATM, ATR has been found to be responsive to UV-induced replication stress [4]. Cr[VI] is capable of inducing DNA double strand breaks (DSBs) possibly due to reactive oxygen species generated during cellular reduction. However, at higher Cr[VI] doses these oxidatively induced DSBs may be dwarfed by other genomic lesions such as intra-strand crosslinks and Cr–DNA adducts that result from direct interaction between Cr and DNA. ATM has not been shown to play a significant role in the response to these types of lesions, and indeed our data show that as Cr[VI] doses increase the role of ATM is diminished. On the contrary, ATR is known to be an important player in response to UV radiation and HU, both of which are capable of generating replication stress through the formation of intra-strand crosslinks and DNA adducts. Therefore, it is not surprising that ATM plays less important roles in response to higher dose Cr[VI] exposure than ATR. It is likely that ATM is activated by Cr-induced DNA DSBs and ATR is activated by DNA intra-strand crosslinks and Cr–DNA adducts.

We find that, in addition to ATM, ATR is responsible for SMC1 phosphorylation in response Cr[VI] exposure. As with its role in IR and UV response, SMC1 phosphorylation is required for optimal S-phase arrest after Cr[VI] exposure. The observation that ATM-dependent SMC1 phosphorylation occurs at lower doses and that ATR-dependent SMC1 phosphorylation occurs at higher doses is consistent with its role in S-phase checkpoint activation after Cr[VI] exposure. However, the mechanisms by which SMC1 phosphorylation influences these processes remain to be elucidated.

Rad17 has been implicated as a DNA damage sensor that is required for optimal response to DNA damage. Rad17 binds to chromatin to recruit DNA damage protein complexes in the presence of DNA damage. Depletion of Rad17 leads to a defective ATR function [18]. Previously, it was reported that Rad17 associates with ATM and ATR in response to DNA damage (induced by IR and UV) and replication blockage (induced by hydroxyurea). Here, we find that Rad17 chromatin binding ability is required for ATR-mediated SMC1 phosphorylation and S-phase checkpoint activation following Cr[VI] exposure; however, the mechanism responsible for these observations remains to be elucidated. It will be interesting to determine if there is a direct interaction between Rad17 and ATR following Cr[VI] exposure, or if other protein interactions function to facilitate this response.