ATM-dependent Dissociation of B55 Regulatory Subunit from Nuclear PP2A in Response to Ionizing Radiation*

Ionizing radiation (IR) is known to activate multiple cell cycle checkpoints that are thought to enhance the ability of cells to respond to DNA damage. Protein phosphatase 2A (PP2A) has been implicated in IR-induced activation of checkpoints; therefore, Jurkat cells were exposed to an activating dose of IR or sham treatment as control, and nuclear extracts were analyzed for PP2A by Mono Q anion exchange chromatography and microcystin affinity chromatography. PP2A exists in eukaryotic cells both as a heterodimer consisting of a 65-kDa scaffolding subunit (A) plus a 36-kDa catalytic subunit (C) and as ABC heterotrimers, containing one of a variety of regulatory (B) subunits. Here we show that IR produces a transient and reversible reduction in the amount of nuclear AB55C heterotrimer without affecting the AB′C heterotrimer or AC heterodimer. In ataxia telangiectasia-mutated (ATM)-deficient cells the amount of nuclear PP2A heterotrimer relative to heterodimer was not reduced by radiation, but the radiation response was restored by transfection of these cells with plasmids encoding ATM. Wortmannin, an inhibitor of kinases such as phosphatidylinositol 3-kinase, also prevented the IR-induced reduction in nuclear PP2A heterotrimer. The changes in nuclear PP2A occurred without a noticeable difference in the carboxyl-terminal methylation of the C subunit, which is known to influence association with B subunits. We conclude a novel ATM-dependent mechanism is regulating association of B55 subunits with nuclear PP2A in response to IR.

Ionizing radiation (IR) is known to activate multiple cell cycle checkpoints that are thought to enhance the ability of cells to respond to DNA damage. Protein phosphatase 2A (PP2A) has been implicated in IR-induced activation of checkpoints; therefore, Jurkat cells were exposed to an activating dose of IR or sham treatment as control, and nuclear extracts were analyzed for PP2A by Mono Q anion exchange chromatography and microcystin affinity chromatography. PP2A exists in eukaryotic cells both as a heterodimer consisting of a 65-kDa scaffolding subunit (A) plus a 36-kDa catalytic subunit (C) and as ABC heterotrimers, containing one of a variety of regulatory (B) subunits. Here we show that IR produces a transient and reversible reduction in the amount of nuclear AB55C heterotrimer without affecting the ABC heterotrimer or AC heterodimer. In ataxia telangiectasia-mutated (ATM)-deficient cells the amount of nuclear PP2A heterotrimer relative to heterodimer was not reduced by radiation, but the radiation response was restored by transfection of these cells with plasmids encoding ATM. Wortmannin, an inhibitor of kinases such as phosphatidylinositol 3-kinase, also prevented the IRinduced reduction in nuclear PP2A heterotrimer. The changes in nuclear PP2A occurred without a noticeable difference in the carboxyl-terminal methylation of the C subunit, which is known to influence association with B subunits. We conclude a novel ATM-dependent mechanism is regulating association of B55 subunits with nuclear PP2A in response to IR.
Ionizing radiation (IR) 1 is known to activate multiple signaling pathways, resulting in diverse stress responses including apoptosis, cell cycle arrest, and gene induction (1,2). Many of these responses are controlled by phosphorylation and dephosphorylation of Ser and Thr residues in proteins. PP2A is a major protein Ser/Thr phosphatase conserved among eukaryotic species that regulates a variety of key steps in metabolism, replication, transcription, and cell cycle (3)(4)(5)(6). PP2A has been implicated in the G 1 /S transition as well as in S phase progression, and genetic and pharmacological evidence shows that it plays a role in mitosis (7)(8)(9). Mutations in the B55 regulatory subunit in both the budding yeast (10) and Drosophila (11) show defects in the spindle assembly checkpoint and sister chromatid separation, respectively. The precise mechanism by which PP2A regulates the mitotic progression is unclear, although it has been implicated both in the dephosphorylation of substrates of cyclin-dependent kinases (CDK) as well as in the negative regulation CDK complexes (4,12). It is, therefore, not surprising that PP2A has been implicated in radiation-induced responses, including activation of checkpoints (13,14).
Thus, PP2A is a potentially important downstream effector of damage-sensing pathways. PP2A-associated proteins have been shown to function in checkpoint pathways. Hox11, an orphan homeobox protein, binds directly to the catalytic subunit of PP2A, and when it is expressed in Jurkat cells, Hox11 completely abrogates the radiation-induced G 2 arrest (13). Similarly, one isoform of the B56 regulatory subunit of PP2A interacts with paxillin at focal adhesions. When an N-terminally truncated form of this B56 subunit is expressed in 3T3 cells, it abrogates radiation-induced G 2 arrest (14). PP2A has also been implicated in the UV-induced activation of the G 1 checkpoint (15), and a novel PP2A regulatory B subunit, termed PR59, has been shown to control rapid dephosphorylation of p107 (16), leading to repression of E2F-dependent transcriptional activity in response to UV. These studies show involvement of PP2A in various damage-sensing pathways.
Although PP2A is involved in multiple checkpoints, the mechanism by which it is targeted to diverse substrates in response to radiation is unknown. Within cells there is essentially no catalytic subunit of PP2A (C) in a free, monomeric state (17), because C binds to a 65-kDa scaffolding unit (called A or PR65) made up of 15 imperfect HEAT sequence repeats (18,19). This PP2A heterodimer (AC) can bind to one of several different B subunits to form a variety of ABC heterotrimers. The B subunits are discrete families of proteins (e.g. B55, B56, and B72) that all recognize similar segments of the scaffolding subunit without having significant sequence identity to one another (7, 20 -24). The B55 subunit is widely distributed, and the other B subunits show differential tissue and developmental distribution. Yeast has two B subunits called Cdc55 and Rts1, and deletion of these genes produces different phenotypes that are not rescued by the other gene (25,26), supporting the concept that the role of the B subunits is to specify the function of different ABC heterotrimers within cells, probably by directing the phosphatase activity to different substrates in different intracellular locales (27).
Little is known about the upstream signals regulating the association/dissociation of the B subunits with the AC heterodimer of PP2A. Recent studies in both yeast and mammalian cells indicate that reversible methyl esterification of the carboxyl-terminal Leu residue in the C subunit of PP2A affects the affinity of the AC heterodimer for B subunits (28,29). A separate methyltransferase (30) and methylesterase (31,32) have been identified, and deletion of these genes altered the methylation of the C subunit of PP2A and the formation of ABC heterotrimers. Antibodies that selectively recognize the methylated form of the PP2A carboxyl terminus have been a useful tool in studying this reaction. There are reports that methylation of the C subunit of PP2A fluctuates during phases of the cell cycle (33) and differs in the cytoplasmic and nuclear compartments, but signals that regulate the pathway of methylation remain unknown. Because the balance between heterotrimer and the heterodimer is expected to play a fundamental role in regulation of the diverse activities of PP2A, we investigated the effect of IR on PP2A distribution between these forms.
Here we show that nuclear PP2A is a downstream effector of an ataxia telangiectasia-mutated (ATM)-dependent damagesensing pathway. Modest doses of ionizing radiation (10 Gy) lead to an acute loss of heterotrimeric PP2A in an ATM-dependent and wortmannin-sensitive manner. Regulatory subunit exchange in response to DNA-damaging agents provides a mechanism by which upstream sensors of DNA damage, such as ATM, could regulate PP2A activity, potentially leading to activation of cell cycle checkpoints.

MATERIALS AND METHODS
Cell Culture-Jurkat cells (a human T cell lymphoma cell line) were grown in RPMI 1640 medium (Invitrogen) with penicillin and streptomycin and 10% fetal bovine serum. FT/pEBS7 and FT/pEBS7-YZ5 cells were both derived from the AT22IJE-T line (34), an immortalized fibroblast line containing a homozygous frameshift mutation at codon 762 of the ATM gene. AT22IJE-T cells were transfected with the mammalian expression vector pEBS7 (35) containing either the hygromycin resistance marker to yield FT/pEBS7 cells or with full-length ATM open reading frame to yield FT/pEBS7-YZ5 cells. FT/pEBS7 and FT/pEBS7-YZ5 were generously provided by Y. Shiloh (Tel Aviv University) and grown in Dulbecco's modified Eagle's medium with 15% fetal bovine serum and 100 g/ml hygromycin B. All cells were in an exponential growth phase at the time of radiation.
Radiation Treatment-Cell cultures were irradiated with a Varian linear accelerator at a dose rate of 9 Gy per min. During irradiation, the cultures were maintained in a container designed to mimic the conditions of the cell culture incubator (5% CO 2 and 95% air at 37°C).
Preparation of Nuclear Extracts-Cells were collected by centrifugation in a nuclear preparation buffer consisting of 20 mM HEPES pH 7.4, 110 mM potassium acetate, 2 mM magnesium acetate, and 5 mM EGTA. The cell pellet was resuspended in the same buffer containing 1 g/ml aprotinin, 1 mM Pefabloc TM , and 1 mM dithiothreitol (complete buffer) at 5 ϫ 10 5 cells/ml. Digitonin was added to a final concentration of 50 g/ml to permeabilize the plasma membrane and release the cytosol. The cell suspension was placed on ice for 5 min, then diluted 10-fold in complete nuclear preparation buffer. Following centrifugation, the pellet containing intact nuclei was extracted with nucleus lysis buffer (1% Nonidet P-40, 0.15 M NaCl, 0.01 M sodium phosphate, pH 7.2, 2 mM EDTA, 50 mM sodium fluoride, 0.2 mM sodium vanadate, and 1 g/ml aprotinin). Insoluble material was removed by centrifugation and the supernatant used as nuclear extract.
Mono Q Chromatography-A Mono Q HR5/5 (Amersham Biosciences) column was equilibrated with Buffer A (25 mM Tris-HCl, pH 7.4, 150 mM NaCl, 30 mM 2-mercaptoethanol). After loading 3 mg of protein of nuclear extract, the column was washed for 16 min (flow rate 0.5 ml/min) with Buffer A followed by a linear gradient to Buffer B (25 mM Tris-HCl, pH 7.4, 500 mM NaCl, 30 mM 2-mercaptoethanol) more than 29 min, after which the column was washed with buffer B for 15 min. Fractions were collected every min (0.5 ml/fraction). For Western analysis, the protein in each fraction was precipitated with 20% trichloroacetic acid and 0.1% deoxycholate, redissolved in 1ϫ SDS sample buffer, and resolved by SDS-PAGE.
Microcystin Affinity Purification and Immunoprecipitation-A 50% slurry of microcystin-Sepharose (Upstate Biotechnology) or 4 g of anti-PP2A catalytic subunit (Upstate Biotechnology) coupled to agarose beads were added to nuclear extract containing 500 g of total protein and incubated for 3 h at 4°C after which the beads were washed three times with lysis buffer. The protein was then dissociated from the beads with SDS-PAGE sample buffer.
Phosphatase Assays-Aliquots (0.1 ml) of each fraction were incubated with 0.3 ml of reaction mixture (20 mM p-nitrophenyl phosphate in 50 mM MOPS, pH 7.4, 20 mM MgCl 2 , 2 mM MnCl 2 , and 1 mM dithiothreitol at 30°C as previously described (36). Reactions were stopped by the addition of 0.17 ml of 2.5 M Na 2 CO 3 , and optical density was measured at 405 nm.

RESULTS
Effects of IR on Nuclear PP2A in Jurkat and AT cells-We prepared nuclear extracts from Jurkat cells 30 min after exposure to either 0 or 10 Gy of IR and fractionated the PP2A by Mono Q chromatography. Fractions 30 -60 contained two peaks of PP2A activity, measured as okadaic acid-inhibitable phosphatase with p-nitrophenyl phosphate as substrate (Fig.  1). This pattern was consistent with the chromatographic behavior of PP2A reported by us and others (36,37). These peaks were the ABC trimers of PP2A (fractions 32-40) preceding the PP2A dimer (fractions 41-60). As shown in Fig. 1A, in multiple independent experiments, radiation of wild-type Jurkat cells caused the loss of most of the peak of activity of the ABC trimer of PP2A. On the other hand, with AT cells lacking the ATM protein, radiation did not significantly diminish the peak of activity of PP2A trimer, relative to a nonirradiated control (Fig.  1B). The reduction in PP2A trimer activity was likely because of the known radiation activation of the ATM kinase, because stable transfection of these AT cells with a vector encoding ATM restored the radiation-induced loss of activity of the PP2A heterotrimer (Fig. 1C). Thus, ATM was required for the reduction in the activity of heterotrimer PP2A in response to IR.
Redistribution of PP2A Subunits in Response to IR-To assay for distribution of PP2A subunits, proteins in the Mono Q column fractions were acid precipitated and subjected to Western analysis. IR caused a parallel loss in recovery of A, B55␣, and C from the peak of heterotrimeric PP2A, as shown in Fig.  2, consistent with the IR-induced loss of the peak of PP2A activity seen in Fig. 1A. The IR-induced loss of the PP2A trimer peak could be because of several possible mechanisms: 1) nuclear export of the trimer, removing it from the nuclear fraction, or 2) binding of an additional protein that would greatly change or eliminate binding to Mono Q, or 3) dissociation of the B55␣ subunit from heterotrimeric PP2A because of a covalent modification, competition by another subunit, or proteolytic degradation of the B55.
To distinguish between these possibilities, and to examine the kinetics of IR-induced loss of heterotrimer, we utilized additional assays to track the recovery of B55␣ at various times following IR. Nuclear extracts prepared following irradiation of cells with 10 Gy were subjected to microcystin affinity purification, direct Western analysis, and immunoprecipitation with anti-PP2Ac antibody. Microcystin binds with nanomolar affinity to the catalytic cleft of the PPP class of protein phosphatases (38) and rapidly and quantitatively recovers both heterotrimeric as well as heterodimeric PP2A from extracts (36). Microcystin affinity purification of protein phosphatases from nuclear extracts (Fig. 3, top panel) showed a loss of B55␣ from PP2A heterotrimer that was evident at 15 min, maximal at 30-min postirradiation, and was then completely reversed by 60-min postirradiation. The recovery of the C subunit was the same at all times, serving as a normalization control for protein recovery and loading of the gel. These results were consistent with the reduction of the peak of heterotrimer PP2A seen in Mono Q chromatography in Figs. 1 and 2. There was no change in the total amount of B55 or C subunit of PP2A in the nucleoplasm from cells at any of these times postradiation (Fig. 3). This made it unlikely that either export or degradation of the B55 occurred in response to IR. Additionally, B55 was missing from PP2A recovered by microcystin via binding to the C subunit, excluding the possibility that the change in elution from Mono Q was because of radiation-induced recruitment of another protein onto the AB55C heterotrimer. Thus, the IR-induced loss of B55␣ seemed to involve a radiation-induced signal leading to transient dissociation of B55␣ from heterotrimeric PP2A.
Interestingly, the association of B56␥, which chromatographs in fractions 9 -12 of the Mono Q column, with PP2A showed no change in association with PP2A in the same samples exhibiting a time-dependent loss of B55␣ after IR (Fig. 3,  bottom panel). These results are consistent with IR inducing a specific dissociation of B55␣-containing PP2A heterotrimer without affecting the B56 heterotrimer. However, we did not Nuclear extracts were prepared as previously described (50) from irradiated and mock-irradiated cells, and 3 mg of nuclear extract protein was analyzed by chromatography on a Mono Q HR5/5. Elution was at a flow rate of 0.5 ml/min for 60 min, and fractions were collected every minute. Aliquots were assayed for phosphatase activity as described under "Experimental Procedures." Error bars show the mean Ϯ S.D. from three separate experiments.
detect an increase in activity of the peak of PP2A dimer (see Fig. 1) or an increase in abundance of the A and C subunits in fractions 41-60 from the Mono Q column, as measured by immunoblot analysis (see Fig. 2). Possibly because the mass and activity of PP2A dimer was much greater than the trimers in these nuclear extracts, we might not have been able to detect a relatively small increase in dimer resulting from conversion of trimer to dimer. Alternatively, a selective subunit exchange reaction could have promoted the dissociation of B55, not B56, and produce a different heterotrimer that did not bind to Mono Q but was still recovered by microcystin or immunoprecipitation.
IR Effects on Nuclear PP2A Are Inhibited by Wortmannin and Require ATM-We tested whether wortmannin at the doses known to nonspecifically block checkpoint kinases including DNA-PK and ATM (39) interfered with the IR-induced loss of nuclear PP2A heterotrimer. Jurkat cells in the presence and absence of wortmannin were irradiated with either 0 or 10 Gy, and 30-min postirradiation nuclear extracts were prepared and subjected to microcystin affinity chromatography followed by immunoblot analysis. Fig. 4A shows that wortmannin treatment of the cells completely abrogated IR-induced loss of B55␣ from heterotrimeric PP2A. Treatment of cells with wortmannin also completely blocked the IR-induced loss of B55␣ analyzed by Mono Q anion exchange chromatography and immunoblotting (Fig. 4B). As before, these results concurred from the two independent assays for the AB55C heterotrimer of PP2A. The observation that wortmannin blocked the IR-induced loss of PP2A heterotrimer suggested that dissociation of B55␣ from ABC trimer might be controlled by checkpoint signaling pathways.   Fig. 4A were irradiated, and nuclear extracts were fractionated by Mono Q chromatography. Aliquots of fractions were processed for immunoblotting with antibodies for PP2A subunits, as described in Fig. 2, which served as the parallel positive control for the effect of IR.
To demonstrate that the IR-induced loss of B55␣ regulatory subunit was ATM-dependent, AT fibroblasts transfected with either an empty vector or with a vector containing full-length wild-type ATM (35) were irradiated with 10 Gy. Nuclear extracts were subjected to Mono Q chromatography and analyzed by Western blotting. AT fibroblasts demonstrated minimal loss of B55␣ from nuclear PP2A heterotrimer, as shown in Fig. 5. However, AT cells ectopically expressing full-length ATM behaved like Jurkat cells (see Fig. 2), with a significant loss of the B55␣ subunit from the PP2A heterotrimer in response to IR.
IR-induced Loss of Heterotrimeric PP2A Is Methylation-independent-Evidence suggests that methyl esterification of the conserved C-terminal residue Leu plays an important role in promoting assembly of PP2A heterotrimer (28,29). AC dimers are the preferred substrate of PP2A methyltransferase, and methylation promotes the stability of ABC holoenzymes. The IR-induced loss of PP2A trimer in an ATM-dependent manner suggested the possibility that ATM could be affecting the activity of either PP2A methyltransferase or methylesterase. We tested if the IR-induced loss of PP2A B55␣-containing trimer involved methylation/demethylation of Leu-309. Jurkat cells were radiated or sham-irradiated. Nuclear extracts were prepared and subjected to microcystin affinity chromatography. Eluted proteins were immunoblotted with a Leu-309 (methylester)-specific PP2A antibody. Even at IR doses far exceeding the dose required to cause loss of the B55␣ subunit from the ABC heterotrimer, IR did not change methylation of C subunit of PP2A, as shown in Fig. 6. AT cells and reconstituted AT cells expressing ATM also failed to show changes in methylation following IR (data not shown). Thus, IR-induced changes in AB55C heterotrimer cannot be explained by changes in PP2A methylation.

DISCUSSION
PP2A participates in a diverse set of cellular events including DNA replication, metabolism, transcription, and activation of cell cycle checkpoints. How PP2A plays such diverse roles is unknown, although it is thought that the regulatory B subunits serve a critical function by targeting the PP2A heterotrimer and affecting substrate specificity (12,40). Although controversial, there is evidence that both ABC trimers and AC dimers exist in cells (41). The loss of B55␣ subunit from heterotrimers provides a possible mechanism for regulating PP2A in response to IR. As shown in the model in Fig. 7, AC has activity in the absence of B and the activity as well as substrate specificity of AC differs from ABC (42). Loss of the B55␣ regulatory subunit from the ABC heterotrimer in response to IR may interrupt normal cell cycle progression leading to checkpoint activation. Either depletion of this heterotrimer or formation of another PP2A holoenzyme could be involved. It is possible that the AC dimer from the AB55C could form a complex with another regulatory or targeting subunit "X," which could in turn enhance PP2A activity in a substrate-specific manner (42). Recent discoveries show that the repertoire of subunits that can associate with the AC heterodimer of PP2A goes beyond the known B subunit families (e.g. B55 and B56) and includes proteins such as stratin and S/G 2 nuclear autoantigen (SG 2 NA) (43,44). If these ACX trimers did not bind to Mono Q, they would not have been detected. PP2A is known to regulate the activity of several families of protein kinases involved in radiation-signaling pathways, including protein kinase B, protein kinase C, extracellular signal-related kinase (ERK), mitogenactivated protein (MAP), and cyclin-dependent kinases (27). Hence, following IR the AC or ACX forms of PP2A could alter kinase activities as a means of transmitting the signal. Identifying substrates for which PP2A has altered activity following IR may reveal novel targets for radiation sensitizers.
What is the mechanism by which IR causes dissociation of PP2A heterotrimer containing B55␣Ј3f phosphorylation is a Nuclear extracts were prepared 30 min following the indicated doses of radiation (upper panels) and at various timepoints following 10 Gy (lower panels). Samples of PP2A affinity purified by microcystin-Sepharose were analyzed by immunoblotting with anti-PP2A-C and methyl-specific anti-PP2A-C antibodies. Loss of B55 subunit from these samples is shown in Fig. 3, which reveals the effects of IR.
FIG. 7. Model for IR-induced changes in PP2A. IR activates ATM through an unknown mechanism leading through a signaling pathway to loss of the B55␣ regulatory subunit from the heterotrimeric ABC complex. AC can either function as a dimer or associate with another regulatory subunit X in mediating DNA damage responses including checkpoint activation. Alternatively, the loss of B55␣ from heterotrimeric PP2A may interrupt cell cycle progression. likely possibility because IR-activated ATM is known to phosphorylate a number of downstream proteins including p53, BRCA1, NBS1, and chk2 (45)(46)(47)(48). In addition, ATM-dependent dephosphorylation of both p53 and histone H1 has been observed, suggesting that ATM signals could activate phosphatases (49,50). Recent observations offer possibilities for IR regulation of PP2A heterotrimer assembly through phosphorylation. The PI 3-like kinase m-TOR, which has an important role in G 1 progression, has been reported to regulate the protein translational regulator PHAS-1 through the phosphorylation of the catalytic subunit of PP2A (51). Alternatively, substitution of the putative phosphorylation site Thr-304 or the known phosphorylation site at Tyr-307 in the PP2A catalytic subunit with a negatively charged amino acid abolished binding of the B subunit to the dimeric PP2A and altered substrate specificity (40). IR-induced phosphorylation of PP2A is a possibility that is currently under investigation. Phosphorylation of the B55 subunit or A subunit are also possible, but never yet observed. Phosphorylation of a protein that competes away and replaces the B55 is an attractive possibility. We show that the mechanism for displacement of the B55 subunit in response to IR is highly specific, without effects on B56, and is reversible with a maximal effect around 30 min and recovery within an hour post-IR. These characteristics are an important step toward definition of the mechanism by which IR leads to dissociation of B55␣ from heterotrimeric PP2A. Regardless of the mechanism, subunit exchange might be the way PP2A is directed to execute its particular functions in response to IR.