The ataxia-telangiectasia related protein ATR mediates DNA-dependent phosphorylation of p53

Article metrics

Abstract

Levels of the tumour suppressor protein p53 are increased in response to a variety of DNA damaging agents. DNA damage-induced phosphorylation of p53 occurs at serine-15 in vivo. Phosphorylation of p53 at serine-15 leads to a stabilization of the polypeptide by inhibiting its interaction with Mdm2, a protein that targets p53 for ubiquitin-dependent degradation. However, the mechanisms by which DNA damage is signalled to p53 remain unclear. Here, we report the identification of a novel DNA-activated protein kinase that phosphorylates p53 on serine-15. Fractionation of HeLa nuclear extracts and biochemical analyses indicate that this kinase is distinct from the DNA-dependent protein kinase (DNA-PK) and corresponds to the human cell cycle checkpoint protein ATR. Immunoprecipitation studies of recombinant ATR reveal that catalytic activity of this polypeptide is required for DNA-stimulated phosphorylation of p53 on serine-15. These data suggest that ATR may function upstream of p53 in a signal transduction cascade initiated upon DNA damage and provide a biochemical assay system for ATR activity.

Introduction

Cellular DNA damage leads to either apoptosis or cell cycle arrest at the S phase, G1/S, or G2/M cell cycle checkpoints, allowing repair of DNA before either replication or segregation of the genome (Elledge, 1996; Paulovich et al., 1997; Weinert, 1998). When a cell is exposed to DNA damaging agents, such as ultra-violet light (UV) or ionizing radiation (IR), a rapid but transient increase in levels of the tumour suppressor protein p53 is observed, resulting in the activation of genes that initiate cell cycle arrest, DNA repair, or induction of apoptosis (Gottlieb and Oren, 1997; Ko and Prives, 1996; Levine, 1997). The N-terminus of p53 is phosphorylated at serine 15 in response to DNA damage induced by IR and UV (Siliciano et al., 1997). Furthermore, phosphorylation of p53 at serine 15 has been proposed to lead to its stabilization by inhibiting interaction with Mdm2, a protein that targets p53 for ubiquitin-mediated degradation (for reviews see Giaccia and Kastan, 1998; Prives, 1998).

Although much is known about how p53 activates genes that lead to cell cycle arrest, repair, or apoptosis, the mechanisms by which DNA damage is detected and signalled to p53 remain unclear. One attractive candidate for a kinase that signals to p53 is the DNA-dependent protein kinase (DNA-PK). DNA-PK consists of a large catalytic subunit (DNA-PKcs) that becomes activated when targeted to DNA by Ku, a heterodimer consisting of Ku70 and Ku80 polypeptides (for reviews see Jackson, 1996; Smith and Jackson, 1999). DNA-PK-mediated kinase activity is stimulated strongly by DNA termini and is capable of phosphorylating a number of substrates in vitro, including p53 at serines 15 and 37 (Jackson, 1996; Steegenga et al., 1996). Indeed, it has recently been reported that phosphorylation of p53 by DNA-PK inhibits the interaction between p53 and Mdm2 in vitro (Shieh et al., 1997). DNA-PK is also involved in the stimulation of p53 DNA binding activity after DNA damage, although this appears to be mediated via a distinct mechanism (Woo et al., 1998). Significantly, however, cells lacking DNA-PK do not exhibit aberrant p53 stabilization in response to DNA damage nor defects in serine-15 phosphorylation, suggesting that other kinases are capable of targeting this site (Shieh et al., 1997; Woo et al., 1998). Here, we report the identification of a novel DNA-activated protein kinase that is capable of phosphorylating p53 at serine-15. We demonstrate that this kinase is distinct from DNA-PK and corresponds to the cell cycle checkpoint protein ATR, a protein that is related in sequence to DNA-PKcs and ATM, the protein deficient in ataxia-telangiectasia (Hoekstra, 1997; Jackson, 1996; Rotman and Shiloh, 1997). Furthermore, catalytic activity of ATR is required for DNA-activated phosphorylation of p53 at serine-15. These data suggest additional mechanisms by which DNA damage may be signalled to p53 and provide a biochemical assay system for the study of ATR catalytic activity.

Results

Identification of a novel DNA-activated protein kinase that phosphorylates p53 on serine-15

To investigate the phosphorylation of p53 on serine-15, we generated antisera against a synthetic p53-derived peptide phosphorylated at this residue (see Materials and methods). As revealed by Western blot analysis, such antibodies (P-Ser15 antibodies) detect recombinant p53 protein that has been incubated with purified DNA-PK in the presence of linear DNA but do not recognize untreated p53 and only weakly detect p53 treated with DNA-PK in the absence of DNA (Figure 1a). Furthermore, these antibodies do not cross-react with DNA-PK phosphorylated p53 that contains a point mutation converting serine-15 to alanine, demonstrating that the antibodies specifically recognize only p53 phosphorylated on serine-15 (Figure 1a). P-Ser15 antibodies are capable of detecting serine-15 phosphorylation of p53 mediated by crude HeLa cell nuclear extract (data not shown, Figure 3b). To determine whether all of this serine-15 kinase activity corresponds to DNA-PK, we fractionated HeLa nuclear extract by chromatography on Q-sepharose and then tested the resulting fractions for p53 serine-15 phosphorylation in the presence of DNA (Figure 1b). In parallel, these fractions were tested by immunoblotting for the presence of DNA-PK components – DNA-PKcs and the two subunits of the Ku70/Ku80 heterodimer. This approach revealed two distinct peaks of activity, termed Serine Fifteen Kinase 1 (SFK1) and SFK2 (Figure 1b). Notably, although SFK1 contains both DNA-PKcs and Ku, neither of these components is detected in fractions containing SFK2. These data therefore provide evidence that the p53 serine-15 kinase function of SFK2 is not mediated by DNA-PK.

Figure 1
figure1

Identification of a novel DNA-activated protein kinase that phosphorylates p53 on serine-15. (a) Antisera raised against phospho-serine-15 of p53 recognize p53 after it is phosphorylated by DNA-PK. Reactions contained either wild-type or Ser15->Ala mutant p53, as indicated. Reactions were performed with p53 alone, or p53 and DNA-PK in the presence or absence of 25 ng of sheared genomic DNA, as indicated. Proteins were separated by SDS – PAGE and p53 serine-15 phosphorylation detected by Western blotting using phospho-specific antibodies raised against serine-15 of p53. (b) Fractionation of HeLa nuclear extract by ion-exchange chromatography reveals two peaks of activity capable of phosphorylating serine-15 of p53. HeLa nuclear extract was fractionated over Q-sepharose and fractions employed in kinase reactions in the presence of 25 ng of sheared genomic DNA using p53 as a substrate. p53 phosphorylation was detected as in (a). (c) Q-sepharose derived SFK2 is not inhibited by low concentrations of wortmannin that inhibit SFK1 (DNA-PK). Q-sepharose derived SFK1 (corresponding to DNA-PK) or SFK2 was employed in kinase reactions using p53 as a substrate in either the presence or absence of 1 μM wortmannin. Phosphorylated p53 was detected as in (a). (d) Q-sepharose derived SFK2 is stimulated by DNA. Kinase assays were performed in the absence (−) or presence of 1, 10, 50, 100 and 250 ng of either sheared genomic, circular plasmid or linear plasmid DNA as indicated. Phosphorylated p53 was detected as in (a)

Figure 3
figure2

Characterization of the novel DNA-activated p53 serine-15 kinase SFK2. (a) SFK2 is inhibited by wortmannin but at higher concentrations than required to inhibit DNA-PK. Kinase assays contained p53 and either enriched SFK2 (upper panel) or DNA-PK (lower panel) in the absence or presence of increasing concentrations of wortmannin. Proteins were separated by SDS – PAGE and p53 serine-15 phosphorylation detected by Western blotting and probing with antibodies raised against p53 phospho-serine-15. (b) ATR but not DNA-PKcs nor ATM co-purifies with SFK2. Equivalent amounts of serine-15 kinase activity (lower panel) from each of the purification steps outlined in Figure 2 were subjected to SDS – PAGE followed by detection of ATR, ATM or DNA-PKcs by immuno-blotting (upper panels)

Previous studies have revealed that DNA-PK and several other members of the PI 3-kinase family are inhibited effectively by sub-micro molar concentrations of the fungal metabolite wortmannin (Hartley et al., 1995; Powis et al., 1994; Ui et al., 1995). Consistent with this, SFK1 is strongly inhibited by 1 μM wortmannin, providing additional evidence that this kinase activity is mediated by DNA-PK (Figure 1c). In contrast, under the assay conditions employed in this study, SFK2 is affected only slightly by wortmannin at this concentration (Figure 1c). This observation therefore provides further support for SFK2 being distinct from DNA-PK. To determine the cofactor requirements for SFK2, we conducted p53 phosphorylation assays in the absence or presence of DNA. As shown in Figure 1d, whereas little or no p53 serine-15 phosphorylation by SFK2 is detected in the absence of DNA, phosphorylation at this site occurs effectively by SFK2 in the presence of either linear or supercoiled plasmid DNA molecules, or sheared genomic DNA. In addition to revealing that SFK2-mediated phosphorylation of p53 is DNA-dependent, these data provide further evidence that the activity does not represent DNA-PK, which has a marked preference for linear DNA in such studies (Gottlieb and Jackson, 1993).

Further purification of SFK2

To analyse the novel DNA-activated p53 kinase(s) comprising SFK2 further, we subjected it to a variety of chromatographic fractionation techniques. In no case did this DNA-PK independent activity resolve into multiple distinct peaks, suggesting that it corresponds to one predominant protein or protein complex (data not shown). Through such approaches, it was possible to devise a purification protocol (outlined in Figure 2a) that yields fractions substantially enriched for SFK2. Analysis of fractions resulting from the final chromatographic step revealed one major polypeptide of >250 kDa and two minor polypeptides of 180 and 120 kDa whose elution profiles all match that of the p53 kinase activity (Figure 2b). Furthermore, the >250 kDa polypeptide could be observed to co-elute with DNA-activated p53 serine-15 kinase activity upon gel-filtration chromatography (data not shown).

Figure 2
figure3

Partial purification of SFK2. (a) Strategy to generate highly enriched preparations of SFK2 employing sequential chromatography on Q-sepharose, DNA-cellulose and heparin-agarose (see Materials and methods). (b) Elution profile of SFK2 activity from heparin-agarose. p53 kinase assays were performed in the presence of sheared genomic DNA. Proteins were separated by SDS – PAGE, and phosphorylation of p53 at serine-15 was detected by Western blotting and probing with antibodies raised against phospho-serine-15 of p53 (lower panel). The same fractions were subjected to SDS – PAGE and proteins visualized by silver staining (upper panel). (c) Phosphorylation of p53 by purified SFK2 is stimulated by DNA. Kinase reactions containing purified SFK2 were conducted in the presence of [γ-32P]ATP in addition to either p53, p53 and sheared genomic DNA, or sheared genomic DNA, as indicated. Proteins were separated by SDS – PAGE and radiolabelled p53 detected by autoradiography

To verify the DNA-dependence and specificity of the highly enriched SFK2 preparation, we tested it for its ability to phosphorylate p53 in the presence of [γ-32P]ATP. These studies revealed that radiolabel incorporation into p53 by the enriched SFK2 preparation is stimulated greatly by DNA, suggesting that other, DNA-independent p53 kinases are not present, or are only minor components of the SFK2 preparations (Figure 2c). Furthermore, phosphorylation of p53 containing a serine-15 to alanine mutation is greatly diminished when compared to wild-type p53, suggesting that serine-15 is the predominant site of phosphorylation by SFK2 (data not shown).

SFK2 contains the PI 3-kinase family member ATR

Although the above studies reveal that DNA-PK and SFK2 are distinct from one another, the DNA-dependence of the two activities are strikingly reminiscent, suggesting that the polypeptides involved may be related. We therefore tested the effect of various concentrations of wortmannin, a selective inhibitor of DNA-PK and certain other members of the PI 3-kinase family, on SFK2 activity. Consistent with previous data (Figure 1c), no inhibition of SFK2 is achieved at low concentrations of wortmannin that inhibit DNA-PK. However, effective inhibition of SFK2 to levels of activity obtained in the absence of DNA is achieved when concentrations of wortmannin are raised to 10 – 30 μM, concentrations that are ineffective against a variety of other protein kinase types that have been tested (Figure 3a, upper panel; Powis et al., 1994; Ui et al., 1995). In contrast, and as shown previously, much lower concentrations of wortmannin effectively inhibit DNA-PK (Figure 3a, lower panel). These data therefore support the possibility that SFK2 corresponds to another member of the PI 3-kinase family. Other members of this kinase family include the human proteins ATM and ATR, both of which are >250 kDa in size and function in the detection and signalling of DNA damage (Hoekstra, 1997; Jackson, 1996; Rotman and Shiloh, 1997). We therefore tested whether either of these proteins co-purifies with SFK2 by equalizing each of the purification steps for p53 SFK2 activity and then probing for ATM, DNA-PKcs and ATR by Western immunoblotting. As shown in Figure 3b, ATR but not DNA-PKcs nor ATM co-purifies with SFK2 activity throughout our purification scheme. Quantitative silver staining and Western blot analysis of fractions generated during the purification scheme reveal that the yield of ATR from HeLa nuclear extract is approximately 30% and that ATR is relatively non-abundant, comprising approximately 0.005% of nuclear protein by weight. Taken together, these data suggest strongly that the predominant polypeptide of >250 kDa detected in these studies is ATR.

ATR catalytic activity is required for DNA-stimulated phosphorylation of p53 at serine-15

The above data suggest strongly that ATR is a component of SFK2. To validate this hypothesis, we tested the activity of full-length recombinant ATR that had been expressed in mammalian cells. For this purpose, we utilized previously described cell lines that express a FLAG epitope-tagged version of full-length ATR under the control of a tetracycline-inducible promoter (Cliby et al., 1998). As a control, we employed an equivalent cell line that expresses a kinase-dead ATR derivative containing a point mutation converting Asp2475 to Ala in the presumptive kinase catalytic domain of the polypeptide (Cliby et al., 1998). Wild-type (wt) and kinase-dead (kd) ATR derivatives were immunoprecipitated from cell extracts and, after washing extensively, the immunoprecipitates were tested for ATR immuno-reactivity and for their ability to phosphorylate p53. Significantly, when incubated with purified p53 in the presence of [γ-32P]ATP, only the wild-type ATR and not the kinase dead ATR derivative catalyzes significant levels of radiolabel incorporation into p53 (Figure 4a). These results indicate that ATR has an intrinsic kinase activity and that this is required for mediating p53 phosphorylation.

Figure 4
figure4

ATR catalytic activity is required for phosphorylation of p53 at serine-15. (a) Wild-type (wt) but not kinase-dead (kd) ATR immunoprecipitates contain a kinase activity capable of phosphorylating p53. Immunoprecipitations were performed using whole cell extracts prepared from cells expressing either recombinant FLAG-tagged ATR (wt), or recombinant FLAG-tagged ATR containing a kinase-dead mutation (kd). Immunoprecipitates were employed in kinase reactions containing p53 and [γ-32P]ATP in the presence of sheared genomic DNA. Proteins were separated by SDS – PAGE and radiolabelled p53 detected by autoradiography. (b) Kinase activity mediated by immunoprecipitated ATR is stimulated by DNA. Immunoprecipitations were performed using whole cell extracts prepared from cells expressing recombinant wt FLAG-tagged ATR. Immunoprecipitates were employed in kinase reactions containing p53 and [γ-32P]ATP, in either the absence or presence of sheared genomic DNA, as indicated. Proteins were separated by SDS – PAGE and radiolabelled p53 detected by autoradiography. (c) Wt but not kd ATR immunoprecipitates contain a kinase activity capable of phosphorylating p53 on serine-15. Immunoprecipitations were performed using whole cell extracts prepared from cells expressing no recombinant ATR (untransfected), or cells expressing recombinant FLAG-tagged wt ATR, or recombinant FLAG-tagged kd ATR. Immunoprecipitates were employed in kinase reactions containing p53 and sheared genomic DNA. Proteins were separated by SDS – PAGE and serine-15 phosphorylation detected by performing Western blots using antibodies raised against p53 phospho-serine-15. (d) ATR immunoprecipitates contain a kinase activity that is inhibited by wortmannin at higher concentrations than required to inhibit DNA-PK. Immunoprecipitations were performed using whole cell extracts prepared from cells expressing recombinant FLAG-tagged ATR. FLAG-ATR immunoprecipitates (upper panel), purified cyclin A/cdk2 (middle panel), or purified DNA-PK (lower panel) were employed in kinase reactions containing [γ-32P]ATP, p53 and sheared genomic DNA, in either the absence or presence of increasing concentrations of wortmannin. Proteins were separated by SDS – PAGE and radiolabelled p53 detected by autoradiography

To establish whether DNA is required for efficient phosphorylation of p53 by immunoprecipitated ATR, kinase reactions using immunoprecipitated wild-type ATR were performed in either the absence or presence of DNA. As illustrated in Figure 4b, ATR-mediated phosphorylation of p53 is stimulated significantly by addition of sheared genomic DNA. Thus, as with purified SFK2, immunoprecipitated ATR also requires DNA as a cofactor for efficient p53 phosphorylation. To determine whether this DNA-activated phosphorylation is directed towards serine-15, we conducted a parallel study in which FLAG-ATR immunoprecipitates were incubated with p53 and unlabelled ATP, then phosphorylation was assessed by immunoblotting with phospho-specific anti-p53 antibodies. These studies revealed that serine-15 phosphorylation activity is detected in immunoprecipitates containing wild-type ATR, but not in those containing the kinase-dead ATR mutant, nor in those generated from cells not expressing recombinant ATR (Figure 4c). Finally, when wortmannin inhibition studies were performed on ATR-containing immunoprecipitates using incorporation of radioactively labelled phosphate as an assay for kinase activity, we found that inhibition of p53 kinase activity occurs at concentrations of 20 – 40 μM (Figure 4d). The lower levels of activity remaining after wortmannin inhibition may correspond to either residual ATR function, or alternatively, to an associated wortmannin insensitive kinase activity. No inhibition of p53 phosphorylation by an unrelated kinase (cyclin A/cdk2; Figure 4d, middle panel) was observed at these wortmannin concentrations, consistent with previous reports that this inhibitor is highly selective towards members of the PI 3-kinase family. Thus, the activity, cofactor requirements, specificity, and inhibition characteristics of the immuno-affinity purified ATR protein parallel those of biochemically purified SFK2.

Discussion

Our results reveal that, in addition to DNA-PK, there exists another protein kinase in human cell extracts that targets serine-15 of p53. Moreover, we find that this novel activity is stimulated by DNA, and thus represents the only characterized example, besides DNA-PK itself and possibly ATM (Gately et al., 1998), of a kinase that is activated by DNA. Our data indicate that this novel p53 kinase is ATR, which is consistent with recent reports that immunoprecipitated ATR is capable of phosphorylating p53 on serine-15 (Canman et al., 1998; Tibbetts et al., 1999). Although it is clear that ATR catalytic activity is necessary for p53 serine-15 phosphorylation in our assays, at this stage we cannot exclude the possibility that the ATR-mediated activity requires additional tightly-associated polypeptides. Indeed, as the inherent catalytic activity of DNA-PKcs is stimulated by Ku, an attractive possibility is that ATR will be regulated through its association with other polypeptides. Consistent with this, gel-filtration studies reveal that SFK2 fractionates at a size greater than the molecular weight of the ATR polypeptide, suggesting that it may be associated with other proteins (data not shown). The availability of purified ATR and assays for ATR activity should allow the identification and characterization of such potential polypeptides.

Significantly, the substrate preferences and cofactor requirements of DNA-PKcs and ATR are very similar, which is consistent with the fact that these two enzymes are related evolutionarily. In line with this, it has recently been shown that the ATR-related polypeptide ATM possesses a kinase activity that targets serine-15 of p53 (Banin et al., 1998; Canman et al., 1998). Furthermore, we have found that fractions lacking ATR and DNA-PKcs but which contain ATM, also display DNA-stimulated p53 serine-15 kinase activity (unpublished data). Recent work has revealed that disruption of ATR function in human cells is manifested by defects in executing DNA damage-induced cell cycle checkpoint controls, and defects in DNA damage-induced p53 serine-15 phosphorylation, although the fact that these cells are SV40 transformed has precluded investigations into p53-dependent G1 checkpoint responses (Cliby et al., 1998; Wright et al., 1998; Tibbetts et al., 1999). Furthermore, whilst disruption of ATR function by expressed kd ATR does not induce sensitivity to UV, it does result in hypersensitivity and intermediate sensitivity to MMS and IR respectively (Cliby et al., 1998). An attractive model, therefore, is that perturbations in genomic integrity are sensed by an ATR-containing complex, resulting in activation of ATR catalytic activity and the phosphorylation of downstream targets, including p53. In turn, phosphorylation of p53 on serine-15 would lead to its activation and stabilization, at least in part, through disrupting its interactions with Mdm2. In a similar manner, the DNA-PK and ATM systems might also target p53. In this regard, A-T cells exhibit defective p53 up-regulation in response to IR, but not to MMS or UV (Artuso et al., 1995; Kastan et al., 1992; Khanna and Lavin, 1993), and defects in the yeast ATR homologue, Mec1p, result in hypersensitivity and impaired cell cycle checkpoint responses towards a range of DNA damaging agents (for reviews see Hoekstra, 1997; Longhese et al., 1998; Weinert, 1998). It is tempting to speculate, therefore, that ATM and ATR, and possibly also DNA-PK, signal different but possibly partially overlapping types of DNA damage to p53. Thus, a diversity of signals could be simply yet effectively integrated into the common p53 effector pathway.

Material and methods

Antibody preparation and Western blotting

Anti-phospho-Ser 15 rabbit antiserum was generated by immunizing with an 11 residue phosphopeptide (corresponding to residues 10 – 20 of human p53 and containing a phospho-serine at residue 15) coupled to KLH (Monrovian Antibody Co., Brno, Czech Republic). Bleeds were assayed by ELISA against the same peptide coupled to BSA and serum collected on days corresponding to peak antibody response. Serum was shown to recognize full-length p53 phosphorylated in vitro with DNA-PK but not unphosphorylated protein nor p53 that had been phosphorylated with a variety of other kinases (Figure 1a and data not shown). The nature of the proteins detected that migrate slower than p53 in certain instances is unknown, but blotting with other p53 antibodies has shown that they do not correspond to p53 isoforms. Polyclonal anti-ATR antisera were raised against recombinant protein spanning residues 2122 – 2380, whilst ATM and DNA-PKcs antibodies have been described previously (Lakin et al., 1996; Song et al., 1996). Antibody raised against residues 11 – 25 of human p53 (Do-1) was purchased from Santa Cruz Biotechnology Inc. Western blotting was performed as described previously (Lakin et al., 1996).

Lysate preparation and immunoprecipitations

Cells were induced to express wild-type (wt) FLAG-tagged ATR or kinase-dead (kd) FLAG-tagged ATR as described previously (Cliby et al., 1998). Cells were harvested and whole cell extracts prepared as described previously (Lakin et al., 1996). Relative expression levels of recombinant protein were determined by Western blotting using an anti-FLAG antibody. Wt and kd FLAG-tagged ATR were diluted in immunoprecipitation buffer (25 mM HEPES-KOH, pH 7.6, 20% glycerol, 2 mM MgC12, 0.2 mM EDTA, 250 mM KC1, 0.1% NP-40, 2 mM microcystin-LR, 1 mM sodium ortho-vanadate and protease inhibitor cocktail [Boehringer Mannheim]) and anti-FLAGM2 sepharose beads (Sigma) added prior to incubation at 4°C with gentle rocking for 1 h. Precipitates were washed seven times in immunoprecipitation buffer and two times in kinase buffer before being split in half and analysed for either kinase activity or by Western blotting with anti-ATR serum.

Kinase assays

Kinase assays containing either DNA-PK, cyclin A/cdk2, SFK2, or anti-FLAG sepharose beads from immunoprecipitations were performed in 20 μl reactions containing 50 ng of p53, 50 ng of plasmid or sheared DNA, and 10 μl of Z′ buffer (25 mM HEPES-KOH pH 7.9, 50 mM KC1, 10 mM MgC12, 2 mM MnC12, 20% glycerol, 0.1% NP-40, 1 mM DTT, 2 mM microcystin-LR). Reactions were assembled and incubated on ice for 3 min prior to ATP addition. ATP was added to a final concentration of 100 μM when performing kinase reactions for Western blotting with phospho-specific antibodies, or to a final concentration of 50 μM along with 5 μCi [γ-32P]ATP when performing radioactive kinase assays. Reactions were incubated at 30°C for 10 min. Wortmannin (Alexis Corporation, Switzerland) was preincubated with kinase for 3 min at 30°C prior to ATP addition. Phosphorylated proteins were subjected to 7.8% SDS – PAGE and visualized either by Western blotting with phospho-specific antibodies or by autoradiography.

Protein purification

All steps were performed at 4°C. HeLa cell nuclear extract (50 ml; Computer Cell Culture Centre, Mons, Belgium) was applied to a Q-Sepharose column equilibrated in D* Buffer (25 mM HEPES-KOH, pH 7.6, 20% glycerol, 2 mM MgC12, 0.2 mM EDTA) containing 50 mM KC1. After washing with two column volumes of 50 mM KC1 D* buffer, proteins were eluted in a continuous salt gradient from 50 – 500 mM KC1 in D* buffer. SFK2 eluted at 70 – 120 mM KC1, whereas SFK1 (DNA-PK) started to elute from 180 mM KC1. Fractions were assayed for p53 serine-15 kinase activity and for the presence of DNA-PKcs and Ku70/Ku 80 by Western blotting. Peak SFK2 fractions were pooled, diluted to 50 mM KC1 in D* buffer and applied to a DNA-cellulose column equilibrated in 50 mM KC1 D*. After washing with two column volumes of 50 mM KC1 D*, proteins were eluted in a continuous KC1 gradient from 50 – 500 mM. SFK2 eluted between 170 – 200 mM KC1. SFK2 fractions were pooled and diluted to 100 mM KC1 D* prior to loading onto Heparin-agarose. After washing with two column volumes of 50 mM KC1 D*, proteins were eluted with a continuous KC1 gradient from 50 – 500 mM in D*. Peak SFK2 activity eluted at between 315 – 350 mM KC1. Fractions were aliquoted, frozen on dry ice and stored at −70°C. DNA-PK was purified as described previously (Dvir et al., 1993). Recombinant p53 was bacterially expressed and purified as described previously (Hupp et al., 1992).

References

  1. Artuso M, Esteve A, Bresil H, Vuillaume M and Hall J. . 1995 Oncogene 11: 1427–1435.

  2. Banin S, Moyal L, Shieh S-Y, Taya Y, Anderson CW, Chessa L, Smorodinsky NI, Prives C, Reiss Y, Shiloh Y and Ziv Y. . 1998 Science 281: 1674–1677.

  3. Canman CE, Lim D-S, Cimprich KA, Taya Y, Tamai K, Sakaguchi KA, Apella E, Kastan MB and Siliciano JD. . 1998 Science 281: 1677–1679.

  4. Cliby WA, Roberts CJ, Cimprich KA, Stringer CM, Lamb JR, Schreiber SL and Friend SH. . 1998 EMBO J. 17: 159–169.

  5. Dvir A, Stein LY, Calore BL and Dynan WS. . 1993 J. Biol. Chem. 268: 10440–10447.

  6. Elledge SJ. . 1996 Science 274: 1664–1672.

  7. Gately DP, Hittle JC, Chan GK and Yen TJ. . 1998 Mol. Biol. Cell. 9: 2361–2374.

  8. Giaccia AJ and Kastan MB. . 1998 Genes Dev. 12: 2973–2983.

  9. Gottlieb TM and Jackson SP. . 1993 Cell 72: 131–142.

  10. Gottlieb TM and Oren M. . 1997 Biochim. Biophys. Acta 1287: 77–102.

  11. Hartley KO, Gell D, Smith GCM, Zhang H, Divecha N, Connelly MA, Admon A, Lees-Miller SP, Anderson CW and Jackson SP. . 1995 Cell 82: 849–856.

  12. Hoekstra MF. . 1997 Curr. Opin. Genet. Dev. 7: 170–175.

  13. Hupp TR, Meek DM, Midgley CA and Lane DP. . 1992 Cell 71: 875–886.

  14. Jackson SP. . 1996 Cancer Surv. 28: 261–279.

  15. Kastan MB, Zhan Q, El-Deiry WS, Carrier F, Jacks T, Walsh WV, Plunkett BS, Vogelstein B and Fornace AJJ. . 1992 Cell 71: 587–597.

  16. Khanna K and Lavin MF. . 1993 Oncogene 8: 3307–3312.

  17. Ko LJ and Prives C. . 1996 Genes Dev. 10: 1054–1072.

  18. Lakin ND, Weber P, Stankovic T, Rottinghaus ST, Taylor AM and Jackson SP. . 1996 Oncogene 13: 2707–2716.

  19. Levine AJ. . 1997 Cell 88: 323–331.

  20. Longhese MP, Foiani M, Muzi-Falconi M, Lucchini G and Plevani P. . 1998 EMBO J. 17: 5525–5528.

  21. Paulovich AG, Toczyski DP and Hartwell LH. . 1997 Cell 88: 315–321.

  22. Powis G, Bonjouklain R, Berggren MM, Gallegos A, Abraham R, Ashdel C, Zazkow L, Matter WF, Dodge J, Grindley G and Vlamos CJ. . 1994 Cancer Res. 95: 2419–2423.

  23. Prives C. . 1998 Cell 95: 5–8.

  24. Rotman G and Shiloh Y. . 1997 Cancer Surv. 29: 285–304.

  25. Shieh S-Y, Ikeda M, Taya Y and Prives C. . 1997 Cell 91: 325–334.

  26. Siliciano JD, Canman CE, Taya Y, Sakaguchi K, Appella E and Kastan MB. . 1997 Genes Dev. 11: 3471–3481.

  27. Smith GCM and Jackson SP. . 1999 Genes Dev. (in press).

  28. Song Q, Lees-Miller SP, Kumar S, Zhang N, Chan DW, Smith GCM, Jackson SP, Alnemri ES, Litwack G and Lavin MF. . 1996 EMBO J. 15: 3238–3246.

  29. Steegenga WT, van der Eb AJ and Jochemsen AG. . 1996 J. Mol. Biol. 263: 103–113.

  30. Tibbetts RS, Brumbaugh KM, Williams JM, Sarkaria JN, Cilby WA, Shieh SY, Taya Y, Prives C and Abraham RT. . 1999 Genes Dev. 13: 152–157.

  31. Ui M, Okado T, Hazeki K and Hazeki O. . 1995 Trends Biochem. Sci. 20: 303–307.

  32. Weinert I. . 1998 Curr. Opin. Genet. Dev. 8: 185–193.

  33. Woo RA, McLure KG, Lees-Miller SP, Rancourt DE and Lee PWK. . 1998 Nature 394: 700–704.

  34. Wright WA, Keegan KS, Herendeen DR, Bentley NJ, Carr AM, Hoekstra MF and Concannon P. . 1998 Proc. Natl. Acad. Sci. USA 95: 7445–7450.

Download references

Acknowledgements

We thank members of the SPJ laboratory for their advice and support. In particular, we thank Graeme Smith, John Rouse, Daniel Durocher and Susan Critchlow for helpful discussions. We also thank David Lane for advice and support. Cell lines expressing ATR derivatives were a generous gift from Stephen Friend, and cyclin A/cdk2 was a gift from Mark Jackman and Jon Pines. This work was funded by grants from the Cancer Research Campaign, the Kay Kendall Leukaemia Fund and the A-T Children's Project.

Author information

Correspondence to Stephen P Jackson.

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • p53
  • ATR
  • ATM
  • DNA-PK
  • ataxia telangiectasia
  • DNA damage

Further reading