Binding of RNA to p53 regulates its oligomerization and DNA-binding activity

Article metrics

Abstract

The C-terminus of p53 is responsible for maintaining the latent, non-DNA-binding form of p53. However, the mechanism by which the C-terminus regulates DNA binding is not yet fully understood. We show here that p53 interacts with RNA via its C-terminal domain and that disruption of this interaction, by RNase A treatment, truncation or phosphorylation of the C-terminus, restores DNA-binding activity. Furthermore, the oligomerization of p53 is significantly enhanced by disrupting the interaction between p53 and RNA. These findings suggest that binding of RNA to p53 is involved in the mechanism of p53 latency.

Introduction

p53 activity is required for tumor suppression, and the cellular response to DNA damage involves the ability of the protein to bind to specific DNA sequences and to function as a transcription factor (El-Deiry et al., 1992; Funk et al., 1992). p53 has also been reported to bind both RNA (Samad and Carroll, 1991) and single-stranded DNA (Bakalkin et al., 1994), in a sequence-independent manner, and has been shown to repress translation of its own mRNA and that of cdk4 mRNA, through a direct interaction (Mosner et al., 1995; Miller et al., 2000). Indeed, p53 is covalently linked to 5.8S ribosomal RNA (Fontoura et al., 1992); it can be associated with a complex containing 5S ribosomal RNA, L5 ribosomal protein and MDM2 (Marechal et al., 1994), and is copurified with polysomal fractions (Fontoura et al., 1997). In addition to binding to RNA and DNA, p53 promotes annealing of single-stranded RNA and DNA into duplex hybrids (Bakalkin et al., 1994).

The C-terminus of p53 protein plays an important role in controlling the structure and function of the entire molecule. DNA-binding activity is regulated by the dimerization and tetramerization domains within the C-terminus of p53 (Hupp et al., 1992). Hupp et al. were the first to discover that phosphorylation or deletion of the C-terminus stimulates DNA binding of inert bacterially expressed p53. This domain was shown to exert a negative effect on the DNA sequence-specific binding activity mediated by the central core domain of the molecule. C-terminally deleted p53 (Hupp et al., 1992, 1993; Marston et al., 1998) or p53 with an altered C-terminus generated by alternative splicing (Kulesz-Martin et al., 1994) loses this negative control. Furthermore, the C-terminus of the p53 molecule was shown to interact directly with DNA in a sequence-nonspecific manner and thus to serve as a sensor for damaged DNA (Jayaraman and Prives, 1995; Lee et al., 1995). However, the mechanism by which the C-terminus regulates DNA binding is not yet fully understood. We have previously shown, using GST-p53 expressed in bacteria, that p53 enhances HMG1 binding to cisplatin-modified DNA (Imamura et al., 2000).

In this study, we investigated the interactions between RNA and p53. We found that binding of p53 to RNA regulates its oligomerization and DNA-binding activity. Our findings suggest that interaction of p53 with RNA may play an important role in the p53 pathway.

Results

P53 binds RNA

MCF7 cells express wild-type p53 and a significant amount of p53 is detected in the cytoplasmic fraction (Figure 1a). To test whether p53 interacts with RNA in vivo, the cytoplasmic fraction prepared from MCF7 cells was immunoprecipitated using a p53-specific antibody, Do-1, and nucleic acids purified from immune complexes were then radio-labeled with [γ-32P] ATP and polynucleotide kinase. The radio-labeled fraction, separated on an acrylamide gel, were of various sizes (Figure 1b, lane 1), and no radio-labeled fraction was observed when mouse IgG (lane 3) was used. Further, the radio-labeled fractions purified from the immunoprecipitate disappeared with the addition of RNase A (lanes 4–8), indicating that they were cytoplasmic RNAs. Figure 1d shows that the labeled RNA was immunoprecipitated from both nuclear and cytoplasmic fractions of MCF7 cells incubated with or without cisplatin (20 μ M for 24 h). Expression of p53 and p21 was observed in untreated MCF7 cells (Figure 1c, lane 2). However, when the cells were treated with cisplatin, p53 levels increased significantly in the nuclear as well as the cytoplasmic fraction (Figure 1c, lanes 3–4). Interestingly, no labeled RNA was detected when the nuclear fraction was immunoprecipitated with anti-p53 antibody (Figure 1d, lanes 5–8), suggesting that p53 only binds to RNA in the cytoplasm. To exclude the possibility that RNA was directly immunoprecipitated with Do-1, we repeated the experiment with another anti-p53 antibody, namely Do-7 (Figure 1e, lane 5). Moreover, we showed that immunoprecipitates from p53-null cells (PC3) do not contain any RNA (Figure 1e, lanes 1–2) and repeated the experiment with cells transfected with Flag-p53 as a positive control (Figure 1f, lane 4).

Figure 1
figure1

In vivo interaction between p53 and RNA. (a) Cytoplasmic fractions of MCF7 cells were immunoprecipitated with anti-p53 antibody (lane 1) or mouse IgG (lane 2) and immunoprecipitates electrophoresed and immunoblotted with antibody to p53. HC indicates heavy chain. (b) After immunoprecipitation, the RNA binding to p53, and cytoplasmic RNA, were isolated using Sepasol reagent, radio-labeled, resolved by polyacrylamide gel electrophoresis in TBE and analysed using a bioimaging analyzer BAS2000. Increasing amounts of RNase A (0.1, 1, 10, 100 ng/ml, and 1 μg/ml) (lanes 5–9) were incubated with cytoplasmic RNA. The amount of RNase A in the lane 2 is 1 μg/ml. Arrows indicate RNA. (c) Western blots of fractionated protein from MCF7 cells with or without 20 μ M cisplatin. The arrows denote p53 (53 kDa) and p21 (21 kDa). (d) Nuclear (lanes 5–8) and cytoplasmic (lanes 1–4) fractions from MCF7 cells, incubated with or without 20 μ M cisplatin, were immunoprecipitated with anti-p53 antibody (lanes 1, 3, 5, 7) or mouse IgG (lanes 2, 4, 6, 8) and isolated using Sepasol reagent, radio-labeled, resolved by polyacrylamide gel electrophoresis in TBE and analysed using a bioimaging analyzer BAS2000. (e) Cytosol fractions from MCF7 cells (lanes 3–6) or PC3 cells (lanes 1–2) were immunoprecipitated with anti-p53 antibody (Do-1 or Do-7) or mouse IgG and isolated using Sepasol reagent, radio-labeled, resolved by polyacrylamide gel electrophoresis in TBE, and analysed using a bioimaging analyzer BAS2000. (f) MCF7 cells were transfected with Flag (lanes 1–2) or Flag-p53 (lanes 3–4) expression plasmids, and whole cell extracts (1 mg) were immunoprecipitated by anti-mouse IgG or anti-Flag antibody. The immune complexes were isolated using Sepasol reagent, radio-labeled, resolved by polyacrylamide gel electrophoresis in TBE and analysed using a bioimaging analyzer BAS2000

Oligomerization of p53 is enhanced by RNase A treatment

Wild-type p53 forms tetramers and higher order oligomeric structures in solution, and the domain responsible for oligomerization is the carboxyl terminus (Wang et al., 1993). In order to determine whether the interaction between p53 and RNA can influence the oligomerization of p53, a pull-down and immunoprecipitation assays were performed. Figure 2a shows a schematic representation of the various expression constructs of p53. The p53 fragments were fused to GST and Flag-tagged plasmids for expression in bacteria and mammalian cells, respectively. The pull-down assay showed that ThioHis-p53 interacted with GST-p53, and this interaction was enhanced by RNase A treatment (Figure 2b, lanes 5–7). ThioHis-p53 can interact with mutant p53 (Figure 2c, lanes 8–9), which contains a point mutation in the codon for amino acid 392 (392 serine to alanine). RNase A treatment again enhanced the interaction with p53 (Figure 2c, lanes 5, 9). As expected, C-terminal truncation (p53 N346) abolished the interaction (lanes 6–7). Glutaraldehyde cross-linking experiments were carried out to determine whether the presence of RNA results in a decrease in the formation of p53 tetramers. GST-p53 was highly purified by the gel-purification method as described previously (Figure 2d, lane 2) (Yoshida et al., 2002). As shown in Figure 2d, formation of tetramers and dimers of the highly purified GST-p53 was reduced in the presence of RNA. To confirm the effect of RNase A treatment on in vivo interaction, we performed an immunoprecipitation assay following transfection of p53-expression plasmids into p53 null PC3 cells. The co-immunoprecipitation assay showed that Flag-p53 interacted with HA-p53 (Figure 3a, lane 4), and this interaction was enhanced by RNase A treatment (lane 5). C-terminal truncation p53 abolished interaction with full-length p53 in vivo (Figure 3b, lanes 6–7). However, full-length p53 can interact with mutant Fag-p53 S392A and the interaction was increased by the RNase A treatment (Figure 3b, lanes 8–9). RNase A treatment did not affect the amount of Flag-p53 in the immune complex (Figure 3B, lanes 10–18).

Figure 2
figure2

Oligomerization of p53 with RNase A in vitro. (a) Schematic representation of GST- and Flag-p53 fusion protein and its deletion mutants used in this assay. A schematic representation of the functional domains of p53 is also shown (top panel). (b) GST (lanes 2–4) or GST-p53 (lanes 5–7) fusion protein (1 μg) immobilized on glutathione-sepharose 4B beads were incubated with ThioHis-p53 expressed in bacteria and increasing amounts of RNase A (0, 100 ng/ml, and 5 μg/ml). Bound protein samples representing 10% of input (lane 1) were electrophoresed and analysed by immunoblotting with anti-Thio antibody. (c) GST fusion proteins (1 μg) immobilized on glutathione-sepharose 4B beads were incubated with ThioHis-p53 expressed in bacteria and RNase A (0 ng/ml and 5 μg/ml). Bound protein samples were subjected to immunoblotting with anti-Thio antibody as described above. (d) The right panel shows CBB staining of highly purified GST-p53 (lane 2). The highly purified GST-p53 was crosslinked by addition of glutaraldehyde to a final concentration of 0.025% with or without yeast tRNA (1 μg/μl). Glutaraldehyde crosslinking samples were subjected to SDS–PAGE and analysed by immunoblotting with antibody to GST

Figure 3
figure3

Oligomerization of p53 with RNase A in vivo. (a) PC3 cells were transfected with both Flag-p53 and HA-p53 expression plasmids (lanes 1–5), and whole cell extracts (1 mg) were immunoprecipitated by anti-mouse IgG (lanes 2–3) or anti-Flag antibody (lanes 4–5) with or without RNase A (5 μg/ml). The immune complexes and 10% of input (lane 1) were electrophoresed and analysed by immunoblotting with antibody to HA. (b) PC3 cells were transfected with HA-p53 expression plasmids (lanes 1–9) and Flag (lanes 2–3), Flag-p53 (lanes 4–5), Flag-p53 N346 (lanes 6–7) or Flag-p53 S392A (lanes 8–9), and whole cell extract (1 mg) were immunoprecipitated by anti-Flag antibody with or without RNase A (5 μg/ml). The immune complexes and 10% of input were electrophoresed and analysed by immunoblotting with antibody to HA or Flag

RNase A treatment activates sequence-specific DNA binding of p53

The C-terminal amino acids of p53 play a key role in the regulation of sequence-specific DNA-binding activity (Hupp et al., 1992; Jayaraman and Prives, 1995). Since the RNA-binding site mapped to this domain, it was important to examine the effect of RNase A treatment on sequence-specific DNA binding by p53. The sequence-specific DNA binding of fixed amounts of purified GST-p53 was activated by RNase A treatment (Figure 4a, lane 4); purified GST-p53 expressed in bacteria possessed only slight DNA-binding activity without RNase A treatment (Figure 4a, lane 3). A competition assay was performed to test whether it is sequence-specific DNA binding of p53 that is influenced by RNA. A p53 consensus oligonucleotide was used as specific competitor (Figure 4a, lanes 6–7), and CRE (lanes 9–10) and NF-κB (lanes 12–13) consensus oligonucleotides were utilized as nonspecific competitors. The signal was substantially decreased by the p53 consensus oligonucleotide but not by the CRE or NF-κB consensus oligonucleotides (Figure 4a, lower panel). These data indicate that the interaction between RNA and the C-terminus of p53 exerts a negative effect on DNA-binding. As shown in Figure 4b, the enhanced DNA-binding activity after RNase A treatment was reduced by addition of nonspecific yeast tRNA (lanes 6–8). Next, we examined the DNA-binding activity of p53 expressed in SaOS-2 cells (p53 null cells) after transfection with the p53 expression plasmid. As shown in Figure 4c, DNA binding by immunoprecipitated p53 was also stimulated by RNaseA treatment when either Flag-p53 (lane 4) or Flag-p53 S392A (lane 6) were introduced by transfection. However, DNA binding was not stimulated by RNaseA treatment in the case of Flag-p53 N346 (lane 8). RNase A treatment did not affect the amount of Flag-p53 in the immune complex (lanes 9–16).

Figure 4
figure4

Effect of RNase A on sequence-specific DNA-binding activity. (a) GST fusion proteins (1 μg) immobilized on glutathione-sepharose 4B beads were mixed with 32P-labeled double-stranded oligonucleotides (22-mers, 0.4 ng/μl) and RNase A (0 ng/ml and 1 μg/ml) (lanes 1–4). Bound oligonucleotides were purified and electrophoresed on a polyacrylamide gel in TBE and analysed using bioimaging analyzer BAS2000. p53 consensus oligonucleotide (lanes 6–7) was used as a specific competitor. A CRE consensus oligonucleotide (lanes 9–10) and a NF-κB consensus oligonucleotide (lanes 12–13) were utilized as nonspecific competitors. (b) Left panel: GST fusion proteins (1 μg) immobilized on glutathione-sepharose 4B beads were incubated with or without RNase A (1 μg/ml) for 60 min (lanes 1–4). After washing the RNase A with binding buffer, RNA (0 μg/ml or 1.6 μg/ml) were incubated with the GST-fusion proteins immobilized on beads for 60 min and further mixed with 32P-labeled double-stranded oligonucleotides (22-mers, 0.4 ng/μl). Bound oligonucleotides were purified and analysed as described in (a). Right panel: GST fusion proteins (1 μg) immobilized on glutathione-sepharose 4B beads were incubated with RNase A (1 μg/ml) for 60 min (lanes 5–8). After washing the RNase A with binding buffer, increasing amounts of RNA (0, 0.1, 0.4, and 1.6 μg/ml) were incubated with the GST-fusion proteins immobilized on beads for 60 min and further mixed with 32P-labeled double-stranded oligonucleotides (22-mers, 0.4 ng/μl). Bound oligonucleotides were purified and analysed as described in (a). (c) SaOS-2 cells were transfected with Flag (lanes 1–2), Flag-p53 (lanes 3–4), Flag-p53 S392A (lanes 5–6) and Flag-p53 N346 (lanes 7–8) expression plasmids, and whole cell extract (1 mg), were immunoprecipitated by anti-Flag antibody with or without RNase A (5 μg/ml). The immune complexes were mixed with 32P-labeled double-stranded oligonucleotides (22-mers, 0.4 ng/μl) and bound oligonucleotides were analysed as described in (a). The immune complexes were electrophoresed and analysed by immunoblotting with anti-Flag antibody (lanes 9–16)

Phosphorylation at serine 392 reduces RNA binding

Phosphorylation of serine 392 by CKII is a well-established p53 modification (Meek et al., 1990; Minamoto et al., 2001). Equal amounts of GST-p53 treated with CKII or PKC using [γ-32P] ATP were resolved by SDS-PAGE and analysed using a bioimaging analyzer. One CKII site and three PKC sites are present in the C-terminal domain of p53 (Levine, 1997). Figure 5a shows that purified GST-p53 is an effective substrate for both CKII and PKC in vitro (lanes 2, 4). Since phosphorylation of p53 by CKII and PKC appears to occur within the C-terminus of p53, we tested whether it affects the interaction with RNA and DNA. In vitro studies of CKII phosphorylation have shown that wild-type p53 binding to a consensus DNA sequence is enhanced by this modification (Hupp et al., 1992). As indicated in Figure 5b, phosphorylation of p53 by CKII, but not PKC, activates the DNA binding of GST-p53 (lane 3). Substitution of alanine at the CKII site in codon 392 (GST-p53 S392A) prevented enhancement of DNA-binding activity by CKII (Figure 5b, lane 3 and 11). As shown in Figure 5c, CKII phosphorylation reduced the RNA-binding activity of p53 (lane 5), but PKC phosphorylation did not (lane 6). RNA-binding of p53 is not affected by the addition of CKII alone without ATP (Figure 5c, lane 9). On the other hand, GST-p53 could bind to RNA even when the CKII phosphorylation site was mutated (Figure 5d, lane7).

Figure 5
figure5

Phosphorylation of p53 enhances the DNA-binding activity and disrupts interaction between RNA and p53. (a) GST-p53 phosphorylated with CKII (lane 2) or PKC (lane 4) was resolved by SDS–PAGE and analysed using a bioimaging analyzer BAS2000. The lower panel is CBB staining (lanes 5–8). (b) GST fusion proteins (1 μg) immobilized on glutathione-sepharose 4B beads were phosphorylated with CKII or PKC, and mixed with 32P-labeled double-stranded oligonucleotides (22-mers, 0.4 ng/μl). Bound oligonucleotides were purified and electrophoresed on a polyacrylamide gel in TBE. Asterisk indicates that phosphorylation of p53 by CKII activates the DNA binding of GST-p53 (lane 3). (c) GST fusion proteins (1 μg) immobilized on glutathione-sepharose 4B beads were phosphorylated with CKII or PKC, and bound RNAs of E. coli origin were purified and electrophoresed on an agarose gel. The arrow indicates RNA. (d) RNAs of E. coli origin binding to GST fusion proteins (1 μg) immobilized on glutathione-sepharose 4B beads were purified and treated with or without 1 μg/ml of RNase A. Bound RNA was purified and electrophoresed on an agarose gel. Arrow indicates RNA

In conclusion, the oligomerization and sequence-specific DNA-binding activities of p53 are significantly enhanced by disruption of its interaction with RNA (Figure 6).

Figure 6
figure6

Schematic summary, and possible roles of RNA binding and phosphorylation

Discussion

Cytoplasmic p53 has been found in association with ribosomes (Fontoura et al., 1997), and p53 has been reported to regulate its own translation via binding of its 5′ leader sequence to cdk4 mRNA (Mosner et al., 1995). The C-terminal region of p53 is dispensable for RNA binding and translation. p53 has also been detected in complexes containing the L5 ribosomal protein (Marechal et al., 1994), and found to be covalently bound to the 5.8S ribosomal RNA (Fontoura et al., 1992). Thus, p53 possesses intrinsic RNA-binding activity and is involved in translation. Previously, PML and p53 were shown to co-localize in the PML nuclear body, leading to inactivation of p53 (Guo et al., 2000). As the nuclear body is a kind of store of ribosomal RNA, it would be interesting to see if the binding of RNA to p53 is related to this localization of p53 in the nuclear body.

It is generally accepted that p53 is a dimer of dimers (Clore et al., 1994, 1995; Lee et al., 1994). Since each monomeric subunit of a dimer binds to a quarter-site, a dimer could theoretically bind to two contiguous quarter-sites that make up a half-site, or to alternating quarter-sites within a full consensus sequence. Since the functional DNA-binding form of p53 is tetrameric, the RNA-binding activity of p53 will affect the regulatory functions of this molecule, such as DNA binding and dimerization activity in addition to translational control. To determine if RNA-binding affects tetramer formation, we performed pull-down and immunoprecipitation assays. We confirm the RNA-binding activity of p53 expressed in mammalian cells (Figure 1). The stability of oligomerization was enhanced by RNase A treatment (Figure 2). Interestingly, p53 treated with RNase A can interact well with p53, indicating that homodimerization is stabilized by RNase A treatment.

The C-terminus of p53 has been described as a region whose function is to control sequence-specific DNA binding. Previous studies have shown that p53 translated in vitro is in a latent, non-DNA-binding form that can be activated by interaction with monoclonal antibody Pab421 (Halazonetis et al., 1993). Deletion of the C-terminus of p53 was also shown to activate DNA-binding activity (Hupp et al., 1992). However, the mechanism by which the C-terminus regulates DNA binding is not yet fully understood. Regulation of DNA-binding activity is mediated by the dimerization and tetramerization domains within the C-terminus (Hupp et al., 1992). As shown in Figure 3, we showed that RNase A treatment can stabilize the mutual interaction of p53. Therefore, to further analyse the contribution of RNase A treatment to the regulation of DNA binding function, we analysed DNA-binding activity with or without RNase A. As expected, sequence-specific DNA-binding activity was enhanced by RNase A treatment. Hupp et al. reported that DNA binding of p53 expressed in E. coli was cryptic. Our results suggest that p53 expressed in bacteria binds to bacterial RNA to inhibit DNA-binding activity. The idea of ‘latent’ p53 has been questioned recently by the evidence that p53 is an active DNA-binding protein that does not require activation either in vitro or in vivo (Espinosa and Emerson, 2001; Kaeser and Iggo, 2002; Yakovleva et al., 2002). Our data show that binding of the C-terminal domain of p53 to RNA almost completely prevents the interaction of the core domain with specific DNA sequences. However, we detected no RNA when nuclear fractions of MCF7 cells pre-incubated with or without cisplatin were immunoprecipitated with anti-p53 antibody (Figure 1d). This suggests that nuclear p53 can bind to DNA but not to nuclear RNA, and that the damage-inducible modification of p53 may not interfere with DNA-binding activity. It has been shown recently that the DNA-binding activity of p53 is strongly dependent on structural features of the target DNA (Ahn and Prives, 2001). On the basis of these findings, we hypothesize that the binding of RNA by p53 in the cytoplasm may control p53 latency.

Phosphorylation of the serine 392 by CKII is a well-established p53 modification and has been reported to increase sequence-specific DNA binding (Hupp and Lane (1995); Hao et al., 1996). Indeed, several experimental approaches in vitro and in intact mice support this notion (Crook et al., 1994; Keller et al., 2001), contributing to the evidence that phosphorylation of p53 fine-tunes the protein to respond to specific stresses. On the other hand, several groups have demonstrated that phosphorylation of residue 392 has no significant impact on p53 transcriptional activity (Fiscella et al., 1994; Fuchs et al., 1995). Thus, the function of the C-terminal domain of p53 remains uncertain. Furthermore, only phosphorylation of serine 392 significantly enhanced tetramer stability (Sakaguchi et al., 1997). We phosphorylated GST-p53 with CKII and PKC in vitro and examined the effect on RNA- and DNA-binding activity. Phosphorylation by CKII enhanced the DNA binding of p53 (Figure 5b). Furthermore, phosphorylation of p53 by CKII reduced the RNA binding of GST-p53 expressed in bacteria, but phosphorylation by PKC did not (Figure 5c, d). This suggests that enhanced tetramer stability and sequence-specific DNA-binding activity by CKII-dependent phosphorylation are caused by disruption of RNA binding. We next introduced a mutation at amino acid 392. Substitution of alanine for serine did not affect the binding of RNA by p53. This also suggests that phosphorylation of C-terminal serines plays a critical role in DNA/RNA binding and oligomerization. CKII is ubiquitously expressed in mammalian cells and is present in both nuclei and cytoplasm, indicating that C-terminal phosphorylation by CKII may be observed in both nuclei and cytoplasm. It has been shown that CKII interacts with p53 (Filhol et al., 1992). This interaction was significantly reduced in the presence of ATP. The addition of CKII without ATP could not affect the RNA binding of p53 (Figure 5c). These results again suggest that phosphorylation by CKII but not interaction with CKII is critical for release from RNA. One interesting possibility is that p53 translated in the cytoplasm first binds to RNA, after which its phosphorylation releases it from RNA and prevents it from entering the degradation cascade while enhancing tetramerization to facilitate nuclear translocation when CKII is activated (Figure 6).

In conclusion, our results provide experimental evidence of RNA-regulated activation of p53 oligomerization and DNA binding. A salient feature of the results is that C-terminal phosphorylation by CKII plays a pivotal role in DNA binding via release from RNA.

Materials and methods

Cell culture and antibodies

SaOS-2 human osteosarcoma cells and MCF7 human breast cancer cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum. PC3 human prostate cancer cells were grown in modified Eagle's medium supplemented with 10% fetal bovine serum.

Anti-Thio antibody and anti-Flag (M2) antibody were purchased from Invitrogen (San Diego, CA, USA) and Sigma, respectively. Anti-HA monoclonal antibody, anti-Bcl2 antibody, and anti-p53 (Do-1 and Do-7) antibody were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA).

Plasmid preparation

Full-length human p53 cDNA was amplified from total RNA extracted from the human epidermoid cancer cell line, KB, by reverse transcription–PCR. The construction of GST-p53, GST-p53 N346, and ThioHis-p53 has been described previously (Yoshida et al., 2003). GST-p53 S392A (392 serine to alanine) was obtained by PCR using the following primer pairs: 5′-IndexTermCCATGGAGGAGCCGCAGTCAGATCC-3′ and 5′-IndexTermTCAGTCTGCGTCAGGCCCTTCTG-3′.

Expression and purification of GST and ThioHis fusion proteins

GST and ThioHis fusion proteins were induced by 1 mM isopropyl-1-thio-ß-D-galactopyranoside as described previously (Yoshida et al., 2002). The E. coli cells were sonicated for 10 s in binding buffer (50 mM Tris–HCl pH 8.0, 1 mM EDTA, 120 mM NaCl, 10% glycerol, 0.5% Nonidet P-40, 1 mM DTT, and 0.5 mM PMSF), and the soluble fraction obtained by centrifugation at 21 000 g for 10 min at 4°C. GST fusion proteins binding to 15 μl of glutathione-sepharose 4B beads were eluted with 50 mM Tris–HCl (pH 8.0) and 20 mM reduced glutathione, according to the manufacturer's protocol (Pharmacia), and stored at −80°C.

Purification and analysis of nucleic acids

The cytosolic fraction of MCF7 cells was prepared as described previously (Yoshida et al., 2003). In all, 500 μg of the cytosol fraction was immunoprecipitated with 20 μl of protein A/G-agarose and 2 μg of antibodies, as indicated. RNA included in immunoprecipitates from MCF7 cells was isolated using Sepasol reagent (Nacalai Tesque, Kyoto, Japan). The purified RNA was labeled with [γ-32P] ATP using T4 Polynucleotide Kinase (Takara, Tokyo, Japan) in kinase reaction buffer (50 mM Tris-HCl pH 8.0, 10 mM MgCl2, 5 mM DTT) according to the manufacturer's protocol, resolved by acrylamide gel electrophoresis in TBE buffer and analysed using bioimaging analyzer BAS2000 (FJIX, Tokyo).

GST-fusion proteins expressed in bacteria were immobilized on 15 μl glutathione-sepharose 4B beads. Nucleic acid was isolated from the purified GST-fusion protein with phenol–chloroform and the resulting supernatant was purified by ethanol precipitation. The nucleic acid fraction, with or without RNase A or DNase I treatment, was resolved by agarose gel electrophoresis in TBE buffer.

In vitro phosphorylation

GST-fusion proteins were phosphorylated with CKII or PKC according to the manufacturer's protocol (Promega). In brief, GST-fusion proteins were incubated in 50 μl of kinase reaction buffer (CKII: 25 mM Tris-HCl pH 7.4, 10 mM MgCl2, 200 mM NaCl; PKC: 20 mM HEPES pH 7.4, 1.0 mM DTT, 10 mM MgCl2, 1.7 mM CaCl2), supplemented with 1 mM ATP or [γ-32P] ATP, in the presence or absence of CKII or PKC.

DNA-binding assay

DNA-binding assays were performed as described previously (Torigoe et al., 2003), with a few modifications. GST-fusion proteins (100 ng) immobilized on 15 μl glutathione-sepharose 4B beads were washed three times with binding buffer (50 mM Tris-HCl (pH 8.0), 1 mM EDTA, 120 mM NaCl, 10% glycerol, 0.5% Nonidet P-40, 1 mM DTT and 0.5 mM PMSF), and incubated with 1 × 106 cpm 32P-labeled oligonucleotide probes (50 ng) for 1 h at 4°C. The bound complexes were washed three times with binding buffer. 32P-labeled oligonucleotides recovered from the bound complex were isolated with phenol–chloroform and the supernatant was purified by ethanol precipitation. The oligonucleotides were analysed on nondenaturing 15% polyacrylamide gels in TBE buffer using a bioimaging analyzer BAS2000. The following double-stranded p53 consensus oligonucleotide (p53C), nuclear factor κB (NF-κB), and CRE were used in the DNA-binding assay (Okamoto et al., 2000; Izumi et al., 2003):

Co-immunoprecipitation

PC3 cells were seeded in six-well plates at a density of 1 × 105 cells/well. The following day, the cells were co-transfected with 1.5 μg of Flag-p53 and HA-p53 expression plasmids along with 6 μl of SuperFect, according to the manufacturer's protocol (Qiagen). At 3 h after transfection, the cells were washed with PBS, and the medium was replaced with fresh medium. After 48 h, the cells were washed twice with PBS and lysed in binding buffer (described above). After incubating for 30 min on ice, the lysates were centrifuged at 21 000 g for 10 min at 4°C. The supernatants (whole cell extract: 1 mg) were incubated with or without 5 μg/ml RNase A (Sigma) in binding buffer, and incubated for 2 h at 4°C with 20 μl of protein A/G-agarose and antibodies as indicated. Immunoprecipitates were washed three times with binding buffer, subjected to SDS–PAGE, analysed by immunoblotting using anti-Thio antibody, and visualized by chemiluminescence according to the ECL protocol (Amersham Pharmacia Biotech).

Pull-down assay

Pull-down assays were performed as described previously (Yoshida et al., 2003). Briefly, GST fusion proteins binding to 15 μl of glutathione-sepharose 4B beads in 50% slurry were mixed with ThioHis fusion proteins in binding buffer. The mixtures were incubated for 2 h at 4°C with gentle inversion, and washed three times with binding buffer. Pull-down samples representing 10% of the starting material were subjected to SDS-PAGE and developed as described above.

Glutaraldehyde crosslinking

After RNase A treatment, GST fusion proteins binding to 15 μl of glutathione-sepharose 4 B were eluted with 50 mM Tris–HCl (pH 8.0). and 20 mM reduced glutathione, according to the manufacturer's protocol (Pharmacia). GST fusion proteins were heated for 10 min at 37°C in SDS–PAGE loading buffer, and separated in a SDS 10% polyacrylamide slab gel. Gel strips containing the fractionated proteins were cut transversely with a razor blade as described previously (Yoshida et al., 2002). Each gel piece was then placed into a tube containing two volumes (per weight of the polyacrylamide slice) of elution - renaturation buffer (1% Triton X-100, 20 mM HEPES pH 7.6, 1 mM EDTA, 100 mM NaCl, 2 mM DTT, 0.1 mM PMSF) and homogenized with a small stainless steel pestle. After 24 h incubation period at 4°C, the residual polyacrylamide was sedimented by centrifugation, and the supernatant transferred to a new tube. The purified proteins were dialysed in dialysis buffer (50 mM Tris–HCl pH 7.5) with PlusOne Mini DialysisKit (Amersham). The purified proteins were crosslinked by addition of glutaraldehyde to a final concentration of 0.025% (Du and Maniatis, 1994), which contained 5% glycerol, 10 mM Tris–HCl (pH 7.5), 50 mM NaCl, 0.1 mM EDTA, 1 mM DTT for 5 min at room temperature. Glutaraldehyde crosslinking samples were subjected to SDS–PAGE and developed as described above.

References

  1. Ahn J and Prives C . (2001). Nat. Struct. Biol., 8, 730–732.

  2. Bakalkin G, Yakovleva T, Selivanova G, Magnusson KP, Szekely L, Kiseleva E, Klein G, Terenius L and Wiman KG . (1994). Proc. Natl. Acad. Sci. USA, 91, 413–417.

  3. Clore GM, Ernst J, Clubb R, Omichinski JG, Kennedy WM, Sakaguchi K, Appella E and Gronenborn AM . (1995). Nat. Struct. Biol., 2, 321–331.

  4. Clore GM, Omichinski JG, Sakaguchi K, Zambrano N, Sakamoto H, Appella E and Gronenborn AM . (1994). Science, 265, 386–391.

  5. Crook T, Marston NJ, Sara EA and Vousden KH . (1994). Cell, 79, 817–827.

  6. Du W and Maniatis T . (1994). Proc. Natl. Acad. Sci. USA, 91, 11318–11322.

  7. El-Deiry WS, Kern SE, Pietenpol JA, Kinzler KW and Vogelstein B . (1992). Nat. Genet., 1, 45–49.

  8. Espinosa JM and Emerson BM . (2001). Mol. Cell, 8, 57–69.

  9. Filhol O, Baudier J, Delphin C, Mackenbach PL, Chambaz EM and Cochet C . (1992). J. Biol. Chem., 267, 20577–20583.

  10. Fiscella M, Zambrano N, Ullrich SJ, Unger T, Lin D, Cho B, Mercer WE, Anderson CW and Appella E . (1994). Oncogene, 9, 3249–3257.

  11. Fontoura BM, Atienza CA, Sorokina EA, Morimoto T and Carroll RB . (1997). Mol. Cell Biol., 17, 3146–3154.

  12. Fontoura BM, Sorokina EA, David E and Carroll RB . (1992). Mol. Cell Biol., 12, 5145–5151.

  13. Fuchs B, O'Connor D, Fallis L, Scheidtmann KH and Lu X . (1995). Oncogene, 10, 789–793.

  14. Funk WD, Pak DT, Karas RH, Wright WE and Shay JW . (1992). Mol. Cell Biol., 12, 2866–2871.

  15. Guo A, Salomoni P, Lou J, Shih A, Zhong S, Gu W and Pandolfi PP . (2000). Nat. Cell Biol., 2, 730–736.

  16. Halazonetis TD, Davis LJ and Kandil AN . (1993). EMBO J., 12, 1021–1028.

  17. Hao M, Lowy AM, Kaopoor M, Deffie A, Liu G and Lozano G . (1996). J. Biol. Chem., 271, 29380–29385.

  18. Hupp TR and Lane DP . (1995). J. Biol. Chem., 270, 18165–18174.

  19. Hupp TR, Meek DW, Midgley CA and Lane DP . (1992). Cell, 71, 875–886.

  20. Hupp TR, Meek DW, Midgley CA and Lane DP . (1993). Nucleic Acids Res., 21, 3167–3174.

  21. Imamura T, Izumi H, Nagatani G, Ise T, Nomoto M, Iwamoto Y and Kohno K . (2001). J. Biol. Chem., 276, 7534–7540.

  22. Izumi H, Ohta R, Nagatani G, Ise T, Nakayama Y, Nomoto M and Kohn K . (2003). Biochem. J., 373, 713–722.

  23. Jayaraman J and Prives C . (1995). Cell, 81, 1021–1029.

  24. Kaeser MD and Iggo RD . (2002). Proc. Natl. Acad. Sci. USA, 99, 95–100.

  25. Keller DM, Zeng X, Wang Y, Zhang QH, Kapoor M, Shu H, Goodman R, Lozano G, Zhao Y and Lu H . (2001). Mol. Cell, 7, 283–292.

  26. Kulesz-Martin MF, Lisafeld B, Huang H, Kisiel ND and Lee L . (1994). Mol. Cell Biol., 14, 1698–1708.

  27. Lee S, Elenbaas B, Levine A and Griffith J . (1995). Cell, 81, 1013–1020.

  28. Lee W, Harvey TS, Yin Y, Yau P, Litchfield D and Arrowsmith CH . (1994). Nat. Struct. Biol., 1, 877–890.

  29. Levine AJ . (1997). Cell, 88, 323–331.

  30. Marechal V, Elenbaas B, Piette J, Nicolas JC and Levine AJ . (1994). Mol. Cell Biol., 14, 7414–7420.

  31. Marston NJ, Ludwig RL and Vousden KH . (1998). Oncogene, 16, 3123–3131.

  32. Meek DW, Simon S, Kikkawa U and Eckhart W . (1990). EMBO J., 9, 3253–3260.

  33. Miller SJ, Suthiphongchai T, Zambetti GP and Ewen ME . (2000). Mol. Cell Biol., 20, 8420–8431.

  34. Minamoto T, Buschmann T, Habelhah H, Matusevich E, Tahara H, Boerresen-Dale AL, Harris C, Sidransky D and Ronai Z . (2001). Oncogene, 20, 3341–3347.

  35. Mosner J, Mummenbrauer T, Bauer C, Sczakiel G, Grosse F and Deppert W . (1995). EMBO J., 14, 4442–4449.

  36. Okamoto T, Izumi H, Imamura T, Takano H, Ise T, Uchiumi T, Kuwano M and Kohno K . (2000). Oncogene, 19, 6194–6202.

  37. Sakaguchi K, Sakamoto H, Lewis MS, Anderson CW, Erickson JW, Appella E and Xie D . (1997). Biochemistry, 36, 10117–10124.

  38. Samad A and Carroll RB . (1991). Mol. Cell Biol., 11, 1598–1606.

  39. Torigoe T, Izumi H, Yoshida Y, Ishiguchi H, Okamoto T, Itoh H and Kohno K . (2003). Nucleic Acids Res., 31, 4523–4530.

  40. Wang Y, Reed M, Wang P, Stenger JE, Mayr G, Anderson ME, Schwedes JF and Tegtmeyer P . (1993). Genes Dev., 7, 2575–2586.

  41. Yakovleva T, Pramanik A, Terenius L, Ekstrom TJ and Bakalkin G . (2002). Trends Biochem. Sci., 27, 612–618.

  42. Yoshida Y, Izumi H, Ise T, Uramoto H, Torigoe T, Ishiguchi H, Murakami T, Tanabe M, Nakayama Y, Itoh H, Kasai H and Kohno K . (2002). Biochem. Biophys. Res. Commun., 295, 945–951.

  43. Yoshida Y, Izumi H, Torigoe T, Ishiguchi H, Itoh H, Kang D and Kohno K . (2003). Cancer Res., 63, 3729–3734.

Download references

Acknowledgements

This work was supported by MEXT. KAKENHI (13218132).

Author information

Correspondence to Kimitoshi Kohno.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Yoshida, Y., Izumi, H., Torigoe, T. et al. Binding of RNA to p53 regulates its oligomerization and DNA-binding activity. Oncogene 23, 4371–4379 (2004) doi:10.1038/sj.onc.1207583

Download citation

Keywords

  • p53
  • RNA
  • DNA binding
  • oligomerization
  • RNase

Further reading