Molecular Signaling Laboratory, Baker Medical Research Institute, Commercial Road, Prahran, Victoria, Australia

Correspondence to: Jun-Ping Liu, Molecular Signaling Laboratory, Baker Medical Research Institute, PO Box 6492, St. Kilda Road Central, Melbourne 8008, Victoria, Australia

The telomere DNA polymerase (telomerase) and the tumor suppressor protein p53 are frequently associated with human cancers, and activation of telomerase and inactivation of p53 involved in cancer cell immortalization. In this report, we demonstrate a direct interaction of telomerase with p53 in the nuclear lysates of human breast cancer cells, and with recombinant human p53, by affinity chromatography and immunoprecipitation. On activity criteria, the interaction is between the carboxyl-terminal region of p53 and a region close to the amino-terminus of human telomerase-associated protein 1 (hTEP1). Incubation of recombinant p53 with nuclear telomerase extracts results in inhibition of telomerase activity, with the C-terminal region of p53 being essential for inhibition. This effect is not mediated by binding to telomerase substrate DNA, but requires the region near the N-terminus of hTEP1, in that a synthetic peptide derived from this region of hTEP1 similarly inhibits telomerase activity. Together, these in vitro interactions between telomerase and p53 suggest that the activity of telomerase may be regulated by p53, down-regulation of which in turn would favor up-regulation of telomerase activity in cancer cell development.

telomerase reverse transcriptase (TERT); telomerase-associated protein 1 (TEP1); p53; telomerase inhibitory polypeptide 1 (TEIPP1); breast cancer cells

Introduction

Telomerase, a specialized RNA-directed DNA polymerase that extends telomeres of eukaryotic chromosomes, has been implicated as playing an important role in cell survival. The activity of telomerase is repressed in many human somatic tissues, whereas the enzyme is activated during tumor progression in most human cancers (Blackburn, 1992; de Lange, 1994, 1998; Greider, 1996, 1998; Harley et al., 1995; Kim et al., 1994; Sedivy, 1998; Shay and Bacchetti, 1997; Zakian, 1995). Although little is known of how telomerase is regulated in telomerase-positive cells, activation is thought to be involved in cellular replicative lifespan extension and cancer cell immortalization (Bodnar et al., 1998; Counter et al., 1998; Kondo et al., 1998; Meyerson et al., 1997; Nakamura et al., 1997; Nakayama et al., 1998; Vaziri and Benchimol, 1998). Telomerase is a ribonucleoprotein complex containing an RNA subunit and several protein components. The RNA subunit contains the template for telomeric DNA addition and is essential for telomerase activity (Blasco et al., 1996; Broccoli et al., 1996; Feng et al., 1995; Greider and Blackburn, 1989; Kondo et al., 1998; Lee et al., 1998; Niida et al., 1998; Shippen-Lentz and Blackburn, 1990; Tsao et al., 1998). The protein subunits of telomerase include telomerase reverse transcriptase (TERT) and telomerase associated protein 1 (TEP1). TERT is the catalytic subunit of telomerase (Counter et al., 1997; Harrington et al., 1997b; Kilian et al., 1997; Lingner and Cech, 1996; Lingner et al., 1997; Meyerson et al., 1997; Nakamura et al., 1997; Nakayama et al., 1998), whereas TEP1 has been shown to co-purify with telomerase activity and bind to both telomerase RNA and TERT (Harrington et al., 1997a,b; Nakayama et al., 1997).

Of the telomerase subunits, TERT is concomitantly expressed with the activation of telomerase during cellular immortalization and tumor progression (Harrington et al., 1997b; Kilian et al., 1997; Meyerson et al., 1997; Nakamura et al., 1997; Nakayama et al., 1998). Introduction of TERT into telomerase negative cells leads to expression of telomerase activity (Bodnar et al., 1998; Counter et al., 1998; Nakayama et al., 1998; Vaziri and Benchimol, 1998; Wen et al., 1998), elongate telomeres (Bodnar et al., 1998; Vaziri and Benchimol, 1998), and extend cellular lifespans (Bodnar et al., 1998; Counter et al., 1998; Vaziri and Benchimol, 1998). In vitro reconstitution of telomerase also shows that TERT and the telomerase RNA subunit constitute a minimum core structure of telomerase (Beattie et al., 1998; Greenberg et al., 1998; Weinrich et al., 1997). Moreover, recent studies suggest that proto-oncogene c-myc induces expression of TERT and telomerase activity in both normal human mammary epithelial cells and normal human diploid fibroblasts (Takakura et al., 1999; Wang et al., 1998; Wu et al., 1999), and that anti-c-myc antisense treatment causes inhibition of telomerase activity in three human leukemic cell lines (Fujimoto and Takahashi, 1997). Thus, de novo activation of TERT gene expression is a first rate-limiting step in telomerase activation in cancer.

To date, little is known of post-translational mechanisms of telomerase regulation. Recent studies show that alteration of the C-terminal structure of hTERT does not affect telomerase enzymatic activity but impedes the ability of this enzyme to maintain telomeres and induce cell immortalization (Counter et al., 1998; Ouellette et al., 1999). These data suggest that the catalytic activity of hTERT does not always correlate with telomerase function in cells, and some structures of hTERT other than those required for the catalytic activity are also essential for telomerase function in vivo probably through interactions with other molecules. Identifications of telomerase interacting proteins and roles of other protein components of telomerase complex such as hTEP1 (Harrington et al., 1997a; Nakayama et al., 1997) and Tetrahymena p95 (Collins et al., 1995) in the regulation of telomerase activity are therefore of importance in our understandings of the control mechanisms of telomerase activity in human cancer.

Studies in our laboratory have shown that activation of telomerase requires phosphorylation of telomerase protein subunits. Both TERT and TEP1 are phosphoproteins in human breast cancer cells, and telomerase exists in two configurations with TERT and TEP1 being phosphorylated/dephosphorylated (Li et al., 1997, 1998). Phosphorylation is associated with high telomerase activity, and dephosphorylation with low. Protein phosphatase 2A and protein kinase C have been shown to be the phosphatase and kinase involved in regulating telomerase activity in human breast cancer cells (Li et al., 1997, 1998). To further delineate the post-translational mechanisms of telomerase regulation, the present study was undertaken to characterize the potential role of the tumor suppressor protein p53 in the regulation of telomerase activity. We show by affinity chromatography and immunoprecipitation that p53 binds to TEP1 and inhibits telomerase activity. This inhibition requires an intact C-terminal region of the tumor suppressor, and can in turn be blocked by a synthetic peptide derived from a region near the N-terminus of TEP1. Thus, p53 and telomerase may interact directly in the regulation of cell survival.

Results

To identify potential telomerase regulatory proteins, we set up affinity chromatography columns with synthetic peptides derived from the sequences of human telomerase-associated protein 1 (hTEP1) and of human telomerase reverse transcriptase (hTERT). These peptides were selected on the basis of prediction of sequence hydrophilicity and charge characteristics. Elution of nuclear proteins from human breast cancer (PMC42) cells in three hTEP1 peptide affinity columns showed p53 tumor suppressor protein reactivity. As shown in Figure 1a, p53 from nuclear protein lysates bound selectively to an affinity column coupled with hTEP1 peptide (Trp943-Cys959); the tumor suppressor protein Rb did not bind to the column (not shown). The binding appeared to be of high affinity with dissociation requiring high concentrations of salt or low pH. Nuclear p53 also bound to affinity columns coupled with the hTEP1 peptide (His385-Gly399) or hTEP1 peptide (Asp2538-Arg2551), but not to columns coupled with scrambled peptides or three peptides from human telomerase reverse transcriptase (hTERT) (data not shown). Given that all of the three peptides have previously been suggested to be able to enrich hTEP1 and telomerase activity, these data suggest specific complex interactions possibly involving hTEP1 oligomers (Li et al., 1998). To further explore the specificity of the interaction between p53 and the hTEP1 peptide affinity column, nuclear protein lysates were mixed with the peptide hTEP1^943-959 before loading onto the same peptide affinity column. In the presence of excess hTEP1 peptide, binding of p53 to the column was inhibited, with much lesser amounts eluting in the low pH glycine buffer (Figure 1b).

To determine whether or not p53 directly interacts with hTEP1, either purified GST or GST-p53 fusion proteins were loaded onto the peptide affinity column coupled with the peptide hTEP1^943-959, followed by sequential elution with high salt and low pH buffer. As shown in Figure 1c, GST did not bind to the column, whereas GST-p53 bound to the column and was eluted by low pH glycine buffer (Figure 1d, indicated by arrow). The low molecular weight band associated with GST-p53 in lanes P and 18 - 26 of Figure 1d is likely to be a degraded fragment of GST-p53. When GST-p53 was mixed with soluble hTEP1^943-959 peptide and then loaded onto the column in the presence of this peptide, the binding of GST-p53 fusion to the column was markedly inhibited with the fusion protein mainly in the flow through and wash fractions (not shown). These data suggest that p53 directly and specifically interacts with the hTEP1 peptide.

To determine the potential interaction between p53 and intact hTEP1 protein in human breast cancer cells, proteins were immunoprecipitated from nuclear protein lysates with specific antibodies against either p53 or hTEP1. Although hTEP1 could be readily detected by immunoblot analysis of anti-hTEP1 immunoprecipitates (Li et al., 1998), no hTEP1 was detectable in immunoprecipitates of p53 (data not shown); in contrast, however, significant amounts of p53 (20 - 35% of total) were found in the immunoprecipitates of hTEP1 (Figure 1e). This suggests that only a fraction of p53 was associated with hTEP1, and that those hTEP1 bound p53 molecules were inaccessible to p53 antibody in the immunoprecipitation studies (or were relatively weekly associated with the large telomerase complex). Addition of recombinant wild-type GST-p53 fusion protein into cell lysates before immunoprecipitation also showed co-immunoprecipitation of the fusion protein with hTEP1 (Figure 1e, lane 6). The association of both endogenous and exogenous p53 proteins with hTEP1 also appeared to be specific; addition of the hTEP1 peptide used for antibody generation blocked the immunoprecipitation of hTEP1 (not shown) and thereby co-immunoprecipitation of p53 (Figure 1e, lanes 5 and 7). The intermediate-sized bands between GST-53 and nuclear p53 in Figure 1e are likely to be degraded GST-p53 fragments. Thus, given that hTEP1 co-immunoprecipitates with hTERT and telomerase activity (Harrington et al., 1997a; Li et al., 1998; Nakayama et al., 1997), these data demonstrate that a subpopulation of p53 is physically in association with the ribonucleoprotein complex of telomerase, and that association is likely to be mediated at least in part by hTEP1 possibly in cultured human breast cancer PMC42 cells.

Since it is yet to be determined of potential mutations in the amino acid sequence of p53 expressed in the PMC42 breast cancer cells, using three different GST-p53 mutants we investigated the potential regions of p53 involved in association with telomerase by immunoprecipitation. While the GST-p53 mutants with a single amino acid residue mutation in the DNA binding domain (R175H) (not shown) or with a truncation in the amino terminus (2C) retained binding activity for telomerase (Figure 1e, lane 8), the GST-p53 mutant with a deletion of the C-terminus (N5) was less effectively immunoprecipitated relative to nuclear p53 compared with GST-p53 wild-type and GST-p53 mutant (2C) (Figure 1e, lane 10 versus lanes 6 and 8), suggesting an importance of the C-terminal region of p53 in telomerase regulation.

To determine roles for the binding of p53 to telomerase, nuclear telomerase activity was assessed as a function of differential incubation with different recombinant GST-p53 proteins. Incubation with wild-type GST-p53 produced marked inhibition of telomerase activity, with GST itself having no effect (Figure 2a). Inhibition occurred immediately after GST-p53 was included (not shown), with an IC₅₀ for GST-p53 of ~40 nM. Since such inhibition might reflect non-specific DNA binding of the tumor suppressor, following incubation with telomerase in the presence or absence of GST-p53 de novo synthesized telomere DNA was extracted for specific PCR. As shown in Figure 2b, telomerase inhibition by GST-p53 was confirmed not to be due to inhibiting PCR, and thus specific. In separate experiments, increasing concentrations of the telomerase substrate (TS) oligonucleotide were added to the telomerase activity incubates in the presence or absence of GST-p53 (Figure 2c). Under these conditions, inhibition of telomerase by GST-p53 was similarly not affected, suggesting that the inhibition of telomerase by p53 is not mediated by an interaction of p53 with TS oligonucleotide.

To explore the structure-function relationships of GST-p53 in regulating telomerase activity, different GST-p53 mutants were tested in the telomerase activity assay. Consistent with the data shown in Figure 2b,c, a single amino acid mutation in the DNA binding domain (R175H) did not alter the inhibitory effect of p53. N-terminal truncation only slightly attenuated the inhibitory effect of p53 compared with wild-type, whereas deletion of the C-terminal region (Figure 2d) or both C- and N-terminal regions (not shown) abrogated the inhibitory effect of GST-p53. This is consistent with the finding from the immunoprecipitation studies that the C-terminal region is probably involved in telomerase association, and adds weight to the possibility that the interaction mediated through the C-terminal region of p53 is important in p53 regulation of telomerase activity.

Since only a small amount of endogenous p53 co-immunoprecipitates with hTEP1, and added extra amounts of GST-p53 are able to co-immunoprecipitate with hTEP1 and inhibit telomerase activity, we hypothesized that phosphorylation of endogenous p53 might regulate its interaction with telomerase. To test this hypothesis, wild-type GST-p53 was stoichiometrically phosphorylated on glutathione sepharose 4B beads in the presence of [-³²P]ATP by purified cdc2 protein kinase, protein kinase C and casein kinase II, respectively. These protein kinases have previously been shown to phosphorylate p53 in the C-terminal region (reviewed by Ko and Prives, 1996). When phosphorylation reached to plateaus, the differentially phosphorylated GST-p53 proteins were extensively washed and eluted, and their effects on telomerase activity tested. Neither cdc2, protein kinase C (Figure 2e) nor casein kinase II (not shown) affected the inhibitory activity of GST-p53 on telomerase, suggesting that phosphorylation of p53 by these protein kinases plays little role in p53 inhibition of telomerase activity in vitro.

Alternatively, the inhibition of telomerase activity by GST-p53 could reflect modulation of telomerase tertiary/quaternary structures affecting intramolecular allosteric reactions. We therefore tested several synthetic peptides derived from hTEP1 and hTERT for their potential effects on both basal and p53-inhibited telomerase activities from nuclei of human breast cancer cells. While none of the three peptides from hTERT, nor the two hTEP1^943-959 and hTEP1^2538-2551 peptides used for affinity chromatography had any effect (not shown), the peptide hTEP1^385-399 clearly interacted with telomerase and GST-p53 (Figure 3). This peptide, termed telomerase inhibitory polypeptide 1 (TEIPP1), was capable of inhibiting telomerase activity in a concentration-dependent manner, and scrambling TEIPP1 markedly lowered its inhibitory activity (Figure 3a). In addition, co-incubation of TEIPP1 at varying concentrations with telomerase extracts and GST-p53 showed that TEIPP1 titrated p53 away from the telomerase complex, resulting in a restoration of telomerase activity at fixed concentrations of TEIPP1 (Figure 3b,c). Near complete restoration of p53 inhibition was seen at ~1 M TEIPP1; at higher concentrations of TEIPP1, however, the peptide itself showed inhibitory effects on telomerase activity (Figure 3b).

Discussion

The p53 tumor suppressor protein is activated by DNA damage; activation then arrests chromosome replication allowing DNA repair (for reviews, see Levine, 1997; Oren, 1998). Inactivation of p53 through specific interactions with oncogenic proteins, dephosphorylation or genetic mutations (Levine, 1997; Oren, 1998) is involved in cancer cell immortalization (Atadja et al., 1995; Bond et al., 1994; Gollahon et al., 1998; Rogan et al., 1995; Roos et al., 1998; Serrano et al., 1997; Wynford-Thomas, 1996). Similarly, activation of telomerase also plays an important role in tumor cell immortalization via preventing telomeres from shortening (Bodnar et al., 1998; de Lange, 1998; Greider, 1998; Harley et al., 1995; Kondo et al., 1998; Sedivy, 1998; Shay and Bacchetti, 1997), although there are scant data on the regulation of telomerase at molecular levels. The present study using affinity isolation and recombinant proteins suggests for the first time a possibly direct link between p53 and telomerase and a potential pathway whereby p53 and telomerase interact in the regulation of cellular senescence.

Our results show clearly that both nuclear p53 of human breast cancer cells and recombinant wild-type p53 interact directly with hTEP1 and thus presumably the telomerase complex in vitro. The binding of p53 to the affinity chromatography columns coupled with three different hTEP1 peptides suggests that p53 interacts with the telomerase complex during chromatography, as both immunoreactive hTEP1 and telomerase activity have been specifically eluted from the columns (Li et al., 1998). This interpretation is also supported by the findings that p53 (Figure 1e) and telomerase activity (Harrington et al., 1997a; Li et al., 1998) co-immunoprecipitate with hTEP1, and that higher amounts of recombinant p53 specifically inhibit telomerase activity in vitro (Figure 2). The specificity of p53 inhibition of telomerase activity was demonstrated by excluding potential non-specific interactions with PCR reagents (Figure 2b) and telomerase substrate oligonucleotide primer (Figure 2c). Thus, it is possible that a fraction of nuclear p53 may interact with the telomerase complex and thus play an inhibitory role in the regulation of telomerase activity in human breast cancer cells.

Structure-function analysis reveals that the C-terminal region of p53 is important in regulating telomerase activity. Since this region has been shown to be able to bind RNA and to be phosphorylated by cdc2 protein kinase, protein kinase C and casein kinase II (Ko and Prives, 1996), it is possible that p53 regulation of telomerase activity might be mediated by interacting with the telomerase RNA moiety and regulated by protein phosphorylation. However, in vitro phosphorylation of recombinant p53 by either cdc2 protein kinase, protein kinase C or casein kinase II elicited no change in p53 inhibition of telomerase, suggesting that p53 phosphorylation by the protein kinases tested plays no role in the p53 regulation of telomerase activity, although our data do not exclude the possibility that other sites are involved. Further study is also required to determine if p53 might interact directly with the telomerase RNA subunit.

In exploring the possible mechanism of p53 inhibition of telomerase activity through interacting with hTEP1, however, we have shown that while the hTEP1^943-959 and hTEP1^2538-2551 peptides have no effect on either basal or p53 inhibited telomerase activity, the peptide hTEP1^{385 - 399} reverses p53 inhibition of telomerase at low dose and inhibits telomerase activity itself at high dose (Figure 3b,c). The peptide hTEP1^385-399 may thus prove very useful in the study of telomerase structure and function, and for convenience we have called it telomerase inhibitory polypeptide 1 or TEIPP1. These data are consistent with an intimate, possibly direct involvement of a sequence including hTEP1^385-399, or TEIPP1 itself, in mediating p53 inhibition of telomerase activity. In addition, the inhibitory effect of TEIPP1 on p53 inhibition of telomerase provides further evidence that the inhibition of telomerase activity by p53 is not non-specific, but rather a specific protein interaction with hTEP1. Furthermore, the effects of TEIPP1 on both telomerase and p53 activities also confirms a regulatory role of hTEP1 in telomerase activity, and provide a novel insight into the inter- and intramolecular control mechanisms of telomerase.

Given that hTEP1 is associated with telomerase activity (Li et al., 1998) and p53 (Figure 1) and that additional TEIPP1, the sequence near the N-terminus of hTEP1 antagonizes the inhibition of telomerase activity by exogenously added p53, it is possible that hTEP1, TERT, telomerase RNA, and p53 form a complex and in the presence of excess amounts of TEIPP1 or p53 telomerase activity is then inhibited, although the possibility cannot be excluded of mutually exclusive binding of hTEP1 to p53 or TERT. As depicted in Figure 4, TEIPP1 may play an essential role in forming and maintaining an active telomerase configuration through interacting with another region of the telomerase complex. p53 may interact with this sequence to disrupt telomerase activity, whereas additional TEIPP1 may competitively bind elsewhere, perhaps to inhibit the correct folding of telomerase and leading to inhibition of telomerase activity in a manner similar to that seen with p53 (Figure 4). When p53, TEIPP1 and telomerase are all together, TEIPP1 may competitively bind to the tumor suppressor to block its inhibition on telomerase, and at higher concentrations inhibit telomerase activity by interacting with different regions of the telomerase complex (Figure 4).

Taken together, the data that p53 binds to and inhibits telomerase in vitro argue strongly in favor of the possibility that p53 is an important regulatory factor modulating activity of telomerase in the nucleus of human breast cancer cells. It is possible that among potentially multiple factors, p53 modulates telomerase activity post-translationally through interacting with hTEP1 following TERT and telomerase RNA expression and holoenzyme assembly. Previous studies have shown that human papillomavirus type 16 E6 protein that targets p53 for degradation activates telomerase in a p53-independent manner in early pre-crisis passage human keratinocytes and mammary epithelial cells (Klingelhutz et al., 1996). In addition, occasional normal human mammary epithelial cells transfected with a p53 mutant become immortalized and upon additional genetic events reactivated for telomerase (Gollahon and Shay, 1996). These studies support the idea that p53 loss of function alone is inefficient in activating telomerase and other factors are also involved in telomerase activity. Given that p53 function is lost in ~50% (Levine, 1997; Oren, 1998) and telomerase activated in ~85% (Kim et al., 1994; Shay and Bacchetti, 1997) of human cancers, and the direct in vitro interactions between telomerase and p53, whether or not down-regulation of p53 may be frequently involved in up-regulating telomerase activity in vivo during cancer development requires further investigation. Identification of the peptide sequences of p53 and hTEP1 involved in their mutual interaction and with TEIPP1 in telomerase inhibition may provide both a testable system for investigating molecular interactions and the tertiary structures of telomerase proteins, and a potential therapeutic avenue for altering telomerase activity in anti-cancer therapy.

Materials and methods

Cells, peptides and affinity chromatography

The human mammary carcinoma cell line, PMC42, was grown in vitro in a hormone-dependent manner (Whitehead et al., 1984) and lyzed for nuclear extracts before affinity chromatography (Li et al., 1997). Several peptides corresponding to regions of human TEP1 (hTEP1) and human TERT (hTERT) were synthesized and purified to >70% purity by Chiron Technologies Pty. Ltd. (Victoria, Australia). The peptides were coupled onto Affigel-10/15 beads as described elsewhere (Li et al., 1998) to isolate potential regulatory proteins of telomerase. The three peptides used from hTEP1 were ³⁸⁵HRAKRHPRRPPRSPG³⁹⁹, ⁹⁴³WGVTEEETRRNRQLEVC⁹⁵⁹, and ²⁵³⁸DSEPTPHLKTRQRR²⁵⁵¹. Approximately 10 mg protein of nuclear extract from human breast cancer PMC42 cells, or 10 g purified protein, were loaded onto the column coupled (bed volume, 10 ml) at a flow rate of ~0.5 ml/min. Following extensive washing (5´bed volumes), proteins were eluted from the column with KCl (0 - 300 mM) and then glycine buffer (0.1 mM, pH 2.4). The collected fractions were subject to SDS - PAGE followed by Coomassie blue staining or immunoblotting using specific antibodies. Alternatively, in some experiments nuclear extracts or purified GST-fusion protein was mixed with synthetic peptide before loading the sample onto the affinity column coupled with the same peptide.

Antibodies, immunoprecipitation and immunoblotting

Anti-hTEP1 antibody was produced by immunizing rabbits with the hTEP1 peptide ²⁵³⁸DSEPTPHLKTRQRR²⁵⁵¹ coupled to BSA and then affinity-purified on the same peptide-coupled column (Li et al., 1998). The antibody has been shown to be specific in immunoblotting, immunoprecipitation (Li et al., 1998) and immunocytochemical studies (unpublished). Anti-p53 and anti-Rb antibodies were from Santa Cruz Biotechnology, Inc., CA, USA. The anti-p53 antibody (DO-1) was covalently coupled to agarose for immunoprecipitation, and the anti-p53 antibody to the central DNA binding domain (Pab240) was used for immunoblotting. To determine interactions between hTEP1 and endogenous as well as exogenous p53, equal amounts of nuclear proteins (100 g) plus or minus different GST-p53 fusion proteins (1 g) in the presence or absence of the immunogenic hTEP1 peptide were subjected to immunoprecipitation with either normal IgG as a control, anti-p53 antibodies, or anti-hTEP1 antibody. Proteins in the immunoprecipitates were separated by SDS - PAGE, transferred onto nitrocellulose membranes, and probed with either anti-hTEP1, anti-p53 or anti-Rb antibodies, followed by ECL detection (Amersham) and autoradiography.

Recombinant GST-fusions, protein phosphorylation and purification

Wild-type and various mutants of GST-p53 were produced as described previously (Ruppert and Stillman, 1993). Briefly, for wild-type and the mutants with deletion of the N-terminal regions (2C: 95 - 393 and 3C: 155 - 393) or a single DNA-binding domain mutation (R175H), pET11GTK plasmids containing relevant p53 cDNAs (kindly provided by Dr Bruce Stillman) were amplified and purified. For the C-terminal deletion (N5: 2 - 293), a DNA fragment was generated by specific PCR using a wild-type human p53 cDNA in pC53-SN₃ plasmid as template and the synthetic oligonucleotides (gcaggtaccggaggagccgcag (jp33) and taggatccccctttcttgcgg (jp36)) as primers. It was then subcloned into the pET11GTK plasmid and the recombinant plasmid DNA was amplified and then purified. For double N- and C-terminal deletions, the GST-p53 mutant cDNA was constructed by subcloning a PCR cDNA fragment into the pET11GTK plasmid with the synthetic oligonucleotides (gctggtaccgtcttctgtccct (jp35)) and jp36 as primers. The recombinant plasmid DNA was then amplified and purified as above. After sequence verification, the cloned plasmids were individually transformed into E. coli BL21 (DE3) to produce the fusion proteins under IPTG induction. Recombinant GST-p53 fusion proteins were purified and identified by SDS - PAGE as described elsewhere (Ruppert and Stillman, 1993). For phosphorylation of the recombinant proteins, GST and GST fusions on glutathione sepharose 4B beads were incubated with purified cdc2 protein kinase, protein kinase C and casein kinase II separately in phosphorylation buffer containing ATP (~40 M, 3 Ci [-³²P]ATP), Mg²⁺ (1 mM) and other kinase specific activators, and the reaction allowed to proceed at 30°C for ~20 min followed by extensive washing of beads with 0.5 M NaCl and cold PBS. All recombinant GST fusions were eluted from the beads in reduced glutathione buffer and examined for their actions with reduced glutathione solution as control. Purified cdc2 protein kinase was from Upstate Biotechnology, Inc. (New York, NY, USA), protein kinase C was from Biomol Research Laboratories (Plymouth Meeting, PA, USA) and casein kinase II from Promega Corporation (Madison, WI, USA).

Telomerase activity assay

A TRAP assay, performed essentially as described previously (Kim et al., 1994), was employed to determine effects of p53 on telomerase activity. Briefly, equal amounts of nuclear telomerase extracts (1 g) were incubated in the presence or absence of recombinant proteins or/and synthetic peptides as indicated in particular experiments. Following incubation, de novo synthesized telomeres were amplified by specific PCR and the resultant ³²P-labeled telomeres resolved by polyacrylamide slab gel electrophoresis followed by autoradiography. To exclude any potential effect of proteins on PCR, de novo synthesized telomeres were extracted with phenol and chloroform post-incubation with telomerase and the isolated telomeres used in PCR amplification. To verify the specificity of telomerase, negative controls consisting either of added RNase-A or sample heated to 80°C to inactivate telomerase were included in each telomerase activity assay. To establish the specificity of actions of GST-p53 proteins and synthetic peptides, GST alone, relevant mutated proteins and scrambled peptides were compared as indicated in individual experiments.

Acknowledgements

We thank Drs Arnold J Levine and Bruce Stillman for kindly providing p53 cDNAs, Dr Roger Reddel for discussion and Ms Linlin Zhao for technical assistance. This work was supported by grants from the Australia Research Council and National Health and Medical Research Council of Australia.

Atadja P, Wong H, Garkavtsev I, Veillette C and Riabowol K. (1995). Proc. Natl. Acad. Sci. USA 92, 8348-8352. MEDLINE

Beattie TL, Zhou W, Robinson MO and Harrington L. (1998). Curr. Biol. 8, 177-180. MEDLINE

Blackburn EH. (1992). Ann. Rev. Biochem. 61, 113-129. MEDLINE

Blasco MA, Rizen M, Greider CW and Hanahan D. (1996). Nature Genetics 12, 200-204. MEDLINE

Bodnar AG, Ouellette M, Frolkis M, Holt SE, Chiu C-P, Morin GB, Harley CB, Shay JW, Lichtsteiner S and Wright WE. (1998). Science 279, 349-352. Article MEDLINE

Bond JA, Wyllie FS and Wynford-Thomas D. (1994). Oncogene 9, 1885-1889. MEDLINE

Broccoli D, Godley LA, Donehower LA, Varmus HE and de Lange T. (1996). Mol. Cell. Biol. 16, 3765-3772. MEDLINE

Collins K, Kobayashi R and Greider CW. (1995). Cell 81, 677-686. MEDLINE

Counter CM, Hahn WC, Wei W, Caddle SD, Beijersbergen RL, Lansdorp PM, Sedivy JM and Weinberg RA. (1998). Proc. Natl. Acad. Sci. USA 95, 14723-14728. Article MEDLINE

Counter CM, Meyerson M, Eaton EN and Weinberg RA. (1997). Proc. Natl. Acad. Sci. USA 94, 9202-9207. Article MEDLINE

de Lange T. (1994). Proc. Natl. Acad. Sci. USA 91, 2882-2885. MEDLINE

de Lange T. (1998). Nature 392, 753-754. Article MEDLINE

Feng J, Funk WD, Wang SS, Weinrich SL, Avilion AA, Chiu CP, Adams RR, Chang E, Allsopp RC, Yu J, Le S, West MD, Harley CB, Andrews WH, Greider CW and Villeponteau B. (1995). Science 269, 1236-1241. MEDLINE

Fujimoto K and Takahashi M. (1997). Biochem. Biophys. Res. Commun. 241, 775-781. MEDLINE

Gollahon LS, Kraus E, Wu TA, Yim SO, Strong LC, Shay JW and Tainsky MA. (1998). Oncogene 17, 709-717. MEDLINE

Gollahon LS and Shay JW. (1996). Oncogene 12, 715-725. MEDLINE

Greenberg RA, Allsopp RC, Chin L, Morin GB and DePinho RA. (1998). Oncogene 16, 1723-1730. MEDLINE

Greider CW. (1996). Annual Review of Biochemistry 65, 337-365. MEDLINE

Greider CW. (1998). Curr. Biol. 8, R178-R181. MEDLINE

Greider CW and Blackburn EH. (1989). Nature 337, 331-337. MEDLINE

Harley CB, Kim NW, Prowse KR, Weinrich SL, Hirsch K, West MD, Bacchetti S, Hirte HW, Counter CM, Greider CW, Wright WE and Shay JM. (1995). Cold Spring Harbor Symp. Quant. Biol. 59, 307-315.

Harrington L, McPhail T, Mar V, Zhou W, Oulton R, Bass MB, Arruda I and Robinson MO. (1997a). Science 275, 973-977. Article MEDLINE

Harrington L, Zhou W, McPhail T, Oulton R, Yeung DS, Mar V, Bass MB and Robinson MO. (1997b). Genes Dev. 11, 3109-3115. MEDLINE

Kilian A, Bowtell DD, Abud HE, Hime GR, Venter DJ, Keese PK, Duncan EL, Reddel RR and Jefferson RA. (1997). Hum. Mol. Genet. 6, 2011-2019. Article MEDLINE

Kim NW, Piatyszek MA, Prowse KR, Harley CB, West MD, Ho PL, Coviello GM, Wright WE, Weinrich SL and Shay JW. (1994). Science 266, 2011-2015. MEDLINE

Klingelhutz AJ, Foster SA and McDougall JK. (1996). Nature 380, 79-82. MEDLINE

Ko LJ and Prives C. (1996). Genes Dev. 10, 1054-1072. MEDLINE

Kondo S, Tanaka Y, Kondo Y, Hitomi M, Barnett GH, Ishizaka Y, Liu J, Haqqi T, Nishiyama A, Villeponteau B, Cowell JK and Barna BP. (1998). FASEB J. 12, 801-811. MEDLINE

Lee HW, Blasco MA, Gottlieb GJ, Horner II WJ, Greider CW and DePinho RA. (1998). Nature 392, 569-574. Article MEDLINE

Levine AJ. (1997). Cell 88, 323-331. MEDLINE

Li H, Zhao LL, Funder JW and Liu J-P. (1997). J. Biol. Chem. 272, 16729-16732. MEDLINE

Li H, Zhao L, Yang Z, Funder JW and Liu J-P. (1998). J. Biol. Chem. 273, 33436-33442. MEDLINE

Lingner J and Cech TR. (1996). Proc. Natl. Acad. Sci. USA 93, 10712-10717. MEDLINE

Lingner J, Hughes TR, Shevchenko A, Mann M, Lundblad V and Cech TR. (1997). Science 276, 561-567. Article MEDLINE

Meyerson M, Counter CM, Eaton EN, Ellisen LW, Steiner P, Caddle SD, Ziaugra L, Beijersbergen RL, Davidoff MJ, Liu Q, Bacchetti S, Haber DA and Weinberg RA. (1997). Cell 90, 785-795. MEDLINE

Nakamura TM, Morin GB, Chapman KB, Weinrich SL, Andrews WH, Lingner J, Harley CB and Cech TR. (1997). Science 277, 955-959. Article MEDLINE

Nakayama J, Saito M, Nakamura H, Matsuura A and Ishikawa F. (1997). Cell 88, 875-884. MEDLINE

Nakayama J, Tahara H, Tahara E, Saito M, Ito K, Nakamura H, Nakanishi T, Tahara E, Ide T and Ishikawa F. (1998). Nat. Genet. 18, 65-68. MEDLINE

Niida H, Matsumoto T, Satoh H, Shiwa M, Tokutake Y, Furuichi Y and Shinkai Y. (1998). Nat. Genet. 19, 203-206. Article MEDLINE

Oren M. (1998). Nature 391, 233-234. Article MEDLINE

Ouellette MM, Aisner DL, Savre-Train I, Wright WE and Shay JW. (1999). Biochem. Biophys. Res. Commun. 254, 795-803. MEDLINE

Rogan EM, Bryan TM, Hukku B, Maclean K, Chang AC, Moy EL, Englezou A, Warneford SG, Dalla-Pozza L and Reddel RR. (1995). Mol. Cell. Biol. 15, 4745-4753. MEDLINE

Roos G, Nilsson P, Cajander S, Nielsen NH, Arnerlov C and Landberg G. (1998). Int. J. Cancer 79, 343-348. Article MEDLINE

Ruppert JM and Stillman B. (1993). Mol. Cell. Biol. 13, 3811-3820. MEDLINE

Sedivy JM. (1998). Proc. Natl. Acad. Sci. USA 95, 9078-9081. Article MEDLINE

Serrano M, Lin AW, McCurrach ME, Beach D and Lowe SW. (1997). Cell 88, 593-602. Article MEDLINE

Shay JW and Bacchetti S. (1997). Eur. J. Cancer 33, 787-791. Article MEDLINE

Shippen-Lentz D and Blackburn EH. (1990). Science 247, 546-552. MEDLINE

Takakura M, Kyo S, Kanaya T, Hirano H, Takeda J, Yutsudo M and Inoue M. (1999). Cancer Res. 59, 551-557. MEDLINE

Tsao DA, Wu CW and Lin YS. (1998). Gene 221, 51-58. MEDLINE

Vaziri H and Benchimol S. (1998). Curr. Biol. 8, 279-282. MEDLINE

Wang J, Xie LY, Allan S, Beach D and Hannon GJ. (1998). Genes Dev. 12, 1769-1774. MEDLINE

Weinrich SL, Pruzan R, Ma L, Ouellette M, Tesmer VM, Holt SE, Bodnar AG, Lichtsteiner S, Kim NW, Trager JB, Taylor RD, Carlos R, Andrews WH, Wright WE, Shay JW, Harley CB and Morin GB. (1997). Nat. Genet. 17, 498-502. MEDLINE

Wen J, Cong YS and Bacchetti S. (1998). Hum. Mol. Genet. 7, 1137-1141. MEDLINE

Whitehead RH, Quirk SJ, Vitali AA, Funder JW, Sutherland RL and Murphy LC. (1984). J. Natl. Cancer Inst. 73, 643-648. MEDLINE

Wu KJ, Grandori C, Amacker M, Simon-Vermot N, Polack A, Lingner J and Dalla-Favera R. (1999). Nat. Genet. 21, 220-224. Article MEDLINE

Wynford-Thomas D. (1996). J. Pathol. 180, 118-121. MEDLINE

Zakian VA. (1995). Science 270, 1601-1607. MEDLINE

Figure 1 Identification of p53 interaction with telomerase hTEP1. (a, b) Binding of p53 from nuclear lysates of human breast cancer cells to a telomerase peptide affinity chromatography column. Nuclear protein extracts were loaded onto an Affigel-10/15 column coupled with hTEP1^{943 - 959} peptide in the presence (b) or absence (a) of the same hTEP1 peptide, and eluted with a linear KCl gradient and then low pH glycine buffer. p53 was detected by immunoblotting. P represents the load onto the column, T the flow-through fraction and W the wash fraction. (c, d) Elution profiles for GST and GST-p53 fusion proteins from the telomerase peptide affinity chromatography column. Equal amounts (10 g) of GST (c) or GST-p53 (d) were loaded onto the hTEP1^{943 - 959} peptide affinity column and eluted with KCl and glycine buffer as above. The proteins eluted from the column were detected by silver staining. (e) Binding of nuclear p53 and GST-p53 to hTEP1 as assessed by immunoprecipitation analysis. Nuclear protein extracts were mixed without (lanes 1 - 4), or with wild-type GST-p53 (lanes 6 and 7), or GST-p53 with an N-terminal deletion (lanes 8 and 9) or GST-p53 with a C-terminal deletion (lanes 10 and 11) in the presence (lanes 3, 5, 7, 9 and 11) or absence (lanes 1, 2, 4, 6, 8 and 10) of the immunogenic hTEP1 peptide. Samples were immunoprecipitated with normal IgG, anti-p53 or anti-hTEP1 antibodies as indicated. Immunoprecipitated proteins were separated by SDS - PAGE and examined for p53 immunoreactivity by immunoblotting using anti-p53 antibodies. Co-immunoprecipitated p53 and GST-p53 with hTEP1 are indicated with arrows

Figure 2 Inhibition of telomerase activity by GST-p53. (a) Concentration-dependent inhibition of telomerase activity by recombinant p53. Purified GST-p53 or GST proteins were included in incubations for assessment of nuclear telomerase activity using TRAP assays (Kim et al., 1994). An RNase-A control is included as a specificity control. Results are representative of four separate experiments. (b) The inhibition of telomerase activity by p53 is not an artifact of p53 inhibition of PCR amplification. After incubation of telomerase extracts with p53, de novo synthesized telomeres were phenol/chloroform extracted to remove all proteins and then subjected to telomere-specific PCR. RNase-A and heat-treated controls (to deactivate telomerase) were included and the telomerase activity was determined in duplicate. (c) Effects of telomerase substrate (TS) oligonucleotide primer on basal and GST-p53-inhibited telomerase activity in the nuclear lysates of human breast cancer cells. Varying concentrations of TS primer were included in the incubations of telomerase extracts with or without GST-p53 as indicated, which was then subjected to telomere-specific PCR. (d) Effects of different GST-p53 mutants on telomerase activity. Recombinant GST-p53 fusion proteins were individually included at three different concentrations in the TRAP assay with nuclear telomerase extracts from human breast cancer cells. In contrast to the N-terminal deletions (2C, 3C) and DNA-binding domain mutation (R175H), deletion of the C-terminal region (N5) abolished the inhibitory effect of p53 on telomerase. (e) Effects of differentially phosphorylated GST-p53 on telomerase activity. Wild-type GST-p53 bound to glutathione sepharose 4B beads was incubated with or without purified cdc2 protein kinase or protein kinase C under appropriate phosphorylation conditions separately for 20 min at 32°C. Following washing to remove the protein kinases and other reagents, differentially phosphorylated and control GST-p53 were eluted and tested in TRAP assays. Different concentrations of the proteins were used and an RNase-A control for telomerase specificity was included as indicated

Figure 3 TEIPP1 not only inhibits telomerase activity but also antagonizes p53 inhibition of telomerase. (a) Concentration-dependent inhibition of telomerase activity by TEIPP1. Nuclear telomerase extracts were incubated with different concentrations of TEIPP1 or scrambled TEIPP1 as indicated. RNase-A controls for specificity of telomerase activity are also indicated. Results are representative of four separate experiments. (b, c) TEPP1 reverses p53 inhibition of telomerase activity. Nuclear telomerase extracts were incubated with or without GST-p53 (100 ng in b and 200 ng in c) plus or minus TEIPP1 or scrambled control peptide at different concentrations. Telomerase activity was determined using the TRAP assay. Results are representative of three separate experiments

Figure 4 A cartoon illustrating the mechanisms postulated to underlie the interactions between telomerase, p53 and TEIPP1. Telomerase (globule shaped) may maintain its activity with the N-terminal region of hTEP1 being involved in forming and/or maintaining a potentially active configuration. Added p53 (pentagon) may interact with this region resulting in a conformational change and low telomerase activity. Added TEIPP1, corresponding to this N-terminal region of hTEP1, may competitively inhibit the natural folding of hTEP1 leading to lowered activity similar to that induced by p53. When both p53 and TEIPP1 are added together, inhibition by either can be prevented at appropriate concentrations or return at higher concentrations of either p53 or TEIPP1