Background
p53, a major tumor suppressor or guardian of the genome is mutated, deleted or inactivated in various cancers [
1‐
4]. Almost all human papillomavirus (HPV) infected cancer cells contain wild-type p53. p53 is non-functional as HPVE6 protein abrogates its function either by ubiquitin-dependent and independent degradation [
5], by inhibition of acetylation or by repressing p53-dependent downstream molecular pathways [
6]. Though, E6 associates with p53 for its degradation [
4]; there are contradictory reports on the inhibition and activation of p53 pathways by E6 [
7,
8].
Ectopic expression of p53 in cancer cells lacking p53 or harboring mutant and/or abrogated wild-type p53, have contrasting effects on cell-fate. In p53 null cancer cells, p53 overexpression causes cell cycle arrest and apoptosis [
9]. However, in virus infected cells harboring wild-type p53, overexpression of p53 does not induce cell cycle arrest and apoptosis [
10]. Till date there are only three reports describing the consequences of p53 overexpression in HPV-positive cells and results obtained leave ample scope for debate [
10‐
12]. Disparity among these reports may be due to differences in adenoviral multiplicity of infection. Taken together, the role of p53 overexpression in HPV-positive cells remains obscure. In HPV-positive cells, E6 works at different hierarchal levels in p53 pathway. It degrades p53, p21 and Bax causing impairment in cell cycle arrest/apoptosis [
13,
14] and making p53 activation more difficult.
With recent developments in efficient gene delivery systems and the prospect of gene therapy making a come-back [
15] it's likely that p53 based therapy may become a reality [
2]. p53 executes its tumor suppressor activity by triggering cell cycle arrest and apoptosis. However, the factors that facilitate selection between cell cycle arrest and/or apoptosis are not well-understood. It has been reported that p21 is most important transcriptional targets of p53 for causing cell cycle arrest [
16] and p53 executes apoptosis through Bax transcription [
17]. To study the role of p53 in E6-positive cells, we developed a novel isogenic HeLa cells with Tet-On-regulated p53 expression. Tet-On system exhibits tight-on/off regulation and is devoid of pleiotropic effects. Moreover, rapidly high induction levels are achievable and the inducer, doxycycline (Dox), is well-characterized.
p53 overexpression does not promote cell cycle arrest and apoptosis in HeLa cells. We demonstrate that protein phosphatase 2A (PP2A) controls p53 functions and its inhibition activates p53, causing cell cycle arrest/apoptosis
in vitro and tumor growth inhibition
in vivo. Interestingly, cyclin dependent kinase 5 (Cdk5) regulates p53 phosphorylation essential for its activation. Taken together, we propose that non-genotoxically overexpressed p53 can be activated by inhibiting its dephosphorylation in HPV-positive cervical cancer cells. This strategy may be of therapeutic importance in p53 associated gene therapy [
18‐
20].
Materials and methods
Chemicals and cell-lines
Antibodies against p53 (DO-1 and FL-393), GFP, pCdk5, Cdk5, p35, Bcl-2, cytochrome-C, PARP, COX-IV, β-Tubulin, β-Actin, HRP-linked secondary antibodies, Control, p53 and PP2A siRNAs were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Phospho-specific p53 antibodies were purchased from Cell Signalling Technology (Danvers, MA). FITC and rhodamine-conjugated secondary antibodies were purchased from KPL (Gaithersburg, MD). APO-Direct TUNEL kit, Matrigel and Bax antibody was purchased from BD (Franklin Lakes, NJ). Dox was purchased from Sigma (St. Louis, MO). Hygromycin B and Tet system approved serum was purchased from Clontech (Mountain View, CA). G418 and MTT were purchased from USB (Cleveland, OH). Okadaic acid (OA), mitotracker dye, DMEM, FBS and Lipofectamine2000 were purchased from Invitrogen (Carlsbad, CA). Cdk2/5 inhibitor, PFTα and U0126 were purchased from Calbiochem (Gibbstown, NJ). Cdk5 siRNA was purchased from Dharmacon Inc (Lafayette, CO). Development of cell-lines is described in additional file
1.
Plasmids and transfection
pC53-SN3 and pG13CAT were a kind gift from Dr. Bert Vogelstein, John Hopkins, Baltimore, MD. p53 fragment of pC53-SN3 was sub cloned in BamH1 site of pTRE and renamed as pTREp53. pG13CAT contains 13 repeats of p53 consensus binding site inserted in the 5' end to polyomavirus basal promoter linked to CAT reporter gene. Cells were co-transfected with 2 μg of pG13CAT and 0.5 μg of pEGFPC1 which serves as an internal control for transfection. Bcl-2 fragment from pRc/CMVBcl-2 (kind gift from Dr. S. Soddu, Regina Elena Cancer Institute, Italy) was excised by HindIII and cloned into pTRE2 to obtain pTRE2Bcl-2. Cells were transfected with either 2.0 μg (for 35 mm plate) or 0.5 μg (for 96 well plate) plasmid by Lipofectamine2000 transfection reagent as per manufacturer's instructions.
Clonogenic-survival assay
Cells (500) were treated with indicated concentrations of Dox, OA or Cdk2/5 inhibitor based on the experimental design and incubated for 48 h. Cells were further grown for 21 days and thereafter colonies on the plate were stained with crystal-voilet.
Electrophoretic mobility shift assay (EMSA)
To visualize the DNA-binding activity of p53 in nuclear extracts of HTet23p53, HTet26p53, HTet43GFP and HeLa cells, EMSA was performed. After treatment with Dox, cells were harvested for the preparation of cytoplasmic and nuclear fractions by using nuclear extraction kit as per manufacturer's instructions (Chemicon, Billerica, MA). Nuclear lysates were incubated for 45 min at 4°C and cleared by centrifugation at 15,000 ×g for 15 min at 4°C. Equal amount of nuclear proteins were used for the binding reaction. Complementary oligonucleotides containing the sequences corresponding to putative p53 binding site (forward, 5'-GAACATGTCTAAGCATGCTG-3'; reverse, 5'-CAGCATTCTTAGACATGTTC-3') were annealed and 5'-end-labeled with 2 micro curie (ΔCi) [γ-32P] ATP using 10 U of T4 polynucleotide kinase (Invitrogen) for 90 min. Binding reaction was carried out in a final volume of 20 μl consisting of 10 mM Tris.HCl (pH 7.5), 50 mM NaCl, 1 mM DTT, 1 mM EDTA, 2.5% glycerol, 1 μg deoxyinosinic deoxycytidylic acid [poly(dI-dC)], 300 ng BSA, 5 μg nuclear extract, and 2 μl of [γ-32P] labeled oligonucleotide probe. Reaction mixtures were incubated for 20 min at room temperature. Samples were resolved on a native polyacrylamide gel. Gel was dried under vacuum at 80°C for 45 min by gel dryer (Bio-Rad) and DNA-protein complex were visualized by autoradiography.
Chloramphenicol acetyl transferase assay
Cells were co-transfected with pG13CAT and pEGFPC1 expression vector using Lipofectamine2000 as described in transfection section. After 18 h post-transfection, p53 was induced with Dox for 48 h with or without PFTα pretreatment for 1 h. CAT assay was performed as described earlier [
2] except that the reaction time was reduced to 30 min at 37°C. Spots were quantified by phosphoimager (Bio-Rad). GFP intensity was directly measured from the cell lysates to check or correct for equal transfection efficiency as well to normalize the reporter activity. The fluorescence intensity of GFP in equal amount of lysate was measured by fluorimeter (Fluoroskan Ascent FL, Fisher Scientific) with excitation at 485 nm and emission at 510 nm.
SiRNA transfections
Cells were transfected with 100 nM control or p53 siRNA using Lipofectamine2000 [
21]. Eighteen hour post-transfection, Dox was added with or without OA and further incubated for 48 h. Thereafter, western blot or MTT assay was performed. To knock-down PP2A and Cdk5; Cdk5 siRNA was transfected 12 h prior to PP2A siRNA transfection and then incubated with Dox for 48 h.
Immunoprecipitation and Chromatin-immunoprecipitation (ChIP) assay
After indicated treatment cells were lysed in RIPA buffer. Equal amount of protein (400 μg) was taken and lysates were pre-cleared with 50 μl protein A/G-plus agarose for 30 min. Fifty microgram lysates were run as input. Agarose beads were pelleted and supernatant was incubated with p53 specific antibody overnight at 4°C. Fifty microliter protein A/G-plus agarose was added in antibody-antigen complex with gentle shaking for 4 h at 4°C. The protein A/G-plus was separated by centrifugation at 4,000 rpm. Target and its associated proteins were disrupted and resolved on SDS-PAGE. The expression of Cdk5 and p53 was detected by western blotting.
For chromatin-immunoprecipitation assay cells or homogenized tumors which were earlier fixed with 1% para-formaldehyde for 15 min, were lysed with 500 μl of lysis buffer [5 mM PIPES (pH 8.0), 85 mM KCl and 1% NP-40]. After centrifugation (5000 rpm), nuclear pellets were resuspended in 150 μl buffer [50 mM Tris-Cl (pH 8.0), 10 mM EDTA and 1% NP-40]. To fragment DNA to approximately 500 bps, samples were sonicated and centrifuged for 10 min. Samples were diluted 10-fold in IP buffer [16.7 mM Tris-Cl (pH 8.0), 167 mM NaCl, 1% NP-40 and 1 mM EDTA]. Samples (400 μg) were incubated with anti-p53 or anti-goat IgG overnight. Remaining solutions (10-times diluted) were used as input. Protein A/G-plus agarose beads pre-blocked with salmon-sperm DNA were added to antibody-antigen complexes and incubated for 4 h. Immune-complexes were centrifuged and washed with buffer [20 mM Tris-Cl (pH 8.0), 150 mM NaCl, 0.5% SDS, 1% Triton X-100 and 2 mM EDTA] twice and with buffer containing 250 mM NaCl. Immune-complexes were eluted by 50 μl of buffer [1% SDS and 0.1 M NaHCO3] twice. Then 20 μl of 5 M NaCl was added and incubated at 65°C overnight. DNA was precipitated with ethanol. RT-PCR was performed with promoter primer pairs for p21 (F) 5'-GGC TGG TGG CTA TTT TGT CC-3', (R) 5'-TCC CCT TCC TCC CTG AAA AC-3', and bax (F) 5'-AGC GTT CCC CTA GCC TCT TT-3' and (R) 5'-GCT GGG CCT GTA TCC TAC ATT CT-3' at annealing temperature 57°C and 59°C respectively.
Mitochondrial and cytosolic fractionation
HTet26p53 cells were swelled in ice-cold hypotonic HEPES buffer [10 mM HEPES (pH 7.4) 5 mM MgCl2, 40 mM KCl, 1 mM PMSF and protease inhibitor cocktail] for 30 min and centrifuged at 1500 rpm to pellet the nuclei. The resulting supernatant was centrifuged at 10,000 rpm to pellet mitochondrial fraction. Supernatant was used as cytosolic fraction and mitochondrial pellet was washed with PBS twice. This pellet was lysed in mitochondrial buffer [10 mM MOPS (pH 7.4), 1 mM EDTA, and 4 mM KH2PO4, 1% NP-40, protease inhibitor cocktail] and centrifuged at 12,000 rpm for 30 min.
Immunostaining
Cells grown on Labtek chamber slides were treated with Dox for 48 h and processed for immunofluorescene study as described earlier [
21]. Primary antibody against p53 (1:50) was added and incubated for 2 h at room temperature. Following incubation, cells were washed 5 times. Fluorescein isothiocyanate (FITC) or Rhodamine conjugated secondary antibodies (1:100) were added and incubated for 1 h at room temperature. After five washes, vectashield mounting medium containing DAPI was added and slides were examined by a confocal microscope (LSM510, Carl Zeiss, Germany). For mitotracker deep red staining, after indicated treatments cells were incubated with 200 μM of mitotracker dye for 20 min. These were then fixed and processed for immunofluorescence study by incubating with a Bax specific primary antibody and FITC conjugated secondary antibody. Slides were mounted with DAPI containing medium and images were acquired in confocal microscope. Terminal deoxynucleotidyltransferase dUTP nick end labeling (TUNEL) staining was performed as per manufacturer's protocol (BD) except the reaction time was increased to 3 h at room temperature. Cells were washed twice with binding buffer and PI solution was added. Slides were washed, mounted and observed under confocal microscope (META, Carl Zeiss).
Tumor growth
HTet23p53 or HTet43GFP cells (5 × 106) in 100 μl PBS mixed with 100 μl matrigel were injected s.c. into 4-6 week-old female NOD/SCID mice (Jackson Laboratories). Total 12 mice were injected with HTet23p53 cells on the right flank and 4 mice were injected with HTet43GFP cells on both the flanks. Out of two groups, one was fed on 500 ng/ml Dox in drinking water. Tumor development was monitored. After tumor-size reached to 5-10 mm in diameter, OA (40 pg/mice) was administered at the tumor site. Tumor-sizes were measured weekly by digital Vernier Caliper (Sigma) and tumor volume was calculated by formula V = [1/2 × (large diameter) × (small diameter)2.
MTT assay, FACS analysis and western blotting
For methylthiazole tetrazolium (MTT) assay, 7,500 cells were treated with Dox, OA and/or Cdk2/5 inhibitor as per experimental requirement and assayed for cell survival. For western blotting following indicated treatments, cells were washed thrice with ice-cold phosphate buffered saline (PBS) and lysed in ice-cold lysis buffer (50 mM Tris-Cl, pH 7.5, with 120 mM NaCl, 10 mM NaF, 10 mM sodium pyrophosphate, 2 mM EDTA, 1 mM Na3VO4, 1 mM PMSF, 1% NP-40 and protease inhibitor cocktail (Roche Diagnostics, Penzberg, Germany). Equal amount of protein was resolved on a polyacrylamide gel. Where ever possible blots were stripped by incubating the membranes at 50°C for 30 min in stripping buffer (62.5 mM Tris-Cl pH 6.7, 100 mM mercaptoethanol, 2% SDS) with intermittent shaking. Membranes were washed thoroughly with TBS and reprobed with required antibodies. Otherwise gels run in duplicates were probed for the desired proteins by western blotting and then compiled.
For FACS analysis cells were plated at a density of 5 × 10
5 cells in 35 mm plates and allowed to adhere for 24 h. Cells treated as per experimental requirement were harvested by trypsinization and processed for flow cytometric analysis. The fluorescence of propidium iodide (PI) was measured through a 585 nm filter in a flowcytometer (FACS Calibur, BD) for 10,000 cells. Data were analyzed using cell quest software (BD). Details of these are as published earlier [
21,
22].
TUNEL staining
To detect apoptotic cells APO-DIRECT TUNEL assay kit (BD) was used followed by flow cytometric analysis as per the manufacturer's instructions with some modifications. Cells were incubated in DNA-labeling solution for 2 h at 37°C and analyzed by FACS Calibur (BD). PI stains total DNA and FITC conjugated dUTP stains apoptotic cells.
Reverse-Transcription-PCR
Total RNA from the cells or tumor samples was extracted using TRIzol™ reagent and PCR was performed as described [
21] with following primers; p53 (F) 5-CTG AGG TTG GCT CTG ACT GTA CCA CCA TCC-3', (R) 5'-CTC ATT CAG CTC TCG GAA CAT CTC GAA GC-3'; e6 (F) 5'-TGT GTA TGG AGA CAC ATT GG-3', (R) 5'-ATA GTG CCC AGC TAT GTT GT-3'; β-actin (F) 5'-ATC TGG CAC CAC ACC TTC TAC AAT GAG CTG CG-3', (R) CGT CAT ACT CCT GCT TGC TGA TCC ACA TCT GC-3', at annealing temperature of 55°C and p21 (F) 5'-GGC GTT TGG AGT GGT AGA AA-3' (R) 5'-GAC ACC ACT GGA GGG TGA CT-3' at annealing temperature of 59°C for 25-30 cycles.
Statistical analysis
Statistical comparisons are made using student's paired t-test using SPSS10.0 (SPSS Inc., IL) and P- value < 0.05 was considered significant.
Discussion
This study highlights the activation of overexpressed p53 and its effect on cell cycle arrest and apoptosis in HPV-positive HeLa cells. Under stress conditions p53 is stabilized by phosphorylation and acetylation at serine/threonine/tyrosine and lysine residues respectively. The serine phosphorylation at residues 6,9,15,20,33,37,46,315 and 392 plays a crucial role depending upon the nature of stress thereby causing cell cycle arrest and/or apoptosis [
26,
27]. Unlike stress condition wherein p53 induction promotes cell cycle arrest or apoptosis, this study demonstrates that p53 overexpression in HPV-positive cells does not induce cell cycle arrest or apoptosis though; it is reported to do so in other cancer cell types [
16,
17,
28]. The reason for this difference could be inhibition of cellular machinery necessary for performing critical posttranslational modifications which are required for sequence specific promoter selection of the genes responsible for the induction of cell cycle arrest or apoptosis by HPV [
5,
29].
Equilibrium between phosphorylation and dephosphorylation of a protein like p53 is essential for its normal functioning in the cells. Therefore, conditions causing shift in the equilibrium between phosphorylated and non-phosphorylated states will dictate the functionality of a protein and subsequently the cells fate [
30,
31]. Protein phosphatases inactivate p53 by dephosphorylating it. Very recently Lu
et al., reported that PP2A inhibition also decreases p53 protein and its phosphorylation at Ser15 through activation of its negative regulator MDM2 [
32]. In contrary, we herein demonstrate that inhibition of phosphatase stabilizes and activates overexpressed p53 probably because of impairment in functional MDM2 pathway in HPV-positive cells [
33]. Phosphorylation of p53 at specific serine residues is essential for the induction of cell cycle arrest and apoptosis. Under stress conditions p53 is phosphorylated at Ser20 located in the transactivation domain [
26], thereby stabilizing and triggering downstream pathways. Ser46 phosphorylation, located in the DNA-binding domain of p53 plays a crucial role in sequence specific DNA-binding required for the induction of cell cycle arrest and apoptosis [
34]. In this study, we confirm that phosphorylation at these residues fully restores p53 functionality and induces cell-death even under non-stress conditions.
Stress-induced p53 is stabilized and activated by various kinases such as ATM, ATR, Chk1, HIPK2 and Chk2 by phosphorylation [
26,
27,
34,
35]. However, very little is known about the kinases that phosphorylate p53 under non-stressed conditions. Cdk5 was originally discovered in HeLa cells [
36] and its functional role as p53 upstream kinase has been documented in neuronal cells [
37]. Involvement of Cdk5 in growth of breast and prostate cancers cells has been reported [
38‐
40]. Recently, we reported that Cdk5 transactivates p53 in breast cancer cells under positive regulation of ERK following carboplatin treatment [
40]. Cdk5-inhibition promotes survival of p53 expressing cells. As PP2A-inhibition restores the ability of overexpressed p53 to promote cell-death, the upstream kinase that phosphorylates overexpressed p53 under non-stress conditions was investigated. In the present study we demonstrate that p35, a Cdk5 activator levels diminishes following inhibition of PP2A and simultaneous increase in the levels of more sustainable Cdk5 activator p25 following p35 cleavage [
41]. Thus, increased level of Cdk5 activator (p25) may facilitate Cdk5-mediated phosphorylation of overexpressed p53, which causes cell-growth inhibition. The decreased level of p35 protein in HTet43GFP cells does not cause cell-growth inhibition because of unavailability of its substrate (in this model overexpressed p53). Though, Cdk5 plays an important role in activating overexpressed p53, as such it is not involved in the proliferation of parental HeLa cells
per se in spite of the fact that E6 expression leads to increase in Cdk5 protein expression.
p53 executes its apoptotic function through intrinsic or extrinsic pathways [
42,
43]. To further confirm the pathway involved, we investigated Bax, an important transcriptional target of p53 involved in promoting intrinsic mitochondrial apoptosis. Bax translocates to mitochondrial outer membrane causing MOMP and releases cytochrome-C into cytosol. Cells lacking Bax or those overexpressing Bcl-2 are profoundly resistant to a broad range of apoptotic stimuli, including chemotherapeutic drugs treatment and serum starvation [
17]. In HPV-positive cancers Bcl-2 overexpression and Bax degradation by E6 facilitates cancer progression [
14]. Here, we demonstrate that upregulated Bax translocates to mitochondria upon PP2A-inhibition in p53 overexpressing cells which is dependent on Cdk5 activity. Thus, only phosphorylated p53 triggers Bax transcription to increase its levels and cause apoptosis. In addition, the cell cycle arrest caused by inhibition of PP2A in p53 overexpressing cells may be dependent on transcriptional upregulation of p21 gene. Collectively these data also provide evidence for reactivation of E6 disrupted p21 and Bax pathways in HPV positive cells.
Finally, we propose that Cdk5 interacts with p53 and phosphorylates Ser20 and Ser46 residues. Phosphorylation restores the ability of overexpressed p53 to specifically bind on p21 and bax promoters (Figure
5C). These findings provide novel insight into the regulation of p53 transactivation functions and propose PP2A to be a key player in modulating p53 functionality. The phosphorylated status of specific residues may be involved in promoter selection and this proposition needs further investigations. Also, this is the first report which provides mechanism for functional activation of p53, and details the essential modifications necessary for non-genotoxically overexpressed p53 to be able to execute its tumor suppressor functions in HPV-positive cells. Moreover, activation of overexpressed p53 without targeting viral oncogenes may have implication in the treatment of virus infected carcinomas. The efforts towards the newer approaches to target p53 pathway and usefulness of reactivation of p53 pathways in treatment of cancers are encouraging. Therefore, these findings could have therapeutic importance for the treatment of cervical cancers as well as other cancers types in which p53 is functionally abrogated.
Acknowledgements
We thank Dr. G.C. Mishra, Director, NCCS for giving encouragement to accomplish this work. We also thank Department of Biotechnology, Government of India, for providing financial support. A.K.A thanks ICMR, A.K.U and R.K thank UGC and V.P thanks CSIR for fellowships.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
AKA performed most of the experiments and prepared the manuscript. AKU and SS helped in manuscript preparation. MVV repeated animal experiment. RK and VP helped in tumor weight and excision. BR helped in mice daily maintenance. MKB conceived the study, participated in its design and coordination, corrected the manuscript and supervised the project. All authors read and approved the final manuscript.