p 53-dependent suppression of the human calcyclin gene ( S 100 A 6 ) : the role of Sp 1 and of NFκB

Calcyclin (S100A6) is believed to participate in cell cycle control. It was, however, unclear if its expression depends on p53, a key regulator of apoptosis and cell cycle. We therefore performed transcription regulation assays in HeLa cells and found that wild type p53 suppressed the S100A6 promoter up to 12-fold in a dose-dependent manner. In contrast, the well-characterized V143A, R175H, R249S, and L344A p53 mutants cloned from human cancers suppressed this promoter with a 6 to 9-fold lower efficiency. All the sites mediating the p53-dependent suppression were contained in the –167 to +134 fragment of the S100A6 promoter. Separate overexpression of either Sp1 or of NFκB only partially counteracted the p53 inhibitory effect on the S100A6 promoter, while simultaneous overexpression of both these transactivators resulted in a complete abolishment of the p53 inhibitory effect on this promoter. Sp1 and NFκB binding to the probes resembling their putative binding sites present in the S100A6 promoter was decreased in the presence of wild type p53. We propose that the suppression of S100A6 is yet another mechanism by which p53 inhibits proliferation. Insufficient suppression of this gene by p53 mutants could well be responsible for calcyclin overexpression and cell cycle deregulation observed in cancer tissues.


IntRoduCtIon
Calcyclin (also known as S100A6) seems to be involved in early phase of cell differentiation (Tonini et al., 1995), exocytosis (Thordarson et al., 1991;Okazaki et al., 1994) and, above all, in cell cycle regulation, but there are still some doubts about the precise function of this protein.Calcyclin belongs to the S100 family of calcium-binding proteins (Heizmann et al., 2002;Donato, 2003;Santamaria-Kisiel et al., 2006) and under physiological conditions it is mainly expressed in epithelial cells and fibroblasts (Kuznicki et al., 1992), in neurons (Filipek et al., 1993), and in lymphocytes (Ferrari et al., 1992).Its putative role in cell cycle regula-tion is supported by the following findings: first, the amount of calcyclin mRNA in the cell undergoing mitosis is not uniform: it is high in the late G1 phase and decreases in the S phase of the cell cycle (Hirschhorn et al., 1984); second, during late prophase, calcyclin relocates from the nuclear envelope to the cytoplasm and then it disappears (Stradal & Gimona, 1999;Tomas & Moss, 2003;Farnaes & Ditzel, 2003); third, calcyclin mRNA and protein levels increase in the renal cortex during recovery after acute tubular necrosis, and the protein co-localizes with the proliferating cell nuclear antigen (PCNA) (Cheng et al., 2005).In addition, calcyclin is overexpressed in many different human cancers, such as melanoma (Weterman et al., 1993),
Expression of the gene encoding calcyclin, S100A6, is activated by growth factors.Serum-inducible and platelet-derived growth factor-responsive sequences are contained within a 164 bp promoter fragment just upstream of the transcription start site (Ghezzo et al., 1988).To further support the hypothesis on calcyclin involvement in cell cycle control, we decided to establish if the activity of its gene depends on the action of p53, one of the major regulators of proliferation and apoptosis (Szymańska & Hainaut, 2003).While p53-dependent activation of transcription is mediated by direct p53 binding to its recognition sequences within the target promoters, transcription suppression by p53 is exerted via a number of different mechanisms: direct p53 binding to target promoter (St Clair et al., 2004), competition with transactivators for binding to DNA (Subbaramaiah et al., 1999;Li & Lee, 2001), or formation of complexes between p53 and other transcriptional activators leading to their sequestration or to a change in their activity or affinity for DNA (Xu et al., 2000;Sengupta et al., 2005).The transcription factor most commonly involved in the latter mechanism of transrepression by p53 is Sp1 (Webster et al., 1996;Bargonetti et al., 1997;Ohlsson et al., 1998;Xu et al., 2000;Zhang et al., 2000).Other transcription factors include C/EBP (Webster et al., 1996), NFκB (Bargonetti et al., 1997), as well as members of the basal transcriptional machinery, such as the TATA-binding protein (Seto et al., 1992;Truant et al., 1993;Subbaramaiah et al., 1999) and other general transcription factors (Ragimov et al., 1993;Xiao et al., 1994;Farmer et al., 1996).
In this paper we show that the promoter of the S100A6 gene encoding calcyclin is suppressed by wild type p53 in a dose-dependent manner, while suppression by p53 mutants cloned from human cancers is less efficient.We also provide evidence that this is a result of a dual mechanism: p53 interference with Sp1 and with NFκB function on the S100A6 promoter.

MAteRIALS And MetHodS
Cloning of Sp1 transcription factor.The 2355 bp human Sp1 coding sequence (GeneBank NM138473) was cloned from human cDNA (made from healthy thyroid mRNA) with Platinum Pfx polymerase (Invitrogen Life Technologies, Carlsbad, CA, USA) and using the following primers: 5'Sp1 5'-CAGGtACCATGAGCGACCAAGATCACTC-3' (incorporated KpnI site in bold), 3'Sp1 5'-CACtC-GAGTCAGAAGCCATTGCCACTGA-3' (incorporated XhoI site in bold) in a PCR reaction supplemented with MgSO 4 up to 2 mM.After a 3 min initial denaturation at 94 o C, five cycles of 94 o C for 20 s, 55 o C for 30 s and 68 o C for 3 min, and then 30 cycles of 94 o C for 20 s, 58 o C for 30 s and 68 o C for 3 min were performed, followed by a final extension at 68 o C for 5 min.Once the PCR reaction was completed, 2 units of Taq polymerase was added to generate A-overhangs, and the sample was incubated at 72 o C for 20 min.The PCR product was then cloned into the pGEM-T vector (Promega, Madison, WI, USA), cut with KpnI and XhoI endonucleases and re-cloned into the pcDNA3.1(+)expression vector (Invitrogen Life Technologies, Carlsbad, CA, USA) prepared with the same enzymes.The Sp1 coding sequence was verified by restriction and sequencing.
Cloning of the -66 to +134 deletion mutant of the S100A6 promoter.The shortest, 200 bp long deletion mutant of the S100A6 promoter (pGL2-CACY(-66/+134)) was cloned by PCR on a template of the vector containing the -1731 to +134 fragment of the S100A6 promoter (pGL2-CACY(-1731/+134) (a gift from Dr. Wiesława Lesniak, Nencki Institute of Experimental Biology, Warszawa, Poland; Lesniak et al., 2000), with proofreading Platinum Pfx polymerase, the forward primer 5'TACACGAGCTT-GGCCGAGCTGGCCT3' (XhoI restriction site in bold) and the reverse primer 5'-AGAAGCttGATC-CAGCGGCTGAACTGG-3' (HindIII restriction site in bold).The cycles were as follows: 94 o C for 3 min, five cycles of 94 o C for 20 s, 56 o C for 30 s, 68 o C for 1 min, 30 cycles of 94 o C for 20 s, 58 o C for 30 s, 68 o C for 1 min, final extension at 68 o C for 5 min.To produce A-overhangs at 3' ends of PCR product, 2 units of Taq polymerase was then added, and the sample was incubated at 72 o C for 20 min.The product of the reaction was cloned into the pGEM-T vector, restricted out with XhoI and HindIII enzymes, and recloned into the pGL2-Basic vector (Promega, Madison, WI, USA) cut with the same enzymes.
Cloning of the peGFP-CACY(-1731/+134) reporter plasmid.The -1731 to +134 fragment of the S100A6 promoter was amplified on the template of pGL2-CACY(-1731/+134) with Platinum Pfx polymerase, and with 5'-TCGAACGTAGAGtACtCGGT-GTTCCTGAG-3' forward and 5'-TCGAACGTAG-GGAtCCAGCGGCTGAAC-3' reverse primers (ScaI and BamHI restriction sites in bold, respectively).The cycles were: 94 o C for 2 min, 20 cycles of 94 o C for 20 s, 68 o C for 2 min, and the final extension at 68 o C for 5 min.To produce A-overhangs at the 3' ends of the PCR product, 2 units of Taq polymerase was added, and the sample was incubated at 72 o C for 20 min.The product of the reaction was cloned into pGEM-T vector, and restricted out with ScaI and BamHI.pEGFP-N1 vector (Clontech Laboratories, Mountain View, CA, USA) was restricted with AseI, and the overhanging ends were blunted with the Klenow enzyme.After gel purification, the vector was restricted with BamHI.The PCR product was ligated into this vector, producing a pEGFP-CACY(-1731/+134) reporter plasmid.
Mutagenesis of the S100A6 promoter.A 38 bp GC-rich fragment located at positions -100 to -63 of the S100A6 promoter (5'-GGGTGGGGCGCG-GGCGGGACTTGGGCGGGCGGGGTGGG-3'), containing putative Sp1 binding sites, was removed by PCR from the 1.5 kb promoter fragment.Initially, two fragments of this promoter were cloned.The first one flanking the GC-rich sequence from the 5'-side was cloned with the following primers: 5'cacy 5'-TCGAACGTAGAGtACtCGGTGTTCCT-GAG-3' (ScaI restriction site in bold) and -3'cacyint 5'-AGCTCGGCCAAG//TGCGCCGTGGCTTACAC-3' consisting of the sequences flanking the GC-rich promoter fragment from both sides (// shows the border of the two sequences), while the second one located at the -3' side of the GC-rich fragment was cloned with the following primers: 5'-cacyint 5'-AGCCACG-GCGCA//CTTGGCCGAGCTGGCCT-3' consisting of the sequences flanking promoter fragment to be removed (// shows the border of the two sequences) and -3'cacy 5'-TCGAACGTAGGGAtCCAGCGGCT-GAAC-3' (BamHI restriction site in bold).The products of these PCRs were resolved on 1% agarose gel.The bands of approx.1.25 kb (the first S100A6 promoter fragment) and 200 bp (the second fragment) were excised and incubated separately in 100 µl of dH 2 O for 4 h.Aliquots of 4 µl from each fragment were pooled and used as a template in a second round of PCR performed with 5'-cacy and -3'cacy primers.All PCRs were performed with proofreading Platinum Pfx polymerase.The cycles of the first round of PCR were as follows: 94 o C for 2 min, five ycles of 94 o C for 20 s, 58 o C for 30 s, 68 o C for 2 min, 25 cycles of 94 o C for 20 s, 68 o C for 2 min, and the final extension at 68 o C for 5 min.The cycles of the second PCR were: 94 o C for 2 min, 30 cycles of 94 o C for 20 s, 68 o C for 2 min, and the final extension at 68 o C for 5 min.The product of the second round of PCR was cut with ScaI and BamHI enzymes, while pGL2 basic vector was cut with SmaI and BglII enzymes.Both DNAs were ligated producing pGL2-CACY(-1371/+134)∆38 plasmid.pGL2-CACY(-167/+134)∆38 was prepared from pGL2-CACY(-1371/+134)∆38 by restriction of a 263 bp long promoter fragment (-167 to +134) with SmaI and HindIII enzymes, and its ligation into the pGL2-basic vector prepared with the same enzymes.
Cell culture.HeLa cells were grown in high glucose Dulbecco Modified Eagle's Medium (DMEM, Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS), with or without 100 u/ml penicillin G and 100 µg/ml streptomycin.Cells were incubated in a humidified incubator at 37 o C in 5% CO 2 .
Immunofluorescence.HeLa cells (10 4 ) were seeded onto glass coverslips (soaked in 70% ethanol for 3 h and air-dried) placed into a 24-well dish.The cells were transfected 24 h later with 250 ng pEGFP-CACY(-1371/+134) reporter plasmid containing EGFP reporter gene placed under the control of the -1371 to +134 fragment of the S100A6 promoter, and with 100 ng CMV-p53 plasmid or with 100 ng of 'empty' pcDNA3.1(+)vector.After 24 h incubation, the cells were washed 3 times with PBS, fixed for 10 2008 W. Kroliczak and others min at 4 o C with 2% paraformaldehyde, incubated for 10 min at room temperature in 0.1% Triton X-100 in PBS, and washed again 3 times in PBS for 1 min.The cells were then incubated in 2% BSA in PBS for 1 h at room temperature, incubated for 1 h at room temperature in a humidified chamber with a primary anti-p53 mouse monoclonal antibody (1:500 in 2% BSA in PBS; DO1, Oncogene Science, Inc., Cambridge, MA, USA), washed 3 times in 2% BSA in PBS, and incubated for 1 h in a humidified chamber in the dark with Alexa Fluor 594 goat anti-mouse secondary antibody (1:1000 in 2% BSA in PBS; Invitrogen Molecular Probes, Eugene, OR, USA).After 3 further washes with 2% BSA in PBS (twice for 40 min at room temperature, and once for 16 h at 4 o C), the coverslips were mounted with SlowFade Light Antifade Reagent (Invitrogen Molecular Probes, Eugene, OR, USA) and sealed with acetone-free nail polish.The cells were analyzed under a fluorescence microscope.
Whole-cell protein isolation from HeLa cells and Western blotting.HeLa cells (2×10 5 ) were suspended in 70 µl of lysis buffer consisting of 75 mM Tris/HCl, pH 8.0, 2% SDS, 15% glycerol, and boiled for 5 min.Forty micrograms of protein extract was supplemented with β-mercaptoethanol to 5% and bromophenol blue to 0.01%, boiled, and loaded onto a 10% polyacrylamide gel.Proteins were transferred onto a nitrocellulose membrane (Bio-Rad Laboratories, Inc., Hercules, CA, USA).The protein loading and integrity were monitored by Ponceau Red staining.The destained membrane was blocked overnight at 4 o C in 5% nonfat dry milk solution in TBS-T (Tris-buffered saline (TBS) supplemented with 0.1% Tween 20), washed at room temperature in TBS-T once for 15 min and twice for 5 min, incubated with a mouse monoclonal anti-p53 antibody (1:10 000 in TBS-T; DO1, Oncogene Science, Inc., Cambridge, MA, USA), or a rabbit polyclonal anti-Sp1 antibody (1:2500 in TBS-T; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), or a rabbit polyclonal anti-p65 NFκB antibody (1:5 000 in TBS-T; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), or a monoclonal anti-β-actin antibody (1:10 000 in TBS-T; Sigma-Aldrich, St. Louis, MO, USA) for 1 h at room temperature, washed as before, then incubated at room temperature for 1 h either with a goat anti-mouse horseradish peroxidase-conjugated polyclonal antibody (1:10 000 in TBS-T; Calbiochem, San Diego, CA, USA), or a goat anti-rabbit horseradish peroxidaseconjugated polyclonal antibody (1:10 000 in TBS-T; Calbiochem, San Diego, CA, USA), respectively.Specific bands were visualized by chemiluminescent reaction performed with an ECL kit (Amersham Biosciences UK Limited, Little Chalfont, England).The blots were exposed against the film (Biomax MS, Eastman Kodak Company, Rochester, NY, USA) for 15 s to 3 min.Relative amounts of the receptor proteins were estimated from the densitometric measurements of the intensity of the specific bands normalized against the intensity of β-actin band.
Isolation of nuclear proteins.A 10 cm-diameter dish with HeLa cells (approx.5×10 6 cells) was washed twice with PBS.The cells were then scrapped off, suspended in 500 µl of buffer A consisting of 10 mM Hepes, pH 7.9, 10 mM KCl, 1.5 mM MgCl 2 , supplemented with a protease inhibitors mix (Complete Protease Inhibitor Cocktail, Roche Applied Science, Basel, Switzerland) and PMSF (phenylmethylsulfonyl fluoride) to 400 µM, and incubated on ice for 1 h.After incubation, the cells were homogenized in an ice-cold glass-teflon homogenizer.The resulting homogenate was transferred to an Eppendorf tube and centrifuged at 400 × g for 5 min at 4 o C. The pellet was resuspended in an equal volume of buffer A, centrifuged as before, and again resuspended in two volumes of buffer B consisting of 20 mM Hepes, pH 7.9, 10% glycerol, 420 mM NaCl, 1.5 mM MgCl 2 , 0.2 mM EDTA, supplemented with protease inhibitors as above.After a 30 min incubation on ice, the sample was centrifuged at 15000×g for 20 min at 4 o C, then the supernatant was transferred to a new tube and supplemented with an equal volume of buffer C consisting of 20 mM Hepes, pH 7.9, 30% glycerol, 1.5 mM MgCl 2 , 0.2 mM EDTA, and protease inhibitors.The sample containing soluble nuclear proteins was aliquoted into pre-chilled Eppendorf tubes, flash-frozen in liquid nitrogen, and stored at -80 o C.

ReSuLtS p53 suppresses the human S100A6 promoter
Initially, to verify that p53 expressed from CMV-p53 plasmid was functional in the transfected HeLa cells that weakly express endogenous p53 (Hoppe-Seyler & Butz, 1993), we performed immunoblot analysis of the cells transfected with this plasmid (Fig. 1A), and control transcription regulation assays with this plasmid and with pGL3-p53BS reporter plasmid containing a p53BS synthetic promoter with consensus p53 binding sites (a gift from Professor Maciej Zylicz, International Institute of Molecular and Cell Biology, Warszawa).They showed that in the presence of overexpressed p53, the activity of the reporter gene increased approx.16-fold (Fig. 1B).
To address the question of regulation of the human S100A6 promoter by p53, the pGL2-CACY(-1371/+134) reporter plasmid, containing a 1505 bp fragment of the human S100A6 promoter, was transfected into HeLa cells together with increasing amounts of the CMV-p53 expression vector.These experiments showed that the inhibition of the (A) Immunoblots of whole-cell extracts made from control HeLa cells, and from HeLa cells overexpressing p53.The blots were probed with anti-p53 antibody and re-probed with anti-β-actin antibody.(B) HeLa cells transfected with CMV-p53 vector overexpress transcriptionally active p53 protein.HeLa cells were transfected with pGL3-p53BS reporter vector with the promoter containing consensus p53 binding sites, as well as with CMV-p53 expression vector encoding wild type p53.(C) The S100A6 promoter is inhibited by p53 in a dose-dependent manner.HeLa cells were transfected with pGL2-CACY(-1371/+134) reporter vector, with increasing amounts of CMV-p53 expression vector or with equal amounts of 'empty' pcDNA3.1(+)vector, and with phRL-CMV internal control vector.Firefly luciferase activity was measured 24 h post-transfection, and the results normalized against Renilla luciferase activity.Bars represent the mean result (± standard deviation) of 9 experiments performed with two different DNA preparations.

2008
W. Kroliczak and others S100A6 promoter was p53 dose-dependent and that the S100A6 promoter was suppressed up to 12-fold (8% of its activity without p53).In the presence of an 'empty' pcDNA3.1(+)vector, the S100A6 promoter activity remained unaltered, providing that no more than 100 ng of this vector was used (Fig. 1C).Transfection with higher amounts of the expression construct resulted in a slight, non-specific promoter suppression.Based on the above results, we decided to use 100 ng of CMV-p53 in all subsequent experiments.This amount of expression vector caused an 8-fold suppression of the S100A6 promoter after 24 h incubation.
To corroborate the above result with a different method, we designed an experiment in which HeLa cells were transfected with pEGFP-CACY(-1731/+134) reporter vector encoding EGFP protein under the control of the S100A6 promoter, and either with the CMV-p53 expression vector, or with an equal amount of 'empty' pcDNA3.1(+).The relative amount of EGFP protein was assessed 24 h later under a fluorescence microscope in the cells transfected with the reporter and pcDNA3.1(+)vectors vs. cells transfected with the reporter and p53 expression plasmids.A direct comparison of both cell types was not possible, because the single and double transfections had to be performed separately, and, for this reason, the cells were grown on coverslips placed in separate wells.However, both coverslips were identically processed for immunofluorescence.Therefore, both cell types had identical controls, i.e. neighboring, not-transfected cells.Using them as go-between controls, we found that the cells transfected with the EGFP reporter vector and with pcDNA3.1(+)expressed high levels of EGFP protein in the cytoplasm (Fig. 2B, black arrows).In contrast, the cells transfected with the EGFP reporter vector and the CMV-p53 expression plasmid, and, therefore, overexpressing p53 in the nucleus (Fig. 2C, black arrows), did not express EGFP, judging by the absence of a specific immunofluorescent signal in their cytoplasm (Fig. 2D, black arrow), a situation identical to that observed in control cells that had not been transfected (Fig. 2D, white arrows).

the -167 to +134 fragment of the S100A6 promoter contains all sites mediating p53-dependent suppression
In order to find the region of the S100A6 promoter mediating the suppression exerted by p53, we used promoter deletion mutants (-1371/+134, -588/+134, -365/+134, -167/+134, and -66/+134, Fig. 4A) in transcription regulation assays.The assays were performed as described above.Deletion of subsequent fragments from the 5' end of the S100A6 promoter had a weak influence on the promoter's activity; only shortening of the promoter to the -167 position resulted in a 40% decrease of its activity.Deletion of the subsequent 101 bp to the -66 position resulted in the complete silencing of the S100A6 promoter.With the exception of the inactive -66 to +134 fragment, each promoter fragment was suppressed by p53 to the same extent, i.e. approx.8-fold (Fig. 4B).This data indicate that the three deleted putative p53-binding sites (DR-1, DR13, DR17, located at positions -846 to -828, the -837 to -805, and -814 to -778, respectively, Fig. 4A) did not mediate the S100A6 promoter inhibition by p53.Fur-thermore, lack of a direct involvement of p53 in this process was confirmed by the negative result of electrophoretic mobility shift assays performed with HeLa nuclear extracts containing overexpressed p53, and with the probes resembling the four putative p53-binding sites present in the S100A6 promoter (the three described above, and the fourth located at position -65 to -46, DR0, Fig. 4A).No specific binding was observed (not shown).
Taken together, these results indicate that the -167 to +134 promoter fragment contains all the sequences mediating the p53-dependent suppression of the S100A6 transcriptional activity.

deletion of a GC-rich region located at position -100 to -63 decreases the level of p53-induced S100A6 promoter suppression
The -167 to +134 S100A6 promoter fragment differs from the -66 to +134 fragment by 101 bp including the 38 bp long GC-rich sequence located at the -100 to -63 position, containing six overlapping putative Sp1-binding sites.To determine if this sequence is important for p53-dependent suppression of the S100A6 activity, transcription regulation assays were performed with pGL2-CACY(-1371/+134)∆38, and with pGL2-CACY(-167/+134)∆38 plasmids devoid of this sequence (Fig. 5).The deletion of the GC-rich sequence decreased the overall promoter activity by approx.50%.Importantly, deletion of the putative Sp1-binding sites decreased

Sp1 only partially counteracts the p53-dependent suppression of the S100A6 promoter
The role of Sp1 in the S100A6 promoter suppression by p53 was further analyzed using transactivation assays performed in the presence of overexpressed p53 and of increasing amounts (20-250 ng) of the pcDNA3.1(+)-Sp1expression construct.As shown before (Fig. 1C), in the presence of 100 ng of the CMV-p53 expression plasmid, the activity of the S100A6 promoter was inhibited 8-fold.In the presence of overexpressed Sp1 (Fig. 6A) only a partial relief of the p53-dependent suppression of the S100A6 promoter was achieved.This effect was dose-dependent (Fig. 6B).In the presence of 200 ng of the Sp1 expression construct, the p53-dependent suppression of the S100A6 promoter activity was only 1.8-fold (to 55-60% of the promoter activity without p53).This effect was not magnified by further increase of the amount of the Sp1 expression vector.

NFκB only partially counteracts the p53-dependent suppression of the S100A6 promoter
We then turned our attention to the role of the NFκB transcription factor (another factor indicated by others as mediating the p53 inhibitory effect on transcription) in the S100A6 promoter suppression by p53.Transcription regulation assays were performed as described before, in the presence of increasing amounts of pcDNA3.1(+)-NFκBp50 (5-150 ng) and of pcDNA3.1(+)-NFκBp65 (2.5-100 ng) expression plasmids.Transfection performed with 25 ng or higher amounts of the NFκB p65 expression construct resulted in a similar production of the encoded protein.Again, in the presence of overexpressed NFκB (Fig. 7A), only a partial, dose-dependent reversal of the p53-dependent suppression of the S100A6 promoter was achieved.In the presence of optimal amounts of the NFκB expression  constructs, the activity of the S100A6 promoter was suppressed by p53 only 1.8-fold (to 55-60% of its activity without p53).The plateau of de-inhibition was reached with 25 ng of NFκB p50 and 12.5 ng of NFκB p65 expression vectors (Fig. 7B).

Simultaneous action of Sp1 and NFκB fully counteracts the S100A6 promoter suppression by p53
To establish if Sp1 and NFκB are the only factors mediating p53-dependent suppression of the S100A6 promoter, its activity was analyzed in the presence of overexpressed p53 and of the lowest amounts of expression plasmids pcDNA3.1(+)-Sp1,pcDNA3.1(+)-NFκBp50 and pcDNA3.1(+)-NFκBp65 (200 ng, 25 ng and 12.5 ng, respectively) that, as shown in the previous experiments, maximally counteracted the p53-dependent suppression of the S100A6 promoter.Simultaneous overexpression of both transcription factors completely abolished the inhibitory action of p53 on the S100A6 promoter (Fig. 8).

Sp1 and NFκB binding to the S100A6 promoter is decreased in the presence of p53
To further elucidate the role of Sp1 and NFκB in the p53-dependent suppression of the S100A6 promoter, electrophoretic mobility shift as-says were performed using probes identical to the putative Sp1-binding sites present in the GC-rich region of the S100A6 promoter (two overlapping probes covering together the whole GC-rich region), and to the putative NFκB-binding site located downstream of the transcription start site (+66 to +77).The other NFκB-binding site located at the -460 to -451 position (Joo et al., 2003) was not included in this analysis, since we had previously shown that deletion of this site had no influence on the S100A6 promoter suppression by p53.Experiments were performed with HeLa nuclear extracts made from cells transfected with p53 expression vector and from control cells weakly expressing endogenous p53 (Fig. 9A).A weak-to-strong binding of endogenous transcription factors to the putative Sp1-binding sites was detected (Fig. 9B, lanes 1 and 5, black arrow).The presence of overexpressed p53 in the binding reaction decreased the amount of Sp1-DNA complexes (Fig. 9B, lanes 1 vs. 3, 5 vs. 7, black arrow).No other changes in the binding pattern were observed in the pres-  ence of p53.A specific binding to the NFκB putative binding site was also detected (Fig. 9C, lane 1, black arrow).As described above for Sp1 binding to DNA, in the presence of overexpressed p53 the binding of NFκB to its recognition site was decreased (Fig. 9C, lane 1 vs., 3, black arrow).The results of the assays performed with HeLa cell extract containing high amount of endogenous p53 induced by 6 µM camptothecin (Houser et al., 2001) were identical to these described above (not shown).

dISCuSSIon
In this work we show that the promoter of the S100A6 gene encoding calcyclin, a member of the S100 family of Ca 2+ -binding proteins, is suppressed by p53 in a dose-dependent manner.Such dependence is an important additional argument in the discussion concerning the role of calcyclin in cell cycle regulation.
Calcyclin is overexpressed in many types of human cancers, and sometimes is considered a tumor marker and diagnostic factor (Wojda & Kuznicki, 1993;Filipek & Kuznicki, 1993;Weterman et al., 1993;Berta et al., 1997;Komatsu et al., 2000;Kim et al., 2002;Jang et al., 2004;Vimalachandran et al., 2005).Since p53 accumulates in more than 50% of all cancers, the elevation of p53 should be accompanied by a corresponding decrease of calcyclin.However, as shown by us in this paper, such correlation might be true only for the cases with accumulation of the wild type p53, which is not common.This could be so because p53 is mutated in more than 50% of malignant tumors (Hernandez-Boussard et al., 1999;Strano et al., 2007;Petitjean et al., 2007).The majority of these mutations are missense mutations located in the DNAbinding domain (Martin et al., 2002;Olivier et al., 2004).Since trans-repression by p53 usually does not require DNA binding, these mutants may retain their ability to suppress transcription.However, the majority of the mutants analyzed so far by other authors have lost, at least in part, this ability (Unger et al. 1993;Harris et al., 1996;Wang & Beck, 1998;Subbaramaiah et al., 1999;Blagosklonny, 2000).Our data is consistent with their results, as we also showed that certain p53 mutants suppress the A100A6 promoter 6 to 9-fold weaker than the wild type p53.Therefore, an absence of correlation should be expected in cancer tissues, rather than an inverse correlation between p53 and calcyclin levels.In accordance with this expectation, in pancreas cancer, the only cancer type analyzed so far for the association between p53 and calcyclin, there was no correlation between nuclear calcyclin expression and p53 (Vimalachandran et al., 2005).This is consistent with the fact that in 50-80% of cancers of this organ, p53 mutants have been reported (Redston et al., 1994;Iwao et al., 1998).
It has been shown by other authors that the suppression of transcription by p53 might be a result of the diverse mechanisms listed in the introduction (Subbaramaiah et al., 1999;Xu et al., 2000;Li & Lee, 2001;St Clair et al., 2004;Sengupta et al., 2005), and might be mediated by different proteins (Seto et al., 1992;Truant et al., 1993;Ragimov et al., 1993;Xiao et al., 1994;Farmer et al., 1996;Webster et al., 1996;Bargonetti et al., 1997;Ohlsson et al., 1998;Subbaramaiah et al., 1999;Xu et al., 2000;Zhang et al., 2000).Our data indicate that the S100A6 promoter is suppressed by p53 due to its interference with the function of Sp1 and NFκB.Therefore, the S100A6 promoter suppression by p53 is, in fact, the result of insufficient activation by the Sp1 and NFκB transcription activators.Electrophoretic mobility shift assays point to disturbed binding of these transcription factors to the S100A6 promoter in the presence of p53: lower amounts of the Sp1-DNA and NFκB-DNA complexes are formed in the presence of p53 than in its absence.This indicates that either Sp1 and NFκB decrease their affinity for DNA or are sequestered by p53, most possibly due to the formation of p53-Sp1 and p53-NFκB heterocomplexes.Such interaction between p53 and Sp1 has been previously described by other authors studying suppression of transcription by p53 (Bargonetti et al., 1997;Ohlsson et al., 1998;Xu et al., 2000;Sengupta et al., 2005).
In conclusion, we propose that the suppression of the S100A6 promoter and the resulting decrease of the amount of calcyclin could well be a mechanism by which p53 inhibits cell proliferation.The weak or no suppression of the S100A6 promoter by mutant p53 may be unable to initiate such a mechanism in cancers, leading to overexpression of calcyclin and, possibly, to the deregulation of cell cycle control.However, to confirm this hypothesis, a detailed analysis of p53-calcyclin interrelationship throughout the cell cycle should be performed.

Figure 1 .
Figure 1.Human calcyclin (S100A6) promoter is inhibited by p53.(A)Immunoblots of whole-cell extracts made from control HeLa cells, and from HeLa cells overexpressing p53.The blots were probed with anti-p53 antibody and re-probed with anti-β-actin antibody.(B) HeLa cells transfected with CMV-p53 vector overexpress transcriptionally active p53 protein.HeLa cells were transfected with pGL3-p53BS reporter vector with the promoter containing consensus p53 binding sites, as well as with CMV-p53 expression vector encoding wild type p53.(C) The S100A6 promoter is inhibited by p53 in a dose-dependent manner.HeLa cells were transfected with pGL2-CACY(-1371/+134) reporter vector, with increasing amounts of CMV-p53 expression vector or with equal amounts of 'empty' pcDNA3.1(+)vector, and with phRL-CMV internal control vector.Firefly luciferase activity was measured 24 h post-transfection, and the results normalized against Renilla luciferase activity.Bars represent the mean result (± standard deviation) of 9 experiments performed with two different DNA preparations.

Figure 3 .
Figure 3. p53 mutants suppress the S100A6 promoter 6 to 9-fold less than wild type p53.(A) Immunoblots of protein extracts made from control HeLa cells as well as HeLa cells overexpressing wild type and mutant p53.The blots were probed with anti-p53 antibody and re-probed with anti-β-actin antibody.(B) p53 mutants weakly inhibit the S100A6 promoter.HeLa cells were transfected with pGL2-CACY(-1371/+134) reporter plasmid, with CMV-p53 expression plasmid encoding either wild type or mutant p53, and with phRL-CMV internal control vector.Firefly luciferase activity was measured 24 h post-transfection, and the results were normalized against Renilla luciferase activity.Bars represent the mean result (± standard deviation) of 6 experiments.

Figure 4 .
Figure 4. deletion of three out of four promoter fragments similar to p53-specific sequences has no effect on S100A6 promoter inhibition by p53.(A) Schematic presentation of the -1371 to +134 S100A6 promoter fragment and the localization of the selected putative transcription factor binding sites.The arrows point to 5' ends of the deletion mutants.+1, transcription start site.(B) The -167 to +134 fragment contains all sites mediating p53 inhibitory effect on the S100A6 promoter.HeLa cells were transfected with pGL2 reporter construct containing different deletion mutants of the S100A6 promoter, with CMV-p53 expression plasmid encoding wild type human p53 or 'empty' pcDNA3.1(+)vector, and with phRL-CMV internal control vector.Firefly luciferase activity was measured 24 h post-transfection and the results were normalized against Renilla luciferase activity.Bars represent the mean result (± standard deviation) of 9 experiments performed with two different DNA preparations.

2008W.
Kroliczak and others    the level of promoter suppression by p53 from 8 to 2-fold.

Figure 6 .
Figure 6.overexpression of Sp1 partially counteracts the p53 inhibitory effect on the S100A6 promoter.(A) Immunoblots of whole-cell extracts made from control HeLa cells as well as HeLa cells overexpressing p53 and different amounts of Sp1 were probed with anti-Sp1 antibody.Duplicate blots were re-probed with anti-β-actin antibody.(B) Overexpression of Sp1 counteracts the p53-dependent inhibition of the S100A6 promoter in a dose-dependent manner.HeLa cells were transfected with pGL2-CACY(-1371/+134) reporter plasmid, with CMV-p53 expression plasmid encoding wild type human p53, with different amounts of pcDNA3.1(+)-Sp1expression plasmid encoding wild type human Sp1, with phRL-CMV internal control vector, and, to equalize the total amount of DNA in each sample, with different amounts of pcDNA3.1(+).Firefly luciferase activity was measured 24 h post-transfection and the results were normalized against Renilla luciferase activity.Bars represent the mean result (± standard deviation) of 9 experiments.

Figure 7 .
Figure 7. Overexpression of NFκB partially counteracts the p53 inhibitory effect on the S100A6 promoter.(A) Immunoblots of whole-cell extracts made from control HeLa cells as well as HeLa cells overexpressing p53 and NFκB p50 and p65.The blots were probed with anti-NFκB p65 antibody.Duplicate blots were probed with anti-β-actin antibody.(B) Overexpression of NFκB counteracts the p53-dependent inhibition of the S100A6 promoter in a dose-dependent manner.HeLa cells were transfected with pGL2-CACY(-1371/+134) reporter plasmid, with CMV-p53 expression plasmid encoding wild type human p53, with different amounts of pcDNA3.1(+)-NFκBp50 and of pcD-NA3.1(+)-NFκBp65 expression plasmids encoding wild type human NFκB p50 and p65 proteins, respectively, with phRL-CMV internal control vector, and, to equalize the total amount of DNA in each sample, with different amounts of 'empty' pcDNA3.1(+).Firefly luciferase activity was measured 24 h post-transfection and the results were normalized against Renilla luciferase activity.Bars represent the mean result (± standard deviation) of 9 experiments.

Figure 9 .
Figure 9. p53 decreases Sp1 and NFκB binding to their recognition sites in the S100A6 promoter.(A) Immunoblot of protein extracts made from control HeLa cells as well as HeLa cells overexpressing p53.The blot was probed with anti-p53 antibody and re-probed with anti-β-actin antibody.(B) Electrophoretic mobility shift assays performed with control HeLa nuclear extracts and with nuclear extracts made from HeLa cells transfected with p53 expression plasmid and the two overlapping probes identical to the S100A6 promoter fragment containing putative Sp1-binding sites (GC-rich region located at -100 to -63 position).(C) Electrophoretic mobility shift assays performed with control HeLa nuclear extracts or with nuclear extracts made from HeLa cells transfected with p53 expression plasmid, and with the probe identical to the putative NFκB-binding site located at +66 to +77 position of the S100A6 promoter.S, specific competitor (cold probe).Arrows, specific shifted bands.