Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The efficient transcription and replication of DNA requires changes in its double-strand topology at specific times during the cell cycle. These topological alterations are carried out by the ubiquitously expressed, homodimeric nuclear protein, topoisomerase II (topo II). Topo II can relieve supercoiling that results during DNA replication and is an essential enzyme for decatenation of sister chromatids at mitosis. The action of topo II involves the cleavage of one DNA double strand, the passage of a transfer double strand through the break, and religation of the cleaved DNA. During the decatenation cycle, topo II is covalently bound to the cleaved DNA strand, forming an intermediate topo II-DNA cleavable complex. The expression of topo II is lowest in the G₁ phase, increases as the cells traverse S phase, and reaches a maximum at the G₂/M phase interface (Woessner et al., 1991). This observation is consistent with the association of increased topo II with DNA replication, mitosis, and proliferation.

Various transcriptional mechanisms regulate the expression of topo II. The promoter region of the topo II gene contains five inverted CCAAT boxes (ICBs), two GC boxes, and an ATF site (Hochhauser et al., 1992). Topo II expression is increased in rapidly proliferating cells after exposure to heat shock (Matsuo et al., 1993) and by the ras oncogene (Chen et al., 1999). On the other hand, topo II expression is down-regulated by wild-type p53 (Wang et al., 1997) or confluence-induced growth arrest (Isaacs et al., 1996). Specific ICBs have been implicated in the regulation of topo II expression by confluence arrest (Isaacs et al., 1996) heat-shock response (Furukawa et al., 1998), p53 (Wang et al., 1997), and cell cycle dependence (Falck et al., 1999).

Topo II is an important target for a variety of clinically useful anticancer agents. The topo II poisons (i.e., etoposide, VP-16) act by stabilizing the topo II-DNA cleavable complex (Corbett and Osheroff, 1993), which results in DNA double strand breaks. The topo II catalytic inhibitors (i.e., aclarubicin) inhibit topo II activity at a step other than the formation of the cleavable complex (Drake et al., 1989).

Whereas increased levels of topo II correlate with increased cellular proliferation, induction of the tumor suppressor p53 leads to negative regulation of proliferation. After DNA damage, the activity and level of p53 increases significantly, which may lead to either cycle arrest at the G₁ checkpoint or apoptotic cell death (Kastan et al., 1991). A structural change in p53 facilitates its sequence-specific binding to DNA (Cho et al., 1994) and increases expression of genes involved in cycle arrest or apoptosis. Gene expression is up-regulated by wild-type (wt) p53 for genes such as GADD45 (Kastan et al., 1992), p21^Waf1/Cip1 (El-Deiry et al., 1993), mdm-2 (Momand et al., 1992), and cyclin G (Okamoto and Beach, 1994) that contain a p53 binding site. In contrast, genes lacking a p53 consensus binding site, like c-fos (Kley et al., 1992), mdr1 (Chin et al., 1992), hsp70 (Agoff et al., 1993), and O⁶-methylguanine-DNA methyltransferase (Harris et al., 1996) are down-regulated by wt p53. Evidence suggests the repression by wt p53 results from its direct interaction with factors such as TATA-binding protein (Liu et al., 1993), Sp1 (Borellini and Glazer, 1993), CCAAT binding factor (Agoff et al., 1993), and transcriptional coactivators such as p300/cAMP-response element-binding protein (Ravi et al., 1998) and p300/cAMP-response element-binding protein-associated factor (Scolnick et al., 1997). On the other hand, mutant p53 exhibits an attenuated transcriptional repression activity that may reflect a lack of association/interaction with these or other transcription factors (Zambetti and Levine, 1993).

Nuclear factor-Y (NF-Y) is a heterotrimeric protein composed of NF-YA, NF-YB, and NF-YC subunits (Maity and de Crombrugghe, 1998). It functions as a transcription factor whose DNA binding domain is created by the interaction of highly conserved regions located in the three subunits (Maity et al., 1992). NF-Y specifically recognizes a CCAAT box motif found in the promoter and enhancer regions of many genes (Mantovani, 1998). These sites are typically located in the proximal promoter region from 80 to 60 bp upstream of the transcription start site. Biochemical studies demonstrate that the DNA binding domain of the NF-YB and NF-YC subunits associate through a protein-protein histone-fold "handshake" motif (Sinha et al., 1996). The NF-YA subunit interacts only with the NF-YB:NF-YC heterodimer, suggesting that the NF-YB:NF-YC histone-fold is critical in creating a functional NF-Y CCAAT box DNA binding complex. In this report, we describe experiments in which wt p53 and certain anticancer drugs induce the down-regulation of topo II gene expression through inhibition of the binding of NF-Y to specific ICBs in the topo II promoter.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials. All cell culture media and supplements were purchased from Invitrogen (Carlsbad, CA), BioWhittaker (Walkersville, MD), or Atlanta Biologicals (Norcross, GA). ³²P-Labeled deoxycytidine triphosphate was purchased from PerkinElmer Life Sciences (Boston, MA). Mouse anti-NF-YA (clone YA-1a) was purchased from BD Pharmingen (San Diego, CA). The ICBp90 antibody was a gift from Dr. Christian Bronner (Strasbourg, France) (Hopfner et al., 2000). Unless otherwise specified, all other antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). VM-26 (teniposide) was a gift from Bristol-Myers Squibb Co. (Princeton, NJ). Unless otherwise indicated, all other chemicals were purchased from Sigma Chemical Co. (St. Louis, MO).

Plasmids and Cell Culture. The human p53 expression plasmids (a gift from Gerard Zambetti, St. Jude Children's Research Hospital, Memphis, TN) and the pCMV-Neo-Bam control vector used in this study have been described previously (Wang et al., 1997). The mutant p53-22/23 vector contains mutations at amino acids 22 and 23 of the expressed p53 protein. The dominant negative NF-YA vector (4-YA13m29) (a gift from Dr. Roberto Mantovani, University of Milan, Italy) contains mutations in three amino acids in the DNA binding domain (Mantovani et al., 1994). The (10)1 cell line (a gift from Gerard Zambetti) is a spontaneously immortalized murine BALB/c embryo fibroblast line, containing large deletions in both p53 alleles; consequently, it is completely deficient in p53 protein. The (10)1val cell line was developed by transfection of (10)1 cell with a temperature-sensitive p53 expression vector (Wang et al., 1997, and references within). The p53 protein is predominately in the wt conformation at 32°C, and in the mutant conformation at 39°C. The mouse NIH3T3 cell line was obtained from American Type Culture Collection (Manassas, VA). All cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (Atlanta Biologicals), 2 mM glutamine, 100 U/ml of penicillin, and 100 µg/ml of streptomycin, in humidified 5% CO₂/95% air atmosphere at 37°C.

Transient Transfection of (10)1 Cells. (10)1 cells were cultured in 150-mm plates at a cell density of 2.5 × 10⁶ cells/plate for 24 h. A mixture of 2 µg of the control vector, wt p53 vector, or mutant p53-22/23 vector DNA and 50 µl of LipofectAMINE (Invitrogen) (1 µg/50 µg) were incubated at room temperature in 10 ml of DMEM without serum for 30 min. Culture media on the cells was replaced with the DNA-LipofectAMINE mixture, the cells were incubated for 5 h at 37°C, and then 20 ml of DMEM with serum was added without aspirating off the DNA-LipofectAMINE mixture. Twenty-four hours after transfection, the media was aspirated off, the cells washed once with 1× phosphate-buffered saline, and fresh DMEM with serum was added. The cells were allowed to grow for an additional 24 h before being harvested with trypsin-EDTA for preparation of nuclear extract.

Preparation of Nuclear Extracts. After transient transfection, cells were harvested and centrifuged at 1000 × g for 4 min. The cell pellet was resuspended in hypotonic buffer and the nuclear proteins were extracted as described previously (Danks et al., 1988). In all buffers, protease inhibitors were added just before use: phenylmethylsulfonyl fluoride and benzamidine at 1 mM each, aprotinin, soybean trypsin inhibitor and leupeptin at 10 µg/ml, and pepstatin A at 1 µg/ml. Protein concentrations were determined by the Bio-Rad protein assay (Bio-Rad, Hercules, CA). The nuclear extracts were stored in aliquots at 80°C.

Preparation of Cell Lysates. After the designated drug treatment, cells were harvested and collected by centrifugation at 1000 × g for 4 min. The cell pellet was resuspended in 120 µl RIPA buffer (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 0.1% SDS, 0.5% deoxycholic acid and 1% NP-40). The lysed cells were homogenized by multiple passes through a pipette tip, incubated on ice for 30 min, and then centrifuged at 12,500 rpm for 10 min. The supernatant was collected as the cell lysate and the protein concentration determined by the Bio-Rad protein assay.

Electrophoretic Mobility Shift and Supershift Assays. The human topo II CCAAT box oligomers used in these experiments were as shown in Fig. 1. The oligomers were annealed to their respective complementary strands designed to create 2 to 4 base overhangs at the 5' end. The resulting overhangs were filled in with ³²P-dCTP using Klenow fragment of DNA polymerase and the incorporated dCTP determined by scintillation counting. Nuclear proteins (1.5 µg) from the (10)1 cell line transfected with either the wt p53, mutant p53 or control vector, along with poly dIdC (1 µg) (Roche Molecular Biochemicals, Indianapolis, IN) and 70,000 cpm/lane of labeled ICB oligonucleotide were incubated at room temperature for 30 min in a binding buffer consisting of 20 mM HEPES pH 7.6, 0.1 mM EDTA, 1 mM DTT, 10% glycerol and 50 mM NaCl. For the supershift assays 1 µl of the specific antibody was also included in the binding reaction with the labeled oligomer and the nuclear extract. To check for background protein binding, 10 µg of BSA was added to specific control samples. For competition studies, 1 µl (10 fold) of the unlabeled oligo was included in the binding reaction. The DNA-protein complexes were resolved on a 6.5% nondenaturing polyacrylamide gel run in Tris-glycine buffer (25 mM Tris, pH 8.5, 200 mM glycine and 1 mM EDTA) at 4°C. The gels were dried and exposed to Kodak BIOMAX MR film (Eastman Kodak Co.) with intensifying screens at 80°C. In the studies with drug-treated cells, the following drugs were used: vinblastine, etoposide (VP-16), amsacrine (mAMSA), teniposide (VM-26), aclarubicin, cisplatin (CDDP), camptothecin, mitoxantrone, or ellipticine. The cells were exposed to the drugs for 20 h.

View larger version (34K): [in this window] [in a new window]   Fig. 1.   Oligomers containing ICBs for topo II. Numbers indicate base positions relative to the topo II transcription start site (see Hochhauser et al., 1992). ICBs are underlined and mutations are indicated by boxed small letters.

SDS-Polyacrylamide Gel Electrophoresis and Western Blotting. For detection of p53 protein, 30 µg of nuclear extract protein was denatured by boiling in loading buffer for 5 min and loaded onto a 12% polyacrylamide gel containing 0.1% sodium dodecyl sulfate (SDS). The gel was run at 200 V for 30 min. The separated proteins were transferred onto a nitrocellulose membrane by electroblotting. The gel was subsequently stained to confirm the equivalent loading and transfer of proteins. After being blocked in 5% nonfat dry milk, the nitrocellulose membrane was incubated overnight in a 1:500 dilution of rabbit anti-human p53 antibody. After incubation with goat anti-rabbit alkaline phosphatase antibody (1:500 dilution) the bound antibody was visualized by reaction with 5-bromo-4-chloro-3-indolyl phosphate and nitroblue tetrazolium. For NF-Y protein, 40 mg of cell lysate protein was loaded onto a 7.5% polyacrylamide minigel and run at 150 V for 1.0 h. The nitrocellulose membrane was incubated overnight in a 1:1000 dilution of mouse anti-NF-YA antibody. After incubation with an anti-mouse alkaline phosphatase antibody (1:1000 dilution), the bound antibody was visualized as above.

Luciferase Assays. (10)1 cells were cultured in 12 well plates (7 × 10⁴ cells/well) for 24 h before transfection as described above with slight modifications. Briefly, 1 µg of the topo II-luciferase reporter construct DNA and 0.5 µg of the dominant negative NF-YA vector (4YA13m29) and/or 0.1 µg of the wt p53 vector was mixed with 4 µl LipofectAMINE (1 µg/2.7 µl) in 0.5 ml of DMEM without serum for 30 min at room temperature. The DNA-LipofectAMINE mixture was added to the cells for 2.5 h and the transfections completed as described above. After 24 h in fresh media, cells were washed in phosphate-buffered saline, lysed with 200 µl of passive lysis buffer (Promega), and centrifuged for 10 s to remove cell debris. The cell lysate was analyzed for luciferase activity using the Promega luciferase substrate as described by the supplier. Luciferase activity was measured on a luminometer (Turner Designs). The lysate protein concentration was determined using the Bio-Rad protein dye reagent.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Binding of Transcription Factors to the ICBs of the Topo II Promoter Is Inhibited by the Expression of wt p53. Several laboratories have documented the vital role of multiple ICBs in the topo II promoter for expression of the topo II gene in both human and rodent cell lines (Ng et al., 1995; Herzog and Zwelling, 1997; Wang et al., 1997; Takano et al., 1999). Earlier studies from our lab have demonstrated that successive deletions of each of the five ICBs from the topo II promoter results in progressively reduced promoter activity in mouse embryo fibroblast (10)1 cells and that the activity of the topo II promoter is specifically inhibited by wt p53 but not by mutant p53 (Wang et al., 1997 and Table 1). Specific mutations disrupting the CCAAT sequence of certain ICBs eliminated the p53-induced inhibition of topo II promoter activity. The effect of p53 expression on the binding of transcription factors to the topo II promoter was studied using nuclear extracts from the p53-null (10)1 cells transiently transfected with control, wt, or mutant p53-expressing vector. Electrophoretic mobility shift assays were conducted with labeled oligomers representing the five ICBs of the topo II promoter to ascertain distinctions in factor binding.

View this table: [in this window] [in a new window]   TABLE 1 Effects of wt and mutant p53 on topo II promoter constructs in (10)1 cells (10)1 cells were cotransfected with topo II promoter constructs and either wt or mutant p53 expression vector. p53 values are expressed as a percentage of the luciferase activity in lysates from cells cotransfected with the control CMV-neo-Bam vector only. The values are the mean ± 1 S.D. of activities from triplicate samples in a representative experiment.

View larger version (57K): [in this window] [in a new window]   Fig. 2.   wt p53, but not mutant p53, inhibits binding of NF-Y to topo II ICBs. Nuclear extracts were prepared from p53-null (10)1 cells that had been transfected with the control vector (no p53), the wt p53 expression vector, or the mutant p53-22/23 expression vector. In three separate experiments, nuclear extract proteins (1.5 µg) were incubated with labeled duplex oligomers that contained the sequence for the indicated ICB, and gel mobility shift assays were performed as described under Materials and Methods. The arrow indicates the position of the NF-Y band, and the arrowhead indicates the position of the NF-Y band supershifted with antibody to NF-YA.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   wt p53, but not mutant p53, inhibits binding of NF-Y to topo II ICB-2. Nuclear extracts were prepared from p53-null (10)1 cells that had been transfected with the control vector (no p53), the wt p53 expression vector, or the mutant p53-22/23 expression vector. Nuclear extract proteins (1.5 µg) were incubated with labeled duplex oligomers that contain the sequence for ICB-2, and gel mobility shift assays were performed as described under Materials and Methods. The arrow indicates the position of the NF-Y band, and the arrowhead indicates the position of the NF-Y band supershifted with antibody to NF-YA.

View larger version (36K): [in this window] [in a new window]   Fig. 4.   wt p53, but not mutant p53, inhibits binding of a transcription factor to ICB-5 of the topo II promoter. Nuclear extracts were prepared from p53-null (10)1 cells that had been transfected with the control vector (no p53), the wt p53 expression vector, or the mutant p53-22/23 expression vector. Nuclear extract proteins (1.5 µg) were incubated with labeled duplex oligomers that contained the sequence for ICB-5, and gel mobility shift assays performed were as described under Materials and Methods. The arrow indicates the position of the major unknown band binding to ICB-5 that is inhibited by p53.

Inhibition of Topo II Promoter Activity by Dominant-Negative NF-Y. The present study shows that transient transfection with wt p53 inhibits the binding of NF-Y to the ICBs of the topo II promoter. Our previous results demonstrate that transient transfection with wt p53 inhibits topo II promoter activity (Wang et al., 1997). We wanted to determine whether the inhibition of topo II promoter activity by wt p53 was related to the ability of p53 to inhibit the binding of NF-Y to the ICBs of the topo II promoter. To further confirm the relationship of inhibition of NF-Y binding to decreased topo II promoter activity, a dominant negative NF-Y vector (4YA13m29) was transfected into (10)1 cells with or without cotransfection of wt p53. Luciferase assays were performed to examine the effect of the mutant NF-Y vector on topo II promoter-luciferase reporter expression. The dominant-negative NF-YA can associate with the endogenous NF-YB and NF-YC, but the resultant heterotrimer does not bind to ICBs to activate transcription. Thus expression of the dominant-negative NF-YA should inhibit the binding of a functional NF-Y complex to the ICBs of the topo II promoter in a manner independent of the effect of wt p53.

View larger version (54K): [in this window] [in a new window]   Fig. 5.   Both wt p53 and dominant-negative NF-Y decrease topo II promoter activity. Cell lysates were prepared from p53-null (10)1 cells that had been transfected with a topo II promoter- luciferase vector and the wt p53 expression vector, the dominant negative NF-Y expression vector, or both. The luciferase activity was determined as described under Materials and Methods, and is expressed as a percentage of the activity of the specific topo II promoter construct alone. The positions of the ATTGG comprising ICB-1 to -5 are 68, 108, 175, 259, and 389, respectively (see Fig. 1). The number of the topo II construct indicates the 5'-most base included in that construct. The number of ICBs in each of the constructs is indicated at the bottom of the graph.

Effect of Anticancer Drugs on the Binding of NF-Y to the ICBs of Topo II in Cells Containing Endogenous p53. Many of the topo II-targeted anticancer drugs result in DNA double-strand breaks. Other anticancer drugs cause DNA damage by alternative means, such as the blocking of mitosis or inhibition of topoisomerase I activity. These types of DNA damage are known to induce p53, leading to cell-cycle arrest and apoptosis. Exogenous expression of wt p53 inhibits the binding of NF-Y to the ICBs of the topo II promoter and decreases topo II expression. Therefore, one might predict that treatment of cells containing a functional wt p53 gene with anticancer drugs would have similar inhibitory effects on NF-Y binding. Nuclear extracts were prepared from NIH3T3 cells treated with various anticancer drugs for 20 h. Mobility shift assays with the ICB-1 oligomer of the topo II promoter and nuclear extracts from NIH3T3 cells treated with vinblastine, aclarubicin, VP-16, or VM-26 exhibited a significant decrease in NF-Y binding (Fig. 6). Treatment of NIH3T3 cells with m-AMSA, mitoxantrone, ellipticine, and camptothecin induced a moderate decrease in NF-Y binding. Nuclear extracts from cisplatin-treated NIH3T3 cells did not exhibit any decrease in NF-Y binding to ICB-1. In lane 7, antibody to NF-Y was added to the nuclear extract to confirm that the decreased band represents NF-Y. This decrease in NF-Y binding with treatment of NIH3T3 cells with anticancer agents correlates directly with the effect of VM-26 treatment on topo II promoter activity. We have found that treatment of NIH3T3 cells with 5 µM VM-26 can inhibit promoter activity up to 75%, as measured by relative luciferase activity with the topo II promoter constructs containing three or four ICBs (see Table 1).

View larger version (33K): [in this window] [in a new window]   Fig. 6.   Treatment of NIH3T3 cells with anticancer drugs can block the binding of NF-Y to ICBs of the topo II promoter. Nuclear extracts were prepared from NIH3T3 cells that had been treated for 20 h with the indicated drug. The nuclear proteins (1.5 µg) were incubated with labeled duplex oligomer that contain the sequence for ICB-1, and gel mobility shift assays were performed as described under Materials and Methods. The arrow indicates the position of the NF-Y band, and the arrowhead indicates the position of the NF-Y band supershifted with antibody to NF-YA. The lanes are: 1, no nuclear extract; 2, no drug; 3, vinblastine (10 µM); 4, VP-16 (10 µM); 5, m-AMSA (5 µM); 6, VM-26 (5 µM); 7, VM-26 (5 µM) plus NF-YA antibody; 8, aclarubicin (5 µM); 9, cisplatin (10 µM); 10, camptothecin (5 µM); 11, mitoxantrone (1 µM); 12, ellipticine (5 µM).

Effect of Anticancer Drugs on the Binding of NF-Y to the ICBs of Topo II in Nuclear Extracts from p53-Null Cells. From our initial experiments, it seemed that the inhibition of NF-Y binding to the ICBs of the topo II promoter occurs because of the expression of wt p53. However, with the exception of vinblastine- and aclarubicin-treated cells, western blots with nuclear extracts from drug-treated NIH3T3 cells did not show a detectable induction of p53 protein expression (Fig. 7). This unanticipated result led us to question whether the induction of p53 in drug-treated cells was essential for the inhibition of NF-Y binding to ICBs. Gel-shift assays were conduced with the ICB-1 oligomer and nuclear extracts from the p53-null (10)1 cells treated with anticancer drugs for 20 h. The drug concentrations used were the same as those in the NIH3T3 experiments, except that 0.1 µM aclarubicin was used instead of 5 µM. Interestingly enough, nuclear extracts from the drug-treated p53-null (10)1 cells also exhibited an inhibition of NF-Y binding to ICB-1 as seen in the studies with drug-treated NIH3T3 cells (Fig. 8). In lane 7, antibody to NF-Y was added to the nuclear extract to confirm that the decreased band represents NF-Y.

View larger version (16K): [in this window] [in a new window]   Fig. 7.   Effect of drug treatment on p53 expression of NIH3T3 cells. Cell lysates were prepared from NIH3T3 cells that had been treated for 20 h with the indicated drug. A Western blot with antibody to p53 was performed as described under Materials and Methods. The lanes are: 1, no drug; 2, VP-16 (15 µM); 3, m-AMSA (10 µM); 4, cisplatin (30 µM); 5, ellipticine (1 µM); 6, vinblastine (10 µM); and 7, aclarubicin (0.1 µM).

View larger version (25K): [in this window] [in a new window]   Fig. 8.   Treatment of p53-null (10)1 cells with anticancer drugs can block the binding of NF-Y to ICBs of the topo II promoter. Nuclear extracts were prepared from (10)1 cells that had been treated for 20 h with the indicated drug. The nuclear proteins (1.5 µg) were incubated with labeled duplex oligomer that contain the sequence for ICB-1, and gel mobility shift assays were performed as described under Materials and Methods. The arrow indicates the position of the NF-Y band, and the arrowhead indicates the position of the NF-Y supershifted with antibody to NF-YA. The lanes are: 1, no drug; 2, vinblastine (10 µM); 3, VP-16 (10 µM); 4, m-AMSA (5 µM); 5, VM-26 (5 µM); 6, VM-26 (5 µM) plus NF-YA antibody; 7, cisplatin (10 µM); 8, aclarubicin (0.1 µM); 9, camptothecin (5 µM); 10, mitoxantrone (1 µM); 11, ellipticine (5 µM).

Effect of Anticancer Drugs and wt p53 Expression on the Endogenous Levels of NF-Y Protein. We have demonstrated thus far that both cellular expression of wt p53 and treatment of cells with specific anticancer drugs results in a substantial decrease in the binding of NF-Y to the ICBs of the topo II promoter. We next checked the protein levels of NF-Y in the treated cells to determine whether the decreased binding was the result of decrease levels of available NF-Y protein. Western blots were performed with cell lysates from the p53-null (10)1 cells treated with various drugs for 20 h or with lysates from the (10)1val cells incubated at either 32°C (wt p53) or 39°C (mutant p53) for 20 h. We did not observe a significant change in endogenous NF-Y protein levels in the drug-treated cells compared with the untreated cells (Fig. 9A). Similarly, there was no significant difference in the NF-Y protein levels in cells expressing wt p53 compared with those expressing mutant p53 (Fig. 9B), indicating that anticancer drugs or wt p53 expression do not affect the endogenous protein levels of NF-Y. For independent confirmation of the effect of cell-cycle arrest on NF-Y protein levels, (10)1 cells were serum-starved or treated with aphidicolin or nocodazole to block the cell cycle at distinct points. Serum-starved cells were arrested in G₁, before entry into S (71% in G₁), and aphidicolin-treated cells were blocked in both G₁ (52%) and S (42%) phases. Nocodazole treatment arrested 86% of cell in G₂. Western blots of cell lysates indicate that cell cycle arrest in any of these states does not significantly affect the protein levels of NF-Y (Fig. 9C).

View larger version (23K): [in this window] [in a new window]   Fig. 9.   A, treatment of (10)1 cells with anticancer drugs does not alter the endogenous expression levels of NF-Y protein. Cell lysates were prepared from (10)1 cells that had been serum-starved or -treated for 20 h with the indicated drug. A Western blot with antibody to NF-YA was performed as described under Materials and Methods. The arrow indicates the position of the NF-Y protein band. The lanes are: 1, no drug; 2, vinblastine (10 µM); 3, VP16 (10 µM); 4, mAMSA (10 µM); 5, cisplatin (10 µM); 6, aclarubicin (0.1 µM); 7, camptothecin (5 µM); 8, mitoxantrone (1 µM); and 9, ellipticine (5 µM). B, wt and mutant p53 expression does not affect the endogenous expression levels of NF-Y protein. Cell lysates were prepared from (10)1val cells that had been incubated at either 32°C (wt p53) or 39°C (mutant p53). A Western blot with antibody to NF-YA was performed as described under Materials and Methods. The arrow indicates the position of the NF-Y protein band. The lanes are: 1, (10)1val39; 2, (10)1val32. C, drug treatment or serum starvation of (10)1 cells does not change the endogenous expression levels of NF-Y protein. Cell lysates were prepared from (10)1 cells that had been serum-starved or -treated for 20 h with the indicated drug. A Western blot with antibody to NF-YA was performed as described under Materials and Methods. The arrow indicates the position of the NF-Y protein band. The lanes are: 1, untreated; 2, nocodazole (15 µM); 3, aphidicolin (5 µg/ml); 4, no serum.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Using promoter deletion constructs and mutated ICB consensus sites, it has been well documented that the ICBs in the topo II promoter are required for optimal transcriptional activity (Hochhauser et al., 1992; Ng et al., 1995; Park et al., 1995; Herzog and Zwelling, 1997). Isaacs et al. (1996) first indicated that the transcription factor complex NF-Y binds to the ICBs of the topo II promoter, and this NF-Y-specific binding has been confirmed by others (Herzog and Zwelling, 1997; Wang et al., 1997), including this present report. Work in our laboratory and the laboratory of Ian Hickson has demonstrated that expression of wt p53, but not mutant p53, represses the transcription of the topo II promoter (Sandri et al., 1996; Wang et al., 1997). The present report provides data showing that the p53-induced decrease in topo II promoter activity results from the wt p53-dependent inhibition of transcription factor NF-Y binding to the four proximal ICBs of the topo II promoter. With each of the ICB oligomers tested, wt p53 was shown to decrease binding of NF-Y, but no decrease in binding was observed with mutant p53. The p53-22/23 mutant protein we have used in this study has a disruption in the transactivation domain of the protein. Previous studies in our lab with two other p53 mutants (p53-175 and p53-281) having alterations in their DNA binding domains also showed no significant inhibitory effect on topo II promoter activity (Wang et al., 1997). It follows that the ability to inhibit NF-Y binding to ICBs requires a p53 protein with functional transactivation and DNA binding domains. wt p53 transrepresses the human heat-shock promoter and is reported to physically interact with a CCAAT binding factor (Lum et al., 1990). The inability of mutant p53 to inhibit NF-Y binding may reflect a lack of association with such CCAAT binding factors.

The results shown in Fig. 5 confirm, through two independent means, that inhibition of NF-Y binding to the ICBs causes a decrease in topo II promoter activity. In this experiment we used the NF-YA plasmid, 4YA13m29, which expresses a dominant-negative mutant form of the NF-YA subunit containing alterations of three amino acids in the DNA binding domain. The dominant-negative NF-YA subunit forms a heterotrimer with the endogenous NF-YB and NF-YC subunits, but the resultant NF-Y complex cannot bind to the ICBs (Mantovani et al., 1994). Thus, the decrease in topo II promoter activity seen with coexpression of the dominant-negative NF-YA can be directly attributed to the decreased availability of viable NF-Y complex to bind the ICBs. The inhibition of topo II promoter activity seen with transfection of the dominant-negative NF-YA vector is strikingly similar in result, if not in precise mechanism, to the inhibition seen with transfection of wt p53. It is of note that simultaneous expression of both dominant-negative NF-YA and wt p53 does not produce significantly more promoter inhibition than either factor alone, indicating that either action can maximally reduce promoter activity to the basal promoter activity seen in the absence of NF-Y binding.

The inhibition of NF-Y binding to the ICBs of topo II is not caused by a decrease in the endogenous expression levels of NF-Y protein because both the anticancer drug-treated and wt p53-expressing cells exhibited no significant changes in NF-YA protein levels (Fig. 9, A and B). An earlier work by Bolognese et al. (1999) suggests that the regulation of the cyclin B2 promoter depends on cell-cycle regulated CCAAT-binding activity of NF-Y. The NF-Y protein, but not its mRNA, was found to be maximal in the mid-S phase and decreased in the G₂/M phase of the cell cycle. We used the drugs nocodazole and aphidicolin in addition to serum starvation to arrest cells in various phases of the cell cycle and found that NF-YA protein levels did not vary significantly under these conditions (Fig. 9C). Similar results recently reported by Jung et al. (2001) showed that the protein levels of all three subunits of NF-Y were unchanged by the expression of p53. These results indicate that inhibition of NF-Y binding induced by wt p53 or anticancer drugs does not occur because of a cell cycle-related decrease of NF-Y protein.

It is obvious from a comparison of Figs. 2 and 3 that there is a distinct difference in the strong binding affinity of NF-Y for ICB-1, -3, and -4, and the very weak affinity of NF-Y for ICB-2. For the detection of NF-Y binding to ICB-2 in Fig. 3, the autoradiogram was exposed for more than five times longer then was necessary for detection of binding to ICB-1. It is interesting to note that studies of the hamster topo II promoter show NF-Y to have a weaker binding affinity for ICB-2 and ICB-4 than the other ICBs, and mutations in ICB-2 cause only minimal decreases in promoter activity (Ng et al., 1995). This is not to say that ICB-2 is of lesser importance in the regulation of topo II. To the contrary, Isaacs et al. (1996) presented evidence that decreased binding of NF-Y to ICB-2 may play a key role in mediating down-regulation of topo II transcription in confluence-arrested cells, which is relieved in proliferating cells through the binding of NF-Y to the ICB-2. Thus, the relative contribution of an ICB to the regulation of topo II may depend on factors other than its binding affinity for NF-Y. The possibility of distinct roles for the ICBs is also supported by Falck et al. (1999), who report that ICB-1 may play an important role in the S phase-specific induction of topo II expression, and by Morgan and Beck (2001), who report that ICB-3 may play a specific role in topo II up-regulation in ICRF-187-resistant CEM leukemic cells.

The protein factor binding to ICB-5 is unknown at the present time. Despite having the required ATTGG sequence, NF-Y does not bind to the ICB-5 oligomer. However, the ICB-5-binding factor responds similarly to NF-Y in that its binding is reduced by the expression of wt p53 but not mutant p53. The presence of an AT-rich flanking sequence decreases the ability of NF-Y to bind to an ICB (Dr. Mantovani, personal communication). The ICB-5 of the human topo II promoter has such an AT-rich flanking sequence. ICB-5 is also unique with respect to its position in the promoter sequence. A sequence alignment of the human and hamster topo II promoters will show no comparable ATTGG site in the hamster to match the ICB-5 site in the human. The fifth ICB in the hamster coincides with ICB-4 in the human, whereas the positions of the first three ICBs are equivalent in both sequences (Ng et al., 1995).

To examine the possibility that treatment of cells with anticancer drugs could induce the expression of p53 and thus down-regulate topo II expression, we analyzed the effect of a panel of drugs on the binding of NF-Y to ICB-1 (Fig. 6). Nuclear extracts from NIH3T3 cells treated with the topo II-targeting drugs VM-26 or VP-16 exhibited decreased binding of NF-Y to the ICB. Treatment with cisplatin, which does not induce DNA breaks, did not result in any change in NF-Y binding to the ICB. Treatment with vinblastine, a mitotic spindle inhibitor, or aclarubicin, an inhibitor of topo II catalytic activity and not a cleavable complex-forming drug, caused inhibition of NF-Y binding. Aclarubicin has been shown to stabilize topo I cleavage resulting in single strand breaks (Nitiss et al., 1997). It has also been reported that vincristine, a mitotic spindle inhibitor similar to vinblastine, induces p53 in MCF7 cells (Vayssade et al., 2002). We would have expected that p53 would be expressed after treatment with the topo II-targeted drugs because of double strand breaks resulting from stabilization of the cleaved topo II-DNA complex. Western blot analysis of lysates from drug-treated cells confirmed that vinblastine and aclarubicin could induce the expression of p53 protein in the NIH3T3 cells. However, it was unexpected that p53 was not expressed at detectable levels after treatment with the topo II-targeted drugs under the conditions of this experiment.

To follow-up this observation of drug-induced inhibition of NF-Y binding, we treated the p53-null (10)1 cells with these same anticancer drugs. Even though the (10)1 cells are completely deficient for p53, we saw similar inhibition of NF-Y binding to ICBs of topo II when the cells were treated with the anticancer agents. These results are substantiated in an earlier study by Goldwasser et al. (1999) suggesting that down-regulation of topo II expression by ionizing radiation can occur independent of the p53 status of the cell. Thus, it seems that the inhibition of topo II by anticancer drugs could occur either by a p53-dependent and/or -independent mechanism. We are presently exploring the possibility that p53 transcriptional targets, such as p21^Waf1/Cip and 14-3-3, which function in DNA damage-induced G₂/M arrest, may be activated by alternative means in the drug-treated cells.