Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

In response to DNA damage induced by cytotoxic agents, the cell cycle is interrupted to prevent the replication of genomic errors. Cell cycle checkpoints are controlled by cyclin-dependent kinases (Cdks), cyclins, and Cdk inhibitors (Peter and Herskowitz, 1994). One of these inhibitors, p21, also called WAF1 (wild-type p53-activated factor) or CIP1 (Cdk-interacting protein), inhibits Cdk2, Cdk3, Cdk4, and Cdk6 to stop the cell cycle. Human p21 contains a highly conserved cysteine-rich region (amino acids 21-60) with a potential zinc finger domain. A second conserved region (amino acids 130-164) contains several putative nuclear localization signals (El-Deiry et al., 1993). p21 can also interact with proliferating cell nuclear antigen (PCNA), Gadd45, Cdk2, and cyclins, suggesting that it may coordinate DNA repair and replication in damaged cells (Xiong et al., 1993; Zhang et al., 1993). The expression of p21 is predominantly induced by p53, for example, after DNA damage (El-Deiry et al., 1993, 1994).

p21 and p53 are implicated in the response to chemotherapeutic agents, such as anthracyclines. These drugs intercalate into the DNA and induce distortion of the double helix, stabilization of the cleavable complex formed between DNA and topoisomerase II, and finally DNA breaks (Tewey et al., 1984). In response to DNA damage, p53 accumulates and functions as a sequence-specific DNA-binding protein, which positively regulates expression of several genes, including p21. Cells then undergo p21-dependent cell cycle arrest, which allows DNA damage repair (Fritsche et al., 1993; El-Deiry et al., 1994). However, in addition to direct transcriptional induction by p53, p21 gene expression can also be regulated by p53-independent mechanisms in response to serum, ultraviolet C (Macleod et al., 1995; Haapajarvi et al., 1999), cytokines, oxidative stress (Russo et al., 1995; Qiu et al., 1996), or growth factor stimulation (Michieli et al., 1994). On the other hand, p53 can inhibit cell cycle progression without inducing p21 expression (Hirano et al., 1995).

The rat, mouse, and human p21 promoters contain conserved putative p53 recognition sequences and one region that could possibly bind MyoD (El-Deiry et al., 1995). The human p21 promoter also harbors a TATA box, three STAT transcription factor binding sites, several Sp1 sites, and a vitamin D-responsive element (El-Deiry et al., 1993).

Nuclear factor-B (NF-B) is a transcriptional factor involved in the response to various stimuli and plays a central role in immune response and inflammatory reactions (for review, see Ghosh et al., 1998). NF-B activation in response to DNA-damaging agents might protect cells against apoptosis (Beg and Baltimore, 1996; Van Antwerp et al., 1996; Wang et al., 1996). In a cell type-specific manner, NF-B can also be implicated in cell cycle arrest or cell proliferation (Baldwin et al., 1991; Perkins et al., 1997; Guttridge et al., 1999).

We identified a novel potential NF-B binding site at position 2008 of the human p21 promoter and we showed the binding of NF-B proteins to this B site and the transactivation of this promoter by NF-B subunits. However, in breast and colorectal cancer cells, the transcriptional activation of p21 promoter and the cell cycle arrest induced by daunomycin is regulated by p53 and is NF-B-independent. Interestingly, the drug induces apoptosis through a p53- and NF-B-independent pathway.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Recombinant Plasmids. The p21-LUC (wild-type p21/WAF1 promoter-luciferase reporter), which contains a 2.4-kb genomic fragment from the wild-type p21 promoter in front of the luciferase gene, was kindly provided by Dr. B. Vogelstein (Johns Hopkins University, Baltimore, MD). The PMT2T expression vectors encoding the p65 (RelA), p50, p52, and c-Rel subunits of NF-B were previously described (Bours et al., 1992). The expression vector for wild-type p53 (pC53-SN3) was provided by Dr. U. Gullberg (University of Lund, Sweden), and the expression vector pCMV-HPV-16 E6 was provided by Dr. B. Vogelstein (Johns Hopkins University).

Cell Culture and Transfection. HCT15 human colon carcinoma cells (ATCC CCL225); HTM29 human colon adenocarcinoma cells (a gift from Dr. E. Kohn, National Institutes of Health, Bethesda, MD); SK-OV-3 human ovarian adenocarcinoma cells (ATCC HTB-77); HL-60 human promyelocyte cells (ATCC CCL-240); Jurkat human T lymphoma cells; LN18 glioma cells; and the human breast cancer cell lines MCF7, MCF7 MAD, and MCF7E6 were grown in RPMI-1640 medium supplemented with 1% L-glutamine, 1% antibiotics, and 10% fetal bovine serum (Life Technologies, Grand Island, NY). HT1080 human fibrosarcoma cells were grown in Dulbecco's modified Eagle's medium supplemented with 1% L-glutamine, 1% antibiotics, and 10% fetal bovine serum (Life Technologies). MCF7 MAD and MCF7E6 cell lines were grown in medium containing G418 (geneticin, 0.5 mg/ml active concentration; Roche, Mannheim, Germany). The MCF7 MAD and MCF7E6 cells, generous gifts from Dr. S. Chouaib (Institut Gustave Roussy, Villejuif, France), were derived from the MCF7 cell line stably transfected with an IB expression vector mutated at serines 32 and 36 (Cai et al., 1997b) or transfected with the HPV-E6 expression vector (Cai et al., 1997a), respectively. The HCT116 (ATCC CCL247), HCT116 MT9 human colon carcinoma cells and MDA-MB435 human breast cancer cell line were previously described (Hellin et al., 1998; Bentires-Alj et al., 1999).

Reporter Plasmid Assay. CAT activities were determined by the diffusion assay as previously described (Bours et al., 1992). The CAT activities were expressed as initial rates of chloramphenicol acetylation and normalized to the protein amounts quantified by the Micro BCA protein assay reagent (Pierce, Rockford, IL). LUC activities were determined by the Luciferase reporter gene assay kit (Roche) as recommended by the manufacturer and normalized to the amount of protein.

Protein Extraction and Western Blotting. Nuclear and cytoplasmic protein extracts were prepared as previously described (Hellin et al., 1998). Protein amounts were quantified with the Micro BCA protein assay reagent (Pierce). Western blotting was previously described (Hellin et al., 1998). Human p53 was detected with the monoclonal antibody Pab 1801 (Ab2; Oncogene Science, Cambridge, MA) and human p21 with the monoclonal antibody WAF1 (Ab1; Oncogene Science).

Electrophoretic Mobility Shift Assay (EMSA). EMSA and supershifting experiments were performed as previously described (Bentires-Alj et al., 1999).

Determination of Cellular Viability. Stably transfected cells or untransfected cells were seeded at the concentration of 7 × 10³ cells (HCT116, HCT116 MT9, and HCT15 cells) or 10⁴ cells (MCF7, MCF7 MAD, MCF7E6) per well in flat-bottomed 96-well plates in 0.2 ml of medium. After 48 h of incubation with the drug, cell viability was measured by a colorimetric assay based on the cleavage of the tetrazolium salt WST-1 by mitochondrial dehydrogenases (cell proliferation reagent WST1; Roche). For long-term cell survival, cells were seeded at the concentration of 500 cells (HCT116, HCT116 MT9) or 3000 cells (HCT15, MCF7, MCF7E6) per well in flat-bottomed 96-well plates in 0.2 ml of medium. After 7 days of treatment, cell viability was measured by the colorimetric assay WST-1.

FACS Analysis. Cells were washed in phosphate-buffered saline (Life Technologies) and harvested after dispase treatment (Dispase II; Roche), washed again in PBS, fixed in ice-cold 70% ethanol, and stored at 4°C overnight. After two other washes with PBS, cells were stained with propidium iodide in presence of RNase as described by the manufacturer (Cycle Test Plus DNA reagent kit; Becton Dickinson, San Jose, CA). Cell cycle analysis was performed using a FACStar Plus (Becton Dickinson) with a 100-mW air-cooled argon laser (Spinnaker 1161; Spectra Physics, Mountain View, CA) and the CellQuest software (Macintosh, Facstation; Becton Dickinson).

Clonogenic Survival Assay. Cells were seeded in six-well plates at the concentration of 30,000 cells (HCT116) or 50,000 cells (HCT15, MCF7, and MCF7E6) and were treated with daunomycin at different concentrations (0.1, 0.5, and 1 µM) for 48 h. Cells were then washed and incubated with fresh medium for 12 days. The number of colonies (>50 cells) was counted after staining with hematoxylin solution (0.1% v/v).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Enhanced p21 Expression in Daunomycin-Treated HCT116 and MCF7 Cells. To investigate p21 expression after a genotoxic stress, HCT116 and MCF7 cells, which harbor a wild-type p53 gene, were treated with daunomycin (1 µM). Western blotting analysis was performed with nuclear protein extracts and revealed with anti-p53 and anti-p21 monoclonal antibodies (Fig. 1A) and with an anti--actin antibody to confirm that equivalent amounts of protein extracts were loaded in each lane (data not shown). Daunomycin treatment induced a marked time-dependent increase in p53 with a peak occurring 6 (MCF7) or 24 h (HCT116 cells) after drug addition. This p53 accumulation was accompanied by a substantial increase in p21, already detectable after 3 h of daunomycin treatment and reaching a maximum after 6 h in MCF7 and 24 h in HCT116 cells.

View larger version (34K): [in this window] [in a new window]   Fig. 1.   Enhanced p21 expression in daunomycin-treated HCT116 and MCF7 cells. A, p53 and p21 protein accumulation after daunomycin treatment. HCT116 and MCF7 cells were treated with daunomycin (1 µM) for 3, 6, and 24 h. Nuclear protein extracts (10 µg) were analyzed by Western blotting using an anti-p53 or an anti-p21 antibody. B, effect of PDTC or NAC treatment on p21 induction by daunomycin. MCF7 cells were incubated with PDTC (60 µM) or NAC (10 mM) for 1 h before treatment with daunomycin (1 µM). Nuclear extracts were prepared after 3 and 6 h of daunomycin treatment and analyzed as described in A.

NF-B DNA Binding to p21 Promoter Sequences. We localized a novel NF-B binding site in the human p21 promoter at position 2008. To determine whether NF-B was involved in daunomycin-induced p21 expression, we studied first the binding of NF-B complexes to p21 promoter sequences. EMSA showed that nuclear extracts from daunomycin-treated HCT116 cells contained NF-B complexes that bound to the B site from the p21 promoter (Fig. 2A, lanes 1-4). Supershifting experiments performed with antibodies directed toward NF-B subunits (p50, p65, p52, c-Rel, RelB) demonstrated that the complex was mainly formed of p50/p65 heterodimers (Fig. 2A, lanes 8-13). EMSA was repeated with a consensus NF-B probe and confirmed that NF-B binding was increased after 3 and 6 h of daunomycin treatment and returned to basal level after 24 h (Fig. 2B). The affinity of the binding is higher with the consensus probe compared with a probe including the B site from the p21 promoter, according to competition assays performed with 10-, 50-, and 100-fold excess of each probe (data not shown).

View larger version (53K): [in this window] [in a new window]   Fig. 2.   NF-B DNA binding to p21 promoter. A, HCT116 cells were stimulated with daunomycin (1 µM) for 3, 6, and 24 h (lanes 1-4). Nuclear protein extracts (5 µg) were analyzed by EMSA with a labeled probe corresponding to the B site from the human p21 promoter. For competition assays, we added to the binding reactions a 100-fold excess of unlabeled oligonucleotides corresponding either to the p21 wild-type (wt, lane 5) or mutated (mt, lane 6) sequence or to an irrelevant OCT probe (oct, lane 7). Competitions and supershifting experiments were performed on nuclear extracts from cells treated for 3 h with daunomycin. Supershifting experiments were performed with specific antibodies directed against p50, p65, p52, c-Rel, and RelB (lanes 8-13) and supershifted complexes are indicated by an asterisk (*). The arrow indicates the specific p50/p65 complex. B, the same extracts as in A were analyzed by EMSA with a labeled probe corresponding to a B palindromic sequence (lanes 16-19) and competitions were performed with an excess of unlabeled B wild-type (wt, lane 14) or mutated (mt, lane 15) probe. C, to test the integrity of the protein extracts, the same extracts were analyzed by EMSA with a labeled probe corresponding to an OCT binding sequence (lanes 20-23).

Regulation of the p21 Gene Promoter by p53, NF-B, and Daunomycin. To unravel the mechanisms involved in the control of p21 expression, we transfected, in MCF7 cells, the p21-LUC reporter plasmid that contains a 2.4-kb genomic fragment of the p21 gene promoter upstream from the LUC gene. As illustrated in Fig. 3, daunomycin (1 µM) induced LUC activity 5-fold compared with untreated cells. Higher concentrations of daunomycin (2-3 µM) did not further induce the reporter gene expression (data not shown).

View larger version (47K): [in this window] [in a new window]   Fig. 3.   Regulation of the p21 gene promoter by p53, NF-B, and daunomycin. MCF7 cells were cotransfected with p21-LUC (0.5 µg) and RSVCAT (0.5 µg) reporter plasmids alone or together with expression plasmids for different NF-B subunits (0.5 µg of each subunit) or for p53. Cells were left untreated or were treated with daunomycin (1 µM) for 6 h before determination of LUC and CAT activities. LUC activity is expressed in fold induction relative to the activity observed with the reporter plasmid alone. Each value represents the mean of a minimum of three independent experiments after normalization to the protein amount of the extracts and to the CAT activity for transfection efficiency. Statistical analysis (Student's t test) demonstrated a significant induction (P < .05 for all conditions).

p53-Dependent p21 Induction by Daunomycin. To examine whether p53 was required for p21 induction after daunomycin treatment, MCF7 cells were transfected with the p21-LUC reporter plasmid together with an expression vector for the HPV-16 E6 oncoprotein. The E6 protein binds to p53 and induces its degradation by the proteasome complex (Scheffner et al., 1990). The transcription of the p21-driven reporter gene was reduced by E6 expression (Fig. 4A) in both control and daunomycin-treated cells.

View larger version (24K): [in this window] [in a new window]   Fig. 4.   p53-dependent p21 induction by daunomycin. A, MCF7 cells were cotransfected with p21-LUC (0.5 µg) and RSVCAT (0.5 µg) reporter plasmids alone or together with an expression vector for the E6 oncoprotein (CMV-E6, 1 µg). Cells were left untreated or were treated with daunomycin at 1 µM. LUC and CAT activities were determined as described in Fig. 3. Statistical analysis (Student's t test) demonstrated a significant induction (P < .05 for all conditions). B, MCF7E6 cell line and HCT15 cells were transfected with p21-LUC (0.5 µg) and RSVCAT (0.5 µg) reporter plasmids and treated with daunomycin at 1 and 10 µM. LUC and CAT activities were measured as described in Fig. 3. Statistical analysis (Student's t test) did not show any significant difference between treated and untreated cells.

p21 Induction in Cells Expressing the IB Mutant. To investigate the role of NF-B in p21 induction by daunomycin, an expression vector coding for a dominant-negative IB protein mutated at serines 32 and 36 was stably transfected in MCF7 cells (MCF7 MAD cells). Consequently, NF-B was sequestered in the cytoplasm of these cells and could not be induced by tumor necrosis factor- or daunomycin, as previously reported (Cai et al., 1997b; Bentires-Alj et al., 1999). MCF7 MAD cells were transfected with the p21-LUC reporter plasmid alone or together with a p53 expression vector and were then treated or not with daunomycin at 1 µM. Under these conditions, the transactivation of p21 promoter by p53 expression vector or after daunomycin treatment was still observed in cells expressing the mutant IB (Fig. 5A), indicating that NF-B was not essential for p21 promoter-driven gene expression. Similar results were obtained with daunomycin-treated MCF7 cells transiently transfected with the same expression vector coding for the dominant-negative IB protein mutated at serines 32 and 36 (data not shown).

View larger version (28K): [in this window] [in a new window]   Fig. 5.   p21 induction in cells expressing the IB mutant. A, MCF7 MAD cell line was transfected with p21-LUC (0.5 µg) and RSVCAT (0.5 µg) reporter plasmids alone or together with p53 expression plasmid at increasing concentrations. Cells were left untreated or were treated for 6 h with daunomycin (1 µM). LUC and CAT activities were measured as described in Fig. 3. Statistical analysis (Student's t test) demonstrated a significant induction (P < .01) for all conditions except for the transfection with 1 ng of pC53-SN3. B, HCT116 MT9 cells were treated with daunomycin (1 µM) for 3, 6, and 24 h and nuclear protein extracts (10 µg) were analyzed by Western blotting using an anti-p21 monoclonal antibody.

Role of NF-B and p53 in Cell Viability after Daunomycin Treatment. To determine whether stable NF-B inhibition or p53 degradation or mutation could modify the cytotoxic effect of daunomycin at short- or long-term treatment, HCT116, HCT116 MT9, HCT15, MCF7, MCF7 MAD, and MCF7E6 cells were incubated in the presence of increasing daunomycin concentrations (0.1, 0.5, 1, and 2 µM) and cell viability was measured after 48 h or 7 days of treatment (Fig. 6). Under these conditions, we could demonstrate that NF-B inhibition did not significantly modify the number of viable cells. E6-expressing MCF7 cells are more resistant to the drug treatment only at short treatment. HCT15 cells were more resistant to short and long daunomycin treatment than HCT116 and HCT116 MT9 cells (Fig. 6).

View larger version (26K): [in this window] [in a new window]   Fig. 6.   Role of NF-B and p53 in cell viability after daunomycin treatment. A, MCF7, MCF7 MAD, and MCF7E6 cells were seeded in 96-well plates at the concentration of 104 cells and treated with daunomycin at different concentrations (0.1, 0.5, 1, and 2 µM) or with control medium for 48 h. Cell viability was estimated with the colorimetric WST-1 test. Each value represents the mean of three independent experiments. B, HCT116, HCT116 MT9, and HCT15 cells were seeded in 96-well plates at the concentration of 7 × 103 cells and treated with daunomycin at different concentrations (0.1, 0.5, 1, and 2 µM) or with control medium for 48 h. Cell viability was estimated with the colorimetric WST-1 test. Each value represents the mean of three independent experiments. C, for long-term cell survival, cells were seeded at the concentration of 500 cells (HCT116, HCT116 MT9) or 3000 cells (HCT15) per well in flat-bottomed 96-well plates in 0.2 ml of medium. After 7 days of treatment with daunomycin at different concentrations (0.1, 0.5, and 1 µM), cell viability was measured with the colorimetric assay WST-1.

Role of NF-B and p53 in Cell Cycle Arrest and Apoptosis after Daunomycin Treatment. The main role of the p21 protein is to induce cell cycle arrest by inhibiting Cdk2 activation. To investigate the role of NF-B and p53 in daunomycin-induced cell cycle arrest and apoptosis, MCF7, MCF7 MAD, and MCF7E6 cells were treated with the drug for 72 h. DNA content was then analyzed by FACS (Fig. 7). No significant differences were observed between the three untreated cell lines. Treatment with 1 µM daunomycin for 72 h induced apoptosis in the three cell lines; these results were confirmed by DNA laddering and TUNEL staining (data not shown). The apoptosis levels increased with higher daunomycin concentrations in the three cell lines. The proportion of apoptotic cells were similar in MCF7 and MCF7 MAD cells, indicating that NF-B was not implicated in daunomycin-induced apoptosis. In MCF7E6 cells, where p53 is degraded by the E6 oncoprotein, the apoptosis rate was not significantly modified compared with the other two cell lines. After 72 h of daunomycin treatment, about 60% of the MCF7 and MCF7 MAD cells were in G_o/G₁ phase, suggesting again that NF-B did not play a role in cell cycle arrest. In MCF7E6 cells, the percentage of cells in G_o/G₁ arrest was decreased and the cells seemed either to be stopped in G₂ phase and/or to enter into a new cell cycle, showing a major role of p53 in p21-induced cell cycle arrest. These results were compatible with the distinct roles of p53 and NF-B in p21 induction by daunomycin.

View larger version (29K): [in this window] [in a new window]   Fig. 7.   Role of NF-B and p53 in cell cycle arrest and apoptosis after daunomycin treatment. MCF7, MCF7 MAD, and MCF7E6 cells were incubated with daunomycin (1 µM) or with control medium for 72 h. Cells were collected for propidium iodide staining and DNA contents were analyzed by FACS. One of two independent experiments is shown.

Role of NF-B and p53 in Clonogenic Survival after Daunomycin Treatment. To determine cell resistance to long-term daunomycin treatment and clonogenic ability of cells harboring wild-type, mutated, or degraded p53, we treated HCT116, HCT116MT9, HCT15, MCF7, and MCF7E6 cells with daunomycin at different concentrations (0.1, 0.5, and 1 µM) for 48 h and then incubated them with fresh medium for 12 days. The number of colonies (>50 cells) was counted after staining with hematoxylin solution. Under these conditions, we detected a very low number of colonies in daunomycin-treated HCT116 cells, whereas HCT15 cells were significantly more resistant to the drug (Table 1; Fig. 8). However, the experiment could not be reproduced with the MCF7 cells because our E6 stably transfected clones had lost the ability to form colonies. Moreover, we did not observe any difference in the number and size of the clones between HCT116 and HCT116 MT9 cells.

View larger version (92K): [in this window] [in a new window]   Fig. 8.   Clonogenic survival assay. HCT116 and HCT15 cells were seeded in six-well plates and treated for 48 h with daunomycin at increasing concentrations (0.1, 0.5, and 1 µM). The figure shows the colonies after staining with hematoxylin solution.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The molecular mechanisms leading to cell death or to cell cycle arrest after daunomycin treatment of cancer cells remain partially unknown. Their characterization would be of upmost importance to understand and circumvent drug resistance. In the present article, we confirmed that daunomycin was a potent inducer of p53, NF-B, and p21 in human colorectal and breast cancer cells. Because the p21 promoter contains functional p53 and NF-B binding sites, we investigated 1) the NF-B-dependent transactivation of a reporter gene under the control of the p21 promoter; 2) the role of p53 and NF-B in p21 induction by daunomycin; and 3) the role of NF-B and p53 in apoptosis and cell cycle arrest after treatment with daunomycin.

In response to DNA damage, the expression of p21 is predominantly induced by activated wild-type p53 (El-Deiry et al., 1993), but recent studies have shown the existence of a p53-independent pathway, after stimuli such as ultraviolet C, oxidative stress, serum, and growth factors (Michieli et al., 1994; Macleod et al., 1995; Russo et al., 1995; Qiu et al., 1996; Haapajarvi et al., 1999). Under our experimental conditions, we confirmed the central role of p53 in p21 induction by daunomycin, but did not observe any p53-independent mechanism.

We investigated whether the drug-dependent induction of p21 protein could be mediated by free radicals. Indeed, anthracycline drugs intercalate into DNA bases and induce topoisomerase II-dependent DNA strand breaks. The putative agents responsible for this DNA damage have been suggested to be superoxide radical, hydrogen peroxide, or hydroxyl radical (Powis, 1987). Moreover, H₂O₂ can activate NF-B and antioxidants such as N-acetylcysteine and pyrrolidine dithiocarbamate inhibit NF-B activation in a number of experimental settings (Schreck et al., 1992). It has also been reported that NF-B activation by daunomycin and mitoxanthone in HL60 and Jurkat T lymphoma cells was inhibited by PDTC (Boland et al., 1997). Finally, the p53 activity is dependent on the redox status and the presence of oxidizing agents. Indeed, p53 is stabilized by metal ions, such as zinc and copper, and in reducing conditions (Hainaut et al., 1995). In our experiments, the two antioxidants PDTC and NAC were not able to block daunomycin-induced NF-B activation (Hellin et al., 1998) or p53 and p21 induction, suggesting that radical oxygen species were not involved. We had previously reported that the role of radical oxygen species in interleukin-1-mediated NF-B activation was cell type-dependent (Bonizzi et al., 1997). These results could explain the apparent discrepancy between our results and those published by Boland et al. (1997) because the role of radical oxygen species in NF-B activation by daunomycin could also be cell specific.

Our data indicated that daunomycin-induced NF-B complexes were able to bind the B site of the p21 promoter in vitro and to activate the transcription of a reporter gene in transient transfection assays. However, p21 protein induction was NF-B-independent. Several hypotheses could account for these observations. First, p50/p65 heterodimers, which were activated after daunomycin treatment (Fig. 2), only weakly induced transactivation of the p21-LUC reporter plasmid (Fig. 3A); this rather limited in vivo effect could thus explain the absence of a significant biological answer. Other heterodimers, such as p52/p65, which bind more strongly to the promoter, did not seem to be activated by the drug. Also, the NF-B site in the p21 promoter might be covered by histones and would thus not be accessible to transcription factors. Another hypothesis is a competition between NF-B and p53 for p300/CBP coactivators (Ravi et al., 1998; Wadgaonkar et al., 1999). Indeed, p300 has been reported to be involved in p21 induction (Billon et al., 1996). However, we did not observe any competition between p53 and NF-B for the transactivation of the p21 promoter (data not shown). Moreover, other activators of NF-B, such as tumor necrosis factor-, H₂O₂, interleukin-1, and the phorbol ester phorbol-12-myristate-13-acetate, also failed to generate any p53-independent p21 induction in HCT116 and MCF7 cells (data not shown).

Several observations suggest that NF-B and IB play a role in cell proliferation. It has indeed been reported that NF-B was activated during the G_o/G₁ transition (Baldwin et al., 1991) and could regulate cyclin D1 expression and G₁-to-S-phase transition (Guttridge et al., 1999). Conversely, other investigators showed that p65 or c-Rel can stop cell cycling (Bash et al., 1997; Sheehy and Schlissel, 1999). Thus, according to the cell type and the stimulus pathway, NF-B activation can induce cell cycle progression or arrest. However, in our experimental model, the cell cycle arrest was p53-dependent and NF-B-independent, thereby suggesting a role of p53 in the mechanism of Cdks inhibition by p21. Similarly to its cell type-specific effect on cell cycle, it has been reported that NF-B could be an inhibitor or an activator of apoptosis. Indeed, NF-B activation by DNA-damaging agents or cytotoxic cytokines protects cells against apoptosis (Beg and Baltimore, 1996; Van Antwerp et al., 1996; Wang et al., 1996) and NF-B inhibition in cancer cells sensitizes them to cytotoxic drug-induced cell death in vitro and in vivo (Wang et al., 1999). However, we and others failed to observe an increased cell toxicity in response to daunomycin or other cytotoxic drugs after NF-B inhibition in MCF7, HCT116, and other cell lines (Cai et al., 1997b; Bentires-Alj et al., 1999). Although NF-B protected HT1080 cells against apoptosis, this transcriptional factor is not involved in p21 induction after daunomycin treatment of this cell line (data not shown). Moreover, NF-B can be required for the onset of apoptosis in other experimental systems (Lin et al., 1995; Grilli et al., 1996). It has also been recently demonstrated that p53 activation induced apoptosis through an NF-B-dependent pathway (Ryan et al., 2000). In the cells investigated in the present study, the apoptotic response to daunomycin was relatively weak and, in spite of a clear activation of both transcription factors, p53- and NF-B-independent. Finally, NF-B inhibition was not sufficient to restore the apoptotic pathways, suggesting that other mechanisms of resistance are activated.

Our study thus indicates that in HCT116 and MCF7 cells, daunomycin treatment induces p21 expression and cell cycle arrest through p53 induction. NF-B, despite its in vitro DNA-binding to the p21 gene promoter, does not have any effect on the cell cycle in these cells. Interestingly, the apoptotic response to daunomycin was not influenced by p53 nor NF-B. To increase cancer cell response to cytotoxic drugs, the in vivo manipulation of the NF-B signaling pathways will thus have to consider these highly different and cell-specific situations.