The Oncogene DEK Promotes Leukemic Cell Survival and Is Downregulated by both Nutlin-3 and Chlorambucil in B-Chronic Lymphocytic Leukemic Cells

Purpose: To characterize the role of the oncogene DEK in modulating the response to either Nutlin-3, a small-molecule inhibitor of the MDM2/p53 interaction, or chlorambucil in primary B-chronic lymphocytic leukemia (B-CLL) cells.

Experimental Design: DEK mRNA and protein levels were evaluated in primary B-CLL samples (n = 21), p53^wild-type SKW6.4, p53^mutated BJAB lymphoblastoid cell lines, and normal CD19⁺ B lymphocytes–treated Nutlin-3 or chlorambucil (10 μmol/L, each). Knocking down experiments with either p53 or DEK small interfering RNA (siRNA) were done to investigate the potential role of p53 in controlling the expression of DEK and the role of DEK in leukemic cell survival/apoptosis.

Results: Both Nutlin-3 and chlorambucil downregulated DEK in primary B-CLL samples (n = 21) and SKW6.4 but not in BJAB cells. Knocking down p53 attenuated the effect of Nutlin-3 on DEK expression, whereas knocking down DEK significantly increased both spontaneous and Nutlin-3–induced apoptosis. Conversely, counteracting DEK downmodulation by using p53 small interfering RNA reduced Nutlin-3–mediated apoptosis. On the other hand, Nutlin-3 potently induced p53 accumulation, but it did not affect DEK levels in normal CD19⁺ B lymphocytes.

Conclusions: These data show that the downregulation of DEK in response to either Nutlin-3 or chlorambucil represents an important molecular determinant in the cytotoxic response of leukemic cells, and suggest that strategies aimed to downregulate DEK might improve the therapeutic potential of these drugs. Clin Cancer Res; 16(6); 1824–33

Human DEK is a 375 amino acid (43 kDa) abundant nuclear protein with important functions in the architectural regulation of chromatin assembly (1, 2). It was originally discovered as part of a fusion protein discovered as the target of a chromosomal translocation event [t(6;9)(p23;q34)] in a subset of acute myeloid leukemias (3, 4), whereas only one previous study reported altered expression of DEK in a subset of B-chronic lymphocytic leukemia (B-CLL; ref. 5). It has also been reported that the expression of DEK is increased in bladder cancer, hepatocellular carcinoma, glioblastoma, melanoma, uterine cervical cancers, and ovarian cancers (6–8). Several studies have implicated DEK in a variety of cellular processes, such as DNA replication, splice site recognition, and gene transcription, as well as in the control of cell viability, differentiation, and cell-to-cell signaling (9–20). Although the mechanisms through which DEK mediates its oncogenic effects are only partially understood, it has been proposed that the pro-oncogenic role of DEK is mediated by its ability to destabilize p53 protein and to inhibit p53 activity and p53-mediated apoptosis (12, 20). On the other hand, a recent study carried out on a series of melanoma cell lines showed that DEK exerted its oncogenic activity through the p53-independent transcriptional activation of the antiapoptotic Bcl-2 family member Mcl-1 (21).

Taking into consideration of the lack of data about the modulation of DEK in B-CLL in response to therapeutically relevant drugs, the aim of this study was to investigate the role of DEK in the cytotoxic response to either chemotherapeutic drugs or nongenotoxic activators of the p53 pathway. For this purpose, primary B-CLL samples as well as p53^wild-type and p53^mutated B lymphoblastoid cell lines were treated in vitro with either chlorambucil, which represented the treatment of choice for B-CLL for several decades (22, 23), or Nutlin-3, which showed promising cytotoxic activity against B-CLL (24–28).

Peripheral blood samples were collected in heparin-coated tubes from 21 B-CLL patients (Table 1) and from 4 healthy blood donors following informed consent, in accordance with the Declaration of Helsinki and in agreement with institutional guidelines (University-Hospital of Udine). All patients had been without prior therapy at least for 3 mo before blood collection. B-CLL samples were characterized for IgVH and p53 mutational status, ZAP-70 expression levels, and by interphase fluorescence in situ hybridization, as previously described (29). For CD19⁺ B-cell purification, T lymphocytes, natural killer lymphocytes, granulocytes, and monocytes were negatively depleted from total peripheral blood mononuclear cells with immunomagnetic microbeads (MACS microbeads, Miltenyi Biotech), with a purity of >95% of resulting CD19⁺ B-cell population. The p53^wild-type SKW6.4 and the p53^mutated BJAB B lymphoblastoid cell lines were purchased from the American Type Culture Collection. Cells were cultured in RPMI 1640 (Life Technologies Bethesda Research Laboratories) containing 10% fetal bovine serum, l-glutamine, and penicillin/streptomycin (Life Technologies Bethesda Research Laboratories).

Both primary B-CLL and lymphoblastoid cell lines were seeded at a density of 1 × 10⁶ cells/mL before treatment with either Nutlin-3 (10 μmol/L; Cayman Chemical) or chlorambucil (10 μmol/L; Sigma Aldrich). Cell viability was examined at different time points after treatment by trypan blue dye exclusion, whereas the induction of apoptosis was quantified by Annexin V-FITC/propidium iodide staining (Immunotech) followed by flow cytometry analysis, as previously detailed (30).

Total RNA was extracted from cells by using the Qiagen RNeasy mini-kit (Qiagen) according to the supplier's instructions. The quality of the RNA preparations was verified by agarose gel and, when necessary, further purification was done with the RNeasy cleanup system (Qiagen) to remove chromatin DNA. Total RNA was transcribed into cDNA, using the AMV Reverse transcriptase (Finnzyme). Modulation of the DEK and MDM2 gene expression upon Nutlin-3 treatment was assessed with the Real Time Thermal Analyzer Rotor-Gene 6000 (Corbett) by SYBR Green real-time PCR detection method, using the SABioscience's RT² Real-Time Gene Expression Assays, which include specific validated primer sets and PCR master mixes (SABioscience Corp.). All samples were run in triplicates.

Cells were lysed in ice-cold RIPA buffer [50 mmol/L Tris (pH 7.5), 150 mmol/L NaCl, 0.1% SDS, 1% Nonidet P-40, 0.25% sodium desoxycholate] supplemented with protease inhibitors (ROCHE) on ice for 1 h. Cell lysates were passed through 27G needles, were added with a loading buffer [250 mmol/L Tris (pH 6.8), 2% SDS, 40% Glycerin, 20% β-mercaptoethanol], and boiled for 2 min. Because DEK protein is predominantly found associated to the nuclear fraction (1), we cannot completely rule out the possibility that part of nuclear DEK protein is lost with the procedure adopted in our study. Equal amounts of protein for each sample were migrated in acrylamide gels and blotted onto nitrocellulose filters, as previously described (31). The following monoclonal antibody were used in our experiments: anti p53 (DO-1), anti-MDM2 (both purchased from Santa Cruz Biotechnology), anti-DEK (BD Transduction Laboratories), and anti-tubulin (Sigma-Aldrich). After incubation with peroxidase-conjugated anti-mouse IgG, specific reactions were revealed with the enhanced chemiluminescence detection kit (Amersham Pharmacia Biotech) and densitometry values were estimated by the ImageQuant TL software (Amersham).

Cells (1.25 × 10⁶) were resuspended into 0.1 mL of Nucleofector solution V of human nucleofector kit V (Amaxa). Two micrograms of plasmid DNA (green fluorescent protein construct) or 1 μg of small interfering RNA (siRNA) were mixed with the 0.1 mL of cell suspension, transferred into a 2.0-mm electroporation cuvette, and nucleofected using an Amaxa Nucleofector II apparatus, following the manufacturer's guidelines. After transfection, cells were immediately transferred into complete medium and cultured in six-well plates at 37°C. Transfection efficiency was estimated in each experiment by scoring the number of green fluorescent protein–positive cells by flow cytometry analysis. SiRNA were designed and manufactured by Ambion, Inc., according to the current guidelines for effective gene knockdown by this method. Based on preliminary validation experiments, siRNA selected were as follows: target #1 for TP53 5′-GGGAGUUGUCAAGUCUUGCtt-3′ (sense) and 5′-GCAAGACUUGACAACUCCCtc-3′ (antisense); target #2 for TP53, 5′-GGGUUAGUUUACAAUCAGCtt-3′ (sense) and 5′-GCUGAUUGUAAACUAACCCtt-3′ (antisense); target #1 for DEK, 5′-GGUGUGCACUGUGAGAUCAtt-3′ (sense) and 5′-UGAUCUCACAGUGCACACCct-3′ (antisense); target #2 for DEK, 5′-GGACUGUGCGGAGCAUGUAtt-3′ (sense) and 5′-UACAUGCUCCGCACAGUCCag-3′ (antisense); and target #3 for DEK, 5′-UCGUCUACCUGGAGAUUGAtt-3′ (sense) and 5′-UCAAUCUCCAGGUAGACGAtg-3′ (antisense). A cocktail of three different negative control siRNAs, each composed of a 19 bp-scrambled sequence with 3′ dT overhangs (Ambion's Silencer negative control siRNA), was used to show that the transfection did not induce nonspecific effects on gene expression.

The results were evaluated by using ANOVA with subsequent comparisons by Student's t test and with the Mann-Whitney rank-sum test. Statistical significance was defined as P < 0.05.

Primary leukemic cells, obtained from 21 B-CLL patients and characterized for Rai stage, ZAP70 levels, and fluorescence in situ hybridization analysis (Table 1), were treated for 24 to 48 hours with Nutlin-3 (10 μmol/L), a nongenotoxic activator of the p53 pathway (32). Nutlin-3 significantly downmodulated the amount of the DEK protein, assessed by Western blot followed by densitometric analysis (Fig. 1A), as well the steady-state mRNA levels of DEK, analyzed by quantitative reverse transcription-PCR (RT-PCR; mean ± SD, 0.55 ± 0.14; fold of decrease, P < 0.01; Fig. 1B). On the other hand, Nutlin-3 potently upregulated the steady-state mRNA levels of MDM2 (mean ± SD, 3.43 ± 2.04; fold of increase, P < 0.01; Fig. 1B), which represents a major transcriptional target of p53 (33). In parallel, a small subset of B-CLL samples (n = 5) was treated with chlorambucil (10 μmol/L), which induced a cytotoxicity comparable with Nutlin-3. In addition, chlorambucil downregulated DEK at both the protein and mRNA levels (data not shown).

To ascertain whether DEK downmodulation might involve p53, the effect of Nutlin-3 or chlorambucil was next analyzed on the levels of DEK protein as well as of p53 and MDM2 proteins in p53^wild-type and p53^mutated B lymphoblastoid cell lines. Nutlin-3 induced the rapid (from 2 hours onwards) and progressive accumulation of both p53 and MDM2 proteins in the p53^wild-type SKW6.4 cells (Fig. 2). Conversely, the levels of DEK protein showed a progressive and significant (P < 0.05) decline, starting 24 hours after Nutlin-3 treatment (Fig. 2).

When analyzed comparatively, we observed that chlorambucil was less potent than Nutlin-3 in promoting the accumulation of both p53 and MDM2 proteins, as well as in inducing the downregulation of DEK protein (Fig. 3A). On the contrary, in the p53^mutated BJAB cells, characterized by constitutive high basal levels of p53, neither Nutlin-3 nor chlorambucil induced p53 and MDM2 protein accumulation and did not affect DEK protein levels (Fig. 3A). Consistent with the results obtained at the protein level, exposure of SKW6.4 cells to either Nutlin-3 or chlorambucil downregulated the steady-state mRNA levels of DEK, while increasing the steady-state mRNA levels of MDM2 (Fig. 3B). Of note, the ability of Nutlin-3 to activate the p53 pathway fully correlated with its cytotoxic activity, as shown by the progressive and significant (P < 0.05) decrease of cell viability, observed at 24 and 48 hours after Nutlin-3 treatment in p53^wild-type but not in the p53^mutated cell lines (Fig. 3C). On the other hand, chlorambucil was less potent than Nutlin-3 in downregulating DEK, an event occurring only after treatment of SKW6.4 cells for 48 hours (Fig. 3A-B), as well as in inducing p53 and its transcriptional target MDM2 (Fig. 3A-B). However, in spite of its lower efficacy in activating the p53 pathway and in downregulating DEK in SKW6.4 cells, chlorambucil exhibited cytotoxic activity against SKW6.4 comparable with that of Nutlin-3 (Fig. 3C). In addition, at variance to Nutlin-3, chlorambucil significantly (P < 0.05) reduced cell viability also in p53^mutated BJAB cells (Fig. 3C), clearly suggesting that chlorambucil promotes leukemic cytotoxicity also through p53- and DEK-independent pathways.

To further investigate whether the Nutlin-3–mediated downregulation of DEK requires a functional p53 pathway, in the next experiments, we have used predetermined optimal experimental conditions to specifically knock down TP53 gene expression by siRNA transfection of SKW6.4 cells (Fig. 4A). Taking into consideration that in our experimental conditions the transfection efficiency was approximately 40% to 50%, it is noteworthy that knocking down p53 partially, but significantly (P < 0.05), counteracted the ability of Nutlin-3 to downregulate DEK expression (Fig. 4B).

Because a previous study suggesting that DEK confers resistance to apoptosis induced by genotoxic stress in solid tumors (34), it was of crucial interest to investigate whether DEK plays a role in modulating cell survival/apoptosis in response to Nutlin-3 or chlorambucil. For this purpose, after having obtained a significant downmodulation of DEK expression after 48 hours of transfection with DEK siRNA, as documented by quantitative RT-PCR and by Western blot (Fig. 5A), cells were either left nontreated or exposed to Nutlin-3 before analyzing the amount of apoptosis. In SKW6.4 cells transfected with DEK-specific siRNAs, both the spontaneous (white columns) and Nutlin-3–mediated (gray columns) cytotoxicity were significantly (P < 0.05) increased with respect to cells transfected with scrambled control siRNA (Fig. 5B). On the other hand, transfection with p53 siRNA significantly counteracted (from 25.5 ± 3 to 11 ± 2.5, P < 0.05) the ability of Nutlin-3 to induce cell death, consistently with the effect of hampering Nutlin-3–mediated downregulation of DEK (Fig. 4B).

Because DEK overexpression has been shown previously to mediate antiapoptotic effects through destabilization of p53 protein and inhibition of p53 activity (12, 20), we have also analyzed the potential effect of DEK siRNA on p53 levels. As shown in Fig. 5C, the accumulation of p53 induced by Nutlin-3 in SKW6.4 cells was not affected by DEK downmodulation (Fig. 5C).

In the last group of experiments, we have investigated the effect of Nutlin-3 on DEK expression in primary normal CD19⁺ B lymphocytes, purified from normal blood donors. As shown in Fig. 6, Nutlin-3 potently induced p53 accumulation in primary CD19⁺ B lymphocytes but, in striking contrast to what observed in primary B-CLL cells and leukemic cell lines, it did not induce DEK downregulation either at the protein (Fig. 6A) or mRNA (Fig. 6B) level, indicating that p53 activation in nontransformed primary cells was insufficient to induce DEK downregulation. Moreover, in keeping with an important role of DEK in promoting cell survival, Nutlin-3 did not exhibit cytotoxic activity in primary normal B-cell cultures, compared with p53^wild-type leukemic cell lines or primary B-CLL cells (Fig. 6C).

In response to a variety of stimuli, such as cellular stress induced by chemotherapeutic drugs, the p53-MDM2 interaction is disrupted and p53 rapidly accumulates within the cell (32). Alternatively, p53 can accumulate in response to selective small-molecule inhibitors of the p53-MDM2 interaction, which binds MDM2 in the p53 binding pocket with high selectivity and can release p53 from negative control leading to effective stabilization of p53 and activation of the p53 pathway (35). In this study, we have shown for the first time that both chemotherapeutic drugs (chlorambucil) and nongenotoxic activators of the p53 pathway (Nutlin-3) significantly downregulated DEK in primary p53^wild-type B-CLL and lymphoblastoid SKW6.4 cells but not in p53^mutated (BJAB) B lymphoblastoid cells. Although these data clearly indicate that a p53^wild-type status is necessary to observe the DEK downregulation in response to either Nutlin-3 or chlorambucil in leukemic cells, the exact role of p53 in DEK regulation remains to be determined. In fact, by using bioinformatic programs (TESS³

and TFSEARCH)⁴, no p53 consensus sequences were found in the DEK promoter. Moreover, whereas knocking down p53 partially counteracted the ability of chlorambucil or Nutlin-3 to downregulate DEK in leukemic cells, Nutlin-3 potently upregulated p53 accumulation in primary CD19⁺ cells without inducing DEK downregulation. Therefore, although p53^wild-type status seems to be necessary, but not sufficient for the Nutlin-3–mediated downmodulation of DEK, further work is needed to clearly establish the role of additional transcription factors in mediating the downregulation of DEK by either Nutlin-3 or chemotherapeutic drugs.

The major conclusion of our study is that DEK plays an important role in promoting the survival/counteracting apoptosis in lymphoblastoid B cells. This was shown by the experiments done with siRNA specific for DEK. In fact, knocking down DEK resulted in a significant increase of both spontaneous and Nutlin-3– or chlorambucil-induced apoptosis in SKW6.4 cells, thus suggesting that the downregulation of DEK is an important molecular mechanism involved in the proapoptotic activity of both chemotherapeutic drugs and novel activators of the p53 pathway in leukemic cells. Thus, although it has been argued that transcriptional independent mechanisms play a key role in mediating the cytotoxic activity of p53 (36, 37), perhaps due to the presence of negative feedback loops (38), our data show for the first time that the transcriptional downregulation of the oncogene DEK represents an important additional mechanism of action by which Nutlin-3 promotes apoptosis in leukemic cells. However, although previous studies have proposed that the overexpression of DEK inhibits the transcriptional activity of p53 (12, 20), silencing of DEK expression by siRNA in SKW6.4 lymphoblastoid cells did not induce any accumulation of p53 protein, nor of p53 target genes, such as MDM2, in response to either Nutlin-3 or chlorambucil. In keeping with a recent study done on melanoma cell lines, showing that siRNA specific for DEK did not change the levels of p53 in the absence or presence of doxorubicin (21), the contribution of DEK to promote tumor cell survival is likely due to its ability to upregulate antiapoptotic genes of the Bcl-2 family (21), rather than to directly control the p53 pathway.

In conclusion, we have shown for the first time that the activation of p53 represses the transcription of DEK in p53^wild-type B-CLL cells and we propose that such downregulation is an important mediator of Nutlin-3 and chlorambucil cytotoxic activity. The relevance of our findings is underscored by the fact that, in contrast to most solid tumors, p53 is mutated in approximately 10% to 15% of both myeloid and lymphoid leukemias at diagnosis (39). Therapeutic strategies able to downregulate DEK should be further explored to improve the antileukemic activity of both conventional and novel antileukemic drugs.

No potential conflicts of interest were disclosed.

Grant Support: The Italian Association for Cancer Research and Beneficientia Foundation (G. Zauli) and by CariFe Foundation (P. Secchiero).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.