|
|
Cancer Res. Author manuscript; available in PMC 2009 September 22. Published in final edited form as: doi: 10.1158/0008-5472.CAN-07-5817. | PMCID: PMC2748725 NIHMSID: NIHMS126308 |
Sequential transcription factor targeting for diffuse large B-cell lymphomas Leandro C Cerchietti,1 Jose M Polo,1 Gustavo F Da Silva,1 Pedro Farinha,2 Rita Shaknovich,3 Randy D Gascoyne,2 Steven F Dowdy,4 and Ari Melnick1 1 Department of Developmental and Molecular Biology, Albert Einstein College of Medicine. 1300 Morris Park Ave, Bronx, NY 10461. 2 Department of Pathology, British Columbia Cancer Agency, 600 W 10th Avenue, Vancouver, BC V5Z 4E6, Canada 3 Department of Pathology. Albert Einstein College of Medicine. 1300 Morris Park Ave, Bronx, NY 10461. 4 Howard Hughes Medical Institute and Department of Cellular and Molecular Medicine. University of California at San Diego School of Medicine. 9500 Gilman Drive, Dept 0686. La Jolla, CA 92093−0686. Transcription factors play a central role in malignant transformation by activating or repressing waves of downstream target genes. Therapeutic targeting of transcription factors can reprogram cancer cells to lose their advantages in growth and survival. The BCL6 transcriptional repressor plays a central role in the pathogenesis of diffuse large B-cell lymphomas (DLBCLs) and controls downstream checkpoints including the p53 tumor suppressor gene. We report that a specific inhibitor of BCL6 called BPI can trigger a p53 response in DLBCL cells. This was partially due to induction of p53 activity and partially due to relief of direct repression by BCL6 of p53 target genes. BPI could thus induce a p53-like response even in the presence of mutant p53. Moreover, sequential BCL6 peptide inhibitors followed by p53 peptide or small molecule activators provided a more powerful anti-lymphoma effect than either treatment alone by maximally restoring p53 target gene expression. Therefore, tandem targeting of the overlapping BCL6 and p53 transcriptional programs can correct aberrant survival pathways in DLBCL and might provide an effective therapeutic approach to lymphoma therapy. Diffuse large B- cell lymphomas (DLBCL) are the most common type of B-cell lymphomas. The BCL6 ( B- Cell Lymphoma 6) proto-oncogene is frequently constitutively expressed in DLBCL due to translocation of heterologous promoters to the BCL6 coding region, or point mutations of negative regulatory elements due to misdirected somatic hypermutation( 1). Constitutive expression of BCL6 in mouse models that mimic human BCL6 translocations can induce formation of DLBCL( 2, 3). Many DLBCLs express BCL6 even in the absence of genetic lesions, suggesting that other factors that drive BCL6 expression could also be oncogenic hits. The lymphomagenic effects of BCL6 may be related to its ability to directly repress critical cell cycle checkpoint genes including TP53 (tumor protein p53), ATR (ataxia telangiectasia and Rad3 related), CHEK1 (CHK1 checkpoint homolog) and p21CDKN1A ( cyclin- dependent kinase inhibitor 1A (p21, Cip1)( 4- 7). DLBCL cells become dependent on BCL6 since its continued presence is required to maintain proliferation and survival of DLBCL cells( 6, 8). BCL6 is a transcriptional repressor and member of the BTB/POZ ( bric à brac- tramtrack- broad complex/pox virus zinc finger) family of proteins. The BCL6 BTB domain plays a key role in repression by recruiting the SMRT ( silencing mediator of retinoid and thyroid receptors), N-CoR ( nuclear receptor corepressor) and BCoR ( BCL6 corepressor) corepressors( 9, 10). These three corepressors can bind to a groove located on the surface of BCL6 BTB homodimers( 11). We designed a BCL6 peptide inhibitor (BPI), which mimics the 18 amino acid region of SMRT that binds to BCL6. BPI readily penetrates DLBCL cells and blocks BCL6 from recruiting N-CoR, SMRT and BCoR, resulting in chromatin remodeling and re-activation of BCL6 target genes( 8). BPI specifically inhibited BCL6 and not other transcriptional repressor proteins( 8). BPI specifically killed DLBCL cells that express BCL6 but had no effect on BCL6 negative DLBCL cells( 6- 8, 12). Finally, BPI abrogated the biological activity of BCL6 on B-cells in vivo, demonstrating that effective levels of this inhibitor can be readily achieved in an animal model ( 8). Targeting of oncogenic transcription factors like BCL6 is a feasible strategy for cancer therapy ( 13, 14). However, a single targeted agent is unlikely to be curative since tumors typically contain multiple oncogenic “hits”. One way to overcome this obstacle is to combine a transcription therapy agent with chemotherapy drugs, as routinely employed in the treatment of acute promyelocytic leukemia ( 15). We propose another more specific route towards enhancing the activity of specific transcription therapy agents. In particular, we hypothesized that is possible to enhance the anti-tumor effects by combining agents that target successive waves of transcriptional programming. In the case of BCL6, an attractive downstream network for therapeutic targeting is that of the p53 tumor suppressor, which regulates genes involved in cell death and proliferation( 16). Restoration of defective p53 function could potentially facilitate induction of cell death and growth arrest in tumor cells. Recent studies carried out in lymphoma and other tumor cells demonstrated that p53 function could be enhanced or restored by specific peptides or small molecules, which rescue certain types of mutant as well as wild type p53 proteins (reviewed in( 17)). Herein we show that BPI can induce p53 expression and biological functions in DLBCL cells, including those with certain p53 mutations. These effects were due both to induction of TP53 as well as loss of BCL6 repression on p53 target genes. Sequential administration of BPI followed by a p53 activating peptide or a small molecule cooperatively enhanced p53 activity and killed DLBCL cells. p53 and BCL6 are thus intricately functionally linked in DLBCL and can be therapeutically harnessed as a form of tandem transcription therapy with potent anti-lymphoma activity. Cell lines, peptides and drugs The DLBCL cell lines Ly1, Ly4 and Ly10 were grown in Iscove's medium with 10% FCS, and penicillin G/streptomycin. The Farage, Ly3, SU-DHL6 and SU-DHL4 cell lines were grown in RPMI with 10% FCS, penicillin G/streptomycin, glutamine and HEPES. p53C’-TAT (GSRAHSSHLKSKKGQSTSRHKKGYGRKKRRQRRR), p53-control peptide (CP2) (GSRAHSSHLESAEGQSTSRHKKGYGRKKRRQRRR) ( 18), BPI (YGRKKRRQRRRGGRSIHEIPR) and control peptide (CP) (YGRKKRRQRRRG) were obtained from Biosynthesis, Inc (Lewisville, TX). The purity determined by HPLC-MS was 98% or higher for each peptide. Unless noted, peptides were used at the following concentrations: BPI and CP 5 μM five times per day, and p53C’-TAT and CP2 10 μM once per day. The P53-DN-(His)6-pTAT-HA( 18) was expressed in Escherichia coli cells BL21 (D3) (EMD Biosciences, Inc., San Diego, CA), and affinity purified by Ni-NTA Hi-Trap column (APBiotech, Piscataway, NJ) using an AKTA Purifier 10 (APBiotech). Cyclic-pifithrin- ![[proportional, variant]](/corehtml/pmc/pmcents/x221D.gif) (2-(4-methylphenyl)imidazo[2,1-b]-5,6,7,8-tetrahydrobenzothiazole) (EMD Biosciences, Inc.) was resuspended in DMSO immediately before use. For combination experiments pifithrin- 20 μM (or DMSO) were added 12 h before the BPI or CP and every 12 h thereafter until analysis. PRIMA-1 (2,2-bis(hydroxymethyl)-1-azabicyclo(2,2,2)octan-3-one) was obtained from Tocris Cookson Inc, Ellisville, MO. Growth inhibition determination Ly1, Ly3, Ly4, Ly10, SU-DHL4, SU-DHL6 and Farage cells were grown at concentrations to maintain exponential growth over the 48 h drug/peptide exposure time. We determined cell viability using a MTS tetrazolium-based method (Cell Titer 96 Aqueous One, Promega Corp., Madison, WI) and a Polarstar Optima microplate reader (BMG Labtechnologies, Germany). We performed 6-point standard curves for each cell line (correlating the number of viable cells by Trypan blue with optical density). For combination studies, viability was determined using a fluorometric resazurin reduction method (CellTiter-Blue, Promega Corp.) and the Polarstar Optima microplate reader. The number of viable cells in each treated well was calculated by using the linear least-squares regression of the standard curve. Optical density or fluorescence was determined for 6 replicates per treatment condition or standard. We verified cell viability by the Sulforhodamine B assay (Sigma, Milwaukee, WI) with minor modifications for suspended cells. Experiments were carried out in triplicates and data are represented as percentage of growth inhibition to respective control. We used the Calcusyn software (Biosoft, Great Shelford, Cambridge, UK) to determine the concentration of peptides and PRIMA-1 that inhibits 50% the growth of cell lines compared to control treated cells (GI50) and to determine the combination index (CI). For the CI experiments cells were treated with BPI at 5, 15 and 25 μM and/or PRIMA-1 at 25, 75 and 100 μM (1:5 constant ratio). We assume a mutually exclusive effect for the combination of BPI and PRIMA-1. Cell cycle analysis Cells were fixed with a formaldehyde-based fixation buffer (Santa Cruz, Santa Cruz, CA), resuspended in permeabilization buffer (Santa Cruz) and stained with anti-Ki-67 antibody (Santa Cruz) and 50 μL propidium iodide 1 mg/ml (Sigma) or 7-AAD (BD Biosciences, San Diego, CA) with 20 μL ribonuclease A 10 μg/ml (Sigma). Samples were analyzed by flow cytometry in a FASCalibur (BD Bioscience) using CellQuest software (BD Bioscience). Detection of Ki-67 positive cells, combined with simultaneous measurement of DNA content followed by bivariate analysis allowed identification of cells in G0/G1 (low DNA content values, Ki-67 negative), G2/M (DNA twice that of G0/G1, Ki-67 negative) and S (intermediate DNA content, Ki-67 positive). ATP determination ATP levels were determined using the ApoSENSOR ADP/ATP assay kit (BioVision, Mountain View, CA). Briefly, cells were lysed to free ATP into media containing luciferase and luciferin, which produces light in the presence of ATP. The light emission is proportional to the amount of ATP present and was measured using the Polarstar Optima microplate reader). Apoptosis/necrosis determination Morphological features of cell death we use the differential uptake of acridine orange (Sigma) and the DNA binding dye ethidium bromide (Sigma) (EB/AO) ( 19, 20). After 24 h incubation with the respective active and control peptides, cells were gently centrifuged at 4C and resuspended in 20 μl of PBS plus 2 μl of a 1:1 mix of 100 μg/ml of EB/AO. Stained cell suspension (10 μl) were transferred to a microscope slide and viewed under a fluorescent light microscope (Axioskop 2, Carl Zeiss AG, Germany) with 100, 200 and 400X magnifications. A total of 100 cells were counted in triplicates per treatment condition. Caspase 7/3 activity The activity of caspase-7 and caspase-3 was determined using the Apo-ONE caspase 3/7 assay (Promega). DLBCL cell lines were treated for 6 h and 18 h with drug/peptides followed by 1 hr exposure to the pro-fluorescent Z-DEVD-R110 substrate. Activation of Z-DEVD-R110 by the activity of caspases 3/7 allows the R110 group to become intensely fluorescent (Ex499nm / Em521nm), which was measured using the Polarstar Optima microplate reader. Caspase 7/3 activity was related to the cell number determined by CellTiter-Blue (Promega) in a multiplex assay. DNA fragmentation An ELISA kit (Cell death detection ELISA, Roche Diagnostics, Penzberg, Germany) was used to determine the presence of mono and oligonucleosomes in the cytoplasmatic fraction of cell lysates. DLBCL cells were exposed to BPI and/or p53C’-TAT and/or control peptides for 24 hours. Following the manufacturer instructions, the cell lysates were applied in triplicates onto an anti-histone coated plate. After saturation of non-specific binding with a blocking solution, the remaining immobilized complexes were incubated with a peroxidase-conjugated anti-DNA antibody. The amount of peroxidase retained was colorimetrically determined by using ABTS as substrate (absorbance A405nm - A490nm, Polarstar Optima microplate reader). The enrichment of nucleosomes in the cytoplasm of treated cells was calculated by comparing the absorbance of the treated cells over the corresponding control. Tissue microarrays, chromatin Immunoprecipitation (ChIP) analysis, annexin-V / propidium iodide staining, western blotting, real time quantitative PCR (RT-PCR) and TP53 sequencing Are provided as supplementary data. BPI induces p53 expression in DLBCL cells Since BPI can reactivate BCL6 target genes and kill DLBCL cells ( 6- 8, 12), and p53 is a BCL6 target gene ( 4), we wondered whether BPI could induce p53 expression in DLBCL cells. For this analysis we selected three representative DLBCL cell lines: Ly1, which expresses BCL6 and has biallelic point mutations in the p53 DNA-binding domain ( 21- 23); Ly10, which expresses BCL6 and has wild type (WT) p53( 23); and Ly4, which expresses little BCL6, is BCL6 independent and has only one allele of p53 which is inactivated by two mutations( 22, 23) ( See also Supp Table 1). We examined the kinetics of BPI effects on p53 mRNA abundance by qRT-PCR over a time course of 48 hours. BPI caused rapid 3.5-fold induction of p53 transcription in Ly10 (p53 WT) cells, peaking at 6 hours, but had different kinetics in Ly1 cells, reaching a maximum of 2-fold increase at 24 hours (). There was no induction of p53 in the BCL6-independent Ly4 cell line. p53 mRNA was baseline expressed at similar levels (ratio to GAPDH: Ly4: 0.63, Ly1: 0.74, Ly10: 0.68) in the three cell lines regardless of mutational status or BCL6 expression. BPI also induced p53 protein levels () with a 1.7 peak fold increase in Ly10 cells at 12 h followed by a decline in protein abundance at 48 h, whereas in Ly1 cells the protein steadily accumulated during the observation time, peaking at 24 h, possible because the mutated protein may not as efficiently induce its own degradation. In contrast, p53 protein was not increased in Ly4 cells. BPI induces a p53 response in DLBCL cells To determine the functional impact of BPI-mediated induction of p53 we examined the kinetics of p53 target gene expression in DLBCL cells exposed to BPI vs. control peptide by qRT-PCR. In Ly10 cells BPI induced the p53 target genes p21, NOXA, PUMA ( p53- upregulated modulator of apoptosis) and GADD45A ( growth arrest and DNA- damage-inducible 45 alpha) within 6−12 h and PIG3 ( p53 induced gene 3) within 24 h (). p21 induction was slightly biphasic, peaking first at 6 h and increasing again at 24 h. Since p21 is also a direct target of BCL6 ( 5), it is possible that its induction by BPI was partially independent of p53. Accordingly, BPI induced p21 in Ly1 (p53 mutant) cells with a similar biphasic pattern and level of expression (). In contrast, PUMA, NOXA and PIG3 were unaffected by BPI in Ly1 cells. Like p21, GADD45A was also induced in Ly1 cells to a 2-fold maximum at 24 hours. We wondered whether GADD45A is also a BCL6 target gene. The GADD45A promoter contains a BCL6 consensus sequence located within a highly conserved region at −500. Chromatin immunoprecipitation assays confirmed strong enrichment of BCL6 to this site ( Supplementary Fig 1). None of these genes were induced by BPI in Ly4 cells. These results indicate that BPI can increase p53 transcriptional function in DLBCL cells and show that the BCL6 and p53 transcriptional programs are at least partially overlapping. We next examined DLBCL cells for cell cycle progression 24 h after BPI exposure (at which point BPI-treated cells are still viable). The Ly10 cell line displayed a reproducible doubling in G2/M phase cells from 8 to 15%, with a corresponding reduction in S-phase, (consistent with biological effects of p53 ( 24)) (). In contrast, Ly1 cells showed a small but reproducible increase in the S-phase fraction of cells from 50 to 60% with a corresponding reduction in G2/M (consistent with our work showing that BCL6 suppresses an ATR dependent S-phase checkpoint) ( 7). Ly4 cells were not affected by BPI. BPI impaired viability of both the Ly10 and Ly1 DLBCL cells within 48 h, as shown in metabolic assays (). In contrast, Ly4 cell viability and survival were unaffected. The cellular effects of BPI in DLBCL cells thus vary with the mutational status of p53. p53 induction contributes to BPI-mediated cell death To determine the contribution of p53 to effects of BPI, we used pifithrin-α, an inhibitor of p53( 25). We first examined the effect of pifithrin-α on expression of p21, GADD45A and PIG3. BPI induced p21 and GADD45A in a pifithrin-α independent manner in (p53 mutant) Ly1 cells, consistent with their direct regulation by BCL6, but not PIG3 (). In contrast, pifithrin-α markedly impaired the induction of all three of these genes in p53 wild type Ly10 cells, suggesting that the mechanism controlling expression of these genes is dependent on whether p53 was wild type or mutated (). As expected, BPI and pifithrin-α had no effect in Ly4 cells. Pifithrin-α rescued Ly10 cells from BPI-mediated loss of viability, but had no such effect on Ly1 cells (). In order to confirm the specificity of these results, we repeated these experiments using a dominant negative form of p53 delivered by TAT protein transduction, or a mutant inactive form of p53 as a control( 18). Like pifithrin-α, dominant negative p53 could at least partially overcome the effects of BPI in Ly10 but had no impact on Ly1 cells (). Dominant negative p53 had a mild viability effect in Ly4 cells, the reason for which is unknown but could be related to interference with gain of function effects of the mutant p53 variants expressed in these cells. Taken together, these data suggest that BPI-mediated release of the p53 pathway from BCL6 repression plays an important role in its ability to kill DLBCL cells with wild type p53. Release of p53 targets also regulated by BCL6 may contribute in part to BPI effects on DLBCL cells with mutant p53. BCL6 inhibition plus p53 activation cooperate to kill DLBCL cells A TAT-p53 C-terminal peptide (p53C’-TAT) was shown to enhance the transcriptional activity of wt p53( 18) and to rescue the activity of certain p53 DNA contact mutants( 26), without significantly increasing the p53 protein levels. We tested the impact of p53C’-TAT on p53 target gene expression in DLBCL cells. In Ly10 cells p53C’-TAT induced p21, GADD45A and PIG3 by 2.5 fold within 6 hours, and PUMA and NOXA at later time points (). p53C’-TAT also induced these target genes in Ly1 cells, although with generally delayed kinetics () but had negligible activity in Ly4 cells. Consequently, p53C’-TAT modestly reduced the number of viable cells in Ly1 (by 16%), Ly10 (by 22%) and Ly4 cells (by 4%) after 48 hours of exposure compared to control ( Supp Fig 2). Since BPI can induce p53 expression, and considering that the downstream genes induced by BPI and p53 partially overlap, we asked whether BPI and p53C’-TAT might cooperate in a transcriptional response that could lead to enhance anti-lymphoma effects. DLBCL cells were exposed to BPI to release p53 from BCL6 mediated suppression, followed 24 hours later by p53C’-TAT to further induce the transcriptional activity of p53. We first measured the fold induction of p53 target gene mRNA transcripts relative to their respective controls. Remarkably, BPI and p53C’-TAT mediated supra-additive induction of p21, PUMA and NOXA in both Ly10 and Ly1 cells and of PIG3 in Ly1 cells ( and Supp Fig 3). Induction of GADD45A was additive in both cell lines. None of the p53 target genes were synergistically activated in Ly4 cells. Based on these results we wondered whether the combination of BPI and p53C’-TAT might also cooperate in killing lymphoma cells. DLBCL cells were exposed to BPI or control peptide followed 24 hours later by p53C’-TAT or its control peptide for another 24 hours. BPI alone mediated 40% loss of viability in Ly1 cells and 50% loss of viability in Ly10 cells while had no effect in Ly4 cells. The sequential administration of p53C’-TAT after BPI resulted in markedly enhanced anti-lymphoma activity to a roughly 80% and 70% loss of viability in Ly1 and Ly10 cells respectively (). BPI and p53 activation can induce different types of cell death In order to better understand BPI and p53 anti-lymphoma effects we explored death mechanisms triggered by these agents alone or in combination. For example, in addition to apoptosis, p53 can also activate a necrosis-like cell death program in distinct tumor cell types ( 27). In the search for more effective therapies, the induction of alternative death pathways may be important to bypass resistance mechanisms associated with a particular type of cell death ( 27, 28). In order to determine whether 24 h exposure to BPI and p53C’-TAT induce the same or alternative p53-dependent cell death pathways, Ly4, Ly1 and Ly10 were examined microscopically by ethidium bromide/acridine orange (EB/AO) staining, and by flow cytometry using annexin V and propidium iodide ( and Supp Fig 4). Cells undergoing necrosis can be distinguished from apoptotic cells based on morphologic criteria including cellular shape, cellular volume, chromatin condensation, nuclear fragmentation and status of the plasma membrane ( 29). In accordance with a recent report on p53 C-terminal peptide effects in prostate cancer cells( 27), we found that p53C’-TAT induced mostly a necrotic-type cell death in DLBCL cells regardless of the mutation status of p53 (first and second rows in and Supp Fig 4). Ly1 cells exposed to BPI for 24 h showed a morphologically apoptotic-type of death, while Ly10 cells exhibit a necrotic-type cell of death similar to that induced by p53C’-TAT (third row in and Supp Fig 4). The concurrent administration of BPI and p53C’-TAT increased both types of death, with a higher proportion of necrosis in Ly10 cells than in Ly1 cells (fourth row in and Supp Fig 4). Moreover, BPI alone or in combination with p53C’-TAT strongly induced DNA fragmentation and caspase 7 and 3 activity in Ly1 cells, consistent with death by apoptosis (). In contrast, Ly10 and Ly4 cells displayed minimal changes in DNA fragmentation or caspase activation (). Apoptosis is an energy/ATP dependent process, while, in contrast, ATP levels dissipate rapidly in cells undergoing necrosis( 30), perhaps due to cellular depletion of NAD+ by PARP( 31, 32), with consequent depletion of ATP. On the contrary, in most apoptotic cells PARP is inactivated by caspase-dependent cleavage( 33), and the cellular ATP pool is not depleted. We found that after 18 h of exposure of Ly4, Ly1 and Ly10 cells to p53C’-TAT there is a sharp decline in the cellular levels of ATP while the PARP protein remains intact (). A similar effect was observed after BPI exposure in Ly10 cells. However, Ly1 cells treated with BPI maintained levels of ATP similar to control treated cells while the PARP protein was cleaved in this cell line (). Concomitant administration of BPI and p53C’-TAT yielded a mixed result with partial ATP depletion and a lesser extent of PARP cleavage. Taken together, these results indicate that the type of cell death induced depends on the mechanism of action of the therapeutic agent (p53C’-TAT induces mostly necrosis) and on the genetic background of the tumor cell. Co-targeting of BCL6 and p53 enhances anti-lymphoma activity Recent screening efforts identified small molecules that reactivate a greater spectrum of p53 mutants and with higher efficiency than p53-derived C-terminal peptides. One such molecule is called PRIMA-1 ( 34). The mechanism of action of PRIMA-1 is not yet fully understood, but may involve the promotion of refolding of mutant forms of p53 into active conformations ( 35). Our data suggest that combinatorial targeting of BCL6 and p53 could provide enhanced anti-lymphoma activity by inducing either apoptosis or necrosis in DLBCL cell lines. In order to confirm and expand these results we examined BPI and PRIMA-1 effects in a larger panel of DLBCL cell lines including one BCL6-independent (Ly4) and six BCL6-dependent cell lines. We analyzed by direct sequencing the TP53 mutational status in this panel of cells and found that on top of the reported mutations, Ly4 and Ly1 have additional mutations affecting the DNA binding region of p53 (G245D and R249K respectively) ( Supp Table 1). The SU-DHL4 and Farage cell lines contained R273C and E285Q mutations respectively, affecting the DNA binding region of p53. Ly10, Ly3 and SU-DHL6 had wild type p53 ( Supp Table 1). PRIMA-1 has been reported to preferentially reactivate and induce cell death in tumor cells with mutant p53 proteins( 34). We determined the GI 50 for PRIMA-1 in our panel of DLBCL cell lines () and found that in fact the GI 50 for cell lines with mutated p53 were significantly lower than for those cell lines with wt p53 (p=0.0004, for triplicates, Wilcoxon's rank sum test) ( and Supp Table 2). In order to determine whether PRIMA-1 cooperates with BPI in killing lymphoma cells we exposed BPI responsive DLBCL cells to each drug alone or in combination. The dose of BPI used was the median of the GI 25 values of the cell lines (for a single dose exposure) and the dose of PRIMA-1 was the median of the GI 50 of the most responsive cell lines. We found that the concurrent administration of BPI and PRIMA-1 decreased viability for all the cell lines to a higher degree than either compound alone (). With the exception of Ly10, this effect was preceded by an increase in the activity of the executioner caspases 7 and 3, suggesting that apoptosis contributes to the anti-lymphoma effect of combined therapy in most cell lines (). Since BPI can induce p53 expression in DLBCL cells, which is the therapeutic target of PRIMA-1, we predicted that the most effective sequencing strategy would consist of first treating with BPI followed by administration of PRIMA1 to maximally up-regulate p53 activity. This might be particularly critical in the case of mutant p53, which is more functionally enhanced by PRIMA1 than wild type. In order to determine whether there is a schedule-dependent activity for the combination of BPI and PRIMA-1, cells were exposed for 48 h to the sequences of BPI at 0h followed by PRIMA-1 at 24 h and PRIMA-1 at 0 h followed by BPI at 24 h. We found that the sequence of administration did not greatly affect the combination index values for cells with wt p53, whereas cells with mutant p53 tend to respond better to the sequence BPI → PRIMA-1 ( and Supp Table 3). Our study addresses the therapeutic targeting and functional interaction between BCL6 and p53 in the setting of DLBCL. Previous reports have shown that p53 expression is relatively common in DLBCL, and that its presence is not necessarily correlated with mutations of its coding sequence (reviewed in ( 36). Although p53 is a BCL6 target gene, a recent report showed that p53 was expressed in BCL6 positive germinal center B-cell type DLBCLs ( 37). Similarly, we found that median p53 transcript abundance was greater in the BCL6 positive B-cell receptor (BCR) type DLBCL and that BCL6 and p53 proteins can be co-expressed in DLBCL ( Suppl Fig 5). p53 was also expressed in all DLBCL cell lines examined, regardless of whether it was mutated. However, inhibition of BCL6 in DLBCL cells induced higher expression of p53. These results suggest that p53 expression is maintained below a critical threshold by BCL6 and/or that p53 is functionally impaired in the presence of BCL6, as otherwise p53 would be expected to counteract BCL6 mediated survival and proliferation. Our data support this scenario since BPI induced the amount and activity of p53 as judged by transcriptional activation of target genes and biological effects. Many mechanisms could account for the functional impairment of p53 in the presence of BCL6 in DLBCLs. One of them could be a BCL6-dependent repression of genes upstream of p53 that control its activity through post-translational modifications. We have recently shown that BCL6 directly represses ATR, a protein kinase and master regulator of DNA damage sensing and genomic integrity ( 7). Accordingly, BPI could induce ATR expression and activity, resulting in phosphorylation and activation of p53( 7). Our most recent preliminary data suggest that BCL6 may also impair p53 acetylation( 38). Another mechanism through which BCL6 might attenuate the activity of p53 is by downstream interference with the p53 transcriptional program. It was reported that BCL6 can directly repress p21( 5) and we find that BPI can induce p21 in a p53-independent manner. BCL6 disruption of the p53 program is even more extensive, since we find that BCL6 can directly repress GADD45A and reported previously that BCL6 directly regulates additional DNA damage checkpoint genes including GADD45B and CHEK1( 6). BCL6 can thus suppress genomic integrity checkpoints at multiple levels, an effect that is presumably necessary for survival of GC B-cells but that also contributes to lymphomagenesis. Our current data help to link the biochemical mechanism of action of BCL6 with its biological effects in lymphoma cell survival. DLBCL survival requires binding of the SMRT and N-CoR corepressors to the BCL6 BTB domain, which is the target of BPI( 6, 8). The fact that BPI upregulates p53 indicates that repression of p53 is dependent on SMRT and N-CoR and explains in part how blockade of the BCL6 BTB domain can kill lymphoma cells. In contrast, BPI has no effect on BCL6 target genes involved in differentiation such as PRDM1( 8). Instead, BCL6 represses PRDM1 through the MTA3/NuRD corepressor complex ( 12, 39). Opposite to BPI, MTA3 depletion induced differentiation but not cell death( 12). Therefore, these results demonstrate the functional separation of BCL6 controlled pathways due to differential corepressor requirements. Taken together, our data reveal a complex functional relationship between p53 and BCL6 in DLBCL and show that although p53 is often expressed in association with BCL6, it is in a relatively inactive state that can be switched on by therapeutically targeting BCL6. Restoration of p53 function in tumor cells has been the focus of intense research, either through gene-therapy approaches or through small molecule or peptide therapies aiming to stabilize p53 protein or enhance its transcriptional activity( 40). The fact that p53 is usually wild type in BCL6-positive DLBCL, and can be switched on by blocking BCL6 presents an excellent opportunity for harnessing the therapeutic potential of p53. The ability to sequentially target successive waves of transcription factors introduces the concept of tandem transcription therapy. In this case inhibiting BCL6 makes available a second transcription factor (p53), which can also be therapeutically targeted by enhancing its activity. Since blocking BCL6 and enhancing p53 is harmless to normal tissues( 8, 18), it may be possible to extensively reprogram death pathways in lymphoma cells with minimal effects on the host. Thus, given sufficient information on the transcriptional networks that sustain specific tumor types, it could be possible in the future to deliver combinatorial therapy that would eliminate or reduce the need for chemotherapy drugs. 1. Pasqualucci L, Bereschenko O, Niu H, et al. Molecular pathogenesis of non-Hodgkin's lymphoma: the role of Bcl-6. Leuk Lymphoma. 2003;44(Suppl 3):S5–12. [PubMed] 2. Cattoretti G, Pasqualucci L, Ballon G, et al. Deregulated BCL6 expression recapitulates the pathogenesis of human diffuse large B cell lymphomas in mice. Cancer Cell. 2005;7(5):445–55. [PubMed] 3. Baron BW, Anastasi J, Montag A, et al. The human BCL6 transgene promotes the development of lymphomas in the mouse. Proc Natl Acad Sci U S A. 2004;101(39):14198–203. [PubMed] 4. Phan RT, Dalla-Favera R. The BCL6 proto-oncogene suppresses p53 expression in germinal-centre B cells. Nature. 2004;432(7017):635–9. [PubMed] 5. Phan RT, Saito M, Basso K, Niu H, Dalla-Favera R. BCL6 interacts with the transcription factor Miz-1 to suppress the cyclin-dependent kinase inhibitor p21 and cell cycle arrest in germinal center B cells. Nat Immunol. 2005;6(10):1054–60. [PubMed] 6. Polo JM, Juszczynski P, Monti S, et al. Transcriptional signature with differential expression of BCL6 target genes accurately identifies BCL6-dependent diffuse large B cell lymphomas. Proc Natl Acad Sci U S A. 2007;104(9):3207–12. [PubMed] 7. Ranuncolo SM, Polo JM, Dierov J, et al. Bcl-6 mediates the germinal center B cell phenotype and lymphomagenesis through transcriptional repression of the DNA-damage sensor ATR. Nat Immunol. 2007;8(7):705–14. [PubMed] 8. Polo JM, Dell'Oso T, Ranuncolo SM, et al. Specific peptide interference reveals BCL6 transcriptional and oncogenic mechanisms in B-cell lymphoma cells. Nat Med. 2004;10(12):1329–35. [PubMed] 9. Huynh KD, Bardwell VJ. The BCL-6 POZ domain and other POZ domains interact with the co-repressors N-CoR and SMRT. Oncogene. 1998;17(19):2473–84. [PubMed] 10. Huynh KD, Fischle W, Verdin E, Bardwell VJ. BCoR, a novel corepressor involved in BCL-6 repression. Genes Dev. 2000;14(14):1810–23. [PubMed] 11. Ahmad KF, Melnick A, Lax S, et al. Mechanism of SMRT corepressor recruitment by the BCL6 BTB domain. Mol Cell. 2003;12(6):1551–64. [PubMed] 12. Parekh S, Polo JM, Shaknovich R, et al. BCL6 programs lymphoma cells for survival and differentiation through distinct biochemical mechanisms. Blood. 2007;110(6):2067–74. [PubMed] 13. Melnick A. Predicting the effect of transcription therapy in hematologic malignancies. Leukemia. 2005;19(7):1109–17. [PubMed] 14. Prive GG, Melnick A. Specific peptides for the therapeutic targeting of oncogenes. Curr Opin Genet Dev. 2006;16(1):71–7. [PubMed] 15. Melnick A, Licht JD. Deconstructing a disease: RARalpha, its fusion partners, and their roles in the pathogenesis of acute promyelocytic leukemia. Blood. 1999;93(10):3167–215. [PubMed] 16. Schmitt CA, Fridman JS, Yang M, Baranov E, Hoffman RM, Lowe SW. Dissecting p53 tumor suppressor functions in vivo. Cancer Cell. 2002;1(3):289–98. [PubMed] 17. Selivanova G, Wiman KG. Reactivation of mutant p53: molecular mechanisms and therapeutic potential. Oncogene. 2007;26(15):2243–54. [PubMed] 18. Snyder EL, Meade BR, Saenz CC, Dowdy SF. Treatment of terminal peritoneal carcinomatosis by a transducible p53-activating peptide. PLoS Biol. 2004;2(2):E36. [PubMed] 19. Renvoize C, Biola A, Pallardy M, Breard J. Apoptosis: identification of dying cells. Cell Biol Toxicol. 1998;14(2):111–20. [PubMed] 20. Cohen JJ. Apoptosis. Immunol Today. 1993;14(3):126–30. [PubMed] 21. Farrugia MM, Duan LJ, Reis MD, Ngan BY, Berinstein NL. Alterations of the p53 tumor suppressor gene in diffuse large cell lymphomas with translocations of the c-MYC and BCL-2 proto-oncogenes. Blood. 1994;83(1):191–8. [PubMed] 22. Mehra S, Messner H, Minden M, Chaganti RS. Molecular cytogenetic characterization of non-Hodgkin lymphoma cell lines. Genes Chromosomes Cancer. 2002;33(3):225–34. [PubMed] 23. Chang H, Blondal JA, Benchimol S, Minden MD, Messner HA. p53 mutations, c-myc and bcl-2 rearrangements in human non-Hodgkin's lymphoma cell lines. Leuk Lymphoma. 1995;19(1−2):165–71. [PubMed] 24. Taylor WR, Stark GR. Regulation of the G2/M transition by p53. Oncogene. 2001;20(15):1803–15. [PubMed] 25. Gudkov AV, Komarova EA. Prospective therapeutic applications of p53 inhibitors. Biochem Biophys Res Commun. 2005;331(3):726–36. [PubMed] 26. Wang W, El-Deiry WS. Targeting p53 by PTD-mediated transduction. Trends Biotechnol. 2004;22(9):431–4. [PubMed] 27. Dinnen RD, Drew L, Petrylak DP, et al. Activation of targeted necrosis by a p53 peptide: a novel death pathway that circumvents apoptotic resistance. J Biol Chem. 2007;282(37):26675–86. [PubMed] 28. Fingrut O, Reischer D, Rotem R, et al. Jasmonates induce nonapoptotic death in high-resistance mutant p53-expressing B-lymphoma cells. Br J Pharmacol. 2005;146(6):800–8. [PubMed] 29. Kroemer G, El-Deiry WS, Golstein P, et al. Classification of cell death: recommendations of the Nomenclature Committee on Cell Death. Cell Death Differ. 2005;12(Suppl 2):1463–7. [PubMed] 30. Eguchi Y, Shimizu S, Tsujimoto Y. Intracellular ATP levels determine cell death fate by apoptosis or necrosis. Cancer Res. 1997;57(10):1835–40. [PubMed] 31. Ha HC, Snyder SH. Poly(ADP-ribose) polymerase is a mediator of necrotic cell death by ATP depletion. Proc Natl Acad Sci U S A. 1999;96(24):13978–82. [PubMed] 32. Ying W, Alano CC, Garnier P, Swanson RA. NAD+ as a metabolic link between DNA damage and cell death. J Neurosci Res. 2005;79(1−2):216–23. [PubMed] 33. Los M, Mozoluk M, Ferrari D, et al. Activation and caspase-mediated inhibition of PARP: a molecular switch between fibroblast necrosis and apoptosis in death receptor signaling. Mol Biol Cell. 2002;13(3):978–88. [PubMed] 34. Bykov VJ, Issaeva N, Shilov A, et al. Restoration of the tumor suppressor function to mutant p53 by a low-molecular-weight compound. Nat Med. 2002;8(3):282–8. [PubMed] 35. Rehman A, Chahal MS, Tang X, Bruce JE, Pommier Y, Daoud SS. Proteomic identification of heat shock protein 90 as a candidate target for p53 mutation reactivation by PRIMA-1 in breast cancer cells. Breast Cancer Res. 2005;7(5):R765–74. [PubMed] 36. Sanchez-Beato M, Sanchez-Aguilera A, Piris MA. Cell cycle deregulation in B-cell lymphomas. Blood. 2003;101(4):1220–35. [PubMed] 37. Iqbal J, Greiner TC, Patel K, et al. Distinctive patterns of BCL6 molecular alterations and their functional consequences in different subgroups of diffuse large B-cell lymphoma. Leukemia. 2007;21(11):2332–43. [PubMed] 38. Cerchietti L, Figueroa M, Meyers D, Cole P, Bhalla K, Melnick A. Connectivity mapping of BCL6 targeted therapy guides rational design of potent and specific non-chemotherapy combinatorial regimens in DLBCL. Blood. 2007;110(11):161a. [PubMed] 39. Fujita N, Jaye DL, Geigerman C, et al. MTA3 and the Mi-2/NuRD complex regulate cell fate during B lymphocyte differentiation. Cell. 2004;119(1):75–86. [PubMed] 40. Wang W, Rastinejad F, El-Deiry WS. Restoring p53-dependent tumor suppression. Cancer Biol Ther. 2003;2(4 Suppl 1):S55–63. [PubMed]
|
The publisher's final edited version of this article is available free at 
