Cancer Cell Biology
Inhibition of CDK4 impairs proliferation of pancreatic cancer cells and sensitizes towards TRAIL-induced apoptosis via downregulation of survivin
Article first published online: 15 FEB 2007
Copyright © 2007 Wiley-Liss, Inc.
International Journal of Cancer
Volume 121, Issue 1, pages 66–75, 1 July 2007
How to Cite
Retzer-Lidl, M., Schmid, R. M. and Schneider, G. (2007), Inhibition of CDK4 impairs proliferation of pancreatic cancer cells and sensitizes towards TRAIL-induced apoptosis via downregulation of survivin. Int. J. Cancer, 121: 66–75. doi: 10.1002/ijc.22619
- Issue published online: 24 APR 2007
- Article first published online: 15 FEB 2007
- Manuscript Accepted: 12 JAN 2007
- Manuscript Received: 8 SEP 2006
- Deutsche Forschungsgemeinschaft. Grant Number: SFB456
- pancreatic cancer;
Pancreatic ductal adenocarcinoma is one of the most common causes of cancer death in Western countries with an average survival after diagnosis of 3–6 months and a five-year survival rate under 5%. Because of the lack of effective therapies, there is the need to characterize new molecular treatment strategies. Abnormal regulation of the cell cycle is a hallmark of neoplasia. Cyclin-dependent kinase 4 (CDK4), a key regulator of G1-phase of the cell cycle, has been shown to be overexpressed in pancreatic cancer. Until now, the contribution of CDK4 to tumor maintenance of pancreatic cancer has not been investigated. In this study, we used the chemical CDK4 inhibitor 2-bromo-12,13-dihydro-5H-indolo[2,3-a]pyrrolo[3,4-c]carbazole-5,7(6H)-dione, as well as RNA interference, to investigate the function of CDK4 in pancreatic cancer cells. Both approaches led to a reduction of pancreatic cancer cell proliferation due to G1-phase cell cycle arrest and Rb activation. Furthermore, we observed increased sensitivity of G1-arrested pancreatic cancer cells towards TRAIL-induced apoptosis. Sensitization towards TRAIL was due to the transcriptional downregulation of survivin. These findings show that a combined sensitizer/inducer strategy may be a potential therapeutic strategy for pancreatic ductal adenocarcinoma. © 2007 Wiley-Liss, Inc.
Pancreatic cancer is the fourth leading cause of cancer death in Western countries in both men and women and has a dismal prognosis. Approximately 32,000 people develop pancreatic adenocarcinoma each year in the United States.1 Despite surgical and medical management, the 5-year survival stays under 5%, demonstrating the insufficiency of the current therapeutic strategies. Hence, there is the urgent need to characterize new molecular therapeutic strategies in preclinical settings.
Disturbed regulation of the cell cycle is a hallmark of neoplasia. Cell cycle progression is regulated by concerted action of cyclins with cyclin-dependent kinases (CDKs). D-type cyclins interact with their catalytic subunits, CDK4 and CDK6, to control the progression through the G1-phase followed by the activation of cyclin E-CDK2 complexes during the late G1-phase.2 Sequentially, CDK4/6 and CDK2 complexes collaborate to phosphorylate and inactivate the retinoblastoma gene product Rb, to initiate an E2F-dependent transcriptional program, needed to enter S-phase.3 Overexpression of D-type cyclins and CDK4 is frequently found in pancreatic ductal adenocarcinomas.4, 5 Together with the biallelic inactivation of the INK4A locus, present in up to 95% of pancreatic ductal adenocarcinomas, D-type cyclin/CDK4 complexes contribute to the functional inactivation of the Rb-dependent G1-phase checkpoint.6, 7, 8, 9
Oncogenic stress induces intrinsic tumor suppressive pathways, such as oncogene-induced apoptosis and oncogene-induced premature senescence. During carcinogenesis, cells have to adopt the ability to conquer this intrinsic surveillance program.10 As a result, resistance towards apoptosis, an important factor for the survival and drug resistance of cancer cells, occurs.11, 12 Pancreatic ductal adenocarcinoma cells are characterized by a robust apoptosis-resistance, leading to the known treatment failure.13, 14 They express death receptors such as TNF-R, TRAIL-R (DR4 and DR5) and CD95, but resist towards death-receptor ligand induced apoptosis.15, 16, 17 Upon others, the IAPs (inhibitor of apoptosis proteins) family, including survivin, ML-IAP, XIAP, cIAP1, cIAP2, NIAP and Apollon, are important mediators of resistance towards death-receptor ligand induced apoptosis.18 IAPs have a widespread anti-apoptotic potential by blocking terminal effector caspases.19
Survivin, the smallest member (16.5 kD) of the IAP protein family, is selectively expressed in most human neoplasms including pancreatic cancer, but not in normal adult tissues.20, 21 Survivin has been implicated in the control of cell division and inhibition of apoptosis.22, 23, 24 It is expressed in a cell cycle-regulated manner with high levels in G2/M and low levels in the G1-phase. Downregulation of survivin is associated with enhanced sensitivity towards TRAIL-induced apoptosis in SHEP neuroblastoma cells, hepatoma cells, human glioma cells, and human breast cancer cells.25, 26, 27, 28 Furthermore, downregulation of survivin by RNA interference induces apoptosis and enhances radiosensitivity of pancreatic ductal adenocarcinoma cells.29
In the current study, we validate CDK4 as a therapeutical drug target in pancreatic ductal adenocarcinoma. Inhibition of CDK4 activity or CDK4 downregulation results in Rb-activation and G1-phase arrest. Furthermore, blocking CDK4 sensitizes pancreatic ductal adenocarcinoma cells towards TRAIL-induced apoptosis by transcriptional downregulation of survivin.
Material and methods
2-Bromo-12,13-dihydro-5H-indolo[2,3-a]pyrrolo[3,4-c]carbazole-5,7(6H)-dione (EMD Biosciences, San Diego, CA) and zVAD-fmk (Biomol, Hamburg, Germany) were dissolved in DMSO and stored at −20°C. Thymidine and mimosine were purchased from Biomol, Hamburg, Germany. TRAIL was purchased from EMD Biosciences, San Diego, CA.
Cell culture, transfection, siRNAs, plasmids
Pancreatic cancer cells were cultivated as described.30 Double stranded siRNAs were transfected in a concentration of 200 nM using oligofectamine (Invitrogen, Karlsruhe, Germany) as recently described.30 For the simultaneous transfection of siRNAs directed against two different genes, the total amount of siRNA (400 nM) was kept constant using scramble siRNA. siRNA sequences used (target sequence, sense strand): scramble control siRNA 5′AACAGTCGCGTTTGCGACTGG3′, CDK4 siRNA-1 5′AAGGCCCGTGATCCCCACAGT3′, CDK4 siRNA-2 5′AAGCCGACCAGTTGGGCAAAA3′, CDK4 siRNA-3 5′AAGTTCGTGAGGTGGCTTTAC3′; survivin siRNA-1 5′AAGGACCACCGCATCTCTACA3′; survivin siRNA-2 5′AAGCATTCGTCCGGTTGCGCT3′. siRNAs were purchased from Ambion, Austin, TX, and were stored in a 20 μM stock at −80°C.
The survivin luciferase reporter gene constructs SpII (−939/−15), SpV (−565/−15) and SpVI (−105/−15) were kindly provided by Dr. Maureen Murphy.31 The pGL3control vector was used as a negative control. For simultaneous transfections of the reporter gene (500 ng/well) and double stranded siRNA (200 nM), oligofectamine (Invitrogen, Karlsruhe, Germany) was used. After 48 hr, the cells were incubated in lysis buffer (Promega, Mannheim, Germany) for 15 min, harvested, and cleared by centrifugation for 15 min. Lysates were normalized for protein content. Luciferase activity was determined in a LB 9501 luminometer (Berthold, Bad Wildbad, Germany) using a luciferase assay system (Promega, Mannheim, Germany). At least 3 independent transfection experiments were performed in triplicate. Data are presented as mean and standard error of the mean (SEM). The GFP-survivin expression vector (pReceiver-M03-GFP-survivin) was purchased from RZPD (RZPD, Berlin, Germany). pEGFP-C1 (Invitrogen, Karlsruhe, Germany) was used as a control. One microgram of the GFP constructs was transfected using Fugene (Roche Applied Sciences, Mannheim, Germany). All plasmids were verified by sequencing.
Preparation of total cell lysates
Whole-cell lysates were prepared by incubating cell pellets for 30 min at 4°C in immunoprecipitation buffer (50 mM HEPES, 150 mM NaCl, 1 mM EDTA, 0.5% NP-40, 10% glycerol, 1 mM dithiothreitol, 1 mM phenylmethysulfonylfluoride and 5 mM NaF). Insoluble material was removed by centrifugation, and lysates were aliquoted and stored at −80°C.
Western blot analysis
Extracts were normalized for protein and heated at 95°C for 5 min in Lämmli buffer. Proteins were resolved on 10% SDS-polyacrylamide gels and electrophoretically transferred to polyvinylidene difluoride (Millipore, MA) membranes in a semidry blotting system. Membranes were blocked in phosphate-buffered saline (PBS) supplemented with 5% skimmed milk and 0.1% NP-40 and were incubated with antibody for 1 hr at room temperature. The membranes were incubated with antibodies against CDK4, cyclin A, cyclin D1 (Santa Cruz Biotechnology, Santa Cruz, CA), Rb, p27Kip1, PARP (BD Pharmingen, San Diego, CA), survivin (R&D Systems, Wiesbaden-Nordenstadt, Germany), and against beta-actin (Sigma-Aldrich, Munich, Germany). Proteins recognized by the antibodies were detected by the Odyssey Infrared Imaging System (Licor, Bad Homburg, Germany) using Alexa680-coupled (Molecular Probes, Leiden, Netherland) or IRDeye800-coupled (Rockland, Gilbertsville, PA) secondary antibodies.
BrdU incorporation assay and apoptosis stain
BrdU incorporation was measured using the colorimetric BrdU assay according to the manufacturer's protocol (Roche Applied Science, Mannheim, Germany). DNA was stained with 4 μM Hoechst 33342 and apoptotic changes were visualized using fluorescence microscopy. Three hundred cells were counted per individual experiment. Three independent experiments were performed as triplicates. Data are presented as mean and standard error of the mean (SEM).
Quantitative reverse-transcriptase PCR
Total RNA was isolated from pancreatic carcinoma cell lines using the RNeasy kit (Qiagen, Hilden, Germany) according to the manufactures instructions. One hundred nanograms of total RNA were used to perform reverse transcriptase PCR using the TaqMan Reverse Transcription reagents (Applied Biosystems, Norwalk, CT). Quantitative mRNA analysis was performed using real-time PCR analysis as previously described.30 Primer sequences are available upon request. At least 3 independent transfection experiments were performed in triplicate. Data are presented as mean and standard error of the mean (SEM).
Cell cycle analysis
For cell cycle analysis, cells were washed twice in PBS and redissolved in propidium iodide (PI) staining buffer containing 0.1% sodium citrate, 0.1% Triton X-100, and 50 μg/ml PI. After 1 h of incubation at 4°C, flow cytometry was performed using a BD Biosciences FACScan. The distribution of cells in different cell cycle stages (subG1/G1/S/G2+M) was determined according to their DNA content. Thus, cells possessing 2n (diploid) DNA were assigned to be in G1, those having a DNA content between 2n and 4n (tetraploid) were defined as S-phase cells and those having a 4n DNA content were assumed to be in G2+M. Apoptotic cells with internucleosomal DNA fragmentation possessing a hypoploid DNA content were assigned to be in sub G1. The sum of all cells in FACS analysis was set to 100% and the distribution into sub-G1, -G1, -S, and -G2+M phases was determined by counting the cells in each window according to the definitions mentioned earlier, using the CellQuest software.
Knockdown of CDK4 impairs proliferation of pancreatic cancer cells by inducing G1 cell cycle arrest
Abnormalities in the normal regulation of the cell cycle lead to uncontrolled proliferation and tumor formation. Since CDK4 is overexpressed in pancreatic ductal adenocarcinoma, we investigated the function of CDK4 in pancreatic ductal adenocarcinoma cells using RNA interference.5 CDK4 protein abundance is reduced 48 hr after the transfection of MiaPaCa2 and DanG with CDK4 specific siRNAs compared with untransfected or control siRNA transfected cells (Fig. 1a). Next, we examined the proliferation of MiaPaCa2 and DanG cells following knockdown of CDK4. Forty-eight hours after transfection with the CDK4 specific siRNAs, BrdU incorporation was reduced to 42–49% in MiaPaCa2 cells and to 62–66% in DanG cells compared with untransfected controls (Fig. 1b). Reduced proliferation is due to a G1-phase cell cycle arrest. After the knockdown of CDK4, 74–76% of MiaPaCa2 were found in G1-phase compared with 54% in untransfected controls (Fig. 1c). In line, transfection of CDK4-specific siRNAs into DanG cells leads to an accumulation of 56–62% of the cells in G1-phase compared with 47% in untransfected controls (Fig. 1c). As shown in Figure 1d, Rb is activated after the CDK4 knockdown in MiaPaCa2 and DanG cells.
These findings demonstrate that CDK4 activity contributes to inactivation of a Rb-dependent G1-phase checkpoint in pancreatic ductal adenocarcinoma cells.
Inhibition of CDK4 by the chemical inhibitor 2-bromo-12,13-dihydro-5H-indolo[2,3-a]pyrrolo[3,4-c]carbazole-5,7(6H)-dione impairs proliferation of pancreatic cancer cells by inducing G1 cell cycle arrest
To further validate CDK4 as a specific drug target in pancreatic ductal adenocarcinoma cells, we used a specific chemical inhibitor. 2-Bromo-12,13-dihydro-5H-indolo[2,3-a]pyrrolo[3,4-c]carbazole-5,7(6H)-dione (BIPCD) is a cell-permeable, potent and selective ATP-competitive inhibitor of CDK4.32 As shown in Figure 2a, BIPCD dose-dependently reduced BrdU incorporation of MiaPaCa2 cells to 49% compared with untreated controls. In DanG cells, BrdU incorporation was dose-dependently reduced to 42% 24 hr after the treatment with BIPCD compared with untreated controls (Fig. 2a). Figure 2b demonstrates that after 24 hr of BIPCD treatment, MiaPaCa2 and DanG cells accumulate in G1-phase of the cell cycle in a dose-dependent manner. Again, we found activation of Rb after the treatment of MiaPaCa2 and DanG cells with BIPCD (Fig. 2c).
Knockdown or inhibition of CDK4 sensitizes the pancreatic cancer cell line MiaPaCa2 to TRAIL-induced apoptosis
Several studies show cell cycle-dependent sensitivity of cancer cells towards induction of apoptosis.33, 34 To examine if G1 cell cycle arrest sensitizes pancreatic cancer cells towards TRAIL-induced apoptosis, MiaPaCa2 cells were treated with increasing concentrations of TRAIL after CDK4 knockdown or BIPCD treatment. Forty-eight hours after CDK4 knockdown or 24 hr after CDK4 inhibition, MiaPaCa2 cells were treated with increasing concentrations of TRAIL for 24 hr and the apoptotic fraction after Hoechst stain was determined. Forty-eight hours after transfection with CDK4 siRNA, the apoptotic fraction of MiaPaCa2 cells was 7–21% (depending on the siRNA) (Fig. 3a). After treatment with increasing concentrations of TRAIL, the apoptotic fraction of MiaPaCa2 cells increased to 30–37% (CDK4 siRNA-1 and -3) at a TRAIL concentration of 5 ng/ml, to 49–62% (CDK4 siRNA-1 and -3) at a TRAIL concentration of 10 ng/ml and to 64–67% (CDK4 siRNA-1 and -3) at a TRAIL concentration of 30 ng/ml. In comparison, 11% of untransfected MiaPaCa2 cells were apoptotic after treatment with 30 ng/ml TRAIL. Simultaneous treatment with TRAIL and zVAD-fmk reduced the apoptotic fraction to 2–3%, suggesting caspase-dependent apoptosis. As shown in Figure 3b, 24 hr after treatment of MiaPaCa2 cells with the CDK4 inhibitor in a concentration of 2.5, 4, or 5 μM, the apoptotic fraction of MiaPaCa2 cells was 3%, 6%, and 7% respectively. After treatment with increasing concentrations of TRAIL, the apoptotic fraction of MiaPaCa2 cells treated with the CDK4 inhibitor in a concentration of 5 μM increased to 10% at a TRAIL concentration of 5 ng/ml, to 17% at a TRAIL concentration of 10 ng/ml and to 54% at a TRAIL concentration of 30 ng/ml. Simultaneous treatment with TRAIL and zVAD-fmk again reduced the apoptotic fraction to 1%. As apoptotic cells accumulate in the sub-G1-phase of the cell cycle, we next performed cell cycle analysis. Forty-eight hours after CDK4 knockdown or 24 hr after CDK4 inhibition, MiaPaCa2 cells were treated with TRAIL for 24 hr. Consistent with the results, the aforementioned treatment with increasing concentrations of TRAIL resulted in a dose-dependent accumulation of MiaPaCa2 cells transfected with CDK4 siRNA or treated with the CDK4 inhibitor in the sub-G1-phase of the cell cycle (Figs. 3c and 3d). These data demonstrate that G1 cell cycle arrest sensitizes MiaPaCa2 cells to TRAIL-induced caspase-dependent apoptosis.
G1-phase arrest sensitizes MiaPaCa2 cells to TRAIL-induced apoptosis
To discriminate between a cell cycle- and a CDK4-specific mechanism, explaining TRAIL sensitization, we performed synchronization experiments in MiaPaCa2 cells. Thymidine induced S-phase arrest is controlled by high cyclin A expression and mimosine induced late G1-phase arrest demonstrates low p27Kip1 and low cyclin A expression (Fig. 4a). The TRAIL-induced apoptotic fraction was nearly equal in random cycling (10%) and S-phase (14%) MiaPaCa2 cells 24 hr after the treatment with 30 ng/ml TRAIL (Fig. 4b). Significant apoptosis was observed in mimosine treated MiaPaCa2 cells. The apoptotic fraction was further increased to 52% 24 hr after the treatment of late G1-phase arrested MiaPaCa2 cells with 30 ng/ml TRAIL (Fig. 4b). This suggests cell cycle-dependent sensitivity towards TRAIL-induced apoptosis in pancreatic cancer cells.
CDK4 knockdown or inhibition induces down-regulation of survivin protein and survivin mRNA
Next we examined the molecular mechanisms linking cell cycle regulation and apoptosis sensitivity after CDK4 knockdown or inhibition. Since survivin is a member of the inhibitor of apoptosis protein family and is expressed in a cell-cycle regulated manner with low levels in G1-phase, we analyzed survivin protein abundance after CDK4 knockdown or inhibition by immunoblotting. As shown in Figure 5a, survivin protein abundance is clearly reduced in MiaPaCa2 cells 48 hr after CDK4 siRNA transfection. Furthermore, CDK4 inhibition by BIPCD resulted in a dose-dependent decrease in survivin protein in MiaPaCa2 cells 24 hr after treatment (Fig. 5b). In addition, mimosine induced G1-phase arrest also leads to a downregulation of survivin abundance in MiaPaCa2 cells (Fig. 4a). To determine whether the cell cycle-dependent decrease of survivin expression in G1 is mediated via a transcriptional mechanism, we analyzed survivin mRNA expression. MiaPaCa2 cells were transfected with CDK4 siRNAs and assessed for survivin mRNA expression by real-time PCR 48 hr after transfection. As shown in Figure 5c, survivin mRNA expression was downregulated to 18–34% compared with untreated control cells. Next we examined survivin promoter activity after CDK4 knockdown in MiaPaCa2 cells, using the recently described survivin luciferase reporter gene constructs SpII (−939 to −15, where +1 represents the ATG), SpV (−566 to −15), and SpVI (−105 to −15) in luciferase reporter gene assays.31 As shown in Figure 5d, CDK4 knockdown led to a significant reduction of survivin promoter activity. A reduction to 23–50% in promoter activity after the CDK4 knockdown compared with control siRNA transfected MiaPaCa2 cells was observed for the SpII survivin promoter construct (Fig. 5d). The activity of the SpV construct was reduced to 20–40% after the CDK4 knockdown and the activity of the SpVI construct was reduced to 24–54% after the CDK4 knockdown, respectively (Fig. 5d). These data suggest that the survivin gene is transcriptionally downregulated after CDK4 knockdown and that the proximal survivin promoter harbors a CDK4 responsive element.
Knockdown of survivin sensitizes MiaPaCa2 cells to TRAIL-induced apoptosis
To gain further insight into the function of survivin as an anti-apoptotic molecule and to investigate the contribution of downregulation of survivin in sensitizing MiaPaCa2 cells to TRAIL-induced apoptosis, we used RNA interference. In the pancreatic ductal adenocarcinoma cell line, MiaPaCa2 survivin protein abundance is clearly reduced 48 hr after survivin siRNA transfection compared with control siRNA transfection (Fig. 6a). To determine if the knockdown of survivin enhances the sensitivity of MiaPaCa2 cells to TRAIL-induced apoptosis, MiaPaCa2 cells were treated with TRAIL (30 ng/ml) and the apoptotic fraction after Hoechst stain was determined. Forty-eight hours after transfection with survivin siRNAs, the apoptotic fraction of MiaPaCa2 cells was 20–25% (Fig. 6b). After treatment with TRAIL (30 ng/ml) the apoptotic fraction of MiaPaCa2 cells transfected with survivin siRNA increased to 45–52%. In comparison, the apoptotic fraction of control MiaPaCa2 cells was about 6% at a TRAIL concentration of 30 ng/ml. Simultaneous treatment with TRAIL and zVAD-fmk reduced the apoptotic fraction of MiaPaCa2 cells transfected with survivin siRNA to 11–12%, suggesting caspase contribution. To demonstrate that CDK4 inhibition induced sensitization towards TRAIL depends on survivin down-regulation, we transfected MiaPaCa2 cells with a survivin-specific siRNA and simultaneously inhibited CDK4 using BIPCD. As shown in Figure 6c, we observed no additive or synergistic effects of the simultaneous knockdown of survivin and BIPCD treatment for sensitization towards TRAIL-induced apoptosis. In addition, no additive or synergistic effect of the simultaneous knockdown of survivin and CDK4 towards TRAIL-induced apoptosis was observed. As shown in Figure 6d, no increased TRAIL-induced PARP cleavage was noticed in MiaPaCa2 transfected simultaneously with the survivin and CDK4 siRNA. In line, we found no significant difference in the TRAIL-induced apoptotic fraction in MiaPaCa2 cells transfected with the survivin siRNA, CDK4 siRNA, or both siRNAs (Fig. 6e). These data suggests that cell-cycle dependent sensitization towards TRAIL works via survivin downregulation.
Survivin overexpression confers TRAIL resistance to pancreatic cancer cells
To further confirm that survivin transfers TRAIL resistance in pancreatic cancer cells, we overexpressed GFP-tagged survivin in BIPCD treated MiaPaCa2 cells and measured the TRAIL-induced apoptotic fraction of GFP positive cells using Hoechst stain. In BIPCD treated and control GFP vector transfected MiaPaCa2 cells, 59% of GFP positive cells were apoptotic 24 hr after the treatment with 10 ng/ml TRAIL. The apoptotic fraction was further increased to 82% 24 hr after the treatment with 30 ng/ml TRAIL (Fig. 7a). In contrast, in BIPCD treated and GFP-survivin transfected MiaPaCa2 cells, only 36% of GFP positive cells were apoptotic 24 hr after the treatment with 10 ng/ml TRAIL, and 56% of GFP positive cells were apoptotic 24 hr after the treatment with 30 ng/ml TRAIL (Fig. 7a). These data demonstrates that survivin contributes to the sensitization towards TRAIL observed after the CDK4 inhibition in MiaPaCa2 cells.
The pancreatic cancer cell line Capan1 was recently characterized by TRAIL sensitivity and low survivin expression levels.35 In control GFP vector transfected Capan1 cells, 41% of GFP positive cells were apoptotic 24 hr after the treatment with 10 ng/ml TRAIL, and 56% of GFP positive cells were apoptotic 24 hr after the treatment with 30 ng/ml TRAIL (Fig. 7b). In GFP-survivin transfected Capan1 cells, the TRAIL-induced apoptotic fraction was reduced to 33% 24 hr after the treatment with 10 ng/ml TRAIL and to 39% 24 hr after the treatment with 30 ng/ml TRAIL, again disclosing contribution of survivin towards TRAIL resistance of pancreatic cancer cells (Fig. 7b).
In the current work, we demonstrate that interfering with CDK4 induces a cytostatic-response of pancreatic cancer cells, validating CDK4 as a drug target in pancreatic ductal adenocarcinoma. The impaired proliferation is due to a Rb-dependent G1-phase arrest. Furthermore, we demonstrate that CDK4 inhibition sensitizes pancreatic ductal adenocarcinoma cells towards TRAIL-induced apoptosis in a preclinical in vitro model. Therefore, a sensitizer (CDK4 inhibition)/inducer (TRAIL) strategy may be a feasible approach to treat pancreatic ductal adenocarcinoma in the future.
CDK4 is a key regulator of normal and cancerous cell cycle progression.36, 37 The importance of CDK4 in tumorigenesis is seen in sporadic and familiar melanoma, where miscoding mutations in the CDK4 locus, rendering CDK4 insensitive towards INK4A inhibition, were observed.38, 39 Further experimental evidence for the fundamental role of CDK4 in tumor-formation is provided by CDK4 knock-in mice, harboring an Arg24 to Cys mutation. Homozygous CDK4R24C/R24C mice develop multiple tumors with a high penetrance.40 CDK4 is overexpressed in pancreatic ductal adenocarcinoma.5 Inappropriate or enhanced CDK4 activity results in loss of proliferative control by Rb and consequent activation of an E2F-dependent transcriptional program, needed to enter S-phase.36 Though CDK4 is dispensable for proliferation in some mammalian cell types, our data show that CDK4 plays an important role in controlling a Rb-dependent G1-phase checkpoint in pancreatic ductal adenocarcinoma cells.41 CDK4 knockdown or inhibition leads to hypophosphorylation of Rb, G1-phase cell cycle arrest, and reduced proliferation. In CDK4 and CDK6, double knock-out MEFs CDK2 can compensate for CDK4/6 loss.41 Since we observed a clear activation of Rb after interfering with CDK4, CDK6 and CDK2 seem to be dispensable for Rb-regulation and G1- to S-phase progression in pancreatic ductal adenocarcinoma cells.
Survivin regulates both cell cycle control and apoptosis.23 Although survivin expression is a hallmark of various cancers, regulation of its expression is not entirely clear. Survivin displays a cell cycle regulated expression that peaks in the G2/M-phase, and the TATA-less survivin promoter is typically transactivated in M-phase, resulting in an approximately 40-fold increase of survivin mRNA.42, 43 In our cellular model, survivin expression is also cell-cycle-dependent and decreased expression was found in G1-phase. Although survivin abundance is regulated by proteasomal degradation and posttranslational modifications, we provide evidence that survivin is mainly regulated by transcription in G1 arrested pancreatic ductal adenocarcinoma cells.44, 45 Several transcription factors, like SP1, TCF/β-catenin, NF-κB, E2F family members, STAT3, DEC1, KLF5 and p53 are known to regulate the survivin promoter.31, 46, 47, 48, 49, 50, 51 We demonstrate that the proximal survivin promoter responded in cis to the G1 arrest achieved by the CDK4 knockdown. Since this promoter part is known to be regulated by E2F, Rb is activated by interfering with CDK4 activity in pancreatic ductal adenocarcinoma cells and hypophosphorylated Rb binds to E2F and converts E2F from a transcriptional activator to a repressor, the Rb/E2F system is likely to play a significant role in the observed survivin promoter regulation.46 This explanation awaits further experimental validation.
TRAIL is a cytokine and potent inducer of apoptosis in various cancers. TRAIL appears to be a relatively safe and promising agent in cancer therapy, because it induces apoptosis in a variety of human cancer cell lines, but not in normal cells.52 Despite intact receptors and signaling machineries, pancreatic ductal adenocarcinoma cells show reduced susceptibility to cell death induced by death receptors.15, 53, 54 Although 50% of Jurkat cells die after treatment with 7 ng/ml TRAIL, the LD50 of most pancreatic cancer cell lines, including MiaPaCa2, is greater than 85 ng/ml.17 Recently, TRAIL resistance in pancreatic ductal adenocarcinoma cells was correlated with increased c-Flip expression and was regulated by XIAP.35, 55 In addition to c-Flip and XIAP, we now show that survivin is important for TRAIL resistance of pancreatic ductal adenocarcinoma cells. We demonstrate that interfering with survivin expression either by arresting pancreatic ductal adenocarcinomas in the G1-phase (via impairing CDK4 function) or by direct targeting survivin expression by RNAi sensitizes pancreatic ductal adenocarcinoma cells towards TRAIL-induced apoptosis. Furthermore, overexpression of survivin reduces TRAIL-induced apoptosis of the TRAIL sensitive pancreatic cancer cell line Capan1 and the G1-phase arrested MiaPaCa2 cells. These results confirm recent studies showing increased TRAIL sensitivity after lowering survivin abundance.25, 26, 27, 28 In addition to CDK4 knockdown induced G1-phase arrest, we demonstrate that arresting pancreatic cancer cells by mimosine in G1-phase of the cell cycle sensitizes towards TRAIL-induced apoptosis. Since mimosine is known to arrest the cell cycle in the late G1-phase and as we observed Rb hypophosphorylation as a marker of late G1-phase checkpoint activation after interfering with CDK4 function, a narrow window in late G1-phase is TRAIL sensitive in pancreatic cancer cells.56 Downregulation of survivin after the mimosine induced G1-phase arrest also points to an important contribution of survivin for sensitization towards TRAIL.
In conclusion, our results demonstrate that the combined sensitizer (CDK4 knockdown/inhibition/G1-arrest)/inducer (e.g. TRAIL) strategy may be a novel approach for the treatment of pancreatic ductal adenocarcinoma.
We thank Ms. Birgit Kohnke-Ertel and Ms. Konstanze Geiger for excellent technical support and Dr. Maureen Murphy for generously providing the survivin promoter constructs. This work was funded by grants from the Deutsche Forschungsgemeinschaft (SFB456 to G.S. and R.S.).