Abstract
Background/Aim: This study was designed to analyse the effects of the novel, orally bioavailable CDK9-inhibitor Atuveciclib (BAY 1143572) in combination with tumour necrosis factor-related apoptosis-inducing ligand (TRAIL) on pancreatic ductal adenocarcinoma (PDAC) cancer cells. Materials and Methods: To assess the effect of combinatorial use of atuveciclib and TRAIL on pancreatic cancer cells, we used an MTT assay, colony formation assay, flow cytometry, and western blot analysis. Results: Atuveciclib combined with TRAIL significantly reduced the viability of pancreatic cancer cells and their colony formation potential by inducing apoptosis and cell-cycle arrest. Atuveciclib sensitised PDAC cells to TRAIL-induced cell death through the concomitant suppression of cFlip and Mcl-1. A gemcitabine-resistant PDAC cell-line and patient-derived xenograft (PDX) cell lines were also suppressed by this combinatorial approach. Conclusion: This study provides the basis for further preclinical and clinical evaluation of combined treatment with atuveciclib and TRAIL.
For all stages combined, pancreatic ductal adenocarcinoma has a five-year survival rate of 9%, which has hardly improved over recent years (1). This type of cancer thus continues to be a major oncological challenge. Treatment options are considerably limited since pancreatic adenocarcinoma, which is characterised by non-specific symptoms and a tendency for early local invasion and distant metastasis, is normally diagnosed at a late stage (2). At the time of diagnosis, surgery is possible in only 15-20% of pancreatic carcinomas. Radical surgical resection, however, is the only potential cure, which is thus difficult to achieve (3-5). Adequate and effective chemotherapeutic options are therefore urgently required. Apart from early invasion and distant metastasis, resistance to conventional chemotherapy (2, 4, 6-9) and a tendency for local recurrence and metachronous metastases (2, 10) impede adjuvant treatment. Adjuvant treatment currently consists of two major different schemes: The more effective but more toxic modified FOLFIRINOX scheme, which is now recommended for patients with good performance status and alternatively gemcitabine/capecitabine, which should be used for patients with contraindications for mFOLFIRINOX (11).
For this reason, new innovative treatment approaches are urgently required. The induction of apoptosis is such an approach to cancer treatment. In the 1990s, TRAIL, whose C-terminal domain shows clear homology to other TNF family members, was identified on human cells. Very low concentrations of TRAIL were found to selectively induce apoptosis in transformed cells but not in normal epithelial cells (12, 13). Unfortunately, studies have shown that approximately 50% of all established cancer cell lines and most primary tumour cell lines are resistant to TRAIL (14, 15). For this reason, it is important to identify substances that can be used in combination with TRAIL to overcome the resistance of cancer cells to TRAIL.
Three major cellular mechanisms were found to lead to TRAIL resistance. First, high levels of the X-linked inhibitor of apoptosis protein (XIAP), which can prevent the induction of apoptosis along the extrinsic pathway. Apoptosis induction then requires the release of a second mitochondria-derived activator of caspase (SMAC)/DIABLO, an inhibitor of XIAP. This can be achieved through the Bid-tBid-Bax/Bak-MOMP pathway (14, 16-21). Second, Mcl-1, an anti-apoptotic protein of the Bcl-2 family, can interact with truncated Bid (tBid) and thus inhibit mitochondrial outer membrane permeabilisation (MOMP) (22-25). This component of the signalling cascade is of particular importance for type II cells that require the intrinsic pathway for the activation of apoptosis. This explains why high levels of Mcl-1 can contribute to TRAIL resistance (16-19). Third, cFlipS and cFlipR exert anti-apoptotic functions by competing with procaspase 8 for binding to FAS-associated death domain (FADD), thereby preventing the conversion of procaspase 8 to active caspase 8 (26, 27). The third splice variant, cFlipL, exerts pro-apoptotic effects at low concentrations by facilitating the activation of caspase-8 in the death-inducing signalling complex (DISC), whereas at intermediate or high levels, which are probably commonly found in tumours, acts as an inhibitor of the DISC and, thus, has anti-apoptotic effects (28). Selective CDK9 inhibition was reported to suppress not only Mcl-1 but also cFlip, thereby sensitizing TRAIL-resistant cells to TRAIL-induced apoptosis (14). The combination of CDK9 inhibitors and TRAIL is thus a promising approach to sensitise cancer cells to TRAIL-induced apoptosis (29).
Atuveciclib (BAY 1143572) is a novel, orally bioavailable and highly selective CDK9 inhibitor (30). This substance is currently being investigated in phase I clinical studies for advanced solid tumours and acute leukaemia (NCT01938638, NCT02345382). It has shown promising anti-tumour activity in acute T-cell leukaemia (31, 32). The study presented here investigates the effects of atuveciclib alone or in combination with TRAIL on pancreatic cancer cells with a focus on cell viability, apoptosis induction and molecular effects at the protein level. We were able to show that atuveciclib in combination with TRAIL significantly reduces the viability of pancreatic cancer cells. Mechanistically, this combination treatment induces apoptosis and cell-cycle arrest. Western blot analysis confirmed that atuveciclib-mediated CDK9 inhibition sensitises cancer cells to TRAIL-induced cell death through the concomitant suppression of cFlip and Mcl-1. In addition, we showed that this combination treatment significantly reduces the colony formation potential of pancreatic cancer cells and potently suppresses gemcitabine-resistant cells as well as patient-derived xenograft (PDX) cell lines.
Materials and Methods
Cell culture. Human pancreatic ductal adenocarcinoma (PDAC) cell lines (Panc89, PancTu-1 and Colo357) were cultured in RPMI 1640 medium containing 1 mM sodium pyruvate, 2 mM L-glutamine, 100 units/ml penicillin/streptomycin, and 10% fetal calf serum (FCS). Cell lines 609, 722 and 1157 (kindly provided by Andrew S. Liss, PhD, Massachusetts General Hospital, Boston, MA, USA) were established from a PDX model using primary PDAC cells and were cultured in Dulbecco’s modified Eagle’s medium (DMEM)/F-12 (1:1) supplemented with 100 units/ml penicillin/streptomycin and 10% FCS. A gemcitabine-resistant clone of cell line Panc89 (Panc89-GR3), which was established by our research group, was cultivated in the same medium (supplemented with 1 μM gemcitabine) as normal Panc89 cells.
MTT assay. The MTT assay was performed as described previously (33). In brief, cells were seeded in quadruplets at 15,000 cells/well for Panc89 and Colo357 and 20,000 cells/well for PancTu-1, Panc89-GR3 and 609, 722 and 1157 cell lines. Cells were cultivated for 24 h before treatment with atuveciclib (MedChemExpress, Monmouth Junction NJ, USA), izTRAIL (kindly provided by Henning Walczak, UCL Cancer Institute, London, UK) or z-VAD-FMK (Selleck Chemicals, Houston TX, USA) with the concentrations indicated in the results section. Cell viability was evaluated after 24 h.
Colony formation assay. Colony formation assay was performed as described (34) according to the protocol of Millipore (Catalog No. ECM570, Merck KGaA, Darmstadt, Germany). In brief, 10,000 cells suspended in 0.5 ml of a 0,4% top agar (Biozym Plaque GeneticPure Agarose, Biozym Scientific GmbH, Hessisch Oldendorf, Germany) containing media and supplements were added to one ml of 0,8% base agar containing the media and supplements per well in a 6-well-format. After 24 h of incubation, 1 ml of culture medium was added to each agar plate, containing 1 ng/ml izTRAIL, 10 ng/ml izTRAIL, 1 μM atuveciclib, 1 μM atuveciclib+1 ng/ml izTRAIL, 1 μM atuveciclib+10 ng/ml izTRAIL, or DMSO (control). Cells were then cultivated for 24 h. Following incubation, cultivation media were carefully removed, the wells were washed twice with Dulbecco’s phosphate buffered saline, followed by an incubation period of 28 days with complete medium without inhibitors. After 28 days of incubation, the plates were examined microscopically for colony formation. Six photos were taken per well. Maximum colony size was assessed in every photo using ImageJ. Cells were treated with an MTT solution for 4 h for colony counting. After 4 hs, viable colonies (positively stained with MTT) were counted manually.
Cell-cycle analysis by flow cytometry. Apoptosis was quantified using propidium iodide staining and flow cytometry, as described by Riccardi and Nicoletti (35). Panc89 cells were diluted and incubated in 6-cm cell culture dishes for 48 h at 37°C and 5% CO2 to 70-80% confluence. Cells were then treated for 24 h as follows: izTRAIL at concentrations of 0 ng/ml, 0.1 ng/ml, 1 ng/ml, 10 ng/ml, and 100 ng/ml in the presence of atuveciclib at concentrations of 0 μM, 0.1 μM, 0.5 μM, and 1 μM. Following this treatment, cells were prepared for flow cytometry and stained with propidium iodide, as described by Riccardi and Nicoletti. Detached or suspended cells were collected by centrifugation. Stained cells were then analysed by flow cytometry using a MACSQuant® X (Miltenyi Biotec, Bergisch Gladbach, Germany). Propidium iodide was detected in the B3 channel (655-733 nm).
Western blot analysis. Cells and supernatants were lysed in lysis buffer (30 mM Tris-HCl (pH 7.4), 150 mM NaCl, 2 mM EDTA, 2 mM KCl, 10% glycerol, 1% Triton X-100, fresh 1% Phosphatase Inhibitor Cocktail 2 (Sigma-Aldrich, St. Louis, MO, USA) and 1% Halt™ Protease Inhibitor Cocktail (ThermoFisher Scientific, Waltham, MA, USA). Proteins were separated by SDS polyacrylamide gel electrophoresis and transferred onto a polyvinylidene difluoride (PVDF) membrane. Primary antibodies were diluted 1:1000. The following primary antibodies were used: anti-β-actin (1:1,000, Sigma-Aldrich), anti-Bid (Cell Signaling, Frankfurt am Main, Germany), anti-caspase-3 (R&D Systems, Minneapolis MN, USA), anti-caspase-8 (Enzo, Lörrach, Germany), anti-caspase-9 (Cell Signaling), anti-CDK9 (Cell Signaling), anti-cFlip (NF6) (AdipoGen, Liestal, Switzerland), anti-cyclin T1 (Cell Signaling), anti-Mcl-1 (Cell Signaling), anti-PARP (Cell Signaling), anti-pSer2 RNA Pol II (Cell Signaling), anti-RNA Pol II (Cell Signaling). The following secondary antibodies were used at a dilution of 1:10,000 for the detection of immunocomplexes: anti-mouse IgG, HRP-linked antibody (Cell Signaling), anti-rabbit IgG, HRP-linked antibody (Cell Signaling), goat IgG horseradish peroxidase-conjugated antibody (R&D Systems). Chemiluminescence was detected using ECL solutions. (Solution A: 1 ml Tris-HCl pH 8.5+9 ml H2O+100 μl luminol (250 mM in DMSO)+44 μl coumarin (90 mM in DMSO). Solution B: 1 ml Tris-HCl pH 8.5+9 ml H2O+15 μl H2O2. Solutions were prepared just before application).
Statistical analysis. Data were analysed using GraphPad Prism 7 (GraphPad Software, San Diego, CA, USA). Statistical significance was determined with one-way analysis of variance (ANOVA). Significant p-Values are marked with asterisks: *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.
Results
Atuveciclib can suppress the viability of established pancreatic ductal adenocarcinoma (PDAC) cell lines. First, the effects of atuveciclib monotherapy on the established cell lines Panc89, PancTu-1 and Colo357 were examined (Figure 1). All three cell lines showed little response towards monotherapy with more than half of the cells still viable at the maximum concentration of atuveciclib.
The treatment of pancreatic cancer cells with atuveciclib leads to a concentration- and time-dependent decrease in phosphorylation of the RNA polymerase II at Serine 2. Concentration-dependent molecular effects on pancreatic cancer cells after treatment with atuveciclib were analysed using Western blot (Figure 2A). CDK9-mediated phosphorylation of RNA polymerase II (RNA Pol II) at Serine 2 (pSer2) showed a concentration dependent decrease with increasing atuveciclib concentrations. This effect was detected even at the low concentration of 0.1 μM. Likewise, there was a slight concentration-dependent decrease in the expression of total RNA polymerase II. CDK9 expression was not altered by atuveciclib treatment. The same applies to the CDK9 co-factor cyclin T. Atuveciclib thus led to an impressive concentration-dependent decrease in the phosphorylation RNA Pol II at Ser2 and slightly decreased protein expression levels whereas the expression of CDK9 and cyclin T remained unchanged. Time-dependent effects of the treatment of pancreatic cancer cells with atuveciclib were analysed using Western blot (Figure 2B). Similarly to the previous experiment, the Figure 2 shows a time-dependent decrease in pSer2-RNA Pol II expression upon treatment with atuveciclib. In this experiment, the maximum decrease in pSer2-RNA Pol II expression was seen after 12 h. The maximum inhibition of CDK9 was thus achieved after 12 h. As in the concentration-dependent experiment, the expression of total RNA polymerase II decreased in a time-dependent manner too. Treatment with atuveciclib did not have time-dependent effects on CDK9 and cyclin T. Atuveciclib thus led to a time-dependent decrease in pSer2 RNA Pol II at the protein level.
Atuveciclib sensitises established pancreatic cancer cell lines to TRAIL-induced cell death. Next, we analysed the extent to which a combination of atuveciclib and izTRAIL sensitises pancreatic cancer cell lines to TRAIL-induced cell death using the MTT assay (Figure 3). Figure 3 clearly shows a concentration dependent, atuveciclib-mediated sensitisation of the cell lines to izTRAIL. The results showed the greatest sensitisation in Panc89 and Colo357 cells. Interestingly, higher concentrations of atuveciclib achieved marked sensitisation of cells to TRAIL-induced loss of viability, especially when combined with low izTRAIL concentrations.
Combined treatment with atuveciclib and izTRAIL leads to a substantial reduction in colony formation by established PDAC cell lines. The colony formation potential of PDAC cell lines PancTu-1 and Colo357 was assessed using a colony formation assay (Figure 4). Colony formation potential were evaluated based on maximum colony size and number of colonies. The colony formation assay showed results even more impressive than the 2D-MTT assay. Combination of atuveciclib and TRAIL led to highly efficient suppression of colony formation up to the complete loss of Colo357 colonies. Even more, in PancTu-1 cells, 2D growth did not show a strong response towards combination treatment, but the more representative 3D growth was significantly suppressed. A potential adverse effect of the treatment of PancTu-1 cells with izTRAIL alone, which was reflected by increased colonies, was reversed.
Combined treatment with atuveciclib and izTRAIL induces apoptosis and causes cell-cycle arrest in established pancreatic cancer cell lines. We then assessed the effects of combined treatment with atuveciclib and izTRAIL by analysing cell-cycle phases with a focus on mechanistic aspects. For this purpose, propidium iodide staining and flow cytometry were performed.
In the absence of atuveciclib, there was a continuous increase in cell populations in the Sub-G1 phase, which is indicative of apoptosis. Increasing izTRAIL concentrations led to a decrease in the number of cells in the G1 phase and at the same time to an increase in G2 cells, suggesting a G2 cell-cycle arrest (Figure 5A). Like indicated previously, 0.1 μM atuveciclib were insufficient to produce a sensitising effect (Figure 5A). The combination of higher concentrations of atuveciclib and izTRAIL led to a substantial increase in apoptotic cells. Atuveciclib thus sensitised PDAC cell lines to TRAIL-induced apoptosis. Relative proportions of cells in the sub G1 group are shown in Figure 5B. The dose-dependent increase of the sub G1 proportion after atuveciclib and TRAIL is highly significant (Figure 5B). The combination of higher concentrations of atuveciclib and izTRAIL was associated with a relative increase in the G1 phase versus the G2 phase, suggesting a G1 arrest. Combined treatment with atuveciclib and izTRAIL thus had a dual mode of action: it induced apoptosis and caused cell-cycle arrest.
The pan-caspase inhibitor zVAD-fmk can reverse cell death induced by atuveciclib and TRAIL. For further assessment of apoptosis mechanisms, the pan-caspase inhibitor zVAD-FMK was added to Pan89 cells upon combined treatment with atuveciclib and izTRAIL to prevent apoptosis (Figure 6). zVAD caused a highly significant increase in the viability of cells that were treated with a combination of atuveciclib and izTRAIL. These results support apoptosis as a mechanism of cell death. zVAD, however, did not completely reverse cell death.
Down-regulation of cFlip and Mcl-1 is responsible for atuveciclib-mediated sensitisation to TRAIL. Data so far show that atuveciclib considerably sensitises PDAC cells to TRAIL-induced cell death. The precise mechanisms can be revealed in the western blot analysis of the TRAIL-induced apoptotic cascade (Figure 7). In general, pre-incubation with atuveciclib over a period of 12 h led to a substantial increase in apoptosis, as indicated by a major increase in cleaved PARP. The cleavage of PARP is mediated by active effector caspase-3 in the apoptotic cascade. Treatment with atuveciclib led to an almost complete loss of cFlip expression at the protein level, which was accompanied by synchronous cleavage and thus activation of procaspase-8. Atuveciclib treatment was associated with increased levels of the active p18 subunit of caspase-8. In addition, atuveciclib suppressed Mcl-1, which led to a higher activation of caspase-9. Accordingly, atuveciclib sensitises PDAC cells to TRAIL-induced cell death by suppressing cFlip and Mcl-1.
The combination of atuveciclib and TRAIL is an effective second-line treatment for gemcitabine-resistant PDAC cells. We also assessed whether the combination of atuveciclib and TRAIL is an effective treatment for tumour cells that have become resistant to first-line chemotherapy (Figure 8). For this purpose, a gemcitabine-resistant clone of the Panc89 cell line (Panc89-GR3), that was established by our research group, was treated with atuveciclib and izTRAIL. Compared with wild-type Panc89 cells (Figure 1), gemcitabine-resistant Pan89-GR3 cells were also sensitive to combined treatment with atuveciclib and izTRAIL. A higher concentration of izTRAIL, however, was required to produce a reduction in viability similar to that observed for wild-type cells. Nevertheless, the combination of izTRAIL and atuveciclib appears to be an effective treatment option for chemotherapy resistant PDAC cells.
Atuveciclib sensitises two of three cell lines established from a PDX model to TRAIL-induced cell death. Finally, we analysed whether the promising effects of combined treatment with atuveciclib and izTRAIL can also be detected in a translational model (Figure 9). Atuveciclib sensitised cell lines 609 and 1157 especially at low concentrations of izTRAIL. Cell line 722 appeared to be resistant to this combination treatment. We were nevertheless able to demonstrate that combined treatment with atuveciclib and izTRAIL is also effective in primary patient derived cell lines.
Discussion
It has been shown that CDK9 is over-expressed in pancreatic ductal adenocarcinoma and that a high level of expression is associated with poorer outcome (36). Inhibition of CDK9 induces apoptosis by shifting the balance of anti-apoptotic proteins and pro-apoptotic proteins towards apoptosis (37). This shift in apoptosis-associated proteins has been reported to overcome resistance to TRAIL-induced apoptosis in TRAIL-resistant cell lines by concomitant suppression of cFlip and Mcl-1 (14). This suggests that combined treatment with TRAIL and CDK9 inhibitors is a particularly promising strategy for the treatment of highly resistant carcinomas such as pancreatic carcinomas. In the present study, we were able to show that the novel and orally bioavailable CDK9 inhibitor atuveciclib is such a promising CDK9 inhibitor and that atuveciclib alone can suppress PDAC cell lines. The EC50 values in this study were similar to those reported for cell lines derived from haematological neoplasms, which, however, have been reported to be associated with generally greater maximum suppression by atuveciclib (32). Western blot analyses demonstrated both time-dependent and concentration-dependent decrease in pSer2-RNA polymerase II because of atuveciclib-mediated CDK9 inhibition. As a side effect of treatment with atuveciclib, total RNA polymerase II decreased as well. This effect, which has already been reported for the selective CDK9 inhibitor SNS-032 by other research groups, may be the result of the prolonged inhibition of transcriptional elongation in response to CDK9 inhibition (36, 38, 39).
Combined treatment with atuveciclib and izTRAIL is highly potent in reducing cell viability. We showed that higher concentrations of atuveciclib effectively sensitise PDAC cell lines to TRAIL-induced cell death, especially in the presence of low izTRAIL concentrations. Interestingly, PancTu-1 cells showed a much poorer response to this novel combination treatment in 2D cultures but was similarly effective in the three-dimensional colony formation assay.
Flow cytometry demonstrated that combined treatment with atuveciclib and TRAIL has a dual mode of action: it induces apoptosis, as signalled by an increase in the Sub-G1 population, and causes cell-cycle arrest. Selective CDK9 inhibition has already been reported to induce G1 phase cell-cycle arrest (36, 40). Sensitisation of PDAC cell lines to TRAIL-induced cell death is partially attributable to the induction of cell-cycle arrest. Other research groups have reported that cells with an arrested cell cycle exhibit increased sensitivity to TRAIL (41-43). ZVAD was not able to completely reverse the loss of viability induced by atuveciclib and izTRAIL. Other forms of cell death such as necroptosis, which too can be induced by TRAIL, may also play a role in this context (44, 45). Further research is required to address CDK9 inhibition in the treatment of solid tumours. A research group reported in an abstract, however, that TRAIL combined with the CDK9 inhibitor dinaciclib led to morphological changes in HPV+ cervical cancer cells, which were indicative of necroptosis (46). This finding warrants further research.
We then investigated the mechanisms involved in the atuveciclib-mediated sensitisation of PDAC cells to TRAIL-induced cell death at the protein level. We showed that the suppression of Mcl-1 and cFlip may be the underlying mechanism. According to Lemke et al. (14), this phenomenon plays an integral role in overcoming TRAIL resistance. The ability of cells to form colonies in soft agar suggests malignant transformation (47). TRAIL has already been shown to promote adverse effects such as metastasis formation and proliferation (48-50). In our study, this was supported by the results of colony formation assay by PancTu-1 cells. The combination of TRAIL with atuveciclib was found to be a promising strategy for overcoming the adverse effects of TRAIL and for producing opposite effects. Our study demonstrated in vitro that the combination of izTRAIL and atuveciclib is an effective treatment option when PDAC cells have become resistant to gemcitabine. This is another reason why combined treatment with atuveciclib and izTRAIL holds great promise since PDACs often and rapidly develop resistance to gemcitabine within a few weeks of initiation of treatment with this first-line medication. The prognosis of patients with gemcitabine-resistant pancreatic cancer is extremely poor (51, 52).
In the final part of our study, we evaluated combined treatment with atuveciclib and izTRAIL in a translational model using PDX-derived cell lines (53). We obtained excellent and promising results using the 609 and 1157 cell lines. Atuveciclib was only unable to sensitise the 722 cell line to TRAIL-induced cell death. This might be due to the following reasons: Bax and Bak deficiency severely impairs the apoptotic programme (54). In addition, Bax/Bak-deficient cells have been found to be completely resistant to sensitisation to TRAIL-induced cell death mediated by CDK9 inhibition since MOMP cannot be executed effectively (14). Another possible explanation is the over-expression of anti-apoptotic proteins of the Bcl-2 family. Bcl-2 over-expression has been shown to be associated with increased metastatic potential and to result in resistance to TRAIL-induced apoptosis in certain types of cells (55-57). These mechanisms of resistance will be the focus of a future study.
In conclusion, the study presented here provides a basis for further preclinical and clinical evaluation of combined treatment with atuveciclib and TRAIL. Our results show that atuveciclib sensitises PDAC cells to TRAIL-induced cell death through the induction of apoptosis and cell-cycle arrest. Further studies should focus on the molecular analysis of cells that are resistant to this combination treatment. Other studies should focus on transferring this new and innovative combination treatment to the in vivo domain.
Acknowledgements
J.-P. R. was supported by the University of Ulm via the Promotionsprogramm Experimentelle Medizin. J.L. received funding from Deutsche Forschungsgemeinschaft (390780490).
Footnotes
Authors’ Contributions
J.-P. R., J.L., D. H.-B. and B.T. designed the study. J.-P. R. performed the experiments. J.-P. R., A.-L. K. and J.L. prepared the figures. J.-P. R., M.K. and B.T. wrote the manuscript. All Authors contributed to the preparation of the manuscript and read and approved the manuscript.
Conflicts of Interest
The Authors declare no conflicts of interest in relation to this study.
- Received September 13, 2021.
- Revision received September 27, 2021.
- Accepted September 28, 2021.
- Copyright © 2021 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.