Background: Pancreatic ductal adenocarcinomas (PDACs) are highly resistant to treatment due to changes in various signalling pathways. CK1 isoforms play important regulatory roles in these pathways.
Aims: We analysed the expression levels of CK1 delta and epsilon (CK1δ/∊) in pancreatic tumour cells in order to validate the effects of CK1 inhibition by 3-[2,4,6-(trimethoxyphenyl)methylidenyl]-indolin-2-one (IC261) on their proliferation and sensitivity to anti-CD95 and gemcitabine.
Methods: CK1δ/∊ expression levels were investigated by using western blotting and immunohistochemistry. Cell death was analysed by FACS analysis. Gene expression was assessed by real-time PCR and western blotting. The putative anti-tumoral effects of IC261 were tested in vivo in a subcutaneous mouse xenotransplantation model for pancreatic cancer.
Results: We found that CK1δ/∊ are highly expressed in pancreatic tumour cell lines and in higher graded PDACs. Inhibition of CK1δ/∊ by IC261 reduced pancreatic tumour cell growth in vitro and in vivo. Moreover, IC261 decreased the expression levels of several anti-apoptotic proteins and sensitised cells to CD95-mediated apoptosis. However, IC261 did not enhance gemcitabine-mediated cell death either in vitro or in vivo.
Conclusions: Targeting CK1 isoforms by IC261 influences both pancreatic tumour cell growth and apoptosis sensitivity in vitro and the growth of induced tumours in vivo, thus providing a promising new strategy for the treatment of pancreatic tumours.
Statistics from Altmetric.com
Pancreatic ductal adenocarcinoma (PDAC) is one of the most aggressive malignant tumours and has an extremely poor prognosis.1 Within tumorigenesis multiple checkpoints in cell proliferation, differentiation and apoptosis are affected, finally resulting in uncontrolled growth of tumour cells.2 3 Recently, several mechanisms responsible for apoptosis resistance, such as an over-expression of anti-apoptotic proteins (TRAF2, FAP-1, XIAP, Bcl-xL and PKCμ/PKD1), activation of PKC and NFκB and downregulation of pro-apoptotic proteins (FADD and Bid) have been identified.4–8 Furthermore, in apoptosis-resistant pancreatic tumour cells stimulation of CD95 leads to strongly enhanced invasiveness.9 The high resistance to the induction of apoptotic pathways and the high metastatic potential of pancreatic tumours result in the generally known low efficiency of common adjuvant therapies. Therefore, research interests are focused on increasing the efficacy of agents like gemcitabine in single and combined approaches, on developing new strategies for an effective immunotherapy, and on generating potential agents against newly identified target molecules.10 11 Within the past few years more and more protein kinases and phosphatases have become targets for drug development and, recently, interest in specifically targeting members of the CK1 (formely, casein kinase 1) family has increased.
CK1 represents a family of highly conserved pleiotropic serine/threonine specific protein kinases existing of at least seven isoforms (α, β, γ1–3, δ and ∊) in mammals.12 13 CK1 phosphorylates various substrates playing key roles in diverse physiological processes, such as DNA repair, cell cycle progression, cytokinesis, differentiation and apoptosis (reviewed by Knippschild et al13 and Price14). Their important role within various signalling pathways is supported furthermore by reports linking CK1 isoforms, especially CK1α, δ and ∊ to key regulator proteins, among them p53 and β-catenin, which act as signal integration molecules in stress situations.13 15
Considering the importance of signals mediated by CK1δ/∊ to finally ensure genome stability, it is obvious that mutations and/or changes in the activity of CK1δ/∊ or mutations of CK1 specific phosphorylation sites of their substrates can contribute to the development of tumours.13 16 At present, data are accumulating regarding the mechanisms by which CK1 can act in an anti-apoptotic manner (reviewed by Knippschild et al13).
Sensitising resistant tumour cells to apoptosis by inhibiting CK1 has been effectively achieved by siRNA mediated knockdown of CK1α or by using the CK1 specific small molecule inhibitor N-(2-aminoethyl)-5-chloroisoquinoline-8-sulfonamide dihydrochloride (CKI-7).17 18
In addition to CKI-7 several CK1 specific competitive small molecule inhibitors have been developed,13 19 among them is IC261,20 which shows a high selectivity towards CK1δ/∊ in the micromolar range. IC261 can induce predominantly cell cycle arrest or apoptosis dependent on the cellular background.21 Since changes in the activity and/or expression of CK1 isoforms have been detected in various tumours (reviewed by Knippschild et al13) the use of IC261 as a therapeutic drug in cancer treatment, especially of highly resistant tumours like pancreatic tumours, is an attractive possibility. In the present study we therefore analysed the expression levels of CK1δ/∊ in various pancreatic tumour cell lines and in pancreatic normal and tumour tissues. Furthermore, we characterised the effects of IC261 on cell growth of established, well-characterised pancreatic tumour cell lines, its ability to sensitise apoptosis resistant cells towards CD95-mediated apoptosis in vitro and its effects on inhibition of tumour growth in a xenograft model.
ASPC-1,22 BxPc3,23 Capan-1,24 Colo357,25 MiaPaCa-2,26 Panc89,27 Panc1,28 PancTu-125 and Panc89-H2B-clone 1 cells were grown in a equal mixture of Dulbecco’s-modified Eagle’s medium (DMEM) and RPMI medium (both Invitrogen, Karlsruhe, Germany) supplemented with 10% fetal calf serum (FCS) and incubated at 37°C in a humidified 5% carbon dioxide atmosphere. The characteristics of these cell lines have been described previously.25 29
The following antibodies were used: anti CK1∊ (mAb, BD Biosciences Pharmingen, San Diego, California, USA; and the polyclonal rabbit antiserum 712 (see supplementary data); anti CK1δ (mAb 128A; ICOS Corporation, Bothell, Washington, USA; and the polyclonal goat antiserum ab10877, abcam, GB), anti γ-tubulin (mAb; Sigma, Taufkirchen, Germany), anti-CD95 (mAb, clone CH11; Coulter Immunotech, Krefeld, Germany), anti-PARP (mAb, Ab-2, Merck, Calbiochem, Darmstadt, Germany), anti-c-IAP1, anti c-IAP2, anti XIAP, anti Bid (mAbs; R&D Systems, Wiesbaden, Germany), anti-Bcl-xL (BD Biosciences Pharmingen, USA), anti-caspase 8 (StressGen Biotechnology, Ann Arbor, Michigan, USA), anti β-actin (Sigma, Germany), anti-Ki-67 (Merck, Germany and Epitomics, Burlingame, California, USA) and a polyclonal rabbit antiserum against α/β-tubulin (kindly provided by G. Rutter, HPI, Germany).
Five million PancTu-1 cells resuspended in 100 µl of a solution containing 50% Matrigel (BD Pharmingen) and 50% DMEM/RPMI-1640 (1:1) were injected into the dorsolateral site of 6-week-old C.B-17/IcrHsd-scid-bg mice (Harlan Winkelmann, Borchen, Germany). After 17 days, mice were randomised to the control group (n = 5), the IC261 treatment group (n = 5), the gemcitabine group (n = 5) and to the IC261/gemcitabine group (n = 5). Injection of dimethylsulfoxide (DMSO; control group), IC261 (20.5 mg/kg), gemcitabine (0.6 mg/kg) alone or in combination (20.5 mg/kg IC261/0.6 mg/kg gemcitabine) (treatment groups) was performed daily for 8 days. Mice were sacrificed by asphyxiation with CO2 the day after the last treatment. Tumours were measured before and during treatment. Finally, the tumours were excised, measured, weighed and fixed in formalin or shock frozen. Tumour volume was calculated according to the formula for a rotational ellipsoid (length × height × width × 0.5236).
Human pancreatic tissue
In summary, 27 patients suffering from PDAC, whose informed consent was obtained prior to surgery, were chosen. The specimens included 27 sporadic PDACs of different gradings (G) (G1, one patient; G1–G2, two patients; G2, 11 patients; G2–3, five patients; G3, eight patients).
Formalin-fixed paraffin-embedded sections of the pancreatic tissues were used for immunohistochemical studies.
Deparaffinised and re-hydrated tissue sections were microwaved with CitraPlus solution (Biogenex, San Ramon, California, USA). After blocking the endogenous peroxidase activity (peroxidase blocking agent; DAKO, Glostrup, Denmark) the sections were incubated either with a monoclonal anti Ki-67 antibody (Calbiochem, 1:200, or Epitomics, 1:1200), the polyclonal goat anti-CK1δ antibody (ab10877 abcam, 1:1600), or with the polyclonal rabbit anti-CK1∊ antibody 712 (1:1200) followed by incubation with anti-rabbit or anti-goat immunoglobulins conjugated to peroxidase-labelled dextran polymers (N-histofine®; Nichirei Corporation, Tokyo, Japan). The sections were treated either with the chromogen 3,3′-diaminobenzidine (DAB; DAKO) (ab10877, 712), or with 3-amino-9-ethylcarbazol (AEC; DAKO) (Ki-67), counterstained with haematoxylin, and covered permanently with a cover slip.
Gene expression analysis
Total RNA was isolated from untreated and treated mice using the RNAeasy Kit (Qiagen, Hilden, Germany). Total RNA (1.5 µg) was reverse transcribed into complementary DNA using the RT2 First Strand Kit (SuperArray Bioscience Corp., Frederick, Maryland, USA). Gene profiling was done as described by the manufacturer using the RT2 profiler PCR array human p53 Signalling Pathway (84 genes) and a customer designed RT2 profiler PCR array (10 genes). The reactions were carried out in an Applied Biosystems 7500 Fast-Real Time PCR System (Applied Biosystems, Foster City, California, USA). The results were read out with the 7500 Fast System SDS Software.
The significance of differences was evaluated by one-way analysis of variance (ANOVA) followed by the Student–Newman–Keuls test. A p value <0.05 was considered significant. Calculations were performed by commercial software (GraphPad Prism, version 3.0; GraphPad Software Inc., San Diego, CA, USA).
Details of the generation of stable GFP-histon2B expressing Panc89 GFP-H2B cells, treatment of cells and cell lysis, western blot analysis, expression and purification of glutathione-S-transferase (GST) fusion proteins, flow cytometry, antibody production, and the TUNEL assay for detection of apoptosis are given in the supplementary data.
CK1δ/∊ are strongly expressed in established pancreatic tumour cell lines
Members of the CK1 family exhibiting anti-apoptotic features are interesting targets for drug targeting. Thus, the expression levels of CK1δ/∊ were first analysed in pancreatic tumour cell lines. The data presented in fig 1 indicate that a slightly lower expression level of CK1δ was detected in ASPC-1 and BxPc3 cells and additionally for CK1∊ in Capan-1 cells, compared with the CK1δ/∊ expression levels in MiaPaCa-2, Panc1, Panc89 and PancTu-1 cells (fig 1). Although no great differences in the CK1δ/∊ protein levels were detectable, the CK1δ/∊ specific kinase activity in fractionated pancreatic cell extracts differed by up to 6-fold.30
IC261 inhibits growth of human pancreatic tumour cell lines
The small molecule inhibitor IC261 is classified as a class I inhibitor, which shows a high selectivity towards CK1δ/∊ at the micromolar range. The selectivity and affinity of IC261 for CK1 result from an induced fit mechanism through which the dissociation rate of IC261 is decreased. As a non-charged molecule at physiological pH it can diffuse through cell membranes.20 It is highly active within cells and has been shown to induce predominantly cell cycle arrest or apoptosis dependent on the cellular background.21 31 Therefore, the effects of IC261 on cell growth of various pancreatic tumour cell lines were determined. The results showed that IC261 (1.25 μmol/l) inhibited the proliferation of all analysed pancreatic tumour cell lines (fig 2A). FACS analyses of propidium (PI)-stained pancreatic tumour cells revealed that the effects of IC261 differed in a cell line- (fig 2B) and dose-dependent manner (supplement fig 1). As shown in fig 2B IC261 treatment predominantly resulted in a strong increase of G2/M cells after 24 h and within the next 24 h the amount of dead cells increased.
Immunofluorescence analyses revealed that IC261 also induces mitotic spindle failure in all analysed pancreatic tumour cell lines, shown in Panc89 cells as an example (supplement fig 2A). To study the morphological changes during IC261 treatment in living cells, IC261-treated Panc89-H2B-clone1 cells were observed for 24 h. Within the first 18 h of IC261 treatment cells accumulate in the prophase of the cell cycle indicated by their condensed chromatin. Then within the next 2 h some of the cells seem to undergo apoptosis indicated by cell shrinkage, membrane blebbing and the appearance of apoptotic bodies (supplement fig 2B).
IC261 abolishes the CD95 resistance of pancreatic tumour cells
Recently, we have shown that the majority of pancreatic tumour cells are resistant to CD95-mediated apoptosis despite the expression of CD95 on the cell surface. To study the effects of CK1 inhibition on CD95-mediated apoptosis the apoptosis-resistant Panc89 cells were chosen. The data presented in fig 3A demonstrate that consistent with our previous data,7 Panc89 cells were almost completely resistant to CD95 triggering. However, pre-treatment of these cells with IC261 strongly sensitised them to CD95-mediated cell death and more than 50% of the cells were killed by combined treatment with IC261 and CH11 (fig 3A). The observed cell death was defined as apoptotic since it could be almost completely abolished by the caspase inhibitor zVAD (fig 3A). Furthermore, IC261 strongly increased the CD95-mediated processing of caspase 8 as well as activation of caspase 3 as demonstrated by enhanced cleavage of the caspase 3 substrate PARP (fig 3C). In addition, analysis of the cleavage pattern of Bid revealed that IC261 in combination with CH11 led to efficient Bid cleavage, whereas no Bid cleavage could be observed in lysates from cells either treated with CH11 or IC261 alone. Similar results were obtained when using the CK1 specific inhibitor CKI-7 (200 μmol/l, 22 h) underlining the role of CK1 in establishing the resistant phenotype of Panc89 cells towards CD95-mediated apoptosis (data not shown).
To further study the mechanism of IC261 sensitising effects of CD95-mediated apoptosis we analysed the expression levels of anti-apoptotic and pro-apoptotic proteins in untreated and IC261 treated Panc89 cells. We found that Bcl-xL, a member of the Bcl-2 family which mediates its anti-apoptotic effects through preventing cytochrome C release from mitochondria, and the IAP family members c-IAP1, c-IAP2 and XIAP, acting mainly through inhibition of caspases, were downregulated after IC261 treatment (fig 3C). Similar changes in the expression levels of c-IAP2, XIAP and Bcl-xL were obtained using CKI-7 (200 μmol/l, 22 h), which inhibits the activity of all CK1 isoforms (data not shown).
In contrast to CD95-mediated apoptosis gemcitabine-mediated cell death could not be increased by IC261 in gemcitabine-resistant PancTu-1 cells (supplement fig 3).
Immunohistochemical staining of CK1δ in PDACs
So far, our results revealed that inhibition of CK1δ/∊ by small molecule inhibitors can affect the growth of established pancreatic tumour cell lines and sensitise resistant Panc89 cells towards CD95-mediated apoptosis. These results point to the possibility to use CK1 specific inhibitors in new treatment concepts for PDAC. This would be especially important if alterations in the CK1δ/∊ expression levels could be observed in tissue specimens of PDAC. To detect possible changes in the CK1δ/∊ immunoreactivity, paraffin-embedded sections of 27 patients with PDAC were analysed. Histologically well-differentiated tumours (G1 tumours) consisting of tubular and glandular structures formed by mucus-secreting columnar cells, exhibited a low proliferation rate as indicated by the staining with the proliferation marker Ki-67 (fig 4E and H). Whereas in mucus secreting columnar cells CK1δ (fig 4D) and ∊ (fig 4G) staining was not detectable, a cytoplasmic staining of CK1δ (fig 4C) and ∊ (fig 4F) was observed in tumour cells in the glandular structures. In addition, fibroblasts in the stromal area around the mucus-secreting columnar cells showed a strong CK1δ and ∊ immunoreactivity. Normal pancreatic ducts and pancreatic acini, all of which exhibit moderate cytoplasmic expression of CK1δ/∊, served as internal controls in each of the serial sections (fig 4A and B).
As in G1 tumours, a cytoplasmic staining of CK1δ (fig 4I) and ∊ (fig 4J) was observed in G2 tumours. In contrast, a heterogeneous cytoplasmic staining intensity for CK1δ (fig 4L) and ∊ (fig 4M) ranging from weak to strong was observed in poorly differentiated G3 tumours. Detection of Ki-67 indicated high proliferation rates of G2 (fig 4K) and G3 (fig 4N) tumours.
IC261 inhibits tumour growth in a xenotransplantation model
The above results strongly suggest that inhibition of CK1 by small molecule inhibitors affects the growth of pancreatic tumour cell lines. To compare the effects of IC261 on pancreatic tumour cell growth in vivo with those of gemcitabine, which is commonly used in pancreatic tumour treatment, PancTu-1 cells were immobilised in matrigel and subcutaneously injected into the hind flank of 6-week-old SCID mice to obtain xenografts. Seventeen days after the tumours had been implanted, the mice were randomised into four treatment groups (n = 5): the control group (MOCK treated), the IC261 group, the gemcitabine group, and the IC261/gemcitabine treated group. The mice were treated daily for 8 days. Clinical observations of the mice during the treatment period and by gross pathological observation of all the vital organs at the termination of the study did not reveal any significant effects either on their weights or on their clinico-pathological disposition. Furthermore, the results of these in vivo experiments indicated that IC261 and gemcitabine led to a significant reduction of tumour growth. A combined treatment did not further reduce tumour growth (fig 5). Reduced tumour rates were matched to decreased tumour cell proliferation, as demonstrated by immunohistological staining for Ki-67. In the treated groups the positive Ki-67 staining was reduced (IC261 group, 11.3%; gemcitabine group, 14.7%; IC261/gemcitabine group, 10.5%) compared to that in the control group (21.0%) (fig 5B). In addition, an increased apoptosis rate was observed in the IC261-treated group and the other treated groups compared to the untreated control group. A representative example is shown in fig 5C.
Real-time PCR analyses revealed changes in the expression levels of various genes in tumours of IC261- and gemcitabine-treated animals compared to their expression levels in tumours of MOCK-treated animals (fig 5D). Downregulation of several anti-apoptotic proteins, CK1δ/∊, KRAS, and IL6 and upregulation of p21, ATM, CHEK1 and STAT1 were observed in IC261-and gemcitabine-treated animals. Interestingly, FASLG was downregulated in gemcitabine-treated animals, but upregulated in IC261-treated animals.
Constitutive proliferation and profound resistance to apoptosis are characteristic features of pancreatic tumour cells. Thus, much effort has been made to find the major players/proteins controlling both pathways. Recent evidence has revealed that several proteins involved in the regulation of cell growth and cell death are frequently affected in pancreatic tumour cells, among them PKCµ/PKD1,7 p5332 33 and Ras.34–36 So far, only a few specific pharmacological substances exist which can efficiently inhibit the growth and improve apoptosis sensitivity of these cells in parallel. Therefore, it is essential to test drugs against alternative targets for their use in pancreatic tumour treatment. Members of the CK1 kinase family are one of these new targets. In the present work we demonstrate that CK1δ/∊ are highly expressed in pancreatic tumour cell lines. Inhibition of CK1δ/∊ by IC26120 strongly reduced the growth of all the cells analysed. Our FACS analyses clearly revealed that most of the cells arrest in G2/M 24 h after IC261 treatment and that the amount of dead cells increased within the following 24 h. However, cell line and dose-dependent effects were detected, underlining the role of cellular factors in modulating the effects of IC261 on cell cycle progression and cell death.21 31 Furthermore, we showed here that inhibition of CK1δ/∊ by IC261 restores the sensitivity of Panc89 cells to CD95-induced apoptosis. PDAC cells are so-called type II cells regarding death receptor mediated apoptosis.4 In such cells efficient induction of apoptosis needs a mitochondrial amplification loop. One of the major players in this pathway is Bid, which has to be cleaved by caspase 8 to activate mitochondria.37 Recently, it has been shown that CK1 and CK2 can phosphorylate Bid and that the phosphorylation sites are in the direct neighbourhood of the caspase 8 cleavage site.17 Thus, the observed IC261- or CK1-7-mediated enhancement of Bid cleavage and, consequently, CD95-mediated apoptosis may be due to underphosphorylation of CK1-specific phosphorylation sites. The ability of CK1 inhibitors to sensitise Panc89 cells to CD95-mediated apoptosis might further be related to their ability to downregulate anti-apoptotic proteins, among them Bcl-xL and members of the IAP family (c-IAP1, c-IAP2 and XIAP). It has been shown that apoptosis-resistant pancreatic tumour cells show multiple and synergistic changes in the expression of anti-apoptotic proteins.4 7 Among them IAPs and Bcl-xL are most frequently over-expressed and contribute substantially to the pronounced resistant phenotype. Furthermore, CD95 stimulation results in apoptosis resistant cells and in invasiveness instead of cell death.9 This may have important consequences in vivo since in serum probes from patients with a pancreatic tumour an elevated CD95L level has been detected.38 39 Moreover, the cells of the immune system producing CD95L could induce the increase in malignancy of tumour cells. Therefore, sensitisation of pancreatic tumour cells against CD95-mediated apoptosis by CK1 inhibitors, especially IC261 along with its strong anti-proliferative action and its ability to downregulate anti-apoptotic proteins might provide an additional therapeutic advantage.
Our immunohistochemical analyses of 27 PDACs show that the expression of CK1δ and ∊ are slightly increased in higher graded tumours. These findings suggest that over-expression of CK1 contributes to aggressive tumour growth. These results are in line with Affimetrix GeneChip experiments showing that CK1 isoforms are upregulated in PDAC.40 41
Our data indicate that treatment of PDAC cells with IC261 also reduced their growth in a murine xenotransplantation model. The extent of tumour growth reduction was similar to that observed upon gemcitabine treatment. This was reflected by both a decreased proliferation rate as well as an increased apoptosis rate in treated tumours. However, combined treatment with IC261 and gemcitabine did not further reduce the tumour growth. Thus, there is a marked difference between the effects of IC261 on death receptor- and chemotherapy-mediated apoptosis. Since chemotherapy affects only fast growing cells, it is very likely that IC261-mediated cell cycle arrest is responsible for a failure of synergistic or even additive effects of IC261 and gemcitabine.
Analysis of gene expression in tumours of untreated IC261- and gemcitabine-treated animals revealed downregulation of CK1δ/∊, KRAS and anti-apopotic proteins, among them BCL2/1, BCL2A1, c-IAP1 and c-IAP2, and upregulation of ATM, CHK2, Stat1 and p21. Transcriptional active Stat1 is required for the induction of apoptosis in some cell types and can enhance drug-induced apoptosis by inducing the expression of caspases, FAS, FASL, TRAIL and p21.42 43 ATM kinase activation contributes to the activation of a number of regulatory proteins, leading to the activation of cell cycle checkpoints, DNA repair or to the induction of apoptotic processes.44
Our results show the potential of IC261 in inhibiting tumour growth in vivo. Being aware that multiple factors influence the effects of IC261 main issues for further characterisation are (1) to decrease its hydrophobic character in order to increase its inhibitory effects on tumour growth and (2) to analyse the usefulness of IC261 for different combined therapies. The use of CH11 in combination with IC261 is restricted by the toxicity of CH11 and combined IC261/gemcitabine did not enhance apoptosis in vitro nor reduce the tumour growth rate in vivo. However, inhibition of CK1γ by siRNA in combination with a low dose of the Akt kinase specific small molecule inhibitor A443654 significantly increased the death of MiaPaCa-2 cells.45 Therefore, CK1 inhibition in combination with other kinase inhibitors might provide a powerful tool in treating multiple drug resistant pancreatic tumours.
We would like to thank Tony De Maggio, Gabriel Rutter, David Meek, Jochen Heukeshoven and Michael Marzinzik for providing reagents and Arnhild Grothey, Annette Blatz, Nadine Süßner, Beate Rimmel and Bernhard Schmidt for technical assistance.
Funding: This work was supported by grants from the Deutsche Forschungsgemeinschaft and Deutsche Krebshilfe to Uwe Knippschild (SFB 518/B15, 10-2237 Kn), and to Holger Kalthoff (SFB 415/A3).
Competing interests: None.
Ethics approval: Animal studies were performed in accordance with the guidelines of the authority of Animal Use and Care Committee.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.