PKC inhibitor Go6976 induces mitosis and enhances doxorubicin-paclitaxel cytotoxicity in urinary bladder carcinoma cells

Cancer Letters 253 (2007) 97–107 www.elsevier.com/locate/canlet

PKC inhibitor Go6976 induces mitosis and enhances doxorubicin-paclitaxel cytotoxicity in urinary bladder carcinoma cells Vesa Aaltonen a, Jussi Koivunen a, Matti Laato

b,c

, Juha Peltonen

a,d,e,f,*

a

e

Department of Anatomy and Cell Biology, University of Oulu, Oulu, Finland b Department of Surgery, University of Turku, Turku, Finland c Department of Medical Biochemistry, University of Turku, Turku, Finland d Department of Dermatology, University of Oulu, Oulu, Finland Department of Anatomy, University of Turku, Kiinamyllynkatu 10, 20520 Turku, Finland f Department of Dermatology, University of Turku, Turku, Finland Received 9 November 2006; accepted 15 January 2007

Abstract Protein kinase C (PKC) a/bI isoenzyme inhibitor Go6976 has been suggested to be a G2 checkpoint abrogator by direct Chk1 inhibition. In the present study, we demonstrate that Go6976 induces mitosis in doxorubicin treated G2-arrested 5637 urinary bladder transitional cell carcinoma cells and interestingly also in non-synchronized 5637 cells. Importantly, the results demonstrated that both doxorubicin treated and non-synchronized cancer cells are forced to mitosis by Go6976. However, part of the cells avoid the death in mitosis and continue in the cell cycle which may increase the probability of genomic instability. Cytotoxicity of Go6976 alone and in combination with chemotherapeutic agents was further studied. Go6976 treatment alone induced apoptotic cell death. Cytostatic doxorubicin pre-treatment induced G2 arrest and inhibited the cytotoxic effects of mitosis specific drug paclitaxel. Cytotoxicities of doxorubicin-paclitaxel and doxorubicinGo6976 sequences could be markedly enhanced by combining Go6976 with paclitaxel after doxorubicin pre-treatment. In doxorubicin-Go6976+paclitaxel sequence, paclitaxel arrested the cells to mitosis and unfavourable progression of the cell cycle was inhibited. Analyzes of the molecular mechanisms underlying Go6976 induced mitosis showed that PKC inhibiting concentrations of Go6976 induced cdc2 activation concentration-dependently in non-synchronized and in DNA damaged cells. Simultaneously, Chk1/2 became deactivated and cdc25C activated in DNA damaged cells, indicating regulatory events upstream. In non-synchronized cells, activation of cdc25C, but not Chk1/2, was observed,

Abbreviations: ATM, ataxia-telangiectasia mutated; ATR, ataxia-telangiectasia mutated- and Rad3-related; BSA, bovine serum albumin; cdc, cell division cycle; DMEM, Dulbecco’s modified Eagle’s medium; DMSO, dimethyl sulfoxide; ECL, enhanced chemiluminescence; FACS, fluorescence activated cell sorting; FCS, fetal calf serum; HRP, horse radish peroxidase; MPF, mitosis promoting factor; PBS, phosphate-buffered saline; PKC, protein kinase C; PVDF, polyvinylidene difluoride; TCC, transitional cell carcinoma; Rb, retinoblastoma gene/protein. * Corresponding author. Address: Department of Anatomy, University of Turku, Kiinamyllynkatu 10, 20520 Turku, Finland. Tel.: +358 2 3337350; fax: +358 2 3337352. E-mail address: juhpel@utu.fi (J. Peltonen). 0304-3835/$ - see front matter  2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2007.01.011

98

V. Aaltonen et al. / Cancer Letters 253 (2007) 97–107

suggesting inactivation of c-TAK1. The results of the current study suggest that Go6976 has a synergistic cytotoxic effect when combined with doxorubicin and paclitaxel.  2007 Elsevier Ireland Ltd. All rights reserved. Keywords: Go6976; Protein kinase C; Cell cycle; Bladder; Carcinoma

1. Introduction Cancer cells often display non-functional cell cycle checkpoints, which is often caused by deregulation or mutations in tumor suppressor genes such as Rb and p53. Rb is a key controller of the G1-S transition and its inactivation promotes progression to S-phase [1]. In contrast, p53 functions as a key regulator of both G1 and G2 checkpoints and loss of p53 function disables cells to arrest cell cycle to G1 [2,3]. However, even these cells can arrest to Sor G2-phases of the cell cycle in response to DNA damage. Mutations of p53 have been associated with 50% of cancers, including bladder cancer [4,5]. G2 checkpoint is controlled by mitosis promoting factor (MPF), composed of activated cdc2/ Cyclin B1 complex, which promotes transition to M-phase of the cell cycle. Activity of cdc2 is in part adjusted by activating dephosphorylation on Tyr15 by active cdc25C. Phosphorylation and inactivation of cdc25C at Ser216 during interphase depends on active c-TAK1 and in G2 checkpoint on activated Chk1 and/or Chk2. Chk1 and Chk2 are, in turn, activated by DNA damage and subsequent ATM/ ATR-kinase activation [6]. In normal mitosis, MPF activity has to be sustained until metaphase, but its degradation is a prerequisite for entry into anaphase and Cyclin B1 must be destroyed before progression of mitosis from this phase [7–9]. Cytotoxic cancer chemotherapeutic agents can be roughly divided in DNA damaging agents (alkylating agents, anti-tumor antibiotics, anti-metabolites, platinum derivates, and topoisomerase inhibitors) and mitosis inhibitors (anti-microtubule agents; vinca-alkaloids and taxanes). In cancer chemotherapy, combination of various cytotoxic agents is used to increase the response rates compared to a single agent. The treatment with DNA-damaging agents generally induces cell death in the S-phase and cell cycle arrest to S- and G2-phases while mitosis inhibitors induce cell death in the M-phase and arrest to M-phase. An interesting cancer chemotherapeutic strategy has been suggested which involves combined use of DNA damaging agents and certain compounds which function as cell cycle checkpoint

abrogators. Checkpoint abrogators are substances that, for instance, inhibit Chk1- and/or Chk2-kinases in cells that contain DNA damage, and thus induce activation of cdc2, and entry to mitosis. This treatment eventually destroys the cells by mitotic catastrophe. However, when driving the DNA damaged cells to mitosis by checkpoint abrogators, there may be a risk that part of the cells avoid mitotic catastrophe and continue in the cell cycle. This may increase genomic instability and favour development of new harmful mutations [10]. Using mitosis inhibitors in the combination could prevent this disadvantage by their M-phase specific toxicity. In addition, it is worth noting that DNA damaging agents can inhibit the effect of mitosis inhibitors by induction of S- and G2-arrest and inhibit the cells from entering to M-phase. It is feasible to speculate that the use of G2 checkpoint abrogators could be effective in the combination of DNA damaging agents and mitosis inhibitors. Protein kinase C (PKC) family of serine–threonine kinases has been linked to cancer progression since most of the tumor promoters are PKC activators in two stage carcinogenesis models [11]. The PKC family members are classified into three major groups: classical (a, bI, bII, and c), novel (d, e, g, and h) and atypical (l, f, and i). PKCs are involved in various cellular processes such as regulation of gene expression, proliferation, cell junctions, apoptosis, and migration [12–14]. PKC inhibitors such as PKC412, Go6976, and UCN-01 have been shown have anti-cancer activity as single agents and in combination with classical chemotherapeutic agents [15]. As a single agent, Go6976 has been shown to inhibit invasion and induce apoptosis in vitro [16,17], and also be active in in vivo models [18,19]. UCN-01 and Go6976 have been shown to be potent abrogators of G2 checkpoint initially activated by DNA damage [20–23]. Go6976 has previously been suggested to directly inhibit Chk1 and possibly Chk2 since no change in the phosphorylation status of these kinases was observed in response to treatment with Go6976 in concentration of 30 nM, a concentration which however activated cdc25C in cells that were treated with topoisomerase I inhibitor SN38 [20]. In addi-

V. Aaltonen et al. / Cancer Letters 253 (2007) 97–107

tion, a previous paper showed in in vitro kinase assay that Go6976 is a potent Chk1 inhibitor at micromolar concentration, but data for ATM-, ATR-, or Chk2inhibitory function was not provided [24]. Our aim was to test if PKC a- and bI-isoenzyme inhibitor Go6976 functions as a checkpoint abrogator/activator of mitosis, in asynchronous, untreated cells, and in the presence of DNA damage using 5637 TCC cell culture as a model. 5637 TCC cells, being p53 and Rb null [25], provided an excellent tool to study classical chemotherapy and cell cycle checkpoint abrogators. We found out that DNA damage is not a prerequisite for Go6976 induced mitosis, which may result from c-TAK1 inhibition. Furthermore, in the presence of DNA damage, the activation of MPF results from inactivation upstream of Chk2, in concentrations that are readily capable of inactivating PKC a- and bI-isoenzymes. The results suggest that cells that are forced into mitosis with or without externally induced DNA damage can survive and continue in the cell cycle. Furthermore, using Go6976 as a checkpoint abrogator after DNA damage leads to increased cell death, which can be augmented by combining a mitosis inhibitor, paclitaxel into treatment. 2. Materials and methods 2.1. Antibodies Rabbit antibodies against human phospho-cdc2 (Tyr15), phospho-Chk1 (Ser345), phospho-Chk2 (Thr68), phospho-cdc25C (Ser216), cdc2, Chk1, Chk2, cleaved caspase-3 and mouse antibodies against Cyclin B1 (V152) were obtained from Cell Signaling Technology (Beverly, MA) and rabbit anti-b-actin from Sigma (St. Louis, MO). All antibodies were used at dilution 1/1000, except anti-cleaved caspase-3 and anti-Cyclin B1 antibodies which were used at dilutions 1/200 and 1/2000, respectively. 2.2. TCC cell culture Cell line 5637 (grade 2–3, p53 negative, Rb null) was obtained from American Type Culture Collection (ATCC, Rockville, MD) [25]. Cell line was maintained at DMEM supplemented with 10% FCS, penicillin (100 U/ml) and streptomycin (100 lg/ml) (DMEM + 10% FCS). 2.3. Chemicals PKC a- and bI-isoenzyme inhibitor Go6976 [26,27] was obtained from Calbiochem (La Jolla, CA) and dissolved in DMSO. All the control reactions were done with equal volumes of DMSO (1/1000 v/v) and ethanol

99

(1/10,000 v/v) as in drug treatments. Topoisomerase II inhibitor doxorubicin hydrochloride in 0.9% NaCl and microtubule active semi-synthetic paclitaxel dissolved in ethanol were obtained from Calbiochem. 2.4. Cell growth analyses In one time point growth analyses (doxorubicin-, paclitaxel- and Go6976-dose–responses) equal amounts of cells were seeded on opaque walled lClear 96-well plates (Greiner bio-one, Frickenhausen, Germany), allowed to attach and proliferate for 24 h. Subsequently, the cells were treated with doxorubicin, paclitaxel or Go6976 using indicated concentrations (4-wells/concentration) for 48 h. After incubation, the relative amount of living cells in each well was addressed using CellTiter Glo Luminescent Cell Viability Assay (Promega Corporation, Madison, WI) according to manufacturer’s guidelines. In growth analyses that required multiple time points (doxorubicin-Go6976-paclitaxel-sequence), trypan blue exclusion and cell counting was performed. Trypan blue exclusion was used additionally in Go6976-dose-response experiment to confirm proper function of CellTiter Glo assay. Equal amounts of cells were seeded on 3.5 cm petri dishes, allowed to attach and proliferate for 24 h. Subsequently, the cells were treated with different combinations and concentrations of chemicals for 0, 48, or 96 h. First, all samples (except 0 h time point) were treated with 100 ng/ml doxorubicin. At 48 h time point, the medium was changed to medium containing different concentrations and combinations of Go6976 and/or paclitaxel and treated for another 48 h. For cell counting, the cells were detached with Trypsin–EDTA. Cell suspension was treated with 1/5 (V/V) FCS and 0.2% trypan blue (final concentration), incubated in 37 C for 10 min and viable cells were counted using haemocytometer. All experiments were performed triplicate. 2.5. Cell cycle analysis The cells were cultured in DMEM + 10% FCS supplemented with different combinations of Go6976, doxorubicin and paclitaxel. Samples that were treated only with vehicle served as controls. In each time point, the cells were detached using trypsin–EDTA, treated with DMEM + 10% FCS, centrifuged 1000 rpm for 5 min, and washed with PBS. The cells were fixed with 85% ethanol under gentle agitation, washed with PBS, and resuspended in propidium iodide staining solution (40 lg/ml propidium iodide, 0.1% Triton X-100 and 200 lg/ml RNase A), and incubated at 37 C for 30 min. After incubation, the samples were analyzed using FACSCalibur (Becton Dickinson, Franklin Lakes, NJ). The results were analyzed and the graphs reproduced using Mod-Fit software (Venty Software House, Topsham, ME). All cell cycle analyses were performed triplicate.

100

V. Aaltonen et al. / Cancer Letters 253 (2007) 97–107

2.6. Analysis of caspase activation To analyze cleaved caspase-3 activation, the cells were treated with 100 or 1000 nM Go6976 or vehicle only, detached with Trypsin–EDTA, and fixed with 4% paraformaldehyde in PBS for 10 min at 37 C. The fixed cells were chilled on ice, resuspended in cold 90% methanol and incubated at +4 C for 30 min. The cells were then washed with a buffer containing 0.5% BSA in PBS (0.5% BSA/PBS), and diluted in the same buffer 1 · 106 cells/ml/tube. The cells were incubated with rabbit anti-cleaved caspase-3 antibody, or in control immunoreactions with normal rabbit IgG. The cells were subsequently washed twice with 0.5% BSA/PBS, incubated with goat anti-rabbit Alexa Fluor 488 conjugated secondary antibody (Molecular Probes, Eugene, OR) in 0.5% BSA/PBS for 30 min, washed twice with 0.5% BSA/PBS and analyzed with FACSCalibur. All experiments were performed triplicate. 2.7. Western transfer analysis The cells were treated with different combinations of Go6976, doxorubicin and paclitaxel. After treatment, the cells were rinsed with PBS, lysed in boiling buffer containing 1% SDS, 10 mM Tris pH 7.4 and 1 mM sodium orthovanadate, heated 5 min in 95 C and centrifuged for 16,000g for 5 min, to remove the debris. Protein concentration was measured using DC Protein Assay (Bio-Rad, Hercules, CA) and equal amounts of protein (10 lg) were subjected to SDS–PAGE on 10% gel. The proteins were then electrophoretically transferred to polyvinylidene difluoride (PVDF) membrane and processed for immunoblotting. Membranes were first blocked with 5% BSA/ PBS + 0.05% Tween 20 and immunolabelled using primary antibodies. Goat anti-rabbit or horse anti-mouse HRP conjugated antibodies (Cell Signaling Technology, Beverly, MA) were used as secondary antibodies and detected with enhanced chemiluminescence (ECL) (Amersham Life Sciences, Little Chalfont, England). Equal loading of each lane was evaluated by immunoblotting the same membranes with b-actin antibodies after detachment of previous primary antibodies. The films were photographed and densitometric analyses were performed using ImageJ 1.37 v (National Institutes of Health, USA). 3. Results 3.1. PKCa/bI inhibitor Go6976 induces apoptotic cell death of cultured TCC cells The cell growth was analyzed using trypan blue exclusion and cell counting, and with CellTiter Glo growth assay. Low concentrations (10–100 nM) of Go6976 had no marked effect on growth of 5637 cells. However, 100 nM concentration seemed to increase cell number slightly (by 6–8%) within 48 h, which was evident in

both growth assays (Fig. 1a). In contrast, 1000 nM Go6976 induced a major growth inhibition within 48 h (by 35–41%) (Fig. 1a). To test if cell death Go6976 resulted from apoptosis, the cells were cultured for 24 or 48 h under influence of 100 or 1000 nM Go6976 or vehicle alone, labelled with propidium iodide or with anti-cleaved caspase-3 antibodies and studied with flow cytometry. The results showed that treatment of the cells with 1000 nM Go6976 induced an increase in sub-G1 population (Fig. 1c) and increased cleavage (activation) of caspase-3 within 24 h (Fig. 1b) when compared to the cells treated with 100 nM Go6976 or vehicle alone. At 24 h time point, 2.3% and at 48 h 5.1% of the cells were positive for cleaved caspase-3, indicating increased apoptosis in response to 1000 nM Go6976 treatment (Fig. 1b). In contrast, cells treated with 100 nM Go6976 or vehicle alone, showed markedly smaller number of apoptotic cells (0.5–0.7%) (Fig. 1b). The cell cycle analysis showed an increase in sub-G1 population when the cells were treated with 1000 nM Go6976 for 24 h (2.9%) or 48 h (3.1%), which also suggests indirectly that apoptosis may be increased in response to the treatment. 3.2. PKCa/bI inhibitor Go6976 induces of mitosis in nonsynchronized TCC cells Flow cytometry analysis of propidium iodide stained 5637 cells were performed to test possible changes in the cell cycle distribution in response to Go6976 treatment (Fig. 1c). Treatment of the cells with 1000 nM Go6976 induced a prominent change in the cell cycle distribution within 24 h. During the first 24 h, the cells appeared to concentrate to G0/1 phase of the cell cycle. Specifically, the percentage of the cells in G0/1-phase increased dosedependently from 35.9% to 54.0% (Fig. 1c) The percentage of the cells in S- and G2/M-phases decreased in the same proportion as a result from treatment with 1000 nM Go6976 (Fig. 1c). Subsequently, during the next 24 h a G2/M-phase accumulation was seen in samples treated with 1000 nM Go6976 (Fig. 1c). These findings suggest that treatment with Go6976 affects the G2-M transition in non-synchronized cells. To study changes in MPF activation in response to Go6976 treatment, Western transfer analysis using antibodies against phosphorylated cdc2 (Tyr15) was performed (Fig. 1d). The results showed that after treatment with 100 or 1000 nM Go6976 for 12 h, cdc2 phosphorylation decreased concentration-dependently at Tyr15 (45% and 65% decrease, respectively), suggesting activation of MPF (Fig. 1d). This dephosphorylation continued until 48 h (24 h: 40% and 82% decrease; 48 h: 43% and 70% decrease, respectively). This suggests persisting activation of MPF. Changes in activity of controllers of cdc2 were studied using Western transfer analysis. The results showed a slight but detectable and concentration dependant

V. Aaltonen et al. / Cancer Letters 253 (2007) 97–107

101

Fig. 1. Characterization of the effects of PKC a/bI inhibitor Go6976 on cell growth, caspase activation, and cell cycle progression. (a) The percentage of viable 5637 cells after 48 h treatment with different concentrations of Go6976 was analyzed using trypan blue exclusion and cell counting (black bars) and CellTiter Glo assay (bars with diagonal lines); +SD are shown. (b) Flow cytometry analysis in which the 5637 cells are labelled with antibodies against cleaved caspase-3 after treatment with different concentrations of Go6976 for 24 or 48 h. The percentage of cleaved caspase positive cells +/ SD is shown. Rabbit IgG serves as negative control. Cell number is given in vertical axes and fluorescence in horizontal axes. (c) Cell cycle analysis of 5637 cells labelled with propidium iodide after treatment with different concentrations of Go6976 for 0, 24, or 48 h. The percentage of cells in G0/1-, S-, and G2/M-phases of the cell cycle and the percentage of sub-G1 population and +/ SD are shown above the histograms. Cell number is given in vertical axes and channels in horizontal axes. (d) Western transfer analysis of 5637 cells treated with different concentrations of Go6976 for 12, 24, or 48 h: the levels of phosphorylated cdc2 (Tyr15), cdc25C (Ser216), Chk1 (Ser345) and Chk2 (Thr68), and total cdc2, Chk1 and Chk2 were analyzed. b-Actin labelling shows equal loading.

102

V. Aaltonen et al. / Cancer Letters 253 (2007) 97–107

decrease in cdc25C phosphorylation in response to 100 nM and 1000 nM Go6976 treatments in time points 12–48 h (12 h: 8% and 13% decrease; 24 h: 16% and 31% decrease; 48 h: 13% and 32% decrease, respectively). This suggests activation of cdc25C. The results showed that the expression of phosphorylated Chk1 and Chk2 was low in all samples. However, increased phosphorylation (activation) in Chk1 was detected in 100–1000 nM Go6976 treated samples especially at 24 h time point (8% and 63% increase, respectively), but it was present also in 48 h samples (46% and 26% increase, respectively). The changes in Chk2 phosphorylation were minor, since the amounts of phosphorylated Chk2 were almost undetectable. However, minor phosphorylation in Chk2 was detected in samples that were treated with 1000 nM Go6976 for 12 and 24 h. 3.3. Cytotoxicity of doxorubicin to 5637 TCC cells 5637 TCC cells were incubated with 0–1000 ng/ml doxorubicin to analyze the cytotoxic profile of the agent. The results showed a concentration dependent cell death (Fig. 2a). Our results suggested that 100 ng/ml doxorubicin does not induce significant cell death, but instead arrests these cells to G2-phase of the cell cycle within 48 h (Fig. 2b and c). Furthermore, the results showed that after removal of the doxorubicin from culture medium, the G2 arrest persisted up to 48 h, with only moderate cell death (Fig. 2b and c).

decreased expression of Cyclin B1 by 47% and 54%, respectively (Fig. 2d). This indicates G2 arrest abrogation, and moreover that the cells are not arrested at metaphase of mitosis even though they may contain severe DNA damage. This is supported by findings in cell cycle analysis, which displayed a concentration dependant increase in the G0/1 population after treatment of the G2 arrested cells with 100–1000 nM Go6976. Chk1 and Chk2 regulate the phosphorylation of cdc2 through cdc25C and their activity was studied following doxorubicin and Go6976 induced G2 checkpoint abrogation. Western transfer analysis using anti-phospho-Chk1 (Ser345), anti-phospho-Chk2 (Thr68), and anti-phosphocdc25C (Ser216) antibodies showed that 48 h cytostatic doxorubicin treatment induced phosphorylation of Chk1 (2.2-fold increase), Chk2 (1.5-fold increase) and cdc25C (1.9-fold increase) (Fig. 2d). This indicates Chk1/2 activation, cdc25C inactivation and activation of G2 checkpoint. After removal of doxorubicin from culture medium, Chk1 and Chk2 remained active and cdc25C inactive. However, 100 nM–1000 nM Go6976 treatment induced a decrease in Chk2 phosphorylation (inactivation) concentration dependently (54% and 64% decrease, respectively). In contrast, decrease in Chk1 phosphorylation was less prominent (28% and 33% decrease, respectively). These results suggest that in the presence of externally induced DNA damage and Chk1/2 activation, Go6976 induced G2 checkpoint abrogation is in part mediated by inactivation of Chk2 and to lesser extent by inactivation of Chk1.

3.4. PKCa/bI inhibitor Go6976 abrogates doxorubicin induced G2 arrest, increases doxorubicin cytotoxicity and inactivates predominantly Chk2

3.5. Doxorubicin pre-treatment inhibits paclitaxel cytotoxicity

Since Go6976 proved to be a potent activator of mitosis in non-synchronized cells, we wanted to test if it could function similarly in cells arrested at G2-phase of the cell cycle. Flow cytometry analysis of propidium iodide stained 5637 cells showed that treatment with 100 or 1000 nM Go6976 induced a concentration dependent abrogation of the doxorubicin induced G2 arrest within 24 h (Fig. 2c). This was accompanied with moderate increase in the cytotoxicity suggesting additive effects of these agents (Fig. 2b). Western transfer analysis showed 1.4-fold increase in phosphorylation of cdc2 at Tyr15 (inactivation) in response to doxorubicin treatment, simultaneously with the G2 arrest seen in cell cycle analysis (Fig. 2c and d). When the cells were treated with 100–1000 nM Go6976, cdc2 Tyr15 became dephosphorylated (activated) in a concentration dependant manner (47% and 60% decrease, respectively), when compared to untreated control (Fig. 2d). This indicates onset of mitosis in cultures. Furthermore, cultures displayed an 2.9-fold increase in Cyclin B1 expression in response to doxorubicin treatment, suggesting G2 or M arrest (Fig. 2d). Interestingly, 100–1000 nM Go6976 treatment

Since small concentrations of doxorubicin induced G2 arrest, one could speculate that this would inhibit the cytotoxic effect of mitosis specific drug paclitaxel. When 5637 TCC cells were incubated with 0–1000 ng/ml paclitaxel as a single agent for 48 h, the cells showed a concentration-dependent cell death (Fig. 2a). To study simultaneous cytotoxic effects of doxorubicin and paclitaxel, 5637 TCC cells were incubated with cytostatic (100 ng/ml) concentration of doxorubicin for 48 h to induce G2 arrest. After the treatment, the cell culture medium was replaced with medium containing paclitaxel in concentrations of 5 or 100 ng/ml which were earlier shown to display major cytotoxicity (Fig. 2a). The results showed that pre-treatment with cytostatic concentrations of doxorubicin strongly inhibited the cytotoxic effect of paclitaxel, with only minor increase in the cytotoxicity when compared to the untreated cells (Fig. 2b). As a single agent, 100 ng/ml paclitaxel reduced the amount of living cells in the cultures by 78%, but when the same concentration was used after cytostatic doxorubicin treatment the amount of living cells decreased only by 14% (Fig. 2b).

V. Aaltonen et al. / Cancer Letters 253 (2007) 97–107

103

Fig. 2. Characterization of the effects of PKC a/bI inhibitor Go6976 on G2 checkpoint in doxorubicin-paclitaxel sequence. (a) The percentage of viable 5637 cells after 48 h treatment with different concentrations of doxorubicin or paclitaxel was analyzed with CellTiter Glo assay; +SD are shown. (b) The number of viable cells in three different time points (0, 48, and 96 h) was analyzed using trypan blue exclusion and cell counting. The cells were first treated with 100 ng/ml doxorubicin (Doxo) for 48 h, after which doxorubicin was removed and replaced with different concentrations of Go6976 (Go) or paclitaxel (Pacli), or with combinations of these two for 48 h (96 h time point). (c) Cell cycle analysis of 5637 cells labelled with propidium iodide at four time points. First, all samples were treated with 100 ng/ml doxorubicin for 48 h (Doxo 48 h). After removal of doxorubicin, the cells were treated with different concentrations of Go6976 (Go 100 nM, Go 1000 nM) or paclitaxel (Pacli 5 ng/ml, Pacli 100 ng/ml), or with combinations of these two for 24 h (24 h after Doxo) or 48 h (48 h after Doxo). (d) Western transfer analysis of untreated cells (0 h), cells treated with 100 ng/ml doxorubicin (Doxo 48 h), or cells that were first treated with 100 ng/ml doxorubicin for 48 h and subsequently treated with different concentrations of Go6976 (Go 100 nM, Go 1000 nM) or 5 ng/ml paclitaxel (Pacli), or with combinations of these for 24 h. The amount of Cyclin B1 and phosphorylated cdc2 (Tyr15), cdc25 C (Ser216), Chk1 (Ser345) and Chk2 (Thr68), and total cdc2, Chk1 and Chk2 were analyzed. b-Actin labelling shows equal loading.

104

V. Aaltonen et al. / Cancer Letters 253 (2007) 97–107

3.6. Go6976 enhances paclitaxel cytotoxicity after G2 arrest induced by doxorubicin Our results had indicated that doxorubicin pre-treatment counteracts paclitaxel cytotoxicity and that Go6976 effectively abrogates doxorubicin induced G2 arrest. Thus, we reasoned that Go6976 treatment could enhance paclitaxel cytotoxicity after doxorubicin pre-treatment. The 5637 TCC cells were pre-treated with 100 ng/ml doxorubicin and subsequently with 100 or 1000 nM Go6976 combined with 5 or 100 ng/ml paclitaxel for 48 h. The results showed a marked increase in cytotoxicity when compared to vehicle, 5 ng/ml paclitaxel, 100 ng/ml paclitaxel, 100 nM Go6976, or 1000 nM Go6976 treatments (Fig. 2b). Flow cytometry analysis of propidium iodide stained cells displayed that treatment with a combination of Go6976 and paclitaxel arrested the doxorubicin pre-treated cells to M-phase before cell death since G2/M-phase fraction increased when compared to treatment with Go6976 alone (Fig. 2c). In addition, there was no increase in G0/1 population when paclitaxel was combined with Go6976, which was however seen in cells treated with Go6976 alone. This indicates that paclitaxel treatment arrests the cells in mitosis very effectively. This was further supported by findings in Western transfer analyses using phospho-cdc2 (Tyr15) and Cyclin B1-antibodies (Fig. 2d). The results showed that treatment of 5637 TCC cells with cytostatic concentrations of doxorubicin induced cdc2 phosphorylation, apparently as a result from DNA damage to the cells. This phosphorylation of cdc2 persisted after removal of doxorubicin from culture medium, and replacement of doxorubicin with 5 ng/ ml paclitaxel did not affect the phosphorylation status. When doxorubicin was replaced with a combination of 100 or 1000 nM Go6976 with 5 ng/ml paclitaxel, a decrease in phosphorylation of cdc2 on Tyr15 was still seen (44% and 42% decrease, respectively). Furthermore, combining 100–1000 nM Go6976 to paclitaxel was still able to induce a decrease in Chk2 (67% and 68% decrease, respectively) and cdc25C (45% and 57% decrease, respectively) phosphorylation, while Chk1 dephosphorylation could not be detected. However, Western transfer analysis using Cyclin B1-antibodies showed that treatment of the cells with 100 or 1000 nM Go6976 in combination with 5 ng/ml paclitaxel did not reduce the amount of Cyclin B1 in the cultures, indicating metaphase arrest. This further supports the findings obtained from cell cycle analysis which suggested that combining paclitaxel to Go6976 treatment restrains the cells into mitosis, specifically to metaphase, and inhibits the cell cycle progression from that phase after G2 checkpoint abrogation.

4. Discussion The results of the present study demonstrate that the cytotoxicity of DNA damaging agent

doxorubicin can be increased by PKC inhibitor Go6976. Specifically, cytostatic concentrations of doxorubicin induced G2 checkpoint activation and subsequent Go6976 treatment resulted in checkpoint abrogation and cell death. Reaching cytotoxic concentrations of chemotherapeutic drugs in tumors is major problem in cancer therapy. Since, low concentrations of DNA damaging agents may lead to cell cycle arrest instead of cell death, checkpoint abrogators may provide an advantage by enhancing cell death. However, we showed that checkpoint abrogation could have adverse effects since part of the DNA damaged cells continue in the cell cycle after they are forced into mitosis from G2 arrest. This may increase genomic instability of the cell population. Our results suggest that introducing a mitosis specific drug paclitaxel to the doxorubicin-Go6976 sequence may provide a solution to the problem described above. Paclitaxel induced marked mitotic arrest and increased cell death. Previously, doxorubicin-UCN-01+paclitaxel sequence has been shown to result in augmented cytotoxicity [28]. Cancer therapies frequently include combinations of drugs, such as doxorubicin and paclitaxel which affect different phases of the cell cycle. Relapse of the disease is often caused by a subpopulation of drug resistant cancer cells. Drug resistance may arise from e.g., drug induced cell cycle arrest which protects the cancer cells from other phase specific drugs. Our results showed that when used alone, mitosis specific drug paclitaxel resulted in major cytotoxicity to the 5637 TCC cell line. Pre-treatment with cytostatic concentration of DNA damaging agent doxorubicin induced G2 arrest and almost completely inhibited the paclitaxel associated cytotoxicity. Adding a checkpoint abrogator, such as Go6976, to a combination therapy which involves DNA damaging agent and mitosis specific drug may inhibit drug resistance in the case of cell cycle arrest mediated drug resistance, as shown in the present study. On DNA damage, G1 arrest is p53 dependant. In contrast, G2 checkpoint activation is p53-independent as shown in Fig. 3. Thus, p53 null cells are capable of arresting to G2 [2,3,29]. One could assume that manipulating activated G2 checkpoint would affect not only cancer cells but also normal cells. Indeed, wild type p53 becomes active also on DNA damage induced G2 checkpoint, and wild type p53 has been shown to inhibit UCN-01 induced cell cycle progression from G2. However, if both

V. Aaltonen et al. / Cancer Letters 253 (2007) 97–107

Fig. 3. Proposed cell cycle-related changes in cancer cells following Go6976 treatment. (a) Black arrows and signs represent molecular interactions and phosphorylations during normal interphase of non-synchronized cells. Red indicates changes following Go6976 treatment. (1) Go6976 inhibits C-TAK1 function, resulting in activating dephosphorylation of cdc25C and subsequent activating dephosphorylation of cdc2. This leads to premature aberrant mitosis. (2) Cells forced to mitosis may stall replication, which in turn activates ATR. (3) Activated ATR phosphorylates Chk1 and to lesser extent, Chk2. Both are, however, directly inhibited by Go6976, as suggested by Kohn et al. [20]. Consequently, cdc25C and cdc2 remain dephosphorylated and active. (b) Doxorubicin induced G2 checkpoint activation is shown in black and its abrogation by Go6976 in red. (1) DNA damage activates ATR and ATM, which activate Chk1 and Chk2. Go6976 appears to inactivate ATM. (2) As a result form direct Chk1 and Chk2 inhibition [20] and suggested ATM inhibition by Go6976, cdc25C remains dephosphorylated and active despite of the DNA damage. (3) Active cdc25C phosphorylates and activates cdc2, and thus induces G2 checkpoint abrogation.

transactivation and repression functions of p53 are lost, p53 cannot inhibit UCN-01 mediated G2 checkpoint abrogation [29]. This leads to the follow-

105

ing conclusions: (1) p53 negative cells, such as 5637 cells, cannot arrest to G1 and are susceptible to G2 checkpoint manipulation, and (2) p53 wild type (normal) cells are capable of arresting to all cell cycle phases and are not as predisposed to G2 checkpoint manipulation as p53 negative cells. It is unlikely that Rb status of 5637 cells would affect the results of the present study, since p53 null status makes the cell line already G2 checkpoint dependant. The drug combination used in the current study may provide a tool to target cytotoxicity more selectively to the cancer cells. However, further studies are needed to analyze if combining these three substances increases or decreases toxicity to normal cells. Previous studies have suggested that low, 10– 30 nM, concentrations of Go6976 directly inhibit Chk1 and perhaps Chk2 in DNA damaged cells. It should be noted that these low concentrations of Go6976 are below those needed for PKC inhibition. Go6976’s inhibitory effect on Chk1 would explain its ability to abrogate G2 checkpoint [20]. In the present study, cell cultures were subjected to combinations of doxorubicin, paclitaxel and different concentrations of Go6976 and further analyzed using Western transfer analysis. More specifically, selected key players controlling entry to mitosis were analyzed in cells with and without DNA damage. The results prompted the following suggestions and speculations as outlined in Fig. 3. The results imply that molecular changes leading to G2 checkpoint abrogation in DNA damaged cells take place upstream of Chk1 and Chk2, when high, PKC inhibiting concentrations of Go6976 (>100 nM) are used [16]. This notion is based on the finding that these kinases undergo inactivating dephosphorylation in response to Go6976 treatment after DNA damage. Thus, ATR-Chk1 and/or ATM-Chk2 pathways may be inhibited by Go6976 either directly or indirectly. However, the results on non-synchronized cells representing cells at different phases of cell cycle revealed that DNA damage is not required for Go6976 induced activation of mitosis. Furthermore, Go6976 concentrations above 100 nM are needed for cdc2 activation and subsequent mitosis of non-synchronized cells. Interestingly, non-synchronized cells that were forced to mitosis using Go6976, showed moderate Chk1 activation, while changes in Chk2 phosphorylation were only minor. This result is seemingly conflicting since the results obtained from cells treated with doxorubicin-Go6976 sequence suggested that Go6976 inhibits ATR-Chk1 and/or ATM-Chk2

106

V. Aaltonen et al. / Cancer Letters 253 (2007) 97–107

pathways. It should be noted, however, that ATM and ATR kinases react to different types of DNA damage, ATM responds to DNA double-strand breaks, while ATR also reacts to stalled replication. [30,31]. Furthermore, there is limited crosstalk between ATM-Chk2 and ATR-Chk1 pathways [32]. We speculate that non-synchronized cells that are forced to premature mitosis contain incompletely replicated DNA. Stalled replication would then activate ATR and subsequently Chk1, while Chk2 activation would be only minor. In the case of doxorubicin induced DNA double strand breaks, ATM and ATR, and subsequently Chk2 and Chk1 will be activated. This data allows us to speculate that Go6976 inhibits ATM either directly or indirectly, since (i) Go6976 induced Chk2 targeted inhibition in DNA damaged cells, suggesting ATM inhibition by Go6976, see Fig. 3b, box 1; (ii) Go6976 treatment of non-synchronized cells preferably induced activation of Chk1. This suggests ATR activation as a result from premature mitosis and incompletely replicated DNA, see Fig. 3a, box 3; (iii) Chk1 activation was possible in non-synchronized cells, thus Go6976 is likely to inhibit pathway other than ATR-Chk1, such as ATM-Chk2. See Fig. 3a, box 3 and 3B, box 1. Our results suggest that Go6976 inhibits also cTAK1, the action of which is to phosphorylate and inactivate cdc25C during interphase (see Fig. 3a, box 1). This notion is based on the following findings: Exposure to Go6976 resulted in activation of cdc25C and cdc2 in non-synchronized cells as demonstrated by Western transfer analysis. Furthermore, the same cells showed some Chk1 activation but only minor Chk2 activity. There are three potential Go6976 targets whose inhibition results in activation of cdc25C and cdc2; namely Chk1 or Chk2 and c-TAK1. Previous data showed full inhibition of Chk1 and Chk2 with 10–30 nM concentrations but here we show a concentration dependant activation of cdc25C and cdc2 with Go6976 when concentrations of 100–1000 nM were used. Furthermore, non-synchronized cells were not affected by any cellular stress prior to mitosis induction by Go6976, ruling out primary Chk1/2 activation and its inhibition by Go6976. In addition, Go6976 shares significant homology with UCN-01 which has been previously reported to inhibit c-TAK1 at high concentrations [33]. Based on these considerations, c-TAK1 is a likely target of Go6976. In conclusion, the present study further characterized the possibilities of PKC a- and bI-inhibitor Go6976 as an anti-cancer agent using in vitro model.

We did show that Go6976 has anti-cancer function as a single agent with growth retardation and apoptosis of the cancer cells. Interestingly, Go6976 can increase cytotoxicity of both DNA damaging agent and mitosis specific drug to cancer cells by induction of G2 checkpoint abrogation. Furthermore, adding a mitosis specific drug to a DNA damaging agentG2 checkpoint abrogator combination arrested the cells in mitosis and protected cancer cells from entering to new cell cycle. The results by us and others would suggest that Go6976 is a superior G2 checkpoint abrogator since: (i) it affects G2 checkpoint control in three levels (ATM, Chk1/2, and C-TAK1), and (ii) it does not show significant binding to plasma proteins as does UCN-01 [20]. Acknowledgements This work was supported by Oulu University Hospital Grant H01139, Cancer Society of Northern Finland, Emil Aaltonen Foundation, and Maud Kuistila Memorial Foundation and Oulu University Scholarship Foundation. We thank Marja Paloniemi for expert technical assistance. Johanna and Emmi Aaltonen are thanked for their support. References [1] B.N. Chau, J.Y. Wang, Coordinated regulation of life and death by RB, Nat. Rev. Cancer 3 (2003) 130–138. [2] A. Eastman, Cell cycle checkpoints and their impact on anticancer therapeutic strategies, J. Cell. Biochem. 91 (2004) 223–231. [3] T. Kawabe, G2 checkpoint abrogators as anticancer drugs, Mol. Cancer Ther. 3 (2004) 513–519. [4] A. Blanes, J. Rubio, A. Martinez, H.J. Wolfe, S.J. DiazCano, Kinetic profiles by topographic compartments in muscle-invasive transitional cell carcinomas of the bladder: role of TP53 and NF1 genes, Am. J. Clin. Pathol. 118 (2002) 93–100. [5] M. Sanchez-Carbayo, N.D. Socci, E. Charytonowicz, M. Lu, M. Prystowsky, G. Childs, C. Cordon-Cardo, Molecular profiling of bladder cancer using cDNA microarrays: defining histogenesis and biological phenotypes, Cancer Res. 62 (2002) 6973–6980. [6] A. Sancar, L.A. Lindsey-Boltz, K. Unsal-Kacmaz, S. Linn, Molecular mechanisms of mammalian DNA repair and the DNA damage checkpoints, Annu. Rev. Biochem. 73 (2004) 39–85. [7] E.A. Nigg, Mitotic kinases as regulators of cell division and its checkpoints, Nat. Rev. Mol. Cell Biol. 2 (2001) 21–32. [8] V.A. Smits, R.H. Medema, Checking out the G(2)/M transition, Biochim. Biophys. Acta 1519 (2001) 1–12. [9] J.M. Peters, The anaphase-promoting complex: proteolysis in mitosis and beyond, Mol. Cell 9 (2002) 931–943.

V. Aaltonen et al. / Cancer Letters 253 (2007) 97–107 [10] M. Castedo, J.L. Perfettini, T. Roumier, K. Andreau, R. Medema, G. Kroemer, Cell death by mitotic catastrophe: a molecular definition, Oncogene 23 (2004) 2825–2837. [11] S.H. Yuspa, Cutaneous chemical carcinogenesis, J. Am. Acad. Dermatol. 15 (1986) 1031–1044. [12] Y. Nishizuka, Protein kinase C and lipid signaling for sustained cellular responses, FASEB J. 9 (1995) 484–496. [13] Y. Nishizuka, Intracellular signaling by hydrolysis of phospholipids and activation of protein kinase C, Science 258 (1992) 607–614. [14] V. Aaltonen, J. Koivunen, M. Laato, J. Peltonen, Heterogeneity of cellular proliferation within transitional cell carcinoma: correlation of protein kinase C {alpha}/{beta}I expression and activity, J. Histochem. Cytochem. (2006). [15] J. Koivunen, V. Aaltonen, J. Peltonen, Protein kinase C (PKC) family in cancer progression, Cancer Lett. (2005). [16] J. Koivunen, V. Aaltonen, S. Koskela, P. Lehenkari, M. Laato, J. Peltonen, Protein kinase C alpha/beta inhibitor Go6976 promotes formation of cell junctions and inhibits invasion of urinary bladder carcinoma cells, Cancer Res. 64 (2004) 5693–5701. [17] D.K. Biswas, A.P. Cruz, E. Gansberger, A.B. Pardee, Epidermal growth factor-induced nuclear factor kappa B activation: A major pathway of cell-cycle progression in estrogen-receptor negative breast cancer cells, Proc. Natl. Acad. Sci. USA 97 (2000) 8542–8547. [18] D.K. Biswas, S.C. Dai, A. Cruz, B. Weiser, E. Graner, A.B. Pardee, The nuclear factor kappa B (NF-kappa B): a potential therapeutic target for estrogen receptor negative breast cancers, Proc. Natl. Acad. Sci. USA 98 (2001) 10386–10391. [19] D.K. Biswas, K.J. Martin, C. McAlister, A.P. Cruz, E. Graner, S.C. Dai, A.B. Pardee, Apoptosis caused by chemotherapeutic inhibition of nuclear factor-kappaB activation, Cancer Res. 63 (2003) 290–295. [20] E.A. Kohn, C.J. Yoo, A. Eastman, The protein kinase C inhibitor Go6976 is a potent inhibitor of DNA damageinduced S and G2 cell cycle checkpoints, Cancer Res. 63 (2003) 31–35. [21] A.N. Tse, G.K. Schwartz, Potentiation of cytotoxicity of topoisomerase i poison by concurrent and sequential treatment with the checkpoint inhibitor UCN-01 involves disparate mechanisms resulting in either p53-independent clonogenic suppression or p53-dependent mitotic catastrophe, Cancer Res. 64 (2004) 6635–6644. [22] K. Sugiyama, M. Shimizu, T. Akiyama, T. Tamaoki, K. Yamaguchi, R. Takahashi, A. Eastman, S. Akinaga,

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30] [31] [32]

[33]

107

UCN-01 selectively enhances mitomycin C cytotoxicity in p53 defective cells which is mediated through S and/or G(2) checkpoint abrogation, Int. J. Cancer 85 (2000) 703– 709. R.T. Bunch, A. Eastman, Enhancement of cisplatin-induced cytotoxicity by 7-hydroxystaurosporine (UCN-01), a new G2-checkpoint inhibitor, Clin. Cancer Res. 2 (1996) 791– 797. S.P. Davies, H. Reddy, M. Caivano, P. Cohen, Specificity and mechanism of action of some commonly used protein kinase inhibitors, Biochem. J. 351 (2000) 95–105. K.M. Rieger, A.F. Little, J.M. Swart, W.V. Kastrinakis, J.M. Fitzgerald, D.T. Hess, J.A. Libertino, I.C. Summerhayes, Human bladder carcinoma cell lines as indicators of oncogenic change relevant to urothelial neoplastic progression, Br. J. Cancer 72 (1995) 683–690. G. Martiny-Baron, M.G. Kazanietz, H. Mischak, P.M. Blumberg, G. Kochs, H. Hug, D. Marme, C. Schachtele, Selective inhibition of protein kinase C isozymes by the indolocarbazole Go 6976, J. Biol. Chem. 268 (1993) 9194– 9197. M. Gschwendt, S. Dieterich, J. Rennecke, W. Kittstein, H.J. Mueller, F.J. Johannes, Inhibition of protein kinase C mu by various inhibitors. Differentiation from protein kinase c isoenzymes, FEBS Lett. 392 (1996) 77–80. M.V. Blagosklonny, Sequential activation and inactivation of G2 checkpoints for selective killing of p53-deficient cells by microtubule-active drugs, Oncogene 21 (2002) 6249–6254. A.A. Levesque, E.A. Kohn, E. Bresnick, A. Eastman, Distinct roles for p53 transactivation and repression in preventing UCN-01-mediated abrogation of DNA damageinduced arrest at S and G2 cell cycle checkpoints, Oncogene 24 (2005) 3786–3796. C.E. Canman, Replication checkpoint: preventing mitotic catastrophe, Curr. Biol. 11 (2001) R121–R124. R.T. Abraham, Cell cycle checkpoint signaling through the ATM and ATR kinases, Genes Dev. 15 (2001) 2177–2196. X.Q. Wang, J.L. Redpath, S.T. Fan, E.J. Stanbridge, ATR dependent activation of Chk2, J. Cell Physiol. 208 (2006) 613–619. E.A. Kohn, N.D. Ruth, M.K. Brown, M. Livingstone, A. Eastman, Abrogation of the S phase DNA damage checkpoint results in S phase progression or premature mitosis depending on the concentration of 7-hydroxystaurosporine and the kinetics of Cdc25C activation, J. Biol. Chem. 277 (2002) 26553–26564.