ERK mediated upregulation of death receptor 5 overcomes the lack of p53 functionality in the diaminothiazole DAT1 induced apoptosis in colon cancer models: efficiency of DAT1 in Ras-Raf mutated cells

Many of the anticancer compounds need p53 for their mechanism of action. However, from the light of numerous studies conducted recently, it can be concluded that correlation of drug sensitivity with p53 status depends very much on the drug and cell type. To test the activity of DAT1 in cell lines where the action of p53 is suppressed, we have used colon cancer cell lines HT29, SW620 and SW480 with mutated p53 [38] and HCT116 p53−/−, an isogenic p53 knock out (p53 −/−) line and compared with p53 wild type cell line HCT116 (p53 +/+). The half inhibitory concentration (IC₅₀) of DAT1 was determined in these cell lines by MTT assay and was compared with the widely used chemotherapeutic agents paclitaxel (taxol®), vinblastine and 5-fluorouracil. DAT1 was effective in all these cell lines in a comparable manner irrespective of their p53 status (Table 1). As reported earlier [39‐42], taxol maintained its efficacy in cell lines with nonfunctional p53, whereas vinblastine and 5-fluorouracil had decreased efficiency in most of these cell lines.

DAT1 was earlier shown to induce apoptosis in different cell lines [30]. Treatment of DAT1 to the cell lines with nonfunctional p53 followed by staining with the DNA binding dye DAPI detected condensation of chromosomes (Fig. 1a) which is a hallmark of apoptosis. Quantitation of apoptosis showed that DAT1 was quite effective in all these cell lines (Fig. 1b). The effector caspase 3 activation was also observed in these cell lines (Fig. 1c). However, another microtubule targeting anticancer drug vinblastine, failed to induce apoptosis in p53 knock out (p53 −/−) line at a concentration of 0.01 μM, a concentration in which it displayed large apoptotic response in the p53 wild type HCT116 cells (Additional file 1: Figure S1).

A number of earlier studies indicate that p53 is crucial in the mitochondria mediated apoptotic pathway. However, the extrinsic pathway can be both p53 dependent or independent. Previous studies from our laboratory showed the involvement of both apoptotic pathways in DAT1 mediated apoptosis with major contribution from an independent extrinsic pathway. Since DAT1 was also found to be highly efficient in cells where the intrinsic pathway was blocked [30], it seemed worthwhile to analyze whether the extrinsic pathway of apoptosis caused by DAT1 would be p53 mediated.

Western blotting of lysates prepared from cells treated with DAT1 for different time periods showed that DAT1 caused caspase 8 activation in cell lines irrespective of their p53 status (Fig. 1c).

Several downstream targets of p53 are involved in apoptosis. Bax, PIGs (p53 induced genes), Fas/APO-1, DR5 are some of the downstream targets of apoptosis that play major roles in apoptosis [34]. Previous studies from our laboratory have shown that DAT1 mediated apoptosis depends mainly on the DR5 activation [30]. DR5 is a member of the TNFR family and is homologous to the other death receptor DR4 [43]. Both DR4 and DR5 gets activated when bound by its substrate TRAIL and recruits Fas associated death domain (FADD) which in turn recruits caspase 8 and initiates extrinsic pathway of apoptosis. DAT1 treatment caused DR5 upregulation in cells with non functional p53 confirming that DAT1 mediated DR5 induction is p53 independent (Fig. 2a). There was no considerable change in the levels of DR4 and the other member of the TNF receptor family, Fas receptor, after DAT1 treatment (Fig. 2b).

Mitogen activated protein kinases are serine/threonine specific protein kinases and comprises of Extracellular signal Regulated stress Kinases (ERK), c-Jun N-terminal kinase/stress activated protein kinases (JNK) and p38 kinases [44, 45]. Among the many isoforms, ERK1/2 is the most extensively studied isoform and is activated by phosphorylation in response to various stress stimuli through diverse mechanisms involving the Ras-Raf-MEK pathway. A variety of biological responses (apoptosis, proliferation, migration and differentiation) are associated with ERK activation depending on the cell type, duration of activation and stimulus [46, 47]. Thus we were interested to know whether MAPK signaling has any role in the DAT1 mediated apoptosis. Both HCT116 and HCT116 p53−/−cells were treated with DAT1 for the indicated time periods, and P-ERK, P-p38, P-JNK levels were checked. DAT1 caused P-ERK activation in HCT116 and HCT116 p53−/− cells, while P-p38 and P-JNK levels remained same (Fig. 3a). Further, immunofluorescence and western blotting experiments also revealed increase in nuclear localization of P-ERK following DAT1 treatment (Fig. 3b, c) indicating an increase in their role for transcription of molecules involved in further cellular processes.

U0126 is a selective inhibitor of MEK1/2 and hence prevents ERK1/2 activation by its phosphorylation. HCT116 and HCT116 p53−/−cells were treated with U0126 followed by DAT1 treatment and the chromatin condensation was checked by DAPI staining. It was seen that the percentage of apoptotic cells decreased from 67 to 3 % and 60 to 5 % in HCT116 and HCT116 p53 −/− respectively in the inhibitor treated cells, as compared to cells treated with DAT1 alone (Fig. 4a). The reduction of apoptosis was also observed by spectrofluorometric measurement of active caspase 3 in cells treated with DAT1 either in the presence or absence of the ERK inhibitor U0126. The relative fluorescence intensity (which indicates caspase 3 activation) decreased in the presence of U0126 as compared to treatment with DAT1 alone (Fig. 4b). The findings were also supported using RNAi inhibition of ERK1/2. Prominent reduction of apoptosis as measured by chromatin condensation was observed using siRNA against ERK1/2 as compared to scrambled siRNA (Fig. 4c).

There are many studies which show that ERK signaling can be an upstream regulator of DR5 activation [48‐50]. To check whether there is any link between ERK activation and DR5 upregulation in DAT1 mediated apoptosis, we tested for DR5 induction following DAT1 treatment either in the presence or absence of U0126 in HCT116 and HCT116 p53−/− cells. DR5 induction by DAT1 was significantly hampered in presence of U0126 (Fig. 4d) proving that DAT1 induces ERK mediated DR5 activation. Reduced DR5 induction was also observed using siRNA against ERK1/2 as compared to scrambled siRNA (Fig. 4e).

Thus ERK activation was found to be a key signaling mechanism upstream DR5 mediated apoptosis by DAT1 in colon cancer cells with both wild type and nonfunctional p53.

Ras/Raf/MEK pathway plays an important role upstream of ERK kinase activation. Mutations in the several isoforms of the small GTPase Ras or Raf kinase superfamily of proteins were found to be associated with different types of human cancers [51, 52]. We thus wanted to see the effect of Ras/Raf mutation on the signaling and DAT1 activity in colon cancer cell lines. The IC₅₀ values of DAT1 and Ras/Raf mutational status in five different cell lines are shown in Table 2. The apoptosis induction, as measured by chromosome condensation, was also done in the cell lines Colo 320 DM and Caco2 (Additional file 2: Figure S2) with wild type Ras/Raf. It was observed that at approximately 1.5 times the IC₅₀ concentration, DAT1 was able to induce only 38 and 32 % apoptosis respectively in these cells (Fig. 1 for comparison with other cell lines). Further, as shown in Fig. 5, DAT1 was found to activate ERK in SW 480 and HCT116 cell lines where Ras mutations [53‐55] were noted but it was unable to activate ERK in Colo 320 DM or Caco2 cell lines with wild type Ras or Raf [54, 56, 57]. Consequently, DR5 was also found to be upregulated upon DAT1 treatment in SW480 and HCT116 cells but not in Colo320 DM or Caco2 cells showing that DAT1 was more effective to induce apoptosis by modulating Ras/Raf/MEK/ERK mediated death receptor pathway in cells with Ras/Raf mutation.

Tumours were grown in Severe Combined immunodeficiency/Non obese diabetic (SCID/NOD) mice by injecting HCT116 or HCT116 p53 −/− cells subcutaneously. As shown in Fig. 6, DAT1 treatment for 28 days reduced the mean tumour volume effectively in a concentration dependent way for both HCT116 and HCT116 p53−/− xenograft models.

To rule out the possibility that the efficacy of DAT1 in tumours is not due to toxicity of the compound in mice, we determined both the acute and sub-acute toxicity of DAT1 in Swiss Albino mice for the range of concentrations used for xenograft studies. In acute toxicity studies, where single doses were administered to different groups, there were no signs of weight loss, dehydration, mortality or behavioural change after 14 days upto 50 mg/kg of DAT1 (Additional file 3: Table S1).

In sub-acute toxicity studies, animals were administered with 4, 8 or 16 mg/kg of DAT1 on alternate days for the first 10 days and twice weekly thereafter for a period of 3 months and then compared with the control group of animals. At the end of 3 months, there was no significant change in the weight. Further, as shown in Table 3, hematological analysis detected no significant change in the RBC and WBC content of the treated mice from the control mice indicating that there was no hematotoxicity. The serum AST, ALP, BUN and creatinine levels of the treated mice also did not vary much from the control values indicating that the drug did not induce hepato and renal toxicity (Table 3). The serum ALT level remained similar to the control for the groups that were administered with 4 or 8 mg/kg DAT1. However, the ALT level increased for mice that were administered 16 mg/kg of DAT1. But the increase could be attributed to solvent toxicity as the vehicle control containing similar amount of cremaphor/ethanol showed similar toxicity for ALT.

We further checked whether the reduction of tumour sizes were due to apoptosis induction by DAT1. Apoptosis induction in the tumour tissues was analyzed by tunnel assay. As seen in (Fig. 7), there were much more staining in the treated tissues as compared to the control tissues indicating prominent apoptosis in DAT1 treated colon xenografts with both wild type and nonfunctional p53. This was also confirmed by immunohistochemistry, where the tissues were probed with active caspase 3 antibody. Here also, more cells showed positive staining for active caspase 3 in the treated tissues as compared to the control tissues (Fig. 7) confirming more apoptosis in the treated tissues.

Experiments in colon cancer cell lines have shown that DAT1 is causing DR5 and P-ERK activation. In order to confirm these observations in vivo, DR5, P-ERK, P-JNK/SAPK and P-p38 levels were checked in tissues from both p53 wild type and knock out mice by immunohistochemistry. Clearly, as indicated by the increase in blue staining, DR5 and P-ERK levels increased in the treated tissues as compared to the control tissues showing activation of the respective proteins by DAT1 (Fig. 8a -b). However, consistent with the in vitro results, the tissues labeled with P-p38 and P-JNK/SAPK antibodies displayed similar staining pattern in both treated and control tissues showing that these pathways were not activated by DAT1 in colon tumours.

DR5 and P-ERK activation was also confirmed by western blotting. Blots from three different mice are shown in Fig. 9. The levels of DR5 and P-ERK were found to be more in the treated tissues as compared to the control tissues, thus supporting our findings from cell line studies.

Cell lines HCT116 and SW480 were from ATCC. SW620 and HT29 were from NCI. HCT116 p53 −/− was a kind gift from Dr. Bert Vogelstein. Caco2 and Colo 320 DM cell lines were procured from the national repository of National Centre for Cell Sciences, Pune, India. DAT1 was synthesized by slight modification of the method described earlier [63]. Vinblastine, taxol, 5-Fluoro uracil, DAPI, propidium iodide and U0126 were from Sigma, USA. siRNA construct for ERK1/2 was obtained from Santacruz Biotechnology. DMEM and PSN antibiotic mixture were purchased from Invitrogen, USA. MTT and Caspase 3 fluorogenic substrate were from USB and BD pharmingen, respectively. Tunnel assay kit was from R & D Biosystems, USA and immunohistochemical kit was purchased from Vector Labs, USA. Kits for detection of ALP, AST, ALT, Creatinine and BUN levels in serum were from Aspen Laboratories, India.

DR5 antibody was from Imgenex, Canada or Abcam, UK, DR4 antibodies were procured from Imgenex, Canada. Antibody against Fas was from Sigma and antibodies against P-ERK, P-p38, P-JNK/SAPK, Caspase 3 and 8 were from Cell Signaling Technologies, USA or Abcam, UK. Lamin, β actin, ERK, p38 and JNK, SAPK antibodies were purchased from Santacruz Biotechnology, USA. Alexafluor 488, 568 and 633 antibodies were from Molecular Probes (Invitrogen). Anti rat tubulin marker antibody was from Abcam, UK.

The cell lines used in the study were maintained in DMEM or RPM1 containing 10 % FBS with 1X Penicillin Streptomycin Neomycin antibiotic mixture. The cells were incubated at 37 °C in a CO₂ incubator in humid condition containing 5 % CO_2. From the reference stock, frozen stocks of cells were made within passage 3 and stored in liquid nitrogen. For experiments, cells were used within 3 months after revival.

The cells were grown on coverslips in 24 well plates and incubated with respective drug or inhibitor for 24 h. They were then washed with PBS, fixed with chilled methanol-EDTA for 10 min and rehydrated with PBS for 10 min at room temperature. Successively, the cells were treated with DAPI (0.5 μg/ml) for 2 min in the dark. The coverslips were then mounted on fluoromount G and viewed at 40X in the UV region using a fluorescence microscope (Olympus IX71).

Cells were grown to 70–80 % confluency in 60 mm culture dishes. After drug treatment, they were harvested and lysed. Protein samples from tissues were collected by homogenizing the tissue in the presence of lysis buffer, followed by centrifugation for 20 min. The protein samples were run in a 10–12 % gel and transferred on to a PVDF membrane. The membranes were probed with specific primary antibodies (1:1000) overnight followed by secondary antibodies conjugated with horse radish peroxidase/alkaline phosphatase (1:2000) for 1 hr at room temperature and were developed by chemiluminescence or alkaline phosphatase method.

Cells were seeded in 96 well plates. At about70 % confluency, they were treated with different concentrations of drugs either individually or in combination. After 48 h, MTT assay was done (Mossmann 1983) and absorbance was measured at 570 nm (Biorad Microplate reader). Percentage viability was plotted against the respective concentrations of the drugs and half inhibitory concentration (IC₅₀) was calculated with the non linear regression programme of Origin.

Cells were seeded on coverslips in 35 mm dishes and were treated with DAT1 for the indicated time period. After fixing with methanol-EDTA and blocking, they were probed with the primary antibodies overnight followed by alexafluor 488. Nuclei were stained with propidium iodide and were imaged in a confocal microscope (Leica SP2) or in Olympus IX71.

The enzymatic activities of caspase 3 was assayed spectrofluorimetrically using fluorogenic substrates from BD Pharmingen according to manufacturer’s instructions and then quantitated using a fluorescence plate reader (Tecan Infinite M200) with the excitation and emission wavelengths of 400 and 505 nm, respectively. The values of relative fluorescence units with respect to the controls were plotted against different treatments.

10⁵ cells were treated with DAT1 for the indicated time period after attachment. Following incubation, 0.5 mL Buffer A (10 mM HEPES, pH 7.9, 10 mM KCl, 0.1 mM EDTA, protease inhibitors) was added, and cells were collected by scraping. After centrifugation at 13000 rpm for 3 min, the supernatant was preserved as cytoplasmic extract. To the pellet 150 μL of Buffer B (20 mM HEPES, pH 7.9, 0.4 M NaCl,1 mM EDTA, 10 % Glycerol) was added and kept in ice for 1 h with intermittent shaking followed by centrifugation at high speed. The supernatant was collected as the nuclear extract.

Cells were seeded in 35 mm dishes and after attachment, they were transfected with ERK or control siRNA using the transfection reagent lipofectamine 2000. The cells were further incubated for 72 h before addition of drug and processing for western blotting or DAPI staining.

Swiss Albino mice and SCID/NOD mice were used for the experiments. Animals weighing 25–30 g and age of 9–11 weeks were used in the experiments. All the animal experiments were conducted as per the approved guidelines of Institute Animal Ethics Committee which is under the Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA), Govt. of India. The mice were handled and housed in conventional plastic cages and maintained on an automatic 12 h lighting cycle at a temperature of 22-24 °C.

In acute toxicity studies, Swiss Albino mice were divided in to six groups of six animal each and DAT1 (12–50 mg/kg) was administered intraperitoneally (i.p) and observed for 14 days for signs of weight loss, dehydration or mortality. DAT1 was dissolved in vehicle cremaphor:ethanol (1:1) and diluted in 0.9 % saline.

In sub-acute toxicity studies, Swiss albino mice were divided into five groups of eight animals each. Three groups of Swiss Albino mice were administered (i.p injection) with 4, 8 or 16 mg/kg DAT1, and the control and vehicle control groups were administered with 0.9 % saline or vehicle respectively. Injections were given every alternate day for the first 10 days and after that, twice weekly for a period of 3 months. After 3 months animals were sacrificed and whole blood, liver and kidney were collected. Hematotoxicity was checked by RBC and WBC count. Hepatotoxicity was measured by monitoring the serum ALT and AST levels. Nephrotoxicity was measured by monitoring BUN and creatinine levels in the serum. ALP activity was measured for detection of liver or bone toxicity.

SCID/NOD mice were divided into groups of ten and allowed to acclimatize. After 5 days, tumours were raised by injecting HCT116 or HCT116 p53 −/− cells (1.5 × 10⁵ cells suspended in PBS) subcutaneously in the inner flank region. Once the tumour volumes reached 20–80 mm³, drug was administered intraperitoneally, thrice a week at different concentrations.

Fixed tissues were stored in sucrose and sectioned in a cryostat. The sections were collected in gelatin coated slides and tunnel assay was done according to manufacturer’s protocol (R & D Biosystems, USA). Diaminobenzidine (DAB) staining was used to detect the apoptotic cells.

The number of apoptotic cells was determined by counting from at least five fields. Results are shown as mean ± standard deviation. Standard deviations were determined from three independent experiments. Biorad quantity one software was used for densitometric analysis of blots. Blots were normalized with actin, ERK, p38 or JNK/SAPK wherever necessary. For animal experiments, results are shown as mean ± standard error of mean. Data for at least six animals for sub-acute toxicity studies and eight animals for xenograft studies were analyzed. Unpaired student’s t test was used for finding the statistical significance and p ≤ 0.05 were considered as statistically significant.