Background
The interplay between oncogenic signal transduction pathways and their downstream mediators has been extensively characterized over the past two decades. These signaling events are transmitted by protein-protein interactions that are frequently regulated by phosphorylation events [
1]. PI3K/Akt signaling is a major signal transduction cascade involved in the regulation of a number of cellular processes including cellular proliferation, survival, and metabolism. PI3K/Akt signaling has been implicated in the progression and metastasis of a wide range of cancers [
2]. The Akt protein kinase, comprised of 3 isoforms (Akt1, 2 and 3), is a direct downstream effector of PI3K, which becomes fully activated by phosphorylation at the T308 and S473 sites [
2,
3]. Activated Akt is frequently observed in poorly differentiated tumors where it bridges the link between various oncogenic receptors and pro survival cellular functions making the tumor cells highly invasive and less responsive to chemotherapeutic drugs [
2,
4].
The Akt effects on aberrant cell survival are mediated by the regulation of a number of critical downstream proteins that have been implicated in apoptosis and anoikis including Bad, Caspase9, IKK, Mdm2 and FHKR [
1,
5,
6]. Akt is also involved in cell cycle regulation by phosphorylation and inactivation of the cyclin dependent kinase inhibitors p21 and p27/kip1 [
1,
7]. Constitutively activated Akt has been linked to epithelial-to-mesenchymal transition (EMT) by regulating MMPs resulting in reduced cell-to-cell adhesion, increased motility and invasion. It has also been reported that Akt–driven EMT may confer the motility required for malignant progression and dissemination of cancer cells to distant organs [
8,
9]. Recently, we identified a new pathway by which TGFβ/PKA/PP2A signaling deactivates Akt phosphorylation leading to downregulation of IAPs, XIAP and survivin in colorectal cancer (CRC) cells [
10,
11].
The broad roles of this enzyme in cancer have established Akt as an attractive therapeutic candidate in cancer. Small molecule inhibitors of the PI3K/Akt pathway are being developed for clinical use. Several Akt inhibitors have been synthesized, including MK-2206, a novel allosteric kinase inhibitor of Akt [
12‐
14]. MK-2206 binds to the pleckstrin-homology (PH) domain of Akt and thereby inhibits PDK1 binding and activation of Akt. This results in change in confirmation of Akt and its inability to localize to the plasma membrane [
12‐
14]. MK-2206 has shown promising preclinical activity and is currently undergoing phase II clinical evaluation. Although the specific mechanisms underlying the anti-cancer activity of MK-2206 remain to be fully elucidated, MK-2206 has been shown to induce cell cycle arrest and apoptosis [
12‐
14].
We now report that MK-2206 induces anti-tumor activity in a subset of human CRC cell lines characterized by their dependence on IGF1R signaling which leads to PI3K/Akt upregulation for cell survival [
15]. Strikingly, exposure to MK-2206 resulted in the generation of 2 mechanisms of cell death, which have not previously been documented, for this drug. The MK-2206-dependent cell death of IGF1R-dependent CRC cells
in vitro and
in vivo was characterized by Apoptosis Inducing Factor (AIF) induction and its mitochondria-to-nuclear translocation, which is known to induce caspase-independent cell death [
16,
17]. Additionally, MK-2206-dependent cell death was also characterized by the inactivation of the cytoskeletal organizing protein Ezrin at T567 leading to the loss of Inhibitor of Apoptosis (IAP) family protein XIAP. It has been reported that aberrant increase of Ezrin phosphorylation at the T567 site generates increased cell survival and metastatic capabilities of cancer cells [
18‐
21]. In summary, our results indicate that MK-2206 is a promising therapeutic candidate for treatment of IGF1R-dependent CRC characterized by PI3K/Akt signaling upregulation.
Methods
Cell lines and reagents
GEO [
22] and CBS [
23] colon carcinoma cells were cultured in serum free (SF) medium (McCoy’s 5A with pyruvate, vitamins, amino acids and antibiotics) supplemented with 10 ng/ml epidermal growth factor, 20 μg/ml insulin and 4 μg/ml transferrin at 37°C in a humidified atmosphere of 5% CO
2 [
10,
11] . When the cells were under GFDS (growth factor deprivation stress) [
24], they were cultured in Supplemental McCoy’s (SM) medium in the absence of growth factor or serum supplements for the indicated times as described previously [
24]. HCT 116 (IGF1R- independent colon cancer cell lines) [
25,
26] and MiaPaCa (pancreatic cancer cells with constitutive activation of IGF-1R) cells [
27] were used as control to demonstrate the specificity of the dose of the kinase inhibitor. The
in vivo experiments were carried out with GEO cells stably transfected with a GFP vector to visualize the tumor size. MK-2206 was provided by Merck and Co., Inc. MK-2206 was dissolved in DMSO for
in vitro experiments. However, for
in vivo experiments 30% Captisol (Cydex Pharmaceuticals) was used as a vehicle for the drug. In
in vitro experiments, the control cells were treated with DMSO. The control animals also received 30% Captisol. AIF inhibitor, N- Phenylmaleimide was purchased from sigma.
Proliferation assay
GEO cells were plated at a density of 8 × 10
3 cells per well in a 96 well plate. After 72h the cells were treated with increasing concentrations of MK-2206. Cell proliferation was measured after 48h by 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide (MTT) assay as described previously [
15].
DNA fragmentation assay
Cells were seeded in 96 well plates at the same density as for proliferation assays. MK-2206 was treated 72 h after plating the CRC cells. DNA fragmentation assays were performed after 48 h of treatment using a Cell Death Detection ELISA plus kit (Roche) according to the manufacturer’s protocol as described previously [
24]. To confirm AIF mediated cell death, DNA fragmentation was performed by pretreating the cells with AIF inhibitor (50 μM/L) for 1 h prior to treatment with MK-2206 for 48 hrs. Additionally a DNA fragmentation assay was performed after siRNA-mediated knockdown of AIF followed by treatment with MK-2206 for 48 hrs. GEO cells were treated with XIAP siRNA for 48 h and then DNA fragmentation was performed to confirm the effect of XIAP on cell death.
Subcellular fractionation
Cells were washed with ice-cold phosphate buffer saline (PBS) twice. The cells were suspended in 1 ml of PBS and centrifuged for 1 min at 4°C. The supernatant was removed, the pellet was dissolved in 1ml of CE buffer, and samples were vortexed for about 15 sec. The samples were kept on ice for an hour, passed through a syringe every 20 minutes and centrifuged for 1 min. The supernatant was collected and the pellet was left out to isolate nuclear extract. The supernatant was centrifuged again for 1 min to get rid of any debris. The supernatant isolated now was designated as the cytoplasmic extract and was stored at −80°C. Nuclear extract buffer was added to the pellet and the sample was vortexed for 20 seconds. The samples were kept on ice for an hour and sonicated twice for 10 seconds at 60% amplitude. The samples were centrifuged for 20 min at 4°C and supernatant collected was stored at −80°C.
Western blot analysis and immunoprecipitation
Cells were lysed in a buffer consisting of 50 mmol/L Tris–HCl (pH 7.4), 150 mmol/L NaCl, 0.5% NP40, 50 mmol/L NaF, 1 mmol/L NaVO3, 1 mmol/L phenylmethylsulfonyl fluoride, 1 mmol/L DTT, 25 μg/mL aprotinin, 25 μg/mL trypsin inhibitor, and 25 μg/mL leupeptin. The supernatants were cleared by centrifugation at 4°C. Protein concentration was measured by bicinchoninic acid assay (Pierce) using a Biotek 96 well plate reader. Protein (30–100 μg) was fractionated by electrophoresis on a 10% acrylamide denaturing gel and transferred onto a nitrocellulose membrane (Life Science, Amersham) by electroblotting. The transfer on the nitrocellulose membrane was routinely confirmed by Ponceau S staining. The membrane was blocked with 5% nonfat dry milk in TBST [50 mmol/L Tris (pH 7.5), 150 mmol/L NaCl, 0.05% Tween 20] for 1h at room temperature or overnight at 4°C and washed in TBST. The membrane was then incubated with primary antibodies at 1:200–1:1000 in TBST overnight at 4°C. After washing with TBST for 15 min, the membrane was incubated with horseradish peroxidase–conjugated secondary antibody (Life Science, Amersham) at 1:1000 dilutions for 1h at room temperature. The proteins were detected by the enhanced chemiluminescence system (Amersham). Immunoprecipitation was performed with 500 μg of protein samples using magnetic beads (Millipore) according to manufacturer’s protocol. Antibodies were purchased from Cell Signaling for tAkt, pAkt (S473), pAkt(T308), AIF (Apoptosis Inducing Factor), pEzrin (Thr567), Akt1, Akt2, Akt3 survivin, Bad and pBad (S136). Ezrin antibody was purchased from Santa Cruz. XIAP antibody was purchased from Abcam.
Retroviral knockdown of Akt1, Akt2 and Akt3
Small hairpin RNA sequence for Akt1si, Akt2si, Akt3si and scramblesi were cloned and expressed in a retroviral expression vector pSUPER.Retro.Puro (Oligoengine). 293T derived Phi-NX cells were used for transfection. A 19-nucleotide sequence for Akt1, Akt2 and Akt3 were designed from Dharmacon si design center. The target sequence for Akt1 5′GAGACTGACACCAGGTATT 3′ was 1634 bases while that for Akt2 5′TGAATGAGGTGTCTGTCAT 3′ selected was 301 bases downstream of 5′UTR. Akt3 target sequences selected were 5′GCAAAATGCCAGTTAATGA 3′. Another non-targeting small hairpin siRNA was used as an experimental control. The GEO cells were stably transfected with siRNA to reduce the expression of Akt1, Akt2 and Akt3. The cells were selected with Puromycin (4 μg/ml) and the resistant cells were pooled. Stable cell lines with Akt1, Akt2 and Akt3 knockdown were maintained in serum free medium with puromycin (4 μg/ml).
RNA interference studies
XIAP siRNA (ON-TARGET plus Human XIAP (331) siRNA smart pool), AIF si RNA (ON-TARGET plus Human AIF siRNA smart pool) were purchased from Thermo scientific and transient transfections were done as per manufacturer’s protocol.
Immunofluorescence
The translocation of AIF from mitochondria to nucleus was determined by, immunofluorescence assay. GEO cells were plated on a cover slip in a six well plate. When the cells were 60-70% confluent, culture medium with 400 nm of Mitotracker (CMX Ros, Invitrogen) was added to the cells. The cells were checked for red fluorescence under the microscope after one hour. The cells were stained, washed with growth medium and fixed by placing in ice-cold methanol for 5 minutes. The cells were washed with PBS, permeabilized by incubating with PBS containing 0.1% Triton X-100 for 15minutes and subsequently blocked with 10% normal goat serum. After one hour of blocking, the cells were incubated with primary antibody for AIF (1:100) for 2 h. Fluorescein isothiocyanate- conjugated anti rabbit antibody (FITC) was used as the secondary antibody. Nuclei were counter stained with 4′-6 diamino-2- phenylindole and mounted on glass slide in anti fade vecta shield mounting medium (vector labs). An LSM 510 microscope (Carl Ziess GmbH, Oberkochen, Germany) was used to perform laser confocal microscopy.
Xenograft studies
All the experiments involving animals were approved by the University of Nebraska Medical Center Institutional Animal Care and Use Committee. 3–5 week old athymic nude mice (N = 16) were purchased from NCI. 7×106 GEO GFP labeled cells were subcutaneously injected on one side in the right flank pad of mice and allowed to form xenografts. When the tumor size was approximately 100 mm3, 120 mg/kg body weight of MK-2206 was administered orally. Captisol was used as a vehicle for the drug and the control animals were treated with vehicle only. MK-2206 was given orally for 3 weeks on alternate days. The dose and the duration mentioned in the study have been provided by Merck and Co. based on standard mono therapy efficacy studies on mice. Tumor growth and body weight were measured every other day. The tumor size was measured manually with calipers, and the tumor volume was calculated using the formula (l2 × h × π/6). We used Near-IR enhanced Macro Imaging System Plus Cooled with the LT-99D2 with the Dual Tool dual excitation upgrade for viewing the 2D image of the tumor as well as to image the mice. All in vivo characterizations were confirmed in at least 3 independent control and MK-2206 treated animals.
Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay
The mice were euthanized after 21 days of treatment with MK-2206. The xenograft tumors were harvested after imaging to determine the size of the tumor using a microimaging system and immediately placed in 10% neutral buffered formalin fixative for 24 h. This was followed by lysate preparation and embedding in paraffin. Sections (4 μM) from paraffin embedded blocks were stained according to the Apotag terminal nucleotidyl transferase mediated nick end labeling (TUNEL) assay kit. The apoptotic rates were determined by counting the number of positively stained apoptotic bodies at 40× magnification. Fifteen different fields were randomly selected per slide for analysis. The ratio of the average number of apoptotic cells to the total number of cells counted (4000 cells each for control and treated groups) was used to determine apoptotic rates [
28].
Hematoxylin and Eosin staining and Ki67 staining
Sections (4 μM) from paraffin embedded blocks were used for H and E staining and for Ki67 IHC using antibody for Ki67 from BD biosciences. Ki67 is a non-histone nuclear antigen present in late G1, G2 and S phase of cell cycle but absent in G
0. The dilution of Ki67 antibody used was 1:100. The proliferation rate was determined quantitatively by utilizing NIH Image J software (public domain software). Ten different, but histologically similar fields, were selected for analysis [
28].
Immunohistochemistry
The slides were deparafinized by keeping them at 60°C for 1 h and then rehydrated using graded alcohol for 5 min each. Subsequently the slides were treated with 0.3% H2O2/methanol for 10 min and then submerged in 95°C citrate buffer (pH = 7.8) for 15 min. Blocking was performed in 5% normal goat serum for 1h at room temperature and then incubated with primary antibody for tAkt (1:100) and pAkt (1:25) at 4°C overnight. The slides were treated with Biotinylated secondary antibody for 30 min at RT, followed by incubation with streptavidin peroxidase complex (Invitrogen). Reaction products were developed using diaminobenzidine containing 0.3% H2O2 as a substrate for peroxidase (Dako). Nuclei were counterstained with hematoxylin (Protocol). To determine the difference in staining intensity for total and phospho Akt, 10 different but histologically similar fields were selected per sample and the slides were analyzed using NIH image J software. The staining intensity measured by the software was plotted using Graph pad 5.0.
Statistical analysis
Statistical analysis was performed using Graph pad 5.0 software for student’s t test. A P value of less than 0.05 was considered significant.
Discussion
Extensive drug development efforts and clinical evaluations are underway targeting the aberrant cell survival properties associated with PI3K/Akt signaling in regulating cancer progression and metastasis [
1]. Inhibition of Akt activation by small molecule kinase inhibitors is an attractive candidate for targeting aberrant cell survival associated with malignant progression and metastasis and could be effective in the treatment of CRC. MK-2206 is a novel Akt allosteric kinase inhibitor, which is currently in clinical evaluation [
12‐
14].
Several studies have described MK-2206 effects as a single agent or in combination with other inhibitors (e.g. PI3K or mTOR inhibitors) on cell proliferation and/or cell death. Gorlick et.al. [
32] demonstrated a significant reduction in tumor volume
in vivo and decreased cell survival
in vitro in pediatric cancer cell lines following MK-2206 treatment. Simoni et.al. [
33] studied the effect of MK-2206 in T cell acute lymphoblastic leukemia demonstrating cell cycle arrest in G0/G1 phase, apoptosis and autophagy. Ma et.al. [
34] showed that MK-2206 treatment in nasopharyngeal carcinoma cells (NPC) induced cell cycle arrest and apoptosis. Similarly, we observed that MK-2206 treatment in the IGF1R-dependent GEO cells reduced cell proliferation and increased cell death in a concentration dependent manner (Figure
2A, B) while MK-2206 has been shown to be effective in causing cell death in different types of cancer. However, specific mechanisms associated with MK-2206-mediated cell death have not been characterized. This study identifies molecular mechanisms involved in MK-2206-mediated cell death in IGF1R- dependent CRC cells in response to Akt inhibition. Identification of specific mechanisms may generate new therapeutic targets that offer potential for enhancing cell death of CRC cells. The mechanistic novelty of this study is our identification of 2 pathways whereby MK-2206 treatment leads to control of aberrant cell survival and induction of cell death
in vitro and
in vivo.
We studied the expression of various apoptosis-regulators following exposure to MK-2206. As expected, a reduction in phospho-Bad (pBad) at the Ser 136 site was observed (Figure
2D), which is known to be regulated by Akt signaling [
29]. It is known that pBad interacts with 14-3-3, a major mediator of cell survival providing an anti-apoptotic milieu to the cellular environment [
35]. We observed that treatment with MK-2206 results in reduced 14-3-3 interaction with pBad (Ser136) indicating that MK-2206 results in reduction in cell survival through this mechanism. The protein expression of Bad remained unchanged following MK-2206 treatment; however, there was an increase in the interaction of Bad with Bcl-x
L. Bad inactivates Bcl-x
L thus leading to increases in cell death. Additionally, we observe a decrease in the interaction of Bad with 14-3-3 on treatment with MK-2206. This might suggest that Bad remains activated leading to apoptosis of colorectal cancer cells.
Strikingly, we made the observation that MK-2206 exposure led to an induction of pro-apoptotic protein AIF and its translocation from mitochondria to the nucleus of the GEO cells (Figure
6A, C). It has been reported that AIF is responsible for caspase-independent death in ovarian cancer cells [
30,
36,
37]. AIF is localized in the mitochondria but upon activation it translocates to the nucleus and causes DNA fragmentation [
38]. However, the mechanism that regulates AIF induction leading to its caspase-independent apoptotic functions is not well understood. Treatment with AIF inhibitor resulted in reduced cell death thus indicating that AIF is responsible for cell death mediated by MK-2206.
MK-2206 treatment of GEO cells reduced survivin and XIAP levels both
in vivo and
in vitro (Figure
2D, Additional file
1: Figure S3). Survivin and XIAP are key cell survival-associated proteins that have been characterized as having an important role in metastasis [
39]. XIAP binds to caspases 3, 7 and 9 thereby inhibiting their pro-apoptotic activity [
39,
40]. During stress conditions, mitochondrial XIAP and survivin migrate to the cytosol forming a survivin/XIAP complex, which inhibits caspases and promotes cytoprotection [
40]. Dan et al. [
41] made the novel finding that Akt phosphorylates XIAP at a stabilizing Ser87 site. We demonstrated that TGFβ/PKA signaling regulates aberrant cell survival in IGF1R-dependent CRC cells by disengaging survivin/XIAP complex formation thus causing caspase activation and inducing cell death. We sought to determine the mechanism by which MK-2206 increased XIAP loss and cell death. It was observed that MK-2206 treatment dephosphorylates Ezrin at the Thr567 site (Figure
7A, B). However, no change in total Ezrin protein expression was observed. Ezrin is a member of Ezrin-radixin-moesin (ERM) protein family that plays a key role in cancer progression and metastasis in a wide range of cancers, including CRC [
42]. Ezrin is found in a closed confirmation in the cytosol. Ezrin phosphorylation at Thr567 leads to its activation and conformational change to an open conformation resulting in its localization to the plasma membrane for its oncogenic-associated functions [
43]. Several kinases are known to phosphorylate Ezrin at T567 including Rho kinase and PI3K/Akt [
44]. We performed siRNA knockdown of Ezrin and observed a complete loss of XIAP and survivin (Additional file
1: Figure S11). Thus, we have found that MK-2206 treatment inhibits the Akt-pEzrinT567-XIAP cell survival-signaling axis leading to a caspase-dependent cell death in the IGF1R-dependent CRC cells, in addition to caspase independent cell death accompanying AIF translocation from the mitochondria to the nucleus.
Stable knockdown of Akt2 in the IGF1R-dependent and highly metastatic colon cancer cell line GEO was performed to give a better understanding of the mechanism of cell death mediated by loss of pEzrin. Loss of Akt2 resulted in decreased the activation of Ezrin since there was a loss of phosphorylation of Ezrin at the T567 site. Besides loss of pEzrin we also observed a reduction in the expression of XIAP on knockdown of Akt2. However, there was no such loss of pEzrin on knockdown of Akt1 and Akt3 in GEO cells. Thus we can conclude that loss of the Akt2 isoform is responsible for Akt-pEzrin-XIAP mediated cell death.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
EA carried out the majority of the in vitro and in vivo experiments in the study analyzed the data and drafted the manuscript. ABC and PDL helped with the in vivo studies. KLH helped with performing western blots. MGB and SC participated in the conception and design of the study and helped to draft the final manuscript. All authors read and approved the final manuscript.