IJC International Journal of Cancer

PDGFR-b-activated ACK1-AKT Signaling Promotes Glioma Tumorigenesis Jiannan Zhang1*, Tao Chen2*, Qin Mao3, Jinbo Lin2, Jun Jia2, Shanquan Li3, Wenhao Xiong3, Yingying Lin4, Zhiqiang Liu2, Xiaoyu Liu5, Hailiang Zhao2, Guisong Wang3, Duo Zheng5, Shuqi Qiu2 and Jianwei Ge3 1

Operation Room, Shanghai Children’s Medical Center, School of Medicine, Shanghai Jiaotong University, Shanghai 200127, People’s Republic of China ENT Hospital, Longgang Central Hospital, Otolaryngology Institute of Shenzhen University, Shenzhen, Guangdong 518116, People’s Republic of China 3 Department of Neurosurgery, Renji Hospital, School of Medicine, Shanghai Jiaotong University, Shanghai 200127, People’s Republic of China 4 Institute of Health Sciences, Shanghai Institutes for Biological Sciences of Chinese Academy of Sciences-Shanghai Jiao-Tong University School of Medicine (SJTU-SM), Shanghai 200025, People’s Republic of China 5 School of Medicine, Shenzhen University, Shenzhen, Guangdong 518060, People’s Republic of China

Aberrant PDGF-PDGFR signaling and its effects on downstream effectors have been implicated in glioma development. A crucial AKT regulator, ACK1 (TNK2) has been shown to be a downstream mediator of PDGF signaling; however, the exact underlying mechanisms in gliomas remain elusive. Here, we report that in glioma cells, PDGFR-b activation enhanced the interaction between ACK1 and AKT, resulting in AKT activation. PDGF treatment consistently promoted the formation of complexes containing PDGFR-b and ACK1. Mutational analysis suggested that Y635 of ACK1 is a PDGFR-b phosphorylation site and that the ACK1 Y635F mutant abrogated the sequential activation of AKT. Moreover, PDK1 interacted with ACK1 during PDGF stimulation, which is required for the binding of ACK1 to PDGFR-b. Further mutational analysis showed that T325 of ACK1 was crucial for the ACK1 and PDK1 interaction. ACK1 Y635F or T325A mutants abolished PDGFR-b-induced AKT activation, the subsequent nuclear translocation of bcatenin and the expression of cyclin D1. Glioma cell cycle progression, proliferation and tumorigenesis were accordingly blocked by ACK1 Y635F or T325A. In glioblastoma multiforme samples from 51 patients, increased ACK1 tyrosine phosphorylation correlated with upregulated PDGFR-b activity and AKT activation. Taken together, our data demonstrate that ACK1 plays a pivotal role in PDGF-PDGFR-induced AKT signaling in glioma tumorigenesis. This knowledge contributes to our understanding of glioma progression and may facilitate the identification of novel therapeutic targets for future glioma treatment.

Key words: PDGFR, glioma, ACK1, AKT, tumorigenesis *J.Z. and T.C. contributed equally to this work This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. Grant sponsor: National Natural Science Foundation of China; Grant number: 81372705; Grant sponsor: Shanghai Science and Technology Development Fund; Grant number: 10JC1409802; Grant sponsor: Wu Jieping Medical Foundation; Grant number: 320.6750.11092; Grant sponsor: National Natural Science Foundation of China; Grant number: 81402963; Grant sponsor: China Postdoctoral Science Foundation; Grant number: 2013T60791; Grant sponsor: Science & Technology Project of Shenzhen Longgang District; Grant number: YLWS20140609111127924; Grant sponsor: Municipal Science and Technology Project of Shenzhen; Grant number: 201404113000346; Grant sponsor: National Natural Science Foundation of China; Grant numbers: 81372149, 81071655, 30871247; Grant sponsor: Science & Technology Innovation Commission of Shenzhen; Grant numbers: JCYJ20130329102515481, ZDSY20130329101130496 DOI: 10.1002/ijc.29234 History: Received 29 Apr 2014; Accepted 10 Sep 2014; Online 25 Sep 2014 Correspondence to: Jianwei Ge, PhD, MD, Department of Neurosurgery, Renji Hospital, School of Medicine, Shanghai Jiaotong University, Shanghai 200127, People’s Republic of China, Tel.: 1[86-21-6838-3768], Fax: 1[86-21-5875-2311], E-mail: [email protected], or Shuqi Qiu, MD, Longgang Central Hospital; ENT hospital of Longgang Central Hospital; Otolaryngology Institute of Shenzhen University; Otolaryngology Institute of Shenzhen, Shenzhen, Guangdong 518116, People’s Republic of China, Tel.: 1[86-0755-2898-0366], Fax: 1[860755-2898-0366], E-mail: [email protected] or Duo Zheng, Ph.D., School of Medicine, Shenzhen University, Shenzhen, Guangdong 518060, People’s Republic of China, Tel.: 1[86-0755-8667-4681], Fax: 1[86-0755-8667-1906], E-mail: [email protected], or Guisong Wang, PhD, Department of Neurosurgery, Renji Hospital, School of Medicine, Shanghai Jiaotong University, Shanghai, 200127, People’s Republic of China, Tel.: 1[86-21-6838-3768], Fax: 1[86-21-8522-1983], E-mail: [email protected]

C 2014 The Authors. Published by Wiley Periodicals, Inc. on behalf of UICC. Int. J. Cancer: 136, 1769–1780 (2015) V

Carcinogenesis

2

1770

PDGFR-b-activated ACK1-AKT promotes glioma tumorigenesis

Carcinogenesis

What’s new? Gliomas are the most common malignant tumors in the central nervous system, and molecular insight into their establishment and maintenance is urgently needed. Here the authors identify the non-receptor tyrosine kinase ACK1 (also known as TNK2) as an important link between the platelet-derived growth factor receptor (PDGFR) beta and the downstream activation of the growth-regulating kinase AKT, thus facilitating glioma cell proliferation and survival. The demonstration of a dedicated PDGFinduced PDGFR-ACK1-AKT signaling axis that also involves the kinase PDK1 could be an important step towards more targeted treatments of this deadly cancer.

Gliomas are the most common form of primary malignancy arising from the central nervous system (CNS) in humans.1,2 Based on distinct clinical and histological traits, gliomas are classified into astrocytomas, oligodendrogliomas, oligoastrocytomas or ependymomas.3 Several studies have reported a link between the status of growth factors and glioma development with potentially important implications for clinical therapy.4,5 For instance, the epidermal growth factor (EGF) and the platelet-derived growth factor (PDGF) family of growth factors can promote glioma cell transformation via the activation of the tyrosine kinase receptors EGFR and PDGFR, which in turn activate their downstream effectors.6–10 Therefore, investigation of the aberrant effects of EGFR and PDGFR on glioma development could facilitate the identification of new therapeutic targets. ACK1 (TNK2), which belongs to a non-receptor tyrosine kinase family, is expressed in several mammalian tissues types.10,11 Together with other well-known non-receptor tyrosine kinases, such as Src and Csk, ACK1 has been reported to be involved in the development of different types of tumors,12–14 due to its role in the coordination of various important signaling pathways.11,15 In prostate cancer, ACK1 promotes tumorigenesis by blocking the function of the tumor suppressor Wwox.14,16,17 Ack1 is also critically involved in the radiation resistance in castration-resistant prostate cancer through regulating the androgen receptor phosphorylation.18 Distinct stimuli, including EGF or PDGF, can induce ACK1 activation.19 Phosphorylation at Y284 is required for ACK1 kinase activity,17 and ACK1 is required for AKT activation.15,20–22 EGF stimulation activates ACK1, which in turn directly phosphorylates Y176 of AKT.20 The robust activation of ACK1 can facilitate the development of prostatic intraepithelial neoplasias in mice and has been correlated with the survival outcome of breast cancer patients.20 Given the potential importance of the relationship between ACK1 and diverse growth factors, it is essential to further elucidate the mechanisms underlying these effects. Under normal conditions, PDGFs are involved in cellular growth and differentiation during embryonic development and normal tissue repair.23,24 However, the dysfunction of PDGFs and their cognate receptors has been shown to play an important role in human carcinogenesis.7,25,26 Studies of PDGFs and their receptors have provided valuable insights into glioma progression. PDGFs and their receptors, PDGFRa and PDGFR-b, can be detected in human gliomas and are most highly expressed in glioblastomas.6 PDGFR-a primarily

contributes to the growth and survival of glial tumors, whereas PDGFR-b stimulates angiogenesis in gliomas.6 In this study, we demonstrated that PDGFR-b could bind to ACK1 during PDGF treatment, promoting ACK1 activation, for which PDK1 functions as a crucial adaptor. The activation of ACK1 induced by PDGF promoted AKT activation, thereby facilitating glioma cell proliferation and survival. Our study suggests that ACK1 plays a pivotal role in PDGFPDGFR-induced AKT signaling in glioma tumorigenesis.

Material and Methods Tumor samples and plasmids

A total of 51 glioblastoma multiforme (GBM) specimens with adjacent normal brain tissues were selected from consenting patients with primary gliomas who had been operated on at the Department of Neurosurgery, Renji Hospital, School of Medicine, Shanghai Jiaotong University China (Ethical Approval Letter No.: 2010-AN-3). The samples were harvested from the tumors at the time of surgery and were snap-frozen and stored at 280 C. Full-length human ACK1 and PDGFR-b cDNAs were obtained from HEK293T cells using reverse transcription PCR (RT-PCR), confirmed using DNA sequencing and subcloned into the p3XFlag-CMV-10 expression vector. Ack Y635F, Y859F, T325A, T899A, rACK1 WT, rACK1 Y635F, rACK1 Y859F, rACK1 T899A, rACK1 T325A and PDGFR-b T681M were constructed using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). The transcriptional activity of b-catenin was measured using the TCF/LEF reporter kit (SABiosciences, Frederick, MD). Reagents, cell culture and transfections

AKT inhibitor VIII (Calbiochem, Gibbstown, NJ), plateletderived growth factor receptor (PDGFR) inhibitor AG1296 (Sigma, St. Louis, MO) and recombinant human glutathione S-transferase-AKT (GST-AKT) (Novus Biologicals, Littleton, CO) were used in this study. The U87 (human glioblastomaastrocytoma, epithelial-like cell line) and U251 human glioma cell lines were obtained from the Cell Bank of Type Culture Collection of the Chinese Academy of Sciences (Shanghai Institute of Cell Biology, Chinese Academy of Sciences, Shanghai, China) and grown in dulbecco’s minimum essential medium (DMEM) supplemented with 10% fetal calf serum (Life Technologies, Gaithersburg, MD), penicillin G (100 U/ ml) and streptomycin (100 lg/ml) at 37 C in a 5% CO2 atmosphere. The U87 cells were transfected with Lipofectamine 2000 (Invitrogen, Carlsbad, CA).

C 2014 The Authors. Published by Wiley Periodicals, Inc. on behalf of UICC. Int. J. Cancer: 136, 1769–1780 (2015) V

1771

Zhang et al.

Immunofluorescence

In vitro kinase assay

For immunofluorescence, cells were washed with cold phosphate-buffered saline (PBS) and fixed in paraformaldehyde solution (4% in PBS, pH 7.4) with a permeabilizing agent (0.1% Triton X-100) for 10 min at room temperature. Following incubation in blocking buffer [3% bovine serum albumin (BSA) in Tris-buffered saline-tween (TBST) solution] for 1 hr, the cells were incubated with primary antibodies in TBST containing 3% BSA overnight at 4 C. Next, the cells were washed with PBS and then incubated with the indicated fluorescent-labeled secondary antibodies in the dark at 37 C for 1 hr. The cells were then examined via fluorescence microscopy. Confocal images were captured via confocal microscope (Leica, TCS SP5).

A total of 100 ng of recombinant PDGFR-b (Biovision, Milpitas, CA) was mixed with 100 ng of ACK1 (Abcam) in 20 ll of kinase buffer [25 mM MOPS(3-Morpholinopropanesulfonic acid), 12.5 mM b-glycerolphosphate, 20 mM MgC12, 25 mM MnC12, 5 mM EGTA, 2 mM EDTA] at 25 C for 1 hr. The final reaction products were then used for SDSPAGE analyses.

Tissue or cell lysates were pre-cleared with protein A-sepharose for 1 hr at 4 C. After pre-clearing, the indicated proteins were immunoprecipitated with 2 lg of the appropriate antibody with gentle rocking at 4 C overnight. Next, protein A-sepharose was added, and the mixture was incubated for 4 hr at 4 C. The beads were washed five times with lysis buffer. Immunoprecipitation was performed with the indicated antibodies. For immunoblotting, sodium dodecyl sulfate-polyacrylamide gelelectrophoresis (SDS-PAGE)-resolved proteins were transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, MA). The membranes were developed via enhanced chemiluminescence (ECL, Amersham Pharmacia Biotech, Buckinghamshire, UK). Antibodies

Antibodies for regular tyrosine phosphorylation, PDGFR, phospho-PDGFR-b (Y751), b-catenin, cyclin D1, tubulin, total AKT, phospho-AKT (S473) and PDK1 were obtained from Cell Signaling Technology (Danvers, MA). ACK1 antibody was obtained from Abcam (Cambridge, MA). Antibodies for Foxo1, Bcl2 and HRP-labeled goat anti-mouse, as well as goat antirabbit IgGs and anti-myc were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Alexa Fluor 546-labeled goat antimouse antibody and Hoechst 33342 were purchased from Molecular Probes (Eugene, OR). RNA interference

For ACK1 knockdown, the oligonucleotide ACTCATCCACGACTTGCGT was selected and inserted into the FG12 vector after its knockdown efficiency was compared with the other oligonucleotide, TTCTCTGGAAGAAGCAGGT. A scramble oligonucleotide TTGGCACATTGGCCAGATC was also inserted into the FG12 vector. To knock down the expression of PDK1, the oligonucleotide TTATTACAAGACTACTGAC was selected and inserted into the FG12 lentiviral shRNA vector after its knockdown efficiency was compared with the other oligonucleotide, ATAAGATACTCGTTTCCAG. A scramble oligonucleotide, GGAATCTTGCATAAGCATA, was also inserted into the FG12 vector. Cell lysates were prepared, and ACK1 and PDK1 protein levels were assessed by Western blot analysis to determine the RNAi knockdown efficiency.

R reagent (Invitrogen, Total RNA was extracted using TRIzolV Carlsbad, CA) and then reverse transcribed with M-MLV reverse transcriptase (Promega, Madison, WI). Real-time PCR was performed as previously described27 on an ABI 7500 fast sequence detection system (Applied Biosystems, Carlsbad, CA) with the fluorescent dye SYBR green. The human b-actin transcript was used as an internal reference to control for variations in the total quantity of mRNA in each sample. Each RNA sample was analyzed in triplicate. The following primers were used in this study:

1. Human ACK1 primers: forward, 50 -TCACAAGCCAGAGGA CAGAC-30 , reverse 50 -ATCTGGATGTGCAGCTTGTC-30 ; 2. Human PDGFR-b primers: forward, 50 -ACGGAGAGTG TGAATGACCA-30 , reverse 50 -GATGCAGCTCAGCAAATT GT-30 ; 3. Human b-actin primers: forward, 50 -CACTCTTCCAGCC TTCCTTC-30 , reverse 50 -GGATGTCCACGTCACACTTC-30 . Luciferase assay

For luciferase assays, U87 cells were seeded in 24-well plates and transfected with the indicated plasmids according to the manufacturer’s protocol for the TCF/LEF reporter kit (SABiosciences, Frederick, MD). Triplicate transfections were performed for each treatment. After transfection, the cells were harvested in lysis buffer containing 1 mM phenylmethanesulfonyl fluoride (PMSF) and 0.1% Triton X-100 in PBS. Flow cytometric analysis

For flow cytometric analyses, 1 3 106 U87 cells transfected with the indicated plasmids were treated with platelet-derived growth factor subunit B homodimer (PDGF-BB) (50 ng/ml), fixed in cold 70% ethanol and incubated with DNase-free RNase A (100 lg/ml) for 30 min and propidium iodide (50 lg/ ml) for 15 min. Next, the cell cycle phase distribution of the treated cells was assessed via fluorescence-activated cell sorting (FACS). The data represent the means 6 standard deviation (SD) of five independent experiments. Proliferation rate Assay

The U87 cell proliferation rate was assessed using an methylthiazolyldiphenyl tetrazolium bromide (MTT) assay. U87 cells (3 3 104 cells /well) transfected with indicated plasmids were seeded into 96-well plates. After 4 days, the cells were fixed with MTT

C 2014 The Authors. Published by Wiley Periodicals, Inc. on behalf of UICC. Int. J. Cancer: 136, 1769–1780 (2015) V

Carcinogenesis

Immunoprecipitation and immunoblotting

Real-time PCR

1772

PDGFR-b-activated ACK1-AKT promotes glioma tumorigenesis

for 4 hr at 37 C; the sediment was then dissolved with 150 ml of dimethyl sulfoxide (DMSO). The absorbance at 570 nm was used to record the relative value using a microplate reader. The data represent the means 6 SD of five independent experiments. Subcutaneous injection

We subcutaneously injected 4-week-old athymic nude mice with 5 3 105 U87 cells that expressed the indicated plasmids. Six mice were assigned to each group. Four weeks after tumor cell injection, tumor formation was monitored, and the tumor volume was measured.

Carcinogenesis

Statistical analyses

Each experiment was repeated three or more times. Values in study are represented as the means 6 SD of at least three independent experiments and (two-tailed, unpaired) Student’s t-test was used to compare two groups of independent samples. Analyses of Student’s t-test met the normal distribution; the F-test was used to compare variances, and the assured variances are not significantly different. The chi-squared test was used to analyze the correlations between the tyrosine phosphorylation of ACK1, p-AKT-S473 and p-PDGFR-b Y751 in the glioma specimens. All cell line-derived data were evaluated with Student’s t-test.

Results ACK1 mediates PDGF-induced AKT activation

To determine the potential effects of ACK1 on AKT activation during PDGF stimulation in glioma cells, we treated glioma cells with PDGF for various lengths of time. PDGFR-b activation was validated by analyzing the PDGFR-b pY751 levels (Fig. 1a). Consequently, treatment with PDGF in U87 cells enhanced the interaction between ACK1 and AKT, which was accompanied by increased levels of phospho-AKT S473, an indication of AKT activation28 (Fig. 1a). In particular, the peak levels of phospho-AKT S473 appeared at 1 hr (Fig. 1a). In contrast, the amount of ACK1/AKT complex induced by PDGF reached a maximum at 12 hr (the fourth lane), as did the (notable) upregulation of cyclin D1 levels (Fig. 1a). Additionally, treatment of U251 cells with PDGF also enhanced the interaction between ACK1 and AKT, which was accompanied by increased levels of phospho-AKT S473 (Fig. 1b, left panel). Immunofluorescent staining revealed that ACK1 exhibited a diffuse localization pattern within the cytosol at basal levels, whereas it translocated to the plasma membrane and to inner membrane structures after PDGF treatment (Fig. 1b, right panel). The AKT activation, cyclin D1 up-regulation and decrease of FOXO1 levels induced by PDGF treatment (12 hr) could be blocked by either ACK1 knockdown or a PDGFR inhibitor (Figs. 1c and 1d), which were rescued by the expression of a WT-ACK1 shRNA-resistant vector (rACK1 WT, Figs. 1c and 1d). In contrast, ACK1 knockdown only slightly suppressed AKT activation after cells were treated with PDGF for 1 hr (data not shown), which suggested that ACK1 has a more critical role in maintaining or prolonging the AKT activation

induced by PDGF stimulation. The effects of ACK1 knockdown on the levels of the AKT downstream target genes cyclin D1 and Foxo1 indicate that ACK1 is critical for PDGF-induced AKT activation.29 Furthermore, both the PDGFR inhibitor and ACK1 knockdown resulted in the suppression of the PDGFinduced tyrosine phosphorylation of AKT (Fig. 1e). Consistent with previous reports indicating that the tyrosine phosphorylation of AKT primes AKT for complete activation,20 our results showed that the expression of the Y176F mutant of AKT attenuated the EGF- and PDGF-induced increase in the levels of AKT tyrosine phosphorylation and AKT S473 phosphorylation (Fig. 1f). Moreover, treatment with a PDGFR inhibitor in U87 cells inhibited the formation of AKT/ACK1 complexes and the tyrosine phosphorylation of ACK1 (Fig. 1g). More importantly, we found that ACK1 bound to PDGFR-b in PDGF-treated U87 and U251 cells (Fig. 1h), implying the potential direct effects of PDGFR on ACK1 activity. ACK1 Y635 phosphorylation is required for the PDGFinduced activation of AKT

To determine the precise role of ACK1 in the PDGF-induced activation of AKT, we first investigated the phosphorylation of ACK1 on residues potentially regulated by PDGFR-b. As shown in Figure 2a, PDGF treatment resulted in an increased tyrosine phosphorylation of ACK1. Amino acid sequence analysis using the “Scansite” program revealed a potential PDGFR-b phosphorylation site, Y635, which is contained within the following ACK1 sequence: 628-PLPPPPAYDDVAQDE-642. Intriguingly, mutational analysis confirmed the importance of Y635 for the PDGF-induced tyrosine phosphorylation of ACK1, as indicated by the inhibitory effects elicited by the expression of the Y635F mutant but not the Y859F mutant (Fig. 2a). In contrast, the tyrosine phosphorylation of ACK1 induced by EGF was intact in cells with expression of ACK1 Y635F and ACK1 Y859F (Fig. 2b), suggesting that Y635 of ACK1 would exclusively affect PDGFR-b-mediated AKT activation. Additionally, the kinase assay also indicated that PDGFR-b could phosphorylate ACK1 on its tyrosine residue and that the tyrosine phosphorylation was abolished by the ACK1 Y635F mutant (Fig. 2c). Furthermore, we examined whether Y635 of ACK1 is responsible for the enhanced PDGF-induced interaction between ACK1 and AKT. The Y635F mutation abolished the formation of ACK1- and AKT-containing complexes (Fig. 2d) and inhibited PDGF-induced AKT activation (Fig. 2d). Interestingly, the expression of the ACK1 Y635F mutant suppressed both the PDGF-induced tyrosine phosphorylation and S473 phosphorylation of AKT (Fig. 2e). GST pull-down assays demonstrated that GST-AKT binds WT but not Y635F ACK1 in PDGF-treated U87 cells, indicating that the phosphorylation of Y635 is required for the interaction between ACK1 and AKT (Fig. 2f). PDK1 is required for the ACK1-mediated activation of AKT

Furthermore, a co-immunoprecipitation analysis showed that ACK1 bound to PDK1 (a kinase required for AKT

C 2014 The Authors. Published by Wiley Periodicals, Inc. on behalf of UICC. Int. J. Cancer: 136, 1769–1780 (2015) V

1773

Carcinogenesis

Zhang et al.

Figure 1. PDGFR-b activation leads to ACK1-mediated AKT activation. (a) PDGFR activation promotes the interaction between ACK1 and AKT, as well as AKT activity and cyclin D1 levels. U87 cells were treated with PDGF-BB (50 ng/ml) for different lengths of time as indicated. Whole-cell lysates were subjected to immunoprecipitation (IP) and immunoblotting (IB) using the indicated antibodies. (b) PDGFR activation promotes the interaction between ACK1 and AKT in U87 and U251 cells. U87 and U251 cells were treated with or without PDGF-BB (50 ng/ ml) for 12 hr. Whole-cell lysates were subjected to immunoprecipitation (IP) and immunoblotting (IB) using the indicated antibodies. Rabbit IgG was used as a negative control for the IP (left panel). The U87 cells were treated with or without PDGF-BB (50 ng/ml) for 12 hr. The distribution of ACK1 was assessed via immunofluorescence staining using an ACK1 antibody (right panel). (c) ACK1 shRNA knockdown efficiency. U87 cells were transfected with scramble shRNA plasmids (control) or with two ACK1 shRNA plasmids labeled #1 and #2 (upper panel). ACK1 shRNA plasmid #1, with a higher ACK1 shRNA knockdown efficiency, was selected and expressed in U87 cells. ACK1 expression was rescued by the expression of WT-ACK1 shRNA-resistant plasmid (rACK1 WT, bottom panel). Whole-cell lysates were assessed for the expression of the indicated proteins. (d) ACK1 depletion abrogates the PDGF-induced activation of AKT and changes in AKT downstream protein levels. U87 cells were transfected with or without the ACK1 shRNA plasmid. The cells were then treated with or without PDGF-BB (50 ng/ml) for 12 hr in the presence or absence of the PDGFR inhibitor AG1296 (10 mM). Whole-cell lysates were assessed for the expression of the indicated proteins. (e) ACK1 depletion reduces PDGF-induced AKT tyrosine phosphorylation and activation. U87 cells were transfected with or without the ACK1 shRNA plasmid. The cells were then treated with or without PDGF-BB (50 ng/ml) for 12 hr in the presence or absence of the PDGFR inhibitor AG1296 (10 mM). Whole-cell lysates were assessed for the expression of the indicated proteins. (f) The AKT Y176 residue is crucial for the EGF- and PDGF-induced activation of AKT. U87 cells were transfected with or without the plasmid as indicated. The cells were then treated with or without PDGF-BB (50 ng/ml) for 12 hr, or treated with or without EGF (100 ng/ml) for 15 min in the presence or absence of the AKT inhibitor VIII (10 lM). Whole-cell lysates were assessed for the expression of the indicated proteins. (g) PDGF stimulation results in the increased tyrosine phosphorylation of ACK1. U87 cells were treated with or without PDGF-BB (50 ng/ml) for 12 hr. Whole-cell lysates were subjected to immunoprecipitation (IP) and immunoblotting (IB) using the indicated antibodies. Rabbit IgG was used as a negative IP control. (h) PDGF stimulation results in the formation of PDGFR-b/ACK1 complexes. U87 and U251 cells were treated with or without PDGF-BB (50 ng/ml) for 12 hr. Whole-cell lysates were subjected to immunoprecipitation (IP) and immunoblotting (IB) using the indicated antibodies. Rabbit IgG was used as a negative IP control.

activation30) in U87 cells upon stimulation with PDGF (Fig. 3a). To determine whether PDK1 is involved in the ACK1mediated AKT activation induced by PDGF, we transfected U87 cells with PDK1 shRNA (Fig. 3b). Interestingly, PDK1

depletion in U87 cells abrogated the binding of ACK1 to PDGFR and ACK1 tyrosine phosphorylation as well as the interaction between ACK1 and AKT; the increased AKT S473 phosphorylation resulted from PDGF administration

C 2014 The Authors. Published by Wiley Periodicals, Inc. on behalf of UICC. Int. J. Cancer: 136, 1769–1780 (2015) V

1774

Carcinogenesis

PDGFR-b-activated ACK1-AKT promotes glioma tumorigenesis

Figure 2. The ACK1 Y635 residue mediates the PDGF-induced activation of AKT. (a) The ACK1 Y635 residue is important for the ACK1 tyrosine phosphorylation induced by PDGFR-b activation. U87 cells were transfected with or without the plasmids as indicated. Next, the cells were treated with or without PDGF-BB (50 ng/ml) for 12 hr in the presence or absence of the PDGFR inhibitor AG1296 (10 mM). Whole-cell lysates were subjected to immunoprecipitation (IP) and immunoblotting (IB) using the indicated antibodies. Rabbit IgG was used as a negative IP control. (b) The ACK1 Y635 residue is not required for the EGF-induced tyrosine phosphorylation of ACK1. U87 cells were transfected with or without the plasmids as indicated. The cells were then treated with or without EGF (100 ng/ml) for 12 hr. Whole-cell lysates were subjected to immunoprecipitation (IP) and immunoblotting (IB) using the indicated antibodies. Rabbit IgG was used as a negative IP control. (c) In vitro kinase assays were performed by mixing purified GST-PDGFR-b with WT or Y635F recombinant ACK1 in the presence of ATP. (d) The ACK1 Y635 residue is required for the ACK1-mediated activation of AKT. U87 cells were transfected with or without the plasmids as indicated. Next, the cells were treated with or without PDGF-BB (50 ng/ml) for 12 hr in the presence or absence of the PDGFR inhibitor AG1296 (10 mM). Whole-cell lysates were subjected to immunoprecipitation (IP) and immunoblotting (IB) using the indicated antibodies. Rabbit IgG was used as a negative IP control. (e) The expression of the ACK1 Y635F mutant abolishes the PDGF-induced tyrosine phosphorylation of AKT. U87 cells were transfected with or without plasmids as indicated. The cells were then treated with or without PDGF-BB (50 ng/ml) for 12 hr in the presence or absence of the PDGFR inhibitor AG1296 (10 mM). Whole-cell lysates were subjected to immunoprecipitation (IP) and immunoblotting (IB) using the indicated antibodies. Rabbit IgG was used as a negative IP control. (f) The expression of the ACK1 Y635F mutant abolishes the ability of GST-AKT to bind to ACK1 during PDGFR-b activation. U87 cells were transfected with or without plasmids as indicated. The cells were then were treated with or without PDGF-BB (50 ng/ml) for 12 hr in the presence or absence of the PDGFR inhibitor AG1296 (10 mM). Whole-cell lysates were subjected to a GST pull-down assay.

(Fig. 3c), suggesting the important role of PDK1 in PDGFACK1-AKT signaling. The GST pull-down assay consistently revealed that the binding of GST-AKT to WT ACK1 but not ACK1 Y635F was blocked by PDK1 knockdown (Fig. 3d). To further investigate the relationship between ACK1 and

PDK1, we used the Scansite program to analyze the amino acid sequence of ACK1. We uncovered a potential PDK1binding motif, 318-VTLWEMFTYGQEPWI-332, in which 325T constitutes a critical binding site. As shown in Figure 3e, the interaction between PDK1 and AKT induced by

C 2014 The Authors. Published by Wiley Periodicals, Inc. on behalf of UICC. Int. J. Cancer: 136, 1769–1780 (2015) V

Figure 3. PDK1 is required for the PDGFR-ACK1-AKT signaling cascade. (a) PDGFR activation induces the interaction between PDK1 and ACK1. U87 cells were treated with or without PDGF-BB (50 ng/ml) for 12 hr. Whole-cell lysates were subjected to immunoprecipitation (IP) and immunoblotting (IB) using the indicated antibodies. Rabbit IgG was used as a negative IP control. (b) PDK1 shRNA knockdown efficiency. U87 cells were transfected with scramble shRNA plasmids (control) or with the PDK1 shRNA plasmid. Whole-cell lysates were assessed for the expression of the indicated proteins. (c) PDK1 is required for the interaction of PDGFR-b and ACK1. U87 cells were transfected with or without the plasmids as indicated. The cells were then were treated with or without PDGF-BB (50 ng/ml) for 12 hr. Wholecell lysates were subjected to immunoprecipitation (IP) and immunoblotting (IB) using the indicated antibodies. Rabbit IgG was used as a negative IP control. (d) PDK1 depletion abrogates the ability of GST-AKT to bind to ACK1 under PDGFR activation. U87 cells were transfected with or without plasmids as indicated. The cells were then treated with or without PDGF-BB (50 ng/ml) for 12 hr. Whole-cell lysates were then subjected to a GST pull-down assay. (e) The PDK1 T325 residue is required for the binding of ACK1 to PDK1. U87 cells were transfected with or without plasmids as indicated. The cells were then treated with or without PDGF-BB (50 ng/ml) for 12 hr. Whole-cell lysates were subjected to immunoprecipitation (IP) and immunoblotting (IB) using the indicated antibodies. Rabbit IgG was used as a negative IP control. (f) The expression of the PDK1 T325A mutant abrogates the ability of GST-AKT to bind to ACK1 during PDGFR-b activation. U87 cells were transfected with or without the plasmids as indicated. The cells were then were treated with or without PDGF-BB (50 ng/ml) for 12 hr. Whole-cell lysates were then subjected to a GST pull-down assay.

PDGF was significantly impaired in cells expressing the FlagACK1 T325A mutant but not the T750A or T899A mutants. The amount of ACK1/AKT complexes and AKT S473 phosphorylation was also reduced by the expression of the ACK1 T325A mutant (Fig. 3e). Additionally, a GST-pull down assay indicated that GST-AKT could only bind to WT ACK1 but not ACK1 T325A after PDGF treatment in U87 cells (Fig. 3f). These results suggest that PDK1 is responsible for the sequential activation of AKT by regulating the upstream PDGFR-ACK1 signaling axis. ACK1 is critical for downstream signaling of PDGF-PDGFR

Because ACK1 is important for the PDGF-induced activation of AKT, we next investigated how ACK1 affects downstream signaling through AKT. b-catenin and cyclin D1 can be positively regulated by AKT signaling and in turn regulate the G1–S phase transition during cell cycle progression.31,32 The expression of the ACK1 Y635F mutant (higher than

endogenous ACK1 levels) but not the ACK1 Y859F mutant (higher than endogenous ACK1 levels) dramatically blocked the PDGF-induced upregulation of cyclin D1 protein levels and nuclear levels of b-catenin in U87 and U251 cells (Figs. 4a left panel and 4b left panel). Similarly, the expression of the PDK1 binding mutant (ACK1 T325A mutant) abrogated PDGF-promoted b-catenin and cyclin D1 signaling (Figs. 4a right panel and 4b right panel), supporting the critical role of PDK1 in AKT activation by PDGF. To avoid the effects of endogenous ACK1 on the related downstream molecules, ACK1 was depleted in cells with reconstitution of shRNAresistant ACK1 mutants (re-ACK1, Figs. 4c and 4d). Only reACK1 WT but not re-ACK1 Y635F could reverse the decreased b-catenin and cyclin D1 protein levels caused by ACK1 knockdown (Fig. 4c left panel). The results from realtime PCR also indicated the tyrosine 635 of ACK1 is critical for the cyclin D1 expression induced by PDGF (Fig. 4c right panel). As shown in Figure 4d, the inhibitory effects of

C 2014 The Authors. Published by Wiley Periodicals, Inc. on behalf of UICC. Int. J. Cancer: 136, 1769–1780 (2015) V

Carcinogenesis

1775

Zhang et al.

1776

Carcinogenesis

PDGFR-b-activated ACK1-AKT promotes glioma tumorigenesis

Figure 4. ACK1 regulates the effects of PDGFR-b activation on cell-cycle progression. (a) U87 cells were transfected with or without plasmids as indicated (left and right panels, respectively). The cells were then treated with or without PDGF-BB (50 ng/ml) for 12 hr in the presence or absence of the AKT inhibitor VIII (10 lM). Whole-cell lysates were assessed for the expression of the indicated proteins. (b) U251 cells were transfected with or without the plasmids as indicated (left and right panels, respectively). The cells were then treated with or without PDGF-BB (50 ng/ml) for 12 hr in the presence or absence of the AKT inhibitor VIII (10 lM). Whole-cell lysates were assessed for the expression of the indicated proteins. (c) U87 cells were transfected with or without ACK1 shRNA and reconstituted with shRNA-resistant plasmids as indicated. The cells were then treated with or without PDGF-BB (50 ng/ml) for 12 hr. Whole-cell lysates were assessed for the expression of the indicated proteins. The extracted RNA was assessed for cyclin D1 relative mRNA levels by real-time PCR. Student’s t-test (two-tailed, unpaired) was used to compare two groups of independent samples. The data shown represent the mean values 6 standard deviation (SD) from three independent experiments. * indicates a p value

PDGFR-β-activated ACK1-AKT signaling promotes glioma tumorigenesis.

Aberrant PDGF-PDGFR signaling and its effects on downstream effectors have been implicated in glioma development. A crucial AKT regulator, ACK1 (TNK2)...
706KB Sizes 0 Downloads 6 Views