Inhibition of EP2/EP4 signaling abrogates IGF-1R-mediated cancer cell growth: involvement of protein kinase C-θ activation.

Associations between growth factor receptor-mediated cell signaling and cancer cell growth have been previously characterized. Receptors for prostaglandin E2, such as EP2, and EP4, play roles in cancer growth, progression and invasion. Thus, we examined the interactions between EP2/EP4- and IGF-1R-mediated cellular signaling in human pancreatic cancer cells. Selective antagonists against EP2 and EP4 abrogated IGF-1-stimulated cell growth and suppressed MEK/ERK phosphorylation. In subsequent experiments, phospho-antibody arrays indicated increased phosphorylation levels of protein kinase C-θ (PKC-θ) at the Thr538 position following the inhibition of EP2/EP4-mediated signaling. Inhibition of PKC-θ activity impaired cell viability compared with EP2/EP4-antagonized IGF-1-stimulated cells. PKC-θ kinase MAP4K3, which plays a pivotal role in PKC-θ activation, also affected growth signaling in the presence of EP2/EP4 antagonists. Administration of EP2 and EP4 antagonists significantly inhibited the growth of an orthotopic xenograft of IGF-1-secreting pancreatic cancer cells, with increased phospho-PKC-θ and decreased phospho-ERK. Clinico-pathological analyses showed that 17.4% of surgical pancreatic cancer specimens were quadruple-positive for IGF-1R, EP2 (or EP4), MAP4K3, and PKC-θ. These results indicate a novel signaling crosstalk between EP2/EP4 and IGF-1R in cancer cells, and suggest that the MAP4K3-PKC-θ axis is central and could be exploited as a molecular target for cancer therapy.


Effect of AH6809/GW627368X on the activation of MEK and ERK in BxPC-3 cells
BxPC-3 cells (3 × 10 5 cells/well) were plated onto six-well plates, preincubated overnight at 37°C, and then starved for 24 h in 1.5 mL of serum-free medium.
Subsequently, the cells were treated with AH6809 and GW627368X (0.5 and 5 M, respectively) for 3 h at 37°C. After treatment, the cells were subjected to immunoblotting for phospho-MEK, MEK, phospho-ERK, ERK, and actin.

EP4 knockdown studies
BxPC-3 cells (3 × 10 5 cells/well) were seeded onto six-well plates and preincubated overnight at 37°C. The following day, the cells were transfected with negative universal control siRNA and EP4 siRNA (ID SASI_Hs01_00105507; Sigma) using Lipofectamine RNAiMAX, according to the manufacturer's protocol. After transfection, the cells (5 × 10 3 or 3 × 10 5 cells/well) were re-plated onto 96-well microplates or six-well plates. After preincubation overnight, the cells were starved for 24 h in serum-free medium and then stimulated with IGF-1 for 48 h or 20 min. Cell growth and the phosphorylation of MEK and ERK were determined by MTT assays and immunoblotting, respectively.

IGF-1-induced cellular growth and signaling
The effect of BX912 on cell viability was determined by MTT assays. BxPC-3 cells (5 × 10 3 cells/well) were plated onto 96-well microplates, preincubated overnight at 37°C, and then treated with 1, 10, or 100 nM BX912 (SYNkinase, San Diego, CA) for 48 h in serum-free medium at 37°C. After treatment, the viable cells were counted using the MTT method. The effect of BX912 on the AH6809/GW627368X-mediated abrogation of IGF-1-induced cellular growth and signaling was determined based on growth stimulation assays and immunoblotting. Briefly, BxPC-3 cells (5 × 10 3 or 3 × 10 5 cells/well) were plated onto 96-well microplates or six-well plates. After preincubation overnight, the cells were starved for 24 h in serum-free medium and then stimulated with IGF-1 for 48 h or 20 min in the absence or presence of AH6809/GW627368X and BX912 (100 nM) pretreatment for 3 h. Cell growth and the phosphorylation of PKC-, MEK, and ERK were then determined by MTT assays and immunoblotting, respectively.

Establishment of stable transfectants expressing hmIGF-1
Total RNA was isolated from human Ewing sarcoma SK-ES-1 cells and cDNA that contained the complete coding sequence of human mature IGF-1 (hmIGF-1) was synthesized and amplified by PCR, as previously described [47]. The PCR products were digested with EcoRI and BamHI, and then subcloned into pFUSE-hFc2 (IL2ss) vectors (Invivogen, San Diego, CA). Secondary PCR was then performed using this construct as a template. The PCR products were then digested with AgeI and NotI, and the products were subcloned into pIRESneo3 vectors (Clontech, Mountain View, CA).

In vivo characteristics of BxPC-hmIGF-1
In vivo growth assays were performed in an orthotopic nude mouse xenograft model.
The mice were anesthetized with ketamine/xylazine and BxPC-mock or BxPC-hmIGF1 cells (1 × 10 6 cells/mouse) were injected intrapancreatically (n = 5 in the BxPC-mock group and n = 5 in the BxPC-hmIGF1 group). After 42 days of observation, the mice were euthanized and the tumor lesions were collected and weighed. The tumors were then fixed in 10% phosphate-buffered formaldehyde for the histological analyses.
Whole blood was also collected from all the mice and sera were obtained. hIGF ELISA was performed to assess the persistence of hmIGF-1 expression. Formaldehyde-fixed tissues were embedded in paraffin and sectioned at 4 m. All of the sections were subjected to H&E staining and immunohistochemical staining for IGF-1 and Ki-67.
Quantitative analyses were microscopically performed by counting the number of Ki-67-positive cells per field.