Fig 2: EP4 is essential for PGE2-induced pancreatic cancer cell migration. (a) Immunodetection of EP4 protein expression in six human pancreatic cancer cell lines and a murine pancreatic cancer cell line, Pan02. GAPDH was used as a loading marker. (b) RT-PCR analysis of mRNA levels of EP4 in the above seven pancreatic cancer cell lines. ACTB was used as a loading marker. (c) Cell migration analysis of Pan02 and BxPC-3 cells treated with indicating concentrations of PGE2 for 16 h via transwell assay. Scale bar, 200 µm. (d,e) Quantification of cell migration of Pan02 (d) and BxPC-3 (e) cells treated with indicating concentrations of PGE2 as in (c) (n = 3). (f) Cell migration analysis of Pan02 and BxPC-3 cells transfected with a scramble shRNA (Control) or shRNAs targeting EP4 (EP4-KD), followed by a 24-h treatment with DMSO or 0.1 µM PGE2 via transwell assay. Scale bar, 200 µm. (g,h) Quantification of cell migration of Pan02 (g) and BxPC-3 (h) cells in (f) (n = 4). The p value was calculated by one-way ANOVA with multiple comparison test. ** indicates p < 0.01; *** indicates p < 0.001. All data are presented as mean ± SEM.

Fig 3: L001 is a potent and selective antagonist of EP4. (a) Chemical structure of L001. (b,c) Dose–response curve of L001 (b) or E7046 (c) against EP4 in HEK293 cells in CRE reporter assay (n = 3). (d) Schematic diagram of TANGO assay. (e) Dose–response curve of L001 against EP4 in CHO-K1 cells via TANGO assay (n = 3). (f) Immunodetection of ERK1/2 phosphorylation. Serum-starved cells were pre-incubated with indicating concentrations of L001 or E7046 for 20 min and then stimulated with DMSO or 30 nM PGE2 subsequently for 5 min. Fetal bovine serum (FBS) was used as a positive control. GAPDH was used as a loading marker. (g) Dose–response curves of L001 on human EP1, EP2, EP3 and EP4 receptors in calcium flux assay (n = 3). (h–j) Dose–response curves of L001 on monkey (h), mouse (i) and rat (j) EP4 receptors in calcium flux assay (n = 3). All data are presented as mean ± SEM of three independent experiments.

Fig 4: EP4 receptors were down-regulated in a cellular density-dependent manner in HCA-7 cells. Cells were cultured at low (L; 3 × 104 cells/each 6 well), middle (M; 1 × 105 cells/each 6 well), and high (H; 3 × 105 cells/each 6 well) densities. Each cellular density was then subjected to immunoblot analysis with an antibody against EP4 (A) or EGFR (B), quantitative PCR (C), the EP4 promoter-luciferase assay (D), PGE2-stimulated cAMP assay (E), or EP2 or EP3 promoter-luciferase assay (F), as described in the Materials and Methods. (A, B) The blots were stripped and re-probed with an antibody against ß-tubulin. The bar graphs represent the ratio of COX-2 to ß-tubulin as assessed with pooled densitometric data (mean ± SD) from more than three independent experiments. Data are normalized to the ratio of COX-2 to ß-tubulin at the low cellular density as 100%. (C) The RNA of HCA-7 cells at each cellular density was isolated and used for quantitative real-time PCR with primers specific for either human EP4 or GAPDH, as described in the Materials and Methods. Data were analyzed by comparative methods relative to the expression of GAPDH and were normalized to the results obtained at the low cellular density as 100%. Data are the mean ± SD from three independent experiments. (D, F) Each cellular density was transiently transfected with human EP4 receptor promoter- (D), human EP2 promoter-, or EP3 promoter- (F) responsive luciferase reporter genes and control pRL-CMV or pGL3-CMV plasmids, and luciferase activity was determined as described in the Materials and Methods. Data were analyzed by comparative methods relative to the results of control luciferase activity levels, and were normalized to the results obtained at the low cellular density as 100%. Data are the mean ± SD from three independent experiments. (E) Each cellular density was treated with the indicated concentrations of PGE2 for 60 min. cAMP formation was determined as described in the Materials and Methods. Data are the means ± SEM of three independent experiments, with each being performed in duplicate. Data were normalized to the vehicle-treated control at the low cellular density as 100%. *P, †P < 0.05, ANOVA, significantly different from the vehicle-stimulated HCA-7 cells at the low cellular density. ‡P < 0.05, ANOVA, significantly different from PGE2-stimulated HCA-7 cells at the low cellular density (the analyzed results were shown for 1 µmol/L PGE2 only).

Fig 5: Effects of EP4 knock-down on migration and tube formation by RMLEC. a Real-time qPCR of EP4 mRNA expression (normalized to GAPDH) in EP4 knock-down RMLEC. b Cellular migration of EP4 KD RMLEC in response to PGE2 (20µM), EP4 agonist PGE1OH (20µM), and 24h C3L5-CM. c Tube formation by EP4 KD RMLEC on GFR-Matrigel at 24 h in the presence of the stimulating agents at same concentrations used for migration assay. Bars equal 100 µm. d Quantification of the number of branching points in the captured area at 10 different FOV for each experimental condition as determined by imageJ. Error bar represents mean ± SEM of three different experiments. (*) and (**) represent p < 0.05 and p < 0.001 respectively