Fig 2: EP2 receptor inhibition blocks BDNF production after pilocarpine SE in mice. A, SE was induced in mice by pilocarpine (280 mg/kg, ip), allowed for 60 min, and terminated by sodium pentobarbital (30 mg/kg, ip). Mice were then treated by EP2 receptor antagonist TG6-10-1 (5 mg/kg, ip) twice daily. mRNA of BDNF in the hippocampus was measured by qPCR 1 d or 4 d after SE. The seizure-induced BDNF mRNA was decreased by TG6-10-1 treatment at 1 day (N = 5-6, P = .348 compared to vehicle treatment, unpaired t test) and 4 d (N = 8-9, P = .046 compared to vehicle treatment, unpaired t test) after SE. The chemical structure of EP2 antagonist TG6-10-1 is displayed. B, The BDNF protein in the hippocampus was measured by ELISA 4 d after pilocarpine SE in mice. TG6-10-1 treatment prevented SE-induced BDNF at the protein level (N = 8-9, *P < .05; **P < .01; ***P < .001, one-way ANOVA with post hoc Bonferroni's multiple comparisons test). C, The BDNF protein in the hippocampus was also measured by Western blot analysis. Two representative samples from each of the 4 experimental groups are shown on the blots. The immunoblots were scanned, and the band intensities were quantified using ImageJ. The relative protein levels of mature BDNF were normalized to GAPDH and then to their basal levels. TG6-10-1 treatment prevented SE-induced mature BDNF (N = 8-9, **P < .01; ***P < .001, one-way ANOVA and post hoc Bonferroni's multiple comparisons test). All data shown as mean + SEM

Fig 3: Effects of EP2 knockdown on the apical extrusion of YAP (5SA) cells. (a) Immunoblot to detect EP2 protein in normal MDCK cells and YAP (5SA) cells at 0, 12 and 24 hr after Dox treatment. GAPDH, loading control. (b,c) Quantitation of percentages of apically extruded cells in cocultures of labeled YAP (5SA) cells, or YAP (5SA) cells expressing Luciferase shRNA (shLuc; control), or EP2 shRNA (shEP2), that were mixed 1:50 with unlabeled normal MDCK cells, or MDCK cells expressing shLuc or shEP2, as indicated. Cocultures were treated for 24 hr with Dox starting at 24 hr postseeding. (c) The EP2 knockdown cocultures in (b) were treated for 24 hr with Dox plus forskolin (10 µM) or dibutyryl-cAMP (3 mM) (as indicated) starting at 24 hr postseeding. For (b) and (c), data are the mean + SD (n = 3/group) of three independent experiments

Fig 4: Effects of prostaglandin E2 and various receptor agonists or antagonists on the apical extrusion of YAP (5SA) cells. (a–d) Quantification of percentages of apically extruded cells in cocultures of labeled YAP (5SA) cells that were mixed 1:50 with unlabeled normal MDCK cells and treated with Dox plus the indicated compounds for various times starting at 24 hr postseeding. Controls were cocultures incubated for the appropriate times with/without Dox in the absence of the indicated compounds. (a) Cells were treated for 24 hr with Dox plus ONO-8711 (EP1 antagonist, 10 µM), PF-04418948 (EP2 antagonist, 10 µM), L-798106 (EP3 antagonist, 10 µM) or ONO-AE3-208 (EP4 antagonist, 10 µM). (b) Cells were treated for 16 hr with Dox plus butaprost (EP2 agonist, 10 µM). (c) Cells were treated for 24 hr with Dox and NS398 (50 µM) plus butaprost (10 µM), U-46619 (TP agonist, 100 nM), Fluprostenol (FP agonist, 10 µM), sulprostone (EP1/3 agonist, 10 µM), BW-245C (DP agonist, 10 µM) or Cicaprost (IP agonist, 10 µM). (d) Cells were treated for 24 hr with Dox and NS398 (50 µM) plus PGE2 (1 µM or 5 µM). For (a)–(d), data are the mean + SD (n = 3/group) of three independent experiments

Fig 5: mPGES-1 derived PGE2 and EP4 activation limit leukocyte-EC interactions. PGE2 levels were measured for culture supernatant of mouse (mPges-1 WT/KO) lung endothelial cells (ECs) (a) and peritoneal myeloid cells (b). Cells were treated with 10 ng/ml IL-1ß for 12 h (a) or with 100 nM C5a for 30 min (b). Effect of mPGES-1 inhibition on cell adhesion was analyzed. WT ECs were treated with 10 µM Cay10526 for 12 h before peritoneal myeloid cells were applied, and cell adhesion was quantified in absence or presence of 100 nM C5a. Representative photos of cell adhesion for each treatment condition (c) and the quantitative results (d) were shown. Adhesion of blood leukocytes to ECs was determined in vitro in presence of 100 nM C5a (e). The fold change of adherent cells over the number for WT leukocyte to WT EC was presented, with cell genotype labeled below each column. Leukocyte–endothelial cell interactions were videotaped by intravital microscopy during hindlimb I/R, and rolling cells were counted prior to and after femoral artery ligation and reperfusion at indicated time points (f), as detailed in the Methods. Effects of agonists for EP1/3 (Sulprostone), EP2 (Butaprost), and EP4 (AE1-329), each at 1 µM, on in vitro blood leukocyte adhesion to WT (g) or mPges-1 KO (h) EC monolayer were examined in presence of 100 nM C5a. Adhesion of purified blood neutrophils was similarly examined (i). One-way ANOVA with Tukey’s multiple comparison tests (a, n = 3; e, n = 29, 28, 27, 32; g, n = 23, 24, 22, 23, 23; h, n = 11, 11, 11, 11, 10; i, n = 12, 12, 11, 11, 11) or Bonferroni’s multiple comparison tests (b, n = 5, 5, 5, 8; d, n = 11, 12, 17, 14). Two-way ANOVA with Bonferroni’s tests (f, n = 6) and unpaired Student’s t test (e, n = 29, 28, 27, 32, #p = 0.045; ##p = 0.0011). Error bar indicates SEM