(A) Expression of PGE₂ biosynthetic enzymes and PGE receptors in mPGES-1-KD and control (mock) cells was assessed by immunoblotting (left-hand panel). Equal amounts of cell lysates (20 μg of protein equivalents) were separated by SDS/PAGE and analysed by Western blotting at the same time with the corresponding antibodies to allow for a direct comparison. Representative results of at least three experiments are shown. Expression levels of EP2 and EP4 mRNAs in KD and mock cells were evaluated by quantitative RT–PCR, with GAPDH mRNA used for normalization (right-hand panel). (B–D) Effects of mPGES-1 silencing and 10 nM NS-398, a COX-2 selective inhibitor, on production of PGE₂ (B), 6-keto PGF_1α and PGF_2α (C), and on cell proliferation (D). mPGES-1-KD and mock cells were seeded at 6×10⁴ cells/well in six-well culture dishes in the presence or absence of 10 nM NS-398. After culture for 3 days, the cells were collected and counted in a Bright-line haemocytometer in the presence of Trypan Blue, and the supernatants were taken for enzyme immunoassay of several PGs. (E) Effect of dmPGE₂ on cell proliferation. mPGES-1-KD and mock cells were seeded in T-25 culture flasks at 1.5×10⁵ cells/flask in culture medium in the presence or absence of 1 μM dmPGE₂. After culture for 3 days, the cells were collected and counted. Values are the means±S.E.M. Similar results were obtained in three independent cell lines of mPGES-1-KD and control cells.

(A) Expression of COX-2 and mPGES-1 in mPGES-1-overexpressing and mock cells was assessed by Western blot analysis. Representative results of at least three experiments are shown. (B and C) Effects of mPGES-1 overexpression on PGE₂ generation (B) and cell growth (C). mPGES-1-overexpressing and mock cells were seeded at 6×10⁴ cells/well in six-well culture dishes. After culture for 2 days, the cells were collected and counted, and the supernatants were taken for a PGE₂ enzyme immunoassay. Values are the means±S.E.M. of five independent experiments. (D) mPGES-1-overexpressing and mock cells (1.2×10⁶ cells in the presence or absence of 10 nM NS-398) that had invaded through BD BioCoat™ Matrigel™ Invasion Chamber inserts over 16 h were counted. Values are the means±S.E.M. of five independent experiments. Similar results were obtained in two independent cell lines of mPGES-1-overexpressing and mock cells.

A total of 10⁵ cells were injected into the subcutaneous tissue of female mPGES-1-KO and WT mice. (A) Tumour volumes were scored on the indicated days as described in the Experimental section. (B) On day 14 after implantation, the tumour tissues were dissected and weighed (left-hand panel). Amounts of PGE₂ and PGF_2α in homogenates of the tumour tissues were quantified by enzyme immunoassay (right-hand panel). Results are presented as the means±S.E.M. (n=9). (C) Tumour tissues were cut out, fixed in formalin, and stained with silver, haematoxylin and eosin for histochemical analysis. Microvessels and necrotic regions are shown by arrowheads and an asterisk respectively. (D) Vascular density (left-hand panel) and haemoglobin content (right-hand panel) were determined as described in the Experimental section. Results are presented as the means±S.E.M. (n=9). HPF, high-power field. (E) Expression levels of VEGF protein (upper panel) and mRNA (lower panel) in the tumour tissues were determined by Western blot and RT–PCR analysis respectively. Representative results of six independent experiments are shown. On Western blot analysis, the expression level of VEGF protein was quantified by a densitometer, with the expression level of α-tubulin used for normalization (means±S.D., n=6) (right-hand panel).

(A) mPGES-1-KD and mock cells [1.2×10⁶ cells in DMEM containing 3% (v/v) FCS] were seeded on to the upper wells of BD BioCoat™ Matrigel™ Invasion Chambers. DMEM containing 10% (v/v) FCS was added as a source of chemoattractants into the bottom wells of the plates. After 16 h of incubation, cells that had invaded on to the lower surface of the chambers were fixed, stained with Crystal Violet and counted (left-hand panel). Values are the means±S.E.M. of five independent experiments. Representative photographs of mPGES-1-KD and mock cells that invaded across the Matrigel™-coated inserts are shown (right-hand panel). (B) The expression of VEGF protein in mPGES-1-KD and mock cells was assessed by immunoblotting (upper panel). The cell lysates (20 μg of protein equivalents) were separated by SDS/PAGE and then subjected to Western blot analysis using anti-VEGF and anti-α-tubulin antibodies. The activities of MMP-2 and -9 in mPGES-1-KD and mock cells were assessed by gelatin zymography (lower panel). Lysates of cells (27 μg of protein equivalents) were subjected to SDS/PAGE containing 1 mg/ml gelatin. Following electrophoresis, the gels were incubated in a bath containing 1 μM ZnCl₂ at 37 °C for 18 h. The gels were then stained with 0.1% Coomassie Blue. The expression levels of mRNAs for VEGF, MMP-2 and MMP-9 were evaluated by quantitative RT–PCR, with GAPDH mRNA used for normalization (means±S.D.; n=3) (right-hand panel). (C) Adhesion of mPGES-1-KD and mock cells to ECM proteins. mPGES-1-KD or mock cells (2×10⁵ cells) were seeded on to 35-mm dishes coated with collagen, fibronectin or laminin in culture medium in the presence or absence of 1 μM dmPGE₂. After allowing cells to attach for 60 min at 37 °C, non-adherent cells were removed by washing. Adherent cells were fixed and stained with Giemsa solution and counted in three fields at ×40 magnification using a microscope and J image software. Values are the means± S.E.M. of five independent experiments. (D) Expression of α5 and β1 integrins in mPGES-1-KD and mock cells. Total RNA was isolated from mPGES-1-KD and mock cells and subjected to quantitative RT–PCR using specific primers of α5 (left-hand panel) and β1 (right-hand panel) integrins and those of GAPDH as a reference. The results of quantitative RT–PCR were normalized with their expression in mock cells being regarded as 1 (means±S.D., n=3). Similar results were obtained in two independent cell lines of mPGES-1-KD and control cells.

(A–C) mPGES-1-KD or mock cells (10⁶ cells) were injected into the subcutaneous tissue of female BALB/c mice. On day 14 after implantation, tumour volume was scored according to the formula V=(L×W²)×0.5 (A, left-hand panel). The tumour tissues were photographed at ×100 magnification (B), dissected and weighed (A, right-hand panel). Values are means± S.E.M. (n=9). (C) Amounts of PGE₂, PGF_2α and 6-keto PGF_1α in homogenates of the tumour tissues were quantified by enzyme immunoassay (means±S.E.M., n=9). (D) Expression of VEGF in tumour xenografts of mPGES-1-KD and mock cells was assessed by RT–PCR followed by electrophoresis (left-hand panel) and quantitative RT–PCR (right-hand panel), with the expression of GAPDH as a reference (means±S.D., n=4).

A total of 10⁵ cells were injected into the lateral tail veins of female mPGES-1-KO and WT mice on the BALB/c background. On day 14 after implantation, the mice were killed, and the removed lungs were weighed (B, left-hand panel) and then stained with Bouin's solution (A). Amounts of PGE₂ in homogenates of the metastasized lung tissues were measured by enzyme immunoassay (B, right-hand panel). (C) Expression of VEGF and GAPDH mRNAs were assessed by RT–PCR. (D) Activities of MMP-2 and -9 were evaluated by gelatin zymography. Results are presented as the means±S.E.M. (n=4) in (B) and representative results of four independent experiments are shown in (A, C and D). (E) The expression levels of mRNAs for VEGF, MMP-2 and MMP-9 were evaluated by quantitative RT–PCR, with GAPDH mRNA used for normalization (means±S.D., n=4).