Figure 5: Inflammatory signalling is required for stress-induced lymphatic remodelling.

(a) qRT–PCR analysis of COX2 gene expression in MDA-MB-231 primary tumours of control versus stressed mice. Expression normalized to RPL30 (n=5). (b) Relationship between COX2 and VEGFC expression in MDA-MB-231 tumours of control or stressed mice (n=5). Pearson coefficient 0.820 (P=0.004). (c) Representative images and (d) quantification of LVD (LYVE-1⁺, green; nuclear, blue) immunostaining of MDA-MB-231 primary tumours and bioluminescence imaging of metastasis in vehicle- or celecoxib-treated mice (n=5). Scale bar, 200 μm. (e) VEGFC protein levels in MDA-MB-231 tumour cells in response to PGE2, measured by ELISA (n=5). (f) qRT–PCR analysis of Cox2 (Ptgs2) expression in mouse primary macrophages (MØ) treated with isoproterenol (Iso)±propranolol (Prop) (n=3). (g) PGE2 production in mouse primary macrophages treated with vehicle or isoproterenol in vitro, measured by ELISA (n=3). (h) Cox2 (Ptgs2) expression in CD11b⁺F4/80⁺ TAMs isolated from 66cl4 primary mammary tumours from control or stressed mice (n=5). (i) qRT–PCR analysis of tumour Vegfc gene expression and representative in vivo images of lung and LN metastasis in control or stressed mice with 66cl4 primary tumours (not shown, fourth mammary fat pad) 28 days post tumour cell injection treated with GW2580 or vehicle (n=5). All data represent mean±s.e. *P<0.05, **P<0.01 and ***P<0.001 by two-way analysis of variance (post hoc Tukey’s adjustment) or Student’s t-test.