Figure 2.

Both branches of WNT signaling can be induced in HGSC cells, but only WNT/PCP pathway is important for acquisition of metastatic traits. A) Schematic of the WNT signaling pathways and their outcomes including WB analysis used to assess the activation of WNT canonical (left part, in blue) vs. WNT non-canonical/PCP signaling (right part, in red). Active WNT signaling results in phosphorylation-induced electrophoretic shift of DVL proteins. Phosphorylation of LRP6 on S1490 and electrophoretic shift of ROR receptors, respectively, differentiates between activation of canonical (WNT/β-catenin) by WNT3A and non-canonical (WNT/PCP) signaling by WNT5A. B) WNT3A activates WNT/β-catenin signaling, while WNT5A activates WNT/PCP signaling in Kuramochi cell line. C) Kuramochi cells are able to activate transcription of TOPFLASH luciferase reporter, which is dependent on full activation of WNT/β-catenin pathway, upon stimulation by WNT3A CM. D) Group A ascites activates both canonical and non-canonical WNT signaling, while group B ascites do not activate either of the pathways. E) WNT5A (but not WNT3A) increases the self-renewal potential of single cells in sphere forming assay. F-G) WNT5A induces migration potential (F) as well as the invasion capacity of HGSC cells in a transwell assay (G). Results are expressed as sphere, migration or invasion index, which is a value normalized to the mean of triplicates of the DMSO control in each experiment. Representative images of transwell membranes are shown on the right side of the corresponding graph, scale bar = 100 μm. Data information: In (C, E-G), data is presented as mean ± SD. ** = P ≤ 0.01; *** = P ≤ 0.001. **** = P < 0.0001 (Tukey post-hoc test of one-way ANOVA). Experiments were done in triplicate, N = 3. Quantification of WBs is available in Supplementary Fig. S1A' and S1B' and S2A-F.