Fig. 3.

TAZ drives an inflammatory gene signature and an increase in regulatory T cells (Tregs). (A) Tumours induced by in vitro treatment of shC or shTAZ cells for 48 h hours with 6 μm doxycycline, followed by mammary fat pad injection and continuous administration of 625 mg·kg⁻¹ doxycycline for 6 weeks (4 tumours of each genotype), were subjected to RNA‐seq analysis. Shown are pathways enriched in shC tumours compared to shTAZ tumours (FDR ≤ 0.05), determined by GSEA. (B) The RNA‐seq data in (A) was analysed using the CIBERSORT analytical tool [33]. Shown is the predicted percentage of Tregs in the shC vs. shTAZ tumours. (C) Representative images at 10× magnification of WT and TAZ‐KO tumours (n = 5 per group) harvested 2 weeks post injection and stained for CD3 (red), FOXP3 (green). Scale bar = 100 μm. (D) Quantification of CD3‐positive cells in WT vs. TAZ‐KO tumours. Values were compiled from entire slide scans of 5 WT and 5 TAZ‐KO tumours harvested 2 weeks post injection. Analysis was performed using the imagej software MOMENT threshold. Error bars indicate mean ± SEM from all repeats. *P < 0.05, Unpaired T‐test. (E) Quantification of Tregs (CD3/FOXP3 double positive cells) relative to the total number of CD3⁺ T cells. 15 random tiles were picked from 5 WT and 5 TAZ‐KO tumours. CD3‐positive cells and FOXP3‐positive cells were identified using the imagej MOMENT threshold. Double positive cells were counted manually, and their number was normalized to the total number of CD3‐positive cells. Error bars indicate mean ± SEM from all repeats. *P < 0.05, Unpaired T‐test. (F) Gene Set Enrichment Analysis (GSEA) of a CD8 T cell cytotoxic activity gene signature [39] in the in vivo shC vs. ShTAZ RNA‐seq data. (G) GSEA of a CD8 T cell exhaustion gene signature [39] in the in vivo shC vs. ShTAZ RNA‐seq data. FOXP3, forkhead box P3; GSEA, gene set enrichment analysis; KO, knockout; ShC, scrambled shRNA; ShTAZ, TAZ‐targeting shRNA; WT, wild type.