SFRP1 in transition state fibroblasts regulates invasion, RHOA activity and cell shape. a) Primary human fibroblasts were isolated from lungs of human patients and cultured on rigid two-dimensional surfaces. SFRP1 expression was diminished in these cells by application of three different siRNAs targeting various exons. The successful knockdown of SFRP1 protein was confirmed by Western blot analysis. b) Primary human fibroblasts were applied on top of a collagen extracellular matrix (ECM) and left to invade for 72–96 h. Cells were stained for nuclei (4′,6-diamidino-2-phenylindole (DAPI), blue) and filamentous actin (phalloidin, red). A maximum intensity projection of a z-stack from one entire well is depicted as xz-view. Scale bar=200 µm. c) Transforming growth factor (TGF)β1-treatment as well as SFRP1-depletion augmented the ECM invasion capacity of primary human lung fibroblasts. Data are presented as mean±sd. One-way ANOVA with Tukey's multiple comparison test, n=4 (fibroblasts from four different patients displayed as distinct colours). Logarithmic (log10) y-axis. d) Addition of recombinant SFRP1 (R) rescued the increased invasion of SFRP1-depleted human lung fibroblasts (siS1R−TGFβ1) as well as partially the reduced invasion of TGFβ1-treated and SFRP1-depleted fibroblasts (siS1R+TGFβ1). Sc shows pooled data from Sc1-2. SiRNA shows pooled data from siRNA1-3. One-way ANOVA with Tukey's multiple comparison test, n=1–3 (fibroblasts from up to three different patients). Logarithmic (log10) y-axis. e) Ingenuity pathway analysis (IPA) canonical pathway analysis based on bulk transcriptome data analysis of SFRP1-siRNA-depleted compared to SFRP1-expressing primary human lung fibroblasts. IPA identified RHOA-signalling as a highly affected pathway in SFRP1-siRNA-depleted fibroblasts. f) Transcript analysis by quantitative PCR confirmed the successful reduction of Sfrp1 mRNA in Sfrp1-siRNA-treated primary human lung fibroblasts. RHOA mRNA was found to be considerably reduced in SFRP1-depleted fibroblasts. Data are presented as mean±sd. Unpaired two-tailed t-test. n=3 (fibroblasts from three different patients). g) Consistent with IPA analysis shown in (e) as well as reduction of RHOA mRNA displayed in (f), RHOA-GTPase activity was significantly reduced in SFRP1-depleted primary human lung fibroblasts. Data are presented as mean±sd. Unpaired two-tailed t-test. n=4 (fibroblasts from four different patients). h) Immunofluorescence labelling of filamentous actin cytoskeleton (phalloidin in red) indicated a morphological switch towards smaller and elongated cell shapes in SFRP1-depleted primary human lung fibroblasts. Scale bar=100 µm. i) Detailed cell morphological analysis using an automatised workflow in CellProfiler software. The unbiased quantification of 1000 cells from untreated (UT), scrambled-siRNA-treated (Sc) and SFRP1-siRNA-treated (siSfrp1) human lung fibroblasts confirmed a substantial switch towards more elongated (smaller area extent, higher aspect ratio) cell morphologies in SFRP1-depleted primary human lung fibroblasts. In total we analysed up to 15 000 single cells from each condition and patient. Data are presented as mean±sd. Unpaired t-test. p=0.08, n=3 (fibroblasts from three different patients). j) RHOA-GTPase inhibition by CT04 (C3 transferase) triggered an increase in the invasive capacity of primary human lung fibroblasts. Sc shows pooled data from Sc1-2. SiRNA shows pooled data from siRNA1-3. One-way ANOVA with Tukey's multiple comparison test, n=4 (fibroblasts from four different patients). Logarithmic (log10) y-axis. *: p<0.05, **: p<0.01, ***: p<0.001, ****: p<0.0001.