Figure 2. Recombinant soluble GARP and enforced soluble or membrane-bound GARP expression drives epithelial-mesenchymal cell transition (EMT) and invasion of normal mammary gland epithelial cells.

(A) Immunoblot analysis for GARP, E-cadherin, Vimentin and Smad2/3 phosphorylation in NMuMG cells stably transfected with GARP or empty vector. (B) Immunoblot analysis for GARP in NMuMG cells stably transduced with expression vector for GARP, sGARP or empty vector. (C) NMuMG cells were treated for the indicated time points and with increasing doses of soluble GARP in serum-free culture medium, followed by immunoblot for vimentin and β-actin. (D) In vitro scratch assay to indicate the difference in the gap closure at 24 hours. (E) Summary statistics of three independent scratch assays. (F) In vivo imaging of the luceferin-enhanced bioluminescence in mice after injection of GARP, sGARP and control NMuMG cells, with images from Week 4. (G) Quantification of luciferase signal in the region of interests (boxed); data were weekly quantified by average radiance (photons/s/cm²/steradian). Thin lines represent data from individual mouse, whereas the three thick curves reflect mean values of each group. (H) Week 4 H&E and IHC analysis of Vimentin and E-cadherin expression (brown color) for GARP-expressing NMuMG tumors. Scale bar: 20 μm. *P<0.05; **P<0.01. Statistical significance was determined by the Wald test (panel G) and two-tailed T-test (panel E). Western Blot images are representative of 2 independent experiments and scratch assay are representative of 3 independent experiments.