Fig 2: TLR2 serves as a chemotactic receptor for SAA1 and mediates migration of HSCs(A) Schematic of the TLR2 gene knockout strategy. The Cas9/sgRNA(s) target site(s) are indicated in red and were confirmed by sequencing.(B) Sequencing result of targeted region.(C) Confirmation of TLR2 gene KO at protein level determined by western blot.(D and E) WT and TLR2-/- LX-2 cells (D) and WT and siPRHSCs (E) were transfected with reporter plasmid for NF-?B. Luciferase activity was measured as fold induction in comparison to untreated control (See also STAR methods section for detail information).(F and G) Representative images showing agarose spot and Transwell migration assays of SAA1-treated WT LX-2, TLR2-/- LX-2 cells, and non-treated control.(H and I) Representative images for immunostaining of TLR2 (H) and a-SMA+ cells (I) in CU-CPT22- or VL-treated samples as shown in CCl4 and CI injury models. Bar graph represents quantification of IHC images per 5 field(s).(J and K) Immunofluorescence images showing co-localization of SAA1 and HSCs in CU-CPT22-treated and control (VL) samples in CCl4 (J) and CI injury models (K).(Scale bar represents 100 µm for CCl4 injury, 200 µm for CI injury, and 50 µm for inset). Data represent mean ± SEM *p < 0.05, **p < 0.01 and ***p < 0.0001 (n = 3). (For detailed TLR2 knockout strategy, see STAR methods section).

Fig 3: SAA1/TLR2 axis mediates increased deposition of ECM at injury sites and induces chemokines secretion in activated HSCs(A and B) Sirius red staining of ECM deposition at injury locus in CCl4 and Cl-induced models.(C) ELISA detection of MCP-1, IL-8, and RANTES secretion from LX-2 cells after treatment of the cells with rhSAA1 for 24 hr.(D) mRNA levels of MCP-1, IL-8, and RANTES.(E) Protein-level detection of MCP-1, IL-8, and RANTES. (Scale bar represents 100 µm). Data represent mean ± SEM *p < 0.05, **p < 0.01 (n = 3).

Fig 4: Hepatic SAA1 suppression in mice via siRNA(A) Experimental design representing suppression of SAA1 in CCl4 and CI injury models and timescale for sample collection.(B and C) Relative expression of liver SAA1 mRNA after 2 days of transfecting five different siRNA(s) in CCl4 and CI injury models, respectively.(D and E) IHC and western blot analyses of selected SAA1-siRNA2 in CCl4 and CI injury models, respectively.Neg-siRNA represents non-specific control. Scale bar represents 100 µm. Where applicable, data represent mean ± SEM *p < 0.05, **p < 0.01 and ***p < 0.001 (n = 3).

Fig 5: SAA1/TLR2 axis induces Rac GTPase-mediated actin reorganization and migration of HSCs(A) Time course of GTPase-Rac1 activation(s) in LX-2 cells at indicated time interval(B and C) Basal and stimulated levels of GTPase-Rac1 in WT and TLR2-/- cells as determined by pull-down assays(D and E) WT and TLR2-/- LX-2 cells were pretreated with NSC 23766 (50 µ?), and activation of GTP-Rac1 was determined by pull-down assay (D), and MLCK and p-MLC at Ser19 activity was determined by western blotting (E).(F and G) Migration assays showing pretreatment of the cells with NSC 23766 (50 µ?) attenuated their migration(s) in agarose spot (F) and Transwell (G) assays. (H) Time course of PI3K activations after treatment of LX-2 cells at indicated time points.(I and J) WT and TLR2-/-cells were pretreated with LY294002 (10 µ?), and the phosphorylation of PI3K at p85 subunits was determined by western blot analysis, and the activations of Rac GTPase were determined by pull-down assay.(K and L) Migration assays showing pretreatment of the cells with LY294002 attenuated their migration(s) in agarose spot (K) and Transwell (L) assays. Photographs are representatives of (n = 6). (Scale bar represents 200 µm).Data represent mean ± SEM *p < 0.01 and **p < 0.001 (n = 3). Photographs are representatives of (n = 6). (Scale bar represents 200 µm). Data represent mean ± SEM *p < 0.01 and **p < 0.001 (n = 3).