Fig 1: S100A4 mediates stiffness-dependent dysregulation of both NMII-A and NMII-B expression but selectively drives central accumulation of NMII-A in response to pathophysiologic range substrate stiffness (1–25 kPa). MLFs were treated as in Figure 2. Representative immunoblots with quantification of area density of (A) NMII-A and (B) NMII-B relative to GAPDH. Data from three independent experiments.∗denotes p < 0.05 compared to WT on corresponding stiffness by t test. Representative photomicrographs (20x orig. mag) of WT versus S100A4 KO fibroblasts labeled for (C) NMII-A or (E) NMII-B. Red lines indicate where plot profile was taken. Plot profiles of (D) NMII-A and (F) NMII-B intensity in WT (top) versus S100A4 KO (bottom) fibroblasts on 25 kPa gels as shown in (C) and (E). Comparison of central to peripheral (G) NMII-A and (H) NMII-B ratio in WT versus S100A4 KO fibroblasts. Data pooled from three experiments, n >10 cells/condition. ∗denotes p < 0.05 compared to WT at each stiffness by t test.
Fig 2: Loss of S100A4 selectively locks peripheral actin in filamentous bundles. WT or S100A4 KO MLFs were plated on 25 kPa fibronectin-coated gels in the presence of SiR-actin (75 nM) and TGFβ (2 ng/ml, 24 h). A, representative photomicrographs (20x orig. mag.) of time-lapse microscopy. Yellow lines indicate the location of the mean fluorescence intensity measurements as reflected in the plot profile below the photomicrograph. Red scale bar = 75 μm. B, quantification of central:peripheral F-actin stress fiber ratio over time from (A). ∗ denotes p < 0.05 of S100A4KO + TGFβ compared to WT +TGFβ at the indicated time points using t test. C, WT or S100A4 KO fibroblasts were plated as before in the presence of SiR-actin for 4 h, which was then washed off, then incubated in TGFβ (2 ng/ml, 6 h). Mean fluorescence intensity was measured over time in peripheral (left panel) and central (right panel) regions. Inset shows mean fluorescence intensity at 105 min ∗ denotes p < 0.001 compared to WT by t test. All experiments were performed in triplicate, n >30 cells/condition per experiment.
Fig 3: Endogenous fibroblast S100A4 is necessary for cell spreading in response to pathophysiological range substrate stiffness. MLFs were plated and allowed to spread (24 h) on fibronectin-coated polyacrylamide gels of indicated stiffnesses or fibronectin-coated glass. A, representative photomicrographs (20x/0.4NA orig. mag) of WT versus S100A4 KO fibroblasts phalloidin-labeled for F-actin. B, quantification of cell area from (A). ∗ denotes p = 0.02 compared to WT 8 kPa, + denotes p < 0.001 compared to WT 25 kPa, # denotes p < 0.001 compared to WT glass; two-tailed t test. C, quantification of cell circularity (4π × area/perimeter2) from (A). ∗ + # all denote p < 0.001 compared to WT for each stiffness; two-tailed t test. Data reflect means of six independent experiments, n >10 cells/condition per experiment.
Fig 4: S100A4 protein localizes to the cytoplasm and is upregulated with increasing ECM stiffness in primary lung fibroblasts. NL HLFs were plated on fibronectin-coated polyacrylamide gels of indicated stiffnesses, or fibronectin-coated tissue culture-treated plastic. A, representative immunoblots of S100A4 and GAPDH from cell lysates. B, quantification of S100A4/GAPDH ratio from (A). Data represent mean ratio from three independent experiments. ∗denotes p < 0.05 compared to 1 kPa by one-way ANOVA. C, representative immunoblots from nuclear and cytoplasmic extracts; Lamin A/C was used as a nuclear marker and GAPDH was used as a cytoplasmic marker. There was no detection of nuclear S100A4 with increasing stiffness or addition of TGFꞵ. Each experiment was performed in triplicate.
Fig 5: Endogenous fibroblast S100A4 mediates myofibroblast transdifferentiation in response to physiologic range stiffness in a calcium-dependent manner. S100A4 KO MLFs were plated on 25 kPa gels following transfection with lentiviral vectors expressing GFP only (Control LV), S100A4 and GFP (S100A4-GFP LV), or a mutant S100A4 that cannot be activated by calcium and GFP (mut-S100A4-GFP LV), or an untransfected control with no lentivirus (No LV). A, representative photomicrographs (20x/0.4NA orig. mag.). Green represents cells that were successfully transfected. Red indicates α-SMA. Inset demonstrates α-SMA incorporated into stress fibers only in the S100A4 KO cells expressing the wild-type S100A4-GFP (third row). Red scale bar = 75 μm; inset red scale bar = 25 μm. B, quantification of the percentage of cells with α-SMA in stress fibers under the conditions of (A). Only S100A4 KO cells rescued with LV expressing wild type S100A4 demonstrated stiffness-dependent increase in α-SMA incorporation into stress fibers (red squares). ∗ denotes p < 0.05 compared to other conditions within same stiffness by one-way ANOVA/Student-Newman-Keuls test. The calcium-inactivatable mutant was unable to form stress fibers on any stiffness (red diamonds). ∗ denotes p < 0.001 by one-way ANOVA/Student-Newman-Keuls test. C, representative photomicrographs (20x/0.4NA orig. mag.) of WT versus S100A4 KO fibroblasts on 25 kPa gels in calcium-containing media (no TGFꞵ, left panel) or in medium without calcium (no TGFꞵ, right panel). Merged image demonstrates α-SMA (green), F-actin (red), and α-SMA colocalized to stress fibers (yellow) only in WT + calcium condition. Red scale bar = 75 μm; inset red scale bar = 25 μm. D, quantification of the percentage of cells with α-SMA colocalized to stress fibers in WT versus S100A4 KO fibroblasts on varying stiffness substrates ± extracellular calcium (no TGFꞵ added). ∗ denotes p = 0.005 WT vs all other groups; # denotes WT p = 0.022 vs all “no calcium” groups (ANOVA/Dunnett’s test).
Supplier Page from Abcam for Recombinant Mouse S100A4 protein (His tag N-Terminus)