Myosin-independent stiffness sensing by fibroblasts is regulated by the viscoelasticity of flowing actin

Figures

• 

  Fig. 1. Stiffness-dependent differential traction is transmitted independently of myosin-II contractility in F-actin retrograde flow-dependent manner. a Average traction integrated over 1-μm-thick cell perimeter of WT NIH 3T3 fibroblasts as a function of different gel stiffness (red). Sample sizes, n, are denoted on top of each stiffness value. Markers with error bars: mean ± s.e.m. b Representative traction vector fields (top) and traction magnitude maps (bottom) of WT-control cells. Arrow scale: 150 Pa, 300 Pa, 500 Pa, 1500 Pa and 1500 Pa of traction for gel stiffness of 0.6 kPa, 1.3 kPa, 2.6 kPa, 6 kPa and 12.7 kPa, respectively. Scale bar: 10 μm. c Average traction integrated over cell perimeter of cells treated with 20 μM blebbistatin (BBS) as a function of a gel stiffness (grey dotted in a and red in c). d Representative traction vector fields (top) and traction maps (bottom) of BBS-treated cells. Arrow scale: 50 Pa, 75 Pa, 90 Pa, 100 Pa and 150 Pa of traction for gel stiffness of 0.6 kPa, 1.3 kPa, 2.6 kPa, 6 kPa and 12.7 kPa, respectively. Power-law curve fits (Traction = b * (Stiffness)c) were added in (a) and (c). See Supplementary Table 1 for fit parameters. e Average F-actin…

• 

  Fig. 2. Myosin-independent, stiffness-dependent differential traction depends on actin polymerization mediated by Arp2/3 and formin. a Average traction integrated over cell perimeter of cells treated with 100 μM CK666 in addition to 20 μM BBS as a function of a gel stiffness. Sample sizes, i.e., the number of cells, n, are denoted on top of each stiffness value. Markers with error bars: mean ± s.e.m. b Representative traction vector fields (top) and traction magnitude maps (bottom) of CK666- and BBS-treated cells. Arrow scale: 10 Pa, 30 Pa, 40 Pa, 50 Pa and 75 Pa of traction for gel stiffness of 0.6 kPa, 1.3 kPa, 2.6 kPa, 6 kPa and 12.7 kPa, respectively. c Average traction of cells treated with 20 μM SMIFH2 in addition to 20 μM BBS as a function of a gel stiffness. d Representative traction vector fields (top) and traction magnitude maps (bottom) of SMIFH2- and BBS-treated cells. Arrow scale: 5 Pa, 15 Pa, 20 Pa, 30 Pa and 50 Pa of traction for gel stiffness of 0.6 kPa, 1.3 kPa, 2.6 kPa, 6 kPa and 12.7 kPa, respectively. e Average traction of cells treated with 1 μM Latrunculin-A (LatA) in addition to 20 μM BBS as a function of a gel stiffness. f Representative traction vector fields…

• 

  Fig. 3. Myosin-independent F-actin retrograde flow is still stiffness-dependent in the absence of Arp2/3 but becomes negligible in the absence of formin and actin polymerization. a Average F-actin flow speed as a function of the gel stiffness of CK666-BBS treated cells (red, n=9, 10, 10, 9, 9 cells for increasing stiffness, collected from m=67937, 75571, 69670, 85238, 65224 windows). Markers with error bars: mean ± s.e.m. where the statistics was based on the number of cells. b Representative interpolated flow vectors (top) and speed maps (bottom) of SNAP-actin of CK666-BBS-treated fibroblasts on 0.6 kPa, 1.3 kPa, 2.6 kPa, 6 kPa and 12.7 kPa gel. Arrow scale: 3 μm min−1 of actin flow. c Average F-actin flow speed as a function of the gel stiffness of SMIFH2-BBS-treated cells (red, n=8, 11, 12, 14, 9 cells for increasing stiffness, collected from m=57937, 42861, 79091, 99196, 80533, 54022 windows). d Representative interpolated flow vectors (top) and speed maps (bottom) of SNAP-actin of SMIFH2-BBS-treated fibroblasts on 0.6 kPa, 1.3 kPa, 2.6 kPa, 6 kPa and 12.7 kPa gel. Arrow scale: 1 μm min−1 of actin flow. e Average F-actin flow speed as a function of the gel stiffness of LatA-BBS…

• 

  Fig. 4. A molecular clutch model assuming rigid actin flow insufficiently explains Arp2/3′s role in stiffness-dependent force-flow behaviors. a A schematic of the molecular clutch model modified with addition of actin polymerization. Actin polymerization-powered retrograde flow velocity as a function of force, vactin=vmax,actin(1−F∕Fstall,actin), was added to the myosin motor-generated actin retrograde flow velocity, vmyosin=vmax,myosin(1−F∕Fstall,myosin). Myosin-independent, stiffness-dependent traction was simulated by considering only vactin without vmyosin (see materials and methods for details). b Simulated results of traction as a function of a substrate stiffness. c Simulated results of actin flow speed as a function of a substrate stiffness. WT-control (black) was simulated using both vmyosin+vactin but BBS (orange) using only vactin. Traction data in (b) were fitted using a power law function, i.e., f=aEb where a=60.8 and b=0.38×10−10 for WT-control and a=6.650 and b=0.14×10−10 for the BBS condition. Flow speeds in (c) were fitted using a negative exponential relationship, V=a∗exp(R0E)+V0 where a=0.60, R0=−0.21, V0=0.33 for WT-control and a=0.18, R0=−0.50, V0=0.21 for the BBS…

• 

  Fig. 5. Actin-elasticity-based, actin-polymerization-powered molecular clutch model. a An overview of the new molecular clutch model that models F-actin as a viscoelastic material. b, c Magnified views of a cell leading edge before (b) and after (c) addition of an elastic actin unit at a polymerizing tip. The elasticity of an elastic actin unit, Kactin, is an meso-scale modulus and comes from the architecture and morphology of actin fiber network in addition to thickness of individual fibrils and fibers. The added actin unit leads to compressive force, Factin=kactinΔL, by the boundary conditions at the membrane tension (right) and at the clutch (left) where ΔL is the change in length of individual actin units. The membrane was assumed to be a rigid wall (see Supplementary Fig. 4 for justification). The compressive force displaces the clutch and is transmitted to the substrate as traction, FECM. The displacement and the force balance are damped by a viscous damper, η, in the actin. d, e Illustration that compares the model with high (a) vs. low (b) actin elasticity, kactin. d With high kactin, addition of a new actin unit results in high force, Factin, which leads to high clutch…

• 

  Fig. 6. F-actin elasticity decreases with inhibition of Arp2/3 and formin. a A schematic of live cell atomic force microscopy (AFM) force spectroscopy using a modified AFM microcantilever with an attached 10 μm spherical probe in WT-control (DMSO), BBS, CK666-BBS and SMIFH2-BBS conditions. b A top view brightfield image with a triangular AFM microcantilever and spread NIH-3T3 fibroblasts. c Representative force-distance curves after the contact between cell and AFM microcantilever tip in WT-control and inhibitor conditions. The force curve was fitted using Hertz contact mechanics to calculate Young’s modulus. d Young’s modulus of the cell as a function of different conditions calculated from the force-distance curve. N=20 cells for each condition. Mean ± S.D. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

• 

  Fig. 7. Both model and experiments show that actin flow speed frequency increases with the substrate stiffness. a Actin flow velocity simulated using the actin-elasticity-based clutch model as a function of time on substrates with increasing stiffness, i.e., 0.6 kPa (blue), 1.3 kPa (purple), 2.6 kPa (green), 6.0 kPa (black) and 12.7 kPa (orange). Note that more frequent unclutching events, represented by velocity drop to zero, occurs as the stiffness increases. b, c Frequency analysis of the simulated time-series of velocities with a power spectrum distribution plot (b) and 2D-histogram of frequency power as a function of stiffness (c). d, e Representative images of SNAP-SiR647 actin in 3T3 fibroblasts treated with BBS on soft (0.6 kPa) (d) and stiff (12.7 kPa) substrates (e). Right: Montage of SNAP-actin over time with color-coded flow vectors in the yellow-boxed window on the full image. f, g, h, i Time-series plots of flow velocities (f, h) and power of the flow frequency (g, i) of a cell on a 0.6 kPa gel (f, g) vs. on a 12.7 kPa gel (h, i), sampled from 1×1 μm window in d and e, respectively. j Average frequencies of hundreds of windows of multiple cells as a function of stiffness.…