Average displacement of SFs for all cells examined over time in response to locally applied forces. (A) The amount of SF movement is proportional to the applied force and occurs within 20 seconds. Displacement occurring in cells exposed to 0 nN force represents normal remodelling dynamics of SFs. A further control was performed on cells fixed with paraformaldehyde. Any displacement in this control can be attributed to microscope drift or error within the tracking method. (B) In some experiments cells were pre-treated with 10 μM nocodazole and either left unstimulated or exposed to a 20 nN force. The data for a cell experiencing a 0 or 20 nN applied force, without treatment with nocodazole, is re-plotted from A (squares). Following treatment with nocodazole, cells experiencing a 20 nN force only exhibited a minor increase in displacement compared with cells experiencing no force (with or without nocodazole pre-treatment; circles). This indicates the importance of an intact microtubule cytoskeleton for the transmission of force. All values are means ± s.e.m., n=10 cells for 5–20 nN, n=5 cells for 0 nN, n=3 cells for the PFA control, n=4 for nocodazole-treated cells at 20 nN.

Spatial heat maps of SF displacement in representative NIH3T3 cells experiencing 0 nN or 20 nN of applied force. The magnitude of the displacement vectors were plotted as spatial heat maps. It is clear that far more displacement occurs in response to 20 nN than to 0 nN. Displacement of SFs generally takes place immediately (within 20 seconds) and near to the point of force application (the contact point of the AFM cantilever is marked with a white ‘x’ and is over the centre of the nucleus). At later times retraction sometimes takes place at distant cell edges (red areas at 120 and 240 seconds after the application of 20 nN). Scale bar: 15 μm.

Spatial heat maps of SF displacement over time in representative NIH3T3 cells experiencing 0 nN or 20 nN of applied force after treatment with 10 μM nocodazole. Spatial heat maps were constructed by plotting the magnitude of displacement vectors over time. It is clear that the SF displacements that occur in response to 20 nN without drug treatment is largely eliminated following treatment with nocodazole, as the heat map for 20 nN strongly resembles that for 0 nN. As before, the displacement that does occur is apparent as both a slight response near to the point of force (the contact point of the AFM cantilever is marked with a white ‘x’ and is over the centre of the nucleus) and as a retraction along the edges of the cell. Moreover, although the cells have a lower cortical elasticity after nocodazole treatment (supplementary material Fig. S3) no isotropic deformation is observed near the point of contact. Note that the displacement colour scale is identical to that in Fig. 3. Scale bars: 25 μm (0 nN); 15 μm (20 nN).

Membrane deformation and AFM tip indentation. In order to compare the amount of indentation that occurs during force application in untreated (A,B) and nocodazole-treated (C,D) cells we transiently expressed PH-PLCδ–EGFP (green) and co-stained with Hoechst 33342 (blue). PH-PLCδ–EGFP is found at the plasma membrane and provides visual evidence of membrane deformation and tip indentation (A–D). In each case a cell was imaged before and during indentation over the course of 240 seconds. In the ZX projections taken along the dashed line (B,D) the triangular tip indentation is clearly visible after only 20 seconds of force application (20 nN for both). Importantly, the indentation does not noticeably increase over time. (E) The distance between the top and bottom of the cell at 0 seconds and the distance from the tip apex to the bottom of the cell after 240 seconds of an applied 20 nN force was plotted (white bars, untreated cells; grey bars, nocodazole-treated cells). Nocodazole treatment does not result in a significant increase in cell height (P>0.84). Moreover, the indentation in response to a 20 nN load after 240 seconds has no dependence on nocodazole treatment (P>0.97). All values are means ± s.e.m., n=5 for control and n=3 for nocodazole-treated cells Scale bars: 15 μm.

Photobleaching and strain quantification of SFs. (A–C) Actin SFs were photobleached to produce a regular pattern of dark spots. The boxed region on A is shown at higher magnification before photobleaching in B and after photobleaching in C. (D) After a cell was chosen for measurement, linear regions of interest were laid out to produce a series of lines, as perpendicularly as possible to SF orientation and spaced every 5 μm. The cell here is expressing actin–EGFP and zyxin–mRFP and was photobleached along the dashed lines. The dashed lines are not shown at the top of the image in order to reveal the effect on intact SFs that have active EGFP in short segments along their length (note that the SF has not been physically altered). (E) In order to characterize changes in segment lengths we measured the fluorescence intensity profile along each SF (red line). We then developed a simple script (described in the text) that identified each peak in the profile and fitted a sum of approximate Heaviside functions to the data (black dashed line). The script was very successful at determining segment widths, however, in some cases, because of intensity noise, some peaks were not a good fit. In all cases the fits were checked by eye and poorly fitted peaks were removed from the analysis.

Strain dynamics along SFs in control and mechanically stimulated cells. (A) A diagram of a cell experiencing an applied force of 20 nN. The strain along each SF is shown as a coloured spot according to magnitude (the contact point of the AFM cantilever is marked with a black ‘x’ and is over the centre of the nucleus). The strain magnitude and sign is heterogeneous along the length of any given filament and fluctuates in time, indicating rapid changes in stretch and contraction dynamics. Similar results were found for control cells (data not shown). Scale bar: 20 μm. (B,C) In order to compare SFs directly, we normalized their lengths between −1 and +1 and plotted the average strain observed as a function of position along the fibre after 240 seconds (similar plots were generated after 60, 120 and 180 seconds; see supplementary material Fig. S5). The measured strain tended to fluctuate around zero. However, the standard deviation (red lines) for control cells (B) is 0.15 and mechanically stimulated cells (C) is 0.22; i.e. ~50% higher in the stimulated cells. These strain fluctuations possibly reflect an increase in actomyosin activity in response to applied force. For the control, n=25 SFs in five cells. For the 20 nN stimulation, n=32 SFs in eight cells. Values are means ± s.e.m.