Local application of stress gives rise to cell-scale intracellular flow.

A. Schematic diagram of the experiment. Collagen-coated fluorescent beads are bound to the cell surface. An AFM cantilever is brought into contact with one side of the cell and bead movement in the z-direction is imaged over time. B. Representative image showing a combined phase contrast and fluorescence image of the beads on the cell. Top panels: left: The AFM cantilever appears as a dark shadow on the left of the image. The bead is visualized by fluorescence. When the plane of focus is moved higher than the cell, a halo of fluorescence centered on the bead appears (middle panel, right). The diameter of the halo of the bead reports on the distance between the bead and the plane of focus (see SI, Fig S1A). Variations in halo radius indicate changes in height caused by indentation. The distance between the bead and the AFM tip is indicated by a red line. Middle panels: left: Profile of a cell before indentation. A cell- impermeable fluorescent indicator has been added to the medium and the cell appears dark. The AFM cantilever was imaged by reflectance and appears bright. Right: Representative image of the halo before indentation. Bottom panels: left: Profile of the same cell as in the middle panel during indentation by an AFM cantilever. Right: Halo of the same bead as in the middle panel during indentation. Scale bar= 10μm. C. Change in bead height as a function of time for a total of 12 beads on 7 cells. Beads from the same cell appear in the same colour. Inset shows normalized displacement of three beads on the same cell located at different distances from the AFM tip. This highlights the slower response in the second phase for more distant beads. The color code in the inset is the same as the main figure. D. Characteristic relaxation time τp of the second phase for control cells (n=7 cells) and cells treated with sucrose (n=6 cells) and latrunculin (n=7 cells). In the box plot, the black line is the median, the bottom and top edges of the box indicate the 25th and 75th percentiles, respectively. The whiskers extend to the most extreme data points that are not outliers. Data points appear as black dots. E. Characteristic relaxation time τp as a function of distance from the AFM tip for control cells (black), sucrose (light grey), and latrunculin (dark grey).

Intracellular fluid propagation in response to pressure gradients.

A. Schematic diagram of the experiment. Collagen-coated fluorescent beads (yellow) are bound to the surface of a cell in fluidic communication with a micropipette. At time t=0s, pressure is increased within the pipette leading to injection of fluid into the cell (dark blue). After a time lag δt, the fluid propagation front reaches a bead in δx resulting in an increase of its height by δz. B. Top: Representative image showing a combined DIC and fluorescence image of a cell. The micropipette appears at the top of the image and a fluidic connection is established. The distance from the pipette tip to each bead is indicated by red lines. Scale bar=10μm. Bottom: Defocused image of the fluorescent beads tethered to the cell surface before (left) and after (right) propagation of the fluid flow through the cell. C. Temporal evolution of height for beads situated at different distances from the pipette tip. Data from N=12 beads from n=6 cells. The distance of each bead is listed in the inset. Beads from the same cell are in the same colour. The time at which microinjection starts is t=0 s (indicated by the arrow). The beads respond to injection with a time lag that increases with increasing bead-pipette distance. The dashed black line corresponds to a 0.1 𝜇𝑚 displacement of the beads from their initial position. This threshold is used to calculate the time lag δt between injection and bead response (see SI methods). D. Bead displacement as a function of distance from the pipette after t=2s. E. Time lag δt as a function of distance from the pipette. F. Velocity as a function of estimated pressure gradient. (D-F) Beads from the same cell appear as solid or open markers of the same colour. Colour code is indicated on the right of panel F.

The cytoplasm of metaphase cells displays a porous behavior.

A. Schematic diagram of the experiment examining the change in diameter of metaphase HeLa cells due to application of a step increase in pressure through the micropipette. To detect the arrival of fluid flow at the cell periphery following application of a step pressure in the micropipette, we monitored changes in cell diameter. An increase in cell diameter was detectable after a time delay δt that depended on the amplitude of the pressure step. B. DIC images of representative experiments. The step change in pressure was applied at t=0s. Top: equatorial plane of cells blocked in metaphase at t=0s. The micropipette tip is out of the plane of focus. Bottom: cells at t=2.6s after pressure application. The initial diameter is indicated by the red dashed circle. Scale bar=10µm. C. Temporal evolution of the relative diameter of metaphase HeLa cells subjected to different pressure steps. Error bars represent standard deviations. N=3 cells per pressure. The timing of pressure application is indicated by a vertical black line. The pressure corresponding to each curve is indicated on the graph. D. Time lag δt between pressure application and the onset of diameter increase as a function of the amplitude of the pressure step. The dashed line indicates a hyperbolic fit. Whiskers indicate the standard deviation. n=3 cells per pressure. E. Velocity as a function of estimated pressure gradient. The dashed line indicates a linear fit. Whiskers indicate the standard deviation. N=3 experiments per data point.

A poroelastic cytoplasm enables the emergence of steady state pressure gradients.

A. Schematic representation of a quasi two-dimensional slab of poroelastic material with a deformable upper surface. The material is parameterized by its elasticity E, and its poroelastic diffusion constant D. The bottom surface is uniformly tethered to an impermeable infinitely stiff material. The top surface is subjected to an indentation δ applied by a spherical indenter of radius R. The surface profile after deformation is indicated by the dashed black line. B. Schematic representation of a two-dimensional slab of poroelastic material (grey). The cytoplasmic material is parameterized by its elasticity E, its poroelastic diffusion constant D, and its pressure Pin. One part of the surface is permeable and fluid is injected into the cell through this region at a pressure Papp. The bottom surface is uniformly tethered to an impermeable infinitely stiff material. The surface profile after fluid injection is indicated by the dashed grey line. C. Experimental (grey) and predicted steady-state vertical displacement of beads tethered to the cell surface in response to localized indentation. Inset shows the predicted force relaxation as a function of time. D. Vertical displacement of beads in response to fluid microinjection after 2s as a function of distance from the micropipette. Grey data points indicate experiments and blue data points the simulation. The red line indicates the trendline. Inset shows a zoom on the further distances. E. Time lags of the onset of vertical displacement as a function of distance from the micropipette. Grey data points indicate experiments and blue data points the simulation. F. Schematic diagram of depressurization experiment. A two-dimensional slab of poroelastic material (grey) surrounded by a less permeable, thin, outer layer (yellow) representing the cortex. The cytoplasmic material is parameterized by its elasticity E2, its poroelastic diffusion constant D2, and its pressure Pin. The outer layer is parameterized by E1, D1, and Pin. One part of the surface is permeable and fluid is released from the cell through this region to atmosphere pressure Patm. At t=0s, the cell is subjected to a suction Pin through the micropipette. G. Pressure distribution immediately after depressurization (top) and at steady state (bottom). H. Intracellular pressure as a function of distance from the micropipette for a range of internal pressures Pin. I. Intracellular pressure profiles from H normalised to the pressure at x=30𝜇m from the micropipette. J. Intracellular pressure distribution as a function of distance from the micropipette for different membrane-cortex poroelastic diffusion constants D1.

Whole cell patch clamp and pressure release give rise to an outflow of fluid from the cell.

A. Differential interference contrast image of a typical experiment. The top cell is subjected to pressure release, while the bottom cell is not. Scale bar=5µm. B. Temporal evolution of CMFDA cytoplasmic fluorescence intensity in control cells and cells subjected to depressurization. The solid lines indicate the average and the whiskers the standard deviation. N=5 experiments were averaged for each condition. C. Fluorescence intensity of CMFDA in the cells in A prior to and after 575s of depressurization. Scale bar=5µm.

Cells can accommodate sustained intracellular pressure gradients.

A. Intracellular pressure in Filamin-deficient blebbing M2 melanoma cells and HeLa cells. Cells were treated with Y27632 for 30 minutes prior to measurement. Each condition represents n=34 cells from N=4 experiments. B. Schematic diagram of the pressure release experiment. At time t=0s, a fluidic communication is established between a cell and a micropipette, resulting in a small suction pressure at the tip of the pipette. This leads to the establishment of a pressure gradient within the cell. As blebs are pressure driven protrusions, they can be used as pressure gauges to report on the effect of pressure release. C. Computational prediction of the intracellular pressure profile in response to depressurization. The dashed line represents the pressure below which blebbing cannot occur based on the pressure measured in control M2 cells and cells treated with the Rho-kinase inhibitor Y27632. D. Representative pressure release experiments in M2 cells. Top row: DIC images of a representative experiment. Pressure is released at the pipette tip in a blebbing cell at t=0s and maintained constant thereafter. The location of the pipette tip is indicated by a white arrow. A second cell in the field of view serves as a negative control. Scale bar=10μm. Bottom row: Fluorescence images of the F-actin cytoskeleton in blebbing cells during a pressure release experiment. The location of the micropipette is indicated by the black arrow. Scale bar=10μm. Summary statistics over n=19 cells are presented in Fig S10. E. Pressure release experiments in interphase (top) and metaphase (bottom) HeLa cells. Pressure is locally released in the cells at t=0s through a micropipette. The presence of intracellular pressure is detected by the emergence of blebs in response to partial depolymerization of the F-actin cytoskeleton by latrunculin treatment (t=80s and t=150s). Scale bar=10μm. F. Pressure release experiments in interphase (top) and metaphase (bottom) HeLa cells treated with the Rho-kinase inhibitor Y27632. (E-F) Drugs were added at t=0+s. G. Laser ablation of the cortex of metaphase HeLa cells expressing GFP-actin. The target region for laser ablation is indicated by the red circle in the before images and by a red arrow in the after images. Control cells, cells treated with the inhibitor of contractility Y27632 for 30 minutes, and cells in which a suction was applied through a pipette are shown. (E-G) Summary statistics are presented in Fig S12B-C.

Experimental setup.

A. Principle of defocusing microscopy. Left: Collagen coated fluorescent beads are bound to the cell surface and are imaged by optical microscopy using a high magnification objective. The plane of focus is purposely set above the cell such that beads appear to have multiple concentric rings around them. The diameter of the outer ring reports on the distance between the bead and the plane of focus. The diameter of the outer ring changes linearly with changes in bead height. The top halo illustrates the rings visualised at t=t0 and the bottom halo illustrates them at t=t1 once a change in surface height has taken place. Calibration of the outer ring diameter as a function of bead distance from the plane of focus allows the detection of small movements in z3. Middle: representative intensity profile of an out-of-focus bead. The central peak denotes the position of the bead centre. Right: detection of outer peak positions. The outer peaks are detected by fitting Gaussian functions (red) to the fluorescence intensity profile of the halo. Blue data points represent the intensity of the outer halo. B. Experimental data acquisition setup. An AFM cantilever is positioned above the cell of interest just above the cell membrane. Recording is started and the AFM is brought into contact with the cell to generate an indentation with a depth of 2-4 um. During this time, the position of beads at the surface is monitored using defocusing microscopy. The digitiser simultaneously records the cantilever deflection, the height of the piezoelectric ceramic, and pulses sent by the camera each time an image is acquired. These data allow synchronisation of AFM data and imaging data. Images are recorded separately and are acquired using micromanager.

Definition of 𝝉𝒑 in AFM experiments.

Representative plot of the vertical displacement of a bead tethered to the cell surface as a function of time. The displacement consists of two phases, a first rapid and linear displacement d1 followed by a slower displacement of amplitude Δ𝑑. As previous work has shown that poroelastic relaxation is approximately exponential1, we measured the characteristic time 𝜏𝑝 . This characteristic time can be defined as the time for which the displacement of the second phase decreases by 37% (1/e) of its initial value: 𝑑(𝑡1 + 𝜏𝑝) = 𝑑1 + 0.37. Δ𝑑.

Surface movement in response to application of extrinsic force by AFM indentation.

In all panels, experiments in control conditions are shown in red, hyperosmotic conditions in black, and latrunculin treatment in yellow. The number of cells examined (n) and the number of beads examined (N). A. Bead displacement δz at steady state as a function of distance to the AFM tip δx. B. Bead displacement δz for the first fast phase of displacement. C. Bead displacement δz for the second slow phase of displacement.

Fluid injection experiments and experimental setup.

A. Principle of the fluid injection experiments. An electrophysiology setup was used with a whole-cell recording configuration. In patch-clamp electrophysiology, the tip of a micropipette (∼2μm diameter) is first approached towards the cell surface (left) and a high resistance (GΩ) seal is generated by suction of the membrane into the pipette (a configuration known as cell-attached patch clamp, middle). A short pulse of suction can then rupture the membrane patch and create a fluidic and electrical connection between the cell and the micropipette (a configuration known as whole-cell clamp, right). The response to a voltage step can be used to determine the resistance between the two electrodes and reports on the configuration at any given time (indicated as inset in the top right of each panel). B. Experimental setup. A cell is brought into fluidic communication with a glass micropipette containing medium with an ionic concentration mimicking intracellular composition. An electrode within the micropipette and within the Petri dish enables measurement of the electrical resistance of the cell-pipette assembly. Pressure can be applied to the fluid within the micropipette using a pressure reservoir and a manifold of computer controlled pinch-valves. Within the micropipette, an electrode is used to monitor the combined resistance of the glass micropipette and cell. Currents recorded by the electrode are amplified before being digitised. The digitiser simultaneously records electrical currents from the electrode in the micropipette, pressure from the pressure transducer, and pulses sent by the camera each time an image is acquired. These data allow synchronisation of electrophysiological data, pressure data, and imaging data. Images are recorded separately and are acquired using micromanager.

The length-scale of surface deformations is controlled by cell surface tension.

A. Overlay of the zx profiles of a mitotic cell before (green) and during indentation (red). The cell membrane is labelled with CellMask DeepRed. The arrowhead indicates the position of the AFM tip. Scale bar 10μm. B. Segmentation of the membrane along the top half of the cell before (green) and during (red) indentation. Membrane position is derived from segmentation of the data in A. The tip position is marked by an *. C. Difference in membrane height between pre-indentation and indentation profiles plotted in B. The tip is located at x=0. D. Schematic of the cell surface profile during indentation and the corresponding length-scale of the deformation induced by indentation. E. Measured length-scale for an indentation ∼2μm in depth for DMSO control l=1.2±0.2μm (n=8) and for blebbistatin treatment (100μM) l=0.8±0.4μm (n=9) (p= 0.016, student t-test).

A positive pressure must be applied to the pipette to generate outflow.

In all experiments, pipettes were filled with fluid up to mid-height and attached to the micromanipulator at an angle of ∼40°. The pipette medium contained a fluorescent dye and epifluorescence images were acquired by wide-field microscopy. The pipette tip was positioned at a similar height in the Petri dish as during measurements on cells. Scale bar=5µm. A. When a sufficiently large positive pressure was applied to the pipette rear, fluorophore leaked out of the pipette. B. When no pressure was applied to the rear of the pipette, no fluorophore leaked out of the pipette due to capillary forces resulting from surface tension in the pipette. The position of the pipette is indicated by the yellow dashed lines. The same pipette is shown in A and B. On average, the pressure needed to observe outflow of fluorophore from the micropipette tip was Pc=0.2kPa (N=5 pipettes).

The pipette access resistance stays constant during whole-cell patch clamp configuration.

Representative data showing the temporal evolution of the pipette access resistance as a function of time during an experiment in which a suction was applied through the pipette. The vertical axes show the measured access resistance and the horizontal axis shows time in seconds. Two different scales are used in the top and bottom graphs to allow monitoring of gigaseal formation (necessitating GΩ scale, top) and access resistance in whole cell configuration (necessitating tens of MΩ scale, bottom).

In the top graph, the micropipette internal fluid is initially just in contact with the cell, a gigaseal is formed by aspiration of a patch of cell membrane into the pipette (see Fig S4A), leading to a tight seal between the cell membrane and micropipette interior demonstrated by a large increase of the resistance observed at ∼150s (red arrow, top row). The membrane patch is then ruptured to create a fluidic connection between the cell and the micropipette (while gigaseal is maintained), leading to formation of whole cell configuration with access resistance in the order of Mega Ohms at ∼180s (red arrow, bottom row). In this configuration, a tight seal is maintained between the cell membrane and the micropipette interior.

In the bottom graph, the temporal evolution of the access resistance is monitored after formation of the fluidic connection between the cell and the micropipette and during the rest of the experiment. The timing of creation of the fluidic connection is indicated by the red arrow (∼180s). When the fluidic connection is generated, the access resistance (Ra∼13MΩ) is large compared to the pipette resistance (∼6MΩ). This access resistance reports on how easily current can flow between the cell and the micropipette and is extremely sensitive to any clogging or obstacles in the vicinity of the micropipette tip. After generation of the fluidic connection, the access resistance stays constant over the whole duration of the experiment, indicating that a tight seal is maintained between the pipette and the cell and that the fluidic connection does not get progressively clogged by cellular material or active cellular processes. Similar results were obtained for all whole-cell clamp experiments, indicating that no obstruction occurred due to cellular debris.

F-actin localisation at the interface between the cell and the micropipette during a pressure release experiment.

The cells stably expressed the F-actin reporter Life-Act-Ruby. Pressure release was applied at t=0s through the micropipette. The position of the micropipette is indicated by the yellow dashed lines. All images were acquired by epifluorescence microscopy and show the top of the cell. The distribution of F-actin at the interface between the micropipette and the cell stayed approximately constant over the duration of the pressure release experiment, consistent with the constant access resistance observed (Fig S7). Scale bar=10µm.

Cell volume remains constant during pressure release experiments.

A. Temporal evolution of relative volume derived from measurement of the radius R(t)/R(0) for 3 metaphase cells during a pressure release experiment.

Angular distribution of blebs in a blebbing melanoma cell during a pressure release experiment.

Representative data relating to the cell shown in Fig 6D, top row. The position of the micropipette is indicated. The concentric circles indicate the number of blebs at each angular position appearing over the duration of the experiments. The experiment lasted a total of 5 minutes. This data is representative of experiments on N=19 cells.

Effect of latrunculin treatment on F-actin and myosin distribution within interphase and metaphase HeLa cells.

The first two columns show DIC images of wild-type cells before (column 1) and after Latrunculin treatment (column 2). The 3rd and 4th column show epifluorescence images of cells stably expressing the F-actin reporter Life-act Ruby before (column 3) and after treatment with Latrunculin (column 4). The 5th and 6th column show epifluorescence images of cells stably expressing Myosin Regulatory Light Chain (MRLC) tagged with GFP before (column 5) and after treatment with Latrunculin (column 6). All images were acquired by wide-field microscopy. In all experiments, latrunculin was added at t=0+s. In (C-D), cells were pretreated with Y27632 for 30 minutes prior to the start of the experiment. In all panels, blebs are indicated by red arrows. Scale bars= 10um. A. Interphase HeLa cells treated with Latrunculin. Treatment with latrunculin gives rise to blebs in all cases but a well defined actomyosin cytoskeleton is still present in the cells after 3 min incubation. B. Metaphase HeLa cells treated with Latrunculin. Treatment with latrunculin gives rise to blebs in all cases but a well-defined actomyosin cytoskeleton is still present in the cells after 3 min incubation. C. Interphase HeLa cells pretreated with Y27632 to block contractility prior to Latrunculin exposure. No blebs can be observed in response to latrunculin treatment. D. Metaphase HeLa cells pretreated with Y27632 to block contractility prior to Latrunculin exposure. No blebs can be observed in response to latrunculin treatment. (B-D). Insets show an overlay of the main image with a fluorescent DNA stain.

A. Proportion of cells displaying blebs in response to latrunculin treatment with or without pretreatment with Y27632 for 30 minutes. Cells stably expressing Life-Act Ruby (LFR) or Myosin Regulatory Light Chain (MRLC) are used. No differences in response could be observed between LFR cells, MRLC cells, or WT cells (in B). B. Percentage of wild-type HeLa cells in interphase (IP) or metaphase (MP) displaying blebs in response to latrunculin treatment for different treatments with (WCR) or without pressure release. Pretreatment with Y27632 was carried out for 30 minutes prior to the start of the experiment. C. Proportion of cells displaying a bleb in response to laser ablation for different conditions. WCR indicates pressure release experiments. Pretreatment with Y27632 was carried out for 30 minutes prior to the start of the experiment. On each bar chart, N indicates the number of experimental days and n indicates the number of cells examined.