How blebs and pseudopods cooperate during chemotaxis

Edited by Herbert Levine, Rice University, Houston, TX, and approved July 7, 2014 (received for review November 29, 2013)
July 29, 2014
111 (32) 11703-11708


We show that blebs—pressure-driven extensions of the plasma membrane—preferentially form in concave regions of the plasma membrane of migrating Dictyostelium cells and propose that the underlying physical reason is due to membrane tension, which provides an outward force at concavities. Blebs can transform into pseudopods by continued actin polymerization and pseudopods can trigger blebs on their flanks. Thus the two forms of protrusion can cooperate during chemotaxis.


Two motors can drive extension of the leading edge of motile cells: actin polymerization and myosin-driven contraction of the cortex, producing fluid pressure and the formation of blebs. Dictyostelium cells can move with both blebs and actin-driven pseudopods at the same time, and blebs, like pseudopods, can be orientated by chemotactic gradients. Here we ask how bleb sites are selected and how the two forms of projection cooperate. We show that membrane curvature is an important, yet overlooked, factor. Dictyostelium cells were observed moving under agarose, which efficiently induces blebbing, and the dynamics of membrane deformations were analyzed. Blebs preferentially originate from negatively curved regions, generated on the flanks of either extending pseudopods or blebs themselves. This is true of cells at different developmental stages, chemotaxing to either folate or cyclic AMP and moving with both blebs and pseudopods or with blebs only. A physical model of blebbing suggests that detachment of the cell membrane is facilitated in concave areas of the cell, where membrane tension produces an outward directed force, as opposed to pulling inward in convex regions. Our findings assign a role to membrane tension in spatially coupling blebs and pseudopods, thus contributing to clustering protrusions to the cell front.
Crawling cells must restrict protrusions to a limited part of their periphery if they are to move efficiently, and when these cells chemotax, the location of projections must be further controlled by the chemotactic gradient (13). Cellular protrusions are of two main types: those driven by actin polymerization, such as pseudopods or lamellipods, and those driven by fluid pressure, which are usually called blebs. Blebs form when the cell membrane locally detaches from the underlying cortex and is driven outward by hydrostatic pressure, created by myosin-II–driven contraction of the cortex (4, 5). When blebs form, the cortex is left behind as an “F-actin scar,” which depolymerizes, while a new actin cortex forms at the freshly exposed membrane.
Blebbing is important in cells migrating in three-dimensional environments, such as during tumor invasion (6, 7), zebrafish primordial germ cell migration (8, 9), or migration of the pathogen Entamoeba histolytica in the liver (10). Dictyostelium amoebae can also move with blebs (11). In standard conditions on a 2D surface under buffer, they move mainly with F-actin–driven pseudopods, but switch progressively to bleb-driven motility when faced with mechanical resistance to their movement (12). This can be conveniently applied by inducing the cells to migrate under an elastic overlay, such as agarose, which they must deform to progress (13). Blebbing is stimulated by acute treatment with the chemoattractant cyclic AMP (14), and blebs can be chemotactically orientated by cyclic-AMP gradients (11, 12).
Actin-driven pseudopods are preferentially formed up-gradient by chemotaxing cells, and they can be induced on the flanks of cells by applying a steep gradient of chemoattractant from a micropipette, showing, along with much other evidence, that the site of actin polymerization and pseudopod formation can be controlled by a signaling pathway from the chemotactic receptors (15). Despite intensive investigation, many aspects of this pathway remain to be established.
Because blebs, like pseudopods, are preferentially orientated to the front of moving Dictyostelium cells and can be similarly orientated by gradients of cyclic AMP, the questions arise: How are bleb sites selected and how is this selection influenced by chemotactic gradients? And how do blebs and pseudopods interact to produce sustained movement, rather than potentially interfering with each other?
There are several proposals for bleb site selection, including by local contraction of the cortex (8) or local weakening of it (16, 17) and by local weakening of the attachment between cortex and membrane (18, 19). In cells of the parasite E. histolytica, where blebs expand rapidly, there are, however, no visible signs of a weakening of the actin cortex before a bleb forms (10) and likewise blebs in Dictyostelium form without detectable weakening of the cortex.
One clear interaction between blebs and pseudopods is where blebs give rise to pseudopods by continued actin polymerization: The F-actin–denuded membrane of a newly formed bleb is an excellent template for actin polymerization, which rapidly restores the cortex. In many cases, actin polymerization continues after the cortex has been restored, transforming the bleb into a pseudopod and creating hybrid structures, which we call “blebbopods” (12). Thus, through this route of continued actin polymerization, blebs can determine where pseudopods form.
To study bleb formation and bleb–pseudopod interactions in detail we developed an accurate method for automatically analyzing blebs in image time series. We describe here a second interaction, mediated by mechanical means, in which pseudopods (and potentially other F-actin projections) as well as other blebs can influence where blebs form by creating areas of local negative membrane curvature.


To seek general rules governing bleb formation, we examined Dictyostelium cells in three situations. In each, wild-type Ax2 cells move on glass and are attracted under an agarose overlay, which provides mechanical resistance. We used cells starved for 5–6 h chemotaxing to cyclic AMP under 0.7% agarose (Young’s modulus E = 39 kPa) (12), where both pseudopods and blebs form in equivalent numbers, along with some filopods; cells chemotaxing to cyclic AMP under 2% (wt/vol) agarose (E = 295 kPa), where blebs but very few pseudopods form, and filopods are not detected at all [using the cyclic AMP receptor 1 fused to green fluorescent protein (cAR1-GFP) membrane marker]; and finally, vegetative cells freshly harvested from growth medium and chemotaxing to folic acid under 2% agarose, where blebs and pseudopods form in roughly equal numbers, although chemotaxis is less efficient.

Characterization and Chemotactic Orientation of Cellular Protrusions.

Fig. 1 shows the major types of protrusion made by cells migrating under 0.7% agarose toward cyclic AMP. Cells are observed using GFP-ABD (actin binding domain of Dictyostelium filamin fused to GFP) to mark F-actin and with fluorescent dextran included in the agarose to reveal their outlines (Fig. 1A and Movie S1). F-actin–driven pseudopods (henceforth pseudopods) advance steadily with a continuous zone of F-actin beneath the plasma membrane, whereas blebs advance in a saltatory fashion, with very little F-actin beneath the membrane as they expand. The hybrid blebbopods start life as a bleb and then transform into a pseudopod by continued actin polymerization.
Fig. 1.
Characteristics of the projections produced by Dictyostelium cells under an agarose overlay. The cells express an F-actin marker (green), and RITC-dextran, included in the agarose, shows cell outlines (red). (A) Typical bleb: Arrows show freshly polymerized actin of the reconstituting bleb cortex; starred arrow shows residual cortex. Differential interference contrast (DIC) (gray) shows that blebs are organelle-free. (B) Typical pseudopod: F-actin is associated with the membrane at all times. (C) Two adjacent blebs produced in less than 1 s: Initially there is little F-actin (first bleb, 1 s), which builds up quickly (2 s), while the actin scar breaks down, becoming visibly ruptured (4 s). (D) Blebbopods: Fast, bleb-like initial extension is followed by slower extension driven by actin polymerization. (E–G) Kymographs of protrusion along yellow lines shown in B–D, highlighting the continuous association of actin during pseudopod extension, sudden membrane detachment in blebbing, and two-stage extension in blebbopods. (H) Chemotactic orientation of projections (green circle is up-gradient). Pseudopods dominate the leading edge of cells chemotaxing to cyclic AMP under 0.7% agarose whereas blebs have a bimodal orientation (30 cells; Npseud = 304; Nblebs = 194, Hartigan’s dip test: P < 0.05). Distribution of blebbopods is a mix of that of blebs and pseudopods, Nbp = 49. Blebs are also bimodal on cells chemotaxing to cyclic AMP under 2% agarose (13 cells; Nblebs = 530, Hartigan’s dip test: P < 0.01) or folate (10 cells; Nblebs = 212, Hartigan’s dip test: P < 0.01). Bleb distributions differ between 0.7% and 2% agarose (Kuiper two-sample test, P < 0.001) and mode peaks appear closer together under 2% agarose. In all cases projections align with the gradient (zero mean), although cells chemotaxing to folate are significantly less directed (circular one-sample t test). (Adapted from ref. 12.)
We enhanced our QuimP software (20, 21) to semiautomatically determine the orientation of blebs and pseudopods made by cells moving in chemotactic gradients (details in SI Text and Figs. S1S4). As shown previously (12), polar plots of cellular projections under 0.7% agarose reveal that actin protrusions dominate the leading edge, whereas blebs also nucleate at the cell front, but preferentially on the flanks of the leading edge, forming a bimodal distribution. The distribution of blebbopods is a mix of that of blebs and pseudopods, suggesting that blebs formed at the cell front are more likely to elicit sustained actin polymerization and so broaden the focus of actin-driven protrusions.
We confirmed that the bimodal distribution of blebs also holds for the other two experimental conditions (Fig. 1H). Under 2% agarose, few pseudopods form and the orientation of blebs is more dispersed, but still bimodal. Cells also produce blebs preferentially toward folic acid in a bimodal manner, although the distribution of pseudopods is less well defined, consistent with the less efficient chemotaxis to this agent.
The tendency of blebs to form on the flanks of pseudopods, apparent under 0.7% agarose, suggests that the two forms of protrusion may interact. Because the flanks of pseudopods are characterized by negative curvature, we hypothesized that this interaction might be controlled by membrane geometry.
To investigate this further we performed detailed mappings of pseudopods and blebs, using a novel method—electrostatic contour migration (21)—where charged virtual markers migrate in an electric field between successive contour segments to which opposite charges have been applied (Fig. 2A). Markers on the membrane can be assigned velocities (Fig. 2B) and marker tracks can be stitched together for any point on the cell contour (Fig. 2C). Changes in cell morphology and local curvature (Methods) are visualized in 2D spatiotemporal maps, in which selected marker tracks show that our algorithm can closely follow deformations due to blebs, protrusions, and contractions (Fig. 2C and Movies S2–S5), enabling us to automatically characterize their velocity, curvature, and intensity.
Fig. 2.
Tracking protrusions suggests that blebs originate in areas of negative curvature. (A) Principle of the tracking method, which is based on migration of charged markers in an electrostatic field. A sector is defined by the intersection of successive cell contours and here sector ϒT is positively charged, and ϒT+1 is negatively charged, forming an electrostatic field in which positively charged markers migrate from ϒT to ϒT+1. The method minimizes the total path integral and thus the energy required to deform ϒT to yield ϒT+1; as field lines never cross, it creates a unique and smooth mapping. (B) Tracking applied to pairs of successive frames (0.5-s intervals) showing example blebs (b.1b.3), pseudopods (p.1 and p.2), and a contraction (c.1). Colors indicate node speed (red, high; green, low). Positions on the membrane can be followed over extended periods by interpolation of mappings. (C) Spatiotemporal velocity map. The fastest moving points within the example regions labeled in B are automatically traced through the entire sequence (black traces) and clearly follow patterns of cellular deformations (black frames; complete tracking shown in Fig. S3 and Movie S2). (D) Corresponding curvature map. It is apparent that the selected blebs originate in areas of negative curvature and the pseudopods in areas of positive curvature. (E) Zoomed curvature map demonstrates b.1 originates from a region of negative curvature and induces negative curvature from which bleb b.2 nucleates.
The algorithm located 519 projections in 30 cells under 0.7% agarose with a false positive rate of 4.1% (determined by visual inspection and excluded from further analysis) and a sensitivity of 96.4% correctly identified projections.

Blebs Preferentially Nucleate at Regions of Negative Membrane Curvature: Qualitative Observations.

To test the hypothesis that blebs form preferentially from areas of negative membrane curvature, we constructed velocity and curvature maps of cells moving toward cyclic AMP under 0.7% agarose (Fig. 2 C and D). As expected, blebs and pseudopods have distinctly different velocity signatures, with the short, sharp expansion of blebs contrasting with the slower, more prolonged expansion of pseudopods. The curvature map shows that the two pseudopods (Fig. 2B, p.1 and p.2) each expand in areas of positive curvature, whereas consistent with the hypothesis, the three blebs form in areas of negative curvature: b.1 is, for example, created between two pseudopods, and b.2 is from the negative curvature created on the flank of b.1. Higher resolution shows the abrupt switch from negative to positive curvature as a bleb forms and the induction of negative curvature on its flank (Fig. 2E).

Blebs Preferentially Nucleate at Regions of Negative Membrane Curvature: Population Data.

To test the generality of the correlation between negatively curved membrane and bleb formation, we analyzed populations of protrusions. Projections were manually classified as blebs or pseudopods, depending on the presence of an actin scar and actin at the leading edge, and the relevant parameters were extracted automatically.
A global view reveals a wide distribution in both speed and displacement for blebs and pseudopods under all conditions (Fig. 3 A–C, Left). Cells moving under 2% agarose are flatter (12) and bleb 10 times more than those moving under 0.7% agarose (Table 1). In contrast, the rate at which pseudopods are produced remains unchanged, although at the higher agarose concentration they more often emerge directly from blebs, thus forming blebbopods.
Fig. 3.
Population studies show blebs preferentially nucleate in areas of negative curvature. (A) Cells moving toward cyclic AMP under 0.7% agarose (30 cells, 144 blebs, 304 pseudopods) (adapted from ref. 12). (B) Cells moving toward cyclic AMP under 2% agarose (13 cells, 454 blebs, 37 pseudopods). (C) Cells moving toward folic acid under 2% agarose (10 cells, 118 blebs, 231 pseudopods). For each condition, blebs and pseudopods are characterized by their peak projection speed and total displacement (excluding blebbopods) and the curvature of the membrane from which they originate (blebbopods included as blebs). In A and C there is a significant difference in distribution of blebs and pseudopods with curvature (Kolmogorov–Smirnov test, P < 0.001; Mann–Whitney–Wilcoxon test, P < 0.001; insufficient pseudopods in B to test). To test the significance of the increased frequency of blebs at negative curvature and decreased frequency at positive curvature, the data were split into six classes of curvature (chosen to maintain required minimum frequencies for testing), and χ2 tests were made between observed and expected bleb distributions (stars). In addition, blebbing is 10-fold increased in cells chemotaxing to cyclic AMP under 2% compared with 0.7% agarose, and the blebs expand faster (0.7% = 1.35 ± 0.92 μm/s, 2% = 1.88 ± 0.99 μm/s).
Table 1.
Projection counts
ExperimentNo. cellsPseudopodsBlebsSingular blebs*Blebbopods
cAMP, 0.7% agarose30 (42.1)304 (0.106)194 (0.068)144 (0.050)49 (0.017)
cAMP, 2.0% agarose13 (15.3)37 (0.047)530 (0.667)454 (0.571)62 (0.078)
Folate, 2.0% agarose10 (15.5)231 (0.211)212 (0.193)118 (0.108)94 (0.086)
Excludes blebs that transform into blebbopods and stacked blebs. Definitions are in SI Text.
Total imaging time, minutes.
Protrusion rate: protrusions per micrometer per minute.
During chemotaxis toward cyclic AMP under 0.7% agarose, pseudopods clearly originate from regions of positive curvature (91% of cases), but at first sight blebs do not appear so clearly directed to regions of negative curvature (63% of cases; Fig. 3A, Center). For a more rigorous statistical analysis, we compared actual curvatures in regions where blebs originate to a test distribution based on random sampling of curvatures from the same cells and found that significantly more blebs form below −27° than would be expected by chance. Likewise, the number of blebs in regions with more than +27° of curvature is significantly reduced.
Blebs are more strikingly associated with areas of negative curvature in cells under 2% agarose, where the induction of negative curvature is almost entirely due to the blebs themselves (Fig. 3B). This argues that the correlation between blebbing and negative curvature is independent of the type of protrusion causing the negative curvature. Similarly, blebs in vegetative cells chemotaxing toward folic acid show a significant bias toward areas of negative curvature (Fig. 3C and Fig. S5A). Consideration of curvature also helps us to understand why blebs are more common in the rear of cells under 2% than 0.7% agarose (Fig. 1H): Fig. S6 shows that cells under 2% agarose have a less pointed rear. Because this is equivalent to a reduction of positive curvature, the chances of blebbing at the rear are increased.
Surprisingly, given the increased mechanical resistance, blebs are projected faster by developed cells under 2% than under 0.7% agarose (Table 2, 0.7%, 1.35 μm/s; 2%, 1.88 μm/s; P < 0.001, two-sample Kolmogorov–Smirnov test), suggesting a greater motive force, presumably due to increased intracellular pressure. However, these faster blebs expand less far than their slower counterparts (0.7%, 1.14 μm; 2%, 0.97 μm; P < 0.01). Fast (10 fps) imaging of cells under 2% agarose (Fig. S7 and Movie S6) shows there is a 3% local area gain during bleb expansion, offset by a global loss of 2%, so that on average the cell area remains relatively constant, with a SD of less than 1% [we take projected cell area because cells under 2% agarose are essentially flat and of uniform height (12); global contraction concomitant with blebbing has previously been described in ref. 11]. Despite the differences in modes of movement observed under 0.7% and 2% agarose, cell speeds do not differ significantly [0.7%, 10.8 μm/min (±8.1); 2%, 7.6 μm/min (±4.6); two-sample Kolmogorov–Smirnov test].
Table 2.
Displacement, peak speed, and curvature measurements, ± SD
ExperimentDisplacement, μmPeak speed, μm/sCurvature, °
cAMP, 0.7% agarose2.52 ± 2.011.14 ± 0.550.67 ± 0.281.35 ± 0.9240.5 ± 30.1−5.8 ± 25.63.6 ± 24.6
cAMP, 2.0% agarose1.24 ± 0.820.97 ± 0.410.65 ± 0.231.88 ± 0.998.82 ± 14.6−10.26 ± 20.9−6.84 ± 22.3
Folate, 2.0% agarose2.26 ± 1.511.19 ± 0.490.80 ± 0.331.66 ± 0.910.72 ± 20.3−28.62 ± 30.9−16.56 ± 19.3
Excludes blebs that transform into blebbopods and stacked blebs.

Blebbing in Fundulus Deep Cells.

The literature contains many examples where blebs seem to originate preferentially from regions of negative curvature, but the possibility of a causal link has received little attention to date (ref. 10, figure 2B; ref. 22, figure 10A; and ref. 23, figure 3A). Fundulus deep cells provide a beautiful example, which we have analyzed with our software (24) (data kindly provided by Rachel Fink, Mount Holyoke College, Holyoke, MA; Movie S7 and Fig. S5B). Blebs form regularly from the front of these cells, and in the one shown, it can be seen that they arise mainly from flat or negatively curved areas and not at all from strongly positively curved areas. It is also apparent that blebs do not form at the negatively curved waist separating the uropod from the main body of the cell, suggesting that membrane–cortex connections are reinforced in this area.

A Simple Physical Model.

To make our ideas explicit, we made a simple physical model relating membrane curvature to blebbing propensity. In this we assume that the plasma membrane is attached to the cortex by evenly spaced linkers and that fluid pressure on the membrane is everywhere the same. On the timescale of bleb nucleation the cortex is considered to be fixed. Membrane tension produces a force whose direction depends on curvature: In regions of positive curvature it is directed inward (Fig. 4A) and in regions of negative curvature outward (Fig. 4B). Similar considerations apply to membrane bending, but its contribution is an order of magnitude less. In these conditions a bleb will form anywhere where the net outward force on the membrane due to fluid pressure and membrane tension is strong enough to stretch and eventually break linkers.
Fig. 4.
Proposed mechanism for bleb nucleation. Red, cell membrane; green, cortex; black, tethers. (A) In regions of positive curvature intracellular pressure is opposed by membrane tension and tethering. (B) In regions of negative curvature tension is inverted, thus putting more pull on tethers. During detachment, a surplus of membrane becomes available for expansion. (C) Model simulation (Movie S8 and SI Text for model details). A protrusion, formed from initially circular geometry, induces concavities at its flanks from which blebs form spontaneously. (D) Space–time map of the net force due to membrane tension and bending acting on tethers. Force is high at the tip of the simulated pseudopod and directed inward, preventing bleb formation. In the neck region membrane tension is of similar magnitude but is directed outward and no longer opposes fluid pressure. This causes linkers to rupture and the high pulling forces on the basal bleb linkers (red streaks) promote unzipping of further linkers. Stopping bleb expansion is discussed in SI Text. (E) Curvature map of simulated blebbing shows a distinct pattern of negative curvature being flipped to positive curvature at the bleb site and induced negative curvature at bleb flanks. This simple model gives a feedback mechanism in which F-actin–driven pseudopods promote blebbing at their flanks. (F and G) A bleb cascade (Movie S3). A cell chemotaxing to cyclic AMP (source at bottom) under 2% agarose produces a cascade of blebs at the leading edge in which the first (0.0 s) produces negative curvature to spawn flanking blebs (0.5 s) that in turn spawn further blebs along the cell sides (1.0–2.5 s), consistent with the model proposed in A–E.
Fig. 4C shows a snapshot of a simulation where blebs originate from the negatively curved neck region of a protrusion (Movie S8; details in SI Text and Table S1). Fig. 4D shows that the force acting on linkers is directed outward in the position where the two blebs form; the values in the piconewton range are indicative only of the expected order of magnitude. The corresponding curvature map (Fig. 4E) shows that blebs themselves generate negative curvature on their flanks. That this mechanism can actually spawn new blebs is apparent with real cells under 2% agarose, which sometimes extend consecutive blebs in a chain reaction because of successively induced concavities (Fig. 4 F and G and Movie S3). The characteristic jumps in curvature (from blue to red) seen in the model (Fig. 4E) agree well with the patterns marked by the short traces of blebs in the experimental curvature map (Fig. 4G). It is notable that daughter blebs in the real cell do not form at the apex of the earlier bleb, where the F-actin cortex is at its thinnest, but the membrane curvature is positive, but at the flank: Thus curvature is likely the dominant factor in this case. The propagation of a bleb around a cell in “circus movement” may have a similar explanation, in that the bleb continuously expands into areas of negative membrane curvature (25, 26).


Bleb-driven motility is a feature of cells moving in tissues and other mechanically resistive environments, and in Dictyostelium it seems to be provoked by the mechanical resistance itself (12). Here we have addressed the questions of how the site of blebbing is selected and how blebs can cooperate with pseudopods when both are present and thus ultimately of how a cell moving with blebs is steered.
In each of the three physiological situations examined, bleb formation is correlated with areas of moderate-to-strong negative curvature and decreased in areas of positive curvature. This is particularly striking in cells under 2% agarose, where chains of blebs often form, with each new bleb triggered from the concavity formed at the flank of the preceding one. The correlation does not depend on the chemoattractant used or the nature of the projection causing curvature—bleb or pseudopod—consistent with the idea that it is a purely physical interaction. The underlying reason is readily apparent: Membrane tension provides an inward force opposing bleb formation in convex areas and an outward one, favoring it, in concave areas.
To test whether membrane curvature can alone predict where blebs will form, we asked whether blebs always form in the region of the cell with the greatest negative (or least positive) curvature (Fig. S8). Analyzed in this way, the data confirm the strong correlation between negative curvature and bleb formation, but also show that blebs do not necessarily form in the areas of highest negative curvature. Similarly, our analysis of Fundulus deep cell movement shows that although blebs form from areas of negative curvature, they do not form in the concave waist between the cell body and the uropod.
These observations show that other factors apart from membrane curvature must also regulate bleb site selection. Such factors could include the density of membrane-cortical linkers and the local concentration of PIP2 in the membrane, as this can modulate the attachment of these linkers (18, 27). Consistent with this, in melanoma cells, it has been argued that a high level of the membrane–cytoskeleton linker ezrin inhibits blebbing from the uropod (28).
Because concavities often form on the flanks of pseudopods, pseudopods can trigger adjacent blebs and provide indirect chemotactic steering to them. However, when cells reorientate toward a moving cyclic-AMP source, they follow a distinct series of events in which actin microspikes first form in the new direction of the gradient, followed by blebs, and pseudopods only later (12). In this case, blebs are not triggered by a preceding pseudopod, but could be by the actin microspikes, which can also create local negative membrane curvature. Filopods could also play a role in triggering blebs, but not an essential one, because they are not detected in strongly blebbing cells under 2% agarose (Movie S9). Conversely, blebs can determine where pseudopods form, through continued actin polymerization and the formation of blebbopods. Thus, each type of projection can trigger formation of the other, helping to focus protrusive activity to the front of the cell. However, because cells can move chemotactically primarily with either pseudopods (vegetative cells under buffer) or blebs (developed cells under 2% agarose), it seems likely that in addition to their chemotactic cooperation, blebs and pseudopods also have independent chemotactic steering.
It is notable that blebs are projected faster than actin-driven pseudopods in each condition we investigated, suggesting that they are impelled by a greater force; this greater force provides a rationale for why blebs often predominate in resistive conditions. The global propensity of cells to form blebs is regulated by cortical contractility, which is mediated through myosin II. Cells might thus adapt to resistive conditions by increasing their expression of myosin II, as seen in Dictyostelium development, or by acute increases in myosin-II activity (29).
Our observations add to the range of cellular processes that can be controlled by membrane tension. As well as blebbing, these include endocytosis and, in the field of cell motility, the long-range suppression of pseudopods by tension produced by a dominant pseudopod (30), all emphasizing the role of mechanics in controlling cell physiology (31).


Cells and Microscopy.

The axenic strain Ax2 of Dictyostelium discoideum, used throughout (R.R.K. laboratory strain; DBS0235521 at, was grown on HL5 medium (ForMedium). All experiments were at 22 °C. Ax2 was transformed with markers for F-actin (GFP–ABD120, ref. 32) or the plasma membrane/filopods (cAR1-GFP, ref. 33).
Cells for folate chemotaxis experiments were diluted 10-fold from growth medium into KK2 buffer (16.5 mM KH2PO4, 3.9 mM K2HPO4, 2 mM MgSO4, 0.1 mM CaCl2, pH 6.1) and plated, and the medium was replaced with fresh KK2 after 10 min. Cells for cyclic-AMP chemotaxis experiments were washed free of growth medium and starved for 5.5 h with pulses of cyclic AMP (around 90 nM final) added every 6 min after the first hour. Imaging was performed under KK2: Cells were placed into wells cut in an agarose (SeaKem GTG; Lonza) overlay on Lab-Tek coverslips (Thermo Fisher Scientific) and attracted under the agarose toward an adjacent well, containing either 4 μM cyclic AMP or 1 mM folic acid (12). Rhodamine-B-isothiocyanate-dextran (RITC-Dx), 0.5 mg/mL, was included in the agarose for negative staining, as indicated. Dual channel images were acquired at 2–10 fps, using either a Zeiss 710 laser scanning confocal microscope with a 63× oil-immersion objective or a spinning-disk confocal microscope (Ultraview; PerkinElmer) with an 100× oil-immersion objective.

Image Processing and Protrusion Analysis.

GFP-ABD120 images were background corrected and intensity normalized. For the purpose of segmentation, RITC-Dx images were inverted and a Gaussian convolution (σ = 1 pixel) was applied to smooth high-frequency noise. Cells were segmented and analyzed using the QuimP11 software ( and passed through our automatic protrusion detection algorithm (overview of the QuimP11 software in SI Text).

Quantifying Projection Peak Velocity.

Maxima in protrusion velocity profiles suffer from membrane fluctuations and segmentation noise. To alleviate this, we compute the associated displacement profile and fit a sigmoid curve of the form f(x)=a(1exp(xc/b))+d to it, which allows for asymmetrical acceleration and deceleration and is differentiable. Peak velocity is determined as the maximal gradient, max(df(x)/dx) (Fig. S4). A sample of peak velocities computed from automatically extracted projection paths was verified using manually obtained projection paths and found to be in good agreement (Fig. S4E).

Computing Projection Rates.

Given a sequence of N cell outlines with K detected projections, the projection rate PR is computed as the number of observed projections per micrometer, per minute, PR=K/IP, where IP=i=1N(Ti+1Ti)((Pi+Pi+1)/2) is the integrated perimeter, T is time, and P the cell perimeter.

Defining Curvature and χ2 Tests.

Curvature in the range (−180, 180) was determined by summing up the angles formed between markers on the cell contour and their neighbors over 1 μm. Zero is flat, positive values are convex, and negative values are concave. Curvature at bleb nucleation was computed by interpolating between closest markers in the frame immediately before an expansion by 0.2 μm. Significance testing used a test distribution of randomly sampled curvatures (>4,000) drawn from actual cell contours at times when blebs appear. Comparisons using χ2 tests were performed, binning the data into six classes with boundaries [−Inf, −45, −27, −9, 9, 27, Inf]. A Bonferroni correction for multiple comparisons was applied to final P values.


We thank the Medical Research Council (MRC) for core support (MRC file reference no. U105115237). E.Z. was supported by a Herchel Smith Fellowship. R.A.T. was supported by the Wellcome Trust Project Grant WT094131MA (to T.B.).

Supporting Information

Supporting Information (PDF)
Supporting Information


RR Kay, P Langridge, D Traynor, O Hoeller, Changing directions in the study of chemotaxis. Nat Rev Mol Cell Biol 9, 455–463 (2008).
RH Insall, Understanding eukaryotic chemotaxis: A pseudopod-centred view. Nat Rev Cell Biol 11, 453–458 (2010).
AJ Ridley, Life at the leading edge. Cell 145, 1012–1022 (2011).
G Charras, E Paluch, Blebs lead the way: How to migrate without lamellipodia. Nat Rev Mol Cell Biol 9, 730–736 (2008).
OT Fackler, R Grosse, Cell motility through plasma membrane blebbing. J Cell Biol 181, 879–884 (2008).
K Wolf, et al., Compensation mechanism in tumor cell migration: Mesenchymal-amoeboid transition after blocking of pericellular proteolysis. J Cell Biol 160, 267–277 (2003).
H Yamaguchi, J Condeelis, Regulation of the actin cytoskeleton in cancer cell migration and invasion. Biochim Biophys Acta 1773, 642–652 (2007).
H Blaser, et al., Migration of zebrafish primordial germ cells: A role for myosin contraction and cytoplasmic flow. Dev Cell 11, 613–627 (2006).
A Schepis, D Sepich, WJ Nelson, αE-catenin regulates cell-cell adhesion and membrane blebbing during zebrafish epiboly. Development 139, 537–546 (2012).
B Maugis, et al., Dynamic instability of the intracellular pressure drives bleb-based motility. J Cell Sci 123, 3884–3892 (2010).
K Yoshida, T Soldati, Dissection of amoeboid movement into two mechanically distinct modes. J Cell Sci 119, 3833–3844 (2006).
E Zatulovskiy, R Tyson, T Bretschneider, RR Kay, Bleb-driven chemotaxis of Dictyostelium cells. J Cell Biol 204, 1027–1044 (2014).
G Laevsky, DA Knecht, Under-agarose folate chemotaxis of Dictyostelium discoideum amoebae in permissive and mechanically inhibited conditions. Biotechniques 31, 1140–1142, 1144, 1146–1149 (2001).
PD Langridge, RR Kay, Blebbing of Dictyostelium cells in response to chemoattractant. Exp Cell Res 312, 2009–2017 (2006).
KF Swaney, C-H Huang, PN Devreotes, Eukaryotic chemotaxis: A network of signaling pathways controls motility, directional sensing, and polarity. Annu Rev Biophys 39, 265–289 (2010).
E Paluch, M Piel, J Prost, M Bornens, C Sykes, Cortical actomyosin breakage triggers shape oscillations in cells and cell fragments. Biophys J 89, 724–733 (2005).
J-Y Tinevez, et al., Role of cortical tension in bleb growth. Proc Natl Acad Sci USA 106, 18581–18586 (2009).
D Raucher, et al., Phosphatidylinositol 4,5-bisphosphate functions as a second messenger that regulates cytoskeleton-plasma membrane adhesion. Cell 100, 221–228 (2000).
A Diz-Muñoz, et al., Control of directed cell migration in vivo by membrane-to-cortex attachment. PLoS Biol 8, e1000544 (2010).
D Dormann, T Libotte, CJ Weijer, T Bretschneider, Simultaneous quantification of cell motility and protein-membrane-association using active contours. Cell Motil Cytoskeleton 52, 221–230 (2002).
R Tyson, D Epstein, K Anderson, T Bretschneider, High resolution tracking of cell membrane dynamics in moving cells: An electrifying approach. Math Model Nat Phenom 5, 34–55 (2010).
K Yoshida, K Inouye, Myosin II-dependent cylindrical protrusions induced by quinine in Dictyostelium: Antagonizing effects of actin polymerization at the leading edge. J Cell Sci 114, 2155–2165 (2001).
M Bergert, SD Chandradoss, RA Desai, E Paluch, Cell mechanics control rapid transitions between blebs and lamellipodia during migration. Proc Natl Acad Sci USA 109, 14434–14439 (2012).
JP Trinkaus, Surface activity and locomotion of Fundulus deep cells during blastula and gastrula stages. Dev Biol 30, 69–103 (1973).
N Fujinami, Studies on the mechanism of circus movement in dissociated embryonic cells of a teleost, Oryzias latipes: Fine-structural observations. J Cell Sci 22, 133–147 (1976).
GT Charras, M Coughlin, TJ Mitchison, L Mahadevan, Life and times of a cellular bleb. Biophys J 94, 1836–1853 (2008).
L Fets, JME Nichols, RR Kay, A PIP5 kinase essential for efficient chemotactic signaling. Curr Biol 24, 415–421 (2014).
A Lorentzen, J Bamber, A Sadok, I Elson-Schwab, CJ Marshall, An ezrin-rich, rigid uropod-like structure directs movement of amoeboid blebbing cells. J Cell Sci 124, 1256–1267 (2011).
V Sanz-Moreno, CJ Marshall, The plasticity of cytoskeletal dynamics underlying neoplastic cell migration. Curr Opin Cell Biol 22, 690–696 (2010).
AR Houk, et al., Membrane tension maintains cell polarity by confining signals to the leading edge during neutrophil migration. Cell 148, 175–188 (2012).
NC Gauthier, TA Masters, MP Sheetz, Mechanical feedback between membrane tension and dynamics. Trends Cell Biol 22, 527–535 (2012).
KM Pang, E Lee, DA Knecht, Use of a fusion protein between GFP and an actin-binding domain to visualize transient filamentous-actin structures. Curr Biol 8, 405–408 (1998).
Z Xiao, N Zhang, DB Murphy, PN Devreotes, Dynamic distribution of chemoattractant receptors in living cells during chemotaxis and persistent stimulation. J Cell Biol 139, 365–374 (1997).

Information & Authors


Published in

Go to Proceedings of the National Academy of Sciences
Proceedings of the National Academy of Sciences
Vol. 111 | No. 32
August 12, 2014
PubMed: 25074921


Submission history

Published online: July 29, 2014
Published in issue: August 12, 2014


We thank the Medical Research Council (MRC) for core support (MRC file reference no. U105115237). E.Z. was supported by a Herchel Smith Fellowship. R.A.T. was supported by the Wellcome Trust Project Grant WT094131MA (to T.B.).


This article is a PNAS Direct Submission.



Richard A. Tyson
Warwick Systems Biology Centre, University of Warwick, Coventry CV4 7AL, United Kingdom; and
Medical Research Council Laboratory of Molecular Biology, Cambridge CB2 0QH, United Kingdom
Robert R. Kay
Medical Research Council Laboratory of Molecular Biology, Cambridge CB2 0QH, United Kingdom
Till Bretschneider1 [email protected]
Warwick Systems Biology Centre, University of Warwick, Coventry CV4 7AL, United Kingdom; and


To whom correspondence should be addressed. Email: [email protected].
Author contributions: R.A.T., E.Z., R.R.K., and T.B. designed research; R.A.T. and E.Z. performed research; R.A.T. analyzed data; and R.A.T., R.R.K., and T.B. wrote the paper.

Competing Interests

The authors declare no conflict of interest.

Metrics & Citations


Note: The article usage is presented with a three- to four-day delay and will update daily once available. Due to ths delay, usage data will not appear immediately following publication. Citation information is sourced from Crossref Cited-by service.

Citation statements



If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

Cited by


    View Options

    View options

    PDF format

    Download this article as a PDF file


    Get Access

    Login options

    Check if you have access through your login credentials or your institution to get full access on this article.

    Personal login Institutional Login

    Recommend to a librarian

    Recommend PNAS to a Librarian

    Purchase options

    Purchase this article to access the full text.

    Single Article Purchase

    How blebs and pseudopods cooperate during chemotaxis
    Proceedings of the National Academy of Sciences
    • Vol. 111
    • No. 32
    • pp. 11569-11906







    Share article link

    Share on social media