Conceived and designed the experiments: LH ABS AB. Performed the experiments: YI YBK. Analyzed the data: YI YBK. Wrote the paper: YBK ABS AB.
The authors have declared that no competing interests exist.
During cellular migration, regulated actin assembly takes place at the cell leading edge, with continuous disassembly deeper in the cell interior. Actin polymerization at the plasma membrane results in the extension of cellular protrusions in the form of lamellipodia and filopodia. To understand how cells regulate the transformation of lamellipodia into filopodia, and to determine the major factors that control their transition, we studied actin self-assembly in the presence of Arp2/3 complex, WASp-VCA and fascin, the major proteins participating in the assembly of lamellipodia and filopodia. We show that in the early stages of actin polymerization fascin is passive while Arp2/3 mediates the formation of dense and highly branched aster-like networks of actin. Once filaments in the periphery of an aster get long enough, fascin becomes active, linking the filaments into bundles which emanate radially from the aster's surface, resulting in the formation of star-like structures. We show that the number of bundles nucleated per star, as well as their thickness and length, is controlled by the initial concentration of Arp2/3 complex ([Arp2/3]). Specifically, we tested several values of [Arp2/3] and found that for given initial concentrations of actin and fascin, the number of bundles per star, as well as their length and thickness are larger when [Arp2/3] is lower. Our experimental findings can be interpreted and explained using a theoretical scheme which combines Kinetic Monte Carlo simulations for aster growth, with a simple mechanistic model for bundles' formation and growth. According to this model, bundles emerge from the aster's (sparsely branched) surface layer. Bundles begin to form when the bending energy associated with bringing two filaments into contact is compensated by the energetic gain resulting from their fascin linking energy. As time evolves the initially thin and short bundles elongate, thus reducing their bending energy and allowing them to further associate and create thicker bundles, until all actin monomers are consumed. This process is essentially irreversible on the time scale of actin polymerization. Two structural parameters,
Actin polymerization at the plasma membrane results in the formation of cellular protrusions known as lamellipodia or filopodia, which mediate cell migration
Most cultured animal cells assemble both lamellipodia and filopodia. Some cells, like dendritic cells, are dominated by filopodia
Notwithstanding recent advances in understanding the dynamic organization of lamellipodia and filopodia protrusions, it is still not fully understood how cells control the transition between these structures, what directs the localization of filopodia formation along the cell leading edge, and how their thickness is regulated. It is expected that the emergence of filopodia from lamellipodia would be strongly affected by the properties (e.g., the density and length of filaments) of the branched lamellipodial network from which they emanate. The concentration of Arp2/3, which strongly affects these structural properties, was recently observed to have a dramatic influence on the formation of filopodia
To resolve these issues, it is essential to understand the mechanisms underlying the formation of lamellipodia and filopodia and evaluate the factors controlling their structure and dynamics. To this end, in the present work we study a system containing the major proteins participating in the assembly of lamellipodia and filopodia: a) the constitutively active VCA
Our goal in the present work is to study in detail the self-assembly characteristics of actin in the presence of variable amounts of Arp2/3 complex and fascin. We will show that in the absence of fascin, actin organizes into dense 3D, aster-like, structures composed of branched networks of actin filaments
This picture of stepwise bundle elongation and thickening underlies our structural-energetic model of bundle formation and growth. Using reasonable approximations for the bending energies of actin filaments and bundles, and the cohesive energy due to fasin-actin bonds we can explain the dependence of the thickness and length of the mature bundles upon [Arp2/3]. More explicitly, using Kinetic Monte Carlo (KMC) simulations we can model the nucleation and growth of asters and derive their structure as a function of time thus obtaining two central structural parameters: the length,
The first part of this section describes the three types of structures, i.e., asters, stars and network of bundles (a phase of actin-fascin bundles) that were experimentally observed and the transition between them. We also present a phase diagram of the system to illustrate the regions of existence of the three types of structures as a function of the initial [Arp2/3] and [fascin]. In the second part, we focus specifically on the transition from asters to stars, as it mimics the transition from lamellipodia to filopodia in cells. Special attention is given to evaluating the roles of [Arp2/3] and [fascin] on filopodia formation and dimensions. In the last part of this section we describe our theoretical model for bundle assembly and apply it to explain our experimental results.
First, we aimed at studying the self–assembly of actin in the presence of increasing amounts of fascin, at a given [Arp2/3]. At low Arp2/3 complex concentration of 6.25 nM (the [VCA]/[Arp2/3] ratio was kept equal to two in all experiments) two structural transitions are observed, the first, from ‘asters’ to ‘stars’, and the second, from ‘stars’ to ‘network of bundles’ (
Conditions are 7 µM G-actin, 6.25 nM Arp2/3, and 12.5 nM GST-VCA. Fascin concentration is: (a) 3 nM, (b) 4 nM, (c) 5 nM, (d) 7 nM, (e) 50 nM, (f) 200 nM, (g) 500 nM, and (h) 3 µM. At very low concentrations of fascin asters are formed. Transition to star-like structures occurs above a fascin concentration of 4 nM (b). The density and the length of the bundles emanating from the star core increase with fascin concentration (b–f). Above a certain fascin concentration of 0.5 µM (g) the size and the number of stars decreases; the stars coexist with an entangled network of actin/fascin bundles, seen in the image background. At 3 µM stars do not form anymore; the system is composed solely of entangled network of actin bundles (h). Bar is 10 µm.
Gradual addition of fascin, from 0.005 to 0.2 µM, increases the number of bundles emanating from the asters core as well as their length (
Structural transformations mediated by fascin seem to be facilitated when the initial [Arp2/3] was reduced. The top and bottom panels in
Conditions are: 7 µM G-actin, upper line (a–d) 12.5 nM Arp2/3 complex, and 25 nM GST-VCA and bottom line (e–h) 100 nM Arp2/3, and 200 nM GST-VCA. Transition from aster to stars and from stars to network: occurs between 3 to 6 nM and between 0.2 to 3 µM at 12.5 nM [Arp2/3] (blue and red arrows, respectively). At 100 nM [Arp2/3] only the transition from aster to star is visible; transition to a network structure is not visible in these fascin concentration ranges. Bars are 10 µm (a–h).
Further addition of fascin induces the formation of additional and longer bundles, resulting in the appearance of fully developed stars, as shown in
Finally, when the amount of fascin was further elevated to 3 µM (red arrow), a transition from ‘stars’ to ‘a network of bundles’ was observed in the 12.5 nM Arp2/3 system (
In
Condition is: 7 µM G-actin, Arp2/3 complex and fascin concentration was changed according to the graph. Condition for images is: 3 nM, 200 nM and 3 µM fascin; 6.25 nM, 12.5 nM and 6.25 nM Arp2/3 complex (a, b and c, respectively). Experimental points demonstrate different type of structures formed; these are represented by squares - entangled network, triangles - stars and circles - asters. Phase changes from asters to stars (blue line) and from stars to entangled bundle network (red line). Lines of phase separation are plotted as averages of two points of different structures. For low concentration of Arp2/3 complex phase transition is more abrupt for the same change in fascin concentration. No transition to entangled network was visible for high concentrations of Arp2/3 complex.
To test our assumption that each star originates from a preformed aster, we have measured the number of “aggregates” (i.e., asters or stars) per unit area (surface density), within the 1 µm section of the bulk sample observed. The surface density of aggregates
Conditions are: 7 µM G-actin; fascin concentration of 5 nM (triangles) and 200 nM (circles); and variable amount of Arp2/3 complex: 6.25, 12.5, 25, 40, and 100 nM. The [VCA]/[Arp2/3 complex] = 2 in all experiments. We observe a monotonic growth in density of objects with Arp2/3 complex concentration. The density of objects doesn't show a dependence on fascin concentration (inset, a half log plot is given in order to see clearly all experimental data points). Error bars represent standard deviation from average values.
For all the concentrations of Arp2/3 that we have tested, the mean final length of bundles,
(a) Conditions are: [G-actin] = 7 µM; [Arp2/3] = 6.25 nM (green dots), 25 nM (blue dots), and 100 nM (red dots). For all three cases, the mean bundles' length,
Our experiments show that upon mixing actin monomers with Arp2/3 complex, VCA and fascin molecules, the first structures to appear in solution are dense and highly branched aster-like networks of actin filaments
Bending the filaments emanating from the surface of the aster is easier, yet two conditions must be fulfilled to enable the onset of bundle formation. First,
Our goal in this section is to explain the experimental observation that bundles are thicker and longer when the initial [Arp2/3] is low (see
The structural properties of the asters at the onset of bundle formation, primarily the distances
(a). Left: snapshot of an aster taken from the simulations. Filaments originating in the periphery shell of the aster might bend towards each other and link into bundles. Right: Values of
At the onset of the aster-to-star transition the radius of the aster is generally much larger than both
Assuming that all the
Based on this notion we propose here an alternative, stepwise, mechanism whereby single filaments first associate into thin bundles (
Suppose for simplicity that all possible bundles have a circular cross section consisting of a central straight filament surrounded by
Results are shown for two of the initial [Arp2/3] studied experimentally (12.5 and 100 nM), and several representative values of
Association of two adjacent bundles of thickness
To derive the results shown in
The energy change upon linking two bundles is proportional to the length,
We conclude this analysis by noting that from some point in time onwards, the bundles observed experimentally appear to continue elongating without changing their thickness. Based on our model we may conclude that this happens when the time scale of bundle bending fluctuations which lead to bundle-bundle association becomes long compared to the experimental time scale, or simply long relative to the time it takes to all actin monomers to join the growing bundles.
In this work we studied in detail the steady state structures formed in a simple
We found that in the absence or at very low concentration of fascin the system is dominated by Arp2/3 complex nucleation and branching activities, resulting in the appearance of dense 3D aster-like networks of actin. Increasing fascin concentration induces phase changes, first to stars and then to network of bundles. A star is composed of a dense aster core with actin/fascin bundles emanating radially from its surface. Our experiments show that in the early stages of actin polymerization fascin is passive, while Arp2/3 mediates dense aster-like structures of actin, whose structure is very similar to the one observed in the absence of fascin. Fascin comes into play when actin filaments in the periphery of the aster get long enough, and can thus bend and associate with each other into bundles of parallel filaments, held together by fascin linkers. This is in accordance with
The competition between Arp2/3 and fascin is critical in determining the actin structures formed. This is because these two proteins are nucleators of different actin geometries; Arp2/3 complex promotes the formation of branched actin seeds
In formulating our model for bundle formation and growth we assumed that only filaments originating at the surface of the aster take part in this process (similar to
The evolution of asters to stars is of direct biological relevance because of its similarity to lamellipodia to filopodia transition in cells. Recall that two structural parameters,
To conclude, using a minimal model system containing only Arp2/3 complex, actin and fascin we were able to mimic complex events which take place at the leading edge of cells. Compared to a cell this system is simple and easily controlled, making it possible to characterize the roles of a limited set of proteins in the higher order assembly of actin filaments. In the future, it will be interesting to measure the size of the actin/fascin bundle using high resolution electron microscopy, and correlate it with the properties of the aster network structure (e.g., density of the actin network, filament length and orientation at the aster periphery). It is worthwhile to test the effects of mDia2 and capping proteins concentrations on bundle formation and dimensions. Finally, quantitative information from
Actin was purified from rabbit skeletal muscle acetone powder
The motility medium contained 10 mM HEPES, pH = 7.7, 1.7 mM Mg-ATP, 5.5 mM DTT, 0.12 mM Dabco [an anti-bleaching reagent], 0.1 M KCl, 1 mM MgCl2, 1% BSA and various concentrations of G-actin, Arp2/3 complex, VCA, and fascin.
Actin assembly was monitored for about an hour by fluorescence with an Olympus IX-71 microscope. The labeled actin fraction was 1/40 and the temperature was 22°C. Time-lapse images were acquired using an Andor DV887 EMCCD camera (Andor Co., England). Data acquisition and analysis was performed using METAMORPH (Universal Imaging Co.).
(0.38 MB DOC)
Conditions are: 7 µM G-actin, (a–b) 12.5 nM Arp2/3 complex, and 25 nM GST-VCA and (c–d) 100 nM Arp2/3, and 200 nM GST-VCA. The amount of fascin required inducing transition from aster (a and c) to stars (b and d) ranges between 5 to 6 nM and between 6.5 to 7 nM at 12.5 nM and 100 nM Arp2/3 complex, respectively. The insets in b and d show zoom-in images of the bundles emanating from the stars at each [Arp2/3]. At the transition, the bundles emanating from the stars at lower [Arp2/3] (b) are thicker than those originating from the aster core at 100 nM Arp2/3 (d). Bars are 10 µm.
(2.78 MB TIF)
We would like to thank Nir Gov for useful discussions.