Reconstituting the actin cytoskeleton at or near surfaces in vitro

Rodrigo Cáceres, Majdouline Abou-Ghali, Julie Plastino

PII: S0167-4889(15)00258-X
DOI: doi: 10.1016/j.bbamcr.2015.07.021
Reference: BBAMCR 17634

To appear in: BBA - Molecular Cell Research

Received date: 9 March 2015

Revised date: 15 July 2015
Accepted date: 16 July 2015

Please cite this article as: Rodrigo Cáceres, Majdouline Abou-Ghali, Julie Plastino,
Reconstituting the actin cytoskeleton at or near surfaces in vitro, BBA - Molecular Cell
Research (2015), doi: 10.1016/j.bbamcr.2015.07.021

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Reconstituting the actin cytoskeleton at or near surfaces in vitro

Rodrigo Cáceres a,b,c,d, Majdouline Abou-Ghali a,b,c, Julie Plastino*a,b,c

T
a
Institut Curie, Centre de Recherche, Paris, F-75248 France

IP
b
Centre National de la Recherche Scientifique, Unité Mixte de Recherche 168, Paris, F-75248
France

R
c
Université Pierre et Marie Curie, Paris F-75248, France

SC
d
Université Paris Descartes, Paris F-75248, France

NU
*correspondance julie.plastino@curie.fr MA
D
P TE
CE
AC

1
 ACCEPTED MANUSCRIPT

Abstract
Actin filament dynamics have been studied for decades in pure protein solutions or in cell
extracts, but a breakthrough in the field occurred at the turn of the century when it became
possible to reconstitute networks of actin filaments, growing in a controlled but physiological

T
manner on surfaces, mimicking the actin assembly that occurs at the plasma membrane during

IP
cell protrusion and cell shape changes. The story begins with the bacteria Listeria
monocytogenes, the study of which led to the reconstitution of cellular actin polymerization

R
on a variety of supports including plastic beads. These studies made possible the

SC
development of liposome-type substrates for filament assembly and micropatterning of actin
polymerization nucleation. Based on the accumulated expertise of the last 15 years, many

NU
exciting approaches are being developed, including the addition of myosin to biomimetic
actin networks to study the interplay between actin structure and contractility. The field is
MA
now poised to make artificial cells with a physiological and dynamic actin cytoskeleton, and
subsequently to put these cells together to make in vitro tissues.
D
P TE
CE
AC

2
 ACCEPTED MANUSCRIPT

1. Introduction
Actin is a protein that exists in a globular soluble form and in an assembled filamentous form,
echoing a common theme observed in other types of cytoskeleton like microtubules and
intermediate filaments. Cell shape changes in general, including cell motility, cell division

T
and cancer cell invasion, are due in part to the controlled assembly of actin into filamentous

IP
networks that can push membranes or contract in the presence of the molecular motor myosin
thus leading to cell shape changes. The fact that actin filaments are polar, with a dynamic

R
barbed end that grows and shrinks more quickly than the pointed end, is important for the

SC
directionality of network growth and for myosin motor activity.

NU
Actin has been studied since the 1940s when it was first isolated from muscle. By the time
the last century was drawing to a close, the dynamics of individual actin filaments had been
MA
well characterized in vitro [1] and much had been discovered about other factors that
interacted with both the globular and filamentous forms of actin [2]. The great step forward at
the turn of the century was the successful recreation of dynamic actin networks growing at
D

surfaces in a controlled fashion using cellular components, a departure from previous single
TE

filament studies where polymerization was generally occurring in the bulk solution. This
review will be about the progress over the last 15 years in the field of reconstitution of
P

dynamic actin and acto-myosin networks at surfaces or under confinement, and how
CE

technological advances have been used to further our understanding of cellular actin
dynamics. Other excellent reviews on reconstitution have been published over the last 5 years
concentrating on actin and adhesion, membrane-bound actin and single filament dynamics [3-
AC

7]. The focus here is actin and acto-myosin networks at or near surfaces in vitro, to mimic
cellular confinement and geometry.

2. The beginnings of actin network reconstitution

2.1. Listeria in cells
Somewhat surprisingly, most modern approaches to studying actin networks in vitro can trace
their inspiration back to the food-borne pathogen Listeria monocytogenes (Fig. 1). This
bacterium propels itself in the host cell cytosol not by swimming with a flagellum, but by
building a network of filamentous actin behind itself, dubbed an actin tail or actin comet due
to its appearance by electron and light microscopy (reviewed in [8]). What made this motility
mode interesting to the general cell biology community was the discovery that the bacteria
produced a single factor necessary for its motility, the ActA protein, which was displayed on

3
 ACCEPTED MANUSCRIPT

its surface and was responsible for forming the actin comet from host cell components
(reviewed in [9]). In addition landmarking experiments in the actin network of moving cells
and in Listeria tails showed that both processes involved insertion of newly polymerized actin
at the cell membrane or bacterial surface, and this was hypothesized to be the driving force for

T
propulsion in both cases [10, 11]. It was quickly realized by pioneers in the field that the

IP
Listeria actin network could be a powerful tool to study the biochemical basis of mammalian
actin assembly, in isolation from cell signaling and adhesion. This discovery also opened up

R
new avenues for studying how actin assembly created movement from a physical perspective

SC
since bacterial movement was a more tractable object to manipulate and model than an entire
cell [12, 13]. We will discuss here Listeria motility, but other pathogens with similar motility

NU
mechanisms have also been useful in the study of actin-based motility [14].
MA
Initial experiments involved observation of Listeria movement in living cells. Such studies
revealed that many host cell actin-binding proteins were present in the Listeria comet tail
([15] and references therein). Further this type of experiment led to more unexpected results,
D

such as the fact that the actin tail composition changed depending on the intracellular
TE

location: in the cell body, comets contained -actinin, while in cell protrusions, comets shed
-actinin concomitant with an evolution of the comet structure toward an aligned unbranched
P

array of long filaments [16]. Information about how the actin network was constructed was
CE

also gleaned from altering the ActA protein itself and observing how this changed Listeria
motility in cells, notably identifying the Arp2/3 complex and Ena/VASP binding domains as
AC

important motility motifs [17, 18]. However the limitations of this approach quickly became
apparent. For example, a back-to-back study of Listeria motility in cells expressing different
forms of Ena/VASP proteins as compared to the movement of the cells themselves showed
that cell movement and Listeria movement required different domains of Ena/VASP [19, 20].
This perplexing result could have resulted from off-target effects, including mislocalization of
the mutant proteins in the host cells, and changes in the internal structure of the host cell that
could have decreased or enhanced Listeria motility. Indeed other studies showed that the
mechanical inhomogeneity of the cell interior altered the motile behavior of Listeria [21].

2.2. Listeria in cell extracts and pure protein mixes

The cell interior was too complex of a place to conduct controlled biochemical motility
assays, and physical manipulations were rendered difficult. The solution to the confounding
effects of the biochemical and mechanical heterogeneity of the cell interior was the use of cell
4
 ACCEPTED MANUSCRIPT

extracts, homogenous cytosolic preparations lacking organelles and cell membrane. Although
not without its own challenges, mostly associated with obtaining cell extracts sufficiently
concentrated in cytoskeleton factors that were not even entirely known at the time, cell
extracts were successfully used to perform some first quantitative physical and biochemical

T
characterizations. For example Listeria actin tail elasticity was measured using optical

IP
tweezers, and the roles of profilin and Ena/VASP proteins in Listeria movement were
examined [22-24]. At about the same time, great advances were being made in the

R
understanding of how actin assembly was catalyzed in cells. A major step was the discovery

SC
of the Arp2/3 complex as a weak catalyzer or “nucleator” of actin assembly that made
branches from the sides of existing filaments, and the subsequent finding that the Listeria

NU
ActA protein and the mammalian nucleation promoting factors (NPFs) WASp and Scar
activated the activity of the Arp2/3 complex [25-27]. All together these findings paved the
MA
way for the next great advance: the reconstitution of Listeria motility in a mix of pure proteins
[28]. The purified protein mix provided tight control of biochemical parameters, and is still
today the method of choice for studying actin-based motility, especially for attaining the
D

reproducibility needed for quantitative measurements.

TE

However cell extracts should not be neglected. The study of a pure protein can reveal its
P

mechanism in isolation, but not necessarily its mode of action in vivo in association with other
CE

proteins. A case in point is ADF/cofilin, an actin filament fragmenting protein. When pure
ADF/cofilin was mixed with pure actin filaments in conditions where ADF/cofilin fully
decorated the filaments, ADF/cofilin lost its ability to sever [29]. This was perplexing since
AC

high ratios of ADF/cofilin to actin are in fact physiological in some cell types. Recent results
using cell extracts showed that an additional factor, Aip1, was present in cytosol that
permitted ADF/cofilin to efficiently sever and disassemble actin at high ratios [30], although
the exact mode of action of Aip1 is the subject of some controversy [31-33]. Use of cell
extracts also permitted other exciting developments such as the reconstruction of complex
actin structures like the cleavage furrow in cytokinesis [34]. Recent advances make possible
the production of mutant extracts to study individual proteins while retaining the complexity
of the cell cytosol and the preparation of staged extracts to examine how actin assembly varies
with the cell cycle [35, 36].

5
 ACCEPTED MANUSCRIPT

3. The next generation

3.1. Replacing Listeria with beads

The first reconstituted motility systems using Listeria set the stage for the next generation of
in vitro systems where the pathogen was replaced by a bead or other particle coated with the

T
ActA protein (Fig. 1). This allowed for control of the size and properties of the cargo and the

IP
density and nature of the activating protein on the surface, including, importantly, the use of
mammalian factors (next section).

R
SC
The first successful bead systems were performed with ActA-coated particles in cell extracts
[37]. This study brought to light one of the stumbling blocks of working with particles in the

NU
place of Listeria: homogenous distribution of the ActA protein on the bead surface led to
homogenous actin growth, which had to undergo “symmetry breaking” to form a polarized
MA
actin network and directional motility. Symmetry breaking was shown to depend on particle
size, coating density and the concentration of the cell extract, and could be circumvented by
preparing artificially asymmetric beads via silicon monoxide shadowing [37, 38]. Studies of
D

such comets allowed for the important demonstration that actin comet tails observed by
TE

electron microscopy had a similar dendritic organization to that found in the lamellipodia of
moving cells, thus further validating the use of the bead system as a mini-lamellipodium
P

mimic [39].
CE

Although an impediment to forming actin comets, symmetry breaking was an interesting topic
AC

in and of itself, and much was learned about actin network mechanics by observing the
growth and rupture of actin networks on spherical beads. In particular it was demonstrated
that the network had elastic properties, due to its entangled nature, and stresses could develop
in the network and affect growth dynamics [40, 41]. Later with the purified protein mix,
symmetry breaking on beads was thoroughly characterized and it was shown that stress build-
up drove the polarization of the actin network and that stress development depended in
predictable ways on the biochemical components of the protein mixture and the balance
between nucleation of new filaments, capping and crosslinking [42-44].

3.2. What to coat the beads with?

ActA-coated beads are less employed today, but these original studies opened the door to
grafting beads with the mammalian equivalent of ActA, the WASP/WAVE/Scar proteins.

6
 ACCEPTED MANUSCRIPT

Reconstitution of actin comet tails and motility of beads coated with the NPF WASP in
bovine brain extracts was the first entirely mammalian reconstitution of actin-based motility
[45]. Subsequently the WASP proteins and the related Scar/WAVE molecules were picked
apart by absorbing different protein fragments to bead surfaces and observing which domains

T
gave optimal actin network growth and optimal motility in cell extracts and pure protein

IP
mixes [46-48]. Different domains from different actin-binding proteins were also absorbed
simultaneously and in different proportions to bead surfaces, for example to recruit and

R
activate the Arp2/3 complex in varying proportions with Ena/VASP proteins [49]. When

SC
formin proteins were identified as actin polymerization nucleators that produced unbranched
networks, in contrast to the Arp2/3 complex-based branched networks, formin-based actin

NU
assembly and movement were also reproduced on bead surfaces [50, 51]. Given this history,
it is remarkable that no one has yet recreated Arp2/3 complex-based and formin-based
MA
nucleation together on a bead surface, despite the biological relevance to the lamellipodium
where both nucleation systems co-exist and actin networks are generally mixes of branched
and unbranched filaments [52, 53]. This is particularly pertinent given a recent study that
D

showed that the Arp2/3 complex and formin worked together in a mechanism where the new
TE

filament ends created by the Arp2/3 complex were captured and elongated by the formin
FMNL2 [54]. However other studies showed that formin and Arp2/3 complex compete for
P

actin monomers in cells [55], and are not favored by the same conditions in profilin in vitro
CE

[56], so reconstitution of the two activities together may be a challenge.

In general exotic surface coatings remain rare in the biomimetic field, and the predominant
AC

activating proteins used today in in vitro systems are human WASP protein fragments, in
particular the VCA domain that binds and activates the Arp2/3 complex or its variant pVCA
that additionally encompasses the proline-rich portion of WASP that binds profilin actin.
VCA is also called WA, due to vocabulary created simultaneously by different labs [57-59].
The pVCA construct is more effective for Arp2/3 complex activation than VCA when
monomeric actin is bound with profilin [27]. Indeed most modern reconstitution studies use
high concentrations of globular actin bound with profilin to prevent spontaneous nucleation, a
closer mimic of actual conditions in cell cytosol and a departure from the original pure protein
reconstitution system which used a reservoir of prepolymerized filamentous actin to maintain
a low but stable concentration of actin monomers via depolymerization [28, 60].

7
 ACCEPTED MANUSCRIPT

The choice of pVCA from WASP as the most-used NPF is more motivated by history than by
physiology. WASP is in fact a protein that is only found in hematopoietic cells, while the
closely-related N-WASP protein is ubiquitous, but was discovered later (reviewed in [61]).
N-WASP-coated beads were used in some studies [62, 63], and it is the VCA domain of

T
human N-WASP that is currently commercially available. N-WASP is a more effective

IP
Arp2/3 complex activator than either WASP or WAVE/Scar due to the enhanced acidity of
the A domain in the case of N-WASP, not as originally believed due to the extra V domain

R
that N-WASP proteins contain [64]. WAVE/Scar-derived bead coatings have been used for

SC
some studies, but less extensively than the other NPFs [46, 65]. WAVE proteins exist in
regulatory complexes, which are impossible to mimic in pure protein mixtures although the

NU
WAVE regulatory complex has been successfully recruited to membrane-coated glass beads
to form actin comets in cell extracts [66]. In the cell, NPFs have very different roles
MA
downstream of signaling cascades: WAVE/Scar proteins are involved in lamellopodial
protrusion, while WASP proteins are implicated in filopodia formation and endocytosis (for
review [67]). However, as far as biomimetics are concerned, where the regulatory portions of
D

the NPFs are removed, the different NPFs can be used interchangeably since the VCA portion
TE

of the different NPFs give the same end product: an Arp2/3 complex-branched network.
P

3.3. The power of the bead system in the pure protein mix
CE

The combination of the bead system with the pure protein mix changed the face of how actin
polymerization was studied. Most importantly it made possible a type of biophysical
experiment that had been impossible before, namely varying biochemical and physical
AC

parameters and observing how that changed actin assembly and motility. For example it was
observed that simply changing particle size or bead-coating density could completely change
how the actin comet created movement, switching between continuous and periodic, even
though biochemical conditions were identical [68]. Controlled force measurements also
became possible in a variety of different experimental set-ups [62, 69]. Bead/pure protein
mixes were also used to study the role of the important actin factor, capping protein, showing
that capping protein restricted polymerization to the surface via promotion of Arp2/3 complex
activity [70, 71].

Bead speeds were a particularly easy parameter to measure while changing the biochemistry
of the mix. As one example, this approach was used to resolve the confusion concerning
Ena/VASP proteins and Listeria motility mentioned previously. When recruited to the bead

8
 ACCEPTED MANUSCRIPT

surface, Ena/VASP proteins were shown to indeed increase bead speed and different mutants
of Ena/VASP showed concordant effects on beads and on an in vivo cell motility event [49,
65, 72]. However the relation between actin polymerization and particle speed is a complex
one. It has been observed since the conception of the pure protein mix that movement

T
velocity has a bell-curve dependence on the concentration of polymerization factors: both too

IP
much and too little of a given component can reduce speed [28]. In the case of Ena/VASP for
example, under different conditions than the study cited above, it was observed that a bead

R
that was already moving very efficiently displayed drastically reduced motility when treated

SC
with Ena/VASP, concomitant with the production of a much denser comet tail (Fig. 2). So it
seems that when motility is optimal, adding factors that increase polymerization (like

NU
Ena/VASP or even the Arp2/3 complex) can slow bead motility and this is something to keep
in mind when using bead velocity as a read-out of protein function.
MA
4. Polymerization from soft, fluid and deformable substrates
The work on beads spawned a whole other branch of the reconstitution family (Fig. 1)
D

involving polymerization on an assortment of fluid and sometimes deformable substrates like

TE

oil droplets, liposomes, lipid-coated beads or supported bilayers, moving one step closer to
the real conditions for actin polymerization at a cell membrane bilayer.
P
CE

The first of such studies involved the absorption of a His-tagged form of ActA to liposomes
containing nickel lipids and incubation in cell extracts or cell extracts supplemented in the
Arp2/3 complex to form actin comets [73, 74]. Several interesting observations came out of
AC

these studies, observations that were corroborated subsequently under different conditions:
using the mammalian NPFs VCA-WASP and N-WASP absorbed to liposomes or non-
specifically to oil droplets and incubated in either cell extracts or purified protein mixes [75-
77]. Although liposomes were more physiological, the advantage of oil droplets was that the
surface tension was known so the curvature of the droplet surface could be used to calculate
stresses exerted by the growing actin cytoskeleton. One of the main findings from such
studies was, first of all, a direct visual proof of the elastic squeezing effect evoked to explain
symmetry breaking, mentioned previously. The growth of an actin gel on a convex surface
created compressive or squeezing stresses, and this could be clearly seen with both liposomes
and oil droplets as a deformation from spherical shape (Fig. 3a, 3b). Furthermore it was
shown that the actin comet exerted retarding or pulling forces on its substrate, presumably due
to transient attachments between the actin network and the surface-bound NPFs mediated by

9
 ACCEPTED MANUSCRIPT

the Arp2/3 complex. As a result, the NPFs on the fluid surface were convected under the
comet (Fig. 3c). In line with this, another study using the bead system showed that cortactin
enhanced motility by releasing NPF molecules from new branches [78]. Another proposed
mechanism for transient network-surface attachment was the binding of the WH2 (or V)

T
domain of NPFs to filament barbed ends, an interaction that was mediated by monomeric

IP
actin, giving convection of NPFs on lipid-coated glass beads [79]. WASP/WAVE WH2
domains do not bind profilin-actin [80], the predominant form of actin in vivo so, in the cell, a

R
combination of attachment via the Arp2/3 complex and WH2 domains may be occurring.

SC
From all this, it is clear that actin growth exerts both protrusive and braking forces on the
objects it acts upon.

NU
However much was also gleaned from biomimetic membrane systems in conjunction with
MA
actin polymerization in the absence of motility (for review [7]). For example actin
polymerization was shown to induce phase separation of lipids in giant vesicles grafted with
N-WASP, incubated in actin and the Arp2/3 complex [81]. In a similar experiment, the
D

branched actin network produced by Arp2/3 complex-based polymerization was observed to

TE

be reorganized into bundled filopodia-type structures by the deformable lipid bilayer [82].
Even simpler, and in a continuum with approaches using lipid-coated glass beads, actin
P

polymerization was reproduced on supported lipid bilayers. In particular filopodia formation

CE

was recreated on such bilayers, showing that recruitment of biochemical factors from the cell
extract gave spontaneous self-assembly of the bundled structure in the absence of membrane
deformation [83].
AC

Overall the actin network-on-liposome/droplet systems were a great advance in the field
because they brought information as to the interplay between actin assembly and lipid bilayer
properties and also opened the door to looking at actin-based deformations. Supported
bilayers as a subset of this family have the advantage that they are easier to manipulate
physically and image by techniques such as Total Internal Reflection Microscopy (TIRF), but
give up the deformability of the liposome system and reduce the mobility of factors in the
membrane via friction with the support [7].

10
 ACCEPTED MANUSCRIPT

5. Expanding the biomimetic repertoire

5.1. Confining physiologically nucleated dynamic actin networks
There is nothing new about encapsulating actin polymerization. For decades people have
been incorporating monomeric actin into liposomes, triggering polymerization and then

T
observing shape changes. A non-exhaustive list of such studies include [84-88]. Some

IP
studies included non-physiological bonds between the encapsulated actin network and the
liposome inner leaflet, such as the linking of biotin actin to biotin lipids via streptavidin [89].

R
Similar experiments have been performed with pure actin and actin-binding proteins or with

SC
cell extracts confined in stabilized aqueous-in-oil emulsions, two examples of which are [90,
91]. More recently actin polymerization has been confined in microchambers [92]. In all

NU
cases restricting actin polymerization led to interesting phenomena including self-
organization, which were not seen in unconfined solutions. This can be understood in the
MA
larger framework of how confinement changes biological processes, including cytoskeleton
dynamics [93].
D

A new development concerning confined actin polymerization builds on these experiments,

TE

but with several additional characteristics that were previously absent. Namely, to truly
reproduce cellular dynamics, the actin network should be growing from the surface via
P

localized actin polymerization nucleation. This means that there are transient attachments
CE

between the network and the surface, and the barbed ends are growing mostly toward or near
the surface. The actin network should also be depolymerizing, and monomers continually
recharging with ATP and repolymerizing to make a dynamic network. These aspects are
AC

important for mimicking not only lamellipodia-type protrusions, but also for reconstituting
other cytoskeletal organelles as we will see in the next section.

Advances have been made in this direction over the last few years. Liposomes were made
from native membranes and swelled in the presence of actin, with or without the membrane-
actin crosslinking proteins ankyrin/spectrin. In the presence of ankyrin/spectrin, polymerized
actin was anchored and bundled at the membrane [94]. This was a physiological link,
however the filaments were not dynamic. At about the same time, liposomes were made by a
different technique, the inverted emulsion technique, whereby the reconstituted motility mix
of pure proteins described earlier was encapsulated in low salt conditions that prevented
polymerization and then polymerization was triggered by inserting pores in the membrane to
allow passage of salts [95]. Importantly polymerization occurred preferentially at the

11
 ACCEPTED MANUSCRIPT

membrane because a VCA protein was specifically bound there by interaction of its histidine
tag with nickel lipids in the membrane, and additionally this actin layer was shown to be
actively turning over due to the presence of actin depolymerizing and recycling factors in the
liposome interior. This study produced for the first time a dynamic membrane-associated

T
actin structure in a liposome, polymerized in a physiological manner. Subsequently the

IP
inverted emulsion technique was used for actin/actin-binding protein encapsulation and
micropipette aspiration experiments to show that the membrane-associated actin layer was

R
determinant for the mechanical properties of the liposome [96, 97]. Additionally membrane-

SC
bound actin layers have since been formed in aqueous-in-oil emulsions, using interface-
targeted ActA protein and cell extracts [98]. These actin networks were shown to not only be

NU
actively turning over, but also were capable of auto-organization to break symmetry. An
added motivation to use liposome-type biomimetic systems is to study proteins that recognize
MA
or impose membrane curvature and also interface with the actin cytoskeleton, such as BAR
domain proteins [99].
D

5.2. Patterning actin assembly

TE

Another innovation in the actin biomimetics field is that of making defined actin structures
via micropatterning of nucleation sites [100]. In some ways this is similar to the previous
P

challenge, but the confinement is imposed by the filament source instead of being created by
CE

the envelope. A pioneering study showed that the angle and distance between nucleation sites
for actin assembly determined the proportion of parallel bundles versus anti-parallel structures
within a given actin network although the biochemistry of the networks was identical [101].
AC

This showed that the geometry of filament growth could determine macroscopic structure
formation, something that had previously been ascribed to actin-binding proteins. However in
cells there is surely a mixture of both geometrical and biochemical control, for when the anti-
parallel actin bundler -actinin was added in high concentrations into the actin polymerization
mix, antiparallel filament structures were favored even though the geometry dictated
predominant parallel bundle formation [101].

6. Reconstituting acto-myosin contractility in vitro in cell-like systems

The stage is now set for one of the next big challenges in actin biomimetics: reproducing the
acto-myosin contractile structure found in non-muscle cells juxtaposed to the plasma
membrane, an organelle commonly called the cell cortex. This mixed network of actin
filaments and myosin motors dynamically polymerizes, depolymerizes and contracts, while at
12
 ACCEPTED MANUSCRIPT

the same time being transiently linked to the plasma membrane that it deforms to produce cell
shape changes. In the well-studied contractile system of the muscle sarcomere, unbranched
actin filaments are arranged in an anti-parallel manner so as to enable myosin-based
contraction. In non-muscle cells, the actin network in the cell cortex is a random array of

T
branched and unbranched actin filaments, not organized like in a muscle sarcomere [52, 53].

IP
The question then is: how does the cortex contract efficiently? To answer this, the
previously-described techniques are being used to produce cell-like dynamic actin networks,

R
but now containing myosin.

SC
6.1. Interplay between actin organization and myosin contractility

NU
As would be predicted from consideration of how myosin functions, it has been shown
experimentally that the overall actin architecture can modify where and how effectively
MA
myosin contracts the actin network. The micropatterning approach described above was used
to create different network geometries, mixed parallel bundles and anti-parallel structures.
When myosin was added to this network, it preferentially contracted anti-parallel structures
D

although it decorated parallel bundles as well [102]. Myosin was capable of contracting
TE

entangled branched networks, albeit much more slowly. However this appeared to be due to
the spontaneous occurrence of anti-parallel structures within such networks that were the real
P

substrate for myosin function [102]. A very different experimental approach involving acto-
CE

myosin layers near but not attached to supported lipid bilayers also showed that a disordered
actin network was efficiently contracted by myosin, but only above a critical myosin
concentration [103].
AC

When a static disordered acto-myosin network was attached to the outside or the inside of a
lipsome, the outcome of contraction was modulated by the attachment to the bilayer [104]. In
the “outside geometry”, the balance between contraction and membrane attachment
determined whether the acto-myosin network compacted and peeled off the exterior of the
lipsome or whether the network contracted and crushed the liposome. In the “inside
geometry”, contraction either occurred on the bilayer or pulled off the bilayer depending on
attachment strength. Taking this experiment one step further, actin was polymerized in the
outside geometry with a physiological attachment to the bilayer via a membrane-bound VCA
molecules, with the Arp2/3 complex, capping protein and profilin to mimic cellular actin
polymerization [105]. It was observed that both myosin contraction and actin polymerization
contributed to stress build-up in this system, and importantly, that the cocktail of actin-

13
 ACCEPTED MANUSCRIPT

binding proteins determined the window where myosin produced contraction. All together,
these results emphasize the importance of the geometry of the network, its attachment to the
bilayer and the biochemistry of network formation for determining myosin contractility. This
is why there is much to be learned by performing biomimetic experiments, which could give

T
very different behavior from pure acto-myosin networks in the absence of constraints,

IP
attachments and physiological polymerization.

R
Another aspect of actin architecture that could affect myosin contractility efficiency is the

SC
presence of crosslinkers. The contraction of the anti-parallel regions of the actin network
grown from micropatterns was slower in the presence of the anti-parallel cross-linking protein

NU
-actinin, presumably due to resistance to filament sliding imposed by the cross-links [102].
However a macroscopic contraction assay using suspended actin layers showed that the
MA
connectivity conferred by actin cross-linking proteins was necessary for a global contraction
[106]. These biomimetic studies show that cross-linking may play a role in controlling how
the network contracts. Indeed cross-linking proteins are abundant in the acto-myosin cell
D

cortex [107], and myosin-regulatory roles for the actin-binding proteins fascin and
TE

ADF/cofilin, sometimes contradictory in the latter case, have been recently reported in cells
[31, 108-110]. These issues will be one of the many questions to address in the future with
P

biomimetics,
CE

6.2. Myosin contractility as a disassembly agent

Contraction was expected to change the organization of the actin network by compacting it.
AC

What was somewhat unexpected was the observation that motor activity also severed and
dismantled the network. This had been observed with actin bundles in bulk assays [111].
However as concerns biomimetic networks, this depolymerization effect was most clearly
demonstrated with the micropatterning experiments where contraction of the anti-parallel
portions of the network led to their disappearance, and seemingly liberated monomeric actin,
as evidenced by an enhanced growth of the parallel bundles in the assay [102]. This
macroscopic effect reflected what was happening on the single filament level, where myosin
activity was observed to buckle and fragment filaments that were attached to a lipid bilayer
[112, 113].

14
 ACCEPTED MANUSCRIPT

7. Conclusion
One of the next challenges for biomimetics is to put together all that we have learned over the
last 15 years in order to produce the ideal artificial acto-myosin in vitro system (Fig. 4). The
goal is to reconstitute inside a cell-like confinement the acto-myosin network, while

T
preserving the architecture of the network as found in living cells, its attachment to the bilayer

IP
and the biochemistry of network formation, all of which appear to be important for
determining myosin contractility. Such systems should allow for the in vitro study of shape

R
changes and spontaneous oscillations. Down the road, the next step will be to include

SC
adhesion to the substrate to make motile biomimetic cells, and adhesion to adjacent “cells” to
build up artificial tissues in order to mimic and study collective shape changes.

ACKNOWLEDGEMENTS
NU
MA
We thank Cécile Sykes for discussions and reading of the manuscript. This work
was funded by the Fondation pour la Recherche Médicale (Grant DEQ20120323737). R.C.
was funded by a PhD fellowship from ITMO Cancer.
D
P TE
CE
AC

15
 ACCEPTED MANUSCRIPT

FIGURE LEGENDS

Fig. 1: The family tree of biomimetic systems of actin motility and dynamics.
The original inspiration came from Listeria motility in cells a), which led to studies of
Listeria in cell extracts and pure protein mixes b). The next generation of in vitro systems can

T
be split into two groups, one involving reconstitution on solid supports such as beads c) and

IP
the other involving the use of fluid, deformable substrates such as liposomes d). ActA from
Listeria was used to coat the beads and liposomes, but also mammalian nucleation promoting

R
factors (NPFs) of the WASP/WAVE/Scar family. The recent innovations in each branch of

SC
the family consist of reconstitution of actin dynamics on micropatterns on one hand e), and
reconstitution of actin cortices inside liposomes on the other hand f). The lateral double-

NU
headed arrows indicate cross-talk between the different systems. a) Reprinted from [114]:
Cell, vol. 68, C. Kocks, E. Gouin, M. Tabouret, P. Berche, H. Ohayon, P. Cossart, L.
MA
monocytogenes-induced actin assembly requires the actA gene product, a surface protein, 521-
531(1992), with permission from Elsevier. b) Adapted by permission from Macmillan
Publishers Ltd: Nature, [28], 1999. c) Adapted by permission from Macmillan Publishers
D

Ltd: Nature, [68], 2002. d) Adapted by permission from the National Academy of Sciences:
TE

PNAS, [74], 2003. e) Adapted by permission from Macmillan Publishers Ltd: Nature
Materials, [101], 2010. f) Reprinted from [95]: Biophysical Journal, vol. 96, L.-L. Pontani, J.
P

Van der Gucht, G. Salbreaux, J. Heuvingh, J.-F. Joanny, C. Sykes, Reconstitution of an actin
CE

cortex inside a liposome, 192-198 (2009), with permission from Elsevier.

AC

Fig. 2: Enhancing polymerization does not always increase bead motility.

a) When motility is very fast (2-3 m/min), addition of VASP b) slows the beads down
(below 1m/min) even though the comet is denser. So the effect of VASP on motility seems
to depend on the initial state of the system, and when speed is already optimal, adding an
enhancing molecule like VASP does not have the expected effect. Images taken at about 10-
15 minutes reaction time of PRD-VCA-WAVE-coated beads in reconstituted motility mix as
described in [65], but with Arp2/3 complex purchased from Cytoskeleton. Phase contrast
microscopy. Comet appears as a dark streak behind the white bead. Since there is no
depolymerization in this system, comet length is proportional to bead velocity. Images M.
Abou-Ghali, 2014.

16
 ACCEPTED MANUSCRIPT

Fig. 3: Actin polymerization on deformable, fluid supports.

a) and b) Oil droplets are deformed by the actin comet, depending on how the comet is
organized. When the oil droplet is grafted with VCA a), motility is slow, comets are uniform

T
and dense and the droplet is deformed in a pear shape. When the droplet is coated with a mix

IP
of VCA and PRO b), a fragment of the ActA protein that recruits VASP, movement is rapid,
the comet is partially hollow and the droplet is therefore deformed differently than in a) into a

R
kiwi shape. See also [77]. Phase contrast microscopy. c) On the fluid surface of the oil

SC
droplet, VCA (green) is enriched under the comet (actin in red), as observed by the dimmer
intensity of VCA at the front of the droplet. The droplet is undergoing jumping movement.

NU
For more details see [77]. Confocal fluorescence microscopy. All images Léa Trichet, 2004-
2005.
MA
Fig. 4: The ideal artificial acto-myosin in vitro system.
The main characteristics include 1) the system has a cell-like geometry confined by a lipid
D

bilayer to mimic the cell membrane, 2) actin filament nucleation occurs at the membrane by
TE

physiological factors such as the Arp2/3 complex or another nucleator such as formin, 3)
attachment to the membrane is ensured by transient links via the Arp2/3 complex and
P

physiological actin filament-membrane linkers such as ezrin, 4) non-muscle myosins are

CE

included in the artificial cell interior, 5) actin filaments disassemble either due to the activity
of proteins such as ADF/cofilin or to the buckling/severing action that results from myosin
contraction and 6) the actin monomers thus liberated are recycled to the cell membrane for
AC

subsequent rounds of nucleation. Like in cells, spontaneous formation of filaments in the

“cell” interior is inhibited by maintaining free actin in its profilin-bound form. In grey are
depicted the future of such systems where, in addition to all the characteristics listed above,
the artificial cell is also capable of adhering to its substrate and to its neighboring “cells” via
its cytoskeleton and transmembrane proteins, thus mimicking epithelial tissues.

17
 ACCEPTED MANUSCRIPT

REFERENCES
[1] T.D. Pollard, Rate constants for the reactions of ATP- and ADP-actin with the ends of
actin filaments, J. Cell Biol., 103 (1986) 2747-2754.
[2] T.D. Pollard, L. Blanchoin, R.D. Mullins, Molecular mechanisms controlling actin
filament dynamics in nonmuscle cells, Annu. Rev. Biophys. Biomol. Struct., 29 (2000) 545-
576.

T
[3] D. Brüggemann, J.P. Frohnmayer, J.P. Spatz, Model systems for studying cell adhesion

IP
and biomimetic actin networks, Beilstein J. Nanotechnol., 5 (2014) 1193-1202.
[4] S.K. Vogel, P. Schwille, Minimal systems to study membrane–cytoskeleton interactions,

R
Curr. Opin. Biotech., 23 (2012) 758–765.
[5] R.D. Mullins, S.D. Hansen, In vitro studies of actin filament and network dynamics, Curr.

SC
Opin. Cell Biol., 25 (2013) 6-13.
[6] A.P. Liu, D.A. Fletcher, Biology under construction: in vitro reconstitution of cellular
function, Nat. Rev. Mol. Cell Biol. , 10 (2009) 644-650.

NU
[7] M. Loose, P. Schwille, Biomimetic membrane systems to study cellular organization, J.
Struct. Biol., 168 (2009) 143-151.
[8] L.G. Tilney, M.S. Tilney, The wily ways of a parasite: induction of actin assembly by
Listeria, Trends in Microbiology, 1 (1993) 25-31.
MA
[9] P. Cossart, Actin-based bacterial motility, Curr. Opin. Cell Biol., 7 (1995) 94-101.
[10] J.A. Theriot, T.J. Mitchison, Actin microfilament dynamics in locomoting cells, Nature,
352 (1991) 126-131.
[11] J.A. Theriot, T.J. Mitchison, L.G. Tilney, D.A. Portnoy, The rate of actin-based motility
D

of intracellular Listeria monocytogenes equals the rate of actin polymerization, Nature, 357
(1992) 257-260.
TE

[12] C.S. Peskin, G.M. Odell, G.F. Oster, Cellular motions and thermal fluctuations: The
Brownian ratchet, Biophys. J., 65 (1993) 316-324.
[13] A. Mogilner, G. Oster, Cell Motility Driven by Actin Polymerization, Biophys. J., 71
P

(1996) 3030-3045.
CE

[14] E. Gouin, H. Gantelet, C. Egile, I. Lasa, H. Ohayon, V. Villiers, P. Gounon, P.J.

Sansonetti, P. Cossart, A comparative study of the actin-based motilities of the pathogenic
bacteria Listeria monocytogenes, Shigella flexneri and Rickettsia conorii, J. Cell Sci., 112
(1999) 1697-1708.
AC

[15] S. Pistor, T. Chakraborty, U. Walter, J. Wehland, The bacterial actin nucleator protein
ActA of Listeria monocytogenes contains multiple binding sites for host microfilament
proteins, Curr. Biol., 5 (1995) 517-525.
[16] A. Sechi, J. Wehland, J.V. Small, The isolated comet tail pseudopodium of Listeria
monocytogenes: a tail of two actin filament populations, long and axial and short and
random., J. Cell Biol., 137 (1997) 155-167.
[17] J. Skoble, D.A. Portnoy, M.D. Welch, Three regions within ActA promote Arp2/3
complex-mediated actin nucleation and Listeria monocytogenes motility, J. Cell Biol., 150
(2000) 527-537.
[18] J. Skoble, V. Auerbuch, E.D. Goley, M.D. Welch, D.A. Portnoy, Pivotal role of VASP in
Arp2/3 complex-mediated actin nucleation, actin branch-formation, and Listeria
monocytogenes motility, J. Cell Biol., 155 (2001) 89-100.
[19] J.J. Loureiro, D.A. Rubinson, J.E. Bear, G.A. Baltus, A.V. Kwiatkowski, F.B. Gertler,
Critical roles of phosphorylation and actin binding motifs, but not the central proline-rich
region, for Ena/Vasodilator-stimulated Phosphoprotein (VASP) function during cell
migration, Mol. Biol. Cell, 13 (2002) 2533-2546.

18
 ACCEPTED MANUSCRIPT

[20] M. Geese, J.J. Loureiro, J.E. Bear, J. Wehland, F.B. Gertler, A.S. Sechi, Contribution of
Ena/VASP proteins to intracellular motility of Listeria requires phosphorylation and proline-
rich core but not F-actin binding or multimerization, Mol. Biol. Cell, 13 (2002) 2383-2396.
[21] P.A. Giardini, J.A. Theriot, Effects of intermediate filaments on actin-based motility of
Listeria monocytogenes, Biophys. J., 81 (2001) 3193-3203.
[22] F. Gerbal, V. Laurent, A. Ott, M.-F. Carlier, P. Chaikin, J. Prost, Measurement of the

T
elasticity of the actin tail of Listeria monocytogenes, Eur. Biophys. J., 29 (2000) 134-140.
[23] J.A. Theriot, J. Rosenblatt, D.A. Portnoy, P.J. Goldschimdt-Clermont, T.J. Mitchison,

IP
Involvement of profilin in the actin-based motility of L. monocytogenes in cells and cell-free
extracts, Cell, 76 (1994) 505-517.

R
[24] V. Laurent, T.P. Loisel, B. Harbeck, A. Wehman, L. Gröbe, B.M. Jockusch, J. Wehland,
F.B. Gertler, M.-F. Carlier, Role of proteins of the Ena-VASP family in actin-based motility

SC
of Listeria monocytogenes, J. Cell Biol., 144 (1999) 1245-1258.
[25] R.D. Mullins, J.A. Heuser, T.D. Pollard, The interaction of Arp2/3 complex with actin:
nucleation, high affinity pointed end capping, and formation of branching networks of

NU
filaments, Proceedings of the National Academy of Sciences, 95 (1998) 6181-6186.
[26] M.D. Welch, J. Rosenblatt, J. Skoble, D.A. Portnoy, T.J. Mitchison, Interaction of human
Arp2/3 complex and the Listeria monocytogenes ActA protein in actin filament nucleation,
MA
Science, 281 (1998) 105-108.
[27] L.M. Machesky, R.D. Mullins, H.N. Higgs, D.A. Kaiser, L. Blanchoin, R.C. May, M.E.
Hall, T.D. Pollard, Scar, a WASp-related protein, activates nucleation of actin filaments by
the Arp2/3 complex, Proc. Natl. Acad. Sci., 96 (1999) 3739-3744.
[28] T.P. Loisel, R. Boujemaa, D. Pantaloni, M.F. Carlier, Reconstitution of actin-based
D

motility of Listeria and Shigella using pure proteins, Nature, 401 (1999) 613-616.
TE

[29] E. Andrianantoandro, T.D. Pollard, Mechanism of actin filament turnover by severing

and nucleation at different concentrations of ADF/Cofilin, Mol. Cell, 24 (2006) 13-23.
[30] A.V. Nadkarni, W.M. Brieher, Aip1 destabilizes cofilin-saturated actin filaments by
P

severing and accelerating monomer dissociation from ends, Curr. Biol., 24 (2014) 2749-2757.
[31] Q. Chen, N. Courtemanche, T.D. Pollard, Aip1 promotes actin filament severing by
CE

cofilin and regulates constriction of the cytokinetic contractile ring, J. Biol. Chem., 290
(2015) 2289-2300.
[32] L. Gressin, A. Guillotin, C. Guérin, L. Blanchoin, A. Michelot, Architecture dependence
AC

of actin filament network disassembly, Curr. Biol., 25 (2015) 1437-1447.

[33] S. Jansen, A. Collins, S.M. Chin, C.A. Ydenberg, J. Gelles, B.L. Goode, Single-molecule
imaging of a three-component ordered actin disassembly mechanism, Nat. Commun., 6
(2015) 7202.
[34] P.A. Nguyen, A.C. Groen, M. Loose, K. Ishihara, M. Wühr, C.M. Field, T.J. Mitchison,
Spatial organization of cytokinesis signaling reconstituted in a cell-free system, Science, 346
(2014) 244-247.
[35] Y. Miao, C.C.L. Wong, V. Mennella, A. Michelot, D.A. Agard, L.J. Holt, J.R.I. Yates,
D.G. Drubin, Cell-cycle regulation of formin-mediated actin cable assembly, Proc. Natl.
Acad. Sci. U.S.A., 110 (2013) E4446–E4455.
[36] A. Michelot, A. Grassart, V. Okreglak, M. Costanzo, C. Boone, D.G. Drubin, Actin
filament elongation in Arp2/3-derived networks is controlled by three distinct mechanisms,
Dev. Cell, 24 (2013) 182-195.
[37] L.A. Cameron, M.J. Footer, A. Van Oudenaarden, J.A. Theriot, Motility of ActA protein-
coated microspheres driven by actin polymerization, Proceedings of the National Academy of
Sciences, 96 (1999) 4908-4913.
[38] L.A. Cameron, T.M. Svitkina, D. Vignjevic, J.A. Theriot, G.G. Borisy, Dendritic
organization of actin comet tails, Curr. Biol., 11 (2001) 130-135.

19
 ACCEPTED MANUSCRIPT

[39] T.M. Svitkina, G.C. Borisy, Arp2/3 complex and actin depolymerizing factor/cofilin in
dendritic organization and treadmilling of actin filament array in lamellipodia, J. Cell Biol.,
145 (1999) 1009-1026.
[40] V. Noireaux, R.M. Golsteyn, E. Friederich, J. Prost, C. Antony, D. Louvard, C. Sykes,
Growing an actin gel on spherical surfaces, Biophys. J., 278 (2000) 1643-1654.
[41] J. Plastino, I. Lelidis, J. Prost, C. Sykes, The effect of diffusion, depolymerization and

T
nucleation promoting factors on actin gel growth, Eur. Biophys. J., 33 (2004) 310-320.
[42] J. van der Gucht, E. Paluch, J. Plastino, C. Sykes, Stress release drives symmetry

IP
breaking for actin-based movement, Proc. Natl. Acad. Sci. U.S.A., 102 (2005) 7847-7852.
[43] M.J. Dayel, O. Akin, M. Landeryou, V. Risca, A. Mogilner, R.D. Mullins, In silico

R
reconstitution of actin-based symmetry breaking and motility, PLOS Biol., 7 (2009) 1000201.
[44] A. Kawska, K. Carvalho, J. Manzi, R. Boujemaa-Paterski, L. Blanchoin, J.-L. Martiel, C.

SC
Sykes, How actin network dynamics control the onset of actin-based motility, Proc. Natl.
Acad. Sci. U.S.A., 109 (2012) 14440–14445.
[45] D. Yarar, W. To, A. Abo, M.D. Welch, The Wiskott-Aldrich syndrome protein directs

NU
actin-based motility by stimulating actin nucleation with the Arp2/3 complex, Curr. Biol., 9
(1999) 555-558.
[46] D. Yarar, J.A. D'Alessio, R.L. Jeng, M.D. Welch, Motility determinants in WASP family
MA
proteins, Mol. Biol. Cell, 13 (2002) 4045–4059.
[47] F. Castellano, C. Le Clainche, D. Patin, M.-F. Carlier, P. Chavrier, A WASP-VASP
complex regulates actin polymerization at the plasma membrane, EMBO J., 20 (2001) 5603-
5614.
[48] J. Fradelizi, V. Noireaux, J. Plastino, B. Menichi, D. Louvard, C. Sykes, R.M. Golsteyn,
D

E. Friederich, ActA and human zyxin harbour Arp2/3-independent actin-polymerization

TE

activity, Nat. Cell Biol., 3 (2001) 699-707.

[49] J. Plastino, S. Olivier, C. Sykes, Actin filaments align into hollow comets for rapid
VASP-mediated propulsion, Curr. Biol., 14 (2004) 1766-1771.
P

[50] S. Romero, C. Le Clainche, D. Didry, C. Egile, D. Pantaloni, M.-F. Carlier, Formin is a

processive motor that requires profilin to accelerate actin assembly and associated ATP
CE

hydrolysis, Cell, 119 (2004) 419-429.

[51] A. Michelot, J. Berro, C. Guérin, R. Boujemaa-Paterski, C.J. Staiger, J.-L. Martiel, L.
Blanchoin, Actin-filament stochastic dynamics mediated by ADF/Cofilin, Curr. Biol., 17
AC

(2007) 825-833.
[52] M. Bovellan, Y. Romeo, M. Biro, A. Boden, P. Chugh, A. Yonis, M. Vaghela, M.
Fritzsche, D. Moulding, R. Thorogate, A. Jégou, A.J. Thrasher, G. Romet-Lemonne, P.P.
Roux, E. Paluch, G. Charras, Cellular control of cortical actin nucleation, Curr. Biol., 24
(2014) 1628–1635.
[53] N. Morone, T. Fujiwara, K. Murase, R.S. Kasai, H. Ike, S. Yuasa, J. Usukura, A.
Kusumi, Three-dimensional reconstruction of the membrane skeleton at the plasma membrane
interface by electron tomography, J. Cell Biol., 174 (2006) 851-862.
[54] J. Block, D. Breitsprecher, S. Kühn, M. Winterhoff, F. Kage, R. Geffers, P. Duwe, J.L.
Rohn, B. Baum, C. Brakebusch, M. Geyer, T.E.B. Stradal, J. Faix, K. Rottner, FMNL2 drives
actin-based protrusion and migration downstream of Cdc42, Curr. Biol., 22 (2012) 1005-
1012.
[55] T.A. Burke, J.R. Christensen, E. Barone, C. Suarez, V. Sirotkin, D.R. Kovar,
Homeostatic actin cytoskeleton networks are regulated by assembly factor competition for
monomers, Curr. Biol., 24 (2014) 579-585.
[56] C. Suarez, R.T. Carroll, T.A. Burke, J.R. Christensen, A.J. Bestul, J.A. Sees, M.L. James,
V. Sirotkin, D.R. Kovar, Profilin regulates F-actin network homeostasis by favoring formin
over Arp2/3 complex, Dev. Cell, 32 (2015) 43-53.

20
 ACCEPTED MANUSCRIPT

[57] R. Rohatgi, L. Ma, H. Miki, M. Lopez, H. Kirchausen, T. Takenawa, M. Kirschner, The

interaction between N-Wasp and the Arp2/3 complex links Cdc42-dependent signals to actin
assembly, Cell, 97 (1999) 221-231.
[58] L.M. Machesky, R.H. Insall, Scar1 and the related Wiskott-Aldrich syndrome protein,
WASP, regulate the actin cytoskeleton through the Arp2/3 complex, Curr. Biol., 8 (1998)
1347-1356.

T
[59] H. Miki, S. Suetsugu, T. Takenawa, WAVE, a novel WASP-family protein involved in
actin reorganization induced by Rac, EMBO J., 17 (1998) 6932-6941.

IP
[60] L. Blanchoin, R. Boujemaa-Paterski, C. Sykes, J. Plastino, Actin dynamics, architecture
and mechanics in cell motility, Physiol. Rev., 94 (2014) 235-263.

R
[61] H. Miki, T. Takenawa, Regulation of actin dynamics by WASP family proteins, J.
Biochem., 134 (2003) 309-313.

SC
[62] Y. Marcy, J. Prost, M.-F. Carlier, C. Sykes, Forces generated during actin-based
propulsion: a direct measurement by micromanipulation, Proc. Natl. Acad. Sci. U.S.A., 5992-
5997 (2004).

NU
[63] S. Wiesner, E. Helfer, D. Didry, G. Ducouret, F. Lafuma, M.-F. Carlier, D. Pantaloni, A
biomimetic motility assay provides insight into the mechanism of actin-based motility, J. Cell
Biol., 160 (2003) 387-398.
MA
[64] J. Zalevsky, L. Lempert, H. Kranitz, R.D. Mullins, Different WASP family proteins
stimulate different Arp2/3 complex-dependent actin-nucleating activities, Curr. Biol., 11
(2001) 1903-1913.
[65] S. Havrylenko, P. Noguera, M. Abou-Ghali, J. Manzi, F. Faqir, A. Lamora, C. Guérin, L.
Blanchoin, J. Plastino, WAVE binds Ena/VASP for enhanced Arp2/3 complex-based actin
D

assembly, Mol. Biol. Cell, 26 (2015) 55-65.

TE

[66] V. Koronakis, P.J. Hume, D. Humphreys, T. Liu, O. Horning, O.N. Jensen, E.J. McGhie,
WAVE regulatory complex activation by cooperating GTPases Arf and Rac1, Proc. Natl.
Acad. Sci. U.S.A., 108 (2011) 14449-14454.
P

[67] K.G. Campellone, M.D. Welch, A nucleator arms race: cellular control of actin
assembly, Nat. Rev. Mol. Cell Biol., 11 (2010) 237-251.
CE

[68] A. Bernheim-Groswasser, S. Wiesner, R.M. Golsteyn, M.-F. Carlier, C. Sykes, The

dynamics of actin-based motility depend on surface parameters, Nature, 417 (2002) 308-311.
[69] T. Pujol, O. du Roure, M. Fermigier, J. Heuvingh, Impact of branching on the elasticity
AC

of actin networks, Proc. Natl. Acad. Sci. U.S.A., 109 (2012) 10364-10369.
[70] O. Akin, R.D. Mullins, Capping protein increases the rate of actin-based motility by
promoting filament nucleation by the Arp2/3 complex, Cell, 133 (2008) 841-851.
[71] V. Achard, J.-L. Martiel, A. Michelot, C. Guérin, A.-C. Reymann, L. Blanchoin, R.
Boujemaa-Paterski, A "Primer"-Based Mechanism Underlies Branched Actin Filament
Network Formation and Motility, Curr. Biol., 20 (2010) 423-428.
[72] S. Samarin, S. Romero, C. Kocks, D. Didry, D. Pantaloni, M.-F. Carlier, How VASP
enhances actin-based motility, J. Cell Biol., 163 (2003) 131-142.
[73] A. Upadhyaya, J.R. Chabot, A. Andreeva, A. Samadani, A. van Oudenaarden, Probing
polymerization forces by using actin-propelled lipid vesicles, Proceedings of the National
Academy of Sciences, 100 (2003) 4521-4526.
[74] P.A. Giardini, D.A. Fletcher, J.A. Theriot, Compression forces generated by actin comet
tails on lipid vesicles, Proceedings of the National Academy of Sciences, 100 (2003) 6493-
6498.
[75] H. Boukellal, O. Campas, J.-F. Joanny, J. Prost, C. Sykes, Soft Listeria: actin-based
propulsion of liquid drops, Phys. Rev. E, 69 (2004) 061906/061901-061906/061904.
[76] V. Delatour, E. Helfer, D. Didry, K.H.D. Lê, J.-F. Gaucher, M.-F. Carlier, G. Romet-
Lemonne, Arp2/3 controls the motile behavior of N-WASP-functionalized GUVs and

21
 ACCEPTED MANUSCRIPT

modulates N-WASP surface distribution by mediating transient links with actin filaments,
Biophys. J., 94 (2008) 4890-4905.
[77] L. Trichet, O. Campas, C. Sykes, J. Plastino, VASP governs actin dynamics by
modulating filament anchoring, Biophys. J., 92 (2007) 1081-1089.
[78] O. Siton, Y. Ideses, S. Albeck, T. Unger, A.D. Bershadsky, N.S. Gov, A. Bernheim-
Groswasser, Cortactin releases the brakes in actin-based motility by enhancing WASP-VCA

T
detachment from Arp2/3 branches, Curr. Biol., 21 (2011) 2092-2097.
[79] C. Co, D.T. Wong, S. Gierke, V. Change, J. Taunton, Mechanism of actin network

IP
attachment to moving membranes: barbed end capture by N-WASP WH2 domains, Cell, 128
(2007) 901-913.

R
[80] F. Ferron, G. Rebowski, S.H. Lee, R. Dominguez, Structural basis for the recruitment of
profilin-actin complexes during filament elongation by Ena/VASP, Embo J, 26 (2007) 4597-

SC
4606.
[81] A.P. Liu, D.A. Fletcher, Actin Polymerization Serves as a Membrane Domain Switch in
Model Lipid Bilayers, Biophys. J., 91 (2006) 4064-4070.

NU
[82] A.P. Liu, D.L. Richmond, L. Maibaum, S. Pronk, P.L. Geissler, D.A. Fletcher,
Membrane-induced bundling of actin filaments, Nat. Phys., 4 (2008) 789-793.
[83] K. Lee, J.L. Gallop, K. Rambani, M.W. Kirschner, Self-assembly of filopodia-like
MA
structures on supported lipid bilayers, Science, 329 (2010) 1341-1345.
[84] J.D. Cortese, B. Schwab III, C. Frieden, E.L. Elson, Actin polymerization induces a
shape change in actin-containing vesicles, Proc. Natl. Acad. Sci. U.S.A., 86 (1989) 5773-
5777.
[85] L. Limozin, M. Bärmann, E. Sackmann, On the organization of self-assembled actin
D

networks in giant vesicles, Eur. Phys. J. E 10 (2003) 319-330.

TE

[86] L. Limozin, E. Sackmann, Polymorphism of cross-linked actin networks in giant

vesicles, Phys. Rev. Lett., 89 (2002) 168103.
[87] H. Miyata, H. Hotani, Morphological changes in liposomes caused by polymerization of
P

encapsulated actin and spontaneous formation of actin bundles, Proc. Natl. Acad. Sci. U.S.A.,
89 (1992) 11547-11551.
CE

[88] H. Miyata, S. Nishiyama, K.-I. Akashi, K. Kinosita Jr., Protrusive growth from giant
liposomes driven by actin polymerization, Proc. Natl. Acad. Sci. U.S.A., 96 (1999) 2048-
2053.
AC

[89] F.-C. Tsai, B. Stuhrmann, G. Koenderink, Encapsulation of active cytoskeletal protein

networks in cell-sized liposomes, Langmuir, 27 (2011) 10061–10071.
[90] M. Pinot, V. Steiner, B. Dehapiot, B.-K. Yoo, F. Chesnel, L. Blanchoin, C. Kervrann, Z.
Gueroui, Confinement induces actin flow in a meiotic cytoplasm, Proc. Natl. Acad. Sci.
U.S.A., 109 (2012) 11705-11710.
[91] M.M.A.E. Claessens, M. Bathe, E. Frey, A.R. Bausch, Actin-binding proteins sensitively
mediate F-actin bundle stiffness, Nat. Mater., 5 (2006) 748-753.
[92] J. Alvarado, B.M. Mulder, G.H. Koenderink, Alignment of nematic and bundled
semiflexible polymers in cell-sized confinement, Soft Matter, 10 (2014) 2354-2364.
[93] M.D. Vahey, D.A. Fletcher, The biology of boundary conditions: cellular reconstitution
in one, two, and three dimensions, Curr. Opin. Cell Biol., 26 (2014) 60-68.
[94] D. Merkle, N. Kahya, P. Schwille, Reconstitution and anchoring of cytoskeleton inside
giant unilamellar vesicles, ChemBioChem, 9 (2008) 2673-2681.
[95] L.-L. Pontani, J. Van der Gucht, G. Salbreaux, J. Heuvingh, J.-F. Joanny, C. Sykes,
Reconstitution of an actin cortex inside a liposome, Biophys. J., 96 (2009) 192-198.
[96] T. Luo, V. Srivastava, Y. Ren, D.N. Robinson, Mimicking the mechanical properties of
the cell cortex by the self-assembly of an actin cortex in vesicles, Appl. Phys. Lett., 104
(2014) 153701.

22
 ACCEPTED MANUSCRIPT

[97] C. Campillo, P. Sens, D. Köster, L.-L. Pontani, D. Lévy, P. Bassereau, P. Nassoy, C.

Sykes, Unexpected membrane dynamics unveiled by membrane nanotube extrusion, Biophys.
J., 104 (2013) 1248-1256.
[98] E.A. Shah, K. Keren, Symmetry breaking in reconstituted actin cortices, eLife, 3 (2014)
e01433.
[99] S. Suetsugu, S. Kurisu, T. Takenawa, Dynamic shaping of cellular membranes by

T
phospholipids and membrane-deforming proteins, Physiol. Rev., 94 (2014) 1219-1248.
[100] A.-C. Reymann, C. Guérin, M. Théry, L. Blanchoin, R. Boujemaa-Paterski,

IP
Geometrical control of actin assembly and contractility, Methods Cell Biol., 120 (2014) 19-
38.

R
[101] A.-C. Reymann, J.-L. Martiel, T. Cambier, L. Blanchoin, R. Boujemaa-Paterski, M.
Théry, Nucleation geometry governs ordered actin networks structures, Nat. Mater., 9 (2010)

SC
827-832.
[102] A.-C. Reymann, R. Boujemaa-Paterski, J.-L. Martiel, C. Guérin, W. Cao, H.F. Chin,
E.M. De La Cruz, M. Théry, L. Blanchoin, Actin network architecture can determine myosin

NU
motor activity, Science, 336 (2012) 1310-1314.
[103] M. Murrell, M.L. Gardel, Actomyosin sliding is attenuated in contractile biomimetic
cortices, Mol. Biol. Cell, 25 (2014) 1845-1853.
MA
[104] K. Carvalho, F.-C. Tsai, E. Lees, R. Voituriez, G. Koenderink, C. Sykes, Cell-sized
liposomes reveal how actomyosin cortical tension drives shape change, Proc. Natl. Acad. Sci.
U.S.A., 110 (2013) 16456-16461.
[105] K. Carvalho, J. Lemière, F. Faqir, J. Manzi, L. Blanchoin, J. Plastino, T. Betz, C. Sykes,
Actin polymerization or myosin contraction: two ways to build up cortical tension for
D

symmetry breaking, Phil. Trans. R. Soc. B, 368 (2013) 20130005.

TE

[106] J. Alvarado, M. Sheinman, A. Sharma, F.C. MacKintosh, G.H. Koenderink, Molecular

motors robustly drive active gels to a critically connected state, Nature Phys., 9 (2013) 591-
597.
P

[107] G.T. Charras, C.-K. Hu, M. Coughlin, T.J. Mitchison, Reassembly of contractile actin
cortex in cell blebs, J. Cell Biol., 175 (2006) 477-490.
CE

[108] N. Elkhatib, M.B. Neu, C. Zensen, K.M. Schmoller, D. Louvard, A.R. Bausch, T. Betz,
D.M. Vignjevic, Fascin plays a role in stress fiber organization and focal adhesion
disassembly, Curr. Biol., 24 (2014) 1492-1499.
AC

[109] O. Wiggan, A.E. Shaw, J.G. DeLuca, J.R. Bamburg, ADF/Cofilin regulates actomyosin
assembly through competitive inhibition of Myosin II binding to F-actin, Dev. Cell, 22 (2012)
530-543.
[110] I. Mendes Pinto, B. Rubinstein, A. Kucharavy, J.R. Unruh, R. Li, Actin
depolymerization drives actomyosin ring contraction during budding yeast cytokinesis, Dev.
Cell, 22 (2012) 1247-1260.
[111] L. Haviv, D. Gillo, F. Backouche, A. Bernheim-Groswasser, A cytoskeletal demolition
worker: myosin II acts as an actin depolymerization agent, J. Mol. Biol., 375 (2008) 325-330.
[112] M.P. Murrell, M.L. Gardel, F-actin buckling coordinates contractility and severing in a
biomimetic actomyosin cortex, Proceedings of the National Academy of Sciences, 109 (2012)
20820-20825.
[113] S.K. Vogel, Z. Petrasek, F. Heinemann, P. Schwille, Myosin motors fragment and
compact membrane-bound actin filaments, eLife, 2 (2013) e00116.
[114] C. Kocks, E. Gouin, M. Tabouret, P. Berche, H. Ohayon, P. Cossart, L. monocytogenes-
induced actin assembly requires the actA gene product, a surface protein, Cell, 68 (1992) 521-
531.

23
 ACCEPTED MANUSCRIPT

T
IP
R
SC
NU
MA
D
PTE
CE
AC

24
 ACCEPTED MANUSCRIPT

T
IP
R
SC
NU
MA
D
PTE
CE
AC

25
 ACCEPTED MANUSCRIPT

T
IP
R
SC
NU
MA
D
PTE
CE
AC

26
 ACCEPTED MANUSCRIPT

T
IP
R
SC
NU
MA
D
PTE
CE
AC

27
 ACCEPTED MANUSCRIPT

Highlights

We review the reconstitution of actin polymerization and myosin contraction at or near

surfaces.

Actin and acto-myosin networks at or near surfaces mimic cellular confinement and

T
geometry.

IP
The field is poised to make artificial cells and in vitro tissues.

R
SC
NU
MA
D
P TE
CE
AC

28