Actomyosin contractility spatiotemporally regulates actin network dynamics in migrating cells.

Abstract

Coupling interactions among mechanical and biochemical factors are important for the realization of various cellular processes that determine cell migration. Although F-actin network dynamics has been the focus of many studies, it is not yet clear how mechanical forces generated by actomyosin contractility spatiotemporally regulate this fundamental aspect of cell migration. In this study, using a combination of fluorescent speckle microscopy and particle imaging velocimetry techniques, we perturbed the actomyosin system and examined quantitatively the consequence of actomyosin contractility on F-actin network flow and deformation in the lamellipodia of actively migrating fish keratocytes. F-actin flow fields were characterized by retrograde flow at the front and anterograde flow at the back of the lamellipodia, and the two flows merged to form a convergence zone of reduced flow intensity. Interestingly, activating or inhibiting actomyosin contractility altered network flow intensity and convergence, suggesting that network dynamics is directly regulated by actomyosin contractility. Moreover, quantitative analysis of F-actin network deformation revealed that the deformation was significantly negative and predominant in the direction of cell migration. Furthermore, perturbation experiments revealed that the deformation was a function of actomyosin contractility. Based on these results, we suggest that the actin cytoskeletal structure is a mechanically self-regulating system, and we propose an elaborate pathway for the spatiotemporal self-regulation of the actin cytoskeletal structure during cell migration. In the proposed pathway, mechanical forces generated by actomyosin interactions are considered central to the realization of the various mechanochemical processes that determine cell motility.

Full text

Page 1

Actomyosin contractility spatiotemporally regulates actin network dynamics
in migrating cells
Kennedy Omondi Okeyoa, Taiji Adachia,b,?, Junko Sunagab, Masaki Hojoa
aDepartment of Mechanical Engineering and Science, Kyoto University, Sakyo, Kyoto, Japan
bComputational Cell Biomechanics Team, VCAD System Research Program, RIKEN, Wako, Japan
a r t i c l e i n f o
Article history:
Accepted 7 July 2009
Keywords:
Cell migration
Actomyosin contractility
Actin filament network
Cytoskeletal dynamics
Fluorescent speckle microscopy
Particle imaging velocimetry
Cell biomechanics
Mechanobiology
a b s t r a c t
Coupling interactions among mechanical and biochemical factors are important for the realization of
various cellular processes that determine cell migration. Although F-actin network dynamics has been
the focus of many studies, it is not yet clear how mechanical forces generated by actomyosin
contractility spatiotemporally regulate this fundamental aspect of cell migration. In this study, using a
combination of fluorescent speckle microscopy and particle imaging velocimetry techniques, we
perturbed the actomyosin system and examined quantitatively the consequence of actomyosin
contractility on F-actin network flow and deformation in the lamellipodia of actively migrating fish
keratocytes. F-actin flow fields were characterized by retrograde flow at the front and anterograde flow
at the back of the lamellipodia, and the two flows merged to form a convergence zone of reduced flow
intensity. Interestingly, activating or inhibiting actomyosin contractility altered network flow intensity
and convergence, suggesting that network dynamics is directly regulated by actomyosin contractility.
Moreover, quantitative analysis of F-actin network deformation revealed that the deformation was
significantly negative and predominant in the direction of cell migration. Furthermore, perturbation
experiments revealed that the deformation was a function of actomyosin contractility. Based on these
results, we suggest that the actin cytoskeletal structure is a mechanically self-regulating system, and
we propose an elaborate pathway for the spatiotemporal self-regulation of the actin cytoskeletal
structure during cell migration. In the proposed pathway, mechanical forces generated by actomyosin
interactions are considered central to the realization of the various mechanochemical processes that
determine cell motility.
& 2009 Elsevier Ltd. All rights reserved.
1. Introduction
Cell migration involves a continuous cycle of spatiotemporally
highly coordinated processes including protrusion by actin
polymerization at the leading edge (Watanabe and Mitchison,
2002; Pollard and Borisy, 2003; Ananthakrishnan and Ehrlicher,
2007), attachment to the substrate via focal adhesions (Palecek
et al., 1997; Hu et al., 2007), and tension-dependent retraction of
the trailing edge (Lauffenburger and Horwitz,1996; Mitchison and
Cramer, 1996). It is recognized that coupling interactions among
mechanical and biochemical factors are fundamental to the
realization of many of these processes (Li et al., 2005). For
instance, studies have shown that tensile forces are necessary
for stress fiber (SF) formation and stability (Costa et al., 2002; Sato
et al., 2006) and focal adhesion dynamics (Choquet et al., 1997;
Kaverina et al., 2002; Tan et al., 2003). Furthermore, it has been
suggested that the interactions between F-actin and myosin II are
essential to the organization and dynamics of F-actin network in
migrating cells (Svitkina et al., 1997; Verkhovsky et al., 1999).
Although notable progress has been made toward elucidating the
roles and functions of mechanical forces, more work is still needed
particularly to reveal the relationship between cytoskeletal force
generation and the organization of cytoskeletal functions during
cell migration. This calls for focus on the relationship between
mechanical force generation by actomyosin contractility and
spatiotemporal dynamics of the F-actin network.
Probing the dynamics of the F-actin network during cell
migration using such quantitative tools as fluorescent speckle
microscopy (FSM) (Waterman-Storer et al., 1998; Danuser and
Waterman-Storer, 2006) combined with computer tracking algo-
rithms can provide useful insights into the organization of
cytoskeletal functions. Moreover, information from F-actin net-
work flow can be utilized to define the mechanical properties of
the network, and even how it interacts with other components
ARTICLE IN PRESS
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/jbiomech
www.JBiomech.com
Journal of Biomechanics
0021-9290/$-see front matter & 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jbiomech.2009.07.002
?Corresponding author at: Department of Mechanical Engineering and Science,
Kyoto University, Yoshida-honmachi, Sakyo, Kyoto 606-8501, Japan, Tel./fax:
+81757535216.
E-mail address: adachi@me.kyoto-u.ac.jp (T. Adachi).
Journal of Biomechanics 42 (2009) 2540–2548

Page 2

ARTICLE IN PRESS
like myosin II or focal adhesion proteins (Ji et al., 2008).
Previously, a quantitative study using FSM and image intensity
correlation evaluated the strain field in the lamellipodial F-actin
network and proposed the selective depolymerization model to
correlate negative strain with network depolymerization (Adachi
et al., 2009). Since the cytoskeleton is a complex system whose
dynamics involves various interacting components, systematic
perturbation of the actomyosin machinery is considered a suitable
method for elucidating the relationship between actomyosin-
based force generation and F-actin dynamics.
In this study, we examine the consequences of perturbing
actomyosin contractility on F-actin network flow organization and
deformation in the lamellipodia of migrating fish keratocytes.
These cells are excellent motility models because of their simple
shape, rapid and persistent movement. We perturb the actomyo-
sin system by activating or inhibiting myosin II activity, and, using
a combination of FSM and particle imaging velocimetry (PIV)
(Willert and Gharib, 1991), we quantitatively determine the effect
of actomyosin contractility on cytoskeletal actin dynamics.
2.Materials and methods
2.1. Cell culture and actomyosin perturbation
Black tetra keratocytes used in this study were cultured and handled
essentially as described previously (Adachi et al., 2009). After isolation, freely
moving keratocytes were microinjected with an 8mM conjugate solution of
quantum dot-phalloidin (Qdot-phalloidin) and incubated for 30min before use.
We estimated the volume of the labeled phalloidin delivered into the cell to be
?0.1pL. Moreover, the final cytoplasmic concentration of the conjugate was
determined to be ?0.4mM, where cell volume was taken to be ?1.5pL. Although
phalloidin has deleterious effects on cell migration (Wehland et al., 1977), it has
been observed that microinjecting cells with tiny amounts of the drug does not
induce detectable changes in cell morphology or migration behavior (Wehland
et al., 1980; Wang, 1987). Moreover, phalloidin does not interfere with F-actin
translocation or its ability to interact with myosin II (Dancker et al., 1975).
Perturbation of the actomyosin system was achieved by treating microinjected
keratocytes with either a 2nM solution of calyculin A (Biovision) or a 25mM
solution of blebbistatin (Toronto Research Chemical) diluted in DMEM. Calyculin A
is known to induce actomyosin contractility by inhibiting the activity of myosin
light chain phosphatases (Ishihara et al., 1989; Henson et al., 2003), and
blebbistatin is a well characterized specific inhibitor of myosin II activity (Straight
et al., 2003). These concentrations were chosen to achieve a partial perturbation of
actomyosin contractility without adversely affecting cell versatility. Microscopic
imaging was performed 15min after drug addition, and was continued for 30min
in the presence of the drug. Microinjected but untreated cells served as control.
2.2. Actomyosin staining
For F-actin staining, keratocytes were fixed with 5% paraformaldehyde,
permeabilized with 0.2% Triton-X, and then incubated for 150min at 371C with
25nM rhodamine-phalloidin (Invitrogen) diluted in PBS. For myosin II labeling by
immunofluorescence, fixed and permeabilized cells were incubated for 60min
with 11mg/mL rabit polyclonal myosin IIA primary antibody (GeneTex) in PBS, and
then incubated for 150min with 20mg/mL Alexa Fluor488 goat anti-rabbit IgG
secondary antibody (Invitrogen) in PBS at room temperature. Prior to incubation
with the primary antibody, blocking with 3% BSA in PBS was performed for 60min
at room temperature to reduce non-specific staining. Finally, cells were rinsed and
observed in the presence of PBS to avoid drying.
2.3.Microscopic imaging
Sequential FSM images (15–25 frames) with a spatial sampling rate of 120nm
per pixel were acquired using Zeiss 200M (Zeiss) with a Plan-APO 100?/1.4NA oil
immersion objective lens and a highly sensitive CCD camera (Axiocam, Zeiss) for
use in the analysis of F-actin network flow in the lamellipodia of treated and
control keratocytes. Exposure time was 0.8s and acquisition interval was 2.0s for
calyculin, and 4.0s for blebbistatin and control. Acquired images were filtered
using a linear convolution low pass and high Gauss filters (Image-Pro Plus, Media
Cybernetics) both with a matrix size of 7?7 pixels to eliminate high frequency
noise and to enhance speckle appearance. Similarly, images of actomyosin-labeled
fixed-cells were acquired using the laser scanning mode of Olympus Fluorview
FV1000 system fitted with UPLSAPO 100?/1.4N.A oil objective lens at an
excitation wavelength of 473nm and 559nm for Alexa Fluor488 and rhodamine–
phalloidin, respectively.
2.4.Analysis of F-actin network flow and deformation
Quantitative analysis of F-actin network dynamics in the lamellipodia was
performed by PIV-based image correlation method essentially as described
previously (Adachi et al., 2009). This involved using a PIV software (Flow-Vec32,
Library) to track the motion of numerous F-actin speckles on the sequential FSM
images by intensity correlation to obtain F-actin network flow maps. To do this,
images were meshed using square grids of side 0.6mm, and then cross-correlated
using a square reference sub-window and a corresponding square interrogation
sub-window of sides 3.0 and 0.8mm, respectively. To improve search accuracy, a
relatively high correlation coefficient of 0.7 was used, and the lowest threshold
intensity was set higher than the predetermined image background intensity to
minimize the influence of background noise. Correlation data was processed using
Micro-AVS (KGT) to obtain F-actin flow maps.
Network deformation was computed based on the F-actin flow fields over a
5.4mm wide region at the center of the lamellipodia as outlined previously (Adachi
et al., 2009). Mean strain rate, _ e, defined as network strain per unit time, was
determined as the group average of individual sample means for each
experimental group (n ¼ 9 per group). The components of _ e parallel to the
direction of cell movement, _ ep, and in the normal direction, _ en, were also
determined and plotted. Mean strain rate results for perturbed and control cells
were statistically compared by t-test (po0.05). Since the length of the lamellipodia
varied from cell to cell, a mean length of 5.1mm (for n ¼ 9) was determined and
used in the plots of mean strain rates.
3. Results
3.1. Effect of actomyosin perturbation on actin cytoskeletal structure
To examine the effect of actomyosin perturbation, we deter-
mined the distributions of F-actin and myosin II in the
lamellipodia based on the intensity of fluorescent images. As
shown in Fig. 1, F-actin distribution was characterized by a high
density at the leading edge, followed by a steady decline in
density toward the back of the lamellipodia. However, for the
cases of calyculin and control, a second peak in density (marked
by arrows in Fig. 1B and D) indicating the presence of SFs was
observed, and this was more prominent in the case of calyculin
(compare Fig. 1B with 1D). In deed, calyculin-treated cells
exhibited higher F-actin density over the entire lamellipodia,
and their SFs were more pronounced and could be observed even
at the lateral edges (Fig. 1A, left panel). Contrarily, F-actin in
blebbistatin-treated cells appeared ruffled and disrupted, and SFs
were barely visible (Fig. 1E, left panel). Remarkably, treated cells
also displayed high F-actin density at the leading edge suggesting
that actomyosin perturbation did not significantly interfere with
F-actin organization in this region (Fig. 1B and F).
Myosin II appeared as distinct bright spots on the images of
stained cells (Fig. 1, middle panel), and as expected, its distribu-
tion in calyculin-treated cells and control cells contrasted greatly
(compare Fig.1A with C, middle panel). Whereas myosin II motors
were widely distributed at a higher density over the entire
lamellipodia in calyculin-treated cells, myosin spots were mainly
localized at the back of the lamellipodia and cell body region in
control cells. Moreover, comparatively higher myosin II density
around the SFs region was observed for calyculin, and, to a lesser
extent, for control (Fig. 1B and D), suggesting that myosin II
colocalized along the SFs at the back of the lamellipodia as
previously reported (Svitkina et al., 1997; Kolega, 2006). Contra-
rily, myosin II density was considerably depleted in the lamelli-
podia and cell body region of blebbistatin-treated cells, as
expected. Nevertheless, sparsely distributed myosin II spots could
still be observed (Fig. 1E, middle panel), suggesting that the
intended partial inhibition was achieved. We also observed
saturated staining of the cell nucleus for all the three cases,
consistent with the observation that myosin II motors are present
K.O. Okeyo et al. / Journal of Biomechanics 42 (2009) 2540–2548
2541

Page 3

ARTICLE IN PRESS
in the nucleus (Li and Sarna, 2009), and in agreement with the
results of myosin II staining by Svitkina et al. (1997).
Taken together, these results demonstrate that we achieved the
intended purpose of perturbing the actomyosin system by altering
the distribution of the myosin II motors. Moreover, the correlative
distributions of F-actin and myosin II suggest that actomyosin
contractility is important for the structural organization and
integrity of F-actin network and SFs, in agreement with previous
reports (Lin et al., 1996; Henson et al., 2003; Schaub et al., 2007).
3.2. Effect of actomyosin perturbation on F-actin network flow and
cell speed
The flow vector fields shown in Fig. 2 were obtained by PIV
analysis of FSM images, and they describe F-actin network
kinematics in the lamellipodia of migrating fish keratocytes.
As can be seen from the magnified flow maps, the flow fields
consist of centripetally organized flow vectors that tend to
converge around the middle of the lamellipodia (Fig. 2A and B).
Retrograde flow, marked as RF in Fig. 2A and B and indicated by
backward oriented flow vectors, was mainly limited to the front
region of the lamellipodia and progressively decreased in intensity
toward the back. On the other hand, anterograde flow, indicated
by forward oriented flow vectors and marked AF in Fig. 2A and B,
was mainly localized around the cell body region and trailing
edge. The two opposing flows converged around the middle of the
lamellipodia, resulting in a region of highly reduced flow marked
as CZ in Fig. 2A and B.
Actomyosin perturbation produced a marked alteration in both
the intensity and organization of F-actin network flow, as can be
ActinMyosin IIMerged
Myosin II
Actin
0
20
40
60
80
0.02.04.0 6.0 8.010.0
Distance from the leading edge [μm]
SF
Fluorescence intensity [a.u.]
Calyculin
0
20
40
60
80
0.0 2.04.06.08.0 10.0
Distance from the leading edge [μm]
Fluorescence intensity [a.u.]
Blebbistatin
0
20
40
60
80
0.0 2.0 4.06.0 8.0 10.0
Distance from the leading edge (μm)
SF
Fluorescence intensity [a.u.]
Control
Fig. 1. Actomyosin distribution in the lamellipodia of migrating fish keratocytes. A, C and E: laser scanning microscopy images of keratocytes labeled for actin (left panel),
myosin II (middle panel) and merged (right panel) corresponding to calyculin (A), control (C), and blebbistatin (E). B, D, and F: density distribution profiles of F-actin and
myosin II determined as the average image intensity over the white rectangular boxes shown in the respective merged images. The arrows marked SF in B and D indicate the
intensity rise due to the presence of SFs. Scale bar is 5.0mm.
K.O. Okeyo et al. / Journal of Biomechanics 42 (2009) 2540–2548
2542

Page 4

ARTICLE IN PRESS
appreciated by comparing the flow fields of calyculin (Fig. 2A) and
blebbistatin (Fig. 2C) with that of control (Fig. 2B). Activating
actomyosin contractility resulted in an increase in the intensity of
both retrograde and anterograde flows (compare Fig. 2A with B).
The two opposing flows gradually decreased in intensity toward
the middle of the lamellipodia where they merged to form a
clearly visible convergence zone characterized by a markedly
weak flow at the middle of the lamellipodia. A similar zone of flow
convergence has been reported in other studies using fish
keratocytes (Schaub et al., 2007) and epithelial cells (Vallotton
et al., 2004), and the zone has been associated with increased
network contraction (Vallotton et al., 2004). On the other hand,
the flow fields of blebbistatin-treated cells were characterized by
a significant reduction in the intensity of F-actin network flow and
lack of a clearly defined convergence zone. In some cells, F-actin
network flow was completely attenuated in most region of the
lamellipodia, although some remnant flow could still be observed
at the retracting rear and lateral edges.
To elucidate the relationship between F-actin dynamics and
cell speed, we determined cell speed for both treated and control
cells. As shown in Fig. 3, F-actin network flow and cell speed were
affected similarly by actomyosin perturbation; whereas calyculin
treatment dramatically increased both F-actin network flow and
cell speed, blebbistatin treatment resulted in a marked reduction
in both the flow and cell speed. Assuming that the perturbation
drugs did not affect the contribution by polymerization to both
F-actin flow and cell speed (Henson et al., 1999; Watanabe and
Mitchison, 2002), these perturbation results indicate that these
parameters are determined by actomyosin contractility.
3.3. Influence of actomyosin contractility on F-actin network
deformation
Based on the flow velocity fields outlined above, continuum-
based analysis of F-actin network strain was performed as
described previously (Adachi et al., 2009). The concept is
schematically illustrated in Fig. 4, which also shows the
coordinate system used (Fig. 4A), and the deformation of a
square lamellipodial element during a time interval, Dt (Fig. 4B).
Qdot-phalloidin attached to F-actin (Fig. 4B) can serve as pseudo-
fiduciary markers for evaluating element deformation in both
x and y directions using PIV-based image correlation method.
RF
AF
RF
AF
0.08
0.06
0.02
0.00
μm/s
Calyculin
Control
Blebbistatin
CZ
CZ
t
d
Fig. 2. Flow fields of F-actin network dynamics in the lamellipodia of migrating
fish keratocytes obtained by PIV-based cross-correlation of FSM images. (A)
Calyculin, (B) control, and (C) blebbistatin. Each composite figure (A, B and C)
consists of an FSM image overlaid with a flow field (left side), a magnification
(bottom right side) of the region indicated by a white square on the left image, and
a kymograph (top right side). In A and B, RF denotes retrograde flow, AF:
anterograde flow and CZ: convergence zone. The kymographs were constructed
from slices cropped from the center of each corresponding FSM image, as shown
by the white narrow rectangle in C. Time and distance axes are indicated on the
kymographs by t and d, respectively. Horizontal scale bar is 5.0mm, except that of
the magnified flow fields which is 2.5mm, and the vertical scale bar is 2.0s.
0.0
0.1
0.2
0.3
0.4
0.5
CalyculinControlBlebbistatin
Cell speed
Retrograde flow
Cell speed and retrograde flow velocity [μm/s]
p < 0.05
p < 0.001
Mean ± S.D, n = 6
vs. control
Fig. 3. Comparison of cell speed and retrograde flow. Cell speed was determined
from time-lapse movies of migrating fish keratocytes as the time–average
displacement of center of mass. Mean retrograde flow was determined over a
6.0mm wide region at the center of the lamellipodia (gray shaded region in Fig. 4A)
based on the obtained flow maps (refer to Fig. 2). It is clear that calyculin-treated
cells exhibit higher cell speed and retrograde flow velocity than control cells. In
contrast, blebbistatin-treated cells show a marked reduction in both migration
speed and retrograde flow velocity.
K.O. Okeyo et al. / Journal of Biomechanics 42 (2009) 2540–2548
2543

Page 5

ARTICLE IN PRESS
The results of network strain analysis are shown in Fig. 5.
Parallel strain rate, _ ep, in the direction of cell migration was found
to be predominantly negative for all the experimental groups, in
agreement with our previous analysis (Adachi et al., 2009).
Importantly, _ ep values were found to be significantly higher for
calyculin than for control (Fig. 5A), and strain rate values in this
case consistently increased toward the back of the lamellipodia,
suggesting that the F-actin network is under a state of increasing
compression toward the back of the lamellipodia. In deed, the
value of _ ep at the leading edge (x ¼ 0.3mm) was found to be
almost half its value at the back of the lamellipodia (x ¼ 5.1mm).
Thus, the response of _ epto actomyosin perturbation suggests that
the network deformation in the direction of cell movement is
dependent on actomyosin contractility.
Furthermore, for calyculin and control, _ epwas dominant over
_ en(Fig. 5A and B), implying that F-actin network deformation is
predominant in the direction of cell movement. This is expected in
the case where F-actin network in the lamellipodia undergo
isomeric contraction as predicted by the network contraction
model (Svitkina et al., 1997). We therefore focused on _ ep and
integrated it over the 5.4mm wide centre region of the
lamellipodia to obtain its total distribution, _ ep total, presented in
Fig. 5C. The figure shows that compared with control, _ ep totalfor
calyculin was dominant and significantly negative, particularly
toward the back of the lamellipodia.
Contrarily, inhibiting actomyosin resulted in markedly small
strain rates (Fig. 5A and B), in accordance with the flow
attenuation illustrated in Fig. 2C. Since strain rates were
computed as spatial gradients of the flow vector fields, the
gradients of very small vectors such as those describing the flow
field for blebbistatin (Fig. 2C) would yield infinitesimal strain
rates. The variations, if any, in the values of such strains would be
equally infinitesimal and undetectable, thus both parallel and
normal strain rates in the case of blebbistatin do not show
appreciable change with distance (Fig. 5A and B).
4. Discussion
We have shown that actomyosin perturbation by up- and
down-regulation of myosin II activity affects F-actin dynamics and
deformation. Below we discuss the consequences of these
observations with regard to cell motility.
4.1. Actomyosin contractility determines F-actin network dynamics
and deformation
The F-actin flow velocity maps presented in Fig. 2 and the
strain distributions shown in Fig. 5 correlate well with actomyosin
distributions shown in Fig.1. Higher flow intensity and strain rates
in the case of calyculin are consistent with the relatively higher
myosin II density and enhanced SFs observed for calyculin. In
contrast, highly reduced F-actin flow and strain rates correspond
to the low myosin II density in the case of blebbistatin. Thus, these
parameters of F-actin dynamics depend on the level of actomyosin
activity, as previously reported (Svitkina et al., 1997).
In fish keratocytes, actomyosin interactions especially along
the SFs located at the boundary between the lamellipodia-cell
body region are believed to generate contractile forces that
initiate network contraction (Svitkina et al., 1997; Verkhovsky et
al., 1997; Kolega, 2006). Since the stability and formation of SFs
depend on actomyosin tension, these structures were enhanced in
most cells treated with calyculin, but were disrupted in cells
treated with blebbistatin (Fig. 1). Thus, in the case of calyculin,
increased contractility due to actomyosin activation would result
in increased SF tension and network contraction thereby leading
to increased retrograde flow intensity (Figs. 2 and 3). Conse-
quently, increased contractility would result in increased retrac-
tion and cell body translocation, with associated increase in the
intensity of anterograde flow of F-actin network (Fig. 2A).
The converse of this would be expected in the case of blebbistatin
(Fig. 2C). Overall, the observed effects of actomyosin perturbation
on F-actin flow suggest that actomyosin contraction drives
actin network dynamics, in agreement with previous reports
(Henson et al., 1997; Svitkina et al.,1997; Verkhovsky et al.,1999).
Quantitative analysis of F-actin network deformation in the
lamellipodia of migrating cells has revealed that F-actin network
deformation is predominantly negative in the parallel direction
(Fig. 5A). The network is expected to undergo compressive
deformation by the virtue of being sandwiched between two
opposing flows. In fact, the convergence of retrograde and
anterograde flows at the interior of the lamellipodia (Fig. 2A and
y
x
lp (t)
lp (t + Δt)
ln (t)
ln (t + Δt)
Lamellipodia
Center
region
t = t
t = t + Δt
Qdot
phalloidin
F-actin
o
Fig. 4. Schematic illustration of continuum-based strain analysis in the lamellipodia of migrating cells. (A) Schematic of a keratocyte showing the analyzed center region
and the coordinate system with the origin taken at the point the point marked ‘‘O’’. Parallel and normal strains rates,_ epand_ en, defined in the text were determined over the
gray-shaded center region in x and y directions, respectively. (B) Strain rate analysis based on continuum deformation of a square lamellipodial element during a time
interval, Dt. The magnified portions of the elements show Qdot-phalloidin attached to F-actin. Speckles of Qdot-phalloidin serve as pseudo-fiduciary markers for evaluating
network strain using PIV-based image correlation method.
K.O. Okeyo et al. / Journal of Biomechanics 42 (2009) 2540–2548
2544

Page 6

ARTICLE IN PRESS
B) is accompanied with a steep change in the gradient of flow
velocity, which is reflected by the significantly negative strain
rates shown in Fig. 5A. In deed, the zone of convergence has
previously been attributed to increased network contraction and
depolymerization (Vallotton et al., 2004; Ponti et al., 2005), and
the results of our strain measurements under actomyosin
perturbation further elucidates this fact. Nonetheless, it remains
to be investigated whether flow convergence occurs passively or
actively in the sense of being influenced by localized F-actin-focal
adhesion interactions.
As pointed out previously (Adachi et al., 2009), the negative
strain rates implies that the actin network is undergoing
compressive deformation due to anisotropic network contraction
generated by actomyosin contractility along the transverse SF
bundles (Svitkina et al., 1997; Verkhovsky et al., 1999; Kolega
2006). Therefore, perturbing the actomyosin system would
influence the level of contractility and therefore affect the level
of network deformation. Furthermore, considering that myosin II
colocalizes with SFs at the back of the lamlellipodia (Fig. 1), a
slight increase in parallel strain rate toward the back of the
lamellipodia, as in the case of calyculin (Fig. 5A), indicates that the
forces that cause network deformation are mainly generated in
the contractile module. Thus, the deformation can be regarded as
a direct consequence of actomyosin contractility, mainly along the
contractile SF bundles. Nevertheless, we should mention that
the continuum-based approach of strain analysis adopted in this
study could not capture infinitesimal strains expected at the level
of an individual filament.
Distance from the leading edge, x [μm]
-0.12
-0.08
-0.04
0.00
Total strain rate, εp total
-0.02
-0.01
0.00
0.01
0.61.2 1.82.4 3.0 3.64.24.85.40.0
∗
∗
∗
∗
∗
∗
Strain rate, εn
-0.04
-0.03
-0.02
-0.01
0.00
0.01
∗
Blebbistatin
Control
Calyculin
∗
∗
∗
∗
∗
∗
∗
∗
∗ p < 0.05 vs. control (t-test)
Mean ± S.D., n = 9
0.6 1.21.82.4 3.03.6 4.24.8 5.4 0.0
Strain rate, εp
x
0.61.21.8 2.4 3.03.6 4.2 4.8 5.40.0
Calyculin
Control
Blebbistatin
Fig. 5. Distribution of mean strain rates in the F-actin structure forming the lamellipodia of migrating fish keratocytes. (A) Distribution of parallel strain rate, _ ep, from the
leading edge toward the back of the lamellipodia for the three experimental groups. This strain rate is predominantly negative for all the groups, with the values for
calyculin being significantly higher than for both control and blebbistatin, particularly toward the back of the lamellipodia. (B) Distribution of normal strain rate, _ en, from
the leading edge toward the back of lamellipodia. This strain rate is nominal compared to that in the migration direction, especially for the case of calyculin. (C) Distribution
of total parallel strain rate, _ ep total, in the lamellipodia as obtained by integrating in the x direction the parallel strain rate shown in A over the center region shown in Fig. 4A.
K.O. Okeyo et al. / Journal of Biomechanics 42 (2009) 2540–2548
2545

Page 7

ARTICLE IN PRESS
4.2. F-actin network as a spatiotemporally self-regulating
mechanical system
Based on the results of this study, the previously proposed selective
depolymerization model (Adachi et al., 2009), and other supportive
evidences in literature, we propose the pathway illustrated in Fig. 6 for
the mechano-self-regulation of the actin cytoskeletal structure.
Starting with tension generation by actomyosin contractility, we
elucidate how various motility processes are spatiotemporally
regulated by mechanical factors during cell migration.
Actin filament
Contractile force
Myosin II
Retrograde flow
Retraction
Network contraction
filament
realignment
focal
adhesion
SF
polymerization
actomyosin
contraction/ tension
generation *
network
contraction *
retrograde
flow *
negative
deformation *
filament
realignment
retraction *
anterograde
flow *
flow
convergence *
depolymeri-
zation
polymerization
FA dynamics
monomer
supply
calyculin
blebbistatin
SF formation &
stability *
* Investigated in this paper
Pathway link
Proposed pathway
activation
inhibition
monomer
supply
flow
convergence
anterograde
flow
negative
strain
FA dynamics
retraction
Protrusion
Antirograde flow
1a
2b
2a
2b2c
2d
34
5
678a9
8b1b
1c
10
11
12
2c
5
6
9
1c
SF formation
& stability
12
depolymerization
network
contraction
1d
retrograde
flow
2a
2d
actomyosin
contractility
1b
8a
10
1d
3
4
11
7
Network contraction region
1a
8b
Fig. 6. Proposed mechano-self-regulatory pathway of the actin cytoskeletal structure. (A) Cell level representation of the various cellular processes involved in cell motility.
Tension generation by actomyosin contractility triggers a series of spatiotemporally coordinated mechanical events inside the cell that collectively result in cell movement.
Coupling interactions among mechanical and biomechanical factors are important for the realization of most of these events. (B) A simplified flow chart of the motility
processes outlined in A. Various pathways by which mechanical forces generated by actomyosin interactions can regulate cytoskeletal functions are illustrated. In both A
and B, arrows link correlated processes and indicate the direction of regulation.
K.O. Okeyo et al. / Journal of Biomechanics 42 (2009) 2540–2548
2546

Page 8

ARTICLE IN PRESS
Actomyosin interactions generate tensile forces that contribute
to the formation and stability of the SFs, as demonstrated in this
study and others (Verkhovsky et al., 1999; Costa et al., 2002;
Hayakawa et al., 2005). The contraction of actin fiber bundles, in
turn, modulates focal adhesion dynamics (Choquet et al., 1997;
Kaverina et al., 2002; Tan et al., 2003) and retraction of the cell
rear edge (Lee et al., 1993; Small et al., 1996; Anderson and Cross,
2000). Moreover, contractile forces in the SFs causes the
contraction of F-actin network in the lamellipodia and the cell
body region (Svitkina et al., 1997; Verkhovsky et al., 1999),
resulting in the generation of retrograde and anterograde flows, as
shown in this study and others (Vallotton et al., 2004, 2005;
Schaub et al., 2007). The two opposing flows finally merge at the
convergence zone, resulting in increased network compression, as
revealed in this study and elsewhere (Vallotton et al., 2004).
In addition, as suggested by the selective depolymerization model
(Okeyo et al., 2006; Adachi et al., 2009), negative strain
contributes to the depolymerization of actin filaments by coupling
with other biochemical factors such as ADF/Cofilin. Depolymer-
ization in turn provides new monomers for continued polymer-
ization (Mitchison and Cramer, 1996; Pollard and Borisy, 2003),
which leads to protrusion. Meanwhile, newly formed actin
filaments are transported by retrograde flow toward the back of
the lamellipodia (Henson et al., 1999; Verkhovsky et al., 1999),
where they undergo realignment before they are incorporated
into new fiber bundles, facilitated by actomyosin tension
(Verkhovsky et al.,1995; Kolega, 2006). The newly formed bundles
resume the role of tension generation, and the cycle repeats. Thus,
the cytoskeletal actin structure can be regarded as a spatiotem-
porally self-regulating mechanical system.
In conclusion, this study has highlighted the roles played by
actomyosin contractility in driving F-actin network flow and
deformation. We have elaborated on the role of the actomyosin
system as the mechanical force generator, and placed it into
perspective in the proposed mechano-self-regulatory pathway for
cell motility regulation. The proposed mechanical pathway can be
considered alongside other known molecular pathways (Pollard
and Borisy, 2003) to achieve a better understanding of the
underlying mechanisms of cell motility.
Conflict of interest statement
There is no conflict of interest.
Acknowledgements
This work was partially supported by the Grant-in-Aid for
Specially Promoted Research (20001007) from the Ministry of
Education, Culture, Sports, Science and Technology of Japan, and
by the Japan Society for the Promotion of Science under the
Research Fellowship for Young Scientists.
References
Adachi, T., Okeyo, K.O., Shitagawa, Y., Hojo, M., 2009. Strain field in actin filament
network in lamellipodia of migrating cells: implication for network reorgani-
zation. Journal of Biomechanics 42, 297–302.
Ananthakrishnan, R., Ehrlicher, A., 2007. The forces behind cell movement.
International Journal of Biological Sciences 3, 303–317.
Anderson, K.I., Cross, R., 2000. Contact dynamics during keratocyte motility.
Current Biology 10, 253–260.
Choquet, D., Felsenfeld, D.P., Sheetz, M.P., 1997. Extracellular matrix rigidity causes
strengthening of integrin–cytoskeleton linkages. Cell 88, 39–48.
Costa, K.D., Hucker, W.J., Yin, F.C., 2002. Buckling of actin stress fibers: a new
wrinkle in the cytoskeletal tapestry. Cell Motility and the Cytoskeleton 52,
266–274.
Dancker, P., Low, I., Hasselbach, W., Wieland, T., 1975. Interaction of actin
with phalloidin: polymerization and stabilization of F actin. Biochimica
et Biophysica Acta 400, 407–414.
Danuser, G., Waterman-Storer, C.M., 2006. Quantitative fluorescent speckle
microscopy of cytoskeleton dynamics. Annual Review of Biophysics and
Biomolecular Structure 35, 361–387.
Hayakawa, K., Tatsumi, H., Sokabe, M., 2005. Mechanical tension in actin filaments
prevents cofilin from disassembling the filaments. Cell Structure and Function
30, 113.
Henson, J.H., Kolnik, S.E., Fried, C.A., Nazarian, R., McGreevy, J., Schulberg, K.L.,
Detweiler, M., Trabosh, V.A., 2003. Actin-based centripetal flow: phosphatase
inhibition by calyculin-A alters flow pattern, actin organization, and actomyo-
sin distribution. Cell Motility and the Cytoskeleton 56, 252–266.
Henson, J.H., Svitkina, T.M., Burns, A.R., Hughes, H.E., MacPartland, K.J., Nazarian,
R., Borisy, G.G., 1999. Two components of actin-based retrograde flow in sea
urchin coelomocytes. Molecular Biology of the Cell 10, 4075–4090.
Henson, J.H., Svitkina, T.M., Hughes, H.E., Mendola, R.J., Borisy, G.G.,1997. Structural
organization of actin and myosin II in sea urchin coelomocytes undergoing
centripetal/retrograde flow. Molecular Biology of the Cell 8, 969.
Hu, K., Ji, L., Applegate, K.T., Danuser, G., Waterman-Storer, C.M., 2007. Differential
transmission of actin motion within focal adhesions. Science 315, 111–115.
Ishihara, H., Martin, B.L., Brautigan, D.L., Karaki, H., Ozaki, H., Kato, Y., Fusetani, N.,
Watabe, S., Hashimoto, K., Uemura, D., Hartshorne, D.J., 1989. Calyculin A and
okadaic acid: Inhibitors of protein phosphatase activity. Biochemical and
Biophysical Research Communications 159, 871–877.
Ji, L., Lim, J., Danuser, G., 2008. Fluctuations of intracellular forces during cell
protrusion. Nature Cell Biology 10, 1393–1400.
Kaverina, I., Krylyshkina, O., Small, J.V., 2002. Regulation of substrate adhesion
dynamics during cell motility. The International Journal Biochemistry and Cell
Biology 34, 746–761.
Kolega, J., 2006. The role of myosin II motor activity in distributing myosin
asymmetrically and coupling protrusive activity to cell translocation. Mole-
cular Biology of the Cell 17, 4435–4445.
Lauffenburger, D.A., Horwitz, A.F., 1996. Cell migration: a physically integrated
molecular process. Cell 84, 359–369.
Lee, J., Ishihara, A., Theriot, J.A., Jacobson, K., 1993. Principles of locomotion for
simple-shaped cells. Nature 362, 167–171.
Li, Q., Sarna, S.K., Chien, S., 2009. Nuclear Myosin II Regulates the Assembly of
Preinitiation Complex for Icam-1 Gene Transcription. Gastroenterology 137 (3),
1054–1063.
Li, S., Guan, J.L., Chien, S., 2005. Biochemistry and biomechanics of cell motility.
Annual Review of Biomedical Engineering 7, 105–150.
Lin, C.H., Espreafico, E.M., Mooseker, M.S., Forscher, P., 1996. Myosin drives
retrograde F-actin flow in neuronal growth cones. Neuron 16, 769–782.
Mitchison, T.J., Cramer, L.P., 1996. Actin-based cell motility and cell locomotion.
Cell 84, 371–379.
Okeyo, K.O., Shitagawa, Y., Adachi, T., Hojo, M., 2006. Quantitative evaluation of
strain field in the lamella region of cellular fragments from fish keratocytes.
Journal of Biomechanics 39 (Supplement 1), S244.
Palecek, S.P., Loftus, J.C., Ginsberg, M.H., Lauffenburger, D.A., Horwitz, A.F., 1997.
Integrin–ligand binding properties govern cell migration speed through
cell–substratum adhesiveness. Nature 385, 537–540.
Pollard, T.D., Borisy, G.G., 2003. Cellular motility driven by assembly and
disassembly of actin filaments. Cell 112, 453–465.
Ponti, A., Matov, A., Adams, M., Gupton, S., Waterman-Storer, C.M., Danuser, G.,
2005. Periodic patterns of actin turnover in lamellipodia and lamellae of
migrating epithelial cells analyzed by quantitative fluorescent speckle
microscopy. Biophysical Journal 89, 3456–3469.
Sato, K., Adachi, T., Shirai, Y., Tomita, Y., 2006. Local disassembly of actin stress
fibers induced by selected release of intracellular tension in osteoblastic cell.
Journal of Biomechanical Science and Engineering 1, 204–214.
Schaub, S., Bohnet, S., Laurent, V.M., Meister, J.J., Verkhovsky, A.B., 2007.
Comparative maps of motion and assembly of filamentous actin and myosin
II in migrating cells. Molecular Biology of the Cell 18, 3723–3732.
Small, J.V., Anderson, K., Rottner, K., 1996. Actin and the coordination of
protrusion, attachment and retraction in cell crawling. Bioscience Reports 16,
351–368.
Straight, A.F., Cheung, A., Limouze, J., Chen, I., Westwood, N.J., Sellers, J.R.,
Mitchison, T.J., 2003. Dissecting temporal and spatial control of cytokinesis
with a myosin II inhibitor. Science 299, 1743–1747.
Svitkina, T.M., Verkhovsky, A.B., McQuade, K.M., Borisy, G.G., 1997. Analysis of the
actin–myosin II system in fish epidermal keratocytes: mechanism of cell body
translocation. Journal of Cell Biology 139, 397–415.
Tan, J.L., Tien, J., Pirone, D.M., Gray, D.S., Bhadriraju, K., Chen, C.S., 2003. Cells lying
on a bed of microneedles: an approach to isolate mechanical force. Proceedings
of the National Academy of Sciences of the United States of America 100,
1484–1489.
Vallotton, P., Danuser, G., Bohnet, S., Meister, J.J., Verkhovsky, A.B., 2005. Tracking
retrograde flow in keratocytes: news from the front. Molecular Biology of the
Cell 16, 1223–1231.
Vallotton, P., Gupton, S.L., Waterman-Storer, C.M., Danuser, G., 2004. Simultaneous
mapping of filamentous actin flow and turnover in migrating cells by
quantitative fluorescent speckle microscopy. Proceedings of the National
Academy of Sciences of the United States of America 101, 9660–9665.
Verkhovsky,A.B.,Svitkina,T.M.,Borisy,
assemblies in the active lamella of fibroblasts—their morphogenesis and role
G.G.,1995.Myosin-IIfilament
K.O. Okeyo et al. / Journal of Biomechanics 42 (2009) 2540–2548
2547

Page 9

ARTICLE IN PRESS
in the formation of actin filament bundles. Journal of Cell Biology 131,
989–1002.
Verkhovsky, A.B., Svitkina, T.M., Borisy, G.G., 1997. Contraction of actin–myosin II
dynamic networkdrivescell translocation.
Cell 8, 974.
Verkhovsky, A.B., Svitkina, T.M., Borisy, G.G., 1999. Network contraction model for
cell translocation and retrograde flow. Biochemical Society Symposium 65,
207–222.
Wang, Y., 1987. Mobility of filamentous actin in living cytoplasm. Journal of Cell
Biology 105, 2811–2816.
Watanabe, N., Mitchison, T.J., 2002. Single-molecule speckle analysis of actin
filament turnover in lamellipodia. Science 295, 1083–1086.
Molecular Biology of the
Waterman-Storer, C.M., Desai, A., Bulinski, J.C., Salmon, E.D., 1998. Fluorescent
speckle microscopy, a method to visualize the dynamics of protein assemblies
in living cells. Current Biology 8, 1227–1230.
Wehland, J., Osborn, M., Weber, K., 1977. Phalloidin-induced actin polymerization
in the cytoplasm of cultured cells interferes with cell locomotion and growth.
Proceedings of the National Academy of Sciences of the United States of
America 74, 5613–5617.
Wehland, J., Osborn, M., Weber, K.,1980. Phalloidin associates with microfilaments
after microinjection into tissue culture cells. European Journal of Cell Biology
21, 188–194.
Willert, C.E., Gharib, M., 1991. Digital particle image velocimetry. Experiments in
Fluids 10, 181–193.
K.O. Okeyo et al. / Journal of Biomechanics 42 (2009) 2540–2548
2548