Am J Physiol Lung Cell Mol Physiol 285: L456-L463, 2003.
First published April 18, 2003; doi:10.1152/ajplung.00329.2002
1040-0605/03 $5.00
Mechanical strain increases cell stiffness through cytoskeletal filament reorganization
Paul G. Smith,1
Linhong Deng,2
Jeffrey J. Fredberg,3 and
Geoffrey N. Maksym2
1Department of Pediatrics, Case Western Reserve
University, Cleveland, Ohio 44106; 2School of
Biomedical Engineering, Dalhousie University, Halifax, Nova Scotia B3H 3J5,
Canada; and 3Physiology Program, Harvard School of
Public Health, Boston, Massachusetts 02115
Submitted 30 September 2002
; accepted in final form 17 April 2003
 |
ABSTRACT
|
|---|
We tested the hypothesis that cytoskeletal reorganization induced by cyclic
strain increases cytoskeletal stiffness (G'). G' was measured by
optical magnetic twisting cytometry in control cells and cells that had
received mechanical strain for 10–12 days. G' was measured before
and after both contractile and relaxant agonists, and in the strained cells
both parallel (Para) and perpendicular (Perp) to the aligned cytoskeleton.
Before activation, G' Para was 24 ± 5% (± SE) greater
compared with Perp (P < 0.05), and 35% ± 6 greater compared
with control (Cont, P < 0.01). The difference between strained and
control cells was enhanced by KCl, increasing G' 171 ± 7% Para
compared with 125 ± 6% Perp and 129 ± 8% Cont (P <
10-5 both cases). The decrease in G' from baseline
due to relaxant agonists isoproterenol and dibutyryl cAMP was similar in all
groups. Long-term oscillatory loading of airway smooth muscle (ASM) cells
caused stiffness to increase and become anisotropic. These findings are
consistent with the hypothesis that cytoskeletal reorganization can enhance
ASM stiffness and contractility. They imply, furthermore, that oscillatory
loading of ASM may contribute to airway narrowing and failure of airway
dilation in asthma.
smooth muscle contraction; plasticity; optical magnetic twisting cytometry; anisotropy; airway smooth muscle cell
BRONCHIAL HYPERRESPONSIVENESS leads to excessive narrowing of
the airway and is ultimately driven by the contractile activity of the airway
smooth muscle. Why the airway narrows excessively in asthma is not clear,
however. The evidence for alteration of airway smooth muscle in asthma and
other diseases, both structurally and functionally, is equivocal
(26,
27), and the cellular
mechanisms responsible for changes that have been reported are uncertain
(13,
18). One factor that could
alter smooth muscle behavior may be the mechanical stress experienced by the
smooth muscle cell. For example, in asthma, there are changes in the
mechanical microenvironment of the smooth muscle cell due both to connective
tissue remodeling, which alters the stresses and strains impinging on the cell
from the tidal action of breathing
(13,
28), and to the excessive
activation and shortening of the airway smooth muscle itself.
We have shown previously using cultured airway smooth muscle cells that
long-term periodic mechanical strain induces changes in cytoskeletal and
contractile filament organization as well as contractile function. Strain
induces dramatic differences in the formation and organization of focal
adhesions and stress fibers
(19). Strain also induces
increased force production and greater shortening capacity
(22,
23). Whether such cytoskeletal
changes translate into changes in active or passive cell mechanics in vivo is
unknown, but such changes might explain mechanisms of abnormal function of
smooth muscle.
Stiffness is an important determinant of stress generation in any tissue
experiencing mechanical strain and, in muscle, is also a good index of active
force development (29). Here
we report the changes in smooth muscle cytoskeletal stiffness (G')
induced by cyclic deformational strain before and after application of
contractile and relaxant agonists. Stiffness was measured by optical magnetic
twisting cytometry (OMTC). OMTC is a technique whereby microscopic (4.5 µm)
ferromagnetic beads are passively attached to the cytoskeleton of adherent
cells via focal adhesion linkages. We measured G' by optically tracking
bead motions through an inverted microscope during application of an
oscillating magnetic field.
We used identical protocols because we had earlier employed using cyclic
deformational strain to induce parallel alignment of the cells and
organization of the cytoskeleton
(19). We found that,
functionally, airway smooth muscle cells subjected to mechanical strain
displayed increased G' at baseline and that these differences in
stiffness were greatly enhanced after exposure of the cells to contractile
agonists. In addition, the G' was found to be anisotropic, being greater
when measured parallel to the cell long axis compared with when measured in
the transverse direction. By selectively using relaxant agonists and
contractile agonists and by probing cell stiffness both parallel and
perpendicular to the cell long axes, we show that filament reorganization
greatly enhanced the generation of G' with contractile activation.
|
METHODS
|
|---|
Cell culture and strain apparatus. Canine trachealis muscle was
harvested and digested in collagenase and elastase with soy trypsin inhibitor
as previously described (20).
Freshly dissociated cells were seeded into flasks at a density of 5 x
104 cells/cm2 in Ham's F-12/DMEM with 10% fetal bovine
serum, penicillin, streptomycin, and amphotericin. Cells were passaged to
collagen type I-coated silastic membranes in a six-well plate (Flexcell,
McKeesport, PA) when 80–90% confluent. First- and second-passage cells
were used for these studies. To subject cells to mechanical stress, we
positioned the plates over a manifold connected to a vacuum source. The vacuum
was programmed by computer software (Flexcell) to cause a 10% increase in
surface area of the membranes for a quasisinusoidal positive half-wave
(tensile only) strain at 0.25 Hz. Cells were subjected to strain as described
above for 10–12 days.
The mechanical strain on the membranes is nonuniform, varying with position
on the membrane. At the center, the strain is uniform and isotropic, but at
the periphery the strain is directed radially. This radial strain is maximal
near the rigid boundary of the well and decreases toward the center where it
is a minimum (7). Cultured
cells migrate and reorient in response to this cyclic strain gradient,
resulting in circumferential alignment of the cells around the periphery of
the wells, giving the impression of a ring of aligned cells, which is absent
in unstrained control cells
(19). In all experiments,
medium was replaced with serum-free medium for 24–48 h before
measurements of G'.
OMTC. To determine G' using OMTC, we grew cells to 
90%
confluence and then changed them to serum-free media supplemented with insulin
and transferrin as previously described
(21). After a rinse with PBS,
membranes were cut from the six-well plates with a scalpel and placed in
serum-free media. Sections were cut from the membranes for experimental
measurement and maintained at 37°C with 5% CO2. Depending on
membrane size and cell density, we placed 100–120 µl of solution
containing 1 mg/ml beads on the cells for 15 min to maintain about two beads
per cell to allow bead attachment. The beads were Fe3O4
ferrimagnetic (diameter 4.5 µm) and coated with an Arg-Gly-Asp-containing
peptide (5). After 15 min, the
cells with beads attached were washed once to remove any unbound beads. The
35-mm dish containing the membrane section was placed on the heated (37°C)
stage of an inverted microscope (Nikon Diaphot). We magnetized beads in the
horizontal plane by using Helmholtz coils by a brief magnetic-field pulse
(<10 ms, >0.1 T). An oscillatory magnetic twisting field was then
applied in the vertical direction (Fig.
1). The specific torque (mechanical torque per unit bead volume)
applied by the magnetic field was thus T = cH cos
,
where is the angle between the horizontal plane and the magnetic
moment of a bead as a function of time. The bead constant, c, is the
torque per unit bead volume per mT and is determined by placing the beads in a
sample of known viscosity and measuring the angular velocity when a magnetic
twisting torque is applied. H is the twisting field amplitude and was set at
either 2 mT [baseline, cytochalasin D, isoproterenol, and dibutyryl (db)-cAMP]
or 5 mT (baseline and KCl experiments). Twisting forces of the lower strength
(2 mT) were used during relaxant and cytoskeletal disruption experiments to
reduce bead displacements when cytoskeletal forces were weakened with these
two agents. Bead movement was recorded with a JAI CV-M10 (JAI, Glostrup,
Denmark) progressive-scan, triggerable, black-and-white charge-coupled device
camera. Images of 780 x 560 pixel size were digitized at eight-bit
resolution and then transferred to a personal computer with an Eltec PC-Eye4
frame grabber (Eltec Electronik, Mainz, Germany). Gray-scale images were
analyzed in real time, and bead positions or images were stored on disc. Bead
positions were computed within a rectangular window (12 x 12 pixels) by
using an intensity centroid algorithm
(5). Position resolution noise
was <5 root mean nm2.
View larger version (8K):
[in this window]
[in a new window]
|
Fig. 1. Schematic of bead twisting attached to cell attached to silastic membrane
substrate. The beads are initially magnetized in the horizontal direction, M,
and are then subjected to a vertical oscillatory magnetic field. The resulting
torque, T, causes the bead to pivot back and forth resisted by the cell
mechanics, and the resultant bead motion is observed as displacement via the
inverted microscope (not shown).
|
|
The camera was phase locked to the sinusoidal twisting field, to obtain 16
images per cycle for frequencies <1 Hz. Because the maximum frame rate of
the camera was limited, heterodyning was employed for twisting frequencies
>1 Hz (which would normally require >16-Hz frame rate). For example, at
twisting frequencies of 100 Hz, the beads were oscillated for 160 cycles, and
the 16 phases for a single were recorded from every 10th twisting cycle,
giving an 10-Hz frame rate.
Twisting protocols. To determine the response of cell stiffness to
varying twisting frequencies, we used a protocol of increasing and decreasing
oscillation frequencies of 0.1, 0.3, 0.76, 4.2, 9.1, 30, 100, and 300 Hz, for
five cycles each, excepting 300 Hz, which was recorded for 10 cycles. This
ascending pattern of frequencies was repeated in reverse order. The total
measurement required <5 min. To determine the time course of drug effects
on cell stiffness, we fixed the oscillation frequency at 0.76 Hz, and either
control measurements were made or a drug was administered at 60 s, and the
response was recorded for 4 min. Before and after every recording (varying
frequency and time course), a single cycle of 0.3-Hz oscillation was recorded
to avoid large transients during data analysis. For all samples, a varying
frequency measurement was recorded first, which we refer to as baseline. After
the baseline measurement, a time-course response to one of the
stiffness-altering drugs was recorded. After the time course, we made a
repeated measurement of the varying frequency experiment, which we refer to as
the plateau measurement. Only one chemical was added to any single preparation
of cells, and separate samples of each population were tested with each
chemical.
Cytoskeletal organization and OMTC stiffness measurements. We
sought additional evidence for cytoskeletal contributions to cell stiffness by
taking advantage of the fact that >90% of the cells under cyclic
deformational strain are oriented perpendicular to the direction of the strain
(19). This orientation is
accompanied by increased alignment of actin filaments parallel to the long
axis of the cell. In contrast, cells not grown under the influence of cyclic
strain lie in random directions with less organized stress fibers. We reasoned
that if cyclic deformational strain increased cell stiffness by organizing
cytoskeletal elements, stiffness measurements should be greatest when the
twisting forces were parallel to the stress fibers (along the long axis of
strained cells). Therefore, we made baseline measurements of cell stiffness by
OMTC in three different manners (Fig.
2): direction of twisting force parallel to long axis of strained
cells (Para), direction of twisting force perpendicular to the long axis of
the cells (Perp), and OMTC measurements from control (randomly oriented) cells
(Cont). To further demonstrate the contribution of the stress fibers to the
stiffness measurements, we exposed cells to cytochalasin D (2 µM) to
disrupt actin filaments. To determine the level of baseline activation of
contractile elements of the cells contributing to stiffness, we made OMTC
measurements before and after administration of the relaxant agonists db-cAMP
(1 mM) or isoproterenol (10 µM). To determine whether baseline increases in
cell stiffness that might be caused by cytoskeletal reorganization might
enhance cell stiffening to contractile agonist, we measured stiffness before
and after KCl (80 mM). Significant differences between baseline and plateau
values were determined by t-test with P < 0.05.
Statistical tests were conducted at either 0.76 or 0.3 Hz. These frequencies
are physiologically relevant, as they are close to breathing frequencies, and
0.76 Hz is the sampling rate chosen in the time-course recordings limited by
the camera frame rate. In any case, the choice was largely arbitrary because
the stiffness dependencies on frequency were parallel among groups
(Fig. 3).
View larger version (48K):
[in this window]
[in a new window]
|
Fig. 2. Images of cells with beads showing the alignment of the cells relative to
the twisting direction (arrows). Images are bright field, with contrast
digitally enhanced to reveal the cell edges. A: control cells (Cont)
with no apparent alignment; B: strained cells aligned in the parallel
twisting direction (Para); C: strained cells aligned in the
perpendicular direction (Perp).
|
|
Reagents. All tissue culture reagents and chemicals were from
Sigma Chemical (St. Louis, MO) with the exception of the trypsin-EDTA
solution, which was purchased from GIBCO (Grand Island, NY). KCl was diluted
in PBS at 240 mM, and 1 ml was added to the 35-mm dish for a final
concentration of 80 mM. Db-cAMP was dissolved at 10-1 M
in distilled water, frozen in aliquots, and thawed on the day of use.
Isoproterenol was prepared 1 h before use, and cytochalasin D was frozen in
aliquots and thawed on the day of use. Solutions of db-cAMP, cytochalasin D,
and isoproterenol were added in volumes of 1 ml of medium to 2 ml of medium
containing the cells.
Data analysis. The recorded data were digitally filtered with a
band-pass filter 48 points in length to remove baseline drift and
high-frequency noise in the bead positions. For frequency recordings, the data
were then separated according to oscillation frequency and separately analyzed
by Fourier transform. The Fourier-transformed bead motion signal from a single
frequency, 
, and the specific torque
, were used to compute a complex modulus,
defined here as
 | (1) |
where G has dimensions of Pascals per nanometer and is related by a geometric
factor to the complex shear modulus of the cell, and j is the unit
imaginary number 
-1. The component of bead displacement that is in
phase with T is the real part of G and is denoted G'. G' is the
measure of the elasticity or stiffness, which we use to quantify the cell
stiffness. G'' is a measure of the friction, representing the
out-of-phase bead motion to the applied torque and is typically <0.2
G' (3). For time-course
recordings, each cycle of the 0.76-Hz oscillations was analyzed cycle by
cycle, giving 1.3-s time resolution. Because the distribution of G' from
OMTC is approximately log-normal, results are reported as the medians,
together with either SE or geometric standard deviation (GSD) where indicated.
GSD of G' was calculated as the SD of the natural logarithm of the
G' from all beads for a given treatment group (Para, Perp, or Control).
Measurements were recorded from 4–8 membranes, yielding from a minimum
of 600 beads (300 cells) to as many as 1,800 beads, which was more than
sufficient to accurately determine the median G' for each group
(5).
Bead acceptance criteria. Beads with erratic motions below the
noise level of the OMTC system were eliminated according to the following
criteria. Because all frequencies were recorded twice, beads were eliminated
if G' changed by more than a factor of two at any frequency. Beads were
also eliminated if the waveform response of the bead motion approached a
square wave rather than that of a sinusoidal wave. This was done by
eliminating beads with a 2nd harmonic of >0.18 of the fundamental
magnitude. The choice of 0.18 of the fundamental was because this is the
halfway point between the 2nd harmonic magnitude of a sinusoid (zero) and a
square wave (0.36 of the fundamental). Beads were eliminated if they did not
maintain the elastic component of the bead motion within 30 degrees of the
twisting direction (at 0.76 Hz). The top 5% and lower 5% of all beads (motions
measured at 0.76 Hz) for any measurement were also rejected (this had no
effect on the median for any recording). The most common rejection was for
bead angle outside the ±30-degree range followed by the upper and lower
5%. The 2nd harmonic test eliminated the least number of beads. Together,
these criteria eliminated 1/3 of the beads.
|
RESULTS
|
|---|
Under baseline conditions, G' measured parallel to the cell's long
axes was significantly greater than G' measured either perpendicular in
the strained cells or G' measured from the control unstrained cells
(P > 0.05 Para vs. Perp, and P > 0.01 Para vs. Cont
determined at 0.76 Hz, see Fig.
3 insert). All G' data exhibited a weak power law
f b with exponent b = 0.172 ± 0.008 (±SE),
with no significant difference of the exponents between groups as indicated by
the parallel regressions on the log-log plot
(Fig. 3). The rank ordering of
baseline stiffness among groups (Cont, Perp, Para, respectively) was the same
regardless of oscillatory frequency (Fig.
3).
The response to contractile agonists KCl is shown in
Fig. 4. There was a rapid rise
in stiffness after administration of the KCl at 60 s, followed by a slower
decline to a plateau level after 140 s. Because the G' distribution
was approximately log-normally distributed, the dotted trace shows the
location of median/GSD of G' corresponding to approximately the lower
16th percentile rank (analogous to the 16th percentile rank described by the
mean minus the SD for a normal distribution).
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 4. Time course of response to KCl obtained parallel to long axes of the cells.
Median displacement (D, solid top trace) decreased after
administration of KCl at 60 s (arrow), then increased to a plateau below
baseline, and the stiffness (G', bottom panel) increased then
decreased to a plateau above baseline. The dotted traces show the median
divided by geometric SD.
|
|
After activation with KCl, G' was increased at all frequencies in all
groups. The increase in G' as percentage of baseline was not
significantly different in control cells from strained cells Perp, 125
± 8% (Fig. 5,
left and center). However, the increase in G' due to
KCl was significantly greater Para at 171 ± 7% than either Perp, which
was 125 ± 8%, or Cont, which was 129 ± 8% (P <
10-5, Fig.
5).
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 5. Plateau response of cell stiffness to 80 mM KCl (open symbols) vs. baseline
(solid symbols) as a function of bead oscillation frequency on semilog axes.
The control cells (Cont) and strained cells with beads oscillated
perpendicular to the long axis (Perp) stiffen comparably in response to KCl,
whereas the beads twisted parallel to the long axis (Para) respond
significantly more (P < 10-5, 0.3 Hz). The
lines are best-fit power-law regressions, and error bars show SE. There was no
significant difference in power-law exponent among Cont, Perp, or Para groups
or between baseline and stimulated in each group.
|
|
Comparing drug responses at a single frequency (0.3 Hz), we found the
following (Fig. 6): 1)
isoproterenol and db-cAMP significantly decreased G' from baseline
measured with P < 0.05 for all groups, excepting db-cAMP in the
Cont; 2) in a comparison of decreases from db-cAMP and isoproterenol,
there were no significant differences in the decrease as a percentage of
baseline for any group (Cont, Para, or Perp); 3) cytochalasin D
decreased stiffness from baseline significantly for each group; and
4) cytochalasin D also decreased G' significantly more in
strained cells Para and Perp compared with Cont (P < 0.05,
Fig. 6).
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 6. Plateau changes in G' as a percentage of baseline G' for each
group obtained at 0.3-Hz bead oscillation frequency. After KCl, G' was
significantly more increased in the strained cells measured in the parallel
direction compared with that measured in the perpendicular direction
(P < 10-5) or compared with control cells
(P < 10-5), whereas there was no difference
in increase in the perpendicular direction relative to control. Similar
decreases in stiffness were observed for both dibutyryl cAMP (db-cAMP) and
isoproterenol, whereas there was a greater decrease after cytochalasin D in
the strained cells compared with the decrease in control cells measured both
parallel (P < 0.01) and perpendicularly (P < 0.05) to
the cell long axes.
|
|
|
DISCUSSION
|
|---|
The primary results of this study are that cultured airway smooth muscle
cells exposed to 10–12 days of mechanical strain have increased
G', and this stiffness is greatly enhanced by contractile agonists
compared with unstrained, time-matched controls. Furthermore, we found that
the stiffness after stimulation was greatest when measured in the parallel
direction to the cell long axis, whereas contractile activation only modestly
increased stiffness measured in the perpendicular direction. In the following
sections, we discuss the methodology and similarities of our results with
others. We then discuss the relevance and implications of increased airway
smooth muscle stiffness in asthma.
These results are consistent with and extend previous results. We have
previously shown that mechanical strain causes several changes in smooth
muscle cell structure and function that suggest enhancement of contractile
function. In addition to increases in smooth muscle-specific protein content
(21) and activity
(20), mechanical strain causes
cytoskeletal filaments to be more aligned and the cells more elongated
(Fig. 2 and Ref.
19). Functionally, strain
increases the shortening capacity and velocity and increases force generation
(22,
23). In those studies, force
production increases could not be entirely accounted for by increases in
myosin or myosin light chain kinase activity, and our speculation was that
enhanced contractile filament organization may have led to the enhanced force
production. The results we now report support this.
Stiffness anisotropy. We were able to probe G' and
enhancement of stiffness during cell activation relative to cell orientation.
Using strain regimens identical to those applied here, we had previously
demonstrated that stress fibers of airway smooth muscle cells become aligned
in thick bundles with a unified vector along the long axis of the
spindle-shaped cells (Ref. 19
and shown in Fig. 2). Because
the beads were magnetized along a particular vector, the twisting direction
could be chosen relative to cell orientation, and we took advantage of this
and tested stiffness both parallel and perpendicular to cytoskeletal
organization in strained cells and made comparisons before and after
stimulation in both strained and control cells. We found a large difference
between stiffness measured in the parallel direction compared with the
perpendicular direction, relative to the cell long axis, both at baseline and
after stimulation with KCl. On the other hand, KCl only modestly increased
stiffness measured in the perpendicular direction (Figs.
5 and
6). In control cells that had
much less alignment and less apparent cytoskeletal organization compared with
the strained cells (19), only
modest increases in stiffness were observed. Together, these findings are
important primarily because they imply that the orientation of the contractile
lattice is aligned to produce efficient contraction and stiffness generation.
This interpretation of G' from OMTC is in agreement with earlier
interpretations that stiffness reflected the number of attached acto-myosin
cross bridges (5) and also
reflects the stiffness of the actin scaffolding
(2). Furthermore, we found that
there were no significant differences in the magnitude of decrease in baseline
stiffness measured after either db-cAMP or isoproterenol when comparing the
decreases between strained and unstrained cells
(Fig. 6). This would seem to
indicate that baseline stiffness is more dependent on the cytoskeletal
organization than contractile protein activation.
However, organization is not likely to have been the only contributor to
enhancing baseline stiffness, because control cells were not as stiff as the
cells measured perpendicular to filament direction. If organization of
cytoskeletal elements was the only difference between unstrained cells and the
strained cells, then stiffness of control cells would be between the stiffness
perpendicular and the stiffness parallel to the cytoskeletal alignment,
because in randomly oriented cells some lie parallel and some lie
perpendicular to the bead twisting. Other changes that we have previously
documented in the contractile protein content may account for this finding
(21). Additionally, strain may
activate other signaling events that enhance contractile function. For
example, we have recently described strain-induced activation of RhoA, a small
GTPase that is known to regulate smooth muscle contractility through several
effects (24). Strain has an
immediate effect enhancing contractile response in smooth muscle through
stretch-activated calcium channels
(1,
16,
25). We previously found that
strain causes an increase in calcium sensitivity of force production in
permeabilized cells that had been subjected to strain. However, even when
calcium concentration was controlled for, increases in force production
demonstrated that other effects of strain were responsible
(23). Thus strain apparently
leads to increased contractile function through different mechanisms, which
may or may not be independent: increased cellular organization, increased
contractile-associated proteins, and changes in regulation of contractile
function.
OMTC. We used OMTC
(3,
5) to measure the stiffness of
the cytoskeleton of canine airway smooth muscle cells. This technique probes
the mechanical properties of the cytoskeleton through sinusoidal mechanical
twisting of a ferromagnetic bead that is linked to the cytoskeleton through
ligand-receptor binding (3,
5). We found that stiffness was
highly heterogeneous and was distributed approximately log-normally, as has
previously been reported with this technique
(3,
5) and its progenitor, magnetic
twisting cytometry (4,
15). Heterogeneity from OMTC
has been largely attributed to differences in bead attachment characteristics,
such as the number of binding sites linking the bead to the focal adhesion and
the focal adhesion linkage to the cytoskeleton, but a significant portion of
the heterogeneity is likely from variations in properties among cells
(5). The heterogeneity inherent
in the method and in the cells themselves requires that a sufficient number of
beads (and cells) be probed
(5). We were able to show
significant differences in our data from groups including from 600 to
>1,800 beads, as we describe in METHODS.
With contractile agonist, both the time course of stiffening and degree of
stiffening we report in control canine smooth muscle cells (Figs.
4 and
5) were similar to previous
results in human airway smooth muscle cells
(5,
15). Similarly, the time
courses (not shown) and decrease in stiffness with relaxant agonists
(Fig. 6) are in agreement with
earlier findings (4,
10,
15). We attribute the
stiffness and its changes measured here and obtained previously to the elastic
properties of the airway smooth muscle cytoskeleton and contractile machinery
(3,
15). However, it may have been
that other components, such as the lipid membrane, sub-membrane organelles,
etc., contributed to the stiffness. Possibly these structures would impede
bead motion along the cell axis compared with that perpendicular to it,
leading to measured stiffness anisotropy, but for a number of reasons we
believe this to be unlikely. The stiffness of these structures is thought to
be much less than that of the actin cytoskeleton
(3). Also with disruption of
the cytoskeleton, we found that G' decreased to <20% of its original
value, in agreement with earlier studies
(3). Furthermore, the percent
decreases with cytochalasin D were comparable whether measured parallel or
perpendicular to the cell long axis (Fig.
6), implying that an intact actin cytoskeleton contributed the
dominant part of the measured stiffness. Moreover, in studies where the beads
are bound only to the lipid membrane with nonspecific binding or via
acetylated LDL receptors, the G' recorded are much lower than with
integrin binding (3) and are
similar to the levels we find after actin disruption. Finally, the
disproportionate increases in cell stiffness we measured parallel to the cell
long axis compared with the perpendicular direction or with the control cells
also implicates underlying cytoskeletal and contractile filaments as the
origin of the measured stiffness rather than membrane structures.
We explored the frequency-dependent stiffness of the stiffness between 0.1
and 300 Hz and found a power-law dependence of G' on frequency. This
observation has been previously reported and is generally consistent with the
notion that the cytoskeleton behaves as a glassy system as reported by Fabry
et al. (3). In these systems,
changes in the structural state are mechanistically linked to changes in a
single parameter, the "noise temperature," which is defined by the
exponent of the power-law stiffness minus one
(3). We observed a power-law
frequency dependence for G' (Figs.
4 and
6) with exponent 0.17, which is
in agreement with the report by Fabry et al. and found, as they did
(3), that the exponent was
increased after administration of muscle relaxants and cytoskeletal
disruption. The notion that cytoskeletal remodeling follows the framework of
glassy systems may have important implications. We had speculated that the
changes in cytoskeletal organization induced by mechanical strain would alter
the mechanical state of the cells and that we would see differences in
exponent between groups. However, we did not find any significant differences
in the slope of the G' frequency dependence between groups (Cont, Perp,
or Para) at baseline (Fig. 4)
or after contractile or relaxant agonists; the reasons for this are
unclear.
Mechanical plasticity of the airway smooth muscle cell and chronic
strain. It is now well established that the contractile function of the
airway smooth muscle cell is highly dependent on the length-tension history
and is highly adaptable on the time scale of minutes
(18). For example, the muscle
is able to generate near maximal force and stiffness over more than a twofold
change in length, and oscillatory length perturbations are able to decrease
average force and stiffness to 20%. These and similar length-tension
adaptation behaviors have been described as mechanical plasticity, and a
number of molecular mechanisms have been postulated
(9,
17,
18). For example, Pratusevich
et al. (17) postulated that
with length changes, changes occur in the organization of contractile units in
series to optimize force production, implying a change in cytoskeletal
organization. Gunst and colleagues
(8,
9) postulated that changes in
length may stimulate remodeling of the actin lattice by either changing the
length of the filaments or changing the sites of attachment to the cell
membrane or both in a manner that optimizes force production. Additionally,
airway smooth muscle exhibits changes in myosin filament density when
contracted at different lengths, indicating that myosin filaments may reform
between activations, permitting changes in force generation and stiffness
(14). Each of these mechanisms
occurs within minutes. In comparison, the differences we observe in cell
structure and function from mechanical strain occurred after several days of
continuous cyclic strain of a magnitude approximating physiological movement
of the airways. In this regard, the strain cells more closely mimic
physiological conditions than the control cells grown in static conditions.
Thus the changes in mechanical stiffness, shortening velocity and capacity,
and force generation due to long duration mechanical strain we report here and
previously are likely due to different mechanisms than the changes seen after
an acute length change in other studies. However, it is likely that mechanical
strain has altered the mechanical plasticity of these cells, although this
remains to be tested. The present data indicate that alignment of cytoskeletal
elements contributes to cell stiffness and the efficiency of contraction.
Airway smooth muscle stiffness and asthma. Therefore, the data we
present here demonstrate an important means whereby increased smooth muscle
G', in response to chronic strain, might contribute to the observed
failure of deep inspirations to induce bronchodilation in asthmatics. It is
known that a normal individual can dilate constricted airways with a deep
breath, whereas an asthmatic fails to do so
(11,
12). The mechanism for airway
dilation may reside within the smooth muscle. The normal response of activated
airway smooth muscle to acutely delivered oscillatory strain is a rapid
decrease in stiffness and force generation
(6). This decrease requires
that activated smooth muscle receive sufficient oscillatory loading to
adequately stretch the acto-myosin cross bridges. Independent of changes in
passive parenchymal lung tissue components, increases in smooth muscle
stiffness would decrease the oscillatory stretch imparted to the smooth
muscle. Thus, if chronic activation and/or mechanical stress increases
cytoskeletal organization, this could translate into increased stiffness, with
reduced stretch precipitating a vicious cycle of stiffness preventing
relaxation and enhancing further contractility.
|
ACKNOWLEDGMENTS
|
|---|
We thank Ben Fabry and Chaity Roy for technical assistance.
DISCLOSURES
This work was supported by National Institutes of Health Grants P01
HL-33009, RO1 HL-59682, RO1 HL/AI-65960, and HL-0340905; the Nova Scotia
Health Research Foundation; and the Whitaker Foundation.
|
FOOTNOTES
|
|---|
Address for reprint requests and other correspondence: G. N. Maksym, School of
Biomedical Engineering, Dalhousie Univ., Halifax, N. S. B3H 3J5, Canada
(E-mail:
geoff.maksym{at}dal.ca).
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
|
REFERENCES
|
|---|
- An SS and Hai
CM. Mechanical signals and mechanosensitive modulation of intracellular
[Ca2+] in smooth muscle. Am J Physiol Cell
Physiol 279:
C1375–C1384, 2000.[Abstract/Free Full Text]
- An SS, Laudadio
RE, Lai J, Rogers RA, and Fredberg JJ. Stiffness changes in cultured
airway smooth muscle cells. Am J Physiol Cell Physiol
283: C792–C801,
2002.[Abstract/Free Full Text]
- Fabry B, Maksym
GN, Butler JP, Glogauer M, Navajas D, and Fredberg JJ. Scaling the
microrheology of living cells. Phys Rev Lett
87: 148102,
2001.[Medline]
- Fabry B, Maksym
GN, Hubmayr RD, Butler JP, and Fredberg JJ. Implications of heterogeneous
bead behavior on cell mechanical properties measured with magnetic twisting
cytometry. J Magnetism Magnetic Materials
194: 120–125,
1999.
- Fabry B, Maksym
GN, Shore SA, Moore PE, Panettieri RA Jr, Butler JP, and Fredberg JJ. Time
course and heterogeneity of contractile responses in cultured human airway
smooth muscle cells. J Appl Physiol
91: 986–994,
2001.[Abstract/Free Full Text]
- Fredberg JJ,
Inouye D, Miller B, Nathan M, Jafari S, Raboudi SH, Butler JP, and Shore
SA. Airway smooth muscle, tidal stretches, and dynamically determined
contractile states. Am J Respir Crit Care Med
156: 1752–1759,
1997.[Abstract/Free Full Text]
- Gilbert JA,
Weinhold PS, Banes AJ, Link GW, and Jones GL. Strain profiles for circular
cell-culture plates containing flexible surfaces employed to mechanically
deform cells in-vitro. J Biomech
27: 1169–1177,
1994.[ISI][Medline]
- Gunst SJ, Meiss
RA, Wu MF, and Rowe M. Mechanisms for the mechanical plasticity of
tracheal smooth muscle. Am J Physiol Cell Physiol
268: C1267–C1276,
1995.[Abstract/Free Full Text]
- Gunst SJ and Wu
MF. Plasticity of airway smooth muscle stiffness and extensibility: role
of length-adaptive mechanisms. J Appl Physiol
90: 741–749,
2001.[Abstract/Free Full Text]
- Hubmayr RD,
Shore SA, Fredberg JJ, Planus E, Panettieri RA Jr, Moller W, Heyder J, and
Wang N. Pharmacological activation changes stiffness of cultured human
airway smooth muscle cells. Am J Physiol Cell Physiol
271: C1660–C1668,
1996.[Abstract/Free Full Text]
- Jensen A,
Atileh H, Suki B, Ingenito EP, and Lutchen KR. Airway caliber in healthy
and asthmatic subjects: effects of bronchial challenge and deep inspirations.
J Appl Physiol 91:
506–515, 2001.[Abstract/Free Full Text]
- Kapsali T,
Permutt S, Laube B, Scichilone N, and Togias A. Potent bronchoprotective
effect of deep inspiration and its absence in asthma. J Appl
Physiol 89:
711–720, 2000.[Abstract/Free Full Text]
- King GG, Pare
PD, and Seow CY. The mechanics of exaggerated airway narrowing in asthma:
the role of smooth muscle. Respir Physiol
118: 1–13,
1999.[ISI][Medline]
- Kuo KH, Wang L,
Pare PD, Ford LE, and Seow CY. Myosin thick filament lability induced by
mechanical strain in airway smooth muscle. J Appl
Physiol 90:
1811–1816, 2001.[Abstract/Free Full Text]
- Maksym GN,
Fabry B, Butler JP, Navajas D, Tschumperlin DJ, Laporte JD, and Fredberg
JJ. Mechanical properties of cultured human airway smooth muscle cells
from 0.05 to 0.4 Hz. J Appl Physiol
89: 1619–1632,
2000.[Abstract/Free Full Text]
- Mehta D, Wu MF,
and Gunst SJ. Role of contractile protein activation in the
length-dependent modulation of tracheal smooth muscle force. Am J
Physiol Cell Physiol 270:
C243–C252, 1996.[Abstract/Free Full Text]
- Pratusevich VR,
Seow CY, and Ford LE. Plasticity in canine airway smooth muscle.
J Gen Physiol 105:
73–94, 1995.[Abstract/Free Full Text]
- Seow CY and
Fredberg JJ. Historical perspective on airway smooth muscle: the saga of a
frustrated cell. J Appl Physiol
91: 938–952,
2001.[Abstract/Free Full Text]
- Smith PG,
Garcia R, and Kogerman L. Strain reorganizes focal adhesions and
cytoskeleton in cultured airway smooth muscle cells. Exp Cell
Res 232:
127–136, 1997.[ISI][Medline]
- Smith PG,
Janiga KE, and Bruce MC. Strain increases airway smooth muscle cell
proliferation. Am J Respir Cell Mol Biol
10: 85–90,
1994.[Abstract]
- Smith PG,
Moreno R, and Ikebe M. Strain increases airway smooth muscle contractile
and cytoskeletal proteins in vitro. Am J Physiol Lung Cell Mol
Physiol 272:
L20–L27, 1997.[Abstract/Free Full Text]
- Smith PG, Roy
C, Dreger J, and Brozovich F. Mechanical strain increases velocity and
extent of shortening in cultured airway smooth muscle cells. Am J
Physiol Lung Cell Mol Physiol 277:
L343–L348, 1999.[Abstract/Free Full Text]
- Smith PG, Roy
C, Fisher S, Huang QQ, and Brozovich F. Mechanical strain increases force
production and calcium sensitivity in cultured airway smooth muscle cells.
J Appl Physiol 89:
2092–2098, 2000.[Abstract/Free Full Text]
- Smith PG, Roy
C, Zhang YN, and Chauduri S. Mechanical stress increases RhoA activation
in airway smooth muscle cells. Am J Respir Cell Mol
Biol 28:
436–442, 2003.[Abstract/Free Full Text]
- Somlyo AP and
Himpens B. Cell calcium and its regulation in smooth muscle.
FASEB J 3:
2266–2276, 1989.[Abstract]
- Stephens NL. Airway smooth muscle.
Lung 179:
333–373, 2001.[ISI][Medline]
- Thomson RJ and
Schellenberg RR. Increased amount of airway smooth muscle does not account
for excessive bronchoconstriction in asthma. Can Respir
J 5: 61–62,
1998.[Medline]
- Tschumperlin DJ, Shively JD, Swartz MA, Silverman ES, Haley KJ,
Raab G, and Drazen JM. Bronchial epithelial compression regulates MAP
kinase signaling and HB-EGF-like growth factor expression. Am J
Physiol Lung Cell Mol Physiol 282:
L904–L911, 2002.[Abstract/Free Full Text]
- Wang N,
Tolic-Norrelykke IM, Chen J, Mijailovich SM, Butler JP, Fredberg JJ, and
Stamenovic D. Cell prestress. I. Stiffness and prestress are closely
associated in adherent contractile cells. Am J Physiol Cell
Physiol 282:
C606–C616, 2002.[Abstract/Free Full Text]
Copyright © 2003 by the American Physiological Society.
TOP
ABSTRACT
METHODS
) and strained cells measured perpendicular (
) and parallel
(
) to cell long axes. Insert: difference between cell groups
on nonlog scale at 0.76 Hz. Control and strain were measured perpendicular and
parallel to the cell long axes. Para (*) was significantly stiffer
than both Perp (P < 0.05) and Cont (P < 0.01).