Am J Physiol Cell Physiol 286: C831–C839, 2004.
First published November 26, 2003; 10.1152/ajpcell.00224.2003.
Depletion of plasma membrane cholesterol dampens hydrostatic pressure and
shear stress-induced mechanotransduction pathways in osteoblast cultures
Jeffrey T. Ferraro,1 Mani Daneshmand,1 Rena Bizios,2 and Victor Rizzo1
1
Center for Cardiovascular Sciences, Albany Medical College, Albany, New York 12208; and
Department of Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, New York 12180
2
Submitted 30 May 2003; accepted in final form 20 November 2003
Ferraro, Jeffrey T., Mani Daneshmand, Rena Bizios, and Victor Rizzo. Depletion of plasma membrane cholesterol dampens hydrostatic pressure and shear stress-induced mechanotransduction pathways in osteoblast cultures. Am J Physiol Cell Physiol 286:
C831–C839, 2004. First published November 26, 2003; 10.1152/
ajpcell.00224.2003.—The preferential association of cholesterol and
sphingolipids within plasma membranes forms organized compartments termed lipid rafts. Addition of caveolin proteins to this lipid
milieu induces the formation of specialized invaginated plasma membrane structures called caveolae. Both lipid rafts and caveolae are
purported to function in vesicular transport and cell signaling. We and
others have shown that disassembly of rafts and caveolae through
depletion of plasma membrane cholesterol mitigates mechanotransduction processes in endothelial cells. Because osteoblasts are subjected to fluid-mechanical forces, we hypothesize that cholesterol-rich
plasma membrane microdomains also serve the mechanotransduction
process in this cell type. Cultured human fetal osteoblasts were
subjected to either sustained hydrostatic pressure or laminar shear
stress using a pressure column or parallel-plate apparatus, respectively. We found that sustained hydrostatic pressure induced protein
tyrosine phosphorylation, activation of extracellular signal-regulated
kinase (ERK)1/2, and enhanced expression of c-fos in both time- and
magnitude-dependent manners. Similar responses were observed in
cells subjected to laminar shear stress. Both sustained hydrostatic
pressure- and shear stress-induced signaling were significantly reduced in osteoblasts pre-exposed to either filipin or methyl-␤-cyclodextrin. These mechanotransduction responses were restored on reconstitution of lipid rafts and caveolae, which suggests that cholesterol-rich plasma membrane microdomains participate in the
mechanotransduction process in osteoblasts. In addition, mechanical
force-induced phosphoproteins were localized within caveolin-containing membranes. These data support the concept that lipid rafts and
caveolae serve a general function as cell surface mechanotransduction
sites within the plasma membrane.
lipid rafts; caveolae; extracellular signal-regulated kinase
in response to mechanical
loading of the skeletal system play an important role in bone
development and homeostasis; removal of physical stimuli
proves detrimental to the skeletal tissue in vivo (17, 51).
Osteoblasts are primary regulators of bone metabolism and
appear to be particularly sensitive to changes in their physical
environment. In response to fluid flow-induced shear stress,
osteoblasts generate (5, 24–27, 39) prostaglandin E2 (PGE2),
inositol 1,4,5-trisphosphate (IP3), cAMP, cyclooxygenase-2
(COX-2), and nitric oxide (NO). Likewise, application of
compressive forces to osteoblast cultures stimulates prostacyclin production and enhances expression of the early response
BIOMECHANICAL STRESSES GENERATED
Address for reprint requests and other correspondence: V. Rizzo, Center for
Cardiovascular Science, Albany Medical College, 47 New Scotland Ave.,
Albany, NY 12208 (E-mail: [email protected]).
http://www.ajpcell.org
gene c-fos (9) and alkaline phosphatase (19). Although the
effects of mechanical stimulation on osteoblast function pertinent to new bone formation continue to be the center of study,
the basic mechanisms by which biomechanical forces are
detected and translated into biological signals within these cells
are not clearly defined.
There is accumulating evidence to support the concept that
mechanochemical signaling occurs in spatially discrete subcompartments within cells. Perhaps the best characterized of
these are the adhesions and focal contacts present on cell
membranes. In endothelial cells, shear stress has been shown to
activate integrin complexes that trigger localized accumulation
of a wide variety of signaling molecules (2, 18, 48). Similar
evidence for integrin-mediated mechanotransduction in osteoblasts exposed to shear stress has been observed, suggesting
that focal contacts may serve as mechanosensitive sites within
this cell type (22).
The recent discovery that many receptors and signaling
molecules preferentially localize to plasma membrane regions
composed of glycerolphospholipids and enriched in cholesterol
and sphingolipids, termed lipid rafts and caveolae, has led to
the hypothesis that these membrane microdomains also serve
as signaling compartments (4, 38, 40). The importance of
plasma membrane microdomain signaling was highlighted in a
recent study that identified caveolin-enriched membrane fractions as important structures necessary for proper growth
factor-mediated signaling in osteoblasts (43). Interestingly,
mechanotransduction in osteoblasts appears to involve many of
the same signaling components that are necessary for growthfactor signaling. On the basis of these observations and our
past studies demonstrating that acute mechanosignaling events
are localized within and depend on the structural integrity of
caveolae in endothelial cells (28, 30), we tested whether
cholesterol-rich caveolin-containing plasma membrane domains are necessary for efficient mechanotransduction in cultured osteoblasts.
METHODS
Cell culture. Human fetal osteoblasts (hFOB 1.19, American Type
Culture Collection) were cultured in a 1:1 mixture of Ham’s F-12
medium/DMEM without phenol red and supplemented with 10%
FBS. Cells were cultured in a humidified 5% CO2-95% air environment at 37°C.
Hydrostatic pressure experiments. Osteoblasts were seeded onto
35-mm round glass coverslips and cultured to confluency. Before
placement in the hydrostatic pressure chamber, cells were deprived of
serum for 2 h to reduce the level of growth factor-stimulated signaling
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.
0363-6143/04 $5.00 Copyright © 2004 the American Physiological Society
C831
C832
PLASMA MEMBRANE CHOLESTEROL AND MECHANOTRANSDUCTION
and proliferation. Coverslips were fixed to the base of the sterile
polypropylene cylinder (height, 11 cm; diameter, 3 cm) with a small
amount of sterile vacuum grease placed on the side opposite of the
cells. Osteoblasts were then exposed to sustained hydrostatic pressures of either 5 or 10 mmHg above ambient atmospheric pressure for
either 10 or 30 min. In some experiments, osteoblasts were pretreated
with methyl-␤-cyclodextrin (CD) or filipin as described (see Lipid raft
and caveolae manipulation with cholesterol-interacting agents). After
pressure exposure, the osteoblasts were lysed in 1 ml of HEPESbuffered sucrose that contained protease inhibitors [0.25 M sucrose,
25 mM HEPES at pH 7, 10 ␮g/ml leupeptin, 10 ␮g/ml pepstatin A, 10
␮g/ml o-phenanthroline, 10 ␮g/ml 4-(2-aminoethyl)benzenesulfonyl
fluoride, and 50 ␮g/ml trans-epoxysuccinyl-L-leucinamido(4-guanidino)butane] and were processed for SDS-PAGE and Western blot
analysis.
Shear-stress experiments. Osteoblasts were seeded onto 75 ⫻
38-mm glass slides and cultured to confluency. After serum deprivation for 2 h, cells were subjected to 10 dyn/cm2 of laminar shear stress
in a parallel-plate apparatus for 10 min. Serum-free culture medium
was recirculated using a flow circuit composed of a variable-speed
peristaltic pump, a fluid capacitor that damped pulsation, and a
reservoir with culture medium. Laminar shear stress was achieved by
circulating media through a flow chamber created by separating the
glass slide and a Plexiglas top plate with a 12-␮m silicon gasket.
Temperature was maintained at 37°C, and pH and oxygen levels were
controlled with a humidified 95% air-5% CO2 gas mixture flowing
over the medium in the reservoir. Shear stress was quantified using the
following equation: shear stress ⫽ 6␮Q̇/wh2, where ␮ is fluid viscosity, Q̇ is the volumetric flow rate, and w and h are the chamber width
and height, respectively. In some experiments, osteobalsts were incubated with CD during the final 30 min of the serum-deprivation period
(as described in Lipid raft and caveolae manipulation with cholesterol-interacting agents).
Lipid raft and caveolae manipulation with cholesterol-interacting
agents. Cholesterol is an essential component of lipid rafts and
caveolae and is required for the structural integrity of these microdomains. CD is a cyclic oligosaccharide that effectively extracts cholesterol from the plasma membrane, which results in lipid raft and
caveolae disassembly (4, 38). To evaluate the role of cholesterol on
mechanotransduction processes, osteoblasts were incubated in serumfree media that contained 1 mM CD for 30 min at 37°C before
exposure to either hydrostatic pressure or shear stress. In a separate set
of experiments, cholesterol was added back to cholesterol-depleted
osteoblast cell cultures to reconstitute disassembled rafts. Briefly,
repletion with cholesterol was accomplished by incubating cells in the
presence of a cholesterol-CD mixture for 1 h at 37°C. A stock solution
of 0.4 mg/ml cholesterol and 10% CD was prepared by vortexing at
40°C in 10 ml of 10% CD with 200 ␮l of cholesterol (20 mg/ml in
ethanol solution; Ref. 20).
Filipin is a polyene macrolide antibiotic. This class of drugs binds
sterols such as cholesterol and can cause reversible disassembly of
rafts and caveolae (33, 37). As a complementary set of experiments to
the CD studies, cells were pretreated with 5 ␮g/ml filipin for 5 min at
37°C before pressure treatment.
Preparation of caveolin-containing lipid raft membranes. Caveolin-enriched, light-buoyant-density membranes were prepared by detergent extraction essentially as described (23). Briefly, osteoblasts
were lysed on ice for 30 min in 1 ml of extraction buffer [25 mM
HEPES, pH 7.0, 150 mM NaCl, 1% Triton X-100, 10 ␮g/ml leupeptin, 10 ␮g/ml pepstatin A, 10 ␮g/ml o-phenanthroline, 10 ␮g/ml
4-(2-aminoethyl)benzenesulfonyl fluoride, and 50 ␮g/ml trans-epoxysuccinyl-L-leucinamido(4-guanidino)butane]. Cell lysates were collected by scraping and were passed through a 21-gauge needle 10
times. The lysates were mixed with an equal volume of 80% sucrose
in 25 mM HEPES and placed in the bottom of a centrifuge tube. After
a continuous layer of sucrose (35–5%) was overlaid on the lysates, the
tubes were centrifuged at 100,000 g for 18 h at 4°C in an SW 55 rotor.
AJP-Cell Physiol • VOL
Fractions (0.5 ml) were collected beginning at the top of the gradient.
A 20-␮l sample from each fraction was prepared for Western blot
analysis as described below.
Western blot analysis. Proteins were prepared in sample buffer
[170 mM Tris䡠HCl, pH 6.8, 3% (wt/vol) SDS, 1.2% (vol/vol) ␤-mercaptoethanol, 2 M urea, and 3 mM EDTA] and separated by SDSPAGE (10% gel) before electrotransfer to nitrocellulose filters for
immunoblotting. Filters were incubated with indicated primary antibodies followed by either anti-mouse or anti-rabbit horseradish peroxidase secondary antibody. Proteins were detected using enhanced
chemiluminescence substrate (ECL, Amersham). Densitometric quantification of immunoblots using ImageQuant software (Molecular
Dynamics, Sunnyvale, CA) was performed to enable direct comparisons of ECL signals between each group. Protein content of each
sample was determined by bicinchoninic acid analysis.
Statistical analysis. At least three independent experiments were
conducted for each study. Raw data from individual experiments were
collected and pooled according to group (i.e., control, pressuretreated, shear-treated, CD-filipin-pretreated, or CD-filipin-pretreated
cells exposed to pressure or shear). Means and standard deviations
were calculated for each group and analyzed with unpaired two-tailed
Student’s t-test using Statgraphics 4.0 software (Statistical Graphics).
Differences between control and experimental groups were deemed to
be significant at P ⬍ 0.05.
RESULTS
Hydrostatic pressure activates signaling transduction in
osteoblasts. Osteoblast responsiveness to fluid mechanical
stimulation was measured by analysis of specific signal transduction parameters that describe mechanotransduction events.
We observed that total protein tyrosine phosphorylation increased in osteoblasts exposed to sustained hydrostatic pressures of 5 and 10 mmHg compared with cells maintained at
ambient pressure. Compared with results obtained from control
cells, protein tyrosine phosphorylation increased nearly 3.5fold in cells subjected to 5 mmHg pressure and 7.3-fold at 10
mmHg for 10 min (Fig. 1). This pressure-induced increase in
protein tyrosine phosphorylation appeared to be transient;
osteoblasts subjected to 5 mmHg of hydrostatic pressure sustained for 30 min showed no significant difference in phosphotyrosine levels compared with control (Fig. 1). Although a
higher magnitude of hydrostatic pressure (10 mmHg) could
still induce significant levels of protein tyrosine phosphorylation (3.6-fold at 10 mmHg for 30 min), these values were
⬃50% lower than those observed at the 10-min observation
point (Fig. 1). These data demonstrate that osteoblasts are
sensitive to both the magnitude and duration of externally
applied sustained hydrostatic pressures.
A convergence point for many of the second-messenger
signaling pathways is the mitogen-activated protein kinases
(MAPKs). Activation of both 42- and 44-kDa MAPKs [extracellular signal-regulated kinase (ERK)1/2] occurs in endothelial cells in response to increased fluid shear stress (12, 46).
Similar to our observations of pressure-induced protein tyrosine phosphorylation, 10 min of sustained hydrostatic pressure enhanced ERK1/2 activity 3.2- and 8.3-fold under sustained pressures of 5 and 10 mmHg, respectively (Fig. 2). After
a 30-min exposure to 5 mmHg of pressure, ERK1/2 phosphorylation measured only 1.6-fold over control, whereas a pressure of 10 mmHg enhanced induced ERK1/2 activation by
fourfold (Fig. 2). Hence activation of downstream ERK1/2
MAPKs in osteoblasts also appears to be sensitive to both time
and pressure magnitude.
286 • APRIL 2004 •
www.ajpcell.org
PLASMA MEMBRANE CHOLESTEROL AND MECHANOTRANSDUCTION
Fig. 1. Enhanced protein tyrosine phosphorylation in osteoblasts in response
to hydrostatic pressure. Osteoblasts were exposed to either 5 or 10 mmHg of
hydrostatic pressure sustained for 10 or 30 min. Whole cell lysates were
resolved, transferred, and immunoblotted with phosphotyrosine antibody
4G10. A: Western blot shows that the profile of tyrosine-phosphorylated (pY)
proteins increased after 10- or 30-min exposure to hydrostatic pressure. B: total
pY detected after exposure to each pressure magnitude for 10 and 30 min was
densitometrically quantified. These results indicate that pressure-induced protein tyrosine phosphorylation occurred in a transient manner. Extent of protein
tyrosine phosphorylation was also dependent on the magnitude of the sustained
hydrostatic pressure applied to the cells. For these experiments, ␤-actin
showed no change in expression and thus served as a load control. Data are
reported as percent fold increases over cells maintained at ambient pressure
and are expressed as means ⫾ SD from three independent experiments; *P ⬍
0.05.
The early-response gene c-fos is believed to play an important role in bone cell physiology through regulation of transcription factors such as activator protein-1, which can engage
genes for collegen type I, osteocalcin, osteopontin, and others
(31, 45). Here we show that hydrostatic pressure enhanced
c-fos expression in both a time- and pressure-dependent manner. Expression of c-fos increased 2.2-fold in osteoblasts exposed to 5 mmHg of pressure for 10 min (Fig. 3). By 30 min,
expression of c-fos was 3.5-fold greater than that observed in
control cells (Fig. 3). Under sustained hydrostatic pressure of
10 mmHg, c-fos expression increased 3.8-fold after 10 min,
and by 30 min, expression was enhanced by nearly fivefold
compared with cells exposed to ambient pressure (Fig. 3).
CD and filipin alter hydrostatic pressure-induced mechanotransduction. Because cholesterol is essential for structural
maintenance of lipid rafts and caveolar-signaling complexes,
cholesterol-binding agents provide a useful pharmacological
tool for evaluating cholesterol-rich plasma membrane microdomain function. In this study, we pretreated osteoblasts with the
cholesterol-sequestering compound CD before exposing the
AJP-Cell Physiol • VOL
C833
cells to hydrostatic pressure. Incubation of osteoblasts with 1
mM CD for 30 min did not alter cell morphology or integrity
as measured by Trypan blue exclusion (data not shown). Figure
4 shows that pressure-induced protein tyrosine phosphorylation, ERK1/2 activation, and enhanced c-fos expression were
significantly reduced and approached nontreated, unstimulated
baseline values measured in control cells after depletion of
plasma membrane cholesterol. To further investigate the role
of cholesterol in spatiotemporal mechanosignaling, CD-pretreated osteoblasts were incubated with cholesterol-loaded CD
to replenish the plasma membrane cholesterol content. Figure
5 illustrates that cholesterol repletion restored pressure-induced
signaling.
In a complementary set of experiments, osteoblasts were
pretreated with the alternative raft-disrupting agent filipin.
Similar to CD, filipin pretreatment significantly reduced mechanosignaling in these cells (see Fig. 4). These data together
with the CD data suggest that intact cholesterol-rich plasma
membrane compartments (i.e., lipid rafts and caveolae) are
important components of the pressure-induced mechanotransduction pathway in osteoblasts.
Pressure-induced mechanosignaling events occur within
caveolin-containing membrane microdomains. To begin to
assess whether mechanotransduction events occur within
caveolar membranes, we used sucrose-gradient centrifugation
to separate detergent-resistant rafts, which include caveolae,
from osteoblast cell lysates. Consistent with past reports (23,
35), the caveolar marker protein caveolin was primarily de-
Fig. 2. Hydrostatic pressure-induced extracellular signal-regulated kinase
(ERK)1/2 mitogen-activated protein kinase (MAPK) activation within osteoblasts. Osteoblasts were exposed to sustained hydrostatic pressures of either 5
or 10 mmHg for 10 and 30 min. Whole cell lysates were resolved, transferred,
and immunoblotted with a phosphospecific antibody for ERK1/2. A: Western
blot shows that sustained hydrostatic pressures rapidly activated ERK1/2. B:
phosphorylated ERK1/2 (pERK1/2) detected after each time point of applied
pressure was densitometrically quantified. Similar to the results obtained for
protein tyrosine phosphorylation, pressure-induced activation of ERK1/2
MAPK occurred in a transient manner. Extent of activation was dependent on
the magnitude of the hydrostatic pressure applied to the cells. ERK2 served as
a load control for these experiments. Data are reported as percent fold increases
over cells maintained at ambient pressure and are expressed as means ⫾ SD
from three independent experiments; *P ⬍ 0.05.
286 • APRIL 2004 •
www.ajpcell.org
C834
PLASMA MEMBRANE CHOLESTEROL AND MECHANOTRANSDUCTION
To evaluate whether intact cholesterol-rich plasma membrane microdomains participate in the shear-induced mechanotransduction response, osteoblast membrane cholesterol was
extracted (see Lipid raft and caveolae manipulation with cholesterol-interacting agents). Similar to our observations for
sustained hydrostatic pressure-induced mechanotransduction,
lipid raft and caveolae ablation significantly reduced the capacity of the osteoblast to propagate a shear-induced mechanotransduction signaling cascade. Compared with nontreated
cells, shear-induced protein tyrosine phosphorylation (see Fig.
7), ERK1/2 activation (see Fig. 8), and enhanced c-fos expression (see Fig. 9) were stimulated to approximately half of the
value in cells treated with CD. Taken together, these data
suggest that osteoblasts use cholesterol-rich lipid rafts and/or
caveolae signaling compartments for activation of hydrostatic
pressure- and shear stress-responsive mechanotransduction
pathways.
DISCUSSION
Fig. 3. Hydrostatic pressure enhances expression of the early response gene
c-fos within osteoblasts. Osteoblasts were exposed to sustained hydrostatic
pressures of either 5 or 10 mmHg for 10 and 30 min. Whole cell lysates were
resolved, transferred, and immunoblotted with c-fos antibody. A: Western blot
shows the c-fos expression induced by exposure of osteoblasts to sustained
hydrostatic pressures for either 10 or 30 min. B: total expression detected at
each time point and for each magnitude of pressure was densitometrically
quantified. Expression of c-fos was sensitive to the magnitude of pressure
applied to osteoblasts. Each level of sustained hydrostatic pressure tested
enhanced c-fos expression as early as 10 min and as late as 30 min. ␤-Actin
showed no change in expression and thus served as a load control. Data are
reported as percent fold increases over cells maintained at ambient pressure
and are expressed as means ⫾ SD from three independent experiments; *P ⬍
0.05.
tected in a light-buoyant-density fraction collected from the
20% sucrose level of the continuous gradient (Fig. 6). As
observed in Fig. 1, pressure induced the tyrosine phosphorylation of a variety of cellular proteins (Fig. 6). Of these
phosphoproteins, bands corresponding to ⬃115, 85, 60–58,
and 24 kDa were enriched in the caveolin-enriched lipid raft
fraction of the sucrose gradient. In addition, proteins of apparent 180, 120, 70, and 40 kDa were sedimented in higherdensity nonraft cellular compartments. Thus a portion of the
proteins that were tyrosine-phosphorylated in response to increased hydrostatic pressure were detected within caveolinenriched plasma membrane rafts. This demonstrates that upstream signaling events can occur within these microdomains.
Shear stress activates mechanosignaling in osteoblasts
through cholesterol-rich signaling compartments. Because
fluid shear stress is also considered to be an important factor in
regulation of osteoblast function in vivo, we examined mechanotransduction responses in osteoblasts subjected to laminar
shear stress. Exposure to shear stress (10 dyn/cm2) for 10 min
induced a 5.2-fold increase in protein tyrosine phosphorylation
over sham-manipulated osteoblasts (Fig. 7). In addition,
ERK1/2 activity increased 2.5-fold (Fig. 8) and c-fos expression was enhanced by fourfold (Fig. 9) in response to laminar
shear stress. Thus both shear stress and sustained hydrostatic
pressure initiate similar mechanotransduction responses that
lead to enhanced gene expression in osteoblasts.
AJP-Cell Physiol • VOL
We first examined responsiveness of osteoblasts to changes
in the fluid-mechanical environment by using in vitro systems
designed to mimic biomechanical forces generated during bone
loading. Previous in vitro studies showed that hydrostatic
pressure stimulates production of prostaglandins and other
Fig. 4. Depletion of plasma membrane cholesterol attenuates pressure-induced
mechanosignaling in osteoblast cells. Osteoblasts were incubated with either
methyl-␤-cyclodextrin (CD) for 30 min or filipin for 5 min before exposure to
10 mmHg hydrostatic pressure sustained for 10 min. Whole cell lysates were
resolved, transferred, and immunoblotted with respective primary antibodies.
Lipid raft/caveolae disassembly attenuated tyrosine phosphorylation, ERK1/2
activation, and c-fos expression in response to hydrostatic pressure. ␤-Actin
and ERK2 served to demonstrate equivalent amount of protein loaded in each
lane. Western blot is representative of three independent experiments.
286 • APRIL 2004 •
www.ajpcell.org
PLASMA MEMBRANE CHOLESTEROL AND MECHANOTRANSDUCTION
Fig. 5. Reconstitution of cholesterol-enriched rafts/caveolae restores pressureinduced mechanotransduction. A: to restore cholesterol-rich raft domains,
osteoblast cell cultures were incubated with a cholesterol-loaded CD for 1 h at
37°C as described in METHODS. After raft reconstitution, osteoblasts were
subjected to 10-mmHg hydrostatic pressure sustained for 10 min. Whole cell
lysates were resolved, transferred, and immunoblotted with anti-phosphotyrosine antibody 4G10. Allowing lipid rafts/caveolae to reassemble restores the
pressure-induced mechanotransduction responses attenuated by acute CD
treatment. Western blot is representative of three independent experiments.
signaling molecules, activity of alkaline phosphatase, and expression of c-fos and the osteoblastic marker genes collagen
and osteopontin (9, 14, 32). Here we show that exposure of
osteoblasts to sustained hydrostatic pressure stimulated additional mechanosignaling pathways that included protein tyrosine phosphorylation and ERK1/2 MAPK activation (see
Figs. 1 and 2). Consistent with past reports (9), we also
C835
observed that osteoblasts subjected to increased magnitudes of
pressure responded by enhancing transcriptional expression of
c-fos, the early regulatory gene, in a time-dependent manner
(see Fig. 3).
Increasing evidence shows that fluid shear stress also plays
an important role in regulating osteoblast metabolism (13, 16,
24–27, 39). Osteoblasts derived from a variety of bone types
produce both NO and PGE2 in response to shear stress (13, 16,
39). Application of shear stress to mouse osteoblastic
MC3T3-E1 cells enhanced expression of COX-2 and c-fos (5,
22) through a mechanotransduction pathway that involved
IP3-mediated intracellular Ca2⫹ release (5). In the present
study, exposure to laminar shear stress (10 dyn/cm2 for 10 min)
activated additional signaling pathways in osteoblasts. Both
protein tyrosine phosphorylation (see Fig. 7) and activation of
ERK1/2 MAPK were enhanced by shear stress (see Fig. 8). In
agreement with previous studies (5, 22), we observed that fluid
shear stress induced a mechanotransduction cascade that led to
enhanced transcription of c-fos (see Fig. 9). Thus osteoblasts
are capable of sensing compressive and shear forces, which are
subsequently converted into biochemical signaling through
activation of second-messenger signaling molecules.
We observed that both hydrostatic pressure and laminar
shear stress activated comparable mechanotransduction pathways including a similar pattern of protein tyrosine phosphorylation, ERK1/2 activation, and enhanced c-fos expression.
However, osteoblasts showed greater mechanosensitivity to
hydrostatic pressure than shear stress. After 10 min, 10 mmHg
of hydrostatic pressure increased protein tyrosine phosphorylation by 7.3-fold, whereas 10 min of laminar shear stress
applied at 10 dyn/cm2 stimulated a 5.2-fold increase in phosphorylation. Likewise, hydrostatic pressure induced an eightfold activation of ERK1/2 after 10 min, whereas osteoblasts
subjected to shear stress for the same duration of time enhanced ERK1/2 activity by 2.5-fold. Although these observa-
Fig. 6. Pressure-induced tyrosine-phosphorylated proteins reside in caveolin-enriched membranes. Control and pressurechallenged osteoblasts were lysed by detergent extraction and
fractionated on a continuous sucrose gradient to yield caveolinenriched membranes as described in METHODS. A: a sample of
each fraction was loaded by equivalent volume onto a 10%
SDS-PAGE gel. Proteins were then separated and transferred to
nitrocelluose membranes. Western blot analyses showed caveolin-1 to be enriched in the low-buoyant-density fractions. The
same fractions probed with anti-phosphotyrosine (4G10)
showed several proteins tyrosine-phosphorylated in response to
hydrostatic pressure localized in the same membrane fraction as
caveolin. Experiments shown are representative of at least three
independent experiments. B: a protein concentration profile for
each fraction of the sucrose gradient collected from control and
pressure-treated osteoblast cultures is shown.
AJP-Cell Physiol • VOL
286 • APRIL 2004 •
www.ajpcell.org
C836
PLASMA MEMBRANE CHOLESTEROL AND MECHANOTRANSDUCTION
Fig. 7. Shear stress-induced protein tyrosine phosphorylation is dependent on
intact lipid raft/caveolae. Osteoblasts were pretreated with either CD (1 mM)
or vehicle at 37°C for 30 min, placed in a parallel device, and exposed to
laminar shear stress (10 dyn/cm2) for 10 min. Whole cell lysates were resolved,
transferred, and immunoblotted with phosphotyrosine antibody 4G10. A: shear
stress induced tyrosine phosphorylation of several cellular proteins. CD pretreatment substantially reduced this response. B: densitometric quantification
of total protein tyrosine phosphorylation shows a fivefold increase in osteoblasts exposed to shear stress. Lipid raft/caveolae ablation reduced this response by ⬃50%. ␤-Actin showed no change in expression and thus served as
a load control. Data are reported as percent fold increases over shams and are
expressed as means ⫾ SD from three independent experiments. *P ⬍ 0.05;
#
P ⬍ 0.05.
tions indicate that osteoblasts are more responsive to changes
in pressure, whether changes in shear magnitude would alter
the degree of the mechanotransduction response in these cells
remains to be tested. Because osteoblasts are subjected to both
pressure and shear forces in vivo, direct comparisons between
the effects that these forces exert on osteoblast function would
provide greater insight into the role that fluid mechanical forces
play in bone homeostasis.
Although our present observations in conjunction with previous studies investigating the mechanotransduction process in
osteoblasts and other cell types serve to elucidate the secondmessenger signaling molecules that propagate force transmission within the cell, the initial force-sensing element(s) or
mechanoreceptor(s) remains elusive. To date, the most intensively studied possible mechanosensing structures has been the
focal adhesion regions of the cell membrane (for overview, see
Ref. 6). Focal adhesions play a role in the adaptations of
cultured endothelial cells. These sites undergo remodeling
when shear stress is applied to the luminal surface of cultured
endothelial cells and realign in the direction of flow (8).
Cytoskeletal elements similarly rearrange and form stress fibers parallel to the direction of flow, which suggests an
association with focal adhesions (8). In addition, several cytoAJP-Cell Physiol • VOL
solic proteins localized to focal adhesions, particularly focal
adhesion kinase (34) and paxillin (47), undergo phosphorylation and initiate signal transduction events in response to flow.
In osteoblasts, shear-induced recruitment of ␣-actinin and
␤1-integrin to focal adhesion sites that correlated with enhanced expression of COX-2 and c-fos (22) suggested a link
between focal adhesion signaling events and gene expression.
Although focal adhesion complexes can clearly participate in
the cellular responses to flow, they do not give a complete
account of the mechanism(s) involved in the mechanotransduction process within either endothelial cells or osteoblasts.
It is becoming increasingly appreciated that the plasma
membrane contains organized regions or microdomains that
have a unique lipid and protein composition. These domains,
known as lipid rafts, are assembled primarily of glycerolphospholipids and enriched with cholesterol and sphingolipids (23).
Caveolae are considered specialized rafts due to the association
of caveolin proteins with raft lipids (40, 44). These structures
are particularly prominent in adipocytes and endothelial,
smooth muscle, and epithelial cells and have recently been
described in osteoblasts (15, 42). Both rafts and caveolae are
purported to participate in diverse cellular functions including
cholesterol transport (7, 41), endocytosis (36), and potocytosis
(1). In addition, several laboratories have reported that many
receptors and signaling molecules are concentrated within
these cholesterol-enriched membrane compartments (4, 38,
40). This latter observation has led to the hypothesis that these
Fig. 8. Shear stress-induced ERK1/2 activation is dampened after cholesterol
sequestration. Osteoblasts were pretreated with either CD (1 mM) or vehicle at
37°C for 30 min, placed in a parallel device, and exposed to laminar shear
stress (10 dyn/cm2) for 10 min. Whole cell lysates were resolved, transferred,
and immunoblotted with an antibody that detects activated ERK1/2. A: shearinduced ERK1/2 activation was attenuated by CD pretreatment. B: laminar
shear stress enhanced ERK1/2 activity by ⬃2.5-fold but only by 1.5-fold for
osteoblasts pretreated with CD. ERK2 served as a load control. Data are
reported as percent fold increases over shams and are expressed as means ⫾
SD from three independent experiments. *P ⬍ 0.05; #P ⬍ 0.05.
286 • APRIL 2004 •
www.ajpcell.org
PLASMA MEMBRANE CHOLESTEROL AND MECHANOTRANSDUCTION
Fig. 9. Shear stress-induced c-fos expression is reduced after CD pretreatment.
Osteoblasts were pretreated with either CD (1 mM) or vehicle at 37°C for 30
min, placed in a parallel device, and exposed to laminar shear stress (10
dyn/cm2) for 10 min. Whole cell lysates were resolved, transferred, and
immunoblotted with c-fos antibody. A: CD pretreatment significantly reduced
shear stress-induced c-fos expression. B: densitometric quantification of c-fos
expression shows a fourfold increase in c-fos expression in osteoblasts exposed
to shear stress. Lipid raft/caveolae ablation reduced this response by ⬃35%.
␤-Actin showed no change in expression and thus served as a load control.
Data are reported as percent fold increases over shams and are expressed as
means ⫾ SD from three independent experiments. *P ⬍ 0.05.
membrane regions serve as platforms for organizing and integrating signal transduction processes within the cell. Indeed,
this concept has now been demonstrated in a wide variety of
cell types including osteoblasts (43). Solomon et al. (43)
showed that platelet-derived growth-factor receptors, nonreceptor tyrosine kinases, tyrosine adapter proteins, G proteins,
and NO synthases were clustered as a signaling complex in
caveolin-enriched membrane fractions derived from human
and murine osteoblasts. More importantly, ligand-stimulated
activation of platelet-derived growth-factor receptor-induced
phosphorylation and dynamic regulation of MAPK effectors
located within these membrane fractions (43) suggest that
caveolin-enriched membrane domains are important sites for
signaling transduction in osteoblasts.
Because many of the signaling molecules that reside in lipid
rafts and caveolae have been implicated in the transduction of
fluid mechanical forces into chemical signals, we hypothesized
that cholesterol-rich plasma membrane microdomains, particularly caveolae, serve as logical sites for initiation of a mechanotransduction signaling cascade. We first tested this concept
in an in situ lung perfusion model (30). We reported that
increasing flow and pressure in rat lungs in situ stimulated
protein tyrosine phosphorylation at the luminal endothelial
plasma membrane specifically within caveolae (30). Enhancing
flow through the rat lung vasculature also activated endothelial
NO synthase located in caveolae (28) and a signaling cascade
that involved Ras-Raf-ERK1/2 (30). In addition, we and others
recently observed that flow preconditioning of cultured endothelial cells enhanced the density of cell surface caveolae and
AJP-Cell Physiol • VOL
C837
recruited them into mechanotransduction pathways (3, 29).
From these studies, we concluded that caveolae serve as
mechanotransduction sites at the endothelial cell surface. Because osteoblasts are subjected to similar biomechanical forces
(i.e., pressure and shear stress) and appear to respond to these
forces using similar signaling mechanisms as endothelial cells,
we tested whether cholesterol-rich plasma membrane microdomains play a role in the mechanotransduction process in this
cell type.
Cholesterol binding agents such as cyclodextrins and filipin
sequester cholesterol and cause disassembly of cholesterol-rich
rafts and caveolae (11, 21, 33). Thus these compounds provide
useful tools for study of lipid raft and caveolae functions.
Consistent with previous reports that demonstrated that CD
acts as a nonintrusive agent for reduction of plasma membrane
cholesterol (11), osteoblasts incubated with CD showed no
change in cell morphology, viability, or baseline levels of the
signaling parameters monitored in this study (Figs. 4 and 7–9,
as well as data not shown). Here we provide evidence that
disruption of cholesterol-rich plasma membrane compartments
significantly reduce both pressure and shear-induced mechanotransduction events in osteoblasts (Figs. 4 and 7–9). More
importantly, mechanotransduction responses could be restored
in osteoblasts where rafts were reconstituted (see Fig. 5). These
results are in support of our past studies, which showed that
abrogation of lipid rafts and caveolar architecture with cholesterol-sequestering compounds served to prevent mechanotransduction in rat lungs (30). Furthermore, Park et al. (21) reported
that cholesterol depletion with CD also prevented shear-induced activation of MAPK in cultured endothelial cells. Taken
together, these data support the concept that lipid rafts and
caveolae function as important mechanotransduction sites in
both endothelial cells and osteoblasts.
To begin to evaluate whether the observed mechanotransduction responses occurred directly in caveolar raft domains,
we isolated caveolin-enriched membranes from control and
pressure-treated osteoblasts and probed for mechanotransduction events. Here we show that several proteins phosphorylated
on tyrosine in response to enhanced pressure were localized in
caveolin-containing rafts (see Fig. 6) including an apparent
24-kDa protein. Although the identities of these proteins remain unknown, caveolin-1 is rapidly phosphorylated on Tyr14
in response to shear stress in endothelial cells (29). Thus it is
tempting to speculate that the apparent 24-kDa protein is
tyrosine-phosphorylated caveolin-1. The presence of mechanosensitive proteins, potentially including caveolin-1, demonstrates that at least some mechanotransducing elements reside
within these microdomains. Interestingly, we observed the
caveolin-1 signal trending toward the lipid raft fractions of the
sucrose gradient after pressure exposure. Whether increased
caveolin levels in the lipid raft fraction represent a recruitment
of caveolae into mechanotransduction pathways similar to that
observed for endothelial cells subjected to shear stress (3, 29)
remains to be examined.
We also observed a substantial number of phosphoproteins
that were not associated within caveolin-enriched rafts (see
Fig. 6). The ability of fluid mechanical forces to enhance
tyrosine phosphorylation of these proteins, however, was significantly dampened by treatment with both cyclodextrin and
filipin. The most obvious interpretation of this finding is that
these proteins are localized to noncaveolar lipid raft domains
286 • APRIL 2004 •
www.ajpcell.org
C838
PLASMA MEMBRANE CHOLESTEROL AND MECHANOTRANSDUCTION
that are mechanosensitive. Another possibility is that these
proteins are part of the downstream mechanotransduction process and reside in cellular compartments other than caveolincontaining membranes. In this senario, fluid mechanical forces
activate signaling molecules targeted to caveolin-enriched rafts
such as Src-like kinases, which then induce phosphorylation of
proteins that are not necessarily localized to raft domains.
Therefore, ablation of the raft mechanosignaling domain with
cholesterol-sequestering compounds may not allow for the
subsequent activation of these proteins. Yet another alternative
is the association of these phosphoproteins with other mechanosensitive sites within the cell such as focal contacts. Emerging evidence (10, 49, 50) suggests that caveolae and integrin
sites possess the ability to cross-talk. In support of this concept,
we find that ␤1-integrin-mediated mechanotransduction is regulated by plasma membrane cholesterol and caveolin-1 in
endothelial cells (our unpublished observations).
In summary, hydrostatic pressure and laminar shear stress
applied to osteoblast cell cultures stimulate a mechanotransduction cascade that is partially dependent on the integrity of
cholesterol-enriched plasma membrane compartments. These
conclusions are supported by the observation that two independent and well-characterized pharmacological agents for
disrupting cholesterol-rich domains, CD and filipin, blunt fluid
mechanical force-induced signal transduction in osteoblasts.
Furthermore, osteoblasts regain their mechanotransducing
properties after the restoration of plasma membrane rafts. The
localization of at least part of the observed mechanotransduction event to caveolin-enriched lipid rafts supports our previous
findings that these domains can serve as mechanotransduction
sites. Because a fair degree of mechanostimulated molecules
appeared to be unassociated with rafts and caveolae but sensitive to cholesterol-sequestering compounds, we do not rule
out that other cellular compartments or regions, perhaps influenced by cholesterol-enriched membrane domains, play an
additional role in the mechanotransduction process. Our data
support the general concept that proper organization of the
lipid milieu within plasma membranes allows the cell to
compartmentalize lipids and protein-signaling mediators that
provide a more efficient means to respond to fluid-mechanical
stimulation. Additional studies are necessary to determine
whether raft- and caveolae-mediated mechanotransduction
events require these structures as a whole or are regulated
through a single molecule or multiple effectors located in these
microdomains such as G proteins, endothelial NO synthase,
and calcium transport proteins. Such studies would be relevant
toward understanding the mechanisms by which osteoblasts
respond to mechanical forces and thereby influence bone tissue
adaptation to the physical environment.
ACKNOWLEDGMENTS
We thank Sheila Dela Cruz (Dept. of Biomedical Engineering, Rensselaer
Polytechnic Institute) for instruction and assistance in using the sustained
hydrostatic pressure chamber and Tin N. Oo for excellent technical support.
GRANTS
This work was supported by American Heart Association, New York
Affiliate Grant 0030300T (to V. Rizzo).
AJP-Cell Physiol • VOL
REFERENCES
1. Anderson RG, Kamen BA, Rothberg KG, and Lacey SW. Potocytosis:
sequestration and transport of small molecules by caveolae. Science 255:
410–411, 1992.
2. Berk BC, Corson MA, Peterson TE, and Tseng H. Protein kinases as
mediators of fluid shear stress stimulated signal transduction in endothelial
cells: a hypothesis for calcium-dependent and calcium-independent events
activated by flow. J Biomech 28: 1439–1450, 1995.
3. Boyd NL, Park H, Yi H, Boo YC, Sorescu GP, Sykes M, and Jo H.
Chronic shear induces caveolae formation and alters ERK and Akt
responses in endothelial cells. Am J Physiol Heart Circ Physiol 285:
H1113–H1122, 2003.
4. Brown D and London E. Functions of lipid rafts in biological membranes. Annu Rev Cell Dev Biol 14: 111–136, 1998.
5. Chen NX, Chen NX, Ryder KD, Pavalko FM, Turner CH, Burr DB,
Qiu J, and Duncan RL. Ca2⫹ regulates fluid shear-induced cytoskeletal
reorganization and gene expression in osteoblasts. Am J Physiol Cell
Physiol 278: C989–C997, 2000.
6. Davies PF. Flow-mediated endothelial mechanotransduction. Physiol Rev
75: 519–560, 1995.
7. Fielding PE and Fielding CJ. Plasma membrane caveolae mediate the
efflux of cellular free cholesterol. Biochemistry 34: 14288–14292, 1995.
8. Girard PR and Nerem RM. Shear stress modulates endothelial cell
morphology and F-actin organization through the regulation of focal
adhesion-associated proteins. J Cell Physiol 163: 179–193, 1995.
9. Glantschnig H, Varga F, and Rumpler M. Prostacyclin (PGI2): a
potential mediator of c-fos expression induced by hydrostatic pressure in
osteoblastic cells. Eur J Clin Invest 26: 533–548, 1996.
10. Gopalakrishna P, Chaubey SK, Manogaran PS, and Pande G. Modulation of alpha5beta1 integrin functions by the phospholipid and cholesterol contents of cell membranes. J Cell Biochem 77: 517–528, 2000.
11. Ilangumaran S and Hoessli D. Effects of cholesterol depletion by
cyclodextrin on the sphingolipid micromains of the plasma membrane.
Biochem J 335: 433–440, 1998.
12. Ishida T, Peterson TE, Kovach NL, and Berk BC. MAP kinase
activation by flow in endothelial cells. Circ Res 79: 310–316, 1996.
13. Johnson DL, McAllister TN, and Frangos JA. Fluid flow stimulates
rapid and continuous release of nitric oxide in osteoblasts. Am J Physiol
Endocrinol Metab 271: E205–E208, 1996.
14. Klein-Nulend J, Roelofsen J, Semeins CM, Bronckers AL, and Burger
EH. Mechanical stimulation of osteopontin mR15 expression and synthesis in bone cell cultures. J Cell Physiol 170: 174–181, 1997.
15. Lofthouse RA, Davis JR, Frondoza CG, Jinnah RH, Hungerford DS,
and Hare JM. Identification of caveolae and detection of caveolin in
normal human osteoblasts. J Bone Joint Surg Br 83: 124–129, 2001.
16. McAllister TN and Frangos JA. Steady and transient fluid shear stress
stimulate NO release in osteoblasts through distinct biochemical pathways. J Bone Miner Res 14: 930–936, 1999.
17. Morey-Holton E, Whalen RT, Arnaud SB, and Van Der Meulen MC.
The skeleton and its adaptation to gravity. In Handbook of Physiology.
Environmental Physiology. Bethesda, MD: Am. Physiol. Soc., 1996, sect.
4, vol. I, chapt. 31, p. 691–719.
18. Muller J, Chilian W, and Davis M. Integrin signaling transduced
shear-stress dependent vasodilation of coronary arterioles. Circ Res 80:
320–326, 1997.
19. Nagatomi J, Arulanandam BP, Metzger DW, Meunier A, and Bizios
R. Frequency- and duration-dependent effects of cyclic pressure on select
bone cell functions. Tissue Eng 7: 717–728, 2001.
20. Oliferenko S, Paiha K, Harder T, Gerke V, Schwarzler C, Schwarz H,
Beug H, Gunthert U, and Huber LA. Analysis of CD44-containing lipid
rafts: recruitment of annexin II and stabilization by the actin cytoskeleton.
J Cell Biol 146: 843–854, 1999.
21. Park H, Go YM, St John PL, Maland MC, Lisanti MP, Abrahamson
DR, and Jo H. Plasma membrane cholesterol is a key molecule in shear
stress-dependent activation of extracellular signal-regulated kinase. J Biol
Chem 273: 32304–32311, 1998.
22. Pavalko FM, Chen NX, Turner CH, Burr DB, Atkinson S, Hsieh YF,
Qiu J, and Duncan RL. Fluid shear-induced mechanical signaling in
MC3T3-E1 osteoblasts requires cytoskeleton-integrin interactions. Am J
Physiol Cell Physiol 275: C1591–C1601, 1998.
23. Pike LJ, Han X, Chung KN, and Gross RW. Lipid rafts are enriched in
arachidonic acid and plasmenylethanolamine and their composition is
286 • APRIL 2004 •
www.ajpcell.org
PLASMA MEMBRANE CHOLESTEROL AND MECHANOTRANSDUCTION
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
independent of caveolin-1 expression: a quantitative electrospray ionization/mass spectrometric analysis. Biochemistry 41: 2075–2088, 2002.
Reich KM and Frangos JA. Effect of flow on prostaglandin E2 and
inositol trisphosphate levels in osteoblasts. Am J Physiol Cell Physiol 261:
C428–C432, 1991.
Reich KM and Frangos JA. Protein kinase C mediates flow-induced
prostaglandin E2 production in osteoblasts. Calcif Tissue Int 52: 62–66,
1993.
Reich KM, Gay CV, and Frangos JA. Fluid shear stress as a mediator of
osteoblast cyclic adenosine monophosphate production. J Cell Physiol
143: 100–104, 1990.
Reich KM, McAllister TN, Gudi S, and Frangos JA. Activation of G
proteins mediates flow-induced prostaglandin E2 production in osteoblasts. Endocrinology 138: 1014–1018, 1997.
Rizzo V, McIntosh DP, Oh P, and Schnitzer JE. Flow activates eNOS
in caveolae at the luminal cell surface of endothelium in situ with rapid
caveolin dissociation and calmodulin association. J Biol Chem 273:
34724–34729, 1998.
Rizzo V, Morton C, DePaola N, Schnitzer JE, and Davies PF. Recruitment of endothelial caveolae into mechanotransduction pathways by flow
conditioning in vitro. Am J Physiol Heart Circ Physiol 285: H1720–
H1729, 2003.
Rizzo V, Sung A, Oh P, and Schnitzer JE. Rapid mechanotransduction
in situ at the luminal cell surface of the microvascular endothelium and its
caveolae. J Biol Chem 273: 26323–26329, 1998.
Rodan GA and Noda M. Gene expression in osteoblastic cells. Crit Rev
Eukaryot Gene Expr 1: 85–98, 1991.
Roelofsen J, Klein-Nulend J, and Burger EH. Mechanical stimulation
by intermittent hydrostatic compression promotes bone-specific gene expression in vitro. J Biomech 28: 1493–1503, 1995.
Rothberg KG, Rothberg KG, Ying YS, Kamen BA, and Anderson
RG. Cholesterol controls the clustering of the glycophospholipid-anchored membrane receptor for 5-methltetrahydrofolate. J Cell Biol 111:
2931–2938, 1990.
Schaller MD, Borgman CA, Cobb BS, Vines RR, Reynolds AB, and
Parsons JT. pp125FAK, a structurally distinctive protein-tyrosine kinase
associated with focal adhesions. Proc Natl Acad Sci USA 89: 5192–5196,
1992.
Schnitzer JE, McIntosh DP JE, Dvorak AM, Liu J, and Oh P.
Separation of caveolae from associated microdomains of GPI-anchored
proteins. Science 269: 1435–1439, 1995.
Schnitzer JE, Oh P, and McIntosh DP. Role of GTP hydrolysis in
fission of caveolae directly from plasma membranes. Science 274: 239–
242, 1996.
AJP-Cell Physiol • VOL
C839
37. Schnitzer JE, Oh P, Pinney E, and Allard J. Filipin-sensitive caveolaemediated transport in endothelium: reduced transcytosis, scavenger endocytosis, and capillary permeability of select macromolecules. J Cell Biol
127: 1217–1232, 1994.
38. Simons K and Toomre D. Lipid rafts and signal transduction. Nat Rev
Mol Cell Biol 1: 31–39, 2000.
39. Smalt R, Mitchell FT, Howard RL, and Chambers TJ. Induction of NO
and prostaglandin E2 in osteoblasts by wall-shear stress but not mechanical
strain. Am J Physiol Endocrinol Metab 273: E751–E758, 1997.
40. Smart EJ, Graf GA, McNiven MA, Sessa WC, Engelman JA, Scherer
PE, Okamoto T, and Lisanti MP. Caveolins, liquid-ordered domains and
signal transduction. Mol Cell Biol 19: 7289–7304, 1999.
41. Smart EJ, Ying Y, Donzell WC, and Anderson RG. A role for caveolin
in transport of cholesterol from endoplasmic reticulum to plasma membrane. J Biol Chem 271: 29427–29435, 1996.
42. Solomon KR, Adolphson LD, Wank DA, McHugh KP, and Hauschka
PV. Caveolae in human and murine osteoblasts. J Bone Miner Res 15:
2391–2401, 2000.
43. Solomon KR, Danciu TE, Adolphson LD, Hecht LE, and Hauschka
PV. Caveolin-enriched membrane signaling complexes in human and
murine osteoblasts. J Bone Miner Res 15: 2380–2390, 2000.
44. Sowa G, Pypaert M, and Sessa W. Distinction between signaling
mechanisms in lipid rafs vs. caveolae. Proc Natl Acad Sci USA 98:
14072–14077, 2001.
45. Stein GS, van Wijnen AJ, Stein JL, Lian JB, McNeil S, and Pockwinse
SM. Transcriptional control of osteoblast growth and differentiation.
Physiol Rev 76: 593–629, 1996.
46. Tseng H, Peterson TE, and Berk BC. Fluid shear-stress stimulated
mitogen-activted protein kinase in endothelial cells. Circ Res 77: 869–878,
1995.
47. Turner CE, Glenny JR, and Burridge K. Paxillin: a new vinculinbinding protein present in focal adhesions. J Cell Biol 111: 1059–1068,
1990.
48. Tzima E, del Pozo MA, Shattil SJ, Chien S, and Schwartz MA.
Activation of integrins in endothelial cells by fluid shear stress mediates
Rho-dependent cytoskeletal alignment. EMBO J 20: 4639–4647, 2001.
49. Wary KK, Mariotti A, Zurzolo C, and Giancotti FG. A requirement for
caveolin-1 and associated kinase fyn in integrin signaling and anchoragedependent cell growth. Cell 94: 625–634, 1998.
50. Wei Y, Yang X, Liu Q, Wilkins JA, and Chapman HA. A role for
caveolin and urokinase receptor in integrin-mediated adhesion and signaling. J Cell Biol 144: 1285–1294, 1999.
51. Whedon G and Heaney R. Effects of physical inactivity, paralysis and
weightlessness on bone growth. In: Bone—Bone Growth B, edited by Hall
K. Melbourne, FL: CRC, 1993, p. 57–77.
286 • APRIL 2004 •
www.ajpcell.org