A quantitative study on morphological responses of osteoblastic cells to fluid shear stress.

Abstract

Fluid shear stress (FSS) is widely explored regarding its influence on osteoblasts. In vitro studies have shown that the cytoskeleton is very important in cellular responses to FSS. However, morphological changes, which would reflect the cytoskeleton changes as well as other cellular responses, were rarely quantitatively studied in the past years. Therefore, FSS-induced morphological changes in osteoblasts were quantified in this study. Real-time rapid morphological responses were observed by exposing osteoblasts to FSS with magnitude of 1.2, 1.6, and 1.9 Pa for 1 h. Afterward, osteoblast actin cytoskeleton was labeled with rhodamine phalloidin and observed using fluorescence microscopy. The results showed that 1.6 and 1.9 Pa FFS resulted in significant cellular elongation and reorientation along the direction of fluid flow. Besides, along with the enhancement of FSS magnitude, cytoskeleton aggregated more remarkably. Furthermore, extracellular Ca(2+)-depleted fluid flow was also used to stimulate osteoblasts for 1 h with magnitude of 1.6 and 1.9 Pa. No morphological change was observed after removing extracellular calcium. Our study suggested that the level of FSS from 1.2 to 1.9 Pa is capable of influencing cellular morphology, and extracellular calcium might play a role in osteoblasts' response to FSS stimulation.

Full text

Page 1

A quantitative study on morphological responses of osteoblastic cells to fluid
shear stress
Xiaoli Liu, Xu Zhang, and Imshik Lee*
State Key Laboratory of Bioactive Materials, School of Physics, Nankai University, Tianjin 300071, China
*Correspondence address. Tel: þ86-22-23494411. E-mail: ilee@nankai.edu.cn
Fluid shear stress (FSS) is widely explored regarding its
influence on osteoblasts. In vitro studies have shown that
the cytoskeleton is very important in cellular responses to
FSS. However, morphological changes, which would
reflect the cytoskeleton changes as well as other cellular
responses, were rarely quantitatively studied in the past
years. Therefore, FSS-induced morphological changes in
osteoblasts were quantified in this study. Real-time rapid
morphological responses were observed by exposing osteo-
blasts to FSS with magnitude of 1.2, 1.6, and 1.9 Pa for
1 h. Afterward, osteoblast actin cytoskeleton was labeled
with rhodamine phalloidin and observed using fluor-
escence microscopy. The results showed that 1.6 and
1.9 Pa FFS resulted in significant cellular elongation and
reorientation along the direction of fluid flow. Besides,
along with the enhancement of FSS magnitude, cytoskele-
ton aggregated more remarkably. Furthermore, extra-
cellular Ca21-depleted fluid flow was also used to
stimulate osteoblasts for 1 h with magnitude of 1.6 and
1.9 Pa. No morphological change was observed after
removing extracellular calcium. Our study suggested that
the level of FSS from 1.2 to 1.9 Pa is capable of influen-
cing cellular morphology, and extracellular calcium might
play a role in osteoblasts’ response to FSS stimulation.
Keywords
cytoskeleton; image analysis
fluid shear stress; osteoblast; morphology;
Received: September 9, 2009Accepted: December 14, 2009
Introduction
It is well known that bone cells are exposed to a complex
environment which exerts several types of mechanical stress
such as stretching stress [1,2], fluid shear stress (FSS) [3],
and hydrostatic pressure [4]. All these types of stress have
influence on bone metabolism, especially on bone formation
and remodeling [5,6].Since Reich et al. [7]revealed the bio-
logical response of bone cells under FSS in 1990, how FSS
takes effects on osteoblasts has been widely explored and
significant results have been obtained, including the upload-
ing of Ca2þsignals [8,9]; increase in cAMP, prostaglandins
[10], and NO [11]; increase in gene expressions of COX-2
and c-fos [12]; overexpression of cross-linking proteins such
as vimentin, a-actinin, and filamin [13]; enhancement of
osteoblasts proliferation and differentiation [14,15]; and
enhancement of integrin-regulated gene expression which is
related to bone formation [16]. In particular, cytoskeleton
was considered to be very important in cellular responses to
FSS stimulation. Tensegrity theory published in 1997 indi-
cated that cytoskeleton was the key of cells to sense mechan-
ical stimulations [17]. Likewise, it is well accepted that
the cytoskeleton plays important roles in keeping cell
morphology and transferring molecular signals [18,19].
Therefore, investigation in changes of cytoskeleton is useful
for clarifying how osteoblasts respond to FSS stimulation.
Since cell morphology was regulated by cytoskeleton,
thus certain morphological changes may indirectly reflect
the responses of upstream signals at molecular level,
e.g. calcium signals. Accordingly, biological responses,
especially changes of cytoskeleton, could be quantified by
morphological measurement. Although some studies have
reported morphological changes in osteoblasts under stimu-
lation of FSS [3,12,13,20], little information on quantitative
study was available in the past years. Horikawa et al. [21]
reported the relationship between duration of FSS and mor-
phological changes in osteoblasts in 2000. However, no
study to date has included the analysis of the relationship
between magnitude of FSS and morphological changes.
Since quantitative investigation of morphological changes
would be helpful in clarifying the influences of FSS on
osteoblasts, the aim of this study is to explore the effects of
FSS on cell morphology quantitatively.
Theoretical modeling predicted that shear stress in the
range 0.8–3.0 Pa most approaches to FSS in vivo [22].
Additionally, FSS from 1.0 to 2.0 Pa was usually applied to
experiments in vitro [14,16,20]. Thus, FSS of 1.2, 1.6, and
1.9 Pa were chosen to examine the morphological changes
in osteoblasts in this study. Rapid morphological responses
were observed during the stimulations. Besides, FSS-induced
Acta Biochim Biophys Sin (2010): 195–201 | ª The Author 2010. Published by ABBS Editorial Office in association with Oxford University Press on behalf of the
Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences. DOI: 10.1093/abbs/gmq004.
Acta Biochim Biophys Sin (2010) | Volume 42 | Issue 3 | Page 195

Page 2

actin-cytoskeleton changes were quantified by using rhoda-
mine-phalloidin imaging techniques. Osteoblasts elongated
and reoriented along the direction of the fluid flow. Also,
with the increase in FSS magnitude, actin-cytoskeleton aggre-
gated more apparently. Furthermore, morphological change
in osteoblasts was not observed after removing extracellular
calcium, implying the significance of extracellular calcium in
osteoblasts’ response to FSS stimulation.
Materials and Methods
Cell culture
Thecalvariae
Academy of Medical Sciences, Tianjin, China) was used
for isolating osteoblast-like cells. Procedures of cell iso-
lation and cell culture were described in our earlier report
[2]. Cells were grown in RPMI 1640 medium (10.4 g/L
RPMI 1640, 2 g/L NaHCO3, 5.96 g/L HEPES, 0.05 g/L
penicillin G, 0.1 g/L streptomycin sulfate, pH 7.2), and cul-
tured at 378C in an MCO175 incubator (SANYO Electric
Biomedical Co., Ltd, Osaka, Japan) containing 5% CO2. In
the present experiments, cells were subcultured on glass
slides (32 mm ? 24 mm) at 2.5–3 ? 104cells/cm2. Fluid
flow was applied 48 h after subculture so that cells were
80–90% confluent at the time of experimentation.
of Wistar rat(3–5-day-old,Chinese
Flow chamber and experiment
As shown in Fig. 1, a parallel-plate flow chamber was used
to introduce fluid flow over the cells. The fluid-flow setup
consisted of a parallel-plate flow chamber and a recirculating
flow circuit. This circuit included a variable-speed peristaltic
pump and a reservoir with RPMI 1640 medium maintained
at 378C with 5% CO2. To investigate the role of extracellular
calcium in response to shear stress, 1.5 mM ethylene glycol
tetra-acetic acid (EGTA) was supplemented into RPMI 1640
medium to produce Ca2þ-depleted medium. This system pro-
duces laminar flow over a cell monolayer. A flow rate was
chosen to yield a t-value of 1.2, 1.6 or 1.9 Pa by using t ¼
6Qm/bh2, where Q is the flow rate, m the medium viscosity,
b the channel width, and h the channel height. Control cells
were kept under static conditions with the same culture
medium at 378C with 5% CO2.
In real-time observation, this system was equipped with an
Axio Observer D1 inverted microscope (Zeiss, Berlin,
Germany) and pictures were recorded every 5 min for 1 h
using DU-897D-CS0-BV CCD system (Andor, London, UK).
Fluorescence observation and image analysis
Rhodamine phalloidin was used to label cytoskeletal
F-actin according the method described [23]. Cells were
washed in PBS, fixed with 4% paraformaldehyde (diluted
in PBS) for 30 min and then rinsed twice in PBS. PBS
containing 0.5% Triton X-100 (v/v) was used to permeate
the membrane for 30 min. Then cells were rinsed twice in
PBS. Rhodamine-phalloidin mixed solution (2.4%) was
used to label osteoblasts actin cytoskeleton (200 ml for
each glass slide). After labeling (10 min, 258C), the
coverslips were washed twice in PBS and observed
with fluorescence microscope using a 40? oil lens
with the MetaMorph 7.1 software (Universal Imaging,
Figure 1 A schematic illustration of the parallel-plate flow chamber used to induce FSS
an inlet and a outlet, (B) a thin rubber silastic gasket with uniform thickness, (C) a cover slide on which osteoblasts were cultured, and (D) a quartz slide of
the same size as (A). (E) An overhead view of the assembled fluid cell. (F) The assembled system. The sizes of all parts were shown in this figure.
The parallel plates are comprised of (A) a quartz slide with
A quantitative study on morphological responses of osteoblastic cells to FSS
Acta Biochim Biophys Sin (2010) | Volume 42 | Issue 3 | Page 196

Page 3

Downingtown, USA) capable of cell-by-cell analysis. The
exposure time was 100 ms. Measurement was recorded for
20–25 cells selected randomly within one slide of a
culture and repeated for several cultures.
We used two shape factors, r and Circularity, to quantify
cell shape. r is the ratio of cell length (defined along the
major axis) to cell width (defined along the minor axis).
Circularity is defined as 4pA/P2for quantifying the degree
of cell’s circularity, where A is the area and P the per-
imeter. The other two parameters, Angle and AngleSD,
were used to represent cell orientation. Parameter Angle is
the absolute value of the angular orientation when com-
pared with the direction of the FSS (a smaller number in
Angle indicates that cell oriented near the direction of the
shear stress). AngleSD is the standard deviation of Angle
in one group, which was used to reflect the orientational
distribution of the cells. Additionally, the fluorescence
intensity (FI), which is proportional to the concentration of
labeled F-actin for not too large absorption of the emitted
fluorescence light, was used to quantify the content of
labeled F-actins.
Statistical analysis
Statistical differences in the cell morphological parameters
were determined by the Student’s t-test and one-way
ANOVA. Statistical significance was established at the P ,
0.05 and all experimental analyses were performed blind.
Results
Real-time observation of cell morphology
In an applied stimulation of FSS, some cells (Fig. 2)
elongated along the long axis of cell in the first 15 min and
no dose-dependent characteristics were seen. Although not
very distinct in phase contrast pictures, the trends of these
cells to bulge and become more spindle-shaped could be
seen, particularly in the highlighted regions (red arrows).
FSS exerted in this experiment was 1.2 Pa [Fig. 2(A–C)],
1.6 Pa [Fig. 2(D–F)], and 1.9 Pa [Fig. 2(G–I)], respect-
ively. Photos were recorded chronologically at a 5-min
interval. After 15 min, no apparent morphological change
was observed (data not shown). Since real-time morpho-
logical changes were observed, quantitative analysis was
then conducted by using fluorescence microscopy.
Fluorescence observation and image analysis
F-actins of control cells were of clearly visible filamentous
configuration [Fig. 3(A)]. After FSS stimulation for 1 h,
the F-actins tended to arrange parallel to the long axis of
cell [Fig. 3(B)], which was consistent with the result of
real-time observation.
Figure 2 Real-time observation of cell shape under the stimulation of FSS with different magnitudes
5, 10 and 15 min, respectively. (D–F) show the cells under 1.6 Pa FSS for 5, 10, and 15 min, respectively. (G–I) show the cells under 1.9 Pa FSS for 5,
10, and 15 min, respectively. Cells tended to bulge and become more spindle shape (arrow) in the first 15 min of stimulation. Scale bar ¼ 50 mm.
(A–C) show the cells under 1.2 Pa FSS for
A quantitative study on morphological responses of osteoblastic cells to FSS
Acta Biochim Biophys Sin (2010) | Volume 42 | Issue 3 | Page 197

Page 4

Meanwhile, morphological information of the cells (r,
Circularity, Angle, AngleSD) was measured (Table 1). In
thecontrol group,
r
was
Circularity was 0.312+0.027 (a circularity value of 1.0
indicating a circular shape, and a smaller value indicating a
more spindle-shaped cell). Correspondingly, cell Angle
was 40.222+16.3458. And less SD indicates more
uniform cellular orientation, thus no evidence of preferred
cell orientation was found, with the AngleSD being
51.6868.
Statistical analysis of the image data revealed significant
increase in r of 1.6 Pa group (P ¼ 0.042) and 1.9 Pa group
(P ¼ 0.011) (Fig. 4 and Table 1). Also, 1.9 Pa stimulation
resulted in apparent decrease (P ¼ 0.03) in cell Circularity
when compared with the control group (Table 1). Both
enhancement of r and reduction in Circularity suggested
that cell elongated and became more narrowly spindle-
shaped after stimulation, especially under the stimulation of
1.9 Pa FSS. The same trend (i.e. increase in r and decrease
in Circularity), though did not approach statistical signifi-
cance (P ¼ 0.182 for r, P ¼ 0.694 for Circularity), could
be seen in the 1.2-Pa group. On the other hand, significant
reduction in Angle values after stimulations of 1.6 Pa (P ¼
0.0005) and 1.9 Pa (P ¼ 0.001) are shown in Table 1 and
Fig. 5 (versus control group). The Angle values of 1.6 Pa
(0.906+4.281) and 1.9 Pa (3.474+5.573) were much
smaller than that of control (40.222+16.345), indicating
that cells preferentially aligned along the direction of the
2.619+0.249, whereas
shear stress. In addition, distinct decrease in AngleSD
(control: 51.686; 1.6 Pa: 13.538; 1.9 Pa: 17.623) suggested
that cells reoriented from random distribution to uniform
orientation (Figs. 3 and 5).
Table 1 Morphological measurements of rhodamine-phalloidin stained osteoblasts after 1-h stimulation of FSS with different magnitudes
Group
r
Circularity Angle (8) AngleSDa(8)
Control
1.2 Pa
1.6 Pa
1.9 Pa
2.619+0.249
3.281+0.419
3.426+0.271*
3.960+0.414*
0.312+0.027
0.295+0.031
0.273+0.013
0.229+0.020*
40.222+16.345
25.778+12.233
0.906+4.281**
3.474+5.573**
51.686
38.684
13.538**
17.623**
Data are presented as the mean+SE. *P , 0.05, **P , 0.01 versus control group.aAngleSD is the standard deviation of Angle in one group, see the
‘Materials and Methods’ section for detail. r is the ratio of cell length (defined along the major axis) to cell width (defined along the minor axis).
Figure 3 Fluorescence micrographs of cells labeled with rhodamine-
phalloidin
(A) control group, cells arranged randomly. (B) Cells after
1-h stimulation of 1.6 Pa FSS. The F-actins tended to arrange parallel to the
long axes of cells and reorient into uniform direction. Scale bar ¼ 50 mm.
Figure 4 The length–width ratio r given as a function of FSS
intensity
After 1-h stimulation, both 1.6 and 1.9 Pa FSS resulted in
significant increase in r, indicating the elongation of the osteoblastic cells.
Data are the mean+SE from three individual experiments. *P, 0.05
compared with control group.
Figure 5 Comparison in changes of orientation parameters (Angle and
AngleSD) between normal group and calcium-depleted group
1-h stimulation, neither Angle nor AngleSD reduced distinctly in
extracellular calcium-depleted group as in the normal group, which
suggested that osteoblasts did not reorient under FSS without extracellular
calcium. Data are the mean+SD from three individual experiments.
After
A quantitative study on morphological responses of osteoblastic cells to FSS
Acta Biochim Biophys Sin (2010) | Volume 42 | Issue 3 | Page 198

Page 5

The FI of cytoskeleton was also measured (Fig. 6).
Consistent with the influences on cell morphology, both
1.6 and 1.9 Pa stimulation resulted in significant increases
(P ¼ 0.017 in 1.6 Pa group, P ¼ 0.003 in 1.9 Pa group) in
FI compared with controls, which was indicative of aggre-
gation of F-actin filaments. Overall, the data showed that
the morphological change is dependent on the magnitude
of FSS (Table 1, Figs. 4 and 6).
Extracellular Ca21depletion experiment
Since intensities of 1.6 and 1.9 Pa seemed to have robust
effect on cell morphology as shown above, these two stimu-
lations were applied in calcium-depleted experiment. As a
result, no significant differences were found in cell shape
factors in both 1.6 and 1.9 Pa groups comparing with the
control group (P . 0.05, Table 2), nor were any changes in
FI value observed (data not shown). Similarly, orientation
parameters (Angle and AngleSD) showed no apparent
change when compared with the control group (P . 0.05,
Table 2). To more thoroughly illustrate the role of calcium,
comparison in orientation parameters between normal
mediumand calcium-depleted
(Fig. 5). The figure showed that both Angle and AngleSD
did not reduce distinctly in Ca2þ-depleted group as in the
mediumwasprepared
control group, i.e. cells stimulated by 1.6 and 1.9 Pa FSS
still distributed randomly without a preferential orientation.
Discussion
Cytoskeleton is well known to be important as the mechan-
ical sensor of osteoblasts to FSS [3]. On the other hand, it
is widely accepted that cytoskeletal elements, including
microfilaments, microtubules, and intermediate filaments,
are important in maintaining the normal shape of cell.
There are two types of actins: F-actins in filamentous form
and G-actins in free form, both of which are the basic com-
ponent of microfilaments. Osteoblasts presented a trend of
extension under the influence of FSS [12,13,21,24].
However, little work has been done to make further analy-
sis on the change of cytoskeleton or morphology. Recently,
it was shown that the cause of cytoskeleton changes might
be the increase of anchor points [25], the release of intra-
cellular Ca2þinduced by inositol 1,4,5-triphosphate (IP3)
[20] or the increase in linking proteins [13,26]. Also, it was
reported that the changes in cytoskeleton might exert a
function of transferring mechanical signaling into osteo-
blasts [12]. Though cytoskeleton of osteoblasts has been
emphasized and studied, limited quantitative morphological
research work, which would reflect cytoskeleton changes
as well as other biological responses of cells, was pub-
lished in the past years. Horikawa et al. [21] measured
morphological changes in osteoblasts by exerting 10 dyn/
cm2FSS on three groups of cells for 1, 6, and 12 h,
respectively. As a result, cell elongation and [Ca2þ]i
enhancement were observed in 1 h group. Besides, long-
term FSS stimulation was concluded to have destructive
effects on osteoblasts. In view of the report of Horikawa,
1-h stimulation was chosen in our experiment. Moreover,
we explored morphological changes in osteoblasts by exert-
ing FSS with different magnitudes. Also, we investigated
the role of extracelluar Ca2þon morphological changes in
osteoblasts. Since we could only observe filamentous actins
by phalloidin staining, other cytoskeletal elements were not
studied in the present work.
Real-time observation suggested that the osteoblasts
tended to elongate in the first 15 min of stimulations.
Though not very obvious, the tendency of elongation
Figure 6 The FI given as a function of FSS intensity
stimulation, both 1.6 and 1.9 Pa FSS caused significant increase in FI,
which implied the aggregation of the actin cytoskeleton. Data are the
mean+SE from three individual experiments. *P, 0.05 and **P, 0.01
compared with control group.
After 1-h
Table 2 Morphological measurements of rhodamine-phalloidin stained osteoblasts after 1-h stimulation of FSS with different magnitudes under
extracellular calcium-depleted condition
Condition
r
Circularity Angle (8)AngleSDa(8)
Control
1.6 Pa
1.9 Pa
2.619+0.249
2.4193+0.209
2.4637+0.183
0.312+0.027
0.353+0.034
0.345+0.021
40.222+16.345
30.757+10.821
53.517+11.390
51.686
41.913
46.961
Data are presented as the mean+SE.aAngleSD is the standard deviation of Angle in one group, see the ‘Materials and Methods’ section for detail.
r is the ratio of cell length (defined along the major axis) to cell width (defined along the minor axis).
A quantitative study on morphological responses of osteoblastic cells to FSS
Acta Biochim Biophys Sin (2010) | Volume 42 | Issue 3 | Page 199