ARTICLE

3D Culture of Osteoblast-Like Cells by

Unidirectional or Oscillatory Flow for Bone
Tissue Engineering
Dajiang Du,1 Katsuko S. Furukawa,1,2,3 Takashi Ushida3,4
1
Department of Mechanical Engineering, Laboratory of Biomedical Engineering,
Graduate School of Engineering, University of Tokyo, 2nd Building, 7-3-1 Hongo,
Bunkyo-ku, Tokyo 113-8656, Japan; telephone: 81-3-5841-6331; fax: 81-3-5841-6447;
e-mail: furukawa@mech.t.u-tokyo.ac.jp
2
Department of Bioengineering, Graduate School of Engineering, University of Tokyo,
2nd Building, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan
3
Center for NanoBio Integration, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku,
Tokyo 113-8656, Japan
4
Division of Biomedical Materials and Systems, Center for Disease Biology and Integrative
Medicine, School of Medicine, University of Tokyo, Tokyo, Japan
Received 31 July 2008; revision received 10 November 2008; accepted 17 November 2008
Published online 2 December 2008 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/bit.22214

ABSTRACT: A medium perfusion system is expected to be
beneficial for three-dimensional (3D) culture of engineered
bone, not only by chemotransport enhancement but also by
mechanical stimulation. In this study, perfusion systems
with either unidirectional or oscillatory medium flow were
developed, and the effects of the different flow profiles on 3D
culturing of engineered bone were studied. Mouse osteo-
blast-like MC 3T3-E1 cells were 3D-cultured with porous
ceramic scaffolds in vitro for 6 days under static and
hydrodynamic conditions with either a unidirectional or
oscillatory flow. We found that, in the static culture, the cells
proliferated only on the scaffold surfaces. In perfusion
culture with the unidirectional flow, the proliferation was
significantly higher than in the other groups but was very
inhomogeneous, which made the construct unsuitable for
transplantation. Only the oscillatory flow allowed osteogenic
cells to proliferate uniformly throughout the scaffolds, and
also increased the activity of alkaline phosphatase (ALP).
These results suggested that oscillatory flow might be better
than unidirectional flow for 3D construction of cell-seeded
artificial bone. The oscillatory perfusion system could be
a compact, safe, and efficient bioreactor for bone tissue
engineering.
Biotechnol. Bioeng. 2009;102: 16701678.
2008 Wiley Periodicals, Inc.
KEYWORDS: tissue engineering; perfusion culture; in vitro;
osteogenesis
Correspondence to: K.S. Furukawa
Contract grant sponsor: Japanese Ministry of Education, Culture, Sports, Science, and
Technology
Contract grant sponsor: Center for NanoBio Integration
Introduction
Bone tissue engineering aims to provide functional bone
substitutes for clinical treatment of bone defects by
integrating osteoprogenitor cells into porous scaffolds in
vitro (Goldstein et al., 2001; Sikavitsas et al., 2003b). As the
diffusional limitation of nutrients prevents the uniform
growth of seeded cells on the scaffolds after prolonged three-
dimensional (3D) culture in vitro, various hydrodynamic
bioreactors have been designed to enhance chemotransport
for 3D culture in vitro (Freed et al., 1997; Holtorf et al.,
2005b; Qiu et al., 1999; Vunjak-Novakovic et al., 1998).
The bioreactors physical environment, which simulates
in vivo conditions, has also been proposed to be helpful for
fabricating desired tissues. For the generation of bone tissue,
it appears that bone cells in vivo are subjected mainly to fluid
shear stress, from mainly two kinds of fluid profile at the
cellular level: an outward radial unidirectional flow driven
by a hydrostatic pressure drop across the cortex (McAllister
et al., 2000; Winet, 2003) and an oscillatory flow induced by
mechanical loading (Cowin et al., 1995; Kufahl and Saha,
1990; Rubin et al., 2006; Weinbaum et al., 1994; Zhang et al.,
2007).
Two-dimensional flow chamber studies have shown that
the shear stress induced by unidirectional flow could elicit
a flow-induced osteogenenic response from bone cells by
increasing ALP activity (Jessop et al., 2002; Kapur et al.,
2003; McAlliste and Frangos, 1999; Pavalko et al., 1998), and
the expression of osteocalcin (Lan et al., 2003), osteopontin
(Kreke and Goldstein, 2004), and bone sialoprotein (Kreke

1670 Biotechnology and Bioengineering, Vol. 102, No. 6, April 15, 2009 2008 Wiley Periodicals, Inc.
et al., 2005). The unidirectional flow has also been applied to
the 3D culture of engineered bone. Unidirectional perfusion
bioreactors conveyed culture media throughout the inter-
connected pores of scaffolds to continuously introduce
nutrients and remove wastes within the scaffolds, and
applied a certain physical stimulation by shear stress as well
(Cartmell et al., 2003; Porter et al., 2005). Studies with
unidirectional perfusion systems have suggested that
unidirectional flow has many biological effects (Batra
et al., 2005; Botchwey et al., 2003; Meinel et al., 2004; van
den Dolder et al., 2003; Wang et al., 2003), such as
prostaglandin E2 release (Vance et al., 2005), osteogenic
differentiation enhancement (Holtorf et al., 2005a) and
mineralized matrix production increase (Bancroft et al.,
2002; Sikavitsas et al., 2003a) on bone marrow stromal
osteoblasts seeded in 3D scaffolds.
On the other hand, mechanical loading of bone is a potent
stimulus for new bone formation, and the fluid shear stress
induced on bone cells in vivo is thought to be dynamic in
nature. Therefore, oscillatory fluid flow may be a more
physiologic fluid flow profile than unidirectional fluid flow
(Ponik et al., 2006). Accordingly, a bioreactor with
oscillatory fluid flow might provide a more beneficial
physical environment for bone tissue engineering than one
with unidirectional flow. Some 2D flow chamber studies
have shown that the oscillatory flow profiles have biological
effects on cultured bone cells (Li et al., 2004; Qin et al.,
2003; Wu et al., 2006; You et al., 2000), such as increased
osteopontin expression (Batra et al., 2005; You et al., 2001).
However, very few studies have examined the application of
oscillatory fluid flow to 3D construction of cell-seeded bone
grafts. Likewise, few studies have directly compared the
differences in long-term durability between unidirectional
and oscillatory flow in the construction of artificial bone. It
should be determined which kind of fluid profile is better for
constructing engineered bone. A study of the difference in
biological effects between unidirectional and oscillatory flow
on 3D bone development in vitro would be not only
important for determining the suitable hydrodynamic
conditions of bioreactors, but would also provide useful
information about how bone cells respond to the two
different fluid profiles in 3D culture environments.
Certainly, when we consider the favorability of a
hydrodynamic culture system for clinical application of
bone tissue engineering, we should consider not only the
most suitable fluid profile but also how well the inoculation
method, convenience, safety, and economic efficiency are
balanced.
Our recently developed compact perfusion system with
oscillatory flow appears to enhance early osteogenesis and
the uniformity of cultured bone constructs (Du et al.,
2008). Here, we develop a perfusion culture system with
unidirectional flow and compare its biological effects on the
3D construction of cell-seeded bone grafts against those of a
perfusion system with oscillatory flow. Mouse osteoblast-
like cells, MC 3T3-E1, were cultured in porous ceramic
scaffolds by either static, unidirectional perfusion, or
oscillatory perfusion culture for 6 days. Cell proliferation,
early osteogenic effects, and viability were then evaluated.
Materials and Methods
Materials
Porous b-TCP ceramic scaffolds, OSferion, were provided
by Olympus Terumo Biomaterials Corp. (Tokyo, Japan).
The scaffolds were cylindrical (10 mm in diameter and 8 mm
in height) and had a porosity of 75%.
Cell Seeding and Culture
A mouse immature osteoblast-like cell line, MC 3T3-E1
(American Type Culture Collection, Manassas, VA), was
used for this study, which could be further differentiated by
osteogenic factors (Jadlowiec et al., 2004; Miyahara et al.,
1991) and widely used for studies on biological effects of
fluid flow (Jackson et al., 2006; Liu et al., 2007). The
cells were subcultured in a culture medium consisting of
a-minimum essential medium (a-MEM, Gibco BRL, Grand
Island, NY), 1% penicillinstreptomycin, and 10% fetal
bovine serum (FSB; Gibco BRL) in a 378C incubator with a
humidified atmosphere of 95% air and 5% CO2.
A unidirectional perfusion system was designed as shown
in Figure 1A. A media tank was set on each end of the
perfusion chamber as both media reservoirs and gas bubble
trappers. A gas-permeable bag was set in the circuit loop as a
gas exchanger. The unidirectional media flow was driven by
a Pharmacia P-500 pump (Pharmacia LKB, Freiburg,
Germany). A couple of silicon tubes led from Tank 2 to
the clean bench, so that the medium could be exchanged in a
sterile environment without moving the whole complicated
system. The perfusion system with the scaffolds set inside
was sterilized by ethylene oxide gas.
Before cell seeding, the whole perfusion circuit except the
perfusion chamber was filled with culture medium by
sequentially and well controlling of the connector switches
to avoid gas bubbles. After aspiration of the medium above
the scaffold, 1.5
106 MC 3T3-E1 cells in 100 mL medium
were seeded into the scaffold from above. The cell
suspension was then driven up and down oscillatorily
10 times with a 1 mL syringe through a three-way switch
below the chamber. After a 30 min static period, additional
medium was added from above to fill the circuit before the
chamber was closed, and the system was then set to a 378C
incubator. In total, 250 mL of culture medium was used for
the unidirectional perfusion culture. The flow rate for this
culture was 1.0 mL/min. Every other day, 100 mL of the
culture medium was changed.
The oscillatory perfusion culture bioreactor used in this
study was described in detail previously (Du et al., 2008) as
shown in Figure 1B. Briefly, the oscillatory flow in the
perfusion well was driven by a flexible bottom made of a
0.3 mm-thick silicon film controlled by a programmable
Du et al.: Bone Graft by Unidirectional/Oscillatory Flow 1671
Biotechnology and Bioengineering
Figure 1. Illustrations of the 3D perfusion culture systems for tissue engineering bone by either unidirectional flow or oscillatory flow. A: The unidirectional perfusion system;
(B) the oscillatory perfusion system. [Color figure can be seen in the online version of this article, available at www.interscience.wiley.com.]

syringe pump (KDS-270, KD Scientific, Holliston, MA) as
shown in Figure 1B. A suspension of 1.5
106 MC 3T3-E1
cells in 100 mL medium was seeded into each scaffold in the
perfusion assembly from the top, with the bottom silicon
membrane in its highest position. After the cell suspension
was perfused through the scaffolds 10 times, the system
incubator was set for 30 min statically for the initial
attachment of cells to the scaffolds. After a resting period,
1.4 mL culture medium was added to make 1.5 mL total
volume of medium. Then 0.5 mL/min and 1.0 mL/min were
selected to perfuse the scaffolds. The culture medium was
completely refreshed every day.
After seeding by the oscillatory perfusion system, another
group of scaffolds was released into a 24-well plate and
cultured statically in 1.5 mL medium as a static culture
group.
The scaffolds were extracted after 6 days of culture under
different conditions, and carefully cut into halves long-
itudinally from the midline by a knife. Half was used for
microscopic study, and the other half was weighed and
immediately frozen at 308C until it was needed for DNA
content and ALP activity assays.
DNA Content Assay
The frozen half-scaffolds were homogenized, and the
homogenate was lysed in 1 mL buffer (50 mM TrisHCl,

1672 Biotechnology and Bioengineering, Vol. 102, No. 6, April 15, 2009
0.1% (v/v) Triton X-100, 100 mM phosphate buffer,
pH 7.6). The phosphate buffer was used to reduce the
interference of Ca-P on DNA measurement. The samples
were then sonicated on ice for 30 s, vortexed for 510 s, after
centrifuged at 1,300 rpm for 10 min. The supernatant was
collected for both DNA content measurement and ALP
activity assay. Standards of lambda DNA were prepared
from 0 to 6 mg/mL. Fifty microliters of standard or sample
supernatant was placed into each well of a 96-well plate.
Trisethylenediaminetetraacetic acid buffer (TE buffer) and
PicoGreen dye solution were prepared according to the
manufacturers instructions and added at 100 and 150 mL/
well, respectively. After shaking in the dark and 10 min
incubation at room temperature, the fluorescence was
measured with a Wallac 1420 ARVO SX multilabel counter
(Perkin-Elmer Life Sciences, Waltham, MA) using an
excitation wavelength of 480 nm and an emission
wavelength of 520 nm.
ALP Activity Assay
The ALP activity was measured by a colorimetric endpoint
assay. The same supernatant samples as those prepared for
the DNA content assay were used. Standards of p-
nitrophenol in concentrations ranging from 0 to 2 mM
were prepared from dilutions of a 10 mM stock solution,
and 80 mL of the standard or supernatant sample was placed
into each well of a 96-well plate. An alkaline buffer solution
consisting of 1.5 M 2-amino-2-methyl-1-propanol at
pH 10.3 was then added at 20 mL/well. Substrate solution
was prepared by dissolving 100 mg of 4-nitrophenyl
phosphate disodium salt hexahydrate in 25 mL of
ddH2O, and was added at 100 mL/well. The microplate
was incubated for 1 h at 378C. The reaction was stopped by
the addition of 100 mL 0.3 M NaOH per well. The
absorbance of each well at 405 nm was then measured by a
plate reader.
Cell Viability Analysis
The scaffold halves for microscopic study were further cut
longitudinally into approximately 3 mm-thick trapezoidal
sections to provide stability during observation. The
scaffolds were rinsed with pre-warmed PBS prior to analysis.
One milliliter PBS containing 2 mL of 1 mM calcein-AM and
2 mL of 1.5 mM PI was added to each well with a scaffold of a
24-well plate. After 15 min incubation at 378C, the plate was
set on ice. The scaffolds were observed under a fluorescence
stereomicroscope (Leica MZ FL b, Leica Microsystems,
Glattbrugg, Switzerland) to assess the regional distribution
of the living cells in green (calcein staining) (lex 480/40
nm, lem 510 nm) and dead cells in red (PI staining)
(lex 546/10 nm, lem 590 nm). Images were captured
for each location using a 3CCD digital camera (C7780,
Hamamatsu Photonics, Hamamatsu City, Japan). Live and
dead cells in the upper, middle, and lower regions of the
scaffolds were quantified, and the percentage of cells with
exclusively green fluorescence (interpreted as viable cells) in
each region was calculated.
Statistics
All values are presented as means  standard deviation.
Differences among experimental groups were assessed by
ANOVA residual analyses and were considered statically
significant when P < 0.05.
Results
As shown in Figure 2, the calcein-AM/PI double staining of
the midline section of the samples after 6 days of culture
demonstrated that the living cells grew only on the surfaces
of the scaffolds cultured statically (Fig. 2A); the cells grew
extremely inhomogeneously in unidirectional perfusion
culture, where there was a clear inverted arch-shaped
interface between living cells and the non-living area
(Fig. 2B). In contrast, the scaffolds cultured by the
oscillatory flow had a relatively uniform distribution of
living cells throughout the scaffolds. The inhomogeneous
living cell distribution of the unidirectional perfusion
culture was further verified by higher magnification, as
shown in Figure 3. In the unidirectional perfusion culture,
the top surfaces of the scaffolds were covered by a thick layer
of cells and, above the arch-shaped interface, there were
dense living cells over the pore surface. However, very few
cells were living below the interface or at the bottom of the
scaffolds.
As shown in Figure 4, the quantitative study of cell
viability in the section further confirmed the findings
presented above. In static culture, there were few living cells
in the section view, and these were accumulated only near
the top surface. In the unidirectional perfusion culture, the
living cells and dead cells were distributed extremely
inhomogeneously, with living cells favoring the upper
positions (Fig. 4A). However, although there were fewer cells
in the lower part of the scaffold in the unidirectional
perfusion culture, their viability was not low (Fig. 4B); in the
oscillatory perfusion culture, the cells proliferated uniformly
throughout the scaffolds.
The total cellularity of the scaffolds after 6 days of culture
was evaluated by DNA content analysis, as shown in
Figure 5. The total cell number in unidirectional perfusion
was significantly higher than in any of the other groups
( P < 0.05).
The average ALP activity did not differ significantly
between the unidirectional perfusion culture and the static
culture ( P > 0.05), and was significantly higher in the
oscillatory flow perfusion group than in the unidirectional
flow perfusion group ( P < 0.05), although that of the
oscillatory perfusion at 1 mL/min was also not significantly
higher than that of the static group (Fig. 6).
Du et al.: Bone Graft by Unidirectional/Oscillatory Flow 1673
Biotechnology and Bioengineering
Figure 2. Calcein-AM (green; live cells)/propidium iodide (red; dead cells) double fluorescence staining of the scaffolds cultured with MC 3T3-E1 cells for 6 days by static and
perfusional methods with either unidirectional flow (1 mL/min) or oscillatory flow (0.5 mL/min or 1 mL/min): (A) static culture; (B) unidirectional perfusion culture at 1 mL/min;
(C) oscillatory perfusion culture at 0.5 mL/min; (D) oscillatory perfusion culture at 1 mL/min. [Color figure can be seen in the online version of this article, available at
www.interscience.wiley.com.]
Discussion
We cultured osteoblast-like cells on porous ceramic
scaffolds by unidirectional flow, oscillatory flow, or no
flow for 6 days to then compare the biological effects of each
of these three 3D culture methods.
A unidirectional perfusion system was fabricated and
operated in a manner similar to that of conventional designs
(Cartmell et al., 2003). As we had anticipated that the
unidirectional system would be complicated to operate, we
tried to make the operation safer through measures such as
pre-setting the scaffolds inside the chamber before
sterilization and cell seeding instead of assembling them
after cell seeding, using tubes set in the clean bench to
exchange the medium, and so on. Even then, however, it was
still very difficult to safely perform the sterile manipulation
of the unidirectional perfusion system. For example, when
the system was being assembled before seeding, many
connectors must be switched on and off frequently in order
to remove the trapped bubbles in the different parts of the
circuit; on this occasion, it was very easy for the finger-
touched areas outside the tubes to contaminate the culture
medium filling the circuit. Moreover, during perfusion
culture, once gas bubbles occurred inside the perfusion
circuit, which might alter the flow profile, it was almost
1674 Biotechnology and Bioengineering, Vol. 102, No. 6, April 15, 2009
impossible to release them safely without breaking the closed
circuit entity full of culture medium. In contrast, the
oscillatory perfusion system was very compact, without any
switches along the medium pass-way. This allowed the
device to be handled safely and easily in the clean bench and
incubator. The trapped gas bubbles under the scaffolds
could also be easily removed by pushing the silicon bottom
to its highest position several times.
It was supposed that the unidirectional perfusion system
would not allow the adjustment of cell distribution during
the prolonged culture due to the one-way flow profile once
inhomogeneous seeding occurred. Unfortunately, in the
conventional unidirectional perfusion culture method, most
researchers seeded the cells from the top of the scaffolds by a
static method outside the system, which might lead to the
perfusion culture starting with an inhomogeneous cell
inoculation. Our former work showed that oscillatory flow
could allow homogeneous cell inoculation into the porous
3D scaffolds (Du et al., 2008). Therefore, the cell suspension
in the unidirectional perfusion system was also perfused
oscillatorily 10 times by a 1 mL syringe after seeding, as was
also the case with oscillatory seeding in the oscillatory
perfusion system. However, after double staining by calcein-
PI, we found that the osteoblast-like cells proliferated in a
relatively uniform manner only within the upper 4 mm
Figure 3. Calcein-AM (green; live cells)/propidium iodide (red; dead cells) double fluorescence staining of the scaffolds cultured with MC 3T3-E1 cells for 6 days by static and
perfusional methods with either unidirectional flow (1 mL/min) or oscillatory flow (0.5 mL/min). Images of scaffolds observed at 108C objective of a fluorescence stereomicroscope
are shown according to the regions indicated by the broad views on the left. [Color figure can be seen in the online version of this article, available at www.interscience.wiley.com.]
region of the scaffolds under the unidirectional flow. This
suggested that the unidirectional perfusion culture might be
inefficient for 3D culture of bone grafts thicker than 4 mm.
In contrast, the oscillatory flow cultured the engineered
bone graft uniformly even at a thickness of 8 mm. A possible
explanation for this is that the cells on the lower part of the
scaffolds were lost in the medium below during the 30 min
static period in the unidirectional perfusion system. Once
they became detached, it would be difficult to return them to
the scaffold surface; more likely they would become lost in
the large circulation medium reservoir during the one-way
flow. However, the oscillatory flow profile in the oscillatory
perfusion system might have the potential to redistribute
the cells during the culture and reseed the cells lost
in the medium back to the scaffolds after inoculation, due to
the back-and-forth nature of the oscillatory flow profile.
This, in turn, would enable a homogeneous construction of
artificial bone.
The total cell number was significantly higher after 6 days
of culturing by unidirectional flow than oscillatory flow.
This might be attributable to the enhancement of osteoblast
proliferation by the unidirectional fluid profile in the three-
dimensional culture. Another possibility is that the large
volume of culture medium used in the unidirectional
perfusion culture could provide more nutrients and thus
increase cell proliferation, whereas the small volume of
culture medium in the oscillatory perfusion culture could
promote the accumulation of cellular products and enhance
osteogenic differentiation, which was shown in the ALP
activity results. Thus, as the proliferation of osteogenic cells
decreased when differentiation increased, the proliferation
rates in the oscillatory perfusion groups were lower than
those in the unidirectional perfusion culture.
A recent study investigated the effects of intermittent
steady, pulsatile, and oscillatory fluid flow on bioreactor
cultured osteoblast-like cells, and indicate that intermittent
unidirectional flow favorable for bone tissue engineering
(Jaasma and OBrien, 2008). However, the flow rates used
was 1 mL/min, which was unfavorable for oscillatory
perfusion culture by the data in this research. Moreover, the
culture period was only 48 h, which were too short for a
tissue engineering purpose. On the other hand, it might be
useful to study the biological effects of intermittent
oscillatory flow in the future.
Cell-seeded artificial bone with homogeneous osteogenic
potential, similar to autograft, was favorable for clinical
application. Although the unidirectional flow enhanced
bone cell proliferation, the proliferation was extremely
inhomogeneous and limited to the upper part of the
construct. This made the effects of proliferation enhance-
ment by the unidirectional flow less meaningful at a 3D
level. On the other hand, the oscillatory flow could not only
allow uniform proliferation of bone cells throughout the
scaffolds, but also could induce early osteogenesis. More-
over, the oscillatory perfusion system was much more
compact, easier, and safer for sterile manipulation. It was
Du et al.: Bone Graft by Unidirectional/Oscillatory Flow 1675
Biotechnology and Bioengineering
Figure 4. Quantitative study of cell viability distribution according to the calcein-AM/PI double staining of the middle section after 6 days of culture observed under
fluorescent stereomicroscope under 10
objective. The numbers 0.5 and 1.0 refer to the flow rates of these groups in mL/min. A: Living and dead cell numbers per view of respective
regions were counted at the same magnification as in Figure 3. U: upper; M: middle; L: lower; (B) viability of respective region derived from (A); Error bars represent means  SD,
n 6.

also economically more efficient by virtue of the small

volume of culture medium required for perfusion culture,
especially when conditioned culture was applied, such as
gene vectors and growth factors. Therefore, the oscillatory
fluid flow might be better than unidirectional flow for 3D
culture of engineered bone in vitro. The oscillatory
perfusion system could be a very useful bioreactor for bone
tissue engineering.

Figure 5. The cell number per scaffold after 6 days of culture. The numbers 0.5
and 1.0 refer to the flow rates of the groups in mL/min. Error bars represent
means  SD, n 6. The asterisk () indicates a statistically significant difference
between the static group and all other groups ( P < 0.05).
Conclusion
Osteoblast-like cells were cultured with porous ceramic
scaffolds three-dimensionally in vitro for 6 days under static
and hydrodynamic conditions with either unidirectional or
oscillatory flow. We found that although the unidirectional
flow increased cell proliferation, the proliferation was
extremely inhomogeneous, which rendered the engineered
bone unsuitable for transplantation. On the other hand, the
oscillatory flow enabled uniform proliferation of osteogenic
cells and increased early osteogenesis. This suggested that

1676 Biotechnology and Bioengineering, Vol. 102, No. 6, April 15, 2009

Figure 6. The average ALP activity per cell after 6 days of culture. The numbers
0.5 and 1.0 refer to the flow rates of the groups in mL/min. Error bars represent
means  SD, n 6. The asterisk () indicates a statistically significant difference
between the static group and the perfusion group ( P < 0.05). # indicates a statistically
significant difference between the unidirectional perfusion group and the oscillatory
perfusion group ( P < 0.05).
the oscillatory fluid flow might be better than unidirectional
flow for 3D culture of engineered bone in vitro. The
oscillatory perfusion system could be a compact, safe, and
efficient bioreactor for bone tissue engineering.
This work was supported by Grants-in-Aid for Scientific Research (S
and A) from the Japanese Ministry of Education, Culture, Sports,
Science, and Technology; and by funding from the COE Program,
NEDO project, and the Center for NanoBio Integration in the
University of Tokyo. We gratefully acknowledge Olympus Terumo
Biomaterials Corp. for the supply of scaffolds.
References
Bancroft GN, Sikavitsas VI, van den Dolder J, Sheffield TL, Ambrose CG,
Jansen JA, Mikos AG. 2002. Fluid flow increases mineralized matrix
deposition in 3D perfusion culture of marrow stromal osteoblasts in a
dose-dependent manner. Proc Natl Acad Sci USA 99(20):1260012605.
Batra NN, Li YJ, Yellowley CE, You L, Malone AM, Kim CH, Jacobs CR.
2005. Effects of short-term recovery periods on fluid-induced signaling
in osteoblastic cells. J Biomech 38(9):19091917.
Botchwey EA, Pollack SR, El-Amin S, Levine EM, Tuan RS, Laurencin CT.
2003. Human osteoblast-like cells in three-dimensional culture with
fluid flow. Biorheology 40(13):299306.
Cartmell SH, Porter BD, Garca AJ, Guldberg RE. 2003. Effects of medium
perfusion rate on cell-seeded three-dimensional bone constructs in
vitro. Tissue Eng 9(6):11971203.
Cowin SC, Weinbaum S, Zeng Y. 1995. A case for bone canaliculi as the
anatomical site of strain generated potentials. J Biomech 28(11):1281
1297.
Du D, Furukawa K, Ushida T. 2008. Oscillatory Perfusion Seeding and
Culturing of Osteoblast-like Cells on Porous Beta-Tricalcium Phos-
phate Scaffolds. J Biomed Mater Res A 86(3):796803.
Freed LE, Langer R, Martin I, Pellis NR, Vunjak-Novakovic G. 1997. Tissue
engineering of cartilage in space. Proc Natl Acad Sci USA 94(25):
1388513890.
Goldstein AS, Juarez TM, Helmke CD, Gustin MC, Mikos AG. 2001. Effect
of convection on osteoblastic cell growth and function in biodegradable
polymer foam scaffolds. Biomaterials 22(11):12791288.
Holtorf HL, Jansen JA, Mikos AG. 2005a. Flow perfusion culture induces
the osteoblastic differentiation of marrow stroma cell-scaffold con-
structs in the absence of dexamethasone. J Biomed Mater Res A 72(3):
326334.
Holtorf HL, Sheffield TL, Ambrose CG, Jansen JA, Mikos AG. 2005b. Flow
perfusion culture of marrow stromal cells seeded on porous biphasic
calcium phosphate ceramics. Ann Biomed Eng 33(9):12381248.
Jaasma MJ, OBrien FJ. 2008. Mechanical stimulation of osteoblasts using
steady and dynamic fluid flow. Tissue Eng Part A 14(7):12131223.
Jackson RA, Kumarasuriyar A, Nurcombe V, Cool SM. 2006. Long-term
loading inhibits ERK1/2 phosphorylation and increases FGFR3 expres-
sion in MC3T3-E1 osteoblast cells. J Cell Physiol 209(3):894904.
Jadlowiec J, Koch H, Zhang X, Campbell PG, Seyedain M, Sfeir C. 2004.
Phosphophoryn regulates the gene expression and differentiation of
NIH3T3, MC3T3-E1, and human mesenchymal stem cells via the
integrin/MAPK signaling pathway. J Biol Chem 279(51):5332353330.
Jessop HL, Rawlinson SC, Pitsillides AA, Lanyon LE. 2002. Mechanical
strain and fluid movement both activate extracellular regulated kinase
(ERK) in osteoblast-like cells but via different signaling pathways. Bone
31(1):186194.
Kapur S, Baylink DJ, Lau KH. 2003. Fluid flow shear stress stimulates
human osteoblast proliferation and differentiation through multiple
interacting and competing signal transduction pathways. Bone 32(3):
241251.
Kreke MR, Goldstein AS. 2004. Hydrodynamic shear stimulates osteocalcin
expression but not proliferation of bone marrow stromal cells. Tissue
Eng 10(56):780788.
Kreke MR, Huckle WR, Goldstein AS. 2005. Fluid flow stimulates expres-
sion of osteopontin and bone sialoprotein by bone marrow stromal
cells in a temporally dependent manner. Bone 36(6):10471055.
Kufahl RH, Saha S. 1990. A theoretical model for stress-generated fluid flow
in the canaliculi-lacunae network in bone tissue. J Biomech 23(2):171
180.
Lan CW, Wang FF, Wang YJ. 2003. Osteogenic enrichment of bone-marrow
stromal cells with the use of flow chamber and type I collagen-coated
surface. J Biomed Mater Res A 66(1):3846.
Li YJ, Batra NN, You L, Meier SC, Coe IA, Yellowley CE, Jacobs CR. 2004.
Oscillatory fluid flow affects human marrow stromal cell proliferation
and differentiation. J Orthop Res 22(6):12831289.
Liu D, Genetos DC, Shao Y, Geist DJ, Li J, Ke HZ, Turner CH, Duncan RL.
2007. Activation of extracellular-signal regulated kinase (ERK1/2) by
fluid shear is Ca(2)- and ATP-dependent in MC3T3-E1 osteoblasts.
Bone 42(4):644652.
McAlliste TN, Frangos JA. 1999. Steady and transient fluid shear stress
stimulate NO release in osteoblasts through distinct biochemical path-
ways. J Bone Miner Res 14(6):930936.
McAllister TN, Du T, Frangos JA. 2000. Fluid shear stress stimulates
prostaglandin and nitric oxide release in bone marrow-derived pre-
osteoclast-like cells. Biochem Biophys Res Commun 270(2):638
643.
Du et al.: Bone Graft by Unidirectional/Oscillatory Flow 1677
Biotechnology and Bioengineering
Meinel L, Karageorgiou V, Fajardo R, Snyder B, Shinde-Patil V, Zichner L,
Kaplan D, Langer R, Vunjak-Novakovic G. 2004. Bone tissue engineer-
ing using human mesenchymal stem cells: Effects of scaffold material
and medium flow. Ann Biomed Eng 32(1):112122.
Miyahara T, Nemoto M, Tukamoto S, Yamada H, Kozuka H, Kuze S, Sudo
H, Yamamoto S. 1991. Induction of metallothionein and stimulation of
calcification by dexamethasone in cultured clonal osteogenic cells,
MC3T3-E1. Toxicol Lett 57(3):257267.
Pavalko FM, Chen NX, Turner CH, Burr DB, Atkinson S, Hsieh YF, Qiu J,
Duncan RL. 1998. Fluid shear-induced mechanical signaling in
MC3T3-E1 osteoblasts requires cytoskeleton-integrin interactions.
Am J Physiol 275(6 pt 1):C1591C1601.
Ponik SM, Triplett JW, Pavalko FM. 2006. Osteoblasts and osteocytes
respond differently to oscillatory and unidirectional fluid flow profiles.
J Cell Biochem 100(3):794807.
Porter B, Zauel R, Stockman H, Guldberg R, Fyhrie D. 2005. 3D computa-
tional modeling of media flow through scaffolds in a perfusion
bioreactor. J Biomech 38(3):543549.
Qin YX, Kaplan T, Saldanha A, Rubin C. 2003. Fluid pressure gradients,
arising from oscillations in intramedullary pressure, are correlated with
the formation of bone and inhibition of intracortical porosity.
J Biomech 36(10):14271437.
Qiu QQ, Ducheyne P, Ayyaswamy PS. 1999. Fabrication, characterization
and evaluation of bioceramic hollow microspheres used as microcar-
riers for 3D bone tissue formation in rotating bioreactors. Biomaterials
20(11):9891001.
Rubin J, Rubin C, Jacobs CR. 2006. Molecular pathways mediating
mechanical signaling in bone. Gene 367:116.
Sikavitsas VI, Bancroft GN, Holtorf HL, Jansen JA, Mikos AG. 2003a.
Mineralized matrix deposition by marrow stromal osteoblasts in 3D
perfusion culture increases with increasing fluid shear forces. Proc Natl
Acad Sci USA 100(25):1468314688.
Sikavitsas VI, van den Dolder J, Bancroft GN, Jansen JA, Mikos AG. 2003b.
Influence of the in vitro culture period on the in vivo performance of
cell/titanium bone tissue-engineered constructs using a rat cranial
critical size defect model. J Biomed Mater Res A 67(3):944951.
1678 Biotechnology and Bioengineering, Vol. 102, No. 6, April 15, 2009
van den Dolder J, Bancroft GN, Sikavitsas VI, Spauwen PH, Jansen JA,
Mikos AG. 2003. Flow perfusion culture of marrow stromal osteoblasts
in titanium fiber mesh. J Biomed Mater Res A 64(2):235241.
Vance J, Galley S, Liu DF, Donahue SW. 2005. Mechanical stimulation of
MC3T3 osteoblastic cells in a bone tissue-engineering bioreactor
enhances prostaglandin E2 release. Tissue Eng 11(1112):1832
1839.
Vunjak-Novakovic G, Obradovic B, Martin I, Bursac PM, Langer R, Freed
LE. 1998. Dynamic cell seeding of polymer scaffolds for cartilage tissue
engineering. Biotechnol Prog 14(2):193202.
Wang Y, Uemura T, Dong J, Kojima H, Tanaka J, Tateishi T. 2003.
Application of perfusion culture system improves in vitro and in vivo
osteogenesis of bone marrow-derived osteoblastic cells in porous
ceramic materials. Tissue Eng 9(6):12051214.
Weinbaum S, Cowin SC, Zeng Y. 1994. A model for the excitation of
osteocytes by mechanical loading-induced bone fluid shear stresses.
J Biomech 27(3):339360.
Winet HE. 2003. A bone fluid flow hypothesis for muscle pump-driven
capillary filtration: II. Proposed role for exercise in erodible scaffold
implant incorporation. Eur Cell Mater 6:110.
Wu CC, Li YS, Haga JH, Wang N, Lian IY, Su FC, Usami S, Chien S. 2006.
Roles of MAP kinases in the regulation of bone matrix gene expressions
in human osteoblasts by oscillatory fluid flow. J Cell Biochem 98(3):
632641.
You J, Yellowley CE, Donahue HJ, Zhang Y, Chen Q, Jacobs CR. 2000.
Substrate deformation levels associated with routine physical activity
are less stimulatory to bone cells relative to loading-induced oscillatory
fluid flow. J Biomech Eng 122(4):387393.
You J, Reilly GC, Zhen X, Yellowley CE, Chen Q, Donahue HJ, Jacobs CR.
2001. Osteopontin gene regulation by oscillatory fluid flow via intra-
cellular calcium mobilization and activation of mitogen-activated
protein kinase in MC3T3-E1 osteoblasts. J Biol Chem 276(16):
1336513371.
Zhang P, Su M, Liu Y, Hsu A, Yokota H. 2007. Knee loading dynamically
alters intramedullary pressure in mouse femora. Bone 40(2):538
543.