Home
Search
Collections
Journals
About
Contact us
My IOPscience
A droplet-based building block approach for bladder smooth muscle cell (SMC) proliferation
This content has been downloaded from IOPscience. Please scroll down to see the full text.
2010 Biofabrication 2 014105
(http://iopscience.iop.org/1758-5090/2/1/014105)
View the table of contents for this issue, or go to the journal homepage for more
Download details:
IP Address: 83.171.31.36
This content was downloaded on 28/10/2015 at 05:01
Please note that terms and conditions apply.
IOP PUBLISHING
BIOFABRICATION
doi:10.1088/1758-5082/2/1/014105
Biofabrication 2 (2010) 014105 (9pp)
A droplet-based building block approach
for bladder smooth muscle cell (SMC)
proliferation
F Xu1,4 , S J Moon1,4 , A E Emre1 , E S Turali1 , Y S Song1 , S A Hacking1 ,
J Nagatomi2 and U Demirci1,3
1
Department of Medicine, Bio-Acoustic-MEMS in Medicine (BAMM) Laboratory, Center for
Biomedical Engineering, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA
2
Department of Bioengineering, Clemson University, Clemson, SC, USA
3
Harvard-MIT Health Sciences and Technology, Cambridge, MA, USA
E-mail: [email protected]
Received 16 September 2009
Accepted for publication 2 December 2009
Published 10 March 2010
Online at stacks.iop.org/BF/2/014105
Abstract
Tissue engineering based on building blocks is an emerging method to fabricate 3D tissue
constructs. This method requires depositing and assembling building blocks (cell-laden
microgels) at high throughput. The current technologies (e.g., molding and photolithography)
to fabricate microgels have throughput challenges and provide limited control over building
block properties (e.g., cell density). The cell-encapsulating droplet generation technique has
potential to address these challenges. In this study, we monitored individual building blocks
for viability, proliferation and cell density. The results showed that (i) SMCs can be
encapsulated in collagen droplets with high viability (>94.2 ± 3.2%) for four cases of initial
number of cells per building block (i.e. 7 ± 2, 16 ± 2, 26 ± 3 and 37 ± 3 cells/building
block). (ii) Encapsulated SMCs can proliferate in building blocks at rates that are consistent
(1.49 ± 0.29) across all four cases, compared to that of the controls. (iii) By assembling these
building blocks, we created an SMC patch (5 mm × 5 mm × 20 μm), which was cultured for
51 days forming a 3D tissue-like construct. The histology of the cultured patch was compared
to that of a native rat bladder. These results indicate the potential of creating 3D tissue models
at high throughput in vitro using building blocks.
S Online supplementary data available from stacks.iop.org/BF/2/014105/mmedia
(Some figures in this article are in colour only in the electronic version)
assembly. There are existing methods to fabricate microscale
building blocks [7–11]. Micromolding is a method where
live mammalian cells are encapsulated within microscale
hydrogels (i.e. microgels) using various materials such as
methacrylated hyaluronic acid (meHA), collagen, fibrin,
agarose, PEG and chitosan [12]. However, micromolding
offers limited control over various aspects, such as building
block size and cell density within individual building blocks.
Microscale building blocks have also been fabricated using
photolithographic techniques. Light (e.g., UV) has been used
1. Introduction
Building block, or ‘LEGO’ approach (e.g., cell-laden
microgels [1] and cell aggregates [2]), is a directed assembly
method to build 3D tissue constructs. These building blocks
are assembled using manual handling [3, 4], microfluidics
[5, 6] or self-assembly [1] to form the basis of an organized
tissue. Following the sub-assembly steps, the components
form a tissue construct using component-by-component
4
Both authors contributed equally to this work.
1758-5082/10/014105+09$30.00
1
© 2010 IOP Publishing Ltd Printed in the UK
Biofabrication 2 (2010) 014105
F Xu et al
to crosslink cell-laden hydrogels (e.g., PEG) to build microgels
[13]. This method offers high spatial resolution (<10 μm) of
encapsulated cells within a hydrogel [13]. However, both
methods are low throughput. Further, exposure of cells to UV
light is not biologically desired [14]. Further, current methods
(e.g., self-assembly) to assemble these building blocks into a
3D architecture are challenging.
A technology that could simultaneously generate and
position cell-laden microgels would offer advantages to
address the above-listed challenges.
Cell-encapsulating
droplet generation is a promising technology that can pattern
cells encapsulated in scaffolding materials (e.g., natural
hydrogels such as collagen).
This technology enables
fabrication of building blocks at high throughput. Several
droplet generation technologies are reported, such as inkjet
[15–17], laser [18, 19] and acoustics [20–22]. We have
previously used a valve-based ejector to generate cellencapsulating droplets [23] and investigated large cell-laden
gel constructs built by this technology [24]. To optimize the
properties of 3D tissue constructs created by a droplet-based
approach, it is beneficial to explore the properties of individual
building blocks focusing on cell counts, proliferation rates and
viability.
Smooth muscle cells (SMCs) are associated with
various diseases, such as atherosclerosis [25], cancer [26],
obstructive bladder disease [27], as well as gastrointestinal and
reproductive disorders [28]. 400 million people worldwide
are affected by bladder diseases [29], such as urinary bladder
cancer [30] and interstitial cystitis [31]. Further, high rates of
SMC proliferation have been associated with atherosclerosis
and hypertension [32]. Therefore, investigation of the role and
behavior of SMCs is helpful to better understand the pathology
of disease and its potential treatment. In vitro tissue models
have been used to study smooth muscle diseases using porous
polymer scaffolds [33]. Despite advances in the scaffoldbased approaches [34], these models are still not optimal
due to several limitations: cell distribution control and low
throughput.
There are studies showing that printed cells proliferate
in vitro (e.g., endothelial cells and fibroblast). However,
hepatocellular carcinoma (HepG2) cells were reported not
to proliferate within 7 days post-printing via a pneumatic
microvalve [35]. To the best of our knowledge, no studies
have addressed SMC proliferation post-droplet formation in a
single hydrogel building block. Besides, challenges exist in
the field with long-term culture of printed structures, such as
sterility, viability and contamination.
Here, we present that cell-encapsulating droplet
generation can be used to fabricate building blocks with
measurable, controllable and repeatable properties (e.g., cell
density, viability, proliferation) at high throughput. These
building blocks can be assembled to create 3D tissue structures.
We studied (i) SMC viability and proliferation in individual
building blocks (collagen droplets encapsulating cells) and (ii)
a SMC patch created by overlapping these building blocks
over long-term culture.
2. Materials and methods
2.1. Building block fabrication system
We developed a cell-encapsulating droplet generation system
enclosed within a sterile hood (Cleanroom International,
13202, Grand Rapids, MI) that features controlled humidity
and temperature of the printing process and enables long-term
culture (figure 1(a)). The system has been described in our
earlier study [24] and is briefly summarized here. The system
consists of two key subsystems: a 3D stage movement system
controlled by a stage controller (Newmark Systems, Mission
Viejo, CA) and a droplet deposition system controlled by a
pulse generator (Hewlett Packard, 8112 A, Houston, TX).
The cell-encapsulating droplet generation system is employed
to generate building blocks (i.e. cells encapsulated in media
and hydrogel droplets), guided by computer-aided design
R
, Autodesk Inc.). The building
software (e.g., AutoCAD
block generator (solenoid valve ejector, model G100–150300,
from TechElan, Mountainside, NJ) faces a receiving plate (e.g.,
culture dish or hydrogel substrate). The receiving plate is
placed on top of a XYZ stage that allows precise deposition
of building blocks (stage resolution of 1 μm). This system
provides precise spatial control over droplet deposition, and
accuracy of droplet placement is less than 20 μm [24].
2.2. Cell suspension for building block generation
2.2.1. Cell isolation and preparation. Primary SMCs
were isolated from rat bladder tissue (Sprague-Dawley,
female, 250–300 g) following reported methods [36]. Rats
were euthanized by CO2 aphyxiation, and the bladder was
excised and cleaned of fat and connective tissue. Following
morcelization, the fine tissue was collected in 2 ml SMC
medium (RPMI 1640, Invitrogen, Carlsbad, CA) and filtered
through a micrometer filter (Falcon mesh sits on a 50 ml conical
tube). SMCs were placed into individual cell culture dishes
with 2 ml SMC medium (sterile, humidified, 37 ◦ C, 5% CO2 )
and trypsinized after the culture reached confluency. SMCs at
a low passage number were used (passage 5) for this study. The
SMC phenotype of these cells has been confirmed in an earlier
study by the expression of three different contractile marker
proteins: α-smooth muscle actin, SM22 and h-caldesmon [37].
Initial cell concentration of SMC suspensions was calculated
using hemocytometer and diluted with the SMC medium
to reach the ejector loading cell concentration (0.1 × 106 ,
0.25 × 106 , 0.5 × 106 and 1 × 106 cells ml−1 ).
2.2.2. Cell/collagen mixture preparation. Reconstituted
collagen solutions (500 μl) were prepared by dissolving
250 μl type I rat tail collagen (BD Biosciences, San Jose,
CA) in 50 μl FBS, 50 μl 10× PBS, 50 μl SMC medium,
and by subsequently neutralizing with 50 μl 0.1 M NaOH
to a pH value of 7.4. The SMC suspension was mixed
with reconstituted collagen with 1:1 volume ratio before the
building block generation step. The final cell concentrations
(concentration in the cell–collagen mixture) in this study were
controlled to be between 0.1 and 1 million cells ml−1 . The
final collagen concentration was 2 mg ml−1 (0.2%).
2
Biofabrication 2 (2010) 014105
F Xu et al
(a)
(b)
Figure 1. Cell-laden droplet (building block) generation system. (a) The system consists of two key subsystems, a 3D stage movement
system and a droplet generation system, controlled by a stage controller and pulse generator, respectively. The droplet generation system is
attached to the Z axis of the XYZ stage, while the receiving plate moves within the XY plane. The XYZ stage allows all three axes to be
simultaneously controlled during the building block fabrication process through computer control. (b) Relationship between cell
concentration in the ejection reservoir and the number of cells in a single fabricated building block. Tests were performed at four cell
concentrations (0.1 × 106 , 0.25 × 106 , 0.5 × 106 and 1 × 106 cells ml−1 ) and ten building blocks were chosen randomly from each cell
concentration. Mean and standard deviation at each concentration were 7 ± 2, 16 ± 2, 26 ± 3 and 37 ± 3 cells/droplet for cell
concentrations of 0.1 × 106 , 0.25 × 106 , 0.5 × 106 and 1 × 106 cells ml−1 , respectively.
frequency (10 Hz), so that two neighboring droplets did not
overlap. The fabricated building blocks were ∼30 nl in
volume with a smallest diameter of ∼380 μm and had a
large surface area, which made them vulnerable to drying.
To avoid drying, a small volume of medium (∼50 μl) was
placed around the single building blocks (without contact
during gelling) immediately after ejection to keep a humid
environment. Fabricated cell-encapsulating building blocks
were allowed to gel at 37 ◦ C for 5 min in an incubator before
more SMC medium was added to cover the building blocks.
Then, these building blocks were cultured (37 ◦ C, 5% CO2 ,
sterile, Forma Scientific, CO2 water jacketed incubator).
2.3. Experimental procedure
2.3.1. Building block generation process. The cell/collagen
mixture (2 ml) was placed into a 10 ml syringe reservoir
connected to the valve-based ejector under constant air
pressure (5 psi). Before and after each ejection run, the
ejector was sterilized with 70% ethanol and flushed with
DI water. This step cleans the ejector, minimizes clogging
and increases the lifetime of the ejector. The ejector and
collagen/cell mixture were constantly cooled during ejection
to avoid collagen gelling. The droplet size was controlled by
adjusting pulse width (70 μs) and/or valve pressure (5 psi).
Fabrication of single building blocks (i.e. cell-laden
hydrogel droplets) was achieved by increasing the stage
movement speed (20 mm s−1 ) and decreasing the ejection
2.3.2. Post-ejection staining and tracking of individual
building blocks. Post-ejection cell viability in individual
3
Biofabrication 2 (2010) 014105
F Xu et al
building blocks was assessed using a fluorescent live/dead
viability staining kit (Invitrogen, Carlsbad, CA). The building
blocks were washed with PBS and stained for 10 min at
37 ◦ C with a live/dead staining solution (0.5 μl calcein and
2.0 μl ethidium homodimer-1 (ETH) diluted in 1 ml PBS).
Controls included cell-less collagen droplets and cells cultured
in conventional 2D tissue culture dishes. The droplets were
washed with PBS prior to imaging. Cell proliferation was
monitored over 5 days. Forty cell-encapsulating building
blocks were ejected onto a Petri dish (figure 1(a)). Four tests
of ten cell-laden droplets were used to evaluate the effect of
initial number of cells per droplet (7 ± 2, 16 ± 2, 26 ± 3
and 37 ± 3 cells/droplet) on cell viability. Images were taken
under a bright-field microscope (Nikon TE2000) from day
0 to day 5. The number of cells was counted manually from
the obtained 4× images.
better evaluate this, the average cell viability of SMCs was
assessed (n = 10 per concentration), figure 2(i). Average cell
viability (mean ± standard deviation) for each initial number
of cells per building block was 98.8 ± 3.9%, 94.2 ± 3.2%,
94.3 ± 2.6%, and 94.3 ± 3.7%. This indicates that the overall
cell viability was >94.2 ± 3.2% (absolute values) repeatably
and reliably for 40 building blocks tested. On average, the
cell viability before printing was 97.1% in culture. The cell
viability by our cell-encapsulating building block fabrication
system is comparable to the highest values reported by other
cell printing methods, i.e. inkjet [38] and laser [39]. Our
SMC post-ejection viability results are higher than reported for
coaxial aerodynamically assisted bio-threading (∼64%) [40],
aerodynamically assisted bio-jetting (∼87%) [41] and BES
(∼86%) [42]. Also, in our results, we observed that lesser
initial number of cells per building block (<10 cells) gives
higher cell viability. Similar results have also been observed
in the literature for printed cell-laden collagen lines [43].
The building block fabrication process generally
comprises several sub-processes including droplet formation,
ejection and landing. These sub-processes may induce damage
to cells, especially to fragile mammalian cells, mainly due
to the mechanical and thermal effects (e.g., thermal inkjet)
during droplet formation [44]. In the inkjet method, the
droplet experiences shear strain rate close to 104 s−1 due to
the small nozzle size (10–80 μm diameter) [45, 46]. The
potential damage induced by this high strain impedes postejection cell viability and proliferation. Therefore, we used a
valve-based ejector with a wide nozzle (with nozzle diameter
150 μm) to achieve high post-ejection cell viability and
consistent proliferation rates compared to cell culture controls
[22–24]. The relatively high cell viability in the present study
is due to the large orifice size (150 μm in diameter) and reduced
local shear force created within the droplet during generation.
This mechanical effect is local near the valve and is more
pronounced at higher cell densities. Overall, the system was
able to achieve >94.2 ± 3.2% viability.
3. Results and discussion
3.1. Number of cells per building block
The number of cells per building block can be controlled
by changing the droplet size or the initial loading cell
concentration. In this study, by adjusting the cell/collagen
concentration in the ejection reservoir while keeping the
droplet size constant (∼30 nl), we achieved control over the
number of cells in a single droplet. The mean and standard
deviation at each concentration were 7 ± 2, 16 ± 2, 26 ± 3
and 37 ± 3 cells/droplet (n = 50), at 0.1 × 106 , 0.25 × 106 ,
0.5 × 106 and 1 × 106 cells ml−1 , respectively. Figure 1(b)
shows that the number of cells per building block increases as
cell loading concentration is increased. These results prove
that fabrication of cell-laden building blocks using our system
is a repeatable and controlled process. This is important
because the number of cells per building block determines
the cell densities in the later assembled larger 3D construct, as
discussed in section 3.4.
3.2. Cell viability
3.3. Cell proliferation
Cell viability in single building blocks after fabrication was
assessed using a live/dead assay to investigate the effect of cell
concentration on cell viability (figure 2). We observed higher
cell viability at a lower cell concentration. Bright field and
fluorescent images for four different cases of initial number of
cells (7 ± 2, 16 ± 2, 26 ± 3 and 37 ± 3 cells per building block)
are shown in figures 2(a–b), (c–d), (e–f ), (g–h), respectively.
There were 7 cells in total in the droplet with no dead cells
in figures 2(a), (b); 14 cells in total with 1 dead cell (red
fluorescent stained) in figures 2(c), (d); 27 cells in total with 2
dead cells in figures 2(e), (f ); 39 cells in total with 2 dead cells
in figures 2(g), (h). The cell viability was analyzed for four
initial numbers of cells per building block (see supplementary
figure 1 at stacks.iop.org/BF/2/014105/mmedia) to investigate
the effect of cell concentration on cell viability. Most cells
remained alive post-ejection in ten single building blocks
for initial number of cells 7 ± 2 per building block. With
increasing number of cells per building block, there were dead
cells in almost every building block, but only a few (0–4). To
The total cell count within the building blocks increased in
culture over 5 days. The images of a typical building block
from day 0 to day 5 are given in figure 3(a). From the images
(figure 3(a)), we can also see the cell-laden building block
boundary. For each of four different cases of initial cell density
(7 ± 2, 16 ± 2, 26 ± 3 and 37 ± 3 cells/droplet), ten cellladen building blocks (drops #1, #2, . . . , #10 in supplementary
figure 2(a) at stacks.iop.org/BF/2/014105/mmedia) were
chosen randomly and monitored. Proliferation of SMCs was
observed in all four cases. However, the number of cells per
building block after days of culture differed in each case, even
with the same starting initial number of cells per building block
(e.g., droplets #6 and #7 for the case of 7 ± 2 cells per building
block). This may be due to the fact that cell distribution
(i.e. cell–cell distance) inside each droplet was not the same.
The average number of cells per building block over a 5 day
culture period was calculated (figure 3(b) and supplemental
table 1 at stacks.iop.org/BF/2/014105/mmedia). Larger initial
4
Biofabrication 2 (2010) 014105
F Xu et al
(a)
(b)
(c)
(d)
(e)
( f)
(g)
(h)
(i)
Figure 2. Post-generation cell viability in single building blocks. (a)–(h) Typical live/dead test results for four different cases of initial
number of cells (7 ± 2, 16 ± 2, 26 ± 3 and 37 ± 3 cells/building block). For example, (e) and (f ) are bright field and fluorescent images of
printed single cell-laden collagen droplets. There are 27 cells in this building block. Green dots are live cells (25) whereas red dots are dead
cells (2). (i) Average cell viability (mean + standard deviation) were 98.8 ± 3.9%, 94.2 ± 3.2%, 94.3 ± 2.6%, and 94.3 ± 3.7% for four
different initial number of cells (7 ± 2, 16 ± 2, 26 ± 3 and 37 ± 3 cells/building block), respectively.
number of cells give more cells per building block at the end
of the culture period. This does not mean higher proliferation
rates. We determined the proliferation rate at each day, defined
as NDay(i) /NDay(i−1) with i = 1, 2, 3, 4, 5, supplementary
figure 2(b) at stacks.iop.org/BF/2/014105/mmedia.
The
average proliferation rates were 1.50 ± 0.26, 1.45 ± 0.26,
1.58 ± 0.37 and 1.44 ± 0.21 for four different cases of initial
number of cells per building blocks 7 ± 2, 16 ± 2, 26 ± 3 and
37 ± 3, respectively. The controls were SMCs in culture
dish with a proliferation rate of ∼1.62. These results indicate
that the average proliferation rates were similar, 1.49 ± 0.29,
despite differences in the initial number of cells. This can also
be shown by fitting the data to an exponential proliferation
model, figure 3(b). To investigate the effect of initial number
of cells per building block on cell proliferation at four cell
densities, we used the following proliferation model to fit our
data assuming an exponential increase of cell number:
NDayi = NDay0 × e(a×Dayi )
(1)
where NDayi is the number of cells per droplet at day i (i = 0,
1, . . . , 5), NDay0 is the initial number of cells per droplet at
day 0 and a is a factor related to cell proliferation. It should
be pointed out here that the model used is phenomenological
and we assume that the cell growth is unconstrained. There
are often limitations for how much a cell population can grow,
such as oxygen, nutrient supply and space limitations [47].
All the fabricated building blocks were cultured within the
medium in an incubator, which supplied cells with enough
nutrients. Also, from figure 3(a), the images showed that even
at day 5 the building blocks were not at full confluence. This is
5
Biofabrication 2 (2010) 014105
F Xu et al
(a)
(b)
Figure 3. Cell proliferation in a single building block. (a) Day 0–5 images of the same single building block in culture. A clear increase in
cell number can be observed from these images. (b) The increase of average number of cells per building block (n = 10) for four different
cases of initial number of cells. The data were fitted to the proliferation model NDayx = NDay0 × e(a×Dayx ) . The proliferation-related
parameter, a, for four different cases of initial number of cells was 0.38, 0.45, 0.41, 0.41, respectively.
mainly due to the available space in the building block for cells
to divide. The spread diameter of the cell-laden building block
is ∼950 μm on the Petri dish surface. The highest cell density
(1 × 106 cells ml−1 ) in a single building block monitored in
our study at day 5 is ∼186 cells mm−2 .
The model (equation (1)) indicates exponential growth
and the fitting results are given in figure 3(b).
The
proliferation-related factor, a, is 0.38, 0.45, 0.41, 0.41 for four
different cases of initial number of cells per building block
7 ± 2, 16 ± 2, 26 ± 3 and 37 ± 3, respectively. Both the cell
proliferation rate and proliferation-related factor results show
that SMCs grow at a consistent rate in the printed collagen
droplets from day 0 to day 5 for the four cases of initial
number of cells per building block in this study. At day 5,
the initial number of cells of 7 ± 2, 16 ± 2, 26 ± 3 and
37 ± 3 cells/building block correspond to an average cell
density of 15, 52, 70 and 102 cells mm−2 , respectively. It was
reported in the literature that lower initial number of cells gives
a higher SMC proliferation rate and thus improves the tissue
growth and development process [48]. In our study beyond
5 days, the cells became confluent in building blocks
and lower cell proliferation rates can be expected, due
to higher cell density (>300 cells mm−2 ) in the building
blocks [24].
6
Biofabrication 2 (2010) 014105
F Xu et al
(a)
(b)
(c)
(d )
Figure 4. Long term culture of a cultured SMC-laden collagen patch. (a) Single-layer patch fabrication by overlapping individual building
blocks. (b) Patch structure at day 51. (c), (d) H&E-stained sections of bioprinted smooth tissue (c) and native rat bladder smooth muscle
tissue (d). Two challenges can be observed between (c) and (d), cell alignment and cell size. This can be due to the different
micro-environmental conditions in vitro and in vivo. The cultured patch is cultured under no external mechanical loading, while there exists
a cyclic mechanical strain in vivo.
create a 5 mm × 5 mm patch. So, the total time consumed
is ∼4 s with the control stage. The created patches were
cultured in an incubator for 51 days forming a tissue-like 3D
construct, figure 4(b). The thickness of the created patches (at
day 51) ranged between 100 and 400 μm. This was measured
by changing the microscope focus from the patch surface to
the patch bottom (Nikon TE2000). The uniformity of cell
distribution in a cultured patch has been shown in our earlier
study [24]. This is enabled by the capability of our system to
fabricate individual building blocks with repeatable number of
cells per building block, which proliferate at consistent rates
as we demonstrated here. This indicates that we can generate
building blocks which can be used for the next assembly step,
and with a high degree of freedom to culture each building
block.
Besides the cell-laden microgel, there are also other types
of building blocks, e.g., cell aggregate. Jakab et al fabricated
cellular structures of ring shape using self-assembling cell
aggregates [2], which offer the advantage of high cell
concentration and cell viability. The current technology gives
better spatial resolution in ejection direction (∼10 μm), which
offers advantages for patterning and fabricating layered tissue
structures, e.g., bladder.
3.4. Long-term culture of printed SMC patches
Single building blocks were overlapped (at 50%) to create
patches, figure 4(a). Our goal here is to show that these
building blocks can overlap seamlessly, and a larger SMC
construct can be created at high throughput and cultured for
a long term. We also performed histological comparison of
these 3D constructs with the SMC layer of a native rat bladder
to get an initial view of the morphological assessment of the
created constructs.
3.4.2. Histology comparison. We limited our studies to
morphological assessment, since we mainly focused on cell
viability, proliferation and concentration in this study. We
also isolated a native rat bladder to act as a native control.
Hematoxylin and eosin (H&E)-stained cross sections of
engineered tissue and native rat bladder tissue are shown in
figure 4(c). In our results, differences were observed between
engineered and native smooth muscle tissues (cell alignment,
cell size and micro-architecture), mainly due to the different
micro-environmental conditions in vitro and in vivo. The
3.4.1. High throughput capability. SMC-laden building
blocks (i.e. collagen droplets) were first ejected as a line
by overlapping individual building blocks. This line pattern
was repeated to create a 5 mm × 5 mm square patch. We
generated 20 building blocks per second with our system (up to
160 droplets s−1 ). It took 80 cell-laden building blocks to
7
Biofabrication 2 (2010) 014105
F Xu et al
cell sizes of the native tissue were observed to be larger.
Further, the micro-architecture of the native bladder presented
aligned SMCs, which would lead to better biomechanical
characteristics. The created SMC patch was cultured under
no external mechanical loading, while there exists a cyclic
mechanical strain in vivo. This cyclic mechanical strain
has been shown to regulate the development of engineered
smooth muscle tissues [49]. Investigation of these differences,
functional and biomechanical studies on the printed constructs
and in vivo validation experiments will be the focus of our
future studies to better mimic the native tissue.
Collagen hydrogel was used as the primary scaffold to
create the smooth muscle patch. The low mechanical strength
of the collagen gel might make in vivo suturing and handling
difficult in future clinical applications. This may be addressed
by using additional supporting base scaffolds (e.g., electrospun
fibers) that might be needed for future animal studies with
better mechanical properties.
of such biosensors [50]. Future studies will address the
functional aspects of created tissue constructs.
Acknowledgments
This work was supported by The Randolph Hearst Foundation
and Brigham and Women’s Hospital Department of Medicine
Young Investigator in Medicine Award. We would like to thank
Dr Ali Khademhosseini for discussions. UD was also partially
supported by R21 (EB007707). This work was performed
at the Bio-Acoustic MEMS in Medicine (BAMM) Labs at
the HST-BWH Center for Bioengineering, Harvard Medical
School.
References
[1] Du Y, Lo E, Ali S and Khademhosseini A 2008 Directed
assembly of cell-laden microgels for fabrication of 3D
tissue constructs Proc. Natl Acad. Sci. USA 105 9522–7
[2] Jakab K, Neagu A, Mironov V, Markwald R R and Forgacs G
2004 Engineering biological structures of prescribed shape
using self-assembling multicellular systems Proc. Natl
Acad. Sci. USA 101 2864–9
[3] Yeh J, Ling Y, Karp J M, Gantz J, Chandawarkar A, Eng G,
Blumling J IIIrd, Langer R and Khademhosseini A 2006
Micromolding of shape-controlled, harvestable cell-laden
hydrogels Biomaterials 27 5391–8
[4] McGuigan A P and Sefton M V 2007 Design and fabrication
of sub-mm-sized modules containing encapsulated cells for
modular tissue engineering Tissue Eng. 13 1069–78
[5] Ling Y, Rubin J, Deng Y, Huang C, Demirci U, Karp J M
and Khademhosseini A 2007 A cell-laden microfluidic
hydrogel Lab Chip 7 756–62
[6] Panda P, Ali S, Lo E, Chung B G, Hatton T A,
Khademhosseini A and Doyle P S 2008 Stop-flow
lithography to generate cell-laden microgel particles
Lab Chip 8 1056–61
[7] Nerem R M 1991 Cellular engineering Ann. Biomed. Eng.
19 529–45
[8] Langer R and Vacanti J P 1993 Tissue engineering Science
260 920–6
[9] Freed L E and Vunjak-Novakovic G 1997 Principles of Tissue
Engineering (Austin, TX: Landes)
[10] Bancroft G N, Sikavitsas V I, Dolder J V D, Sheffield T L,
Ambrose C G, Jansen J A and Mikos A G 2002 Fluid flow
increases mineralized matrix deposition in 3D perfusion
culture of marrow stromal osteoblasts in a dose-dependent
manner Proc. Natl Acad. Sci 99 12600–5
[11] Yang S, Leong K F, Du Z and Chua C K 2002 The design of
scaffolds for use in tissue, and engineering: part II. Rapid
prototyping techniques. Tissue Eng. 8 1–11
[12] Khademhosseini A, May M H and Sefton M V 2005
Conformal coating of mammalian cells immobilized onto
magnetically driven beads Tissue Eng. 11 1797–806
[13] Albrecht D R, Tsang V L, Sah R L and Bhatia S N 2005
Photo- and electropatterning of hydrogel-encapsulated
living cell arrays Lab Chip 5 111–8
[14] Sharma M and Jain V 1990 Microcalorimetric studies on cell
survival and repair after UV induced damage Indian J. Exp.
Biol. 28 762–5
[15] Boland T, Xu T, Damon B and Cui X 2006 Application of
inkjet printing to tissue engineering Biotechnol. J. 1 910–7
[16] Ringeisen B R, Othon C M, Barron J A, Young D and
Spargo B J 2006 Jet-based methods to print living cells
Biotechnol. J. 1 930–48
4. Conclusions
Cell-encapsulating droplet generation has potential to create
microscale cell-laden building blocks to engineer complex
3D tissue structures. This method offers high-throughput
capability to fabricate building blocks and to assemble
them as 3D constructs. The unit building block of a
3D structure is a single cell-encapsulating hydrogel droplet.
We monitored individual building blocks (collagen droplets
containing SMCs) created by the droplet method for cell
viability, proliferation and histology after long-term culture.
The results showed that (i) SMCs were encapsulated in
collagen droplets with high viability (>94%) under four
conditions (7 ± 2, 16 ± 2, 26 ± 3 and 37 ± 3 cells/building
block); (ii) after fabrication, SMCs can divide and proliferate
in building blocks, and larger initial number of cells per droplet
gives a larger increase in the number of cells per droplet, but the
proliferation rates are similar (1.49 ± 0.29); (iii) created patch
was cultured for a long term (51 days) and observed to form
a tissue-like 3D construct. These results indicated that cellladen droplet generation techniques can create building blocks
containing living cells at high throughput. These constructs
can be assembled in an automated and programmable way, and
cultured for a long term. The gelled building blocks sustain
the designed geometry of 3D tissue structures. The overall
patterning process using the building blocks can be shortened
by the high-throughput building block generation capability.
As a future vision, cell-encapsulating droplet generation
has several applications. It could serve to advance tissue
engineering by generating 3D viable tissue constructs with
similar architecture and cell densities in a high throughput
manner. This method presents a possibility of carrying out
in vitro drug testing by using similar tissue model constructs
and would decrease the load on animal experiments for drug
testing. These single building blocks with well-controlled cell
density could also be used as efficient cell-based biosensors.
Cell-based biosensors are important tools for biosecurity
applications and rapid diagnostics, where the deposition of
cells at well-defined locations is essential for the development
8
Biofabrication 2 (2010) 014105
F Xu et al
[17] Sanjana N E and Fuller S B 2004 A fast flexible ink-jet
printing method for patterning dissociated neurons in
culture J. Neurosci. Methods 136 151–63
[18] Stuhrmann B, Jahnke H G, Schmidt M, Jahn K, Betz T,
Muller K, Rothermel A, Kas J and Robitzki A A 2006
Versatile optical manipulation system for inspection, laser
processing, and isolation of individual living cells Rev. Sci.
Instrum. 77 063116
[19] Pirlo R K, Dean D M D, Knapp D R and Gao B Z 2006 Cell
deposition system based on laser guidance Biotechnol. J.
1 1007–13
[20] Demirci U 2006 Acoustic picoliter droplets for emerging
applications in semiconductor industry and biotechnology
J. Microelectromech. Syst. 15 957–66
[21] Demirci U and Toner M 2006 Direct etch method for
microfludic channel and nanoheight post-fabrication by
picoliter droplets Appl. Phys. Lett. 88 053117–9
[22] Demirci U and Montesano G 2007 Single cell epitaxy by
acoustic picolitre droplets Lab Chip 7 1139–45
[23] Demirci U and Montesano G 2007 Cell encapsulating droplet
vitrification Lab Chip 7 1428–33
[24] Moon S et al 2009 Layer by layer 3D tissue epitaxy by cell
laden hydrogel droplets Tissue Eng.
doi:10.1089/ten.tec.2009.0179
[25] Shimizu K, Sugiyama S, Aikawa M, Fukumoto Y,
Rabkin E, Libby P and Mitchell R N 2001 Host
bone-marrow cells are a source of donor intimal smoothmuscle-like cells in murine aortic transplant arteriopathy
Nature Med. 7 738–41
[26] Morikawa S, Baluk P, Kaidoh T, Haskell A, Jain R K
and McDonald D M 2002 Abnormalities in pericytes on
blood vessels and endothelial sprouts in tumors Am. J.
Pathol. 160 985–1000
[27] Levin R M, Haugaard N, O’Connor L, Buttyan R, Das A,
Dixon J S and Gosling J A 2000 Obstructive response of
human bladder to BPH vs. rabbit bladder response to partial
outlet obstruction: a direct comparison Neurourol. Urodyn.
19 609–29
[28] Golder M, Burleigh D E, Belai A, Ghali L, Ashby D,
Lunniss P J, Navsaria H A and Williams N S 2003 Smooth
muscle cholinergic denervation hypersensitivity in
diverticular disease Lancet 361 1945–51
[29] Oberpenning F, Meng J, Yoo J J and Atala A 1999
De novo reconstitution of a functional mammalian
urinary bladder by tissue engineering Nature Biotechnol.
17 149–55
[30] Ploeg M, Aben K K and Kiemeney L A 2009 The present and
future burden of urinary bladder cancer in the world World
J. Urol. 27 289–93
[31] Theoharides T C, Whitmore K, Stanford E, Moldwin R
and O’Leary M P 2008 Interstitial cystitis: bladder pain and
beyond Expert Opin. Pharmacother. 9 2979–94
[32] Ross R 1993 The pathogenesis of atherosclerosis: a
perspective for the 1990s Nature 362 801–9
[33] Lee M, Wu B M and Dunn J C 2008 Effect of scaffold
architecture and pore size on smooth muscle cell growth J.
Biomed. Mater. Res. 87 1010–6
[34] Atala A, Bauer S B, Soker S, Yoo J J and Retik A B 2006
Tissue-engineered autologous bladders for patients needing
cystoplasty Lancet 367 1241–6
[35] Chang R, Nam J and Sun W 2008 Effects of dispensing
pressure and nozzle diameter on cell survival from solid
freeform fabrication-based direct cell writing Tissue Eng.
14 41–8
[36] Kropp B P, Zhang Y, Tomasek J J, Cowan R, Furness P D
IIIrd, Vaughan M B, Parizi M and Cheng E Y 1999
Characterization of cultured bladder smooth muscle cells:
assessment of in vitro contractility J. Urol. 162 1779–84
[37] Roby T, Olsen S and Nagatomi J 2008 Effect of sustained
tension on bladder smooth muscle cells in
three-dimensional culture Ann. Biomed. Eng. 36 1744–51
[38] Saunders R E, Gough J E and Derby B 2008 Delivery of
human fibroblast cells by piezoelectric drop-on-demand
inkjet printing Biomaterials 29 193–203
[39] Barron J A, Ringeisen B R, Kim H S, Spargo B J and
Chrisey D B 2004 Application of laser printing to
mammalian cells Thin Solid Films 453/454 383–7
[40] Arumuganathar S, Irvine S, McEwan J R and Jayasinghe S N
2008 A novel direct aerodynamically assisted threading
methodology for generating biologically viable
microthreads encapsulating living primary cells J. Appl.
Polym. Sci. 107 1215–25
[41] Arumuganathar S, Irvine S, McEwan J R and Jayasinghe S N
2007 Aerodynamically assisted bio-jets: the development
of a novel and direct non-electric field driven methodology
for engineering living organisms Biomed. Mater.: Mater.
Tissue Eng. Regenerative Med. 2 158–68
[42] Patel P, Irvine S, McEwan J R and Jayasinghe S N 2008
Bio-protocols for directly forming active encapsulations
containing living primary cells Soft Matter 4 1219–29
[43] Smith C M, Christian J J, Warren W L and Williams S K 2007
Characterizing environmental factors that impact the
viability of tissue-engineered constructs fabricated by a
direct-write bioassembly tool Tissue Eng. 13 373–83
[44] Ringeisen B R, Kim H, Barron J A, Krizman D B, Chrisey D
B, Jackman S, Auyeung R Y and Spargo B J 2004 Laser
printing of pluripotent embryonal carcinoma cells Tissue
Eng. 10 483–91
[45] Saunders R, Derby B, Gough J and Reis N 2004 Ink-jet
printing of human cells Architecture and Application of
Biomaterials and Biomolecular Materials, MRS Extended
Summary ed J Y Wong, A L Plant, C E Schmidt, L Shea,
A J Coury and C S Chen (Warrendale, PA: Materials
Research Society) 95–7
[46] Roth E A, Xu T, Das M, Gregory C, Hickman J J and
Boland T 2004 Inkjet printing for high-throughput cell
patterning Biomaterials 25 3707–15
[47] Palsson B O and Bhatia S N 2003 Tissue Engineering (Upper
Saddle River, NJ: Prentice-Hall)
[48] Ross J J and Tranquillo R T 2003 ECM gene expression
correlates with in vitro tissue growth and development in
fibrin gel remodeled by neonatal smooth muscle cells
Matrix Biol. 22 477–90
[49] Kim B S, Nikolovski J, Bonadio J and Mooney D J 1999
Cyclic mechanical strain regulates the development of
engineered smooth muscle tissue Nature Biotechnol.
17 979–83
[50] Banerjee P and Bhunia A K 2009 Mammalian cell-based
biosensors for pathogens and toxins Trends Biotechnol.
27 179–88
9