Anchorage-Dependent Cell Expansion in Fiber-shaped Microcarrier Aggregates

Kazuhiro Ikeda
Inst. of Industrial Science, The University of Tokyo, Tokyo, Japan
Shoji Takeuchi
Inst. of Industrial Science, The University of Tokyo, Tokyo, Japan
International Research Center for Neurointelligence (WPI-IRCN), UTIAS, The University of Tokyo, Tokyo, Japan

DOI 10.1002/btpr.2755
Published online December 22, 2018 in Wiley Online Library (wileyonlinelibrary.com)

This article describes a three-dimensional culture system for the expansion of anchorage-
dependent cells using fiber-shaped microcarrier (MC; Cytodex3) aggregates, termed “MC
fibers.” The fiber encapsulates the cells, the MC aggregates, and collagen and is covered with
a poly-L-lysine membrane. The thin structure of the fiber enables sufficient supply of O2 and
nutrients to the cell. Using the MC fiber, we demonstrated the efficient expansion of C2C12
cells with high viability through serial passaging. Therefore, our culture system is useful for
various applications where large-scale cell expansion is required, such as in pharmaceutical
technologies, regenerative medicine, and cultured meat production. © 2018 American Institute
of Chemical Engineers Biotechnol. Prog., 35: e2755, 2019.
Keywords: three-dimensional cell culture, microcarrier, anchorage-dependent cell, myoblast

Introduction establish a retrieval method of the cells from MC fibers. Finally,

The microcarrier (MC) suspension culture system provides
anchorage-dependent cells with a floating surface for their expan-
sion and promises high cell yields due to the high surface-to-
volume ratios.1–4 Therefore, this system is used to obtain a large
number of cells required for various applications such as pharma-
ceutical applications,1 regenerative medicine,3–5 and production
of cultured meat.6,7 However, in conventional MC suspension
culture systems, uncontrolled and detrimental agglomeration of
cells and MCs is often observed, resulting in cell death by insuf-
ficient O2 and nutrition supply to the core of the aggregates.4 To
prevent excessive aggregation, agitation is often used in MC sus-
pension culture, leading to cell damage by shear stress.4,8–10
In this study, we described a method to expand cells using
fiber-shaped MC (Cytodex3, GE healthcare Life Sciences)
aggregates. We formed a hydrogel fiber encapsulating the cells,
collagen gel, and highly packed MC aggregates and then coated
the fiber with a poly-L-lysine (PLL) membrane; we named this
fiber “MC fiber” (Figure 1A). In the MC fiber, the cells adhere
onto and expand on the MC surface by three-dimensional cell
migration between the MCs. During the culture, the PLL mem-
brane maintains the fiber structure, that is, the thickness of the
MC aggregates. This microenvironment allowed us to supply
sufficient nutrition and O2 to the cells leading to efficient cell
expansion with high viability.11,12 As the fiber prevents the
excessive MC aggregation, the cells are viable even in static
condition, resulting in decreased shear stress. Here, we investi-
gate the viability of cells cultured in the MC fibers using
C2C12 cells as a model of anchorage-dependent cells. We then
we investigate the fold-change in cell expansion of C2C12
cells, the myogenesis by induction of differentiation after serial
passaging, and optimize the culture conditions by changing the
seeding density of C2C12 cells.
Results and Discussion
Encapsulation of MC aggregates and cells into MC fibers
After extrusion of the alginate and collagen solutions with
cells and MC aggregates into a CaCl2 solution followed by the
PLL coating, the fiber-shaped MC aggregates were formed as
shown in Figure 1B. To obtain enough surface area for cell
culture, MCs were packed in the structure at extremely high
density (approximately 80 mg [dry weight]/mL); the total sur-
face area of MCs in 200 μL of the pregel solution reached
almost 70–80% of a 100 mm tissue culture dish. The micro-
scopic observation also showed that MCs were encapsulated in
PLL membranes. To confirm cell adhesion and survival on
MCs during culture, we observed the viable and dead cell
morphology by calcein AM and ethidium homodimer staining
(live/dead staining) on Day 1. Without alginate lyase treatment,
stained cells maintained a rounded shape. On the other hand,
with alginate lyase treatment, the cells showed flat and elon-
gated shapes on MCs (Figure 1C and Supporting Information
S1). In addition, most of cells were viable in both of the condi-
tions. These results indicate that the removal of the alginate gel
is essential for cell adhesion and sequential expansion.

Additional supporting information may be found online in the Sup- Cell expansion and viability in MC fibers
porting Information section at the end of the article.
Correspondence concerning this article should be addressed to To confirm cell expansion in the MC fibers with high via-
S. Takeuchi at takeuchi@iis.u-tokyo.ac.jp bility, we performed MC fiber culture under static conditions,

© 2018 American Institute of Chemical Engineers 1 of 5

2 of 5
Figure 1. (A) Fabrication of MC fibers. (B) An image of the MC fibers encapsulating C2C12 cells immediately after fabrication.
(C) Fluorescence images of C2C12 cells in MC fibers by live/dead staining on Day 1. Live and dead cells are stained by calcein
AM (green) and ethidium homodimer (red), respectively. Treatment with alginate lyase allowed cell adhesion on MC surface.
MC suspension culture under static and agitation conditions
(conventional control), and then compared them for viability
and fold-change in cell expansion. We confirmed the cell via-
bility using live/dead staining of cells on Day 4 (Figure 2A).
MC suspension cultures under static conditions formed huge
and heterogeneous sized aggregates consisting of the cells and
MCs. Although the size of the aggregates in the MC suspen-
sion culture under agitation conditions was relatively smaller,
a considerable number of MCs without adhesion of cells were
found. In the MC fiber culture (cell seeding density: 1.0 × 106
cells/mL), the thickness of MC fiber was sustained. Moreover,
almost all the MCs were covered with cells. Live/dead stain-
ing showed dead cells at the core of the aggregates and on the
isolated MCs in the MC suspension culture under static and
agitation conditions, respectively. On the other hand, dead
cells in MC fiber cultures were rarely found even under agita-
tion condition (Supporting Information Figure S2A). These
results indicate that uncontrollable cell and MC agglomeration
induced the cell death in the core of the aggregates by poor
supply of nutrition and O2. Moreover, shear stress might dam-
age the cells on isolated MCs in the MC suspension culture
under agitation. The high viability of cells in MC fiber culture
is considered to be due to the restriction of the thickness of
aggregates and the low shear stress during culture. As the thin
layers consisting of PLL-alginate and collagen existing on the
MCs and attached cells could have high diffusion rates that
allow the transport of molecules with large molecular weight
(> 75 kDa), the nutrition and O2 can be sufficiently supplied
to the cellular aggregates.11,12 Furthermore, the reason that the
cells in the MC fibers were protected from shear stress under
static and agitation conditions might be the presence of PLL
Biotechnol. Prog., 2019, Vol. 35, No. 2
membrane and collagen on the MC fibers as previously
reported.13,14
Next, we calculated the fold-change in cell expansion by
counting retrieved cells from these aggregates. The results
showed that the difference of fold-change in cell expansion of
MC fiber and MC suspension cultures under static condition
was not statistically significant while that of the MC suspen-
sion culture under static condition was significantly higher
than that under agitation conditions (Figure 2B). Furthermore,
the difference of cell expansion in MC fiber culture under
static and agitation conditions was not statistically significant,
suggesting that the MC fiber culture can protect cells from
shear stress (Supporting InformationFigure S2B). These results
demonstrated efficient cell expansion with high viability in
MC fiber cultures. The thin structure of cells and MC aggre-
gates after culturing might provide cells with sufficient expo-
sure to nutrients and O2 to prevent cell death in the core.
Lower cell attachment and fold-change in cell expansion in
MC suspension cultures under agitation conditions compared
to those under other cultures might be caused by shear stress
during culturing. In large-scale cultures, MC suspension cul-
tures under static and agitation conditions might cause cell
loss and unwanted differentiation; however, MC fiber culture
did not form detrimental agglomeration without agitation.
These results indicate that MC fiber culture would be suitable
for large-scale expansion of anchorage-dependent cells.
Optimization of MC culture for C2C12 cells
Initial cell density is known to be an important factor for the
expansion efficiency of myoblasts on MCs.7,8 To achieve highly
Biotechnol. Prog., 2019, Vol. 35, No. 2
Figure 2. (A) Phase contrast and fluorescence images of the three types of MC cultures (MC fiber under static condition (MC fiber), MC
suspension cultures under static (Static MC), and agitation (Agitation MC) conditions on Day 1 and Day 4. The agitation cul-
ture was started from Day 1 using the cells in the static culture. Viable cells and dead cells were stained by calcein AM (green)
and ethidium homodimer (red), respectively. (B) Fold-change in cell expansion of the three types of MC cultures on Day 4
(N = 3). *P < 0.05.
efficient expansion and maintenance of the differentiation poten-
tial through serial passaging using the MC fiber culture, we opti-
mized the cell seeding density by culturing cells seeded on MC
fibers using three different seeding densities (0.5 × 106,
1 × 106, and 2 × 106 cells/mL). To compare these fibers, we
calculated their fold-changes in cell expansion until three pas-
sages (total 12 days; passaging every 4 days), and then evalu-
ated their differentiation potential. Microscopic observation
showed that C2C12 cells expanded in all three types of MC
fibers (Figure 3A). In the 0.5 × 106 cells/mL fiber, considerable
numbers of MCs were not coated by the cells, while the cells in
the 2 × 106 cells/mL fiber were over-confluent on the MC sur-
face. Next, the cell expansion in three types of MC fibers was
calculated until three passages (Figure 3B). The results showed
that the expansion was maintained at a high level in only the
1 × 106 cells/mL fiber. Moreover, the estimated accumulative
fold-change in cell expansion in the 1 × 106 cells/mL fiber was
the highest value among the three types of MC fibers until three
passages (437, 2217, and 696-fold expansion, in the 0.5 × 106,
1 × 106, and 2 × 106 cells/mL fiber, respectively). Finally,
retrieved cells from passage 3 fibers were differentiated on a
two-dimensional (2D) surface to evaluate the maintenance of
differentiation potential by immunostaining. The nuclei and
myotubes were stained by Hoechst 33342 and myosin heavy
chain (MF20) antibody, respectively (Figure 3C). As all three
types of MC fiber-derived C2C12 cells showed myogenesis, we
calculated the fusion indices (percentage of nuclei inside the
MF20 [+] myotubes) from three independent experiments
(Figure 3D). The results indicate that the cells in the 1.0 × 106
cells/mL fiber had the highest efficiency of myogenesin among
these three types of MC fiber cultures. Based on these results,
we conclude that the optimal cell density is 1.0 × 106 cells/mL
and optimal passaging frequency is every 4 days for efficient
expansion and maintenance of differentiation potential of
C2C12 cells. This optimization process is essential for MC fiber
culture to expand stem cells and somatic cells while sustaining
the expansion and the characteristics of the cells.
We have developed an MC fiber culture for efficient
expansion of anchorage-dependent cells with high viability.
3 of 5
Our culture system maintained the thickness of the cell and
MC aggregates, thereby improving the cell viability compared
with the conventional MC suspension culture systems. Fur-
thermore, optimization of cell seeding density leads to the
maintenance of the expansion and the differentiation potential
through serial passaging. These results suggest that our culture
system is useful for large-scale culture of other types of
anchorage-dependent cells, such as mesenchymal stem cells
and primary skeletal muscle cells.
Experimental Section
Maintenance culture of C2C12 cells
C2C12 cells were obtained from American Type Culture
Collection, Rockville, Md. The cells were maintained in culture
medium (Dulbecco’s modified Eagle’s medium [DMEM;
1195092, Thermo Fisher Scientific, Waltham, MA] containing
10% fetal bovine serum [BioWest, Miami, FL]) under 5%
CO2 at 37 C. For passaging on a 2D surface, the cells were
washed twice in phosphate-buffered saline (PBS [−]) (Sigma-
Aldrich, St. Louis, MO), incubated with 0.25 w/v % trypsin–
EDTA for 5 min at 37 C and then dissociated into single cells
by pipetting. The detached cells suspended in the culture
medium were centrifuged at 400g for 3 min at room tempera-
ture. The cells were seeded into culture dishes with culture
medium at 2.0 × 105 cells/100 mm culture dish. The medium
was replaced on Day 2. Passaging was performed approxi-
mately every 4 days. In all experiment, we used the cells before
reaching 10 passages. The viable cells were counted using try-
pan blue (Life Technologies, Carlsbad, CA) exclusion assay.
MC suspension culture
We seeded C2C12 cells in ultra-low attachment six-well
plates with culture medium (2.0 × 105 cells/well). Approxi-
mately 16 mg (dry weight) MCs were added in the well. In
static culture, the cells were cultured without agitation for
4 of 5 Biotechnol. Prog., 2019, Vol. 35, No. 2

Figure 3. (A) Images of MC fiber cultures on Day 0 and 4 in the three different types of MC fiber cultures (cell seeding density:
5.0 × 105, 1.0 × 106, and 2.0 × 106 cells/mL in the fiber). (B) Fold-change in cell expansion of the three types of MC fiber cul-
ture. (N = 3) (C) Phase contrast and fluorescence images of differentiated C2C12 cells after passage 3 in the three types of MC
fiber cultures. Myotubes and nuclei were stained by MF20 (green) and Hoechst 33342 (blue). (D) Fusion indices of the three
different types of MC fiber cultures (N = 3). *P < 0.05.

4 days. In the agitation culture, the six-well plates were agi-
tated using an orbital shaker at approximately 100 rpm from
Day 1. Half the medium was changed on Day 2 and Day 3 in
both the MC suspension cultures.
medium supplemented or not with alginate lyase (Sigma-
Aldrich) at 40 μg/mL and incubated for 5 min at 37 C. After
the treatment, MC fibers were floated in fresh medium and
cultured in six-well plates or culture dishes for 4 days. The
half medium was changed on Day 2 and Day 3.

MC fiber culture
To fabricate MC fibers, first, we prepared MC suspensions
in collagen and alginate solutions. Collagen-coated MC
(Cytodex3 [GE healthcare Life Sciences, Velizy-Villacoublay,
France]) was swollen and sterilized by following the instruc-
tions for use. For preparation of MC pellets, MC suspensions
in culture medium were centrifuged at 400g for 3 min at room
temperature. The pellets were washed twice by 3 mg/mL col-
lagen (AteloCell®, IAC-50, KOKEN, Tokyo, Japan) and
1 w/v % alginate (I-1G, KIMICA, Tokyo, Japan) solution and
centrifuged at 400g for 3 min. Next, we mixed the MC pellets
(approximately 80 mg [dry weight]/mL) with C2C12 cells on
ice. Then, the cell and MC suspension was extruded into a
100 mM CaCl2 (Kanto Chemical, Tokyo, Japan) bath using a
syringe (Terumo syringe® 1 mL, Terumo, Tokyo, Japan) and
a blunt needle tip (LS22, Instech Solomon, Plymouth Meeting,
PA). The CaCl2 solution was changed to culture medium and
incubated for 5 min at 37 C. Then, the fiber-shaped MC
aggregates were coated with the PLL solution for 5 min at
37 C. Finally, the PLL solution was changed to culture
Retrieval of cells from the MC fibers and serial passaging
After culturing, MC fibers were mechanically dispersed by
rough pipetting in order to break the PLL membrane. The sus-
pension was centrifuged at 400g for 3 min and washed by
PBS (−). Trypsin-EDTA treatment for 5 min at 37 C allowed
cell detachment from the MC surface. Then MCs were
removed by filtering with a 70 μm cell strainer (Falcon, NJ).
Harvested viable cells were also counted by the trypan blue
exclusion assay. For serial passaging, the harvested cells were
mixed with MC suspensions in collagen and alginate solu-
tions. Fabrication process was the same as above.
Viability assay
Viability assay was performed on MC fibers after culture
using a Live/Dead Viability/Cytotoxicity kit (Molecular
Probes, Eugene, OR). We stained the cells with 0.1 v/v % cal-
cein AM and 0.2 v/v % ethidium homodimer solutions and
imaged them by phase-contrast and fluorescence microscopy.
Biotechnol. Prog., 2019, Vol. 35, No. 2
Differentiation of C2C12 cells
Harvested C2C12 cells were reseeded on collagen type I
coated six-well plates at 2.0 × 105 cells/well. After 3 days of
growth in the culture medium, the cells were cultured in the
differentiation medium (DMEM containing 2 v/v % horse
serum [26050070, Thermo Fisher Scientific]) for 10 days.
Immunostaining
Samples were washed with PBS (−), fixed with 4 w/v %
paraformaldehyde solution (Wako Pure Chemical Industries),
permeabilized with 1 w/v % Triton X-100 (Sigma-Aldrich) in
PBS for 20 min, stained with MF20 antibody (14-6503
(MF20), Thermo Fisher Scientific) for 30 min, and then
stained with Alexa 488 secondary antibody (A-11001, Thermo
Fisher Scientific) for 30 min. This immunostaining was done
at room temperature. The cells were imaged and counted by
phase-contrast and fluorescence microscopy.
Scanning electron microscopy imaging
MC fibers were washed with PBS (−), fixed with 4 w/v %
paraformaldehyde solution for 30 min at room temperature.
The samples were then imaged by HITACHI miniscope
TM3030.
Statistical processing
We used one way ANOVA were used for statistical analysis
in Figures 2B and 3D. Also, Bonferron’s post hoc test was
used to determine the difference between groups. We used
Student’s t-test in the Supporting Information Figure S2. All
calculation was performed by Microsoft Excel.
Acknowledgments
The authors would like to acknowledge Jun Sawayama for
helping with the SEM imaging. Also, the authors are grateful
to Mai Furuhashi for useful discussion. This work was par-
tially supported by Grant-in-Aid for Scientific Research(S)
(Grant number: 16H06329) from Japan Society for the Promo-
tion of Science.
5 of 5
Literature Cited
1. Chu L, Robinson DK. Industrial choices for protein production
by large-scale cell culture. Curr Opin Biotechnol. 2001;12(2):
180–187.
2. Merten OW. Advances in cell culture: anchorage dependence.
Philos Trans R Soc Lond B Biol Sci. 2015;370(1661):
20140040.
3. Oh SK, Chen AK, Mok Y, Chen X, Lim UM, Chin A, Choo AB,
Reuveny S. Long-term microcarrier suspension cultures of human
embryonic stem cells. Stem Cell Res. 2009;2(3):219–230.
4. Serra M, Brito C, Correia C, Alves PM. Process engineering of
human pluripotent stem cells for clinical application. Trends Bio-
technol. 2012;30(6):350–359.
5. Rafiq QA, Coopman K, Nienow AW, Hewitt CJ. Systematic
microcarrier screening and agitated culture conditions improves
human mesenchymal stem cell yield in bioreactors. Biotechnol J.
2016;11(4):473–486.
6. Bhat ZF, Fayaz H. Prospectus of cultured meat—advancing meat
alternatives. J Food Sci Technol. 2011;48(2):125–140.
7. Verbruggen S, Luining D, van Essen A, Post MJ. Bovine myo-
blast cell production in a microcarriers-based system. Cytotech-
nology. 2018;70(2):503–512.
8. Boudreault P, Tremblay JP, Pépin MF, Garnier A. Scale-up of a
myoblast culture process. J Biotechnol. 2001;91(1):63–74.
9. Leung HW, Chen A, Choo AB, Reuveny S, Oh SK. Agitation
can induce differentiation of human pluripotent stem cells in
microcarrier cultures. Tissue Eng Part C Methods. 2011;17(2):
165–172.
10. Cherry RS, Papoutsakis ET. Physical mechanisms of cell damage
in microcarrier cell culture bioreactors. Biotechnol Bioeng. 1988;
32(8):1001–1014.
11. Darrabie MD, Kendall WF Jr, Opara EC. Characteristics of poly-
L-ornithine-coated alginate microcapsules. Biomaterials. 2005;
26(34):6846–6852.
12. Leung BM, Moraes C, Cavnar SP, Luker KE, Luker GD,
Takayama S. Microscale 3D collagen cell culture assays in con-
ventional flat-bottom 384-well plates. J Lab Autom. 2015;20(2):
138–145.
13. Horiguchi I, Chowdhury MM, Sakai Y, Tabata Y. Proliferation,
morphology, and pluripotency of mouse induced pluripotent stem
cells in three different types of alginate beads for mass produc-
tion. Biotechnol Prog. 2014;30(4):896–904.
14. Rebelo SP, Costa R, Silva MM, Marcelino P, Brito C, Alves PM.
Three-dimensional co-culture of human hepatocytes and mesen-
chymal stem cells: improved functionality in long-term bioreactor
cultures. J Tissue Eng Regen Med. 2017;11(7):2034–2045.
Manuscript received May 26, 2018, revision received Nov. 18, 2018,
accepted Nov. 20, 2018.