Cotton wool-like poly(lactic acid)/vaterite composite scaffolds

releasing soluble silica for bone tissue engineering

Akiko Obata

Hiroki Ozasa

Toshihiro Kasuga

Julian R. Jones
Received: 7 January 2013 / Accepted: 13 April 2013 / Published online: 20 April 2013
Springer Science+Business Media New York 2013
Abstract Cotton wool-like poly(L-lactic acid) and silox-
ane-doped vaterite (SiV) composite scaffolds were pre-
pared with a modied electrospinning system for bone
tissue engineering applications. The effects of changing the
SiV content in the materials from 10 to 30 wt% on elas-
ticity and the ability to release calcium ions and soluble
silica were evaluated. The elasticity of the cotton wool-like
composites was almost the same as that of the PLLA from
the results of compressibility and recovery tests. The
materials released calcium ions for more than 56 days and
soluble silica for 2856 days in a tris buffer solution (pH
7.4). Mouse osteoblast-like cells (MC3T3-E1 cells) were
cultured on/in the cotton wool-like materials or the bre-
mats out of the same composite materials as that used for
the cotton wool-like materials. The cells penetrated into
and proliferated inside the cotton wool-like materials,
although they mainly adhered on the bremat surface.
1 Introduction
Electrospinning is often used to fabricate brous membranes
using polymer-based solution. Electrospun bremats have
been widely investigated for tissue engineering applications
because they possess exibility and a high interconnected
porosity [13]. Fibre diameter is known to inuence cell func-
tions, such as adhesion and proliferation, and is controllable
by optimizing system parameters and process parameters [4,
5]. Biodegradability is regarded to be one of the important
factors of the scaffolds for tissue engineering, withthe aimthat
regenerated tissues gradually replace the scaffold template as
it degrades in the body. Hence, electrospun biodegradable
polymers, such as poly(lactic acid), poly(lactide-co-glycolic
acid) and poly(caprolactone), have been expected to be good
candidate for the scaffold [3, 4, 68].
In bone tissue engineering, polymer-bioceramic com-
posites are good candidates for the scaffold because the bi-
oceramics component can improve mechanical properties,
and impart bioactivity, e.g. bonding with bone. The bioce-
ramics can also provide dissolution products that can stim-
ulate bone growth, such as calcium, phosphate, silicon, zinc
etc. [913]. Poly(L-lactic acid) (PLLA)/siloxane-doped
vaterite (SiV) composites, which release a trace amounts of
soluble silica and calcium ions in aqueous solution, were
developed in our previous work [1417]. The SiV particles
had a spherical shape with around 1 lm in diameter and
contained soluble silica derived from aminopropyltriethox-
ysilane (APTES). The proliferation and differentiation of
mouse osteoblast-like cells were enhanced on the composite
surface compared with them cultured on a PLLA/vaterite
composite made without APTES [14]. A bremat of the
PLLA/SiV composite was successfully prepared by the
electrospinning and evaluated in their cell and tissue com-
patibilities by in vitro and in vivo tests [15]. Mineralized
tissues were generated in the PLLA/SiV bremat implanted
in a defect (8 mm in diameter) in rabbit calvaria within
4 weeks of implantation. This generation was more rapid in
the PLLA/SiV bremat than a commercial biodegradable
membrane (PLGA). The PLLA/SiV bremat was expected
to be a good candidate for the scaffold material used in the
bone tissue engineering. The bremat morphology has lim-
itations in the establishment of a three-dimensional
A. Obata (&) H. Ozasa T. Kasuga
Graduate School of Engineering, Nagoya Institute of
Technology, Gokiso-cho, Showa-ku, Nagoya 466-8555, Japan
e-mail: obata.akiko@nitech.ac.jp
J. R. Jones
Department of Materials, Imperial College London,
London SW7 2AZ, UK
1 3
J Mater Sci: Mater Med (2013) 24:16491658
DOI 10.1007/s10856-013-4930-5
engineered tissue, because its thickness is *300 lmand has
little space to allow cells to migrate into its pores.
Fibremats are usually produced with a conventional
electrospinning system which consists of a syringe pump,
power supply and a metallic plate (collector). A polymer
solution is put in the syringe and electrically charged with a
power supply. The charged polymer solution is sprayed
onto the collector, resulting in the formation of polymer
bres. Pham et al. [8] reported that poly(e-caprolactone)
microber scaffolds with 1 mm in thickness could be
produced with a modied electrospinning system with a
ring conguration between the syringe and a collector.
They revealed that pore size and porosity of the brous
scaffold was controllable by regulating the bre diameter.
The prepared scaffolds had 90 % porosity but pore sizes
were less than 50 lm. We hypothesized that an electrospun
material with much larger pores, which has a structure like
cotton wool, could be fabricated by controlling the bre
entanglement through the electrospinning process. In
addition, pressing and heat treatments were expected to
contribute to the control of the pore size and porosity of the
cotton wool-like scaffolds.
The aim was to produce elastic PLLA/SiV brous
scaffolds with a cotton wool-like structure through a
modied electrospinning system. The elasticity and the
dissolution rate of calcium ions and soluble silica were
evaluated as a function of SiV content. The cell penetra-
tion, adhesion and proliferation in the samples were
examined by a cell culture test using mouse osteoblast-like
cells (MC3T3-E1 cells). A pressing and heat treatment
above the glass transition temperature (T
g
) of PLLA (T
g
;
57 C) was performed on the samples used for the culture
test to control their porosity and shape.
2 Materials and methods
2.1 Preparation of cotton-wool like scaffolds
SiV particles with 2 wt% of silicon were prepared with the
carbonation method reported previously [15]. 150 g of
Ca(OH)
2
and 60 ml of aminopropyltriethoxysilane (AP-
TES) were mixed in 2,000 ml of methanol with the addi-
tion of CO
2
gas for 75 min at a rate of 120 l/h, resulting in
the slurry. The slurry was dried at 110 C to get the SiV
powder. The silicon content in the prepared SiV was esti-
mated to be 2 wt% by X-ray uorescence analysis
(RIX3000, Rigaku, Japan) using a calibration curve which
was established by measuring limestone containing various
amounts of SiO
2
.
The SiV particles were mixed with PLLA (LACEA,
molecular weight of 140 kDa, Mitsui Chemicals, Japan) by
a melt-blending method using a kneader at 200 C for
10 min, resulting in the SiV/PLLA composite pellets. The
weight ratio of SiV in the composites was set to 10, 20 and
30 wt%. The resulting composites containing 10, 20 or
30 wt% were denoted by PLLA/SiV10, PLLA/SiV20 and
PLLA/SiV30, respectively. The composite pellets were
dissolved in chloroform at 10 wt% of PLLA to prepare the
solution for electrospinning. Figure 1 shows a schematic
image of the electrospinning system used in this work. The
system was modied to obtain cotton wool-like structured
composite scaffolds; air was blown using a fan (wind
speed; *1 m/s) against the electrospun bres to immedi-
ately evaporate the solvent (chloroform) contained in the
bres and to prevent the bre sticking to each other. A
styrol box with dimensions of 145 9 145 9 75 mm was
set at opposite side of the fan to collect the prepared cotton
wool-like structured scaffold. The composite solution
was loaded into a syringe pump (FP-W-100, Melquest,
Japan) with a syringe needle (22 gauge). The feeding rate
was set at 12 ml/h. A high voltage supply (Power Supply
HARb-40 P0.75, Matsusada Precision Inc., Japan) was
used to apply voltages of 20 kV to the needle tip. The
distance between the needle tip and the collector was kept
at 200 mm. The electrospinning was carried out at room
temperature.
2.2 Sample characterization
The morphology of the prepared samples was observed by
eld emission gun scanning electron microscopy (SEM)
(JSM-6301F, JEOL, Japan) with the accelerating voltage of
4 kV after coating the samples with amorphous osmium
layer using osmium coater (Neoc, Meiwafosis Co. Ltd.,
Japan).
Elasticity of the prepared samples was evaluated, fol-
lowing JIS method (JISL 1097). 0.05 g of the sample was
xed inside a glass tube (23 mm in diameter and 31 mm in
length). The height of the sample (h
1
) was measured and
then a cover glass and a weight (10 g) were placed on the
sample and remained for 30 s. The height of the sample
(h
2
) was measured. The weight was removed and then the
height of the sample (h
3
) was measured 30 s after the
removing. Compressibility (Z
1
) and recovery ratio (Z
2
)
were estimated respectively as follows:
Z
1
h
1
h
2
=h
1
f g 100 1
Z
2
h
3
h
2
= h
1
h
2
f g 100 2
Molecular weights (weight - average molecular weight:
Mw) of PLLAmatrix in the prepared samples were measured
by gel permeation chromatography (GPC) (Shimadzu,
Japan) using two columns (K-806L, Shodex, Japan). The
samples were dissolved in chloroform at a concentration of
2.0 g/l and then ltrated using a syringe lter (25CR, Pall
1650 J Mater Sci: Mater Med (2013) 24:16491658
1 3
Co., USA). The measurements were performed at 40 C at a
ow rate of 60 ml/h.
2.3 Ion release behaviour in tris buffer solution
Tris buffer solution (TBS) was prepared by mixing 6.118 g
of tris(hydroxymethyl)aminomethane and 1,000 ml of
distilled water. The pH of the solution was adjusted to 7.4
by adding an appropriate amount of 1.0 M HCl at 36.5 C.
50 mg of the cotton wool-like structured scaffold were
soaked in 10 ml of TBS and incubated at 36.5 C for
*56 days. Silicon and calcium concentrations in the TBS
after the soaking were measured by inductively coupled
plasma atomic emission spectroscopy (ICP-AES) (ICPS-
500, Shimadzu, Japan). Three samples of each sample were
tested at each time point. Molecular weights of PLLA
matrix in the samples after the soaking test were measured
by GPC. The sample pretreatment and the GPC measure-
ment were performed in the same condition as that afore-
mentioned. One sample of each sample was tested at each
time point.
2.4 Cell culture
Twenty mg of the cotton wool-like PLLA/SiV20 scaffold
were lled in a mould of 15 mm diameter and then pressed
and heated at 60 C for 5 min to set their porosity at 90 or
96 % (denoted by cotton 1 or cotton 2, respectively).
Fibremat as a comparative sample was prepared using the
same composites by the conventional electrospinning
method without the fan system. The porosity of the pre-
pared bremat was 85 %. Morphologies of the bremat
and cotton wool-like scaffolds were shown in Fig. 2. The
samples used for the cell culture tests were sterilised using
ethylene oxide gas. The samples after the sterilization were
placed in a vacuum to keep residual gas in them to a
minimum.
20 mg of the sterilised sample were placed in a 24-well
plate. A glass tube (outside diameter 15 mm and internal
diameter 13 mm) was placed on the sample to prevent the
Fig. 1 Schematic image of the
electrospinning system used in
this work (a) and an appearance
of a prepared cotton wool-like
scaffold (b). Air was blown
using a fan (wind speed;
*1 m/s) against the electrospun
bres to immediately evaporate
the solvent (chloroform)
contained in the bres and to
prevent the bre sticking to each
other. A styrol box with
dimensions of
145 9 145 9 75 mm was set at
opposite side of the fan to
collect the prepared cotton
wool-like structured scaffold
Fig. 2 SEM images of cotton wool-like scaffold (a) and bremat
(b) used for cell culture test. Bars indicate 100 lm. Fewer bres
entangled with each other and bigger spaces were formed between the
bres in the cotton wool-like scaffold compared with the bremat
J Mater Sci: Mater Med (2013) 24:16491658 1651
1 3
samples oating when a culture medium was added into
the well. The culture medium used was aMEM containing
10 % fetal bovine serum (FBS). Mouse osteoblast-like
cells (MC3T3-E1 cells) were seeded onto the samples at a
density of 50,000 cells/well and then incubated at 37 C in
a humidied atmosphere of 95 % air, 5 % CO
2
for 14 days.
The culture medium was changed after 1 day of culture and
then changed every other day. Three samples of each
sample were tested.
The cell number of MC3T3-E1 cells was measured with
a microplate reader (SUNRISE Remote, TECAN, Swit-
zerland) using Cell Counting Kit-8 (Dojindo, Japan), fol-
lowing its instructions. One reagent in the kit is a water
soluble tetrazolium salt which is reduced by dehydrogen-
ases in live cells to generate water soluble formazan. Water
soluble formazan has an absorption maximum at about
450 nm. The samples after appropriate periods were
removed to new wells, and rinsed with a esh culture
medium twice. 1 ml of esh medium was added to each
well followed with the addition of 0.1 ml of the reagent of
the kit. The samples were incubated at 37 C in a humid-
ied atmosphere of 95 % air, 5 % CO
2
for 2 h. Cell
number was counted by measuring the absorbance of the
resulting medium at 450 nm. Signicant differences
between the samples were determined by Students t test
(P\0.05). A calibration curve was established by mea-
suring the absorbance of the media containing MC3T3-E1
cells with various densities. The samples for SEM obser-
vation were xed in 2.5 % glutaraldehyde at 4 C for
40 min and then dehydrated through a series of increasing
concentrations of ethanol. The resulting samples were
dried with hexamethyldisilazane, coated with amorphous
osmium, and observed by SEM with the accelerating
voltage of 3.2 kV. The samples for uorescence micros-
copy observation were xed in 4 % paraformaldehyde
phosphate buffer solution at 4 C for 30 min and treated
with a permeabilising solution consisting of 1 % of bovine
serum albumin, 0.1 % Triton X, 98.9 % of phosphate
buffer solution (PBS) at 4 C for 25 min. Alexa Fluor
488Phalloidin (Molecular Probes, USA) which was diluted
with PBS was added to the samples well. The samples
were kept in the dark for 30 min at 37 C. The resulting
samples were observed with a uorescence microscope
(BIOREVO BZ9000, KEYENCE, Japan).
3 Results and discussion
3.1 Morphology and elasticity
SEM images in Figs. 2a and 3 show the morphologies of
the cotton wool-like scaffolds. Fewer bres entangled with
each other and bigger spaces were formed between the
bres in the cotton wool-like scaffold compared with the
bremat (Fig. 2b). This is because the air blowing during
the preparation of the cotton wool-like scaffolds with the
modied electrospinning system prevented the bre tan-
gling around the collector and the box placed between the
collector and the needle. The blowing enhanced the evap-
oration of solvent (chloroform) between the needle and the
collector, which contributed to the prevention of bre
entanglement. In general, the electrospinning process is
affected by both process parameters and system parame-
ters. System parameters include volatility and conductivity
of polymer solution [2]. The evaporation of the chloroform
in the bre caused the release of electrical charge, which is
Fig. 3 SEM images of cotton wool-like scaffolds. a PLLA/SiV10, b:
PLLA/SiV20 and c PLLA/SiV30. d, e, f cross-sectional view of bres
of each sample. Bars in (a, b, c) and (d, e, f) indicate 10 lm and
3 lm, respectively. The bres were 1020 lm in diameter. The pores
of *1 lm diameter and vaterite particles were observed on the bre
surfaces
1652 J Mater Sci: Mater Med (2013) 24:16491658
1 3
the driving force for the polymer jet traveling through the
atmosphere toward the collector from the bre jet. In
addition, the electrical charge was previously reported to
relate to a chaotic bending instability of electrospun bres.
Yarin et al. [18] suggested that repulsive interactions
between charges in the polymer jet attribute to this bending
instability. In the present work, the bre jet with a low
electrical charge was not accelerated strongly to the col-
lector and did not exhibit bending instability, which may
also attribute to the prevention of the bre entanglement.
While it was not possible to measure the size and shape of
space formed between the bres, as they were quite ran-
dom, large spaces of [100 lm were observed by SEM
(Figs. 2a and 3). The bres were 1020 lm in diameter.
The pores of *1 lm diameter and vaterite particles were
observed on the bre surfaces. The pores were formed
through the evaporation of chloroform during the the
electrospinning process [2]. The vaterite particles were
well dispersed in the bres, as they were also present
within the bres as shown in the cross-sectional images
(Fig. 3df).
Figure 4 shows the compressibility and the recovery
ratio of the cotton wool-like scaffolds. There were no
signicant differences in the compressibility among all the
samples (including the cotton wool-like PLLA sample); the
value was around 60 %. The recovery ratio of the PLLA/
SiV30 sample was lower than those of the other sample,
but there was no signicant difference in the values
between the PLLA/SiV30 and the PLLA samples. The
molecular weight of the matrix polymer was expected to
inuence the mechanical properties. Figure 5 shows the
molecular weight (Mw) of the PLLA contained in the
cotton wool-like scaffolds. Mw of the used PLLA was
around 140 kDa and decreased to 120 kDa after kneading
at 200 C for 10 min without the SiV. Mw further
decreased by mixing with the SiV and reached around
90 kDa for the PLLA/SiV30. The value decreased with the
increase in the content of the mixed SiV. The reason why
Mw decreased with the increase of the SiV content have
been described in our previous report; the scission of the
PLLA chain was induced by a shear stress generated during
the melt-blending process, which leads the formation of
coordinate bond between calcium ions on the SiV surface
and carboxyl groups in PLLA [15]. The shear stress must
increase with the increase in the content of the mixed SiV.
In addition, water adsorbed on the SiV surface may also
attribute to the Mw decrease.
3.2 Ion release in TBS
All samples maintained their cotton wool-like structure
after the soaking test for 56 days. They did not break even
after holding with tweezers. The concentrations of calcium
and silicon in TBS after soaking the samples are described
in Fig. 6. All the samples continued to release calcium ions
through 56 days. The calcium released from the samples
rapidly increased over the rst 714 days of immersion and
then the release rate decreased, as shown by a more gradual
increase until 56 days. Almost all of the Si contained in the
PLLA/SiV30 (30 ppm) and PLLA/SiV20 (20 ppm) was
released before day 28, while it took until day 56 for all the
Si to be removed from PLLA/SiV10 (10 ppm). The trend
of Si release from each sample was different to that of
calcium ion release after day 28. This is because SiV
particles were aggregates consisting of several primary
particles with the size of several hundreds nm and most of
siloxane was present on the primary particle surfaces [17].
Fig. 4 Compressibility (Z
1
) and press recovery ratio (Z
2
) of cotton
wool-like scaffolds. There were no signicant differences in the
compressibility among all the samples. The recovery ratio of the
PLLA/SiV30 sample was lower than those of the other sample, but
there was no signicant difference in the values between the PLLA/
SiV30 and the PLLA samples
Fig. 5 Changes in weight average molecular weight (Mw) of PLLA
in cotton wool-like structures as a function of SiV content. Mw
decreased with the increase in the content of the mixed SiV
J Mater Sci: Mater Med (2013) 24:16491658 1653
1 3
The ion dissolution from the cotton wool-like scaffolds was
controllable by changing the SiV content in the bre, since
both calcium and silicon element release amounts had a
correlation with the content. The released ion amount also
may be controlled by changing the content of silicon ele-
ment in the SiV, since the content is controllable by the
amount of APTES in the SiV preparation. The Si release
was in line with concentrations found to stimulate human
osteoblast activity by Xynos et al. [10]. The Si released
from PLLA/SiV composite materials have been reported to
enhance the proliferation and differentiation of MC3T3-E1
cells and mineralized tissue formation in our previous work
[14, 15], therefore the composite materials in the present
work are expected to show the same stimulatory effects on
the cells.
Figure 7 shows changes in the Mw of the PLLA con-
tained in the cotton wool-like scaffolds through the
immersion test in TBS. The value of each sample showed
almost no change during 56 days, which was expected
considering the rate of PLLA degradation is low even
in vivo [19, 20]. PLLA is hydrophobic and its solubility is
very slow in an aqueous solution. Therefore most of PLLA
matrix in the sample did not degrade or dissolve during this
soaking test. The ion release from SiV/PLLA bre was
expected to take place through the following process; (1)
the SiV particle dissolving progressed from outside to
inside the bre, (2) the interconnected pores were formed
after the SiV lost into the TBS, (3) the TBS penetrated into
inside the bre through the pores. More SiV particles were
expected to connect to each other directly in the PLLA/
SiV30 bre. Although the SiV dissolution progressed, the
PLLA matrix remained, which maintained the brous
structure for 56 days in TBS.
3.3 Cell compatibility
The PLLA/SiV20 sample was used for the cell culture test,
since it possessed almost the same elasticity as the PLLA
and contained a larger SiV content. Figure 8 shows
appearance of the scaffold samples used for the cell culture
test. The porosity and shape of the cotton wool-like scaf-
fold out of the PLLA/SiV20 was controlled with the
pressing and heat treatments. The bre surfaces are slightly
melted and connected to each other at 60 C, maintaining
the brous structure. The pressing and heat treatment
allows the cotton wool-like scaffold to maintain its porosity
and shape throughout the cell culture test. The cell mor-
phology shown in Fig. 9 demonstrated that the bre
structure inuenced cell adhesion. The cells on the bremat
Fig. 6 Calcium (a) and silicon
(b) concentrations in TBS after
soaking cotton wool-like
scaffolds. The calcium released
from the samples rapidly
increased over the rst
714 days of immersion and
then the release rate decreased,
as shown by a more gradual
increase until 56 days. Almost
all of the Si contained in the
PLLA/SiV30 (30 ppm) and
PLLA/SiV20 (20 ppm) was
released before day 28, while it
took until day 56 for all the Si to
be removed from PLLA/SiV10
(10 ppm)
Fig. 7 Mw of PLLA contained in cotton wool-like scaffolds after ion
release test in TBS. The value of each sample showed almost no
change during 56 days
Fig. 8 An appearance (from a lateral view) of three different
scaffolds used for cell culture test. Bar indicates 5 mm
1654 J Mater Sci: Mater Med (2013) 24:16491658
1 3
sample elongated multilaterally and adhered on several
bres which were adjacent to each other. On the other
hand, the cells on the cotton 1 and 2 adhered on one bre
and elongated along the bre direction, because the bres
were far from each other and junctions of the bres were
very few. This agrees with previous work by Sun et al. [21]
reported that cells bridge over two PLLA bres when the
distance of the two bres was below 200 lm and that cells
can adhere to single bres if the bre has a diameter greater
than 10 lm. Here, both the bremat and the cotton wool-
like scaffolds consisted of bres with diameters greater
than 10 lm but the distances between the bres were dif-
ferent: the cotton wool-like scaffold included many spaces
greater than 200 lm and fewer bres tangled each other
compared to the bremat.
Figure 10 shows the uorescence microscopic images
showing actin in the cells (green) stacked on a bright-eld
image. The uorescence microscopic images were
z-stacked one. These images are typical images of each
sample. Phalloidin staining is generally used for f-actin
staining to observe cytoskeletal architecture. The staining,
however, was used to establish whether the cells certainly
existed inside the samples in the present work. After
7 days-culturing, the cells proliferated well on the bremat
and spread widely. On the other hand, the cells observed
on/in the cotton wool-like scaffolds were few and were
found to proliferate along the bres, since the green area
shows a brous shape. After 14 days-culturing, the cell
number on the bremat decreased. In the case of the cotton
wool-like scaffolds, the cells kept proliferating along the
bres, since the bre shape was more clearly shown with
the green. Figure 11 shows the cross-sectional views (non
y-stacked image) of the images in Fig. 10b, d, f which were
taken at 14 days after the culturing. Most cells were present
inside the scaffolds in the case of the cotton wool-like
scaffolds, and the cells on the bremat were found at the
area close to the top surface on which they were seeded. In
the cotton 2 sample, the cells were present at deep levels
far from the top surface. Most cells were suspected to
initially attach not to the bres on the sample surface but to
those inside the scaffold. The cells penetrated through the
gaps formed between the bres close to the top surface,
when they were seeded, and then adhered and proliferated
on the bres inside the sample. Thus, the cell proliferation
mainly progressed inside only in the case of the cotton
wool-like scaffolds. The bres in the cotton wool-like
scaffolds had more three-dimensional structure than that in
the bremat, as shown in Fig. 2. The bre direction in the
cotton wool-like scaffold was signicantly random and
some bre extended from surface to inside the sample.
Spaces between the bres in the cotton-wool scaffolds may
induce the three-dimensional growth of the cells. These
phenomena were not observed in the bremat structure
where the cells adhered and proliferated mainly at the top
surface (outer edge). The cells did migrate inside the PLLA
bremat (with bres of 10 lm diameter and 70 lm gaps
between them) prepared by conventional electrospinning
method (without air blow) but the depth of the cell
migration only reached 90 lm after 13 days of culture
[22].
The cell numbers on/in each sample measured with a
colorimetric method were shown in Fig. 12. The numbers
mean the average of three samples numbers. The cell
number on the bremat increased until day 7 but decreased
between day 7 and 14. The cells formed a sheet at the outer
edge of the bremat after 7 days and the sheet was
removed through the replacement of culture medium
between day 7 and 14. This may be attributed to the larger
Fig. 9 SEM images of MC3T3-E1 cells cultured for 3 days in/on:
bremat (a), cotton 1 (b) and cotton 2 (c). Bars indicate 10 lm. The
cells on the bremat sample elongated multilaterally and adhered on
several bres, although they on the cotton 1 and 2 adhered on one
bre and elongated along the bre direction
J Mater Sci: Mater Med (2013) 24:16491658 1655
1 3
interaction between the cells forming the sheet than that
between the cell and the sample surface. PLLA/SiV -
bremats possesses excellent cell compatibility, since the
proliferation of MC3T3-E1 cell was better on the bremat
than culture plastic disc (Thermanox

) in our previous
report. On the other hand, the numbers of the cotton wool-
like scaffolds gradually increased until day 7 and showed a
marked increase between day 7 and 14. The cell numbers
in the cotton wool-like scaffolds at day 14 were larger than
that of the bremat at day 7. This indicates that the cotton
wool-like scaffold has the ability to maintain cells in 3-D.
This ability is expected to be signicant for the application
of the sample to the scaffold materials for the tissue
engineering, since cells are pre-cultured in/on scaffold
materials in vitro and then they are implanted into body
with the materials in most case of tissue engineering [23,
24]. An interconnected macroporous network with an
interconnected pore size of at least 100 lm has been
reported to allow bone ingrowth and eventually vasculari-
zation [25, 26] so the interconnected spaces formed
between the bres with over 100 lm in size in the cotton
wool-like scaffolds were expected to allow bone tissue to
grow into their inside in body.
Various types of porous bioactive materials have been
developed and reported already but there are still some
problems encountered when using them for tissue
Fig. 10 Fluorescence microscopic images showing actin in the cells
(green) stacked on a bright-eld image. a, b bremat, c, d cotton 1, e,
f cotton 2. a, c, e after 7 days of culturing, b, d, f after 14 days of
culturing. The uorescence microscopic images were z-stacked one.
After 7 days-culturing, the cells proliferated well on the bremat and
spread widely, although they on/in the cotton wool-like scaffolds
were few and proliferated along the bres. After 14 days-culturing,
the cell number on the bremat decreased, although they kept
proliferating along the bres on/in the cotton wool-like scaffolds
(Color gure online)
1656 J Mater Sci: Mater Med (2013) 24:16491658
1 3
engineering. One of the problems is the rapid formation of
tissue on the outer edge. This sometimes induces the
development of a necrotic core due to the limitations of cell
penetration and nutrient exchange [27, 28]. Electrospun
brous materials certainly have interconnected pores which
are gaps formed between bres. However, in case of a
conventional bremat cells may not penetrate into inside
the material. This is because the bres are layered densely
and the gap between the bres at outer layer and inner one
is too small to allow the cells to migrate. In fact, the
MC3T3-E1 cells proliferated at the top surface of the
bremat mainly, as shown in Figs. 10 and 11. On the other
hand, the cells penetrated into immediately and proliferated
inside the cotton wool-like scaffold. This indicates that
bone tissue can grow from inside to outside of the sample,
when engineered in vitro. The open pore structure may also
allow tissue ingrowth from the surrounding in the cotton
wool-like scaffolds after implanted in body [29].
Fibre diameter and density of brous structured scaf-
folds have been shown to affect the cell functions [8, 30,
31]. Pham et al. reported that the initial adhesion of marrow
stromal cells was enhanced on electrospun poly(e-capro-
lactone) scaffold with 5 lm in bre diameter in compari-
son with that on nanobre scaffold with 600 nm in bre
diameter and there were differences in the cell morphology
between the two scaffolds [8]. Takahashi et al. reported
that the proliferation rate of mesenchymal stem cells was
higher on the non-woven polyethylene terephthalate scaf-
fold with bigger bre diameter and higher porosity from
the culture tests using the samples with bre diameter in
range of 242 lm and porosity of about 93 or 97 %. They
also reported that the osteogenic differentiation was
enhanced on the sample with 9 lm in bre diameter and
was signicantly higher on the sample with low porosity
than that with high porosity [30]. In addition, the cell
behavior has been reported to depend on the environments
of 2D or 3D. Fibroblasts formed different focal adhesion
structures and showed higher proliferation in 3D than 2D
culture [32]. They showed an accelerated rate of acquisi-
tion of a characteristic in vivo-like spindle-shaped mor-
phology on the 3D culture. Therefore, the cell adhesion
system and morphology are regarded to inuence the fol-
lowing cell behavior, such as proliferation and differenti-
ation. These phenomena strongly related to bre diameter
and density in the case of the brous structured scaffold.
The cell morphology on the cotton wool-like scaffolds was
different from that on the bremat; a spindle shape on the
cotton wool-like scaffold and a radial one on the bremat.
In addition, the 3D environment formed in the cotton wool-
like scaffolds is expected to play an important role for
enhancing the cell functions. Thus the cell differentiation
may be different between the cotton wool-like scaffolds
and bremats. This will be claried in future study.
4 Conclusion
The cotton wool-like structured scaffold materials were
prepared using the PLLA and SiV composites with a
modied electrospinning system. The elasticity tests for the
cotton wool-like scaffolds demonstrated that they pos-
sessed almost the same elasticity as that of PLLA sample
and were not broken and collapsed even if they were held
by a tweezers. Although the amounts of calcium and sol-
uble silica released from the cotton wool-like scaffolds
changed with the SiV contents in the samples, all the
samples released calcium ions over 56 days and soluble
silica for 2856 days. The osteoblast-like cells penetrated
into and proliferated at inside the cotton wool-like scaf-
folds, while they mainly adhered on a surface of the br-
emat. The cotton wool-like scaffold was expected to be
Fig. 11 Cross-sectional views of the images in Fig. 9(b, d, f). Cells
were cultured on/in each sample for 14 days. The images were of
center parts of each sample and non y-stacked. Most cells were
present inside the scaffolds in the case of the cotton wool-like
scaffolds. The cells on the bremat were found at the area close to the
top surface on which they were seeded
Fig. 12 Cell numbers during 14 days of culture in/on bremat,
cotton 1 and cotton 2. The number of the bremat increased until day
7 but decreased between day 7 and 14, although the numbers of the
cotton wool-like scaffolds gradually increased until day 7 and showed
a marked increase between day 7 and 14. (*P\0.05)
J Mater Sci: Mater Med (2013) 24:16491658 1657
1 3
good candidate for the scaffold which achieves the three-
dimentional engineered bone tissues.
Acknowledgments This work was supported in part by Grant-in-
Aids for Young Scientists (B) (No. 21700487) from The Ministry of
Education, Culture, Sports, Science and Technology (MEXT) and
Scientic Research (B) (No. 20390497) from Japan Society for Pro-
motion of Science. JRJ acknowledges EPSRC Challenging Engi-
neering grant EP/I020861/1.
References
1. Cui W, Zhou Y, Chang J. Electrospun nanobrous materials for
tissue engineering and drug delivery. Sci Technol Adv Mater.
2010;11(1):111.
2. Sill TJ, von Recum HA. Electrospinning: applications in drug
delivery and tissue engineering. Biomaterials. 2008;29(13):
19892006.
3. Li WJ, Laurencin CT, Caterson EJ, Tuan RS, Ko FK. Electrospun
nanobrous structure: a novel scaffold for tissue engineering.
J Biomed Mater Res. 2002;60(4):61321.
4. Badami AS, Kreke MR, Thompson MS, Rife JS, Goldstein AS.
Effect of ber diameter on spreading, proliferation, and differ-
entiation of osteoblastic cells on electrospun poly(lactic acid)
substrates. Biomaterials. 2006;27(4):596606.
5. Yang F, Murugan R, Wang S, Ramakrishna S. Electrospinning of
nano/micro scale poly(L-lactic acid) aligned bers and their
potential in neural tissue engineering. Biomaterials. 2005;26(15):
260310.
6. Zeng J, Xu X, Chen X, Liang Q, Bian X, Yang L, et al. Biode-
gradable electrospun bers for drug delivery. J Control Release.
2003;92(3):22731.
7. Xin X, Hussain M, Mao JJ. Continuing differentiation of human
mesenchymal stem cells and induced chondrogenic and osteo-
genic lineages in electrospun PLGA nanober scaffold. Bioma-
terials. 2007;28(2):31625.
8. Pham QP, Sharma U, Mikos AG. Electrospun poly (e-caprolac-
tone) microber and multilayer nanober/microber scaffolds:
characterization of scaffolds and measurement of cellular inl-
tration. Biomacromolecules. 2006;7(10):2796805.
9. Hoppe A, Guldal NS, Boccaccini AR. A review of the biological
response to ionic dissolution products from bioactive glasses and
glass-ceramics. Biomaterials. 2011;32(11):275774.
10. Xynos ID, Edgar AJ, Buttery LDK, Hench LL, Polak JM. Ionic
products of bioactive glass dissolution increase proliferation of
human osteoblasts and induce insulin-like growth factor II
mRNA expression and protein synthesis. Biochem Biophys Res
Commun. 2000;276(2):4615.
11. Xynos ID, Hukkanen MVJ, Batten JJ, Buttery LD, Hench LL,
Polak JM. Bioglass

45S5 stimulates osteoblast turnover and
enhances bone formation in vitro: implications and applications
for bone tissue engineering. Calcif Tissue Int. 2000;67(4):3219.
12. Jones JR, Tsigkou O, Coates EE, Stevens MM, Polak JM, Hench
LL. Extracellular matrix formation and mineralization on a phos-
phate-free porous bioactive glass scaffold using primary human
osteoblast (HOB) cells. Biomaterials. 2007;28(9):165363.
13. Gough JE, Jones JR, Hench LL. Nodule formation and miner-
alisation of human primary osteoblasts cultured on a porous
bioactive glass scaffold. Biomaterials. 2004;25(11):203946.
14. Obata A, Tokuda S, Kasuga T. Enhanced in vitro cell activity on
silicon-doped vaterite/poly(lactic acid) composites. Acta Bio-
mater. 2009;5(1):5762.
15. Obata A, Hotta T, Wakita T, Ota Y, Kasuga T. Electrospun mi-
crober meshes of silicon-doped vaterite/poly(lactic acid) hybrid
for guided bone regeneration. Acta Biomater. 2010;6(4):124857.
16. Wakita T, Obata A, Poologasundarampillai G, Jones JR, Kasuga
T. Preparation of electrospun siloxane-poly(lactic acid)-vaterite
hybrid brous membranes for guided bone regeneration. Compos
Sci Technol. 2010;70(13):188993.
17. Wakita T, Nakamura J, Ota Y, Obata A, Kasuga T, Ban S. Effect
of preparation route on the degradation behavior and ion releas-
ability of siloxane-poly(lactic acid)-vaterite hybrid nonwoven
fabrics for guided bone regeneration. Dent Mater J. 2011;30(2):
2328.
18. Yarin AL, Koombhongse S, Reneker DH. Bending instability in
electrospinning of nanobers. J Appl Phys. 2001;89(5):301826.
19. Mainil-Varlet P, Rahn B, Gogolewski S. Long-term in vivo
degradation and bone reaction to various polylactides. 1. One-
year results. Biomaterials. 1997;18(3):25766.
20. Bos RRM, Rozema FB, Boering G, Nijenhius AJ, Pennings AJ,
Verwey AB, et al. Degradation of and tissue reaction to biode-
gradable poly(L-lactide) for use as internal xation of fractures: a
study in rats. Biomaterials. 1991;12(1):326.
21. Sun T, Norton D, McKean RJ, Haycock JW, Ryan AJ, MacNeil
S. Development of a 3D cell culture system for investigating cell
interactions with electrospun bers. Biotechnol Bioeng. 2007;
97(5):131828.
22. Fujikura K, Obata A, Kasuga T. Cellular migration to electrospun
poly(lactic acid) bermats. J Biomater Sci Polym Ed. 2012;23:
193950.
23. Rezwan K, Chen QZ, Blaker JJ, Boccaccini AR. Biodegradable
and bioactive porous polymer/inorganic composite scaffolds for
bone tissue engineering. Biomaterials. 2006;27(18):341331.
24. Ohgushi H, Kotobuki N, Funaoka H, Machida H, Hirose M,
Tanaka Y, et al. Tissue engineered ceramic articial joint
ex vivo osteogenic differentiation of patient mesenchymal cells
on total ankle joints for treatment of osteoarthritis. Biomaterials.
2005;26(22):465461.
25. Freyman TM, Yannas IV, Gibson LJ. Cellular materials as porous
scaffolds for tissue engineering. Prog Mater Sci. 2001;46(34):
27382.
26. Lu JX, Flautre B, Anselme K, Hardouin P, Gallur A, Descamps
M, et al. Role of interconnections in porous bioceramics on bone
recolonization in vitro and in vivo. J Mater Sci Mater Med.
1999;10(2):11120.
27. Silva MMCG, Cyster LA, Barry JJA, Yang XB, Oreffo ROC,
Grant DM, et al. The effect of anisotropic architecture on cell and
tissue inltration into tissue engineering scaffolds. Biomaterials.
2006;27(35):590917.
28. Ishaug SL, Crane GM, Miller MJ, Yasko AW, Yaszemski MJ,
Mikos AG. Bone formation by three-dimensional stromal osteo-
blast culture in biodegradable polymer scaffolds. J Biomed Mater
Res. 1997;36(1):1728.
29. Jones JR. Review of bioactive glass: from Hench to hybrids. Acta
Biomater. 2013;9(1):445786.
30. Takahashi Y, Tabata Y. Effect of the ber diameter and porosity
of non-woven PET fabrics on the osteogenic differentiation of
mesenchymal stem cells. J Biomater Sci Polym Ed. 2004;15(1):
4157.
31. Holtorf HL, Datta N, Jansen JA, Mikos AG. Scaffold mesh size
affects the osteoblastic differentiation of seeded marrow stromal
cells cultured in a ow perfusion bioreactor. J Biomed Mater Res.
2005;74A(2):17180.
32. Cukierman E, Pankov R, Stevens DR, Yamada KM. Taking cell-
matrix adhesions to the third dimension. Science. 2001;294(5547):
170812.
1658 J Mater Sci: Mater Med (2013) 24:16491658
1 3