Biomaterials 30 (2009) 780–788

Contents lists available at ScienceDirect

Biomaterials
journal homepage: www.elsevier.com/locate/biomaterials

Biodegradable poly(3-caprolactone) nanowires for bone tissue engineering

applications
Joshua R. Porter, Andrew Henson, Ketul C. Popat*
Department of Mechanical Engineering, School of Biomedical Engineering, Colorado State University, 1374 Campus Delivery, Fort Collins, CO 80523, USA

a r t i c l e i n f o a b s t r a c t

Article history: Critical-sized defects in bone, whether caused by cancer tumor resection, trauma, or selective surgery
Received 6 August 2008 have in many cases presented insurmountable challenges to the current gold-standard treatment for
Accepted 16 October 2008 bone repair. The primary purpose of a tissue-engineered scaffold is to incite and promote the natural
Available online 14 November 2008
healing process of bone, which does not occur in critical-sized defects. In this work, a solvent-free
template synthesis technique was utilized to fabricate uniform arrays of substrate-bound poly(3-cap-
Keywords:
rolactone) (PCL) nanowires. Biodegradation of PCL nanowire surfaces was characterized using scanning
Nanotopography
electron microscopy (SEM) and matrix-assisted laser desorption/ionization time of flight (MALDI-TOF)
Bone repair
Bone tissue engineering mass spectrometry. Rat bone marrow-derived mesenchymal stem cells (MSCs) were employed to assess
Osseointegration short-term biocompatibility and long-term bioactivity of nanowire surfaces. Short-term cell studies
Mesenchymal stem cell indicated that PCL nanowire surfaces supported enhanced cell adhesion and viability compared with
Polycaprolactone control surfaces. MSCs seeded on nanowire surfaces also displayed increased levels of alkaline phos-
phatase (ALP) after 1, 2, and 3 weeks in culture. Calcium-phosphate mineralization was substantially
accelerated on nanowire surfaces compared to control surfaces as indicated through calcium staining,
von Kossa staining, SEM, and electron dispersive spectroscopy (EDS). Increased levels of inter- and
extracellular levels of osteocalcin and osteopontin were observed on nanowire surfaces using immu-
nofluorescence techniques after 3 weeks of culture. Considering the simplicity of the presented fabri-
cation technique, capacity for solvent-free encapsulation of bioactive molecules or particles, and
enhanced MSC performance on nanowire surfaces, this work presents an excellent foundation for the
development of 3-D scaffolds for bone tissue regeneration.
Published by Elsevier Ltd.

1. Introduction constituents, depend intimately on its micro- and nanoscale

Autogenous cancellous bone is currently the most widely used
bone graft material [1]. However, there are several problems
associated with this procedure such as donor site morbidity, pain,
prolonged rehabilitation, increased risk of deep infection, inflam-
mation, restricted availability, and additional scar tissue formation
[2–4]. These problems have motivated the design of synthetic bone
scaffolds as a replacement for autogenous cancellous bone grafts.
Synthetic bone scaffolds aim to mimic the physiochemical envi-
ronment in native bone through micro- and nanoscale physical
manipulations and encapsulation with signaling molecules such as
growth factors. Bone progenitor cell and osteoblast functionality is
heavily regulated by surface micro- and nanoscale architectures in
vitro and in vivo which is not surprising considering the properties
and function of bone, a complex composite of organic and inorganic
* Corresponding author. Tel.: þ1 970 491 1468.
E-mail address: ketul.popat@colostate.edu (K.C. Popat).
structure [5–8]. In attempts to mimic key chemical cues in bone
physiology, bioactive molecules such as growth, transcription, and
adhesion factors have been incorporated into synthetic bone scaf-
folds and demonstrated enhanced osteoconduction and osteoin-
duction [9–12]. In addition to mimicking the physiochemical
physiologic environment of bone, tissue engineers are faced with
several practical design considerations such as controlled biodeg-
radation rate to promote native tissue ingrowth, viability preser-
vation for bioactive molecules, removal of potentially cytotoxic
processing solutions, and sterilization [13,14].
In this work, we present a solvent-free template synthesis
technique for fabricating controlled arrays of high aspect ratio,
substrate-bound nanowires from poly(3-caprolactone), a biocom-
patible and biodegradable polymer. Template synthesis is a simple
procedure which provides a controlled approach for developing
nanoscale polymer constructs for tissue engineering applications.
Capitalizing on its low biodegradation rate and low melting
temperature, PCL scaffolds can be fabricated and impregnated with
bioactive molecules in solvent-free conditions avoiding toxicity

0142-9612/$ – see front matter Published by Elsevier Ltd.

doi:10.1016/j.biomaterials.2008.10.022
 J.R. Porter et al. / Biomaterials 30 (2009) 780–788
associated with organic solvents typically utilized in polymer drug
delivery systems [15–17]. Biodegradation and bioactive molecule or
drug release profiles may be optimized through simple parametric
changes in the nanowires. Further, template synthesis of nanowire
surfaces can be readily combined with existing microscale fabri-
cation technologies making it an ideal candidate for fabrication of
3-D scaffolds. In this work, nanowire fabrication and biodegrada-
tion, short- and long-term MSC response, and bioactive molecule
encapsulation are reported.
2. Materials and methods
2.1. PCL nanowire fabrication
Nanowires were fabricated via template synthesis from Poly(3-caprolactone)
(PCL) (MW: 37 kDa, Sigma), using 13 mm WhatmanÒ AnoporeÔ inorganic
aluminum oxide membranes (Fig. 1). In brief, PCL was placed flat on the surface of
the membrane at 115  C for 3 min while it extruded into the porous membrane. The
alumina membranes were then dissolved in 1 N NaOH for 75 min. Nanowire
templates were then soaked twice and rinsed profusely with DI water, dried under
vacuum, and stored in a desiccator until further analysis. Scanning electron
microscopy (SEM) and electron dispersive spectroscopy (EDS) were utilized to
confirm the absence of alumina or NaOH residues on nanowire surfaces.
2.2. PCL nanowire degradation
Biodegradation of control and nanowire surfaces was studied in PBS alone and
PBS with lipase (from Pseudomonas cepacia (50 Units/mg), Sigma). Due to the very
low nanowire to bulk mass ratio, standard degradation techniques such as gas
permeation chromatography (GPC) or gravimetric analysis could not be used to
quantify degradation. The two techniques employed in this study to characterize PCL
nanowire degradation were: SEM visual analysis and matrix-assisted laser desorp-
tion/ionization time of flight (MALDI-TOF) mass spectrometry. Briefly, control and
nanowire surfaces were soaked in 1 ml PBS with and without lipase. The concen-
tration of lipase was 10k Units/l. Samples were incubated at physiological conditions
(37  C, 5% CO2) for 1 week. They were extracted from the degradation media at days
1, 4, and 7 and rinsed with DI water, dried, and stored in a desiccator until exami-
nation with SEM. The media in which the samples were soaked was collected for
MALDI-TOF analysis. Due to the extremely low concentration (nmol to mmol range)
and high molecular weight of the degradation products, MALDI-TOF mass spec-
trometry was selected to obtain a mass spectrum of the degradation products in the
soaking media.
2.3. Mesenchymal stem cell isolation and culture
MSCs were isolated from Wistar rats (Rattus norvegicus) supplied by Harlan
Sprague Dawley, Inc. Limbs were aseptically removed from recently euthanized
animals. Soft tissue was removed and bones were briefly stored in cold PBS. Meta-
physeal ends were removed to expose the bone marrow cavity. In a 50-ml conical
tube, marrow was repeatedly flushed with culture media (a-MEM with 10% fetal
bovine serum (FBS, Sigma) and 1% penicillin/streptomycin (pen/strep, Sigma)) using
10 ml syringes with 18 and 25 gauge needles. Media containing cells and debris
was filtered with a 70-mm nylon filter into a clean tube. Cells were counted using
a hemocytometer before seeding. Control (smooth) and nanowire PCL test surfaces
were sterilized by exposing them to UV light for 15 min followed by soaking in 70%
ethanol for 30 min. The surfaces were then washed twice with warm PBS followed
by warm culture media prior to MSC seeding. Cells were seeded on control and
nanowire surfaces (surface area: 0.7 cm2) in a 24-well plate at a density of 2 million/
well (w1E6/cm2). Cells were cultured in the same media described above. Cultures
were incubated at 37.0  C and 5% CO2 for the duration of the study. Half of the media
was changed at day 4. On day 7, all the media was replaced with an osteogenic
differentiation media consisting of a-MEM supplemented with 10% FBS, 1% Penn/
Strep, Dexamethasone (108 M), ascorbic cid (50 mg/ml), and b-glycerolphosphate
(8 mmol). Media was changed every 2 days up to 3 weeks. MSC response was
investigated in two phases: a) Cell adhesion, proliferation, and viability up to 7 days
after initial culture, and b) cell osteogenic differentiation and matrix production for
up to 3 weeks after differentiation media was supplied.
Fig. 1. Schematic of PCL nanowire fabrication. Sintered PCL puck is placed on nanoporous alumina membrane (A). At moderately elevated temperatures PCL nanowires are
gravimetrically extruded. (B, C). The alumina membrane is dissolved in NaOH (D) leaving a nanowire surface (E).
781
Tissue culture polystyrene (PS) and smooth PCL (PCL) surfaces were both used as
control surfaces. All test and control surfaces were cultured and assayed in triplicate
at each time point specified. Short-term studies were conducted in triplicate and
long-term studies were conducted in duplicate using different animals as the MSC
source for each study.
2.4. Short-term MSC response to nanowire surfaces
After 1, 4, and 6 days of culture, cell responses to the surfaces were investigated
through cell adhesion, viability (mitochondrial activity), and morphology. A stan-
dard Live/Dead assay (Invitrogen) was used to image cell survival, adhesion, and
spatial organization. Samples were removed from culture media, rinsed in PBS and
incubated in calcein-AM (2 mM in PBS) and ethidium homodimer-1 (4 mM in PBS) for
45 min. Cells with compromised membranes exhibit red-fluorescence from the live-
cell impermeant nucleic acid stain ethidium homodimer-1. Cells with intact cell
membranes are able to use nonspecific cytosolic esterases to convert nonfluorescent
calcein-AM into bright green-fluorescent calcein. Surfaces were rinsed in PBS and
then viewed using appropriate filters with a Zeiss Axioplan 2 fluorescence micro-
scope (Carl Zeiss).
Cell viability was measured after 1 and 4 days of culture (log phase growth)
using a commercially available MTT assay kit (Sigma). Adhered cells were incubated
at 37  C for 3 h in a (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide
(MTT) solution. Mitochondrial dehydrogenases of viable cells cleave the tetrazolium
ring, yielding purple formazan crystals. Formazan crystals were then dissolved in
the MTT solvent. The optical density (OD) of the solvent is proportional to the
mitochondrial activity of the cells on the surface. OD was measured at 570 nm using
a spectrophotometer (FLUOstar Omega; BMG Labtech, Durham, NC). Background
absorbance at 690 nm was subtracted from the measured absorbance.
To view the morphology of adhered MSCs on control and nanowire surfaces, the
cells were fixed, dehydrated, and viewed using scanning electron microscopy (SEM)
after 1, 4, and 6 days of culture to determine short-term morphological changes; and
after 1, 2, and 3 weeks of differentiation to determine long-term cell morphology
and extracellular matrix production. Briefly, the cells were fixed in a solution of 3%
glutaraldehyde (Sigma), 0.1 M sodium cacodylate (Polysciences, Warrington, PA), and
0.1 M sucrose (Sigma) for 45 min. The surfaces were then soaked in buffer containing
0.1 M sodium cacodylate and 0.1 M sucrose. The cells were dehydrated by soaking the
surfaces in increasing concentrations of ethanol (35%, 50%, 70%, 95%, 100%) for
10 min each. Cells were dehydrated further by soaking the surface in hexame-
thyldisilazane (HMDS, Sigma) for 10 min. The surfaces were then dried under
vacuum and stored in a desiccator until examined with SEM. They were sputter
coated with 10 nm of gold and imaged using a JEOL JSM 6500F at voltages ranging
from 10 to 15 kV.
2.5. MSC long-term response to nanowire surfaces
MSC responses to the nanowire and control surfaces were investigated 1, 2, and
3 weeks after providing the cells with osteogenic differentiation media. Alkaline
phosphatase (ALP) activity, calcium and phosphate deposition, and extracellular
matrix protein production were used to assess the osteoinductivity and osseointe-
grativity of the nanowire surfaces. Nanowire and control surfaces were removed
from the culture media and rinsed twice in PBS prior to analysis.
Cytoplasmic ALP was measured at prescribed time points. Adhered cells were
incubated and shaken in a standard mammalian cell lysis reagent (CelLyticÔ M,
Sigma) for 15 min at room temperature. A commercially available ALP colorimetric
assay kit (QuanticromeÔ, BioAssay Systems) was used to quantify ALP concentration
in the cell lysate. Briefly, ALP catalyzes the reaction of p-Nitrophenolphosphate
(p-NPP) to p-Nitrophenol and phosphate. p-Nitrophenol is measured using a plate
reader (yellow, 405 nm) at 1 min and 5 min in order to determine the concentration
of ALP in the lysate.
The intra- and extracellular calcium was detected using a standard alizarin red
staining procedure. Test and control surfaces were rinsed twice in PBS followed by
a rinse in cold (4  C) ringer solution (Sigma). Adhered cells were fixed in 3.7%
paraformaldehyde in cold PBS for 10 min, and then rinsed in cold DI water. To stain
calcium on the surfaces, a solution of 2% alizarin red (Sigma) in 10% ammonium
hydroxide was used. Calcium forms an alizarin red S-calcium complex in a chelation
process, and the end product is birefringent. Digital images of stained surfaces were
captured using a Canon PowerShot SD1000.
Intra- and extracellular calcium phosphate (PO 4 ) was detected using a standard
von Kossa staining procedure. This is a standard procedure typically utilized for the
detection of calcium-phosphate mineralization on different biomaterials. Samples
782 J.R. Porter et al. / Biomaterials 30 (2009) 780–788

Fig. 2. PCL nanowire surfaces imaged using SEM at 750 (A), 1700 (B), 19,000 (C) and 120,000 (D).

were rinsed with a cacodylic buffer (1 M cacodylic acid and 1 M NaOH in DI water)
and then fixed in 2% paraformaldehyde in cacodylic buffer for 10 min at room
temperature. Samples were then rinsed twice with DI water followed by exposure to
a silver nitrate solution. The silver cation binds with anions such as phosphates or
carbonates and turns a dark brown or black color. Digital images were taken using
a Canon PowerShot SD1000.
Samples prepared for SEM were also examined for surface elemental compo-
sition using an attached EDS probe (Thermo Electron, Noran system) to the JEOL JSM
6500F. EDS was used to detect mineralization (calcium and phosphorus) on the
samples. Surfaces were analyzed for 5 min at 5–15 kV and a magnification of 100–
5000 to provide a complete profile of different elements present. Calcium to
phosphorus ratios were computed to speculate on the type of calcium phosphate
present.
After 3 weeks of culture, MSCs on different surfaces were immuno-labeled for
osteocalcin and osteopontin. Cells were fixed and permeabilized as described earlier.
Samples were incubated with 10% donkey serum (Santa Cruz Biotechnology) in PBS
for 20 min at room temperature to prevent nonspecific binding. After rinsing in PBS,
the samples were incubated with either anti-osteocalcin primary antibody (1:100 in
PBS, V-19 purified goat polyclonal antibody of human origin, Santa Cruz Biotech-
nology) or anti-osteopontin primary antibody (P-18 purified goat polyclonal
antibody of mouse origin, Santa Cruz Biotechnology) for 1 h at room temperature.
Following primary antibody incubation, samples were washed three times with PBS
at an interval of 5 min each. Samples were then incubated in FITC-labeled secondary
antibodies for osteocalcin and osteopontin (1:100 donkey antigoat IgG, Santa Cruz
Biotechnology) for 45 min in the dark. Finally, the samples were rinsed in PBS and
imaged using appropriate filters with a Zeiss Axioplan 2 fluorescence microscope
(Carl Zeiss).
2.6. Hydroxyapatite (HAp) nanosphere encapsulation
PCL nanowires were encapsulated with HAp nanospheres using a solvent-free
technique. Briefly, 1.0 wt% HAp powder (Ø < 200 nm, Sigma) was mechanically
mixed with molten PCL on a hot plate above the melting temperature of PCL. After
samples were cooled, PCL-HAp composite nanowires were extruded as described
earlier.
3. Results and discussion
3.1. PCL nanowire fabrication
Research on polymer-based tissue engineering scaffolds has
revealed that the use of organic solvents during fabrication or
bioactive molecule encapsulation may have cytotoxic effects.
Therefore, in this work, we have developed a solvent-free nano-
templating technique to fabricate PCL scaffolds for tissue engi-
neering applications. The melt-extrusion process described in this
study was simple and consistently produced uniform nanowire
surfaces. Prior to cell culture, 24 surfaces were fabricated and
examined under SEM to determine repeatability and uniformity.
Fig. 2 displays representative SEM images of PCL nanowires
fabricated using nanotemplating technique. SEM images revealed
presence of regular microchannels on the surface, which is likely
due to the surface tension interaction during membrane disso-
lution. Further, EDS revealed that no residual alumina membrane
or NaOH was present on the nanowire surfaces (data not shown).
3.2. Nanowire biodegradation characterization
Understanding and characterizing the biodegradation of
a polymer tissue engineering scaffold is critical in determining its
clinical utility. A scaffold should biodegrade (or bioerrode) at a rate
comparable to tissue generation incited by the scaffold. Addition-
ally, if a scaffold is designed to encapsulate and release drugs or
bioactive molecules, the degradation must also be well character-
ized. Due to the inherent slow degradation rate, biocompatibility,

x 104
4.5
Nanowires w/Lipase
4 Control w/Lipase
3.5 Nanowires in PBS
Control in PBS
3
Intensity (a.u.)

2.5

1.5
1

0.5

0
500 1000 1500 2000 2500 3000 3500 4000 4500 5000
m/z
Fig. 3. MALDI-TOF mass spectra. PCL nanowire surfaces incubated for 7 days in PBS alone (A), and PBS with lipase (B).
 J.R. Porter et al. / Biomaterials 30 (2009) 780–788
Fig. 4. Fluorescence microscopy images of live cells on smooth PCL (A) and nanowire PCL (B) surfaces after 6 days of culture. Cell aggregation on the nanowire surface indicates
a higher level of intercellular communication.
and simplicity of polymerization with polylactides and poly-
glycolides, PCL has been investigated as a drug or bioactive mole-
cule delivery system in a variety of applications [18–21]. However,
the influence of nanotopography on degradation rate has not been
demonstrated in the literature to date. Higher PCL degradation
rates in vivo are attributed to the optimum concentration of
pancreatic enzymes in the human body [22–24]. Bacterial enzymes
such as P. cepacia lipase are typically used at higher than physio-
logical concentrations to study in vitro degradation characteristics
[22,25]. Due to very low concentrations of relatively high molecular
weight polymer degradation products, two techniques were
utilized: SEM to visualize bioerosion and MALDI-TOF to determine
the mass spectra of the degradation products collected from the
soaking media. SEM images revealed that polymer nanowires
soaked in PBS alone showed no visible bioerosion for the duration
of the study (Fig. 3A). However, nanowires soaked in PBS with P.
cepacia lipase visibly bioeroded somewhat after 4 days and signif-
icantly after 7 days (Fig. 3B). The MALDI-TOF analysis validated the
SEM observations. Control PCL samples incubated in PBS with
enzymes produced less but still significant amounts of degradation
products compared to both the control and nanowire samples
incubated in PBS alone. The regularity of the peaks in the spectra
indicates classical polymer degradation signatures. This data indi-
cates that degradation is approximately 35% (average) higher for
nanowire surfaces compared with smooth PCL, but is predomi-
nantly driven by the presence of lipase. Therefore, PCL degradation
rate may be controlled, to some extent, through parametric changes
made in the nanowires.
3.3. Short-term MSC response to nanowire surfaces
Bone tissue injury initiates a cascade of events leading to the
migration of MSCs to the site of injury. Cytokines, transcriptional
factors, growth factors and ECM proteins produced by local cells
regulate the differentiation of MSC along the osteogenic line. In
order to mimic the in vivo environment more closely, MSCs isolated
from Wistar rats were utilized instead of a secondary-derived,
immortalized cell line. Short-term MSC survivability, adhesion, and
viability on these polymer nanowire surfaces were investigated.
MSC survival and adhesion was assessed through Live/Dead fluo-
rescence staining. Images taken on a fluorescence microscope
indicate that the nanowire surface was a favorable template for cell
adhesion. Further, MSCs seeded on nanowire surfaces display
increased communication as indicated by a high degree of cell
aggregation, a signature of natural in vivo behavior (Fig. 4 – data
only shown for live cells). This desirable short-term cell response to
nanotopographical surfaces is strongly supported in the literature
[26,27]. MSCs seeded on smooth and nanowire PCL surfaces dis-
played a high level of viability as assessed using a standard MTT
783
assay (Fig. 5). The viability of cells on the PCL nanowire surfaces is
significantly (p < 0.01) higher than that on polystyrene after 4 days
of culture, indicating that the MSCs seeded on the nanowire surface
are healthy and there are no cytotoxic effects. Further, fluorescence
microscopy images of live cells seeded on nanowire surfaces show
higher degree of spreading compared to that on smooth surfaces
(Fig. 4). It is likely that the cell may have reached confluency by day
4 on nanowire surfaces resulting in decreased MTT absorbance
compared to smooth surface.
To visualize morphological changes in MSCs, SEM images were
taken after 1, 4, and 6 days of culture. SEM analysis supported
fluorescence microscopy results that the MSCs preferentially
adhere, spread, and aggregate on PCL nanowire surfaces compared
to smooth surfaces (Fig. 6) (Data not shown for polystyrene surfaces
due to the similarity seen between them and the smooth PCL
surfaces.) SEM images indicate that the cells seeded on nanowire
surfaces had a rapid morphological response compared to smooth
PCL and polystyrene surfaces. After only 24 h of culture, MSCs on
nanowire surfaces (Fig. 6B) were significantly more spread than
cells seeded on smooth surfaces (Fig. 6A). A high magnification
image of the cells seeded on the nanowire surface after 24 h
(Fig. 6C, magnified from the dotted box in Fig. 6B) revealed that
there is an extensive network of cell lamellopodia and filopodia
woven into and surrounding the nanowire substrate. These actin-
rich cell extensions are involved in cell adhesion, spreading,
migration, and intracellular communication [28–30]. After 4 days in
culture, cells seeded on nanowire surfaces (Fig. 6E) have spread
0.1
TC Polystyrene
Optical Density (450nm-690nm)
Smooth PCL
0.08
Nanowire PCL
0.06
*
0.04
0.02
0
Day 1 Day 4
Fig. 5. MSC viability after 1 and 4 days of culture. Data indicates comparable short-
term MSC viability on polystyrene, smooth, and nanowire PCL surfaces.
784 J.R. Porter et al. / Biomaterials 30 (2009) 780–788

Fig. 6. MSCs seeded on smooth and nanowire PCL surfaces after 1 day (A, B, C), 4 days (D, E, F), and 6 days (G, H, I) in culture. The left column displays images of cells seeded on
smooth PCL, the center and right columns display images of cells seeded on nanowire surfaces.

significantly (the right half of the image is a single spread cell), role in the induction of hydroxyapatite deposition on extracellular
while cells seeded on smooth surfaces have begun to spread but are matrix proteins [31,32]. ALP aids in mineralization by hydrolyzing
still mostly spherical and non-interacting. A high magnification organic phosphate esters, thus producing an excess of free inor-
image of the leading edge of a cell seeded on the nanowire surface ganic phosphate which initiates the biomineralization process.
(Fig. 6F, magnified from the dotted box in Fig. 6E) displays larger Intracellular ALP concentrations were measured after 1, 2, and 3
cell extensions woven into the nanowire substrate. After 7 days in
culture, cells seeded on the smooth surface have begun to spread
more (Fig. 6G). MSCs seeded on nanowire surfaces are nearly
indistinguishable and have become completely integrated with the
nanowire substrate (Fig. 6H and I, magnified from the dotted box in
Fig. 6H). These images suggest that MSCs seeded on nanowire
surfaces spread, migrate, adhere, and communicate much faster
compared with the same cell population seeded on smooth
surfaces.

3.4. Long-term MSC response to nanowire surfaces

The performance and clinical relevance of a tissue engineering

scaffold designed for bone regeneration relies heavily on its ability
to accelerate the cellular production of organic and inorganic
extracellular matrix. Cellular production of alkaline phosphatase,
calcium phosphate and ECM proteins required for osseointegration
and tissue regeneration were examined on PCL nanowire surfaces.
After 7 days of culture, cells were provided with media supple-
mented with dexamethasone, b-glycerolphosphate, and ascorbic
acid as described earlier. Alkaline phosphatase (ALP) activity was
Fig. 7. Intercellular concentration of ALP on polystyrene, smooth PCL, and nanowire
measured for up to 3 weeks of culture. The presence of ALP in PCL surfaces after 1, 2, and 3 weeks of culture. Peak values for all surfaces occur after 2
matrix vesicles, the site of new bone formation, suggests the role of week of culture. ALP concentrations are statistically significant (p < 0.05) on nanowire
ALP in mineralization. Research has shown that ALP plays a direct surfaces compared with polystyrene after 2 and 3 weeks in culture.
 J.R. Porter et al. / Biomaterials 30 (2009) 780–788 785

Fig. 8. SEM images of smooth (A, C, E) and nanowire surfaces (B, D, F); 1 week (A, B), 2 weeks (C, D), and 3 weeks (E, F) in culture. Calcium (upper) and von Kossa (lower) stains
correspond to the SEM image to their left. Nanowire surfaces accelerated ECM production and mineralization. Scale bar: 1 mm.

Fig. 9. SEM (A), EDS color map (B, C) and EDS spectrum (D) of a nanowire surface after 1 week of culture. The center of the image contains an aggregation of calcium-phosphate
spherulites. EDS computed a Ca:P ratio of 1.59. The presence of carbon and oxygen in the spectrum is due to the underlying PCL substrate.
786 J.R. Porter et al. / Biomaterials 30 (2009) 780–788

weeks of culture in the differentiation media. Measured intracel-
lular activity displayed a typical rise-fall pattern (Fig. 7) which has
been reported elsewhere for rat MSCs cultured on bioactive
surfaces [33,34]. ALP is an early transient biomarkers for osteo-
differentiation which is reflected in the increase in its activity
between weeks 1 and 2 for all nanowire and control surfaces. The
decrease in ALP activity by week 3 is typically due to its down-
regulation in more mature osteoblast phenotypes. Our results
indicate intracellular ALP concentration is significantly (p > 0.05)
higher on nanowire surfaces compared to control surfaces after 2
and 3 weeks of culture resulting accelerated mineralization on
nanowire surfaces.
SEM, EDS, calcium staining, and von Kossa staining were used to
determine the ability of PCL nanowire surfaces to induce mineral-
ization. Fig. 8 shows digital images of the stained surfaces (calcium
(red) on top, phosphate (brown) stain on bottom) along with SEM
images after 1, 2 and 3 weeks of culture on both PCL nanowire and
smooth surfaces. SEM images and staining results clearly indicate
that cells seeded on nanowire surfaces display accelerated miner-
alization compared to control surfaces. Calcium-phosphate
minerals in the early form of spherulites or globules were observed
after 1 week of differentiation on nanowire surfaces (Fig. 8B),
whereas very little calcium or phosphate was detected on smooth
surfaces (Fig. 8A). This mineral morphology is supported in the
literature as an early phase of biomineralization [35–37]. After 2
weeks in culture, the nanowire surfaces were completely covered
in calcium-phosphate minerals (Fig. 8D), while there is still only
minimal calcium or phosphate detected on smooth PCL surfaces.
After 3 weeks of culture, the nanowire surfaces are indistinguish-
able and completely calcified (Fig. 8F). Notable amount of calcium
and phosphate were stained on smooth surfaces after 3 weeks in
culture even though the SEM images reveal a lower amount of
mineralization (Fig. 8E). EDS was utilized to determine the
composition, stoichiometry, and spatial organization of the
calcium-phosphate minerals. Fig. 9 displays an elemental compo-
sitional map of a nanowire surface after 1 week of MSC culture. Ca:P
ratio was determined from the EDS scan to be approximately 1.59.
This closely corresponds to the theoretical stoichiometric Ca:P ratio
of 1.67 found in hydroxyapatite. However, further studies are
required definitive characterization of the type of calcium phos-
phate present on the surface.
Intra- and extracellular osteocalcin and osteopontin were
detected on nanowire and control surfaces using immunofluores-
cence as described earlier. Osteocalcin (OC) and osteopontin (OPN)
are hydroxyapatite-binding noncollagenous ECM proteins secreted
by osteoblast. Osteocalcin was selected as a biomarker as its
production indicates the presence of mature osteoblast phenotypes
[38]. OPN is a multifunctional extracellular glycoprotein involved
primarily in cell migration, and regulation of mineral deposition
[39]. Among the noncollagenous matrix proteins found in

Fig. 10. Polystyrene (A, B) and smooth PCL (C, D) surfaces showed fewer bright spots indicating focal adhesion development, but very little substrate-bound or intercellular OC or
OPN. MSCs seeded on PCL nanowire surfaces produced significantly higher amounts of both OC and OPN (E, F).
 J.R. Porter et al. / Biomaterials 30 (2009) 780–788
Fig. 11. EDS detection of calcium overlaid on the corresponding SEM image of a PCL nanowire surface encapsulated with HAp nanoparticles (A). A magnified image of the dotted box
in (A) clearly shows spherical HAp particles (white arrows, B).
mineralized tissue, OPN is unique since it preferentially accumu-
lates at mineralized tissue interfaces (i.e. cement lines and laminae
limitantes) and at mineralized tissue/implant interfaces suggesting
a role as an interfacial adhesion molecule, thereby maintaining
overall structural integrity of bone and bone/implant systems [40].
Immunolabeling for OC and OPN after 3 weeks of culture indicates
that nanowires surfaces induced substantially higher cellular
production of both proteins (Fig. 10). In addition to observing bright
spots, nanowire surfaces also displayed increased substrate signal,
which is likely related to OC and OPN binding to higher levels of
calcium phosphate on nanowire surfaces.
3.5. Hydroxyapatite encapsulation
The inclusion of calcium phosphates (CaPs) in biodegradable
polymers enhances the osseoinductivity of the bone scaffold.
Specifically, the inclusion of CaPs in polymer scaffolds enhances its
mineral deposition rate, mechanical properties, and protein
adsorption, thus improving the overall potential of the bone scaf-
fold [41–43]. Therefore, as a preliminary investigation, a solvent-
free technique described in this study was utilized to encapsulate
HAp nanoparticles in PCL nanowire surfaces. SEM/EDS analysis
indicated that HAp nanoparticles were uniformly distributed
throughout the nanowire surface at variable depths of inclusion.
EDS analysis (Fig. 11A) was used to confirm the presence of calcium
in the nanowire surface. Fig. 11A shows green spots, indicating
calcium detected by EDS, overlaid on the corresponding SEM image
of a nanowire surface. Fig. 11B shows magnified SEM image of white
dotted box on Fig. 11A indicating locations of spherical Hap
nanoparticles.
4. Conclusion
Considering the limitations of the current gold-standard treat-
ment for critical-sized defects, biodegradable synthetic bone scaf-
folds hold a lot of promise for future treatment regimes. In this
work, a solvent-free template synthesis technique was utilized to
fabricate uniform arrays of substrate-bound poly(3-caprolactone)
(PCL) nanowires. In addition to being biocompatible and biode-
gradable, nanowire surfaces induced an enhanced MSC response.
Nanowire surfaces demonstrated enhanced MSC adhesion, prolif-
eration, ALP activity, mineralization, osteocalcin, and osteopontin as
compared to smooth surfaces. Further, PCL nanowire surfaces were
also encapsulated with HAp using a solvent-free technique; thereby
avoiding any cytotoxicity concerns associated with use of organic
solvents in fabrication and encapsulation of typical polymeric
scaffolds. The performance of these PCL nanowire surfaces warrants
future work aimed at in vivo characterization, encapsulation with
therapeutics biomolecules, and fabrication of 3-D scaffolds.
787
Acknowledgments
Partial funding support for this work was provided by National
Science Foundation (CBET- 0827827).
Appendix
Figures with essential color discrimination. Certain figures,
especially parts of Figs. 3, 4, and 8–10, in this article are difficult to
interpret in black and white. The full color images can be found in
the on-line version, at doi:10.1016/j.biomaterials.2008.10.022.
References
[1] Khan SN, Cammisa Jr FP, Sandhu HS, Diwan AD, Girardi FP, Lane JM. The
biology of bone grafting. J Am Acad Orthop Surg 2005;13(1):77–86.
[2] Silber JS, Anderson DG, Daffner SD, Brislin BT, Leland JM, Hilibrand AS, et al.
Donor site morbidity after anterior iliac crest bone harvest for single-level
anterior cervical discectomy and fusion. Spine 2003;28(2):134–9.
[3] Heary RF, Schlenk RP, Sacchieri TA, Barone D, Brotea C. Persistent iliac crest
donor site pain: independent outcome assessment. Neurosurgery 2002;50(3):
510–6. Discussion 516–7.
[4] Arrington ED, Smith WJ, Chambers HG, Bucknell AL, Davino NA. Complications
of iliac crest bone graft harvesting. Clin Orthop Relat Res 1996;329:300–9.
[5] Desai TA. Micro- and nanoscale structures for tissue engineering constructs.
Med Eng Phys 2000;9:595–606.
[6] Kretlow JD, Mikos AG. Review: mineralization of synthetic polymer scaffolds
for bone tissue engineering. Tissue Eng 2007;13(5):927–38.
[7] Lim JY, Dreiss AD, Zhou Z, Hansen JC, Siedlecki CA, Hengstebeck RW, et al.
The regulation of integrin-mediated osteoblast focal adhesion and focal
adhesion kinase expression by nanoscale topography. Biomaterials 2007;
28(10):1787–97.
[8] Pham QP, Sharma U, Mikos AG. Electrospun poly(epsilon-caprolactone)
microfiber and multilayer nanofiber/microfiber scaffolds: characterization of
scaffolds and measurement of cellular infiltration. Biomacromolecules
2006;7(10):2796–805.
[9] Liu SJ, Chi PS, Lin SS, Ueng SW, Chan EC, Chen JK. Novel solvent-free fabrication
of biodegradable poly-lactic-glycolic acid (PLGA) capsules for antibiotics and
rhBMP-2 delivery. Int J Pharm 2007;330(1–2):45–53.
[10] Heckmann L, Fiedler J, Mattes T, Dauner M, Brenner RE. Interactive effects of
growth factors and three-dimensional scaffolds on multipotent mesenchymal
stromal cells. Biotechnol Appl Biochem 2008;49(Pt 3):185–94.
[11] Garcia AJ, Reyes CD. Bio-adhesive surfaces to promote osteoblast differentia-
tion and bone formation. J Dent Res 2005;84(5):407–13.
[12] Wilson CJ, Clegg RE, Leavesley DI, Pearcy MJ. Mediation of biomaterial–cell
interactions by adsorbed proteins: a review. Tissue Eng 2005;11(1–2):1–18.
[13] Albrektsson T, Johansson C. Osteoinduction, osteoconduction and osseointe-
gration. Eur Spine J 2001;10(Suppl. 2):S96–101.
[14] Mistry AS, Mikos AG. Tissue engineering strategies for bone regeneration. Adv
Biochem Eng Biotechnol 2005;94:1–22.
[15] Lin WJ, Yu CC. Comparison of protein loaded poly(epsilon-caprolactone)
microparticles prepared by the hot-melt technique. J Microencapsul 2001;
18(5):585–92.
[16] Mooney DJ, Baldwin DF, Suh NP, Vacanti JP, Langer R. Novel approach to
fabricate porous sponges of poly(D,L-lactic-co-glycolic acid) without the use of
organic solvents. Biomaterials 1996;17(14):1417–22.
[17] Miyai T, Ito A, Tamazawa G, Matsuno T, Sogo Y, Nakamura C, et al. Antibiotic-
loaded poly-epsilon-caprolactone and porous beta-tricalcium phosphate
composite for treating osteomyelitis. Biomaterials 2008;29(3):350–8.
788 J.R. Porter et al. / Biomaterials 30 (2009) 780–788
[18] Sinha VR, Bansal K, Kaushik R, Kumria R, Trehan A. Poly-epsilon-caprolactone
microspheres and nanospheres: an overview. Int J Pharm 2004;278(1):1–23.
[19] Verreck G, Chun I, Li Y, Kataria R, Zhang Q, Rosenblatt J, et al. Preparation
and physicochemical characterization of biodegradable nerve guides con-
taining the nerve growth agent sabeluzole. Biomaterials 2005;26(11):
1307–15.
[20] Li S, Liu L, Garreau H, Vert M. Lipase-catalyzed biodegradation of poly(epsilon-
caprolactone) blended with various polylactide-based polymers. Bio-
macromolecules 2003;2:372–7.
[21] Quaglia F, Ostacolo L, Nese G, De Rosa G, La Rotonda MI, Palumbo R, et al.
Microspheres made of poly(epsilon-caprolactone)-based amphiphilic copoly-
mers: potential in sustained delivery of proteins. Macromol Biosci
2005;5(10):945–54.
[22] Kulkarni A, Reiche J, Kratz K, Kamusewitz H, Sokolov IM, Lendlein A. Enzy-
matic chain scission kinetics of poly(epsilon-caprolactone) monolayers.
Langmuir 2007;23(24):12202–7.
[23] Yeo A, Rai B, Sju E, Cheong JJ, Teoh SH. The degradation profile of novel,
bioresorbable PCL-TCP scaffolds: an in vitro and in vivo study. J Biomed Mater
Res A 2008;84(1):208–18.
[24] Sun H, Mei L, Song C, Cui X, Wang P. The in vivo degradation, absorption and
excretion of PCL-based implant. Biomaterials 2006;27(9):1735–40.
[25] Chawla JS, Amiji MM. Biodegradable poly(epsilon-caprolactone) nanoparticles
for tumor-targeted delivery of tamoxifen. Int J Pharm 2002;249(1–2):127–38.
[26] Dalby MJ, McCloy D, Robertson M, Agheli H, Sutherland D, Affrossman S, et al.
Osteoprogenitor response to semi-ordered and random nanotopographies.
Biomaterials 2006;27(15):2980–7.
[27] Popat KC, Leoni L, Grimes CA, Desai TA. Influence of engineered titania
nanotubular surfaces on bone cells. Biomaterials 2007;28(21):3188–97.
[28] Rottner K, Behrendt B, Small JV, Wehland J. VASP dynamics during lamelli-
podia protrusion. Nat Cell Biol 1999;5:321–2.
[29] Pantaloni D, Le Clainche C, Carlier MF. Mechanism of actin-based motility.
Science 2001;292(5521):1502–6.
[30] Bailly M, Macaluso F, Cammer M, Chan A, Segall JE, Condeelis JS. Relationship
between Arp2/3 complex and the barbed ends of actin filaments at the leading
edge of carcinoma cells after epidermal growth factor stimulation. J Cell Biol
1999;145(2):331–45.
[31] Storrie H, Stupp SI. Cellular response to zinc-containing organoapatite: an in
vitro study of proliferation, alkaline phosphatase activity and biomineraliza-
tion. Biomaterials 2005;26(27):5492–9.
[32] Anderson HC. Molecular biology of matrix vesicles. Clin Orthop Relat Res
1995;314:266–80.
[33] Shin H, Zygourakis K, Farach-Carson MC, Yaszemski MJ, Mikos AG. Modulation
of differentiation and mineralization of marrow stromal cells cultured on
biomimetic hydrogels modified with Arg-Gly-Asp containing peptides.
J Biomed Mater Res A 2004;69(3):535–43.
[34] Bancroft GN, Sikavitsas VI, van den Dolder J, Sheffield TL, Ambrose CG,
Jansen JA, et al. Fluid flow increases mineralized matrix deposition in 3D
perfusion culture of marrow stromal osteoblasts in a dose-dependent manner.
Proc Natl Acad Sci U S A 2002;99(20):12600–5.
[35] Lee JY, Choo JE, Choi YS, Park JB, Min DS, Lee SJ, et al. Assembly of collagen-
binding peptide with collagen as a bioactive scaffold for osteogenesis in vitro
and in vivo. Biomaterials 2007;29:4257–67.
[36] Saruwatari L, Aita H, Butz F, Nakamura HK, Ouyang J, Yang Y, et al. Osteoblasts
generate harder, stiffer, and more delamination-resistant mineralized tissue
on titanium than on polystyrene, associated with distinct tissue micro- and
ultrastructure. J Bone Miner Res 2005;20(11):2002–16.
[37] Murphy WL, Hsiong S, Richardson TP, Simmons CA, Mooney DJ. Effects of
a bone-like mineral film on phenotype of adult human mesenchymal stem
cells in vitro. Biomaterials 2005;26(3):303–10.
[38] Hosseinkhani H, Hosseinkhani M, Tian F, Kobayashi H, Tabata Y. Osteogenic
differentiation of mesenchymal stem cells in self-assembled peptide-amphi-
phile nanofibers. Biomaterials 2006;27(22):4079–86.
[39] de Oliveira PT, Nanci A. Nanotexturing of titanium-based surfaces upregulates
expression of bone sialoprotein and osteopontin by cultured osteogenic cells.
Biomaterials 2004;25(3):403–13.
[40] McKee MD, Nanci A. Osteopontin at mineralized tissue interfaces in bone,
teeth, and osseointegrated implants: ultrastructural distribution and impli-
cations for mineralized tissue formation, turnover, and repair. Microsc Res
Tech 1996;33(2):141–64.
[41] Wei G, Ma PX. Structure and properties of nano-hydroxyapatite/polymer
composite scaffolds for bone tissue engineering. Biomaterials 2004;25(19):
4749–57.
[42] Kim SS, Sun Park M, Jeon O, Yong Choi C, Kim BS. Poly(lactide-co-glycolide)/
hydroxyapatite composite scaffolds for bone tissue engineering. Biomaterials
2006;27(8):1399–409.
[43] Rezwan K, Chen QZ, Blaker JJ, Boccaccini AR. Biodegradable and bioactive
porous polymer/inorganic composite scaffolds for bone tissue engineering.
Biomaterials 2006;27(18):3413–31.