Chemical Engineering Journal 222 (2013) 49–59

Contents lists available at SciVerse ScienceDirect

Chemical Engineering Journal

journal homepage: www.elsevier.com/locate/cej

Strontium substituted hydroxyapatite porous microspheres:

Surfactant-free hydrothermal synthesis, enhanced biological
response and sustained drug release
Kaili Lin a,1, Peiyi Liu b,1, Li Wei b, Zhaoyong Zou a, Weibin Zhang b, Ying Qian b, Yuhui Shen b,⇑, Jiang Chang a,⇑
a
State Key Laboratory of High Performance Ceramics and Superfine Microstructure, Shanghai Institute of Ceramics, Chinese Academy of Sciences, China
b
Department of Orthopaedics, Shanghai Institute of Orthopaedics & Traumatology, Shanghai Ruijin Hospital, Shanghai Jiaotong University School of Medicine, China

h i g h l i g h t s g r a p h i c a l a b s t r a c t

" The mesoporous strontium

substituted hydroxyapatite (SrHAp)
microspheres were fabricated.
" The ionic extracts of the
microspheres promoted the
biological responses.
" The architecture of SrHAp resulted in
favorable drug-loading and
sustained release properties.

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

Article history: In the absence of any surfactants, organic solvents or template-directing reagents, the strontium substi-
Received 5 December 2012 tuted hydroxyapatite (SrHAp) microspheres with hierarchically mesoporous structures and 1.17–
Received in revised form 6 February 2013 5.60 mol.% of Sr substitution were successfully fabricated via hydrothermal method. The morphology
Accepted 12 February 2013
observation showed that the fabricated SrHAp porous microspheres in diameters of 50–65 lm were
Available online 21 February 2013
assembled by two-dimensional single-crystal nano-sheets with 30–70 nm in thickness, and up to 3 lm
in width and length. The ionic extracts of SrHAp porous microspheres promoted the proliferation, oste-
Keywords:
ogenic differentiation and angiogenic factor expression of human osteoblast-like cells (MG-63), and the
Hydroxyapatite
Strontium
microspheres with 3.22 mol.% of Sr substitution showed the best potentially therapeutic applications in
Porous microspheres bone regeneration field. In addition, the novel 3D architectures of SrHAp resulted in favorable drug load-
Synthesis ing and sustained release properties. Our study indicated that the fabricated multifunctional SrHAp por-
Biological response ous microspheres might be a potential candidate as bioactive bone-regeneration and drug-delivery
Drug delivery and release material.
Ó 2013 Elsevier B.V. All rights reserved.

1. Introduction als [1]. Calcium phosphate (Ca–P) materials, such as hydroxyapa-

The material with initiative stimulation capacity in tissue
regeneration is the major character for next generation biomateri-
⇑ Corresponding authors. Tel.: +86 21 52412810; fax: +86 21 52413903.
E-mail addresses: yuhuiss@163.com (Y. Shen), jchang@mail.sic.ac.cn (J. Chang).
1
These authors contributed equally to this article.
tite [Ca10(PO4)6(OH)2, HAp], b-tricalcium phosphate [b-Ca3(PO4)2,
b-TCP] and calcium phosphate cement (CPC), are widely used in
biomedical fields due to their good biocompatibility, osteo-con-
ductivity and similarity to the inorganic component of bone tissues
[2]. However, the traditional Ca–P based materials, including the
HAp, are lack of the ability to stimulate the formation of new bone,
which hinder their clinical applications [3]. Several method are

1385-8947/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cej.2013.02.037
50 K. Lin et al. / Chemical Engineering Journal 222 (2013) 49–59
applied to solve this problem, such as loading bioactive growth fac-
tors [4], grain size and surface morphology/topography design [5],
and the incorporation of the functional trace elements [6,7]. In
which, the incorporation of the functional trace inorganic elements
into the biomaterials to improve the biological responses has been
aroused great interest due to its simplicity and low cost. Recently,
strontium (Sr) ranelate, a newly developed drug treating osteopo-
rosis, has been shown to have dual effects of stimulation osteoblast
differentiation and inhibiting osteoclast activity and bone resorp-
tion, which could reduce the incidence of fractures in osteoporotic
patients [8]. In addition, the partial substitution of Ca by Sr can
apparently improve the biological properties of Ca–P materials [9].
Recently, the nano-structured HAp porous microspheres have
attracted great interests due to their high specific surface area
and novel three-dimensional (3D) hierarchical architectures,
which make it possible to incorporate higher dosages of drugs into
the materials and release them at a control rate [6,10]. Further-
more, the 3D architectured HAp microporous materials can be used
as injectable bone regeneration biomaterials and cell/drug loaded
implants, which is superior to particles [11,12]. Thus, comparing
with traditional particles, the Sr-substituted HAp (SrHAp) porous
microspheres might possess better biological properties for guid-
ing bone tissue regeneration, and have great potential application
for drug delivery. Up to now, the SrHAp porous microspheres with
hierarchical nano-architectures are rarely reported.
Furthermore, the traditional strategy to synthesize the 3D
architectured materials is focused on the organic solvent, surfac-
tant and chelating ligand assistant assembled approach [13–20].
Ma and Zhu synthesized HAp hollow microspheres via solvother-
mal method using N,N-dimethylformamide (DMF) as the solvent
[16]. Zhang et al. fabricated HAp microspheres and microflowers
via hydrothermal method using hexadecyltrimethylammonium
bromide (CTAB) as the surfactant [17]. He et al. reported that in
the hexane–water–bis(2-ethylhexyl) sulfosuccinate (AOT) system
hollow HAp microspheres were presented [18]. The HAp materials
with flower-like morphology assembled from nanosheets consist-
ing of nanorod building blocks was synthesized by Ma using potas-
sium sodium tartrate as chelating ligand and template molecule
[19]. Veljovic et al. prepared the HAp hollow microspheres assem-
bled by rod-shaped nanoparticles using Na2EDTA as chelating li-
gand and template molecule [20]. However, in most of these
cases, large amounts of template and/or organic solvents were
widely used, which might be hazardous to health and environ-
ment. Up to now, self-assembly of nano-units into the 3D architec-
tures in the absence of any surfactants, template-supports or
structure-directing reagents are still a major challenge [21]. Re-
cently, Neira et al. developed a repeated stepwise hydrothermal
method to synthesize micrometer-sized HAp particles with a sharp
faceted hexagonal prism-like morphology [22].
Herein, the SrHAp porous microspheres with 1.17–5.60 mol.% of
Sr substitution was synthesized via hydrothermal method. Then
the effect of Sr substitution on osteoblast cell (MG-63) prolifera-
tion, osteogenic differentiation and angiogenic factor expression
of the microspheres, and the effect of the 3D architectured nano-
structures on drug loading and release capacities were further
investigated.
2. Materials and methods
2.1. Synthesis and characterization of SrHAp porous microspheres
The SrHAp porous microspheres with designed Sr2+/(Ca2+ + Sr2+)
molar ratios of 0.01, 0.03 and 0.05 were hydrothermally synthe-
sized using urea ((NH2)2CO) as homogeneous precipitation re-
agent. The obtained products were labeled as Sr1-HAp, Sr3-HAp
and Sr5-HAp, respectively. In the synthesis process, the aqueous
solutions containing 0.05 mol (Ca2+ + Sr2+), 0.03 mol HPO24 and
0.15 mol urea were prepared by dissolving Ca(NO3)2, Sr(NO3)2,
NH4H2PO4 and urea in 100 mL distilled water, and 0.1 mol/L
HNO3 solution was used to adjust the pH to 2.41–2.45 to obtain
clear solutions. Then 17 mL of the obtained solution were trans-
ferred into 20 mL Teflon autoclaves and heated at 120 °C for 1–
72 h, followed by cooling to room temperature naturally. It is well
known that, with the increase of the hydrothermal temperatures,
the (NH2)2CO continuously decomposes to form CO2 and aqueous
ammonia species based on the following equation:
ðNH2 Þ2 CO þ H2 O ! 2NH3 þ CO2
In which, the released NH3 easily dissolves in water and in-
creases the pH value to alkaline condition in the reaction solution,
wherein HAp becomes the more thermodynamically stable cal-
cium orthophosphate compound, and is thus formed [22]. There-
fore, the continuous and homogeneous decomposition of urea
generates the driving force toward the nucleation and growth of
HAp crystals under moderate supersaturation conditions [22]. In
2
addition, the CO2 dissolves in water to form HCO 3 and CO3 , and
then incorporates as carbonate in the HAp lattice [23]. After hydro-
thermal reaction, the obtained suspension was filtrated and
washed with distilled water and anhydrous ethanol for three
times, respectively. The obtained powders were dried at 120 °C
for 24 h.
To investigate the roles of the chemical composition (Sr-substi-
tution) and porous morphologies on the biological performance
and drug loading/release properties, respectively, the pure HAp
nanoparticles labeled as Sr0-HAp were prepared as the control
sample. The pure HAp nanoparticles can well play dual roles as
chemical component control for biological response, and as mor-
phology control for drug loading/release properties. First, 0.5 M
Ca(NO3)2 solution and 0.3 M (NH4)2HPO4 solution were obtained
by dissolving Ca(NO3)2 and (NH4)2HPO4 in distilled water, respec-
tively, and the pH of both solutions was adjusted to 10.5 using
ammonia solution. The reactant molar ratio of Ca/P was kept at
1.67. The solution of (NH4)2HPO4 was added dropwise into the
Ca(NO3)2 solution to obtain a suspension, and the pH of the sus-
pension was maintained at 10.5 using ammonia solution. Then
the suspension was transferred into autoclaves and heated at
120 °C for 72 h, followed by cooling to room temperature naturally.
After the hydrothermal reaction, the obtained suspensions were
filtrated and washed with distilled water and anhydrous ethanol
for three times, respectively. The obtained powders were dried at
120 °C for 24 h.
The obtained powders were characterized by X-ray diffraction
(XRD: D/max 2550 V, Rigaku, Japan) with mono-chromated Cu
Ka radiation and Fourier transform infrared spectroscopy (FTIR:
Nicolet Co., USA). The lattice constants were calculated from the
well determined positions of the intense XRD diffractions that
were processed by MDI Jade 6.1 software [24]. To evaluate the
crystallinity degree, the reflection (2 1 1) of HAp powders at around
2h = 31.8° (JCPDS No. 09-0432) was selected as the diffraction
plane for crystallinity measurement to calculate the relative index
of crystallinity (IOC, %) defined as follows [25]:
IOC ¼ ðIc =Io Þ  100%
where Ic and Io are the peak intensity of the selected diffraction
plane in the obtained products and the control sample, respectively.
In present study, the fully crystallized HAp structure prepared by
wet chemical precipitation method and then calcined in furnace
at 900 °C for 1.5 h was used as the control sample [26]. The mor-
phology and size of the obtained powders were characterized by
scanning electron microscopy (SEM: JSM-6700F, JEOL, Japan) and
transmission electron microscopy (TEM: JEM-2100F, JEOL, Japan).
 K. Lin et al. / Chemical Engineering Journal 222 (2013) 49–59
The surface area and nitrogen sorption isotherm for the samples
were measured on a Micrometitics Tristar 3000 system. The chem-
ical compositions of the products were quantified by inductively
coupled plasma – optical emission spectrometry (ICP-OES, 715-ES,
Varian, USA).
2.2. Effect of ionic product from SrHAp porous microspheres on the
proliferation, osteogenic differentiation and angiogenic factor
expression of human osteoblast-like cells
The ionic extract method is widely used as the international
standard method to evaluate the effect of the chemical composi-
tions on cell biological responses [27], which can effectively avoid
the additional effects coming from the material morphologies by
directly incubation the materials with cells. In present study, the
human osteoblast-like cells (MG-63 cell, Cell bank, Shanghai, Chi-
na) were used, and the cells were cultured in the Dulbecco’s mod-
ified Eagle’s medium (DMEM, Invitrogen) containing 10% fetal
bovine serum (FBS) (Hyclone) and 1% penicillin–streptomycin
(Invitrogen). To prepare the extracts, a stock solution of 200 mg/
mL was first prepared by immersing each test HAp powders in
DMEM culture medium. After incubation at 37 °C for 24 h, the mix-
ture was centrifuged and the supernatant was collected. Then a se-
rial diluted extracts with concentrations of 100, 50, 25 and
12.5 mg/mL were prepared by diluting from the stock solution of
200 mg/mL with serum-free DMEM. Subsequently, these extracts
were sterilized by filtration through 0.22 lm filter membrane for
cell culture experiments. The ion concentrations of the extracts
were measured by ICP-OES. The proliferation of MG-63 was carried
out using Cell Counting Kit-8 (CCK-8) assay [28].
The MG-63 cells was seeded in 96-well tissue culture plates at a
density of 5  103 cells/well for 1 and 3 days of culture by incuba-
tion at 37 °C for 24 h with 5% CO2, 95% air at 100% RH. The medium
in the well was then replaced by the prepared extracts. After cell
culture, removed the supernatant then added 100 lL DMEM plus
10 lL CCK-8 into each well. After additional incubation for 1 h
and then 20 s of slow shaking, the absorbance was read at
450 nm against the reference value at 600 nm (BIO-RAD Model
680, USA), and the results were expressed as optical density
(OD). All experiments were done in triplicate to obtain the average
data.
Based on the proliferation results, the optimal extract concen-
tration of the SrHAp sample with strongest stimulation property
was selected to further examine the effect of the Sr-substitution
on osteogenic differentiation and angiogenic factor expression of
the osteoblasts using real-time polymerase chain reaction (PCR)
analysis. The extract of the pure HAp nanoparticles with same con-
centration was used as control. After culture MG-63 in 96-well tis-
sue culture plates at a density of 1  105 cells/well for 5, 7 and
10 days, the total RNA was isolated and then the cDNA was synthe-
sized using a Prime-ScriptTM RT reagent kit (Takara Bio, Shiga, Ja-
pan) according to manufacturer’s instructions. Highly purified gene
specific primers (Table 1) for alkaline phosphatase (ALP), collagen
type 1 (Col I), core-binding factor subunit alpha-1 (Cbfa1), vascular
endothelial growth factor (VEGF) and the house-keeping gene, AC-
TIN, were synthesized commercially (Shengong, Co., Ltd. Shanghai,
China). Quantification of all cDNA of bone marker genes was tested
using Applied Biosystems-7500 real-time PCR system. For quanti-
tative PCR, 10 lL SYBR Premix Ex Taq™ 6 lL ddH2O, 0.8 lL of each
forward and reverse primer, 0.4 lL Rox and 2.0 lL cDNA template
were used in a final reaction volume of 20 lL. The thermal profile
for all reactions was as follows: 30 s at 95 °C, followed by 40 cycles
of 5 s at 95 °C, 34 s at 60 °C and 30 s at 72 °C. Data collection was
enabled at 72 °C in each cycle. CT (threshold cycle) values were cal-
culated using the Applied Biosystems 7500 software v 2.0.5 sys-
tem. Each mRNA value was normalized to that of the house-
51
Table 1
Primer sequences for real-time polymerase chain reaction (PCR).
Gene Primer sequences
ALP Forward: GGGAACGAGGTCACCTCCAT
Reverse: TGGTCACAATGCCCACAGAT
Col-I Forward: CCTGCGTGTACCCCACTCA
Reverse: ACCAGACATGCCTCTTGTCCTT
Cbfa1 Forward: AGTGATTTAGGGCGCATTCCT
Reverse: GGAGGGCGGCGTGGGTTCT
VEGF Forward: AGGAGGAGGGCAGAATCATCA
Reverse: CTCGATTGGATGGCAGTAGCT
keeping gene, human ACTIN (DCt = Ct gene of interest – Ct ACTIN).
Results are reported as relative gene expression (2DDCt). All exper-
iments were done in triplicate to obtain the average data.
2.3. In vitro study of drug loading and release properties
2.3.1. Release measurements
To investigate the influence of the porous structures on drug
loading and release properties, the antibiotic vancomycin (New
Drug Research and Development CO., Ltd., North China Pharmaceu-
tical Group) and the HAp nanoparticles were used as the drug
model and the material control sample, respectively. Vancomycin
release measurements were carried out using UV/VIS spectroscopy
at a wavelength of 280 nm. Calibration curve (correlation coeffi-
cient R2 = 0.9999) was made for each set of measurements and
determined by taking absorbance vs. vancomycin concentration
between 0 and 1.0 mg/mL as parameters. Three samples from each
group were measured to obtain the average data of the drug load-
ing and release amount.
2.3.2. Drug loading procedures
Typically, 0.25 g of SrHAp products were added into 1.0 mL of
aqueous solution with a vancomycin concentration of 10 mg/mL
at 37 °C, and soaked for 24 h under stirring in a vial that was sealed
to prevent the evaporation of aqueous solution. The vancomycin-
loaded SrHAp samples were separated by centrifugation, and then
dried in vacuum at 37 °C for 24 h. The amount of loaded vancomy-
cin was measured by the depletion method, by determining the
difference in vancomycin concentration in the loading medium be-
fore and after loading. The amount of drug loading was calculated
according to the following formula:
Amount of drug loaded ¼ ðA  BÞ=A  100%
where A and B represented the initial and final drug concentrations,
respectively.
2.3.3. In vitro drug release
0.25 g of drug-loaded SrHAp samples were placed into 1.0 mL of
phosphate buffer solution (PBS) at pH 7.4 and agitated in a hori-
zontally shaking water bath at 37 °C. At the predetermined time
intervals, the solutions were centrifuged, and then the release
medium was withdrawn and replaced with fresh release medium
(0.50 mL). The amount of the drug release was determined by
UV/VIS spectroscopy.
2.4. Statistical analysis
Data were analyzed for statistical significance using an analysis
of variance. Differences at p values of <0.05 were considered
significant.
52 K. Lin et al. / Chemical Engineering Journal 222 (2013) 49–59
3. Results and discussion
3.1. Characterization of SrHAp porous microspheres
At the end of the hydrothermal process, the pH value of the
reaction solutions reached 7.46–7.48 which was favorable for the
nucleation and growth of HAp crystals. Fig. 1 shows the morphol-
ogy images of the control sample and the synthetic SrHAp products
via hydrothermal treatment at 120 °C for 72 h. It is clear that the
obtained SrHAp products were in 3D architectured porous micro-
spheres with similar morphology. The diameters of the porous
microspheres were between 50 and 65 lm and the porous micro-
spheres consisted of 2D nano-sheets. The high magnification
images (Fig. 1B, D and F) showed that the nano-sheet had smooth
surface with a thickness of about 30–70 nm, and width and length
of up to 3 lm. As expected, the control sample was mono-disper-
sive in the absence of aggregations and had smooth surface and
rod-like shape, 25–30 nm in diameter and 25–100 nm in length
(Fig. 1G). In general, the simply hydrothermal treatment of the
precipitated HAp particles in aqueous solution without using any
template-directing reagents is usually dispersive rod-like nanopar-
ticles because of the preferred orientation growth of HAp crystal
along the c-axis [2].
Fig. 2A presents the XRD patterns of the control sample and the
synthetic SrHAp porous microspheres. It is clear that all of the prod-
ucts were identified as pure HAp phase (JCPDS card: No. 09-0432).
The crystallinity of the hydrothermally synthetic HAp products
was about 69–86% calculated by the area-integration method from
XRD patterns [25,26], indicating the relatively high crystallinity of
the products. The small angle XRD scanning results (Fig. 2B) clearly
confirmed that the corresponding peaks of the synthetic SrHAp por-
ous microspheres significantly shifted to lower degree when com-
pared with the pure HAp materials, indicating an increase of the
lattice constants [29]. The shifting degree increased with the in-
crease of the Sr substitution amount. The lattice constants were fur-
ther calculated based on the XRD characterization and the results
were listed in Table 2. It is clear that most of the lattice constants
of the synthetic SrHAp products were larger than that of the pure
HAp. The increasing of the lattice constants were due to the replace-
ment of the Ca ions by bigger atomic diameter of Sr [29]. In present
study, the deviation of lattice constants suggested that Sr2+ replaced
and occupied the Ca2+ crystal sites of the HAp.
Fig. 3 presents the FITR spectra of the control sample and the
synthetic SrHAp porous microspheres. The broad and strong bimo-
dal peaks at 1458 and 1418 cm1, and the peak at 872 cm1 were
attributed to carbonate ions (CO2 3 ) in B-site [21]. The peaks at 475
(m2), 559 (m4), 602 (m4), 958 (m1), 1032 (m3) and 1106 (m3) cm1 were
the characteristic bands for PO3 4 , and the peaks at 3448 and
1636 cm1 were assigned to the bending mode of the adsorbed
water. In addition, the characteristic OH bands of HAp at around
3571 cm1 (mOH) and 633 cm1 (dOH) were not clearly visible from
the synthetic SrHAp porous microspheres because of the presence
of carbonate ions [21]. Carbonate substitution induces crystal lat-
tice aberrance, and the harmonic oscillator unit of the equilateral
triangle 3Ca(II)–OH is disordered, which means that the resonance
effect of harmonic vibration in OH channels is disturbed and the
higher the content of carbonate is, the more this disturbance vari-
able is. Therefore, the structured OH peaks will disappear in case of
high content of carbonate [21].
To investigate the formation mechanism of the microspheres,
the morphology of the products (using the precursors of Sr3-
HAp) obtained at different hydrothermal time points such as 1, 3,
6 and 72 h (final product) at reaction temperature of 120 °C were
characterized by SEM (Fig. 4). After 1 h of hydrothermal treatment,
only a small amount of the product was obtained. As shown in
Fig. 4A, the morphology of the obtained product in the early stage
was small sheetlike shape with sizes of about 0.1–0.5 lm, and was
free of aggregation. Then the crystal growth followed, and bigger
particles grew at the expense of smaller crystals, as described by
the Gibbs–Thomson equation [30]. With the increase of the hydro-
thermal time to 3 h, the obtained powders grew wider and longer
than those in the early stage at the expense of the small crystal-
lites. The small sheets almost completely disappeared. Due to the
high surface energy of the nano-crystals, the as-formed nano-
sheets aggregated (Fig. 4B) to reduce their surface energy in aque-
ous solution. The optimum shape for minimizing the surface free
energy is a spherical structure due to the lowest surface area of
the spheres [31], which favors the self-assembling of the 2D
nano-sheets into microspheres rather than other shape of aggrega-
tions. Further increase of the reaction time to 6 h, these sheets
assembled as ‘‘petals’’ into 3D structures (Fig. 4C). The morphology
of the final product is shown in Fig. 4D. It was clear that much
more sheets were assembled into the clusters, and the typical 3D
architecture of porous microspheres formed. The degree of assem-
bly increased with the increase of hydrothermal time. From
Fig. 4A–D, one can clearly see the process of crystalline nucleation
and growth, and self-assembling growth from nano-sheets to 3D
microsphere morphology. Based on the results of time-dependent
experiments and the morphology observations (Fig. 4), the process
of the morphology evolution of HAp porous microspheres is sum-
marized in Fig. 5. The similar phenomenon of nucleation–dissolu-
tion–recrystallization–self-assembly growth process has been
widely observed in the synthesis of metal oxide and hydroxide,
sulfide and other kinds of materials [32–35]. However, further
work needs to be done to completely illuminate the self-assemble
mechanism for a better understand the role of all parameters in
structure regulation, such as the ion concentration, pH value,
hydrothermal temperature and period.
The crystalline structure of the apatite porous microspheres
was further confirmed by selected-area electron diffraction (SAED)
and high-resolution transmission electron microscopy (HRTEM)
investigations. Fig. 6 showed the representative SAED (Fig. 6B)
and HRTEM image (Fig. 6C) taken from the single nano-sheet
(Fig. 6A) which was sonicated from the microspheres. The discrete
SAED spots (Fig. 6B) revealed that the final product was well-crys-
tallized hexagonal single crystal. A typical HRTEM image of the sin-
gle nano-sheet in Fig. 6C apparently showed the lattice fringes,
with interplanar spacings of 0.46 and 0.34 nm, corresponding to
the (1 1 0) and (0 0 2) planes of HAp, respectively. The HRTEM im-
age further confirmed that the nano-sheet as the assembled com-
ponent for the microspheres was single crystal.
The elemental analyses of the synthetic HAp powders were pre-
sented in Table 2. The results showed that the (Ca + Sr)/P molar ra-
tio of the fabricated SrHAp porous microspheres was similar to that
of pure HAp (Ca/P = 1.67). In addition, the Sr substitution level
could be facilely tailored by regulation the initial molar ratios of
Sr2+/(Ca2+ + Sr2+).
Moreover, the aggregates of the nanocrystals in the synthetic
SrHAp porous microspheres led to higher specific surface area
(SBET) and formation of mesopores, which were confirmed by the
nitrogen adsorption–desorption isotherm and the density func-
tional theory (DFT) pore size distribution curves (Table 3 and
Fig. 7). The SBET of the SrHAp porous microspheres was about
2–2.6 times higher than that of the control HAp nanoparticles
(Table 3). According to the International Union of Pure and Applied
Chemistry, the data shown in Fig. 7 suggested that the fabricated
microspheres exhibit a type H4 hysteresis loop deriving from
particle aggregates with slit-shaped pores in the range of 0.45–
 K. Lin et al. / Chemical Engineering Journal 222 (2013) 49–59 53

Fig. 1. The morphologies of the synthetic SrHAp porous microspheres: Sr1-HAp (A and B), Sr3-HAp (C and D) and Sr5-HAp (E and F), and the control Sr0-HAp nanoparticles
(G).
54 K. Lin et al. / Chemical Engineering Journal 222 (2013) 49–59
Fig. 2. XRD patterns of the control HAp sample and the synthetic SrHAp porous
microspheres.
1.0 P/P0, which was typical of a mesoporous structure [36]. Calcu-
lated from the adsorption branches of the isotherms using the DFT
method, the pore size is distributed from meso- to macro-scale in a
range of about 2–100 nm (Fig. 7).
3.2. Effect of ionic product from SrHAp porous microspheres on MG-63
proliferation, osteogenic differentiation and angiogenic factor
expression
The CCK-8 evaluation results revealed no appreciable toxicity
when the human osteoblast-like cells (MG-63) were subjected to
control HAp nanoparticles and the SrHAp porous microspheres in
the extract concentration range of 200–12.5 mg/mL (Fig. 8). All
extracts of SrHAp porous microspheres exhibited none cytotoxic-
ity. The cell proliferation was evident for all samples over the
duration of the culture time. A stimulatory effect of the ionic
product from Sr3-HAp porous microspheres on MG-63 prolifera-
tion was extraordinarily higher than other materials at all the
extract concentration range of 200–12.5 mg/mL in culture periods.
In addition, the extract concentration of 100 mg/mL both for
Sr1-HAp and Sr5-HAp also favors to stimulate MG-63 proliferation.
The results suggested that the component of Sr3-HAp porous
microspheres might be the best potentially therapeutic applica-
tions for bone regeneration. Previous studies have confirmed that
the Sr2+ ion plays the most important role in biological perfor-
mances in the extracts of Sr-containing bioceramics [8,9]. The
ICP-OES analysis showed that the Sr2+ ion concentrations of the
Sr3-HAp extracts (200–12.5 mg/mL) for cell culture was 0.1342–
0.0084 mmol/L. Furthermore, the effects of Sr on bone cells were
dose-dependent [37]. Canalis et al. found that the divalent stron-
tium salt S12911 at 103 mmol/L increased the replication of pre-
osteoblastic cells by 30–50% after 24 h and by 60% after 96 h of
treatment [38]. Bonnelye et al. revealed that strontium ranelate
with Sr2+ between 0.1 and 1 mmol/L could stimulate osteoblast
formation [39]. It has been reported that the concentration of Sr
in normal serum varies between 10 lg/L (1.14  104mmol/L)
and 217 lg/L (2.48  103 mmol/L) [40], which is much lower than
Table 2
Crystallinity, lattice constants, 2h value for (2 1 1) diffraction and chemical composition of the synthetic HAp products.
Samples Crystallinity (%) Lattice constantsa
a (Å) c (Å)
Sr0-HAp 69 9.496 6.93
Sr1-HAp 86.5 9.496 6.954
Sr3-HAp 73.2 9.507 6.916
Sr5-HAp 82.4 9.507 6.989
a
Hydroxyapatite ICDD PDF # 09-0432: a = 9.418 Å, c = 6.884 Å, 2h for (2 1 1) diffraction is 31.86°.
the values in present study. It is well known that the amount of Sr
in bone tissue is much higher than that in blood. Furthermore,
99.1% of the absorbed Sr is deposited in bone (36–140 mg/kg)
[37,41]. Therefore, it can be indicated that a certain high dose of
Sr may stimulate the proliferation of osteoprogenitor cells and
benefit new bone regeneration.
Base on the cell proliferation results, the extract concentration
of 100 mg/mL from Sr3-HAp porous microspheres was further
selected to examine the osteogenic differentiation and angiogenic
factor expression of the Sr substituted HAp materials, and the
Sr0-HAp was used as the control sample (Fig. 9). The real-time
PCR analysis results showed that the expression levels of osteo-
blastic markers, such as ALP, Col I and Cbfa1, and the VEGF were
markedly higher for Sr3-HAp porous microspheres than those for
the control Sr0-HAp material at all the culture time points, except
for cbfa1 on day 5. For the Sr3-HAp extract, the expression level of
ALP decreased slightly from day 5 to day 7, and then maintained
with the further increase of the culture time to day 10, while those
of Col I, Cbfa1 and VEGF increased sustainably. As for the control
Sr0-HAp extract, the expression levels of ALP, Col I, Cbfa1 and VEGF
reached the peak at day 7, and then decreased apparently.
Especially, the Cbfa1 known as an osteoblast-specific transcription
factor was remarkably upregulated by SrHAp materials, which
plays an important role as an upstream regulator of osteoprogeni-
tor differentiation [42]. Traditionally, the osteogenic differentiation
of osteoblast cell lines is initiated by the addition of dexametha-
sone, b-glycerophosphate and ascorbic acid, etc. [43]. The CCK-8
and PCR analysis showed that the Sr substitution can stimulate
the proliferation and osteogenic differentiation of MG-63 cells.
Furthermore, the angiogenic factor of VEGF can also be upregu-
lated by the Sr incorporation. The angiogenesis is extraordinarily
important during bone regeneration [44]. Especially for bone
defects with critical size in centimeter size, the difficulty of
inducing rapid vascular ingrowths is harmful for bone regenera-
tion, which limits the diffusion of oxygen and nutrients from the
surrounding tissue to ensure the viability and function of cells
[45]. Up to now, the VEGF has been widely used as growth factor
to stimulate angiogenesis in tissue engineered constructs [12,46].
However, due to the complex of in vivo environment and short
half-life of proteins, the release behavior of proteins in vivo cannot
be easily controlled, which may also result in possible toxicity ef-
fects [47]. Furthermore, the cost of the growth factors is extraordi-
narily high. All of these concerns limit the further application of
growth factors in clinical fields. Up to now, little work has been
done on the influences of Sr on angiogenesis. Herein, our primarily
study suggested that the low cost incorporation of inorganic Sr ion
into HAp bioceramics could induce higher angiogenic expression
level, suggesting that the substitution of the functional Sr element
may be extremely useful in designing osteogenic and angiogenic
bioactive materials for bone regeneration and bone tissue engi-
neering. The enhancement of Sr component on angiogenic expres-
sion was rarely reported up to now. However, the mechanism of
promoted angiogenic behaviors initiated by Sr component remains
unknown, which needs to be further investigated in details.
2h (°) for (2 1 1) reflectiona Chemical composition
Ca replacement by Sr (mol.%) (Ca + Sr)/P molar ratio
31.52 0 1.66
31.5 1.17 1.64
31.48 3.22 1.66
31.44 5.6 1.67
 K. Lin et al. / Chemical Engineering Journal 222 (2013) 49–59 55

Fig. 5. Schematic illustration of the formation and morphology evolution of 3D

architectured HAp microspheres in the synthetic process.

specific surface areas, mesopores and interconnected macropores

Fig. 3. The FTIR spectra of the control sample and the synthetic SrHAp porous
microspheres.
3.3. In vitro study of drug-loading and release properties
Comparing with traditional nano-particles, the SrHAp micro-
spheres with hierarchically mesoporous structures possess larger
(Table 3 and Fig. 7). Therefore, the synthetic SrHAp microspheres
with hierarchically mesoporous structures have great potential
applications for drug delivery system. Herein, the vancomycin
model was used to investigate the capability of the synthetic SrHAp
microspheres as drug carriers, and the results are listed in Table 3.
The drug-loading amount (DLA: mloaded drug/mcarrier, mg/g) and
drug-loading efficiency (DLE: mloaded drug/mtotal drug amount, %) were
apparently higher than that of the traditional HAp nanoparticles.

Fig. 4. SEM images of the obtained products after hydrothermal treatment at 120 °C for 1 h (A), 3 h (B), 6 h (C) and 72 h (D).
56 K. Lin et al. / Chemical Engineering Journal 222 (2013) 49–59

Fig. 8. The effect of ionic product from control Sr0-HAp and SrHAp porous
Fig. 6. A single nano-sheet (A) was selected for undertaking intensive character- microspheres on proliferation of MG-63 after culture for 1 and 3d. The experi-
ization by SAED (B) and HRTEM (C). mental group compared with the control group of Sr0-HAp nanoparticles at the
same concentration and culture time, p < 0.05.

Table 3
The specific surface area (SBET), drug-loading amount (DLA) and drug-loading molecules and HAp crystals [6]. The higher specific surface area
efficiency (DLE) of the nanoparticles (controlling sample Sr0-HAp) and the synthetic (SBET) of the SrHAp porous microspheres possess higher amount
SrHAp porous microspheres. of OH groups, which will improve the interactions between adsor-
Samples SBET (m2/g) DLA (mg/g) DLE (%) bents and vancomycin molecules, and ultimately increase the DLA
(Table 3).
Sr0-HAp 11.17 19.08 ± 2.70 47.70 ± 6.75
Sr1-HAp 22.77 34.00 ± 0.42 84.99 ± 1.05 It is well known that the physical interaction of hydrogen bond-
Sr3-HAp 20.81 33.56 ± 0.22 83.90 ± 0.54 ing is relatively weak comparing with the chemical bonding. After
Sr5-HAp 29.71 33.36 ± 1.33 83.41 ± 3.32 soaking the drug-loaded SrHAp in aqueous solution, the loaded
drug will be released into the medium due to the dissociation of
hydrogen bonding [48]. Fig. 10 shows the cumulative release re-
The DLA of the SrHAp porous microspheres reached 33.36– sults of vancomycin from the SrHAp porous microspheres and con-
34.00 mg/g, which was about 74.8–78.2% higher than that of the trol HAp nanoparticles in PBS. It is clear that the vancomycin
HAp nanoparticles. The high DLA and DLE were attributed to the no- showed similar release behavior during the whole period, which
vel 3D hierarchically nanostructures with mesoporous structures. had obvious two-step release behavior, an initial fast release and
The mesoporous structures in the SrHAp microspheres increase a relatively slow subsequent release. However, the drug release
the specific surface area (SBET), which is a vital factor for high rate of HAp nanoparticles was significantly faster than that of
DLA. Simultaneity, the mesopores among the microspheres not only SrHAp porous microspheres. For HAp nanoparticles, the drug re-
decreased the diffusion limitations for loading drug molecules, but lease rate was remarkably higher than that of SrHAp porous micro-
also provided space for attracting the vancomycin. In addition, the spheres, around 9 wt.% of total drug loading in the first 3 h. After
DLA is greatly related to the intrinsic nature of carrier materials. 5 days, the amount of released vancomycin reached 24.28 wt.%.
Vancomycin can adsorb on HAp through the hydrogen bonding While for SrHAp porous microspheres, an initial burst release in
interactions between the OH group, which both existed in the drug the first 3 h was around 6 wt.% in PBS of total amount of vancomy-
cin. The subsequent release rate remarkably decreased with time,
and the cumulative release crept during the next 5 days, reaching
a maximum value of 16.07–17.68 wt.% in PBS. The lower release
rate was due to the novel 3D hierarchically nanostructures with
mesoporous structures. In clinical field, the minimum inhibitory
concentration, the minimum bactericidal concentration and the
breakpoint sensitivity of vancomycin for Staphylococcus aureus
were 1.01, 1.98 and 5 lg/mL, respectively [49]. The vancomycin
concentrations released from SrHAp porous microspheres during
the research period was around 280–580 lg/mL (Fig. 11), which
was significantly higher than that of the breakpoint sensitivity of
vancomycin and the published data [49–51]. Therefore, it can be
believed that the released concentration of the objective drug is
high enough for the local therapeutic need. However, the in vivo
study needs to be done to further confirm the therapeutics of
SrHAp porous microspheres.
Our results show the most promising applications of the fabri-
cated SrHAp porous microspheres in the fields of drug delivery sys-
tem at a sustained release rate [6,10], injectable bone regeneration
biomaterials and cell/drug loaded implants due to the higher spe-
Fig. 7. The DFT pore size distribution of the synthetic SrHAp porous microspheres. cific surface area and novel 3D hierarchical architectures, which is
 K. Lin et al. / Chemical Engineering Journal 222 (2013) 49–59 57

Fig. 9. Real-time PCR analysis of the expression levels of ALP (A), Col I (B), Cbfa1 (C) and VEGF (D) mRNA after culture of MG-63 cells in the extract concentration of 100 mg/
mL from control Sr0-HAp nanoparticles and Sr3-HAp porous microspheres, p < 0.05.

Fig. 11. The vancomycin concentrations released from HAp nanoparticles and
Fig. 10. The cumulative vancomycin release ratio from HAp nanoparticles and
SrHAp porous microspheres in PBS.
SrHAp porous microspheres in PBS.

superior to particles [11,12]. In addition, the SrHAp porous micro-

spheres might be used as the component to fabricate the compos-
ite biomaterials because of the excellent biological performances.
4. Conclusions
Herein, the Sr-substituted mesoporous HAp microspheres
(SrHAp) were successfully synthesized via hydrothermal method
in the absence of any surfactants, organic solvents or template-
directing reagents. Comparing with the traditional HAp nanoparti-
cles, the ionic product of SrHAp porous microspheres could appar-
ently stimulate the proliferation and osteogenic differentiation of
MG-63 at certain concentrations of Sr2+ ion. In which, 3 mol.% Sr
dosage was optimal for bone regeneration. Moreover, the interest-
ing phenomenon of the incorporation of Sr into HAp bioceramics
leading to the promoting of angiogenic behaviors was observed.
Therefore, SrHAp might be favorable in vivo angiogenesis during
58 K. Lin et al. / Chemical Engineering Journal 222 (2013) 49–59
bone repair, and can be used as a promising scaffold in bone regen-
eration field. In addition, the synthetic SrHAp porous microspheres
exhibited higher drug-loading capacity and sustained drug release
properties due to their novel 3D hierarchical mesoporous struc-
tures, high specific surface area, and hydrogen bonding interac-
tions between the carriers and drugs. The present study suggests
that the fabricated SrHAp porous microspheres might be greatly
suitable for applications as new bioactive materials in bone regen-
eration and drug-delivery applications.
Acknowledgements
The authors gratefully acknowledge the support of Natural Sci-
ence Foundation of China (Nos.: 81171458, 81150007, 81190132,
81170655), Funds of the Shanghai Institute of Ceramics, Chinese
Academy of Sciences for Innovation of Science and Technology
(No.: Y26ZC1110G), and Shanghai Municipal Education Commis-
sion (No.: 12YZ039).
References
[1] L.L. Hench, J.M. Polak, Third-generation biomedical materials, Science 295
(2002) 1014–1017.
[2] J.C. Elliott, Structure and chemistry of the apatites and other calcium
orthophosphates, Elsevier, Amsterdam, 1994.
[3] H.P. Yuan, H. Fernandes, P. Habibovic, J. De Boer, A.M.C. Barradas, A. De Ruiter,
W.R. Walsh, C.A. Van Blitterswijk, J.D. De Bruijn, Osteoinductive ceramics as a
synthetic alternative to autologous bone grafting, Proc. Natl. Acad. Sci. USA 107
(2010) 13614–13619.
[4] N. Saranya, S. Saravanan, A. Moorthi, B. Ramyakrishna, N. Selvamurugan,
Enhanced osteoblast adhesion on polymeric nano-scaffolds for bone tissue
engineering, J. Biomed. Nanotechnol. 7 (2011) 238–244.
[5] K.L. Lin, L. Chen, J. Chang, Fabrication of dense hydroxyapatite nano-
bioceramics with enhanced mechanical properties via two-step sintering
process, Int. J. Appl. Ceram. Technol. 9 (2012) 479–485.
[6] K.L. Lin, Y.L. Zhou, Y. Zhou, H.Y. Qu, F. Chen, Y.J. Zhu, J. Chang, Biomimetic
hydroxyapatite porous microspheres with co-substituted essential trace
elements: surfactant-free hydrothermal synthesis, enhanced degradation and
drug release, J. Mater. Chem. 21 (2011) 16558–16565.
[7] K.L. Lin, J. Chang, X.G. Liu, L. Chen, Y.L. Zhou, Synthesis of the element-
substituted hydroxyapatite with controllable morphologies and chemical
compositions using calcium silicate as precursor, CrystEngComm 13 (2011)
4850–4855.
[8] E. Landi, A. Tampieri, G. Celotti, S. Sprio, M. Sandri, G. Logroscino, Sr-
substituted hydroxyapatites for osteoporotic bone replacement, Acta
Biomater. 3 (2007) 961–969.
[9] C. Capuccini, P. Torricelli, F. Sima, E. Boanini, C. Ristoscu, B. Bracci, Strontium-
substituted hydroxyapatite coatings synthesized by pulsed-laser deposition:
in vitro osteoblast and osteoclast response, Acta Biomater. 4 (2008) 1885–
1893.
[10] L. Wu, Y.D. Dou, K.L. Lin, W.Y. Zhai, W.G. Cui, J. Chang, Hierarchically structured
nanocrystalline hydroxyapatite assembled hollow fibers as a promising
protein delivery system, Chem. Commun. 47 (2011) 11674–11676.
[11] X.Q. Gong, S.L. Peng, W.J. Wen, P. Sheng, W.H. Li, Design and fabrication of
magnetically functionalized core/shell microspheres for smart drug delivery,
Adv. Funct. Mater. 19 (2009) 292–297.
[12] H.J. Chung, T.G. Park, Surface engineered and drug releasing pre-fabricated
scaffolds for tissue engineering, Adv. Drug Deliv. Rev. 59 (2007) 249–262.
[13] G. Wang, H. Wang, Z.Y. Ren, J.T. Bai, J.B. Bai, PVP-assisted assembly of
lanthanum carbonate hydroxide with hierarchical architectures and their
luminescence properties, Chem. Eng. J. 214 (2013) 386–393.
[14] D.H. Jiang, W.B. Hu, H.R. Wang, B. Shen, Y.D. Deng, Synthesis, formation
mechanism and photocatalytic property of nanoplate-based copper sulfide
hierarchical hollow spheres, Chem. Eng. J. 189–190 (2012) 443–450.
[15] L.H. Ai, Y. Zeng, Hierarchical porous NiO architectures as highly recyclable
adsorbents for effective removal of organic dye from aqueous solution, Chem.
Eng. J. 215–216 (2013) 269–278.
[16] M.G. Ma, J.F. Zhu, Solvothermal synthesis and characterization of
hierarchically nanostructured hydroxyapatite hollow spheres, Eur. J. Inorg.
Chem. 36 (2009) 5522–5526.
[17] C.M. Zhang, J. Yang, Z.W. Quan, P.P. Yang, C.X. Li, Z.Y. Hou, J. Lin,
Hydroxyapatite nano- and microcrystals with multiform morphologies:
controllable synthesis and luminescence properties, Cryst. Growth Des. 9
(2009) 2725–2733.
[18] W.H. He, J.H. Tao, H.H. Pan, X.R. Xu, R.K. Tang, A size-controlled synthesis of
hollow apatite nanospheres at water–oil interfaces, Chem. Lett. 39 (2010)
674–675.
[19] M.G. Ma, Hierarchically nanostructured hydroxyapatite: hydrothermal
synthesis, morphology control, growth mechanism, and biological activity,
Int. J. Nanomed. 7 (2012) 1781–1791.
[20] D. Veljovic, E. Palcevskis, A. Dindune, S. Putic, I. Balac, R. Petrovic, D.
Janackovic, Microwave sintering improves the mechanical properties of
biphasic calcium phosphates from hydroxyapatite microspheres produced
from hydrothermal processing, J. Mater. Sci. 45 (2010) 3175–3183.
[21] K.L. Lin, J. Chang, Y.J. Zhu, W. Wu, G.F. Cheng, Y. Zeng, M.L. Ruan, A facile one-
step surfactant-free and low-temperature hydrothermal method to prepare
uniform 3D structured carbonated apatite flowers, Cryst. Growth Des. 9 (2009)
177–181.
[22] I.S. Neira, Y. Kolen’ko, O.I. Lebedev, G.V. Tendeloo, H.S. Gupta, F. Guitian, M.
Yoshimura, An effective morphology control of hydroxyapatite crystals via
hydrothermal synthesis, Cryst. Growth Des. 9 (2009) 466–474.
[23] S.P. Parthiban, I.Y. Kim, K. Kikuta, C. Ohtsuki, Effect of urea on formation of
hydroxyapatite through double-step hydrothermal processing, Mater. Sci. Eng.
31C (2011) 1383–1388.
[24] S. Koutsopoulos, Synthesis and characterization of hydroxyapatite crystals: a
review study on the analytical methods, J. Biomed. Mater. Res. 62 (2002) 600–
612.
[25] C.Y. Yang, B.C. Wang, E. Chang, J.D. Wu, The influences of plasma spraying
parameters on the characteristics of hydroxyapatite coatings: a quantitative
study, J. Mater. Sci.: Mater. Med. 6 (1995) 249–257.
[26] H. Li, K.A. Khor, P. Cheang, Titanium dioxide reinforced hydroxyapatite
coatings deposited by high velocity oxy-fuel (HVOF) spray, Biomaterials 23
(2002) 85–91.
[27] Biological evaluation of medical devices – Part 1. Evaluation and testing within
a risk management process, ISO 10993-1, 2009.
[28] L.Z. Zhao, L. Liu, Z.F. Wu, Y.M. Zhang, P.K. Chu, Effects of micropitted/
nanotubular titania topographies on bone mesenchymal stem cell osteogenic
differentiation, Biomaterials 33 (2012) 2629–2641.
[29] Z.Y. Li, W.M. Lam, C. Yang, B. Xu, G.X. Ni, S.A. Abbah, K.M.C. Cheung, K.D.K. Luk,
W.W. Lu, Chemical composition, crystal size and lattice structural changes
after incorporation of strontium into biomimetic apatite, Biomaterials 28
(2007) 1452–1460.
[30] R. Viswanatha, D.D. Sarma, Growth of Nanocrystals in Solution, in: C.N.R. Rao,
A. Muller, A.K. Cheetham (Eds.), Nanomaterials Chemistry, WILEY-VCH Verlag
GmbH & Co, KGaA, Weinheim, 2007.
[31] D. Vanmaekelbergh, Self-assembly of colloidal nanocrystals as route to novel
classes of nanostructured materials, Nano Today 6 (2011) 419–437.
[32] J.S. Xu, Y.J. Zhu, C-Fe2O3 and Fe3O4 magnetic hierarchically nanostructured
hollow microspheres: preparation, formation mechanism, magnetic property,
and application in water treatment, J. Colloid Interface Sci. 385 (2012) 58–65.
[33] J. Zhang, S. Liu, J. Lin, H. Song, J. Luo, E.M. Elssfah, E. Ammar, Y. Huang, X. Ding,
J. Gao, S. Qi, C. Tang, Self-assembly of flowerlike AlOOH (boehmite) 3D
nanoarchitectures, J. Phys. Chem. 110B (2006) 14249–14252.
[34] B.X. Li, Y. Xie, Y. Xue, Controllable synthesis of CuS nanostrucutres from self-
assembled precursors with biomolecule assistance, J. Phys. Chem. 111C (2007)
12181–12187.
[35] S. Thirumalairajan, K. Girija, I. Ganesh, D. Mangalaraj, C. Viswanathan, A.
Balamurugan, N. Ponpandian, Controlled synthesis of perovskite LaFeO3
microsphere composed of nanoparticles via self-assembly process and their
associated photocatalytic activity, Chem. Eng. J. 209 (2012) 420–428.
[36] K.S.W. Sing, Reporting physisorption data for gas solid systems-with special
reference to the determination of surface-area and porosity, Pure Appl. Chem.
54 (1982) 2201–2218.
[37] M.L. Zhang, W.Y. Zhai, K.L. Lin, H.B. Pan, W. Lu, J. Chang, Synthesis, in vitro
hydroxyapatite forming ability, and cytocompatibility of strontium silicate
powders, J. Biomed. Mater. Res. 93B (2010) 252–257.
[38] E. Canalis, M. Hott, P. Deloffre, Y. Tsouderos, P.J. Marie, The divalent strontium
salt S12911 enhances bone cell replication and bone formation in vitro, Bone
18 (1996) 517–523.
[39] E. Bonnelye, A. Chabadel, F. Saltel, P. Jurdic, Dual effect of strontium ranelate:
stimulation of osteoblast differentiation and inhibition of osteoclast formation
and resorption in vitro, Bone 42 (2008) 129–138.
[40] W.E. Cabrera, I. Schrooten, M.E. De Broe, P.C.D. Haese, Strontium and bone, J.
Bone Miner. Res. 14 (1999) 661–668.
[41] S.C. Skoryna, Metabolic aspects of the pharmacologic use of trace elements in
human subjects with specific reference to stable strontium, Trace Subst.
Environ. Health 18 (1984) 3–20.
[42] C. Banerjee, L.R. McCabe, J.Y. Choi, Runt homology domain proteins in
osteoblast differentiation: AML3/CBFA1 is a major component of a bone-
specific complex, J. Cell Biochem. 66 (1997) 1–8.
[43] S.J. Roberts, Y. Chen, M. Moesen, J. Schrooten, F.P. Luyten, Enhancement of
osteogenic gene expression for the differentiation of human periosteal derived
cells, Stem Cell Res. 7 (2011) 137–144.
[44] S.K.L. Levengood, M.J. Poellmann, S.G. Clark, D.A. Ingram, M.C. Yoder, A.J.W.
Johnson, Human endothelial colony forming cells undergo vasculogenesis
within biphasic calcium phosphate bone tissue engineering constructs, Acta
Biomater. 7 (2011) 4222–4228.
[45] E.C. Novosel, C. Kleinhans, P.J. Kluger, Vascularization is the key challenge in
tissue engineering, Adv. Drug Deliver. Rev. 63 (2011) 300–311.
[46] P. Tayalia, D.J. Mooney, Controlled growth factor delivery for tissue
engineering, Adv. Mater. 21 (2009) 3269–3285.
 K. Lin et al. / Chemical Engineering Journal 222 (2013) 49–59
[47] A.A. Gorustovich, C. Perio, J.A. Roether, A.R. Boccaccini, Effect of bioactive
glasses on angiogenesis: a review of in vitro and in vivo evidences, Tissue. Eng.
16 (2010) 199–207.
[48] J.X. Gu, F. Xia, Y. Wu, X.Z. Qu, Z.Z. Yang, L. Jiang, Programmable delivery of
hydrophilic drug using dually responsive hydrogel cages, J. Control. Release
117 (2007) 396–402.
[49] S.W.N. Ueng, L.J. Yuan, N. Lee, S.S. Lin, S.J. Liu, E.C. Chan, J.H. Weng, In vivo
study of hot compressing molded 50:50 poly (DL-lactide-co-glycolide)
antibiotic beads in rabbits, J. Orthop. Res. 10 (2002) 654–661.
59
[50] X. Liu, Z.P. Xie, C.Q. Zhang, H.B. Pan, M.N. Rahaman, X. Zhang, Q. Fu, W.H.
Huang, Bioactive borate glass scaffolds: in vitro and in vivo evaluation for use
as a drug delivery system in the treatment of bone infection, J. Mater. Sci.:
Mater. Med. 21 (2010) 575–582.
[51] B. Li, K.V. Brown, J.C. Wenke, S.A. Guelcher, Sustained release of vancomycin
from polyurethane scaffolds inhibits infection of bone wounds in a rat femoral
segmental defect model, J. Control. Release 145 (2010) 221–230.