G Model ARTICLE IN PRESS COLSUB-6941; No of Pages Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx Contents lists available at ScienceDirect Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb Biomimetic scaffolds based on hydroxyapatite nanorod/poly(d,l) lactic acid with their corresponding apatite-forming capability and biocompatibility for bone-tissue engineering Nguyen Kim Nga a,∗ , Tran Thanh Hoai a , Pham Hung Viet b a b School of Chemical Engineering, Hanoi University of Science and Technology, Dai Co Viet Road, Hanoi, Viet Nam Research Center for Environmental Technology and Sustainable Development, Hanoi University of Science, 334 Nguyen Trai Street, Hanoi, Viet Nam a r t i c l e i n f o Article history: Received 14 October 2014 Received in revised form 27 February 2015 Accepted March 2015 Available online xxx Keywords: Biomimetic scaffolds Poly(d,l) lactic acid Hydroxyapatite nanorods Apatite Biocompatibility Bone tissue engineering a b s t r a c t This study presents a facile synthesis of biomimetic hydroxyapatite nanorod/poly(d,l) lactic acid (HAp/PDLLA) scaffolds with the use of solvent casting combined with a salt-leaching technique for bone-tissue engineering Field emission scanning electron microscopy, Fourier transform infrared spectroscopy, and energy-dispersive X-ray spectroscopy were used to observe the morphologies, pore structures of synthesized scaffolds, interactions between hydroxyapatite nanorods and poly(d,l) lactic acid, as well as the compositions of the scaffolds, respectively Porosity of the scaffolds was determined using the liquid substitution method Moreover, the apatite-forming capability of the scaffolds was evaluated through simulated body fluid (SBF) incubation tests, whereas the viability, attachment, and distribution of human osteoblast cells (MG 63 cell line) on the scaffolds were determined through alamarBlue assay and confocal laser microscopy after nuclear staining with ,6-diamidino-2-phenylindole and actin filaments of a cytoskeleton with Oregon Green 488 phalloidin Results showed that hydroxyapatite nanorod/poly(d,l) lactic acid scaffolds that mimic the structure of natural bone were successfully produced These scaffolds possessed macropore networks with high porosity (80–84%) and mean pore sizes ranging 117–183 m These scaffolds demonstrated excellent apatite-forming capabilities The rapid formation of bone-like apatites with flower-like morphology was observed after days of incubation in SBFs The scaffolds that had a high percentage (30 wt.%) of hydroxyapatite demonstrated better cell adhesion, proliferation, and distribution than those with low percentages of hydroxyapatite as the days of culture increased This work presented an efficient route for developing biomimetic composite scaffolds, which have potential applications in bone-tissue engineering © 2015 Elsevier B.V All rights reserved Introduction Bone repair and regeneration have become a serious challenge in orthopedic surgery because of the increase in clinical bone diseases (e.g., bone infections, bone tumors, and bone loss through trauma) [1] Current therapies for bone defects include autografts, allografts, and other bone substitutes [2] Autografts (bones obtained from another anatomical site in the same subject) comprise the gold standard for the treatment of bone defects, but this surgical method still has major disadvantages, such as the possibility of donor site morbidity, shortage of donor bone supply, anatomical and structural complications, as well as graft ∗ Corresponding author Tel.: +84 38680 110; fax: +84 38680 070 E-mail address: nga.nguyenkim@hust.edu.vn (N.K Nga) sorption [1] Allografts (bones from donors or another species) can be used as an alternative; however, this method causes inherent problems (e.g., disease transmission and immunogenic responses) [3] Synthetic bone substitutes, which are mostly made of metals or bioceramics and glasses, have osteoconductive properties instead of osteoinductive properties, thus limiting their use [4] Bone-tissue engineering (BTE) has attracted scientific attention for being more effective than conventional methods in terms of bone repair and reconstruction The core of BTE is to combine a biodegradable matrix (scaffold) and living cells to grow tissue in vitro prior to implantation of the subject [5] Scaffolds for BTE should possess several properties, such as high porosity, a macroporous network for in vitro cell migration, adhesion, proliferation, and further tissue growth, biocompatibility and biodegradability to non-toxic products, as well as sufficient mechanical strength [1] To achieve these properties, BTE scaffolds are often designed to mimic the structural http://dx.doi.org/10.1016/j.colsurfb.2015.03.001 0927-7765/© 2015 Elsevier B.V All rights reserved Please cite this article in press as: N.K Nga, et al., Biomimetic scaffolds based on hydroxyapatite nanorod/poly(d,l) lactic acid with their corresponding apatite-forming capability and biocompatibility for bone-tissue engineering, Colloids Surf B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.03.001 G Model COLSUB-6941; No of Pages ARTICLE IN PRESS N.K Nga et al / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx and biological functions of a naturally occurring extracellular matrix (ECM) in terms of both chemical composition and physical structure [6,7] In this regard, biodegradable polymers, such as poly(l-lacticco-glycolic acid) [8], poly(-caprolactone) [9], poly(l-lactic acid) [10], and poly(d,l-lactic acid) (PDLLA) [11], have been processed into three-dimensional (3D) scaffolds These polymers show good mechanical properties, with shapes and degradation rates that are easily adjustable The main disadvantage of these polymeric materials is their pure biocompatibility, given that they not provide a favorable surface for cell attachment and proliferation because of the lack of specific cell-recognizable signals [1] Bioactive inorganic materials, such as hydroxyapatite (HAp) [12], -tricalcium phosphate [13,14], and bioactive glasses [15,16], have been designed as 3D porous scaffolds for bone regeneration The advantages of bioactive ceramics are their excellent osteoconductivity and biocompatibility; however, their inherent brittleness and low mechanical strength (for porous specimens) are major disadvantages in developing 3D scaffolds, thus limiting their applications [13,15,17] Polymer/bioceramic composite scaffolds are attracting increasing attention in the field of bone regeneration because of their mechanical stability and biocompatibility [18,19] Among such scaffolds, HAp/polymer composites have received considerable interest because of their composition and structural similarity to natural bone [20,21] Bone is a complex composite that comprises an organic phase (90% type I collagen, other non-collagenous proteins (e.g., proteoglycans)), minor amounts of lipids and osteogenic factors (e.g., bone morphogenetic proteins), and a mineral phase [22] The mineral phase consists of one or more types of calcium phosphates comprising 65–70% of bone and embedded in a protein matrix [22–24] Among the CaP salts, hydroxyapatite (Ca10 (PO4 )6 (OH)2 , HAp) is the most stable calcium phosphate in body fluids and is the most similar to the mineral part of bone [23] The biological HAps found in physiological hard tissues are irregularly rod-like or plate-like crystals of variable lengths and widths (30–45 nm) with a thickness of approximately nm [22] These HAps are oriented to the c-axis, which is parallel to the collagen fibrils [22,25] in which different cell types, including osteoblasts, osteoclasts, and osteocytes, reside A survey of the literature shows that bioceramics that mimic bone mineral in terms of composition, structure, and morphology can promote osteointegration and subsequent bone tissue formation [26–28] PDLLA is a nontoxic, biocompatible, and biodegradable material that has been used as sutures and tissue-engineering scaffolds [29] However, PDLLA presents strong hydrophobicity owing to the absence of hydrophilic groups and suitable functional groups (e.g., NH2 , OH), which results in the absence of osteoconductivity A recent study focused on fabricating bioactive nanocomposite PDLLA/nano-HAp membranes using an electrospinning method to improve the osteoconductivity of the polymer [30] However, the products are characterized by small pore sizes with a mean diameter of 4.8 m and are thus unsuitable for use as BTE scaffolds, given that the minimum pore size requirement for bone scaffolds is 100 m [31] Several methods have been developed to fabricate 3D scaffolds These methods include fiber bonding, freeze drying, phase separation, super-critical fluid technology, solvent casting combined with particulate leaching, and melt molding [1] Among these methods, solvent casting combined with particulate leaching has been widely used for fabricating 3D scaffolds because of its simplicity and efficiency This method allows to produce highly porous scaffolds with porosity up to 93% and mean pore diameter up to 500 m by varying porogen particles size, and weight ratio of polymer to porogen without the need of the specialized equipment [1,32] As a result, in this study, solvent casting combined with particulate-leaching method has been used as an efficient control to synthesize 3D HAp nanorod/poly(d,l) lactic acid (HAp/PDLLA) scaffolds The rodshaped HAp nanoparticles with the same size as bone minerals were successfully prepared in past studies [33,34] and have since been used as an inorganic phase to incorporate into the PDLLA matrix and to prepare biomimetic HAp/PDLLA scaffolds for BTE The apatite-forming capability of the scaffolds was determined by assessing the formation of bone-like apatite on the surface of the scaffolds after immersing in simulated body fluids (SBFs) Meanwhile, biocompatibility was investigated in direct contact with human osteoblast cell line MG 63 through in vitro tests Experimental 2.1 Chemicals All reagents were of analytical grade and used as received without further purification Calcium chloride dihydrate (CaCl2 ·2H2 O), sodium monophosphate dihydrate (Na2 HPO4 ·2H2 O), NaOH, C2 H5 OH, 1,4-dioxane, hydrochloric acid (HCl), pluronic co-polymer PEO20 –PPO70 –PEO20 (P123), NaCl, NaHCO3 , KCl, K2 HPO4 ·3H2 O, MgCl2 ·6H2 O, Na2 SO4 , Tris-hydroxymethyl aminomethane ((HOCH2 )3 CNH2 ), phosphate buffered saline (PBS) were obtained from Sigma–Aldrich PDLLA was purchased from Boehringer Ingelheim (Ingelheim, Germany) Deionized water was used to prepare all solutions and reagents 2.2 Scaffold production The HAp nanorods used in this study were prepared using the hydrothermal method assisted by a pluronic co-polymer PEO20 –PPO70 –PEO20 The preparation and characterizations of these HAp nanorods were conducted in the same manner as in our previous work [33] HAp nanorod/PDLLA scaffolds were then prepared using solvent casting combined with salt-leaching method with NaCl as the porogen A 7% polymer solution was produced by dissolving 0.455 g of PDLLA in 6.5 mL of 1,4-dioxane for h at 32 ◦ C A suspension of C2 H5 OH with various amounts of HAp nanorods (0–30 wt.% HAp to PDLLA) was added dropwise into the PDLLA solution The resulting mixture was vigorously stirred on a magnetic stirrer at a speed of 500 rpm for h to achieve homogeneity The homogeneous mixture was then cast into a 55 mm glass Petri dish containing g of NaCl with particle sizes of 300–450 m The samples were then air-dried under a chemical hood for 24 h and vacuum-dried for another 24 h to remove all solvents The resulting scaffolds were then immersed in distilled water at 35 ◦ C for d (water was changed 3–4 times each day) to leach out the salt The produced HAp/PDLLA scaffolds were further air-dried and vacuumdried and then stored in a desiccator until use Finally, four scaffolds were obtained according to the percentage of HAp nanorods These scaffolds were labeled as S1, S2, S3, and S4 for wt.% HAp, 10 wt.% HAp, 20 wt.% HAp, and 30 wt.% HAp, respectively 2.3 Scaffold characterization The morphologies and pore structures of the produced scaffolds were examined using field emission scanning electron microscopy (FE-SEM) (Supra 40, Zeiss, Germany) at low (200×) and high (500×, 1000× and 5000×) magnifications, as well as optical microscopy (SMZ800, Nikon, Japan) Prior to FE-SEM observation, the dry samples of the scaffolds were cut into small disks, which were mounted on an SEM stub and then sputter-coated with a thin chromium layer (Quorumtech, Q150T ES Turbo Chromium Sputter-Evaporator) The mean pore diameter and pore wall thickness of the scaffolds were measured by using scanning electron microscopic (SEM) images through ImageJ software The Fourier transform infrared spectra Please cite this article in press as: N.K Nga, et al., Biomimetic scaffolds based on hydroxyapatite nanorod/poly(d,l) lactic acid with their corresponding apatite-forming capability and biocompatibility for bone-tissue engineering, Colloids Surf B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.03.001 G Model COLSUB-6941; No of Pages ARTICLE IN PRESS N.K Nga et al / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx (FTIR) of the scaffolds were recorded on a Nicolet 6700 spectrometer using KBr pellet technique in the range of 4000–400 cm−1 with a resolution of cm−1 All measurements were performed at 25 ◦ C The composition of HAp nanoparticles on the surface of HAp/PDLLA scaffolds was examined through energy-dispersive Xray spectroscopy (EDXS) (Nova NanoSEM 450, FEI) The porosity of the HAp/PDLLA scaffolds was measured using the liquid substitution method [35] Distilled water was used as the displacement liquid In brief, a dry sample of each scaffold was weighed and then immersed in a graduated cylinder containing a known volume V1 of water and left to stand to enable the water to penetrate into the pores of the scaffold sample until no air bubbles emerged from the scaffold The total volume of water and water-impregnated scaffold was recorded as V2 The volume difference (V2 − V1 ) represents the volume of the HAp/PDLLA scaffold skeleton The water-impregnated scaffold was then removed from the graduated cylinder, and the residual water volume was recorded as V3 The total volume of the scaffold is given by V = (V2 − V1 ) + (V1 − V3 ) = (V2 − V3 ) By measuring the initial and final weights Wi and Wf of the scaffolds before and after immersing in water, we can calculate the pore volume of the scaffold as Wf − Wi H2 O (1) The porosity of the scaffold can be calculated using the following equation: Porosity = (Wf − Wi )/ H2 O V2 − V3 (2) At least five measurements were conducted for each scaffold, and the results were averaged from these five measurements 2.4 In vitro mineralization tests In vitro mineralization tests were conducted through the incubation of the HAp/PDLLA scaffolds in SBFs SBF was prepared as described previously [36], filtered using a 0.22 m Millipore filter system to eliminate bacterial contamination, and then stored in a plastic bottle in a refrigerator at ◦ C After disinfection in 70% ethanol at ◦ C, three samples of each scaffold were immersed in 30 mL of SBF solution placed in closed polyethylene containers at 37 ◦ C After being immersed in SBF for the designed days, the samples were removed, gently rinsed with deionized water, and dried under warm flowing air The rinsing process has been used to remove the residual ions of the SBF solutions that remained on the scaffold samples and could affect the scaffold structure All operations were conducted in a laminar airflow hood to avoid bacterial infection The apatite-forming capability of HAp/PDLLA scaffolds was assessed through FE-SEM and EDXS 2.5 In vitro cell culture experiments Human osteoblast cells (MG 63 cell line, IZSLER Biobanking of Veterinary Resources, Brescia, Italy) were used to evaluate biocompatibility of the prepared HAp/PDLLA scaffolds Cells were grown in a minimum essential medium (Gibco) supplemented with 10% (v/v) fetal bovine serum (Gibco), 2% (v/v) l-glutamine (Gibco), 1% (v/v) antibiotic, 1% (v/v) sodium pyruvate, and 1% (v/v) nonessential amino acids (Gibco) at 37 ◦ C and 5% CO2 The medium was changed every d The adherent cells were allowed to reach confluence, then detached with 0.1% trypsin ethylenediaminetetraacetic acid, and suspended in a fresh culture medium for the experiments Prior to cell seeding, the prepared HAp/PDLLA scaffold samples with a diameter of mm were sterilized in 70% ethanol at ◦ C for 24 h, washed with sterile distilled water, and immersed in a culture medium for h before seeding Thereafter, the samples were placed on 48-well plates 20 L of cell suspension at a density of 7.5 × 104 cells per sample was seeded on scaffold samples The samples were then incubated at 37 ◦ C for h to allow cell attachment to the scaffold surfaces, followed by adding 250 L of culture medium into each well Culture plates were then transferred to an incubator at 37 ◦ C and 5% CO2 Culture times were 3, 5, and d, and the fresh culture medium was changed every d Cell viability and proliferation were determined after each incubation time using alamarBlue assay The assays were performed after 3, 5, and d of cell seeding, according to the manufacturer’s protocol Briefly, at the end of each incubation time, the culture medium was removed from the wells and fresh culture medium with 10% (v/v) alamarBlue was added to each well After h of incubation at 37 ◦ C, aliquots of 100 L were pipetted into 96-well plates, and fluorescence was recorded on a microplate reader at an excitation wavelength of 565 nm and an emission wavelength of 595 nm Cell morphology, growth, and distribution were visualized after staining with Oregon Green 488 phalloidin (Life Technologies; Carlsbad, CA, USA) and ,6-diamidino-2-phenylindole (DAPI, Sigma–Aldrich) Cells were fixed with 4% (w/v) formaldehyde in PBS, permeabilized with 0.2% Triton X in PBS, and stained with Oregon Green 488 and DAPI, according to the manufacturer’s protocol After three rinses with PBS, the samples were examined using confocal laser microscopy 2.6 Statistical analysis Results were averaged and expressed as mean ± standard deviation Statistical analysis was performed using ANOVA A value of P < 0.05 was considered statistically significant Results and discussion 3.1 Characterization of the synthesized scaffolds Four HAp/PDLLA scaffolds were synthesized with different weight percentages of HAp (one scaffold without HAp filling and the other three scaffolds with HAp filling) The compositions and other characteristics of the produced scaffolds are summarized in Table Fig I1 (supporting material) show digital camera images of the scaffold samples (S3 and S4 scaffolds) and the optical images of a typical S3 scaffold sample (Fig I2) These scaffolds have stable shapes, thicknesses of mm (Fig I1), and porous structures (Fig I2) Surface morphologies and pore structures were then examined through FE-SEM imaging The SEM images of all synthesized scaffolds are presented in Fig and Fig I3 Observations show that the surface morphologies of the scaffolds become coarser as the HAp content increases A smooth surface morphology was produced for S1 PDLLA scaffolds (Fig 1a and I3A), whereas coarse surface morphologies were obtained for HAp/PDLLA scaffolds HAp nanoparticles were homogeneously deposited within the pore walls of the scaffolds (Fig 1b–d) FE-SEM images with high magnifications (Fig I3B–D) indicated that no aggregates of nanoparticles appeared on the pore walls for all composite scaffolds Moreover, thin pore walls of 1.48 ± 0.28 m were observed (Fig I3B) for S2 scaffold with a low HAp percentage of 10 wt.% A further increase in HAp percentage produced scaffolds with thicker pore walls of 1.81 ± 0.3 and 3.94 ± 0.63 m (Fig I3C and D) for S3 and S4 scaffolds, respectively Fig I3E (FE-SEM image of a higher magnification for S4 scaffold) showed that tiny rod-like HAp particles were embedded on the polymer phase According to Table 1, increasing HAp percentage resulted in a decrease in pore sizes of the produced scaffolds Furthermore, statistical analysis and pore size distribution of S1 PDLLA Please cite this article in press as: N.K Nga, et al., Biomimetic scaffolds based on hydroxyapatite nanorod/poly(d,l) lactic acid with their corresponding apatite-forming capability and biocompatibility for bone-tissue engineering, Colloids Surf B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.03.001 G Model COLSUB-6941; No of Pages ARTICLE IN PRESS N.K Nga et al / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx Table Composition and some characteristics of HAp/PDLLA composite scaffolds Sample codes HAp massa (g) PDLLA mass (g) Weight percent (%) (HAp/PDLLA) Pore sizesb (m) S1 S2 S3 S4 0.0455 0.091 0.136 0.455 0.455 0.455 0.455 10 20 30 274 183 122 117 a b c ± ± ± ± 76 60 58 60 Porosityc (%) 89.39 84.46 83.37 80.18 ± ± ± ± 2.86 2.17 1.81 0.78 Actual amount of HAp in total PDLLA Mean value ± standard deviation (SD); n = 30 Mean value ± standard deviation; n = scaffolds (Fig I4A) demonstrate that these scaffolds have pore sizes, which are significantly larger than those of the HAp/PDLLA scaffolds (I P < 0.001) and are mainly in the range of 100–450 m with a mean pore size of 274 m S2 and S3 HAp/PDLLA scaffolds have mean pore sizes of 183 and 122 m (Table 1) within the range of 100–350 m and 50–300 m, respectively (Fig I4B and C); differences in their pore sizes are statistically significant (I P < 0.001) S4 HAp/PDLLA scaffolds have the smallest pore sizes with a mean pore size of 117 m (Table 1) However, statistical analysis showed that pore sizes of S4 HAp/PDLLA scaffolds are significantly smaller than those of S1 PDLLA and S2 HAp/PDLLA scaffolds (I P < 0.001), but not significantly smaller than those of S3 HAp/PDLLA scaffolds (P > 0.05) The obtained results confirm that HAp content greatly affects the pore structure and overall morphology of the composite scaffolds Higher HAp content resulted in the production of scaffolds with small pore sizes and thick pore walls The past studies demonstrated that pores in scaffolds play an important role in bone tissue formation Large pores (100–150 and 150–200 m) showed substantial bone ingrowth, while pores (75–100 m) resulted in ingrowth of unmineralized osteoid tissue, and smaller pores (10–44 and 44–75 m) were penetrated only by fibrous tissues [37,38] Our results indicated that the produced composite scaffolds have mean pore sizes of 117–183 m, which is believed to be suitable for BTE scaffolds To evaluate the interaction between PDLLA matrix and the inorganic phase, FTIR spectra of nano-HAp (a), typical HAp/PDLLA scaffolds (b and c), and PDLLA scaffolds (d) are presented in Fig 2A The spectra of the HAp/PDLLA scaffolds not exhibit significant differences, which reveal the presence of HAp on the composite scaffolds In fact, the stretching and bending vibrations of the PO4 3− groups for HAp are visible at 1090, 1030, 953, 602, 561, and 469 cm−1 in the spectra of the HAp/PDLLA scaffolds However, these vibrations are detected at 1095, 1032, 955, 604, 563, and 471 cm−1 in the spectrum of the HAp sample A strong band at 1749 cm−1 can be observed in the spectrum of S1 PDLLA scaffold This band is assigned to the vibration of the carbonyl group (C O) of PDLLA However, this band shifts to 1754 and 1752 cm−1 for S3 and S4 HAp/PDLLA scaffolds, respectively In addition, the characteristic peaks for C H vibrations were detected at 2999 and 2951 cm−1 for the S1 PDLLA scaffold, but observed at 2996 and 2946 cm−1 and at 2998 and 2948 cm−1 for S3 and S4 HAp/PDLLA scaffolds These results reveal that some molecular interactions between HAp nanoparticles and PDLLA in the composite scaffolds may have occurred Moreover, the weak peak at 3569 cm−1 is characteristic of the vibration of the OH group of HAp nanoparticles However, this peak appeared at a lower region at 3501 and 3496 cm−1 for the S3 and S4 scaffolds, respectively This result suggests that hydrogen bond was formed between the OH group of HAp and the C O group of PDLLA, thus making the HAp nanoparticles stable in the polymer matrix The presence of HAp on the surface of the HAp/PDLLA scaffolds was verified by EDXS analysis The EDXS spectra and Fig FE-SEM images of scaffolds, synthesized at different percentages of HAp to PDLLA: (a) S1 PDLLA scaffold, (b) S2 HAp/PDLLA scaffold, (c) S3 HAp/PDLLA scaffold, and (d) S4 HAp/PDLLA scaffold at low magnification of (200×) The insets represent FE-SEM images at higher magnification of (500×) Please cite this article in press as: N.K Nga, et al., Biomimetic scaffolds based on hydroxyapatite nanorod/poly(d,l) lactic acid with their corresponding apatite-forming capability and biocompatibility for bone-tissue engineering, Colloids Surf B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.03.001 G Model COLSUB-6941; No of Pages ARTICLE IN PRESS N.K Nga et al / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx 3.2 Apatite-forming capability Fig (A) FTIR spectra of (a) nano-HAp powder, (b) S3 HAp/PDLLA scaffold, (c) S4 HAp/PDLLA scaffold, and (d) S1 PDLLA scaffold and (B) EDXS spectrum of S4 HAp/PDLLA scaffold components of a typical S4 HAp/PDLLA scaffold are illustrated in Fig 2B Three elements, namely, O, P, and Ca, are the major constituents of the synthesized HAp/PDLLA scaffold with 45.03 ± 1.63 at.%, 5.64 ± 0.47 at.%, and 9.58 ± 0.52 at.%, respectively A trace of Cl (0.71 ± 0.06 at.%) was detected, which could be attributed to a residue of the synthesis reaction, and was incompletely eliminated from the scaffold sample through washing Moreover, the presence of C in the scaffold samples can possibly be attributed to the use of carbon adhesive tape to mount the samples or the presence of C in the polymer matrix Since delicate particles could easily be affected, therefore carbon adhesive tape could affect the scaffold morphology However, the Ca/P ratio was 1.69, which was close to the stoichiometric value of 1.67 This result confirms that the HAp nanoparticles were successfully incorporated into the scaffolds The porosity data of the synthesized scaffolds are shown in Table The experimental results indicate that the porosity changes linearly with the increase in HAp content and exhibits a downward trend with HAp loading Within the HAp content range studied, the porosity values varied from 89.39 ± 2.86% to 80.18 ± 0.78%, which show relatively high porosity for all produced scaffolds The high porosity levels of the synthesized scaffolds suggested that these scaffolds would be beneficial for in vitro cell adhesion, ingrowth, and survival An important characteristic of the scaffolds is their capability to form an apatite layer on their surfaces Fig presents the surface morphologies of typical HAp/PDLLA scaffolds after being immersed in SBF for and d The FE-SEM image of S3 HAp/PDLLA scaffold with low magnification (Fig 3a) shows that a mineral layer was already formed This layer covered the scaffold surface after d of soaking in SBF The image with higher magnification (Fig 3b) confirms that the mineral layer consisted of aggregated flower-like particles with a mean diameter of 1.08 m These particles were formed on the surface and inside the scaffold pores The growth of the flower-like particles was observed when the soaking time was prolonged to d As shown in the low-resolution FE-SEM image in Fig 3c, the particles grew larger (their mean diameter increased to 1.81 m) The FE-SEM image with a high resolution (Fig 3d) indicates the formation of the flower-like mineral layer consisting of tiny needle-like crystals on the scaffold surface after d of soaking in SBF The surface morphologies of S4 HAp/PDLLA scaffold after d of soaking in SBF are similar to those of the S3 HAp/PDLLA scaffold However, the intense formation of such particles was observed on the entire surface of the S4 HAp/PDLLA scaffold (Fig 3e) The highresolution FE-SEM image (Fig 3f) showed that a rose-like mineral layer was exclusively produced for the S4 HAp/PDLLA scaffold This mineral layer morphology is typical for bone-like apatite [36,39] The obtained results confirm the complete formation of the mineral layer on the surfaces of both S3 and S4 HAp/PDLLA scaffolds after d of soaking in SBF To explore the effect of HAp nanoparticles on the apatite-forming capability of the composite scaffolds, in vitro mineralization experiments of pure PDLLA and HAp/PDLLA scaffolds with 10 wt.% of HAp were examined Fig I5 shows the FE-SEM images of the S1 PDLLA scaffold after being immersed in SBF for and d and that of S2 HAp/PDLLA scaffold after d Only several spherical-like mineral particles formed on the surface of the S1 PDLLA scaffold after d of soaking in SBF (Fig I5A) The number of the mineral particles increased after d in SBF, and some bundles of aggregated mineral-like crystals (Fig I5B) were found on the scaffold surface because of the conjunction of such particles, indicating poor mineralization-forming capability The addition of HAp at 10 wt.% to the PDLLA scaffold almost did not improve the mineralization-forming capability of the composite scaffolds The FE-SEM image in Fig I5C showed that the severe aggregation of mineral-like crystals was observed on the surface of the S2 HAp/PDLLA scaffold after being immersed in SBF for d Meanwhile, the HAp/PDLLA scaffolds with high HAp amounts (20 and 30 wt.%) already showed full coverage by the mineral layer with flower-like morphology on their surfaces after d, which revealed a significantly higher in vitro mineralization response than those of pure PDLLA scaffolds and scaffolds with 10 wt.% of HAp Mineralization involves the nucleation and growth of bone-like apatite onto biomaterials, which is associated with uptake of calcium and phosphate ions from the physiological environment The HAp/PDLLA scaffolds with high HAp percentages provide more nucleation sites (Ca2+ ) for apatite formation than the same composite scaffolds with low HAp percentages (e.g., 10 wt.% HAp) and accordingly demonstrate better in vitro mineralization Our results indicate that the HAp content in the scaffolds significantly affects the induction of the mineral layer on their surfaces HAp content of 20–30 wt.%, which is relative to PDLLA is optimal for producing HAp/PDLLA scaffolds with high in vitro mineralization The chemical composition of the released mineral layer was further analyzed using EDXS method EDXS analyses were conducted for the typical composite scaffolds after being soaked in SBF Table I1 presents the elemental compositions of the mineral layers released on S3 HAp/PDLLA scaffold after being soaked in SBF for and d Ca, P, and O are three main elements found in the Please cite this article in press as: N.K Nga, et al., Biomimetic scaffolds based on hydroxyapatite nanorod/poly(d,l) lactic acid with their corresponding apatite-forming capability and biocompatibility for bone-tissue engineering, Colloids Surf B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.03.001 G Model COLSUB-6941; No of Pages ARTICLE IN PRESS N.K Nga et al / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx Fig FE-SEM images of representative HAp/PDLLA composite scaffolds after immersion in SBF: (a and b) at d with magnifications of (5000×) and (10,000×), and (c and d) at d with magnifications of (5000×) and (50,000×) for S3 HAp/PDLLA scaffold; (e and f) at d with magnifications of (5000×) and (50,000×) for S4 HAp/PDLLA scaffold released mineral layers after soaking in SBF for and d However, Na and Cl were detected as trace elements (for the mineral layer, released after d in SBF), which was attributed to the residue from the reaction synthesis of the scaffolds According to Table I1, the Ca/P ratios were 1.24 and 1.55 for the mineral layers produced on the scaffold after and d of soaking in SBF, respectively Apart from apatite mineral, a number of other calcium phosphate minerals (e.g., amorphous calcium phosphate, dicalcium phosphate dihydrate, octacalcium phosphate, tricalcium phosphate, as well as ␣- and -Ca3 (PO4 )2 ) can be produced under the same in vitro conditions that form apatite The results suggest that after d in SBF a new bone-like apatite phase was already produced on the S3 scaffold, but after d in SBF dicalcium phosphate and/or octacalcium phosphate as intermediate phases were probably released on the scaffold Moreover, the Ca/P ratio of the produced mineral layer was 1.55, which was lower than the stoichiometric value of 1.67, and was attributed to the calcium-deficient carbonated hydroxyapatite, which suggests that the formed apatite is carbonated Table I2 compares the apatite-forming capability of the HAp/PDLLA scaffolds as synthesized in this study with other composite materials reported in previous studies The complete formation of the bone-like apatite layer was observed after d of incubation in SBF for the HAp/PDLLA scaffolds in this study However, the same layer was obtained after a longer time (21 and 14 d) for the nano-HAp/PDLLA films [40] and Akermanite/PDLLA scaffolds [41], respectively Moreover, the bone-like apatite layer with flower-like morphology was obtained for the HAp/PDLLA scaffolds synthesized in this study (Fig 3e and f) Meanwhile, the bone-like apatite layer formed on nano-HAp/PDLLA films consisted of numerous uniform-orbicular aggregates, and a worm-like morphology was produced on Akermanite/PDLLA scaffolds, which significantly differ from those released on the HAp/PDLLA scaffolds in this study The comparison revealed that the HAp/PDLLA scaffolds demonstrated better in vitro mineralization than the other composite materials, which could be attributed to the similarity (e.g., in morphology and composition) of the inorganic component of the scaffolds to that of natural bone As shown in Table I2, the inorganic component of the HAp/PDLLA scaffolds in this study is composed of rod-shaped HAp particles with diameter ranging from 25 to 30 nm and length ranging from 100 to 130 nm, which is similar to the values for bone minerals The inorganic components of the other materials exhibit morphological and compositional differences from those of the HAp/PDLLA scaffolds and bone minerals (see Table I2) Based on the comparison of the in vitro mineralization in the present and previous studies, the morphology and composition of the inorganic part of the scaffolds have a crucial function in their bioactivity and in promoting bone-like apatite formation Please cite this article in press as: N.K Nga, et al., Biomimetic scaffolds based on hydroxyapatite nanorod/poly(d,l) lactic acid with their corresponding apatite-forming capability and biocompatibility for bone-tissue engineering, Colloids Surf B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.03.001 G Model COLSUB-6941; No of Pages ARTICLE IN PRESS N.K Nga et al / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx Fig Cell viability and proliferation on the scaffolds after 3, 5, and d of culture through alamarBlue assay All data are expressed as means ± SD; n = *P < 0.001 (data compared with those at longer culture time, and with S1 PDLLA scaffold), **P < 0.01 and ***P < 0.05 (data compared with S1 PDLLA) 3.3 In vitro biocompatibility Cell–scaffold interactions are the basis of initial cell attachment and influence cell phenotypes and functions In this study, in vitro cell culture tests were conducted to evaluate biocompatibility of the synthesized scaffolds in terms of cell viability, proliferation, and attachment Fig shows results of the alamarBlue assay, which was performed after MG 63 cells cultured on the scaffolds and the cell culture plates (served as the control group) for 3, 5, and d The alamarBlue reduction indicated the viability and proliferation of MG 63 cells Fig shows no significant difference in the cell viability between all scaffold groups and the control group after d of culture Statistical analysis also showed that differences in the cell viability among all groups were not significant after d (P > 0.05) As shown in Fig 4, the fluorescence values for all scaffold groups (at higher levels than those of the control groups) increased as the culture time increased to and d, which revealed that the cell viability of the scaffolds significantly increased within this culture duration (*P < 0.001) At d, the cell viability on S2 HAp/PDLLA scaffold was significantly higher than that on pure PDLLA scaffold (**P < 0.01), but the cell viability for S3 and S4 HAp/PDLLA scaffolds showed no significant difference with that of pure PDLLA scaffold At d, the cell viability for all HAp/PDLLA scaffolds became significantly higher than that of pure PDLLA scaffold (*P < 0.001) The S2 HAp/PDLLA scaffolds (with 10 wt.% HAp) demonstrated better cell viability than the other scaffolds for different culture times (5 and d) (***P < 0.05), whereas the cell viability for S3 and S4 scaffolds was comparable for all the culture times (Fig 4) Previous studies proved that small pores might prevent cellular penetration and migration within scaffolds [31] Among the scaffolds studied, S1 PDLLA scaffold was characterized by the largest pore sizes, but the cell viability of S1 PDLLA scaffold was lower than that of the HAp/PDLLA scaffolds This result can be attributed to the absence of the HAp component in the composition of S1 PDLLA scaffold As a result, the surface of S1 PDLLA scaffold is lacking in functional groups as cell-recognition signals (e.g., OH groups) and does not support that attachment of many cells to the PDLLA scaffold, even if many MG 63 cells can migrate to the scaffold Among the three composite scaffolds, S2 HAp/PDLLA scaffolds present higher cell viability than the other two scaffolds, which may be attributed to the fact that they possess larger pore sizes (mean pore size of 183 m) than S3 and S4 scaffolds (mean pore sizes of 122 and 117 m, respectively) This result probably caused more beneficial cell migration and higher cell viability on the S2 scaffolds than on the S3 and S4 scaffolds The results of the alamarBlue assay indicate that all the scaffolds exhibit good cell viability, and the addition of HAp nanorods to the PDLLA scaffold significantly enhanced the viability of MG 63 cells However, the high content of HAp in the scaffolds may inhibit the increase in cell viability Confocal laser microscopy was used to examine cell attachment and distribution Fig I6 and Fig show the confocal laser scanning microscopy images of MG 63 cells cultured on the scaffolds after 3, 5, and d The staining methods for nuclei by DAPI and for actin filaments by Oregon Green 488 phalloidin indicate the cell attachment and cytoskeleton distribution on the surfaces of scaffolds In fact, few cells could be observed on the S1 PDLLA and S2 HAp/PDLLA scaffolds at d (Fig I6A and B) Meanwhile, qualitatively higher MG 63 cells were observed on the S3 HAp/PDLLA and S4 HAp/PDLLA scaffolds (Fig I6C and D) Additionally, the anchorage of the cellular cytoskeleton to the scaffolds was observed at d, revealing that cell attachment occurred on these scaffolds early in the culture time A further increase in culture time to and d resulted in a significant increase in the number of cells for all scaffolds (Fig 5) At d, all scaffolds showed a cytoskeleton arrangement of MG 63 cells, which was attached to the scaffold surfaces (Fig 5a, c, e and g) Among them, the S2 HAp/PDLLA scaffold demonstrated a uniform cytoskeleton arrangement with a high density of MG 63 cells interconnecting the entire scaffold at d, but the corresponding cytoskeleton distribution was gradually lost at d (Fig 5d) The S1 PDLLA scaffold exhibited a relatively uniform cytoskeleton distribution at d, however, such a cytoskeleton arrangement in the S1 scaffold was lost at d (Fig 5b) The results suggested that HAp nanorods may have an effect on the cell attachment in these scaffolds The S1 and S2 scaffolds with small HAp percentages (0–10 wt.% HAp) in their composition probably did not support the formation of high amount of stress fibers in these scaffolds, which maintain actin bundles and focal contacts between MG 63 cells and the scaffolds for a longer culture time This resulted in weaker cell attachment and gradual loss of cytoskeleton organization in S1 and S2 scaffolds at d By contrast, a more extended cytoskeleton network was observed in the S3 HAp/PDLLA (Fig 5f) and S4 HAp/PDLLA scaffolds (Fig 5h) at d than at d These observations confirm that high HAp percentages (20–30 wt.% HAp) enabled S3 and S4 scaffolds to produce higher amount of stress fibers, which led to stronger cell attachment and the extended cytoskeleton organization in these scaffolds with increasing culture time to d Moreover, a well-organized cytoskeleton network with a large number of adhered cells was exclusively produced in the S4 HAp/PDLLA scaffold at d, which was attributed to the highest HAp percentage in the S4 scaffold composition among the scaffolds studied The results revealed that all the scaffolds possess good cell adhesion, proliferation, and distribution A high percentage of HAp in the composite scaffolds should be helpful in supporting cell adhesion, proliferation, and distribution for longer culture periods The alamarBlue and confocal laser microscopy results demonstrated that all the scaffolds showed good MG 63 cell affinity in terms of cell viability, proliferation, and adhesion The S4 HAp/PDLLA scaffolds have the inorganic component consisting of HAp nanorods with a mean diameter of 28 nm and a mean length of 120 nm [33] and comprise 30 wt.% of the inorganic component to the polymer, which are the closest similarity to natural bone among the scaffolds studied As a result, the S4 scaffolds showed higher biocompatibility (better cell attachment and well-organized cytoskeleton arrangement), compared to the other scaffolds with lower HAp content with increasing number of culture days The S4 HAp/PDLLA scaffolds thus are promising candidates that will Please cite this article in press as: N.K Nga, et al., Biomimetic scaffolds based on hydroxyapatite nanorod/poly(d,l) lactic acid with their corresponding apatite-forming capability and biocompatibility for bone-tissue engineering, Colloids Surf B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.03.001 G Model COLSUB-6941; No of Pages ARTICLE IN PRESS N.K Nga et al / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx Fig Confocal laser scanning microscopy images of MG 63 cells stained nuclei with DAPI (blue) and actin filaments of cytoskeleton with Oregon Green 488 phalloidin (green) after and d of seeding on (a and b) S1 PDLLA scaffold, (c and d) S2 HAp/PDLLA scaffold, (e and f) S3 HAp/PDLLA scaffold, and (g and h) S4 HAp/PDLLA scaffold Scale bars represent 100 m (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.) promote production and organization of ECM with mineralization and expression of osteopontin, collagen type I, bone sialoprotein, which are typically expressed in native bone and will be further investigated in the next phase of our work Conclusion This study demonstrated that biomimetic 3D hydroxyapatite nanorod/poly(d,l) lactic acid scaffolds were successfully Please cite this article in press as: N.K Nga, et al., Biomimetic scaffolds based on hydroxyapatite nanorod/poly(d,l) lactic acid with their corresponding apatite-forming capability and biocompatibility for bone-tissue engineering, Colloids Surf B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.03.001 G Model COLSUB-6941; No of Pages ARTICLE IN PRESS N.K Nga et al / Colloids and Surfaces B: Biointerfaces xxx (2015) xxx–xxx synthesized using solvent casting combined with salt-leaching technique The synthesized scaffolds are porous, have thicknesses of mm, macropore networks with mean pore sizes of 117–183 m, and high porosity Both in vitro mineralization and in vitro cell culture tests showed that the composite scaffolds with high HAp contents, which have a structure that is the closest to that of natural bone, induced the rapid formation of bone-like apatite after a quick soaking time in SBF and demonstrated better cell adhesion, proliferation, and distribution with increasing culture days Within this study, it is concluded that HAp/PDLLA scaffolds with high HAp percentages are potential biomaterials to be used as BTE scaffolds for further studies that will be aimed at performing long-term in vitro cell culture experiments, degradation studies, and deeper characterizations (mechanical strength, surface roughness) Acknowledgments This study was funded by the Vietnam National Foundation for Science and Technology Development (NAFOSTED) under Grant number 104.02-2012.42 The authors would like to thank Prof C Migliaresi and A Motta, Department of Industrial Engineering and BIOtech Research Centre, University of Trento, Italy, for supporting cell culture experiments Appendix A Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.colsurfb 2015.03.001 References [1] D Puppi, F Chiellini, A.M Piras, E Chiellini, Prog Polym Sci 35 (2010) 403 [2] S Zadegan, M Hosainalipour, H.R Rezaie, H Ghassai, M.A Shokrgozar, Mater Sci Eng C 31 (2011) 954 [3] C.R Perry, Clin Orthop Relat Res 360 (1999) 71 [4] S.N Khan, E Tomin, J.M Lane, Orthop Clin North Am 31 (2000) 389 [5] I.O Smith, X.H Liu, L.A Smith, P.X Ma, WIREs Nanomed Nanobiotechnol (2009) 226 [6] D.W Hutmacher, Biomaterials 21 (2000) 2529 [7] R.Y Zhang, P.X Ma, J Biomed Mater Res 52 (2000) 430 [8] S.J Kim, D.H Jang, W.H Park, B.M Min, Polymer 51 (2010) 1320 [9] J.M Williams, A Adewunmi, R.M Schek, C.L Flanagan, P.H Krebsbach, S.E Feinberg, S.J Hollister, S Das, Biomaterials 26 (2005) 4817 [10] K.H Tan, C.K Chua, K.F Leong, C.M Cheah, W.S Gui, W.S Tan, F.E Wiria, Biomed Mater Eng 15 (2005) 113–124 [11] F Intranuovo, D Howard, L.J White, R.K Johal, A.M Ghaemmaghami, P Favia, S.M Howdle, K.M Shakesheff, M.R Alexander, Acta Biomater (2011) 3336 [12] H.W Kim, J.C Knowles, H.E Kim, J Mater Sci Mater Med 16 (2005) 189 [13] C.F.L Santos, A.P Silva, L Lopes, I Pires, I.J Correia, Mater Sci Eng C 32 (2012) 1293 [14] N Mehrban, J Bowen, E Vorndran, U Gbureck, L.M Grover, Colloids Surf B: Biointerfaces 111 (2013) 469 [15] Q.Z Chen, I.D Thompso, A.R Boccaccini, Biomaterials 27 (2006) 2414 [16] S Yang, J Wang, L Tang, H Ao, H Tan, T Tang, C Liu, Colloids Surf B: Biointerfaces 116 (2014) 72 [17] G Heimke, Adv Mater (1989) [18] F.E Wiria, K.F Leong, C.K Chua, Y Liu, Acta Biomater (2007) [19] R.L Simpson, F.E Wiria, A.A Amis, C.K Chua, K.F Leong, U.N Hansen, M Chandrasekaran, M.W Lee, J Biomed Mater Res B: Appl Biomater 84 (2008) 17 [20] R Murugan, S Ramakrishna, Compos Sci Technol 65 (2005) 2385 [21] A Abdal-hay, F.A Sheikh, J.K Lim, Colloids Surf B: Biointerfaces 102 (2013) 635 [22] R.Z LeGeros, Chem Rev 108 (2008) 4742 [23] M Sadat-Shojai, M.T Khorasani, E Dinpanah-Khoshdargi, A Jamshidi, Acta Biomater (2013) 7591 [24] H Zhou, J Lee, Acta Biomater (2011) 2769 [25] Z Lu, S.I Roohani-Esfahani, G Wang, H Zreiqat, Nanomedicine (2012) 507 [26] Y Cai, Y Liu, W Yan, Q Hu, J Tao, M Zhang, Z Shi, R Tang, J Mater Chem 17 (2007) 3780 [27] S.V Dorozhkin, Acta Biomater (2010) 715 [28] T.J Webster, C Ergun, R.H Doremus, R.W Siegel, R Bizios, Biomaterials 21 (2000) 1803 [29] B Zou, X Li, H Zhuang, W Cui, J Zou, J Chen, Polym Degrad Stabil 96 (2011) 114 [30] I Rajzer, E Menaszek, R Kwiatkowski, W Chrzanowski, J Mater Sci Mater Med 25 (2014) 1239 [31] V Karageorgiou, D Kaplan, Biomaterials 25 (2005) 5474 [32] L Chen, C.Y Tang, D.Z Chen, C.T Wong, C.P Tsui, Compos Sci Technol 71 (2011) 1842 [33] N.K Nga, L.T Giang, P.H Viet, C Migliaresi, Colloids Surf B: Biointerfaces 116 (2014) 666 [34] N.K Nguyen, M Leoni, D Maniglio, C Migliaresi, J Biomater Appl 28 (2013) 49 [35] C.R Kothapalli, M.T Shaw, M Wei, Acta Biomater (2005) 653 [36] T Kokubo, H Takadama, Biomaterials 27 (2006) 2907 [37] S.F Hulbert, F.A Young, R.S Mathews, J.J Klawitter, C.D Talbert, F.H Stelling, J Biomed Mater Res (1970) 433 [38] J.J Klawitter, J.G Bagwell, A.M Weinstein, B.W Sauer, J Biomed Mater Res 10 (1976) 311 [39] H Hu, Y Qiao, F Meng, X Liu, C Ding, Colloids Surf B: Biointerfaces 101 (2013) 83 [40] C Deng, J Weng, X Lu, S.B Zhou, J.X Wan, S.X Qu, B Feng, X.H Li, Mater Sci Eng C 28 (2008) 1304 [41] L Chen, D Zhai, C Wu, J Chang, Ceram Int 40 (2014) 12765 Please cite this article in press as: N.K Nga, et al., Biomimetic scaffolds based on hydroxyapatite nanorod/poly(d,l) lactic acid with their corresponding apatite-forming capability and biocompatibility for bone-tissue engineering, Colloids Surf B: Biointerfaces (2015), http://dx.doi.org/10.1016/j.colsurfb.2015.03.001 ... Nga, et al., Biomimetic scaffolds based on hydroxyapatite nanorod/ poly(d,l) lactic acid with their corresponding apatite-forming capability and biocompatibility for bone-tissue engineering, Colloids... Nga, et al., Biomimetic scaffolds based on hydroxyapatite nanorod/ poly(d,l) lactic acid with their corresponding apatite-forming capability and biocompatibility for bone-tissue engineering, Colloids... et al., Biomimetic scaffolds based on hydroxyapatite nanorod/ poly(d,l) lactic acid with their corresponding apatite-forming capability and biocompatibility for bone-tissue engineering, Colloids