Highly aligned porous Ti scaffold coated with bone morphogenetic protein-loaded silica/chitosan hybrid for enhanced bone regeneration Hyun-Do Jung,1* Se-Won Yook,1* Cheol-Min Han,1 Tae-Sik Jang,1 Hyoun-Ee Kim,1 Young-Hag Koh,2,4 Yuri Estrin3 1
WCU Hybrid Materials Program, Department of Materials Science and Engineering, Seoul National University, Seoul 151-742, Korea 2 Department of Dental Laboratory Science and Engineering, Korea University, Seoul 136-703, Korea 3 Centre for Advanced Hybrid Materials, Department of Materials Engineering, Monash University, Clayton, VIC 3800, Australia 4 Department of of Orthopaedics, Korea University Medical Center, Guro Hospital, Seoul, 152-703, Korea Received 7 July 2013; revised 4 October 2013; accepted 20 October 2013 Published online 21 November 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.b.33072 Abstract: Porous Ti has been widely investigated for orthopedic and dental applications on account of their ability to promote implant fixation via bone ingrowth into pores. In this study, highly aligned porous Ti scaffolds coated with a bone morphogenetic protein (BMP)-loaded silica/chitosan hybrid were produced, and their bone regeneration ability was evaluated by in vivo animal experiments. Reverse freeze casting allowed for the creation of highly aligned pores, resulting in a high compressive strength of 254 6 21 MPa of the scaffolds at a porosity level of 51 vol %. In addition, a BMP-loaded silica/ chitosan hybrid coating layer with a thickness of 1 lm was uniformly deposited on the porous Ti scaffold, which enabled the sustained release of the BMP over a prolonged period of time up to 26 days. The cumulative amount of the BMP released was 4 lg, which was much higher than that released
from the specimen without a hybrid coating layer. In addition, the bone regeneration ability of the porous Ti scaffold with a BMP-loaded silica/chitosan coating layer was examined by in vivo animal testing using a rabbit calvarial defect model and compared with those of the as-produced porous Ti scaffold and porous Ti scaffold with a silica/chitosan coating layer. After 4 weeks of healing, the specimen coated with a BMP-loaded silica/chitosan hybrid showed a much higher bone regeneration volume (36%) than the as-produced specimen (15%) (p < 0.005) and even the specimen coated with a silica/chitosan C 2013 Wiley Periodicals, Inc. J Biomed hybrid (25%) (p < 0.05). V Mater Res Part B: Appl Biomater, 102B: 913–921, 2014.
Key Words: titanium, aligned pores, BMP-2, bone regeneration, hybrid coating
How to cite this article: Jung HD, SW Yook, Han CM, Jang TS, Kim HE, Koh YH, Estrin Y. 2014. Highly aligned porous Ti scaffold coated with bone morphogenetic protein-loaded silica/chitosan hybrid for enhanced bone regeneration. J Biomed Mater Res Part B 2014:102B:913–921.
INTRODUCTION
Autologous and allogenic bone grafts are widely used in clinical practice for bone replacement of fractured or damaged bones.1 However, bone graft treatments often suffer from disease transmission and donor site morbidity after surgery. Therefore, considerable effort has been made to use synthetic materials such as bone substitutes,2,3 which are generally composed of biocompatible ceramics or polymers scaffolds. Because of the brittle fracture behavior of ceramics under tension or bending their use in the medical field is limited to non-load-bearing applications.4 Conversely, ductile polymers generally lack the structural rigidity that is required for successful bone replacement.5,6 Titanium (Ti) is very attractive as a scaffold material owing to is excellent mechanical properties (i.e., high
strength and toughness), ductility and resilience as well as reasonably high biocompatibility.7,8 These unique properties enable the use of Ti metal for long-term load-bearing applications, for which ceramics and polymers are generally unsuitable. In addition, porous Ti can provide a threedimensional space and biocompatible surfaces for desirable bone ingrowth when used as a bone substitute.9–11 These porous scaffolds are preferred to have a pore size >100 lm with a higher porosity >45 vol %.9,12,13 Recently, considerable effort has been made to modify the surface characteristics of porous scaffolds for promoting immobilization of biomolecules, which can accordingly accelerate bone regeneration in vivo.14–16 Bone morphogenetic protein-2 (BMP-2), a well-known growth factor (GF), plays a key role in the osteogenic differentiation of
*Both authors contributed equally to this work. Correspondence to: Y. H. Koh (e-mail: [email protected]) Contract grant sponsor: Korea Healthcare technology R&D Project, Ministry for Health, Welfare & Family Affairs; contract grant number: A121035 Contract grant sponsor: Technology Innovation Program (WPM Biomedical Materials–Implant Materials), Ministry of Knowledge Economy (MKE, Korea); contract grant number: 0037915.
C 2013 WILEY PERIODICALS, INC. V
913
mesenchymal cells and new bone formation.17–20 However, when GFs are simply immobilized on scaffolds, they will be quickly released with negligible long-term efficacy. In addition, an excessively high dose required for the sustained release of GFs may cause undesirable growth of tumors or neovascularization of non-targeted tissues.17,21 Thus, considerable effort has been put in the development of new delivery systems for slow release of GFs over a prolonged period of time in a controlled manner.22,23 Recently, silica xerogel/ chitosan hybrids demonstrated great potential as an effective delivery vehicle with excellent osteoconductivity,24 which can be achieved with the combination of an ordered mesoporous structure of silica xerogel and the flexibility of chitosan. In this study, we produced porous Ti scaffolds with highly aligned pores and coated their surface with a BMPloaded silica xerogel/chitosan hybrid. A highly aligned porous structure was achieved by a reverse freeze casting method, newly developed by our group.25 In addition, BMP2 was directly incorporated into a silica/chitosan hybrid mixture in an in situ manner during the sol–gel process. Subsequently, the surfaces of the porous Ti scaffold were coated with the BMP-2 loaded silica/chitosan hybrid using the dip-coating technique, in which the specimens were immersed in the hybrid mixture and then kept at room temperature for 2 h to induce sufficient gelation of the hybrid coating layer. The porosity characteristics, microstructure, and mechanical properties of the porous Ti scaffold produced were studied along with the release behavior of the BMP-2. In addition, the bone regeneration ability was assessed by in vivo animal testing using a rabbit calvarial defect model. For comparison, as-produced porous Ti scaffolds and Ti scaffolds with a silica/chitosan hybrid coating layer were also examined. MATERIALS AND METHODS
Porous Ti scaffold preparation Porous Ti scaffolds with highly aligned pores were produced by the reverse freeze casting method developed by our group and described elsewhere.25 In brief, a Ti/camphene slurry was prepared by mixing commercial Ti powders (-325 mesh, Alfa Aesar, Ward Hill, MA) and camphene (C10H16, Sigma–Aldrich, St. Louis, MO) at 60 C using ballmilling. The slurry was then poured onto solid camphene that had been unidirectionally solidified at 3 C, followed by immediate quenching to 220 C. The green body obtained was then kept at 45.5 C for 30 h to induce migration of the Ti powders down the channels formed within the unidirectionally frozen camphene body. Subsequently, the green samples were freeze-dried to remove the camphene and sintered at 1300 C for 3 h in vacuum.
triethyl phosphate (Sigma–Aldrich Chem) as precursors, with the assistance of HCl as a catalyst. Separately, a chitosan solution was also prepared by dissolving chitosan powder (85%, deacetylated, Sigma–Aldrich Chem) in 2 wt % acetic acid. The prepared silica sol and the chitosan solution were then mixed together by magnetic stirring to obtain a homogenous silica/chitosan hybrid mixture with a silica content of 30 vol %. Subsequently, 200 lL of the BMP-2 solution with a BMP concentration of 1 mg/mL that had been produced using Escherichia coli (E. coli) was added to 1 mL of the silica/chitosan hybrid mixture and then vigorously stirred for 1 h to obtain a homogenous BMP-loaded silica/chitosan hybrid mixture.21 In addition, to visualize and monitor the loading and release of the BMP from the hybrid coating layer, a green fluorescent protein (GFP) solution, which had been produced using E. coli, was directly incorporated into the BMP-loaded silica/chitosan hybrid mixture during magnetic stirring.27,28 Coating of porous Ti with BMP-loaded silica/chitosan hybrid mixture The surfaces of the highly aligned porous Ti scaffolds produced with reverse freeze casting were coated with the BMP-loaded silica/chitosan hybrid mixture by the dipcoating technique.21 The specimens were immersed in the hybrid mixture under vacuum for 10 min to ensure the complete infiltration of the hybrid mixture into threedimensionally interconnected pores and then kept at room temperature for 2 h to induce sufficient gelation of the hybrid coating layer. Subsequently, the coated specimens were rinsed with 0.1 N NaOH and washed with distilled water to remove residual acetic acid, followed by drying and UV irradiation for sterilization. For comparison purposes, the porous Ti scaffolds were immersed in the BMP solution to directly immobile the BMP without the silica/ chitosan hybrid coating layer. Porous structure and microstructure observations The porous structure and microstructure of the coated scaffolds were observed by field emission scanning electron microscopy (FE-SEM, JSM-6330F, JEOL Techniques, Tokyo, Japan). The porous structure was also examined by microcomputer tomography (l-CT, Skyscan 1173 X-ray Microtomography System, Skyscan, Kontich, Belgium) a resolution of 6 lm with a 0.25 mm brass filter operated at a voltage of 130 kV and a current of 60 lA. Projections were reconstructed using a software (NRecon 1.6.4.8, Skyscan, Kontich, Belgium) and 3D images of the specimen were achieved using a volume rendering program (CTvol 2.2, Skyscan, Kontich, Belgium). The chemical composition of the hybrid coating layer was characterized by energy dispersive spectroscopy (EDS) attached to the FE-SEM.
BMP-loaded silica/chitosan hybrid mixture preparation A silica/chitosan hybrid mixture containing BMP-2 aimed for BMP-loaded coating on porous Ti scaffolds was prepared according to a method reported in the literature.21,26 A silica sol was synthesized by the sol–gel process using tetramethylorthosilicate (TMOS, Sigma–Aldrich Chem), CaCl2, and
Compressive strength test To evaluate the mechanical properties of the highly aligned porous Ti scaffolds coated with the BMP-loaded silica/chitosan hybrid, compression tests were conducted using a screw-driven testing machine (Instron 5565, Instron Corp.,
914
HIGHLY ALIGNED POROUS TI SCAFFOLD FOR ENHANCED BONE REGENERATION
JUNG ET AL.
ORIGINAL RESEARCH REPORT
Canton). As a reference, as-produced porous Ti scaffolds without a coating layer were also tested. Cylindrical specimens 6 mm in diameter and 9 mm in height were compressed at a constant cross-head speed of 2 mm/min. The stress and strain responses of the specimens were monitored throughout the tests. Five specimens were tested to obtain the mean value and the standard deviation of the data. In vitro BMP release evaluation Confocal laser scanning spectroscopy (CLSM, Zeiss-LSM510, Carl Zeiss, NY) was used to identify BMP-2 in the silica/chitosan hybrid coating layer deposited on the porous Ti scaffolds by visualizing the distribution of the green fluorescent protein (GFP).21 The green fluorescence observed can be considered as an indication of the effective incorporation of the BMP into the hybrid coating layer. The release of the BMP-2 from the silica/chitosan hybrid coating layer was evaluated by monitoring changes in its concentration after immersion in a phosphate buffered saline (PBS) solution. The specimens, 8 mm in diameter and 2 mm in height, were placed in a PBS solution with a pH value of 7.4 at 37 C for up to 26 days. The PBS solution was exchanged at specific intervals. For comparison purpose, the BMP-2 immobilized specimen directly immobilized with the BMP-2 without a silica/chitosan hybrid coating layer was also tested. The amounts of the BMP released from the silica/chitosan hybrid coating layer were calculated by measuring the optical absorbance values of the PBS solutions after immersion tests using UV-spectroscopy (ICPAES, Optima-4300 DV PerkinElmer, Wellesley, MA) operated at 215 nm. A calibration curve was obtained by measuring the optical absorbance of BMP solutions with various concentrations in the range of 0.004–2.048 lg/mL. The curve has a linear relationship of BMP concentration (y) 5 1.2249 3 optical absorbance (x) 2 0.1346 (lg/mL). The relative amounts of the BMP released from the specimens were normalized to those in the initially loaded coating layer. Five samples were tested to obtain the mean and the standard deviation of the data. In vivo animal experiment The bone regeneration ability of the highly aligned porous Ti scaffolds coated with the BMP-loaded silica/chitosan hybrid was evaluated by in vivo animal experiments using a rabbit calvarial defect model. Both as-produced porous Ti scaffolds used as a reference and those with a silica/chitosan hybrid coating layer were tested to the efficacy of BMP loading. Nine healthy New Zealand white rabbits (13 weeks old, weight range of 2.7–3.0 kg) were used for the animal experiments. All rabbits were anesthetized through intramuscular injection of a combination of 0.1 mL of 2% Xylazine HCl (Rompun, Bayer Korea, Korea), 0.2 mL of Tiletamine HCl (Zoletil, Virbac Laboratories, France) and Lidocaine (Yuhan Corporation, Korea). A total of four circular defects (each 8 mm in diameter) were created symmetrically on bilateral sides in the calvarial of each rabbit using
trephine drills. Subsequently, the defects were filled with the three different specimens (i.e., as-produced porous Ti scaffold, Ti scaffold with a silica/chitosan hybrid coating layer, and Ti scaffold with a BMP-loaded silica/chitosan hybrid coating layer). For comparison purposes, a porous Ti scaffold with a random porous structure produced by dynamic freeze casting was also tested.10 Each animal was administered an intramuscular injection of gentamycin after the surgical procedure. After 4 weeks of healing, the rabbits were sacrificed to extract their bone defect regions. The harvested bone tissues were scanned by l-CT with a resolution of 6 lm operated at a voltage of 130 kV and a current of 60 lA, to evaluate new bone formation and bone ingrowth into the pores. The 3D images of the porous specimens were obtained using the volume rendering program (CTvol 2.2). Regions in the pores, which was filled with newly formed bone, could be reasonably identified by considering a predetermined threshold with negligible artifacts associated with Ti metal.29,30 The bone regeneration fraction was calculated by considering the areas of the regions filled with the newly formed bone and the original pores before implantation. Seven specimens were examined to obtain the mean value and standard deviation. For histological evaluation, the extraction regions were fixed in a neutral 10% formaldehyde solution and embedded in resin (Technovit 7200 VLC, Kulzer, Germany). Nondecalcified thin ground sections were prepared and reduced to a thickness of
WCU Hybrid Materials Program, Department of Materials Science and Engineering, Seoul National University, Seoul 151-742, Korea 2 Department of Dental Laboratory Science and Engineering, Korea University, Seoul 136-703, Korea 3 Centre for Advanced Hybrid Materials, Department of Materials Engineering, Monash University, Clayton, VIC 3800, Australia 4 Department of of Orthopaedics, Korea University Medical Center, Guro Hospital, Seoul, 152-703, Korea Received 7 July 2013; revised 4 October 2013; accepted 20 October 2013 Published online 21 November 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.b.33072 Abstract: Porous Ti has been widely investigated for orthopedic and dental applications on account of their ability to promote implant fixation via bone ingrowth into pores. In this study, highly aligned porous Ti scaffolds coated with a bone morphogenetic protein (BMP)-loaded silica/chitosan hybrid were produced, and their bone regeneration ability was evaluated by in vivo animal experiments. Reverse freeze casting allowed for the creation of highly aligned pores, resulting in a high compressive strength of 254 6 21 MPa of the scaffolds at a porosity level of 51 vol %. In addition, a BMP-loaded silica/ chitosan hybrid coating layer with a thickness of 1 lm was uniformly deposited on the porous Ti scaffold, which enabled the sustained release of the BMP over a prolonged period of time up to 26 days. The cumulative amount of the BMP released was 4 lg, which was much higher than that released
from the specimen without a hybrid coating layer. In addition, the bone regeneration ability of the porous Ti scaffold with a BMP-loaded silica/chitosan coating layer was examined by in vivo animal testing using a rabbit calvarial defect model and compared with those of the as-produced porous Ti scaffold and porous Ti scaffold with a silica/chitosan coating layer. After 4 weeks of healing, the specimen coated with a BMP-loaded silica/chitosan hybrid showed a much higher bone regeneration volume (36%) than the as-produced specimen (15%) (p < 0.005) and even the specimen coated with a silica/chitosan C 2013 Wiley Periodicals, Inc. J Biomed hybrid (25%) (p < 0.05). V Mater Res Part B: Appl Biomater, 102B: 913–921, 2014.
Key Words: titanium, aligned pores, BMP-2, bone regeneration, hybrid coating
How to cite this article: Jung HD, SW Yook, Han CM, Jang TS, Kim HE, Koh YH, Estrin Y. 2014. Highly aligned porous Ti scaffold coated with bone morphogenetic protein-loaded silica/chitosan hybrid for enhanced bone regeneration. J Biomed Mater Res Part B 2014:102B:913–921.
INTRODUCTION
Autologous and allogenic bone grafts are widely used in clinical practice for bone replacement of fractured or damaged bones.1 However, bone graft treatments often suffer from disease transmission and donor site morbidity after surgery. Therefore, considerable effort has been made to use synthetic materials such as bone substitutes,2,3 which are generally composed of biocompatible ceramics or polymers scaffolds. Because of the brittle fracture behavior of ceramics under tension or bending their use in the medical field is limited to non-load-bearing applications.4 Conversely, ductile polymers generally lack the structural rigidity that is required for successful bone replacement.5,6 Titanium (Ti) is very attractive as a scaffold material owing to is excellent mechanical properties (i.e., high
strength and toughness), ductility and resilience as well as reasonably high biocompatibility.7,8 These unique properties enable the use of Ti metal for long-term load-bearing applications, for which ceramics and polymers are generally unsuitable. In addition, porous Ti can provide a threedimensional space and biocompatible surfaces for desirable bone ingrowth when used as a bone substitute.9–11 These porous scaffolds are preferred to have a pore size >100 lm with a higher porosity >45 vol %.9,12,13 Recently, considerable effort has been made to modify the surface characteristics of porous scaffolds for promoting immobilization of biomolecules, which can accordingly accelerate bone regeneration in vivo.14–16 Bone morphogenetic protein-2 (BMP-2), a well-known growth factor (GF), plays a key role in the osteogenic differentiation of
*Both authors contributed equally to this work. Correspondence to: Y. H. Koh (e-mail: [email protected]) Contract grant sponsor: Korea Healthcare technology R&D Project, Ministry for Health, Welfare & Family Affairs; contract grant number: A121035 Contract grant sponsor: Technology Innovation Program (WPM Biomedical Materials–Implant Materials), Ministry of Knowledge Economy (MKE, Korea); contract grant number: 0037915.
C 2013 WILEY PERIODICALS, INC. V
913
mesenchymal cells and new bone formation.17–20 However, when GFs are simply immobilized on scaffolds, they will be quickly released with negligible long-term efficacy. In addition, an excessively high dose required for the sustained release of GFs may cause undesirable growth of tumors or neovascularization of non-targeted tissues.17,21 Thus, considerable effort has been put in the development of new delivery systems for slow release of GFs over a prolonged period of time in a controlled manner.22,23 Recently, silica xerogel/ chitosan hybrids demonstrated great potential as an effective delivery vehicle with excellent osteoconductivity,24 which can be achieved with the combination of an ordered mesoporous structure of silica xerogel and the flexibility of chitosan. In this study, we produced porous Ti scaffolds with highly aligned pores and coated their surface with a BMPloaded silica xerogel/chitosan hybrid. A highly aligned porous structure was achieved by a reverse freeze casting method, newly developed by our group.25 In addition, BMP2 was directly incorporated into a silica/chitosan hybrid mixture in an in situ manner during the sol–gel process. Subsequently, the surfaces of the porous Ti scaffold were coated with the BMP-2 loaded silica/chitosan hybrid using the dip-coating technique, in which the specimens were immersed in the hybrid mixture and then kept at room temperature for 2 h to induce sufficient gelation of the hybrid coating layer. The porosity characteristics, microstructure, and mechanical properties of the porous Ti scaffold produced were studied along with the release behavior of the BMP-2. In addition, the bone regeneration ability was assessed by in vivo animal testing using a rabbit calvarial defect model. For comparison, as-produced porous Ti scaffolds and Ti scaffolds with a silica/chitosan hybrid coating layer were also examined. MATERIALS AND METHODS
Porous Ti scaffold preparation Porous Ti scaffolds with highly aligned pores were produced by the reverse freeze casting method developed by our group and described elsewhere.25 In brief, a Ti/camphene slurry was prepared by mixing commercial Ti powders (-325 mesh, Alfa Aesar, Ward Hill, MA) and camphene (C10H16, Sigma–Aldrich, St. Louis, MO) at 60 C using ballmilling. The slurry was then poured onto solid camphene that had been unidirectionally solidified at 3 C, followed by immediate quenching to 220 C. The green body obtained was then kept at 45.5 C for 30 h to induce migration of the Ti powders down the channels formed within the unidirectionally frozen camphene body. Subsequently, the green samples were freeze-dried to remove the camphene and sintered at 1300 C for 3 h in vacuum.
triethyl phosphate (Sigma–Aldrich Chem) as precursors, with the assistance of HCl as a catalyst. Separately, a chitosan solution was also prepared by dissolving chitosan powder (85%, deacetylated, Sigma–Aldrich Chem) in 2 wt % acetic acid. The prepared silica sol and the chitosan solution were then mixed together by magnetic stirring to obtain a homogenous silica/chitosan hybrid mixture with a silica content of 30 vol %. Subsequently, 200 lL of the BMP-2 solution with a BMP concentration of 1 mg/mL that had been produced using Escherichia coli (E. coli) was added to 1 mL of the silica/chitosan hybrid mixture and then vigorously stirred for 1 h to obtain a homogenous BMP-loaded silica/chitosan hybrid mixture.21 In addition, to visualize and monitor the loading and release of the BMP from the hybrid coating layer, a green fluorescent protein (GFP) solution, which had been produced using E. coli, was directly incorporated into the BMP-loaded silica/chitosan hybrid mixture during magnetic stirring.27,28 Coating of porous Ti with BMP-loaded silica/chitosan hybrid mixture The surfaces of the highly aligned porous Ti scaffolds produced with reverse freeze casting were coated with the BMP-loaded silica/chitosan hybrid mixture by the dipcoating technique.21 The specimens were immersed in the hybrid mixture under vacuum for 10 min to ensure the complete infiltration of the hybrid mixture into threedimensionally interconnected pores and then kept at room temperature for 2 h to induce sufficient gelation of the hybrid coating layer. Subsequently, the coated specimens were rinsed with 0.1 N NaOH and washed with distilled water to remove residual acetic acid, followed by drying and UV irradiation for sterilization. For comparison purposes, the porous Ti scaffolds were immersed in the BMP solution to directly immobile the BMP without the silica/ chitosan hybrid coating layer. Porous structure and microstructure observations The porous structure and microstructure of the coated scaffolds were observed by field emission scanning electron microscopy (FE-SEM, JSM-6330F, JEOL Techniques, Tokyo, Japan). The porous structure was also examined by microcomputer tomography (l-CT, Skyscan 1173 X-ray Microtomography System, Skyscan, Kontich, Belgium) a resolution of 6 lm with a 0.25 mm brass filter operated at a voltage of 130 kV and a current of 60 lA. Projections were reconstructed using a software (NRecon 1.6.4.8, Skyscan, Kontich, Belgium) and 3D images of the specimen were achieved using a volume rendering program (CTvol 2.2, Skyscan, Kontich, Belgium). The chemical composition of the hybrid coating layer was characterized by energy dispersive spectroscopy (EDS) attached to the FE-SEM.
BMP-loaded silica/chitosan hybrid mixture preparation A silica/chitosan hybrid mixture containing BMP-2 aimed for BMP-loaded coating on porous Ti scaffolds was prepared according to a method reported in the literature.21,26 A silica sol was synthesized by the sol–gel process using tetramethylorthosilicate (TMOS, Sigma–Aldrich Chem), CaCl2, and
Compressive strength test To evaluate the mechanical properties of the highly aligned porous Ti scaffolds coated with the BMP-loaded silica/chitosan hybrid, compression tests were conducted using a screw-driven testing machine (Instron 5565, Instron Corp.,
914
HIGHLY ALIGNED POROUS TI SCAFFOLD FOR ENHANCED BONE REGENERATION
JUNG ET AL.
ORIGINAL RESEARCH REPORT
Canton). As a reference, as-produced porous Ti scaffolds without a coating layer were also tested. Cylindrical specimens 6 mm in diameter and 9 mm in height were compressed at a constant cross-head speed of 2 mm/min. The stress and strain responses of the specimens were monitored throughout the tests. Five specimens were tested to obtain the mean value and the standard deviation of the data. In vitro BMP release evaluation Confocal laser scanning spectroscopy (CLSM, Zeiss-LSM510, Carl Zeiss, NY) was used to identify BMP-2 in the silica/chitosan hybrid coating layer deposited on the porous Ti scaffolds by visualizing the distribution of the green fluorescent protein (GFP).21 The green fluorescence observed can be considered as an indication of the effective incorporation of the BMP into the hybrid coating layer. The release of the BMP-2 from the silica/chitosan hybrid coating layer was evaluated by monitoring changes in its concentration after immersion in a phosphate buffered saline (PBS) solution. The specimens, 8 mm in diameter and 2 mm in height, were placed in a PBS solution with a pH value of 7.4 at 37 C for up to 26 days. The PBS solution was exchanged at specific intervals. For comparison purpose, the BMP-2 immobilized specimen directly immobilized with the BMP-2 without a silica/chitosan hybrid coating layer was also tested. The amounts of the BMP released from the silica/chitosan hybrid coating layer were calculated by measuring the optical absorbance values of the PBS solutions after immersion tests using UV-spectroscopy (ICPAES, Optima-4300 DV PerkinElmer, Wellesley, MA) operated at 215 nm. A calibration curve was obtained by measuring the optical absorbance of BMP solutions with various concentrations in the range of 0.004–2.048 lg/mL. The curve has a linear relationship of BMP concentration (y) 5 1.2249 3 optical absorbance (x) 2 0.1346 (lg/mL). The relative amounts of the BMP released from the specimens were normalized to those in the initially loaded coating layer. Five samples were tested to obtain the mean and the standard deviation of the data. In vivo animal experiment The bone regeneration ability of the highly aligned porous Ti scaffolds coated with the BMP-loaded silica/chitosan hybrid was evaluated by in vivo animal experiments using a rabbit calvarial defect model. Both as-produced porous Ti scaffolds used as a reference and those with a silica/chitosan hybrid coating layer were tested to the efficacy of BMP loading. Nine healthy New Zealand white rabbits (13 weeks old, weight range of 2.7–3.0 kg) were used for the animal experiments. All rabbits were anesthetized through intramuscular injection of a combination of 0.1 mL of 2% Xylazine HCl (Rompun, Bayer Korea, Korea), 0.2 mL of Tiletamine HCl (Zoletil, Virbac Laboratories, France) and Lidocaine (Yuhan Corporation, Korea). A total of four circular defects (each 8 mm in diameter) were created symmetrically on bilateral sides in the calvarial of each rabbit using
trephine drills. Subsequently, the defects were filled with the three different specimens (i.e., as-produced porous Ti scaffold, Ti scaffold with a silica/chitosan hybrid coating layer, and Ti scaffold with a BMP-loaded silica/chitosan hybrid coating layer). For comparison purposes, a porous Ti scaffold with a random porous structure produced by dynamic freeze casting was also tested.10 Each animal was administered an intramuscular injection of gentamycin after the surgical procedure. After 4 weeks of healing, the rabbits were sacrificed to extract their bone defect regions. The harvested bone tissues were scanned by l-CT with a resolution of 6 lm operated at a voltage of 130 kV and a current of 60 lA, to evaluate new bone formation and bone ingrowth into the pores. The 3D images of the porous specimens were obtained using the volume rendering program (CTvol 2.2). Regions in the pores, which was filled with newly formed bone, could be reasonably identified by considering a predetermined threshold with negligible artifacts associated with Ti metal.29,30 The bone regeneration fraction was calculated by considering the areas of the regions filled with the newly formed bone and the original pores before implantation. Seven specimens were examined to obtain the mean value and standard deviation. For histological evaluation, the extraction regions were fixed in a neutral 10% formaldehyde solution and embedded in resin (Technovit 7200 VLC, Kulzer, Germany). Nondecalcified thin ground sections were prepared and reduced to a thickness of
