RGD-Conjugated Copolymer Incorporated into Composite of Poly

ARTICLE pubs.acs.org/Biomac

RGD-Conjugated Copolymer Incorporated into Composite of Poly(lactide-co-glycotide) and Poly(L-lactide)-Grafted Nanohydroxyapatite for Bone Tissue Engineering Peibiao Zhang,†,|| Haitao Wu,‡,|| Han Wu,‡ Zhongwen Lu,‡ Chao Deng,†,§ Zhongkui Hong,† Xiabin Jing,† and Xuesi Chen*,† †

Key Laboratory of Polymer Ecobiomaterials, State Key Laboratory of Polymer Physics and Chemistry, Changchun Institute of Applied Chemistry, Graduate School of Chinese Academy of Sciences, Chinese Academy of Sciences, Changchun 130022, People's Republic of China ‡ China-Japan Union Hospital, Jilin University, Changchun 130033, People's Republic of China § Biomedical Polymers Laboratory, and Jiangsu Key Laboratory of Advanced Functional Polymer Design and Application, College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123, People's Republic of China ABSTRACT: Various surface modification methods of RGD (Arg-Gly-Asp) peptides on biomaterials have been developed to improve cell adhesion. This study aimed to examine a RGD-conjugated copolymer RGD/MPEG-PLA-PBLG (RGDcopolymer) for its ability to promote bone regeneration by mixing it with the composite of poly(lactide-co-glycotide) (PLGA) and hydroxyapatite nanoparticles surface-grafted with poly(L-lactide) (g-HAP). The porous scaffolds were prepared using solvent casting/particulate leaching method and grafted to repair the rabbit radius defects after seeding with autologous bone marrow mesenchymal cells (MSCs) of rabbits. After incorporation of RGD-copolymer, there were no significant influences on scaffold’s porosity and pore size. Nitrogen of RGD peptide, and calcium and phosphor of g-HAP could be exposed on the surface of the scaffold simultaneously. Although the cell viability of its leaching liquid was 92% that was lower than g-HAP/PLGA, its cell adhesion and growth of 3T3 and osteoblasts were promoted significantly. The greatest increment in cell adhesion ratios (131.2
157.1% higher than g-HAP/PLGA) was observed when its contents were 0.1
1 wt % but only at 0.5 h after cell seeding. All the defects repaired with the implants were bridged after 24 weeks postsurgery, but the RGD-copolymer contained composite had larger new bone formation and better fusion interface. The composites containing RGD-copolymer enhanced bone ingrowth but presented more woven bones than others. The combined application of RGD-copolymer and bone morphological protein 2 (BMP-2) exhibited the best bone healing quality and was recommended as an optimal strategy for the use of RGD peptides.

1. INTRODUCTION Recently, more and more researchers have focused interests on bioactive hydroxyapatite (HA) ceramics and their composites for bone repair materials because of their good bone-bonding properties and osteoconductivity.1
3 HA are chemically and structurally similar to the mineral phase of native bone and have been used widely in dental and orthopedic surgery to fill bone defects and to coat metallic implant surfaces. Compared with HA ceramics, the composites of nanosized HA and polyesters, such as poly(lactide) (PLA),4,5 poly-ε-caprolactone (PCL),6 and the copolymer poly(lactide-co-glycotide) (PLGA),1,7 would be promising materials for bone grafts because of their improved mechanical properties, biodegradability, and processability.8 Among them, HA/PLGA nanocomposite is more attractive for bone repair applications because the degradation rate of PLGA can be adjusted by altering the ratio of lactic to glycolic acids to match the rate of bone formation.7,9,10 However, the limited r 2011 American Chemical Society

ability of cell adhesion of these composites would be an obstacle for their application in tissue engineering and orthopedics. Cell adhesion is an initial and important cellular procedure for polymeric scaffolds because it directly influences the following cell growth, differentiation, and migration of osteogenic cells that lead to forming bone tissue.11 In general, cell behavior and its interaction with bioactive material surface depend on physical and chemical properties, such as topography, surface charge, and chemistry.12
14 The tripeptide arginine-glycine-aspatic acid (Arg-Gly-Asp, or RGD) is an amino acid sequence that actively promotes cellular adhesion through binding to integrin receptors.15,16 It is the minimal cell-recognizable sequence found in many extracellular matrix proteins and blood proteins, which is first reported by Pierschbacher and Ruoslahti in the 1980s.17 Received: April 6, 2011 Revised: May 21, 2011 Published: May 23, 2011 2667

dx.doi.org/10.1021/bm2004725 | Biomacromolecules 2011, 12, 2667–2680

Biomacromolecules

ARTICLE

Scheme 1. Schematic Diagram Illustrating the Synthesis of Copolymer PEG-PLA-PGL/RGD (RGD-Copolymer)

Since then, various polymers immobilized with RGD or RGD containing peptides have been studied.18,19 Most of them are undertaken by surface modification chemically or physically and show improved cell adhesion. A number of studies have shown that the immobilization of adhesive RGD peptides on the surface of materials enhances osteoblast adhesion.20
22 This attachment property is due to the presence of integrin receptors implied in all the cellular adhesion phenomena. Kantlehner et al.22 have examined the porous poly(methyl methacrylate) (PMMA) implants coated with cyclic RGD peptide, Cyclo(RGDfK) in a rabbit model for their ability to promote bone regeneration, showing that cancellous bone ingrowth is enhanced and accelerated, and newly formed bone directly contacts the implant surface. However, Hennessy et al.23 have reported that RGD-coated HA disks significantly inhibit total bone formation as well as the amount of new bone directly contacting the implant perimeter. The reason is deduced that HA ceramic is particularly efficient at adsorbing proadhesive proteins which may contribute to HA’s high degree of osteoconductivity.23,24 The direct surface modification of RGD peptide on HA is regarded to have a negative effect on implant integration by competing with adsorbed proteins for integrin receptors. On the other hand, RGD peptides immobilized on the surfaces of biodegradable polymers even if with chemical methods are easy to be absorbed and lost their function as soon as the outer layer materials biodegraded. To avoid this problem, a new strategy for the use of RGD peptide in HA biomaterials was investigated in the present study. A biodegradable RGD-conjugated triblock copolymer was synthesized25 and introduced into HA/PLGA nanocomposite to prepare 3-D porous scaffolds for bone tissue engineering. The copolymer and HA nanoparticles in the composite were expected to play their functions in osteoblasts adhesion and biomineralization, respectively. According to our previous work, the copolymer has

shown some advantages of tunable mechanical properties, tunable degradable properties, and controlled cell adhesion. The cell adhesion and viability of 3T3 cell lines and human chondrocytes were significantly increased on the blend of 5% copolymer in PLGA.25 Meanwhile, in this study, the copolymer was used together with bone morphological protein 2 (BMP-2) to examine their synergistic effects.

2. EXPERIMENTAL SECTION 2.1. Materials. L-lactide (LA) and glycolide (GA) were purchased from Purac, Holland. Stannous octoate (Sn(Oct)2) was obtained from Sigma. Monomethoxy-poly(ethylene glycol) with a molecular weight of 750 (PEG750) was obtained from Aldrich. Prior to use, it was dried by an azeotropic distillation in toluene. The N-carboxy anhydride (NCA) of γ-benzyl-L-glutamate (BGL; BGL-NCA) was prepared according to Daly’s method.26 N-tert-Butoxycarbonyl-L-phenylalanine (Phe-BOC), N-hydroxylsuccinimide (NHS), and dicyclohexylcarbodiimide (DCC) from GL Biochem Ltd. (Shanghai, China) were used as received. 4-(Dimethylamino)pyridine (DMAP, 99%) obtained from Acros and RGD (arginine-glycine-(aspartic amide)) purchased from CL Bioscientific Inc. (Xi’an, China) were used without further purification. Hexane, methylene chloride, and chloroform were refluxed over CaH2 and distilled under nitrogen. Tetrahydrofuran (THF) was dried and distilled in the presence of sodium immediately before use. N,N-Dimethylformamide (DMF) was dried over CaH2 and distilled before use. Other chemical agents were all of analytical grade and used as received without further purification. 2.2. Polymer Synthesis. PLGA (LA/GA = 80:20, viscosity average molecular weight Mv = 1.96  105) was synthesized in our lab by the ring-opening copolymerization (ROP) of LA and GA in the presence of Sn(Oct)2 as catalyst.7 Hydroxyapatite nanoparticles (HAP) with the atomic ratio calcium (Ca)/phosphor (P) ≈ 1.67 and their surface-grafting by poly(L-lactide) (PLLA) (g-HAP) were prepared according to our published methods.4,7 The amount of grafted polymer 2668

dx.doi.org/10.1021/bm2004725 |Biomacromolecules 2011, 12, 2667–2680

Biomacromolecules

ARTICLE

Table 1. Porosity and Average Pore Size of Porous Composite Scaffolds composite scaffolds

porosity (%) avg pore size (μm)

g-HA/PLGA (HP)

86.3 ( 1.1

140.3 ( 80.6

RGD-copolymer/g-HA/PLGA (RHP) BMP-2/RGD-copolymer/g-HA/

87.2 ( 1.9 87.7 ( 2.1

132.1 ( 87.3 129.5 ( 88.2

PLGA (BRHP)

on the surface of g-HAP was determined by thermal gravimetric analysis to be 5.0
5.5 wt %. Poly(ethylene glycol)-b-poly(L-lactide)-b-poly(L-glutamic acid) (PEGPLA-PGL/RGD) was synthesized according to our previous work,25 briefly in the following three steps (Scheme 1): (1) diblock copolymer PEG-PLA-OH was prepared by ROP of LA with PEG-OH as macroinitiator and Sn(Oct)2 as catalyst, and its
OH end-group was converted into
NH2 to obtain PEG-PLA-NH2; (2) triblock copolymer PEG-PLA-PBGL was prepared by ROP of the NCA derived from BGL with the PEG-PLA-NH2 as macroinitiator,27 and then the protective benzyl groups in PEG-PLA-PBGL was removed by catalytic hydrogenation to obtain PEG-PLA-PGL with free pendant carboxyl groups; (3) RGD was attached to the pendant COOH with the help of NHS, DCC, and DMAP. After dialysis, the coupling ratio of RGD in the obtained PEG-PLA-PGL/RGD (RGD-copolymer), determined by the contents of glycine and arginine via amino acid analysis, was 39
49 wt %. 2.3. Preparation of Nanocomposites. The composite g-HAP/ PLGA (HP) was prepared by dispersing g-HAP/chloroform suspension into 10% PLGA/chloroform solution uniformly under magnetic stirring overnight and ultrasonic treatment for 30 min. The g-HAP content was 10 wt % in the composite. The RGD-copolymer was dissolved in DMF and dispersed into a HP chloroform solution to prepare RGD-copolymer/g-HAP/PLGA (RHP) composite with magnetic stirring for 2 h and ultrasonic treatment for 30 min. Finally, human BMP-2 (Beijing Bailingke Bioscientific Inc., China) was dispersed into the solution of RHP to prepare BMP-2/RGD-copolymer/g-HAP/PLGA (BRHP) composite. The contents of RGD-copolymer and BMP-2 were 1.0 and 0.4 wt % in the composites, respectively. The experimental groups of composite scaffolds in this study are shown in Table 1. 2.4. Porous Scaffolds Fabrication. The porous scaffolds were fabricated using the solvent casting/particulate leaching method according to our previous work.8 The sucrose particles, 100
450 μm in diameter, obtained with sieves, were mixed into the 10% chloroform solutions of HP, RHP, and BRHP, and the mass ratio of sucrose/ composite was 6:1 (w/w). The mixtures were cast in glass disks and dried in the air for 3 d. The sucrose particles were subsequently removed from the composites by leaching the composites in distilled water for 5 d, with the water changed every 12 h, and then the composites were dried under air and vacuum. The above-mentioned process was undertaken at 4 °C. The porous composite scaffolds obtained were then cut into small bars (3  5  20 mm), sterilized with UV irradiation for 30 min. 2.5. Characterization of Scaffolds. The microstructure of the scaffolds was observed with an emission scanning electron microscope (ESEM; Philips XL30 ESEM FEG, Japan). The diameter of the material of a total of 400 pores each was measured with NIH Image J software. The composition of surface elements was analyzed with energy dispersive X-ray spectrometry (EDX; Philips, XL-30W/TMP, Japan). The porosity of the scaffolds was measured by the method of modified liquid displacement with absolute ethanol according to a published method.28 The opening porosity of the scaffolds was obtained, and three measurements of each material were taken for an average value. 2.6. Cellular Assessment. Cell Culture. NIH 3T3 cells (Mouse embryonic fibroblast cell line) were purchased from Shanghai Institute of Cell Biology, Chinese Academy of Sciences (CAC), and cultured

in Dulbecco’s Modified Eagle Medium (DMEM, Gibco) supplement with 10% fetal bovine serum (FBS, Gibco), 1.0  105 U/L penicillin (Sigma) and 100 mg/L streptomycin (Sigma). Osteoblasts were isolated from the calvarial bone of a newborn rabbit using a modified method of culturing the bone pieces in a culture medium.29 Briefly, the calvaria was dissected under sterile conditions. After removal of periosteum and endosteum, the calvaria was cut into small pieces (2  2  2 mm) and placed in tissue culture disks for 15 min. The disks were added 5 mL of DMEM supplemented with 10% FBS, 10 mM N-(2-hydroxyethyl) piperazine-N0 -(2-ethanesulfonic acid) sodium salt (HEPES, Sigma), 1.0  105 U/L penicillin, and 100 mg/L streptomycin and cultured in a humidified incubator containing 5% CO2 at 37 °C for 24
48 h. The disks were then refreshed with 10 mL of the medium every two days. The 3T3 cells and osteoblasts were harvested at 70
80% confluence for the following assessment. Cell Adhesion and Spreading. Cell adhesion assessment was undertaken by seeding NIH 3T3 cells on the thin films of the composites according to our established method.30 Briefly, 24  24 mm siliconized cover slides were prepared by treatment with 2% dimethyldichorosilane (DMDC, Fluka) and baking at 180 °C for 4 h. Samples were dissolved in chloroform to a concentration of 1%, respectively. A total of 50 μL of the solution was transferred to a cover slide to form a thin film by evaporation in the air for 30 min and in a vacuum dryer for 48 h at the room temperature. The cover glasses were then sterilized with UV for 30 min and placed in a six-well tissue culture plates (TCPs, Costar). A total of 3 mL of medium was added to the wells to prevent the cover slides from floating during cell seeding. 3T3 cells, 1.0  105 (in 1 mL of medium), were then seeded into each well, and the plates were incubated at 37 °C and 5% CO2. The cover slides were washed three times with PBS and fixed with 3% glutaraldehyde in PBS at room temperature for 30 min. After being washed three times with PBS, the cells were dyed with one drop of toluidine blue staining for 30 min, washed with distilled water, and dried in the air. Cell attachment and cell morphology were observed under the reverse microscope (TE2000U, Nikon). To obtain the optimal content of RGD-copolymer in HP composite, the thin films of HP and RHP composites with different contents of RGD-copolymer (0.1, 0.5, 1, 5, and 10%; w/w) were prepared before scaffold fabrication and seeded with NIH 3T3 cells. The cell numbers per sight were counted at 0.5
3 h after cell seeding. Cytoxicity Measurement of Material Extracted Liquid. Cytoxicity of the materials to osteoblasts was evaluated by extracting the polymer materials with the culture medium and performing the 3-(4,5dimethylthiazoyl-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay in the presence of the extract liquids. Sections of the composites were immersed in the culture medium (10 mL of DMEM per 20 cm2 of material surface area) at 37 °C for 48 h to obtain the leachate, which is believed that the main part of the toxic components had been extracted out. The leachate was then diluted to 1-, 2-, and 4-fold of the original volume, respectively. The harvested osteoblasts of passages 2 and 3 were seeded in the 96-well TCPs with a density of 1  104 cells/200 μL/well and incubated in 200 μL of DMEM at 37 °C under an atmosphere of 5% CO2. After 48 h, the media in the wells were replaced with 200 μL/well of the leachate. Six duplicated wells were used for each liquid specimen. A total of 20 h later, 20 μL of MTT (Sigma, 5 mg/mL in PBS) was added into each well, and the incubation was kept for another 4 h. The medium was removed and replaced with 150 μL of DMSO. The plates were surged at 200 rpm for 15 min to dissolve formazan crystals formed and were measured at 492 nm using a Thermo Electron MK3 μm. Cell viability was expressed as a proportion of the absorbance value of TCPS in the same culture medium. Cell Viability. The cell viability of rabbit osteoblasts on porous scaffolds was also measured using MTT method to quantitatively assess the number of viable cells attached and grown on the substrate surfaces. The discs (5.5 mm in diameter, 2 mm in thickness) of porous scaffolds 2669

dx.doi.org/10.1021/bm2004725 |Biomacromolecules 2011, 12, 2667–2680

Biomacromolecules were prepared and placed into 96-well TCPs after sterilized with UV for 30 min. The bottom surfaces of the wells were fully covered by the discs. Each material was repeated in four parallel wells. The wells of TCPs with no material discs were used as control. Then, the harvested osteoblasts were seeded into 96-well TCPs with a density of 5  104 cells in 200 μL of medium, and incubated at 37 °C and 5% CO2 for 3 and 7 d. The medium was refreshed every 2 days. At the time interval, 20 μL of MTT (5 mg/mL in PBS) was added to each well and incubated at 37 °C and 5% CO2 for 4 h. The medium was removed and replaced with 0.2 mL of 0.04 M HCl in isopropanol to solubilize the converted dye. The solution (150 μL) of each well was mixed and transferred to another 96-well plate, and the absorbance values of the converted dye were measured at 540 nm using a Thermo Electron MK3 μm. The relative proliferation ratio (RPR) was calculated as the absorbance of test well/the absorbance of control well. Cell Cycle Analysis with FCM. The cell cycles of osteoblasts cultured on the thin films of different materials were analyzed using a flow cytometer (FCM, FACS Calibur, BD, U.S.A.). Solutions (1 wt %) of the composites in chloroform were cast on glass disks (75 cm in diameter) pretreated with DMDC, respectively,30 and the coated glass disks were dried and sterilized with UV. Three glass disks of each material were prepared as parallel samples in a certain time interval. Glass disks without any treatment were used as control. The obtained osteoblasts of 2
4 passages were seeded onto the disks in a density of 2  104 cells/cm2 and incubated in a humidified incubator of 37 °C with 5% CO2. The culture medium was changed every 2 days. The cells were harvested at the time intervals of 3 and 6 days and fixed with 75% (v/v) cold ethanol. The cells on three parallel disks of each material were collected together as a sample for FCM analysis. At each time interval, the proportion of cell cycles and apoptosis of at least 1  106 osteoblasts grown on different materials were assessed, respectively. 2.7. Animal Test. Isolation and Proliferation of Autologous Bone Marrow Mesenchymal Cells (MSCs). MSCs from rabbit bone marrow were isolated, cultured, and expanded according to the modified procedure described in ref 31. A total of 16 adolescent rabbits 16 weeks old, weighing 2.96 ( 0.25 kg, were used in this study. Half of the rabbits were selected with stochastic methods for implantation of tissueengineered implants and the others were used for implantation of cellfree scaffolds. The rabbits for implantation of tissue-engineered implants were anesthetized with 0.3% pentobarbital sodium and 2.0 mL of bone marrow was drawn respectively with injectors prewetted with 1% fresh heparin at the upper part of tibia of right hind limb. The bone marrow of each rabbit was immediately injected into a 50 mL tube containing 25 mL serum-free medium, dispersed completely, and rinsed with PBS three times by centrifugation at 1000 rpm for 5 min. The obtained cells were resuspended in DMEM supplemented with 10% FBS, 50 mg/L Lascorbic acid (Sigma), 10 mM HEPES, 1 μg/L basic fibroblast growth factor (bFGF, Gibco), 50 μg/L insulin (Sigma), 1.0  105 U/L penicillin, and 100 mg/L streptomycin, and plated at 8 mL aspirate/ cm2 in TCPs in a humidified incubator at 37 °C with 5% CO2. Nonadherent hematopoietic cells were removed after 4 days culture and the plates were washed gently with the culture medium. Thereafter, the medium was changed every two days. Primary MSCs were detached prior to confluence using 0.25% trypsin (Gibco)/0.02% EDTA and replated at 2  104 cells/cm2. The cells were passaged three to four times until the total MSCs of each rabbit were proliferated up to more than 5  106 cells for tissue engineering construction. Preparation of Tissue-Engineered Implants with Autologous MSCs. The bars (3 mm in width, 20 mm in length) of porous composite scaffolds of HP, RHP, and BRHP were seeded with MSCs using collagen hydrogel as a carrier. A 0.5 wt % bovine collagen solution was prepared by dissolving 10 mg bovine collagen (extracted from bovine achilles tendon according to ref 32) in 20 mL of 0.1% acetic acid solution at 4 °C. The pH value of the solution was adjusted to 7.0
7.2 with 1 M cold NaOH on ice. Thereafter, 1 mL of the solution was added into total a total of 1  107 MSCs of a rabbit in a tube and mixed gently. The

ARTICLE

Scheme 2. Schematic of RGD Conjugates at the Interface between the Cells and the Nanocomposite Scaffold of RGDCopolymer/g-HAP/PLGA (RHP)a

a

The symbols indicate cells (a), RGD-copolymer (b), g-HAP (c), and PLGA matrix (d).

collagen/cells mixture was seeded into two of different materials, and incubated at 37 °C for 10
20 min until the hydrogel formed in the scaffolds. These implants were continued to culture in vitro for 3 days before implantation. Implantation for Repair of Bone Defects. The cell-free scaffolds and the tissue-engineered implants of HP, RHP, and BRHP were grafted to bilateral critical-sized radius defects of 16 rabbits by surgery according to the refs 8 and 33. Briefly, the bone defects of 20 mm in length were created in bilateral radii of a rabbit under anesthesia with 30 mg/kg sodium pentobarbital and replaced with cell-free scaffolds or autologous tissue-engineered implants, respectively. The implants were similar in size to the rabbit radius and placed into the defect areas without any fixation. The wounds were closed in routine ways of surgery. The pure radius defects were used as control. Animals were kept in the Institute of Experimental Animal of Jilin University, in accordance with the institutional guidelines for care and use of laboratory animals. Computer Radiography and Histological Examination. The rabbits were sacrificed at 24 weeks postsurgery, and the biopsies of the repaired bone and the adjacent ulna were obtained together for computer radiography (CR, Kodak CR-400plus, U.S.A.) and histological examination. After CR examination, the biopsies were fixed in 4 wt % buffered paraformaldehyde for 6 days, decalcified in 10% EDTA for 3
4 weeks at 37 °C, then dehydrated in ascending grades of ethanol, and embedded in paraffin. The tissue blocks were sectioned at a 5 μm thickness using a microtome (Leika RM2145 microtome, Germany) and stained with H&E staining, toluidine blue staining, and Masson’s Trichrome staining for light microscopic observation. 2.8. Statistics Analysis. All quantitative data were analyzed with Origin 7.0 (OriginLab Corporation, U.S.A.) and expressed as the mean ( standard deviation. Statistical comparisons were carried out using analysis of variance (ANOVA, Origin7.0). A value of p < 0.05 was considered to be statistically significant.

3. RESULTS AND DISCUSSION Although RGD is widely believed to promote cell/biomaterial interactions, it is reported that direct surface modification of RGD peptide has a negative effect on HA implant performance.23 The competition of RGD peptides with adsorbed proteins for integrin receptors can reduce both the initial attachment and 2670

dx.doi.org/10.1021/bm2004725 |Biomacromolecules 2011, 12, 2667–2680

Biomacromolecules

ARTICLE

Figure 1. SEM micrographs of the porous scaffolds of HP (a, b), RHP (c, d), and BRHP (e, f) fabricated with the solvent casting/particulate leaching method. Pore size distributions of these scaffolds are shown in g. Bar lengths are 200 μm (a, c, and e) and 50 μm (b, d, and f).

survival of osteogenic cells on HA surfaces. On the other hand, RGD peptides will fall off from the HA surface because of the infirm combination RGD with HA or HA dissolving. These free RGD peptides can turn over to inhibit cell adhesion.19 This study aimed to develop a novel RGD-conjugated copolymer as a component incorporated into HP nanocomposite for promoting cell adhesion and application in tissue engineering and bone regeneration. The RGD conjugate was mixed with the composite of g-HAP/PLGA, and RGD peptides are expected to be exposed on the surface of scaffolds any time to play its function in cell adhesion and bone formation until the material biodegraded in vivo completely (as shown in Scheme 2).

3.1. Scaffold Characterization. Microstructure. Using the solvent casting/particulate leaching method to produce tissue scaffolds shows advantages in adjusting pore size and porosity by controlling size and amount of particulates to meet the needs of cell ingrowth and tissue regeneration. SEM micrographs of porous composite scaffolds are shown in Figure 1. The scaffolds were highly porous and their pores were irregular and interconnected. The porosity of scaffolds was in the range 86.3
87.7% (Table 1). The average pore size of HP, RHP, and BRHP was 140.3 ( 80.6, 132.1 ( 87.3, and 129.5 ( 88.2 μm (Table 1). Although the scaffolds incorporated with RGD-copolymer seemed to be more irregular and had more micropores of e50 μm in diameter and more macropores of 2671

dx.doi.org/10.1021/bm2004725 |Biomacromolecules 2011, 12, 2667–2680

Biomacromolecules

ARTICLE

Figure 3. EDX analysis showed the contents of calcium, phosphorus and nitrogen on the surfaces of RGD-copolymer (a), HP (b), RHP (c), and BRHP (d).

Figure 2. SEM micrographs showed surface topography of the porous scaffolds of HP (a), RHP (b), and BRHP (c). Bar lengths are 20 μm (a
c).

50
150 μm in diameter (Figure 1g) compared to that of HP, there were no significant differences in average pore size and porosity among three groups (p > 0.05; Table1). The reason is deduced that RGD-copolymer is amphiphilic and the amount of RGD-copolymer and BMP-2 in the composite was extremely little (only 1.0 and 0.4 wt %, respectively). Surface Properties. Figure 2 shows the surface topography of the pore walls in scaffolds. All composite scaffolds presented rough surfaces with dotted granules of uniform distribution because of the incorporation of g-HAP particles. It is regarded that surface roughness can enhance attachment, proliferation, and differentiation of enchorage-dependent bone forming cells.34 Pore wall roughness plays an important role in osteogenic outcomes as well as macroporosity (pore size > 50 μm).35 There were no apparent differences in the size or distribution of g-HAP particles exposed on the pore wall surface of three scaffolds. The EDX analysis was used to determine the surface elemental composition of the scaffolds (Figure 3 and Table 2). The atom ratio of nitrogen (N) to carbon (C) can provide a quantitative measure of the amount of grafted RGD peptide and BMP-2 at the material surface because N is only presented in the

RGD-copolymer and BMP-2 in these composites. As indicated in Table 2, the N/C ratio of RGD-copolymer was 0.148. It increased slightly from 0.037 for HP to 0.045
0.046 for the composites incorporated with RGD-copolymer (1 wt % of RGDcopolymer in the composite), indicating the actual existence of RGD peptide on these two composite scaffolds. Meanwhile, all composite scaffolds showed a high content of Ca and P (Figure 3 and Table 2), which resulted from the incorporation of HA particles into polymer matrix. Compared to HP, the contents of Ca and P in both RHP and BRHP decreased slightly because of introducing RGD-copolymer and BMP-2 (Table 2). The above-mentioned results indicated that RGD peptide, Ca, and P could be exposed on the surface of scaffolds simultaneously because the material system in the present study was prepared by g-HA/PLGA composite incorporated with RGD-copolymer. 3.2. Effects of RGD-Copolymer Contents in Composites on Cell Adhesion. To obtain the optimal content of RGD-copolymer in HP composite, cell adhesion of NIH 3T3 cells on the thin films of HP and RHP composites with different contents of RGD-copolymer were investigated. Figure 4 shows that the cell densities on all RHP composites were significantly higher than that on HP at 0.5 h (p < 0.05). The composites with the RGD-copolymer contents of 0.1, 0.5, and 1% exhibited higher cell adhesion ratios than others and had the greatest increment in cell adhesion ratios (131.2
157.1% higher than HP) but only at 0.5 h after cell seeding. The cell densities decreased significantly when the RGD-copolymer contents were increased up to more than 5%. The differences of cell density between HP and RHP composites were shortened after 1 h culture. At 3 h cell culture, the cell density on 1% RHP was still significantly higher than those on HP and 10% RHP, although the cells attached on HP increased largely. It is indicated that rapid cell adhesion could be promoted by the incorporation of RGD-copolymer at a suitable content. Thus, in this paper, 1% RHP was selected for scaffold fabrication and its following study. 3.3. Cytoxicity of Material Extracted Liquid. Figure 5a shows the cytotoxicity effects of the leaching liquids of various composites on osteoblasts evaluated by MTT assay. Compared to the culture medium, all leaching liquids did not display appreciable cytotoxicity because the cell viabilities were in the range of 92.0
117.3% for the pristine leaching liquids, in the range of 102.9
114.1% after 2-fold dilution and in the range of 100
109.5% after 4-fold dilution. The cell viability of HP was 15.7% higher than that of the control and only slightly decreased 2672

dx.doi.org/10.1021/bm2004725 |Biomacromolecules 2011, 12, 2667–2680

Biomacromolecules

ARTICLE

Table 2. Surface Elemental Analysis with Energy Dispersive X-ray Spectrometry (EDX)

a

samplesa

C (mol %)

N (mol %)

O (mol %)

P (mol %)

Ca (mol %)

N/C

RGD-copolymer

70.18 ( 2.65

10.34 ( 2.65b

18.50 ( 1.79

0.06 ( 0.023b

0.16 ( 0.05b

0.148 ( 0.035b

HP

65.54 ( 1.37

2.41 ( 0.65

28.29 ( 1.82

0.77 ( 0.14

1.64 ( 0.39

0.037 ( 0.010

RHP

64.49 ( 0.64

2.96 ( 0.62

29.38 ( 0.27

0.70 ( 0.03

1.40 ( 0.11

0.046 ( 0.010

BRHP

65.87 ( 0.83

2.97 ( 0.62

27.70 ( 0.87

0.70 ( 0.03

1.47 ( 0.15

0.045 ( 0.010

n = 5. b Indicates p < 0.05: RGD-copolymer vs the other three groups.

Figure 4. Cell adhesion of NIH 3T3 cells on a cover slide and the thin films of HP (0) and RHP with different contents of RGD-copolymer (0.1, 0.5, 1, 5, and 10%; w/w) after 0.5, 1, and 3 h culture in vitro. * indicated p < 0.05 compared to the other group, n = 9.

at the 2- and 4-fold diluted level. Similar results were reported by Zhang et al. when the cytotoxic effects of β-TCP on L929 qcells were evaluated,36 indicating that some ingredients in HA or β-TCP ceramics seemed to be available for cell viability. Although composite RHP showed relatively lower viability (92.0 ( 7.0%) than the other two scaffolds before dilution (p < 0.05), its viability increased with dilution to 100
102.9%. The cell viability of BRHP increased significantly at the initial concentration compared to RHP (p < 0.05) and was similar to that of HP. The results indicated that the three composites prepared in the present study were not toxic for osteoblast cells and could be used in cell growth or in tissue engineering. 3.4. Cell Proliferation on Composite Scaffolds. Figure 5b shows the cell relative proliferation ratios (RPR) of rabbit osteoblasts on porous scaffolds of HP, RHP, and BRHP compared to the cells on TCPs after 3 and 7 days culture analyzed with the MTT method. After 3 days culture in vitro, the RPRs of RHP and BRHP were higher than that of HP, and there was statistical difference between BRHP and HP (p < 0.05). After 7 days culture in vitro, the RPR of osteoblasts on all of three composite scaffolds were decreased obviously. Among them, there was nearly a half decrease in HP and BRHP compared to 3 days culture. The RPR of RHP scaffold was significantly higher than that of HP (p < 0.05). However, there were no significant differences between BRHP and the other two composites. It was deduced that RGD-copolymer promoted adhesion and proliferation of osteoblasts while the combination of BMP-2 and RGD-copolymer enhanced cell proliferation at an early stage but weakened the cell proliferation at the later stage. In other words, as one of the main growth factors in bone formation, the main function of BMP-2 is to induce differentiation of osteoblasts into osteocytes and accelerate bone matrix formation but not to promote the proliferation of the osteoblasts themselves.37

Figure 5. (a) Cytoxicity analysis of osteoblasts survived in material extract liquids using MTT assay. The initial extract ratios of all three materials were 2.0 cm2/mL; * indicated p < 0.05 compared to the other two groups at the same concentration, n = 6. (b) The relative proliferation ratios (RPR) of rabbit osteoblasts growing on the porous scaffolds of HP, RHP, and BRHP after 7 days culture in vitro. The RPR was calculated by ODexp/ODcontrol  100%. * indicated p < 0.05 compared to the other group, n = 4. (c) FCM analysis showed the cell cycle of osteoblasts grown on the thin film of HP, RHP, BRHP and the control group of glass at 3 and 6 days culture in vitro. 2673

dx.doi.org/10.1021/bm2004725 |Biomacromolecules 2011, 12, 2667–2680

Biomacromolecules

Figure 6. Representative computer radiographs of the critical radius defects (20 mm in length) at 24 weeks postsurgery: (a) pure defect; (b
d) implanted with the cell-free scaffolds of HP (b), RHP (c), and BRHP (d); (e
g) implanted with the tissue-engineered implants of HP (e), RHP (f), and BRHP (g).

3.5. Distribution of Cell Cycles. The cell cycles of osteoblasts cultured on the thin films of HP, RHP, and BRHP for 3 and 6 days analyzed with FCM are shown in Figure 5c. In a cell cycle, the four phases, G1, S, G2, and M make up the classic cell cycle “clock”. Cells can enter G0 only from G1, G0 cells can remain viable for months or even years, and most of the cells in the human body are in fact in this nondividing state. Cells are said to be terminally differentiated if they are unable to return to the cell division cycle.38 The cell cycle distribution of cultured osteoblasts is related to their nutritional condition and interface properties between cells and biomaterials. Especially for dividing or proliferating cells, two points in the cell cycle, including G1/S and G2/M, are particularly critical. These are the major control points, and before crossing them, the cell must be sure that conditions are such that S phase and M phase can be executed successfully.38 At 3 days culture, the distribution of osteoblasts on the control of glass was 76.0% in the G0/G1 phase, 13.5% in the S phase, and 16.5% in the G2/M phase. The cell numbers at G0/G1 and G2/M phases in the groups of RHP and BRHP were

ARTICLE

higher than that of HP. BRHP showed highest percentage of cells at G0/G1 or G2/M phase and lowest percentage of cells at S phase among all groups. At 6 days culture, osteoblasts in all groups synchronized to the specified G0/G1 phase, which the percentages of cells in this phase of all groups were nearly at the same level (83.7
84.6%). The cells of S phase only appeared in BRHP and the glass control, and their distributions were at a lower level of 5.5 and 3.9%, respectively. The percentage of cells in the G2/M phase in the group of BRHP was 9.9%, which was lower than that of 15.7% in HP, 16.1% in RHP, and 12.4% in the control. It indicated that the use of BMP-2 in RHP could retard cell cycle at S phase and decrease the number of cells at the phase of cell division (G2/M). The results of FCM analysis as abovementioned could explain partly the mechanism of different proliferation ratios of osteoblasts grown on the different composite scaffolds at 3 and 7 days culture. 3.6. Bone Healing. The bone healing was observed with macroscopic observation, X-ray examination, and histological analysis. In general, the critical size bone defect in rabbit is 15 mm in length, which cannot be bridged spontaneously without assistance of bone implantation or material replacement.39 In this study, the length of rabbit radius defects was 20 mm and was longer than those reported in the literature. To investigate the application of the composites incorporated with RGDcopolymer and BMP in bone tissue engineering, in this study, the tissue-engineered bone substitutes were constructed by porous scaffolds seeded with autologous MSCs of rabbits. MSCs are capable of undergoing differentiation into a variety of specialized mesenchymal tissues, including bone, tendon, cartilage, muscle, ligament, fat, and marrow stroma.40 They are prevalent in bone marrow even in adults and can be easily isolated and cultured in vitro for tissue engineering application. In this paper, all autologous MSCs from 16 adult rabbits were isolated and proliferated successfully. After 2
3 weeks culture, the cell number of each rabbit reached up to 5  106 cells, which met the demands of tissue engineering construction. With the assistance of collagen hydrogel, the present study explored a rapid reconstruction method of tissue-engineered grafts to carry autologous MSCs into bone defects through porous scaffolds. X-ray Evaluation. The typical radiographs of the repaired areas at 24 weeks postsurgery are shown in Figure 6. The defects in untreated control were clearly discernible without bone formation. This contrasted with the repaired areas in other groups treated with the cell-free scaffolds (Figure 6b
d) or the tissueengineered scaffolds (Figure 6e
g), where all defects were completely bridged and replaced with newly formed bony callus. An uncompleted new bone formation was observed in the groups of HP, although the bone defects could be bridged. The similar results have also been observed in our previous works.8,33 The reason is deduced that the surface modification of HA with PLLA could decrease surface exposure of Ca or P on HA and, thus, weakened the ability of cell adhesion and mineralization of the g-HAP/PLGA composite further to some extent. Among the groups of the cell-free scaffolds or the tissue-engineered scaffolds, both RHP and BRHP showed larger new bone formation and better bone fusion than HP. It implied that the composites incorporated with RGD-copolymer could enhance the bone regeneration whether they were cell-free scaffolds or tissueengineered implants. The differences in X-ray radiographs between the cell-free scaffolds and the tissue-engineered scaffolds were also observed. In the group of RHP, there were no obvious changes in the size 2674

dx.doi.org/10.1021/bm2004725 |Biomacromolecules 2011, 12, 2667–2680

Biomacromolecules

ARTICLE

Figure 7. Macroscopic observation of the representative repaired tissues at the critical-sized defects of rabbit radius in untreated control (A) and implanted with tissue-engineered porous scaffolds of HP (B, E), RHP (C, F), and BRHP (D, G). The bold arrows indicate the areas of radius defect or repaired tissues. (E
G) cross sections at the center of the repaired tissues, and the letters of I and NB indicate the implants and the newly formed bones of radius, respectively. The pictures were taken by a digital camera. Scale bars are 20 mm (a
d) and 2500 μm (e
g).

and X-ray density of a new bone between the cell-free scaffolds and the tissue-engineered scaffolds. However, in the group of HP, a larger volume and higher density of newly formed bony callus was observed at the sites treated with the tissue-engineered scaffold compared to that with the cell-free scaffold. Meanwhile, the tissue-engineered BRHP exhibited highest density of a new bone and best fusion at the interface between newly formed bone and neighboring normal radius compared to the cell-free scaffold itself and the tissue-engineered RHP. Macroscopic Observation. Figure 7 shows a macroscopic observation of the repaired tissues at the critical radius defects treated with tissue-engineered implants after an experimental time of 24 weeks. All defects were bridged with bony tissues in the groups of HP, RHP, and BRHP, while there was still a large area of bone defects in the untreated control group. Among them, the groups incorporated with RGD-copolymer or RGD-copolymer and BMP-2 showed larger new bone formation than the group of HP. Histological Observation. The above-mentioned results were further verified with histological Observation, including H&E

staining, toluidine blue staining and Masson’s trichrome staining. The interfaces between the newly formed bone tissue and the normal bone were observed by the micrographs of the vertical sections in Figure 8. In the group of HP, a snick was observed clearly at the interface between new bone and the normal bone because the diameter of the new bone was less than that of the neighboring normal bone. Except for HP, there were no apparent interfaces between the new bone and normal radius in RHP and BRHP. Figure 9 shows histological micrographs of cross sections at the center of repaired areas with Masson’s trichrome staining. Obviously, except for HP, the gradient new bone formation was still observed in the groups of RHP and BRHP. Among them, only RHP presented a typical woven bone at the areas near to the implants. There were tighter and more mature osseous tissues under the periosteum in BRHP than that in RHP. In the central areas of cortical bones, all groups exhibited mature osseous tissues with a large quantity of adjacent abutting osteons. There were more and larger exceptional cavities between adjacent abutting 2675

dx.doi.org/10.1021/bm2004725 |Biomacromolecules 2011, 12, 2667–2680

Biomacromolecules

ARTICLE

Figure 8. Macroscopic and microscopic observation of histological vertical sections at the repaired areas of the tissue-engineered implants of HP (a, d), RHP (b, e), and BRHP (c, f) at 24 weeks postsurgery. The sections were treated with Masson’s staining (a
c) and HE staining (d
f). The slender arrows indicate the interfaces between newly formed bony callus and the neighboring normal bone of radius, and bold arrows indicate normal ulna. The pictures were taken with a digital camera (a
c) and a light microscope (d
f). Scale bars are 4000 μm (a
c) and 100 μm (d
f).

osteons in the groups of HP and RHP than that in BRHP. From the density of red color (in Masson’s trichrome staining, red color represents higher cross-linked collagen and blue color represents lower cross-linked collagen), it is deduced that the mature degrees of osseous tissues in these areas were in the order of HP < RHP < BRHP. As shown in Figure 10, the interfaces between the new bones and the implants, and the ingrowth of new bone tissues were observed in all three groups. There were still a few of unabsorbed scaffolds observed at some areas after 24 weeks postsurgery. They were mainly on the sides of newly formed radius and some of them were wrapped with new bone callus. The scaffolds were

destroyed into small separated pieces and wrapped with invaded fibrous tissues. Most of them were hard to be distinguished except some areas in the group of BRHP. The infiltration of inflammatory cells at the interfaces was observed in all three groups. Compared to RHP and BRHP, the implant was mainly wrapped with the newly formed radius and there were fewer ingrown bone tissues formed in the group of HP. A typical woven bone structure was discovered in the group of RHP, which resulted from a large quantity of new bone ingrowing into the scaffold. Compared to RHP, the ingrown new bones in the group of BRHP were fewer but larger and most of them were covered by a layer of thick fibrous-like tissues. The fibrous-like tissue was in a 2676

dx.doi.org/10.1021/bm2004725 |Biomacromolecules 2011, 12, 2667–2680

Biomacromolecules

ARTICLE

Figure 9. Representative histological micrographs of cross sections at the center of repaired areas implanted with the tissue-engineered implants of HP (a, b), RHP (c, d), and BRHP (e, f) at 24 weeks postsurgery. The slices were treated with Masson’s Trichrome Staining. The bold arrows indicate the areas of newly formed bone tissue under periosteum and the slender arrows indicate the cavities in new bones. Scale bars are 100 (e) and 25 μm (a
d and f).

regular arrangement and connected the surfaces of a new bone with a brush-like structure. The similar structure also appeared at the same areas of RHP but only at the top of a small ingrown new bone in the group of HP. According to its stained blue in Masson’s trichrome staining, it mainly consisted of collagen fibers and may be associated with the formation of a new bone. Figure 11 shows higher magnification of histological structures under periosteum at the repaired areas treated with toluidine blue staining. In the group of HP, the tissue under periosteum was mature osseous tissue, so that only a few of osteocytes were in bone extracellular matrix. However, both in the groups of RHP and BRHP, there were still a large quantity of osteogenic cells under periosteum. In the group of RHP, the cells under the periosteum were in a slender shape and mostly like osteoblasts, which approached together and arranged regularly along the interface of the newly formed bone tissue. Compared to RHP, there were

more mature and larger osteocyte-like cells in a regular arrangement in BRHP, and the interface between bone tissue and osteogenic cells disappeared. From the above-mentioned results, the possible mechanisms of enhanced bone formation by RGD-copolymer are deduced as follows. First, the RGD copolymer that we employed is synthesized by RGD peptides conjugated with the carboxyl groups of PEG-PLA-PGL and have shown higher stability before it is degraded.25,33 According to the Scheme 2, the molecules of RGD-copolymer inlay in the PLGA matrix and can constantly support cell adhesion and tissue regeneration until the composite materials is absorbed completely. It is a possible reason why there was still a large quantity of osteogenic cells under the periosteum at the repaired areas in the groups of RHP and BRHP 24 weeks after surgery. The conjugates of biodegradable polyester and RGD peptides are regarded to be the best scaffold materials for 2677

dx.doi.org/10.1021/bm2004725 |Biomacromolecules 2011, 12, 2667–2680

Biomacromolecules

ARTICLE

Figure 10. Bone ingrowth at the center of repaired areas implanted with the tissue-engineered implants of HP (a, b), RHP (c, d), and BRHP (e, f) at 24 weeks postsurgery. The slices were of cross sections and treated with Masson’s Trichrome staining. The letters of I, NB, and FC indicate the areas of implants, new bones, and fibrous-like collagen, respectively. Scale bars are 100 (a-1
c-1) and 25 μm (b-2
c-2).

guiding tissue regeneration (GTR) in vivo and ex vivo. The level of cell calcification and the hard tissue compatibility of Ti are reported to be improved by immobilizing RGD through functional molecules which have a long molecular chain.41 Second, HA nanoparticles and RGD-copolymer can be exposed on the surface of the composite scaffolds simultaneously and may play their functions independently. It is supported by current results showing that there were similar high levels of Ca and P on the surface of all composite scaffolds and the N/C ratios on the surface of RGD-copolymer and RHP were higher than HP. Obviously, the improved healing of bone defects by the use of RGD conjugates whether in cell-free scaffolds or tissueengineered scaffolds is dedicated to the integrated effects of HA and RGD-copolymer, which can result in a rapid and stable fixation between bone and implant with the formation of an

interface matrix. The better combination of implant and bone is reached at the initial stage, the more new bone formation at the repaired area and more smooth fusion at the interface between newly formed bone and neighboring normal bone will be guided. It also implies that the competition of RGD peptides with adsorbed proteins for integrin receptors can be avoided through the polymer incorporated with HA and RGD-copolymer. Third, the composite incorporated with RGD-copolymer is biocompatible for osteoblasts and has a suitable microenvironment for supporting bone ingrowth. The biocompatibility or the tissue response of bone-implant interface is an important factor for controlled, guided, and rapid bone healing. It might be induced by chemical and physical properties of the implant itself.42 After the incorporation of RGD-copolymer, there were no obvious influences on the microstructure of the scaffolds and the 2678

dx.doi.org/10.1021/bm2004725 |Biomacromolecules 2011, 12, 2667–2680

Biomacromolecules

ARTICLE

migration. Although the number of attached cells is clearly related to RGD surface density, too high surface density of RGD will lead to reducing cell migration caused by an increased detachment force of the integrins tethering the cell to the surface.19 It is a complicated problem to control the amount of RGD on the surface of material. This has been realized by covalently conjugating of RGDC peptides with different well-controlled densities onto the surface of PET, and these densities were correlated clearly to osteoblast and endothelial cell adhesion and focal contact formation.20 In this paper, we provided another approach for controlling the density of RGD on the material surface easily, that is, to prepare RGD conjugate first, and then to mix it into another polymer matrix in a suitable content. It suggests a new strategy for application of RGD or RGD copolymer in a controlled way. However, the obvious limitation of the RGD-copolymer was observed in present study. Although it can promote cell adhesion and immigration of ostogenic cells into porous scaffolds so as to improve bone ingrowth, it might not be beneficial for bone matrix formation. Thus, except for RGD conjugate, the combination of BMP-2 with RGD conjugate is necessary for osteogenic cells growth and differentiation, as well as bone regeneration. BMP-2 may play an important role in inducing osteoblast differentiation and bone matrix expression. Jeschke et al.43 suggested that the combination of growth factors and adhesion peptides might be a way to use the favorable effect of RGD-peptides on cell attachment but circumvent the negative effect on matrix synthesis. It is verified by the results of animal tests in the present study. The group of tissue-engineered BRHP exhibited the best bone healing, as shown in radiograph examination and histological analysis, while there were more woven bones formed in the group of RHP, although it supported bone ingrowth. Meanwhile, the osteogenic cells under periosteum and bone collagen in cortex at the repaired area in the group of BRHP were more mature than that of RHP. It is suggested that the growth and differentiation of osteogenic cells can be adjusted through the incorporation of BMP-2 at the proper periods of bone reconstruction.

distribution of HA nanoparticles on the composites’ surfaces. The porosity and pore size and the levels of Ca and P exposed on these novel composite scaffolds are similar to those of HP. However, the composite scaffolds of RHP promoted cell adhesion of 3T3 significantly and had a larger new bone and better fusion interface than HP. It is deduced that the improved bone ingrowth observed in the group of RHP may result from the use of RGD-conjugated copolymer, supporting osteoblasts adhesion and migration even if in cell-free scaffolds. Meanwhile, in the use of RGD peptides, RGD surface density is also a critical factor associated with cell spreading and cell

’ AUTHOR INFORMATION Corresponding Author

*Tel.: þ8643185262112. Fax: þ8643185685653. E-mail: xschen@ ciac.jl.cn. Author Contributions

)

Figure 11. Representative histological micrographs of vertical sections of repaired areas implanted with the tissue-engineered implants of HP (a), RHP (b), and BRHP (c) at 24 weeks postsurgery. The slices were treated with toluidine blue staining. The bold arrows indicate the areas of newly formed bone tissue under periosteum. After treatment with toluidine blue staining, fibrous tissue and bone tissue show bluish green, while cartilage matrix shows bluish violet. All scale bars are 25 μm.

4. CONCLUSION The 3-D porous nanocomposite scaffold of g-HAP containing RGD-conjugated copolymer has been fabricated and tested for repairing rabbit bone defects using tissue-engineering technology. Promoted cell adhesion and proliferation and enhanced new bone formation by the use of RGD-copolymer in composite have been demonstrated in this study. RGD-copolymer was mainly to improve bone ingrowth and fusion interface but not increase the bone matrix formation. The combined application of RGDcopolymer and growth factors (BMP-2) is necessary for improving the bone healing quality because it could also induce osteoblast differentiation and bone matrix expression. This study suggests a new strategy for the use of RGD peptides in HA composite scaffolds by the form of RGD-conjugated copolymer combined with growth factors, which can support cell adhesion and tissue regeneration effectively and constantly.

Co-first-authors.

2679

dx.doi.org/10.1021/bm2004725 |Biomacromolecules 2011, 12, 2667–2680

Biomacromolecules

’ ACKNOWLEDGMENT We gratefully acknowledge the financial support from the National Natural Science Foundation of China (Nos. 50733003, 30772209, and 50973109), the ‘‘863’’ Project (No. 2007AA03Z320) from the Ministry of Science and Technology of China, as well as Major Project of International cooperation from the Ministry of Science and Technology of China (2010DFB50890). ’ REFERENCES (1) Kim, S. S.; Sun Park, M.; Jeon, O.; Yong Choi, C.; Kim, B. S. Biomaterials 2006, 27 (8), 1399–409. (2) Tsurushima, H.; Marushima, A.; Suzuki, K.; Oyane, A.; Sogo, Y.; Nakamura, K.; Matsumura, A.; Ito, A. Acta Biomater. 2010, 6 (7), 2751–9. (3) Watanabe, J.; Akashi, M. Biomacromolecules 2007, 8 (7), 2288– 93. (4) Hong, Z. K.; Qiu, X. Y.; Sun, J. R.; Deng, M. X.; Chen, X. S.; Jing, X. B. Polymer 2004, 45, 6705–13. (5) Ren, J.; Zhao, P.; Ren, T.; Gu, S.; Pan, K. J. Mater. Sci.: Mater. Med. 2008, 19 (3), 1075–82. (6) Salerno, A.; Zeppetelli, S.; Di Maio, E.; Iannace, S.; Netti, P. A. Biotechnol. Bioeng. 2010, 108 (4), 963–76. (7) Hong, Z.; Zhang, P.; Liu, A.; Chen, L.; Chen, X.; Jing, X. J. Biomed. Mater. Res. A 2007, 81 (3), 515–22. (8) Zhang, P.; Hong, Z.; Yu, T.; Chen, X.; Jing, X. Biomaterials 2009, 30 (1), 58–70. (9) Jose, M. V.; Thomas, V.; Johnson, K. T.; Dean, D. R.; Nyairo, E. Acta Biomater. 2009, 5 (1), 305–15. (10) Petricca, S. E.; Marra, K. G.; Kumta, P. N. Acta Biomater. 2006, 2 (3), 277–86. (11) Anselme, K. Biomaterials 2000, 21 (7), 667–81. (12) Burridge, K.; Fath, K. Bioessays 1989, 10 (4), 104–8. (13) Heinemann, C.; Heinemann, S.; Bernhardt, A.; Worch, H.; Hanke, T. Biomacromolecules 2008, 9 (10), 2913–20. (14) Wilda, H.; Gough, J. E. Biomaterials 2006, 27 (30), 5220–9. (15) Quirk, R. A.; Chan, W. C.; Davies, M. C.; Tendler, S. J.; Shakesheff, K. M. Biomaterials 2001, 22 (8), 865–72. (16) Ruoslahti, E. Annu. Rev. Cell Dev. Biol. 1996, 12, 697–715. (17) Pierschbacher, M. D.; Ruoslahti, E. Nature 1984, 309 (5963), 30–3. (18) Cook, A. D.; Hrkach, J. S.; Gao, N. N.; Johnson, I. M.; Pajvani, U. B.; Cannizzaro, S. M.; Langer, R. J. Biomed. Mater. Res. 1997, 35 (4), 513–23. (19) Hersel, U.; Dahmen, C.; Kessler, H. Biomaterials 2003, 24 (24), 4385–415. (20) Chollet, C.; Chanseau, C.; Remy, M.; Guignandon, A.; Bareille, R.; Labrugere, C.; Bordenave, L.; Durrieu, M. C. Biomaterials 2009, 30 (5), 711–20. (21) Dettin, M.; Conconi, M. T.; Gambaretto, R.; Bagno, A.; Di Bello, C.; Menti, A. M.; Grandi, C.; Parnigotto, P. P. Biomaterials 2005, 26 (22), 4507–15. (22) Kantlehner, M.; Schaffner, P.; Finsinger, D.; Meyer, J.; Jonczyk, A.; Diefenbach, B.; Nies, B.; Holzemann, G.; Goodman, S. L.; Kessler, H. ChemBioChem 2000, 1 (2), 107–14. (23) Hennessy, K. M.; Clem, W. C.; Phipps, M. C.; Sawyer, A. A.; Shaikh, F. M.; Bellis, S. L. Biomaterials 2008, 29 (21), 3075–83. (24) Weiger, M. C.; Park, J. J.; Roy, M. D.; Stafford, C. M.; Karim, A.; Becker, M. L. Biomaterials 2010, 31 (11), 2955–63. (25) Deng, C.; Tian, H.; Zhang, P.; Sun, J.; Chen, X.; Jing, X. Biomacromolecules 2006, 7 (2), 590–6. (26) Daly, W. H.; Poche, D. Tetrahedron Lett. 1988, 29 (46), 5859–62. (27) Deng, C.; Rong, G.; Tian, H.; Tang, Z.; Chen, X.; Jing, X. Polymer 2005, 46 (3), 653–9. (28) Zhang, R.; Ma, P. X. J. Biomed. Mater. Res. 1999, 44 (4), 446–55. (29) Cao, X. Y.; Yin, M. Z.; Zhang, L. N.; Li, S. P.; Cao, Y. Biomed. Mater. 2006, 1 (4), L16–9.

ARTICLE

(30) Hong, Z.; Zhang, P.; He, C.; Qiu, X.; Liu, A.; Chen, L.; Chen, X.; Jing, X. Biomaterials 2005, 26 (32), 6296–304. (31) Roostaeian, J.; Carlsen, B.; Simhaee, D.; Jarrahy, R.; Huang, W.; Ishida, K.; Rudkin, G. H.; Yamaguchi, D. T.; Miller, T. A. J. Surg. Res. 2006, 133 (2), 76–83. (32) Stone, K. R.; Steadman, J. R.; Rodkey, W. G.; Li, S. T. J. Bone Joint Surg. Am. 1997, 79 (12), 1770–7. (33) Cui, Y.; Liu, Y.; Jing, X.; Zhang, P.; Chen, X. Acta Biomater. 2009, 5 (7), 2680–92. (34) Yuan, H.; Kurashina, K.; de Bruijn, J. D.; Li, Y.; de Groot, K.; Zhang, X. Biomaterials 1999, 20 (19), 1799–806. (35) Karageorgiou, V.; Kaplan, D. Biomaterials 2005, 26 (27), 5474–91. (36) Zhang, Z.; Feng, X.; Mao, J.; Xiao, J.; Liu, C.; Qiu, J. Biochem. Biophys. Res. Commun. 2009, 379 (2), 557–61. (37) Senta, H.; Park, H.; Bergeron, E.; Drevelle, O.; Fong, D.; Leblanc, E.; Cabana, F.; Roux, S.; Grenier, G.; Faucheux, N. Cytokine Growth Factor Rev. 2009, 20 (3), 213–22. (38) Bolsover, S. R.; Hyams, J. S.; Shephard, E. A.; White, H. A.; Wiedemann, C. G., Cell Biology: A Short Course, 2nd ed.; Wiley-Liss: New Jersey, 2003; pp 401
422. (39) Mokbel, N.; Bou, S. C.; Matni, G.; Naaman, N. Oral Maxillofacial Surg. 2008, 12 (2), 73–8. (40) Sharma, P.; Cartmell, S.; Haj, A. J. E. Bone Tissue Engineering. In Applications of Cell Immobilisation Biotechnology; Nedovic, V., Willaert, R., Eds.; Springer: Dordrecht, Netherlands, 2005; pp 153
166. (41) Oya, K.; Tanaka, Y.; Saito, H.; Kurashima, K.; Nogi, K.; Tsutsumi, H.; Tsutsumi, Y.; Doi, H.; Nomura, N.; Hanawa, T. Biomaterials 2009, 30 (7), 1281–6. (42) Puleo, D. A.; Nanci, A. Biomaterials 1999, 20 (23
24), 2311–21. (43) Jeschke, B.; Meyer, J.; Jonczyk, A.; Kessler, H.; Adamietz, P.; Meenen, N. M.; Kantlehner, M.; Goepfert, C.; Nies, B. Biomaterials 2002, 23 (16), 3455–63.

2680

dx.doi.org/10.1021/bm2004725 |Biomacromolecules 2011, 12, 2667–2680