dextran hydrogel for bone tissue regeneration

dextran hydrogel for bone tissue regeneration

Acta Biomaterialia 88 (2019) 503–513 Contents lists available at ScienceDirect Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiom...

4MB Sizes 0 Downloads 12 Views

Acta Biomaterialia 88 (2019) 503–513

Contents lists available at ScienceDirect

Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat

Full length article

A strong, tough, and osteoconductive hydroxyapatite mineralized polyacrylamide/dextran hydrogel for bone tissue regeneration Ju Fang a,b, Pengfei Li c, Xiong Lu c, Liming Fang d, Xiaoying Lü b, Fuzeng Ren a,⇑ a

Department of Materials Science and Engineering, Southern University of Science and Technology, Shenzhen, Guangdong 518055, China State Key Laboratory of Bioelectronics, Southeast University, Nanjing, Jiangsu 210096, China c Key Lab of Advanced Technologies of Materials, Ministry of Education, School of Materials Science and Engineering, Southwest Jiaotong University, Chengdu, Sichuan 621000, China d National Engineering Research Center for Tissue Restoration and Reconstruction, South China University of Technology, Guangzhou, Guangdong 510006, China b

a r t i c l e

i n f o

Article history: Received 5 November 2018 Received in revised form 12 February 2019 Accepted 13 February 2019 Available online 14 February 2019 Keywords: Tough hydrogel Osteoconductive Osseointegration Bone regeneration

a b s t r a c t The design of hydrogels with adequate mechanical properties and excellent bioactivity, osteoconductivity, and capacity for osseointegration is essential to bone repair and regeneration. However, it is challenging to integrate all these properties into one bone scaffold. Herein, we developed a strong, tough, osteoconductive hydrogel by a facile one-step micellar copolymerization of acrylamide and urethacrylate dextran (Dex-U), followed by the in situ mineralization of hydroxyapatite (HAp) nanocrystals. We show that the soft, flexible, and hydrophobically associated polyacrylamide (PAAm) network is strengthened by the stiff crosslinked Dex-U phase, and that the mineralized HAp simultaneously improves the mechanical properties and osteoconductivity. The obtained HAp mineralized PAAm/Dex-U hydrogel (HAp-PADH) has extremely high compressive strength (6.5 MPa) and enhanced fracture resistance (over 2300 J m2), as compared with pure PAAm hydrogels. In vitro, we show that the mineralized HAp layer promotes the adhesion and proliferation of osteoblasts, and effectively stimulates osteogenic differentiation. Through the in vivo evaluation of hydrogels in a femoral condyle defect rabbit model, we show regeneration of a highly mineralized bone tissue and direct bonding to the HAp-PADH interface. These findings confirm the excellent osteoconductivity and osseointegration ability of fabricated HAp-PADH. The present HApPADH, with its superior mechanical properties and excellent osteoconductivity, should have great potential for bone repair and regeneration. Statement of Significance We developed a strong, tough, and osteoconductive hydrogel by a facile one-step micellar copolymerization of acrylamide and urethane methacrylate dextran (Dex-U), followed by the in situ mineralization of hydroxyapatite (HAp) nanocrystals. The hydrophobic micellar copolymerization and introduction of the stiff crosslinked Dex-U phase endowed the soft polyacrylamide (PAAm) network with enhanced strength and toughness. The in situ mineralized HAp nanocrystals on the hydrogels further improved the mechanical properties of the hydrogels and promoted osteogenic differentiation of cells. Mechanical tests together with in vitro and in vivo evaluations confirmed that the HAp mineralized PAAm/Dex-U hydrogel (HAp-PADH) achieved a combination of superior mechanical properties and excellent osseointegration, and thus may offer a promising candidate for bone repair and regeneration. Ó 2019 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction Hydrogels have been considered as promising candidates for tissue engineering scaffolds, because their hydrated 3-D porous structure can mimic the critical properties of native-like extracellular matrix (ECM) for signaling, nutrient transport, and ⇑ Corresponding author. E-mail address: [email protected] (F. Ren). https://doi.org/10.1016/j.actbio.2019.02.019 1742-7061/Ó 2019 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

homeostasis, as well as support cell ingrowth, proliferation, and differentiation [1]. However, the inadequate mechanical properties of hydrogels greatly limit their application in load-bearing tissues, such as bone, tendon, ligament, and cartilage [2]. Many research efforts have been devoted to developing tough hydrogels to satisfy the mechanical requirements for load-bearing conditions [3–7]. Polyacrylamide (PAAm) is a widely used synthetic polymer in hydrogel synthesis because of its superior hydrophilicity and non-toxicity [8]. However, the covalently crosslinked

504

J. Fang et al. / Acta Biomaterialia 88 (2019) 503–513

PAAm-based hydrogels are usually brittle due to the soft nature of PAAm and the lack of a mechanism for energy dissipation [9]. Strategies based on non-covalent crosslinking or a combination of distinct mechanisms have been carried out to improve the mechanical performance of the PAAm hydrogel. It has been reported that hydrogels formed by hydrophobic crosslinking can achieve high toughness and excellent fatigue resistance [10]. Recently, Okay and co-workers developed polyacrylic acid and PAAm hydrogels that were tough and self-healing (i.e., bearing the autonomous capacity for repair) through the copolymerization of acrylic acid and acrylamide, using blocks of large hydrophobes in a micellar system with electrolytes [11,12]. Chen et al. designed a high-strength, self-healing PAAm/agar hydrogel through a combination of hydrophobic interactions and physical associations [13]. These strategies were successful in modifying the soft, brittle PAAm hydrogel into a resilient, high-strength network. However, these hydrogels still lack the toughness required for use in cartilage or bone tissues. An ideal bone scaffold requires not only high mechanical strength to support tissue regeneration, but also should be endowed with biofunctional cues to facilitate cell ingrowth and tissue integration [14]. For the replacement or regeneration of large bone defects, the scaffold should be well anchored, integrate with the nascent tissue to prevent degeneration or deformation of the host tissue, and thus, maintain the load-bearing function during bone remodeling [15]. Conventional polymeric hydrogels are poorly adhesive to cells and cannot directly bond to host bone due to a lack of capacity for protein absorption, osteoconduction, and osteoinduction. To overcome these shortcomings, bioceramic nanoparticles and clay-based platelets have been incorporated into hydrogels to provide mineral/polymeric hybrid interfaces that can trigger mineral deposition [16–20]. Haraguchi et al. fabricated a clay-reinforced poly(N-isopropyl acrylamide) hydrogel with extraordinary mechanical, optical, and swelling/de-swelling properties [21,22]. Thorpe et al. developed an injectable laponitecrosslinked poly-(N-isopropyl acrylamide) hydrogel loaded with hydroxyapatite (HAp) that could effectively induce the osteogenic differentiation of human mesenchymal stem cells, with good biocompatibility and osseointegration in a rat femoral defect [23,24]. Despite these certain achievements, it remains challenging to engineer such multifunctional hydrogel scaffolds with strong mechanical properties and good osseointegration properties for large bone defects [25–28]. In this study, we developed a strong, tough, osteoconductive hydrogel by a facile one-step micellar copolymerization of acrylamide and urethane methacrylate dextran (Dex-U), followed by the in situ mineralization of HAp nanocrystals. The hydrophobic micellar copolymerization and introduction of the stiff crosslinked Dex-U phase endowed the soft PAAm network with enhanced strength and toughness. The in situ-mineralized HAp nanocrystals further improved the mechanical properties of the hydrogels and promoted the later osteogenic differentiation of cells. Mechanical tests together with in vitro and in vivo evaluations confirmed that this HAp mineralized PAAm/Dex-U hydrogel (HAp-PADH) achieved a combination of superior mechanical properties and excellent osseointegration, and thus may offer a promising candidate for bone repair and regeneration.

2. Materials and methods 2.1. Material preparation 2.1.1. Synthesis of Dex-U Three grams of dextran (Mw = 100,000, JK Chemical, China) and 0.05 g dibutyltin dilaurate (Sigma-Aldrich, China) were dissolved

into 200 ml dimethyl sulfoxide (DMSO, Sigma-Aldrich, China). 2isocyanatoethyl methacrylate (1.67 g) (IEMA, Sigma-Aldrich, China) was dissolved into another 30 ml of DMSO. The IEMA solution was added dropwise to the dextran solution under nitrogen protection for 8 h at 35 °C. The product was precipitated by pouring into excess isopropyl alcohol, centrifugation and dialysis against deionized water for 1 week. Dex-U was obtained by lyophilization. Proton nuclear magnetic resonance (1H NMR) spectra of Dex and Dex-U were obtained using an NMR spectrometer (AVANCE III 400M, Bruker, Germany) at room temperature. Deuterated DMSO was used as the solvent. Fourier transform infrared (FTIR) spectra of Dex and Dex-U were recorded using an FTIR spectrometer (Frontier, PerkinElmer, USA). Samples were mixed with KBr and pressed into pellets, and then scanned from 4000 cm1 to 500 cm1. 2.1.2. Preparation of PAAm/Dex-U hydrogel (PADH) The pre-solution for the PAAm gel was synthesized as reported previously [13]. Briefly, 7% w/v sodium dodecyl sulfate (SDS, Aladdin Bio-Tech, China) was mixed with 0.5 M NaCl aqueous solution. Stearyl methacrylate (SMA, 1 mol. % of total monomer, Sigma-Aldrich, China) was added to the mix, and vigorously stirred overnight at 35 °C. Acrylamide (18 wt%, Aladdin Bio-Tech, China), Dex-U (6 wt%), and the initiator (Iragcure 2925, 0.5 wt%, SigmaAldrich, China) were added to the pre-solution. The reactor was then sealed, and the reaction solution was degassed by nitrogen and ultrasonically treated until transparent. The PADH was synthesized with an ultraviolet lamp (k = 365 nm, intensity = 8 W) to irradiate the precursor solution for 1 h at room temperature. The obtained hydrogel was dialyzed against deionized water to ensure no remaining catalyst or monomer. During this period, deionized water was replaced three times each day. Following the same procedure, pure PAAm hydrogel without Dex-U (PAH) was prepared as a control. The swelling ratios of the PAH and PADH were measured using a gravimetric method. The original weight of the hydrogel (Wo) and the weight after immersion in the deionized water at predefined time intervals (Ws) were recorded. The swelling ratio was defined as 100  (Ws/Wo). 2.1.3. In situ mineralization of PADH After purification, the PADH was immersed overnight in 500 mM dipotassium hydrogen phosphate (K2HPO4) followed by 300 mM calcium chloride (CaCl2). The Ca/P atomic ratio was fixed to 1.67 to precipitate HAp nanocrystals. The mineralized hydrogel was transferred into ammonium hydroxide solution (pH = 11) for the stabilization of calcium phosphate precipitate. Finally, the obtained HAp-mineralized PAAm/dextran hydrogel (HAp-PADH) was immersed in phosphate-buffered saline (PBS) for at least 5 days to remove unreacted substances. 2.2. Characterization of hydrogels The hydrated hydrogels were frozen at 80 °C and lyophilized at 55 °C and 15 Pa until all water was sublimed. The network morphology of the freeze-dried hydrogels and the mineralized surface composition were examined using a scanning electron microscope (SEM; MIRA 3, TESCAN, Czech Republic) equipped with an energy dispersive X-ray spectroscopy (EDX) system (X-MaxN 50 mm2 silicon drift detector; Oxford Instruments, UK). The topographic maps of the hydrogel surfaces were analyzed using a 3D optical microscope (ContourGT-K, Bruker; Germany). The phase of the calcium phosphate was analyzed by X-ray diffraction (XRD; Rigaku Smartlab, Japan) using Cu-Ka radiation generated at 45 kV and 200 mA. FTIR and thermogravimetric analysis (TGA; STARe system, Mettler Toledo, USA) were also performed to further

505

J. Fang et al. / Acta Biomaterialia 88 (2019) 503–513

confirm that the deposited calcium phosphate was HAp. For FITR and TGA sample preparation, the PADH and HAp-PADH were completely dried in an oven at 60 °C overnight and then ground into powders. The FTIR spectra were obtained following the same procedure as that mentioned above for Dex and Dex-U. For TGA, the dried hydrogel powders were heated to 700 °C at a heating rate of 10 °C/min under nitrogen atmosphere. 2.3. Mechanical tests The mechanical properties of the hydrogels were measured by uniaxial compression and cyclic compression tests using a universal testing machine (Instron 5967, USA) equipped with a 1000-N load cell. Hydrogels were synthesized in a cylindrical plastic mold that was 10 mm in diameter and 30 mm in height. The obtained cylindrical gels were then placed on the lower plate of the test machine and compressed at a constant speed of 20% height per min to 95% compressive strain (to protect the machine). The fracture energy was obtained by integration of the stress-strain curve. The stress at 90% strain was recorded as the compressive stress. For cyclic compression tests, the hydrogel samples were continuously compressed for five cycles to 80% strain. Energy dissipation was calculated from the area between the loading–unloading curves. All hydrogels were immersed in deionized water for 24 h before testing. 2.4. In vitro cytocompatibility evaluation 2.4.1. Cell culture Mouse calvaria-derived MC3T3-E1 (subclone 14) osteoblast-like cells were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). MC3T3-E1 osteoblasts were cultured in minimal essential medium (MEM; HyClone, Australia) containing 10% fetal bovine serum (FBS; Hyclone) and 1% streptomycinpenicillin (Hyclone) at 37 °C under 5% CO2 atmosphere. The hydrogels were cut into discs (10 mm in diameter and 3 mm in height), washed with deionized water and ethanol in sequence thrice, and immersed in ethanol under ultraviolet light for 24 h. The sterilized gel discs were re-hydrated in PBS and then transferred into 48-well plates. Cells were seeded onto the discs according to the densities indicated in each section below. 2.4.2. Cell adhesion Immunofluorescence staining was performed to study the adhesion of MC3T3-E1 cells on the hydrogels. MC3T3-E1 cells were seeded onto purified hydrogels at a density of 1  104 cells per well. The cell-seeded hydrogels after 1 day of incubation were carefully washed with PBS, fixed with 4% (v/v) paraformaldehyde for 15 min, and then permeabilized with 0.1% Triton X-100 for 20 min. The samples were blocked with 3% BSA for 30 min before incubating with 10 lg/mL Alexa Fluor 568 conjugated to phalloidin for 2 h at room temperature to label actin filaments (F-actin). After washing three times with PBS, 1 lg/mL Hoechst was used to label the cell nuclei. The stained samples were washed three times with PBS. An inverted fluorescence microscope (Zeiss Axio Observer Z1, Germany) was used to observe the cells attached to the hydrogels.

reader (Cytation 3, Biotek, USA) at 570 nm. Five parallel groups were measured. 2.5. Osteogenic analysis of MC3T3-E1 cells on hydrogels 2.5.1. Alkaline phosphate (ALP) activity The osteogenic differentiation of MC3T3-E1 osteoblasts on the hydrogels was analyzed using an ALP activity assay. Cells (1  106) were seeded onto the hydrogel samples in a 48-well plate. After 7 days of incubation, the medium was removed, and the cells were washed twice with PBS. The cells were lysed with 200 ll of 1% Triton-X-100 through three rounds of freezethawing between 80 °C and 4 °C. The total protein concentration of each sample was detected using a BCA Kit (Jiancheng Biotech, China). ALP activity was measured using an ALP Assay Kit (Jiancheng Biotech, China), in accordance with the manufacturer’s instructions, and calculated by normalizing the optical density values to the corresponding total protein content of the same cell lysate. The experiment was performed with five parallel groups. 2.5.2. Quantitative real-time polymerase chain reaction (RT-PCR) analysis Osteogenic gene expression was analyzed by quantitative realtime RT-PCR. MC3T3-E1 cells were seeded onto hydrogels at a density of 1  106 cells per well. Total RNA of cells was extracted after 14 days of incubation using TRIzol reagent (Invitrogen, USA) following the manufacturer’s protocol. RNA was reverse transcribed to complementary DNA (cDNA) using a High Capacity cDNA Reverse Transcription Kit (Applied Biosystem, USA). Four genes closely related to osteogenic differentiation—ALP, collagen I (COL), osteocalcin (OCN) and osteopontin (OPN)—were analyzed by quantitative RT-PCR using a SYBR green qPCR kit (PowerUp SYBR Green Master Mix; Applied Biosystem, USA) and an ABI 7500 qPCR system (Applied Biosystem, USA). Human glyceraldehyde-3phosphate dehydrogenase (GAPDH) was used as the housekeeping gene. The normalized values of the expression level relative to GAPDH are presented. The primers are listed in Table 1. 2.6. In vivo evaluation of bone regeneration 2.6.1. Animal surgical procedure A total of 18 female New Zealand white rabbits (6 months old; weight, 2.8–3.2 kg) were used for the surgical animal model. All animal experiments and procedures were carried out in accordance with the permission and regulations of the Animal Care and Use Committee of Southern University of Science and Technology. Rabbits were divided into two groups: PADH-implantation group and the HAp-PADH-implantation group. Briefly, rabbits were anesthetized with pentobarbital (2%, 30 mg/kg) via an ear vein

Table 1 Gene and primers used in qRT-PCR. Gene

Primer sequence

ALP

F: 50 -TTTAGTACTGGCCATCGGACC-30 R: 50 -TCTCTGGCACAAATGAGTTGG-30 F: 50 -TCAAGATGGTGGCCGTTACT-30 R: 50 -CATCTTGAGGTCACGGCATG-30 F: 50 - ATTGTTTGAGGGGCCTGGGA-30 R: 50 -TATGGACAGCCTCTGACAGC-30 F: 50 -AACAATCCGTGCCACTCACT-30 R: 50 -TTGCAATTTCTTCCTTCGGCA-30 F: 50 -CCGTATTCAGCATTCTATGCTCTC-30 R: 50 -TGCCTCAAGTTCAGGAGGTC-30

COL

2.4.3. Cell proliferation The proliferation of osteoblasts on the hydrogels was assessed by MTT assay. MC3T3-E1 cells were seeded onto the purified hydrogels at a density of 1  104 cells per well. The cell-seeded samples after 1, 3, and 5 days of incubation were carefully rinsed with PBS and then incubated with 5 mg/ml MTT (Sigma-Aldrich, China) for 3 h at 37 °C. DMSO was added to dissolve the formazan crystals produced by the living cells. The optical density (OD) of the formazan solution was measured using a multi-mode microplate

OCN OPN GAPDH

Abbreviations: F, forward; R, reverse; ALP, alkaline phosphatase; COL, collagen 1; OCN, osteonectin; OPN, osteopontin; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

506

J. Fang et al. / Acta Biomaterialia 88 (2019) 503–513

injection. The anesthetized rabbits were positioned on a veterinary surgical table. The left hind limb was shaved and sterilized with 10% povidone-iodine solution. A skin incision was carefully created above the knee joint, and the femoral condyle was exposed by separating the muscles and sarcolemma. A cylindrical defect of 5 mm in depth and 3 mm in diameter was created using a skull drill at the femoral condyle with constant rinsing with saline. A cylindrical hydrogel sample, fabricated to the same size as the defect, was implanted. The synovial capsule and skin on both tibias were carefully sutured. After 0, 30, and 90 days of implantation, 6 rabbits (3 PADH implanted models and 3 HAp-PADH implanted models) were anaesthetized with an injection of pentobarbital. The knee specimens were harvested and fixed with 10% neutral-buffered formalin for 24 h. The fixed knee specimens were first scanned using microcomputer tomography (micro-CT), and then processed for histology. 2.6.2. Micro-CT analysis Micro-CT was performed on a Bruker SkyScan1176 Micro-CT scanner (Bruker, Germany) with a source voltage of 90 kV, a source current of 270 lA, and an exposure time of 550 ms. After scanning, the sagittal and axial planes of each implant were reconstructed using NRecon software. Three-dimensional analyses were performed using CTvol and CTvox software. Each scan was reconstructed using the same calibration parameters. Bone mineral content (BMC) and Bone volume (BV) were calculated using a cylindrical model that was 2 mm larger in diameter than the defect size to quantify the mineralized tissue at the peripheral zone. 2.6.3. Histological analysis Histological analysis was carried out to evaluate healing within the defects. After micro-CT analysis, samples were decalcified

using 10% ethylenediaminetetraacetic acid (EDTA) buffer solution for 4 weeks, dehydrated, and then embedded in paraffin wax. Representative sections of 3–5 lm thickness were prepared and stained with hematoxylin and eosin (H&E) and Masson’s trichrome. The stained sections were observed by light microscopy (Nikon, Japan). The images were captured using a DS-U3 imaging system (Nikon). 2.7. Statistical analysis All experiments were performed at least in triplicate. Data are presented as the mean ± SD. Statistical calculations were performed by Student’s t-test using SPSS ver.22 software (SPSS Inc., USA). Statistical significance was set to *p < 0.05 and **p < 0.01. 3. Results 3.1. Synthesis of Dex-U and PADH The design strategy of the hydrogel is illustrated in Fig. 1. The chemical structure of Dex-U was confirmed by 1H NMR (Fig. S1) and FTIR spectra (Fig. S2). PADH was fabricated by a facile onestep copolymerization method using hydrophobic-associated SMA/SDS micelles as the crosslinking bridge, resulting in a resilient and robust PAAm/Dex-U network. To investigate the gelation process of PAH and PADH, we stopped the gelation reaction after copolymerization for 15 min and 60 min, and used SEM to observe the cross-sectional morphology of the hydrogels. As shown in Fig. S3, after 15 min of gelation, large agglomerates were observed in the SEM images for PAH samples, indicating that the acrylamide monomers were primarily copolymerized (Fig. S3A). As the copolymerization proceeded to 60 min, a fibrous network structure had formed (Fig. S3B). On the contrary, in the SEM image of PADH after

Fig. 1. Schematic illustration of the design strategy of HAp mineralized PAAm/Dex-U hydrogel (HAp-PADH).

J. Fang et al. / Acta Biomaterialia 88 (2019) 503–513

15 min of copolymerization (Fig. S3B), some fibril-like structures between the large stacking agglomerates were observed, which we presume to be the Dex-U phase. After 60 min, this fibril-like phase had interpenetrated the entire hydrogel network. It is reasonable to conclude that the interpenetrating crosslinked Dex-U would greatly restrict the flexibility of the PAAm chains. This can be confirmed by the large difference in the swelling ratios between PAH and PADH. As shown in Fig. S4, unlike the PAH, with a swelling ratio of 400% after 24 h of immersion, the PADH showed a much lower swelling ratio of 170%. 3.2. Characterization of HAp-PADH We analyzed the effect of introducing Dex-U and HAp by measuring the surface profiles of the hydrogels. PAH had a continuous, smooth surface (Fig. 2A1), whereas the PADH had a much rougher and more porous surface (Fig. 2B1). After performing the in situ mineralization assay, the HAp-PADH had a relatively smooth surface when compared with the PADH; this may be because the hydrogel network was filled with deposited HAp nanoparticles (Fig. 2C1). The network morphology of the swollen hydrogels was further investigated by SEM. The SEM images of the freeze-dried PAH revealed a loose, inhomogeneous network structure with large pores (Fig. 2A2). On the contrary, the introduction of Dex-U led to a dense network structure with a smaller pore size (Fig. 2B2). The inhomogeneous network of PAH and PADH may be due to the micellar crosslinking mechanism (Fig. S5). The hard phase of the crosslinkable microspheres differed to the soft phase of the surrounding matrix. The surrounding matrix therefore had a relatively larger swelling ratio than the crosslinkable microspheres. This mismatch between the two phases likely resulted in the heterogeneous network structure of PAH and PADH. In addition, the interpenetrated Dex-U restricted the loose, flexible PAAm chains, resulting in a much denser network with a smaller swelling ratio. As a result, the PADH had smaller pores and a rougher surface when compared with PAH. The presence of the rigid Dex-U chains within the soft PAAm network would also help to enhance the mechanical properties of PADH. After in situ mineralization, as

507

expected, HAp nanoparticles were coated on the hydrogel surface and covered part of the pores, resulting in a relatively smooth surface for HAp-PADH as compared with PADH (Fig. 2C2). The SEM image taken at higher magnification (Fig. 2C3) revealed a uniform distribution of the inorganic nanoparticles on the network skeletons of the hydrogel. The micro/nano pores were still visible, suggesting that deposition of the inorganic nanoparticles did not fill the pores but retained the 3D porous hydrogel network structure. We further analyzed the inorganic phase deposited on the skeletons of HAp-PADH after in situ mineralization. A selected, representative low-magnification SEM image of HAp-PADH, with corresponding EDX elemental maps and spectrum, is presented in Fig. 3(A–E). EDX analysis confirmed that the calcium phosphate nanoparticles were uniformly deposited on the hydrogel surface. To examine the phase and chemical compositions of these nanoparticles, XRD and FTIR analyses were performed. The broad peak at 20° indicated that PADH was non-crystallized. However, after mineralization, characteristic reflections of HAp (JCPDS No. 74-0566) were clearly observed (Fig. 3F), despite the low crystallinity of the deposited HAp. The 2h angles at 25.7°, 31.8°, 39.8°, 46.9°, 49.5°, and 53.3° were indexed to be (0 0 2), (2 1 1), (1 3 0), (2 2 2), (2 1 3) and (0 0 4) reflections of HAp, respectively [29]. In the FTIR spectra, in addition to the bands of PADH, characteristic bands of the phosphate (PO3 4 ) group in HAp were present: 1038, 964, 602 and 564 cm1 (Fig. 3G). Both XRD and FTIR confirmed that the deposited calcium phosphate was HAp. TGA curves (Fig. 3H) showed that the residual weight of HAp-PADH at 700 °C was 41%, which was 10% higher than that of PADH. This further confirmed the presence of inorganic HAp phase. 3.3. Dex-U and HAp significantly enhance the mechanical properties of PAH Fig. 4A shows that the HAp-PADH was strong and tough, able to withstand a compression of 5 N without a large deformation. In comparison, the PAH failed to retain its original shape. The compressive stress-strain curves of the hydrogels are presented in Fig. 4B. The compressive strength of the HAp-PADH reached

Fig. 2. 3D surface profiles and SEM images of the hydrogels.

508

J. Fang et al. / Acta Biomaterialia 88 (2019) 503–513

Fig. 3. Chemical and phase compositions of HAp-PADH. (A) A representative low magnification SEM image; (B–D) EDX elemental maps; (E) a representative EDX spectrum; (F) XRD patterns of PADH and HAp-PADH; (G) FTIR spectra of PADH and HAp-PADH; and (H) TGA curves of PADH and HAp-PADH.

6.5 MPa, over twice that of the PADH (2.65 MPa) and about 13times that of the PAH (0.53 MPa). The fracture energy of the HAp-PADH (2385 J m2) was also 35% higher than that of the PADH (1650 J m2) and nearly four times that of the PAH (400 J m2) (Fig. 4C). The resilient and soft hydrophobic-associated PAAm network of the PAH was reinforced by the stiff crosslinked interpenetrating Dex-U chains within the PAAm. After mineralization, the HAp nanoparticles deposited within the hydrogel further strengthened the PADH network, endowing the HAp-PADH with better mechanical stability and fatigue resistance. The loading–unloading compressive curves of the HAp-PADH showed a larger hysteresis loop as compared with those loops of the PADH and the PAH, suggesting that the HAp-PADH dissipated energy much more efficiently; thus improving its mechanical stability (Fig. 4D). Fig. 4E shows the energy dissipation in five loading–unloading compressive cycles. As expected, the HAp-DAPH showed larger energy dissipation than the PADH or the PAH, suggesting better deformation resistance. Note that the cyclic compression tests were performed continuously. The largest energy dissipation of the hydrogels was detected at the first cycle, and then declined gradually because the hydrogels were unable to recover immediately without a break. Despite the limited recovery efficiency after the first compression cycle, the HAp-PADH still showed a much larger plateau value of energy dissipation (103 KJ m3) than the PADH (18 KJ m3) and the PAH (6.5 KJ m3), which strongly shows that the HAp deposited within the hydrogel network significantly enhanced the fatigue resistance of the hydrogel. The results of the mechanical tests demonstrate that the introduction of the stiff Dex-U phase and HAp effectively improved the mechanical properties of PAH, which is crucial for its use as a bone scaffold.

3.4. HAp-PADH promotes osteoblast adhesion, proliferation and osteogenic differentiation A comparison of osteoblast adhesion on PADH and HAp-PADH is presented in Fig. 5. Using phalloidin staining, we show that the cells on PADH were rounded and formed aggregations within the pores. In contrast, the cells were well spread out on the HApPADH and uniformly distributed within the hydrogel network (Fig. 5B), indicating that the deposited HAp facilitated osteoblast adhesion to the hydrogel network and prevented cell aggregations. Although PAAm is a biocompatible polymer, the soft, flat hydrogel surface of PAH along with the reduced capacity for protein absorption greatly limits cell affinity [30], causing cells to aggregate. The proliferation of osteoblasts evaluated by MTT assay (Fig. 6) showed that there was a steady increase in cell proliferation from day 1 to 5 on all three hydrogels, with HAp-PADH showing significantly higher proliferation than the other two scaffolds. This could probably be attributed to the ability of HAp nanoparticles to absorb serum proteins [30,31] and the micro-topological surface [32] of the HAp-PADH. In vitro osteogenic differentiation of osteoblasts on hydrogels was first analyzed by ALP assay. The ALP activity of the cells on the hydrogels after 7 days is presented in Fig. 7A. Compared with the PAH and PADH, which showed similar levels, there was nearly a two-fold increase in ALP activity for cells on the HAp-PADH. To further reveal the role of HAp on osteogenic differentiation at the molecular level, we examined the expression levels of four osteogenic marker genes (ALP, OCN, OPN and COL) in cells grown for 7 days on the PADH and the HAp-PADH using qRT-PCR (Fig. 7B). The cells on the HAp-PADH expressed higher levels of all four

J. Fang et al. / Acta Biomaterialia 88 (2019) 503–513

509

Fig. 4. Mechanical properties of the fabricated hydrogels. (A) Images of PAH and HAp-PADH compressed by a weight of 5 N; (B) compressive stress-strain curves; (C) fracture energy; (D) the first cyclic loading-unloading stress-strain curves and (E) energy dissipation during five continuous loading-unloading compressive cycles.

Fig. 5. Fluorescence microscopy images of osteoblasts stained for nuclei (blue) and F-actin (red) after 24 h of incubation on PADH (A) and HAp-PADH (B) scaffolds. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

genes as compared with those on the PADH. In particular, the levels of ALP and OCN were over 2-fold higher for cells on the HAp-PADH than on the PADH. These in vitro evaluations suggest that the HAp-PADH could remarkably promote osteoblast adhesion, proliferation, and osteogenic differentiation. 3.5. HAp-PADH accelerate new bone formation and improve osseointegration Our in vitro evaluations showed that HAp-PADH remarkably increased osteoblast proliferation and osteogenic differentiation. Therefore, we next sought to evaluate its effect on osteochondral

tissue regeneration using a femoral condyle defect rabbit model. PADH was used as a control to explore the role of the deposited HAp on subchondral bone regeneration and osseointegration. Micro-CT analysis showed mineralized tissue at the interface between the hydrogel and the defect in the HAp-PADHimplanted rabbits at 30 and 90 days after implantation. After 90 days, a dense ossified layer was formed in the HAp-PADHimplanted defect, whereas less-mineralized tissue was found in the PADH-implanted defect (Fig. 8A). The BMC and BV values of the HAp-PADH-implanted defect were remarkably higher than those for the PADH-implanted defect at day 30. After 90 days, despite showing reduced differences in the BMC and BV values,

510

J. Fang et al. / Acta Biomaterialia 88 (2019) 503–513

Fig. 6. Proliferation of osteoblasts after growth on the three hydrogels for 1, 3 or 5 days.

the values for the HAp-PADH-implanted group were still higher than those for the PADH-implanted group (Fig. 8B). This is likely because mineralization occurs mainly at the periphery of the implanted hydrogel. At the early stage, HAp-PADH effectively induced tissue ossification at the interface. However, at the later stage, the tissues at the periphery were almost mineralized, resulting in a reduced difference in BMC and BV values. Histological analysis (Fig. 9) further demonstrated that the HAp-PADH promoted bone regeneration and ossification at the interface between the hydrogel and the defect. The H&E-stained sections showed that new bone was present in the HAp-PADHimplanted model after 30 days. After 90 days, a dense new layer of bone with osteocytes had formed, indicating excellent osseointegration between the hydrogel and the host tissue at the interface. However, in the PADH-implanted model, there was a predominance of connective fibrous tissue between the hydrogel and the host bone at both time points. Limited new bone was formed after 90 days’ implantation. Masson’s trichrome staining revealed that the mineralized bone was directly bonded to HAp-PADH after 90 days. However, in the PADH-implanted model, almost no osteoid tissue was observed, in line with the micro-CT analysis. The histological results showed that HAp-PADH could effectively induce new bone formation and tissue mineralization. 4. Discussion Hydrogels that display high strength and toughness have attracted interest in recent years owing to the great demand for

bone tissue engineering scaffolds. PAAm-based tough hydrogels have been developed through various strategies, including the creation of interpenetrating or semi-interpenetrating polymer networks [33], as well as through nano-reinforcement [34], ionic crosslinking [35], dipole-dipole interactions [36] and hydrophobic interactions [13]. However, the lack of biofunctional cues of the PAAm-based hydrogels to promote cell adhesion and differentiation, along with their limited osseointegration, has hindered their use in applications in vivo. In this study, we first developed a tough PAAm/dextran hydrogel by hydrophobic micellar copolymerization. Dextran is a bioactive, natural polysaccharide produced by microorganisms. Functional dextran-based hydrogels have been used as tissue adhesives and in the encapsulation of cells or proteins owing to their good biocompatibility and high affinity to cells [37]. The high molecular weight of dextran and the 1,6-linked-Dglucophranosyl backbone structure also help to improve the network stiffness when incorporated through copolymerization with PAAm. Thus, the resilient, flexible and soft hydrophobicassociated PAAm network was enhanced by the interpenetrating rigid dextran phase, resulting in a tough, strong network with a compact, hierarchical porous structure. This PAAm/dextran hydrogel was further enhanced by the in situ mineralization of HAp. The resulted composite hydrogel showed exceptional mechanical properties (toughness over 2000 J m2, strength over 6 MPa) comparable with the native load-bearing tissue (toughness of 1000– 3000 J m2, strength of 3–5 MPa) [35]. Moreover, the strong interfacial interaction of HAp led to high efficiency energy dissipation, thereby allowing the material to provide load-bearing function during bone regeneration [38–42]. Osseointegration is an essential component for bone repair implants. The implanted scaffold should integrate with the surrounding bone tissue to restore the functionality of the reconstructed segment [43]. Good osseointegration of implants involves a series of complex biological processes, including preosteoblast adhesion, differentiation, maturation, and mineralization [44]. Numerous factors contribute to osteogenesis-related cell adhesion and differentiation, including biomolecular signals and substrate mechanical cues. Apart from growth factors and ECM proteins, inorganic nanoparticles are also essential for osteogenic expression [45,46]. From a biomechanical perspective, many studies have shown that a rough, porous substrate can effectively stimulate osteoblast maturation and mineralization, as compared with a smooth, soft substrate [47–50]. In this study, HAp nanoparticles were incorporated into the PADH network, resulting in a synergistic effect on the material’s mechanical properties, biocompatibility and osteoconductivity. Our in situ mineralization approach provided a uniform distribution of HAp nanocrystals in the network, and prevented the agglomeration of inorganic particles. The present HAp-PADH construct showed a microporous topographical

Fig. 7. ALP activity (a) and the expression of four selected osteospecific genes (COL, OCN, OPN and ALP) (b) at 7 days after incubation. The expression levels of all genes are normalized to that of GADPH as the housekeeping gene.

J. Fang et al. / Acta Biomaterialia 88 (2019) 503–513

511

Fig. 8. Micro-CT analysis of the PADH- and HAp-PADH-implanted defects. (A) Representative images of defect regeneration after 0, 30, and 90 days. Note that a is for axial section (scale bar = 2 mm), s is for sagittal section (scale bar = 5 mm) and the defect regions are marked with asterisks (*). (B) Bone mineral contents and (C) bone volumes as measured by Micro-CT.

network with enhanced stiffness that mimics the microenvironment of native bone, resulting in osteoblast spreading and elevated rates of proliferation. In vitro, we found higher levels of ALP activity and the expression of osteogenesis-related genes in cells grown on HAp-PADH. We built on these in vitro results using an in vivo tissue regeneration and osseointegration model, implanting our hydrogel scaffolds into femoral condylar defects in rabbits. Femoral condylar defects caused by osteochondral disease show poor levels of selfrepair due to the low vascular environment of cartilage and the complicated hierarchical structure of subchondral bone [16]. The implanted scaffolds for repairing large subchondral defects must stay in place for weeks to months to support cartilage and bone regeneration. To prevent creep and collapse of the host tissue into the defect site, the scaffold should directly bond with the nascent bone tissue. Thus the scaffold must be osteoconductive and integrate well with the host subchondral bone to promote cartilage regeneration. The histological results revealed mineralized osteoid in the HAp-PADH-implanted models at day 30 post-surgery, which dominated at the interface between the host tissue and the

implanted hydrogel after 90 days. We show that few cells migrated into the hydrogel scaffold due to the low degradability of PAAm. A dense new bone layer at the interface of the native bone and the HAp-PADH successfully affixed the scaffold within the defect site, and this provided load-bearing to the native tissue. The larger BMC and BV values measured by micro-CT further demonstrated that the mineralized layer associated directly with the host bone. Hence, through in situ mineralization, we constructed a bioactive HAp/polymer hybrid layer within the hydrogel network that endowed non-bioactive PADH with its osteoconductive and osseointegrative potential, and thus resulted in direct bonding of the scaffold to the host bone and the stimulation of new bone formation. We believe that such a tough PADH hydrogel with in situ mineralized HAp offers a feasible strategy to obtain PAAm-based hydrogel scaffolds with enhanced mechanical properties, affinity for cells, and a capacity for osteoconductivity. Future work may be directed at introducing bioactive molecules, such as growth factors or stem cells into the scaffolds to fabricate multifunctional hydrogels.

512

J. Fang et al. / Acta Biomaterialia 88 (2019) 503–513

Fig. 9. Histological photomicrographs of the bone-hydrogel interfaces of PADH and HAp-PADH with the proximal tissues. Abbreviations: CF, connective fibrous tissue; NB: new bone; HB, host bone; OC, osteocyte, * refers to the hydrogel or the lacuna due to the detachment or shrinkage of the hydrogel after decalcification. Scale bar = 200 lm.

5. Conclusions

References

In summary, we propose a feasible strategy to synthesize a strong, tough, osteoconductive hydrogel through one-step micellar copolymerization of acrylamide and Dex-U, followed by the in situ mineralization of HAp nanocrystals on the hydrogel network. The obtained HAp-PADH exhibited exceptional compressive strength, toughness, and fatigue resistance, owing to the introduction of the rigid Dex-U phase and the organic–inorganic binding of HAp to the polymeric network. The incorporation of HAp in the hydrogels significantly promoted osteoblast adhesion, proliferation, and osteogenic differentiation in vitro. Moreover, through the in vivo implantation of a HAp-PADH, we show that mineralized HAp can effectively improve osseointegration and induce new bone formation. The present strong, tough, and osteoconductive hydrogel scaffold offers a promising candidate for bone repair and regeneration.

[1] J.L. Drury, D.J. Mooney, Hydrogels for tissue engineering: scaffold design variables and applications, Biomaterials 24 (24) (2003) 4337–4351. [2] M. Mehrali, A. Thakur, C.P. Pennisi, S. Talebian, A. Arpanaei, M. Nikkhah, A. Dolatshahi-Pirouz, Nanoreinforced hydrogels for tissue engineering: biomaterials that are compatible with load-bearing and electroactive tissues, Adv. Mater. 29 (8) (2017) 201603612. [3] M. Farokhi, F. Mottaghitalab, S. Samani, M.A. Shokrgozar, S.C. Kundu, R.L. Reis, Y. Fatahi, D.L. Kaplan, Silk fibroin/hydroxyapatite composites for bone tissue engineering, Biotechno. Adv. 36 (2018) 68–91. [4] G. Tozzi, A. De Mori, A. Oliveira, M. Roldo, Composite hydrogels for bone regeneration, Materials 9 (4) (2016) 267. [5] D.M.R. Gibbs, C.R.M. Black, J.I. Dawson, R.O.C. Oreffo, A review of hydrogel use in fracture healing and bone regeneration, J. Tissue Eng. Regen. Med. 10 (3) (2016) 187–198. [6] I.C. Liao, F.T. Moutos, B.T. Estes, X. Zhao, F. Guilak, Composite threedimensional woven scaffolds with interpenetrating network hydrogels to create functional synthetic articular cartilage, Adv. Funct. Mater. 23 (47) (2013) 5833–5839. [7] G. Yang, H. Lin, B.B. Rothrauff, S. Yu, R.S. Tuan, Multilayered polycaprolactone/ gelatin fiber-hydrogel composite for tendon tissue engineering, Acta Biomater. 35 (2016) 68–76. [8] C.E. Kandow, P.C. Georges, P.A. Janmey, K.A. Beningo, Polyacrylamide hydrogels for cell mechanics: steps toward optimization and alternative uses, Methods Cell Biol. 83 (2007) 29–46. [9] P. Lin, S. Ma, X. Wang, F. Zhou, Molecularly engineered dual-crosslinked hydrogel with ultrahigh mechanical strength, toughness, and good selfrecovery, Adv. Mater. 27 (12) (2015) 2054–2059. [10] M.A. Haque, T. Kurokawa, G. Kamita, J.P. Gong, Lamellar bilayers as reversible sacrificial bonds to toughen hydrogel: hysteresis, self-recovery, fatigue resistance, and crack blunting, Macromolecules 44 (22) (2011) 8916–8924. [11] D.C. Tuncaboylu, A. Argun, M. Sahin, M. Sari, O. Okay, Structure optimization of self-healing hydrogels formed via hydrophobic interactions, Polymer 53 (24) (2012) 5513–5522. [12] D.C. Tuncaboylu, M. Sari, W. Oppermann, O. Okay, Tough and self-healing hydrogels formed via hydrophobic interactions, Macromolecules 44 (12) (2011) 4997–5005.

Acknowledgements This work was financially supported by the National Key Research and Development Program of China (2016YFB0700803) and the Fundamental Research Program of Shenzhen (Grant No. JCYJ20170307110418960), China.

Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.actbio.2019.02.019.

J. Fang et al. / Acta Biomaterialia 88 (2019) 503–513 [13] Q. Chen, L. Zhu, H. Chen, H. Yan, L. Huang, J. Yang, J. Zheng, A novel design strategy for fully physically linked double network hydrogels with tough, fatigue resistant, and self-Healing properties, Adv. Funct. Mater. 25 (10) (2015) 1598–1607. [14] B.-S. Kim, I.-K. Park, T. Hoshiba, H.-L. Jiang, Y.-J. Choi, T. Akaike, C.-S. Cho, Design of artificial extracellular matrices for tissue engineering, Prog. Polym. Sci. 36 (2) (2011) 238–268. [15] I. Khan, S. Gilbert, S. Singhrao, V. Duance, Cartilage Integration: evaluation of the reasons for failure of integration during cartilage repair. A review, Eur. Cells Mater. 16 (2008) 26–39. [16] B. Xu, P. Zheng, F. Gao, W. Wang, H. Zhang, X. Zhang, X. Feng, W. Liu, A mineralized high strength and tough hydrogel for skull bone regeneration, Adv. Funct. Mater. 27 (4) (2017) 4327. [17] S. Wada, N. Kitamura, T. Nonoyama, R. Kiyama, T. Kurokawa, J.P. Gong, K. Yasuda, Hydroxyapatite-coated double network hydrogel directly bondable to the bone: biological and biomechanical evaluations of the bonding property in an osteochondral defect, Acta Biomater. 44 (2016) 125–134. [18] M. Sabzi, N. Samadi, F. Abbasi, G.R. Mahdavinia, M. Babaahmadi, Bioinspired fully physically cross-linked double network hydrogels with a robust, tough and self-healing structure, Mater. Sci. Eng. C 74 (2017) 374–381. [19] L. Han, Y. Zhang, X. Lu, K. Wang, Z. Wang, H. Zhang, Polydopamine nanoparticles modulating stimuli-responsive PNIPAM hydrogels with cell/ tissue adhesiveness, ACS Appl. Mater. Inter. 8 (42) (2016) 29088–29100. [20] A.K. Gaharwar, C.P. Rivera, C.-J. Wu, G. Schmidt, Transparent, elastomeric and tough hydrogels from poly(ethylene glycol) and silicate nanoparticles, Acta Biomater. 7 (12) (2011) 4139–4148. [21] K. Haraguchi, T. Takehisa, Nanocomposite hydrogels: a unique organicinorganic network structure with extraordinary mechanical, optical, and swelling/de-swelling properties, Adv. Mater. 14 (16) (2002) 1120–1124. [22] K. Haraguchi, T. Takehisa, S. Fan, Effect of clay content on the properties of nanocomposite hydrogels composed of poly(N-isopropylacrylamide) and clay, Macromolecules 35 (27) (2002) 10162–10171. [23] A.A. Thorpe, S. Creasey, C. Sammon, C.L. Le Maitre, Hydroxylapatite nanoparticle injectable hydrogel scaffold to support osteogenic differentiation of human mesenchymal stem cells, Eur. Cells Mater. 32 (2016) 1–23. [24] A.A. Thorpe, C. Freeman, P. Farthing, J. Callaghan, P.V. Hatton, I.M. Brook, C. Sammon, C.L. Le Maitre, In vivo safety and efficacy testing of a thermally triggered injectable hydrogel scaffold for bone regeneration and augmentation in a rat model, Oncotarget 9 (26) (2018) 18277–18295. [25] J. Zhuang, S. Lin, L. Dong, K. Cheng, W. Weng, Magnetically actuated mechanical stimuli on Fe3O4/mineralized collagen coatings to enhance osteogenic differentiation of the MC3T3-E1 cells, Acta Biomater. 71 (2018) 49–60. [26] L. Hao, L. Wei, L. Can, T. Jie, W. Hong, H. Bao, C. Hong, L. Hui-Jie, L. Zhong-Jun, S. Chun-Li, Incorporating simvastatin/poloxamer 407 hydrogel into 3D-printed porous Ti6Al4V scaffolds for the promotion of angiogenesis, osseointegration and bone ingrowth, Biofabrication 8 (4) (2016) 045012. [27] J.H. Lee, J. Kim, H.-R. Baek, K.M. Lee, J.-H. Seo, H.-K. Lee, A.-Y. Lee, G.B. Zheng, B.-S. Chang, C.-K. Lee, Fabrication of an rhBMP-2 loaded porous b-TCP microsphere-hyaluronic acid-based powder gel composite and evaluation of implant osseointegration, J. Mater. Sci.-Mater. Med. 25 (9) (2014) 2141–2151. [28] A.R. Armiento, M.J. Stoddart, M. Alini, D. Eglin, Biomaterials for articular cartilage tissue engineering: learning from biology, Acta Biomater. 65 (2018) 1–20. [29] M.S.M. Arsad, P.M. Lee, L. Kong Hung, Synthesis and characterization of hydroxyapatite nanoparticles and b-TCP particles, Colloids Surf. APhysicochem. Eng. Asp. 322 (2008) 29–33. [30] K.L. Kilpadi, P.-L. Chang, S.L. Bellis, Hydroxylapatite binds more serum proteins, purified integrins, and osteoblast precursor cells than titanium or steel, J. Biomed. Mater. Res. 57 (2) (2001) 258–267.

513

[31] S.K. Swain, D. Sarkar, Study of BSA protein adsorption/release on hydroxyapatite nanoparticles, Appl. Surf. Sci. 286 (2013) 99–103. [32] A.B. Faia-Torres, S. Guimond-Lischer, M. Rottmar, M. Charnley, T. Goren, K. Maniura-Weber, N.D. Spencer, R.L. Reis, M. Textor, N.M. Neves, Differential regulation of osteogenic differentiation of stem cells on surface roughness gradients, Biomaterials 35 (33) (2014) 9023–9032. [33] W. Sun, B. Xue, Y. Li, M. Qin, J. Wu, K. Lu, J. Wu, Y. Cao, Q. Jiang, W. Wang, Polymer-supramolecular polymer double-network hydrogel, Adv. Funct. Mater. 26 (48) (2016) 9044–9052. [34] L. Han, X. Lu, K. Liu, K. Wang, L. Fang, L.-T. Weng, H. Zhang, Y. Tang, F. Ren, C. Zhao, G. Sun, R. Liang, Z. Li, Mussel-inspired adhesive and tough hydrogel based on nanoclay confined dopamine polymerization, ACS Nano 11 (3) (2017) 2561–2574. [35] J. Li, W.R.K. Illeperuma, Z. Suo, J.J. Vlassak, Hybrid hydrogels with extremely high stiffness and toughness, ACS Macro. Lett. 3 (6) (2014) 520–523. [36] W. Wang, Y. Zhang, W. Liu, Bioinspired fabrication of high strength hydrogels from non-covalent interactions, Prog. Polym. Sci. 71 (2017) 1–25. [37] D. Diekjürgen, D.W. Grainger, Polysaccharide matrices used in 3D in vitro cell culture systems, Biomaterials 141 (2017) 96–115. [38] S. Dhivya, S. Saravanan, T.P. Sastry, N. Selvamurugan, Nanohydroxyapatitereinforced chitosan composite hydrogel for bone tissue repair in vitro and in vivo, J. Nanobiotech. 13 (1) (2015) 40. [39] J.P. Gleeson, N.A. Plunkett, F.J. O’Brien, Addition of hydroxyapatite improves stiffness, interconnectivity and osteogenic potential of a highly porous collagen-based scaffold for bone tissue regeneration, Eur. Cells Mater. 20 (2010) 218–230. [40] R.K. Roeder, G.L. Converse, R.J. Kane, W. Yue, Hydroxyapatite-reinforced polymer biocomposites for synthetic bone substitutes, JOM 60 (3) (2008) 38– 45. [41] R.J. Kane, H.E. Weiss-Bilka, M.J. Meagher, Y. Liu, J.A. Gargac, G.L. Niebur, D.R. Wagner, R.K. Roeder, Hydroxyapatite reinforced collagen scaffolds with improved architecture and mechanical properties, Acta Biomater. 17 (2015) 16–25. [42] M.C. Darnell, J.-Y. Sun, M. Mehta, C. Johnson, P.R. Arany, Z. Suo, D.J. Mooney, Performance and biocompatibility of extremely tough alginate/ polyacrylamide hydrogels, Biomaterials 34 (33) (2013) 8042–8048. [43] K. Yasuda, J. Ping Gong, Y. Katsuyama, A. Nakayama, Y. Tanabe, E. Kondo, M. Ueno, Y. Osada, Biomechanical properties of high-toughness double network hydrogels, Biomaterials 26 (21) (2005) 4468–4475. [44] F. Loi, L.A. Córdova, J. Pajarinen, T.-H. Lin, Z. Yao, S.B. Goodman, Inflammation, fracture and bone repair, Bone 86 (2016) 119–130. [45] M. Christine, L. Jasmin, O. Michael, C. Jennifer, S. Andre, F.S.-B. Alexis, 3D bioprinting of GelMA scaffolds triggers mineral deposition by primary human osteoblasts, Biofabrication 9 (1) (2017) 015009. [46] F. Sun, H. Zhou, J. Lee, Various preparation methods of highly porous hydroxyapatite/polymer nanoscale biocomposites for bone regeneration, Acta Biomater. 7 (11) (2011) 3813–3828. [47] C. Weinand, I. Pomerantseva, C.M. Neville, R. Gupta, E. Weinberg, I. Madisch, F. Shapiro, H. Abukawa, M.J. Troulis, J.P. Vacanti, Hydrogel-b-TCP scaffolds and stem cells for tissue engineering bone, Bone 38 (4) (2006) 555–563. [48] K. Rezwan, Q.Z. Chen, J.J. Blaker, A.R. Boccaccini, Biodegradable and bioactive porous polymer/inorganic composite scaffolds for bone tissue engineering, Biomaterials 27 (18) (2006) 3413–3431. [49] G.M. Cunniffe, C.M. Curtin, E.M. Thompson, G.R. Dickson, F.J. O’Brien, Contentdependent osteogenic response of nanohydroxyapatite: an in vitro and in vivo assessment within collagen-based scaffold, ACS Appl. Mater. Interfaces 8 (2016) 23477–23488. [50] S. Kumar, D. Azam, S. Raj, E. Kolanthai, K.S. Vasu, A.K. Sood, K. Chatterjee, 3D scaffold alters cellular response to graphene in a polymer composite for orthopedic application, J. Biomed. Mater. Res. B 104 (4) (2016) 732–749.