www.advhealthmat.de www.MaterialsViews.com
FULL PAPER
3D-Printed Atsttrin-Incorporated Alginate/Hydroxyapatite Scaffold Promotes Bone Defect Regeneration with TNF/ TNFR Signaling Involvement Quan Wang, Qingqing Xia, Yan Wu, Xiaolei Zhang, Feiqiu Wen, Xiaowen Chen, Shufang Zhang, Boon Chin Heng, Yong He, and Hong-Wei Ouyang* method.[1] Conventional surgical techniques for reconstruction have utilized High expression levels of pro-inflammatory tumor necrosis factor (TNF)-α autogenous, allogeneic, and prosthetic within bone defects can decelerate and impair bone regeneration. However, materials.[2] Autologous bone is currently there are few available bone scaffolds with anti-inflammatory function. The the best treatment option, but it is in limprogranulin (PGRN)-derived engineered protein, Atsttrin, is known to exert ited supply and often cannot completely antagonistic effects on the TNF-α function. Hence, this study investigates fit the size and shape of the defect sites.[3] whether 3D-printed Atsttrin-incorporated alginate(Alg)/hydroxyapatite(nHAp) With allogeneic bone there is a risk of disease transmission. Prosthetic materials scaffolds can facilitate bone healing through affecting the TNF/TNFR signalmade from polymers or metals have been ing. A 3D bioprinting system is used to fabricate Atsttrin-Alg/nHAp comwidely used, but have certain disadvanposite scaffolds, and the Atsttrin release from this scaffold is characterized, tages, such as metal corrosion and poor followed by evaluation of its efficacy on bone regeneration both in vitro and in mechanical properties.[4] vivo. The 3D-printed Atsttrin-Alg/nHAp scaffold exhibits a precisely defined Tissue engineering techniques combining the use of cells, three-dimensional structure, can sustain Atsttrin release for at least 5 days, has negligible cytobiomaterial scaffolds, and growth factors toxicity, and supports cell adhesion. Atsttrin can also attenuate the suppresare recognized as the most promising sive effects of TNF-α on BMP-2-induced osteoblastic differentiation in vitro. approach for bone repair and reconstrucThe 3D-printed Atsttrin-Alg/nHAp scaffold significantly reduces the number tion, and have been demonstrated to be of TNF-α positive cells within wound sites, 7 days after post-calvarial defect a suitable alternative to autogenous or surgery. Additionally, histological staining and X-ray scanning results also allogeneic bone grafts.[5] Previous studies have investigated the use of a diverse show that the 3D-printed Atsttrin-Alg/nHAp scaffold enhances the regeneraarray of biomaterials to fabricate 3D tion of mice calvarial bone defects. These findings thus demonstrate that the printed scaffolds for calvarial reconstrucprecise structure and anti-inflammatory properties of 3D-printed Atsttrin-Alg/ tion,[6–9] such as poly(lactic acid) (PLA),[10] nHAp scaffolds may promote bone defect repair. poly(glycolic acid) (PGA),[11] the co-polymer polylactic-co-glycolic acid (PLGA),[12,13] poly(methyl methacrylate) (PMMA),[14,15] poly(caprolactone) (PCL),[16] collagen, and hydrogels.[17–19] How1. Introduction ever, only a few of these can be incorporated with biological components through bioprinting. Alginate is one of the most The repair of complex calvarial bone defects is challenging, widely used materials for bioprinting because of its biocomand successful healing depends on the defect size, the quality patibility and ability to form stable hydrogels in the presence of soft tissue covering the defect, and choice of reconstructive Dr. Q. Wang, Dr. Q. Xia, Dr. Y. Wu, Dr. X. Zhang, Dr. S. Zhang, Prof. H.-W. Ouyang Center for Stem Cell and Tissue Engineering School of Medicine Zhejiang University Hangzhou, P. R. China E-mail: [email protected] Dr. Q. Wang, Dr. Q. Xia, Dr. Y. Wu, Dr. X. Zhang, Dr. S. Zhang, Prof. H.-W. Ouyang Zhejiang Provincial Key Laboratory of Tissue Engineering and Regenerative Medicine Hangzhou, P. R. China
DOI: 10.1002/adhm.201500211
Adv. Healthcare Mater. 2015, 4, 1701–1708
Dr. F. Wen, Dr. X. Chen Division of hematology and oncology Shenzhen Children’s Hospital Shenzhen, P. R. China Dr. B. C. Heng Department of Biosystems Science & Engineering ETH-Zurich Mattenstrasse 26, Basel, Switzerland Dr. Y. He The State Key Lab of Fluid Power Transmission and Control School of Mechanical Engineering Zhejiang University Hangzhou, P. R. China
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
1701
www.advhealthmat.de
FULL PAPER
www.MaterialsViews.com
of calcium or other divalent cations, but its biological inertness and poor mechanical properties have largely limited its use.[20,21] Hydroxyapatite (nHAp) scaffolds have also been widely utilized for bone regeneration because of its biocompatibility and osteoconductivity.[22] However, its slow degradation rates impedes the complete regeneration of bone defects. One possible approach to combine the favorable properties of alginate and nHAp is to fabricate an alginate hydrogel with a dispersed phase consisting of nanosized nHAp particles. Such types of composite scaffolds have excellent biocompatibility, structural and mechanical properties, and can be incorporated with bioactive factors that promote bone regeneration.[23,24] Apart from the structural and material properties of the scaffolds, their biological function is another important factor that influences the efficacy of bone regeneration. It has been reported that the high levels of tumor necrosis factor (TNF)-α within bone defects can decelerate and impair bone regeneration.[25,26] However, there are few bone scaffolds available with anti-inflammatory function. Many TNF-α blockers, including etanercept, infliximab, and adalimumab, are effective antiinflammatory drugs. In several previous studies, pre-treatment of stem cells with aspirin can markedly reduce TNF-α and interferon (IFN)-γ levels at the injury site, and improve bone marrow mesenchymal stem cells (BMMSCs)-based bone regeneration and cavarial defect repair in C57BL/6 mice.[27] Compared with aspirin, the growth factor progranulin (PGRN) and its derivative, Atsttrin, can bind directly to the tumor necrosis factor receptor (TNFR) and thus inhibit TNF-α/TNFR interactions more efficiently, thereby reducing inflammation in inflammatory arthritis models.[28–30] Another recent study showed that PGRN can facilitate the bone-healing process by interacting with both the bone morphogenic protein (BMP)-2 and TNF-α signaling pathways.[31] However, under inflammatory conditions, PGRN can be digested by proteases and metalloproteinase, into 6 KDa granulin units that are pro-inflammatory, thus neutralizing the anti-inflammatory effects of intact PGRN. By contrast, Atsttrin contains only partial granulin units and is not expected to release such pro-inflammatory fragments. This feature, coupled with its longer active halflife (ca. 120 hours) than PGRN (ca. 40 hours) and its selective binding to TNFR1/TNFR2, thus makes Atsttrin a promising bioactive factor that can be incorporated within anti-inflammatory scaffolds. In this study, we hypothesize that 3D-printed Atsttrin-incorporated alginate/hydroxyapatite scaffolds can promote bone healing through perturbation of the TNF/TNFR signaling. A 3D bioprinting system was used to fabricate Atsttrin-Alg/nHAp composite scaffolds, and the Atsttrin release profile from the scaffold was characterized, followed by evaluation of its efficacy in bone regeneration, both in vitro and in vivo.
2. Results 2.1. Atsttrin Attenuates the Suppressive Effect of TNF-α on BMP-2-Induced Osteoblastic Differentiation We investigated the effects of TNF-α and TNF-α/Atsttrin on BMP-2-induced osteoblastic differentiation and whether
1702
wileyonlinelibrary.com
additional Atsttrin can affect this process. Murine mesenchymal stem cells (C3H cells) were cultured in recombinant human bone morphogenetic protein-2 (rhBMP-2) (200 ng mL−1) and TNF-α (50 ng mL−1)-supplemented medium, with or without Atsttrin for several days, to study the effects of TNF-α and TNF-α/Atsttrin on BMP-2-induced osteoblastic differentiation. After 7 days, alkaline phosphatase (ALP) staining (Figure 1A) demonstrated that the presence of TNF-α inhibited BMP-2-induced ALP expression in C3H cells, and that the presence of Atsttrin attenuated this inhibitory effect in a dose-dependent manner. Consistent with the observed differences in expression of the early osteogenic marker ALP, the intensity of Alizarin red staining was also lower in the TNF-α treated C3H cells and higher in the TNF-α/Atsttrin treated C3H cells (Figure 1B). Furthermore, quantization of the ALP activity (Figure 1C) demonstrated that Atsttrin could attenuate the suppressive effect of TNF-α on BMP-2-induced osteoblastic differentiation. To further verify these findings, the transcription levels of osteogenic marker genes such as COL1A1, OCN, and OPN in C3H cells were quantified by real-time polymerase chain reaction (PCR) (Figure 1D–F). TNF-α treatment reduced the transcriptional levels of these marker genes under BMP-2 stimulation, whereas Atsttrin increased the expression of these genes. Put together, these data indicate that Atsttrin can attenuate the suppressive effect of TNF-α on BMP-2-induced osteoblastic differentiation. We also employed a non-contacting co-culture (trans-well cell culture) system to investigate the efficiency of Atsttrin that is released from the scaffold. C3H cells were seeded on the bottom of 24-well plates, and cultured in rhBMP-2 (200 ng mL−1) and TNF-α (50 ng mL−1)-supplemented medium. Atsttrin-incorporated scaffolds or scaffolds without Atsttrin were put onto the membrane of suitable trans-wells. The alkaline phosphatase (ALP) staining and a quantization of the ALP activity demonstrated that Atsttrin-incorporated scaffolds have a similar efficiency as separately added Atsttrin, indicating that Atsttrin released from the scaffolds could also attenuate the suppressive effect of TNF-α on BMP-2-induced osteoblastic differentiation (Figure S1, Supporting Information).
2.2. Fabrication and Characterization of Bioactive Atsttrin-Incorporated Composite Scaffolds Alg/nHAp composite scaffolds were fabricated with a pneumatic bioprinting system (Figure 2A). The size of strand extruded from the nozzle was tailored by adjusting the speed of deposition and the pressure of extrusion (Figure 2B). Examination of the macroscopic appearance and morphology revealed that the printed scaffolds have vertically connected pores, with a strand diameter of 100 µm and a clearance of 400 µm between strands (Figure 2C,D). 3D hydrogel scaffolds consisting of five or more layers with regular structure could also be printed (Figure 2E). Scanning electron microscopy (SEM) (Figure 2F) analysis revealed that the nano-hydroxyapatite particles are uniformly distributed on the surface of the scaffold.
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Healthcare Mater. 2015, 4, 1701–1708
www.advhealthmat.de www.MaterialsViews.com
FULL PAPER Figure 1. Atsttrin attenuates the suppressive effect of TNF-α on BMP-2-induced osteoblastic differentiation. C3H cells were treated with rhBMP2(200 ng mL−1) in the presence of TNF-α (50 ng mL−1) or TNF-α (50 ng mL−1)/Atsttrin (10 nM or 50 nM). A) ALP staining assay for detecting ALPase activity on day 7. B) Alizarin red staining assay for detecting calcium deposition on day 21. C) Quantification assay of ALPase activity on day 3, normalized with the untreated cells. D–F) Real-time PCR and relative quantification of COL1A1 (D), OCN (E), OPN (F), normalized with respect to GAPDH. Each bar represents the mean S.D. of at least three independent replicate experiments. *p < 0.05 and **p < 0.01.
Cell proliferation analysis demonstrated that the printed scaffolds exerted negligible a cytotoxic effect on the cells (Figure 3A). The sustained release of Atsttrin protein from the Atsttrin-Alg/nHAp composite scaffolds was observed for about 5 days (Figure 3B and Figure S2, Supporting Information).
Carboxyfluorescein diacetate succinimidyl ester (CFDA-SE) staining and visualization of the C3H cells revealed that the cells were retained within the scaffold, and subsequent SEM images showed that the cells spread well over the scaffold surface (Figure 3C,D).
Figure 2. Schematic representation of the 3D bioprinter and morphology of the 3D printed Atsttrin-bioactive scaffolds. A) Schematic diagram of the pneumatic dispensing system. B) The effects of deposition speed and extrusion pressure on strand size. C) The macroscopic appearance of the single layer printed scaffold. D) The morphology of the printed scaffold viewed under an optical microscope. Scale bar, 500 µm. E) The macroscopic appearance of the multilayer printed scaffold. F) SEM image of the surface morphology of the printed scaffold. Scale bar, 50 µm.
Adv. Healthcare Mater. 2015, 4, 1701–1708
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
1703
www.advhealthmat.de
FULL PAPER
www.MaterialsViews.com
be detected at 16 weeks post-implantation (Figure 4C,D). Quantification of the opacity index within the defect by calculating the integrated optical density showed that the Atsttrin-Alg/nHAp group produced more bone-like tissues than the Alg/nHAp group at 8 and 16 weeks post-implantation (Figure 4I).
2.3.2. Histological Analysis of Defect Regions For each specimen, transverse sections of the entire calvaria were initially examined at low magnification to identify regions of interest. Both H&E staining (Figure 5A) and Masson's trichrome staining (Figure 5D) revealed that the samples of the control group typically exhibited no bone formation within the defect regions at 8 weeks post-implantation. The Alg/nHAp and Atsttrin-Alg/nHAp groups Figure 3. Cytotoxicity, controlled release profile, and cell adhesion analysis of the printed scaf- exhibited significantly more regenerative folds. A) Cytotoxicity analysis of the scaffold. B) Cumulative release curve of Atsttrin from activity within the defect sites. In the defects the scaffold in vitro. C) The CFDA-SE labeled C3H cells adhered onto the scaffolds on day 1. of the Alg/nHAp group only some fibrous D) The SEM image of the morphology of C3H cells on the Atsttrin-Alg/nHAp scaffolds on day tissue and negligible bone tissue was observed 3. *p < 0.05 and **p < 0.01. (Figure 5B,E), compared to this far more new bone-like tissue formation was detected in the Atsttrin-Alg/nHAp group (Figure 5C,F). Fragmented remains of the scaffolds were visible in both groups at 2.3. Efficacy of Bioactive Atsttrin-Incorporated Composite 8 weeks post-implantation, with new bone having been formed Scaffolds for Bone Regeneration in a Mice Calvarial Defect around these fragments. At 16 weeks post-implantation, the conModel trol group still exhibited negligible bone formation (Figure 6A,D). By contrast, much bone formation was observed in the defect 2.3.1. Scaffold Implantation and Evaluation of Bone Regeneration We surgically created a calvarial bone defect model in C57BL/6J mice to investigate the efficacy of the Atsttrin-bioactive scaffold for bone regeneration. Oversized calvarial bone defects of 7 mm × 5 mm were established (Figure 4A). The defects were implanted with the Alg/nHAp composite scaffold, or with the Atsttrin-Alg/nHAp composite scaffold (Figure 4B). The blank group served as the untreated control without scaffold implantation. Animals were sacrificed at 8 and 16 weeks after implantation and whole calvaria samples from each group were collected. As revealed by X-ray images, bone-like tissue was formed within the defect areas of the Atsttrin-Alg/nHAp group at 8 and 16 weeks post-implantation. The new bone tissue was localized at the edges and within the center of the defect at 16 weeks postimplantation (Figure 4G,H). However, much less bone-like tissue was detected within the defects of the Alg/nHAp group at 16 weeks post-implantation (Figure 4E,F). Radiopaque tissue was almost undetectable in the blank groups at 8 weeks post-implantation, and minimal bone-like tissue formation could
1704
wileyonlinelibrary.com
Figure 4. The calvarial-defect surgery procedure and X-ray imaging of whole calvarias after 8 and 16 weeks of implantation of the scaffolds. A,B) The surgical operation of mice calvarial bone defect model. C–H) X-ray imaging of whole calvarias of the blank group (C,D), Alg/nHAp group (E,F) and Atsttrin-Alg/nHAp group (G,H). The yellow arrows indicate the new bone. I) The integrated optical densities of the Alg/nHAp group, Atsttrin-Alg/nHAp group, and the blank group. *p < 0.05 and **p < 0.01.
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Healthcare Mater. 2015, 4, 1701–1708
www.advhealthmat.de www.MaterialsViews.com
FULL PAPER Figure 5. Histological analysis of the repaired calvarias after 8 weeks of implantation in vivo. A–C) H&E staining of the repaired calvarias in the blank group (A), Alg/nHAp scaffold group (B), and Atsttrin-Alg/nHAp scaffold group (C). D–F) Masson's trichrome staining of the repaired calvarias in the blank group (D), Alg/nHAp scaffold group (E), and Atsttrin-Alg/nHAp scaffold group (F).
areas of the Alg/nHAp and Atsttrin-Alg/nHAp groups, with the Atsttrin-Alg/nHAp group exhibiting superior bone regeneration (Figure 6C,F) compared to the Alg/nHAp group (Figure 6B,E). This was confirmed by a semi-quantitative analysis showing the relative amount of bone formation in the calvarial defect area from different groups (Figure S3, Supporting Information).
Subsequently, we examined the effect of Atsttrin on TNF-α expression levels in the different groups at 7 days post-implantation. Immunohistochemical analysis showed that the TNF-α concentration expression is very high in the control group (Figure 7A–C). For each sample, six sections for which five fields each of immunohistochemical staining were selected to
Figure 6. Histological analysis of the repaired calvarias after 16 weeks of implantation in vivo. A–C) H&E staining of the repaired calvarias in the blank group (A), Alg/nHAp scaffold group (B), and Atsttrin-Alg/nHAp scaffold group (C). D–F) Masson's trichrome staining of the repaired calvarias in the blank group (D), Alg/nHAp scaffold group (E), and Atsttrin-Alg/nHAp scaffold group (F).
Adv. Healthcare Mater. 2015, 4, 1701–1708
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
1705
www.advhealthmat.de
FULL PAPER
www.MaterialsViews.com
Figure 7. Immunohistochemical analysis of three different groups. A–C) Immunohistochemical staining of TNF-α positive cells, 7 days after defect surgery in the three different groups. D) Semi-quantitative analysis of the relative amounts of TNF-α-positive cells in the different groups. E–G) Runx2 expression in the calvarial defect regions after 16 weeks of implantation in the three different groups. H) Semi-quantitative analysis of the relative numbers of Runx2-positive cells in the different groups. *p < 0.05 and **p < 0.01.
carry out the semi-quantitative analysis. The number of TNF-αpositive cells was found to be significantly reduced in the Atsttrin-Alg/nHAp group, as compared to the Alg/nHAp and control groups (Figure 7D), thus indicating that Atsttrin released from the scaffold exerted an inhibitory effect on TNF-α formation and accumulation. Immunohistological analysis of runt-related transcription factor 2 (Runx2) was also performed to compare osteogenesis in the three groups (Figure 7E-G). There was a higher positive expression of Runx2 in the Atsttrin-Alg/nHAp group, as compared to the Alg/nHAp and control groups (Figure 7H), thus suggesting that Atsttrin promotes osteogenesis.
3. Discussion In this study, we utilized a 3D bioprinting technology to fabricate an anti-inflammatory scaffold through incorporation of Atsttrin, and evaluate its efficacy in bone regeneration with a mouse calvarial defect model. It was demonstrated that such scaffolds could continuously release Atsttrin and attenuate the suppressive effects of TNF-α on BMP-2-induced osteoblastic differentiation in vitro. The in vivo study also demonstrated that the Atsttrin-incorporated scaffold group can decrease TNF-α accumulation and up-regulate Runx2 expression, achieving better bone formation. 3D printing technologies have been widely utilized in the field of tissue engineering. Natural materials such as alginate, collagen, silk fibroin, and various synthetic materials such as PCL, PLA, and PVA have been fabricated into scaffolds or cell/ hydrogel structures, using appropriate material-forming technologies, such as fused deposition modeling (FDM), stereo lithography apparatus (SLA), and selective laser sintering (SLS). These scaffolds fabricated by 3D printing still have some limitations. For example, the material properties of PCL or PLA
1706
wileyonlinelibrary.com
are unsuitable for incorporating biological components. The mechanical strength of the 3D printed scaffolds is often insufficient for tissue-engineering applications, and their degradation rate is often too fast for optimal bone regeneration. In our study, a pneumatic bioprinting system was used to fabricate Alg/nHAp composite scaffolds with vertically connected pores and a strand diameter of 100 µm. The nHAp was added to the alginate hydrogel within the 3D bioprinted multilayer structure to improve the mechanical properties according to previous research,[23,32] and gelatin was used to help maintain the shape of the printed constructs. Such scaffolds had negligible cytotoxicity, supported cell adhesion, and exhibited sustained release of Atsttrin in vitro. TNF-α/TNFR signaling has been extensively studied because of its integral position at the apex of the pro-inflammatory cytokine cascade. Previous studies showed that Atsttrin, an engineered protein composed of three fragments of the growth factor progranulin (PGRN), exhibited selective TNFR binding, making it a potential antagonist of TNF-α signaling.[28] The administration of recombinant PGRN protein can promote bone healing and enhance BMP-2 function in the presence of high TNF-α levels in a radial bone defect model.[31] Considering that a scaffold is essential for bone-defect regeneration, we therefore utilized the bioprinted Alg/nHAp composite scaffold for controlled sustained release of Atsttrin in the defect region, as an alternative to direct administration of Atsttrin. We showed that Atsttrin could attenuate the suppressive effects of TNF-α on BMP-2 induced osteogenesis in vitro. Upon implantation of this Atsttrin-incorporated 3D-bioprinted scaffold, radiographic analysis demonstrated better bone formation in the Atsttrin-incorporated composite scaffold group, as compared to the scaffold-only and blank groups. Histological analyses of samples at 8 and 16 weeks post-surgery conclusively showed enhanced bone formation in the Atsttrin-incorporated composite scaffold group, thus validating our hypothesis that
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Healthcare Mater. 2015, 4, 1701–1708
www.advhealthmat.de www.MaterialsViews.com
4. Conclusion Our findings are consistent with previous studies that showed that aspirin suppressed TNF-α and IFN-γ expression and reversed the pro-inflammatory cytokine-induced osteogenic deficiency of BMMSCs.[27] Site-specific aspirin treatment was shown to enhance BMMSC-based tissue regeneration through the suppression of TNF-α and IFN-γ in a calvarial bone-defect repair model. The results of this study thus suggest that sitespecific modulation of local inflammatory reactions, particularly the suppression of inflammatory cytokine TNF-α, can promote bone healing, although the repair effect is not that satisfactory. Stem cells seems to be essential for calvarial bone healing, so it is logical to expect that the combination of stem cells and Atsttrin will achieve better calvarial bone healing in future. In summary, our findings demonstrate that a composite of alginate and nHAp forms an appropriate bio-ink for 3D printed scaffolds, and that Atsttrin is a promising bioactive factor that can be incorporated within these 3D printed scaffolds to enhance bone-defect repair with TNF/TNFR signaling involvement.
5. Experimental Section Materials: alginate (Alg), hydroxyapatite (nHAp), gelatin (Type A), CaCl2, N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC) were all purchased from Aladdin Reagent Inc., RhBMP-2 was purchased from Sino Biological Inc. (Beijing, P.R. China), and Atsttrin was synthesized by Sangon Biotech Inc. (Shanghai, P.R. China). In Vitro Osteoblastic Differentiation: C3H cells were seeded into a 24-wells plate and cultured in serum-free Dulbecco’s modified Eagle’s medium (DMEM) containing rhBMP-2 (200 ng mL−1), with or without TNF-α (50 ng mL−1). For rescue treatment, cells were treated with Atsttrin (10 nM or 50 nM) in the presence of BMP-2 stimulation. ALP Staining: Before staining, cells were washed with phosphatebuffered saline (PBS) and fixed with 4% (w/v) paraformaldehyde for 30 min. The cells were then stained with 5-Bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium chloride (BCIP/NBT) alkaline phosphatase color solution for 30 min to visualize the ALP activity. Quantitative Real-Time Polymerase Chain Reaction: Total cellular RNA was isolated by using Trizol (Invitrogen, Carlsbad, CA, USA), according to the manufacturer’s protocol. PCR was performed using a Brilliant SYBR Green QPCR Master Mix (TakaRa, Japan) with a Light Cycler apparatus (ABI 7900HT). The PCR cycling consisted of 40 cycles of amplification of the template DNA with primer annealing at 60 °C. The relative level of each target gene was then calculated using the 2-ΔΔCt method. Each real-time PCR was performed on at least 3 different experimental samples and representative results are shown as target gene expression normalized to reference gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH) . PCR primer pairs were designed
Adv. Healthcare Mater. 2015, 4, 1701–1708
using primer 5.0 based on the sequences of different exons of the corresponding genes. Fabrication and Characterization of Scaffolds: Alginate and gelatin powder was sterilized by UV-radiation. Alginate was dissolved in distilled water at a concentration of 2% (w/v), gelatin and nano-hydroxyapatite particles were added under stirring with or without Atsttrin protein, until a homogeneous solution with a suitable viscosity for 3D-printing was achieved. A pneumatic dispensing system with a syringe nozzle of 100 µm was used for bioprinting the composite scaffolds. The machine was operated within a laminar flow-cabinet and scaffolds were printed into tissue-culture plates. After printing, the scaffolds were cross-linked with 2% CaCl2 (w/v) for 10 minutes, followed by further cross-linking with 1% (w/v) EDC solution overnight. The scaffolds were washed with PBS for subsequent in vitro cell seeding or in vivo implantation. Some of the scaffolds were lyophilized for SEM investigation. Scanning Electron Microscopy: A scanning electron microscope (Hitachi S-3000N, Japan) was used to investigate the surface morphology of the scaffolds and morphology of the cells seeded on the scaffolds. BMSCs were seeded on Alg/nHAp composite scaffolds or Atsttrin-Alg/nHAp composite scaffolds at a density of 2 × 104 cells cm−2 and cultured in DMEM supplemented with 10% (v/v) fetal bovine serum. The medium was replaced every 3 days. After 7 days of culture, the specimens were fixed with 0.25% (w/v) glutaraldehyde solution for 24 h. After rinsing 3 times in PBS, the specimens were immersed in OsO4 (Ted Pella) for 1 h and then rinsed 3 times again in PBS. Then the specimens were dehydrated in increasing concentrations of acetone (30–100%, v/v). After drying, the specimens were mounted on aluminum stubs and coated with gold, and then viewed at an accelerating voltage of 15 kV or 20 kV. For the scaffolds without cells, the specimens were mounted on aluminum stubs and coated with gold directly. Measurements of Atsttrin Release from the Scaffold: We used a MicroBCA protein Assay Kit (Life Technologies) to measure the cumulative release of Atsttrin. Firstly, we used different concentrations of Atsttrin (1 µg mL−1, 2 µg mL−1, 4 µg mL−1, 8 µg mL−1,16 µg mL−1, and 32 µg mL−1) to find the linear working range and draw a standard curve. Scaffolds containing Atsttrin were placed into the top compartments of a trans-well system at 37 °C, and the release medium (PBS) in the lower compartment was collected and refreshed at multiple points in time (1 h, 6 h, 12 h, 24 h, 48 h, 72 h, 120 h, and 168 h). The release medium samples were diluted to the working range and the absorbance at 562 nm was measured on a plate reader according to the manufacturer’s protocol. Data are expressed as cumulative release in% of total input. In Vivo Studies of the Calvarial Bone Defect Model in C57BL/6J Mice: All animals were treated according to standard guidelines approved by the Zhejiang University Ethics Committee. The skin of C57BL/6J mice was cut and the periosteum was elevated. The surface of the calvarial bone was exposed, and oversized bone defects of 7 mm × 5 mm were created. The defects were untreated, implanted with the Alg/nHAp composite scaffolds, or implanted with the Atsttrin-Alg/nHAp composite scaffolds. The calvarial bone defects were completely covered with skin and sutured. Animals were sacrificed after 7 days, 8 weeks, and 16 weeks post-implantation and whole calvarial samples from each group (n = 5) were collected. X-Ray Examination: X-ray scanning was used to evaluate the tissueengineered bone formation in the calvarial bone of the mice animal models. X-ray scanning was detected using a Kodak In-Vivo Imaging System. The integrated optical density (IOD) was used to calculate the density of new bone-like tissue within the defect using Image-Pro Plus software. Histological Examination: The harvested specimens at 8 weeks and 16 weeks post-implantation were fixed in neutral 4% (v/v) paraformaldehyde for 24 h immediately after collection, and decalcified in neutral 10% (v/v) ethylenediaminetetraacetic acid (EDTA) solution for one month at room temperature. Then the samples were dehydrated through an alcohol gradient, cleared, and embedded in paraffin blocks. Histological sections (7 µm) were prepared using a microtome and subsequently stained with Hematoxylin and Eosin (H&E) or Masson’s trichrome. The stained sections were photographed digitally under a
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
FULL PAPER
additional Atsttrin released from the scaffolds can promote bone regeneration. Immunohistochemical analysis showed that 7 days after scaffold implantation, the number of TNF-α positive cells was significantly decreased in the Atsttrin-Alg/nHAp scaffold group, as compared to the other groups. Moreover, there was a higher expression of Runx2 in the Alg/nHAp scaffold group. These results thus indicated that Atsttrin released from scaffolds may promote bone regeneration through modulation of the TNF-α signaling pathways.
1707
www.advhealthmat.de
FULL PAPER
www.MaterialsViews.com microscope. Semi-quantitative analysis was carried out using Image-Pro Plus software to show the relative amount of bone formation. Five fields were selected and newly formed mineralized tissue area in each field was calculated and shown as a percentage to total tissue area. Immunohistochemistry: To analyze the TNF-α concentration around the wound sites at 7 days post calvarial bone-defect surgery, immunohistochemical staining was performed on paraffin sections. Mice paraffin sections (7 µm) were incubated with 0.4% pepsin (Sangon Biotech, Shanghai, P.R. China) in 5 mM HCl at 37 °C for 20 min for antigen retrieval. Endogenous peroxidase was blocked by incubation with 3% (v/v) hydrogen peroxide in methanol for 5 min. Non-specific protein binding was blocked by incubation with 2% (w/v) BSA. After overnight incubation at 4 °C with rabbit anti-mouse TNF-α monoclonal antibody, sections were then incubated with goat anti-rabbit secondary antibodies (Beyotime Institute of Biotechnology Inc., P.R. China) for 2 hours at room temperature. The diaminobutyrate (DAB) substrate system (Solarbio Science & Technology Co., Ltd, Beijing, China) was utilized for color development. Hematoxylin staining was used to reveal the cell nuclei. Immunohistochemical staining for Runx2 was performed utilizing a similar protocol except for different appropriate primary and secondary antibodies.
Supporting Information Supporting information is available from the Wiley Online Library or from the author.
Acknowledgements This work was supported by NSFC grants (81125014, 81330041, 81472115, J1103603), and Zhejiang Provincial Grants (LY13C100001, 2012C33015). Additional sponsorship by the Regenerative Medicine in Innovative Medical Subjects of Zhejiang Province and the National Key Scientific Program (2012CB966604), the National High Technology Research and Development Program of China (863 Program) (No.2012AA020503, No.2015AA020302, No.2015AA020303), International Science & Technology Cooperation Program of China (Grant No. 2011DFA32190), Technology innovation project (CXZZ20130320172336579) supported by the Science Technology and Innovation Committee of Shenzhen Municipality is also greatly acknowledged. The authors thank Hanmin Chen for his assistance with the SEM imaging. Received: March 24, 2015 Revised: May 18, 2015 Published online: June 17, 2015
[1] R. Tevlin, A. McArdle, D. Atashroo, G. Walmsley, K. Senarath-Yapa, E. Zielins, K. Paik, M. Longaker, D. Wan, J. Dental Res. 2014, 93, 1187. [2] M. M. Hanasono, N. Goel, F. DeMonte, Ann. Plastic Surgery 2009, 62, 653. [3] M. D. Maves, B. H. Matt, Otolaryngology–Head and Neck Surgery 1986, 95, 464.
1708
wileyonlinelibrary.com
[4] E. M. Horwitz, P. L. Gordon, W. K. Koo, J. C. Marx, M. D. Neel, R. Y. McNall, L. Muul, T. Hofmann, Proc. Natl. Acad. Sci. 2002, 99, 8932. [5] K. J. Burg, S. Porter, J. F. Kellam, Biomaterials 2000, 21, 2347. [6] S. J. Hollister, Nat. Mater. 2005, 4, 518. [7] J. Tan, C. Chua, K. Leong, Biomed. Microdevices 2013, 15, 83. [8] S. Catros, J.-C. Fricain, B. Guillotin, B. Pippenger, R. Bareille, M. Remy, E. Lebraud, B. Desbat, J. Amédée, F. Guillemot, Biofabrication 2011, 3, 025001. [9] N. E. Fedorovich, J. R. De Wijn, A. J. Verbout, J. Alblas, W. J. Dhert, Tissue Eng. Part A 2008, 14, 127. [10] H. W. Kim, H. H. Lee, J. Knowles, J. Biomed. Mater. Res. Part A 2006, 79, 643. [11] K. A. Athanasiou, G. G. Niederauer, C. Agrawal, Biomaterials 1996, 17, 93. [12] C. Petrie, S. Tholpady, R. Ogle, E. Botchwey, J. Biomed. Mater. Res. Part A 2008, 85, 61. [13] E. Yoon, S. Dhar, D. E. Chun, N. A. Gharibjanian, G. R. Evans, Tissue Eng. 2007, 13, 619. [14] M. Shi, J. D. Kretlow, A. Nguyen, S. Young, L. Scott Baggett, M. E. Wong, F. Kurtis Kasper, A. G. Mikos, Biomaterials 2010, 31, 4146. [15] J. D. Kretlow, S. Young, L. Klouda, M. Wong, A. G. Mikos, Adv. Mater. 2009, 21, 3368. [16] F. Wiria, K. Leong, C. Chua, Y. Liu, Acta Biomater. 2007, 3, 1. [17] C. Rodrigues, P. Serricella, A. Linhares, R. Guerdes, R. Borojevic, M. Rossi, M. Duarte, M. Farina, Biomaterials 2003, 24, 4987. [18] D. Barbieri, H. Yuan, F. De Groot, W. R. Walsh, J. D. De Bruijn, Acta Biomater. 2011, 7, 2007. [19] N. E. Fedorovich, W. Schuurman, H. M. Wijnberg, H.-J. Prins, P. R. Van Weeren, J. Malda, J. Alblas, W. J. Dhert, Tissue Eng. Part C: Methods 2011, 18, 33. [20] J. A. Rowley, G. Madlambayan, D. J. Mooney, Biomaterials 1999, 20, 45. [21] G. Turco, E. Marsich, F. Bellomo, S. Semeraro, I. Donati, F. Brun, M. Grandolfo, A. Accardo, S. Paoletti, Biomacromolecules 2009, 10, 1575. [22] H. Liu, H. Peng, Y. Wu, C. Zhang, Y. Cai, G. Xu, Q. Li, X. Chen, J. Ji, Y. Zhang, Biomaterials 2013, 34, 4404. [23] J. M. Oliveira, M. T. Rodrigues, S. S. Silva, P. B. Malafaya, M. E. Gomes, C. A. Viegas, I. R. Dias, J. T. Azevedo, J. F. Mano, R. L. Reis, Biomaterials 2006, 27, 6123. [24] S. Wüst, M. E. Godla, R. Müller, S. Hofmann, Acta Biomater. 2014, 10, 630. [25] J. Bayry, Trends Molecular Med. 2011, 17, 538. [26] C.-Y. Lin, K.-J. Lin, K.-C. Li, L.-Y. Sung, S. Hsueh, C.-H. Lu, G.-Y. Chen, C.-L. Chen, S.-F. Huang, T.-C. Yen, Biomaterials 2012, 33, 7422. [27] Y. Liu, L. Wang, T. Kikuiri, K. Akiyama, C. Chen, X. Xu, R. Yang, W. Chen, S. Wang, S. Shi, Nat. Med. 2011, 17, 1594. [28] W. Tang, Y. Lu, Q.-Y. Tian, Y. Zhang, F.-J. Guo, G.-Y. Liu, N. M. Syed, Y. Lai, E. A. Lin, L. Kong, Science 2011, 332, 478. [29] Z. He, A. Bateman, J. Molecular Med. 2003, 81, 600. [30] J. Jian, J. Konopka, C. Liu, J. Leukocyte Biol. 2013, 93, 199. [31] Y.-P. Zhao, Q.-Y. Tian, S. Frenkel, C.-J. Liu, Biomaterials 2013, 34, 6412. [32] H. R. Lin, Y. J. Yeh, J. Biomed. Mater. Res. Part B: Appl. Biomater. 2004, 71, 52.
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Healthcare Mater. 2015, 4, 1701–1708
FULL PAPER
3D-Printed Atsttrin-Incorporated Alginate/Hydroxyapatite Scaffold Promotes Bone Defect Regeneration with TNF/ TNFR Signaling Involvement Quan Wang, Qingqing Xia, Yan Wu, Xiaolei Zhang, Feiqiu Wen, Xiaowen Chen, Shufang Zhang, Boon Chin Heng, Yong He, and Hong-Wei Ouyang* method.[1] Conventional surgical techniques for reconstruction have utilized High expression levels of pro-inflammatory tumor necrosis factor (TNF)-α autogenous, allogeneic, and prosthetic within bone defects can decelerate and impair bone regeneration. However, materials.[2] Autologous bone is currently there are few available bone scaffolds with anti-inflammatory function. The the best treatment option, but it is in limprogranulin (PGRN)-derived engineered protein, Atsttrin, is known to exert ited supply and often cannot completely antagonistic effects on the TNF-α function. Hence, this study investigates fit the size and shape of the defect sites.[3] whether 3D-printed Atsttrin-incorporated alginate(Alg)/hydroxyapatite(nHAp) With allogeneic bone there is a risk of disease transmission. Prosthetic materials scaffolds can facilitate bone healing through affecting the TNF/TNFR signalmade from polymers or metals have been ing. A 3D bioprinting system is used to fabricate Atsttrin-Alg/nHAp comwidely used, but have certain disadvanposite scaffolds, and the Atsttrin release from this scaffold is characterized, tages, such as metal corrosion and poor followed by evaluation of its efficacy on bone regeneration both in vitro and in mechanical properties.[4] vivo. The 3D-printed Atsttrin-Alg/nHAp scaffold exhibits a precisely defined Tissue engineering techniques combining the use of cells, three-dimensional structure, can sustain Atsttrin release for at least 5 days, has negligible cytobiomaterial scaffolds, and growth factors toxicity, and supports cell adhesion. Atsttrin can also attenuate the suppresare recognized as the most promising sive effects of TNF-α on BMP-2-induced osteoblastic differentiation in vitro. approach for bone repair and reconstrucThe 3D-printed Atsttrin-Alg/nHAp scaffold significantly reduces the number tion, and have been demonstrated to be of TNF-α positive cells within wound sites, 7 days after post-calvarial defect a suitable alternative to autogenous or surgery. Additionally, histological staining and X-ray scanning results also allogeneic bone grafts.[5] Previous studies have investigated the use of a diverse show that the 3D-printed Atsttrin-Alg/nHAp scaffold enhances the regeneraarray of biomaterials to fabricate 3D tion of mice calvarial bone defects. These findings thus demonstrate that the printed scaffolds for calvarial reconstrucprecise structure and anti-inflammatory properties of 3D-printed Atsttrin-Alg/ tion,[6–9] such as poly(lactic acid) (PLA),[10] nHAp scaffolds may promote bone defect repair. poly(glycolic acid) (PGA),[11] the co-polymer polylactic-co-glycolic acid (PLGA),[12,13] poly(methyl methacrylate) (PMMA),[14,15] poly(caprolactone) (PCL),[16] collagen, and hydrogels.[17–19] How1. Introduction ever, only a few of these can be incorporated with biological components through bioprinting. Alginate is one of the most The repair of complex calvarial bone defects is challenging, widely used materials for bioprinting because of its biocomand successful healing depends on the defect size, the quality patibility and ability to form stable hydrogels in the presence of soft tissue covering the defect, and choice of reconstructive Dr. Q. Wang, Dr. Q. Xia, Dr. Y. Wu, Dr. X. Zhang, Dr. S. Zhang, Prof. H.-W. Ouyang Center for Stem Cell and Tissue Engineering School of Medicine Zhejiang University Hangzhou, P. R. China E-mail: [email protected] Dr. Q. Wang, Dr. Q. Xia, Dr. Y. Wu, Dr. X. Zhang, Dr. S. Zhang, Prof. H.-W. Ouyang Zhejiang Provincial Key Laboratory of Tissue Engineering and Regenerative Medicine Hangzhou, P. R. China
DOI: 10.1002/adhm.201500211
Adv. Healthcare Mater. 2015, 4, 1701–1708
Dr. F. Wen, Dr. X. Chen Division of hematology and oncology Shenzhen Children’s Hospital Shenzhen, P. R. China Dr. B. C. Heng Department of Biosystems Science & Engineering ETH-Zurich Mattenstrasse 26, Basel, Switzerland Dr. Y. He The State Key Lab of Fluid Power Transmission and Control School of Mechanical Engineering Zhejiang University Hangzhou, P. R. China
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
1701
www.advhealthmat.de
FULL PAPER
www.MaterialsViews.com
of calcium or other divalent cations, but its biological inertness and poor mechanical properties have largely limited its use.[20,21] Hydroxyapatite (nHAp) scaffolds have also been widely utilized for bone regeneration because of its biocompatibility and osteoconductivity.[22] However, its slow degradation rates impedes the complete regeneration of bone defects. One possible approach to combine the favorable properties of alginate and nHAp is to fabricate an alginate hydrogel with a dispersed phase consisting of nanosized nHAp particles. Such types of composite scaffolds have excellent biocompatibility, structural and mechanical properties, and can be incorporated with bioactive factors that promote bone regeneration.[23,24] Apart from the structural and material properties of the scaffolds, their biological function is another important factor that influences the efficacy of bone regeneration. It has been reported that the high levels of tumor necrosis factor (TNF)-α within bone defects can decelerate and impair bone regeneration.[25,26] However, there are few bone scaffolds available with anti-inflammatory function. Many TNF-α blockers, including etanercept, infliximab, and adalimumab, are effective antiinflammatory drugs. In several previous studies, pre-treatment of stem cells with aspirin can markedly reduce TNF-α and interferon (IFN)-γ levels at the injury site, and improve bone marrow mesenchymal stem cells (BMMSCs)-based bone regeneration and cavarial defect repair in C57BL/6 mice.[27] Compared with aspirin, the growth factor progranulin (PGRN) and its derivative, Atsttrin, can bind directly to the tumor necrosis factor receptor (TNFR) and thus inhibit TNF-α/TNFR interactions more efficiently, thereby reducing inflammation in inflammatory arthritis models.[28–30] Another recent study showed that PGRN can facilitate the bone-healing process by interacting with both the bone morphogenic protein (BMP)-2 and TNF-α signaling pathways.[31] However, under inflammatory conditions, PGRN can be digested by proteases and metalloproteinase, into 6 KDa granulin units that are pro-inflammatory, thus neutralizing the anti-inflammatory effects of intact PGRN. By contrast, Atsttrin contains only partial granulin units and is not expected to release such pro-inflammatory fragments. This feature, coupled with its longer active halflife (ca. 120 hours) than PGRN (ca. 40 hours) and its selective binding to TNFR1/TNFR2, thus makes Atsttrin a promising bioactive factor that can be incorporated within anti-inflammatory scaffolds. In this study, we hypothesize that 3D-printed Atsttrin-incorporated alginate/hydroxyapatite scaffolds can promote bone healing through perturbation of the TNF/TNFR signaling. A 3D bioprinting system was used to fabricate Atsttrin-Alg/nHAp composite scaffolds, and the Atsttrin release profile from the scaffold was characterized, followed by evaluation of its efficacy in bone regeneration, both in vitro and in vivo.
2. Results 2.1. Atsttrin Attenuates the Suppressive Effect of TNF-α on BMP-2-Induced Osteoblastic Differentiation We investigated the effects of TNF-α and TNF-α/Atsttrin on BMP-2-induced osteoblastic differentiation and whether
1702
wileyonlinelibrary.com
additional Atsttrin can affect this process. Murine mesenchymal stem cells (C3H cells) were cultured in recombinant human bone morphogenetic protein-2 (rhBMP-2) (200 ng mL−1) and TNF-α (50 ng mL−1)-supplemented medium, with or without Atsttrin for several days, to study the effects of TNF-α and TNF-α/Atsttrin on BMP-2-induced osteoblastic differentiation. After 7 days, alkaline phosphatase (ALP) staining (Figure 1A) demonstrated that the presence of TNF-α inhibited BMP-2-induced ALP expression in C3H cells, and that the presence of Atsttrin attenuated this inhibitory effect in a dose-dependent manner. Consistent with the observed differences in expression of the early osteogenic marker ALP, the intensity of Alizarin red staining was also lower in the TNF-α treated C3H cells and higher in the TNF-α/Atsttrin treated C3H cells (Figure 1B). Furthermore, quantization of the ALP activity (Figure 1C) demonstrated that Atsttrin could attenuate the suppressive effect of TNF-α on BMP-2-induced osteoblastic differentiation. To further verify these findings, the transcription levels of osteogenic marker genes such as COL1A1, OCN, and OPN in C3H cells were quantified by real-time polymerase chain reaction (PCR) (Figure 1D–F). TNF-α treatment reduced the transcriptional levels of these marker genes under BMP-2 stimulation, whereas Atsttrin increased the expression of these genes. Put together, these data indicate that Atsttrin can attenuate the suppressive effect of TNF-α on BMP-2-induced osteoblastic differentiation. We also employed a non-contacting co-culture (trans-well cell culture) system to investigate the efficiency of Atsttrin that is released from the scaffold. C3H cells were seeded on the bottom of 24-well plates, and cultured in rhBMP-2 (200 ng mL−1) and TNF-α (50 ng mL−1)-supplemented medium. Atsttrin-incorporated scaffolds or scaffolds without Atsttrin were put onto the membrane of suitable trans-wells. The alkaline phosphatase (ALP) staining and a quantization of the ALP activity demonstrated that Atsttrin-incorporated scaffolds have a similar efficiency as separately added Atsttrin, indicating that Atsttrin released from the scaffolds could also attenuate the suppressive effect of TNF-α on BMP-2-induced osteoblastic differentiation (Figure S1, Supporting Information).
2.2. Fabrication and Characterization of Bioactive Atsttrin-Incorporated Composite Scaffolds Alg/nHAp composite scaffolds were fabricated with a pneumatic bioprinting system (Figure 2A). The size of strand extruded from the nozzle was tailored by adjusting the speed of deposition and the pressure of extrusion (Figure 2B). Examination of the macroscopic appearance and morphology revealed that the printed scaffolds have vertically connected pores, with a strand diameter of 100 µm and a clearance of 400 µm between strands (Figure 2C,D). 3D hydrogel scaffolds consisting of five or more layers with regular structure could also be printed (Figure 2E). Scanning electron microscopy (SEM) (Figure 2F) analysis revealed that the nano-hydroxyapatite particles are uniformly distributed on the surface of the scaffold.
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Healthcare Mater. 2015, 4, 1701–1708
www.advhealthmat.de www.MaterialsViews.com
FULL PAPER Figure 1. Atsttrin attenuates the suppressive effect of TNF-α on BMP-2-induced osteoblastic differentiation. C3H cells were treated with rhBMP2(200 ng mL−1) in the presence of TNF-α (50 ng mL−1) or TNF-α (50 ng mL−1)/Atsttrin (10 nM or 50 nM). A) ALP staining assay for detecting ALPase activity on day 7. B) Alizarin red staining assay for detecting calcium deposition on day 21. C) Quantification assay of ALPase activity on day 3, normalized with the untreated cells. D–F) Real-time PCR and relative quantification of COL1A1 (D), OCN (E), OPN (F), normalized with respect to GAPDH. Each bar represents the mean S.D. of at least three independent replicate experiments. *p < 0.05 and **p < 0.01.
Cell proliferation analysis demonstrated that the printed scaffolds exerted negligible a cytotoxic effect on the cells (Figure 3A). The sustained release of Atsttrin protein from the Atsttrin-Alg/nHAp composite scaffolds was observed for about 5 days (Figure 3B and Figure S2, Supporting Information).
Carboxyfluorescein diacetate succinimidyl ester (CFDA-SE) staining and visualization of the C3H cells revealed that the cells were retained within the scaffold, and subsequent SEM images showed that the cells spread well over the scaffold surface (Figure 3C,D).
Figure 2. Schematic representation of the 3D bioprinter and morphology of the 3D printed Atsttrin-bioactive scaffolds. A) Schematic diagram of the pneumatic dispensing system. B) The effects of deposition speed and extrusion pressure on strand size. C) The macroscopic appearance of the single layer printed scaffold. D) The morphology of the printed scaffold viewed under an optical microscope. Scale bar, 500 µm. E) The macroscopic appearance of the multilayer printed scaffold. F) SEM image of the surface morphology of the printed scaffold. Scale bar, 50 µm.
Adv. Healthcare Mater. 2015, 4, 1701–1708
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
1703
www.advhealthmat.de
FULL PAPER
www.MaterialsViews.com
be detected at 16 weeks post-implantation (Figure 4C,D). Quantification of the opacity index within the defect by calculating the integrated optical density showed that the Atsttrin-Alg/nHAp group produced more bone-like tissues than the Alg/nHAp group at 8 and 16 weeks post-implantation (Figure 4I).
2.3.2. Histological Analysis of Defect Regions For each specimen, transverse sections of the entire calvaria were initially examined at low magnification to identify regions of interest. Both H&E staining (Figure 5A) and Masson's trichrome staining (Figure 5D) revealed that the samples of the control group typically exhibited no bone formation within the defect regions at 8 weeks post-implantation. The Alg/nHAp and Atsttrin-Alg/nHAp groups Figure 3. Cytotoxicity, controlled release profile, and cell adhesion analysis of the printed scaf- exhibited significantly more regenerative folds. A) Cytotoxicity analysis of the scaffold. B) Cumulative release curve of Atsttrin from activity within the defect sites. In the defects the scaffold in vitro. C) The CFDA-SE labeled C3H cells adhered onto the scaffolds on day 1. of the Alg/nHAp group only some fibrous D) The SEM image of the morphology of C3H cells on the Atsttrin-Alg/nHAp scaffolds on day tissue and negligible bone tissue was observed 3. *p < 0.05 and **p < 0.01. (Figure 5B,E), compared to this far more new bone-like tissue formation was detected in the Atsttrin-Alg/nHAp group (Figure 5C,F). Fragmented remains of the scaffolds were visible in both groups at 2.3. Efficacy of Bioactive Atsttrin-Incorporated Composite 8 weeks post-implantation, with new bone having been formed Scaffolds for Bone Regeneration in a Mice Calvarial Defect around these fragments. At 16 weeks post-implantation, the conModel trol group still exhibited negligible bone formation (Figure 6A,D). By contrast, much bone formation was observed in the defect 2.3.1. Scaffold Implantation and Evaluation of Bone Regeneration We surgically created a calvarial bone defect model in C57BL/6J mice to investigate the efficacy of the Atsttrin-bioactive scaffold for bone regeneration. Oversized calvarial bone defects of 7 mm × 5 mm were established (Figure 4A). The defects were implanted with the Alg/nHAp composite scaffold, or with the Atsttrin-Alg/nHAp composite scaffold (Figure 4B). The blank group served as the untreated control without scaffold implantation. Animals were sacrificed at 8 and 16 weeks after implantation and whole calvaria samples from each group were collected. As revealed by X-ray images, bone-like tissue was formed within the defect areas of the Atsttrin-Alg/nHAp group at 8 and 16 weeks post-implantation. The new bone tissue was localized at the edges and within the center of the defect at 16 weeks postimplantation (Figure 4G,H). However, much less bone-like tissue was detected within the defects of the Alg/nHAp group at 16 weeks post-implantation (Figure 4E,F). Radiopaque tissue was almost undetectable in the blank groups at 8 weeks post-implantation, and minimal bone-like tissue formation could
1704
wileyonlinelibrary.com
Figure 4. The calvarial-defect surgery procedure and X-ray imaging of whole calvarias after 8 and 16 weeks of implantation of the scaffolds. A,B) The surgical operation of mice calvarial bone defect model. C–H) X-ray imaging of whole calvarias of the blank group (C,D), Alg/nHAp group (E,F) and Atsttrin-Alg/nHAp group (G,H). The yellow arrows indicate the new bone. I) The integrated optical densities of the Alg/nHAp group, Atsttrin-Alg/nHAp group, and the blank group. *p < 0.05 and **p < 0.01.
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Healthcare Mater. 2015, 4, 1701–1708
www.advhealthmat.de www.MaterialsViews.com
FULL PAPER Figure 5. Histological analysis of the repaired calvarias after 8 weeks of implantation in vivo. A–C) H&E staining of the repaired calvarias in the blank group (A), Alg/nHAp scaffold group (B), and Atsttrin-Alg/nHAp scaffold group (C). D–F) Masson's trichrome staining of the repaired calvarias in the blank group (D), Alg/nHAp scaffold group (E), and Atsttrin-Alg/nHAp scaffold group (F).
areas of the Alg/nHAp and Atsttrin-Alg/nHAp groups, with the Atsttrin-Alg/nHAp group exhibiting superior bone regeneration (Figure 6C,F) compared to the Alg/nHAp group (Figure 6B,E). This was confirmed by a semi-quantitative analysis showing the relative amount of bone formation in the calvarial defect area from different groups (Figure S3, Supporting Information).
Subsequently, we examined the effect of Atsttrin on TNF-α expression levels in the different groups at 7 days post-implantation. Immunohistochemical analysis showed that the TNF-α concentration expression is very high in the control group (Figure 7A–C). For each sample, six sections for which five fields each of immunohistochemical staining were selected to
Figure 6. Histological analysis of the repaired calvarias after 16 weeks of implantation in vivo. A–C) H&E staining of the repaired calvarias in the blank group (A), Alg/nHAp scaffold group (B), and Atsttrin-Alg/nHAp scaffold group (C). D–F) Masson's trichrome staining of the repaired calvarias in the blank group (D), Alg/nHAp scaffold group (E), and Atsttrin-Alg/nHAp scaffold group (F).
Adv. Healthcare Mater. 2015, 4, 1701–1708
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
1705
www.advhealthmat.de
FULL PAPER
www.MaterialsViews.com
Figure 7. Immunohistochemical analysis of three different groups. A–C) Immunohistochemical staining of TNF-α positive cells, 7 days after defect surgery in the three different groups. D) Semi-quantitative analysis of the relative amounts of TNF-α-positive cells in the different groups. E–G) Runx2 expression in the calvarial defect regions after 16 weeks of implantation in the three different groups. H) Semi-quantitative analysis of the relative numbers of Runx2-positive cells in the different groups. *p < 0.05 and **p < 0.01.
carry out the semi-quantitative analysis. The number of TNF-αpositive cells was found to be significantly reduced in the Atsttrin-Alg/nHAp group, as compared to the Alg/nHAp and control groups (Figure 7D), thus indicating that Atsttrin released from the scaffold exerted an inhibitory effect on TNF-α formation and accumulation. Immunohistological analysis of runt-related transcription factor 2 (Runx2) was also performed to compare osteogenesis in the three groups (Figure 7E-G). There was a higher positive expression of Runx2 in the Atsttrin-Alg/nHAp group, as compared to the Alg/nHAp and control groups (Figure 7H), thus suggesting that Atsttrin promotes osteogenesis.
3. Discussion In this study, we utilized a 3D bioprinting technology to fabricate an anti-inflammatory scaffold through incorporation of Atsttrin, and evaluate its efficacy in bone regeneration with a mouse calvarial defect model. It was demonstrated that such scaffolds could continuously release Atsttrin and attenuate the suppressive effects of TNF-α on BMP-2-induced osteoblastic differentiation in vitro. The in vivo study also demonstrated that the Atsttrin-incorporated scaffold group can decrease TNF-α accumulation and up-regulate Runx2 expression, achieving better bone formation. 3D printing technologies have been widely utilized in the field of tissue engineering. Natural materials such as alginate, collagen, silk fibroin, and various synthetic materials such as PCL, PLA, and PVA have been fabricated into scaffolds or cell/ hydrogel structures, using appropriate material-forming technologies, such as fused deposition modeling (FDM), stereo lithography apparatus (SLA), and selective laser sintering (SLS). These scaffolds fabricated by 3D printing still have some limitations. For example, the material properties of PCL or PLA
1706
wileyonlinelibrary.com
are unsuitable for incorporating biological components. The mechanical strength of the 3D printed scaffolds is often insufficient for tissue-engineering applications, and their degradation rate is often too fast for optimal bone regeneration. In our study, a pneumatic bioprinting system was used to fabricate Alg/nHAp composite scaffolds with vertically connected pores and a strand diameter of 100 µm. The nHAp was added to the alginate hydrogel within the 3D bioprinted multilayer structure to improve the mechanical properties according to previous research,[23,32] and gelatin was used to help maintain the shape of the printed constructs. Such scaffolds had negligible cytotoxicity, supported cell adhesion, and exhibited sustained release of Atsttrin in vitro. TNF-α/TNFR signaling has been extensively studied because of its integral position at the apex of the pro-inflammatory cytokine cascade. Previous studies showed that Atsttrin, an engineered protein composed of three fragments of the growth factor progranulin (PGRN), exhibited selective TNFR binding, making it a potential antagonist of TNF-α signaling.[28] The administration of recombinant PGRN protein can promote bone healing and enhance BMP-2 function in the presence of high TNF-α levels in a radial bone defect model.[31] Considering that a scaffold is essential for bone-defect regeneration, we therefore utilized the bioprinted Alg/nHAp composite scaffold for controlled sustained release of Atsttrin in the defect region, as an alternative to direct administration of Atsttrin. We showed that Atsttrin could attenuate the suppressive effects of TNF-α on BMP-2 induced osteogenesis in vitro. Upon implantation of this Atsttrin-incorporated 3D-bioprinted scaffold, radiographic analysis demonstrated better bone formation in the Atsttrin-incorporated composite scaffold group, as compared to the scaffold-only and blank groups. Histological analyses of samples at 8 and 16 weeks post-surgery conclusively showed enhanced bone formation in the Atsttrin-incorporated composite scaffold group, thus validating our hypothesis that
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Healthcare Mater. 2015, 4, 1701–1708
www.advhealthmat.de www.MaterialsViews.com
4. Conclusion Our findings are consistent with previous studies that showed that aspirin suppressed TNF-α and IFN-γ expression and reversed the pro-inflammatory cytokine-induced osteogenic deficiency of BMMSCs.[27] Site-specific aspirin treatment was shown to enhance BMMSC-based tissue regeneration through the suppression of TNF-α and IFN-γ in a calvarial bone-defect repair model. The results of this study thus suggest that sitespecific modulation of local inflammatory reactions, particularly the suppression of inflammatory cytokine TNF-α, can promote bone healing, although the repair effect is not that satisfactory. Stem cells seems to be essential for calvarial bone healing, so it is logical to expect that the combination of stem cells and Atsttrin will achieve better calvarial bone healing in future. In summary, our findings demonstrate that a composite of alginate and nHAp forms an appropriate bio-ink for 3D printed scaffolds, and that Atsttrin is a promising bioactive factor that can be incorporated within these 3D printed scaffolds to enhance bone-defect repair with TNF/TNFR signaling involvement.
5. Experimental Section Materials: alginate (Alg), hydroxyapatite (nHAp), gelatin (Type A), CaCl2, N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC) were all purchased from Aladdin Reagent Inc., RhBMP-2 was purchased from Sino Biological Inc. (Beijing, P.R. China), and Atsttrin was synthesized by Sangon Biotech Inc. (Shanghai, P.R. China). In Vitro Osteoblastic Differentiation: C3H cells were seeded into a 24-wells plate and cultured in serum-free Dulbecco’s modified Eagle’s medium (DMEM) containing rhBMP-2 (200 ng mL−1), with or without TNF-α (50 ng mL−1). For rescue treatment, cells were treated with Atsttrin (10 nM or 50 nM) in the presence of BMP-2 stimulation. ALP Staining: Before staining, cells were washed with phosphatebuffered saline (PBS) and fixed with 4% (w/v) paraformaldehyde for 30 min. The cells were then stained with 5-Bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium chloride (BCIP/NBT) alkaline phosphatase color solution for 30 min to visualize the ALP activity. Quantitative Real-Time Polymerase Chain Reaction: Total cellular RNA was isolated by using Trizol (Invitrogen, Carlsbad, CA, USA), according to the manufacturer’s protocol. PCR was performed using a Brilliant SYBR Green QPCR Master Mix (TakaRa, Japan) with a Light Cycler apparatus (ABI 7900HT). The PCR cycling consisted of 40 cycles of amplification of the template DNA with primer annealing at 60 °C. The relative level of each target gene was then calculated using the 2-ΔΔCt method. Each real-time PCR was performed on at least 3 different experimental samples and representative results are shown as target gene expression normalized to reference gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH) . PCR primer pairs were designed
Adv. Healthcare Mater. 2015, 4, 1701–1708
using primer 5.0 based on the sequences of different exons of the corresponding genes. Fabrication and Characterization of Scaffolds: Alginate and gelatin powder was sterilized by UV-radiation. Alginate was dissolved in distilled water at a concentration of 2% (w/v), gelatin and nano-hydroxyapatite particles were added under stirring with or without Atsttrin protein, until a homogeneous solution with a suitable viscosity for 3D-printing was achieved. A pneumatic dispensing system with a syringe nozzle of 100 µm was used for bioprinting the composite scaffolds. The machine was operated within a laminar flow-cabinet and scaffolds were printed into tissue-culture plates. After printing, the scaffolds were cross-linked with 2% CaCl2 (w/v) for 10 minutes, followed by further cross-linking with 1% (w/v) EDC solution overnight. The scaffolds were washed with PBS for subsequent in vitro cell seeding or in vivo implantation. Some of the scaffolds were lyophilized for SEM investigation. Scanning Electron Microscopy: A scanning electron microscope (Hitachi S-3000N, Japan) was used to investigate the surface morphology of the scaffolds and morphology of the cells seeded on the scaffolds. BMSCs were seeded on Alg/nHAp composite scaffolds or Atsttrin-Alg/nHAp composite scaffolds at a density of 2 × 104 cells cm−2 and cultured in DMEM supplemented with 10% (v/v) fetal bovine serum. The medium was replaced every 3 days. After 7 days of culture, the specimens were fixed with 0.25% (w/v) glutaraldehyde solution for 24 h. After rinsing 3 times in PBS, the specimens were immersed in OsO4 (Ted Pella) for 1 h and then rinsed 3 times again in PBS. Then the specimens were dehydrated in increasing concentrations of acetone (30–100%, v/v). After drying, the specimens were mounted on aluminum stubs and coated with gold, and then viewed at an accelerating voltage of 15 kV or 20 kV. For the scaffolds without cells, the specimens were mounted on aluminum stubs and coated with gold directly. Measurements of Atsttrin Release from the Scaffold: We used a MicroBCA protein Assay Kit (Life Technologies) to measure the cumulative release of Atsttrin. Firstly, we used different concentrations of Atsttrin (1 µg mL−1, 2 µg mL−1, 4 µg mL−1, 8 µg mL−1,16 µg mL−1, and 32 µg mL−1) to find the linear working range and draw a standard curve. Scaffolds containing Atsttrin were placed into the top compartments of a trans-well system at 37 °C, and the release medium (PBS) in the lower compartment was collected and refreshed at multiple points in time (1 h, 6 h, 12 h, 24 h, 48 h, 72 h, 120 h, and 168 h). The release medium samples were diluted to the working range and the absorbance at 562 nm was measured on a plate reader according to the manufacturer’s protocol. Data are expressed as cumulative release in% of total input. In Vivo Studies of the Calvarial Bone Defect Model in C57BL/6J Mice: All animals were treated according to standard guidelines approved by the Zhejiang University Ethics Committee. The skin of C57BL/6J mice was cut and the periosteum was elevated. The surface of the calvarial bone was exposed, and oversized bone defects of 7 mm × 5 mm were created. The defects were untreated, implanted with the Alg/nHAp composite scaffolds, or implanted with the Atsttrin-Alg/nHAp composite scaffolds. The calvarial bone defects were completely covered with skin and sutured. Animals were sacrificed after 7 days, 8 weeks, and 16 weeks post-implantation and whole calvarial samples from each group (n = 5) were collected. X-Ray Examination: X-ray scanning was used to evaluate the tissueengineered bone formation in the calvarial bone of the mice animal models. X-ray scanning was detected using a Kodak In-Vivo Imaging System. The integrated optical density (IOD) was used to calculate the density of new bone-like tissue within the defect using Image-Pro Plus software. Histological Examination: The harvested specimens at 8 weeks and 16 weeks post-implantation were fixed in neutral 4% (v/v) paraformaldehyde for 24 h immediately after collection, and decalcified in neutral 10% (v/v) ethylenediaminetetraacetic acid (EDTA) solution for one month at room temperature. Then the samples were dehydrated through an alcohol gradient, cleared, and embedded in paraffin blocks. Histological sections (7 µm) were prepared using a microtome and subsequently stained with Hematoxylin and Eosin (H&E) or Masson’s trichrome. The stained sections were photographed digitally under a
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
FULL PAPER
additional Atsttrin released from the scaffolds can promote bone regeneration. Immunohistochemical analysis showed that 7 days after scaffold implantation, the number of TNF-α positive cells was significantly decreased in the Atsttrin-Alg/nHAp scaffold group, as compared to the other groups. Moreover, there was a higher expression of Runx2 in the Alg/nHAp scaffold group. These results thus indicated that Atsttrin released from scaffolds may promote bone regeneration through modulation of the TNF-α signaling pathways.
1707
www.advhealthmat.de
FULL PAPER
www.MaterialsViews.com microscope. Semi-quantitative analysis was carried out using Image-Pro Plus software to show the relative amount of bone formation. Five fields were selected and newly formed mineralized tissue area in each field was calculated and shown as a percentage to total tissue area. Immunohistochemistry: To analyze the TNF-α concentration around the wound sites at 7 days post calvarial bone-defect surgery, immunohistochemical staining was performed on paraffin sections. Mice paraffin sections (7 µm) were incubated with 0.4% pepsin (Sangon Biotech, Shanghai, P.R. China) in 5 mM HCl at 37 °C for 20 min for antigen retrieval. Endogenous peroxidase was blocked by incubation with 3% (v/v) hydrogen peroxide in methanol for 5 min. Non-specific protein binding was blocked by incubation with 2% (w/v) BSA. After overnight incubation at 4 °C with rabbit anti-mouse TNF-α monoclonal antibody, sections were then incubated with goat anti-rabbit secondary antibodies (Beyotime Institute of Biotechnology Inc., P.R. China) for 2 hours at room temperature. The diaminobutyrate (DAB) substrate system (Solarbio Science & Technology Co., Ltd, Beijing, China) was utilized for color development. Hematoxylin staining was used to reveal the cell nuclei. Immunohistochemical staining for Runx2 was performed utilizing a similar protocol except for different appropriate primary and secondary antibodies.
Supporting Information Supporting information is available from the Wiley Online Library or from the author.
Acknowledgements This work was supported by NSFC grants (81125014, 81330041, 81472115, J1103603), and Zhejiang Provincial Grants (LY13C100001, 2012C33015). Additional sponsorship by the Regenerative Medicine in Innovative Medical Subjects of Zhejiang Province and the National Key Scientific Program (2012CB966604), the National High Technology Research and Development Program of China (863 Program) (No.2012AA020503, No.2015AA020302, No.2015AA020303), International Science & Technology Cooperation Program of China (Grant No. 2011DFA32190), Technology innovation project (CXZZ20130320172336579) supported by the Science Technology and Innovation Committee of Shenzhen Municipality is also greatly acknowledged. The authors thank Hanmin Chen for his assistance with the SEM imaging. Received: March 24, 2015 Revised: May 18, 2015 Published online: June 17, 2015
[1] R. Tevlin, A. McArdle, D. Atashroo, G. Walmsley, K. Senarath-Yapa, E. Zielins, K. Paik, M. Longaker, D. Wan, J. Dental Res. 2014, 93, 1187. [2] M. M. Hanasono, N. Goel, F. DeMonte, Ann. Plastic Surgery 2009, 62, 653. [3] M. D. Maves, B. H. Matt, Otolaryngology–Head and Neck Surgery 1986, 95, 464.
1708
wileyonlinelibrary.com
[4] E. M. Horwitz, P. L. Gordon, W. K. Koo, J. C. Marx, M. D. Neel, R. Y. McNall, L. Muul, T. Hofmann, Proc. Natl. Acad. Sci. 2002, 99, 8932. [5] K. J. Burg, S. Porter, J. F. Kellam, Biomaterials 2000, 21, 2347. [6] S. J. Hollister, Nat. Mater. 2005, 4, 518. [7] J. Tan, C. Chua, K. Leong, Biomed. Microdevices 2013, 15, 83. [8] S. Catros, J.-C. Fricain, B. Guillotin, B. Pippenger, R. Bareille, M. Remy, E. Lebraud, B. Desbat, J. Amédée, F. Guillemot, Biofabrication 2011, 3, 025001. [9] N. E. Fedorovich, J. R. De Wijn, A. J. Verbout, J. Alblas, W. J. Dhert, Tissue Eng. Part A 2008, 14, 127. [10] H. W. Kim, H. H. Lee, J. Knowles, J. Biomed. Mater. Res. Part A 2006, 79, 643. [11] K. A. Athanasiou, G. G. Niederauer, C. Agrawal, Biomaterials 1996, 17, 93. [12] C. Petrie, S. Tholpady, R. Ogle, E. Botchwey, J. Biomed. Mater. Res. Part A 2008, 85, 61. [13] E. Yoon, S. Dhar, D. E. Chun, N. A. Gharibjanian, G. R. Evans, Tissue Eng. 2007, 13, 619. [14] M. Shi, J. D. Kretlow, A. Nguyen, S. Young, L. Scott Baggett, M. E. Wong, F. Kurtis Kasper, A. G. Mikos, Biomaterials 2010, 31, 4146. [15] J. D. Kretlow, S. Young, L. Klouda, M. Wong, A. G. Mikos, Adv. Mater. 2009, 21, 3368. [16] F. Wiria, K. Leong, C. Chua, Y. Liu, Acta Biomater. 2007, 3, 1. [17] C. Rodrigues, P. Serricella, A. Linhares, R. Guerdes, R. Borojevic, M. Rossi, M. Duarte, M. Farina, Biomaterials 2003, 24, 4987. [18] D. Barbieri, H. Yuan, F. De Groot, W. R. Walsh, J. D. De Bruijn, Acta Biomater. 2011, 7, 2007. [19] N. E. Fedorovich, W. Schuurman, H. M. Wijnberg, H.-J. Prins, P. R. Van Weeren, J. Malda, J. Alblas, W. J. Dhert, Tissue Eng. Part C: Methods 2011, 18, 33. [20] J. A. Rowley, G. Madlambayan, D. J. Mooney, Biomaterials 1999, 20, 45. [21] G. Turco, E. Marsich, F. Bellomo, S. Semeraro, I. Donati, F. Brun, M. Grandolfo, A. Accardo, S. Paoletti, Biomacromolecules 2009, 10, 1575. [22] H. Liu, H. Peng, Y. Wu, C. Zhang, Y. Cai, G. Xu, Q. Li, X. Chen, J. Ji, Y. Zhang, Biomaterials 2013, 34, 4404. [23] J. M. Oliveira, M. T. Rodrigues, S. S. Silva, P. B. Malafaya, M. E. Gomes, C. A. Viegas, I. R. Dias, J. T. Azevedo, J. F. Mano, R. L. Reis, Biomaterials 2006, 27, 6123. [24] S. Wüst, M. E. Godla, R. Müller, S. Hofmann, Acta Biomater. 2014, 10, 630. [25] J. Bayry, Trends Molecular Med. 2011, 17, 538. [26] C.-Y. Lin, K.-J. Lin, K.-C. Li, L.-Y. Sung, S. Hsueh, C.-H. Lu, G.-Y. Chen, C.-L. Chen, S.-F. Huang, T.-C. Yen, Biomaterials 2012, 33, 7422. [27] Y. Liu, L. Wang, T. Kikuiri, K. Akiyama, C. Chen, X. Xu, R. Yang, W. Chen, S. Wang, S. Shi, Nat. Med. 2011, 17, 1594. [28] W. Tang, Y. Lu, Q.-Y. Tian, Y. Zhang, F.-J. Guo, G.-Y. Liu, N. M. Syed, Y. Lai, E. A. Lin, L. Kong, Science 2011, 332, 478. [29] Z. He, A. Bateman, J. Molecular Med. 2003, 81, 600. [30] J. Jian, J. Konopka, C. Liu, J. Leukocyte Biol. 2013, 93, 199. [31] Y.-P. Zhao, Q.-Y. Tian, S. Frenkel, C.-J. Liu, Biomaterials 2013, 34, 6412. [32] H. R. Lin, Y. J. Yeh, J. Biomed. Mater. Res. Part B: Appl. Biomater. 2004, 71, 52.
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Healthcare Mater. 2015, 4, 1701–1708
