An injectable gel constructs with regenerative and anti-infective dual

Jun 28, 2018 - The release kinetics of BMP-2 and Bbr from the microspheres was also ... 0 (0),. Abstract: The increasing incidence of infections in im...
0 downloads 0 Views 4MB Size
Subscriber access provided by UNIV OF DURHAM

Biological and Medical Applications of Materials and Interfaces

An injectable gel constructs with regenerative and anti-infective dual effects based on assembled chitosan microspheres Bin Cai, Qin Zou, Yi Zuo, Quanjing Mei, Jinqi Ma, Lili Lin, Li Chen, and Yubao Li ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b06648 • Publication Date (Web): 28 Jun 2018 Downloaded from http://pubs.acs.org on July 3, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

An

injectable

gel

constructs

with

regenerative

and

anti-infective dual effects based on assembled chitosan microspheres Bin Cai, Qin Zou*, Yi Zuo, Quanjing Mei, Jinqi Ma, Lili Lin, Li Chen and Yubao Li* Research Center for Nano-Biomaterial, Analytical & Testing Center, Sichuan University, Chengdu 610064, China. *Corresponding author: [email protected]; [email protected] (Yubao Li)

Abstract There are increasing demand for biomaterials with dual effects of bone regeneration and anti-infection in clinic application. To achieve this goal, chitosan microspheres with either positive or negative charges were fabricated then assembled as a gel for bone healing. The positively charged chitosan microspheres (CSM) (~35.5 µm) and negatively charged O-carboxymethyl chitosan microspheres (CMCSM) (~13.5 µm) were loaded respectively with bone morphogenetic protein (BMP-2) and berberine (Bbr) via swollen encapsulation and physical adsorption without significant change of the electric charges. The release kinetics of BMP-2 and Bbr from the microspheres was also studied in vitro. The results showed that the Bbr/CMCSM microspheres group possessed high antibacterial activity against the S.Aureus; the BMP-2/CSM microspheres group also owned excellent cytocompatibility, and improved osteoinductivity with assistance of BMP-2. The assembled gel group consisting of Bbr/CMCSM and BMP-2/CSM had a porous structure allowing biological signals transfer and tissue infiltration, and exhibited significantly enhanced bone reconstruction than the respective microspheres groups, which should result from the osteoconductivity of the porous structure and the osteoinduction of the BMP-2 growth factor. The oppositely charged microspheres and their assembled gel provide a promising prospect to make injectable tissue-engineered constructs with regenerative and anti-infective dual effects for biomedical applications. 1

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Keywords: Chitosan, Microspheres, Dual effects, Porous assembled gel, Biomedical applications

1. Introduction Tissue engineering plays an increasing crucial role in the clinical application of bone repairing since the use of autologous grafting and alloplastic implants still remains multiple drawbacks, i.e. donor shortage, risk of infection and immunological response, etc.1 Scaffold is the one of essential elements to reconstruct damaged bone tissue by the ex vivo culturing of cells and transplantation into damaged areas.2 In conventional approach, the bone necessitated scaffolds present a simplex ability to provide mechanical support and mass transport to encourage cell adhesion and proliferation.3 However, they are encountered difficulty in satisfying clinical needs of more functions. The outdated scaffolds expose increasing disadvantages which are always lack of ability to inhibit the bacterial infection within the sites of bone fractures or defects. Taking various functions into account, i.e. the infection treatment and tissue regeneration, bone repairing have an urgent demand for developing materials that can deliver therapeutic and regenerative molecules in an individual and controllable manner.4 Du et al. prepared the antibacterial hydrogels incorporated with poly(glutamic acid)-based vesicles recently.5 They have also reported a kind of antibacterial peptide-mimetic copolymer, which can be self-assembled into polymersomes to encapsulate growth factors and subsequently release them during long-term antibacterial process for facilitating bone repair.6 Growth factors have an ability to regulate cell function and stimulate tissue repairing, thereby developing more effective bone regeneration.7 However, the burst release of growth factors8 and fixed shape of conventional scaffolds are confronted with a variety of problems to be resolved immediately, i.e. cyst-like bone regeneration or soft tissue swelling9 and potential gaps between materials and tissue after surgery. Unlike conventional scaffolds, the colloidal gel based on oppositely charged microspheres can act as in-situ controlled and injectable delivery vehicles for growth 2

ACS Paragon Plus Environment

Page 2 of 40

Page 3 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

factors to repair bone defects of irregular size and shape due to the pseudoplastic property of microspheres fluid.10 In addition, the modular designed gels using microspheres as building blocks exhibit distinctive functions, i.e. duplex and controlled drug delivery behaviors, delivering cells by either cell attachment or encapsulation, and so on.11 Since the oppositely charged microspheres can be induced by electrostatic interaction and assembled to be a colloidal gel without external energy and operation, it is a green and economical method to fabricate the materials for bone repairing. Recombined human bone morphogenetic protein-2 (BMP-2) is one of the most efficient growth factors that can stimulate osteogenic induction12 and promote recruitment and differentiation of bone marrow stem cells (BMSCs) into the fracture site. Berberine chloride (Bbr) is a water soluble isoquinoline alkaloid that has been used in medicinal field with a long history.13 It possesses biological and pharmacological effects of anti-fungal, anti-inflammatory and anti-bacterial, etc.14 Chitosan (CS), derived from natural and nontoxic chitin, is positively charged after dissolving in acid solution and easily processed into different forms such as nanofibers and microspheres.15 O-carboxymethyl chitosan (CMCS) is derived from CS but negatively charged and more hydrophilic than CS since the –COO- groups are introduced into the main polymer chains of CS.16 Hence, the resulting CMCSM may be in favor of delivering and releasing hydrophilic Bbr rapidly via swollen encapsulation or physical adsorption.17 As reported, both CMCSM and CSM can provide a controlled release behavior of drugs or proteins to prolong their biological effect.18, 19 A sustained elution of BMP-2 benefiting from controlled delivery of CSM may be more effective at fracture healing in contrast to a burst release alone.20 Overall, if Bbr and BMP-2 were incorporated with designed gels scaffold composed of CSM and CMCSM respectively, it would become functionalized materials providing a sustainable regenerative effect for bone regeneration and simultaneously offering a protective micro environment against osteoarthritis or osteomyelitis21 especially at earlier stage of repairing. The objective of this work is to develop a novel assembled gel scaffold that 3

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

possesses sustainable regenerative and therapeutic effects for bone healing. To achieve this goal, oppositely charged CMCSM and CSM loaded with Bbr and BMP-2 respectively (Bbr/CMCSM and BMP-2/CSM) were fabricated and assembled into a gel (Shown in Fig. 1). The sustained release behavior and different release kinetics were investigated to better understand the dual effects of Bbr and BMP-2. The dual antibacterial and osteogenic property resulting from the assembled gels were confirmed. The developed materials with dual therapeutic and regenerative effects have potential to promote bone healing meanwhile prevent the infection in clinic application.

Fig. 1 Scheme of preparation of oppositely charged microspheres and their assembled gel

2. Materials and method 4

ACS Paragon Plus Environment

Page 4 of 40

Page 5 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

2.1 Materials Chitosan (CS, Mw=200000 g/mol, degree of deacetylation DD≥90%) was from Jinan Haidebei Marine Bioengineering Co. Ltd. (Shandong, China). Carboxymethyl chitosan (CMCS, Mw=100000 g/mol, degree of deacetylation DD≥90%, degree of substitution DS≥82%) was purchased from Bomei Biotechnology Co.Ltd (Hefei, China). Glutaraldehyde (50%) was used as the cross-linker. Analytically pure chemical reagents (acetic acid, liquid paraffin (C16~C20 alkane), petroleum ether (60~90℃), isopropanol, Tween80, Span80, etc.) were used in the fabrication of microspheres.

2.2 Preparation of BMP-2/CSM and Bbr/CMCSM assembled gels As shown in Fig. 1, both CS microspheres (CSM) and CMCS microspheres (CMCSM) were prepared by the modified methods of Oil/Water emulsification and chemical crosslinking.22 Briefly, 14g water phase (2wt% CS acetic acid aqueous solution) was emulsified with 42g oil phase (liquid paraffin) containing 0.3g Tween80 and 0.8g Span80 for 30 min. For preparation of CMCSM, the ratio of water phase and oil phase was modified to be 1:10. Subsequently, the CS or CMCS emulsion was cross-linked by glutaraldehyde (50%) and homogenized at a speed of 200~400 rpm for 2h. At the end, microspheres were separated from emulsion mixture using petroleum ether and washed by isopropyl alcohol for three times, respectively. Obtained samples were lyophilized and kept in a dried condition for further research. The BMP-2 (Sigma-Aldrich) or Bbr (99.5%, Aladdin) was loaded with CSM or CMCSM via diffusional transferring.23 In brief, the CSM and CMCSM were immersed in BMP-2 (2µg/ml) and Bbr aqueous solution (2mg/ml), respectively. The volume of solution was less than that required for complete swelling of microspheres so that the biological molecules can be absorbed completely. Resulting single loading microspheres (BMP-2/CSM or Bbr/CMCSM) were collected in plastic tubes for in vitro release study and in vivo animal experiments. In order to fabricate colloidal gels, obtained 0.33g BMP-2/CSM or Bbr/CMCSM was immersed in 1ml aqueous solution to make a solid content of 33.3 w/v%. A 5

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

twin-syringe was utilized to mix CSM and CMCSM dispersion system in an equal volume (Fig. 1). In detail, each kind of prepared microspheres mixture was added into the separate chamber of the twin-syringe (DanSiTe Medical Instrument Co. Ltd., China), followed by mixing of single microspheres dispersion system through extrusion to form assemblies that were collected into plastic tubes for further research.24 The surface of CSM was stained by rhodamine B (AR, Aladdin) aqueous solution (1mg/ml) so that CMCSM and CSM in the assembled system can be distinguished.25 After 2 days, the samples were washed by water and collected for further characterization until the supernate became colorless and transparent.

2.3 Characterization of microspheres 2.3.1 Fourier transform infrared spectroscopy analysis The chemical structure of microspheres was investigated by means of Fourier transform infrared spectroscopy (FTIR) using a spectrometer (Thermofisher, Nicolet-5700). The samples were ground with KBr to prepare tablets by means of a hydraulic press. FT-IR spectra were collected in the range of 4000–500 cm−1. 2.3.2 Particle size distribution analysis and SEM observation The microspheres were firstly dispersed by ultrasound in anhydrous alcohol. Their particle size distribution was automatically analyzed and calculated by laser particle analyzer (BaiTe, BT-9300H, China). After sputter-coated with gold, the morphological characterization of microspheres was performed by means of scanning electron microscopy (SEM; HITACH S530) at 20kv. 2.3.3 Zeta potential measurement The microspheres were suspended in deionized water (pH=7.4) and homogenized thoroughly. The zeta potential of the microspheres was measured by laser Doppler electrophoresis26 using a Zetasizer Nano-Z (Malvern Instruments Ltd., United Kingdom) with a folded capillary cell (DTS 1060). 2.3.4 Swelling study Swelling tests were conducted using deionized water (pH=7.4) as contact 6

ACS Paragon Plus Environment

Page 6 of 40

Page 7 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

medium. Microspheres dried to constant weight were immersed in the medium at room temperature for 2h to equilibrium. The swollen microspheres were collected by centrifugalizing and separating from the medium. After having been superficially blotted the excess water with filter paper, they were weighed immediately on an analytical balance. The swelling ratio (Sw) was calculated using Eq. (1):27 Sw (%) = [(Wt -W0)/W0]×100

(1)

where Wt denotes the weight of the swollen microspheres at equilibrium swelling, and W0 is the initial dry weight of the microspheres. Each swelling experiment was repeated three times, and the average value was considered as the result of Sw.

2.4 Preparation and characterization of assembled microspheres 2.4.1 Inverting test A simple test tube inverting method28 was performed to offer a visible evidence for the gel formation. Briefly, the dispersion of similarly charged microspheres (CSM or CMCSM) and oppositely charged microspheres assemblies were transferred into the test tubes, respectively. After inverting the test tube within 30 s, the sample was regarded as a “gel” if no visual flow was observed. 2.4.2 Rheology The viscoelastic properties of the microspheres assembled gel were measured by oscillatory rheological experiments29 that were performed on a rheometer (RheoStress 6000, HAAKE) with a flat stainless steel parallel plate geometry (20mm diameter). All measurements were conducted on the parallel plate geometry with a gap height of 1mm at 25 ℃. For dynamic time sweeps (DTS) (9min, 0.1% strain, 1Pa), the prepared assembled gel was transferred to the rheometer. The storage modulus (G’), loss modulus (G’’) and loss tangent (tan δ) were monitored as a function of time. The creep test of the gels was also performed, in which 10 Pa of shear stress was imposed on the sample and the strain was monitored as a function of time. After 1 min of time period, the shear stress was removed and the sample was allowed to recover. The 7

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 40

strain used for DTS and the stress used for creep test were previously determined to be within the linear viscoelastic region (Fig. 8b-c). All tests were performed three times and presented by the representative results.

2.5 In vitro release study For BMP-2 and Bbr release studies in vitro, 20mg BMP-2/CSM or Bbr/CMCSM was suspended in 200µl PBS buffer solution (pH=7.4), and then placed in an incubator shaker at 37 ℃. The buffer medium was collected at predetermined time intervals and replaced by an equal volume of fresh medium immediately. The concentration of BMP-2 in the collected release medium was determined by the BMP-2 ELISA kit (YAD) according to the instruction. The released Bbr concentration was calculated according to the standard curve analyzed by the UV-Vis spectrometer (UV-1000, AOE Instruments, China) at 421 nm.22 The calibration curves between 2.05

and

10.12

mg/L

were

corresponded

to

the

regression

equation:

C=(-0.00654+A)/0.0138 (r=0.99980) (Where A is the absorbance, C is the concentration (mg/L) of Bbr and r is correlation coefficient). Various cumulative release percentages of BMP-2 or Bbr with respect to the time were calculated from the original release concentration. The cumulative release percentage (Dr) was quantified by Eq. (2):30 Dr (%) = (Ct/C∞) ×100

(2)

where Ct and C∞ represent the cumulative amount at time t and at infinite time, respectively. All experiments were performed in triplicate. As for drug-loaded microspheres, cumulative drug release data before 60% release can be best fitted according to the semi-empirical model developed by Ritger and Peppas.31 The model equation is Mt/M∞=ktn, where Mt denotes the cumulative release at time t, M∞ denotes 100% release, k is the kinetic rate constant and n is the release exponent. The correlation coefficient (r2) was used to evaluate the accuracy of the fitness. Based on the best correlation coefficient values, the most appropriate model was selected to explain the release behavior of BMP-2 and Bbr. 8

ACS Paragon Plus Environment

Page 9 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

2.6 Bacteriostatic activity of Bbr/CMCSM in vitro Antibacterial assay was performed by the disk diffusion experimental procedure.32 First, bacteriological plates were inoculated with Staphylococcus aureus (S. Aureus) and placed in an incubator for 1 day at 37 ℃. The Bbr/CMCSM was dispersed into 0.9% saline solution and dropped on the center of the agar plates with a metal ring (10mm in diameter), and the CMCSM was set as a control. Plates were then placed in the incubator overnight and the zones of inhibition were imaged.

2.7 Cytotoxicity in vitro 2.7.1 Cell culture The human osteosarcoma (MG-63) cells were cultured in F-12 Nutrient Mixture medium (Gibco, Life Technologies, USA) containing 10% (v/v) newborn bovine serum (Gibco, Life Technologies, USA) and 1% (v/v) penicillin/streptomycin antibiotic solution (MP Biomedicals, USA) at 37℃ with 5% CO2.33 In order to verify the morphological changes of MG63 cells cultured on different microspheres (CSM or CMCSM), the 3×104 cells/well were seeded on a 24-well plate at 37 ℃ for 24 h and 10 mg/mL of microspheres dispersed solution was added into the cell dishes. After incubating for 1 and 4 day(s), the microspheres-cells were washed with PBS (pH 7.4) three times. The Mitochondrion-Selective Probes (MitoTracker®, Invitrogen) and Hoechst dyes (Invitrogen) were dissolved in DMSO to prepare 1 mM and 10 mg/ml stock solution, respectively. Then 200 µl prepared Mitochondrion-Selective Probes and Hoechst dyes solution were added to stain mitochondrion34 and nucleus35 followed by sufficient incubation at 37 ℃. After that, cells were rinsed by PBS solution, fixed with 3.7% formaldehyde in PBS for 10 min at room temperature. Then place the Alexa Flour® 532 phalloidin (labeling actin filaments) staining solution on the samples for 20 minutes at room temperature to stain the actin cytoskeleton36 followed by incubation for 30min at 37 ℃. The images of cells with three kinds of fluorescence colors were visualized using a confocal laser microscope (A1R MP+, Nikon). 9

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

2.7.2 Cell Morphology At predetermined time intervals, the cells/microspheres adhering on the glass sheet were harvested, washed with PBS, fixed by glutaraldehyde (2.5%), dehydrated though a series of alcohol concentrations (30, 50, 75, 83, 95 and 100%), replaced by isoamyl acetate and dried in critical point. The morphology of cells was monitored through SEM (HITACH S530). 2.7.3 Cell viability assay For the fluorescence observation of cell viability on different microspheres (CSM or CMCSM), the cells were labeled with the live/dead reagent (LIVE/DEAD, a Viability/Cytotoxicity Kit, Life Technologies) after culture for 1 and 4 days. The working solution was prepared by adding 2µl of the supplied 4mM calcein AM stock solution and 4µl of the supplied 2mM Ethidium homodimer-1 (EthD-1) stock solution to 1mL sterile PBS. All samples were incubated with the staining solution (200µM Calcein-AM and 400µM ethidium) in the dark (37 ℃, 30 min), and the morphology of cells was imaged by an inverted fluorescence microscope (Nikon, TE2000-U). 2.7.4 Cells differentiation The bone marrow stem cells (BMSCs) isolated from SD rats’ (~100g) femur and tibia were grown and maintained in α-MEM media (GIBCO, USA), with 20% calf serum (GIBCO, USA), 1% antibiotic/antimycotic mixture, 5 ml L-glutamine (200 mM) and sodium pyruvate. The cell has been used to evaluate the sequential delivery of growth factors on the cell responses. Around 10mg BMP-2/CSM was soaked into 10ml α-MEM media with a constant shake at 37 ℃ for 7 days, and then filtrated by a membrane filter (Millex® GP, Merck Millipore Ltd.) with pore size 0.2µm to prepare osteogenic culture media. The BMSCs (passage 3) were seeded on 24-well plates followed by incubation for one night. After the cells adhered to the bottom of the plates, the nutrient solution (control group) and osteogenic culture medium (experiment group) in equivalent volume replaced the culture media respectively. After incubation for 2 days, the culture media in both groups were replaced by the nutrient solution. 10

ACS Paragon Plus Environment

Page 10 of 40

Page 11 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

After incubation for 4, 7, 14 and 21 days, the intracellular alkaline phosphatase (ALP) activity and extracellular osteocalcin (OCN) activity were measured respectively. The supernate of the culture medium was collected and the concentration of OCN was determined according to the instruction of OCN ELISA kit (Shanghai MLBIO Biotechnology Co. Ltd, China). To measure the ALP activity, BMSCs were rinsed with PBS solution, trypsin digested, counted and lysed in 2.5% Triton X-100 followed by shaking and centrifugation. The ALP activity was determined by ALP ELISA kit (Shanghai MLBIO Biotechnology Co. Ltd, China). The OD values of six parallel samples were measured at 405 nm (OD405) using an ELISA reader (Victor3, Perkin Elmer). Finally, the specific ALP and OCN activity per cell basis were reported by normalizing the ALP and OCN activity with the cell number.

2.8 In vivo study 2.8.1 Surgical procedure The New Zealand White Rabbits weighting about 2kg were used for this experiment. The protocol was approved by the Animal Ethical Committee of the Sichuan University and Chinese national guidelines for the care and use of laboratory animals were applied. Thirty randomly selected experiment animals were assigned to five groups based on the implants placed into bone defects. After routine treatment, a skin incision was made in the lateral femoral condyle of rabbits. Skin and musculature were dissected and the femoral condyle was exposed. A bone defect model (5 mm in diameter, 6 mm in depth) (Fig. S1a) was created in the middle of femoral condyle using drill. The defects were left empty or filled with drug-loaded microspheres (Fig. S1b) or colloidal gels (Fig. S1c-d) by direct injection using twin syringes in a randomized manner. The musculature and the skin were closed with suture separately. Gentamicin (10 mg/kg) was administered for 3 days postoperatively to prevent infection. After surgery, the rabbits housed in individual cages were randomly divided into 3 groups (group I (4weeks), group II (8weeks) and group III(12weeks) ), totaling 10 rabbits for each group. 2.8.2 Micro-CT analysis 11

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The harvested samples were fixed in 4% buffered paraformaldehyde. The fixed femur bone samples were scanned and examined by a micro-computed tomography (micro-CT, Viva CT80, SCANCO Switzerland). Based on the original CT scanning images of each sample, 3D images were reconstructed using the finite element analysis software. The bone tissue was segmented from air or soft tissue using a threshold range of 180~1000 before it was reconstructed. New bone formation was measured using the software based on reconstructed CT images. To exclude measurement errors at the implant margin, the percentage of bone volume per total volume (BV/TV, %) was calculated within a circle region (4.5 mm in diameter) and 5 mm in depth from the bottom of bone defect.37 These measurements were performed on at least three parallel samples, and the mean value was reported. 2.8.3 Histological observation The fixed femur bone samples were decalcified in 20% Ethylene Diamine Tetraacetic Acid. After complete decalcification, dehydration in graded ethanol solution and cleaned in xylene, samples for histological analysis were embedded in paraffin wax and sectioned. The sections were stained with hematoxylin and eosin (H&E) (ScyTek, USA) and observed under light microscopy (Nikon, Ti-U).

2.9 Statistical analysis The compared results were expressed as mean ± standard deviations. One-way analysis of variance (ANOVA) with turkey test was used to determine the statistical significance among different groups. When P﹤0.05, the statistics were considered significant.

12

ACS Paragon Plus Environment

Page 12 of 40

Page 13 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

3.Results

3.1 Microspheres characterization

3.1.1 The morphology and particle size distribution

Fig. 2 The morphology (a-b) and particle size distribution (c-d) of CSM and CMCSM; (e) Zeta potential of CSM, BMP-2/CSM, CMCSM and Bbr/CMCSM. As shown in Fig. 2a and b, both CS and CMCS are shaped into spheres. After 13

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

emulsification and crosslinking, the morphology of CSM and CMCSM gradually changes from polymer chains networks to scattered and compact microspheres with a few attached small flakes. The particle size is one of significant properties for microspheres. From the results obtained from particle size distribution analyzer, we can see that there is obvious difference of particle size between CSM and CMCSM. The particle size of CSM (Fig. 2c) distributes from 11.90 µm (D10) to 64.32 µm (D90) with median diameter 31.85 µm (D50). However, the size distribution of CMCSM is from 1.22 µm (D10) to 29.52 µm (D90) with median diameter 11.68 µm (D50) (Fig. 2d). Compared with that of CSM, the media diameter of CMCSM becomes apparently smaller. Zeta potential is a parameter that factually reflects the surface charge property of microspheres indirectly.26 As shown in Fig. 2e, the zeta potential of CSM and CMCSM is 19.1±1.2 mv and -42.1±1.9 mv, respectively. Moreover, the Zeta potential of CSM and CMCSM do not change apparently after they are loaded with BMP-2 and Bbr respectively. Whether microspheres are loaded with biological molecules is not the dominant factor influencing their surface charge property.

14

ACS Paragon Plus Environment

Page 14 of 40

Page 15 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

3.1.2 FT-IR analysis

Fig. 3 FT-IR spectra of CS, CMCS, CSM and CMCSM The main function groups of fabricated microspheres can be observed from above FT-IR spectra (Fig. 3). The bands at ~1423 cm-1, ~1382 cm-1, ~1089cm-1 and ~1031 cm-1 belong to the absorption of -CH2- groups and -C-O-C- stretching vibration38 that exist in the framework structure of CS polymer chains. Characteristic peaks at 1648 cm 1 and 1601 cm-1 are attributed to -NH2 absorption.22 The chemical −

crosslinking reaction is revealed by a new absorption peak at 1653cm-1 of -C=Nbond.39 The shifted amide absorption band at ~1567 cm-1 indicates residual free -NH2 groups still remain in CSM. The major function groups of CMCS have two typical broad bands at ~3436 cm-1 and ~2914 cm-1 due to N-H and O-H overlapped stretching vibrations and the characteristic absorption of -CH2-.40 They are similar to CS except for the -COOgroup whose strong peaks are observed at ~1416 cm-1 and ~1326 cm-1 corresponding to the symmetric stretching vibration.19 The broad and intense absorption band at 15

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1605 cm-1 represents the asymmetric stretching of carboxylate -CO-41, confirming carboxymethyl groups in the CMCS polymer chains. The peaks at ~2874 cm-1 and ~2938 cm-1 in CMCSM are corresponding to the introduced -CH2- of glutaraldehyde. The amide II related peak band overlapped at 1605 cm-1 becomes irregular and shifts to be 1597 cm-1 and also the intensity of amide III peak at 1326 cm-1 decreases,22 which implies potential effect of cross-linking on the structure of CMCS macromolecular chains. According to FT-IR analysis, there are abundant free function groups, i.e. amino groups and carboxymethyl groups in CSM and CMCSM, respectively. Therefore, the free function groups can be retained to form -NH3+ on the CSM and -COO- on the CMCSM respectively, which makes their surface charged oppositely. 3.1.3 Cell attachment, spreading and viability of CSM and CMCSM

Fig. 4 (I) SEM images (a-d) of cell attachment; (II) Trichromatic fluorescence staining graphs (e-h) of cell spreading (red: F-actin labeled by phalloidin; green: mitochondria labeled by the MitoTracker® probes or the autofluorescence of microspheres ; blue: nucleus labeled by Hoechst dyes); (III) Fluorescence images 16

ACS Paragon Plus Environment

Page 16 of 40

Page 17 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

(i-l) of living cells (green) growing with CSM and CMCSM after 1 day and 4 days. In general, the process of cell attachment to a substrate includes adsorption of protein serum, cell contact and cell spread. Attachment occurs in the first phase of material/cell construct interactions, which has an impact on the capacity of cells to proliferate on materials.42 It can be regulated by materials surface properties, including chemical composition topography and roughness, etc.43 At day 1, cells are closely attached and sparsely spread along the rough surfaces of CSM and CMCSM with rich of filopodia and typical osteoblastic spindle-like morphology (Fig. 4a-b). We also observe that the actin filaments (red fibers) mainly arrange along the outer edge of cells (Fig. 4e-f), suggesting many filopodia and extracellular matrix are formed at the early stage of cells attachment and spreading. The rich mitochondria in the cytoplasmic matrix (Fig. 4e-f) indicate potential cell viability to proliferate on microspheres. Moreover, few dead cells (red fluorescence) can be observed (Fig. 4i-j), which is a sign of the nontoxicity and good cytocompatibility of CSM and CMCSM. With the increase of incubation time, a considerable cell increase is observed (Fig. 4k-l), also suggesting excellent cells viability on CSM and CMCSM. At day 4, cells develop a denser meshwork of well-defined actin fibers and more spreading extracellular matrix,44 subsequently adhere to the microspheres (Fig. 4c-d). Furthermore, it is visible that more actin fibers are present (Fig. 4g-h) around the microspheres. Compared with spindle-like cell morphology at day 1, it seems that cells adhere to the CSM and CMCSM more tightly and some microspheres are almost completely wrapped by cells with larger lamellipodia and richer long microfilaments (Fig. 4g-h) due to higher attachment force.45 The longer cytoplasmic extensions together with larger lamellipodia imply excellent adhesion and colonization on the microspheres, which are crucial for cells to bridge pores and cover the porous structure of the gel. These results indicate that CSM and CMCSM allow osteoblast-like cells to settle on their surface and proliferate quickly, which is meaningful to further tissue regeneration. 17

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

3.1.4 Sustained release behavior and kinetics of BMP-2/CSM & Bbr/CMCSM

Fig. 5 (a) Swelling degree of CSM and CMCSM after reaching an equilibrium overnight; (b) In vitro cumulative release of Bbr and BMP-2 from CSM and CMCSM respectively in PBS solution (pH=7.4) for about 90 days; Release kinetics of Bbr (c) and BMP-2 (d) from CMCSM and CSM respectively before 20 days Swelling performance of CSM and CMCSM was tested in PBS solution (pH=7.4). The experimental results (shown in Fig. 5a) indicate that both CSM and CMCSM possess relatively higher value of swelling degree at 296±17% and 215± 5%, respectively. It has been reported that the molecular weight, degree of deacetylation, hydrophility and crosslinking degree may result in the difference of the swelling ratio.46 However, since we used the CS and CMCS with a similar range of deacetylation degree (DD≥90%) to fabricate the CSM and CMCSM, the different swelling ratio should be derived from the different molecular weight of CS (Mw=200000 g/mol) and CMCS (Mw=100000 g/mol), their different hydrophility 18

ACS Paragon Plus Environment

Page 18 of 40

Page 19 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

and crosslinking degree. Fig. 5b shows in vitro cumulative release of Bbr and BMP-2 from the two different kinds of microspheres. From release profiles, we can see both release of Bbr and BMP-2 reach a balance after the microspheres are immersed in PBS buffer (pH=7.4) for more than two months. At initial period, they exhibit a typical release since some molecules are located on the surface, but followed by a lag release. However, Bbr displays a burst release and about 40% adsorbed drugs release within initial 6 days. Compared to Bbr, BMP-2 shows lower drug release within 30 days, suggesting a controlled release of BMP-2. After 30 days, they exhibit a reverse release rate but a similar sustainable release behavior. At the 50th day, the release amount of Bbr and BMP-2 reach about 90%. Their drug release nearly levels off after this moment. Actually, when adsorbed drugs come in contact with the release medium, they instantaneously dissolve and release from the surface.47 This type of drug release leads to burst effect. Drug entrapped in the surface layer of particles also follows this mechanism.47 In order to understand the mechanism of release, the release kinetics of Bbr and BMP-2 are investigated based on the release profiles. The fitted equations of Bbr and BMP-2 release before 60% (Fig 5c and d) are Mt/M∞=12.85t0.52 (r2=0.9607) and Mt/M∞=9.10t0.62 (r2=0.9958), respectively. Since both of release exponent are between 0.43~0.85, their release mechanism is governed by the anomalous (non-Fickian) transport,48 suggesting firstly both BMP-2 and Bbr diffuse from the microspheres surface and then release from the swelling polymers network. The smaller kinetic rate constant of BMP-2 could indicate slower diffusion-controlled release compared with Bbr, which is coincident with the more well-controlled release behavior of BMP-2 by CSM at initial release period.

19

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

3.1.5 In vitro osteogenic study of BMP-2/CSM

Fig. 6 Normalized ALP activity (a) and OCN production (b) of BMSCs in control and osteogenic medium groups after 4, 7, 14 and 21 days. (* p