Cellulose NanocrystalsâBioactive Glass Hybrid Coating as Bone

Subscriber access provided by UNIV OF LETHBRIDGE

Article

Cellulose Nanocrystals—Bioactive Glass Hybrid Coating as Bone Substitutes by Electrophoretic Co-Deposition: In-situ Control of Mineralization of Bioactive Glass and Enhancement of Osteoblastic Performance Qiang Chen, Rosalina Pérez Garcia, Munoz Josemari, Uxua Pérez de Larraya, Nere Garmendia, Qingqing Yao, and Aldo R. Boccaccini ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.5b07294 • Publication Date (Web): 13 Oct 2015 Downloaded from http://pubs.acs.org on October 18, 2015

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 free 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 accessible to all readers and 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.

ACS Applied Materials & Interfaces 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 24

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

Cellulose Nanocrystals—Bioactive Glass Hybrid Coating as Bone Substitutes by Electrophoretic Co-deposition: In-situ Control of Mineralization of Bioactive Glass and Enhancement of Osteoblastic Performance Qiang Chena, Rosalina Pérez Garciab, Josemari Munozb, Uxua Pérez de Larrayac, Nere Garmendiac, Qingqing Yaod,*, Aldo R. Boccaccinia,* a. Institute of Biomaterials, Department of Materials Science and Engineering, University of Erlangen-Nuremberg, Cauerstrasse 6, 91058 Erlangen, Germany b. CIDETEC, Parque Tecnológico de Miramón, Paseo Miramón 196, 20009 San Sebastian, Spain MSc. Uxua Pérez de Larraya, Dr. Nere Garmendia c. CEMITEC, Materials Department, Polígono Mocholí, Plaza Cein 4, 31110 Noain, Navarra, Spain Dr. Qingqing Yao d. Institute of Advanced Materials for Nano-Bio Applications, School of Ophthalmology & Optometry, Wenzhou Medical University, Wenzhou, 270 Xueyuan Xi Road, Zhejiang 325027, China *Corresponding authors: [email protected] (Qingqing Yao), [email protected] (Aldo R. Boccaccini) † Electronic supplementary information (ESI) available.

Abstract: Surface functionalization of orthopedic implants is being intensively investigated to strengthen bone-to-implant contact and accelerate bone healing process. A hybrid coating, consisting of 45S5 bioactive glass (BG) individually wrapped and interconnected with fibrous cellulose nanocrystals (CNCs), is deposited on 316L stainless steel from aqueous suspension by a one-step electrophoretic deposition (EPD) process. Apart from the co-deposition mechanism elucidated by means of zeta-potential and SEM measurements, in-vitro characterization of the deposited CNCs-BG coating in simulated body fluid reveals an extremely rapid mineralization of BG particles on the coating, e.g. the formation of hydroxyapatite crystals layer after 0.5 day. A series of comparative trials and characterization methods were carried out to comprehensively understand the mineralization process of BG interacting with CNCs. Furthermore, key factors for satisfying the applicability of an implant coating such as coating composition, surface topography and adhesion strength were quantitatively investigated as a function of mineralization time. Cell culture studies (using 1

ACS Paragon Plus Environment


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 2 of 24

MC3T3-E1) indicate that the presence of CNCs-BG coating substantially accelerated cell attachment, spreading, proliferation, differentiation as well as mineralization of extracellular matrix. This study has confirmed the capability of CNCs to enhance and regulate the mineralization of BG particles leading to mineralized CNCs-BG hybrids for improved bone implant coatings. Keywords: hybrid materials; nanocellulose; bioactive glass; electrophoresis; guided mineralization.

1. Introduction Metallic devices (based on stainless steel and Ti alloys) have been utilized for decades for fixation or replacement of bone defects in orthopedic and dental applications.[1] In spite of notable clinical outcomes, prostheses made of those bio-inert materials possess an overall lifetime of approximate 15 years [2], e.g. secondary surgery is commonly required to replace up to 10% of all hip prostheses after a decade of implantation because of aseptic loosening in-situ.[3] To prevent those negative effects, modification of implant surfaces with bioactive materials and antimicrobial agents is being intensively investigated to improve bone-to-implant contact and reduce risks associated with infections and surface corrosion.[4-6] Bioactive glasses (BG) have been demonstrated to be promising bone substitute materials due to their superior bone-bonding ability, biodegradability and stimulatory effect on osteoblast cell activity through a combination of surface reactivity and ion release during the degradation process.[7, 8] In comparison with conventional

bioceramic

coating

techniques

associated

with

high

temperature

consolidation,

room-temperature processing of composite coatings consisting of a polymer matrix embedding BG particles as the bioactive filler is considered to be an effective alternative.[9] Addition of a polymer phase will not only strengthen the porous inorganic coating structure and the bonding to the surface of the implant, but it also plays a role in retarding the ion release rate from BG particles embedded in the polymer matrix.[10] Moreover, functional groups such as amino, carboxyl and hydroxyl groups introduced by polymer molecules could be preserved at room-temperature to facilitate the entrapment of biological entities for the biofunctionalization of metallic implants.[11, 12] In addition, the presence of the polymer component leads to a ‘soft’ coating, which, in comparison to hard coatings composed of pure ceramics, can be expected to promote improved mechanical contact between the rigid metal implant and the compliant bone tissue. Cellulose is well-known as a renewable and environmental friendly material, representing the most abundant polymer resource on Earth.[13] In recent years, nanocellulose materials, such as nanoﬁbers and 2


Page 3 of 24

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


nanocrystals (also called nanowhiskers) extracted from natural cellulosic sources, are rigorously investigated due to their unique physical and chemical properties.[14] Cellulose nanocrystals (CNCs) have attracted significant interest in biomedical applications because of their abundance, low-density, nanoscale dimensions, high aspect ratio, and especially impressive mechanical properties, e.g. CNCs possess a Young’s modulus in the range of 100-300 GPa.[15-17] The extraction of CNCs involves simply two steps, purification of the cellulose component and, isolation of the nanocrystalline parts by chemical or mechanical treatments.[18] Furthermore, it is of great interest that CNCs can be chemically modified by attachting small molecules or functional groups due to the existence of ample hydroxyl groups on the surface, which is beneficial not only to their stable dispersion in aqueous solvent but also to intensively explore their biomedical applications from mechanical reinforcing nanofillers to functional additives, e.g. application as drug delivery carriers.[17, 19-21] When CNCs are isolated by hydrolysis with sulfuric acid, sulfate groups can be grafted on the surface of nanoscrystals by esterification. The availability of such anionic CNCs suspensions therefore motivated us to develop CNCs-based biomaterials (coatings) by electrophoresis, which was demonstrated for the first time in our previous investigation.[22] Electrophoretic deposition (EPD) is a simple, rapid and versatile coating technique that involves the movement of charged particles/molecules under an appropriate electric field, leading to their consolidation on the oppositely charged electrode to form films and coatings with high microstructural homogeneity and tailored thickness.[23, 24] With the application of EPD, CNCs based nanostructures could be directly deposited from well-dispersed colloidal suspensions. Therefore, unexpected aggregation of CNCs during the redispersion and drying processes could be avoided and high-quality CNCs nanocomposites with significant mechanical reinforcement and reproducibility could be anticipated. Recent studies have highlighted the interest of developing composite materials based on biodegradable polymers and bioceramics to mimic the function of natural bone, which represents a composite structure comprising of fibrous collagen and hydroxyapatite crystals.[25, 26] In the case of the potential combination of CNCs with BG, as investigated in this study, we considered for the first time the use of electrophoretic co-deposition to fabricate CNCs-BG hybrids as potential bone substitutes. CNCs can mimic the structure of fibrous collagen in bone and BG, which presents excellent hydroxyapatite (HA) forming ability, and is designed as the bioactive fillers promoting bone-like HA formation. The co-deposition mechanism was experimentally analyzed by means of zeta-potential measurements and SEM observations. The mineralization of BG particles in simulated body fluid (SBF), influenced by their interactions with CNCs, 3



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

was experimentally investigated by means of a combination of characterization techniques. Finally systematic in-vitro cellular studies were conducted to characterize the osteoblastic performance on the deposited CNCs-BG coatings. 2. Experimental 2.1 Materials Bioactive glass (BG) powder with 45S5 composition (in wt.%): 45% SiO2, 24.5% CaO, 24.5% Na2O and 6% P2O5 and average particle size of 2.0 µm was provided by Schott AG (Germany). Microcrystalline cellulose was purchased from Sigma-Aldrich (Steinheim, Germany). CNCs were extracted from microcrystalline cellulose according to a sulfuric acid hydrolysis process, which is schematically described in Figure S1. It should be mentioned that the obtained CNCs are partially negatively charged (grafted with sulfate groups). 2.2 EPD process The stock CNCs suspension was subjected to ultrasound probe treatment (Branson Sonifier® S250D, USA, time: 3 min, amplitude: 30%) before use. De-ionized water (Purelas option ELGA DV25, ≤0.67 µS/cm) was used as solvent. BG powder was dispersed into CNCs suspension by 5 min magnetic stirring and 3 min sonication bath treatment (BANDELIN RK100, Germany). The final concentrations of CNCs and BG were 1 and 10 mg/ml, respectively. The EPD cell includes two parallel 316L stainless steel (SS, 30×15×0.2 mm3) plates as the deposition and counter electrodes. The deposition area was fixed at 20×15 mm2 with the distance of 10 mm. According to a trial-and-error study, the optimal applied voltage and deposition time were determined as 10 V and 60 s, respectively. After deposition, the coated electrode (anode) was gently removed from the suspension and dried at 37 ºC for 24 h before further characterization. 2.3 Zeta potential measurements The surface charge of BG particles in the presence of CNCs was evaluated by zeta potential measurement (Zetasizer nano ZS equipment, Malvern Instruments, UK) based on laser Doppler velocimetry (LDV) technique. CNCs was added into aqueous BG suspension (0.1 mg/ml) and thoroughly mixed by sonication bath and magnetic stirring (10 min each). The zeta potential values of BG particles with different CNCs/BG ratios (w/w) were measured. After applying the electric field to an U-model cell containing the test suspension, the instrument carried out continuous measurements (10-100 times) and stopped when the obtained results were stable. 4


Page 4 of 24

Page 5 of 24

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


2.4 In-vitro mineralization studies 2.4.1 Microstructure The as-prepared coatings were soaked in 50 ml of SBF prepared according to the literature,[27] and placed in an orbital shaker at 37 ºC at an agitation rate of 90 rpm. The incubation times were 0.5, 1, 2, 4, 7, 14 days and the SBF solution was refreshed every two days. Three samples per each time point were recorded. Then the samples were removed, gently rinsed with de-ionized water for three times and dried at 37 ºC. As a control surface in the absence of CNCs, BG pellets produced by die pressing and subsequent thermal treatment at 400 ºC for 30 min, were also soaked in SBF with identical incubation conditions. The disintegration of BG pellets in SBF could be prevented and crystallization of BG was not expected to occur during thermal treatment at 400 ºC.[28] SEM (model Auriga, Zeiss) and energy dispersive spectroscopy (EDS, X-MaxN Oxford Instruments, UK) were used to analyze the morphology (surface and cross-section) and elemental composition of the samples during mineralization. The wettability/porosity of the incubated samples was evaluated by water contact angle measurement (DSA30, Kruess GmbH, Germany). Five parallel tests were performed for each coating condition. 2.4.2 Composition The coating weight (mg/cm2) as a function of incubation time was recorded by an analytical balance (0.1 mg). X-ray diffraction (XRD) analysis (D8 Philips X’Pert PW 3040 MPD) was performed to investigate the phase transformation, i.e. hydroxyapatite formation on coating surfaces, for different incubation times. The content of CNCs in the coating before and after 14 d of SBF incubation was determined by means of thermogravimetric analysis (TG, TA INSTRUMENTS, Q500) with a heating rate of 5 °C/min until 800 °C in air. 2.4 3 D topography 3D topographical features of the deposited coatings were quantitatively assessed using a confocal microscope (Leica DCM 3D, Germany). An objective EPI 20X-L with 1 µm spatial pinhole was used to detect an image area of 4×4 mm2. The 3D structure was reconstructed by surface imaging and metrology software with a Gaussian filter of 0.8 mm (Leica Map 6.0, Leica, Germany). The measurement also allowed to assess the root mean square height, corresponding to the standard deviation of the height distribution, indicating the roughness distribution on the surface. 2.4.4 Ion release study 5



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

Ion release profiles were studied by means of Inductively Coupled Plasma Emission Spectrometer (ICP-ES, ICPE9000, Shimadzu, Japan). The samples were soaked in 50 ml of ultrapure water (Model Elix 3 from Millipore) or SBF and placed in an orbital shaker at 37 ºC with an agitation rate of 90 rpm. CNCs-coated BG pellets were produced by dipping in a CNCs suspension (1 mg/ml) for 15s followed by drying at 37 ºC for 24 h. The uncoated and coated BG pellets were incubated in SBF with the same solid to liquid ratio (w/v) and incubation conditions. After 5, 10, 30, 60, 120, 240, 360, 480, 1440, 2880 min, 1 ml of the solution was extracted and filtrated immediately with 0.22 µm filter to remove potential solid precipitation from the solution. Absorption wavelengths used for determination of Si, P, Ca were selected according to the literature.[29] Calibration curves were obtained before each measurement by preparing standard solutions containing Si, P and Ca (dilutions of 1000 ppm solutions by Spex-CertiPrep). Three replicates were measured for each element condition. 2.4.5 Coating-substrate adhesion The adhesion strength of the coatings to the substrates as a function of incubation time in SBF was measured using a scratch tester (CSM instruments, Revetest®, Switzerland). In a typical test, a diamond indenter with a tip radius of 22 µm is drawn across the coated surface under a linearly increasing load. A scratch length of 5 mm, scratch speed of 10 mm/min and a load range of 0.09–20 N were used for the measurement. Due to the insignificant response from acoustic emission signals, the load at which failures were first observed to occur along the scratch track was determined as the critical load, which was established with the help of a combined optical microscope. At least seven parallel tests were performed for each coating condition. 2.5 Cellular Studies 2.5.1 Cell viability MC3T3-E1 (pre-osteoblast) cells were cultured in a modification of Eagle’s minimum essential medium (Gibco, Invitrogen, USA), and the medium was refreshed every two days. It should be noted that no pure BG coatings could be deposited in the absence of CNCs, and the surface structure of a CNCs coating control produced by EPD is significantly different from that of CNCs-BG coatings. Therefore, uncoated 316L SS substrates were finally chosen as the control so that the possible effects arising from both the coating components and microstructure could be assessed. Cell proliferation was determined using cell counting Kit-8 assay, which is based on the water-soluble disulfonated tetrazolium salt as introduced in the literature 6


Page 6 of 24

Page 7 of 24

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


[30] (CCK8; Dojindo, Japan). MC3T3-E1 cells were seeded on the sterilized samples (1×1 cm2) with 1×104 cells/well. After culturing, 20 µl of CCK-8 reagent was added per well, and the cells were incubated for additional 2h. The results were expressed as the mean absorbance (optical density at 450 nm) of six parallel tests of coated vs. bare substrates. A Live-Dead staining using calcein AM/PI (Sigma, Germany) was performed for a direct observation of cells on samples. After culturing, cells were labeled with a freshly calcein AM/PI solution for 15 min at 37 ºC, then rinsed with PBS for three times and observed under fluorescence microscope (IX 2-UCB, Olympus, Germany). The chemical composition of the mineralized layer was detected by EDS analysis. 2.5.2 Cell morphology and cytoskeleton 4×104 cells per well were seeded onto the samples in a 24-well plate and cultured for 1 and 4 d. Then the samples were rinsed with PBS and fixed in 2.5% glutaraldehyde for 3 h at 4 ºC, and subsequently dehydrated in ethanol solutions of varying concentrations (30, 50, 70, 90 and 100%) for 15 min, respectively. The morphology of the cells was observed by SEM. The cytoskeleton organization of MC3T3-E1 was analyzed using Filamentous actins (F-actin) staining. The sample was placed into a 24-well plate and then covered with a glass coverslip. 1×104 cell/ml of MC3T3-E1 cells were seeded per well. After 8 h and 24 h of culture, cells were fixed with 4% paraformaldehyde for 10 min, permeabilized with 0.1% Triton X-100 for 5 min, blocked with 1% bovine serum albumin for 20 min, stained with Rhodamine Phalloidin (Sigma, USA) for 20 min and Hoechst 33258 (Beyondtime Bio-Tech, China) for 5 min in the dark. The stained MC3T3-E1 cells were viewed under laser confocal microscopy (LSM 710, Zeiss, Germany). 2.5.3 ALP activity and mineral nodules staining Cell differentiation was assessed by means of alkaline phosphatase (ALP) activity using an ALP assay kit (Beyondtime Bio-Tech, China). 5×103 cells per well were seeded onto the samples and cultured for 7 and 14 days. Samples were rinsed with PBS and lysed with 0.1% Trion X-100 solution for 30 min at 4 °C The solution was then collected and centrifuged at 3000 rpm for 2 min at 4 °C. 50 µl of the obtained supernatant was mixed with 50 µl chromogenic substrate and cultured for 30 min at 37 °C. The reaction was stopped by adding 100 µl terminal liquid. ALP activity was measured at 405 nm using a microplate reader. For the cell mineralization assessment, 5×103 cells were seeded per well and cultured for 21 d. Samples were then rinsed with PBS for 3 times, fixed in ethanol for 10 min, then stained with Alizarin red-S (pH 4.2) for 10 min and 7



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

finally rinsed with de-ionized water to remove unreacted Alizarin red-S. The possible precipitation of mineral nodules was observed by optical microscope. 2.6 Statistical analysis Statistical tests in cell tests (between uncoated and coated samples) were carried out by SPSS 16.0 software based on the one-way ANOVA. All results are expressed as the mean ± SD. P10. The zeta potential values of BG particles with different CNCs/BG ratios (w/w) are shown in Figure 1(a), representing the possible interaction between BG particles and CNCs. The zeta potential of CNCs suspension is −27±1 mV (dash line in Figure. 1(a)), while the zeta potential of BG particles increases (in absolute values) from −24±3 mV (bare BG) to −35±7 mV (CNCs/BG ratio of 0.04). The higher absolute values of zeta potential of BG particles in CNCs-containing suspensions is probably a consequence of the negatively charged CNCs adsorbed on the surface of BG driven by the hydrogen bond between hydroxyl groups of CNCs and BG. With the continuous increase of CNCs/BG ratio from 0.04 to 2, all samples present similar zeta potential values in the range of −35 ~ −31 mV. Therefore, it can be concluded that BG particles in the present EPD suspension (CNCs/BG ratio of 0.1:1) were thoroughly adsorbed by CNCs. A possible model describing the interaction between BG particles and CNCs in the EPD suspension, and their co-deposition under an applied electric field, is schematically shown in Figure 1(b). The co-deposition 8


Page 8 of 24

Page 9 of 24

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


mechanism of BG particles with CNCs can be assigned to two separate processes: (i) the CNCs are adsorbed on the surface of BG particles and incorporated in the deposit accompanying them, and (ii) the non-adsorbed CNCs migrate and incorporate in the deposit separately.

Figure 2. Surface (a,b) and cross-sectional (c, d) SEM images of the deposited CNCs-BG coatings.

Surface and cross-sectional SEM images of the deposited coatings are shown in Figure 2. The particles with diameter of around 1~3 µm are assigned to BG particles. As inferred from Figure 2(a), BG particles appear to be firmly bridged by a polymer phase (a typical “bridge” between particles has been arrowed in Fig. 2(a)) although the deposited coating presents a particle-dominated structure. The surface of single particles (Figure 2(b)) is seen to be completely wrapped with a fibrous mesh. In addition, fairly uniformly distributed nanopores are observed on the surface. A homogeneous coating structure is observed from images of both surfaces (Figure S2) and cross-sections (Figure 2(c)) with uniform thickness of ~28 µm. Moreover, it is found in Figure 2(d) that BG particle inside the coating are also masked by fibrous layers. Therefore, it can be concluded that the surface of all deposited BG particles are individually covered with CNCs meshes, which is probably due to the pre-adsorption of CNCs in suspension. This phenomenon is also a visual confirmation of the hypothesized EPD mechanism explained above (Figure 1b). The micropores in the form of gaps between BG particles are found throughout the coating, suggesting a highly porous microstructure. 3.2 In-vitro mineralization study The characterization of apatite formation on the surface of samples in SBF has been established as a 9



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

convenient alternative to predict their bioactive potential.[27] Bioactive glasses are well-known for their rapid HA forming ability, however, some bioactive glasses exhibit a faster dissolution rate than that required for the remodeling of human bone.[31, 32] Therefore, it is essential to improve the mineralization process (bioreactivity) of bioactive glasses, e.g. by in-situ depositing a biopolymer phase with specific functional groups and structures. Such knowledge will lead to novel bioactive materials with controlled topography and ion release properties, which is particularly promising when applying ion-doped bioactive glasses on stimulating biological performances involving osteogenesis, angiogenesis and antimicrobial applications.[8, 33] On the other hand, the in vitro evaluation of composition, microstructure as well as mechanical stability of the deposited CNCs-BG composite, is essential to understand cell-material interactions and the suitability of the prepared coatings for intended orthopedic and dental applications. Figure S3 shows the coating weight and phase composition as a function of immersion time in SBF. Significant weight loss is found after 0.5 d of incubation due to the degradation of BG. However, the coating weight increases significantly with increasing incubation time that, suggests a continuous mineralization stage.[34] According to XRD measurements, the diffraction peak at 22.6 º is typical for cellulose I crystals indicating the presence of CNCs.[35] The characteristic peaks of HA are detected in all incubated samples, and the relative intensity of HA peaks is increased with longer incubation times. The experimental results from both measurements indicate therefore a rapid and continuous precipitation of HA phase during SBF incubation.

Figure 3. Surface SEM images of BG pellets (used as reference) and CNCs-BG coatings as a function of incubation time in SBF, the inset (CNCs-BG coating after 3 h of incubation) shows the elemental distribution of the marked (circled) region. 10


Page 10 of 24

Page 11 of 24

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 surface morphology of bare BG pellets (control) and CNCs-wrapped BG particles as a function of incubation time in SBF is shown Figure 3. From SEM images of bare BG pellets, a few clusters of spherical particles have grown on the surface after 0.5 d of incubation, which is probably caused by the precipitation of calcium phosphate phase.[36] The BG particle is completely covered with a layer of spherical nanoparticles after 1 d, and an enhanced precipitation of calcium phosphate phase was obtained with continuous incubation. Moreover, crystallization of calcium phosphate phase in terms of the formation of plate-like HA appears after 7 d, and a continuous and highly crystalline HA layer is presented after 14 d. In contrast, a more rapid mineralization process is observed on CNCs-wrapped BG particles. When observing the surface of individual BG particles, a fabric-like structure with randomly distributed nanopores has appeared after 3 h. The fabric structure seems to be more dissociative (compared to Figure 2b/d), which is likely due to the relatively rapid degradation of BG particles. In addition, EDS analysis of the annular region on the fabric structure exhibits an atomic Ca/P ratio of 1.76, which is similar to the Ca/P ratio (1.67) of stoichiometric HA (Ca10(PO4)6(OH)2). Therefore, it is considered that the obtained fabric structure is likely the result of the precipitation and preferential growth of HA phase along the CNCs mesh. The BG particle is completely covered with a porous calcium phosphate layer after 6 h. More interestingly, a highly crystallized HA layer is observed after only 0.5 d, which is comparable to the morphology of the incubated BG pellet after 14 d. The results of the comparative experiments indicate that the presence of CNCs mesh significantly accelerates the in-situ mineralization/crystallization process of BG particles.

11



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

Figure. 4. Ion release profiles of (a) CNCs-BG coatings in ultrapure water and SBF, and (b) BG pellets with and without CNCs coating, (c) schematic model describing the mineralization process of the CNCs-wrapped BG particles. Error bars represent means ± SD for n=3. Ion release profiles of the deposited CNCs-BG coating in ultrapure water and SBF are plotted in Figure 4(a). The concentration of silicon, phosphorous, and calcium increases rapidly during the first 8 hours in ultrapure water. Then an appreciable increase of silicon concentration indicates a continuous but much slower degradation rate of BG, while the concentration of phosphorous and calcium remains constant during incubation. A similar silicon ion release is found in SBF, indicating the similar degradation rate of BG in both ultrapure water and SBF. However, the concentration of both phosphorous and calcium reduced continuously during the whole incubation period, which can be assigned to the formation of a calcium phosphate phase compensating the ion products from BG. Ion release studies of BG pellets with and without CNCs coating is shown in Figure 4(b). It is found that the coated sample showed a higher concentration of silicon and lower concentration of phosphorous compared with the uncoated sample, suggesting a more rapid degradation and mineralization processes occurring on CNCs-coated samples. It should be noted that the uncoated sample presents a gradually decreasing concentration of calcium during the first 6 hours, while the coated sample 12


Page 12 of 24

Page 13 of 24

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


exhibits a more significant reduction of calcium concentration at the same time, indicating a preferential precipitation of calcium ions on the CNCs-coated group. According to the abovementioned experimental results, a possible model describing the mineralization process of BG, in-situ wrapped with a porous CNCs layer, is schematically shown in Figure 4(c). The mineralization mechanism of bare 45S5 BG has been well established by Hench,[34] which starts with the formation of a silica-rich gel, followed by ion-exchange induced precipitation of an amorphous CaO– P2O5-rich film and further crystallization to a carbonated hydroxyapatite phase. Compared to the uncoated BG particle, a nanoporous CNCs layer will not retard the transportation of ion dissolution products from BG particles but it will provide a more hydrophilic surface for water absorption and ion transportation. The accelerated mineralization of BG is likely realized by two approaches. On the one hand, the specific area of the coated BG particle is increased, providing more anchoring points for the precipitation of calcium phosphate phase. On the other hand, the anionic sulfate groups (described as green patterns Figure S1) grafted on the CNCs can assist in capturing cationic minerals. For example the significantly reduced concentration of calcium ions (cationic) at the first 6 h is likely due to a prior attachment of calcium ions on CNCs. Those calcium ions could serve as the mineralization nuclei, leading to a preferential precipitation of calcium phosphate along the fiber mesh, as evidenced by SEM images in Figure 3. It has been reported that carboxyl groups on carbon nanotubes were capable to nucleate HA under ambient and physiological temperature in SBF.[37] A bacterial cellulose pellicle pretreated by soaking in a Ca-containing solution could also trigger HA precipitation in a 1.5× SBF.[38] A more relevant example with respect to the present study is a fibrous poly-DL-lactide coating (diameter of 100-200 nm) electrospun on the surface of bulk BG pellets, which presented a full coverage of PDLLA fibers with mineralized HA nanocrystals after 14 days incubation in SBF.[39] It is therefore concluded that an in-situ wrapping of individual BG particles with CNCs meshes, without pretreatment of raw materials or long-term incubation, is capable of inducing remarkably accelerated mineralization (less than 1 day) and guided growth of the HA phase. Apart from the morphological evolution of individual BG particles, Figure S4 shows the surface topography of the bulk CNCs-BG coating during mineralization. The as-deposited CNCs-BG coating shows a relative smooth surface exhibiting a Ra of 0.731±0.07 µm. A minor increase of Ra is obtained with the increase of incubation time, however, an increasing amount of needle-like pattern is detected on the incubated surface. It has been reported that the increased surface roughness of HA layers can effectively 13



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

promote initial osteoblastic cell adhesion and proliferation directly via enhanced formation of focal contacts or indirectly through selective adsorption of serum proteins specific for cell attachment.[40] Moreover, by applying a precise control of surface nanopatterns, human foreskin fibroblasts exhibited significantly smaller cell size and lower proliferation on needle-like nanoposts.[41] Therefore, it is expected that the topographical control of both individual BG particles and the bulk coating surface during mineralization, will offer a straightforward approach to improve the osteoconductivity of coatings for orthopedic applications.

Figure 5. (a-c) Cross-sectional SEM images and (d) water contact angles of CNCs-BG coatings as a function of incubation time in SBF: (a) 0.5 d, (b) 1 d, (c) 7 d (low and high magnification), (d) the inset shows the surface of the as-prepared CNCs-BG coating immediately after contact angle measurement. Error bars represent means ± SD for n=7. Figure 5(a-c) show cross-sectional SEM images of CNCs-BG coatings in order to visualize their internal microstructure. A drastically densified cross-section is observed after 1 d. Day 7 sample exhibits a rather compact and uniform microstructure. Furthermore, well-distributed nanowhiskers are found in the matrix of the newly formed HA phase (indicated with arrows). Figure 5(d) shows the water contact angles of the incubated coatings which can be conveniently used to determine the progress of the coating densification process. The contact angle of the as-prepared coating could not be measured due to the immediate and complete infiltration of water droplet into the entire area of the coating, as shown by the inset image in Figure 5(d), indicating the highly porous structure obtained. As incubation time increases, probably due to the reduced porosity, the incubated samples exhibit contact angles of 6.7±0.5 ° and 7.3±1.2 ° after 1 and 2 d, 14


Page 14 of 24

Page 15 of 24

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


respectively. It seems that the coating is completely densified after 4 d resulting in a constant contact angle value measured even after 14 d. The densified hybrid coating after mineralization exhibits a moderate contact angle (~50 °), which is expected to be suitable to promote the interaction with surrounding tissue when applied on a metallic implant [42, 43]. The content of CNCs in the coating is quantitatively determined by a combination of coating weight and TG measurements, as shown in Table 1. The net weight values of CNCs, obtained by multiplication of coating weight and percentage of CNCs, are 0.24 and 0.20 mg/cm2 in the two samples, respectively. Thus most of the CNCs in the coating is preserved after mineralization, which is important for the role of CNCs as the reinforcing phase in the mineralized coatings. Table 1. Percentage and net weight of CNCs in the coatings before and after 14 days of SBF incubation Coating condition Coating weight Percentage of CNCs Net CNCs weight

As-prepared Day 14 in SBF

(mg/cm2)

(wt %)

(mg/cm2)

1.18±0.004

20.4

0.24

1.77±0.32

11.1

0.20

The adhesion strength of the CNCs-BG coating to the substrate was quantitatively evaluated by scratch test. Figure 6(a) shows the critical load values of the CNCs-BG coating during mineralization. A typical scratch line on day 4 sample is shown in Figure 6(b). The cross on the scratch line, standing for the position of the first detachment under the linearly increasing load, is used to determine the critical load in this study. An appreciable increase of the critical load, probably due to the densification process, is observed during the first 4 d. However, no significant difference in critical load is found among all incubated samples, suggesting an stable adhesion strength of the deposited CNCs-BG coating during mineralization. Figure 6(c-d) shows SEM images of detached areas on the scratch line from a day 14 sample. A thin polymer layer is found in direct contact with the metallic substrate, which is likely the result of the CNCs phase depositing firstly, e.g. before deposition of CNCs-wrapped BG particles. Furthermore, it is likely that this intermediate polymer layer can act as a binder strengthening the connection between the coating and the substrate, leading to the stable adhesion strength of the mineralized coatings. This behavior seems to be confirmed by observation of Figure 6(d) which shows that the mineralized coating could not detach from the substrate due to the presence of the polymer layer. The critical load of the prepared CNCs-BG coating at different mineralization 15



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

stages is higher than that of an EPD-produced calcium phosphate/chitosan coating (~6 N) reported in literature [44]. However, it seems that coatings which are chemically bonded to the metallic substrate always present relatively high adhesion strengths. For example a suspension plasma sprayed HA coating shows a critical load of around 10−12 N [45], and a biomimetic-assembled HA coating on titanium substrates presents critical loads as high as 13.1 N [46]. It should be noted that in the present study the CNCs-BG coating has been only slightly detached under the critical loads (shown in Figure 6(b)). In addition, the adhesion strength of CNCs-BG coatings is expected to be further enhanced by increasing the relative content of CNCs (considered as a binder) and the substrate roughness for a better mechanical interlocking between the coating and the substrate.

Figure 6. (a) Values of critical load for CNCs-BG coatings as a function of incubation time in SBF, (b) typical scratch line on a day 4 sample (the cross shows the occurrence of the first detachment), (c-d) SEM images of the detached area on the scratch line of a day 14 sample. Error bars represent means ± SD for n=7. 3.3 Cellular studies Cell viability cultured on bare and CNCs-BG coated 316L SS substrates is quantitatively studied by means of CCK-8 assay, as shown in Figure 7. Both samples exhibited comparable cell viability after 1 and 2 d, while significantly higher cell viability (P

Cellulose NanocrystalsâBioactive Glass Hybrid Coating as Bone

Recommend Documents

Cellulose NanocrystalsâBioactive Glass Hybrid Coating as Bone