Effective incorporation of rhBMP-2 on implantable titanium disks with microstructures by using electrostatic spraying deposition

Abstract

Incorporation of bioactive molecules, such as bone morphogenetic proteins (rhBMP-2) is an effective way to improve the surface bioactivity and then enhance the osseointegration of the metallic implants. However, preserving/enhancing the osteogenic capacity of rhBMP-2 is still one of the greatest challenges. In this study, electrostatic spraying deposition was applied to construct a biodegradable chitosan coating loaded with rhBMP-2 on hydrophilic SLA-treated titanium disks. A series of analytical tools including scanning electron microscopy, atomic force microscopy, and contact angle measurements were used to characterize the physical/chemical properties of the coatings. The release behaviour of model proteins were compared and regulated for different surfaces. Biological evaluation in terms of cell adhesion, proliferation, and differentiation was done to study the effects of the coating materials/structure on the osteoblast response. In vitro experiments demonstrated the controlled release of proteins from the ES-P-SLA disks and showed that the released rate/concentration could be regulated by the loading amount and the spraying time. Microscopic visualization of the myblast morphology on the ES-P-SLA disks exhibited enhanced cellular adhesion at the initial incubation. MTT testing and ALP activity results confirmed the enhanced proliferation and differentiation of rhBMP-2 over a two-week period. Therefore, we believe the combination of coating properties/rhBMP-2 bioactivity with the surface topography will speed up bone-formation and improve the implant osseointegration. The key technologies developed in this study could be applied for other biomolecules, undoubtedly benefiting the healthcare sector and quality of life.

---

Introduction

Biomedical titanium and titanium alloys are very reliable replacement materials for orthopedic/dental implants because of the excellent mechanical properties, biocompatibility, and corrosion resistance.^1–3 However, the initial instability after implantation and low bioactivity leading to slow healing, poor osseointegration, and ultimately implant failure, are current major issues for medical applications.^4,5 To improve the surface bioactivity and facilitate the growth of bone tissues on/into the Ti surfaces, many strategies have been utilized to functionalize the surfaces via tuning the physical/chemical properties,⁶ creating micro-/nano-structures,⁷ and incorporating bioactive molecules.⁸ The modification of physical/chemical properties allowed the limited improvement for biocompatibility and bone binding capacity on the metal surfaces. Recently, many researchers have focused on the introduction of the bioactive molecules in extracellular matrix components to the implant surface, leading to actively stronger osteoblast adhesion and higher osseointegration rate. These components usually include ECM peptides/polypeptides, cytokines, growth factors, and DNAs.^9–12 In the view of the osteoinductive potential by the stimulation of growth factors, bone morphogenetic proteins (BMPs) have been integrated on/into the implant surfaces to significantly enhance a wide variety of biological functions including stimulating osteogenesis and angiogenesis.^13–15 Upregulation of BMP-2 has been observed during the first three weeks of osteogenesis.¹⁶ Moreover, the growth factors released during the inflammatory phase have the potential of attracting undifferentiated mesenchymal stem cells to the injured site.¹⁷ On the basis of these findings, BMP-2 must be delivered in a sustained fashion that emulates the natural release profile of BMP-2 in vivo.

BMP-2 molecules can be incorporate on/into metallic implants either by direct adsorption or covalent binding. However, the direct adsorption shows the limited load efficiency, short retention time, and inconsistent release profile with significant burst release characteristics.¹⁸ Meanwhile, the covalent binding usually requires non-physiological conditions and complex procedures.¹⁹ Many researchers have also developed layer-by-layer self-assembly and biomimetic/electrochemical deposition approaches for immobilizing growth factors.^20,21 A variety of biomaterials including calcium phosphates, collagen gels, sponges, hyaluronic acid, dextran, chitosan, etc. have been utilized as the loading materials for prolonged and local release of BMPs.^22–24 However, either time consuming or poor controlling of the coating properties, or low loading efficiency limit their applications.

To achieve the high loading efficiency and maintain the biological stability of BMP-2, the incorporation techniques require the fast operation, mild conditions, as well as the well-controlled coating. Among different techniques, the electrostatic spray deposition (ESD) has shown the great potential for biomolecule incorporation on/into metallic materials because of simple and low-cost, fast deposition rate, protein-friendly, well-controlled, and easy coating for complex geometries.²⁵ To date, this technology has been successfully applied to deposit many inorganic/organic materials on the metallic surfaces. For example, a 90 nm thick co-deposition of collagen and calcium phosphates coating electrosprayed on Ti substrate showed high interface stability and high coating retention, enhancing osteoblast differentiation and improving mineral deposition.²⁶ However, there is no investigation for co-deposition of biomaterials and rhBMP-2 on metallic substrates via the ESD technique. In this work, micro/nano-scale porous structures were first constructed on pristine Ti disks by the widely established sandblasting and acid etching (SLA) technique. After plasma treatment, the chitosan solution containing rhBMP-2 was electro-sprayed on the hydrophilic disks. The surface morphology and release performances of the degradable coating were investigated and controlled via tuning the material components and the coating parameters. Biological evaluation in terms of cellular adhesion, proliferation, and differentiation was done to study the effects of coating materials (rhBMP-2)/structure on the osteoblast response.

Experimental

Materials

Biodegradable chitosan (M_n: 500 000 g mol⁻¹, deacetylation degree 80–90%), Bovine Serum Albumin (BSA) were purchased from Sigma Aldrich (Milwaukee, WI). Growth factor, rhBMP-2 was donated by Rebone Biomaterial Co. Ltd (Shanghai, China). Commercially pristine Ti disks were obtained from Wego Inc. (Shangdong, China). C2C12 cells were purchased from the American Type Culture Collection (ATCC). The high-glucose Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS), and antibiotics (100 U mL⁻¹ penicillin, and 100 μg mL⁻¹ streptomycin) were purchased from GIBCO (Grand Island, NY). All other reagents, unless specified, were used without further purification.

Creation of micro-structures and hydrophilic modification

Pristine Ti disks with a diameter of 10 mm were first treated by the SLA technique, which has been introduced elsewhere.²⁷ Briefly, the Ti disks were sandblasted with 250–500 μm alumina particles, followed by acid etching in HCl/H₂SO₄ solution. After SLA treatment, macro-roughness and micro pits were created on the Ti disks. To turn the hydrophilicity of the SLA-treated surfaces, the oxygen-plasma treatment was carried out at 10.0 KHz using a plasma generator (CTP-2000 K; Nanjing Suman, China). Fig. 1 showed the process schematic for surface modification of Ti disks. Before use, the plasma chamber was cleaned with 2-propanol, dried, and further cleaned using a 20 sccm oxygen plasma at 50 W for 5 min. The SLA-Ti disks were then placed in the reaction chamber, followed by pressure evacuation. Oxygen gas was then introduced at a flow rate of 20 sccm, and the glow discharge was ignited at 30 W. After 90 seconds, the power and oxygen were turned off. The samples (denoted as P-SLA) were taken out and immediately immersed into diluted chitosan solution (0.1%) for 30 min to maintain the hydrophilicity.

Fig. 1 Schematic diagrams of the experiment setup: (a) the pre-treatment of Ti disk and electro-spraying deposition of proteins; (b) the plasma treatment under O₂ for 50 W power and 90 seconds; (c) the detail setup for electrospraying. Here, the spraying solution was a mixture of chitosan and model protein with 2% acetic acid solution.

Co-deposition of chitosan and rhBMP-2 by ESD technique

The chitosan molecules were dissolved in the 0.01 M acetic acid to make a viscous solution with a concentration of 10.0 mg mL⁻¹. The model protein BSA was then dissolved in the chitosan solution with different mass ratios (2 : 1, 1 : 1, 1 : 2). To speed up the evaporation rate of the sprayed solution, a certain amount of ethanol was added into the solution above to obtain 10% ethanol content. Co-deposition process was then carried out in a sterile environment using a commercial ESD system (AST, Bleiswijk, Netherlands). Briefly, under the controlled atmosphere (relative humidity 15%, ambient temperature 25 °C), the P-SLA disks were installed on a movable stage and perpendicular to the nozzle as a deposition target. After controlled the moving speed of the stage and applied an electric field between the nozzle and the stage, the mixed solution (chitosan–BSA or chitosan–rhBMP-2) was supplied by a syringe pump and then passed through a 27-gage needle for spraying onto the P-SLA disks (shown in Fig. 1c). The voltage used was 11.0 KV and the distance between the needle tip and the disk surface was around 0.3 cm. The coating thickness was controlled by adjusting the spraying time from 3 min to 15 min. After deposition, the disks (denoted as ES-P-SLA) were air-dried and stored at 4 °C until further use.

Surface characterization

A series of analytical tools were used to characterize the surface properties of the disks w/wo spraying deposition. Surface wettability, a major parameter affecting protein adsorption and cell adhesion, was measured using the sessile drop method. DI water (around 3 μL) was micro-pipetted on various surfaces at room temperatures. A Nikon high-performance camera was then used to capture the water drop profile on the surface. Subsequently, contact angle data were measured and averaged by using Image J (National Institutes of Health, Washington DC).

To visually examine the surface morphology, Au/Pd sputter-coated disks were imaged using a Field Emission Scanning Electron Microscopy (SEM, JSM-6360LV, JEOL). A working distance about 8–10 mm, an accelerating voltage of 10 kV, and a chamber pressure of 10–8 Torr were found to be suitable for obtaining high-resolution images. The magnification in this study varied from 2000× to 20 000× depending on the dimension of the micro-features.

The surface roughness and coating thicknesses corresponding to different spraying times were determined by Atomic Force Microscopy (AFM; Multimode NanoscopeIIIa, USA). Briefly, the coated disks were scanned in tapping mode at a rate of 1 Hz using silicon cantilevers (NSG10, NT-MDT, Russia) with average nominal resonant frequencies of 250 kHz, spring constants of 15 N m⁻¹, and a radius of curvature of the tip <10 nm. Manoscope imaging software (version 6.13r1, Veeco) was used to analyze the surface characteristics.

In vitro release profiles

To investigate the release profiles in vitro, the BSA-loaded disks were immersed in a vial containing 1 mL Simulated Body Fluid (SBF) solution (pH = 7.4) and incubated at 37 °C. At the pre-determined time point, the release medium was collected and totally replaced with an equal amount of fresh SBF solution. The amount of released BSA was quantified by the bicinchoninic acid protein assay kit (Beyotime, shanghai, China) according to the manufacturer's protocol. The release profile was then calculated in terms of the cumulative released percentage vs. release time.

Cell culture

C2C12, a myoblastic precursor cell, was cultured in growth medium DMEM supplemented with 10% FBS and antibiotics in a humidified incubator at 37 °C under a 5% CO₂/95% air atmosphere. The cultures were passaged (1 : 3) every 2 days and used for experimentation up to passage 20.

Bone Mesenchymal Stem Cells (BMSCs) were flushed out of the tibias and femora of rats with DMEM-low glucose (DMEM-LG). After cultured in DMEM-LG medium supplemented with 10% FBS in a CO₂ incubator at 37 °C for 3 days, the BMSC cells set as passage one were replaced with the fresh medium. When the confluences reached the level of 80–85%, the BMSC cells were trypsinized with 0.25% trypsin. All experiments were performed on 2/3 passage after enlarged cultivation.

Cell adhesion, morphology, and proliferation

The cell adhesion and morphology on the disks were determined by fluorescent staining. Briefly, the disks were placed a 24-well plate under the sterilized conditions. C2C12 cells were then seeded at a density of 1 × 10⁴ per well for incubation. 12 h later, the disks were washed with PBS twice and fixed with formalin solution (3.7% formaldehyde in PBS) for 15 min on ice, subsequently stained by FITC-phalloidin (Sigma, St. Louis, USA) for the cell cytoskeleton. 2-(4-Amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI, Beyotime Biotech, Jiangsu, China) solution was added to stain cell nuclei for 10 min. The specimens were then observed using Confocal Laser Scanning Microscope (CLSM, Nikon A1R, Japan). Area and perimeter of isolated cells were analyzed with the grain analysis tool by using Image J software.

MTT assay was performed to evaluate the cell proliferation on the disks with different coatings. C2C12 cells with a density of 1.5 × 10⁴ cells per well were seeded and cultured with the disks for the pre-determined time. After 1 (or 3) day, 30 μL of 0.5 mg mL⁻¹ MTT solution (Sigma-Aldrich) was added to each well and incubated at 37 °C. After 2 h, the media was removed and the formazan solubilized with 100 μL of dimethyl sulfoxide (DMSO). The absorbance of the formazan–DMSO solution was read at 492 nm using an enzyme-linked immunoadsorbent assay plate reader (SPECTRAmax 384, Molecular Devices, USA). Results were directly reported as the OD values (n = 3).

Alkaline phosphatase (ALP) activity

Osteogenic differentiation was assessed by measuring the alkaline phosphatase (ALP) activity of BMSC cells. The BMSC cells were seeded at a density of 3 × 10⁴ cells per well and cultured with different disks. At day 7 and 14, the culture medium was removed and the cells were lysed in 300 μL of 1% Nonidet P-40 solution at 37 °C for 90 min. 50 μL of cell lysate was then mixed with 100 μL of 2.5 mg mL⁻¹ p-nitrophenylphosphate (PNPP–Na) substrate solution at pH 9 composed of 0.1 mol L⁻¹ glycine and 1 mmol L⁻¹ MgCl₂·6H₂O. After 30 min incubation at 37 °C, the reaction was ended by adding 50 μL of 0.1 M NaOH. Using the cells treated with normal medium as standard, the optical density (OD) was measured at the wavelength of 405 nm on the same plate reader. The total protein content was quantified by the BCA kit, and the ALP activity was expressed as the changed OD values divided by the reaction time and total protein quantity (n = 3).

Data analysis

All numerical data were expressed as the mean ± standard deviation. Statistical analysis was performed with one-way analysis of variance (ANOVA). A value of p < 0.05 was considered as statistical significance.

Results and discussion

Influence of plasma treatment on surface properties

Surface energy/wettability is a very important factor influencing the cell adhesion and differentiation. Osteoblasts grown on high surface energy (hydrophilic) substrates display increased cell adhesion, proliferation and upregulation of various differentiation markers such as ALP activity.^28,29 Therefore, turning wettability is one of the effective ways to improve the bioactivity of the metal implants. Here, oxygen plasma was used to adjust the surface wettability. Fig. 2 compared the surface morphology and water contact angles on the Ti disks w/wo plasma treatment. It is noted that the microscale features were made on the disks after SLA treatment: the grit sand blasting produced the micro cavities with an average diameter of 20–40 μm and the acid etching prepared the micropits approximately 0.5–3 μm in diameter. Because of these features, as well as the nature of the substrate materials, the SLA disks showed hydrophobic features. In comparison, the plasma treatment dramatically made the surface hydrophilic (contact angle: 35°) without obvious morphological changes because of the low plasma power and short treatment time.

Fig. 2 SEM images for microstructural analysis and the contact angles on different titanium surfaces: (a) SLA-treated titanium; (b) SLA-titanium with plasma treatment; (c) directly electrospraying for 5 minutes on SLA-titanium; (d) electrospraying for 5 minutes on P-SLA-titanium. The scale bar: 20 μm.

Plasma treatment not only regulated the surface wettability, but also fascinated the electrospraying coating. The surface morphologies w/wo plasma treatments were compared in Fig. 2c and d. As can be seen, the ES-SLA disks still exhibited the micro-cavities at some areas since it was very difficult for hydrophilic chitosan to deposit on the hydrophobic SLA surface. However, the hydrophilic micro-cavities on the P-SLA disks were easily filled with the chitosan molecules. These can be attributed to the enhanced interaction between the negative P-SLA surface via plasma treatment and positive nature of chitosan, as well as the hydrophilic properties of P-SLA disks and coating materials.

Surface morphology vs. spraying time

Fig. 3 investigated the influence of spraying time on the micro features and wettability on the P-SLA disks. SEM images indicated an increase in coating thickness with prolonged deposition time. After electrospraying for 3 min, the micro-cavities with sharp edges became blunt compared with that of the P-SLA disk. As spraying for 5 or 10 min, the surface cavities were almost filled with chitosan molecules. These findings can be further confirmed by the AFM images shown in Fig. 4a. After SLA treatment, the disk surface exhibited a large number of cavities around 1.63 μm in height and 450 nm in depth. The root mean square roughness (RMS) was around 175 nm. With increasing spraying time, the surface became smoother and the RMS value significantly decreased. For instance, the RMS value on the 5 min disk was around 8.942 nm, approximately 5.0% of RMS value on the SLA disk. Fig. 4b also listed the water contact angles on the corresponding surface. To be summarized, the spraying coating of chitosan molecules made the surface hydrophilic with contact angles ranging from 48° to 55°, which would fascinate the cellular adhesion and proliferation.²⁹

Fig. 3 Influence of spraying time on the microstructures and hydrophilicity on different titanium surfaces: (a) plasma-treated SLA-titanium without electrospraying, and chitosan-electrosprayed P-SLA-titanium disk for (b) 3 min, (c) 5 min, and (d) 10 min. The scale bar: 20 μm.

Fig. 4 (a) Representative AFM images on the electrospraying P-SLA samples and the corresponding surface roughness. With the longer electrospraying time, the surface roughness gradually decreased and leveled off. (b) Comparison of water contact angle on different treated surfaces. Here, data were derived from n = 3 samples. *p < 0.05.

In vitro release studies

To improve the protein activities and achieve the prolong release, the loading approach and conditions are very critical. Here, macromolecule BSA was considered as a model protein to evaluate the release characteristics of rhBMP-2 from the chitosan coating layer. Fig. 5 compared the release profiles of BSA molecules from Ti disks with different loading approaches including physical absorption, dropping, BSA–chitosan immersion, and electrospraying. The original concentration for BSA loading in all cases was maintained as 140 μg cm⁻². The disks with BSA absorption, dropping, and immersion exhibited clear burst release profiles in the first 12 hours. Over the first day, more than 90% BSA were quickly released for the absorption- and dropping-samples, and around 87% BSA were released for the immersion approach. In comparison, the electrospraying approach displayed a sustained release profile over one week. In the first 6 hours, the ES-disks showed a little burst release, followed by a slow phase. In the release time ranging from 10 h to 150 h, the release profile can be considered as a zero-order release profile (i.e. a constant release rate). rhBMP-2 is a macromolecule with a molecular weight of 13KD and its release rate was faster than that of BSA molecules (Fig. S3†). Therefore, the electrospraying coating provided the possibility for regulating the sustained release of proteins (i.e. BSA and rhBMP-2).

Fig. 5 Cumulative release of BSA from sprayed Ti in PBS with different loading approaches including physical absorption, dropping, BSA–chitosan immersion and electrospraying.

In addition to the loading approach, the loading amount and the spraying time also affect the rate of protein release. Fig. 6a and b compared the release profiles of BSA-loaded disks with different loading amount and the spraying time, respectively. The spraying time for the BSA-loaded disks in Fig. 6a was kept 10 minutes and the loading concentration in Fig. 6b was fixed 140 μg cm⁻². As can be seen, the release profiles for different BSA loading concentration followed a similar pattern in Fig. 6a. There is a burst release in the first 10 hours, which was followed by a slow phase. However, the release rates in the slow phase for 35 and 70 μg cm⁻² showed smaller than that of 140 μg cm⁻². In other words, the accumulated BSA released concentration exhibited the positive relationship with the loading amount. When the loading amount was decreased from 70 to 35 μg cm⁻², the ratio of the accumulated released concentration at 156 h was around 1.6. Similarly, when the loading amount was decreased from 140 to 70 μg cm⁻², the ratio of the accumulated released concentration at 156 h was around 2.0, implying a positive relationship between the released concentration and the loading amount. Fig. 6b also demonstrated the positive relationship between the accumulated released concentrations with the spraying time. Therefore, by regulating the loading amount and the spraying time, one can pre-design the released rate and the accumulated released concentration for the application needs.

Fig. 6 Cumulative release of BSA from sprayed Ti in PBS with (a) different BSA loading concentrations (35, 70, 140 μg cm⁻²) for spraying 10 minutes; (b) spraying 3, 5, 10 minutes by using the fixed BSA concentration (140 μg cm⁻²).

In vitro cell adhesion and proliferation

The cellular responses to the chitosan–rhBMP-2 coatings were assessed using C2C12 cells in terms of the cell adhesion, morphology, and proliferation. C2C12 is a model cell line that has been widely used to study the skeletal muscle regeneration. Pristine Ti disks with polished surface were used as a control. The morphologies of the cells cultured on the untreated- and treated-Ti disks for 12 hours are presented in Fig. 7. It was observed that the C2C12 cells loosely adhered on the pristine Ti substrates with low density. SLA treatment allowed more cell adhesion and spread well over the surface because of the micro-porous topography. Compared with SLA disks, P-SLA displayed more cell adhesion, which was attributed to the wettability of the surface via plasma treatment. Within all the disks, the ES-P-SLA was most favourably attached by the C2C12 cells with compact structure and high density.

Fig. 7 Fluorescence and merged images of C2C12 cells cultured on the treated surfaces of Ti, SLA, P-SLA and ES-P-SLA for 12 h (scale bar: 100 μm). The cytoskeletal F-actin fibers of cells were stained with FITC-Phalloidin (green) and the nucleus was labeled with DAPI (blue).

On the fluorescence images with higher magnification (shown in Fig. 8), it is clearly seen the single C2C12 cells attaching to the pristine Ti surface well-spread with normal polygonal shape. On the SLA and P-SLA disks, the cells display the slightly elongated shape with irregular perimeter. Meanwhile, the cells randomly elongated on the ES-P-SLA surface after 12 hour of culture, displaying a spindle-like morphology. Researchers have reported the influence of the surface roughness, micro-/nano-topography, porosity, surface chemistry, and wettingability on the cellular adhesion and spreading at the initial stage.³⁰ It has been also found that the BMP-2 not only promotes the osteoblast differentiation, but also enhances the cellular adhesion.³¹ In our case, fluorescence observation indicated that rhBMP-2 and the surface chemistry had important effects on the cellular adhesion and spreading. On the ES-P-SLA disk, the positively charged chitosan molecules absorbed more cell-adhesive proteins than the pristine Ti surfaces. Together with the enhancement by rhBMP-2, much more cells were thus absorbed on the ES-P-SLA disk in the early stage.

Fig. 8 The spread arrangement of C2C12 cells on the surfaces with qualification of cell body (scale bar: 25 μm). Error bars represent the SD about the means (n = 10).

The cell proliferation on different surfaces after 1 and 3 days of incubation is presented in Fig. 9. As can be seen, there is no big difference for the C2C12 proliferation level on the different substrates after 1 day of incubation. At 3 day after cell seeding, the C2C12 cells on the SLA-Ti substrates proliferated slightly better than those on the pristine Ti, but no significant differences existed between these two samples. On the other hand, a significant increase in the cell proliferation level was observed in the cases of the P-SLA and ES-P-SLA disks. Moreover, the Ti disks coated with the chitosan–rhBMP-2 layer were considered to be more promising than the SLA-Ti equivalents in the bone regeneration field. This could be due to improvement in the surface wettability, the chitosan coating, and incorporation of growth factors. These findings are in accordance with the results reported by other researchers: the hydrophilic coating showed enhanced osteoblastic responses as compared to the pure microporous equivalents.^32,33

Fig. 9 Cell proliferation of myoblast cells grown on pristine Ti (as control), SLA, P-SLA, ES-P-SLA loaded with rhBMP-2 after 1 and 3 day incubation.

In vitro cell differentiation

To evaluate the early differentiation of osteoblasts, ALP activity testing were performed at days 1, 7, and 14. As shown in Fig. 10, pristine Ti, SLA, P-SLA, and ES-P-SLA exhibited similar ALP levels of BMSCs cells at day 1. The enhancement of ALP activity could be observed when the cells were incubated with Ti disks after 7 days. At day 7, ES-P-SLA displayed slightly higher ALP activities than rhBMP-2 lacking groups (i.e. pristine Ti, SLA, P-SLA). For prolonging the culture time (i.e. day 14), significant increase was observed in ALP activity assay compared to the rhBMP-2 lacking groups, indicating the osteogenic differentiation induced by rhBMP-2 over two weeks. Although the SLA and P-SLA showed higher ALP activity levels than pristine Ti at day 14, no statistically significant increase was observed in ALP activity assay between SLA and P-SLA.

Fig. 10 Alkaline phosphatase detection during bone marrow stromal cells (BMSCs) differentiation after 1, 7 and 14 days culture onto titanium disks. Statistically significant difference was observed in BMSCs co-cultured with titanium disks (sprayed for 10 minutes) expressing bone morphogenetic protein (BMP-2) (*p < 0.05).

Discussion

Efficient delivery of bioactive molecules including growth factors to cells is one of the most important challenges for improving the bioactivity of the orthopedic/dental implants. Many researchers have reported the incorporation of osteogenic growth factors (i.e. BMP-2) onto metallic implants as biological coatings to improve its osteoinductivity.^34,35 However, the efficient utilization of BMP-2 and then avoidance of supra-physiological doses still represents a great challenge. In order to achieve optimal results, the technique used for incorporating BMP-2 must be protein-friendly to avoid the changes of the protein structure and thereby denaturation in vitro and in vivo. Furthermore, like other bone morphogenetic proteins, BMP-2 plays an important role in the development of bone, and must be delivered in a sustained fashion for an ideal implant. Therefore, enhancing/maintaining the quality and bioactivity should be the key focus for the application of BMP-2 in metallic biomaterials.

In our studies, electrospraying holds great potential to incorporate the BMP-2 on the microporous metallic surfaces. Compared with other techniques, this simple approach can generate very fine droplets from a solution and effectively immobilize biological molecules including proteins, DNAs, and even living cells.³⁶ Researchers have demonstrated the dehydration period is extremely quick during the process, and thus the bioactivity loss is pretty low and neglected.³⁷ Our group has also developed the protein/cell delivery systems by using this technique, and confirmed the bioactivity preservation of proteins, even the maintenance of functionality of the cell-lines and pancreatic islets.^38,39 Therefore, we believe the rhBMP-2 molecules could maintain its normal bioactive function after this biological benign process. Another advantage provided by EDS is to deposit biomaterials on substrates in a well-controllable way, allowing the deposition of various biomaterials with designed thickness and distribution. As has been discussed in Fig. 6a and b, to slow down the rhBMP-2 release rate, the loading amount and the spraying time were adjusted for achieving the different release rate and the needed concentration. For a specific therapeutic application, these parameters and the release characteristics could be pre-designed and regulated according to the nature of the therapeutic molecule. Additionally, two or more biomolecules could be easily deposited on the substrate with precise position and required concentrations for realizing desired multi-functions.

Beside the incorporation technique, the coating materials are very important for designing an ideal coating for orthopedic/dental implants. The material candidates must be biocompatible, good processing, biodegradable, protein-friendly, and so on. As a natural polysaccharide, chitosan has found as a biocompatible material with anti-bacterial properties in tissue engineering and controlled drug delivery. Because of the positive charged amino groups on the chitosan backbone, more cells were easily attached to the ES-P-SLA surface at the initial stage of incubation as discussed in Fig. 7. Adhesion of cells onto the substrate is a necessary step before cells proliferate and differentiate further. The factors including surface topography, chemistry, and substrate–cell interaction affect the osteoblast adhesion significantly and complicatedly. Previous studies have shown the surface characteristics induced by SLA treatment could achieve the strong mechanical interlock between the substrate and host bone, enhancing cellular adhesion and promoting the bony ingrowth.⁴⁰ Compared with the SLA surface, ES-P-SLA disks in our studies showed much smaller roughness and seemed weaken substrate–cell interaction. However, the plasma treatment gave the negative charge on the surface, as well as the high surface energy (hydrophilic), enhancing the mechanical interlock between the chitosan coating and metal substrate assisted with electro-hydrodynamic interaction. The significant enhancement of cellular adhesion on such an positively-charged surface confirmed the positive charges of coating layer promoted cellular adhesion,⁴¹ indicating the dominant effect of the surface charge other than roughness. Certainly, the rhBMP-2 incorporation and the surface wettability also did the positive effect on the cellular adhesion in this case (Fig. 4b).

The goal of this study is to incorporate rhBMP-2 on metallic substrate, maintaining the bioactivity and achieving a sustained release. To data, several strategies have already been attempted by other groups for incorporating rhBMP-2 in titanium surfaces. For instance, titanium implants self-assembled with polyelectrolyte multilayers consisting of BMP-2, chitosan and gelatine promoted osteogenic-lineage differentiation of MSCs in vitro and increased bone formation in vivo.^42–44 To simplify the process and improve the biological efficiency of rhBMP-2, the EDS technique in this work was applied to construct a degradable osteoinductive coating. Our results demonstrated that rhBMP-2 entrapped in chitosan layer retained its bioactivity and was released over an extended period of time (Fig. 5 & S3†). The enhanced proliferation and cell differentiation could be attributed to the sustained released of rhBMP-2. Compared with the disks treated by other techniques, the coating layer constructed by plasma-ESD potently attenuated the burst release (Fig. 5 & S1†). Importantly, it was verified the chitosan layer without crosslinking was completely degraded after 10 days in SBF solution and only small amount of residue was kept on the disk surface (shown in Fig. S2†), indicating the rhBMP-2 released in the medium. For comparison, we confirmed the ALP production induced by free rhBMP-2 with the dose similar to the released from the chitosan layer in culture medium. Therefore, our results support the hypothesis that the incorporated rhBMP-2 promotes the osteoblast behaviours mainly through the degradation of polymeric layers. Some researchers also propose the mechanism that the cells might sense the signals from the entrapped rhBMP-2, stimulating biological responses.^42,45 Our preliminary data in Fig. S2† verified the crosslinked chitosan could not be degraded over three weeks. When rhBMP-2 molecule is entrapped into the crosslinked chitosan, this mechanism may be applicable for explaining the osteoblast response. In the present case, the released rhBMP-2 from degraded chitosan has high possibility for binding its specific receptor and triggering the cellular response.

Conclusions

In this work, we have described the construction and characterization of rhBMP-2-incorprated coating on metallic disks for controlled release and enhanced biological performance. As a biological benign process, the ESD together with plasma treatment increased surface energy/wettingability, enhanced the mechanical interlock of chitosan–Ti disks, and realized the incorporation of the growth factors on metallic surface with the preserved/enhanced osteogenic capacity. Release studies in vitro confirmed the sustained release of proteins over two weeks and no high initial burst effect. By tuning the preparation parameters (i.e. initial loading amount, spraying time, composition of spaying solution), the released rate and the accumulated concentration can be well-controlled for therapeutic needs. According to the cellular response in vitro, it was confirmed positively-charged chitosan and rhBMP-2 incorporation promoted early cellular adhesion. Furthermore, ES-P-SLA disks entrapped rhBMP-2 were bioactive and significantly promoted osteoblast functions by showing the increased proliferation and cell differentiation. Therefore, we believe that the chitosan–rhBMP-2 implants constructed by EDS will be excellent platform to develop the next generation metal implants for medical applications. The key technologies developed in this work can also be applied to other biomaterial surface and clinical applications. Examples include artificial bone, joint replacement, and heart stents.

Acknowledgements

The authors are grateful for the financial support from National Natural Science Foundation of China (Grant No. 31570971 and No. 31330028), State Administration of Foreign Experts Affairs P. R. China (Grant No. B14018), and Science and Technology Commission of Shanghai Municipality (Grant No. 14ZR1409800).

Notes and references

1. O. E. Pohler, Injury, 2000, 31(4), 7–13.
2. J. Breme, E. Steinhauser and G. Paulus, Biomaterials, 1988, 9, 310–313.
3. M. J. Niinomi, J. Mech. Behav. Biomed. Mater., 2008, 1, 30–42.
4. D. A. Puleo and A. Nanci, Biomaterials, 1999, 20, 2311–2321.
5. D. W. Lee, Y. P. Yun, K. Park and S. E. Kim, Bone, 2012, 50, 974–982.
6. H. Yazici, H. Fong and B. Wilson, Acta Biomater., 2013, 9, 5341–5352.
7. P. Jiang, J. Liang and C. Lin, Appl. Surf. Sci., 2013, 280, 373–380.
8. D. Zheng, K. G. Neoh and Z. J. Shi, J. Colloid Interface Sci., 2013, 406, 238–246.
9. H. Schliephake, C. Boetel and A. Foerster, Biomaterials, 2012, 33, 1315–1322.
10. E. De Giglio, C. Stefania and C. D. Calvano, J. Mater. Sci.: Mater. Med., 2007, 18, 1781–1789.
11. C. Jin, L. Ren and H. Ding, J. Biomed. Mater. Res., Part B, 2012, 100, 2167–2177.
12. Q. Ma, S. Mei and K. Ji, J. Biomed. Mater. Res., Part A, 2011, 98, 274–286.
13. J. M. Kanczler and R. O. C. Oreffo, Eur. Cells Mater., 2008, 15, 110–114.
14. W. G. La, S. Park and H. H. Yoon, Small, 2013, 9(23), 4051–4060.
15. S. S. Lee, A. R. Sharma and B. S. Choi, Biomaterials, 2012, 33, 4251–4263.
16. R. Dimitriou, E. Tsiridis and P. V. Giannoudis, Injury, 2005, 36, 1392–1404.
17. D. Chappard, E. Aguado, G. Hure, F. Grizon and M. F. Basle, Int. J. Oral Maxillofac. Implants, 1999, 14, 189–196.
18. J. Hall, R. G. Sorensen, J. M. Wozney and U. M. Wikesjo, J. Clin. Periodontol., 2007, 34, 444–451.
19. D. Kim and A. E. Herr, Biomicrofluidics, 2013, 7(1–47), 041501.
20. C. del Rosario, M. Rodriguez-Evora and R. Reyes, Biomed. Mater., 2015, 10(4), 045008.
21. C. M. Xie, X. Lu and K. F. Wang, ACS Appl. Mater. Interfaces, 2014, 6, 8580–8589.
22. F. M. Chen, M. Zhang and Z. F. Wu, Biomaterials, 2010, 31, 6279–6308.
23. O. F. Zouani, C. Chollet and B. Guillotin, Biomaterials, 2010, 31, 8245–8253.
24. Y. J. Park, K. H. Kim and J. Y. Lee, Biotechnol. Appl. Biochem., 2006, 43, 17–24.
25. H. Li, A. M. Koenig and P. Sloan, Biomaterials, 2014, 35, 9049–9057.
26. L. T. de Jonge, S. C. Leeuwenburgh, J. J. van den Beucken, J. te Riet, W. F. Daamen, J. G. Wolke, D. Scharnweber and J. A. Jansen, Biomaterials, 2010, 31(9), 2461–2469.
27. D. Buser, T. Nydegger, T. Oxland, D. L. Cochran, R. K. Schenk, H. P. Hirt, D. Snétivy and L. P. Nolte, J. Biomed. Mater. Res., Part A, 1999, 45(2), 75–83.
28. Z. Qu, X. Rausch-Fan, M. Wieland, M. Matejka and A. Schedle, J. Biomed. Mater. Res., Part A, 2007, 82, 658–668.
29. J. Y. Lim, M. C. Shaughnessy, Z. Zhou, H. Noh, E. A. Vogler and H. J. Donahue, Biomaterials, 2008, 29, 1776–1784.
30. B. G. X. Zhang, D. E. Myers, G. G. Wallace, M. Brandt and P. F. M. Choong, Int. J. Mol. Sci., 2014, 15, 11878–11921.
31. J. Zhang, H. Zhou, K. Yang, Y. Yuan and C. Liu, Biomaterials, 2013, 34, 9381–9392.
32. D. Buser, N. Broggini, M. Wieland, R. K. Schenk and A. J. Denzer, J. Dent. Res., 2004, 83(7), 529–533.
33. H. W. Kim, H. E. Kim and V. Salih, Biomaterials, 2005, 26, 5221–5230.
34. U. M. Wikesjo, R. G. Sorensen, A. Kinoshita and J. M. Wozney, Clin. Implant Dent. Relat. Res., 2002, 4, 174–182.
35. D. B. Spagnoli and R. E. Marx, Oral. Maxillofac. Surg. Clin., 2011, 23(2), 347–361.
36. U. A. Mirza and B. T. Chait, Int. J. Mass Spectrom. Ion Processes, 1997, 162(1), 173–181.
37. I. Uematsu, H. Matsumoto, K. Morota, M. Minagawa, A. Tanioka, Y. Yamagata and K. Inoue, J. Colloid Interface Sci., 2004, 269(2), 336–340.
38. H. He, E. Luedke, X. Zhang, B. Yu, A. Schmitt, B. McClarren, V. Grignol, W. E. Carson and L. J. Lee, J. Controlled Release, 2013, 165(3), 226–233.
39. L. J. Wang, L. Li, J. Yu, Y. Wu, H. He, X. Ouyang, X. Zhao, Y. Yen and L. J. Lee, Carbon, 2014, 79, 294–301.
40. Y. Oshida, E. B. Tuna, O. Aktören and K. Gençay, Int. J. Mol. Sci., 2010, 11(4), 1580–1678.
41. I. F. Amaral, A. L. Cordeiro, P. Sampaio and M. A. Barbosa, J. Biomater. Sci., Polym. Ed., 2007, 18, 469–485.
42. R. Guillot, F. Gilde, P. Becquart, F. Sailhan, A. Lapeyrere, D. Logeart-Avramoglou and C. Picart, Biomaterials, 2013, 34, 5737–5746.
43. N. J. Shah, M. L. Macdonald, Y. M. Beben, R. F. Padera, R. E. Samuel and P. T. Hammond, Biomaterials, 2011, 32, 6183–6193.
44. Q. H. Jiang, L. Liu, J. W. Shen, S. Peel, G. L. Yang, S. F. Zhao and F. M. He, J. Biomed. Mater. Res., Part A, 2012, 100, 2766–2774.
45. T. Crouzier, K. Ren, C. Nicolas, C. Roy and C. Picart, Small, 2009, 5, 598–608.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra09421j

---