Synthesis of a graphene oxide/agarose/hydroxyapatite biomaterial with the evaluation of antibacterial activity and initial cell attachment

Various materials are used in bone tissue engineering (BTE). Graphene oxide (GO) is a good candidate for BTE due to its antibacterial activity and biocompatibility. In this study, an innovative biomaterial consists of GO, agarose and hydroxyapatite (HA) was synthesized using electrophoresis system. The characterization of the synthesized biomaterial showed that needle-like crystals with high purity were formed after 10 mA/10 h of electrophoresis treatment. Furthermore, the calcium-phosphate ratio was similar to thermodynamically stable HA. In the synthesized biomaterial with addition of 1.0 wt% of GO, the colony forming units test showed significantly less Staphylococcus aureus. Initial attachment of MC3T3-E1 cells on the synthesized biomaterial was observed which showed the safety of the synthesized biomaterial for cell viability. This study showed that the synthesized biomaterial is a promising material that can be used in BTE.

) and flower-like crystals (Fig. 1D) were formed, respectively. Figure 2 showed the surface of the synthesized biomaterials treated with different duration of mineralization. When the synthesized biomaterial was mineralized for less than 5 h, organic substance was the primary component formed. The SEM showed that formed crystals were wrapped with a layer of gelatine. When the mineralization time increased, more crystals were formed and the synthesized biomaterials became harder. After 8 h of mineralization, needle-like crystals were observed on the synthesized biomaterials (Fig. 2D). With increasing mineralization time, more crystals were formed. The crystal growth might be inhibited when HA crystals came into contact with one another, and finally, the crystals grew parallel to one another (Fig. 3). X-Ray Diffraction is a method based on the constructive interference of monochromatic X-rays and crystal samples. In this study, the XRD analysis was performed on the samples treated with 10 mA/10 h electrophoresis with addition of GO (Fig. 4). The XRD spectrum showed orientation of formed crystal bunches with the crystal plane (001) at 2θ = 9.3° of GO and (200) at 2θ = 22.1° of hydroxyapatite.
Energy dispersive spectroscopy (EDS) was applied to measure the Ca/P ratio of the synthesized materials (Fig. 5). The EDS data showed that after 8 h of mineralization, the mean Ca/P ratio of synthesized materials was 1.68 ± 0.045.  Staphylococcus aureus (S. aureus) strains were evaluated by colony forming units (CFU) method. After 24 h of incubation, the quantity of adherent bacteria on the synthesized biomaterials with the addition of 1 wt% GO ((0.52 ± 0.17)*10 7 CFU/ml) were significantly less than that on the synthesized biomaterials with the addition of 0.5 wt% GO ((1.55 ± 0.21)*10 7 CFU/ml; p = 0.0026, t-test) and the synthesized biomaterials without the addition of GO ((28.60 ± 2.20)*10 7 CFU/ml; p < 0.0001, t-test) (Fig. 6). The quantity of adherent bacteria on the synthesized materials with the addition of 0.5 wt% GO were also significantly less than that on the synthesized materials without the addition of GO (p < 0.0001, t-test).
Evaluation of initial attachment. Cellular morphology was evaluated using laser scanning confocal microscope (LSCM). The morphology of MC3T3-E1 was assessed after it was cultured on the synthetized biomaterial after 3 days (Fig. 7A). It was observed that the cell morphology on the synthesized biomaterial was preserved, showing shutter-like and polygon-shaped cells. SEM images of the cells after 3 days of culture on the synthetized biomaterial was also obtained to observe the interaction. The SEM micrograph reflected the status of MC3T3-E1 cells attachment after 3 days of culture (Fig. 7B). It was observed that the MC3T3-E1 cells were attached, and the surface of the synthetized biomaterial was mostly covered by the cells.

Discussion
Crystal growth could be controlled and regulated by organic template, character of the substance surface and addition of surfactants. In the initial stage of crystallization process, the crystal nucleation, morphology and orientation were dominated by gelatine because of the constant electric current. The synthesized biomaterial  www.nature.com/scientificreports/ was hybrid which mainly composed of organics. Thus, formed crystals were covered by a layer of gelatine, and their morphology was hardly detected ( Fig. 2B and C). In Fig. 3, it was observed that HA crystals grew parallel to one another. The process of initial HA crystal growth required time. The HA crystal adapted and self-adjusted, gradually formed a parallel array along the c-axis. After completing the self-adjusting process and more crystals formed, the regulation of gelatine on crystal growth decreased, especially the crystals located away from the gelatine matrix.
X-ray diffraction has become a common technique in the research field of crystal structure and atomic spacing. The X-rays were generated by the cathode ray tube, filtered to produce monochromatic radiation, collimated and then directed to the sample 46 . XRD analysis is an excellent method that has been widely used to study the ultrastructure of certain materials. XRD is also a powerful technique that can be used to evaluate the structure of biological minerals such as teeth and bones, and has been used by various researchers to examine the crystal structure of biological mineral composites 47 . In this study, the XRD result showed no impurity peaks, indicating that the synthesized biomaterial had high purity. In the earlier stage of mineralization process, the crystals grew in all direction from the nucleation sites, as shown in Fig. 2C and D. Due to limited space available between adjacent sites of growth, only perpendicularly growing crystals could effectively utilize the available space, resulting in the formation of parallelly oriented crystals 48 . Therefore, it was observed that after the formation of un-oriented HA crystals, crystals growth take place in the available free space, producing crystals growing perpendicular to the surface of un-oriented HA crystals (Fig. 2D).
The result of the EDS showed that the synthesized material had a similar calcium/phosphate ratio with that of thermodynamically stable HA (Ca/P = 1.67). In the EDS method, analytical information is derived from depths that are generally typical for thin film analysis. The particle size affects both qualitatively and quantitatively the ratio of Ca and P. As the accelerating voltage increases, the signal strength increases corresponding to the Ca and P elements (up to 15 kV) regardless of the particle size of the Ca/P. In the 15-30 kV range, shape variation is observed for particle sizes with minimum and maximum dimensions, and is uniform in the median case 49 .
In a previous study, it was investigated that OH − could regulate crystal growth by triggering the active sites at crystal facets 50 . Specifically, the pH determined whether crystal growth was affected by the crystal's interior structure or by the exterior mineralization condition. The exterior mineralization condition dominated the crystallization when the pH value was higher. In the electrophoresis-aided mineralization system, the electrochemical reaction caused the water in the vicinity of the cathode hydrolysed. This process resulted in increasing pH due  www.nature.com/scientificreports/ to hydroxyl aggregation. The increase of electric current strength could accelerate the electrochemical reaction, leading to more hydroxyl formation and the increase of local pH value. Thus, different crystal morphologies were observed when different electric current strengths were applied. Besides, at a higher OH − concentration, each crystal facet generates more active sites 51 . With the increased concentration of OH − ions, ion aggregation became faster and Ca/P clusters with 'tentacles' formed 52 . This process resulted in crystals with small branches at their ends. The single needle-like crystal was formed when the applied electric current was 8 mA; while, crystals with branches were formed when the applied electric current was 10 mA.
S. aureus is a Gram-positive, non-motile, non-spore forming grape-like cluster. It is the most important coagulase-positive pathogen of staphylococci due to its combination of toxic-mediated virulence, invasion and antibiotic resistance. S. aureus is a common pathogen found in various oral diseases, such as oral mucositis, periodontitis, peri-implantitis, endodontic infections and even dental caries. Despite its pathogenic potential, S. aureus is rarely associated with acute dento-alveolar infection. Other oral infections that have been associated with S. aureus include infected jaw cysts, oral mucosal lesions and denture-induced stomatitis 53,54 .
Effect of graphene-based materials on microbial cell structure, metabolism and viability have been investigated. The effectiveness depends on the concentration of the substance, exposure time, physico-chemical properties, as well as the characteristics of the microorganisms used in the test. There are various ways for graphenebased materials to destroy the microbial cells, including disruption of cell walls and membranes through sharp edges of GO; generation of reactive oxygen species, which can be a degradation factor for microbial cells. In the study of Thani et al., it showed that GO has antibacterial and antifungal activity against microorganisms. The anti-microbial activity of GO was detected by a spectrophotometer as an indirect method to measure the growth and cell count against these microorganisms, one eukaryotic fungus (Candida albicans) two Gram negative bacteria (Escherichia coli ATCC 41570 and Pseudomonas aeruginosa ATCC 25619) and two Gram positive bacteria (Streptococcus faecalis ATCC 19433 and S. aureus ATCC 11632) 55 .
The unique properties of GO and reduced-GO can be used clinically for broad-spectrum antimicrobial disinfection treatment. There are a few known mechanisms related to the antibacterial activity of GO-based materials, such as: interaction of extremely sharp edges of GO sheets with wall membrane of the bacteria 56 , charge transferring 57 , wrapping the bacteria within the aggregated GO sheets 58 , reactive oxygen species generation 59 , inhibiting the bacterial respiration 60 and/or the glycolysis processes 61 , RNS generation 62 and DNA fragmentation 63 . Antibacterial activity of GO-based materials depends on the material's properties, such as size, solubility, dispersion, duration of interaction and concentration, density of functional groups and the production of cellular oxidative stress 64 . Graphene materials of smaller size (e.g. GO) have higher cytotoxicity compared to those with larger size. Antibacterial activity of GO has been attributed to membrane stress induced by sharp edges of graphene nanosheets. GO nanosheets can go through the cell membranes of bacteria and intensely extract great amounts of phospholipids from the membranes. GO nanosheets also can oxidize glutathione, which acts as redox state mediator in bacteria 31,59 .
The study of Romero et al. showed positive result of using GO on Gram-negative Escherichia coli and Grampositive S. aureus. The study concluded that GO and reduced-GO can be used in dermatological infections, because the effect on human skin fibroblasts from these treatments is low compared to the antibacterial effect 65 . Study by He et al., showed that GO was effective in killing dental pathogens, both Gram-positive and Gramnegative bacteria 31 . In this study, after the introduction of GO, the synthesized biomaterial showed the antibacterial property to S. aureus. In the group with 1.0 wt% GO, the CFU count of the S. aureus was found significantly less than 0.5 wt% GO.
Cell morphology greatly affects various cellular events, such as proliferation, differentiation, cytoskeletal organization, and gene expression. MC3T3-E1 cells are well-known in vitro osteogenic model system and have been widely used in BTE-related research. MC3T3-E1 cells display a sequential developmental pattern of proliferation and differentiation, resulting in calcified bone tissue similar to bone formation 66 .
The study by Yang et al. showed that dose-dependent upregulation of protein-positive green fluorescent green dopamine neurons increased by threefold when treated with 100 g/ml GO, while exposure to 1 g/ml GO did not show significant promotion compared without nanoparticle control and graphene or carbon nanotubes 67 . Another study by Alegria et al. demonstrated an enriched hemangioblast cell population when cultured in GO cover slips compared to standard gelatin-coated plates as control. The cell morphology observation on day 1 revealed increased cell cluster formation in GO-cultured group compared to controls 68 . Another study by Mazaheri et al., used GO-chitosan composite layers to evaluate antibacterial activity and cell proliferation. Significant antibacterial activity against S. aureus was observed and the surface density of the human mesenchymal stem cell cultured on 1.5 wt% GO-chitosan was nearly the same. Their study found that at higher concentration of GO (6 wt%) could decelerate the proliferation of human mesenchymal stem cells 27 . In the present study, the synthesized biomaterial contained 1.0 wt% GO and it was observed that the addition of GO at this concentration improved the biological properties of synthesized biomaterial, allowing the initial attachment of MC3T3-E1 cells.
Agarose is a biocompatible and biodegradable material. It enables the diffusion and transport of oxygen and nutrition, thus allowing cells growth. However, it is unfavorable to cell adhesion 34 . In the synthesized biomaterial, HA enabled the MC3T3-E1 cells to adhere on its surface. HA is a bioactive and biocompatible material. As a major inorganic component of hard tissues, it has been widely used as a scaffold for mineral-associated tissue engineering and as a carrier for several growth factors. HA may be able to maintain optimal osseointegration over time, despite the molecular mechanisms 69 .

Methods
Synthesis of GO. Two GO solutions (0.5 and 1 wt%) were prepared separately by first heating 2.5 and 5 g, respectively, of critic acid powder (Sigma-Aldrich, St. Louis, MO). The citric acid was heated to 200 °C, and the color changed from colorless to yellow after 5 min. The color then changed to orange after 30 min and the heating was kept until it turned into black liquid after 100 min. The transformation into black liquid suggested the formation of GO 70 . This liquid was then added to 500 ml deionised water. Finally, the pH was adjusted to 5.5 with 1 M NaOH solution. Fabrication of biomaterial by electrophoresis. CaCl 2 agarose hydrogel and phosphate-containing-GO agarose hydrogel were put into the two sides of the tube. The tube was then connected to the plastic cells. Electrodes were set into the bottom of the cells, which were filled with 0.9% NaCl solution to enhance the electrical conductivity. The electric current was maintained constant at 5, 8, 10, 12 mA during electrophoresis. The gels and NaCl solution were refreshed every 2 h, and their exchange defined the completion of a cycle. The electrophoresis process was terminated after 5 cycles. The layer formed in the middle was harvested and freezedried overnight.

Preparation of metastable mineralization
Characterization and evaluation of synthetized biomaterial. The surface morphology and the chemical analysis with respect to the calcium/phosphate (Ca/P) ratio of the synthetized biomaterial were evaluated using field-emission scanning electron microscopy (SEM) and energy dispersive spectroscopy (EDS) (Hitachi S4800, Hitachi Ltd., Tokyo, Japan), respectively. The structure of scaffold was identified using X-ray diffraction (XRD) (X'Pert Pro, Philips Almelo, Netherlands).
Evaluation of antibacterial activity. The antibacterial activity of synthesized biomaterials on S. aureus strains was evaluated using Colony Forming Units (CFU) counting method. The synthesized biomaterials (containing 1.0 wt% GO: n = 10; containing 0.5 wt% GO: n = 10; and without the addition of GO: n = 10) with a size of 3*3*1.5 mm 3 were prepared and autoclaved to sterilize. S. aureus (ATCC 6538) cells at a concentration of 10 6 CFU/mL was prepared. Sample with addition of 300 μL of S. aureus (10 6 CFU/mL) was put in 1 mL tube. After incubation at 37 °C for 24 h, samples were vigorously vortexed for two minutes. One hundred μL from each sample was collected and immediately put in 900 μL brain heart infusion to dilute the bacteria concentration in tenfold. The diluted bacterial suspension corresponding to each sample was transferred into a horse blood agar. All the plates were incubated at 37 °C for 24 h and the number of CFU on the plate was counted visually.
The data collected in this study was quantitative, numerical, and directly measured by counting visible S. aureus coliform colonies on plates. Data are presented as mean ± standard deviation. The values of experiment groups are compared to those of the control groups. Differences between two mean values were calculated by paired t-test with statistic software (SPSS Statistic 24; IBM). Differences were considered significant at p < 0.05.
Evaluation of initial attachment. MC3T3-E1 cells (obtained from the Cell Culture Centre of the Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, China) were cultured in high-glucose DMEM supplemented with 10% fetal bovine serum, 100 U mg mL −1 penicillin, and 100 mg mL −1 streptomycin. The culture conditions were maintained at 37 °C in a humidified atmosphere containing 5% CO 2 and the medium was renewed every 2 days.
For cell morphology and attachment evaluation, the synthesized biomaterial treated with 10 mA /10 h of electrophoresis system and with addition of 1.0 wt% GO was prepared to match the inside diameter of a 24-well cell culture plate. The synthesized biomaterial was sterilized by immersing in 70% alcohol for 30 min, and seeded with MC3T3-E1 cells at 3.5 × 10 4 cells per mL per well in a 24-well plate. The medium was changed every 2 days. On the first and third day, the medium was removed from the flask and the pre-warmed CellTracker™ green (C7025 Invitrogen) dye working solution was added and incubated for 30 min in CO2 incubator. The dye working solution was then replaced with fresh, pre-warmed medium and the cells incubated for another 30 min at 37 °C. The cell morphology was observed under laser scanning confocal microscope (LSCM; TCS SP2; Leica, Germany). The attachment of MC3T3-E1 cells and synthesized biomaterial was observed using field-emission scanning electron microscopy (SEM, Hitachi S4800, Hitachi Ltd., Tokyo, Japan).

Conclusion
The biomaterial which consists of GO, AG and HA was synthesized in this study with an electrophoresis system to accelerate the fabrication process. The synthesized biomaterial with 10 mA/10 h treatment of electrophoresis and addition of 1.0 wt% GO showed promising results in antibacterial property and MC3T3-E1 cells initial attachment. We concluded that the synthesized biomaterial consists of GO/AG/HA is a good candidate for BTE.