Carbon Nanotubes-Hydroxyapatite Nanocomposites for an Improved Osteoblast Cell Response

An alternative and simple coprecipitation method was developed to obtain carbon nanotube-hydroxyapatite (CNTs:HAp) based nanocomposites. The incorporation of CNTs (in a concentration of 5% and 10% of total weight of the nanocomposite) and their impact on both structural and biological properties were studied by using a set of standard complementary biological, microscopic, andspectroscopictechniques.ThecharacteristicpeaksofcarbonstructureinCNTswerenotobservedintheCNTs-HApcompositesbyX-raydiffractionanalysis.Moreover,FTIRandRamanspectroscopiesconfirmedthepresenceofHApasthemainphaseofthesynthesizedCNTs:HApnanocomposites.TheadditionofCNTsconsiderablyaffectedthenanocompositemorphologybyincreasingtheaveragecrystallitesizefrom18.7nm(forrawHAp)to28.6nm(forCNTs:HAp-10),confirmingtheirproperincorporation.ThebiocompatibilityevaluationofCNTs:HAp-5andCNTs:HAp-10nanocompositesincludedtheassessmentofseveralparameters,suchascellviability,antioxidantresponse,andlipidperoxidation,onhumanG-292osteoblastcellline.Ourfindingsrevealedgood biocompatibility properties for CNTs:HAp nanocomposites prepared by the coprecipitation method supporting their potential uses in orthopedics and prosthetics.


Introduction
One of the major problems that modern medicine has to face is the development of new materials that can mimic the human tissue in order to minimize the risk of rejection by the human body.According to a report of The American Association of Orthopedic Surgeons in 2002, bone implants are one of the most common transplants, outranked only by blood transfusions.In this context, worldwide, researchers have focused their attention on obtaining new and improved materials with special properties for hard tissue replacement implants.For this purpose, they explored the field of biomaterials.A study has revealed that, in a period of 6 years, between 2008 and 2014, the biomaterials device market has increased with almost 150 billion dollars [1,2].There are two main categories of orthopedic biomaterials: implants usually made of metals, ceramics, or hard polymers and scaffolds for bone tissue regeneration [1].The second type of biomaterials is based on biodegradable polymers and their composites [1].The most frequently used biomaterial in orthopedic surgeries is hydroxyapatite (HAp).Having the chemical composition Ca 10 (PO 4 ) 6 (OH) 2 , hydroxyapatite is found in human bones and teeth, being the main mineral constituent of natural hard tissue [3][4][5].Given the similarity with the human tissue, synthetic hydroxyapatite can bond with natural bones [3].Furthermore, it has outstanding properties, such as 2 Journal of Nanomaterials a remarkable bioactivity and osteoconductivity, which make it a suitable candidate for coating metallic implants, creating an adequate surface for bone growth and integration [1,[6][7][8][9][10].However, recent studies have emphasized the weaknesses of HAp implants or coatings, caused by the fragility of synthetic hydroxyapatite, its low fracture toughness, wear resistance [1, 7-9, 11, 12], the appearance of corrosion, and negative tissue reaction.All these drawbacks can induce loosening of implants.Therefore, in order to improve the mechanical properties of HAp and to increase the corrosion resistance, its structure must be reinforced with materials such as carbon nanotubes (CNTs), TiO 2 , or ZrO 2 [13][14][15][16][17][18][19].Since the discovery of CNTs in 1991 by Iijima [20], they have been studied intensively due to their extraordinary thermal, electrical, and mechanical properties owed to their unique atomic structure [21][22][23].Previous studies have proven that CNTs are some of the toughest materials discovered.Compared to steel, the tensile strength of carbon nanotubes is one hundred times higher, although the density is almost seven times lower than the density of steel [24,25].Another remarkable property is the resistance of CNTs to erosion.They can withstand exposure to strong acids and alkali without having their structure affected [24,[26][27][28].Moreover, the biocompatibility of CNTs in orthopedic applications by in vitro studies has been established, which showed an accelerated bone growth [1,29].Also, several studies [1,[30][31][32][33] have proven that carbon nanotubes induced increased proliferation and differentiation of osteoblasts in vitro.Although the biocompatibility of CNTs was not unanimously recognized due to the contradictory reports, it has been shown that different ceramics or polymer matrices reinforced with carbon nanotubes improve the structural properties, like fracture toughness, strength, elastic modulus, or wear resistance [1,[34][35][36] without jeopardizing the properties of the reinforced material.
In this study, we report the synthesis of CNTs:HAp nanocomposites at low temperature by a simple technique.The phase and the chemical structures of the prepared samples were tested by X-ray diffraction, Fourier transform infrared (FTIR), and Raman spectroscopies.Moreover, the structure of the CNTs:HAp-5 and CNTs:HAp-10 nanocomposites was analyzed by scanning electron microscopy (SEM).Furthermore, the human G-292 osteoblast-like cell line was selected as an in vitro model to assess CNTs:HAp nanotoxicity and biological properties.In order to test the biocompatibility of carbon nanotubes-hydroxyapatite based nanocomposites, several parameters including cell viability, antioxidant response, and lipid peroxidation were assessed.

Preparation of Carbon
Nanotubes.The chemical vapor deposition (CVD) method was used to grow single-wall CNTs onto a quartz substrate.The CVD method synthesis is essentially a two-step process consisting of a catalyst preparation step followed by the actual synthesis of the CNTs.The quartz piece cut into required dimensions was first heat treated in a furnace at 500 ∘ C for about 10 minutes.It was then cooled down to the room temperature and dipped in the catalyst that was basically a solution of ethanol and Fe-Mo in mole ratio of 10 : 1.The quartz substrate thus dipped in catalyst was then placed in a quartz tube in a CVD furnace.The deposited Fe catalyst was then reduced at 800 ∘ C by using hydrogen and argon gases which passed through the furnace.Finally, the substrate was subjected to a source of carbon by the passing of ethylene gas that caused the decomposition of carbon and resulted in the synthesis of CNTs.

Preparation of Hydroxyapatite-Carbon
Nanotube Composite.The synthesis of carbon nanotubes-hydroxyapatite (CNTs:HAp) nanocomposites was achieved by an adapted coprecipitation method.Ca(NO 3 ) 2 ⋅4H 2 O was dissolved in 300 mL deionized water.Then, a quantity of 0.5 g or 1 g of CNTs was added into the Ca(NO 3 ) 2 ⋅4H 2 O solution and was sonicated for 12 hours to obtain a homogeneous mixture.On the other hand, the (NH 4 ) 2 HPO 4 was dissolved in 300 mL deionized water according to the molar ratio of Ca/P 1 : 67.The (NH 4 ) 2 HPO 4 solution was added drop by drop into the [CNTs + Ca(NO 3 ) 2 ⋅4H 2 O] homogeneous mixture with continuous stirring at 50 ∘ C for 12 hours.The pH was constantly adjusted and kept at 10 with ammonia solution during the reaction.The resulting mixture was washed several times with deionized water maintaining a neutral pH to remove the residual NO 3 − and NH 4 + .The resulting CNTs:HAp-5 (0.5 g of CNTs added into the Ca(NO 3 ) 2 ⋅4H 2 O solution) and CNTs:HAp-10 (1 g of CNTs added into the Ca(NO 3 ) 2 ⋅4H 2 O solution) nanocomposite samples were dried at 80 ∘ C for 48 hours in an oven.Furthermore, the synthesis of hydroxyapatite (HAp; Ca 10 (PO 4 ) 6 (OH) 2 ) was carried out as reported [37,38] by the following reaction [39]: 2.4.Sample Characterization.X-ray diffraction characterization was carried out using a Bruker D8 Advance diffractometer, with nickel filtered CuK  ( = 1.5418Å) radiation with a high efficiency one-dimensional detector (Lynx Eye type) operated in integration mode.The diffraction patterns were collected in the 2 range 20 ∘ -90 ∘ , with a step of 0.02 ∘ and 34 seconds measuring time per step.A Perkin Elmer 1720X FTIR spectrometer was used for all spectroscopic studies.The instrument was equipped with a horizontal ATR-FTIR accessory from Spectra-Tech.All the spectra were the results of 256 coadded scans at a resolution of 4 cm −1 and a spectral range of 500-2000 cm −1 .All of the FTIR analyses were conducted at room temperature around 23 ∘ C. The spectrum was recorded in the range of 500 to 4000 cm −1 with 4 cm −1 resolution.Micro-Raman spectra on powders were performed in a backscattering geometry at room temperature and in ambient air, under a laser excitation wavelength of 514 nm, using a Jobin Yvon T64000 Raman spectrophotometer placed under a microscope.The morphology of the HAp, NTCs, NTCs:HAp-5, and NTCs:HAp-10 was analyzed by SEM (HITACHI S2600N microscope).

MTT Assay.
Cell viability was evaluated by MTT (3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) colorimetric assay.After 24 and 48 hours of exposure to CNTs, HAp, and CNTs:HAp-5 and CNTs:HAp-10 nanocomposites, the medium was removed and the cells were washed with PBS and incubated with 1 mg/mL MTT solution for 2 hours at 37 ∘ C. Afterwards, the MTT solution was removed and the purple intracellular formazan crystals were solubilized with isopropanol.The absorbance was read at 595 nm using a Tecan multiplate reader (TECAN GENios, Grödic, Germany) and the optical density for untreated cells was represented as a viability of 100%.In spite of all controversies in using the MTT assay to test cytotoxicity of CNTs and to determine relative cell viability, we did not observe any interference of CNTs with the MTT assay, or any indication of false results, which is supported by previous studies on CNTs [40,41].

Glutathione Assay.
At the end of exposure, cells were detached and total protein extract was obtained by sonication (3 times × 30 seconds on ice) using an ultrasonic processor (Hielscher UP50H, Germany).The cell lysate was further deproteinized with a 5% sulfosalicylic acid (1 : 1), centrifuged at 10,000 rpm for 10 minute at 4 ∘ C. The cellular GSH concentration was determined using the Glutathione Assay Kit (Sigma-Aldrich, USA) following the protocol provided by the manufacturer.Briefly, each sample was incubated with 5,5  -dithiobis-2-nitrobenzoic acid (DTNB) for 5 minutes at room temperature in order to allow the reduction of DTNB into 5-thio-2-nitrobenzoic acid (TNB).The absorption was recorded at 405 nm using a Tecan multiplate reader (TECAN GENios, Grödic, Germany).The GSH levels were calculated as nmoles GSH/mg protein and the results were expressed relative to control.
2.8.Malondialdehyde Assay.Malondialdehyde (MDA) concentration was assessed according to the method described by Dinischiotu et al. [42].A volume of 200 L of sample diluted correspondingly was mixed with 700 L of 0.1 N HCl and incubated for 20 minutes at room temperature.Then, 900 L of 0.025 M thiobarbituric acid (TBA) was added and the total mixture was incubated for 65 min at 37 ∘ C. The MDA-TBA adduct fluorescence was measured using a JASCO FP-6300 spectrofluorometer ( ex / em of 520 nm/549 nm).The MDA concentration was calculated using a standard curve of 1,1,3,3-tetramethoxypropane in the range of 0.05-5 M and expressed as nmoles of MDA/mg protein.The results were represented relative to control.
2.9.Protein Concentration Assay.The protein concentration of the cellular extract was determined by Bradford method [43] using bovine serum albumin as standard.

Statistical Analysis.
All values are expressed as mean value ± standard deviation (SD) of three independent experiments.The data was analyzed for statistical difference by Student's t-test.A  value less than 0.05 was considered significant.The average sizes  of HAp and CNTs:HAp were calculated from the XRD patterns using Scherrer's formula [44]:  = / cos , where  is a constant (shape-factor) whose value is 0.9,  is the wavelength of X-rays (1.5406 Å),  (in rad) is the full width at half maximum (FWHM), and  is the Bragg angle.The grain size increased from 18.7 ± 1 nm (HAp) to 22.5±1 nm (CNTs:HAp-5) and to 28.6±1 nm (CNTs:HAp-10).

Results and Discussion
In good agreement with previous studies [45], the XRD patterns of HAp and CNTs:HAp composites demonstrate that the powders exhibit the apatite characteristics with good crystal structure and no new phase or impurity was found.
The phase and the chemical structures of the CNTs:HAp nanocomposite were analyzed by FTIR and Raman spectroscopies.The ATR-FTIR technique was used to investigate the CNTs:HAp nanocomposites.Figure 2 shows the ATR-FTIR spectra of pure HAp, CNTs:HAp-5, and CNTs:HAp-10 nanocomposites.It has been reported that the vibration peaks at 1645 cm −1 correspond to H-O-H bands of water lattice [37].The band at 630 cm −1 characteristic to Intensity (a.u.) the stretching vibrational mode of [OH] − was present in both HAp and CNTs:HAp (CNTs:HAp-5 and CNTs:HAp-10) composites spectra and indicated a well-crystallized HAp [46,47].According to previous studies conducted by Yang et al. [48] and Pan et al. [49], the band at 1095 cm −1 suggests also the formation of a well-crystallized apatite.
On the other hand, the bands at 600 cm −1 and 560 cm −1 appearing from ] 4 [PO 4 ] 3− and indicated also a wellcrystallized HAp [47].The band at 962 cm −1 is due to ] 1 [PO 4 ] 3− mode of phosphates while the bands at 1015 cm −1 and 1095 cm −1 can be attributed to the ] 3 [PO 4 ] 3− mode of phosphates [50,51].According to Ciobanu et al. [37,46], the characteristic bands for carbonate at 878 and 1421 cm −1 are probably due to the combination of ] 2 -out-of-plane bend vibration and ] 3 -asymmetric stretching vibration.In the ATR-FTIR spectra, the bands of phosphates and the carbonate were observed for all samples.The characteristic peak of CNTs at 2359 cm −1 reported in literature by Niu et al. [52] was not observed in the CNTs:HAp nanocomposites.On the other hand, the bands around 2919 cm −1 and 2852 cm −1 which are due to the asymmetric and symmetric stretching of C-H bonds reported by Krishna Kumar et al. [53] were not observed in the NTCs:HAp nanocomposites spectra.The FTIR spectra of CNTs:HAp-5 and CNTs:HAp-10 nanocomposites exhibited similar peaks to those of pure HAp except for the peak at 1335 cm −1 which appeared in CNTs:HAp-10 nanocomposites spectrum.According to Wahab et al. [54] the peak at around 1335 cm −1 is characteristic to the vibrational mode of carbonate ions in hydroxyl sites.Following these investigations, we can say that the presence of CNTs in the CNTs:HAp nanocomposites does not change the basic structure of HAp.
The vibrational bands attributed to asymmetric ] 3 (P-O) stretching [37] are observed at the wavenumbers of 1045 cm −1 (] 3 ) and 1075 cm −1 (] 3 ) for raw HAp while they shift to lower wavenumber at 1042 cm −1 (] 3 ) and 1070 cm −1 (] 3 ) for CNTs:HAp-5 and CNTs:HAp-10 nanocomposites, respectively.The proper incorporation of CNTs was confirmed by the presence of two common bands, in the region 1300-1600 cm −1 for both the CNTs:HAp-5 and CNTs:HAp-10 spectra, related to the D and G bands.For bulk CNTs, these bands that underline the degree of crystallinity and order in the graphitization process appear at 1345 cm −1 and 1572 cm −1 .Once CNTs were mixed with HAp, these bands shifted to lower frequencies at 1355 cm −1 and 1580 cm −1 for the D and G bands, respectively.Previous studies [55][56][57] showed that the peaks at 1580 cm −1 (in CNTs spectrum) and 1575 cm −1 (in CNTs:HAp-5 and CNTs:HAp-10 nanocomposites spectra) were associated with the tangential displacement of C-C band stretching mode of the well-graphitized nanotubes.
On the other hand, the peaks at 1355 cm −1 (in CNTs spectra) and 1345 cm −1 (in CNTs:HAp-5 and CNTs:HAp-10 nanocomposite spectra) were associated with the vibration of carbon atoms in the disorder mode.Comparing the D-mode and G-mode of CNTs and CNTs:HAp-5 and CNTs:HAp-10 nanocomposites, a little change in the frequency value was observed which can be associated with an increase in the defect density of graphite after being mixed into HAp [58].
The nanostructure and morphology of pure HAp, CNTs, and CNTs:HAp-5 and CNTs:HAp-10 nanocomposites were studied by SEM (Figure 4).The SEM images of CNTs:HAp-5 and CNTs:HAp-10 revealed that the nanoparticles were uniformly distributed without forming any agglomeration.The CNTs packages have been destroyed as a result of the formation of CNTs:HAp nanocomposite.
In previous studies, Zheng et al. [59] showed that the CNTs tended to aggregate in water due to the van der Waals interactions.Compared to pure and nonfunctionalized CNTs samples presented in Figure 4, the CNTs:HAp nanocomposite could be dispersed well due to the presence of HAp.

Biochemical Studies.
As it can be observed from Figure 5(a), a dose-dependent decrease in cell viability was registered for CNTs treated cells after 24 hours of exposure, whereas HAp stimulated cell proliferation confirming its biocompatibility effect.At the same time, the cell viability for low doses of CNTs:HAp-5 was not significantly changed (only a 10% decrease for 25 g/mL compared with control).Regarding the highest doses (100 g/mL and 500 g/mL CNTs:HAp-5), the viability was diminished by 17% and 20%, respectively, compared to the control.Exposure to CNTs:HAp-10 induced a decreased viability only for doses higher than 2.5 g/mL (a 28% decrease for the concentration of 500 g/mL compared to control).
Further, our results (Figure 5 compared to 24-hour one and the cell viability was slightly decreased compared to control after the incubation with HAp.CNTs:HAp-5 nanocomposites stimulated cell proliferation in a dose-dependent manner suggesting a higher biocompatibility compared with CNTs:HAp-10 nanocomposites which were more toxic at higher doses (100 g/mL and 500 g/mL).The significant increase of cell viability for CNTs:HAp-5 after 48 hours could be explained by the fact that there is a time interval necessary for osteoblasts to adapt to the environment created by the nanocomposites.Humans are well equipped with an excellent antioxidant system to fight against oxidative stress induced by a various number of damaging effects, protecting efficiently against the formation of free radicals.Thus, being the most prevalent endogenous intracellular thiol antioxidant, GSH plays a central role in neutralizing free radicals and reactive oxygen species.
After 24 hours, a dose-dependent increase of GSH level was noticed in the cells exposed to simple CNTs, whereas a decrease was observed at 48 hours of treatment, suggesting the CNTs toxicity (Figure 6).Regarding the incubation with CNTs:HAp nanocomposites, the GSH content was near to the control values and HAp levels after the first 24 hours diminished only by 20% compared with control for the highest dose, proving their biocompatibility.The above results are similar to the studies performed by Patel and Kwon [60], in which higher GSH levels were observed after 24 hours of treatment and lower values after 48 hour exposure of A549 epithelial cells to multiwalled carbon nanotubes (MWCNTs).The initial increase could be due to a higher demand of GSH to combat the rapid formation of reactive oxygen species (ROS), followed by the depletion of this tripeptide that was rapidly consumed in the resulted oxidative stress.
The level of MDA was determined as a marker of lipid peroxidation in G-292 cells exposed to 25 and 100 g/mL of CNTs, HAp, CNTs:HAp-5, and CNTs:HAp-10 (Figure 7).
After both time intervals of incubation with simple CNTs or CNTs:HAp nanocomposites, MDA contents increased in a dose-and time-dependent manner compared with the control, but the peroxidation for nanocomposites was below the CNTs level.Thus, the nanocomposites showed a lower toxicity compared to CNTs, suggesting their biocompatibility on human osteoblasts through the presence of HAp.In the case of pure HAp exposure, the MDA levels were near the control values, suggesting the absence of hydroxyapatite toxicity.In addition, the oxidative stress induced by CNTs

Conclusions
In this study, we obtained CNTs:HAp nanocomposites through a simple adapted coprecipitation method, which was followed by a characterization of chemical, structural, and biological properties in order to establish their biocompatibility on human osteoblasts.XRD analysis results revealed a good crystal structure, without any peaks corresponding to the addition of CNTs or impurities.Comparing the D-mode and G-mode of CNTs and CNTs:HAp composites, a minor change in the frequency value was observed.All biochemical parameters investigated with the help of in vitro methods presented a variation dependent on time and concentration.Taken together, low doses of CNTs doped hydroxyapatite nanocrystalline powders did not significantly decrease G-292 cell viability and did not induce oxidative stress after 2 days of exposure.These results support the applicability of CNT:HAp nanocomposites prepared by the coprecipitation method for medical purposes because they exhibit good biocompatibility and have great potential in the future for orthopedics and prosthetics.

3. 1 .
Characterization of Carbon Nanotubes and Hydroxyapatite Composite Materials.X-ray diffraction patterns of pure HAp and CNTs:HAp (CNTs:HAp-5 and CNTs:HAp-10) nanocomposites in the 2 ranges 20 ∘ -90 ∘ and 20 ∘ -30 ∘ , respectively, are shown in Figure 1.The characteristic diffraction pattern of CNTs:HAp-5 and CNTs:HAp-10 nanocomposites shows a similar behavior of the spectrum for pure HAp (Figure 1(b)).The peak observed at 2 = 26.38 ∘ , characteristic to CNTs, made no contribution to the specific peak of hydroxyapatite (2 = 25.84 ∘ ), which shows that the peaks corresponding to the addition of CNTs did not appear.It can also be seen that the diffraction peaks of CNTs:HAp composites are in agreement with those of pure HAp indexed to the hexagonal Ca 10 (PO 4 ) 6 (OH) 2 in  63m space group (ICDD-PDF number 9-432).

Figure 7 :
Figure 7: MDA levels in G-292 cells exposed to CNTs, HAp, and CNTs:HAp-5 and CNTs:HAp-10 at concentrations of 25 g/mL and 100 g/mL for 24 and 48 hours.Values are calculated as means ± SD ( = 3) and expressed relative to control.*  < 0.05 versus control.