Biocomposite scaffold preparation from hydroxyapatite extracted from waste bovine bone

Abstract This study was conducted to fabricate scaffold from polylactic acid (PLA) and hydroxyapatite (HA) extracted from waste bovine bone for enhancing both mechanical and biocompatible properties. After pretreatment in dilute NaOH solution, the bone was calcined at 900°C for 6 h, ball milled and converted to HA. Factors that affect the formation of HA were investigated. Experimental results showed that HA particles with crystal size < 100 nm and 99% crystallinity could be obtained at 90°C, pH 11 and 35 mM H3PO4 solution followed by 4 h calcination at 900°C. By using non-solvent induced phase separation method, PLA scaffolds with pore size and surface area of 22.6 μm and 25.7 m2/g, respectively, containing different hydroxyapatite were successfully prepared. Tensile strength of scaffolds increased due to effective support by HA grafted collagen. PLA scaffolds containing HA were more degradable than PLA scaffolds and PLA scaffolds containing HA grafted collagen. Cell culture results showed that cell density increased significantly on porous scaffolds than that on the dense scaffolds. Moreover, cell density also increased significantly on the scaffold containing HA grafted collagen than that on the scaffold with pure HA.


Introduction
Tissue engineering method consists of harvesting cells from a patient, expanding them in culture, and seeding them into scaffolds which can later be implanted into the patient. The scaffolds known as artificial architecture for

Introduction
Let F denote a eld and let V denote a vector space over F with nite positive dimens pair A, A * of diagonalizable F-linear maps on V, each of which acts on an eigenbasis for irreducible tridiagonal fashion. Such a pair is called a Leonard pair (see [13, De nition 1.1 A, A * is said to be self-dual whenever there exists an automorphism of the endomorphis swaps A and A * . In this case such an automorphism is unique, and called the duality A The literature contains many examples of self-dual Leonard pairs. For instance (i) the L ated with an irreducible module for the Terwilliger algebra of the hypercube (see [4,Coroll Leonard pair of Krawtchouk type (see [10, De nition 6.1]); (iii) the Leonard pair associated module for the Terwilliger algebra of a distance-regular graph that has a spin model in th bra (see [1,Theorem], [3,Theorems 4.1,5.5]); (iv) an appropriately normalized totally bip (see [11,Lemma 14.8]); (v) the Leonard pair consisting of any two of a modular Leonard t De nition 1.4]); (vi) the Leonard pair consisting of a pair of opposite generators for the bra, acting on an evaluation module (see [5,Proposition 9.2]). The example (i) is a special examples (iii), (iv) are special cases of (v).
Let A, A * denote a Leonard pair on V. We can determine whether A, A * is self-dual in By [13,Lemma 1.3] each eigenspace of A, A * has dimension one. Let {θ i } d i= denote an or values of A. For ≤ i ≤ d let v i denote a θ i -eigenvector for A. The ordering {θ i } d i= is whenever A * acts on the basis {v i } d i= in an irreducible tridiagonal fashion. If the ordering then the ordering {θ d−i } d i= is also standard, and no further ordering is standard. Similar A * . Let {θ i } d i= denote a standard ordering of the eigenvalues of A. Then A, A * is self-dual is a standard ordering of the eigenvalues of A * (see [7,Proposition 8.7]).
biosynthetic extracellular matrix (ECM) are used to create a 3D environment in order to promote cellular attachment, migration, proliferation and differentiation. Therefore, finding materials to produce scaffolds that possess not only properties suitable for cell penetration, capillary ingrowth, and adequate solute transport but also good mechanical and biological properties is a very important work. Some synthetic polymers have been applied to create scaffolds such as poly(glycolic acid) (PGA), poly(lactic acid) (PLA) [1], poly(lactic-co-glycolic acid) (PLGA) [2], poly(caprolactone) (PCL) [3], poly(propylene fumarate) (PPF), and poly(ethylene glycol) (PEG) [4]. However, mechanical properties of these synthetic polymers are very weak. Some fillers and reinforced materials such as calcium phosphate, ceramic and glass were mixed with polymers to enhance their mechanical properties [5][6][7][8]. However, fillers and reinforced materials usually have low biocompatibility. Therefore, it is necessary to find potential materials that can be applied in scaffold. Recently hydroxyapatite is considered as a good material to replace ceramic [9].
Hydroxyapatite [Ca 10 (PO 4 ) 6 (OH) 2 ] with molar ratio of Ca/P from 1.58 to 1.67 is well known as the mineral component of animal bones [10]. Hydroxyapatite (HA) was found to have good biocompatibility with cells [11]. HA can be synthesized by numerous techniques such as sol-gel, chemical precipitation, hydrothermal, hydrolysis, and microemulsion [12]. HA synthesized from chemical such as CaO, Ca(NO 3 ) 2 , (NH 4 ) 2 HPO 4 did not possess desired biological characteristics while these produced from natural materials such as egg shells or animal bones are very interesting since it is believed to be more biocompatible and biological safe [2][3][4][5]. However, high temperature treatment during extraction process may cause structure breaking or loss of certain components which will partly affect the biological activity of HA. In Vietnam, the yearly amount of bovine bones produced from food processing industry is huge. These bones are often used to produce low economic value animal feed. In this study, waste bovine bone was converted into HA and then applied to produce biocompatible scaffold by using phase separation method.
A simple nonsolvent induced phase separation method was widely used to prepare polymeric membrane [13,14]. This method is also considered as a potential method to create polymeric porous structure, especially scaffolds used in tissue engineering. In the nonsolvent method, polymer was dissolved in organic solvents such as tetrehydrofuran (THF), chloroform to form homogeneous polymer solution. The solution was then separated by adding nonsolvent such as water, acetone and ethanol to form porous structure. In this work, the nonsolvent induced phase separation method was applied to create porous scaffold from polylactic acid (PLA) and hydroxyapatite for tissue engineering.

Materials
Bone of adult bovine (2-3 years old) was collected from a local slaughterhouse. The bone was first cooked at 100°C for 24 h to remove tissue, then dried at 80-100°C for 24 h and cut into small sizes of about 10 mm × 5 mm × 5 mm. Chemicals such as NaOH, H 3 PO 4 and NH 3 , 3-amino propyl triethoxysilane (APTES), glutaraldehyde (GA), and phosphate buffered saline (PBS) were supplied by Merck Vietnam for use in pretreatment and extraction of HA from bovine bone. Poly lactic acid (PLA) (CAS. 9051-89-2) was purchased from Nature Works. The solvents used in the preparation of scaffold such as chloroform (CHCl 3 ) and ethanol were supplied by Merck, Vietnam. Standard hydroxyappatite (≥ 90%) was supplied by Sigma-Aldrich, Vietnam.

Extraction of hydroxyapatite from bovine bone
After pretreatment with 0.1 M NaOH for 24 h, the pretreated bone was washed with water and acetone to remove fats and other impurities. The bovine bone was then calcined at 900°C for 6 h in a furnace (Nabertherm, LHT 1750°C) to remove all organic components in the bone. The calcined bovine bone was then grinded into powder with particle size of 5 to 10 µm by using high speed mill ball. The bovine powder (10 g) was dispersed in 100 mL distilled water. The mixture was then reacted with aqueous H 3 PO 4 solution.
The reacted mixture was cooled to room temperature and neutralized by NH 3 in 2 h. After reaction, the mixture was aged overnight before separating by filter. The sample was dried at 100°C before continuously calcining at high temperature to stabilize the structure of HA. Factors that affected HA formation such as reaction temperature, reaction time, pH and concentration of H 3 PO 4 were investigated. X-ray diffraction (XRD), scanning electron microscopy (SEM), and Fourier transform infrared spectroscopy (FTIR) were applied for material characterization.

Preparation of hydroxyapatite grafted collagen
HA extracted from bovine bone was dispersed (1 wt%) in a co-solvent of ethanol/water (1:2 v/v). The suspension was added sequentially APTES 2 wt% and NH 3 6 wt% while stirred at 100 rpm in an environment of N 2 for 24 h. HA-APTES was collected and washed with distilled water 3 times. The HA-APTES was dispersed (1 wt%) into aqueous solution of 1% glutaraldehyde at room temperature and stirred at 100 rpm for 24 h. The mixture was filtered and washed by PBS solution to obtain the HA-APTES-GA particles. The particles were dispered 1% in PBS solution containing 100 mg/mL collagen extracted from catfish skin. The mixture was then stirred at 100 rpm for 24 h at 4°C. HA grafted collagen obtained was washed three times with PBS solution.

Preparation of scaffold
Nonsolvent induced phase separation method was applied to prepared scaffold. Poly lactic acid (PLA) was dissolved in chloroform at room temperature for 24 h. After that the polymer solution was casted onto a teflon mold (5 mm × 60 mm × 100 mm). The teflon mold was then immersed in a coagulation bath containing nonsolvent (ethanol/water) and PLA porous scaffold was obtained after 24 h.

Cell culture 2.5.1 Cell seeding
Fibrobast L929 standard cell suspension containing 20,000 cells/mL was prepared by diluting the cell suspension with serum medium. One milliliter of the cell solution was distributed in 24-well plate containing samples and incubated at 37°C in 5% CO 2 atmosphere for 24 h. After a certain incubation time, the culture medium was removed and the unattached cells was washed by PBS solution. Cell density was characterized by both lactate dehydrogenase (LDH) assay and confocal laser scanning microscope.

Confocal laser scanning microscope
Confocal laser scanning microscope (LSCM, Leica-SP2, Germany, 40× 0.75 dry, 60× 0.75 oil or 100× 1.4 oil objective) was used to image the surface morphology of stained cells. Fluorescence with excitation wavelength (Ex) = 364 nm and emission wavelength (Em) = 454 nm were used to image cells stained by DAPI; and fluorescence with Ex = 495 nm and Em = 520 nm were used to capture images of cells stained by Phalloidin-FITC.

Statistical analyses
Variations of the amount of adherent and proliferate cells were statistically analyzed by performing one-way analysis of variance (ANOVA), executed by Minitab statistical software®. Fisher's pairwise comparison test was applied to compare cell density on different samples for p < 0.05.

Hydroxyapatite extraction from waste bovine bone
Bovine bone was pretreated to remove tissue, fats and other impurities. As shown in Figure 1, the color of bone was

Cell density by lactate dehydrogenase (LDH) assay
Cell density was determined by lactate dehydrogenase assay [15][16][17][18]. The attached cells were lysed with 100 μL of 1% Triton X-100 in PBS. The reaction solution contained 0.3% NAD + , 0.27% diaphorase, 0.03% bovine serum albumin, 1.2% sucrose, 0.02% iodonitrotetrazolium, and 3.6% sodium lactate (all chemicals were purchased from Sigma). Optical density values were read at 490 nm by using enzyme-linked immunosorbent assay (ELISA, iMark TM Absorbance Reader, BioRad). A standard curve was obtained by plotting the measured optical density of a series of cell solutions verses known cell densities.

Immunofluorescence staining for microscopy observation
The     main components in bovine bone powder. CaO and P 2 O 5 accounts for 44 wt% and 36 wt% of bovine bone powder, respectively (Table 1).

Effect of concentration of H 3 PO 4 on HA product
Results of XRD diffraction analysis of the calcined bovine bone show peaks of HA at 2θ (degree) at 10 Figure 2). It means that after calcining, crystal structure of HA, β-TCP and CaO still exist. Thus, in order to fully convert β-TCP and CaO that was created during calcination, H 3 PO 4 with different concentrations was used to react with bovine bone powder. Results show that as concentration of H 3 PO 4 increased, the peaks of β-TCP and CaO gradually disappeared and the intensity of peaks for HA increased. The peaks of β-TCP and CaO almost disappeared when using 35 mM H 3 PO 4 and its characteristic peaks is not very different with the peaks of standard HA (Figures 2d and 2e). Calculated base on the Scherrer equation, the size of HA crystal was 83, 85, 86 and 88 nm and degree of crystallinity was 88%, 93%, 94% and 96% corresponding to concentration of H 3 PO 4 0 mM, 25 mM, 30 mM and 35 mM, respectively (Table 2). Therefore, H 3 PO 4 35 mM is considered as a good concentration to convert bovine bone to HA.

Effects of reaction temperature on formation of HA
Reaction temperature was varied from 60°C to 90°C to investigate its effect on HA formation. From the XRD diffraction (Figure 3), HA samples synthesized at 60°C, 70°C and 80°C showed that besides the characteristic peaks of HA, there still existed a few characteristic peaks of β-TCP. At 90°C, only the characteristic peaks of HA appeared. This result shows that reaction rate increased with increasing reaction temperature and at 90°C high impurity HA was created. This agrees with the results of Luo [19]. Moreover, besides increasing intensity, the HA peaks also become sharper which means that crystallinity increased with increasing reaction temperature. As temperature was increased from 60°C to 90°C, the degree of crystallization of HA increased from 93.5% to 97.7%,  accompanying with the increase of crystal size from 80.3 to 95.7 nm (Table 3). This result is consistent with the study of Wijesinghe et al. [20]. Therefore, 90°C is the suitable temperature to produce HA from bovine bone.

Effect of pH on formation of HA
HA was synthesized at different pH values to investigate pH effect on the morphology and structure of HA. The XRD diffraction in all samples showed the characteristic peaks of HA appear (Figure 4). However, at pH 9, pH 10 and pH 12, there were few characteristic peaks of β-TCP. This result is consistent with the previous study of Peipei Wang et al. [21]. HA synthesized at pH 11 showed the characteristic peaks of HA and reach the highest crystal level of 97.7% (Table 4). The size of the crystal tends to increase with increasing pH and this is consistent with the study of Liu et al. [22].

Effect of calcination temperature on HA formation
Temperature was varied from 700°C to 1000°C to investigate its effect on the formation of HA. Results of XRD diffraction analysis showed that the characteristic peaks of HA existed in all samples ( Figure 5). The samples synthesized at 700°C and 800°C still showed small characteristic peaks of β-TCP phase and these peaks disappeared as temperature was increased to 900°C. At 900°C, only the characteristic peaks of HA appear and these peaks become narrower and sharper at 1000°C. Moreover, as temperature was increased from 700°C to 1000°C, crystallization degree increased from 83.7% to 99.0%, accompanying by the increase of crystal size from 90.6 nm to 101.8 nm (Table 5).

Preparation of biocompatible composite scaffold
Scaffolds in different concentrations of HA and HA grafted collagen were prepared by using nonsolvent induced phase separation method. PLA solution (125 g/L) was prepared in chloroform at room temperature. The polymer solution was mixed with HA and casted onto a teflon mold (5 mm × 60 mm × 100 mm). The teflon mold was then immersed into a coagulation bath (25°C, volume ratio ethanol to water 96/4; volume ratio solvent to nonsolvent 1/20). PLA porous scaffolds with different HA contents were obtained after for 24 h. SEM images show porous  structure and its average pore size is 22.6 ± 4 µm ( Figure 8). BET measurement resulted in a surface area of 25.7 m²/g. Mechanical property of scaffolds was evaluated by tensile strength measured on a Zwick/Roell BDO equipment -FB050TN. Tensile strengths were 88.5 MPa, 76.5 MPa, 62.3 MPa, 40.1 MPa, and 10.6 MPa for scaffold samples containing 0%, 5%, 10%, 15%, and 20% HA, respectively ( Figure 9). The results mean that the interface of HA particles and PLA was not compatible leading to decrease in tensile strength when HA content was increased. Compatibility was significantly improved for collagen grafted HA. Tensile strength of scaffold increased with increasing HA grafted collagen from 5% to 10% and after that decreased from 15% to 20%.

Degradation of scaffold in PBS solution
Scaffold samples containing 10% particles were prepared with size of 1 × 1 cm 2 and immersed in 10 mL PBS solution at 37°C. After 7 days, degradation of scaffolds was 0.45%, 0.5% and 0.75% for PLA, PLA-HA-Col and PLA-HA, respectively. The degradation of scaffold increased with time. After 28 days, degradation was 1.75%, 3.4% and 4.3% for PLA, PLA-HA-Col, and PLA-HA, respectively. The results show that all scaffolds could be degraded in PBS solution ( Figure 10). The degradation is highest for the PLA scaffolds containing HA. However, after grafting with collagen the degradation is lower than PLA scaffold, meaning that the interaction between PLA and HA-Coll is better than that of PLA and HA. , (e) and (f) are the images from PLA scaffol ds that contained HA ranging from 5%, 10% to 15% respectively and (g), (h) and (k) are the images from PLA scaffolds that contained HA grafted collagen ranging from 5%, 10% to 15%, respectively (scale bar = 50 μm).

Cell responses on scaffolds
Biocompatibility of scaffold was studied based on cell responses. PLA scaffolds containing different concentrations of HA particle and collagen grafted HA particle were evaluated by directly culturing L929 cell line on the samples and characterized by lactate dehydrogenase assay and confocal microscopy. Cell density is shown in Figure 11. Cell density increased significantly on porous scaffolds compared to that on dense scaffold. Cell density also increased sightly as HA content was increased from 0% to 10% and increased significantly as HA content reached 15%. Moreover, cell density also increased significantly on scaffold containing HA grafted collagen compared to that on scaffold with pure HA. Confocal microscopy images show that fibroblasts attached on PLA dense and PLA scaffold without HA with limited number after 24 h cultivation (Figures 12a and 12b). It is similar to LDH results, confocal microscopy images also show that cell number increased approximately with increasing HA concentration in porous scaffold (Figures 12d-f), suggesting that cell proliferation was promoted by HA. Moreover, it was observed that higher cell density was found on surfaces containing HA grafted with collagen (Figures 12g-k).

Conclusion
An effective method was sucessfully applied to produce hydroxyapatite from waste bovine bone. It was possible to create HA particles with a crystal size < 100 nm with a crystallinity of 99% under conditions of H 3 PO 4 35 mM, reaction temperature 90°C, pH 11, calcination at 900°C for 4 h. Bioccompatible PLA scaffolds prepared by combining HA promoted cell growth effectively depending on HA content. Collagen extracted from catfish was successfully grafted on HA surface and it enhanced cell proliferation compared to pure HA.