Bone regeneration by hydroxyapatite-gelatin nanocomposites

In the current research, the preparation and characterization of a series of new biocompatible injectable bone paste (IBP) nanocomposites, hydroxyapatite-gelatin (HA-Gel), and hydroxyapatite-gelatin-alendronate (HA-Gel-Ald np) are reported. IBP nanocomposites, HA-Gel np, were synthesized from mixing different ratios of gelatin to aqueous solutions of both Ca(NO3)2.4H2O and (NH4)2HPO4, while the target nanocomposites, HA-Gel-Ald np, were obtained by submitting aqueous solution of alendronate (Ald) to HA-Gel np nanocomposites. This composite crystallinity was analyzed by FTIR and XRD, and their morphology was characterized by scanning electron microscopy (SEM) and EDX measurements. XRD patterns, SEM, and EDX presented changes in the crystal and surface structure from HA to HA-Gel np to HA-Gel-Ald np. These physico-chemical measurements indicated the success in isolating the nanocomposites, HA-Gel np and HA-Gel-Ald np, with different ratios. Furthermore, the cytotoxicity of the nanocomposites on stem cells was assessed using MTT assay. Although the cytotoxicity data show significant effect of the prepared IBP nanocomposites (p = 0.00), their interaction together had no significant effect (p = 0.624). • New biocompatible injectable bone paste (IBP) nanocomposites, hydroxyapatite-gelatin (HA-Gel), and hydroxyapatite-gelatin-alendronate (HA-Gel-Ald np). • These composites crystallinity was analyzed by FTIR and XRD, SEM, and EDX measurements. • The cytotoxicity of the nanocomposites on stem cells was assessed using MTT assay.


Introduction
Tissue engineering is considered as an interdisciplinary field and involves biomaterials science, cell biology, cellmaterial interactions, and surface characterization, i.e., to enhance tissue functions [1]. Stem cell-based tissue engineering has the potential to revolutionize medicine with the ability to regenerate damaged and diseased tissues [2,3]. Scaffolds can be laden with cells for implantation [4]. Human bone marrow mesenchymal stem cells (HBMSCs) can be harvested from the patient, induced to differentiate into osteoblasts and combined with a scaffold to repair bone defects [5].
Osteoporosis is bone deformation disease in which bone mineral density (BMD) reduced, with micro-architecture disrupted, bone strength reduction and influenced by increasing of bone fracture risk [1,6]. Osteoporosis is caused by an internal factor, reduction of bone ability to do bone remodeling process, due to unbalanced process of osteoblast and osteoclast, while bone defect is caused by external factor [7]. Osteoporosis could be treated by increasing the bone density or filling the bone defect with a suitable material.
Gelatin (Gel, derivative of collagen; Fig. 1ii) is a natural denatured polymer, composed of amino acids (hydroxyproline, proline, or sequences such as RGD -arginine-glycine-aspartic acid). Gel is denatured product of collagen and shows excellent biocompatibility and biodegradability as well as strong affinity to mammalian cells [13]. It is known as a suitable biomaterial to mimic the extracellular matrix because of its function groups and the possibility to form 3D scaffolds with porous structure, i.e., it can be used in tissue engineering based on its biocompatibility and biodegradability [14]. HA-Gel composites have been reported as a bone substitute material with high biocompatibility and non-toxicity [15].
Bone tissue is composed of a large number of calcium phosphate minerals (about 70%), collagen, and polysaccharides (about 30%); thus, many research articles were focusing on applying composites of biodegradable natural polymers and bioactive ceramics, to mimic the extracellular matrix in the composition and avoid the brittleness of ceramics and low mechanical properties of polymers [16][17][18][19]. Gelatin particularly has always been used as the matrix of the composite due to its highly similar composition but lower costs than collagen, and hydroxyapatite (HA) nanoparticles have been selected as favorable fillers to reinforce the composite [20]. Increasing HA nanoparticle content in HA-Gel composite would promote cell attachment, proliferation, and increase levels of alkaline phosphatase and gene expression of osteogenic differentiation [21].
Bisphosphonates (Bps; Fig. 1iii) are important family of drugs for the treatment of bone tissue diseases, such as osteoporosis, bone metastases, hypercalcemia, and Paget's disease, due to their high affinity to the bone mineral hydroxyapatite [22]. Bps are characterized by the [P-(R1) C (R2)-P unit] (Fig. 1iii), which allows a great number of possible variations, by changing the two lateral chains (R1 and R2) on the carbon atoms. The changes in R1 or R2 moiety can lead to extensive alterations in their physicochemical, biological, therapeutic, and toxicological properties [23]. Extensive structure activity studies showed useful drugs that combine potent inhibition of bone resorption osteoclastic action [24]. Bps are significant and dose-dependent for osteoblast genesis [25].
Oral administration and intravenous injection have some adverse problems, such as gastrointestinal disorders, flulike symptoms, and low oral bioavailability. Therefore, it is necessary to improve the delivery method of Bps [26]. Bps drugs could help bone defect healing. Alendronate (Ald), which has high affinity towards Ca 2+ ions of HA, improves the interaction with bone calcium and inhibits the osteoclast in bone engineering process [27].
Gelatin-hydroxyapatite (Gel-HA) composites are of great interest in biomedical applications due to their highly mimic structure and function of bone extracellular matrix, excellent biocompatibility, biodegradability, and bioactivity [28][29][30].
Porous scaffolds are considered as the key components in bone tissue engineering to repair or replace the defective bone tissues, i.e., provides a temporary space to support cell in-growth and formation of new bone tissue as well as transportation of nutrition and waste. The interconnected porous scaffolds are particularly interesting as they support a comparatively homogeneous formation of new tissue. Three porous gelatin-HA composite-based scaffolds have been prepared, and their microstructure and mechanical property were evaluated [31]. Moreover, hydrogels (three-dimensional polymeric networks) are widely used as polymeric scaffolds in tissue engineering [32].
The sol-gel process technique was applied to synthesize HA-Gel composites. It was applied as nanoparticle aggregation in a mixture of hydroxyapatite and tricalcium phosphate (HA-TCP) to form a ceramic slurry. The scaffolds obtained by polymer foam replication technique showed a morphology with adequate porosity for tissue engineering [33]. Upon adding Ald to (Col-HA np hydrogels), the biological performance of the scaffolds was improved, and the nanoparticle aggregation suppressed due to the steric hindrance [34]. The presence of Bps in the network improved the cytocompatibility of the scaffolds and suppressed the nanoparticles aggregation by inducing steric repulsion [35]. In addition, a scaffold of hydroxyapatitecollagen (HA-Col) with bisphosphonate (Bp)-derivatized liposomes has been prepared and developed as a drug {carboxyfluorescein (CF), doxorubicin (DOX), and lysozyme (LYZ)} delivery system in bone regeneration and repair [36]. It can prolong the in situ residence of model drugs and has the potential to provide a sustained drug release platform in bone regeneration and repair. The effect of hydroxyapatite (HAnp) nanoparticles loaded with the simplest bisphosphonate (medronate) as a bone-targeting agent and JQ1, a small-molecule bromodomain inhibitor, as a chemotherapeutic in different 2D and 3D K7M2 OS in vitro models has been examined and discussed [37]. In addition, the nano hydroxyapatite-chitosan-gelatin (HA np-CS-Gel)-based scaffold have been reported as promising composites for bone tissue engineering and regeneration. These scaffolds are non-cytotoxic based on their test by VERO cells [38].
Recently, we have reported the synthesis and cytotoxicity of new magneto nanocomposites, (Fe 3 O 4 )-HA np incorporated in alendronate {(Ald); (mag-HA np)} scaffolds [39]. These scaffolds are promising and can be applied for various orthopedic applications, in which such particles could be injected, their location controlled using an external magnetic source, and bone growth promoted.
In principle, biodegradable polymers with sufficient mechanical strength and optimized structure are desirable scaffold materials for bone tissue engineering [1,40]. These materials should also be osteoconductive so that osteoblasts can adhere to the scaffold and migrate, differentiate, and eventually form new bone. Thus, the novelty of this study is based on.
-The surface modification of HA can adjust its surface properties and may improve the phase compatibility between HA and polymers. Thus, the modified HA-Gel composites may also show much better mechanical properties and promote their application in bone tissue engineering. -The addition of Ald to HA-Gel composites is characteristic bone filler for osteoporotic bone. -the preparation of series of nanocomposites, HA-Gel np and HA-Gel-Ald np, by applying different ratios of HA:Gel and (HA-Gel):Ald, respectively. -The physical and chemical properties of the nanocomposites and their in vitro cytotoxicity were assessed. -These composites were tested against BMSC adhesion, and the flow cytometric analysis was also applied for the immune-phenotypic profile of the BMSCs collected from the bone marrow.

Material
All manipulations were performed under aerobic conditions using materials and solvents as received. Chemicals and solvents were obtained from Sigma and Merck. Calcium nitrate, diammonium hydrogen phosphate, gelatin, and alendronate sodium trihydrate were purchased from Alfa Aesar (USA).

Instrumentation
Ultrasonication was performed on Digital ultrasonic bath CD-4830, 35 kHZ, China. The centrifuge apparatus used is Safety centrifuge Thermo Fisher Scientific Company, USA. FT-IR spectra were measured on a Matson5000 FT-IR spectrometer. SEM measurements were carried out using a Hitachi S-3000VP-SEM Variable Pressure-SEM. EDX measurement were performed using energy dispersive X-ray unit (EDX, JEOLJSM-5500LV SEM, JEOL Ltd, Japan) [41]. X-ray powder diffraction patterns were assessed using X-ray diffractometer (XRD, PA-NalyticalX' Pert PRO, Netherlands, The Electron Microscope Unit, Mansoura University, Egypt) equipped with mono-chromatized Cu-Kα radiation (λ = 1.542 Å, 50 kV, and 40 mA). Sample analysis was covered the 2θ range from 4° to 80° at a scan rate of 0.02° s −1 [42]. The size of the crystallites was calculated using the XRD data to determine the Full-Width Half Maximum (FWHM). Then, by applying Scherrer formula, D = K ∕BCos , D, K, λ, B, and θ are the crystallite size (nm), Scherer constant (0.9), the wavelength of X-rays (1.54056 A°), the line broadening at FWHM in radians, and the Bragg's angle in degrees, respectively [43]. The absorbance for Elisa data were measured using Accu-Tell Elisa washer (ABEW-2), Faculty of Medicine, Mansoura University, Egypt.

Preparation of IBP nanocomposites (HA-Gel np)
The IBP nanocomposites were prepared by chemical wet precipitation method.

Note:
a) All the synthesized nanocomposites were selected in such amount that Ca/P molar ratio was maintained at 1.67 [48]. b) The IBP nanocomposites were manually pulverized using a mortar and pestle for fine powder formation [49].

Biological characterization and cytotoxicity assay
In vitro cytotoxicity assays are considered as indicators to measure the ability of cytotoxic compounds to cause cell damage or cell death. They are widely used in fundamental research and drug discovery for screening toxic compounds. MTT assay has been widely used to assess cell viability, based on the enzymatic reduction of 3-[4,5-dimethylthiazole-2-yl]-2,5-diphenyltetrazolium bromide (MTT) to MTTformazan, which is catalyzed by mitochondrial succinate dehydrogenase [24]. Hence, the MTT assay is dependent on mitochondrial respiration and indirectly serves to assess the cellular energy capacity of a cell. It is a colorimetric reaction that can be measured from cell monolayers that have been plated in 35 mm dishes or multi well plates. Disks of 5 mm diameter and 2 mm thickness were prepared by compressing the IBP nanocomposites powders into cylindrical mold. The disks were sterilized under UV-lamp for 3 h and washed 3 times with phosphate buffer saline, PSB (10 min for each wash).
The test was performed by using rat bone marrow mesenchymal stem cells (MSCs). The material was compressed in the form of disks and each disk was placed in in 96-well plates, followed by seeding of cells as suspension over the disks (5 × 10 3 cells in each well; the control was cells without any samples' disks). The micro well plate was incubated for 24, 48, and 72 h at 37 °C. MTT solution was added onto each well and incubated for 4 h at 37 °C. After incubation, 200 μl dimethyl sulfoxide solution was added to each well and incubated for 15 min to dissolve the insoluble formazan. The measurement process was performed by Elisa reader showing the violet level as optical density (OD), which represents the cell viability of the material based on Eq. 1 [50].
where OD is the absorbance of sample when MSCs cells were seeded in a 96-well plate and treated with the leach liquor, while blank is the absorbance of the sample which only contained growth medium, and control was the absorbance of the sample which contained cells and growth medium.
Statistical analysis Three repeated tests were performed in all the experiment and the results were presented as means ± standard deviation (SD). One-way analysis of variance (ANOVA) and Tukey's post hoc tests were used to compare between groups, while values with and without magnetic field were compared using Paired (dependent) test. A value of p < 0.05 was considered of statistical significance.

Immunophenotype characterization of bone marrow-derived mesenchymal stem cells (BMSCs)
Phosphate buffer saline (PBS) was used to rinse the bonemarrow-mesenchymal stem cells (BMSCs) of passage three.
Then the cells were resuspended in PBS (0.5 mL). Rabbit polyclonal anti-CD90, mouse monoclonal anti-CD105, and rabbit polyclonal anti-CD45 antibodies (Abcam, Cambridge, United Kingdom) with fluorescein isothiocyanate (as fluorophore) were added in separately, followed incubation 30 min at 4 °C in dark. Labeled MSCs were rinsed in PBS, centrifuged at 2000 rpm for 5 min, and resuspended in PBS. A BD Accuri C6 flow-cytometer with the BD Accuri C6 program software installed was used to determine the immunophenotypes of labeled MSCs.

Results and discussion
Gelatin (Gel; derivative of collagen) is a natural denatured polymer, composed of amino acids (hydroxyproline, proline, or sequences such as RGD -arginine-glycine-aspartic acid). Gel is known as a suitable biomaterial to mimic the extracellular matrix because its function groups and the possibility to form 3D scaffolds with porous structure, i.e., it can be used in tissue engineering based on its biocompatibility and biodegradability [31,32]. Gelatin-based scaffolds, such as polycaprolactone-58S bioactive glass-sodium/alginategelatin [51], have been prepared. HA np (Ca/P molar ration 1.65 ± 0.1 and surface Ca/P atomic ratio 1.30 ± 0.05) has high affinity of interaction of positively charged Ca 2+ with alendronate phosphate (PO 3 2− )-richer surface groups [52].

FTIR spectral studies
FTIR spectral data are used to evaluate the functional groups of the synthesized composites {HA-Gel np (Fig. 2i) and HA-Gel-Ald np (Fig. 2ii)}. HA-Gel np nanocomposites are likely formed by the interaction between the critically small sized HA np and Gel molecule [45]. The FTIR spectrum of HA np showed peaks at 1059 and 988 cm −1 , attributed to ν as (P-C-P) and ν s (P-C-P) stretching vibrations of phosphate groups, respectively [46], while that at 3568 cm −1 is due to ν(OH) stretching vibration. Furthermore, bands at 678, 575, and 525 cm −1 were assigned to the δ(O-P-O) bending of phosphate [21,46,53].
The spectrum of Gel shows two characteristic bands at 2923 and 2851 cm −1 , assigned to ν as (CH 2 ) and ν s (CH 2 ) stretching vibrations, and those at 1647 and 1558 cm −1 corresponding to ν as (COO − ) and ν s (COO − ) stretching vibrations of carboxylate group [21]. The band at 3433 cm −1 attributed to ν(OH) stretching vibrations of hydrogen bond water. Two extra bands at 1683, 1537 and 1240 cm −1 belong to amide I (C = O), amide II (N-H) and amide III band (the plane of ν(C-N) and δ(N-H) group in amide) vibrations, respectively [54].
In the nanocomposite, HA-Gel np, and upon increasing Gel content (from G1 to G2 to G3), the intensities of IR bands of HA and Gel were increased, as well as the stretching vibration ofν(OH)based on H-bonds formation [21,54]. All G1 to G2 to G3 spectra exhibited broad band ~ 3100-3500 cm −1 due to ν(OH) and ν(NH) stretches [21]. In addition, the characteristic band of the amide I in free Gel was shifted to lower wavenumber (1655 cm −1 ) in HA-Gel np, indicating the electrostatic attraction interaction between Ca 2+ ions (in HA) and O-C = O − groups (in Gel) [55], causing elongation on C = O bond, with shift to lower wavenumber. All these features indicate the interaction between HA np and Gel to form HA-Gel np nanocomposites with different ratios [54,55].
In the FTIR spectrum of NaHAld, both ν as (P-C-P) and ν s (P-C-P) stretching vibrations were observed in the range of 800-655 cm −1 with the multiple bands in 3600-3300 cm −1 range, assigned to ν(OH) stretching of P-OH and C-OH groups. The ν as (P-O) and ν s (P-O) stretching vibrations are observed at 1057 and (1021, 920) cm −1 , respectively, while the stretching and bending vibrations of P = O and P-OH groups giving rise to the bands in the fingerprint region below 1320 cm −1 . The bands at 1234 and 1183 cm −1 due to ν(P = O) and ν(P-O) stretches, respectively [56,57]. The spectrum shows also bands at 3350 and 3246 cm −1 attributed to ν as (NH 2 ) and ν s (NH 2 ), respectively [56].
In the FTIR spectra of the composites HA-Gel-Ald np, broadening was observed in the bands at the 1620, 1200-900 and 700-500 cm −1 regions, indicates the formation of interlinked bonds into the layer between hydroxyapatite PO 4 −3 and P-OH/P-O bonds specific to Ald [58], and interaction between Ca 2+ ions in HA and HPO 3 − anion of Ald − [7].

X-ray diffraction spectroscopy (XRD)
The nanocomposites (HA-Gel np) were fabricated by using different ratios of HA:Gel (G1: 7:3; G2: 1:1; G3: 3:7 wt%, respectively. The XRD patterns of HA-Gel np were observed in Fig. 3 [60]. In case of HA-Gel-Ald np nanocomposite, the presence of Ald does not change the lattice constants of HA np phase [61]. The amorphous Gel patterns were interfered with the peaks of crystalline HA and Ald [16]. In addition, the increase of HA-Gel-Ald nps size in comparison to those of HA-Gel nps may be attributed to the high affinity of Ald − (HPO 3 − ) to Ca 2+ ions in HA, which form some pores with increasing its size [7]. The broadness of the XRD peaks indicates the lower in crystallinity upon adding Ald. Table 1 lists the crystal sizes (nm) of HA-Gel np and HA-Gel-Ald np composites based on Scherrer equation (Eq. 2) [62].
where L and K are the nano crystallite size and is a dimensionless shape factor, while λ is the XRD radiation of wavelength (nm), β is the line broadening at half the maximum intensity, and θ is the Bragg angle.
(2) L = K ∕( .Cos )   The addition of Ald greatly affected the composites crystal size, because of the interaction of Ca 2+ ions in HA Ald − anion, which may not alter the crystal structure of HA (Fig. 4). However, Ald − anion affects the crystal dimensions and leads to larger composite crystals' size [63,64]. Moreover, this feature is supported by the high affinity of Ald amine group to HA np phosphate groups, i.e., H 2 N--PO 3 −3 binding [37].

Scanning electron microscopy (SEM) and energy dispersive X-ray (EDX) spectroscopy
The morphology of HA-Gel np nanocomposite was assessed via SEM analysis ( Fig. 5a and b). The porous structure of HA-Gel np is regular arrays of circular hollow channels with particle size of about 67.75 nm. When the HA-Gel was up to 7:3 (G1), the morphology of the surface showed apparent pores in comparison to the pure HA [23]. The porous structure may provide more space for cell growth. Upon adding more Gel, the HA np became more compact, since Gel filled the gaps between the HA np [23]. In the ratio HA-Gel (1:1; G2), no pores were observed on the surface, while in case HA-Gel (3:7 G3), HA np were barely observed on the surface. Upon adding Ald, the size was reduced with diameter of about 36 nm. In addition, Ald particles are randomly distributed in HA-Gel np matrix, i.e., some are embedded in the pore wall, and some are piled together between or within pores, reducing the pores size and shape [65]. It can be observed that a large amount of flaky deposition appeared on the surface, and to clarify the composition of these depositions, EDX analysis was subjected to demonstrate that the deposition mainly contains C, O, P, and Ca elements. The elemental composition of HA, HA-Gel and HA-Gel-Ald was performed by EDX analysis (Fig. 6). HA np has 4.54 (C), 46

Biological characterization and cytotoxicity assay
Rat bone marrow mesenchymal stem cells (MSCs) were used to evaluate the cytotoxicity assay of the fabricated compounds, HA np, HA-Gel nps (G1, G2 and G3) and HA-Gel-Ald nps (G4, G5, and G6), using MTT assay. Figure 7 illustrates the obtained results. The absorbance at 570 nm represented the vitality of the MSCs cells in the MTT assay. As reported in Tables 2 and 3, two ANOVA test revealed that the composition variable had a significant effect on the crystal size (p = 0.00). The concentration, composition, and interaction of composition and concentration variable have significant effects (p = 0.00).
After MSCs were cultured in leach liquor of HA-Gel nps and HA-Gel-Ald np, the cells viability of were around 74  and 94%, respectively. These data showed that the fabricated nanocomposites with low toxicity and could be used as preliminary estimate for the proliferation tests of MSCs on the materials. Cytotoxicity properties are based on the CD 50 , as the cell viability was below 50% and the material is toxic [50]. All studied groups showed cell viability higher than 70%, i.e., all the synthesized composites are biocompatible and enhance cell proliferation.
In addition, the presence of Ald on HA-Gel-Ald nps showed higher cells viability than those of HA-Gel nps. E. Boanini et al. [66] reported that the Ald-modified groups increase cell viability and improve osteoblast proliferation, cell attachment, and spreading [50].
Moreover, HA is the main ingredient of the human bone, could improve the biocompatibility of the nanocomposites. As the concentration of HA decreased, the cell viability increased, which may be attributed to the decrease in Ca 2+ ions which would inhibit the biocompatibility [54,67].

Immunophenotype characterization of Bone marrow-derived mesenchymal stem cells (BMSCs) & flow cytometric analysis for BMSCs
Flow cytometric analysis for the cell surface markers on bone-marrow-mesenchymal stem cells (BMSCs) showed positive results for the antibodies, rabbit polyclonal anti-CD90 (65.2%), and mouse monoclonal anti-CD105 (69.6%), while negative results were obtained for rabbit polyclonal anti-CD45 (3.4%) antibodies. These results confirmed the immune-phenotypic profile of the BMSCs and the adequate isolation and collection of these cells from the samples of bone marrow.

Conclusion
In conclusion, porous scaffolds are playing very important role in bone tissue engineering to repair or replace the defective bone tissues. Thus, in this study, porous nanocomposites scaffolds, HA-Gel np and HA-Gel-Ald np, were fabricated. These composites have been characterized based on spectral (FTIR, XRD, SEM) and EDX measurements. Gel is amorphous in nature, and its crystallinity is affected by HA np indicate the formation of HA-Gel np. The presence of Ald in the nanocomposites, HA-Gel-Ald np, decreases the crystallinity due to the high affinity of Ald amine group to HA np phosphate groups, i.e., H 2 N--PO 3 −3 binding. MTT assay was applied to study the cytotoxicity of the reported nanocomposites. The prepared composites are considered nontoxic, and a positive correlation between (HA-Gel np and Ald) content and cell viability was suggested. These biodegradable nanocomposites are expected to be intelligent compounds, and the biomimetic composite scaffold of HA-Gel-Ald np could be suggested as a promising material to promote osteoblast cell growth in bone tissue engineering. In addition, the flow cytometric analysis showed that cell surface markers on the BMSCs were positive for both CD90 and CD105 while negative for CD45 antibodies, confirming the immune-phenotypic profile of the BMSCs and the adequate isolation and collection of these cells from the samples of bone marrow.
Data availability All data generated or analyzed during this study are included in this article.

Declarations
The authors declare no conflict of interest. This work is original research that has not been published previously and not under consideration for publication elsewhere. All the authors listed have approved the manuscript that is enclosed. All figures in this manuscript are non-published and original. There are no personal circumstances or interest that may be perceived as inappropriately influencing the representation or interpretation of reported research results. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.