Extraction of bone minerals
The diaphysis of the bovine femurs was boiled in tap water for 2–3 hours, followed by washing with tap water to remove the undesirable tissues, including the bone marrow, tendons, and muscles. The bones were cut into small pieces, delipidated using acetone for 2 hours, and washed several times with distilled water. To remove the organic matter, they were dried in 100 °C overnight, followed by incubating at 850 °C for 3 hours. They were finally powdered by Balmill.
To evaluate the purity of the extracted hydroxyapatite, X-ray differentiation (XRD, Bruker D8 Advance) was performed, and the results were compared with the commercially prepared hydroxyapatite (Merck). After pulverizing, the samples were mounted on a holder. XRD was performed with = λ 1.54 A˚ (10˚-90˚ range) and Cu Kα as the radiation source.
Decellularization of Bone Tissue
Sheep scapula was cut into 0.5 cm pieces, and after rinsing with phosphate buffer saline (PBS), they decalcified in 0.5 N HCl for 48 hours. Then, they were incubated in 1% SDS for 48 hours and rinsed with PBS to remove the trace of SDS from the tissue. All steps were performed at room temperature and on the stirrer.
Evaluation of the decellularized tissues
Hematoxylin & Eosin and Hoechst staining were performed to evaluate the presence of nuclei in the tissues. To quantify DNA content, QIAamp® DNA Blood and Tissue Mini Kit (Qiagen GmbH, Hilden, Germany) was used. According to the manufacturer's instruction, 25 mg of lyophilized decellularized and intact bones was incubated in proteinase K at 56 °C until it was completely lysed. The samples were transferred to a microcentrifuge tube and washed by a buffer. DNA was eluted by adding ethanol and centrifuged in a mini spin column and extracted DNA quantified by spectrophotometer at γ = 260 nm, using the NanoDrop® ND-1000 (Nanodrop Technologies Inc., Wilmington, DE, USA).
Histochemical staining was done to evaluate the retention of ECM content after decellularization. Alcian blue (pH 2.5), and PAS staining were performed to detect GAGs and neutral sugars, respectively. Collagen and elastic fiber retention was confirmed by Masson's trichrome and aldehyde fuchsin staining.
Immunohistochemistry
Frozen sections at 5 µm thickness were prepared and fixed with 2.5% paraformaldehyde. The endogenous enzyme was blocked by incubating the samples in 3% H2O2 in methanol. Non-specific binding sites were blocked by incubating the samples in PBS containing 10% goat serum and 5% Bovine Serum Albumin (BSA). The samples were incubated in anti-collagen type I, -laminin and -fibronectin antibodies (All from Abcam PLC, Cambridge, MA, USA) at dilutions 1/250, 1/100 and 1/250, respectively. The sections were then incubated with hours radish peroxidase/streptavidin (1:10000; Abcam, USA) at room temperature for 20 minutes. Finally, diaminobenzidine was added as the chromogen.
Immunofluorescence was performed to detect osteopontin as well. To do this, frozen sections were fixed with 4% paraformaldehyde. Non-specific binding sites were blocked by incubating the samples in PBS containing 10% goat serum and 5% BSA. Then, the samples were incubated in anti-rabbit osteopontin primary antibody (1/100, Abcam, UK) overnight. The sections were incubated in Alexa flour 488-conjugated secondary antibody (ab150077) for 45 minutes and observed by fluorescence microscopy (Olympus, BX61).
Scanning Electron Microscopy (SEM)
The ultrastructure of the decellularized matrix was evaluated by SEM. The samples were fixed with 2.5% glutaraldehyde for 20 min, dehydrated by gradually increasing concentration of ethanol, and dried in gradually increasing concentration of Hexamethyldisilazane (HMDS; Merck, Kenilworth, NJ, USA). A gold replica was prepared from each sample by Q150R- ES sputter coater (Quorum Technologies, London, UK), and micrographs were taken using a VEGA3 microscope (TESCAN, Brno, Czech Republic) at 10 kV accelerating voltage.
Extraction of the Extracellular Vesicles from MG63 cell line
MG63 osteosarcoma cell line (purchase from Pasteur institute, Iran) were grown in RPMI-1640 culture medium (Bioadia) at 37 °C and 5% CO2 for 36–48 hours, and then their supernatant was collected. ECVs were isolated from the culture medium by centrifugation at 300 g for 20 minutes. Then, the supernatant was filtered by 0.2 µm microfilter. The samples were then centrifuged at 100 000 g and 4 °C twice, each one for 90 min by an ultracentrifuge (Beckman, USA). The pellet was mixed with 100 µL PBS and stored at -70 °C until use.
Characterization of Extracellular Vesicles
Transmission electron microscopy was used to evaluate the size and shape of the extracted ECVs. Flow cytometry was also performed to detect the surface markers on the ECV membrane. To do this, magnetic beads were vortexed 30 seconds, and then 20 µL of ECVs containing 16 µg was added to the magnetic beads and mixed on a shaker at 2–8 °C overnight. After washing with PBS, the tube was exposed to a magnetic field, the supernatant was removed, and 400 µL PBS was added. This stage was repeated twice. The ECV-bead complex was transferred to flow cytometry tubes and 20 µL of PE-conjugated anti-CD9, -CD63, and -CD81 antibodies (all from BD Pharmingen™) were added. ECVs were analyzed by FL2 channel of a flow cytometer (BD FACSCaliburTM, BD Biosciences). The data were analyzed by FlowJo software.
Internalization of ECVs
MG63 cell line was cultured in the presence of RPMI containing 10% FBS (Gibco), 1% L-glutamine (Bioidea), and 1% penicillin/streptomycin (Gibco) at 37 °C and 5% CO2. ECVs were labeled with PKH26, using PKH26 red fluorescent cell linker kit (Sigma) according to the manufacturer's protocol. Briefly, 0.5 mL of ECVs mixed with 498 µL of dilution buffer and 2 µL PKH26 and incubated for 5 min and diluted with FBS at a ratio of 50/50. Then, the mixture was centrifuged at 100 000 g for 90 minutes at 4 °C. The pellets were mixed with 1 mL of the culture medium. Labeled ECVs were added to MG63 cell line cultures and incubated for 22 hours. Finally, the internalization of the labeled ECVs was evaluated by fluorescence microscopy. For negative control, the same amount of dilution solution without ECVs was added to the cells.
Dynamic Light Scattering (DLS)
ECV size was evaluated by nanoparticle analyzer (SZ-100 Horiba, Japon) equipped with a 532-nm wavelength, 10 mW power, and operating at an angle of 173. ECVs were transferred to cuvettes (ZEN0040, Malvern, Herrenberg, Germany). The measurements were made at a fixed position and at 25 °C. Three independent measurements were performed for each sample, and three samples were analyzed, and the mean value calculated.
Injectable Hydrogel Fabrication
Collagen extracted from the rat tails was lyophilized, and then 750 µL of 0.005% collagen was reconstituted with 125 µL of RPMI 10X. To neutralize acidity, 125 µL of the reconstructed buffer contained 2.2% sodium bicarbonate, and 4.8% HEPES (both from Sigma) was added. To prepared DB gel, a solution containing 1% ground DB in 3% acetic acid was stirred at 4 °C for three days, and then pepsin (20U/g of DB dry weight) was added to dissolve the DB powder.
To prepare hydrogel, we mixed reconstituted tail rat collagen and solubilized DB at the ratio of 2:1, and then, 0.33 mg of hydroxyapatite powder per mL was added. Then, ECV at a concentration of 110 µg/mL was added to the collagen/DB mixture and incubated at 37 °C for thermogelation.
In vitro Study
MG63 cell line (1.5 × 104 cell/well) was cultured on the 3D scaffolds along with 2D conventional culture conditions in the presence or absence of 100 µg ECVs at 37 °C and 5% CO2 for one, three, and seven days. The culture medium was RPMI-1640 containing 10% FBS, 1% L-glutamine, and 1% penicillin/streptomycin.
Cytotoxicity Assessment
To compare cell viability and proliferation in the hydrogel with/without ECVs, we replaced the culture media with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, 0.5 mg/mL) for three hours on day one, three, and seven. Then, MTT was replaced by dimethyl sulfoxide, and the optical density of the eluted formazan was measured at 595 nm.
Mineralization Assessments
MG63 cell line cultured on 3D scaffolds and 2D conventional conditions with/without ECVs in the same condition as reported for cytotoxicity assay. After one, three, and seven days, the culture media were discarded, and the cell-containing scaffolds were washed with PBS, fixed with 75% ethanol for 20 min, and stained with 500 µL of 2% alizarin red S for 30 min. Then, the dye was eluted by incubating the cultures in 100 mM cetyl pyridinium chloride (Rad kimiagaran) for one hour. The concentration of the eluted dye was measured by spectroscopy at a wavelength of 405 nm.
Besides, the culture media were collected to assay the alkaline phosphates (ALP) activity. To evaluate the enzyme activity, a commercial kit (Parsazmoon, Iran), based on the capability of ALP to convert nitrophenyl to yellow nitrophenol, was used. The culture media were treated according to the manufacturer's instruction, and the intensity of yellow nitrophenol was measured using a microplate reader at 405 nm.
Cell Attachment and phenotype
To evaluate the cell attachment capability, 1.5 × 104 cells were grown on both 3D scaffold, and polystyrene culture dish. After two hours, the amount of non-adherent cells was calculated and subtracted from the original cell number. To evaluate cell phenotype on the scaffolds, the same number of cells was seeded on a 3D scaffold with or without ECVs for two days and prepared in the same way as described above.
In vivo studies
Forty male rabbits weighting 2.5-3 kg were purchased from the comparative animal research center, Shiraz, Iran. All animals were treated according to the ethics committee guidelines (IR.SUMS.REC.1397.277).
Experimental Design and Surgery
Scaffolds were prepared as described for in vitro study. The same ratio of collagen, DB, HA, and ECVs was mixed to form an injectable solution for the experimental group. Forty rabbits were divided into four groups. Each group was divided into two subgroups; each one was followed up for either two or eight weeks.
The animals were anesthetized by 50 mg/Kg ketamine and 3.5 mg/Kg xylazine. An incision was performed in the skin parallel to the base of the mandible, and masseteric muscle was exposed. While avoiding the facial artery, we created a defect size of 1 × 1 cm on the right mandible. The mandibular defects in the control and experimental groups were filled with collagen (control), collagen containing DB + HA + ECVs, collagen containing DB + HA, and collagen containing ECVs, respectively. The gel was inserted into the defect, and then the periosteum and skin were sutured, using Viryl 3 − 0 and Nylon 3 − 0 (Supa Medical Devices Co., Tehran, Iran).
After 2 and 8 weeks of follow up, the rabbits were killed and their mandibles were removed. X-ray radiography was performed with an X-ray machine (Planmeca Intra, Helsinki, Finland).
Morphometrical Analyses
The mandibles were fixed in 10% buffer formalin and decalcified in a decalcifying solution containing 4% HCl and 4% formic acid for 36 hours. Paraffin-embedded samples were cut at 5 µm thickness and stained with H&E. Three sections from the first, middle, and end of each sample were chosen. Photographs were taken from systematic randomly selected fields, started from one edge of the grafted scaffold, and ended at the other edge. The number of osteoblasts, osteocytes, osteoclasts, and also the area occupied by the bone, connective tissue, adipose tissue, and vessels were estimated by the ImageJ software (https:// imagej.nih.gov/ij/index.html) or (NIH, Bethesda, MD).