Hydrothermal extraction and physicochemical characterization of biogenic hydroxyapatite nanoparticles from buffalo waste bones for in vivo xenograft in experimental rats

Hydroxyapatite (HA) can be used in odontology and orthopedic grafts to restore damaged bone due to its stable chemical characteristics, composition, and crystal structural affinity for human bone. A three-step hydrothermal method was used for the extraction of biogenic calcined HA from the buffalo waste bones at 700 °C (HA-700) and 1000 °C (HA-1000). Extracts were analyzed by thermogravimetric analysis, differential scanning calorimetry, X-ray fluorescence, X-ray diffraction, Fourier transform infrared spectroscopy, scanning electron microscopy, and in vivo examination of HA xenografts for femoral lesions in experimental rats. Crystallinity, purity, and morphology patterns showed that the HA main phase purity was 84.68% for HA-700 and 88.99% for HA-1000. Spherical HA nanoparticles were present for calcined HA-700 samples in the range 57–423 nm. Rats with critical bone lesions of 3 mm in diameter in the left femur treated with calcined HA-700 nanoparticles healed significantly (p < 0.001) faster than rats treated with HA-1000 or negative controls. These findings showed that spherical biogenic HA-700 NPs with a bud-like structure have the potential to stimulate both osteoconduction and bone remodeling, leading to greater bone formation potential in vivo. Thus, the calcined biogenic HA generated from buffalo waste bones may be a practical tool for biomedical applications.


Thermogravimetric analysis and differential scanning calorimetry (TGA/DSC)
The thermal breakdown and crystallization behavior of B-raw bone powders were studied from 25 to 900 °C in a nitrogen environment at a heating rate of 15 °C/min using a thermal analysis instrument (SDT Q600DSC-TGA, USA) [18][19][20][21][22][23][24] .Thermogravimetric analysis (TGA) was employed to record the weight loss changes in bone powder with increasing temperature.Differential scanning calorimetry (DSC) was used to quantitatively and qualitatively describe the physicochemical characteristics induced by endothermic and exothermic processes in bone powders as the temperature changed.

X-ray diffraction (XRD) analysis
An X-ray diffractometer (GNR, APD 2000 Pro, Italy) with a CuK target (λ = 1.5406Å) and a voltage and current setting of 40 kV and 30 mA was used to examine the crystalline structure of all three HA powders, as described earlier 18,25,26 .Analysis of the XRD patterns of B-raw, HA-700, and HA-1000 samples was performed in the range 2θ = 10°-80° with a step size of 0.05°.The standard diffraction data of the Joint Committee on Powder Diffraction Standards (JCPDS) reference 00-009-0432 for the HA phase was used for phase identification 28 , while the crystallite size (X c , nm) of the HA samples was calculated using Scherrer's equation: 29 where k is a constant equal to 0.9, λ is the X-ray wavelength (1.54 Å), β is the full width at half maximum (FWHM, radians), and θ is the diffraction angle of the reflection plane at [002].
The sample crystallinity (D c ) was calculated using the Landi equation: 18 where D c is the crystallinity, V 112/300 is the intensity of the hollow between (112) and (300) peaks, and I 300 is the intensity of the (300) diffraction peak of the biogenic HA powders.

X-ray fluorescence (XRF) spectroscopy analysis
A hand-held X-ray fluorescence (XRF) analyzer (Delta Premium, Olympus, Waltham, MA, USA), which can detect elements from 12 Mg through 83 Bi, was used to quantify elemental composition of B-raw buffalo femur (1) X c = k /βCosθ (2) D c = 1 − V 112/300 /I 300 bone, as previously described 23,[27][28][29] .The XRF was calibrated with a standard reference material before any measurements were made, and quality control was monitored using software 28 .Mineral content of B-raw samples was measured by scanning three locations on the femur midshaft for two minutes each.The percentage of elements that has measurement errors less than 5%, converted to mg/g units, was calculated by dividing the peak area of each element by the total area of all elements in the scan 27 .

Experimental design
Twelve adult albino rats were unilaterally lesioned at the left femur bone using a motorized drill and split into three equal experimental groups (n = 4).The animals received sham and xenograft HA treatments for 4 weeks.Negative controls with non-treated femoral lesions made up Group I; animals treated with HA-700 for their bone lesions made up Group II; and animals treated with HA-1000 for their bone lesions made up Group III.

Animals
Albino rats (n = 12) of two-month age and weighing 220-250 g at the start of the experiment were randomly divided into three equal groups (n = 4) with ad libitum drinking water and were kept on standard chow.The rats were housed in large rectangular cages (49 × 34 × 16 cm), each of which could house up to 4 adult rats, and each group was allowed to acclimate to the natural daylight cycle for one week.The Institutional Animal Care and Use Committee (IACUC #0121961521) approved the study's protocol before it began to ensure it followed the US National Institutes of Health's Guide for the Care and Use of Laboratory Animals (NIH publication no.85-23, revised 1996) 35 .In addition, the procedures used and described here are consistent with the standards set forth by ARRIVE guidelines.The Ethics Committee of the Medical Research Institute, Alexandria University, Alexandria, Egypt, approved the study protocol.

Surgical procedure
Rats were anesthetized by injecting ketamine (50 mg/kg) and Lidocaine HCl (20 mg/ml) i.p.After the left femur was shaved, the periosteum and soft tissues were fixed, and a motorized drill was used to create a 3-mm-diameter lesion in the femur bone all the way to the medullary canal (as shown in Fig. 1A and B).Afterwards, xenograft treatment with HA-700 or HA-1000 was used to fill the femur bone lesion (Fig. 1C).After repositioning the skin and muscles, a 5-0 suture was used to close the wound (Fig. 1D).

Histology
After 4 weeks of sham and xenograft HA treatment, the rats were anesthetized once again, and a segment of the bone lesion region was removed from the left femurs.Afterwards, longitudinal serial sections 4-5 μm thick were cut from the samples with a microtome.The samples were then put in a 10% buffered formaldehyde solution for 24 h, and then they were decalcified in a 10% EDTA solution.Hematoxylin and Eosin (H&E) staining was used to analyze tissue sections under the microscope 36,37 .

Blinding and masking
Four different researchers worked on each animal as follows: one researcher (MIB) assigned the animals to treatment groups and administered the sham and xenograft HA treatments following a randomization table.This researcher was the only one who knew which group received which treatment.A second investigator (MHM) administered the anesthesia, and a third (SAA, MIB, MHM) performed the surgery.The lesions and histological findings in the femur bones were finally evaluated by a fourth investigator (RAM, EIM), who was also blinded to the treatment.

Outcome measurements
After 4 weeks of sham and xenograft HA treatment, all the animals showed signs of healing, which was measured both in terms of the size of the inflicted lesion on the femur and its histological appearance under the microscope.

Statistical analysis
An analysis of variance (ANOVA) followed by Tukey's post hoc test of significance was used to compare the size of femur lesions between groups using the SPSS statistical package (Version 16; Chicago, IL, USA).The results were displayed in the form of means and standard deviations (SD).A p-value less than 0.05 was taken to show statistical significance.

Results and discussion
The increasing annual demand for biogenic HA in healthcare is due primarily to the growing number of patients worldwide who require orthopedic treatment.In this study, biogenic HA was extracted from buffalo waste bones using a three-step hydrothermal method at varying temperatures.The following techniques were used to characterize and evaluate biogenic HA extracts for in vivo xenografts in experimental rats.

TGA/DSC analysis
Derivative thermogravimetric analysis (DTGA), TGA, and DSC curves of B-raw bone powder sample decomposition due to temperature are shown in Fig. 2. Table 1 shows the DTGA temperature intervals, weight loss percentage, maximum decomposition peak temperature, and sample residues after combustion.Weight loss was initially seen in the range 38.66-169.57°C of the DSC endothermic peaks due to the release of absorbed water and moisture content, accounting for 7.58% of the total mass with the maximum DTGA at 64 °C (Fig. 2A) 18,19 .A further 28.31% of the total mass is lost between 178.54 and 576.67 °C, which is characterized by a DTGA peak at 338.6 °C and a TGA sharp slope change (Fig. 2A and B), due to the combustion of bone's organic components  including collagen polymer fibrils, which triggers the exothermic HA recrystallization process [20][21][22] .The final weight loss seen in the range 585.63-883.33 °C, accounting for 7.27% of the total mass with a maximum DTGA at 830 °C (Fig. 2A), corresponds to the endothermic peak in the TGA curve at 700 °C (Fig. 2B), owing to the slow elimination of the CO 2− 3 ions from the HA lattice 31,32 .These observations are consistent with those of the subsequent FTIR spectroscopy analysis, which confirms the crystallization of B-raw bone powders and the loss of CO 2− 3 groups due to high temperatures.According to reports, thermal decarbonation commonly occurs between 400 and 600 °C in air and between 500 and 890 °C in a nitrogen environment 23 .No substantial weight loss related to the CO 2− 3 decrease in the DTGA curve (Fig. 2A) was observed between 583 and 700 °C; this loss only becomes significant at 750 °C or higher 38 , reflecting thermal stability.In line with earlier studies, raw bone powder calcined at 700 °C gives the best thermal stability and avoids the sintering effect under drastic changes in temperature for the extraction of HA 37,39 .Further temperature increases from 830 to 1000 °C result in minimal weight loss (⁓ 0.45%, DSC curve Fig. 2C) suggesting that HA crystals have formed entirely at that point.

XRD analysis
Crystallographic structure and phase purity of raw and calcined HA samples extracted at 700 °C and 1000 °C are shown by XRD patterns in Fig. 3. B-raw samples showed broad and poorly defined peaks at 2θ values of 25.85, 31.79,39.83, and 46.74, respectively, which could not be reliably identified due to the presence of any leftover organic components as well as from the elastic and inelastic dispersion of the HA NPs crystals (Fig. 3A) 9,16 .The characteristic diffraction peaks of calcined HA-700 samples (Fig. 3B) found at 2θ = 25.78,31.79,32.21, 32.91, 34.01, 39.83, 46.74, 49.54, and 52.12, respectively, are well matched with the standard reference of the JCPDS 00-009-0432 for the HA phase 25 .The crystallite size (X c ) of HA-700 samples was 44.84 nm, with an 84.68% crystallinity (D c ), based on the diffraction peak at 2θ = 25.78 corresponding to the reflection plane [002].Crystal size increased to 55.74 nm, and crystallinity increased to 88.99%, as shown by the narrower and sharper diffraction peaks of HA-1000 samples (Fig. 3C).In a similar earlier study, calcination was shown to increase the size of the crystals from nano to microscale and decrease the FWHM value, neither of which is directly associated with an improvement in crystalline quality 16,40 .Moreover, we did not observe any structural transformations or signs of secondary phases in the HA-700 and HA-1000 samples into beta-tricalcium phosphate (β-TCP) or calcium oxide (CaO), neither by hydrothermal treatment nor by calcination.
The FWHM and computed X c at the diffraction peak 2θ= 25.78, which corresponds to the reflection plane [002], for B-raw, HA-700, and HA-1000 bone samples are shown in Fig. 3D and E. The higher FWHM value of 0.40 rad was associated with an X c of 20.37 nm for B-raw samples.In similar studies, it was argued that the well-ordered and nanometric nature of biogenic HA nanocrystals from porcine origin resulted in broad XRD peaks, high FWHM values 16,40,41 , and crystallite size of comparable values (i.e., 22 nm) 42 .After the calcination, the values of FMWH dramatically decreased, which was responsible for an increase in the X c for HA-700 and HA-1000 (i.e., 44.84 and 55.74 nm, respectively).Evidence from XRD peaks and a similar decrease in the FWHM suggests that crystal growth is initiated during the calcination process of biogenic bovine HA 43 .

XRF spectroscopy analysis
B-raw bone elements with measurement errors of less than 5% among locations on the femur midshaft were selected and shown in Table 2: Ca, Cu, Fe, Mg, Mn, P, K, S, Si, Zn, and Ca/P ratio.The values of Ca, Fe, Mn, P, and Zn for buffalo femur B-raw bones agree well with those for buffalo humerus bones by Buddhachat et al. 27 , who employed the same technique for studying elemental composition in horns, teeth, and humeral bones of 14 different species, including humans, pigs, sheep, and buffalo.Moreover, the Ca/P ratio for buffalo femur B-raw bones was similar to that for humerus buffalo bones by the same group (i.e., 3.85 ± 0.05 vs. 3.70 ± 0.12, respectively) 27 .Furthermore, the values of Ca, P, K, and other bone minerals as well as the Ca/P ratio were similar to qualitative XRF elemental analysis in cow femur bones by Akindoyo et al. 23 , and recent findings by Yılmaz et al. 44 , in adult male and female guinea pigs.

SEM analysis
The SEM images of surface morphology for the raw and calcinated bone powder samples at a standard magnification of 30,000× are shown in Fig. 5. Like earlier studies 33,34 , B-raw samples showed the presence of HA Table 3. Band positions for B-raw, HA-700, and HA-1000 bone samples and reference band positions of earlier studies by Fourier transform infrared (FTIR) spectroscopy analysis.www.nature.com/scientificreports/microparticles with an average size of 0.58 ± 0.47 μm of irregular shape and rough surface, together with organic components (Fig. 5A).Calcinated HA-700 samples showed a broad bud-like particle structure, with small spherical HA NPs of 57 nm on top of others of 423 nm size (Fig. 5B).Further higher calcination temperatures yielded aggregated particles of irregular shape in the range from 63 to 639 nm, as shown for HA-1000 samples in Fig. 5C.It has been pointed out earlier that after the calcination of raw bone powders at 700 °C, particles become more regular and spherical in shape 34 .However, the majority of the HA extracted from mammalian bone exhibits irregular shapes, with some investigations showing the existence of flakes, rods, needles, and plate-like shapes [52][53][54][55][56] .
During the calcination process, HA samples experience many changes, including crystallite size, shape, and crystalline quality, as well as dehydrogenation and Mg release at high temperatures 57 .It has recently been shown that the reduction in FWHM in biogenic bovine HA samples produced by controlled calcination at temperatures ranging from 400 to 720 °C is due to coalescence mechanisms directly related to crystallite size increase 58 .Furthermore, the XRF analysis in Table 2 shows that the B-raw bone samples have a significant concentration of Mg; thus, the nanosized spherical particles seen in SEM micrographs for the HA-700 and HA-1000 samples correspond to MgO in the calcined samples 42,58,59 .
The bud-like structure of HA-700 NPs of two different sizes is remarkably original and has never been seen previously.Therefore, we suggest using the bud-like HA-700 NPs to treat critical-sized bone lesions.In the section that follows, it will be shown how treating critical-sized femoral lesions in a rat experimental model with in vivo xenografts may affect the biodegradability of bone during bone healing.

In vivo study on experimental animals
Photomicrographs at 200× of the left femur shaft of the negative control animals after four weeks showed large lesion sites filled with irregularly formed, unevenly stained, widely separated trabeculae of newly woven bone with mild inflammatory cell infiltration next to the lesion area (Fig. 6A).Unevenly stained bone trabeculae showing larger, densely grouped, and irregularly shaped osteocyte lacunae were also seen at 400× (Fig. 6B).Moreover, rat's left femur shaft treated with calcined HA-700 NPs after the same period showed lesion sites filled with new bone except for a few small, localized areas, small islands of cartilage, and an acidophilic normal compact bone (Fig. 6C) with normal osteocytes inside their lacunae (Fig. 6D).There were no inflammatory or foreign body reactions when using spherical calcined HA-700 NPs, which were osteoconductive and exhibited early biosorption, according to previous recent investigations, demonstrating healthy bone repair 36,37,60,61 .These results also lend credence to in vivo investigations showing that the substitution of CO 2− 3 ions in the HA-700 NPs' lattice structure induces a favorable affinity for osteoblast cells, boosting cellular adhesion and collagen synthesis [62][63][64][65] .Furthermore, the rat's left femur shaft treated with HA-1000 particles after four weeks showed few widely trabeculae of woven bone, with mineralized and unmineralized areas, at the upper part of the lesion, with a partial bridging bon at the center at 200× (Fig. 6E).A persistent large lesion was also seen at 400× with trabecular bone, including a few osteocytes of irregular shape (Fig. 6F).
Based on these observations, we deduce that calcined HA-700 NPs significantly (p < 0.001) accelerated the healing of experimental rats with critical bone lesions of 3 mm in diameter in the left femur when compared to HA-1000 or negative controls.It has been shown earlier that HA-1000 particles had lower CO 2− 3 ion contents, high crystallinity, and a large crystal size 66,67 , which can explain their low resorption rate and poor integration with the newly formed woven bone in the lesion site during the healing process.Brandt et al. 68 , previously reported that after 12 weeks of nanocrystalline HA implantation for bone repair, rabbit femoral bones showed bone formation, low degradation, and slow resorption.Clinical studies showed that HA could form direct physiological bonds with bone, resulting in good biocompatibility and no inflammatory response 69 .

Conclusion
An efficient three-step hydrothermal approach was proposed to extract the pure crystalline phase of biogenic HA from buffalo waste bones, which was found to be stable throughout the range of temperatures investigated.The optimal temperature for the production of crystalline HA in the bone structure was reported to be 700 °C, which was sufficient for eliminating residual organic matter and enhancing the crystallinity of the HA phase.HA-700 extracts were characterized by their bud-like nonstoichiometric wide nano-range particle size and the presence of divalent PO 3− 4 and CO 2− 3 anions on the non-apatite hydrated layer.Despite their limited biodegradability, HA-700 NPs could form bone-like apatite coatings on their surfaces for better bone bonding, which is important in bone remodeling for in vivo xenograft treatment of critical sized femoral lesions in an experimental rat model.Thus, the study confirmed the effectiveness of manufacturing porous biogenic HA bodies for biomedical applications from buffalo waste bones using the hydrothermal technique.

Data and code availability
Metadata used and/or analyzed during the current study will be made available from the corresponding author on reasonable request.

Figure 1 .
Figure 1.Surgical procedure showing an anesthetized rat with left femur shaved (A), periosteum and soft tissues fixed, and a 3-mm-diameter lesion drilled into femur bone to medullary canal (B), xenograft treatment with HA-700 or HA-1000 used to fill femur bone lesion (C) and wound closed with 5-0 suture (D).

Figure 6 .
Figure 6.Photomicrographs of the left femur shaft of negative control animals (n = 4) showing large lesion sites filled with irregularly formed unevenly stained widely separated trabeculae of woven bone (T) with mild inflammatory cell infiltration (*) (A); unevenly stained bone trabeculae exhibiting larger and more irregularly shaped lacunae (black arrow) at a higher magnification (B); treated animals (n = 4) with HA-700 particles after four weeks showing lesions filled with new bone except for few small localized areas (*) and small islands of cartilage (R) (C); an acidophilic normal compact bone containing normal osteocytes inside their lacunae at a higher magnification (D); treated animals (n = 4) with HA-1000 particles after 4 weeks showing a few widely separated trabecular bone (T) with a persistence large lesion (E); and trabecular bone containing few osteocytes with irregular shape at a higher magnification (F).(H&E Original magnification A, C, and E 200×; B, D, and F 400×).

Table 1 .
Temperature interval, maximum decomposition peak temperature, weight loss, and weight residue of bone powders by differential thermogravimetric analysis (DTGA).

Table 2 .
Elemental mineral content of buffalo B-raw femur bone by X-ray fluorescence (XRF) spectroscopy analysis.