Evaluation of a chitosan/polyvinyl alcohol hydrogel loaded with graphene oxide and nano TiO₂ for bone defect reconstruction in a dog model

Abstract

This study evaluated the application of chitosan/polyvinyl alcohol/graphene oxide/nano titanium oxide (CS/PVA/GO/nano TiO₂) hydrogels for bone defect reconstruction in dogs. Dogs were subjected to mid-diaphyseal circular bone defects (0.8 cm²) in the radius bones. Bone defects were implanted with the hydrogel in the treated group (n = 9), while the control group were subjected to spontaneous healing (n = 9). Dogs were subjected to clinical, radiographic, and scanning electron microscopy (SEM) evaluations at 15-, 30-, and 45-days post-surgery. Dogs in the treated group recorded no lameness by the end of the third week post-surgery, while dogs in the untreated group still exhibited lameness of grade 1. There was a significant decrease (p < 0.05) in the cortical defect (mm) of the treated group (5.46 ± 0.17 and 1.45 ± 0.13) compared with the control group (7.57 ± 0.05 and 7.59 ± 0.06) at 30- and 45-days post-surgery, respectively. The depth of the bone defects (mm) decreased significantly (p < 0.05) in the treated group (2.26 ± 0.12 and 0.008 ± 0.002) compared with the untreated group (4.05 ± 0.05 and 2.16 ± 0.07) at 30- and 45-days post-surgery, respectively. Throughout the period of study, there was a significant increase (p < 0.05) in the radiographic density of the bone defects (px) in the treated group (474 ± 17.88) compared with that in the control group (619.6 ± 6.85). SEM results revealed complete closure of the bone defects in the treated group. Thus, implantation of bone defects with the CS/PVA/GO/nano TiO₂ hydrogel represents a promising bone graft substitute for accelerating bone healing.

---

Introduction

In orthopedic surgery, bone defect repair is an essential process that will inevitably arise in the future. The occurrence of bone tumors and an increase in traffic accidents have led to a recent rise in the number of patients suffering from bone abnormalities. However, it is still challenging to find a suitable bone substitute.^1–3

Autogenous bone grafts are considered to be the gold standard because autografts provide a structure that is same as that of the defect region, and they also provide osteogenic cells and osteoinductive growth factors.⁴ Unfortunately, autogenous bones have certain shortcomings, such as they are not readily available, typically result in discomfort at the donor site, and necessitate further surgical procedures.³ Alternatively, allografts and xenografts have been studied to produce comparable substitute materials. While most of them lack intrinsic osteoinductivity, they share several characteristics and have the structure of an autogenous bone. However, further steps are needed to enhance the osteogenic action of these materials.⁵

Tissue-engineered synthetic graft materials could potentially address these issues with previously utilized bone transplant materials. The perfect scaffold should be biocompatible, structurally sound, and serve as a stopgap for the tissue's cells while they grow into a freshly produced bone. Furthermore, to be osteoconductive, the optimal scaffold must preserve the right ratio of mechanical qualities, porosity architecture, and degradability.⁶

The literature has documented a range of bone substitutes (e.g., cobalt alloys, ceramics, aluminium and zirconium oxides, calcium phosphates, and synthetic and natural polymers) that are intended to enhance bone healing.⁷ Titanium and its alloys are highly sought-after implant materials for biomedical applications owing to their remarkable properties, including good biocompatibility, resistance to bodily fluids, incredible tensile strength, flexibility, and strong corrosion resistance.⁸ However, clinical titanium implants often encounter major complications such as implant-associated infections and delayed osseointegration. Wu et al.⁹ avoided these problems by providing a synergistic photothermal (PTT)/photodynamic (PDT) therapy strategy using the advantages of near-infrared (NIR) responsive biomimetic micro/nano titanate/TiO₂-X heterostructure coatings with a potassium titanate nanofiber-assembled micro/nano porous network (KMNW) and a sodium titanate nanosheet-assembled micro/nanoscale sponge-like structure (NaMNS). They found that these coatings not only achieved a high antibacterial rate (94.45%) under NIR irradiation, but also facilitated osseointegration.

Graphene and its derivatives (GDs), including graphene oxide, possess exceptional mechanical qualities, electrical conductivity, a large specific surface area, and stable atomic structure, making them interesting candidates for use in bone healing.¹⁰ To create scaffolds^11,12 and injectable hydrogels,¹³ functionalized GDs have been mixed with a variety of substrates, including metals,¹⁴ inorganic substances^15,16 natural polymers,¹⁷ and synthetic polymers.¹⁸ To demonstrate the benefits of using graphene oxide in bone regeneration, Yang and his team developed a bone microenvironment-responsive multifunctional two-dimensional GO coating on the surface of microporous sulfonated polyetheretherketone (SPEEK) via polydopamine modification. Their in vivo study showed that the immunomodulatory properties of multifunctional GO nanosheets can facilitate bone regeneration.¹⁹

A biopolymer that is produced through the deacetylation of chitin, chitosan (CS) is one of the most promising polymeric compounds. CS has been reported to be safe, hemostatic, biocompatible, and osteoconductive. Furthermore, it helps to maintain the mineralized bone matrix, and causes the least amount of inflammation following implantation.^20–22

Polyvinyl alcohol (PVA) is a synthetic polymer that exhibits properties of water solubility, biocompatibility, non-toxicity, and high water content; thus, it can be utilized in a wide range of industries, including bone reconstruction. Xiang et al. developed a tough 3D hydrogel that contains PVA, tannic acid (TA), hydroxyapatite (HA), and collagen type I for bone regeneration. They found that their hydrogel provided an effective environment for MC3T3-E1 cell growth and enhanced osteogenic differentiation. Furthermore, the hydrogel exhibited bone reconstruction and osteointegration capabilities in their rat femoral defect model.²³ Zhao et al.²⁴ developed a pH-responsive hydrogel by coupling PVA, hydroxyapatite, and polymerized HEMA-HA with acrylic acid (AA). They found that their hydrogel possessed a porous honeycomb structure and was resistant to water loss. The hydrogel showed a good drug release rate of about 51.35 ± 2.39% at 48 h, making it suitable for bone regeneration engineering.

Seeking to emulate the natural organic–inorganic bony matrix, Chen and coworkers developed a PVA-based hydrogel combined with deferoxamine-modified LAPONITE® nanoplatelets and tannin-modified poly(vinyl alcohol). From their in vitro study, they found improvements in the adhesion and proliferation of human umbilical vein endothelial cells (HUVECs) for angiogenesis and promoted mesenchymal stem cells (MSCs) osteogenic differentiation for osteogenesis. Furthermore, the skull bone reconstruction in their rat model had been promoted in comparison to bare PVA.²⁵

Creating novel therapeutic approaches to promote bone repair is still desperately needed. To the best of our knowledge, there have been no previous research studies on the potential of chitosan/polyvinyl alcohol hydrogel loaded with graphene oxide and nano TiO₂ on bone healing. Therefore, the purpose of this study was to assess the efficacy of chitosan/polyvinyl alcohol hydrogel loaded with graphene oxide and nano TiO₂ (CS/PVA/GO/nano TiO₂) for bone defect restoration in dogs using scanning electron microscopy (SEM), radiography, and clinical data.

Experimental

Ethical approval

All of the procedures in this study have been approved by the Research Ethics Committee (REC) of the Faculty of Veterinary Medicine, Assiut University, Assiut, Egypt, in compliance with Egyptian bylaws and the OIE animal welfare standards for the care and use of animals in research and education, under no. (06/2023/0049). All methods were performed in accordance with relevant guidelines and regulations.

Preparation of the chitosan/polyvinyl alcohol/graphene oxide/nano titanium oxide (CS/PVA/GO/nano TiO₂) hydrogel

The hydrogel was prepared according to the following process, as illustrated in Fig. 1. In detail, 1.5 g of polyvinyl alcohol with molecular weight 15 000 (PVA) was dissolved in 30 ml double distilled water at 90 °C (solution A). 0.5 g of chitosan (CS) was dissolved in 99 ml double-distilled water and 1 ml acetic acid (solution B). Subsequently, 70 mg of graphene oxide (GO) and 15 mg of titanium oxide (TiO₂) were sonicated in 20 ml double-distilled water (solution C). Both solutions (B) and (C) were heated up at 70 °C. Solution (C) was added dropwise into solution (A) with continuous stirring. Solution (B) was added to the mixture with continuous stirring at 90 °C to reduce the amount of water. The remaining amount was transferred to a Petri dish. After cooling to room temperature, the Petri dish was frozen at −20 °C for 12 hours. The Petri dish was later removed from the freezer, and stored at room temperature for another 12 hours. The freezing and melting process was repeated three times. After freezing and thawing the mixture for several cycles, the hydrogel will be solidified at room temperature, in accordance with the following mechanism. Two actions occur through the freezing and thawing process: (1) through the freezing process, the nucleuses are generated within the polymer solution, (2) through the thawing process, the chains in the polymer are shaped close to the nuclei in a systematic trend to form crystalline structures. By cycling these two processes, the hydrogen bonding between polymers chains formed and construct a 3D structure.²⁶

Fig. 1 Preparation of the chitosan/polyvinyl alcohol hydrogel loaded with graphene oxide and nano TiO₂ to regenerate bones in dogs.

Characterization of the CS/PVA/GO/nano TiO₂ hydrogel

Various characterization techniques were employed to investigate the different properties of the CS/PVA/GO/nano TiO₂ hydrogel. X-ray diffraction (XRD) (Bruker D8 Advance, Cu Kα, λ = 0.154 nm) was used to check the phase formation. The morphological properties were assessed using scanning electron microscopy (SEM; Zeiss Supra 40). The molecular orientations in the CS/PVA/GO/nano TiO₂ hydrogel were established by Fourier transform infrared (FTIR) spectra obtained from an Agilent Cary 630 FT-IR spectrometer (Agilent Technologies Inc., Santa Clara, CA, USA).

Experimental animals

The study was conducted on 18 non-medicated, clinically healthy, adult mongrel male dogs (n = 18), obtained from the Experimental Animal House, Faculty of Veterinary Medicine, Assiut University, Assiut, Egypt. The dogs had no orthopedic abnormalities based on clinical examination. Their weights ranged from 14 to 15 kg body weight (BW), and they were aged 2–3 years old. Dogs were housed individually in standard cages with feed and water ad libitum. One of the forelimbs (right or left limb) was selected randomly for conduction of a mid-diaphyseal circular bone defect (0.8 cm in diameter) in the radius bone under general anesthesia, as described below.

Experimental protocol

Dogs were divided randomly into two groups, each comprising 9 dogs. In the treated group (group T, n = 9), bone defects were implanted with the CS/PVA/GO/nano TiO₂ hydrogel. Conversely, in the control group (group C, n = 9), bone defects were allowed for spontaneous healing. Dogs were subjected to clinical, radiographical, and electron microscopical scanning evaluation at 15-, 30-, and 45-days post-bone defect induction. Three dogs (n = 3) were randomly selected from each group (treated and control groups) at the predefined time intervals.

Surgical procedure

Water and food were withheld from dogs for 3 and 12 hours (h), respectively, before surgery. The surgical procedure was conducted under the effect of total intravenous anesthesia (TIVA). Dogs were administered intravenous (IV) 1 mg kg⁻¹ xylazine 2% (Xyla-Ject, ADWIA Co., Egypt) for induction of anesthesia, followed by IV 2 mg kg⁻¹ ketamine 5% (ketamine, Sigma-tec pharmaceutical industries, Egypt) 5 minutes later. Anesthesia was then maintained with 10 mg kg⁻¹ h⁻¹ ketamine 5% and 1 mg kg⁻¹ h⁻¹ xylazine 2%.²⁷

The dog was maintained in the lateral position with the operated forelimb the uppermost and draped except for the surgical site. The site of the operation was prepared surgically (clipped, shaved, and scrubbed several times using 70% isopropyl alcohol and 10% povidone-iodine). A latero-palmar incision (8–10 cm) was conducted at the middle of the forearm through the skin and subcutaneous tissues (Fig. 2A and B). The extensor carpi radialis and common digital extensor muscles were bluntly dissected to expose the radial shaft (Fig. 2C and D). The periosteum was elevated (Fig. 3A). A circular bone defect (0.8 cm²) was constructed at the mid-diaphysis of the radius using a bone drill (Fig. 3B).

Fig. 2 (A) and (B) A latero-palmar incision (8–10 cm) at the midsection of the forearm through the skin and subcutaneous tissues. (C) and (D) Blunt dissection of the extensor carpi radialis and common digital extensor muscles, exposing the radial shaft.

Fig. 3 (A) Elevation of the periosteum. (B) Circular bone defect (0.8 cm) made at the mid-diaphysis of the radius (left for spontaneous healing without implantation in the control group). (C) Bone defect implanted with the titanium oxide/graphene oxide/chitosan nanocomposite in the treated group. (D) Surgical site closed in a simple continuous pattern.

The bone defects were implanted with the CS/PVA/GO/nano TiO₂ hydrogel in the treated group (n = 9) (Fig. 3C). The bone defects were left for spontaneous healing without implantation in the control group (n = 9). The periosteum and muscles were repositioned to their normal location over the bone defects. The subcutaneous tissue and skin were closed separately with 2-0 USP polyglactin 910 (Egysorp, Taisier-Med, Egypt) in a simple continuous pattern (Fig. 3D).²⁸ A dressing bandage was applied after closing the surgical site. Wounds were dressed with sterile non-adherent dressing pads, which were secured by sterile elastic bandages and outer elastic adhesive bandages.

All dogs were administered procaine penicillin and dihydrostreptomycin sulfate (Pen & Strep, Norbrook Pharmaceutical Co., Northern Ireland, UK) at a dose of 1 ml/25 kg BW and meloxicam (Meloxidel 0.5%, Delta Pharma Co., Egypt) at a dose of 0.4 ml/5 kg BW intramuscularly (IM) for five successive days postoperatively.

Clinical evaluation

The dogs were kept under clinical observation for 45 days. Changes in the animals’ behavior, food or water consumption, surgical site, or gait were monitored daily and recorded. The animal's gait was scored on a scale of 0 to 5, in accordance with the method by Goh²⁹ (Table 1). The levels are described as follows: grade 0 – no lameness, grade 1 – barely detectable lameness, grade 2 – mild, weight-bearing lameness, grade 3 – moderate, weight-bearing lameness, grade 4 – severe, weight-bearing lameness, and grade 5: – severe, non-weight-bearing lameness.

Table 1 Gait scale of animals

Grade	Description
0	No lameness
1	Barely detectable lameness
2	Mild, weight-bearing lameness
3	Moderate, weight-bearing lameness
4	Severe, weight-bearing lameness
5	Severe, non-weight-bearing lameness

---

Radiographical evaluation

Digital cranio-palmar and lateral radiographical views were taken for the operated limbs under the effect of intravenous 1 mg kg⁻¹ xylazine HCl 2% (Xyla-Ject, ADWIA Co., Egypt) using (FUJIFILM COMPUTED RADIOGRAPHY, MODEL CR-IR 392) at each time interval. The cortical defect, depth, and radiographic density of the bone defects were measured using the RadiAnt DICOM viewer software (version 1.1.2022) (Fig. 4A–C).

Fig. 4 (A): Cortical defect. (B) Depth. (C) radiographic density of the bone defects.

Scanning electron microscopical analysis (SEM)

Dogs (n = 3), of each group were humanely euthanized at each time interval with intravenous (IV) pentobarbital sodium and phenytoin sodium solution (Beuthanasia-D Special, 1 ml/4.5 kg BW, 1 ml = 390 mg pentobarbital sodium, 50 mg phenytoin sodium, Intervet/Merck Animal Health, Germany). Death was confirmed by the absence of breathing and heart beats.

Representative specimens of the bone defects were harvested for further scanning electron microscopical analysis (SEM). The specimens were immediately washed with normal saline, and fixed in a mixture of 2.5% paraformaldehyde and 5% glutaraldehyde in 0.1 M sodium phosphate buffer with a pH of 7.3 at 4 °C for 24 h. The specimens were washed with 0.1 M sodium phosphate buffer with a pH of 7.3, and then dehydrated using an ascending series of ethanol at 30%, 50%, 70%, and 90% for two hours, 100% for two days, and amyl acetate for two days. Critical point drying was applied to the samples using liquid carbon dioxide. Each sample was stuck to a metallic block using silver paint. By using the gold sputter coating apparatus, samples were evenly gold coated in a thickness of 15 nm.³⁰ The samples were examined using the JEOL (JSM 5400 LV) scanning electron microscope in the range of 15–25 kV, and photographed within the electron microscope unit at Assiut University, Assiut, Egypt. Euthanized animals were then sent to the Department of Anatomy and Embryology, Faculty of Veterinary Medicine, Assiut University for educational and teaching purposes.

Statistical analysis

Data are represented as means ± SE. Values with different small superscript letters in the same row for each parameter at various intervals are compared with time 0, and are significantly different at (P < 0.05) using one-way ANOVA, followed by Dunnett's post hoc test. Values with different large superscript letters in the same column for each parameter between the treated and control groups are significantly different at (P < 0.05), using the unpaired t-test and GraphPad Prism software.

Results

Morphology and elemental analysis of the CS/PVA/GO/nano TiO₂ hydrogel

The crystal structure of the CS/PVA/GO/nano TiO₂ hydrogel was investigated using X-ray diffraction (XRD), as presented in Fig. 5a and b. The detected XRD peaks confirm the presence of PVA, CS, GO, and TiO₂. The peaks located at 2θ = 37°, 48°, and 54° belong to (004), (200), and (105) planes of the anatase TiO₂ tetrahedral phase, respectively, whereas the peak located at 2θ = 25.4° corresponds to the (101) plane of anatase TiO₂ and (002) plane of GO.³¹ Other diffracted peaks located at 2θ = 22.6° and 40.2° are assigned to the (100), (200) planes, and the compound crystalline planes (111), (1−11), (210), and (2−10) of PVA, respectively. However, the peaks located at 11.2° can be attributed to the (100) plane of PVA, (020) plane of chitosan, and (001) plane of GO. Other broad peaks located at 19.7° can be assigned to the (101)/(10−1) planes of PVA and (110) planes of chitosan.^32–35

Fig. 5 (a) XRD patterns and (b) 2D-XRD collected patterns of the chitosan/polyvinyl alcohol hydrogel loaded with graphene oxide and nano TiO₂.

The micrographs of the CS/PVA/GO/nano TiO₂ hydrogel were obtained via SEM (Fig. 6). As shown in Fig. 6a, the hydrogel exhibits porous structures. Fig. 6(b–e) shows the energy-dispersive X-ray images (EDX), which display the elemental mapping for the CS/PVA/GO/nano TiO₂ hydrogel sample. The EDX images confirmed the successful conjugation of PVA, CS, GO, and nano TiO₂, where Ti, O, C, and N were well distributed.

Fig. 6 (A) Hydrogel exhibiting porous structure (SEM image). (B)–(E) Energy-dispersive X-ray images (EDX) displaying the elemental mapping for the CS/PVA/GO/nano TiO₂ hydrogel sample.

The molecular orientations in the CS/PVA/GO/nano TiO₂ hydrogel were investigated using FTIR spectroscopy. The FTIR spectrum is shown in Fig. 7. The FTIR spectral bands at 3250 cm⁻¹ corresponded to NH–stretching in α-chitin, which is involved in intermolecular and intramolecular hydrogen bonds. Furthermore, the amide-I band in the α-chitin spectrum at 1640 cm⁻¹ was attributed to the intermolecular hydrogen bonds –CO⋯HN–. The bands at approximately 1551 cm⁻¹ and 1319 cm⁻¹ were attributed to the chitosan characteristic peaks for amide II and amide III, respectively.^36,37 Meanwhile, the observed peaks at 2915 cm⁻¹, 1409 cm⁻¹, 1070 cm⁻¹, and 831 cm⁻¹ can be ascribed to the stretching vibration of –CH₂, bending vibration of C–H, stretching vibration of C–O, and stretching vibration of C–C, respectively, due to the presence of PVA and GO.^32,38–40

Fig. 7 FTIR of the chitosan/polyvinyl alcohol hydrogel loaded with graphene oxide and nano TiO₂.

Clinical evaluation

There were no recorded deaths or severe complications throughout the whole duration of the study. There were no recorded complications in the surgical site (infection or dehiscence). Healing was achieved by the 1st intention and sutures were removed 10–12 days after the operation. Dogs in the treated and control groups refused to eat on the first day post-surgery. They then began to eat gradually on the second day. By the end of the first week after the surgery, all dogs were consuming their regular food. Dogs in the treated and control groups recorded lameness of grade 4 in the first 48 hours post-surgery, which gradually subsided to grade 3 and 2 by the end of the first and the second weeks after surgery, respectively. By the end of the third week, dogs of the treated group recorded no lameness, while dogs of the untreated group still had lameness of grade 1 (Fig. 8).

Fig. 8 Lameness grades in the treated and control groups.

Radiographical evaluation

Cortical defect of the bone defects

Dogs of the treated group recorded a significant decrease (p < 0.05) in the cortical defects at 30- and 45-days post-surgery (5.46 ± 0.17 and 1.45 ± 0.13 mm, respectively), compared to the baseline value (8.23 ± 0.10 mm). Dogs of the control (untreated) group exhibit a non-significant decrease in cortical defects throughout the study time. There was a significant decrease (p < 0.05) in the treated group compared with the control groups at 30- and 45 days post-surgery (Table 2 and Fig. 9).

Table 2 Cortical defect (mm) of the bone defects in the treated and control groups

Intervals (days)	0	15	30	45
Treated	8.23 ± 0.10^a	7.32 ± 0.23^a,A	5.46 ± 0.17^c,A	1.45 ± 0.13^d,A
Control (Untreated)	8.48 ± 0.04^a	7.42 ± 0.14^a,A	7.57 ± 0.05^a,B	7.59 ± 0.06^a,C

---

Fig. 9 X-ray films of the cortical defects of the bone defects in the control (A)–(C) and treated (D)–(F) groups at 15-, 30-, and 45-days after surgery, respectively.

Depth of the bone defects

There was a significant decrease (p < 0.05) in the depth of the bone defects at 30- and 45-days post-surgery in the treated group (2.26 ± 0.12 and 0.008 ± 0.002 mm, respectively), compared to the baseline value (4.22 ± 0.11 mm). Dogs of the untreated group exhibited a significant decrease (p < 0.05) in the depth of the bone defects at 45 days post-surgery (2.16 ± 0.07 mm). The depth of the bone defects decreased significantly (p < 0.05) in the treated group compared with the untreated group at 30- and 45-days post-induction of the bone defects (Table 3 and Fig. 10).

Table 3 Depth (mm) of the bone defects in the treated and control groups

Intervals (days)	0	15	30	45
Treated	4.22 ± 0.11^a	3.24 ± 0.17^a,A	2.26 ± 0.12^c,A	0.008 ± 0.002^d,A
Control (Untreated)	4.78 ± 0.19^a	4.35 ± 0.08^a,A	4.05 ± 0.05^a,B	2.16 ± 0.07^b,C

---

Fig. 10 X-ray films of the depth of the bone defects in the control (A)–(C) and treated (D)–(F) groups at 15-, 30-, and 45-days after surgery, respectively.

Radiographic density of the bone defects

There was a significant increase (p < 0.05) in the radiographical density (px) of the bone defects throughout the study at 15-, 30-, and 45-days post-implantation of the CS/PVA/GO/nano TiO₂ hydrogel in dogs of the treated group (574.1 ± 9.00, 496.3 ± 11.61, and 474.0 ± 17.88, respectively) compared to the base line value (676.5 ± 30.41). The radiographical density of the bone defects in the untreated group exhibited a non-significant increase throughout the study. There was a significant increase (p < 0.05) in the radiographical density of the bone defects in the treated group compared with the control group throughout the study (Table 4 and Fig. 11).

Table 4 Radiographic density (px) of the bone defects in the treated and control groups

Intervals (days)	0	15	30	45
Treated	676.5 ± 30.41^a	574.1 ± 9.00^b,A	496.3 ± 11.61^c,A	474 ± 17.88^d,A
Control (Untreated)	666.0 ± 29.49^a	650.1 ± 3.258^a,B	638.9 ± 8.67^a,C	619.6 ± 6.85^a,D

---

Fig. 11 X-ray films of the radiographical density of the bone defects in the control (A)–(C) and treated (D)–(F) groups at 15-, 30-, and 45-days after surgery, respectively.

Scanning electron microscopical analysis (SEM)

Scanning electron microscopical examination of the control (untreated) group at 15 days post-surgery showed incomplete healing of the bone defect with the presence of cell activity in the form of osteoblasts. These cells appeared to be regular in shape with cytoplasmic protrusions connected with fine collagen fibrils (Fig. 12A–C). The treated group at the same interval showed closure of about 60% of the bone defect with the presence of a junction between the bone and the nanocomposite material. The treated group also exhibited primary bony callus formation and osteocytes appearing ovoid in the lacuna (Fig. 12D–F). Examination of the bone defects at 30 days post-surgery in the untreated group showed incomplete closure with the growth of bony spicules and osteocytes (Fig. 13A–C). The treated group revealed a complete closure of the bone defect at the same time interval with the presence of the nanocomposite material, primary bone formation, and old bone with calcium deposition. Numerous ovoid osteocytes were present and situated in the lacuna with canaliculi formation (Fig. 13D–F). Scanning electron microscopical examination of the bone defects in the untreated group at 45 days post-surgery showed an incomplete healing of the bone defects with the growth of bony spicules and osteocytes (Fig. 14A–C). The treated group showed a complete closure of bone defects. The composite material was surrounded by primary bone formation and a typical old bone shape with the Haversian canal (Fig. 14D–F).

Fig. 12 Scanning electron micrograph of the bone defects 15 days after surgery: (A)–(C) the control group showed an incomplete healing of the bone effect and presence of osteoblasts (arrowheads). (D)–(F) The treated group showed old bone (yellow arrows), nanocomposite material (red arrow), the junction between the bone and nanomaterial (star), new bone formation (white arrow), and osteocytes (arrowheads).

Fig. 13 Scanning electron micrograph of the bone defect 30 days after surgery: (A)–(C) The untreated group showed the growth of bony spicules and osteocytes. (D)–(F) The treated group showed a complete closure of the bone defect (circle), including the composite material (red arrow), primary bone formation (white arrows), old bone (yellow arrows) and osteocytes (red arrowheads).

Fig. 14 Scanning electron micrograph of the bone defects 45 days after surgery: (A)–(C) The untreated group showed an incomplete healing of the bone defects with the growth of bony spicules and osteocytes. (D)–(F) The treated group showed a complete closure of the whole bone defect (circle), including the composite material (red arrow), primary bone formation (white arrows), and typical bone shape old with Haversian canal (yellow arrows).

Discussion

Trauma or illnesses like cancer can cause significant bone loss. However, the healing procedure for bone abnormalities is a long process. New bone synthesis occurs slowly because of limited blood flow to the fracture site, and inadequate calcium and phosphorus to reinforce and harden new bone. Thus, research into bone engineering techniques offers a way to replace missing bone and restore it to functionality.^41,42 This served as the driving force for this investigation.

Even though bone tissue has an amazing ability to repair itself, critical-sized, load-bearing bone lesions cannot heal on their own without surgery. The best course of action for treating these bone abnormalities is still a clinical problem with significant social and financial ramifications.⁴ Delays in bone regeneration can be caused by a number of causes, including age, aberrant hormone and nutrition status, concomitant illnesses, insufficient mechanical stabilization, further pharmaceutical therapy, genetic variations, and others.¹

Many studies aimed at investigating bone repair have been conducted using animal models. Appropriate animal models are needed. These animal models should be well standardized and, most importantly, should approximate the clinical situation in humans.^43,44 Generally, large animal models are advantageous in terms of the dimensions and biomechanical situations.⁴⁵ It was believed that the way bones are mended in dogs closely resembled how bones heal in people.⁴³ The huge skeletal size and human-like bone structure of dogs, along with the presence of a Haversian rebuilding system, are advantageous.⁴⁶ Drill-hole bone defects have been a more reliable model of bone injury in recent years.⁴⁷

Bone repair and regeneration were monitored by clinical, radiographic, and SEM examinations.⁴⁸ Owing to its low cost, great availability, and safety in increasing widespread specificity during the identification of possible long bone nonunions, radiographic examination is still the gold standard for current diagnosis.⁴⁷ A precisely focused electron beam is traversed across a sample surface in scanning electron microscopy (SEM). There are numerous available imaging modes and automation levels.⁴⁹

It is necessary to produce bone replacement materials (BSM) that can endure significant mechanical loading in situ, and have biological performance similar to that of native bone.⁴ Therefore, the implantation period should be long enough to allow for both bone production and remodelling in order to assess if a biological BSM is successful in regenerating bone under the impact of physiological loading.⁵⁰ This was consistent with the SEM results in the treated group, which showed that after 45 days post-surgery, the existence of composite material residues together with freshly created bone was linked to the complete closure of bone defects.

Bone is a nanostructured compound of collagen and hydroxyapatite fibers.⁵¹ Because of this, the topography of the nanoscale surfaces resembles that of the extracellular matrix of bone tissue, offering cellular support and sufficient porosity in certain situations for the processes of cell adhesion, proliferation, and differentiation that occur throughout bone formation.^52,53 Without a sterile microenvironment surrounding the bone defect, bone remolding and the creation of new bone cannot be fully successful. Treatment for infective bone abnormalities is difficult, and requires recuperation over an extended period of time.⁵⁴ In response to this concern, materials in the graphene family are thought to have antibacterial properties.⁵⁵

Here, bone defects implanted with the CS/PVA/GO/nano TiO₂ hydrogel in the treated group exerted a significant decrease (p < 0.05) in the cortical defect and the depth of the bone defects compared with the untreated group, according to the radiographical measurements. This was associated with the complete closure of the bone defects in the treated group, as demonstrated by the SEM examination at 45 days post-implantation. This may be attributed to the bone regeneration promotor effect in a synergistic action of the titanium oxide, chitosan, and graphene oxide in one nanocomposite.

The topography, charge distribution, and chemistry of titanium's surface all have an impact on the adherence of various cell types to titanium implants. Increased osteoblast adhesion, maturation, and subsequent bone formation are observed with titanium implants, including rough surface topography and free energy.^56,57

Polymer-based nanocomposites have been created by adding graphene to polymers. The most commonly utilized member of the graphene family in current biomedical applications is graphene oxide (GO), which is the oxidized form of graphene. GO maintains its laminar structure upon oxidation. The carbon atom sheet is covered in a variety of oxygen functional groups, the most common of which are the carboxyl and carbonyl groups linked to the edge and the hydroxyl and epoxy groups on the basal plane. In addition to providing hydrophilicity and dispersibility, the presence of these functional groups opens up further possibilities for modifying and customizing the features of GO. Hence, because of its accessibility, hydrophilicity, dispersibility, chemical tunability, and processability, GO is seen as a desirable and affordable substitute for graphene.^2,58,59

It has been demonstrated that GO may enhance osteoblast adhesion, proliferation, and mineralization.⁶⁰ Furthermore, GO encourages the differentiation of soft tissue-derived mesenchymal stem cells into osteogenic cells by keeping them active.^61,62 When used in bone tissue engineering, the excellent osteoinductive effectiveness and comparatively low toxicity of the GO-based composite materials promote quick cell proliferation.¹³

Because the polysaccharide backbone of chitosan (CS) is structurally comparable to that of glycosaminoglycans, the main structural element of bone extracellular matrix, CS can promote cell adhesion and proliferation.^63,64 By introducing pores into the composite material, chitosan can improve its osteoconductive, biocompatible, and biodegradable qualities, resulting in a composite that can withstand pressure and accelerate the formation of bone minerals.⁶⁵

With encouraging outcomes, biocompatible synthetic polymers like polyvinyl alcohol (PVA) have even been applied to tissue regeneration.⁶⁶ Because of their superior water absorption and retention capabilities, hydrogels have been thoroughly investigated as scaffolds.⁵⁸

Throughout the course of the trial, there was a substantial increase (p < 0.05) in the radiographic density of the bone defects in the treated group, as compared to the control group. This could be because the treated group has experienced a rise in the production of new bone, which serves as a barrier and reduces the amount of X-ray radiation that is absorbed, which in turn lowers the density number.

Because of their enormous Golgi apparatuses, which enable them to deposit huge amounts of proteins onto the surface of the bone matrix, osteoblasts are primarily responsible for bone regeneration.⁴⁹ Forty-five days after surgery, the SEM examination revealed that the bone flaws in the treated group had completely closed. An initial bone development and a conventional old bone structure with the Haversian channel encircled the composite material. This could be because of the high biodegradability and superior biological activity of the nano titanium oxide/graphene oxide embedded in the chitosan hydrogel, which can promote bone regeneration by osseointegration.

One of this study's limitations is the lack of live stains (such as calcein) to stain the newly formed bone to identify regeneration/reconstruction. However, in this study, scanning electron microscopy (SEM) could be more helpful in monitoring the healing process of bone defects. SEM provides a high-resolution, three-dimensional view of the bone tissue. It compensates for the lack of certain stains by offering detailed surface visualization, which is not visible with conventional stains, and structural integrity insights as it provides clarity on the spatial organization of tissues and the bone healing process. Incorporating SEM strengthened the study's findings by providing complementary data that overcame some of the challenges associated with histological staining, enhancing the overall interpretation of tissue organization and function. Moreover, further studies still need to be conducted on larger bone defects, especially those associated with infection or diabetes on a larger scale.

Conclusions

Based on clinical, radiographic, and SEM evaluations, implantation of bone defects (8 mm) with the CS/PVA/GO/nano TiO₂ hydrogel significantly decreased the cortical defect (1.45 ± 0.13 mm) and depth of the bone defects (0.008 ± 0.002 mm) at 45 days post-surgery. SEM examination revealed complete closure of the bone defects in the treated group. Dogs in the treated group clinically improved, and recorded no lameness by the end of the third week post-surgery. Implantation of bone defects with the CS/PVA/GO/nano TiO₂ hydrogel represents a promising bone graft substitute for accelerating bone healing.

Abbreviations

CS	Chitosan
PVA	Polyvinyl alcohol
GA	Graphene oxide
nano TiO2	Nano titanium oxide
SEM	Scanning electron microscopy
REC	The Research Ethics Committee
XRDX	Ray diffraction
FTIR	Fourier transform infrared
TIVA	Total intravenous anesthesia

Ethical approval

All of the procedures in this study have been approved by the Research Ethics Committee (REC) of the Faculty of Veterinary Medicine, Assiut University, Assiut, Egypt, in compliance with Egyptian bylaws and the OIE animal welfare standards for the care and use of animals in research and education, under no. (06/2023/0049). All methods were performed in accordance with relevant guidelines and regulations. This study was conducted in compliance with the ARRIVE guidelines.

Author contributions

AI: designed the study. AE: prepared the nanocomposite materials, AI, SI, and MS: conducted the surgical procedures and evaluations, KH: performed the electron microscopical scan analysis, all authors wrote, revised, and approved the paper for publication.

Data availability

All data generated or analyzed during this study are included in this manuscript.

Conflicts of interest

The authors declare that they have no competing interests.

Acknowledgements

This study has not received any external funding or sponsorship. The authors are thankful to The Science, Technology & Innovation Funding Authority (STDF), in cooperation with The Egyptian Knowledge Bank (EKB), for providing open access funding.

References

1. B. Tan, et al. , Int. J. Oral Sci., 2021, 13(1), 9.
2. E. Naumenko, et al. , Clays Clay Miner., 2021, 69(5), 522.
3. H. Qian, et al. , ACS Biomater. Sci. Eng., 2023, 9(1), 1–19.
4. J. Baig, et al. , Curr. Protoc., 2023, 3(1), e631.
5. L. Cheng, et al. , Int. J. Nanomed., 2021, 16, 4289.
6. C. Wang, et al. , ACS Appl. Mater. Interfaces, 2019, 11(43), 39470.
7. G. Funda, et al. , Materials, 2020, 13(1), 201.
8. Z. Wright, et al. , Regener. Eng. Transl. Med., 2019, 5, 190.
9. S. Wu, et al. , ACS Appl. Mater. Interfaces, 2024, 16(33), 43227.
10. Z. Du, et al. , Int. J. Nanomed., 2020, 7523.
11. Y. Y. Qi, et al. , J. Appl. Polym. Sci., 2013, 127(3), 1885.
12. D. Li, et al. , J. Biomed. Nanotechnol., 2018, 14(12), 2003.
13. S. Saravanan, et al. , Biomed. Pharmacother., 2018, 107, 908.
14. H. Zhou, et al. , Mater. Lett., 2017, 192, 123.
15. K. Liu, et al. , Ceram. Int., 2025, 51(1), 1103.
16. P. Yu, et al. , Carbohydr. Polym., 2017, 155, 507.
17. S. Dinescu, et al. , Int. J. Mol. Sci., 2019, 20(20), 5077.
18. Y. Zhang and J. Hu, Polymers, 2020, 12(1), 118.
19. C. Yang, et al. , J. Controlled Release, 2024, 374, 15.
20. S.-L. Bee and Z. A. A. Hamid, Int. J. Biol. Macromol., 2025, 295, 139504.
21. H. Bian, et al. , Mater. Today Bio., 2025, 30, 101448.
22. R. Nartita, et al. , Coatings, 2022, 12(8), 1202.
23. C. Xiang, et al. , Int. J. Biol. Macromol., 2024, 261, 129847.
24. Y. Zhao, et al. , Mater. Today Commun., 2025, 42, 111127.
25. M. Chen, et al. , Appl. Mater. Today, 2024, 40, 102401.
26. M. T. Khorasani, et al. , Int. J. Biol. Macromol., 2018, 114, 1203.
27. A. Ibrahim, Vet. Med. Open J., 2017, 2(2), 38.
28. K. A. Johnson, Piermattei's Atlas of Surgical Approaches to the Bones and Joints of the Dog and Cat, Elsevier Health Sciences, 2013.
29. C. Goh, Efficient Orthopedic Exam. World Small Animal Veterinary Association Congress Proceedings, 2019.
30. J. J. Bozzola and L. D. Russell, Electron microscopy: principles and techniques for biologists, Jones & Bartlett Learning, 1999.
31. C. H. Lee, et al. , Nanoscale Res. Lett., 2012, 7(1), 1.
32. X. Hong, et al., Dry-wet spinning of PVA fiber with high strength and high Young's modulus, IOP Conference Series: Materials Science and Engineering, IOP Publishing, 2018, vol. 439, p. 042011.
33. C.-M. Tang, et al. , Materials, 2015, 8(8), 4895.
34. M. E. Badawy, et al. , Bull. Natl. Res. Cent., 2019, 43(1), 1.
35. R. G. Toro, et al. , Materials, 2020, 13(6), 1326.
36. V. Karthika, et al. , Sci. Rep., 2020, 10(1), 18912.
37. G. Hao, et al. , Sci. Rep., 2021, 11(1), 1646.
38. S. S. Al-Taweel, et al. , Results Phys., 2019, 13, 102296.
39. P. Liu, et al. , Sci. Rep., 2019, 9(1), 9534.
40. I. Faniyi, et al. , SN Appl. Sci., 2019, 1, 1.
41. K. Aktuglu, et al. , J. Orthop. Traumatol., 2019, 20(1), 1.
42. Y.-C. Lee, et al. , Int. J. Mol. Sci., 2019, 20(20), 5015.
43. F. Ghorbani, et al. , Mater. Today Commun., 2020, 22, 100979.
44. D. S. Sparks, et al. , Nat. Protoc., 2020, 15(3), 877.
45. K. M. Zekry, et al. , J. Orthop. Surg., 2019, 27(1), 2309499019832970.
46. P. F. Lei, et al. , Orthop. Surg., 2019, 11(1), 15.
47. Y.-X. He, et al. , Bone, 2011, 48(6), 1388.
48. K. R. Richert-Pöggeler, et al. , Front. Microbiol., 2019, 9, 3255.
49. A. C. Masquelet, et al. , Orthop. Traumatol.: Surg. Res., 2019, 105(1), 159.
50. B. S. Xie, et al. , CNS Neurosci. Ther., 2019, 25(4), 465.
51. A. Tampieri, et al. , J. Biomed. Mater. Res., Part A, 2003, 67A(2), 618.
52. K. Lee, et al. , Chem. Rev., 2014, 114(19), 9385.
53. G. Wang, et al. , Appl. Surf. Sci., 2017, 414, 230.
54. L. Pang, et al. , Nanoscale Res. Lett., 2017, 12, 1.
55. X. Cheng, et al. , Nanoscale Res. Lett., 2018, 13(1), 289.
56. X. Qi, et al. , Smart Mater. Med., 2021, 2, 280.
57. N. López-Valverde, et al. , Coatings, 2021, 11(7), 777.
58. V. Kavimani, et al. , Composites, Part B, 2020, 191, 107911.
59. F. M. Mohammed, et al. , Egyp. J. Vet. Sci., 2023, 54(1), 1.
60. L. Zhai, et al. , Environ. Toxicol. Pharmacol., 2018, 57, 41.
61. T. R. Nayak, et al. , ACS Nano, 2011, 5(6), 4670.
62. A. M. Arnold, et al. , Proc. Natl. Acad. Sci. U. S. A., 2019, 116(11), 4855.
63. Y. Zhang and M. Zhang, J. Biomed. Mater. Res., 2001, 55(3), 304.
64. E. Khor and L. Y. Lim, Biomaterials, 2003, 24(13), 2339.
65. S. Bhumiratana, et al. , Sci. Transl. Med., 2016, 8(343), 343ra83.
66. J. Feng, et al. , Sens. Actuators, B, 2019, 298, 126872.

---