Abstract
This study aimed to investigate the in vivo tissue response of the Biosilicate® scaffolds in a model of tibial bone defect. Sixty male Wistar rats were distributed into bone defect control group (CG) and Biosilicate® scaffold group (BG). Animals were euthanized 15, 30 and 45 days post-surgery. Stereomicroscopy, scanning electron microscopy, histopathological, immunohistochemistry and biomechanical analysis were used. Scaffolds had a total porosity of 44%, macroporosity of 15% with pore diameter of 230 μm. Higher amount of newly formed bone was observed on days 30 and 45 in BG. Immunohistochemistry analysis showed that the COX-2 expression was significantly higher on days 15 and 30 in BG compared with the CG. RUNX-2 immunoexpression was significantly higher in BG on days 15 and 45. No statistically significant difference was observed in RANKL immunoexpression in all experimental groups. BMP-9 immunoexpression was significantly upregulated in the BG on day 45. Biomechanical analysis showed a decrease in the biomechanical properties of the bone callus on days 30 and 45. The implantation of the Biosilicate® scaffolds was effective in stimulating newly bone formation and produced an increased immunoexpression of markers related to the bone repair.
Introduction
Bone healing is a complex biological process that requires the coordinated interaction of signaling molecules, growth factors, bone cells and the synthesis of extracellular matrix in the site of the injury [1,2]. In general, bone tissue has the ability of healing by itself however, in the presence of bone defects of large dimensions or diseases, such as osteoporosis or cancer, this process can be compromised, resulting in a delay of the consolidation and requiring innumerous surgical procedures to treat such non-union fractures [3]. Regenerative therapies to enhance bone repair in fractures of beyond critical size dimension have been emerging as a promising alternative and include the use of autografts, allografts and synthetic bone substitutes [4,5].
Autografts is considered the gold standard and provides osteogenic progenitor cells, osteoinductive growth factors and osteoconductive matrices [5–7]. This procedure is based in the harvesting of a small part of a “donor bone” located in a region of low mechanical load (typically, the iliac crest) and the transplantation to the site of the injury. However, several problems are related to this procedure as tissue necrosis, donor site morbidity, risks of infection, small quantity of material available, the necessity for additional surgeries and even esthetic problems [5,8]. An alternative is the use of allogenic tissues which is easier to obtain but it involves the risk of immune rejection, transmission of infectious diseases and/or lack of osteoconductivity and osteointegration [6,7]. In order to overcome these problems, synthetic bone grafts have been developed as a promising treatment [6,7]. Most of them is a safe and effective option in procedures like repairing great bone loss and non-union fractures, showing osteoconductive properties and proper biodegradability [5,8].
Among the various synthetic bone substitutes, bioactive glasses have been widely used as an alternative to biological grafts in various types of bone surgery [9–12]. They present excellent biocompatibility, osteoconductive properties and can be used in an injectable form [10–12]. One of the most frequently studied biomaterials for tissue engineering is the 45S5 Bioglass®. It is a silica-based melt-derived glass containing 45% SiO2, 24.5% Na2O, 24.5% CaO, 6% P2O5, which has been known as the most bioactive composition among numerous bone-bonding glasses [13]. Despite its well-known stimulatory effects on osteogenesis, the use of monolithic 45S5 Bioglass® for bone engineering applications has been limited due to its relatively poor mechanical properties [14,15].
Considering this important issue, research group has designed specific nucleation and growth thermal treatments to obtain a novel fully-crystallized bioactive glass-ceramic of the same quaternary P2O5-Na2O-CaO-SiO2 system with some compositional modification (Biosilicate®, patent application WO 2004/074199) [16]. The first in vitro experiments demonstrated that Biosilicate® promoted enhanced bone-like matrix formation in comparison to its parent glass and to Bioglass® 45S5 in an osteogenic cell culture system [17]. In addition, Granito et al. (2011) demonstrated that particulated Biosilicate® increased the amount of newly formed bone and biomechanical properties of the callus compared to control specimens and the 45S5 Bioglass® group in an experimental bone defect model in tibia of rats [18].
Although the encouraging in vivo data of the biological response of the use of Biosilicate® on bone tissue, its particulated form does not have macroporous structure and consequently presents a small cement surface area, which results a poor degradability [18,19]. Three-dimensional scaffolds are one alternative to the use of particulated materials [7,17]. Scaffolds present appropriate porosity, which contribute to the increased degradation rate of the material and allow cell attachment and proliferation [20].
Since there is a growing interest in the development of materials with improved osteogenic properties to be used as bone grafts, it was hypothesized that Biosilicate®, used as scaffolds, would have improved bioactive properties and more adequate morphology to facilitate cell migration and vascularization, providing a bone graft with additional advantages for clinical use. Consequently, the present study aimed to make the characterization of Biosilicate® scaffolds and to evaluate the temporal in vivo response of the scaffolds in a model of tibial bone defect in rats. To this end, pre-set scaffolds were implanted on non-critical bone defects in rats. Histopathological, immunohistochemistry and biomechanical analysis were evaluated after 15, 30 and 45 days of implantation.
Materials and methods
Experimental design
Sixty male Wistar rats (aged 12 weeks and weighing 250–300 g) were used in this study. They were maintained under controlled temperature (24 ± 2 °C), light-dark periods of 12 hours, with unrestricted access to water and commercial diet. All animal handling and surgical procedures were strictly conducted according the Guiding Principles for the Use of Laboratory Animals. This study was approved by the Animal Care Committee guidelines of the Federal University of São Carlos (002/2009).
Animals were divided into 2 groups: bone defect control group (CG) – animals with bone defects without any filler and Biosilicate® scaffold group (BG) - animals with bone defects with a Biosilicate® scaffold implant. Each group was divided into 3 different sub-groups (N = 10) euthanized in different periods (15, 30 and 45 days after surgery).
Biosilicate® scaffold preparation
A novel fully-crystallized bioactive glass-ceramic of the quaternary Na2O-CaO-SiO2-P2O5 system (Biosilicate®, patent application WO 2004/074199) was utilized in this study [16]. This compound has been described previously [12–15]. The scaffolds were prepared by a new method, which consisted of the addition of a porogen agent (in this case, polyethylene fibers) into a cement based on powdered Biosilicate®. Initially, isopropyl alcohol (99.5% – Qhemis) was heated to 50 °C and 15% vol. of polyvinil butyral (PVB, Butvar – B98) was added. The mixture was maintained under vigorous stirring during 30 min, until the formation of a homogeneous gel. Biosilicate® powder (particle size ∼5 μm) was added to the gel in a proportion of 40% vol. and homogenized with a glass rod. Thus, 15% vol. of polyethilene fibers (Flinco) with 3.2 mm length and 0.22 mm thick were added to the mixture, and again homogenized. The fiber containing cement was poured in a AZS (alumina-zirconia-silica) crucible and dried at room temperature for 72 h. Blocks of approximately 50 × 80 × 16 mm were obtained. The sintering procedure was performed in an electric furnace (EDG 1800 3P-S), in four stages: (1) heating at 1 °C/min up to 380 °C and a holding time of 4h at this temperature for PVB burn-out; (2) heating at 1 °C/min up to 460 °C and a holding time of 4h at this temperature for PE fibers removal; (3) heating at 5 °C/min up to 1000 °C and a holding time of 4h at this temperature to provide Biosilicate® sintering; (4) cooling at 3 °C/min down to room temperature. The porous blocks were cut in small plaques of 16 × 50 mm, with 1.5 mm thick. An odontologic trephine bur (WMA®) driven by a micromotor (BELTEC®) was used to cut scaffolds of 3 mm in diameter and 1.5 mm thick. The scaffolds employed in the in vivo tests were sterilized in a hot air oven at 130 °C for 12h.
Scaffolds characterization
The apparent density and apparent porosities of the scaffolds were determined through immersion tests, based in the Archimedes’s principle, using a precision balance (Mettler Toledo AB 204). Twenty scaffolds were used to measure the sample dry weight. The scaffolds were kept in ethylic alcohol (99.5% - Synth) for 24h and the immerse weight (Wi) and the wet weight (Ww) were measured. The apparent porosity and the apparent density were calculated, using the equations (1) and (2). The values obtained correspond to the arithmetic mean of 20 samples.
By knowing the apparent density of the scaffold, it was possible to estimate the total porosity of the scaffolds using equations (3) and (4):
For microstructural observation, scaffolds were embedded in epoxy resin (Merck, Darmstadt, Germany) under vacuum. The embedded samples were ground in silicon carbide paper until the grit size 1200 and polished with cerium oxide. Then, the transversal section of the samples was coated with a thin layer of gold and analyzed by scanning electronic microscopy (SEM) (Philips TMP XL-30 and FEG XL-30). The macroporosity, i.e., the percentage of pores above 200 μm, was obtained by the analysis of SEM images using the software Image-J (version 1.45). In this case, the percentage corresponding to the area of the macropores was assumed to be equal in volume.
Bilateral non-critical size bone defects were surgically created at the upper third of the tibia (10 mm distal of the knee joint). Surgery was performed under sterile conditions and general anesthesia induced by intra-peritoneal injection of Ketamine/Xylazine (80/10 mg/Kg). The medial compartment of the tibia was exposed through a longitudinal incision on the shaved skin. A standardized bone defect (3.0 mm diameter and 1.5 mm deep) was created by using a motorized drill under copious irrigation with saline solution. Holes were compressed with gauze for 5 minutes. Immediately afterwards, a sterilized 3.0 mm diameter Biosilicate® scaffold was implanted in the bone cavities, with the exception of the control animal. After implantation, the cutaneous flap was replaced and sutured with resorbable polyglactin, and the skin was disinfected with povidone iodin. The health status of the rats was monitored daily.
On days 15, 30 and 45 post injury, rats were euthanized individually by carbon dioxide asphyxia. The tibias were defleshed and removed for analysis.
Histopathological analysis
After the euthanasia, the right tibias were removed for qualitative histological evaluation and immunohistochemistry. They were fixed in 10% buffer formalin (Merck, Darmstadt, Germany) for 48 hours, decalcified with ethylene diamine tetra-acetic acid (EDTA) (Merck, Darmstadt, Germany) for 45 days and embedded in paraffin blocks. Longitudinal thin slices (5 μm) were obtained in a serially sectioned pattern and stained with hematoxylin and eosin (H.E stain, Merck, Darmstadt, Germany). The descriptive qualitatively (per animal) were performed based on the presence of inflammatory process, granulation tissue, biomaterial presence and newly formed bone [19]. This analysis was performed by an experienced pathologist in a blinded manner.
Immunohistochemistry
Sections were deparafinized and rehydrated, then pretreated in a microwave with 0.01 M citric acid buffer (pH 6) for three cycles of 5 min each at 850 W for antigen retrieval and incubate with prepared 30% hydrogen peroxide diluted in phosphate buffered saline (PBS) for 5 min. Three sections of each specimen received polyclonal primary antibodies anti-RUNX-2, anti-COX-2, anti-RANKL and anti-BMP-9, at a concentration of 1:200 (Santa Cruz Biotechnology, USA). Incubation was carried out overnight at 4 °C within the refrigerator. After this procedure, biotin-labelled secondary antibody (ABC kit, PK-6200, Vector laboratories, Burlingame, CA, USA) was applied at a concentration of 1:200 in PBS for 1 h. The sections were washed twice with PBS followed by the application of preformed avidin biotin complex conjugated to peroxidase (Vector Laboratories) for 45 min. Colourimetric detection with a diaminobenzidine substrate (DAB, SK-4100, Vector laboratories, Burlingame, CA, USA) and hematoxylin. For a negative control, the primary antibody was omitted and PBS alone applied. Additionally, internal positive controls were performed with each staining bath.
The expression of COX-2, RUNX-2, RANKL and BMP-9 was evaluated both qualitatively (presence of the immunomarkers) and semi-quantitatively in five pre-determined fields using a light microscopy (Leica Microsystems AG, Wetzlar, Germany). A modified scoring scale) [21,22] was used as a immunomarker system to quantify positve cells per field. It was performed the scoring scale from 0 to 3 (0 = absent, 1 = weak, 2 = moderate and 3 = intense). The analysis was performed by 2 observers, in a blinded way.
Biomechanical test
For the three-point bending test, an Instron® test machine (USA, model 4444, 1 kN load cell) was used. A 3.8-cm metal support was used to locate the tibias, which provides a 1.8-cm-distant double support on the bone diaphysis (Fig. 1). A 5-N pre-load was applied (perpendicularly at the exact site of the bone defect) in order to avoid specimen sliding. Finally, the bending force was applied at a constant deformation rate of 0.5 cm/min until fracture occurred. From de load-deformation curve, the maximum load at failure (Newton, N) was obtained.
Statistical analysis
The normality of all variables’ distribution was verified using Shapiro-Wilk’s W test. Comparisons among the groups were made using one-way analysis of variance (ANOVA), complemented by Duncan’s test post-test analysis. STATISTICA version 7.0 (data analysis software system - StatSoft Inc.) was used to carry out the statistical analysis. Values of p < 0.05 were considered statistically significant.
Results
Scaffolds characterization
The stereomicroscopy images show that the scaffolds exhibit macropores formed due to the elimination of the polyethylene fibers during the sintering procedure of the Biosilicate® blocks (Fig. 2). The macropores were perpendicularly oriented to the scaffolds surface. The orientation of the macropores occurred due to the natural orientation of the fibers during mixing in the cement. As described before, the scaffolds are approximately 3 mm (diameter) by 1 mm (thickness).
The SEM micrographs of the scaffolds embedded in epoxy resin are shown in Fig. 3. The macropores were oval-shaped (Fig. 3A) and presented an average diameter of 230 μm (Fig. 3B). The oval-shape was caused by a local deformation of the Biosilicate® block during the PE fibers burn-out. It can be noted that the scaffolds were also constituted by a network of interconnected micro-channels (Figs. 3C, 3D). The existence of these micro-channels allowed the complete penetration of the resin inside the scaffold. The structure composed of Biosilicate® grains presented closed micropores smaller than 1 μm (Fig. 3F) in diameter. Additionally, it was possible to observe the presence of microcracks in the Biosilicate® grains (arrows in the Fig. 3E). Nevertheless, the scaffolds had enough mechanical strength for handling and insertion into the bone defects.
The scaffolds showed a total porosity of 44%. The macroporosity, the empty volume left by the PE fibers, was around 15%. The remaining porosity corresponds to the sum of the interconnected micro-channels and the closed pores. The results concerning porosity are shown in Table 1.
Animal experiments
General findings
Neither postoperative complications nor behavioral changes were observed in the animals. The rats returned rapidly to their normal diet and showed no loss of weight during the experimentation (data not shown). None of the animals died during the experiment and no infection in the surgical site was observed.
Histopathological analysis
Fifteen days
Figure 4 shows the histological results of CG and BG during the different experimental periods. Regarding CG, the edges of the defect were clearly visible on day 15 post-surgery. Also, the defects were almost completely filled with inflammatory and granulation tissue. Newly formed bone was seen in the peripherical region of the defect (Fig. 4A). For BG, degradation of the material allowed the ingrowth of granulation tissue accompanied by some newly formed bone. The region between the implant and the edges of the defect was in close contact between bone tissue and the material samples (Fig. 4B).
Thirty days
For CG, the borders of the injury still could be observed in this experimental period. The area of the defect was filled with granulation tissue in the central region and newly formed bone at the periphery (Fig. 4C). In the same period, for BG, the degradation of the material had continued, with granulation tissue and bone ingrowth observed in the degraded area. The pores of the remained scaffold were also filled with newly formed bone (Fig. 4D).
Forty five days
Figures 3E and 3F show the histological findings of the CG and BG 45 days after surgery. For CG, newly formed bone had grown into the area of the defects, from the periphery to the center. Most of the newly bone presented an organized tissue structure, with interconnected trabeculaes (Fig. 4E). For BG, the scaffolds were almost completely degraded. Eventually some particles still could be seen. The area of the defect was filled with mature and organized newly formed bone (Fig. 4F).
Immunohistochemistry
COX-2 expression
Fifteen days post-surgery, COX-2 expression was detected predominantly in the cytoplasm of bone cells. For CG a positive COX-2 immunoexpression was observed mainly at the medullar tissue (Fig. 5A). For BG, COX-2 expression was observed in osteoblasts and fibroblasts cells of the granulation tissue around the Biosilicate® particles (Fig. 5B). Furthermore, 30 days after the surgery, COX-2 expression was predominant in osteoblasts and osteocytes for both experimental groups (Figs. 5C, 5D). Forty five days post-surgery, no COX-2 immunoexpression was observed for both CG and BG (Figs. 5E, 5F). It is important to highlight that there was a temporal decrease of the COX-2 immunoexpression in BG, with the highest expression observed in the first period and no expression observed in the last period.
RUNX-2 expression
In the first and second periods analyzed, the RUNX-2 immunoexpression was observed in the cells of granulation tissue for both CG and BG (Figs. 6A–6D). Interestingly, 45 days post-surgery, the immunohistochemical analysis no immunoexpression of RUNX-2 was observed in CG (Fig. 6E) and a positive immunoexpression was observed in the osteoblasts cells in BG (Fig. 6F).
RANKL expression
Figure 7 shows the immunohistochemical analysis of RANKL. Fifteen day post-surgery, CG showed RANKL immunoexpression mainly in the medullar tissue. For BG, the expression of this immunomarker was evident in the osteoblasts and fibroblasts cells of the granulation tissue, especially around the Biosilicate® particles (Figs. 7A, 7B). Thirty days after the surgery, similar RANKL immunoexpression was observed for both groups, predominantly in the osteoblasts cells of the granulation tissue (Figs. 7C, 7D). In the third experimental period, CG and BG demonstrated RANKL immunoexpression mainly at osteocytes of the osteoide matrix lacunaes of both groups (Figs. 7E, 7F).
BMP-9 expression
Fifteen days post-surgery, CG showed the BMP-9 immunoexpression mainly at the medullar tissue (Fig. 8A). At the same period, BG demonstrated an BMP-9 immunoexpression in cells of the granulation tissue around the Biosilicate® particles (Fig. 8B). Positive BMP-9 expression was observed in osteoblasts and fibroblasts at 30 days post-surgery in both groups (Fig. 8C, 8D). Forty five days after the surgery, the CG presented no expression of the immunomarker (Fig. 8E). However, in the presence of Biosilicate®, BMP-9 synthesis was present in osteoblasts and fibroblastic cells of the surrounding connective tissue (Fig. 8F).
Immunohistochemistry: Semi-quantitative analysis
Fifteen days post-surgery, BG presented a statistically higher immunolabeling for COX-2 (p = 0.0005) and RUNX-2 in comparison to CG (p = 0.0147) (Figs. 9A, 9B). Thirty days after the surgery, a significantly higher value of COX-2 was observed in BG compared to CG (p = 0.0227) (Fig. 9A). At the last set point evaluated in this study, BG showed a significantly higher expression of RUNX-2 than CG (p = 0.0005) (Fig. 9B). No statistically significant differences were detected among the groups when considering the same period for RANKL expression (Fig. 9C) BMP-9 immunoexpression was upregulated on BG after 45 days (Fig. 9D).
Biomechanical analysis
The results of the biomechanical analysis are shown in Table 2. Fifteen days after surgery, no statistically significant difference was observed between BG and CG (p > 0.05). On day 30 after surgery, BG presented statistically lower values compared to the CG. On day 45 post-surgery, BG demonstrated statistically lower values of maximum load in comparison to the CG (p < 0.05).
Discussion
In this study, a novel fully-crystallized bioactive glass-ceramic (Biosilicate®) scaffold was characterized and the in vivo response of the introduction of the scaffolds was evaluated in an experimental model of tibial bone defect in rats. The material characterization showed that the scaffolds present total porosity of 44% with interconnected pores. The main findings from the in vivo study revealed that the scaffold degraded over time in the implantation site, with bone ingrowth at the area of the defect. Furthermore, the Biosilicate samples produced a significant upregulation of COX-2 and RUNX-2 imunoexpression. However, no statistical difference of RANK-L and BMP-9 immunoexpression was observed between both experimental groups. Interestingly, Biosilicate scaffolds produced a decrease in the biomechanical properties of the bone callus.
Scaffolds are three dimensional templates for cell attachment and subsequent tissue formation. The ideal scaffold for bone tissue engineering should mimic bone morphology, present interconnected pore structure and appropriate porosity (40–60%), allowing proper diffusion of nutrients and cells [23–25]. The present study showed that the total porosity and the macroporosity of the Biosilicate scaffold were 44% and 15%, respectively (with an average diameter of 230 μm and the remaining porosity corresponded to the sum of the interconnected micro-channels and the closed pores), which may prove to be efficient for intrapore introduction of cells and to have more appropriate osteogenic ability. Several studies have shown that important parameter for osteoconduction is porosity of the scaffolds. Kuboki et al. (1998) using a rat ectopic model and solid and porous particles of hydroxyapatite for BMP-2 delivery showed no new bone formed on the solid particles, while in the porous scaffolds direct osteogenesis occurred [26]. In the same way, titanium fiber-metal porous coatings (45% porosity and 350 mm average pore size) maximized bone ingrowth and increased the potential for stress-related bone resorption of femoral stems in a canine total hip arthroplasty model [27].
Various types of bioactive glasses have been developed over the last 4 decades for bone regenerative applications. Biosilicate is synthetic silica-based bioactive materials, which have unique ability to bond to bone tissue [13]. Their favorable bone behavior is related to the formation of a silica gel layer and it acts as a template for calcium phosphate (CaP) precipitation, which will directs new bone formation [7]. In this study, Biosilicate® scaffolds implantation resulted in a higher amount of newly formed bone in the area of the defect. These findings could be related to the ion release from the glass ceramic, which may have attracted osteoprogenitor cells and stimulated the differentiation osteoblasts, increasing the rate of bone formation and bone ingrowth into bioactive glass material [19,20,28].
Furthermore, fracture healing requires the cooperation of multiple molecular signaling pathway that involves many different cell types, large numbers of genes, proteins and growth factors [29]. Moreover, the process of fracture healing can be divided into several stages such as immediate injury response, inflammatory phase, intramembranous and endochondral ossification and, finally, bone remodeling [30]. Accumulating evidence suggests that inflammation plays an important role in connective tissue repair [31]. Cyclo-oxygenase is a key enzyme in the conversion of arachidonic acid to prostanoids. The expression of isoform cyclo-oxygenase-2 is relevant to many pathological processes, including inflammation and tissue repair [32]. The up-regulation of COX-2 in the Biosilicate® group was observed on 15 and 30 days post-surgery.
Some studies reported that elevated COX-2 expression increases the osteoblastic potencial of mesenchymal stem cells and supports their differentiation to osteoblasts in response to osteogenic signals [33,34]. This higher COX-2 expression may have influenced the expression of the transcription factors such as osterix and RUNX-2 and increased osteoblast activity, which could culminate in the better histological findings observed in the Biosilicate® treated animals.
In this study, the Biosilicate® group, in the different experimental periods, showed a positive RUNX-2 immunoexpression during all set points evaluated, with a higher expression of this protein on days 15 and 45 post-surgery. RUNX-2 is essential to osteoblast differentiation and regulates the expression of many extracellular matrix protein genes during bone cell differentiation [35]. It can be suggested that newly formed bone observed in the treated animals could be a result of the increased expression of COX-2 and RUNX-2 induced by the release of the positive ions of the bioactive glass biomaterial during the process of bone healing.
In the same way, RANKL is a member of the tumor necrosis factor (TNF) cytokine family, which is a ligand for receptor activator of the nuclear factor 𝜅-𝛽 (RANK) and osteoprotegerin (OPG). It works as a key factor for osteoclast differentiation and activation [36,37]. The immunohistochemical analysis demonstrated tendency to increase the RANKL immunolabeling in BG group at all experimental periods. Also, it was possible to observe immunoexpression in the osteoblasts cells and the granulation tissue, especially around the Biosilicate particles. It may be suggested that the the monocyte-macrophage lineage, including macrophages and osteoclasts, could continuously be adhered to the surface of Biosilicate®, in an attempt of degrading the material and participate in the process of bone turnover. The same results were observed by Kondo et al.(2005), who showed a constant immunolabeling of tartrate-resistant acid phosphatase (TRAP) on days 4, 7, 14, 28 and 56 around b-tricalcium phosphate (b-TCP) particles, after implantation in rat femoral condyle, indicating a continuous action of osteoclasts [38].
Concerning the role of BMPs in bone healing, evidences suggested that these proteins could stimulate differentiation of mesenchymal cells into osteogenic/chondrogenic lineage and increase expression of alkaline phosphatase and osteocalcin [39]. Moreover, BMPs also affect bone remodeling through the regulation of osteoclast bone-resorbing activity [40]. In addition, it has recently been shown that BMP-9 is one of the most potent osteoindutive BMPs [41]. The imunohistochemical analysis showed that BMP-9 immunoexpression could be observed with a similar intensity for both groups, 15 and 30 days post-surgery. However, at the last period evaluated, the BMP-9 immunoexpression was observed only in the BG. From these results, it is possible to hypothesize that the particles of Biosilicate® that still could be observed at the site of the defect in the last experimental period was still exerting an osteogenic potential, stimulating BMP-9 upregulation up to this period and also contributing to increase newly formed bone deposition. To the best of our knowledge, the effects of bioglasses on BMP-9 expression have not been addressed so far. Okabe et al. (2010) found a significantly larger BMP-2 immunolabeling in the defects in the nasal bone of rabbits filled with 𝛼-tricalcium phosphate compared to controls, 4 and 12 weeks after implantation [42]. These findings suggest the osteogenic potential of this material.
Despite the stimulatory effects of Biosilicate scaffolds on the newly formed bone, the scaffolds produced a decrease in the biomechanical properties of the bone callus on days 30 and 45. It may be suggested that the lack of the improved load-bearing capacity and stiffness showed by the biomaterial groups probably mirrors the difference in the bone maturation and/or spatial distribution of newly formed bone into the defect site among the groups.
As this study was limited to histopathological analysis and protein expression evaluation, information about the influence of Biosilicate scaffolds on gene expression needs to be provided. Also, more quantitative analysis should be included in future research such as the quantification of inflammatory cells on immunohistochemical analysis.
Conclusion
In summary, this study demonstrated that Biosilicate® scaffolds presented an osteogenic potential, stimulating the formation of bone tissue into the tibial defects of rats. Moreover, the dissolution of the ions from the biomaterial upregulated factor related to bone cells activity. These data highlight the huge potential of the use of the studied scaffolds to be used as bone regeneration applications. Further long-term studies should be developed to provide additional information concerning the late stages of the bone matrix synthesis and degradation induced by this material.
Footnotes
Acknowledgements
The authors thank the Brazilian funding agencies FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo) and CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) for the financial support of this research.
Conflict of interest
None to report.