Asprin‐loaded strontium‐containing α‐calcium sulphate hemihydrate/nano‐hydroxyapatite composite promotes regeneration of critical bone defects

Abstract Our laboratory originally synthesized strontium(Sr)‐containing α‐calcium sulphate hemihydrate/nano‐hydroxyapatite composite (Sr‐α‐CSH/n‐HA) and demonstrated its ability to repair critical bone defects. This study attempted to incorporate aspirin into it to produce a better bone graft material for critical bone defects. After 5% Sr‐α‐CSH was prepared by coprecipitation and hydrothermal methods, it was mixed with aspirin solution of different concentrations (50 μg/ml, 200 μg/ml, 800 μg/ml and 3200 μg/ml) at a fixed liquid‐solid ratio (0.54 v/w) to obtain aspirin‐loaded Sr‐α‐CSH/n‐HA composite. In vitro experiments were performed on the composite extracts. The tibial defects (3 mm*5 mm) in SD rat model were filled with the composite for 4 weeks and 12 weeks to evaluate its osteogenic capacity in vivo. Our results showed its capability of proliferation, migration and osteogenesis of BMSCs in vitro got improved. In vivo treatment with 800 μg/ml aspirin–loaded Sr‐α‐CSH/n‐HA composite led to significantly more new bone formation in the defects compared with Sr‐α‐CSH/n‐HA composite and significantly promoted the expression of osteogenic‐related genes and inhibited osteoclast activity. In general, our research suggests that aspirin‐loaded Sr‐α‐CSH/n‐HA composite may have a greater capacity of repairing tibial defects in SD rats than simple Sr‐α‐CSH/n‐HA composite.

a very complex process. For decades, with joint efforts by researchers and clinicians, new methods of bone tissue engineering and new bone graft techniques have been introduced to solve this problem.
There are a wide variety of bone graft substitutes (BGS) used in bone tissue engineering, such as autologous bone, allogeneic bone, xenogeneic bone and various biosynthetic bone. As the gold standard treatment for bone defects, autogenous bone grafting has demonstrated advantages in bone healing without immune rejection.
However, limited availability of proper materials and likely complications at donor site like bleeding and infection have forced researchers to discover alternatives. As a substitute for classic grafts, biosynthetic materials can have an unrestricted supply and avoid autogenous bone-related drawbacks. 6 Novel engineered bone strives to meet various regenerative requirements of bone tissue simultaneously.
Previous studies have demonstrated that defects of single-component material are inevitable, indicating a necessity to combine two or more materials to achieve better outcome. 7 Attempts have been made to develop BGSs with antibacterial properties to control infection and increase their rate of successful implantation. It has been reported that antibiotics can be released for 20 days or more continuously in animal models using biodegradable polymer scaffolds containing gentamicin and vancomycin and that local delivery also prevents systemic side effects. 8 Moreover, synthetic materials can be loaded with various osteoinductive factors such as bone morphogenetic protein (BMP), insulin-like growth factor (IGF) and transforming growth factor-β (TGF-β) to enhance their osteoinductive ability for better bone formation and bone remodelling. [9][10][11][12] α-calcium sulphate hemihydrate (α-CSH), a class of highly cementitious bone substitute widely used in clinic, has shown to possess superior biocompatibility, biodegradability and osteoconductivity.
Recent studies have combined α-CSH with other bioactive inorganic materials to improve its osteogenicity and physicochemical properties. [13][14][15] Our laboratory previously synthesized strontium(Sr)-containing α-hemihydrate calcium sulphate (Sr-α-CSH) which had a benefit of osteoinductivity in addition to the bone conductivity and biocompatibility of calcium sulphate. 16 However, its rate of degradation was too fast. To solve this problem, our laboratory conducted a further research where Sr-α-CSH was combined with nano-hydroxyapatite (n-HA). The new Sr-α-CSH/n-HA composite showed good biocompatibility, osteoinductivity and improved degradation rate, but its bone support was weak because its compressive strength decreased significantly with increased concentration of strontium. 17 To obtain a better BGS, we next tried to load aspirin into Sr-α-CSH/ n-HA to decrease its concentration of Sr to improve its performance in dealing with critical bone defects.
Aspirin, the most active component of the non-steroidal anti-inflammatory medication, is a very popular antipyretic, anti-inflammatory and analgesic drug. Its other roles in treating diseases have been found. It has a certain impact on bone metabolism and bone health, promoting osteogenic differentiation of bone marrow mesenchymal stem cells (BMSCs), inhibiting adipogenic differentiation of BMSCs, activating osteoblasts and inhibiting osteoclasts. [18][19][20][21][22] It has been found to promote bone formation by inhibiting the expression of inflammatory factors such as IFN-γ and TNF-α. [23][24][25] It may also improve bone marrow microenvironment and enhance immune regulation of BMSCs. We thus hypothesized that incorporation of aspirin into Sr-α-CSH/n-HA composite might produce a better BGS for critical bone defects.
The purpose of this study is to combine Sr containing α-CSH/nano-hydroxyapatite (n-HA) composite with different concentrations of aspirin, and to test the structure, physicochemical properties, biocompatibility and ability to stimulate the proliferation, migration and differentiation of BMSCs in vivo and in vitro. In addition, the material was implanted into an established tibia bone defect model in SD rats for analysis. Our present study demonstrated that aspirin-loaded strontium-containing α-calcium sulphate hemihydrate/ nano-hydroxyapatite composite has a potential of a proper substitute for classic grafts in treatment of critical bone defects.

| Characterization and physicochemical properties of aspirin-loaded SR-α-CSH/N-HA composite
The crystal structure of the composite material was analysed by XRD (Bruker, D8, Germany). Aspirin-loaded Sr-α-CSH/n-HA composite was ground for testing. Different materials had their specific crystal structure and unique parameters such as lattice type and interplanar spacing. The characteristics of crystallographic reflection Five samples from each experimental group were used to calculate the mean and standard deviation of the results. SEM (Phenom XL G2 Netherlands) was used to observe the microstructure of the composite. The experimental sample used to be tested before SEM was pre-treated according to the following procedures (Vacuum drying and gold spraying).

| Preparation of aspirin-loaded Sr-α-CSH/n-HA composite extracts
The aspirin-loaded Sr-α-CSH/n-HA composite extracts were prepared according to international standards. 16 In brief, the composite

| Isolation and culture of BMSCs
Isolation of BMSCs from the SD rats was done following the established protocol. 24 In brief, after the femurs and tibias were removed, the bone marrow cells were flushed out from the bone cavity of femurs and tibias with growth medium for BMSCs carefully. After all the bone marrow cells passed through a 70-μm cell strainer (BD Bioscience), single-cell suspension of all nuclear cells was obtained. Thirty to fifty million cells were seeded onto 10cm culture dishes (Corning) for initial incubation for 48 hours with a-MEM supplemented with 15% FBS, 2 mM L-glutamine (Invitrogen), 55 mM 2-mercaptoethanol (Invitrogen), 100 U/mL penicillin and 100 mg/mL streptomycin (Invitrogen) at 37℃ and 5% CO2 in a humidified environment. Cells were passaged once they became 70% to 80% confluent. BMSCs at passage 3 were used in our study.

| Osteogenic differentiation evaluation
Osteogenic differentiation evaluation was performed as previously reported. 26 Osteogenesis mineralization of BMSCs by composite extract was detected using Alizarin red staining. The gene expression of RUNX2, OCN and BSP was assayed by real-time PCR (RT-PCR).
All mRNA quantification data represent the mean ± standard error of the mean of triplicate experiments.

| In vitro scratch test
BMSCs (1.5 × 10 5 cells/well) were plated in six-well plates for 24 hours and wounded by scratching with a 10 μL pipette tip. Cellular debris was removed by washing with PBS and then incubated in a-MEM (Invitrogen) supplemented with 5% FBS, 2 mM L-glutamine (Invitrogen), 55 mM 2-mercaptoethanol (Invitrogen), 100 U/mL penicillin and 100 mg/mL streptomycin (Invitrogen) at 37℃ and 5% CO2 in a humidified environment. The width of the scratch was determined microscopically immediately after creation and 6, 24, 48 hours later using a phase-contrast microscope (Olympus, Tokyo, Japan). The wounded areas were quantified as wound width by Photoshop (PS).

| Transwell assay
The migration assay was tested using transwell plates (Corning Costar, USA) that were 6.5 mm in diameter with 8 μm pore filters.  Studies have shown that cortical bone defects in the tibia of SD rats greater than 4 mm × 3 mm cannot be repaired by themselves under normal physiological conditions. We created a bone defect model of 5 mm × 3 mm. The 8-week-old SD rats were randomized into 4 groups (n = 12) subjected to treatments (half on week 4 and half on week 12) by Sr-α-CSH/n-HA composite loaded by aspirin of concentrations of 0, 50 μg/ml, 200 μg/ml and 800 μg/ml, respectively. All animals were given a one-week adaptive feeding before the experiment. In order to avoid the difference in skill between operators, all surgical operations were performed by the same person.

| Imaging evaluation
Immediately after operation, X-ray was performed to exam whether the size of a bone defect was appropriate and whether the filling material was loose or not. 3D microarchitecture of the tibia was evalu-

| Histological evaluation and quantitative analysis of regenerated bone
The tibial samples were harvested at 4 weeks and 12 weeks postsurgery, respectively. Bone specimens were fixed in 4% buffered formalin for 24 hours. The specimens were decalcified and embed-

| Statistical analysis
The results were expressed as mean ± SD of three independent experiments. The significance of variability was analysed by two-tailed Student's t-test or one-way ANOVA followed by Dunnett test. P value <0.05 was considered to be significant in all tests. Statistical analyses were performed with SPSS 20 (IBM).

| Characteristics of aspirin-loaded Sr-α-CSH/n-HA composite
Results of the X-ray diffraction (XRD) analysis of the aspirin-loaded Sr-α-CSH/n-HA composite are shown in Figure 1A. The characteristic crystal diffraction peaks of α-CSH appeared at 15°, 25°, 30°, 31° and 48°2 8 and the characteristic peak of Strontium was at 24.8°. 29 n-HA showed a triplet or broad strong diffraction peak in the range from 31.8° to 34.1°, and an apatite phase diffraction peak at 25.9°. 30 After aspirin loading, the position of the diffraction peak of the Sr-α-CSH/n-HA composite did not change significantly, but the intensity of the diffraction peak increased with the increased concentration of aspirin loaded, indicating the properties of the Sr-α-CSH/n-HA composite did not change significantly after it was loaded with aspirin. FTIR ( Figure 1B) has a high sensitivity for analysis of the inferred material composition, spatial conformation, qualitative and effective functional groups and polymerization crystallization. 31 Figure 1C. With the aspirin concentration in the composite material increasing, the compressive strength of the Sr-α-CSH/n-HA composite did not change significantly, and the differences among groups were not statistically significant (P > 0.05). SEM ( Figure 1D) showed that the aspirin-loaded Sr-α-CSH/n-HA composite was rough and porous. The crystals were arranged in a short rod shape and a lamellar shape, and no obvious difference was observed among groups. The above results showed that the characteristics of Sr-α-CSH/n-HA composites were not significantly changed after aspirin loading.

| Effects of aspirin-loaded Sr-α-CSH/n-HA composite on proliferation and migration of BMSCs
As shown in Figure 2C, Sr-α-CSH/n-HA loaded with aspirin of concentrations of 50 μg/ml, 200 μg/ml and 800 μg/ml showed a promoting effect on BMSCs proliferation while Sr-α-CSH/n-HA loaded with a high concentration of aspirin (3,200 μg/ml) displayed an inhibitory effect on cell proliferation. Differences between groups were statistically significant (P < 0.05). As Sr-α-CSH/n-HA loaded with 3200 μg/ ml aspirin led to cytotoxicity for BMSCs, our following experiments did not take this concentration into consideration. The effects of extracts of aspirin-loaded Sr-α-CSH/n-HA composite on migration of BMSCs were shown in Figure 2A. It was obvious that the BMSCs processed by extracts of aspirin-loaded Sr-α-CSH/n-HA showed an increasing migration ability than those of Sr-α-CSH/n-HA material in a dose-dependent manner in the range of this experiment. The statistical analysis of scratch healing rates between groups showed that the differences among groups were statistically significant ( Figure 2D) (P < 0.05). As the concentration of aspirin increased, the effect of scratch repair increased in the present study. The results of Transwell test were consistent with those of the scratch test. With the increase in aspirin concentration, the migration rate of BMSCs was significantly accelerated ( Figure 2G-H).

| Aspirin-loaded Sr-α-CSH/n-HA stimulates the osteogenesis of BMSCs in vitro
As shown in Figure 2B, when BMSCs were subjected to osteogenic inductive conditions, treatment with aspirin-loaded Sr-α-CSH/n-HA increased their capability of forming Alizarin red-positive calcified deposits. Extracts of aspirin-loaded Sr-α-CSH/n-HA resulted in significantly more Alizarin red-positive cells compared with those of Sr-α-CSH/n-HA in a dose-dependent manner ( Figure 2F) (P < 0.05).
As the concentration of aspirin increased, there were more Alizarin red-positive cells. This was confirmed by up-regulation of osteogenesis-related genes (Runx2, ALP and BSP) after BMSCs were treated with extracts of aspirin-loaded Sr-α-CSH/n-HA in osteoinductive conditions for one week ( Figure 2E). Differences among groups were statistically significant (P < 0.05). All these data indicated that aspirin-loaded Sr-α-CSH/n-HA might stimulate the osteogenesis of BMSCs in vitro.

| Aspirin-loaded Sr-α-CSH/n-HA composite promotes critical bone defect regeneration in SD rats
As shown by the X-ray taken immediately after surgery to observe whether there was loosening or shedding in the aspirin-loaded Sr-α-CSH/n-HA composite filling the tibial bone defects freshly created in SD rats ( Figure 3A and B), the material was well located with no loosening or shedding in all groups. Micro-CT analyses at 4 weeks

| Aspirin-loaded Sr-α-CSH/n-HA stimulates osteogenesis of BMSCs in vivo
At 4 weeks and 12 weeks, sections of tibial specimens were harvested for histological examination using HE staining ( Figure 5A

| D ISCUSS I ON
It is a great challenge to develop a promising BGS which can promote high quality new bone formation. 20 The present study demonstrated that the aspirin-loaded Sr-α-CSH/n-HA composite was capable of promoting regeneration of critical bone defects in SD rats because topical administration of aspirin held a great advantage of promoting osteogenesis of BMSCs. The aspirin-loaded Sr-α-CSH/n-HA may also have some therapeutic effects on bone defects caused by tumour resection, malformation, sports injury and infection. [1][2][3][4] The present study started with analyses of the crystal struc- Similarly, our study demonstrated that Sr-α-CSH/n-HA composite loaded with aspirin of a safe range of concentrations was capable of promoting osteogenesis both in vitro and in vivo with no cytotoxicity to BMSCs and had a better effect on regeneration of critical bone defects than Sr-α-CSH/n-HA material in SD rats. What is more, Sr-α-CSH/n-HA loaded with 800 μg/ml aspirin Our research has certain limitations. First, we did not detect the actual concentration of aspirin in the extract. We believed that testing the aspirin concentration alone was not reasonable enough to explain the effect of composite material in the treatment of bone defect because the synergy effect on osteogenesis should be associated with Sr-α-CSH/n-HA composite and aspirin. However, as the safety of biomaterials must be our priority, we first excluded the group showing a cytotoxic effect by aspirin-loaded Sr-α-CSH/n-HA composite. As our animal experiments showed that 800 μg/ml had the best effect on bone defect repair and 3200 μg/ml aspirin in the cell experiment inhibited the proliferation of BMSCs, we just inferred that the actual concentration of aspirin should have been greater than 200 μg/ml according to previous reports. Another shortcoming of ours is that we only studied the biological and osteogenic properties of aspirin-loaded Sr-α-CSH/n-HA composite but did not further investigate their specific mechanisms and signalling pathways.
Further experiments are required.

| CON CLUS IONS
Aspirin-loaded Sr-α-CSH/n-HA may be capable of promoting tibia bone regeneration in SD rats. Local administration of aspirin, coupled with Sr-α-CSH/n-HA, has a twofold effect on regeneration of critical bone defects, alleviating inflammatory response at sites of disease, increasing bone formation and, to some extent, preventing IAI.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

AUTH O R S' CO NTR I B UTI O N S
Hanjun Qin: Formal analysis (equal); Methodology (equal); Software