Abstract
Hydroxyapatite (HA) possesses similar mineral components to bone and possesses good physicochemical properties. Even though pure HA scaffold is brittle, it is insufficient in promoting vascularization and osteoinductivity. This study was conducted to assess whether lithium (Li) incorporated into HA could improve the scaffolds’ inherent shortcomings. In the experiments, Li-hydroxyapatite scaffolds’ mechanical strength, biocompatibility, and biodegradability were researched primarily. In vivo studies, the Li hydroxyapatite scaffolds were implanted into an animal model to repair the bone defects. Meanwhile, we also evaluated the expression of angiogenic and osteogenic factors. For comparison, autologous bone, hydroxyapatite, and blank control groups were designed. According to the results, Li incorporated with hydroxyapatite did not significantly change the scaffold’s degradation velocity, but it obtained higher compress mechanical strength. After Li was doped, bone regeneration was further enhanced but the angiogenic effect was not improved significantly. The in vivo study, Li-HA scaffolds improved new bone formation with GSK-3𝛽 decreased and 𝛽-catenin increased. In conclusion, doped Li into hydroxyapatite was an alternative strategy for improving hydroxyapatite’s mechanical property and promoting the osteogenesis potential. This method is highly recommended for clinical application based on this study alone.
Introduction
Bone injuries and defects are serious health complications, especially those caused by deformity, osteoporosis, and tumors. This results in over 2.2 million bone transplantation procedures each year in orthopedics, neurosurgery, and dentistry [1]. Therefore, numerous investigations are in progress, making efforts on the development of new materials and processing techniques for bone tissue regeneration [2,3]. Tissue engineering is a promising alternative for treating bone defects. The main operation method is to prepare a scaffold with good biocompatibility, which plays a crucial role in tissue engineering, because they represent an alternative to the conventional implantation of organs and tissues. The main goal of scaffolds is to provide appropriate base for tissue growth and cell proliferatio [4]. Nowadays, in clinical therapies for bone regeneration, autologous bone, allografts, demineralized heterologous bone or bone substitutes are used [5]. Autologous bone, the current gold standard of bone grafts, is often used to treat bone defects in the clinic [6]. But its source is limited and may cause donor site pain and infection [7–9]. Allogenic bone is another material which can be selected; nevertheless, it also has some problems including infection and immunoreaction [10], limited supply and too fast of grafting bone absorption. The ideal bone materials should be made up of the substance that is similar to the natural bone, and they ought to be provided with good mechanical performance and controllable degradation properties [11–13]. In addition, the scaffolds with good osteogenic and angiogenic potential has been reported to be good for bone defect repair and regarded as promising bone substitute in recent years [14,15]. Hydroxyapatite [HA, Ca10(PO4)6(OH)2], the main inorganic component of the vertebrate animals’ teeth and bones (as high as 50%), has been widely used as teeth and bone repairing/replacement materials due to its properties of endophilicity, non-toxic, non-stimulating, non-allergenic, non-mutagenic, and osteoconductive, etc [16]. Beta-tricalcium phosphate (TCP) has been also widely used as bone substitutes [58]. Nevertheless, pure HA scaffold also has some shortcomings, for example, it is brittleness and hard to be degraded, meanwhile HA is insufficient in promoting vascularization and osteoinductivity, all of which limit its application [17]. In order to address these challenges and precisely match the irregular boundaries of bone defects as well as facilitating clinical application, scaffold design should be improved [4]. At present, HA is often combined with other materials to make a kind of bone repair material which is similar to natural bone tissue. This is one of the hot spots in the research of bone repair [18,46].
Lithium (Li) has been used as psychiatric medication for half a century in clinics and is also reported to take effect in increasing bone density [19]. Furthermore, studies have demonstrated that the activation of the Wnt/𝛽-catenin signal pathway may promote osteoblast proliferation and differentiation in vitro [20] and bone mineralization and formation in vivo [21]. However, the activity of this pathway is constitutively suppressed by glycogen synthase kinase-3𝛽 (GSK-3𝛽) [22]. Moreover, the lithium (Li) ion, which can activate the canonical Wnt signal pathway by inhibiting GSK-3𝛽 [23], has proved to promote alkaline phosphatase and osteogenic genes expression in osteoblasts [24]. Therefore, Li is a good choice to modify the HA. In our previous studies [59], we found that 1.5%Li-HA has good mechanical compressive strength and degradation properties. Otherwise, it obtains optimal cell proliferation activity and osteogenesis in vitro, but its osteogenic effects in vivo is not intensively studied. This research is conducted to solve the problem.
This study aims to evaluate the biological activities, degradation and osteogenetic activities of Li-HA scaffold in rabbit tibial defect model.
Materials and methods
Scaffold fabrication and characterization
Lithium-hydroxyapatite (Li-HA) powders were obtained by using chemical precipitation [17]. Briefly, Ca(NO3)2⋅ 4H2O and NH4H2PO4 (Aldrich Chemical, Milwaukee, WI, USA) were dissolved in distilled water separately to form 0.5 M solution, and then each solution was stirred until the material was dissolved completely. Subsequently, the pH of NH4H2PO4 solution was buffered close to 10–11 by NH3⋅ H2O, and then poured the adjusted NH4H2PO4 solution into Ca (NO3)2⋅ 4H2O solution slowly, stirring and adding LiNO3 solid with a molar ratio of Ca2+/Li+ = 98.5:1.5, which formed 1.5%Li-HA. The last mixtures were stirred for 1 h at 60 °C and precipitation was completed. The precipitates were placed at room temperature for 48 h, and were then washed with distilled water until the pH of filtrate was about 7. Li-doped HA precursor was achieved via filtering and drying (at 80 °C) the washed precipitates, and then the precursors were ground and screened to produce powders. The powders mixed with 40 wt.% stearic acid as the porogen, were pressed into a disk under the pressure of 2 MPa. Finally, these disks were sintered at 1100 °C for 4 h to form porous Li-HA scaffolds with 10 mm diameter and 1 mm thickness for in vivo experiments. Besides, the scaffolds with 10 mm diameter and 15 mm thickness were also prepared for compressive strength testing.
The FTIR spectra of the samples were obtained using a FTIR spectrophotometer (Bomem MB-120). The scaffolds were tested by X-ray diffractometer (XRD, 3KWX, Philips, The Netherlands), which was used for the qualitative analysis of element. The liquid displacement was generally employed to measure the porosity value of the scaffolds. The morphology of obtained scaffolds was observed by scanning electron microscopy (SEM, Japan Electronics Co., Ltd., Japan). The scaffolds for implanted experiment were sterilized by 𝛾-rays (60Co) at a dose of 25 kGy.
Degradation property and compressive strengths
Simulated body fluid (SBF) was used for in vitro bioactivity experiments. SBF are nearly the same as those of human blood plasma [47].
Scaffold specimens with dimensions of 10 × 1 mm were immersed in plastic tubes containing 10 mL of SBF. The tubes were sealed to minimize the change in pH and incubated in a water bath at 37 °C under gentle interactive motion. The samples were incubated in SBF for 3, 6, 9, 12, 15, 18, 21, 24, 27 and 30 days, and the SBF was changed every 2 days. After the different periods of immersion time, specimens were removed out from SBF, rinsed with distilled water and dried prior to analysis. Scaffolds weight was measured and the immersion fluid was gathered for measuring ionized lithium/phosphate concentrations at the 3, 6, 9, 12, 15, 18, 21, 24, 27 and 30 days. The concentrations of Li ion and phosphate in SBF after immersion were analyzed by inductively coupled plasma mass spectrometry (ICP-MS, IRIS, Thermo Elemental, USA). The cumulative release amount of Li+ and phosphate was calculated at the predetermined times. The compressive strengths of the dried scaffold were measured by an universal testing machine (Model 4881, Instron, Norwood, MA, USA) at a crosshead speed of 0.5 mm/min. For each index, two samples were measured.
study
Preparation of animals
In order to evaluate the osteogenic potential of the Porous lithium-hydroxyapatite scaffold when implanted into animal bone defect, we tried to establish a lacunar bone defect model based on our previous study [48], which was easy to create in the bearing limbs and cannot heal spontaneously but does not cause fracture. After our preliminary experiments, we found that the tibia lacunar bone defect under 2 cm of the knee joint in rabbits with 10 mm in diameter met our satisfaction.
Forty healthy adult male Japanese white rabbits (Animal Center of Sichuan University, Chengdu, China) weighing 2.8–3.2 kg, and with normal limbs comprised the experimental animal model and were used to build bilateral tibia lacunar bone defect. The rabbits were randomly divided into four groups: group A (autologous bone. The autogenous cancellous bone was taken from the ilium through the same incision.); group B (1.5%Li-HA scaffolds); group C (HA scaffolds) and group D (blank control group). The animals were fed a standard diet and allowed free activity. The animal protocol was reviewed and approved by the Animal Care and Committees at the West China Medical School of Sichuan University.
Surgical procedures
The animals were anesthetized with 0.1 ml/kg of sumianxin II via intramuscular administration. During operation, intravenous injection of 10% of the chloral hydrate was performed to maintain the deep anesthesia. With the supine position with the arms and legs fixed, the areas of the bilateral tibia were shaved before surgery, and the surgical field was prepared with an iodine solution. We made a 2 cm longitudinal incision at the center of the bilateral medial of tibia 2 cm under the knee joint, and exposed the tibia. A symmetrical round 5 mm diameter bony defect was then formed at the medialis face of tibia using 5 mm diameter trephine.
After the bone defect was made, group A was implanted with the autogenous cancellous bone; Group B was implanted 1.5%Li-HA scaffolds; while group C was implanted with HA scaffolds and group D was filled with nothing. To prevent the wound infection, all animals were intramuscularly injected with gentamicin (4 mg/kg) one preoperative dose and two postoperative doses. Five rabbits were killed respectively in each group at 6 and 12 weeks and then the tissue samples were prepared. The right tibias were used for radiological detection and histological detection, while the left tibias were used for qPCR and western blot analysis.
Radiologic analysis
To monitor the bone defect healing, X-ray radiographs of tibials were obtained after animals were sacrificed at 6 weeks and 12 weeks. BI-2000 medical image analysis system was used to analyze the X film, which measured the mean gray value of the defect regions. In radiographs, we observed the scaffolds position, new bone formation and the defects healing. After the samples were obtained, Micro-CT analysis was conducted. Micro-CT system (Inveon MultiModalityGantry-STD CT) was used for evaluating the new bone formation in the defect channel in the tibial in this study. Data sets with isotropic 20 mm voxel spacing was acquired at 0.5° steps over total rotation of 360° at 80 kV. And then, original data was managed using Cobra software for three-dimensional reconstruction of the tibial. At last, we calculated the bone volume fraction (bone volume/total volume) and bone density in Inveon research workplace v2.2.0.
Histological and histometric analysis
The implants were cut and fixed in the 10% formaldehyde for a week. The implants were then decalcified in 20% EDTA solution for about six months and then dehydrated through a gradient ethanol series, cleared in xylene and embedded in paraffin. Sections of 3 mm thickness were made and stored in thermostat of 37°. Specimen sections was stained with haematoxylin and eosin (HE) and Masson’s trichrome stain. At last, each piece of the sections were observed by light microscope (BX41, Olympus, Japan).
Quantitative real-time PCR
In order to evaluate the osteogenic, adipogenic and angiogenic factors (Col-I, ALP, OCN, Runx-2, PPAR-𝛾2, VEGF) expression, and analyze the related signal pathway regulation (GSK-3𝛽, 𝛽-catenin) on molecular level, qPCR was conducted. The specimens obtained at 12th week postoperatively were snap frozen in liquid nitrogen and completely grounded into powder and were homogenized. According to manufacturer’s instructions, total RNA was extracted with TRIzol solution (Invitrogen, Carlsbad, CA, USA) and reverse transcribed to complementary DNA using the QuantiTect Reverse Transcription Kit (Thermo Fisher Scientific Inc, Fremont, CA, USA). The specific transcripts were quantified by quantitative real-time polymerase chain reaction (PCR) with QuantiTect SYBR Green PCR Kit (Takara Bio Inc., Tokyo, Japan) and analyzed using ABI 7500 real-time PCR systems (Applied Biosystems, Foster City, CA, USA). The following primers sequences were GSK-3𝛽 forward primer: 5 ′ -AAT CCC ATG GAG CTA AGC AGT-3 ′ and reverse primer: 5 ′ -CTC CAG GCA GTA CTG GTT CTT-3 ′ ; 𝛽-catenin forward primer: 5 ′ -GGA AAG CAA GCT CAT CAT TCT G-3 ′ and reverse primer: 5 ′ -AGT GCC TGC ATC CCA CCA GCT T-3 ′ ; ALP forward primer: 5 ′ -TGG ACA AGT TCC CCT TC-3 ′ and reverse primer: 5 ′ -CAC AGG TAG GCG GTG GC-3 ′ ; OCN forward primer: 5 ′ -CCC AGG CGC TAC CTG TAT CAA-3 ′ and reverse primer: 5 ′ -GGT CAGCCAA CTC GTC ACA GTC-3 ′ ; Runx-2 forward primer: 5 ′ -TAC TGT CAT GGC GGG TAA CG-3 ′ and reverse primer: 5 ′ -GGT TCC CGA GGT CCA TCT AC-3 ′ ; PPAR-𝛾2 forward primer: 5 ′ -GCT AGG TAA CTC GTG TCC CG-3 ′ and reverse primer: 5 ′ -GAC AGG AAA GAC ACC TCG CA-3 ′ ; Col-1 forward primer: 5 ′ - CTC CTG ACG ATG TTC AGC T-3 ′ and reverse primer: 5 ′ -CTT CAC TGT ACC GGA CGC C-3 ′ ; VEGF forward primer: 5 ′ -TAT ACT CGC GCT ACC TGC C-3 ′ and reverse primer: 5 ′ -GAC ATC TCC CCC AGC TGA C-3 ′ ; GAPDH Forward primer: 5 ′ -GGT CAC CAG GGC TGC TTT TA-3 ′ and Reverse primer: 5 ′ -CCA GCA TCA CCC CAC TTG AT-3 ′ . PCR was performed in 40 cycles at 94 °C for 15 s, 55 °C for 30 s, and 72 °C for 30 s. The relative Mrna expression was calculated.
Immunohistochemistry and immunofluorescence detection
In order to detect to protein expression, immunohistochemistry was used to examine to GSK-3𝛽 (Glycogen synthetase kinase-3𝛽), 𝛽-catenin, Bone glaprotein (OCN) and immunofluorescence was used for Collagen type one (Col-I) analysis. Sections were fixed and the primary antibodies were diluted to the most optimal concentration [1:200 anti-GSK-3𝛽 (ab75814), from Abcam], [1:500 anti-𝛽-catenin (ab32572), from Abcam], [1:200 anti-OCN (ab198228), from Abcam] and [1:500 anti-Col-1(ab34710), from Abcam]. The sections were then stained with anti-antibodies and visualized with a secondary antibody conjugated with Cy5 or without Cy5 for immunohistochemistry. The controls were also stained, but without the primary antibodies. Then the sections were treated with 50 μg/ml 4,6-diamidino-2-phenylindole (DAPI) for nuclear staining. Fluorescence images were acquired using a fluorescence microscope (FluoView 500; Olympus, Tokyo, Japan). While immunohistochemistry images were observed using light microscope (BX41, Olympus, Japan).
Western blot analysis
Western blot analysis was also performed evaluate the proteins including GSK-3𝛽, 𝛽-catenin, Alkaline phosphatase (ALP), OCN, Runt related genes2 (Runx-2), Col-I, Peroxisome proliferator activated receptor-𝛾2 (PPAR-𝛾2) and Vascular endothelial growth factor (VEGF). Total proteins were separated on 10% SDS-PAGE gels and then transferred onto polyvinylidene fluoride blotting membranes. After blocking non-specific binding with 5% BSA in Tween-Trisbuffered saline, the membranes were incubated with the following primary antibodies: anti-GSK-3𝛽 (ab107166; Abcam, USA), anti-𝛽-catenin (ab194120; 1:5000 dilution; Abcam, USA), anti-ALP (ab95462; 1:500 dilution; Abcam, USA), anti-OCN (ab133612; 1:1000 dilution; Abcam, USA), anti-Runx2 (ab114133; 1:100 dilution; Abcam, USA), anti-PPAR-𝛾 (ab59256; 1:1000 dilution; Abcam, USA), anti-Col-1 (ab138492; 1:1000 dilution; Abcam, Cambridge, CA, USA), anti-VEGF (ab150766; 1:1000 dilution; Abcam, USA) and Anti-GAPDH (ab8245; 1:500 dilution; Abcam, USA) antibodies overnight. The membranes were then incubated for 2 h with secondary antibodies labeled with horseradish peroxidase. The immunoreactive proteins on the blots were visualized using ECLTM western blot detection reagents, and the signals were detected using Image Station 4000R (Kodak, Rochester, NY, USA).
Statistical analysis
The SPSS19.0 software (SPSS Inc., Chicago, IL, USA) was used for the statistical analysis. The results were analyzed with one-way variance analysis, the comparison between the two groups was expressed by SNK-q test and LSD test, with p < 0.05 indicating statistical significance.
Results
The composition of Li-HA
Figure 1 indicates the FTIR spectra of HA and 1.5%Li-HA. The two samples in Fig. 1 demonstrate similar characteristic peaks. It reveals the important information that low-content lithium had no effect on the FTIR spectra of scaffolds.
Identification of the materials: FTIR spectra of Li-HA and HA.
Figure 2a shows the surface morphology of the 1.5%Li-HA and we found that the pores were uniformly distributed and we can found that the pores were uniformly distributed and the pore size was 200–300 μm. The pores in the implant are used to help in providing a coherent interaction between the implant and body fluid. Ideally, high porosity has indicated to give better interaction between implant and body fluid which in turn induced better bone growth via osteoinduction and osteoconduction [57]. Practically, various techniques such as Gas Pycnometer, Mercury intrusion porosity, Computer tomography etc. are used to determine the porosity. In this study, Mercury intrusion porosimetry (Micromeritics, AutoPore IV1.08) was used to determine the porosity of the samples. The 1.5%Li-HA scaffolds had porosity of 63.7% ± 6.8%. But HA scaffolds had porosity of 69.2% ± 7.3%, the possible explanations were metal ion Li reduced the porosity of the samples. Figure 2b depicts the higher power lens and a large amount of mesoporous. Furthermore, Fig. 2c depicts HA and a large amount of mesoporous.
SEM of the scaffolds. a. Surface morphology of Li-HA scaffolds (×50). b Surface morphology of Li-HA scaffolds at higher power lens (×2000). c. Surface morphology of HA scaffolds (×2000).
Figure 3a shows the compression strength at different time points during the degradation experiment, we find that the compression strength can be improved when HA doped with lithium, and at 15th day after the experiment, the Li-HA scaffolds can achieve compression strength higher than 2 MPa, which was higher than pure HA scaffold (p < 0.05). When it came to 30th day, the compression strength in Li doped scaffold was 1.35 MPa. The pure HA scaffold had compression strength lower than 1 MPa at 30th day and was inferior to the Li-HA scaffold (p < 0.05). Figure 3b, c and d show the weight loss and cumulative ions release for Li-HA in Cells culture. During experimental sessions, weight loss of Li-HA is decreasing compared to HA. (Fig. 3b). In accordance with the Fig. 3c and d, the concentration of Li+ and phosphate anion in culture medium was higher in Li-HA than pure HA scaffold. Moreover, the release of lithium ions and phosphates continues, and the process is relatively stable. But the continuous Li release could activate Wnt signal pathway, which contributed to enhance osteogenesis in bone defect repairing.
Results of degradation and mechanical testing of the scaffolds in simulated body fluid (SBF). a. The maximum compression strength at different time points for Li-HA and HA. b,c,d. The weight loss and cumulative ions release for Li-HA and HA in Cells culture.
General condition of the animals
In the course of our operation, the tibia defects were well established and the scaffolds were precisely implanted (Fig. 4). Additionally, the muscle, nerve and blood vessel were well protected. There were 2 rabbits that were dead before killed them as schedule, both of which were dead for anesthesia when conducted X radiological examination, but the samples were obtained in time. Other complications, such as infection, incision unhealing, fracture, et al. were not observed.
Surgical procedures. a. Expose the tibia. b. Build bone defect using electric drill. c. Implant the scaffolds. d. Scaffolds were correctly placed in the bone defect.
At the sixth week, X-ray radiographs revealed callus had formed in Autologous bone(A), Li-HA(B) and HA(C) in the defect region with uneven density and remnant materials, while in Blank control(D), the bone defects remained obviously. At the twelfth week, the implanted materials were mostly absorbed with increasing density in the defect regions in Autologous bone(A), Li-HA(B) and HA(C), and apparently bone formation can be seen which had repaired most of the defect area. At the same time, there was a little callus formatted in Blank control(D) (Fig. 5a).
When conducted image analysis of the X ray films, the average gray values of the defect regions in each group increased over time, and the differences were significant (p < 0.05). At each time points, there were higher mean gray values in Autologous bone(A) and Li-HA(B) than HA(C) and Blank control(D) (p < 0.05). However, Autologous bone(A) and Li-HA(B) had a similar gray value and the difference had no statistical significance (p > 0.05). At the same time, HA(C) presented a better result at different time when compared with Blank control(D) (p < 0.05) (Fig. 5b). The average gray values of the defect regions was measured.). Radiologic observation showed excellent healing of the tibia defect in Autologous bone(A) and Li-HA(B), while HA(C) showed moderate healing and Blank control(D) showed minimal healing of the defect.
Figure 6a shows the Micro-CT outcomes of the defect repair in the scaffold transverse section. It demonstrated that large number of bone trabecula has formed in Li-HA(B) and HA(C), while the Blank control(D) was significantly showed less bone trabecular and presented least osteogenesis. However, in the Autologous bone(A), the bone defect completely healed. Figure 6b shows the quantitative results of the micro-CT and it demonstrated that Autologous bone(A) achieved best outcomes of bone volume and bone density (p < 0.05), while the Li-HA(B) took the second place and was better than the HA(C) (p < 0.05). The Blank control(D) had minimum results of these two parameters (p < 0.05).
a. X-ray radiographs of the defect area of rabbits (the arrows indicate the area of callus). b. Mean gray values (MGVs) in the four groups at 6 weeks and 12 weeks respectively. ∗means p < 0.05 between the two groups; # means p > 0.05 between the two groups.
Micro-CT outcomes each groups at 12 weeks. a. Micro-CT pictures showing the new bone trabecula in the bone defects. b. quantitative analysis of the Micro-CT pictures, ∗means p < 0.05 between the two groups # means p > 0.05 between the two groups.
Outcomes of histological detection HE and Masson staining were used to observe the new bone formation. As Fig. 7 presents, the implanted materials had begun to degrade and absorbed in Li-HA(B) and HA(C) at the sixth week, and the formation of new bone, fibroblast, osteoblast and newly formed trabeculae can also be seen. Li-HA(B) produced more newly formed trabeculae than HA(C). Autologous bone(A) a large number of fibroblast, osteoblast and new trabecular appeared. In Blank control(D), the defect regions were filled with fibrous tissue. When it came to the 12th week, the defects were almost completely healed in Autologous bone(A) with dense and normal morphology of trabeculae. In Li-HA(B), the implanted materials had been absorbed mostly and the new trabecular was abundant. However, In HA(C), the implanted materials also had been absorbed mostly, but the new trabecular was sparse and only small part of the specimens showed trabecular junction. In Blank control(D), new bone tissue can be seen on the edge of the defects, and the center was still filled with fibrous tissue.
HE staining and Masson showing the defects repairing. a. HE staining pictures. b. Masson staining.
When quantitatively analyzed the RNA level of the implants, the qPCR was performed in this section. We found that the expression of VEGF mRNA in Autologous bone(A) was significant higher compared with HA(C) and Blank control(D) (p < 0.05). But the difference between Autologous bone(A) and Li-HA(B) was not significant (p > 0.05). Furthermore, the difference between Li-HA(B) and HA(C) was also not significant (p > 0.05). At the same time, the expressions of ALP, collagen I, 𝛽-catenin, OCN and Runx-2 mRNA were higher in Autologous bone(A) and Li-HA(B) too (p < 0.05) but the two groups had a different outcome (p < 0.05). However, the expression of GSK-3𝛽 and PPAR-𝛾2 mRNA in Autologous bone(A) and Li-HA(B) was significant lower compared with HA(C) and Blank control(D) (p < 0.05) and the difference between Autologous bone(A) and B was also statistically significant (p < 0.05). The HA(C) was superior to the Blank control(D) in terms of the expression of ALP, collagen I, 𝛽-catenin, VEGF,OCN, Runx-2 mRNA. But with regard to the expression of GSK-3𝛽 and PPAR-𝛾2 mRNA, the HA(C) was inferior to the Blank control(D) (Fig. 8). According to this outcomes, we find lithium could activate the Wnt/𝛽-catenin signal pathway to increase the secretion the osteogenesis factors such as RUNX2, ALP, Collagen type one, et al., which contributed to enhance the osteogenic differentiation and reduce the adipogenic differentiation. However, the expression of VEGF mRNA in Li-HA(B) was not significantly increased compared with that of HA(C). This indicates that only used Li cannot enhance the HA’s angiogenesis potential.
qPCR analysis of the implant regions at 12 weeks after implantation. ∗means p < 0.05 between the two groups; # means p > 0.05 between the two groups.
In order to evaluate the osteogenic factors, and the Wnt signal pathway protein expression, we conducted immunohistochemistry evaluation. Figure 9a presents GSK-3𝛽 expression. The images demonstrate that the positive expressions of GSK-3𝛽 cells was most in Blank control(D), and the HA(C) also had some positive expression but the other two groups showed little positive cells. According to the Fig. 9b, c, the expression of protein in Autologous bone(A) and Li-HA(B) was lower than those in HA(C) and Blank control(D) (p < 0.05) and the differences in Autologous bone(A) and B were statistical significance (p < 0.05). Moreover, Blank control(D) presented a better result in GSK-3𝛽 protein when compared with HA(C) (p < 0.05).
Figure 10a shows that the positive expression of 𝛽-catenin cells was mostly found in Autologous bone(A), and the Li-HA(B) also had some positive expressions but the other two groups showed little positive cells. According to Fig. 10b and c, the expression of 𝛽-catenin protein in Autologous bone(A) and Li-HA(B) was larger than those in HA(C) and Blank control(D) (p < 0.05). The difference in Autologous bone(A) and Li-HA(B) was also statistical significant (p < 0.05). Simultaneously, with regard to the expression of 𝛽-catenin protein, the HA(C) was superior to the Blank control(D) (p < 0.05).
Figure 11a presents the OCN immunohistochemistry. The images indicated that the Autologous bone(A) had significant more positive cells than the other three groups. Moreover, the Li-HA(B) showed more positive expression and deeper stained cells than HA(C). According to Fig. 11b and c, the expression of OCN protein was higher in Autologous bone(A) and Li-HA(B) too (p < 0.05) but the two groups had a different outcome (p < 0.05). Simultaneously, with regard to the expression of OCN protein, the HA(C) was superior to the Blank control(D) (p < 0.05).
a. Immunohistochemistry analysis of GSK-3𝛽 at 12 weeks after implantation. The yellow color indicates the positive expression and darker color indicates stronger expression. b,c. Western blot analysis of the GSK-3𝛽 expression at 12 weeks after implantation. ∗means p < 0.05 when the two groups were in comparison.
a. Immunohistochemistry analysis of 𝛽-catenin at 12 weeks after implantation. The yellow color indicates the positive expression and darker color indicates stronger expression. b,c. Western blot analysis of the 𝛽-catenin expression at 12 weeks after implantation. ∗means p < 0.05 when the two groups were in comparison.
a. Immunohistochemistry analysis of OCN at 12 weeks after implantation. The yellow color indicates the positive expression and darker color indicates stronger expression. b,c. Western blot analysis of the OCN expression at 12 weeks after implantation. ∗means p < 0.05 when the two groups were in comparison.
Figure 12 shows the Col-I expression, which indicates that there was a large number of positive cells can be seen in Autologous bone(A) and Li-HA(B) at 12 weeks after surgery. Although the HA(C) was a little lesser on Col-I expression, it still showed more intensive cells than the Blank control(D). According to Fig. 12b and c, the expression of Col-I protein was higher in Autologous bone(A) and Li-HA(B) too (p < 0.05) but the two groups had a different outcome (p < 0.05). Moreover, HA(C) presented a better result in Col-I protein when compared with Blank control(D) (p < 0.05).
a. I Immunohistofluorescence staining for the Col-I expression at 12 weeks after implantation. Red color indicated Col-I. b,c. Western blot analysis of the Col-I expression at 12 weeks after implantation. ∗means p < 0.05 when the two groups were in comparison.
To quantitatively analyze the protein expression, western blot analysis was performed. According to Fig. 13, the expressions of ALP and Runx-2 protein in Autologous bone(A) and Li-HA(B) was larger than those in HA(C) and Blank control(D) (p < 0.05). The differences in Autologous bone(A) and Li-HA(B) were statistical significance (p < 0.05). Moreover, HA(C) presented a better result in ALP when compared with Blank control(D) (p < 0.05). Nonetheless, the two groups had a similar outcome (p > 0.05) in terms of Runx-2 protein expression. However, the expression of the PPAR-𝛾2 protein in Autologous bone(A) and Li-HA(B) was lower than those in HA(C) and Blank control(D) (p < 0.05) and the differences in Autologous bone(A) and Li-HA(B) were statistical significance (p < 0.05). In addition, Blank control(D) presented a better result in PPAR-𝛾2 protein when compared with HA(C) (p < 0.05). What’s more, the expression of VEGF protein in Autologous bone(A) was significant higher compared with HA(C) and Blank control(D) (p < 0.05). But the difference between Autologous bone(A) and Li-HA(B) was without significance (p > 0.05). Simultaneously, the difference between Li-HA(B) and HA(C) was also without significance (p > 0.05). But the HA(C) was superior to the Blank control(D) in terms of the expression of VEGF protein (p < 0.05).
a,b. Western blot analysis of the ALP, PPAR-gamma2, Runx-2 and VEGF expression at 12 weeks after implantation. ∗means p < 0.05 and # means p > 0.05 when the two groups were in comparison.
This study was conducted to examine the efficacy of Li-doped HA on repairing the bone defect and evaluate its biological properties, characteristics of degradation and osteogenetic activity. Our results indicated that blank control group had difficulty to completely repair bone defect by itself and it needed bone transplantation. According to the results, HA scaffolds improved new bone formation with Glycogen synthetase kinase-3𝛽 (GSK-3𝛽) decreased and 𝛽-catenin increased. After lithium doped on HA, bone regeneration was furtherly enhanced but the angiogenic effect was not improved significantly. Meanwhile, lithium-doped hydroxyapatite (Li-HA) decreased the activity of GSK-3𝛽 in Wnt signal pathway and increased the 𝛽-catenin expression. In short, Li doped into HA decreased GSK-3𝛽 while highly expressed 𝛽-catenin, which indicated the Wnt signal pathway was up-regulated. But when compared with HA scaffold, the Li-HA scaffold was inferior in terms of the efficacy of adipogenesis. Furthermore, when compared with autogenous bone implantation, the Li-HA scaffold was not inferior in terms of the efficacy of angiogenesis(VEGF), but our composite scaffold can’t cause donor site pain and infection. When the composite scaffolds been implanted into rabbit tibia bone defect, the implants were tightly closed to the peripheral bone tissue six weeks after operation, which indicated the bone defects began to be healed. Besides, in the micro-CT and histological observation, large amount of bone trabecula and new bone were generated respectively in Li-HA scaffold group at twelfth weeks.
Glycogen synthase kinase-3𝛽 (GSK-3𝛽) is a key target regulating the Wnt signal pathway and promoting gene expression. When GSK-3𝛽 is inhibited, Wnt signal pathway could be activated [50,51]. Therefore, regulating Wnt pathway play an important role in the treatment of bone defect. Furthermore, the Wnt signaling pathway has important functions in promoting the osteogenic differentiation of MSCs [52]. It has been reported that the canonical Wnt pathway provoke the progression of MSCs from osteoblastic precursor cells into more mature osteoblasts [53–56].
Scaffold materials have been used in various bone tissue engineering applications and have aroused extensive attention for ages. Autogenous bone, allogenic bone, artificial bone and heterogenous bone are the options, but which one is the optimization is always in controversy [38,39]. In our study, we chose HA, a principal inorganic constituent, constitutes 50–70% of human bone [40], has been extensively studied in bone repair biomaterials, to act as the scaffold. HA has excellent biocompatibility, bioactivity as well as osteoinductivity [40]. These characteristics made HA scaffold conducive to cell attachment, ingrowth and proliferation. Furthermore, HA directly bonds with live bone after implantation in cases of bone defects [41]. This feature enhances appropriate vascularization and stem cell proliferation, and guides bone regeneration without causing any local or systemic toxicity [42]. This also increases the use of HA in bone tissue engineering. So it’s reasonable for HA to perform well in our experiment. At the same time, we chose autogenous cancellous bone as the control group. Autogenous bone is an ideal bone grafting material containing bone marrow cells and with the ability of osteogenesis, which has been used to repair bone defects for ages [7–9,43] and proved to have excellent effectiveness.
There are many elements which can be added to the scaffold, for examples Sr, zinc, Mg, copper and Ag, and they improve its mechanical and biological properties for bone repairing materials applications is also commonly used [27–29]. Lithium has been regarded as Wnt signal pathway activator and reported to have effect in treating bone disease [30–32]. In fact, activating of Wnt pathway could also improve cell survival and promote osteogenic differentiation [19,30,33,34]. Lithium has proved to promote osteogenic differentiation of marrow mesenchymal stem cells (BMMSCs) by activating the Wnt/GSK-3𝛽 signal path [33,34]. Consequently, lithium may be a choice for bone defect treatment. In the previous studies, entangled titanium wire porous (ETP) [35], calcium phosphate cement (CPC) [36], bioactive glass (BAG) [37], were used as Li carrier and reported to have a good biological effect. In our research, HA was provided as the scaffold and was well composite with Li, which ensured the Li maintaining in effective doses on osteogenesis for a long time and also showed excellent biological effects.
The Li-HA scaffolds we fabricated were made up with particles with a large number of micropore and mesoporous through the scaffolds, which were beneficial to the cells’ attachment and ingrowth. Li-HA material had typical HA crystal form and doped proper content of Li did not lead to decomposition. In fact, the radius of Li is smaller than the calcium that in the crystal grain, which makes it easy for Li to reach the clearance position of the lattice and replace or occupy the corresponding ion site [25,26]. The compressive strength of our primary porous scaffolds were 3-4 Mpa, which met the strength requirements of cancellous bone (2–12 MPa) [17]. When Li doped into HA, the maximum compressive strength was 3.6 MPa. The possible explanations were metal ion Li reduced the porosity of the samples which resulted in more compact and sturdy structure, and Li doped into HA materials made the particles closer together.
According to our outcomes, Li-HA scaffold was much better than HA scaffold in terms of promoting bone formation, osteoconduction and degradability. In addition, our experimental group showed a higher expression of collagen I, 𝛽-catenin, ALP,OCN and Runx-2 mRNA, which helped to explain the better bone formation. VEGF is regarded as a key angiogenic factor that has the strongest and most significant biological activity in enhancing vascular formation [44], and it has also been reported to stimulate the osteogenic differentiation of osteoblasts [45]. However, the vascular formation regulated by VEGF can contribute to furnishing nutrition and excrete waste in the defect regions, so it also plays an important role in the growth and repairing of the bone. Nevertheless, compared to HA scaffold, Li-HA scaffold’s angiogenic factor was not increased. This indicates that only used Li cannot enhance the HA’s angiogenesis potential. At the same time, Li-HA scaffold was not inferior to autogenous bone in angiogenesis, but the latter was biologically active bone that possessed porous structure and perfect ingredient, and it owned a excellent biocompatibility. In addition, when compared with autogenous bone, we find Li-HA scaffold’s osteogenesis is inferior to autogenous bone that contained bone marrow cells, which have been confirmed by our outcomes of ALP, OCN, collagen I, 𝛽-catenin and Runx-2 mRNA expression. However, Our Li-HA scaffold has no several drawbacks, including the amount of available bone and pain at the donor site [1], and its structure and ingredient are different from the autogenous bone. Furthermore, our Li-HA scaffold was well shaped and completely located in the defects but the autogenous bone leaked out of the defects dispersedly. So after a series of processing procedures, our Li-HA scaffold obtained a good mechanical property (compressive strength and degradability) and can be shaped randomly. In summary, we prefer Li-HA scaffold.
Furthermore, the HA group showed better outcomes in terms of mean gray values, angiogenesis (VEGF) and new osteogenesis when compared with the blank control groups, which demonstrated that a scaffold implanted in the defect can benefit for ingrowth and crawling of the cells to the defect regions. The Li-HA scaffold and autogenous cancellous bone group were superior to the HA group, which indicated scaffolds loaded with some ion or stem cells were significant in bony tissue engineering.
Our study provides a promising bone scaffold to load with Li ion in the treatment of bone defect and tries to convert this composite material into a product and put into clinical application. However, our present works also has some limitations. First, our previous research [49] about the optimal Li content when doped into HA as well as their osteogenic effects in vitro and in vivo was not included in this paper. Second, the slowly- release effect of lithium salt in vivo cannot be controlled accurately. Third, we have just accomplished animal research, to promote this scaffold into clinic, more important human tails need to be developed. Fourth, the mechanical properties of the repaired bone defect were not detected in this experiment. Fifth, the addition of lithium has improved the degradation of scaffold materials, but its degradation performance can be further improved to reduce the effects of residual materials on bone defect and bone remodeling in future studies.
Conclusions
Li-doped HA is a promising bone grafting material which possesses excellent histocompatibility, degradability, osteoconduction and osteoinduction, and has capacity to promote the repair of bone defects. Moreover, it exerts an effect on up-regulated activate Wnt signal pathway, showed good osteogenesis and angiogenesis potential, and achieve an encouraging result in bone regeneration. It’s a very strong statement to recommend it for clinical application based on this study alone.
Footnotes
Conflict of interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
This study was supported by the National Natural Science Fund of China (grant nos. 81271976/H0605 and 81672165).