In vitro and in vivo studies on the biodegradable behavior and bone response of Mg₆₉Zn₂₇Ca₄ metal glass for treatment of bone defect

1. Introduction

Bone substitutes are used to stimulate the bone-healing when filling defects of bone tumor removal and several congenital diseases [1]. It is estimated that more than 500,000 bone grafting procedures occur annually in the USA and 2.2 million worldwide in orthopaedics, neurosurgery and dentistry [2].

The ideal bone grafts should have properties of osteoconduction, osteoinduction and osteogenesis and these properties are often used as a criterion to evaluate and classify the bone substitutes [3]. Although autografts are the best choice for bone defect regeneration, they still have some disadvantages such as donor supply limitation and donor site pain [4]. Nowadays, some bone substitutes can be obtained for orthopaedic diseases treatment, for example biomaterials (DBM: demineralised bone matrix), ceramics (TCP: tricalcium phosphate) and including some composite grafts (β-TCP/BMA composite). However, all of them cannot totally satisfy the clinical requirements due to their rather weak osteoinduction [3]. Recently, magnesium (Mg) and its alloys have been intensively studied as a new kind of biodegradable material owing to their characteristics of biodegradation and osteogenesis inductivity [[5], [6], [7], [8]]. Many studies have shown that Mg and its alloys have great potential in biodegradable fixation screws [9] and bioresorbable scaffolds [10]. But there are few studies about using Mg alloys as bone substitutes. Recently, our research group used Mg-1.5Sr alloy as bone substitutes to treat bone defect. We found that with the degradation of Mg-Sr alloy, a large amount of hydrogen gas was observed [11]. It is well known that the high degradation rate of Mg alloys will lead to adverse inflammation response, gas cysts and osteolysis, and hence delays bone healing [12]. Therefore, Mg-Sr alloy cannot meet the requirements very well. Exploring new Mg alloys with better corrosion resistance is necessary in the treatment of the bone defect. The corrosion resistance of the crystal Mg alloys is not easy to be improved due to the galvanic corrosion between the second phase and the matrix [13,14]. Compared with the crystal Mg alloys, amorphous Mg alloys usually have excellent degradation resistance and they also exhibit great potential in clinic applications, especially Mg-Zn-Ca metal glasses [15]. Gu et al. [16] studied two kinds of metal glasses, Mg₆₆Zn₃₀Ca₄ and Mg₇₀Zn₂₅Ca₅, respectively. They found that Mg-Zn-Ca metal glasses have lower degradation rate and better cell viability than pure Mg. Monfared et al. [17] found that Mg₇₀Zn₂₆Ca₄ metal glass can be a promising candidate for application in nerve tissue regeneration. Up to now, there are few reports on the use of Mg-Zn-Ca metal glass as bone substitutes to treat bone defect. In this study, Mg₆₉Zn₂₇Ca₄ metal glass (the critical diameter of glass forming ability of Mg₆₉Zn₂₇Ca₄ is 4 mm [18]) was proposed to be used as bone substitutes. In vitro biodegradable behavior and in vivo osteogenesis activity of a Mg₆₉Zn₂₇Ca₄ metal glass were investigated to verify the feasibility to treat bone defect. Pure Mg and β-TCP were set as the control group for the in vitro and in vivo experiments, respectively.

2. Materials and methods

2.1. Materials preparation

A mixture of pure Mg (>99.95%), pure Zn (>99.95%) and Mg-30Ca master alloy were melted in an induction furnace using argon as protective gas with a nominal composition of Mg₆₉Zn₂₇Ca₄. The furnace were heated with a power of 2 kW electrical source. When the alloy was fully melted and the composition was uniform, the alloy cooled to room temperature in the furnace. Then, the composition of the alloy was analyzed according to ICP6300/H01 standard. The above alloy was remelted in a quartz tube in the same induction furnace and then the remelted alloy was injected into a copper mold with a hole size of φ3 mm × 60 mm. The fast quenched ingot was investigated by using a scanning electron microscope (SEM, S-3400 N), X-ray diffraction (XRD, Bruker D8 ADVANCE) and differential scanning calorimetry (DSC, Netzsch DSC 204F1) at a heating rate of 20 K/min. The microstructure of high purity of Mg (>99.99%) which was supplied by Dongguan EONTEC. Co., Ltd. was observed by using an optical microscope (OLYMPUS-GX71).

2.2. Electrochemical test

Samples were molded in epoxy resin with one surface (0.07 cm²) exposed. The electrochemical test was performed in the 37 °C Hank’s solution (Table 1). Gamry Instruments (Reference 600) was employed for the test. A three-electrode cell was adopted for the electrochemical test. The reference electrode was a saturated calomel electrode (SCE), the counter electrode was platinum (size: 10 mm × 10 mm × 0.2 mm) and the sample was the working electrode. Open-circuit potential (OCP) tests were monitored for 30 min. Electrochemical impedance spectroscopy (EIS) tests were performed at the OCP in a frequency range of 100000-0.01 Hz. Potentiodynamic polarization measurements were carried out from -0.25 V with reference to OCP and the sweeping rate was 0.5 mV/s. In order to make the results convincible, each test was repeated for three times.

2.3. Immersion test

Samples with size of φ3 mm × 10 mm were immersed in the Hank’s solution for 14 days. According to ISO 10993-12 standard, the immersion ratio was set as 1.25 (cm² mL^-1). The solution was changed with new solution every day. In addition, the pH value of the solution was measured for 14 days period. After immersion, the samples were cleaned in a chromium trioxide solution to get rid of the degradation product and the degradation rate P_i (mm y^-1) was calculated according to the following equation [19]:

P_i=(K×W)/(A×T×D) (1)

where K is a constant coefficient with value of K = 8.76 × 10⁴, W is the samples mass loss with unit of g, A is the sample area with unit of cm², T is the immersion time with unit of h and D is the density of the material with unit of g cm^-3. Finally, the samples were examined by SEM. Each test was repeated for three times.

2.4. Cytotoxicity assessment

The operation of cytotoxicity assessment was similar to the previous work [20]. Murine calvarial preosteoblasts (MC3T3-E1) were adopted to evaluate the cytotoxicity. The cells were cultured in α-MEM, 10% fetal bovine serum (FBS), 100 U mL^-1 penicillin and 100 μg mL^-1 streptomycin in a humidified atmosphere of 5% CO₂ at 37 °C. The test was conducted using the alloy extract which was prepared at a ratio of the surface area of the alloy to extraction medium (α-MEM medium with 10% FBS) of 1.25 cm² mL^-1 in humidified 5% CO₂ atmosphere at 37 °C for 24 h. A 200-μL cell suspension was seeded onto 96-well plates at a density of 1 × 10⁴ cells mL^-1 and incubated for 24 h. The control group involved the use of α-MEM medium as the negative control. The medium was then replaced with 100 μL of extract. After 1, 3 and 5 days incubation, the samples were incubated with MTT for 4 h at 37 °C, and then the old medium was replaced by 150 μL dimethylsulfoxide (DMSO) in each well. The optical density (OD) values were measured by a microplate reader at 490 nm with a reference wavelength of 570 nm. The cell relative growth rate (RGR) was calculated according to the Eq. (2) [21]. In order to make the results convincible, each test was repeated for three times.

RGR = OD_test/ OD_negative ×100% (2)

2.5. Animal tests

For the animal tests, the surgery and treatment of rabbits were performed strictly according to the regulations and laws of the Standing Committee on ethics in China. The samples with φ3 mm × 3 mm cylinder were sterilized by 75% alcohol and ultraviolet-radiation for 30 min for each side. Since the in vitro studies showed that Mg₆₉Zn₂₇Ca₄ alloy had better corrosion resistance and biocompatibility, β-TCP which is commonly used in clinic as bone substitutes was chosen as the control group in the animal test. Eight adult New Zealand White rabbits, about 2.0 kg in weight, were used in this study. All rabbits were anesthetized with chloral hydrate (3 mL kg^-1) by intravenous injection for surgery. A bone defect model was made in the position of distal femur with a hand operated drill. The size of defects module was 6 mm in diameter and 9 mm in length. Four left legs of every rabbit were implanted with β-TCP and four left legs were implanted with Mg₆₉Zn₂₇Ca₄ metal glass. The rabbits received the subcutaneous injections of penicillin post operation to avoid a wound infection. After 2 months, X-ray was used to monitor the healing process. After the rabbits were killed by euthanasia, micro-CT (μ-CT) and histological sections were carried out for further analysis. Undecalcified thin ground sections were prepared through the central axis of the implant and perpendicular to the longitudinal axis of the femur shaft. They were cut into 30-40 μm thick films and stained with vangieson (VG) (methylene blue stained 10 min, picric acid poinsettia stained 15 min) for microscopy examination.

3. Results

3.1. Material characterization

Table 2 is the actual composition of the fabricated Mg₆₉Zn₂₇Ca₄ alloy. Fig. 1 presents the microstructures of Mg₆₉Zn₂₇Ca₄ metal glass and pure Mg. It can be seen that there was no second phases observed in Mg₆₉Zn₂₇Ca₄ metal glass and the microstructure was uniform. For the pure Mg, the grain size was not very uniform with average size of 4.91 ± 0.31 μm. Fig. 2(a) presents the XRD patterns of Mg₆₉Zn₂₇Ca₄ metal glass and pure Mg. There is no crystalline diffraction peak detected on Mg₆₉Zn₂₇Ca₄ alloy curve. However, pure Mg kept its crystal structure with strong diffraction peaks. Fig. 2(b) shows the corresponding DSC curve of Mg₆₉Zn₂₇Ca₄ alloy. Obvious exothermic peaks were detected at 430 K and 515 K. The glass transition temperature (T_g) and the first crystallization temperature (T_x1) of Mg₆₉Zn₂₇Ca₄ alloy were 414.2 K and 419.7 K, respectively. Therefore, it can be confirmed that the amorphous alloy was fabricated.

3.2. Electrochemical test

Fig. 3(a) shows the OCP curves of Mg₆₉Zn₂₇Ca₄ metal glass and pure Mg. The OCP of Mg₆₉Zn₂₇Ca₄ metal glass shifted toward the positive potential, and was higher than that of pure Mg, implying that the Mg₆₉Zn₂₇Ca₄ metal glass had less tendency to be corroded in Hank’s solution. The potentiodynamic polarization curves are illustrated in Fig. 3(b). Tafel fitting was performed by Gamry Echem Analyst software. Table 3 is the Tafel fitting result. The corrosion potential of Mg₆₉Zn₂₇Ca₄ metal glass was -1.30 V, which was about 0.4 V higher than that of pure Mg (-1.70 V). The corrosion current density i_corr of Mg₆₉Zn₂₇Ca₄ metal glass (0.44 μA cm^-2) was much lower than that of pure Mg (4.40 μA cm^-2). The corrosion rate (P_i, mm y^-1) was calculated by the following equation [22]:

P_i=22.85i_corr (3)

The corrosion rate of pure Mg was ten times faster than that of Mg₆₉Zn₂₇Ca₄ metal glass.

In general, hydrogen evolution occurs in the cathodic polarization process and dissolution of Mg occurs in the anodic polarization process [23]. For Mg and its alloys, the main corrosion product Mg(OH)₂ is not compact and it cannot well protect the matrix, meaning that there is no passivation process on the Mg and its alloys [24,25]. However, an obvious passivation process in the anodic polarization curves of Mg₆₉Zn₂₇Ca₄ metal glass at about -1.25 V was observed. This phenomenon might be related to the corrosion products on Mg₆₉Zn₂₇Ca₄ metal glass. The corrosion products which could act as a dense protective film to protect the matrix when the anode voltage was not large. Further increasing the anode voltage destroyed the film thereby accelerating the corrosion process. Fig. 3(c) presents the EIS curves of Mg₆₉Zn₂₇Ca₄ metal glass and pure Mg. It was found that the diameter of the capacitive loop of Mg₆₉Zn₂₇Ca₄ was much larger than that of pure Mg. The high frequency behavior of EIS is closely related to the corrosion rate [26]. The larger diameter at high frequency means the better corrosion resistance. So it could be concluded that the degradation resistance of Mg₆₉Zn₂₇Ca₄ metal glass was better than pure Mg.

The equivalent circuit models for Mg₆₉Zn₂₇Ca₄ metal glass and pure Mg are plotted in Fig. 3(d) and (e), respectively. The fitting result of the equivalent circuit is shown in Table 4. They were fitted with Gamry Echem Analyst software with the errors less than 10%. There are mainly two parts in the equivalent circuit of Mg₆₉Zn₂₇Ca₄ metal glass. R₁ and CPE₁ are used to describe the film resistance and the film capacity which are closely related to the degradation products. R₂ and CPE₂ (electric double layer) are the charge transfer resistance and the capacity between the substrate and the Hank’s solution, respectively. However, three parts can be observed in the equivalent circuit of pure Mg: the first two parts are similar to the Mg₆₉Zn₂₇Ca₄ metal glass, and the third part (R₃ and L) represents the hydrogen evolution on the metal surface. In general, the larger the film resistance R₁ is, the better the protective effect it has. The corrosion products on the Mg₆₉Zn₂₇Ca₄ metal glass exhibited good protective effect than those on the pure Mg due to the larger value of the film resistance of Mg₆₉Zn₂₇Ca₄ metal glass (2.52 × 10³ Ω cm²) compared to pure Mg (10.34 Ω cm²). Furthermore, the value of R₂ is closely related to the degradation resistance of the alloy, the larger value of the R₂, the higher resistance to the corrosion of the alloy. The R₂ value of Mg₆₉Zn₂₇Ca₄ metal glass (6.81 × 10³ Ω cm²) was much larger than that of pure Mg (0.98 × 10³ Ω cm²), and therefore, it was inferred that the corrosion resistance of Mg₆₉Zn₂₇Ca₄ glass was much higher than that of pure Mg.

3.3. Immersion test

Fig. 4 shows the pH change of Hank’s solution for 14 days immersion period. The pH of Mg₆₉Zn₂₇Ca₄ immersed solution was lower than that of pure Mg during the entire immersion period. After 14 days immersion, the degradation rates of Mg₆₉Zn₂₇Ca₄ and pure Mg were 0.08 mm y^-1 and 0.47 mm y^-1, respectively. The representative SEM micrographs and EDS analyses of Mg₆₉Zn₂₇Ca₄ and pure Mg after immersions for 14 days are shown in Fig. 5. From the cross section observation, it can be seen that the corrosion products layer on the Mg₆₉Zn₂₇Ca₄ metal glass was about 4.00 ± 1.77 μm which was much thinner than that of (16.90 ± 3.00 μm) on pure Mg (Fig. 5(a) and (b)). From the surface observation, it can be seen that a smooth and dense corrosion product layer was formed on the Mg₆₉Zn₂₇Ca₄ metal glass surface. The corrosion product acted as a protective film to inhibit the corrosion process (Fig. 5(c)). However, the surface of pure Mg with corrosion products was coarse and loose, which could not well protect the matrix (Fig. 5(d)). According to the EDS analysis, large amounts of Zn and Ca were observed in the corrosion products compared with the matrix, implying that the composition of the corrosion products are probably zinc phosphate and calcium phosphate (Fig. 5). There were only Mg, Ca and P elements detected on the pure Mg surface. In order to further investigate the composition of the corrosion products on the Mg₆₉Zn₂₇Ca₄ metal glass, XRD was carried out (Fig. 6). It can be seen that there were mainly Mg(OH)₂, Zn₃(PO₄)₂, Zn(OH)₂, Ca₃(PO₄)₂ and Ca(OH)₂ in the corrosion product layer of Mg₆₉Zn₂₇Ca₄ metal glass. After removal of the corrosion products, Mg₆₉Zn₂₇Ca₄ metal glass exhibited uniform corrosion without severe pitting corrosion (Fig. 7(a)). Besides, with the degradation of Mg₆₉Zn₂₇Ca₄ metal glass, many uniform microspores formed on the metal surface which maybe is beneficial to the cells adhesion leading to the good growth of the cells. However, the pure Mg was severely corroded (Fig. 7(b)).

3.4. Cytocompatibility

In order to investigate the cytotoxicity of the experimental materials, cell proliferation was evaluated by MTT test. Fig. 8 presents the cell viability after incubation with material extracts for 1, 3 and 5 days. In the first day of incubation, the cells in both Mg₆₉Zn₂₇Ca₄ metal glass extract and pure Mg extract exhibited lower viability. However, with increase of incubation time to 3 days, the cells viability increased obviously. The cells in Mg₆₉Zn₂₇Ca₄ extract showed higher viability than those in pure Mg extract, with RGR of 94 and 84, respectively. After 5 days incubation, RGR values in Mg₆₉Zn₂₇Ca₄ extract and pure Mg extract were 90% and 81%, respectively, indicating that the cells in the two extracts exhibited good viability (according to ISO10993-5 standard, when the RGR value > 80%, the material can be regarded as biocompatible). Moreover, the cells in Mg₆₉Zn₂₇Ca₄ extract still showed better cell proliferation than that in pure Mg extract.

3.5. Histopathology evaluation

According to the above studies, Mg₆₉Zn₂₇Ca₄ metal glass exhibited better corrosion resistance and biocompatibility than pure Mg. β-TCP has good biocompatibility and osteoinductivity and it was commonly used in bone tissue engineering [27]. Therefore, in the animal test, β-TCP was chosen as the control group to further study the osteogenesis activity of Mg₆₉Zn₂₇Ca₄ metal glass. Both implants of Mg₆₉Zn₂₇Ca₄ glass and β-TCP were well tolerated by the rabbits during 2 months implantation. The biological response to the implants analyzed by X-ray and μ-CT is illustrated in Fig. 9. It can be observed that Mg₆₉Zn₂₇Ca₄ metal glass were still clear after 2 months implantation and no bone resorption was observed. New bone formation showed close contact with the implant at the cortical and medullary cavity site in metal glass group (Fig. 9(a)). However, there was a visible cave in β-TCP group and no obvious connection was observed between the implant and the tissue (Fig. 9(d)). From μ-CT observation, no obvious hydrogen evolution was observed around metal glass. Mg₆₉Zn₂₇Ca₄ metal glass maintain its original morphology very well (Fig. 9(b)). However, for the control group (β-TCP) part of the defect bone caved in and some β-TCP leaked out from the wound (Fig. 9(e)). After three-dimensional reconstruction, the size of the bone defects as illustrated in Fig. 9(c) was only φ3.35 mm hole unhealed in the group of Mg₆₉Zn₂₇Ca₄ glasses while the control group had φ4.14 mm hole unhealed. Histological sections are illustrated in Fig. 10. The interfaces between the Mg₆₉Zn₂₇Ca₄ metal glass implant and the tissue were observed in Fig. 10(a) and (b). There was new bone formation around the Mg₆₉Zn₂₇Ca₄ glasses and histological analysis didnot show obvious adverse tissue reactions around the implants (Fig. 10(a)). With progressing implantation time, the cancellous bone tightly attached to the metal glasses indicating the glasses have a good osteoconductivity (Fig. 10(b)). For the control group, only some new cartilages were formed but there was no mature bone observed during the 2 months postoperation (Fig. 10(c)).

4. Discussion

4.1. Corrosion behaviors of Mg₆₉Zn₂₇Ca₄ metal glass

Mg and its alloys exhibited great potential to be used as biomaterials. However, the high degradation in vivo limits wide clinical applications of Mg alloys [28]. Mg and its alloys are usually corroded by pitting corrosion due to galvanic effect [29]. The Mg-based bulk metal glass possesses good corrosion resistance due to its single-phase and chemically homogeneous alloy system [30].

The Mg₆₉Zn₂₇Ca₄ metal glass showed better corrosion resistance compared with pure Mg. On one hand, this metal glass has single-phase and there is neither second phase and nor grain boundary. Therefore, the galvanic effect was avoided. On the other hand, the dense corrosion products acting as a protective layer can protect the matrix well. According to the XRD analysis (Fig. 6) it can be inferred that the main composition of the protective layer were Mg(OH)₂, Zn₃(PO₄)₂, Zn(OH)₂, Ca₃(PO₄)₂ and Ca(OH)₂. The corresponding mechanism is illustrated in Fig. 11. In the beginning of corrosion process, the substrate was dissolved in Hank’s solution due to the reaction with H₂O (Fig. 11(a)). The second stage was that the metal ions M²⁺ (Zn²⁺, Mg²⁺ and Ca²⁺) reacted with phosphate and OH^-, Mg(OH)₂, Zn₃(PO₄)₂, Zn(OH)₂, Ca₃(PO₄)₂ and Ca(OH)₂ were formed on the metal surface (Fig. 11(b)). The degradation products deposited on the Mg₆₉Zn₂₇Ca₄ metal surface could protect the substrate from further corrosion (Fig. 11(c)). Therefore the alloy showed better corrosion resistance.

However, compared with all the engineering metals, Mg has the lowest standard potential (-2.37 V) [31]. Mg is susceptible to be corroded in Hank’s solution. Although pure Mg showed better corrosion resistance than ZK60 and AZ31 alloys [32,33], the corrosion rate of pure Mg is still 6 times faster than that of Mg₆₉Zn₂₇Ca₄ from the immersion test.

4.2. Cytocompatibility of Mg₆₉Zn₂₇Ca₄ metal glass

The MTT tests showed that both Mg₆₉Zn₂₇Ca₄ metal glass and pure Mg had good biocompatibility. When the culture time were 3 days and 5 days, all the RGRs were above 80%. However, the cells in the Mg₆₉Zn₂₇Ca₄ metal glass extract exhibited better viability as the incubation time prolonged. After 3 days, the RGR in Mg₆₉Zn₂₇Ca₄ metal glass extract reached more than 90%. The higher cells viability in Mg₆₉Zn₂₇Ca₄ metal glass extract than that in pure Mg extract can be explained as follows. One aspect was that there were more nutritious elements in Mg₆₉Zn₂₇Ca₄ metal glass extract compared with the pure Mg extract. It has been proved that Mg²⁺, Zn²⁺ and Ca²⁺ have many physiological functions. Mg plays many roles in the nerve system, muscle function, and so on. The interaction between the tissue and the Mg alloy can stimulate osteogenesis during its degradation [34]. Zinc (Zn) is an essential component of more than 300 various enzymes and it is important for the cellular metabolism [35,36]. Calcium (Ca) can also enhance some enzymes activities in the body, participating the activities of nerve and muscle and regulating hormone secretion. Calcium also plays roles in blood coagulation and cell adhesion [37]. Therefore, a combination of Mg²⁺, Zn²⁺ and Ca²⁺ in Mg₆₉Zn₂₇Ca₄ metal glass extract played an important role in the cell proliferation. Another aspect was the relative lower pH value of Mg₆₉Zn₂₇Ca₄ metal glass extract which was beneficial to the cell growth. For the pure Mg extract, the relative strong alkaline environment was harmful to the cell growth.

4.3. Histopathology evaluation of Mg₆₉Zn₂₇Ca₄ metal glass

In this study Mg₆₉Zn₂₇Ca₄ metal glass presented a low degradation rate without observable hydrogen evolution and not any osteolysis was found during the implants degradation. It was much better than other Mg alloys, such as MgYReZr alloy [38] and Mg-Sr alloy [11]. In addition, a cancellous bone was tightly attached to the metal glass implants after 2 months implantation, however only some new cartilages formed around the β-TCP implants, which is consistent with some researches where the loose fibrous tissue was observed in β-TCP group for 2 months implantation [39]. This indicates that the osteogenesis ability of Mg₆₉Zn₂₇Ca₄ metal glass is better than that of β-TCP in the early 2 months implantation. It is deduced that because the low degradation rate of Mg₆₉Zn₂₇Ca₄ metal glass, the gas cavity formation and osteolysis led by high degradation rate of Mg alloys are prevented. In addition, the nutritious Mg²⁺, Ca²⁺ and Zn²⁺ as the main degradation product of Mg₆₉Zn₂₇Ca₄ metal glass, may exert its high osteogenesis activity to improve the bone defect healing. Moreover, compared with the weak mechanical properties of β-TCP, Mg-Zn-Ca metal glasses usually have high compression strength [40]. During the degradation, Mg₆₉Zn₂₇Ca₄ metal glass can keep their mechanical integrity for a long time to enable them to play a supporting role in defect healing process. So in this study, the collapse of the implant in the defect was observed in β-TCP group, but this was not the case in the metal glass group.

5. Conclusions

In this work, in vitro and in vivo investigations on the biodegradable Mg₆₉Zn₂₇Ca₄ metal glass used as bone substitutes for the treatment of bone defect has been investigated. Some conclusions can be summarized as follows:

1) Mg₆₉Zn₂₇Ca₄ metal glass showed excellent corrosion resistance. The degradation rate of pure Mg was about 6 times higher than that of Mg₆₉Zn₂₇Ca₄ metal glass.

2) The cell cytotoxicity test showed that Mg₆₉Zn₂₇Ca₄ metal glass exhibited better cell viability than pure Mg.

3) During the degradation of Mg₆₉Zn₂₇Ca₄ glass, new bone accumulated along the implants without observable hydrogen evolution. Mg₆₉Zn₂₇Ca₄ glass exhibits much better healing effect than β-TCP. Mg₆₉Zn₂₇Ca₄ glass has potential to be used as bone substitutes due to its good corrosion resistance, biocompatibility and high osteogenesis activity.

Acknowledgements

This work was financially supported by the Key Program of China on Biomedical Materials Research and Tissue and Organ Replacement (Nos. 2016YFC1101804 and 2016YFC1100604) and the Shenyang Key Research & Development and Technology Transfer Program (No. Z18-0-027).

---