A tough injectable self‐setting cement‐based hydrogel for noninvasive bone augmentation

Composite hydrogels with excellent properties can open new opportunities to terminate the need for auto/allografts in bone augmentations. However, their clinical application has been limited by their insufficient mechanical strength and lack of osteoinductivity. Here we report a new strategy to design an injectable bioactive double network hydrogel reinforced by inorganic calcium/magnesium phosphate cement (CMPC) hydrates to meet the mechanical performance requirements for bone regeneration. The engineered CMPC hydration endows the composite hydrogel with an appropriate gelation time and temperature for injection, which shows no harm in the defect site. CMPC hydrates could also provide a lower swelling ratio and higher biodegradation rate fitting the in vivo bone regeneration needs. In vitro and in vivo experiments prove that the ions released from inorganic particles endow biocompatibility, cell migration, adhesion, differentiation, and significantly higher bone regeneration capacity. Taken together, the simple addition of CMPC particles imparts in‐demand features that bring us closer to the clinical utilization of hydrogel‐based materials for bone regeneration.

Bone fractures are an unfortunate fact of our life.[3] The current "gold standard" for therapy of these nonhealing defects is using bone autografts or allografts. [1]owever, these approaches are suboptimal and have their own limitations, including donor site morbidity, lack of sufficient graft, immunologic rejection of the tissue, risk of disease transmission, and substantial postoperative complications. [4]Fortunately, the remarkable progress in the development of tissue engineering makes it an effective strategy for overcoming these obstacles. [5,6]Among the scaffolds for bone tissue engineering applications, injectable hydrogels have gained immense attention clinically due to their ability to form any desired shape in situ to match irregular bone defects with minimally invasive surgical procedures. [2]Nevertheless, many injectable systems typically lack mechanical properties (stiffness and toughness) and are preferably for regenerating a minor and nonloadbearing defect. [7]Moreover, ensuring that injectable hydrogels retain their structural integrity once introduced into the bone defect site is challenging. [8]To tackle these scientific issues, it is essential to engineer mechanically tough and strong injectable hydrogel systems with high bioactivity and convenient operability to accelerate the repair of load-bearing bone defects. [9,10]norganic bone cement incorporated within the polymeric network has become popular to generate functional injectable scaffolds in clinical treatment with the ability to enhance the mechanical properties to meet the requirements mentioned above while promoting bone regeneration simultaneously. [11]Generally speaking, bone cement is based on mixing a powder phase and a liquid phase at room temperature to interlock each other to set and have the ability to harden once implanted within the body.Currently, the most commonly used injectable bone cement in orthopedic surgery is based on using bioinert acrylic bone cement, such as poly(methyl methacrylate) (PMMA) or bioactive calcium phosphate bone cement (CPC). [12,13]lthough bioinert acrylic bone cement may provide sufficient mechanical support for the moderate loadbearing bone-damaged site, they have some drawbacks, including the heat liberated during the setting stage can be led to bone necrosis and tissue damage, and the generation of particulate wear debris from the implantation of acrylic bone cement is also capable of inducing osteolysis. [14]Hence, CPC has been more attention due to its bioactivity and excellent biocompatibility.Calcium and phosphate ions released by CPC are also beneficial to forming bone-like apatite layers and trigger the secretion of extracellular matrix (ECM) with a high hyaluronic acid content, which is essential for growth factors like bone morphogenetic protein-2 (BMP-2) and can improve osteoinductivity. [15]owever, they face various limitations, such as poor initial mechanical properties, a long setting time (30-60 min), and a deficient ability to form an interconnected porous structure in situ to facilitate cell attachment, ingrowth, and subsequent osteogenic differentiation and vascular remodeling. [16]o address some of these issues, a fast-setting magnesium phosphate cement (MPC) with higher initial strength with an increased degradation rate has been introduced.The magnesium ions released from MPC are also advantageous to facilitate vascularization through regulating macrophage polarization to generate a balanced M1/M2 or predominantly M2 phenotype, which can further improve osteogenic potential. [15,17]Moreover, potassium ions from MPC can improve calcium balance, reduce markers of bone resorption and increase markers of bone formation. [18]However, the clinical use of MPC remains challenging due to the exothermic process of its intense acid-base reaction between magnesium oxide and phosphate, which must be strictly controlled to avoid tissue necrosis. [19]To make it practically feasible in clinical settings, the combination of CPC and MPC to generate calcium/magnesium phosphate cement (CMPC) has been suggested to regulate heat release better and provide adequate initial mechanical properties. [19,20]owever, such bone cement combination is also plagued by lacking interconnected macropores for cell migration and nutrient penetration, leading to poor osteoconductivity.Although particle leaching procedures with sodium chloride [21] and other quickly degrading polymers like poly (l-lactide) [22] and carboxymethyl chitosan [23] can induce the formation of porous structure within the cement, they normally require several steps of predegradation treatment, which are too complicated for direct injection and in situ formation.Hence, incorporating such bone cement within polymeric hydrogels is a promising strategy to overcome these challenges.
Herein, inspired by the natural bone structure and composition, we have designed and developed an injectable and biodegradable tough hydrogel composite with in situ generation capability with excellent cell attachment performance for bone repair based on incorporating CMPC particles into hydrogel networks (Figure 1).Our tough double-network hydrogel matrix has been made based on sodium alginate, oxidized sodium alginate (OSA), and poly (acrylamide) (PAM).So that, the calcium/magnesium ions released from the dissolution of CMPC particles could ionically crosslink arginine-glycine-aspartic acid (RGD)grafted OSA to form the first hydrogel network within seconds, therefore avoiding the injected hydrogel precursor from leaking out of the target site under the flow of body fluid.The formation of a second PAM hydrogel network is a free-radical polymerization process that could be accelerated by absorbing the heat released from the hydration reaction of the CMPC particles.Such a design provides inherent triggering factors for hydrogel formation without external stimulations.Moreover, the temperature rise and the gelling time can be well controlled by adjusting the mass ratio of CMPC particles to acrylamide (AM) monomers.The hydration products of CMPC are also exploited to increase the interface roughness due to the formation of a gel/bone hybrid layer at the interface to improve osteoblast adhesion.It therefore holds great promise as a new class of injectable bioactive hydrogel system in clinical practice with the capacity to mend defects within load-bearing bones.

| Gelation process and compositional properties
The gelation of this composite double-network hydrogel is started based on the in situ crosslinking of adjacent guluronate blocks available on SA/OSA polymer chains as a result of the fast dissociation of CaHPO 4 and release of divalent calcium ions. [10,24,25]This is confirmed by the quick rise in the rheological properties of CMPC-reinforced composite hydrogel (denoted as GEL-CMPC afterwards) in comparison to the one without CMPC (OSA/PAM, denoted as GEL afterwards).Supporting Information: Figure S1 shows that the hydrogel with 30% CMPC (denoted as GEL-30%CMPC afterwards) reached a shear stress of 270 Pa only after about 10 min while the GEL sample had a shear stress of 0.79 Pa after the same duration.The primary gelation and rapid increase in hydrogel viscosity improve the hydrogel retention upon injection and minimize the material dispersion to the surrounding tissues and potential embolization. [26]Nevertheless, the GEL-CMPC pregel was still extrudable and did not clog the syringe nozzle and hose (Supporting Information: Figure S2).Another factor in offering such a high rate of strength development was the heat released from the CMPC hydration.It is shown that the GEL-30%CMPC could precede the peak of released heat from 26.8 min to less than 10 min (Figure 2A).This released heat increased the pregel temperature and accelerated the endothermic APS dissociation, free radical formation, and AM polymerization reaction-preceding the formation of the second hydrogel network (PAM). [27]According to some previous research, this 10-min window has been proven sufficient to thoroughly mix the hydrogel precursor and inject it into the defect in clinical application. [28]he rise of shear stress in the GEL-CMPC pregel can also be attributed to the growth and entanglement of forming CMPC crystals. [12]In this step, the formation of polymer chains helped the cohesion between the crosslinking hydrogel fibers and forming cementitious particles. [29]SEM images of the hydrogel network with and without CMPC particles are shown in Figure 2B,C.They both possess the interconnected porous structure, while the surface of GEL-CMPC hydrogel is much rougher than the GEL hydrogel due to the attachment of inorganic particles, which are spindle-shaped with a length ranging from 0.1 to 5.0 μm (Figure 2D).X-ray diffraction (XRD) spectrums confirm that those inorganic particles are the hydration products of CMPC, which are mainly potassium magnesium phosphate (K-struvite, KMgPO 4 •6H 2 O) and calcium hydrogen phosphate hydrate (Ca 1.5 HP 2 O 7 •2H 2 O) (Figure 2D-F).K-struvite is known as a favorable alternative to calcium phosphate cement in bone regeneration in terms of osteoinductive properties, sufficient solubility and fast degradation, mechanical properties, and setting time. [30]The interactions between the hydrates of CMPC particles and the hydrogel networks are investigated by Fourier transform infrared spectroscope (FT-IR) spectroscopy, the strengthened bands at 1600-1700 and 1400-1500 cm −1 are assigned to the asymmetric and symmetric stretching vibrations of carbonyl groups, indicating the formation of metalcoordination bonds [31] (Figure 2G).It has been reported that the metal-coordination bonds contributed to improving the mechanical properties by serving as sacrificial bonds for higher energy dissipation. [16,32]o meet the clinical requirements for the in situ formation of the scaffold, the temperature rise should be controlled to around 42°C to reduce the damage to surrounding tissues. [33]Therefore, both in vivo and in vitro experiments were conducted to monitor the temperature change during the gelation process.As illustrated in Supporting Information: Figure S3, the formation process of GEL hydrogel from a 3.5 mL precursor solution could cause a maximum temperature rise to about 47.0°C, which is harmful to the surrounding cells if applied in vivo.Notably, adding CMPC particles could reduce the maximum temperature to 42.5°C by reducing the polymer content in the sample.The in vivo temperature detected via the infrared thermal imager also demonstrated that heat release and temperature rise of the composite hydrogel are controllable and suitable for injection (Supporting Information: Figure S4).

| Microstructure and mechanical properties
Conventional PAM-based hydrogels are relatively soft with low mechanical strength to be used for bone augmentation.On the other hand, CPCs have been determined to require polymeric or inorganic additives for cohesiveness and injectability. [12]Therefore, in this study, an interpenetrating double-crosslinked polymeric network incorporating CMPC particles is introduced to enhance the compressive strength and the toughness of the injectable bone regeneration hydrogel.The typical compressive stress-strain curves of the hydrogels are show in Figure 3A, and their compressive moduli are summarized in Supporting Information: Table S1 and Figure S5.The maximum compressive strength, compressive modulus, and strain at failure for pure PAM hydrogel were determined to be 0.013 MPa, 0.204 MPa, and 41.2%, respectively.The incorporation of the OSA as the second network slightly increased the compressive strength and modulus to 0.164 and 0.410 MPa, respectively.This is while having 30% CMPC in the hydrogel network significantly improved the compressive strength, compressive modulus, and strain at failure to 6.223 MPa, 1.515 MPa, and 83.3%, respectively.This increase in the mechanical properties of an injectable hydrogel by simply adding CMPC particle during crosslinking was found extraordinary as it is a couple of times higher than what has been reported so far for some injectable hydrogels reinforced by calcium/ magnesium phosphate particles. [16]As FTIR results show, the metal-coordination interactions between the hydrogel network and the CMPC particles can serve as sacrificial bonds for higher energy dissipation that leads to superior mechanical performance. [16,32]However, when CMPC content increased to 50%, the aggregation of CMPC particles and their hydrates was more serious and caused uneven stress distribution through the composite network, exacerbating the matrix brittleness and reducing the toughness of the hybrid hydrogel (Figure 3A and Supporting Information: Figure S5).
The intrinsic hydrophilic nature of hydrogel materials endows them with the ability to absorb and retain a large amount of water, which is favorable for cell proliferation and migration, cell-cell interaction, and the exchange of nutrients and metabolites. [34]However, excessive swelling is likely to cause large deformation after in situ gelation, decrease the mechanical stability and strength of scaffolds, and further affect the binding between the hydrogel composites and the host bone. [35]Therefore, additives that can control the excessive swelling of a scaffold are in great demand for developing injectable biomaterials.Incorporating CMPC particles bestowed regulation of the swelling ratio in the ultimate hydrogel products.As shown in Figure 3B, GEL hydrogel needed nearly 45 h to reach the equilibrium-swollen state, while the GEL-CMPC hydrogels only required 10 h.In contrast to the GEL sample, which showed a swelling ratio of more than 700%, moderate content of simulated body fluid (SBF) solution absorbed and retained in GEL-30% CMPC hydrogel (≈2.5 g/g), which is similar to that reported in previous literature. [36]In the meantime, the equilibrium water content (EWC) was reduced from 88.2% to 71.4% by having 30% CMPC in the hydrogel (Supporting Information: Table S1).The reduced swelling ratio and EWC can be attributed to the stronger bonding introduced by CMPC particles, and a lower polymeric content in GEL-CMPC samples that participates mainly in water absorption when compared to CMPC particles.This observation can also be related to the fact that the GEL-CMPC hydrogels seem less porous since some pores were filled and clogged with formed CMPC hydrates.This is obvious in SEM micrographs of GEL and GEL-30% CMPC hydrogels (Figure 2B,C).
The pore architecture of the scaffolds devised for bone regeneration is also of vital importance to cell proliferation and migration, vascularization, spatial cell organization, as well as nutrient/waste exchange.Therefore, we analyzed the pore size distribution and the total porosity of the hydrogel scaffolds by the mercury intrusion porosimeter (MIP) method (Figure 3C).The porosity of the GEL-30% CMPC immersed in SBF solution for 7 days was about 36.4% and the dominant pore range was from 20 to 60 μm, which is desired for fiber-forming and bone generation. [37,38]In line with this, large amounts of pores with a diameter of 40-80 μm within GEL-30% CMPC hydrogel were observed in SEM images (Figure 2C) and 3D reconstruction images of nano-computed tomography (nano-CT) (Figure 3D).Besides, micropores ranging from 1 to 2 μm are also essential to increase the specific surface area and adsorb more BMP, facilitating the dissolution of minerals from CMPC particles and their reprecipitation. [37,38]When increasing the CMPC particles to 50 wt%, most of the pores were reduced in size to about 0.2 μm, and the total porosity was reduced to 20.4% (Figure 3C), which is consistent with the results of swelling properties.This is because the high amount of CMPC particles agglomerated and clogged the empty places between polymeric chains.
An ideal biomaterial in tissue engineering applications should be fully biodegradable but at a rate synced with the generation of neotissues.As the biodegradation results are shown in Supporting Information: Figure S6, the GEL samples were degraded faster than PAM hydrogel because of the low stability of alginate chains in SBF, where the calcium ions exchange with sodium ions. [25]The biodegradation in GEL-30% CMPC samples was higher than in GEL or PAM samples, which is because of the higher degradation of CMPC particles compared to polymer chains.The fairly quick ions release through time from the GEL-30% CMPC hydrogel as an indication of CMPC biodegradation is also illustrated in Figure 3E.Albeit, this sharp phase in weight loss continued until around Day 6, after which the ions released from the GEL-30% CMPC hydrogel and the ions present in the SBF started to form ion complexes and then precipitate as amorphous calcium phosphates (ACPs) and hydroxyapatite (Ca 5 (PO 4 ) 3 (OH), HAp) as the dominant mineral phase of human bone [39] (Figure 2F).The decrease in Ca 2+ concentration after a long-time incubation also supports the finding of mineral precipitation.These inorganic particles precipitated on the hydrogel network enhance scaffold biocompatibility and osteoinductivity.Moreover, K-struvite degradation provided a continuous release and presence of Mg 2+ ions in the extracellular fluid, which could be beneficial for the regulation of osteogenic differentiation and new bone formation. [30]The Mg 2+ ions are usually distributed along the Haversian canals where it has a key role in HAp crystals formation. [40]The excess Mg 2+ ions are shown to be excreted very efficiently by kidney. [40]he SEM micrographs of the GEL-30% CMPC incubated in SBF for 28 days confirmed the attachment of the generated inorganic particles to the hydrogel network (Figure 3F).Comparing these micrographs with the fresh ones (Figure 2C) revealed changes in the shape and size of composite pores and showed the reprecipitation of inorganic particles, in agreement with the XRD data.The pores distributed in the exterior part of the composite hydrogel that were in direct contact with the SBF solution showed more swelling.This might be attributed to the more facile degradation of CMPC and, therefore, faster water absorption and swelling.The MIP analysis (Figure 3C) revealed that after this incubation period, the total porosity increased to 42.5%, and the dominant pore range enlarged to 40-200 μm, which is beneficial for cell migration and angiogenesis. [41]3 | In vitro cytocompatibility and osteogenic differentiation

| Cytocompatibility
The safety of the GEL-30% CMPC hydrogels was assessed by cell counting kit-8 (CCK-8) assay using L929 and MC3T3-E1 cell lines.Superior to PAM and GEL hydrogels, the composite hydrogel showed no toxicity for L929 even after 7 days (Figure 4A).At this time point, we observed the cell viability of more than 90% for GEL-30% CMPC, while it was less than 70% and 80% for PAM and GEL hydrogels, respectively.This high level of cell survival can be related to the continuous release of Mg 2+ ions which play a key role in the protein adhesion and further cell attachment. [42,43]The higher can be attributed to the cell anchorage points provided by the RGD motif available on RGD-OSA chains. [44,45]A similar assessment using MC3T3-E1 cells showed near 100% cell viability for GEL-30% CMPC hydrogels (Supporting Information Figure S7).Confirming these observations, live/dead staining assay showed improved cell proliferation in GEL-30% CMPC hydrogel (Figure 4B).Compared to PAM hydrogels, few dead cells were observed in GEL and GEL-30% CMPC hydrogels as the RGD peptide grafted on the OSA chains could promote cell proliferation and further improve hydrogels cytocompatibility. [46]otably, CMPC dissolution and reprecipitation of new inorganic phases could directly consolidate toxic ammonium ions released during the gelation process of PAM by forming struvite (NH 4 MgPO 4 •6H 2 O) (Figure 2F).

| Osteogenic differentiation
Alkaline phosphatase (ALP) is an exoenzyme present on the outer surface of osteoblasts membrane and known as an early marker of osteogenic differentiation of L929 and MC3T3-E1 cells. [8,47]Both GEL and GEL-30% CMPC hydrogels increased the ALP expression in L929 cells, while the latter showed stronger ALP upregulation (Figure 4C).The RGD peptide decorated on OSA chains can be accounted for higher ALP level in GEL or GEL-CMPC samples compared to PAM. [48] As for MC3T3-E1 cells, there was no significant difference in the ALP expression level between GEL and GEL-30% CMPC samples in the first 7 days (Supporting Information: Figure S8), but on Day 14, the ALP expression level of MC3T3-E1 cells cocultured with GEL-30% CMPC hydrogel was found to be 33% higher than that of GEL hydrogel.
Next, bone mineralization as an important hallmark for late-stage osteogenic differentiation was assessed using Alizarin Red S (ARS) staining.As demonstrated in Supporting Information: Figure S9, the ARS was obviously more bound to the ECM of MC3T3-E1 cells cocultured with GEL-30% CMPC hydrogel, proving its befitting effect on calcium deposition.Overall, the ALP and ARS results demonstrated that GEL-30% CMPC hydrogel could facilitate the differentiation of cells, and this is because of the RGD motifs of the hydrogel and its releasing bioactive ions, especially Mg 2+ . [49]o further investigate the osteogenic potential of the developed hydrogel, the expression of six representative osteogenic marker genes related to the human bone derived mesenchymal stem cells (hBMSCs) was evaluated by quantitative real-time polymerase chain reaction (qRT-PCR) analysis (Figure 4D and Supporting Information: S10).The results indicated that the mRNA expression of BMP-2, integrin subunit α3 (ITGA 3), twist-related protein 1 (Twist 1), and vascular endothelial growth factor (VEGF) in groups of GEL hydrogels were all upregulated with the control group.BMP-2, Twist 1, and VEGF are important markers of osteoblastic differentiation and angiogenesis, while ITGA 3 is a significant receptor for osteoblast to bind to the osteoinductive protein of NELL-1 in ECM. [50,51]Published reports showed that MSCs exposed to NELL-1 differentiated to osteoblasts with greater bone regeneration potential.It has also been shown that its administration not only has a similar effect on the MSCs in the body but also reduces bone resorption by osteoclasts.RGD functional groups available on GEL and GEL-30% CMPC hydrogel are known to improve the affinity between cells and biomaterials [52] and might be effective in binding osteoblasts to the matrix and, therefore, NELL-1 protein.Furthermore, the gradual generation of HAp in the GEL-30% CMPC hydrogel could create a rougher surface, therefore further enhancing the adhesion and proliferation of osteoblasts. [53]Mg 2+ ions have also been reported with a positive effect on ECM protein deposition and bone mineralization. [51]The mRNA expression levels of BMP-2, BMP-3b, Osterix, and ITGA 3 were significantly higher in GEL-30% CMPC hydrogel compared with the GEL group.BMP-3b is advantageous to the recruitment of stem cells and their osteogenic differentiation, which could further enhance the osteogenic inducing activity of BMP-2. [54,55]Interestingly, it has been reported that the expression level of BMP-3b is at its highest during the initial phase of bone fracture healing, usually in the first 7 days, [56] in agreement with our observation here.
Qualitative and quantitative analyses of western blot analysis results confirmed the higher expression of runtrelated transcription factor 2 (Runx2) and collagen type I (COL-I) proteins in GEL-30% CMPC group compared to its counterparts (Figure 4E and Supporting Information: Figure S11).Runx2 is a vital early osteogenic marker that regulates numerous genes associated with osteoblast differentiation. [36,57]COL-I is also an important early marker for osteogenic differentiation. [58]Ca 2+ and PO 4 3− ions upregulate bone regeneration-associated proteins, including COL-I, Runx2, and OCN by stimulating the BMP-2 and Wnt pathways, leading to the promotion of osteogenesis. [59]Mg 2+ could facilitate the upregulation of Runx2 and ALP through TRPM7/PI3K signaling pathway, improving the osteogenic activity of osteoblasts. [60]

.4 | In vivo bone regeneration
The critical-size rat cranial model was adopted to verify the higher bone regeneration capacity of the GEL-30% CMPC composite hydrogel in vivo.As illustrated in Supporting Information: Figure S12, two circular defects with a diameter of 5 mm were created in the rat cranium, where the left defects were left untreated as the sham control.In some groups, GEL-30% CMPC hydrogel precursor solution was deposited in the right defects using a syringe to simulate the injection process.The right defect in some groups was also used to evaluate the effect of GEL hydrogel.It should be noted that GEL samples should have been crosslinked before applying to the defect site since their low yield stress caused its precursor solution to spread to the adjacent tissues and not fill the defect site.This decreases the cohesion of the tissue surrounding the defect site and the bone scaffold.This is a significant advantage of incorporating CMPC with the polymeric structure rendering an injectable bone regeneration hydrogel with enough yield stress.
The quality and quantity of the regenerated bone were characterized by microcomputed tomography (µCT) and presented in Figure 5A,B.The evaluation of morphometric indices using µCT technique is known as a standard method to assess the callus structure in the bone fracture or defect.These indices are used to measure the strength of the callus, evaluate the efficiency of bone regeneration, and predict the refracture risk by determining the volume and quality of the forming bone. [61]For all the control groups, the defects remained not fully healed even after 12 weeks.The BV/TV quantitative data (Figure 5B) implied that the control defects were mainly filled with soft callus and weak mineralization.This has been clearly observed in the coronal images for control groups in Weeks 4 and 12.The µCT images (Figure 5A) also showed that the newly generated bone tissue in the control defects was formed mainly by the outgrowth from the bone tissue in the defects perimeter.Quite the reverse, the defects filled with hydrogels showed an even distribution of regenerated bone throughout the defect site (Figure 5A, coronal view), highlighting the higher osteoinductive performance of GEL and GEL-CMPC hydrogels.To compare both hydrogels, the GEL-30% CMPC hydrogel group demonstrated a significantly higher mean volume fraction of regenerated bone (BV/TV, 37.9% at Week 4, and 49.2% at Week 12) compared with the GEL groups (27.1% at Week 4 and 31.4% at Week 12).It is also worthwhile to assess bone regeneration from the perspective of normalized data.To this end, the BV/TV for each hydrogel (right defect) was divided into that of the control condition (left defect).The results showed nearly 20% higher BV/TV for GEL groups for both Weeks 4 and 12.It can be concluded that the rate of bone mineralization is almost similar between the control and GEL groups in these 8 weeks.However, BV/TV was 97% higher than control after 4 weeks for GEL-30% CMPC groups, while this morphometric index decreased to 55% higher after 12 weeks.This means that using GEL-30% CMPC, bone regeneration reached near the final stages that occur at a slower rate, while in control or GEL groups, there are still many empty spaces for bone mineralization.Moreover, the images taken from the coronal view in Week 4 showed that the mineralization had been taking place more evenly throughout the defect treated with GEL-30% CMPC when compared to GEL groups.This significantly influences the rate and quality of bone regeneration, leading to more dense and evenly distributed hard callus in GEL-30% CMPC groups after 12 weeks (Figure 5A, coronal view).Interestingly, the aforementioned increase in the mineralized portion can not be attributed to the CMPC in the hydrogel precursor, or even their reprecipitation since the same analysis on the 3D reconstruction images of freshly prepared hydrogel (Figure 3D) showed that the "BV/TV" parameter for this hydrogel is about 6.5%.Additionally, the groups treated with GEL-30% CMPC hydrogel showed a greater trabecular thickness (Tb.Th), trabecular number (Tb.N), and smaller trabecular separation (Tb.Sp) than both GEL-treated and control groups.BV/TV, Tb.Th, and Tb.Sp of the GEL-treated groups almost showed no significant difference to those of the control group at postoperation Weeks 4 and 12, indicating that the CMPC particles were crucial to the in vivo bone repair.
Histological analyses of regenerated tissue were conducted by Hematoxylin and Eosin (H&E) and Masson's trichrome stainings to further investigate the bone regeneration capacity of our composite hydrogels.As illustrated in Figure 5C, only a thin layer of tissue was observed around the defect border in the control group, while the fibrocartilaginous (soft) callus was formed to a greater extent in the defects treated with GEL and GEL-30% CMPC hydrogels.This observation proved the supportive function of the hydrogels in cell migration and deposition of new ECM compared to untreated defects.Although both hydrogels showed better performance than the control group, the cell ingrowth and matrix deposition were significantly higher and denser in the group treated with GEL-30% CMPC.It was also observed that the periphery of the host bone in the defect site was lined with osteogenic cells in a higher density in the group treated with GEL-30% CMPC.This again confirmed the faster and more efficient bone regeneration capacity of the developed bone regeneration hydrogel.Masson's trichrome staining was performed to identify bone maturity in different groups. [62]The significantly more areas stained in blue in the GEL-30% CMPC group implied a greater extent of woven bone in this group.More interestingly, the dark blue portions surrounding the remaining hydrogel proved the deposition of calcified cartilage in the vicinity of the developed hydrogel.This deposition can be related to the inorganic composition of GEL-30% CMPC, which supports a faster calcification and regeneration process. [49]Newly formed osteoids as an indication of mature bone were also observed as red colorstained parts in the deposited matrix. [62,63]Moreover, in the matrix deposited in the GEL-30% CMPC group, a number of hypertrophic chondrocytes were found entrapped in the forming bony matrix that would further develop into osteocytes. [57]Taken together, in agreement with the other analyses provided in this study, in vivo results proved the significant improvement in the regeneration capacity of bone regeneration hydrogel when CMPC is used in combination with an organic constituent.

| CONCLUSION
We have successfully designed and optimized an injectable hydrogel suitable for bone augmentation.To this end, a usual polymeric scaffold made of PAM was composited first with alginate and RGD-grafted alginate, providing a double network system with higher biocompatibility and bioactivity.Then, various concentrations of CMPC particles were added to the hydrogel precursor, resulting in an injectable hydrogel with a suitable cohesion between the precipitating inorganic particles and the crosslinking polymeric chains.The hybrid system with 30% CMPC content could offer fine control over the gelation time without dramatic temperature rise and simultaneously increase both initial and ultimate mechanical strength, which are of great importance clinically when a bone defect should be treated with injectable hydrogel with in situ crosslinking.Other worthy physical features of the GEL-CMPC were interconnected pores, lower swelling ratio, and higher biodegradation rate fitting the in vivo bone regeneration needs.The reprecipitation of inorganic particles during hydrogel remodeling and its ions release offered better biocompatibility, adhesion, and differentiation capacity examined with different cell types in vitro.In vivo assessment in the cranial defect model using µCT technique and histological studies proved the significantly higher bone regeneration capacity of GEL-CMPC hydrogel in terms of higher cell migration, matrix deposition, mineralization, osteoid formation, and osseointegration.Taken together, we believe that the simple addition of CMPC particles during hydrogel preparation provides an injectable hydrogel with great potential for bone regeneration in clinical practice.

| Fabrication process
Preparation of OSA grafted by RGD peptide Sodium alginate was oxidized as previously reported. [65]riefly, 3 SA and 3.25 g sodium periodate were dissolved in 250 mL of deionized water.The reaction was performed at 25°C for 4 h in the dark.To neutralize the excess periodate, 2 mL of ethanediol was added and stirred for at least 30 min.Five hundred milliliters of ethyl alcohol was added to precipitate out the product, which was separated and washed with ethyl alcohol three times.The resulting solid was dissolved in 100 mL of deionized water, dialyzed in a 7000 Da MWCO dialysis bag for 72 h against deionized water, and then lyophilized to obtain OSA.To graft arginine-glycine-aspartic acid (RGD-peptide) on OSA polymer chains, 0.1 g of OSA and 10 mg RGD-peptide were dissolved in 1 and 10 mL of deionized water, respectively.Four milliliters of RGDpeptide solution was mixed with 1 mL of OSA solution and stirred for 36 h in the dark.The resulting solution was lyophilized and the OSA grafted by RGD was obtained for the subsequent fabrication process. [65]The successful oxidation and grafting were confirmed by FT-IR analysis (Supporting Information: Figure S14).

Preparation of GEL-CMPC
All the raw materials were sterilized by ultraviolet irradiation for 30 min, and the fabrication process was performed on an aseptic workstation.Raw materials of the developed composite hydrogels consisted of two main parts: organic hydrogel components (AM/OSA/SA) and inorganic particles (CMPC).CMPC particles contained the dead burned magnesium oxide (MgO), monobasic potassium phosphate (KH 2 PO 4 ), and calcium hydrophosphate (CaHPO 4 ), with a molar ratio of 4:2:1.First, RGD-peptide-grafted OSA and untreated SA were mixed in equal proportions and dissolved in deionized water to prepare a 4% (w/v) solution, denoted as the OSA/SA solution.The organic hydrogel precursor consisted of OSA solution, monomer AM, APS as the activator, and MBAA as the crosslinker with a mass ratio of 4:1:0.1:0.012.The mass ratio of CMPC particles to the organic hydrogel precursor was adjusted according to the experimental design.The sample was denoted as GEL-X % CMPC, in which X% represented the mass fraction of CMPC particles.MgO, CaHPO 4 , and APS were mixed for 15 min at 300 revolutions per minute (RPM) to obtain a uniformly dispersed solid phase, which was then maintained at 35°C for at least 1 h.Meanwhile, AM monomers, MBAA, and KH 2 PO 4 were dissolved in the OSA/SA solution consecutively with an interval of 60 s, then kept similarly at 35°C for at least 1 h.After the temperature stabilization, the solid particles and solution were mixed and stirred at 500-800 RPM for almost 2-5 min.The slurry was deposited into target molds, waiting for complete gelation at 35°C, and used for further assessments unless otherwise stated.
Preparation of GEL GEL hydrogels (OSA/SA/PAM hydrogels without CMPC particles) were fabricated as the control group.Monomer AM and MBAA were dissolved in OSA/SA solution, and APS was added after the solution was placed at 35°C for at least 1 h.After the APS dissolution through mixing, the hydrogel precursor was injected into the molds and kept at 35°C for gelation.To further investigate the effect of OSA/ SA, pure PAM hydrogel was also fabricated where deionized water replaced the 4 wt.%OSA/SA solution.Other procedures were the same as above.For PAM hydrogel, the precursor consisted of deionized water, monomer AM, activator APS, and crosslinker MBAA with a mass ratio of 4:1:0.1:0.012.Monomer AM, APS, and MBAA were dissolved in the deionized water and the prepared solution was kept at 35°C for complete polymerization.

CMPC hydrates synthesis
Dry powders of the dead burned magnesium oxide (MgO), monobasic potassium phosphate (KH 2 PO 4 ), and calcium hydro-phosphate (CaHPO 4 ) were mixed for 15 min at 300 RPM with a molar ratio of 4:2:1.They put at 35°C for at least 1 h.Then, deionized water was added, and the whole was mixed for 2-5 min.The ratio of water to dry powder was 30% w/w.CMPC particles synthesized as such with no polymer involved were only used in temperature measurement and western blot tests as control.The detailed mix proportions of PAM, GEL, GEL-CMPC hydrogels, and CMPC hydrates are listed in Supporting Information Table S3.

| Characterization of gelation and composition of hydrogels
Temperature measurement A type-K thermocouple was utilized to record temperature during the gelation process in vitro.The probe was placed in the mold before pregel deposition and the ambient temperature was controlled at 35°C to mimic the in vivo gelation environment better.The gelation process of OSA/ PAM hydrogel and the hydration process of CMPC particles were also assessed for comparison.A thermal imager (THT60, Italy) was utilized for the in vivo temperature measurement after injecting about 30 μL of the composite hydrogel precursor into a cylindrical calvarium defect (diameter = 5 mm, height = 2 mm) prefabricated in a rat model.

Characterization of gelation process
Eight-channel isothermal calorimetry (TAM AIR, USA) was utilized to characterize the heat release during the gelation process at 35°C.In this regard, the organic and inorganic parts were mixed at 800 RPM for 1.5 min and instantly inserted into the active cell of the calorimeter using a special ampoule.A certain amount of deionized water was inserted in the inert cell of the calorimeter.The comparison of heat release between the active cell and the inert cell was measured by the heat flow sensors.Time to stir and the sample insertion was limited to 2 min.A dynamic rheological analyzer (Thermo Fisher Mars-40) was employed to investigate the gelation time of the pregels mixed for about 5 min before analysis.The rheological measurements were carried out using a parallel plate geometry (25 mm) with a gap size of 900 µm in a steady shear rate of 1 s −1 and constant temperature (35°C).The first derivative of the stress versus time was calculated and the first peak point was defined as the gelation time.

Compositional and structural analyses
All samples for the following tests were freeze-dried (Scientz-18ND) for at least 3 days to remove the water and maintain the microstructure intact.Mineral contents were investigated using an X-ray diffractometer (XRD, D8 Discovery), operating at 30 kV and 30 mA to provide Cu K α radiations.The analysis was performed in the 2θ range of 5-70°at a step size of 0.02°and 0.15 s per step.A FT-IR, Nicolet iS10 was utilized to record the transmittance spectra of the samples over the range of 4000-400 cm −1 .For the following measurements, the samples were swollen for 3 days in SBF before freeze-drying.The morphology of the hydrogels was assessed using a SEM (FEI Sirion) at an acceleration voltage of 20 kV.The hydrogels were sputter-coated with platinum before the assessment.Another set of samples with no coating was analyzed using an energy-dispersive X-ray analyzer (EDX, Element EDS System, EDAX, AMETEK Inc.,) connected to the SEM instrument to investigate their elemental content.APEX-EDS software was employed for this analysis.A transmission electron microscope (TEM, Talos F200X) at an acceleration voltage of 200 kV was utilized to further investigate the morphology of the composite hydrogels.X-ray nano-computed tomography (nano-CT, Zeiss Xradia 510 Versa) was employed to characterize the hydrogel microstructure.The scanning was performed at an acceleration voltage of 70 kV and a source current of 85 μA.The exposure time was set to 10 s, and the acquired data were analyzed using ORS Dragonfly software (Object Research Systems, Canada).A MIP (Micromeritics 9510) was used to measure the porosity and pore size distribution of the hydrogels more precisely.This process consisted of two stages.The first low-pressure stage ranged from 0 to 0.0036 MPa, while the second highpressure one ranged from 0.0036 to 200 MPa, followed by an extrusion stage from 200 to 0.14 MPa.

Compressive strength
Samples were placed in SBF solution to swell for 3 days to ensure their complete swelling before mechanical tests.Next, the compressive strength of relevant samples (10 × 10 × 15 mm) was measured using a universal testing machine (UTM-6503) at a loading speed of 5 mm/min.The compressive modulus was calculated from the slope of the stress-strain curve in the linear region of 0%-40% strain.The measurements were performed in triplicate.

In vitro degradation
Weight loss and releasing ions were monitored to characterize the in vitro degradation properties of the hydrogels.Lyophilized samples were weighed (M 1 ), immersed in SBF solution, and incubated at 37°C.At each time point, a set of samples was washed with deionized water, freeze-dried, and weighed (M 2 ).The mass remaining (%) was then calculated using the below formula.The experiments were conducted in triplicate.
The potentiometric titration method was employed to measure the concentration of Ca 2+ and Mg 2+ as an indication of sample degradation. [66]In this regard, an automatic potential titrator (ZDJ-5, Shanghai Rex Electric Chemical Co., Ltd.) equipped with a Ca 2+ -selective electrode and a calomel reference electrode (232-01) was used.Tris(hydroxymethyl)aminomethane (Tris, ACS, T110600, Aladdin) and acetylacetone (HACAC, AR, A110367, Aladdin) were dissolved together in the deionized water to serve as the auxiliary complexing agent.The concentrations of Tris and HACAC were 0.035 and 0.055 mol/L, respectively, and the pH value was adjusted to 9.5-10.5 using ammonia solution (25%-28%, A112077, Aladdin).The as-prepared hydrogels were immersed in SBF solution at a hydrogel/SBF ratio of 0.1 g/mL for different time intervals.As for each measurement, a working solution was made by mixing 1 mL of extracted liquid with 20 mL of fresh deionized water and 20 mL of the auxiliary complexing agent.After mixing, the pH of the working solution was adjusted to 9.5-10.5 using ammonia solution (25%-28%, A112077, Aladdin).For titration, ethylenediaminetetraacetic acid disodium salt solution (EDTA-Na, 0.05 mol/L, GBW(E) 081128, Aladdin) was automatically added into the working solution drop by drop using the automatic potential titrator, which could simultaneously monitor the continuous change in the electrode potential of the working solution as a function of the added volume of EDTA-Na.As illustrated in Equations ( 2) and (3), Mg 2+ in the working solution initially tends to be complexed with HACAC to form Mg(ACAC) 2 while Ca 2+ was free.With the continuous addition of EDTA, Ca 2+ tends to be complexed with EDTA-Na to form EDTA-Ca, continuously decreasing the electrode potential.The first potential jump appeared when all the Ca 2+ ions were consumed, and the added volume of EDTA was denoted as V 1 (mL).Next, EDTA would react with Mg(ACAC) 2 because EDTA-Mg has a higher complexation equilibrium constant than Mg(ACAC) 2 .Similarly, the second potential jump appeared once the Mg(ACAC) 2 were all converted to EDTA-Mg.The total added volume of EDTA was denoted as V 2 (mL).The concentrations of Ca 2+ and Mg 2+ (mg/l) were calculated using Equations ( 3) and (4), in which M 1 and M 2 (g/mol) were the molar mass of calcium and magnesium, respectively.Additionally, c(EDTA-Na) was the concentration of EDTA-Na solution.V 0 was the volume of the extracted liquid added in the first step and was fixed at 1 mL for all the measurements.

Swelling behavior
Lyophilized hydrogels were immersed in SBF solution and incubated at 35°C.At each time point, the samples were weighed (M w ) after the surface water was removed using tissue paper.Then, they were washed with deionized water and freeze-dried, and their dry weights were measured (M d ).The swelling ratio was then calculated using Equation (5).To calculate the water content in hydrogel samples in their equilibrium state, Equation ( 6) was used.

Cell viability
To prepare the samples for the cell viability tests, the pregels were deposited in 96-well plates and, after complete gelation and swelling in SBF medium, were cut and sterilized for 30 min under UV irradiation.The sample height was about 1 mm after sufficient swelling.L929 cells were seeded and cultured in 24 well plates (NEST) for 6 h (5000 cell/well).The prepared scaffolds were immersed in the cell culture medium and cocultured for 1, 3, 7, 14, and 21 days.The culture medium was replaced every 12 h during these periods.The cell viability was quantified at the indicated time points by Cell Counting Kit-8 assay (CCK-8, Beyotime) following the manufacturer's instructions.Briefly, after adding 10 µL of CCK-8 reagent to each well, cells were cultured for another 4 h, and then 150 μL of the medium was transferred to 96-well plates.The absorbance at 450 nm was measured using a microplate reader (EPOCH2, BIOTEK).Cell viability was expressed in percentage relative to the negative control (untreated cells).For further evaluations, the cells on the culture plate were stained with Calcein-AM and propidium iodide (PI) (Bestbio) according to the company procedure and then observed using a fluorescence microscope (DMI8, Leica), at 488 and 552 nm, respectively.

| In vitro osteogenic differentiation
Hydrogels performance in inducing osteogenesis was assessed by measuring the ALP activity of L929 and MC3T3-E1 cells as well as ECM mineralization of MC3T3-E1 cells cocultured with the hydrogels.The ALP activity and ECM mineralization were detected by an ALP kit (Beyotime) and ARS staining (Beijing Solarbio Science & Technology Co., Ltd.), respectively.

ALP activity assay
The hydrogels were prepared in 24-well plates as explained before and cocultured in 6-well plates (NEST, Wuxi) containing 2 × 10 5 cells per well for 1, 3, 7, and 14 days.At the indicated time points, the medium was removed and the cultured cells were washed with PBS (Procell).PBS was removed and the cells were lyzed by adding 200 μL cell lysis buffer to each well.After 30 min, the cell lysates of each well were collected by centrifugation (12 000 rpm, 10 min).Then, 50 μL of substrate solution was added to 5 μL of supernatant from each condition and they were incubated at 37°C for 30 min.Subsequently, 150 μL of a color development reagent was added, followed by measuring their absorbance on a microplate reader at 405 nm.The results were normalized to the total protein content detected by a BCA protein assay kit (Beyotime).
The experiments were conducted in triplicate.

ARS staining
The hydrogels were prepared in 24-well plates as explained before and cocultured in 6-well plates containing 2 × 10 5 MC3T3-E1 cells per well for 3 and 7 days.After this period, the cells were thoroughly rinsed with PBS and fixed with 4% paraformaldehyde (TCI) for 1 h.Next, the cells were stained with ARS solution at room temperature for 30 min.The cells were observed with an inverted phase-contrast microscope (Zeiss) after being rinsed with water several times to remove any unspecific ARS binding.

Real-time quantitative reverse-transcription PCR (qRT-PCR)
To further analyze the hydrogels' ability to induce osteogenesis, the expression of eight osteogenic-related genes was investigated by qRT-PCR.In this regard, the hydrogels prepared in 24-well plates were cocultured with hBMSCs cultured in 6-well plates (2 × 10 5 cells per well).After 7 days, the cells were rinsed with PBS and lysed with 500 μL of Trizol reagent (Sangong Bioengineering Co., Ltd).Then, chloroform was added, and after thorough mixing, the solutions were centrifugated (12 000 rpm, 15 min) to obtain the total RNA.The extracted RNA was transferred to new tubes, and an equal volume of isopropyl alcohol was added.After thorough mixing by hand, the RNA was precipitated through centrifugation (12 000 rpm, 10 min).Next, the precipitates were rinsed with 75% ethanol and dissolved in the RNase-free diethylpyrocarbonate (DEPC)-treated dd-H 2 O (20 μL).The concentration of the isolated RNA was measured by a small molecule protein-nucleic acid tester (Nano-drop 2000, Thermo Scientific).0.8 g of the isolated RNA was reversetranscribed into the complementary DNA (cDNA) with a First Strand cDNA Synthesis Kit (Jumei Biotechnology) according to the manufacturer's instructions.The reverse transcription reaction was performed at 37°C for 15 min and then terminated by heating to 85°C for 5 min.The real-time PCR was performed using an SYBR Green I Premix ExTapTM (Jumei Biotechnology), gene-specific primers, and 1 μl of cDNA.The signal was amplified by setting 45 reaction cycles (Supporting Information: Table S4), and the mRNAs expression levels were normalized to the house-keeping gene glyceraldehyde-3phosphate dehydrogenase (GAPDH) for each condition.A group without adding any material was set as the control, and the relative fold gene expression was calculated using 2 −ΔΔCT method.The experiments were conducted in triplicate.The primer sequences used in this assessment have been listed in Supporting Information: Table S5.

Western blot analysis
The hydrogels prepared in 24-well plates were cocultured with hBMSCs previously cultured in 6-well plates (3 × 10 5 cells per well).After 7 days, proteins from different groups of cells were extracted using a protein lysis buffer (Beyotime), and the protein concentration was quantified using a colorimetric protein assay kit (Jumei Biotechnology).Sixty microgram/lane of protein was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to a polyvinylidene difluoride (PVDF) membrane (0.22 μm, Millipore).5% w/v nonfat milk in tris buffered saline with tween (TBST) was used for blocking (1 h).Then, the membrane was washed trice and incubated overnight at 4°C with corresponding primary antibodies: anti-Collagen I (AF7001, Affinity Biosciences), anti-RUNX2 (AF5186, Affinity Biosciences), anti-Osteocalcin (DF12303, Affinity Biosciences), anti-GAPDH (AF7021, Affinity Biosciences).All the primary antibodies were diluted at 1:1000 in a dilution buffer (R22367, Shanghai Yuanye Bio-Technology Co., Ltd).The membranes were washed and incubated with the secondary antibody (Goat Anti-Rabbit IgG (H+L) HRP, S0001, Affinity Biosciences) for 2 h.The secondary antibody was diluted at 1:10 000 in a dilution buffer (R21301, Shanghai Yuanye Bio-Technology Co., Ltd).The blots were washed three times and developed with the ECL system (Affinity Biosciences).Films were scanned, and the optical density was analyzed using Amersham Imager 600 (Cytiva).
All the experiments were repeated three times.

Surgical implantation
All the in vivo experiments were approved by the Zhejiang Provincial Department of Science and Technology (Approval No. SYXK 2021-0043).All the rats were housed in cages for at least 7 days at room temperature with 50%-60% humidity before the implantation.Twelve female rats (160-200 g, 8 weeks) were randomly divided into two groups and were anesthetized with Zoletil 50 (Virbac) and Rompun (Bayer).Two circular defects with a diameter of 5 mm and a depth of around 1 mm were created on the rat cranium.The left defects were left untreated, while the right ones were filled with about 30 μL of either OSA/OAM hydrogel (n = 6) or the precursor solution of the composite hydrogel (CMPC-OSA/ PAM) (n = 6).Next, the wound was carefully sutured and sterilized with iodophor.After predetermined periods, the animals were euthanized by CO 2 gas, and the bone parts were fixed in 10% neutral formalin buffer for further analysis.

Micro-CT analyses
The formalin-fixed specimens were scanned using a micro-CT device (Skyscan1174 X-Ray Microtomography) at an acceleration voltage of 50 kV and a source current of 800 μA.
A cylindrical region with a diameter of 5 mm and a height of

Histological analyses
The specimens were decalcified in an EDTA-Decalcifying fluid (AR1071, Boster Biological Technology co.Ltd) for 20 days where the decalcifying fluid was changed every 5 days.Then, the specimens were dehydrated with an ascending ethanol gradient and paraffin-embedded.They were sectioned to 5 µm thickness with a microtome (RM2016, Leica).H&E and Masson's trichrome (MT) stainings were carried out to examine the bone repair.The stained sections were observed using an optical microscope (DM2000, Leica).

F I G U R E 1
Schematic diagrams to show the gelation process and bone regeneration effects of composite hydrogel: (1) Ca 2+ ions from the body fluid ionically crosslink SA/oxidized sodium alginate (OSA) within seconds to prevent the injected hydrogel precursor from leaking out of the target site.(2) The formation of the first ionically crosslinked network within seconds induced by Ca 2+ /Mg 2+ ions from the dissolution of calcium/magnesium phosphate cement (CMPC) particles.(3) Heat released from CMPC hydration accelerates the free-radical polymerization process of poly(acrylamide) (PAM), which means formation of a second network structure within several minutes.(4) The composite hydrogel provides an appropriate microenvironment for cell migration and attachment.(5) The injectable hydrogel promotes new bone formation and defect repair.

F
I G U R E 2 (A) Heat release of the GEL and the GEL-calcium/magnesium phosphate cement (CMPC) hydrogels during the gelation process.Scanning electron microscope (SEM) images of the (B) GEL hydrogel and the (C) GEL-30% CMPC hydrogel.SEM and EDS elemental mapping images of the characteristic mineral (D) K-struvite (KMgPO 4 •6H 2 O) and (E) phosphates containing calcium in the asprepared GEL-30% CMPC hydrogel.(F) X-ray diffraction (XRD) patterns of the GEL-30% CMPC hydrogel after immersion in simulated body fluid (SBF) solution for up to 28 days.(G) Fourier transform infrared spectroscope (FT-IR) spectra of the GEL hydrogel, the GEL-CMPC hydrogel, and the hydrated CMPC particles.

F
I G U R E 3 (A) Comparison of the compressive strength of pure poly(acrylamide) (PAM), GEL, and GEL-calcium/magnesium phosphate cement (CMPC) hydrogels with various CMPC content.(B) Swelling ratios of GEL and GEL-CMPC hydrogels after incubation in simulated body fluid (SBF) solution for various time periods.(C) Pore size distributions and porosity of GEL-CMPC hydrogels after incubation in SBF solution for 7 and 28 days.(D) Three-dimensional reconstruction images of the swollen GEL-30% CMPC hydrogel.The dark blue area represents pore network while the white and green areas represent inorganic particles and hydrogel networks, respectively.(E) In vitro release of Mg 2+ and Ca 2+ ions from the GEL-30% CMPC hydrogel incubated in SBF solution.(F) Scanning electron microscope (SEM) images of GEL-30% CMPC hydrogel after immersion in SBF solution for 28 days.

1. 5
mm was set as the region of interest (ROI).A reconstruction software (N-Recon, Bruker) was employed for the 3D reconstruction.CT-VOL and CT-AN (SKYScan) were used to investigate bone formation indices, including bone volume fraction (BV/TV), trabecular thickness (Tb.Th), trabecular number (Tb.N), and trabecular separation (Tb.Sp).