Physicochemical Properties, In Vitro Degradation, and Biocompatibility of Calcium Phosphate Cement Incorporating Poly(lactic-co-glycolic acid) Particles with Different Morphologies: A Comparative Study.

Calcium phosphate cement (CPC) is one of the most promising synthetic biomaterials for bone defect repair, but its low degradation rate and the lack of macropores restrict its repair effect. Poly(lactic-co-glycolic acid) (PLGA) is commonly used as an in situ pore forming agent in CPC, and the morphology of PLGA would affect the properties of CPC. In this study, three kinds of PLGA particles with different morphologies, including dense PLGA microspheres, dense milled PLGA particles with an irregular shape, and porous PLGA microspheres, were respectively incorporated into CPC matrix. The influences of the morphology of PLGA particles on the setting time, porosity, mechanical properties, in vitro degradation, and cytocompatibility of CPC were comparatively investigated. The results showed that the CPC composites containing dense spherical and irregularly shaped PLGA particles showed proper setting time and better compressive strength, but the CPC composite incorporating porous PLGA microspheres significantly prolonged the final setting time and dramatically decreased the compressive strength of CPC. The CPC composite containing irregularly shaped PLGA particles has shown a slightly faster in vitro degradation rate than that containing dense PLGA microspheres. In addition, the CPC composites containing dense PLGA particles were beneficial for cell proliferation. Taken together, the dense PLGA particles are suitable for use as in situ pore forming agents in the CPC matrix, and meanwhile, the dense irregularly shaped PLGA particles are more easily prepared with low cost.

Introduction

Calcium phosphate cement (CPC) has garnered much attention since it was first synthesized in the 1980s by Brown and Chow.[1,2] CPC is thought to be one of the most promising materials for bone repair, owing to its similar composition with the mineral phase of nature bone.[3] Furthermore, CPC possesses superb properties such as excellent biocompatibility, osteoconduction, self-setting, and easy to shape, which make it a practical candidate for bone grafts.[4,5] However, the lack of macropores and the slow degradation rate of CPC have restricted its bone regeneration effect.[6] Much efforts have been made to accelerate the degradation rate of CPC, such as incorporating biodegradable polymer microspheres,[6] increasing porosity (micro/macro)[7,8] and ion substitution,[9] and so forth. Among them, incorporating polymer microspheres is the most commonly used method due to its good biocompatibility and easily tailored biodegradation rate.[10,11]

When incorporated into the CPC matrix, the poly(lactic-co-glycolic acid) (PLGA) particles degrade prior to the CPC matrix, producing macropores in situ which are beneficial for tissue ingrowth, vascularization and bone formation.[12,13] The spherical PLGA particles have attracted the most attention. Félix Lanao et al. investigated the effects of the molecular weight, chemical characteristics (acid-terminated or end-capped), and density (hollow vs dense) of PLGA microspheres on the degradation of CPC and found that the degradation rate of CPC and PLGA was mainly depended on the end-group functionalization, and the dense microspheres were more efficient porogens than hollow microspheres.[14] Habraken et al. found that CPC with greater content of PLGA microspheres had higher degradation rate.[15]

The fabrication processes of PLGA microspheres are relatively complicated, of higher cost, and involved with toxic organic solvents, such as dichloromethane. By contrast, the preparation of PLGA particles with an irregular shape by the direct milling method is much more convenient, economic, and environmentally friendly. A recent study showed that incorporating dense milled PLGA particles into CPC contributed to bone healing in vivo.[16] To our knowledge, the influence of morphology of PLGA particles on the properties of CPC has never been comparatively studied.

In this study, dense PLGA microspheres, porous PLGA microspheres, and dense irregularly shaped PLGA particles were prepared and incorporated into CPC. The effects of PLGA particles’ morphology on the physicochemical properties, in vitro degradation properties, and biocompatibility of CPC were systematically and comparatively investigated.

Results

Physicochemical Properties

The morphologies of PLGAdm (dense PLGA microspheres), PLGAdis (dense irregularly shaped PLGA particles), and PLGApm (porous PLGA microspheres) are presented in Figure . The PLGAdm showed a regular spherical shape with a smooth surface (Figure a). The PLGAdis were irregular in shape with a wrinkle surface (Figure b). Lots of micropores were uniformly distributed on the surface (Figure c) and in the interior (Figure S2) of PLGApm.

Figure 1

SEM images of PLGAdm (a), PLGAdis (b), and PLGApm (c).

The setting time of CPC containing different contents of PLGAdm are presented in Figure a. The initial and final setting time of PLGAdm/CPC did not change distinctively with increasing content of PLGAdm. The initial setting time and the final setting time hovered at around 15 and 35 min, respectively. Figure b,c shows the initial and final setting time of CPC containing different PLGA particles. There was no significant difference in the initial and final setting time between the composites containing both types of dense PLGA particles (PLGAdm and PLGAdis). However, the final setting time of PLGApm/CPC containing 5 and 10 wt % PLGA particles was 45.40 ± 0.86 and 50.60 ± 0.85 min, respectively; which are both significantly longer than that of pure CPC (33.73 ± 0.57 min).

Figure 2

Influence of the PLGAdm content (a) and particle morphologies (b,c) on the initial and final setting time of PLGA/CPC composites.

The XRD patterns of CPC/PLGA composites are illustrated in Figure . The main phase of all these hydrated PLGA/CPC composites was poorly crystallographic hydroxyapatite. The peak position attributed to hydroxyapatite hardly changed with the increasing content of PLGAdm in the composite. Similarly, the PLGA particle morphologies had no effect on the main phase of hydrated PLGA/CPC as well.

Figure 3

XRD patterns of CPC with different contents of PLGAdm (a) and different PLGA particle morphologies (b, 10 wt % of PLGA).

The compressive strength of PLGA/CPC composites with different contents and morphologies of PLGA particles is presented in Figure . The compressive strength of the cement decreased with the increasing content of PLGAdm. The compressive strength of pure CPC was 15.33 ± 1.36 MPa, while it decreased to 8.8 ± 0.84 MPa when the addition content of PLGAdm increased to 30 wt %. The compressive strength of PLGAdis/CPC and PLGAdm/CPC was at the same level when the addition content was 5 and 30 wt %. Compared to PLGAdm/CPC and PLGAdis/CPC, the compressive strength of PLGApm/CPC was significantly lower.

Figure 4

Influences of PLGAdm content (a) and particle morphologies (b) on the compressive strength.

The porosity of PLGA/CPC composites is plotted in Figure . As the content of dense PLGA microsphere particles increased, the microporosity of PLGA/CPC composites decreased; inversely, the macroporosity and total porosity increased (Figure a). For the same content of PLGA particles, the microporosity (Figure b) and macroporosity (Figure d) of PLGAdis/CPC was similar to that of PLGAdm/CPC, and the microporosity of PLGApm/CPC was a slightly lower than that of PLGAdm/CPC and PLGAdis/CPC; inversely, the macroporosity of PLGApm/CPC was higher than that of PLGAdm/CPC and PLGAdis/CPC.

Figure 5

Influences of PLGAdm content (a) and particle morphologies (b–d) on the porosity of PLGA/CPC.

In Vitro Degradation Behaviors

Because of the low compressive strength and long final setting time of PLGApm/CPC, we only chose PLGAdm/CPC and PLGAdis/CPC for the in vitro degradation test and the in vitro cell biocompatibility test.

The variation of the mass loss of PLGA/CPC composites and the pH value of the immersion solution are given in Figure . An abrupt mass loss appeared for all the PLGA/CPC composites. When incorporating the same content of PLGA particles, the mass loss rate of PLGAdis/CPC composites was slightly higher than that of PLGAdm/CPC composites, as opposed to the variation trend of pH value. The mass loss rate of 30PLGAdm/CPC and 30PLGAdis/CPC was significantly higher than that of the other PLGA/CPC composites since week 6. After immersing for 8 weeks, the mass loss rates of 30PLGAdm/CPC and 30PLGAdis/CPC were 39.5 ± 1.1 and 39.2 ± 2.6%, respectively. In addition, the pH value of immersion solution for all PLGA/CPC composites was below 7 while that of pure CPC gradually and steadily approached an initial value of 7.4. In addition, as the content of PLGA particles increased, the pH value of the immersion solution of PLGA/CPC composites obviously decreased, especially when the content of PLGA particles reached 30 wt %.

Figure 6

Mass loss of the PLGA/CPC composites (a) and pH value of the immersion liquids (b).

Figure shows the compressive strength and XRD patterns of PLGA/CPC composites at different degradation time points. The compressive strength of composites decreased as the immersion time prolonged. The composites containing the same content of PLGA particles with different particle morphologies showed similar compressive strength at the early stage of degradation (before 4 weeks). At week 6, a significant difference in compressive strength between 10PLGAdis/CPC and 10PLGAdm/CPC was observed. However, this difference was not observed between 30PLGAdm/CPC and 30PLGAdis/CPC. The composites containing higher content of PLGA (30PLGAdm/CPC and 30PLGAdis/CPC) showed lower compressive strength compared to those containing lower content of PLGA (10PLGAdm/CPC and 10PLGAdis/CPC). At week 4, the compressive strength was higher than 4 MPa for all the samples, but it was below 2 MPa for 30PLGAdm/CPC (1.1 ± 0.3 MPa) and 30PLGAdis/CPC (1.2 ± 0.2 MPa) at week 6. The XRD patterns of PLGAdm/CPC composites almost did not change during immersion, indicating the high stability of hydrated CPC.

Figure 7

Compressive strength of the PLGA/CPC composites (a) and XRD patterns of 30PLGAdm/CPC at different degradation times (b).

The morphology of fracture surfaces of PLGA/CPC composites at different degradation times are shown in Figure . Before immersion, PLGA particles were well embedded in the CPC matrix, and their surfaces remained smooth. Degraded wrinkles appeared on the PLGAdis surfaces after 2 weeks of immersion, but the surfaces of PLGAdm were still smooth. At week 6, PLGA particles partially degraded in the composite containing 10 wt % PLGA particles. By contrast, in the composite containing 30 wt % PLGA, the PLGA particles almost completely degraded accompanied with the formation of macropores in situ, leaving a thin PLGA film on the CPC matrix.

Figure 8

SEM images of PLGA/CPC composite immersion in PBS buffer solution for different times.

In Vitro Biocompatibility

The morphologies of mBMSCs cultured on the PLGA/CPC composite disks are shown in Figure . Cells adhered well and extended completely with distinct cellular pseudopodia (see the SEM images in Figure ). It should be noted that cells cultured on the composites containing PLGA particles have shown more cellular pseudopodia, especially the cell on the surface of 10PLGAdm/CPC and 10PLGAdis/CPC [see the confocal laser scanning microscopy (CLSM) images in Figure ]. For the same content, the morphologies of PLGA particles in the CPC matrix seemed to have no effect on cell morphology.

Figure 9

Morphology of mBMSCs on the surfaces of PLGA/CPC composites, which are exhibited in SEM images (1 day, cells were painted in red false colored for better visibility, scale bar 20 μm) and CLSM images (1 day, green: cell cytoskeletons, blue: nucleus, scale bar 50 μm).

The proliferation and live/dead staining of mBMSCs cultured on the PLGA/CPC composites are shown in Figure . A constant growth of cell number was observed during the cell culture period. At day 1, cellular proliferation on all the PLGA/CPC composites was comparable with each other. At day 3 and day 7, the cell number on the composites containing PLGA particles, especially on those containing higher content of PLGA, was significantly higher than that of CPC without PLGA particles. However, for the same content of PLGA, there was no significant difference in cell number between the composites containing the PLGA particles with different morphologies. As shown in the live/dead images at Figure b, few dead cells were observed on all the CPC/PLGA composites and more live cells were found on the surface of CPC/PLGA than that of pure CPC, which indicated that CPC containing PLGA particles were highly biocompatible and favored to cell proliferation.

Figure 10

Proliferation of mBMSCs cultured on the surface of PLGA/CPC composites at different times (a) and fluorescence images of live/dead staining of cells on PLGA/CPC composites at day 3 (green: live cells, red: dead cells, scale bar 100 μm) (b) (n = 6, *P < 0.05, **P < 0.01, and ***P < 0.001).

Discussion

This study aimed to investigate the influence of PLGA morphology on the properties of CPC. Different morphologies of PLGA particles, which were used as in situ pore forming agents, were incorporated into CPC. The physiochemical properties, in vitro degradation behavior, and biocompatibility of PLGA/CPC composites were comparatively investigated. The addition of PLGApm extended the final setting time and decreased the compressive strength of CPC, which indicated that porous PLGA microspheres was not suitable for use as a CPC pore forming agent. PLGAdis/CPC showed a slightly faster degradation rate than PLGAdm/CPC, while the physicochemical properties and biocompatibility were similar for both; however, PLGAdis possessed the advantages of easily prepared with lower cost. Though the general pore forming agents used in the CPC matrix are dense microspheres, dense PLGA particles made by machine milling could be powerful competitors.

Setting time is one of the most important characteristics of self-setting CPC for clinical use.[17] In clinic, the initial setting time of CPC should not be shorter than 5 min, and the final setting time is required to be around 30 min. The initial and final setting time of CPC containing PLGAdm and PLGAdis was around 15 and 35 min, respectively, both of which are suitable for clinical use. In the composites containing dense PLGA particles, the liquid (deionized water) to powder (PCCP + DCPA) ratio in our study was kept at 0.4 mL/g. The initial and final setting time of composites containing PLGAdm did not change much with increasing content of dense PLGA particles mainly due to the unchanged liquid to powder ratio, but in Habraken’s work, the setting time decreased with increasing PLGA content.[15] This difference could be caused by using different liquid phases and different CPC powders. Incorporating PLGApm dramatically prolonged the final setting time of CPC. With the same content of PLGA, the volume percentage of porous PLGA particles (PLGApm) was supposed to be higher because of its lower density than that of the dense PLGA particles (PLGAdm and PLGAdis). Furthermore, PLGApm could absorb much more liquid than the dense PLGA particles owing to the micropores inside PLGApm. Therefore, a higher liquid-to-powder (L/P) ratio was required to PLGApm/CPCs (see Table ) to make them operable as plastic pastes. During the hydration process, the higher volume percentage of PLGApm particles and the higher L/P ratio slowed down the hydration rate and prolonged the final setting time. In conclusion, the setting time results have shown that the liquid-to-powder ratio play a more dominant role than PLGA morphologies in the effect on CPC setting time.

Table 1

Components and Amount of Liquid Phase of the PLGA/CPC Composites

PLGA particles	CPC powder (g)	liquid-phase volume (mL)	PLGA particles (g)	label
	1	0.4	0	CPC
PLGAdm	0.9	0.36	0.1	10PLGAdm/CPC
PLGAdm	0.7	0.28	0.3	30PLGAdm/CPC
PLGAdis	0.9	0.36	0.1	10PLGAdis/CPC
PLGAdis	0.7	0.28	0.3	30PLGAdis/CPC
PLGApm	0.95	0.4	0.05	5PLGApm/CPC
PLGApm	0.9	0.4	0.1	10PLGApm/CPC

The porosity of bone grafts is very important for bone conduction and vascularization.[12,13] Therefore, increasing the porosity of CPC can not only accelerate the degradation rate of CPC but also promote tissue ingrowth and new bone formation.[6] When PLGA particles completely degrade, the macroporosity of PLGA/CPC composites containing 10 wt % dense PLGA particles (10PLGAdm/CPC and 10PLGAdis/CPC) and 30 wt % dense PLGA particles (30PLGAdm/CPC and 30PLGAdis/CPC) would reach about 22 and 38%, respectively (Figure ). However, addition of PLGA particles may play a detrimental effect on the mechanical strength of the CPC matrix.[15,18] In addition, the macropores with the size ranging from 100 to 400 μm are favorable for bone tissue ingrowth.[19] Hence, PLGA particles with a size in 106–212 μm were used in this study in consideration of both the mechanical strength and tissue ingrowth. Increasing the porosity of CPC always accompanies with decreasing compressive strength.[17] However, with increasing the content of PLGA, PLGA/CPC composites showed decreasing microporosity and compressive strength. This can be explained by the weak interfacial strength between the inorganic CPC matrix and the organic PLGA particles. Incorporating PLGApm significantly decreased the compressive strength of the CPC matrix. This may be owing to the week interfacial strength, higher volume percentage, and lower elasticity modulus of porous PLGA particles.

Throughout the in vitro degradation test, for the same content of PLGA, the varying tendencies of mass loss rate and pH value profiles of PLGAdm/CPC were similar with those of PLGAdis/CPC. However, PLGAdis/CPC showed a slightly faster degradation rate than PLGAdm/CPC for the high energy milling might decrease the PLGA molecular weight.[20] Significant differences in mass loss rate and pH value were observed between the composites containing 10 and 30 wt % of PLGA after week 6 (Figure ), which indicated an earlier bulk degradation of PLGA particles in the composite containing higher content of PLGA. The higher mass loss and lower pH value of the composite containing higher PLGA content indicated that the higher content of PLGA induced a faster degradation rate. So, the CPC degradation rate can be easily manipulated by changing the content of PLGA and make it fit the regeneration rate of host bone, in the context of remaining enough mechanical strength. Moreover, changing molecular weight or chemical modification can also change the degradation rate of PLGA particles[21,22] and then regulate the degradation rate of PLGA/CPC composites. In general, compared to PLGA morphology, the content of PLGA played a more dominant role in the degradation rate of PLGA/CPC composites.

Noted that, before degradation, the compressive strength of all samples was higher than the critical mechanical requirements (compressive strength: 2–12 MPa) for repairing cancellous bone defects,[23] which means that the composites can provide mechanical support for defected bone healing for at least one month with an addition content of PLGA particles even up to 30 wt % for the mechanical strength of 30PLGAdm/CPC and 30PLGAdis/CPC were both higher than 4 MPa at week 4. A significant decrease in the compressive strength of degraded samples was observed at week 6 (Figure ). This could be explained by the introduction of macropores produced by the degradation of PLGA particles; however, the macropores could be filled with new ingrowth bone tissue and maintain a long-term strength in vivo.[24]

The biocompatibility of PLGA/CPC composites was characterized by cell adhesion morphology, live/dead staining, and proliferation on the material surface, and mBMSCs were used as the cell model. The cell culture studies on the surface of PLGA/CPC composites disks suggested that the PLGA/CPC composites can provide favorable conditions for cell adhesion, spreading, and proliferation. Note that PLGA/CPC showed more well-spread condition and higher proliferation than the pure CPC, but there were few differences in the cell spread condition and proliferation between PLGAdm/CPC and PLGAdis/CPC, which demonstrated that PLGAdis is an effective pore generation agent like the commonly used PLGAdm. For biomaterials, both the physical and chemical characteristics of the material surface were important for bioactivity. A rough material surface contributes to cell adhesion and proliferation.[25,26] Li et al. fabricated the PLGA/CPC composite pellets with sea island structure favored to cell adhesion and cell proliferation.[27] PLGA was thought to be lack of bioactive groups for cell adhesion, cell spreading, and poor osteoconductivity.[10] However, in PLGA/CPC composites, PLGA particles were mainly embedded in the CPC matrix and the bulge parts on the sample surface were also covered by CPC (see Figure ). Therefore, cells were rarely directly contacted with PLGA particles during the cell culture period. In addition, the bulge parts of PLGA particles increased the available surface area of composites disks and facilitated the cell adhesion and proliferation; this is the reason why the composites containing higher content of PLGA possessed higher cck8 OD value. Overall, based on the in vitro experiments above, we can see that both PLGAdm and PLGAdis are suitable pore forming agents for CPC via degradation, and it will be of great interest to study the effects of CPC containing PLGAdm and PLGAdis on repairing the bone defects in vivo.

Conclusions

Dense PLGA microspheres, dense PLGA particles with an irregular shape, and porous PLGA microspheres fabricated by an emulsion solvent evaporation method, machine milling, and double emulsion method, respectively, were incorporated into the CPC matrix, and the influence of PLGA morphology on the properties of CPC was investigated in this study. Incorporating porous PLGA microspheres dramatically prolonged the final setting time and significantly decreased the compressive strength of CPC. Furthermore, the operability of the composite containing porous PLGA microspheres was inferior to that of the composite containing dense PLGA particles, demonstrating that porous PLGA microspheres were not suitable for use as a pore generation agent in CPC. Compared to porous PLGA microspheres, incorporating dense PLGA particles only slightly decreased the compressive strength of CPC. What is more, the setting time of CPC containing dense PLGA microspheres and dense PLGA particles with an irregular shape were both suitable for clinical use. Though dense PLGA particles with an irregular shape and dense PLGA microspheres have shown similar effects on the setting time, porosity, and compressive strength of CPC, the degradation rate of the composite containing dense PLGA particles with an irregular shape was slightly higher than that of the composite containing dense PLGA microspheres. In addition, the composite containing dense PLGA particles could distinctly promote the proliferation of mBMSCs compared to CPC without PLGA particles. Overall, both types of dense PLGA particles are suitable for use as a pore generation agent in CPC. In addition, the dense PLGA particles with an irregular shape possessed the advantages of easily prepared with lower cost.

Materials and Methods

Preparation of CPC and PLGA Particles

The CPC powder was composed of 45 wt % partially crystallized calcium phosphate (PCCP), 45 wt % dicalcium phosphate anhydrous (DCPA), and 10 wt % amorphous calcium phosphate (ACP). The preparation process of PCCP and DCPA was described in our previous work.[28] Briefly, PCCP was synthesized by a chemical precipitation method, dropwise adding an aqueous calcium solution of Ca(NO3)2·4H2O (0.36 mol/L) to a phosphorus solution of (NH4)2HPO4·12H2O (0.15 mol/L). Then, the precipitate was centrifugally separated, lyophilized, and calcined at 450 °C for 2 h in a furnace to obtain PCCP powder. DCPA was obtained by milling the commercially available dicalcium phosphate dihydrate in absolute ethyl alcohol for 2 h and then dehydrating at 120 °C for 12 h. ACP was prepared by the chemical precipitation method. Briefly, 1 L of (NH4)2HPO4·12H2O (0.15 mol/L) was dropwise added into 1 L of Ca(NO3)2·4H2O (0.25 mol/L), and the pH value of the mixed solution was maintained at 10.0 by adding NH3·H2O. Then, the precipitated suspension was aged for 24 h. The precipitate was centrifuged, lyophilized, and sieved to the particle size below 106 μm.

PLGA (lactide-to-glycolide ratio = 50/50, Mw = 3 × 104) was purchased from M.K. Biotechnology (Jinan, China). Dense PLGA microspheres were fabricated by an emulsion solvent evaporation method. Briefly, 1 g of PLGA was dissolved in 10 mL of dichloromethane, and then, the PLGA solution was injected into a beaker containing 400 mL of 0.3% polyvinyl alcohol (PVA) solution. The mixed solution was stirred at a speed of 350 rpm overnight. The PLGA microspheres were washed three times with deionized water, centrifuged, and lyophilized. The microspheres with diameters ranging from 106 to 212 μm were separated by sieving for use in this study.

Porous PLGA microspheres were prepared by a double emulsion method. The preparation process was similar to that of dense PLGA microspheres except adding a second aqueous phase of NH4HCO3 solution. Especially, NH4HCO3 solution was added in the PLGA solution and emulsified at a speed of 8000 rpm for 30 s before being injected into the PVA solution.

Dense milled PLGA particles were fabricated by machine milling (DE-100g, Zhejiang Hongjingtian Co., Ltd, China). The PLGA particles were washed three times with deionized water, centrifuged, lyophilized, and then sieved. The irregularly shaped PLGA particles with sizes ranging from 106 to 212 μm were obtained. The PLGA particles in the forms of dense microspheres, porous microspheres, and dense irregularly shaped particles were labeled PLGAdm, PLGApm, and PLGAdis, respectively.

Preparation of PLGA/CPC Composites

The PLGA particles with different morphologies were uniformly mixed with CPC powder. Then, PLGA and CPC mixtures were homogeneously mixed with deionized water and hydrated at 37 °C with 98% humidity for 3 days. The ingredients of different PLGA/CPC composites are listed in Table .

Characterization of PLGA/CPC Composites

The cement pastes of different PLGA/CPC composites were molded into cylinder columns (Φ6 mm × 12 mm) for mechanical tests, porosity measurements, and in vitro degradation tests. The cement paste was poured into the steel molds and then pressed under 10 kg for 15 s to eliminate air bubbles. Then, the samples were demolded and hydrated at 37 °C with humidity. The compressive strength of the cement columns was measured by a universal material testing machine (Instron 5567, Instron, USA) at a crosshead speed of 1 mm/min. Each test was performed six times and the average value was calculated.

The hydrated PLGA/CPC columns were grinded into powders and the phases were analyzed by X-ray diffraction (XRD; X’Pert Pro, PANalytical, The Netherlands). Data were collected for 2θ ranging from 20 to 60° with a step size of 0.013°. The morphology, microstructure, and fracture surface of the samples were observed using scanning electron microscopy (SEM; Quanta 200, FEI, USA).

The setting time of the PLGA/CPC composites was measured by a Gillmore needle method according to ASTM C266.[29] Cement pastes were poured into a mold and maintained in a humidity chamber (37 °C, 98% humidity). The Gillmore needle was carefully lowered vertically onto the surface of the cement. The initial setting time is the time when the cement can withstand a small fixed pressure applied by the thick needle without notice penetration. The final setting time is the time when the cement can withstand a high fixed pressure applied by the thin needle without notice penetration. Each measurement was performed three times and the average value was calculated.

The porosity was measured by the Archimedes methods. For the cement columns, the micropores were attributed to the inherent pores formed in the hydration process, while the macropores were mainly produced by the degradation of PLGA particles. Therefore, the porosity of PLGA/CPC composite columns before removing the PLGA is regarded as the microporosity and the porosity after removing PLGA is the total porosity which is regarded as the porosity of PLGA/CPC composites when PLGA degrades completely. The PLGA particles in the CPC matrix were removed by calcining the PLGA/CPC columns at 600 °C for 2 h with a heating rate of 5 °C/min according to the thermogravimetric (TG) analysis results of PLGA (Figure S1). The total porosity was determined by testing the porosity of these calcined columns. Herein, the macroporosity of the cement columns after degradation of PLGA particles was calculated according to eq .where Pmac, Ptotal, and Pmic are the macroporosity, total porosity, and microporosity of cement columns, respectively.

In Vitro Degradation Behaviors of PLGA/CPC Composites

PLGA/CPC samples were immersed in phosphate buffered saline (PBS, pH 7.4) at a solution volume-to-sample weight ratio of 30 mL/g and incubated in a shaker (60 rpm) at 37 °C for 3 days and 1, 2, 4, 6, and 8 weeks. The immersion solution was refreshed every week. At each time point, the samples were taken out, washed, and dried.

The mass loss ratio of the samples was calculated according to eq .where ML (%) is the mass loss ratio of the sample, Ws denotes the mass of the sample before immersion, and Wd represents the dry mass of the sample at each datum time point.

The pH value of the immersion solution was measured immediately after the samples were taken out.

Cell Adhesion and Proliferation on PLGA/CPC Composites

The cement pastes of different PLGA/CPC composites were molded into disks (Φ6 mm × 2 mm) for an in vitro cell biological performance test. Mouse bone mesenchymal stem cells (mBMSCs, American Type Culture Collection, USA) at passage 6 were used in this experiment. A complete cell culture medium composed of high-glucose Dulbecco’s modified Eagle’s medium (Gibco, USA) with 10 vol % fetal bovine serum (Gibco, USA) was used for cell culture. The osteogenic inductive medium was further supplemented with 10 mM of sodium β-glycerophosphate, 0.1 μM of dexamethasone, and 50 mg/L of vitamin C. Before cell culture, the sterilized disks were placed into 48-well plates and immersed with the medium for 12 h.

For cell proliferation, 0.5 mL of complete culture medium containing 1 × 104 cells was allocated in a 48-well plate. Then, the cells were cultured in a cell incubator at 37 °C with 5% CO2. The medium was refreshed every 2 days. The Cell Counting Kit-8 (CCK-8, Dojindo, Japan) was used to assay the cell viability and proliferation at days 1, 3, and 7.

For cell viability, cells cultured on the disks for 3 days for live/dead stain were observed by a Live kit (Biotium, USA) under a fluorescence inverted microscope (40FL Axioskop; Zeiss, Germany). For cell attachment, the cells cultured on the disks for 1 day were taken out, washed with PBS twice, and fixed with 2.5% glutaraldehyde solution for 1 h, followed by dehydration with gradient ethanol, air-dried, sputtered with platinum, and observed by SEM. Cell cytoskeleton was observed with a confocal laser scanning microscope (TCS SP5, Leica Microsystem, Germany). Cells cultured for 1 day was stained by FITC-phalloidin (cell navigator F-actin labeling kit, AAT Bioquest, USA) and DAPI (Beyotime, China), respectively, to observe the cell actin cytoskeleton and nuclei.

Statistical Analysis

Quantitative data are presented as mean ± standard deviations, which were obtained from at least three replicates. A comparison between two means was made using the Student t-test with statistical significance set as p < 0.05.