Three-Dimensional Printing of Calcium Carbonate/Hydroxyapatite Scaffolds at Low Temperature for Bone Tissue Engineering

Abstract

Three-dimensional (3D) printing technology has been applied to fabricate bone tissue engineering scaffolds for a wide range of materials with precisely control over scaffold structures. Coral is a potential bone repair and bone replacement material. Due to the natural source limitation of coral, we developed a fabrication protocol for 3D printing of calcium carbonate (CaCO₃) nanoparticles for coral replacement in the application of bone tissue engineering. Up to 80% of CaCO₃ nanoparticles can be printed with high resolution using poly-l-lactide as a blender. The scaffolds were subjected to a controlled hydrothermal process for incomplete conversion of carbonate to phosphate to produce CaCO₃ scaffold covered by hydroxyapatite (HA) to modify the biocompatibility and degradation of CaCO₃/HA scaffolds. X-ray diffraction and Fourier transform infrared spectroscopy showed that HA was converted and attached to the surface of the scaffold, and the surface morphology and microstructure were studied using a scanning electron microscope. To confirm the bone regeneration performance of the scaffold, cell proliferation and osteogenic differentiation of MC3T3 cells on the scaffold were evaluated. In addition, in vivo experiments showed that CaCO₃/HA scaffolds can promote bone growth and repairing process and has high potential in bone tissue engineering. ClinicalTrials.gov ID: SH9H-2020-A603

Introduction

Three-dimensional (3D) printing, which can build 3D scaffolds effectively with computer-aided design as well as data source from micro-computed tomographies (CTs) with well-defined pore size and shape, has been extensively applied in tissue engineering.^1–5 Among various 3D printing techniques, low temperature 3D bioprinting (LT-3DP) has been considered to be the most suitable method for bone defects repairs in the future as it offers a number of essential merits for bone healing process⁶: (i) It can be used to generate 3D structures of the patient's specific defects with high biocompatible materials that are similar to the natural bone component and construction, with high concentration of inorganic compounds combined with essential component of bone extracellular matrix, such as collagens, which cannot be printed with melting extrusion method at high temperature⁷; (ii) bioactive agents, such as bone morphogenetic proteins (BMPs), can be incorporated into the 3D bioprinted scaffolds to improve the bone healing process^8–10; (iii) LT-3DP also provides the ability to print different cell types, which enables the functionality of the 3D bioprinted scaffolds, such as stem cell involvement.^11,12

One of the concerns with LT-3DP is that the scaffolds made by this method tend to be weak in mechanical strength. However, optimized structural design, materials composition, and postprocessing conditions could improve the mechanical properties of the 3D printed scaffold.¹³ In 3D printed structures, mechanical strength is controlled by porosity, the mesh shape also plays an important role. Reasonable post-treatment operations and material composition can further improve the mechanical properties of 3D printed scaffolds.

To date, the scaffolds made of tricalcium phosphate (TCP), calcium sulfate hemihydrate (CSH), nano-hydroxyapatite (nHA), and bioactive glass have been fabricated with LT-3DP.^14–18 In the work of Li et al., nHA and deproteinized bovine bone were dispersed in collagen to prepare the bioink for LT-3DP of bone scaffolds.¹⁴ The results showed that the expression levels of osteogenesis-related genes were significantly increased after 7 days. However, the scaffolds made of nHA and collagen were mechanically weak, and it is difficult to maintain the scaffold structure for new bone growth for a longer period of time due to the fast degradation of collagen. In the work of Chen et al., up to 60% of Mg substituted nHAs were mixed with collagen and gelatin to process porous scaffolds via LT-3DP.¹⁵ A high degree of Mg substitutes leads to a significant enhancement in alkaline phosphatase (ALP) activity and osteogenic-related gene expression level. However, the mechanical properties and degradation behavior of the scaffolds were not evaluated, which would be essential in bone regeneration process. Scaffolds made of 75 wt% of mesoporous calcium silicate and mesoporous bioactive glass in polycaprolactone (PCL) were fabricated by LT-3DP.¹⁸ The scaffolds produced are highly potential for drug-loaded therapeutic bone implants due to the mesoporous structure of the nanoparticles.

Calcium carbonate (CaCO₃) is a bioresorbable bone filling inorganic material, which has been reported to enhance osteoconductivity and provide calcium ions for bone reconstruction.¹⁹ Corals, which mainly consist of CaCO₃ and have a porous 3D structure, have been approved to be an effective bone replacement.^20,21 However, due to the natural source limitation, coral harvest has been banned in several countries. Recent report showed that 50 wt% of pearl powder, mainly consists of CaCO₃, was incorporated into CSH to fabricated bone scaffolds by LT-3DP.¹⁷ The results showed that the pearl powder improved the mechanical, degradation, and biocompatibility of the bone implants. Various forms of CaCO₃ nanoparticles have been deposited on scaffolds via in situ synthesis by mixing salts containing calcium and carbonate ions to enhance osteoblast cell adhesion and proliferation^22,23; however, due to its poor printability, 3D printed scaffolds of CaCO₃ have been rarely reported in the literature.

The chemical similarity of hydroxyapatite (HA) to the mineral phase of a native bone has led to an excellent osteoconductivity and biocompatibility.^24,25 Roohani-Esfahani et al. used the sacrificial template method to prepare porous biphasic calcium phosphate scaffolds after high temperature sintering and then immersed them in HA/PCL chloroform solution, so that the surface of the stents was coated with HA/PCL nanocomposites.²⁴ The results showed that nHA/PCL-coated scaffold has the highest compressive strength and the strongest osteoblast differentiation profile. However, there is a lack of convincing evidence of repairing bone tissue in vivo. The hydrothermal conversion of calcium carbonate to calcium phosphate was carried out with natural corals to tune its degradation rate and biocompatibility.^26,27 Bensaïd et al. obtained scaffolds by cutting natural corals and then converted the calcium carbonate at the surface of the scaffolds into HA by hydrothermal chemical exchange reaction with ammonium phosphate.²⁷ The results of in vivo experiments showed that significantly improved degradation rate remarkably improved the rate of bone healing on coral scaffolds coated with HA compared with natural coral scaffolds.

In this study, up to 80% of CaCO₃ nanoparticles were printed with high resolution at ambient temperature using poly-l-lactide (PLLA) as a blender. The as-printed scaffolds were subjected to a controlled hydrothermal process for incomplete conversion of carbonate to phosphate to produce CaCO₃ scaffolds covered by HA to increase its biocompatibility; meanwhile, the mechanical properties of the CaCO₃ scaffolds were also improved. Cell culture and animal tests showed that the CaCO₃ scaffolds could promote bone repair, and it is highly potential in bone tissue engineering.

Materials and Methods

Materials and scaffold fabrication

The inorganic component of CaCO₃ powders (DK Nano, Beijing, China) with a particle size of 20 nm in the hybrid inks was commercially available. PLLA (molecular weight = 15.3 kg/mol; Daigang, Jinan, China) was used as blender in the ink. To mimic the natural bone matrix composition,^28,29 20–30% of organic compounds and 70–80% of inorganic compounds were chosen for 3D printed scaffolds reported this study. 1.44 g of PLLA were dissolved in 5.4 mL of dichloromethane at 21°C ± 1°C for 4 h using a vortex mixer (VM-T1; Titansci, Shanghai, China). 5.76 g of CaCO₃ were added to the solution, together with 1.8 mL of absolute ethanol (water content <0.1%) to control the viscosity of the slurry and the evaporation kinetics and mixed in a vortex mixer for 10 min. The final quality of the ink was assessed in terms of printability, measured by extruding the ink without clogging with a syringe tip diameter of 410 μm.

Scaffolds of CaCO₃ were printed with a 3D printing machine (3D Bioplotter; EnvisionTec, German). The deposition process was carried out at ambient temperature of 21°C ± 1°C. The 3D periodic scaffolds consisted of a linear array of parallel rods in each layer aligned such that their orientation was orthogonal to the previous layer (as shown in Fig. 1A). The center-to-center rod spacing was 0.8, 1.0, and 1.2 mm (Fig. 1B), respectively. The high-magnification images of the printed structure are shown in Figure 1C. Once a layer was printed, the nozzle was raised by a fixed height, which depends on the tip diameter, and another layer was deposited. The diameter of the printing nozzle was 410 μm and the printing speed was set at 3–8 mm/s. The flow rate and the printing speed were adjusted to the nozzle diameter to print a continuous line with uniform thickness. Samples were printed onto glass slides and were easily removed after overnight at room temperature in air.

FIG. 1.

(A) Schematic diagram representing the preparation of CaCO₃ scaffolds; (B) CaCO₃ scaffolds with different line spacings; (C) the morphology and microstructure of CaCO₃ scaffolds. CaCO₃, calcium carbonate.

Hydrothermal process

The as-printed scaffolds were washed in deionized water and dried. After cleaning, the scaffold was soaked in 2 M phosphate solution and ammonium diacid phosphate (ADP, NH₄H₂PO₄; MACKLIN), and the CaCO₃/ADP mole ratio was maintained constant at 1:1.2. The following Reaction (1) was completed for 12 h, and the reacting temperature was varied from 80°C to 120°C. According to the Reaction (1), the calcium carbonate particles on the surface of the scaffolds undergo hydrothermal reaction with dihydrogen ammonium phosphate to generate HA particles.³⁰ The CaCO₃ scaffolds were coated with HA, named CaCO₃/HA scaffolds.

$$\begin{array}{cc}10CaC{O}_3& +6\left(N{H}_4\right)H_2\left(P{O}_4\right)+2H_{2}O\\ & \to C{a}_{10}{\left(P{O}_4\right)}_6{\left(OH\right)}_2+3{\left(N{H}_4\right)}_{2}C{O}_3\\ & +7H_{2}O+7C{O}_2\end{array}$$ (1)

Characterization of CaCO₃/HA scaffolds

Scanning electron microscopy

The morphology of CaCO₃/HA scaffold was observed by scanning electron microscope (SEM) (S-3400N; Hitachi Ltd., Japan) with the field emission gun operated at accelerating voltage of 2.5–10kV. Dry samples were placed on cuprum stubs and sputter-coated with a layer (3 nm thick) of aurum.

Energy-dispersive X-ray spectroscopy

Energy-dispersive X-ray spectroscopy (EDS) measurement was carried out on a Hitachi microscope S-3400N (Hitachi Ltd.) at a low voltage. EDS mapping technique generates a two-dimensional image using the abundance of an element as the intensity of the image. By mapping a few elements on the same area of a material, distribution and abundance of the elements in the materials can be compared. The CaCO₃/HA scaffolds were inserted into the SEM after coated with aurum and were examined at an acceleration voltage of 10 kV. Distribution maps of Ca, P, C, and O were produced.

X-ray diffraction

X-ray diffraction (XRD) was used to determine the nature and crystal size of the HAs on an X-ray powder diffractometer (max2550VB; Rigaku Ltd., Japan) operated at 40 kV and 100 mA. Fine particles of CaCO₃/HA scaffolds were affixed onto a piece of clean silicon wafer with silicone grease. Data were collected between 5° and 75° 2θ at a scan rate of 0.002° 2θ/s.

Fourier-transform infrared spectroscopy

Fourier-transform infrared (FTIR) spectra were obtained in transmission mode using an FTIR spectrometer (6700; Thermo Nicolet Corporation) from disks containing potassium bromide and poly-lactic-co-glycolic acid. Twenty-five scans over the spectral range of 400–4000 cm⁻¹ were obtained at a resolution of 2 cm⁻¹ with the background scan subtracted.

Mechanical testing

Compression tests of the untreated and degraded scaffolds were performed by means of a cautious adaptation of the ISO 13314 standard that refers to the mechanical testing of porous and cellular metals. Compression tests were performed with a mechanical tester (HY-0350 Shanghai Hengyi Testing Instruments Co., Ltd.) at a compression rate of 1 mm/min with 3D printed samples in a cylinder shape (10 mm in diameter and 5 mm in height corresponding to 16 layers) using a needle tip of 410 μm. In the stress–strain curves, the maximum compressive strength was obtained.

In vitro study

In vitro MC3T3 proliferation and differentiation assay

CaCO₃/HA scaffolds (6 × 6 × 1 mm³) were immersed in 75% ethanol for 2 h and then dipped three times with phosphate-buffered saline to remove residual ethanol. MC3T3 cells were cultured in α-MEM (α-minimum essential medium) with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin in a commercial incubator (160i; Thermo Fisher Scientific) at 37°C with 5% CO₂. The cells were cultured at a density of 10⁵ cells per well in 48-well plates, referred to as control groups. After sterilization, the scaffolds were placed into 48-well culture plates. Thereafter, 10⁵ MC3T3 cells suspended in 50 μL α-MEM medium were seeded on CaCO₃/HA scaffolds (6 × 6 × 1 mm³) to form CaCO₃/HA-MC3T3 constructs, referred to as experimental groups. Nonadherent cells were washed away 1 day after seeding. The culture medium was replaced with fresh medium every 2 days.

All culture media contained penicillin and streptomycin (100 U/mL; Hyclone Company); differentiation media also contained l-ascorbic acid (50 μM; Sigma–Aldrich, China) and sodium β-glycerophosphate (10 mM; Sigma–Aldrich). And the scaffolds were cultured in α-MEM with 10% FBS and 1% penicillin–streptomycin in an incubator at 37°C with 5% CO₂. Cell Counting Kit-8 (Dojindo, China) was used to test cell proliferation at 1, 4, and 7 days.

The activity of ALP was measured on days 1 and 14.³¹ ALP assay kit (Beyotime, China) was used to test cell differentiation effects. The cellular ALP activity was normalized against the total protein concentration measured by a BCA protein assay kit and expressed as U/mg protein. There are three parallel samples for each test.

F-actin and DAPI assay

CaCO₃/HA scaffolds were coated with a layer of collagen I (0.012 mg/mL) (rat tail collagen; Corning) for 24 h. After sterilization, the scaffolds were placed into 48-well culture plates. Thereafter, 10⁵ MC3T3 cells suspended in 50 μL α-MEM medium were seeded on CaCO₃/HA scaffolds (6 × 6 × 1 mm³) to form CaCO₃/HA-MC3T3 constructs. F-actin (Yeasen, China) and DAPI (Biofroxx, China) were used to assess cell morphology. The samples were submerged in dye solution composed of F-actin (200 nM) and DAPI (10 μg/mL) in a 1% FBS solution, followed by incubation at 37°C for 30 min. A fluorescent microscope (Nikon Eclipse Ti; Nikon, China) was utilized to capture the stained samples. The F-actin appeared in green and the DAPI in blue.

In vivo test

Surgical implantation

Animal tests in this study were approved by the Animal Care and Experiment Committee of the Ninth People's Hospital. A12 adult male rabbits were used for CaCO₃/HA scaffold bone regeneration experiments in vivo, and a 15 mm defect of rabbit radius was established as described previously.³² Briefly, all rabbits were anesthetized with 3% pentobarbital. Then, hair from forelimbs of each animal is cleared using commercial hair removal cream. A 4.5 cm longitudinal incision was made in the skin along the radius. Dissection of the muscle exposed the radius. The periosteum was removed and a 15 mm defect was created in the middle of the radius by a burr with a diamond blade under continuous saline cooling. The samples were inserted in the defect without external fixation. The 12 rabbits were all treated for two radius defects (right and left). All animals were injected intraperitoneally with gentamicin (2 mL) for 3 days to prevent infection. In this study, the CaCO₃/HA scaffold (15 × 3 × 3 mm³) was implanted on one side for repair, whereas the other side was used as a blank control to observe the repair effect.

Sequential fluorescent labeling

Polychrome sequential fluorescent labeling was performed postoperatively to assess the time course of bone formation and mineralization as reported previously.³³ Fluorochrome labeling of the bones was performed by intraperitoneal injections of 25 mg/kg tetracycline hydrochloride (13 days before sacrifice; Sigma–Aldrich) and 5 mg/kg calcein (3 days before sacrifice; Sigma–Aldrich). The sections were examined under a confocal laser scanning microscope (A1R; Nikon Corp., Tokyo, Japan). Animal was killed by an intraperitoneally injected overdose of sodium pentobarbital at 4, 8, and 12 weeks, respectively. Results were quantified by the ImageJ software.

Radiography

Standardized anteroposterior and lateral radiographs were taken 4 weeks after surgery to monitor the placement of the graft and the bony integration. Images were obtained by Mobilett X-ray machine (Siemens, Munich, Germany).

Statistical analysis

All data are presented as mean ± standard error. For each variable, effects across treatment groups were compared by one-way analysis of variance. If the overall difference was significant, multiple comparisons were performed between groups using Tukey's post hoc test. Differences were considered significant at a probability p < 0.05 on a two tailed test.

Results and Discussion

Characterization of CaCO₃/HA scaffold

SEM and EDS results of CaCO₃ scaffold and CaCO₃/HA scaffold are shown in Figure 2A. Irregular blocks of particles were bonded together on the surface in the CaCO₃ scaffolds. The particles were calcium carbonate crystals. However, the particles are agglomerated together in the CaCO₃/HA scaffolds. This is because the HA crystals have a high surface energy and therefore tend to agglomerate to reduce the surface energy. The size of the microparticles is about 1–3 μm. The image at high magnification shows that the microparticles are petal-like, it can also be observed that the calcium carbonate crystals originally distributed on the surface of the material have been replaced by needles or rods HA crystals after hydrothermal reaction.

FIG. 2.

(A) SEM and EDS results of CaCO₃ scaffold and CaCO₃/HA scaffold. (B) SEM showing the monoclinic crystal structure on the CaCO₃/HA scaffold after hydrothermal reaction at 100°C. Raw represents the CaCO₃ scaffold. EDS, energy dispersive X-ray spectroscopy; HA, hydroxyapatite; SEM, scanning electron microscope.

The peaks in EDS results in the CaCO₃ scaffolds are Ca and O from calcium carbonate. In the CaCO₃/HA scaffolds, after the hydrothermal reaction at different temperatures, the corresponding peaks of the P element appear in the corresponding energy spectrum, which can be further proved after the hydrothermal reaction at 80°C, 100°C, and 120°C, the calcium carbonate particles on the surface of the CaCO₃ scaffold were replaced with HA crystals.

The crystal structure of calcite-type calcium carbonate is trigonal, and the most commonly mentioned crystal structure of HA is hexagonal, and HA can also exist in a monoclinic state. The monoclinic crystal structure of HA is more ordered, and the thermodynamic state is more stable. The significant difference between these two crystal forms of HA lies in the orientation of the hydroxyl groups. Although the crystal structure difference between the monoclinic and hexagonal HA is small, it has a great influence on the physical and chemical properties of HA. After hydrothermal reaction, most of the replaced HA crystals are needle-like or rod-shaped. In the hydrothermal reaction at the 100°C, monoclinic crystal structure can be found as shown in Figure 2B in the CaCO₃/HA scaffolds. By using the Jade 6.0 to calculate the lattice parameters,³⁴ it is found that α, β, and γ are 90°, 90°, and 119.9°, respectively, which conform to 90°, 90°, and 120° of monoclinic crystals. It also coincides with monoclinic HA (as shown in Table 1).

Table 1.

Hydroxyapatite Lattice Parameter

	a/Å	b/Å	c/Å	γ
Hexagon HAP	9.432	9.432	6.881	120°
Monoclinic HAP	9.421	2a	6.881	120°
Surface of the CaCO3/HA scaffold	9.407	18.87	6.869	119.9°

CaCO₃, calcium carbonate; HA, hydroxyapatite; HAP, hydroxylapatite.

XRD spectra of CaCO₃ scaffold and CaCO₃/HA scaffold are shown in Figure 3A. Strong diffraction peaks were observed at 23.3°, 29.3°, and 33.3°, and this confirms that they are calcite-type calcium carbonate according to the standard spectrum (JCPDS No. 05-0586, calcite form of calcium carbonate). The three sets of peaks existed in the three groups of scaffolds after the hydrothermal reaction, indicating that the CaCO₃/HA scaffold obtained by the hydrothermal reaction under the above three reaction conditions still has residual calcium carbonate. The strong diffraction peaks at 25.6° and 32° were the diffraction peak of monoclinic HA according to the data of JCPDS No. 76-0694. The peaks existed in the three groups of scaffolds after the completion of the hydrothermal reaction, indicating that the HA crystal can be formed after hydrothermal reaction in three groups of reaction conditions. The replacement amount of HA increases as the temperature of the reaction increases.

FIG. 3.

(A) XRD spectra of CaCO₃ scaffold and CaCO₃/HA scaffold. (B) FTIR spectra of CaCO₃, PLLA, and CaCO₃ scaffolds reacting at different temperatures. Raw represents the CaCO₃ scaffold. FTIR, Fourier-transform infrared; PLLA, poly-l-lactide; XRD, X-ray diffraction.

FTIR spectra of CaCO₃ scaffold and CaCO₃/HA scaffold are shown in Figure 3B. The calcium carbonate used in the experiment is calcite-type calcium carbonate with the corresponding ν4 absorption peak of CO₃²⁻ at 712.7 cm⁻¹, the ν2 absorption peak of CO₃²⁻ at 876.2 cm⁻¹, and the ν3 absorption peak of CO₃²⁻ at 1424.5 cm⁻¹. At 1758 cm⁻¹ is the C = O carbonyl absorption peak of PLLA, and the fingerprint region between 1214.2 and 1043.9 cm⁻¹ is the absorption peak of C-O-C in PLLA. The ν4 absorption peak of CO₃²⁻ at 712.7 cm⁻¹ and the ν2 absorption peak of CO₃²⁻ at 876.2 cm⁻¹ decayed after the hydrothermal reaction. At the same time, absorption peaks appeared at 564.4 and 603.5 cm⁻¹ (the ν2 absorption peak of PO₄³⁻). As the reaction temperature increases, the absorption peak of CO₃²⁻ is further attenuated, and the absorption peak of PO₄³⁻ continues to increase, indicating that the replacement of HA is increased with the increase of reaction temperature. The C = O carbonyl absorption peak of PLLA at 1758 cm⁻¹, and the absorption peak of C-O-C stretching vibration in PLLA from 1214.2 to 1043.9 cm⁻¹, shows significant attenuation in the CaCO₃/HA scaffold group, which was reacted at 120°C. The results indicate that the PLLA in the scaffold has undergone a significant hydrolysis under the reaction condition of 120°C.

The stress–strain curves of CaCO₃ scaffold and CaCO₃/HA scaffold are shown in Figure 4. It can be seen that samples break at a certain range of deformation. This is because the scaffold has pores in the compression direction, and the material is squeezed to fill the gap at the beginning of the compression test. The CaCO₃ scaffold exhibits characteristics close to that of the ductile material in the compression test, and it does not break at the final deformation rate of 95% but is compacted into a flat shape. Compared with the CaCO₃ scaffold, the strength of the CaCO₃/HA scaffold after hydrothermal reaction at 80°C has increased dramatically, the characteristics of the brittle material are exhibited in the compression test, and the compressive strength reaches 16.6 ± 1.4 MPa. However, as the reaction temperature increases, the compressive strength of the scaffold gradually decreases. The compressive strength of the hydrothermal reaction at 120°C is close to zero. The reason could be that as the reaction temperature increases, PLLA begins to hydrolyze, this process weakens the bonding effect and weakens the mechanical properties of the scaffolds, this is also indicated in the FTIR results as shown in Figure 2D. In the following in vitro and in vivo tests, the scaffolds reacted at 80°C were chosen due to its good mechanical strength.

FIG. 4.

Stress–strain results of CaCO₃ scaffold and CaCO₃/HA scaffold. Raw represents the CaCO₃ scaffold.

The mechanical strength of cancellous bone in human body is in the range of 2–12 MPa; however, the cortical bone is in the range of 100–230 MPa.³⁵ The mechanical property of the scaffold produced in this study is 16.6 ± 1.4 MPa; therefore, the scaffold could be used to replace nonload-bearing bones. In the previous studies,^36–38 it was found that the mechanical properties of the scaffolds produced with LT-3DP could be improved by optimizing the internal structure of scaffolds, components of building materials, in terms of the ratio of organic and inorganic compounds, nanomaterials reinforcement, and so on. In the work of Cheng et al.,³⁶ the compressive strength of pure PLA scaffold was more than 60 MPa by varying the internal structures of scaffold. Cabral et al.³⁷ used chitosan and gelatin as blender to make a scaffold containing 80% (w/w) of TCP, and 0.27% (w/w) of nano-graphene oxide, the mechanical properties of the scaffold were 22–32 MPa. High ratio of inorganic substances may reduce the mechanical strength of scaffold. In the work of Hung et al., it was found that the compressive strength of PCL scaffold containing 30% (w/w) bone particles was 30 MPa, whereas that of PCL scaffold containing 70% (w/w) bone particles decreased to 10 MPa.³⁸ In this study, we aim to produce scaffolds with highest amount of inorganic compounds and good precision, scaffolds of up to 80% of nano-CaCO₃ were produced with good precision. The ratio of PLLA and CaCO₃ could be adjusted if higher mechanical strength is required.

In vitro experiment

F-actin/DAPI fluorescence images of CaCO₃/HA scaffold are shown in Figure 5A. On the first day, MC3T3 cells spread on the CaCO₃/HA scaffold. On the seventh day, the number of cells on the scaffold was significantly increased compared with the first day, and the cell morphology changed significantly from the original fusiform to the spherical or cylindrical shape. The reason could be that the osteoblasts differentiate under the stimulation of the differentiation medium and become bone cells. Proliferation test result of MC3T3 is shown in Figure 5B. The number of cells showed an exponential increase in the cell culture experiments. Comparing the first and fourth days, the optical density (OD) values of the experimental group and the control group were close. The OD value of the experimental group on the seventh day was significantly higher than that of the control group, indicating that CaCO₃/HA scaffolds have a better biocompatibility compared with the control group. The nano-HA particles might also count for the high number of cells proliferated compared with the control group. This also confirms that the samples after the hydrothermal reaction were not biologically toxic to MC3T3 cells.

FIG. 5.

Data show the cell attachment to the CaCO₃/HA scaffold. (A) F-actin/DAPI fluorescence image of CaCO₃/HA scaffold, the scale bar is 200 μm. (B) Proliferation result of MC3T3. (C) Differentiation result of MC3T3 (n = 3; * indicates significant difference compared with the control group, p < 0.05).

Differentiation result of MC3T3 is shown in Figure 5C. After 14 days of induced differentiation culture, the expression of ALP activity in the experimental group and the control group was positive, which indicated that MC3T3 cells were successfully induced to differentiate into bone cells under the condition of differentiation medium. The specific activity of ALP in the experimental group was higher than that in the control group, indicating that the osteogenic differentiation effect of the experimental group was better than that of the control group. It is because after the hydrothermal reaction, a layer of HA particles is formed on the surface of the scaffold, which has acted as the conducive stimuli to the osteogenic differentiation of MC3T3 cells.³⁹

In vivo experiment

To study dynamic changes in bone remodeling, in the in vivo experiments, the rabbits were injected with fluorochrome markers to label the formation and mineralization of new bone around the implant (as shown in Fig. 6), and undecalcified sections of the constructs were prepared and observed under fluorescence microscopy. The yellow and green fluorescent lines were used to label tetracycline hydrochloride and calcein, respectively, to calculate new bone deposition speed during different periods of time. As shown in Figure 6, the new bone formation and mineralization were observed at 4, 8, and 12 weeks after the operation. At week 4 and 12, the percentage of CA labeling and TE labeling was 0.452% ± 0.103% and 2.16% ± 0.183%, 1.61% ± 0.187% and 4.06% ± 0.92%, respectively. There were significant differences between the week 4 group and the week 12 group (p < 0.05). The laser confocal image (Fig. 6A) shows that the yellow fluorescent lines were deposited on the outer side and the green fluorescent lines were deposited on the inner side of mineralization front, this indicates that new bone formation was from outside to inside along scaffolds. It can be seen that two colors overlap to some degree, indicating that bone mineral deposition occurs simultaneously on both sides of the mineralization front. The results showed that CaCO₃/HA scaffolds exhibit good bone rehealing ability, biocompatibility, and robust osteogenic activity.

FIG. 6.

(A) Bone fluorescent markers of rabbit models were observed at different times: tetracycline (yellow), calcein (green). (B) Percentages of tetracycline and calcein staining in each group assessed after implantation by histomorphometric analysis. (n = 3, * indicates significant differences, p < 0.05).

After 4 weeks, new bone had formed and the gaps were narrowed slightly in the control group (Fig. 7A). However, in the CaCO₃/HA scaffolds, the boundary became illegible, suggesting the occurrence of mineralization and increased density of bone (Fig. 7B). The disappearance of the boundary of material and tissue indicated that the density of newly formed bone was nearly the same as that of host bone. The in vivo study showed that a rapid process of bone repair was observed in the CaCO₃/HA scaffold, the mineralization started at the bilateral ends of the implant near to the neighboring bone tissues, progressively reached toward the center of the implant from. The X-ray microradiographic analysis confirms the results of the sequential fluorescent labeling study that the CaCO₃/HA scaffold are biocompatible and osteoconductive to the host bone and CaCO₃/HA scaffolds can improve the speed of the bone healing process.

FIG. 7.

The X-ray images of radius defect site after 4 weeks. (A, C) Control. (B, D) CaCO₃/HA scaffold. (C, D) Higher magnification X-ray images of (A, B).

Conclusions

In this study, LT-3DP technology combined with the hydrothermal reaction was used to successfully fabricate a HA-coated coral-like bone scaffold with high precision. Up to 80% of CaCO₃ nanoparticles were printed into 3D scaffolds with defined structures. The hydrothermal reaction was carried out at 80°C, 100°C, and 120°C. The mechanical strength and biocompatibility were significantly improved at the reaction temperature of 80°C. Due to the conversion of HA nanoparticles on the surface of the scaffold, 3D printed CaCO₃/HA scaffold showed a good biocompatibility and osteoinduction both in vivo and in vitro. Therefore, the CaCO₃/HA scaffolds reported in this study are promising in bone defect repair and implants. This LT-3DP also offers the ability to fabricate antibacterial drug- and bioactive agent-loaded scaffolds for sustaining release purposes, which can further improve the biocompatibility and promote bone formation, this would be our future work.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This work was supported by Key R&D plan of Hainan Province (No. ZDYF2017093) and the Opening project of Shanghai Key Laboratory of Orthopedic Implant (KFKT2019001).