Functionalized 3D‐Printed PCL/β‐TCP Scaffolds with ZnO for Enhanced Bone Healing

ABSTRACT

Alveolar ridge deficiency remains a significant clinical challenge in dental implantology, often necessitating bone augmentation procedures. While autogenous bone grafting is considered the gold standard, its limitations, including donor site complications and limited availability, have prompted the development of alternative bone graft materials. In this study, we fabricated a novel polycaprolactone (PCL)/β‐tricalcium phosphate (β‐TCP)@zinc oxide (ZnO) composite scaffold via direct ink writing (DIW) three‐dimensional (3D) printing. The scaffold integrates a biomimetic porous architecture with a controlled ion‐release strategy to enhance osteogenic performance. The characterization of the material confirmed excellent chemical stability, hydrophilicity, and tunable degradation behavior. The incorporation of ZnO facilitated the sustained release of bioactive ions (Ca²⁺, PO₄ ³⁻, and Zn²⁺), significantly improving the scaffold's bioactivity. In vitro assessments revealed that the scaffold with 2% ZnO (PTZ‐2) exhibited optimal cytocompatibility and osteogenic differentiation capacity, promoting alkaline phosphatase activity and mineralized nodule formation in MC3T3‐E1 cells. In a rat cranial defect model, the PTZ‐2 scaffold achieved approximately 95% new bone area after 12 weeks. Histological analysis further confirmed the presence of continuous, highly mineralized bone matrix with excellent osseointegration capability. These findings collectively indicate that the 3D‐printed PCL/β‐TCP@ZnO composite scaffold has great potential for clinical translation in bone defect regeneration.

Alveolar ridge defects require bone augmentation therapy, but autologous bone grafting has significant limitations. We prepared 3D‐printed PCL/β‐TCP@ZnO scaffolds using direct ink writing technology, in which ZnO triggers the sustained release of bioactive ions. The PTZ‐2 scaffold, containing 2% ZnO, exhibited the best cellular compatibility and osteogenic activity in vitro, offering hope for clinical bone regeneration applications.

1. Introduction

Alveolar ridge deficiencies, frequently resulting from tooth loss, trauma, or periodontal disease, pose a significant hurdle for successful dental implant placement, as inadequate bone volume compromises primary stability and optimal prosthetic positioning [1]. Current clinical management primarily relies on bone augmentation procedures. Autogenous bone grafts are considered the gold standard owing to their osteoconductive, osteoinductive, and osteogenic properties; however, their application is limited by donor‐site morbidity, limited availability, and the risk of complications [2]. Alternative approaches such as allogeneic grafts, guided bone regeneration (GBR), and distraction osteogenesis have been developed, yet each presents drawbacks including immune rejection, inconsistent osteogenic outcomes, and technical sensitivity [3]. These limitations underscore the urgent need for synthetic bone graft substitutes that can reliably promote bone regeneration while minimizing patient morbidity [4].

Bone tissue engineering has emerged as a promising strategy to meet this demand, focusing on the design and fabrication of bioactive, porous scaffolds that can mimic the structure and function of natural bone [5, 6]. Among various fabrication techniques [7, 8]. 3D printing, especially direct ink writing (DIW), has attracted widespread attention for its outstanding capabilities in precise structural control, patient‐specific customization, and multi‐material integration [9, 10]. This technology enables the construction of scaffolds with tailored porosity, pore interconnectivity, and mechanical properties, thereby providing an optimal microenvironment for cell infiltration, vascularization, and new bone formation [11, 12].

Polycaprolactone (PCL), a synthetic polyester with excellent biocompatibility, biodegradability, and processability [13], is widely used as a scaffold matrix in bone tissue engineering. However, its inherent bioinertness and suboptimal mechanical strength for load‐bearing applications often limit its standalone use. To overcome these shortcomings, PCL is frequently combined with bioactive ceramics. For example, a study used the sol‐gel method to combine polycaprolactone (PCL) with montmorillonite (MNT) and Akermanite (AK) powders, and then constructed composite scaffolds through 3D printing technology. Researchers found that by adjusting the incorporation ratio of AK and MNT, the mechanical properties and bioactivity of the material improved as the ratio increased, providing a promising alternative for bone repair. Among bioceramic materials, β‐tricalcium phosphate (β‐TCP) is widely favored because its chemical structure closely matches the mineral phase of bone and its osteoconductivity is well documented [14, 15]. Previous studies have demonstrated that PCL/β‐TCP composite scaffolds can effectively support osteoblast attachment, proliferation, and differentiation [16, 17]. Lee et al. coated 3D‐printed PCL/TCP scaffolds with bone‐demineralized decellularized extracellular matrix (bdECM) and conducted experiments in a canine mandibular defect model, observing significantly improved bone regeneration [18]. Similarly, Kim et al. applied amine plasma polymerization to 3D‐printed PCL/β‐TCP scaffolds, which enhanced surface characteristics and stimulated osteogenic differentiation of pre‐osteoblasts in vitro [14, 19]. However, the osteoinductivity of such composite materials is limited, which may restrict their effectiveness in critical‐sized or poorly vascularized defects.

In recent years, zinc oxide (ZnO) nanoparticles have garnered widespread attention due to their excellent biocompatibility, antibacterial properties, and osteogenic promotion effect [20]. Zinc ions (Zn²⁺) have been shown to upregulate key osteogenic markers such as Runx2, alkaline phosphatase (ALP), and osteocalcin (OCN), thereby accelerating the differentiation of osteoprogenitor cells and mineral deposition [21]. Moreover, Zn²⁺ can modulate local immune responses and suppress osteoclast activity, further favoring anabolic bone remodeling. Despite these promising attributes, the strategic integration of ZnO into 3D‐printed PCL/β‐TCP composite scaffolds–with precise control over ion release kinetics and its consequent impact on in vitro and in vivo bone regeneration–remains underexplored.

To overcome the above limitations, the present study aimed to develop a novel scaffold system by incorporating ZnO nanoparticles into a PCL/β‐TCP matrix using DIW 3D printing. The resulting PCL/β‐TCP@ZnO composite scaffolds were designed to provide sustained release of Ca²⁺, PO₄ ³⁻, and Zn²⁺ ions. The physicochemical characteristics, ion release profiles, in vitro cytocompatibility and osteogenic differentiation, as well as in vivo bone regeneration in a rat cranial defect model were systematically investigated. The overall fabrication process and evaluation workflow are illustrated in Scheme 1.

SCHEME 1.

Schematic of the preparation process and performance assessment for 3D PCL/β‐TCP@ZnO scaffold.

2. Materials and Methods

2.1. Fabrication of 3D‐Printed Composite Scaffolds

β‐TCP powder (Sigma‐Aldrich) was blended with ZnO nanoparticles (Aladdin) at mass fractions of 0.1, 0.2, 0.4, and 0.8 wt%. The mixed powders were placed into an agate jar together with 300 g of agate grinding media (diameters: 1, 3, and 5 mm). Wet ball milling was carried out in 150 mL of ethanol at 300 rpm for 6 h. After milling, the jar was dried in an electric forced‐air drying oven at 60°C for 24 h, after which the grinding media were removed, and the resulting homogeneous powder was further dried for 48 h. The dried powder was collected and sintered in a tube furnace at 1250°C for 2 h. After cooling, ZnO‐doped β‐TCP composite powders with different ZnO mass fractions were obtained and designated as TZ‐1, TZ‐2, TZ‐4, and TZ‐8, respectively. The composite powders were then mixed with PCL (Mn ≈ 80 kDa, Aladdin) at a mass ratio of 2:8 in dichloromethane and stirred for 12 h to prepare a homogeneous printing ink (labelled PT for 0% ZnO, and PTZ‐1 to PTZ‐8 for the respective ZnO‐doped groups). Scaffolds were fabricated using a direct ink writing (DIW) 3D printing system (Bio‐Architect Pro). A cylindrical model (Ø  =  5 mm, h  =  2 mm) was designed to simulate a critical‐sized defect. Printing parameters were as follows: nozzle diameter  =  500 µm, layer thickness  =  220 µm, line spacing  =  2.0 mm, printing speed  =  2  mm·s⁻¹, and pneumatic pressure  =  2.5 kPa. After printing, scaffolds were dried at room temperature for 48 h to fully evaporate the solvent (Figure S1).

2.2. Material Characterization

The morphology and elemental distribution of the powders and scaffolds were examined using scanning electron microscopy (SEM, JEOL JSM‐6701F) coupled with energy‐dispersive X‐ray spectroscopy (EDS). Phase composition was analyzed by X‐ray diffraction (XRD, Rigaku D/max‐2400) with Cu‐Kα radiation (λ  =  1.5406 Å). Chemical bonds were characterized via Fourier‐transform infrared spectroscopy (FTIR, Bruker IFS66V/S). Surface chemistry was probed by X‐ray photoelectron spectroscopy (XPS, Kratos AXIS Ultra DLD). The rheological properties of the PTZ inks were evaluated using a rotational rheometer (MCR301, Anton Paar) equipped with a parallel‑plate geometry (plate diameter: 25 mm, gap: 1 mm). All measurements were performed at 25°C. Steady‑state viscosity curves were obtained by logarithmically varying the shear rate from 0.1 to 100 s⁻¹ to characterize shear‑thinning behavior. The yield stress was determined through a shear stress sweep (1–1000 Pa), with the yield point defined as the stress at which the viscosity decreased by more than 90% from its low‐stress plateau. Prior to oscillatory tests, an amplitude sweep (strain  =  0.01%–100%) at a constant frequency of 1 Hz was conducted to identify the linear viscoelastic (LVE) region. Subsequently, frequency sweeps were performed from 0.1 to 100 rad s⁻¹ at a strain within the LVE range (0.5%), and the storage modulus (G′) and loss modulus (G″) were recorded to assess the viscoelastic response and shape‑retention ability of the inks. Surface wettability was determined by water contact angle measurements (DSA100, Krüss). In vitro degradation was evaluated by incubating scaffolds in phosphate‐buffered saline (PBS, pH =  7.4) containing 1 U mL⁻¹ lipase at 37°C; samples were collected weekly, dried, and weighed to calculate mass loss.

2.3. In Vitro Cell Culture and Osteogenic Assays

Mouse pre‐osteoblastic MC3T3‐E1 cells (ATCC) were cultured in Dulbecco's Modified Eagle Medium (DMEM,Gibco) supplemented with 10% fetal bovine serum (FBS, Solarbio) and 1% penicillin‐streptomycin (Beyotime). Scaffolds were sterilized with 75% ethanol and UV irradiation before cell seeding. Cell viability and proliferation were evaluated using the CCK‐8 (Beyotime) assay after 24, 48, and 96 h of culture. Live cells were stained with Calcein‐AM and visualized by fluorescence microscopy. To assess osteogenic differentiation, cells were induced in osteogenic medium (DMEM containing 10 mM β‐glycerophosphate, 50 µg mL^{− 1} ascorbic acid, and 100 nM dexamethasone). Alkaline phosphatase (ALP, Beyotime) activity was measured and stained after 3, 7, and 14 days, choose three random non‐overlapping visual fields per group were imported into ImageJ software to quantify the mean fluorescence intensity of ALP. Mineralized nodule formation was evaluated by Alizarin Red S (ARS, Beyotime) staining at 7, 14, and 21 days, followed by quantitative elution with 10% cetylpyridinium chloride.

2.4. In Vivo Bone Regeneration Study

All animal procedures were approved by the Institutional Animal Care and Use Committee of Lanzhou University (Approval No. LZUK‐2022‐092). Twelve male Sprague‐Dawley rats (10 weeks old) were randomly divided into two groups: 3D PT (PCL/β‐TCP) and 3D PTZ‐2 (PCL/β‐TCP with 0.2 wt% ZnO). Under general anesthesia, a full‐thickness critical‐sized defect (Ø  =  5 mm) was created in the calvarium using a trephine bur. The scaffold was press‐fit into the defect, and the wound was closed in layers. At 8 and 12 weeks post‐implantation, new bone formation was monitored by PET‐CT scanning (InliView‐3000). Bone volume/total volume (BV/TV) and bone mineral density (BMD) were quantified using Mimics and ImageJ software. After sacrifice, harvested calvaria were fixed, decalcified, sectioned, and stained with hematoxylin & eosin (H&E) and Masson's trichrome for histological evaluation.

2.5. Statistical Analysis

All quantitative data are presented as mean ± standard deviation (SD). Statistical analyses were performed using GraphPad Prism 8 and SPSS 25. One‐way ANOVA with Tukey's post‐hoc test was used for multiple group comparisons. Student's t‐test was applied for comparisons between two groups. Differences were considered statistically significant at P < 0.05.

3. Results and Discussion

3.1. Phase and Structural Characterization of TZ Composite Powders

A series of β‐TCP@ZnO (designated as TZ) composite powders with different ZnO doping concentrations (0.1, 0.2, 0.4, and 0.8 wt%) were successfully synthesized via high‐temperature solid‐state sintering followed by mechanical ball‐milling (Figure 1a). SEM analysis demonstrated that both the pure β‐TCP, TZ‐1, TZ‐2, TZ‐4, and TZ‐8 powders exhibited good dispersibility without severe agglomeration after sintering at 1250°C (Figure 1b–f). Pure β‐TCP particles possess large sizes with smooth, compact surfaces and intact grain boundaries. After ZnO incorporation, the particles of TZ‐1, TZ‐2, TZ‐4, and TZ‐8 exhibited rough edges with slight surface etching and fine debris, accompanied by moderate overall surface roughness. Their particle sizes were mainly distributed in the range of 10–30 µm. These morphological features demonstrate the uniform incorporation of the ZnO phase within the β‐tricalcium phosphate matrix.

FIGURE 1.

Synthesis and characterization of β‐TCP@ZnO composite powder (a) Schematic illustration of the preparation process of β‐TCP@ZnO composite scaffolds. (b) SEM images of pure β‐TCP powder (c) SEM images of TZ‐1 powder after sintering at 1250°C. (d) SEM images of TZ‐2 powder after sintering at 1250°C. (e) SEM images of TZ‐4 powder after sintering at 1250°C. (f) SEM images of TZ‐8 powder after sintering at 1250°C. (g) XRD patterns of β‐TCP and TZ‐x (x = 1, 2, 4, 8) powders sintered at 1250°C. Characteristic peaks of β‐TCP and α‐TCP are marked. (h1–h4) EDS elemental mapping of TZ‐8 powder showing the distribution of Ca, P, O, and Zn.

The crystalline phases of TZ with different ZnO doping concentrations were analyzed by XRD (Figure 1g). All groups displayed characteristic diffraction peaks of β‐TCP (PDF#09‐0169) at 27.8°, 31.1°, and 34.4°. With increasing ZnO doping, the 30.7°, diffraction peak of the α‐TCP impurity phase (PDF#09‐0348), decreased markedly from 8929 in TZ‐1 to 4023 in TZ‐8, representing a reduction of approximately 54%. This suppression of the β→α phase transformation at high sintering temperatures indicates effective incorporation of Zn²⁺ into the β‐TCP lattice by replacing the Ca²⁺ sites, thereby stabilizing the β‐TCP phase. This observation is consistent with previous reports on β‐TCP stabilization by divalent cation doping.

To confirm the spatial distribution of elements in the TZ‐8 powder, EDS mapping was performed (Figure 1h), which revealed homogeneous distributions of calcium (Ca), phosphorus (P), and oxygen (O). Notably, the zinc (Zn) signals appeared as dispersed punctate spots without forming localized aggregates, further confirming the atomic‐level uniform doping of ZnO in β‐TCP. This homogeneous doping is crucial for achieving controlled and sustained ion release in subsequent biomedical applications and provides a materials‐design foundation for developing advanced functional bone‐graft substitutes.

3.2. Preparation and Multiscale Structure Analysis of 3D‐Printed PTZ Scaffolds

Rheological properties are essential for the printability of inks in Direct Ink Writing (DIW). We systematically investigated the rheological behavior of all inks (PT, PTZ‐1 to PTZ‐8) (Figure 2b–d). All groups showed pronounced shear‐thinning, with apparent viscosity decreasing as shear rate increased. This property ensures smooth nozzle extrusion and allows for rapid structural recovery after deposition, while maintaining dimensional fidelity.

FIGURE 2.

Morphology and elemental composition of 3D‐printed PCL/β‐TCP@ZnO composite scaffolds. (a) Schematic diagram of the direct ink writing (DIW) process for fabricating the PCL/β‐TCP@ZnO (3D PTZ) composite scaffold. (b) Shear stress dependence of the viscoelastic moduli (storage modulus G′ and loss modulus G″) for the PTZ ink. (c) Viscosity of the PTZ ink as a function of shear rate, illustrating shear‐thinning behavior. (d) Viscoelastic moduli (G′ and G″) of the PTZ ink plotted against angular frequency. (e1‐i1) Surface and (e2‐i2) corresponding cross‐sectional SEM images of the scaffolds. (e1, e2) 3D PT (0% ZnO). (f1, f2) 3D PTZ‐1 (0.1 wt% ZnO). (g1, g2) 3D PTZ‐2 (0.2 wt% ZnO). (h1, h2) 3D PTZ‐4 (0.4 wt% ZnO). (i1, i2) 3D PTZ‐8 (0.8 wt% ZnO). (j‐m) EDS elemental mapping images showing the distribution of O, P, Ca, and Zn on the surfaces of (j) 3D PTZ‐1. (k) 3D PTZ‐2. (l) 3D PTZ‐4. (m) 3D PTZ‐8 scaffolds.

Oscillatory shear tests showed that, across the tested frequency range, all inks exhibited a storage modulus (G′) consistently higher than the loss modulus (G″). This indicates that the materials display solid‐like, elasticity‐dominated behavior, which is essential for preserving fiber morphology and scaffold structural integrity during layer‐by‐layer deposition. The crossover point of G′ and G″ defines the yield stress, marking the transition from a gel‐like to a sol‐like state and ensures continuous and uniform flow during extrusion [22]. These rheological profiles align well with established printability criteria, providing a reliable foundation for the fabrication of high‐performance, patient‐specific scaffolds.

Using DIW and a cylindrical defect model, we successfully fabricated composite scaffolds with well‐defined three‐dimensional porous structures. Macroscopic observations showed that all scaffolds exhibited a regular grid‑like structure, with pore sizes precisely controlled within the range of 0.8–1.0 mm (Figure 2a), a size range that has been proven to promote cell infiltration, vascular ingrowth, and nutrient diffusion. High‑magnification SEM images (Figure 2e–i) showed that β‑TCP@ZnO particles are uniformly dispersed in the PCL matrix, generating a microarchitecture that closely mimics the trabecular structure of native bone [23]. As the ZnO content increases, the roughness of scaffold surface and cross‐sectional also increases, thereby enlarging the specific surface area and providing more active sites for cell adhesion and osteogenic differentiation.

To gain deeper insight into the spatial distribution of key elements, EDS mapping is conducted on the scaffold surfaces (Figure 2j–m). Strong, homogeneous signals for oxygen (O), phosphorus (P), and calcium (Ca) confirmed the uniform dispersion of the β‐TCP phase within the PCL matrix. At the same time, the zinc (Zn) signal intensity increased progressively with higher ZnO doping levels, and its uniform distribution across all PTZ groups indicated successful and stable incorporation of ZnO into the composite scaffold, without evident phase segregation during printing.

3.3. Physicochemical Properties and Degradation Behavior of 3D‐Printed PTZ Scaffolds

The chemical composition and crystalline structure of the scaffolds were analyzed by XRD and FTIR (Figure 3a,b). The XRD patterns showed distinct diffraction peaks at 21.3° and 23.7°, which are characteristic of the crystalline phase of PCL [24], while peaks at 27.8°, 31.1°, and 34.4° were assigned to β‐TCP. Weak signals near 19.0°, 24.8°, and 28.8° matched the standard ZnO pattern (PDF#36‐1451). The intensity of these ZnO‐related peaks increases with ZnO content, confirming that crystalline ZnO was successful incorporation into the composite material without affecting its phase purity [25].

FIGURE 3.

Physicochemical properties and degradation behavior of 3D‐printed PCL/β‐TCP@ZnO composite scaffolds. (a) XRD patterns and (b) FTIR spectra of the scaffolds. (c) Wide‐scan XPS survey spectra of 3D PTZ scaffolds. (d) In vitro degradation profiles of scaffolds incubated in lipase‐containing PBS. (e1–e4) High‐resolution XPS spectra of the 3D PTZ‐8 scaffold. (e1) O 1s, (e2) P 2p, (e3) Ca 2p, and (e4) Zn 2p. (f1–f5) Water contact angle measurements on 3D PT (f1), 3D PTZ‐1 (f2), 3D PTZ‐2 (f3), 3D PTZ‐4 (f4), and 3D PTZ‐8 (f5) scaffolds.

FTIR spectral analysis further verified the chemical integrity of each component (Figure 3b). The strong absorption band at 1725 cm^{− 1} is attributed to the ester carbonyl (C═O) stretching vibration of PCL, while the peaks at 1245 and 1160 cm^{− 1} corresponded to C─O─C stretching [26]. Characteristic bands at around 1044 and 971 cm^{− 1} originated from P─O stretching vibrations of PO₄ ³⁻ groups in β‐TCP. Additionally, the weak absorptions around 1049, 1025, and 952 cm^{− 1} are attributed to Zn─O─Zn and Zn─O vibrational modes, confirming the presence of ZnO. Overall, these features show that PCL, β‑TCP, and ZnO are effectively integrated while maintaining their respective chemical structures [27, 28].

The surface chemical states were analyzed by XPS (Figure 3c,e(1)–e(4)), and the results confirmed that all scaffold surfaces contained C, O, Ca, P, and Zn, with the intensity of the Zn signal increasing proportionally with the level of ZnO doping. High‐resolution spectral analysis of the PTZ‐8 scaffold shows that the O 1s region deconvolutes into two peaks, located at 531.9 eV (C═O in PCL) and 530.7 eV (O²⁻ in β‐TCP/ZnO). The characteristic peaks in the Ca 2p region showed at 551.0 and 547.5 eV correspond to Ca²⁺ in β‐TCP. The P 2p spectrum exhibits a double peak at 533.6 and 532.8 eV, corresponding to PO₄ ³⁻ in β‐TCP, while the Zn 2p spectrum exhibited shows a clear double peaks at 1044.5 and 1021.4 eV, which is characteristic of Zn²⁺. These results confirm that bioactive ions such as Ca²⁺, PO₄ ³⁻, Zn²⁺ coexist on the material surface, providing a favorable chemical environment for cellular interactions and osteogenic orientation [29].

Surface wettability is a key factor in the initial interactions between cells and materials, and it can be assessed by measuring the water contact angle (Figure 3f(1)–f(5)).

The contact angle of the pure PCL/β‐TCP scaffold (PT group) was 83.4 ± 3.3, indicating it is relatively hydrophobic. After the addition of zinc oxide, the surface hydrophilicity of all groups was improved, with the PTZ2 group showing the lowest contact angle (80.8 ± 3.2). This enhancement may be attributed to the increased surface polarity caused by Zn²⁺ substitution in the β‐TCP lattice and the inherent hydrophilicity of ZnO nanoparticles. Notably, surfaces with reported contact angles in the range of 65–73 degrees are more favorable for osteoblast adhesion [24, 30]. Among them, the PTZ2 group is close to this optimal range, which may contribute to the improvement of its biological response.

In vitro degradation profiles of the scaffolds were assessed in PBS (containing lipase) over 4 weeks (Figure 3d). All groups showed a gradual decrease in mass over time, but the degradation rate of the PTZ group scaffolds was significantly faster than that of the PT control group. By the fourth week, the weight loss of the PTZ group reached about 80%, while it was only about 40% for the PT group. This accelerated degradation may result from the hydrolysis of ester bonds in the PCL backbone catalyzed by zinc oxide, as well as the enhanced dissolution of the β‐TCP phase due to ionic exchange interactions between Zn²⁺ and Ca²⁺/PO₄ ³⁻. This controlled degradation, combined with continuous ion release, enables the scaffold to provide prolonged bioactive stimulation, thereby potentially contributing to a favorable microenvironment for bone regeneration.

3.4. Cytocompatibility and Osteogenic Differentiation Potential

The cytocompatibility of the fabricated scaffolds was assessed using MC3T3‐E1 pre‐osteoblasts. According to the CCK‐8 assay, cell viability on all scaffold groups remained above 90% at 24, 48, and 96 h, meeting the non‐cytotoxicity criterion defined by ISO 10993–5. As shown in Figure 4b, the PTZ‐2 group consistently supported the highest cell proliferation rates across all time points. After 96 h of culture, the proliferation rate in the PTZ‐2 group was approximately 1.24‐fold higher than that of the PT control group (p < 0.05), indicating that an optimal release of Zn²⁺ can significantly enhance pre‐osteoblast growth [31, 32].

FIGURE 4.

In vitro cytocompatibility and cell proliferation on 3D‐printed composite scaffolds. (a) Schematic of the in vitro cell culture experiment with MC3T3‐E1 (b) Quantitative analysis of cell proliferation by CCK‐8 assay after 24, 48, and 96 h of culture on the scaffolds. Data are presented as mean ± SD (n = 5). (c1‐c5) Fluorescence images of Calcein‐AM stained live cells cultured for 24 h on (c1) 3D PT, (c2) 3D PTZ‐1, (c3) 3D PTZ‐2, (c4) 3D PTZ‐4, and (c5) 3D PTZ‐8 scaffolds. (d1‐d5) Representative cell morphology after 48 h on (d1) 3D PT, (d2) 3D PTZ‐1, (d3) 3D PT‐2, (d4) 3D PTZ‐4, and (d5) 3D PTZ‐8. (e1‐e5) Cell distribution and confluency after 96 h on (e1) 3D PT, (e2) 3D PTZ‐1, (e3) 3D PT‐2, (e4) 3D PTZ‐4, and (e5) 3D PTZ‐8.

Notably, a concentration‐dependent effect is observed that as the ZnO doping level increased further (PTZ‐4, PTZ‐8), cell proliferation gradually declined, though it remained above the level of the control. This trend aligns with the established dual role of Zn²⁺, whereby low concentrations (<2 ppm) promote cell growth, while higher levels (> 5 ppm) may induce cell‐cycle arrest or apoptosis [33]. The results confirm that the ZnO content in PTZ‐2 achieves a kinetically favorable release profile, balancing bioactivity with biosafety.

Live‐cell staining with Calcein AM was performed to assess cell adhesion and morphology on the scaffolds (Figure 4c–e). At 24 h, cells had adhered to all scaffold surfaces and appeared predominantly rounded or oval. With increasing culture time, the cells spread, extended pseudopodia, and developed a characteristic osteoblastic morphology. Among the groups, PTZ‐2 supported the highest cell density and displayed extensive intercellular connections, resulting in an early network‐like structure. By 96 h, the scaffold surfaces were nearly confluent with cells, which was consistent with the CCK‐8 results and further demonstrated the excellent cytocompatibility of the PTZ‐2 scaffold [34].

Alkaline phosphatase and alizarin red are important in vitro markers of osteogenesis. In this study, scaffolds were stained and analyzed to evaluate osteogenic potential after co‐culturing cells with scaffolds for 3–21 days (Figure 5). ALP activity (Figure 5a–c), an early osteogenic marker, showed little significant inter‐group differences at day 3. By day 7, intense staining was evident in the PTZ‐1 and PTZ‐2 groups, with PTZ‐2 displaying the deepest and most uniformly distributed blue‐purple nodules at day 14, indicating stronger early osteogenic differentiation capacity, PTZ‐4 and PTZ‐8 groups showed lighter staining. The ALP fluorescence intensity of all groups increased with culture time(Figure S2) confirm that 3D PTZ‐2 group exhibited superior ALP activity at day 7 and day 14, with significant differences compared with the 3D PT group, demonstrating the strongest capacity to facilitate early osteogenic differentiation. In contrast, the high‐ZnO 3D PTZ8 group showed the lowest osteogenic induction effect, which may be attributed to the inhibitory effect of excessive Zn²⁺ on cell activity [32]. Mineralized nodule formation (Figure 5d–f), a hallmark of late osteogenesis, was evaluated by ARS staining. Red mineralized deposits were visible as early as day 7 in both PTZ‐1 and PTZ‐2 groups. The number and volume of nodules increased substantially by day 14. By day 21, PTZ‐2 scaffolds were almost entirely covered with dense, intensely stained mineralized matrix, indicating significantly superior osteogenic performance compared to all other groups.

FIGURE 5.

In vitro osteogenic differentiation induced by 3D‐printed composite scaffolds. (a1–a5) Alkaline phosphatase (ALP) staining of MC3T3‐E1 cells cultured in osteogenic medium for 3 days on different scaffolds. (b1–b5) ALP staining after 7 days of culture. (c1–c5) ALP staining after 14 days of culture. (d1–d5) Alizarin Red S (ARS) staining of showing mineralized nodules formation after 7 days of osteogenic induction. (e1–e5) ARS staining after 14 days.(f1–f5) ARS staining after 21 days. Scaffold groups from left to right in each row: 3D PT, 3D PTZ‐1, 3D PTZ‐2, 3D PTZ‐4, 3D PTZ‐8.

Mechanistically, the enhanced osteogenic effect of PTZ‐2 is attributed to the controlled release of Zn²⁺ [35], which is known to activate key signaling pathways such as BMP/Smad and Wnt/β‐catenin, thereby upregulating osteogenic transcription factors (e.g., Runx2, Osterix) and extracellular matrix genes (e.g., ALP, OCN). Moreover, Zn²⁺ can suppress osteoclast differentiation, further favoring a net anabolic environment [36].

3.5. In Vivo Bone Regeneration Performance and Mechanism

The osteogenic performance of the scaffolds was assessed in a rat calvarial critical‐size defect model (Figure 6a–c). At 8 weeks post‐implantation, PET‐CT 3D reconstruction images showed that defects in the control group were still largely occupied by fibrous tissue, with only limited new bone at the defect margins. In the 3D PT group, bone had grown appreciably from the edges toward the center, yet the central region remained partially unhealed. By contrast, the 3D PTZ‐2 group displayed markedly enhanced bone regeneration, with the defect region in most samples being largely bridged by newly formed bone tissue. By 12 weeks, healing in the blank control group remained limited. The 3D PT group demonstrated further defect reduction and gradual maturation of the newly formed bone. Remarkably, in the 3D PTZ‐2 group, the defect was almost completely covered by newly formed bone, with mineral density approaching that of the surrounding native bone. Quantitative analysis of the bone volume fraction (BV/TV) confirmed these observations: at 12 weeks, the 3D PTZ‐2 group achieved a BV/TV of 95.2 ± 3.5%, (P < 0.01) significantly higher than that of the 3D PT group (80.0 ± 4.2%) (P < 0.01) and the blank control (25.3 ± 6.8%) (P < 0.01). These data are in line with previous studies on Zn²⁺‐doped bioceramics and further demonstrate the strong in vivo bone‐regenerative capacity of the PTZ‐2 scaffold [37].

FIGURE 6.

In vivo osteogenic performance and histological evaluation in a rat cranial defect model. (a) Schematic of the experimental design for in vivo bone regeneration. (b) Intraoperative photograph showing implantation of a 3D PT scaffold into the calvarial defect. (b2–b4) Representative PET‐CT 3D reconstruction images of the defect site immediately after surgery (b2), at 8 weeks (b3), and at 12 weeks (b4) post‐implantation of the PT scaffold. (c2–c4) Corresponding PET‐CT images at the same time points following implantation of the PTZ‐2 scaffold. (d1–d4) H&E staining of the defect region at 8 weeks (d1,d2) and 12 weeks (d3,d4) after PT scaffold implantation. (e1‐e4) H&E staining at 8 weeks (e1,e2) and 12 weeks (e3,e4) after PTZ‐2 scaffold implantation. Black arrows indicate defect boundaries. HB represents host bone, NB represents new bone, FT represents fibrous tissue, and V stands for blood vessel. (f1–f4) Masson's trichrome staining of the defect area at 8 weeks and 12 weeks after PT scaffold implantation, corresponding to (f1,f2) and (f3,f4), respectively. (g1–g4) Masson's trichrome staining at 8 weeks and 12 weeks after PTZ‐2 scaffold implantation, corresponding to (g1,g2) and (g3,g4), respectively. Black arrows indicate defect margins. HB represents host bone, NB represents new bone, and FT represents fibrous tissue.

H&E and Masson's trichrome staining further characterized the quality and progression of bone regeneration (Figure 6d–g). At 8 weeks, the defects in the 3D PT group were primarily filled with fibrous tissue and immature, disorganized woven bone, with limited vascularization. In contrast, the 3D PTZ‐2 group displayed substantial formation of well‐aligned new bone trabeculae, accompanied by abundant vascular ingrowth. By 12 weeks, the 3D PT group showed advancing bone maturation and early transition toward lamellar bone. Meanwhile, the 3D PTZ‐2 group consisted predominantly of mature bone, with clear evidence of developing Haversian systems. Importantly, direct bone‐to‐material contact without intervening fibrous tissue was observed at the interface, demonstrating excellent osteointegration.

Masson's trichrome staining (Figure 6f,g) distinguishes collagen‐rich immature bone (blue) from mineralized mature bone (red). At 8 weeks, the 3D PT group stained predominantly blue, indicating abundant collagen but limited mineralization. The 3D PTZ‐2 group displayed a mixed blue and red pattern, reflecting concurrent active osteogenesis and mineralization. By 12 weeks, the proportion of red‐stained area increased in the 3D PT group, while the 3D PTZ‐2 group was characterized by extensive red staining, with blue regions confined mainly to remodeling sites around vascular channels, demonstrating advanced mineralization and structural maturity.

The superior in vivo performance of the 3D PTZ‐2 scaffold can be attributed to a synergistic combination of structural and biochemical cues. Its hierarchical porous architecture facilitates cell infiltration, vascularization, and nutrient transport. The sustained co‐release of Ca²⁺ and PO₄ ³⁻ creates a local microenvironment supersaturated with respect to bone‐like apatite, while the controlled delivery of Zn²⁺ directly stimulates osteoblast differentiation and matrix mineralization. Additionally, Zn²⁺ modulates the repair microenvironment by promoting angiogenesis and suppressing osteoclastic activity. This integrated design‐merging 3D‐printed topology with functional ion therapy, aligns with current advances in multifunctional bone tissue engineering [38, 39].

3.6. Perspectives

This study successfully developed a 3D‐printed PCL/β‐TCP@ZnO composite scaffold through a combination of multi‐scale structural design and functional ion‐doping. The approach synergized the patient specific customization capability of direct ink writing (DIW) with a bioactive ion release strategy, resulting in a scaffold with significantly enhanced osteoinductive activity and osseointegration potential. Compared to other advanced bone repair materials, such as ZnO‐modified mesoporous bioactive glass nanospheres [40], multi‐ion doped hydrogels [41], and biomimetic mineralized collagen scaffolds [42], the present work distinguishes itself by integrating precise architectural control with sustained multi‐ionic (Ca²⁺, PO₄ ³⁻, Zn²⁺) delivery, offering a versatile and translatable platform for repairing complex bone defects.

However, several limitations of the current study warrant attention and further investigation. First, the range of ZnO doping concentrations explored remains relatively narrow. Future studies should expand the doping spectrum and investigate the synergistic effects of multi‐ionic doping (e.g., Sr²⁺, Mg²⁺, Cu²⁺) to better recapitulate the complex chemical microenvironment of native bone. Second, the in vitro degradation and ion‐release assays were conducted under static conditions, which may not fully simulate the dynamic physiological milieu. The adoption of microfluidic‐based dynamic culture systems could provide more predictive insights into the long‐term in vivo behavior of the scaffold. Third, while the osteogenic performance was comprehensively evaluated, the antibacterial properties of the ZnO‐doped scaffolds were not systematically assessed. Subsequent work should include rigorous antibacterial tests to validate the dual functionality of anti‐infection and osteogenesis, which is crucial for clinical applications in contaminated or infected defect sites.

From a technological standpoint, the integration of artificial intelligence (AI) and machine learning (ML) could further advance the development of next generation scaffolds. AI driven optimization of printing parameters and real‐time process monitoring would enhance the precision, reproducibility, and scalability of scaffold fabrication. Moreover, ML algorithms could be employed to establish predictive relationships among material composition, printing parameters, structural features, functional performance, thereby accelerating the rational design of intelligent, patient‐specific bone grafts [43].

In summary, this work not only presents a novel and clinically promising bone repair scaffold but also provides a systematic structure, property and mechanism framework for the design of functional bone tissue engineering materials. Through continued interdisciplinary collaboration, the strategy of combining 3D printing with functional ion doping is poised to contribute meaningfully to the advancement of personalized regenerative medicine.

Author Contributions

Ayijiamila Mohetaer: writing – original draft, date curation, software, methodology, writing – review and edtiting. Zhang Jing: software, writing – review and editing. Shan Rong: methodology, data curation, writing – original draft. Wang Yuxin: methodology, data curation. Wan Xiaofei: formal analysis, visualization. Ma Chunquan: writing – original draft, data curation. He Qian: writing – review and editing, validation. Song Fuxiang: funding acquisition, investigation, writing – original draft, writing – review and editing, supervision. Liu Bin: funding acquisition, visualization, supervision.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: mabi70245‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.