3D‐Printed Titanium Implants with Bioactive Peptide‐Polysaccharide Scaffolds for Personalized Bone Reconstruction

ABSTRACT

Large bone defects caused by trauma, tumor resection, or congenital abnormalities remain a major clinical challenge. Standard titanium implants are widely used due to their strength and biocompatibility, but their bioinert surfaces often lead to poor osseointegration. The emergence of 3D printing has enabled patient‐specific titanium implants with tailored architecture and mechanical properties. However, these constructs still lack the bioactivity required for robust and spatially uniform bone integration, particularly within the implant core. To address this limitation, we developed a bioactive, cell‐free strategy that integrates porous titanium implants with a nanofibrillar peptide‐hyaluronic acid scaffold, delivered either as a hydrogel or in lyophilized form. The scaffold exhibited enhanced enzymatic stability and supported osteoblast‐like cell adhesion in vitro. In a rabbit calvarial critical‐size bone defect model, scaffold‐integrated implants significantly outperformed inert controls, with hydrogel integration nearly doubling inner bone volume and improving trabecular architecture. Histological analysis confirmed enhanced bone‐implant integration, active periosteum, healthy marrow, and reduced inflammation. This acellular, growth‐factor‐free approach combines the structural precision of titanium with the regenerative potential of ECM‐mimicking scaffolds, offering a translatable pathway for personalized skeletal repair.

Porous 3D‐printed titanium implants are made bioactive by integration with a supramolecular peptide‐hyaluronic acid nanofibrillar scaffold, without the addition of exogenous cells or growth factors. Uniform filling of the implant architecture promotes vascularized, spatially homogeneous bone regeneration, significantly enhancing osteogenesis throughout the implant, including its core, in a critical‐size bone defect model.

1. Introduction

Bone defects caused by trauma, tumor resection, or failed surgical procedures often affect load‐bearing bones, where structural integrity and long‐term mechanical performance are critical [1, 2]. When such defects exceed the body's natural regenerative capacity, they are considered critical‐size defects and represent a major clinical challenge [3, 4, 5]. Although bone is a dynamic tissue capable of self‐repair when fracture fragments are stabilized and closely aligned [6, 7], regeneration in critical‐size defects is impaired due to disrupted biological and mechanical environments [7, 8].

To restore both structure and function in such cases, surgical intervention is required, often including the implantation of a device that provides stability or replaces the missing bone [1, 9, 10]. Traditional approaches such as autografts, allografts, and standard prosthetics can fall short, especially in anatomically complex cases, highlighting the need for more effective solutions [3, 5].

A widely adopted solution is the implantation of metallic devices, particularly those composed of titanium or titanium alloys, which are favored for their strength, biocompatibility, and corrosion resistance [11, 12]. The emergence of additive manufacturing has enabled the fabrication of patient‐specific, 3D printed titanium implants offering tailored geometry and mechanical properties [4, 9, 10, 13, 14, 15]. In some cases, the use of patient‐specific implants in bone defects adjacent to a joint can serve as an alternative to limb amputation [16]. However, the bioinert nature of titanium limits direct interaction with host tissue, often resulting in poor osseointegration and limited new bone formation within the inner porous regions of the implant. This lack of bone ingrowth into the core often leaves large central voids, weakening the implant's stability and putting the long‐term success of the reconstruction at risk, especially when the implant has to bear load [8, 13, 14, 15, 17, 18, 19].

To address these limitations, numerous strategies have been developed to enhance the bioactivity of porous titanium implants. One major approach has focused on surface modifications of the printed structure. Techniques such as acid etching, alkali treatment, anodization, and plasma spraying have been employed to tailor surface roughness and chemistry, thereby promoting protein adsorption, osteoblast adhesion, and subsequent bone formation [20]. In addition, biofunctional coatings, including calcium phosphate, hydroxyapatite, bioactive glasses, and peptide‐functionalized layers, have been applied onto porous titanium to accelerate osseointegration and provide osteoconductive cues [20, 21, 22, 23, 24, 25]. Beyond surface modifications, another strategy involves incorporating bone graft materials, such as autografts and allografts, into the porous implant structure to provide a direct osteogenic stimulus [26, 27, 28]. While these grafts possess inherent osteoinductive and osteoconductive properties, their limited availability, donor site morbidity, and risk of immune rejection make them a challenging clinical option [29, 30].

Additional efforts have turned to synthetic bone substitutes and hydrogels, which provide a more adaptable and biomimetic microenvironment. For example, silicate‐substituted calcium phosphate was incorporated into a lamellar titanium cage to enhance bone formation and implant fusion in a sheep spine model [31]. Hydrogels, in particular, have gained considerable attention owing to their tunable mechanical properties, injectability, and ability to deliver cells, drugs, or bioactive molecules [32, 33]. Several systems have been investigated, including hydroxyapatite‐loaded calcium alginate hydrogels [34] and silver nanoparticle‐loaded hydrogels [35]. PLGA‐PEG‐PLGA gels have primarily served as drug carriers for cisplatin or simvastatin within the titanium implants post tumor resection, where they suppressed local tumor recurrence and simultaneously promoted bone regeneration [36, 37]. Despite their promise, most hydrogel systems still face significant challenges, including dependence on controlled drug or growth factor release, difficulty in synchronizing their degradation upon new tissue formation, limited and uneven infiltration throughout the porous architecture, and insufficient mechanical stability under load‐bearing conditions [38, 39, 40, 41, 42, 43]. These limitations emphasize the need for next‐generation hydrogel formulations that combine biomimicry, mechanical resilience, and bioactivity to achieve consistent regeneration throughout the implant core [44].

A promising approach involves mimicking the native extracellular matrix (ECM), which consists of a 3D fibrillary network predominantly composed of proteins, with collagen being the most abundant [45, 46, 47]. Self‐assembling amino acids and short peptides can form nanofibrillar structures that resemble the ECM's morphology [46, 48, 49, 50, 51]. These peptides are highly tunable in their mechanical and biochemical properties and can incorporate molecular recognition motifs such as the arginine‐glycine‐aspartic acid (RGD) sequence found in fibronectin, enabling enhanced cell adhesion and signaling [52, 53].

Traditionally, scaffolds designed to mimic the native ECM are fabricated either as hydrogels in a hydrated state or as dry, solid structures [1, 46]. Lyophilization is commonly used to convert hydrogels into dry scaffolds by removing water while preserving the 3D fibrillary architecture [54]. However, the outcome of lyophilization depends on multiple factors, including freezing rate, fiber stability, and processing parameters, which can subtly alter the scaffold's microstructure compared to the original hydrogel form [55]. As a result, although lyophilized scaffolds retain an ECM‐like network, their mechanical and physical properties often differ from those of the hydrated hydrogels, despite originating from the same material [56]. Understanding these differences is crucial, as the scaffold's physical state can influence cell behavior, tissue infiltration, and ultimately, the success of bone regeneration [57, 58].

We and others have previously demonstrated the potential use of peptide‐based nanostructures in a range of biomedical applications [46, 53, 59, 60, 61]. We demonstrated that fluorenylmethoxycarbonyl‐diphenylalanine (FmocFF) peptides can self‐assemble into elongated β‐sheet fibrillar networks that interpenetrate hyaluronic acid (HA), yielding homogeneous composite hydrogels with markedly improved mechanical rigidity and resistance to enzymatic degradation, while preserving biocompatibility and enabling the stabilization and sustained release of bioactive molecules without the need for toxic cross‐linkers [62]. When integrated into polymeric scaffolds, HA‐peptide composites were further shown to promote osteogenic differentiation and mineralization in vitro, highlighting their potential for bone tissue regeneration [53, 63]. The HA component used in this study has a molecular weight of approximately 3 × 10⁶ Da. Such high‐molecular‐weight HA (HMW‐HA) is known to exert immunomodulatory effects and reduce inflammation in surrounding tissues [64]. Accordingly, FmocFF/HA hydrogels have been shown to induce macrophage polarization toward a pro‐regenerative M2 phenotype, attenuate inflammatory responses, and promote vascularized bone formation in vivo [65]. This combination of mechanical robustness with immunomodulatory bioactivity positions the FmocFF/HA as a particularly promising scaffold for biomimetic and regenerative applications.

In this study, we aimed to functionalize porous titanium implants to introduce bioactivity and enhance bone regeneration by utilizing a minimal, self‐assembling FmocFF/HA scaffold delivered either as a hydrogel or a lyophilized scaffold. While various scaffold treatment approaches work very well in terms of bone regeneration, their limited mechanical strength prevents their direct translation to large, load‐bearing bone defects that require long‐term stability. The present work addresses this limitation by integrating the scaffold within 3D‐printed titanium implants, combining the mechanical robustness of the metallic architecture with the bioactivity of the supramolecular hydrogel.

Using a critical‐size calvarial defect model in rabbits, we investigated the scaffold's ability to induce bone regeneration and how the physical state of the scaffold influences spatial bone formation within the implant. While the lyophilized scaffold primarily supported bone formation at the defect periphery, the hydrogel form was shown to promote substantial regeneration across both peripheral and central regions, exhibiting a uniform and volumetric pattern. Unlike previous bone regeneration strategies that rely on complex fabrication, exogenous cells [66], or growth factors [39], this approach employs a simple, cell‐free supramolecular scaffold that self‐assembles within titanium implants, enabling efficient bone regeneration through endogenous repair mechanisms. This novel system thus offers a promising strategy for personalized bone regeneration in various pathologies.

2. Results and Discussion

2.1. Implant‐Scaffold Integration and Characterization

The porous architecture of the 3D‐printed titanium implant plays a crucial role in facilitating bone ingrowth and integration with the host tissue [5, 42]. In this study, the gyroid lattice design was employed, offering a continuous, interconnected pore network that not only supports vascular and cellular infiltration but also provides mechanical stability and uniform stress distribution [4]. This geometry further enables efficient incorporation of bioactive scaffolds via simple infusion techniques. For this purpose, we utilized additive manufacturing via SLM to produce titanium gyroid cylinders with a 5 × 5 × 5 mm cell size and a wall thickness of 0.3 mm. We immersed the implant in an HA solution, followed by the addition of the FmocFF peptide and brief vortexing, which induced an in situ sol–gel transition within the pores. The resulting FmocFF/HA hydrogel was confined within the implant structure or could subsequently be lyophilized, thereby preserving its 3D architecture and facilitating later rehydration (Figure 1A–C). To evaluate the FmocFF/HA hydrogel, in its wet or its lyophilized form, infiltration into the internal architecture of the 3D‐printed titanium mesh, we designed a cylindrical implant that could be longitudinally divided into two semi‐cylinders for direct observation (Figure 1D). Cross‐sectional observation revealed that the hydrogel uniformly infiltrated the entire mesh structure (Figure 1E–J). As an alternative integration strategy, we injected the preformed FmocFF/HA hydrogel directly into the implant using a fine cannula. Despite the hydrogel's self‐healing capabilities [65], this approach resulted in partial leakage. The high shear stress and pressure generated within the narrow lumen of the cannula disrupted the hydrogel network, leading to compromised scaffold integrity. Based on these observations, we selected the two more effective methods for further investigation: in situ gelation within the implant pores and lyophilization following gelation.

FIGURE 1.

Structural and compositional characterization of 3D‐printed implant lattices integrated with FmocFF/HA scaffolds. (A) Molecular structure of the FmocFF peptide. (B) Molecular structure of HA. (C) Schematic illustration showing the incorporation of a hydrogel or lyophilized scaffold into the 3D‐printed implant. (Created in Biorender.com). (D) Illustration of the custom 3D‐printed lattice fabricated for the in vitro scaffold penetration assessment. (E–G) Top‐view images. (H–J) Lateral view images. (K–M) HRSEM images displaying the surface. (N, O) EDS analysis. (E, H, K) Inert implant (F, I, L) Lyophilized FmocFF/HA‐incorporated implant, (G, J, M) Hydrogel‐incorporated Implant, (N) Inert implant, (O) Lyophilized FmocFF/HA‐incorporated implant.

High‐resolution scanning electron microscopy (HRSEM) imaging revealed a network of fine fibrils formed by the self‐assembly of the FmocFF peptide, both in the hydrogel state and after lyophilization (Figure 1K–M). These fibrillar structures closely mimic the 3D microarchitecture of native ECM, which is known to support cell adhesion, signaling, and proliferation [48]. Although macroscopic analysis suggested uneven scaffold distribution across the implant surface (Figure 1F, G, I, J), EDS analysis demonstrated a consistent coating of organic matrix. Even in regions that macroscopically appear uncoated, the surface exhibited a significantly elevated carbon content, indicating successful and uniform scaffold integration (Figure 1N, O; Table S1).

2.2. In Vitro Structural Integrity, Mechanical Properties, and Degradation Behavior

The stability of the scaffolds and the contributions of each component, namely FmocFF and HA, were evaluated by comparing hybrid FmocFF/HA lyophilized scaffolds to implants integrated with each component separately (Figure 2A–I). Pristine FmocFF, pristine HA, and hybrid FmocFF/HA hydrogels were incorporated into the implants and then subjected to lyophilization. Following rehydration in MEMα medium and incubation at 37°C, distinct differences in structural integrity were observed. Implants containing pristine HA dissolved almost immediately upon rehydration, demonstrating the limited stability of HA in aqueous environments (Figure 2A–C). In contrast, the pristine lyophilized FmocFF scaffold remained intact during the first 24 h of rehydration and began to disintegrate into the medium within seven days of incubation (Figure 2D–F). Notably, the hybrid FmocFF/HA scaffold maintained its structure without significant disintegration over seven days of incubation in the medium (Figure 2G–I). These results indicate that the incorporation of HA into the FmocFF fibrillar network enhances structural cohesion through non‐covalent supramolecular interactions, thereby improving aqueous stability and resistance to dissolution.

FIGURE 2.

In vitro degradation dynamics and mechanical characterization of scaffold‐integrated implants. (A–I) Media‐induced rehydration of lyophilized scaffolds embedded within 3D‐printed implants over 24 h and 7 days at 37°C. (A–C) HA‐only lyophilized scaffold, (D–F) FmocFF‐only lyophilized scaffold, and (G–I) composite FmocFF/HA lyophilized scaffold. (J) Time‐sweep rheological analysis showing the storage modulus (G′) of HA, FmocFF, and FmocFF/HA hydrogels (The experiment was repeated three times). (K) Cumulative degradation of fluorescently labeled HA in the presence of hyaluronidase (Hyas) when incorporated alone or as part of FmocFF/HA, in both hydrogel and lyophilized forms (n = 3).

To further evaluate the mechanical stability of the hydrogels, oscillation time‐sweep rheology was performed. The analysis showed a substantial increase in the storage modulus (G′) from 91 Pa for pristine HA and 1,313 Pa for pristine FmocFF to 64,360 Pa for the FmocFF/HA hybrid hydrogel (Figure 2J). This pronounced enhancement reflects the synergistic reinforcement arising from the interpenetration of FmocFF nanofibrils with HA chains. The resulting high stiffness falls within the mechanical range known to promote osteogenic differentiation of pre‐osteoblasts and mesenchymal stem cells (MSCs), indicating that the scaffold's intrinsic rigidity may contribute to its osteoinduction activity [67, 68].

The enzymatic degradation profile of HA was assessed by exposing implants containing fluorescently labeled HA to hyaluronidase and monitoring fluorescence release over 14 days. Lyophilized pristine HA degraded rapidly, with more than 80% lost within the first few hours. Hydrated pristine HA showed a slower yet still substantial degradation rate. In contrast, HA incorporated into the FmocFF/HA hybrid scaffold, either in hydrogel or lyophilized form, exhibited markedly enhanced resistance to enzymatic degradation (Figure 2K). In addition, hydrogel mass loss measurements, conducted in the presence of hyaluronidase (Figure S1) confirmed the improved structural persistence of the hybrid hydrogel system compared with the pristine HA or pristine FmocFF hydrogels. These results indicate that the hybrid scaffold effectively protects HA from rapid enzymatic breakdown, enhancing its structural stability. Importantly, this slower degradation is biologically relevant, as rapid HA fragmentation into low molecular weight (LMW) fragments is known to activate pro‐inflammatory signaling cascades, whereas the preserved HMW‐HA form maintains its immunomodulatory and anti‐inflammatory functions [64]. Together, the mechanical and degradation data indicate that the FmocFF/HA scaffold not only maintains its fibrillar architecture under physiological conditions but also resists premature HA enzymatic breakdown.

2.3. In Vitro Cell‐Scaffold Interaction

Next, we examined the in vitro cytocompatibility of the bioactive implants. The MG63 human osteosarcoma cell line, widely used as an osteoblast‐like model, was employed to evaluate cell‐scaffold interactions relevant to bone regeneration [69]. Cell attachment and cell morphology were assessed using HRSEM following 48 h of MG63 cell culture on inert implants, FmocFF/HA lyophilized‐integrated implants, and hydrogel‐integrated implants. In all groups, cells exhibited adhesion, spreading, and evidence of mitotic activity, indicating initial biocompatibility across the different surfaces (Figure 3).

FIGURE 3.

Structural and biological characterization of cell–scaffold interactions on 3D‐printed implants. (A, B) MG63 osteoblast‐like cells were cultured for 48 h on the inert titanium implant. (C, D) MG63 cells grown on the FmocFF/HA lyophilized scaffold integrated implant. (E, F) MG63 cells grown on the FmocFF/HA hydrogel integrated implant. (G) MG63 viability on the different implants over 1, 3, and 7 days (n = 5). (H) Normalized alkaline phosphatase (ALP) activity of MC3T3‐E1 pre‐osteoblasts 8 days after seeding in osteogenic differentiation medium (n = 3). (I) Normalized absorbance of Alizarin Red staining after 14 days of osteogenic differentiation in osteogenic medium (n = 4). (J–M) Microscopy images of MC3T3‐E1 cells stained with Alizarin Red after 14 days on (J) Control (K) inert implant, (L) implants with lyophilized‐scaffold, and (M) hydrogel‐integrated implant under osteogenic conditions. Data are presented as mean ± SD. One‐way ANOVA was used for ALP and Alizarin Red assays; two‐way ANOVA for cell viability. ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001. Scale bars: (A, C, E) 10 µm; (B, D, F) 1 µm; (J–M) 500 µm.

Although overall cellular morphology appeared similar, differences in cell‐substrate interactions were evident. Unlike the inert implant (Figure 3A, B), cells were found to interact with a dense nanofibrillar network on the hydrogel‐integrated implants and lyophilized‐scaffold integrated implants, which resembles the structural organization of the native extracellular matrix (Figure 3C–F). This fibrillar architecture likely facilitates cell attachment, consistent with previous studies showing that supramolecular peptide hydrogels composed of nanofibrils can support robust cell adhesion [32].

The aromatic stacking and nanoscale fibril structure of FmocFF fibrils promote physical anchorage and focal adhesion formation, as reflected by the filopodia bridging observed in high‐magnification micrographs (Figure 3D, F). This structural organization also enhances the adsorption of serum‐derived adhesion proteins, as confirmed by the BCA assay (Figure S3), which showed measurable protein adsorption on both hydrogel‐integrated implants and lyophilized‐scaffold integrated implants. Adsorbed proteins, such as fibronectin and vitronectin, can mediate integrin engagement and drive subsequent focal adhesion formation [57].

These observations suggest that the hybrid FmocFF/HA scaffold supports cell adhesion and mechanotransduction primarily through its supramolecular fibrillar organization and physical topography, complemented by the adsorption of adhesive serum proteins that mediate integrin engagement. This indirect bioactivity mechanism is consistent with previous reports on peptide‐based hydrogels, where noncovalent nanofibrillar organization, along with protein adsorption, were shown to promote cell anchorage and mechanotransduction via protein‐mediated interfaces [32, 57, 68].

On the smooth, rigid titanium implants, only limited focal adhesion points were typically observed. While titanium implant substrates provide essential mechanical support, they also influence cell behavior through mechanotransduction [46]. In contrast, the supramolecular fibrillar network of the FmocFF/HA scaffold promotes adhesion through structural cues, providing a biomimetic topography that enhances integrin clustering and cytoskeletal organization.

Cell viability and proliferation on the different implants were assessed using the Alamar Blue assay at 1, 3, and 7 days, with values normalized to cells grown on standard tissue‐culture plates (Figure 3G). Viability was highest on the lyophilized scaffold‐integrated implants compared to both the inert titanium and hydrogel‐integrated groups. Although initial cell attachment on the hydrogel‐integrated implants appeared slightly limited, the cell population gradually recovered and proliferated over time. These trends likely reflect differences in available surface area and the microstructure of the scaffold. The inert titanium implant provides a fully exposed surface for cell adhesion, whereas the hydrogel‐integrated implant restricts cell penetration under static in vitro conditions, limiting attachment primarily to the outer surface of the hydrogel. In contrast, the lyophilized scaffold provides an interconnected, porous network that increases the accessible surface area and facilitates the physical entrapment and adhesion of cells within the pores. Additionally, the increased initial viability of the lyophilized scaffold‐integrated implant is likely attributed to its higher mechanical stiffness compared to the hydrated hydrogel, which enhances early‐stage attachment and initial viability. Despite these initial variations, all groups exhibited a comparable proliferation trend over time, confirming the overall cytocompatibility of the hybrid scaffolds.

To further assess the intrinsic bioactivity of the scaffold components independently of the titanium substrate, MG63 cell viability and proliferation were evaluated on the hydrogel and lyophilized forms of pristine HA, pristine FmocFF, or FmocFF/HA (Figure S4). Pronounced differences were observed both between formulations and between the hydrogel and lyophilized scaffold forms. Among the hydrogels, distinct trends emerged: pristine HA showed a gradual decline in viability over time, FmocFF exhibited a modest increase followed by a decrease, whereas the FmocFF/HA hydrogel supported continuous cellular proliferation throughout the 7‐day period. These differences likely stem from the limited mechanical stability of pristine HA and pristine FmocFF, compared to the enhanced robustness and supramolecular integrity of the FmocFF/HA hydrogel, which provides a more favorable microenvironment for sustained cell growth. In contrast, the lyophilized scaffolds supported even higher viability and proliferation, particularly in the FmocFF/HA formulation, which consistently outperformed both single‐component scaffolds at all time points. Notably, the decline in viability observed on the lyophilized FmocFF scaffold correlated with its faster disintegration described above (Section 2.2). Overall, these findings align with the rheological and degradation data, indicating that the mechanically reinforced and enzymatically stable FmocFF/HA scaffold, especially in its lyophilized form, offers an optimized architecture that promotes cell attachment and proliferation.

To assess the osteogenic activity of the bioactive implants, alkaline phosphatase (ALP) enzyme was quantified in MC3T3‐E1 pre‐osteoblasts cultured on the different implants after 8 days of differentiation (Figure 3H). Both bioactive implants exhibited higher ALP activity than the inert titanium control, demonstrating that incorporation of the FmocFF/HA scaffold enhances osteogenic differentiation. This elevated activity likely results from the scaffold's nanofibrillar architecture, which provides mechanical stiffness and topographical cues that favor osteogenic signaling, together with the presence of HMW‐HA known to support cell differentiation [63].

Osteogenic differentiation and ECM mineralization were assessed via Alizarin Red staining after 14 days of culture under osteogenic conditions (Figure 3I–M). Alizarin Red quantification revealed a marked increase in mineral deposition on both FmocFF/HA‐integrated implants compared to the control group, while the cells grown on the inert implant demonstrated a mild elevation. The hydrogel‐integrated implant exhibited the highest normalized absorbance value, indicating that incorporation of the FmocFF/HA hydrogel substantially enhances matrix mineralization, consistent with the elevated ALP activity measured at earlier stages. The superior performance of the bioactive implants likely stems from their nanofibrillar architecture, which provides a mechanically supportive and biomimetic microenvironment conducive to osteogenic maturation, while the presence of HA contributes to a hydrated niche that promotes ion diffusion and ECM mineral deposition. Together, the ALP and Alizarin Red assays demonstrate that both hydrogel and lyophilized FmocFF/HA scaffolds actively support osteogenic differentiation and mineralized matrix formation in vitro.

However, in vitro assays provide only a partial picture, as bone regeneration depends on tightly orchestrated processes, such as vascularization, immune response, and tissue remodeling, that cannot be fully reproduced ex vivo. Therefore, to assess the true regenerative potential of the hybrid implants, their performance was further examined in vivo.

2.4. In Vivo Critical‐Size Rabbit Calvarial Defect for Restored Bone Regeneration

2.4.1. In Vivo Micro‐CT Assessment of the Entire Bone Defect

The regenerative performance of the bioactive implants, utilizing two different incorporation strategies, was evaluated in a rabbit calvarial critical‐size defect model. Three defects 8 mm in diameter were created in the calvaria of nine female New Zealand White rabbits. Each defect was reconstructed with one of the following treatments: (i) an inert porous Ti‐6Al‐4 V implant (control) N = 7, (ii) an implant integrated with a lyophilized FmocFF/HA scaffold N = 8, or (iii) an implant integrated with FmocFF/HA hydrogel N = 7. All implants were fabricated by additive manufacturing using SLM to produce gyroid cylinders with a 5 × 5 × 5 mm cell size and a wall thickness of 0.3 mm (Figure 4A).

FIGURE 4.

Micro‐CT assessment of in vivo bone regeneration in rabbit calvarial defects treated with bioactive titanium implants. (A) Schematic illustration of the gyroid lattice unit cell used for implant fabrication, with 5 × 5 × 5 mm cell size and 0.3 mm wall thickness. (B) 3D micro‐CT reconstruction of a rabbit calvaria illustrating bone regeneration across the three implant types. Regenerated bone is shown in light blue, pristine bone in light brown, and implants in semi‐transparent grey. (C–E) Representative micro‐CT cross‐sections (top and lateral views) of calvarial defects 8 weeks post‐surgery in rabbits treated with: (C) inert titanium implant, (D) titanium implant incorporating lyophilized FmocFF/HA scaffold, and (E) titanium implant incorporating FmocFF/HA hydrogel. (F) Quantification of restored bone volume within the entire defect area for each treatment group. (G) Trabecular number calculation and (H) Trabecular separation measurements of the regenerated bone (mean ± SD; ^* p < 0.05, ^** p < 0.01, one‐way ANOVA, inert and hydrogel‐integrated implants n = 7, lyophilized‐scaffold integrated implants n = 8).

Animals were euthanized at 8 weeks post‐surgery, and calvarial specimens were harvested for micro‐CT and histological analyses. Visual assessment of the micro‐CT sections indicated some degree of implant integration across all groups. However, substantially superior bone formation was observed in the central regions of defects treated with the FmocFF/HA‐integrated implants, both in hydrogel and lyophilized forms. In contrast, bone regeneration in the inert control group was predominantly limited to the periphery of the defect and regions adjacent to native bone. Notably, the hydrogel‐integrated implants exhibited pronounced bone ingrowth extending throughout the inner zones of the implant architecture (Figure 4C–E; Figure S6).

Quantitative micro‐CT and 3D reconstruction analyses confirmed that incorporation of FmocFF/HA, whether as a hydrogel or as a lyophilized scaffold, significantly enhanced overall bone regeneration compared to the inert implants. Although the difference did not reach statistical significance, there was a trend toward higher bone volume in the hydrogel‐integrated group (81%) compared to the lyophilized group (68.6%), both of which outperformed the inert implant group (48.4%) (Figure 4F). Morphometric analysis further demonstrated an increased trabecular number (Tb.N) in both FmocFF/HA‐integrated groups, 0.84 mm^{− 1} for the hydrogel group and 0.86 mm^{− 1} for the lyophilized group, compared to 0.62 mm^{− 1} for the inert control, indicating a denser trabecular architecture and improved bone quality for the scaffold‐treated implants (Figure 4G). Moreover, trabecular separation (Tb.Sp) was notably higher in the inert implant group (0.88 mm) relative to the lyophilized (0.68 mm) and hydrogel (0.65 mm) groups, suggesting poorer bone regeneration and disrupted trabecular connectivity surrounding the inert implant (Figure 4H).

The slightly enhanced regeneration observed in the hydrogel group compared to the lyophilized scaffold may stem from the homogeneous distribution of the former and its close contact with the implant surface, enabling better diffusion of nutrients and cellular infiltration during early healing. In contrast, the lyophilized scaffold, while offering higher initial porosity and surface area, may require more time to rehydrate and integrate with the surrounding tissue before supporting effective cellular migration. These complementary characteristics highlight how the physical state of the scaffold modulates the biological response, with the hydrogel providing immediate bioactive continuity and the lyophilized scaffold form offering structural guidance.

2.4.2. In Vivo Regional Analysis of Bone Regeneration

The central region of a bone defect is often recognized as particularly challenging for successful repair in critical‐size defect models [8, 13, 14, 19]. Thus, although titanium alloys such as Ti‐6Al‐4 V are known to exhibit excellent osseointegration [10, 11, 44], new bone formation often remains concentrated near the defect margins, leaving the central regions inadequately regenerated. This limited ingrowth can compromise long‐term implant stability and hinder full integration with the surrounding bone [19]. To gain deeper insights into the spatial pattern of bone regeneration, the defect volume was virtually divided into two equal‐volume segments, the outer and inner regions. In the outer zone, both FmocFF/HA‐integrated groups improved bone regeneration, with bone volume reaching 106% in the lyophilized‐integrated group and 101% in the hydrogel‐integrated group, compared to only 78% in the inert implant control (Figure 5).

FIGURE 5.

Regional in vivo micro‐CT analysis of regenerated bone volume. (A) Quantification of regenerated bone volume within defined outer and inner regions of the calvarial defect, each occupying half of the total defect volume (two‐way ANOVA: ^* p < 0.05, ^** p < 0.01, inert and hydrogel‐integrated implants n = 7, lyophilized‐scaffold integrated implants n = 8). (B) Representative 3D micro‐CT reconstructions of selected defects from each treatment group, showing digitally segmented regenerated bone in outer and inner regions. The color scheme corresponds to the group legend shown in (A).

Moreover, a more pronounced regenerative effect of the hydrogel was shown in the inner region. In the inner zone, the hydrogel‐integrated group exhibited nearly a twofold increase in bone volume (43.8%) compared to the inert control (21%). The lyophilized‐integrated group also showed an average enhancement in bone formation in this region (32.3%), although the difference compared to the inert group did not reach statistical significance (Figure 5). The superior bone formation observed within the central region of the hydrogel‐integrated implants group may be attributed to the continuous, hydrated nature of the scaffold, which allows better interaction with host cells and signaling molecules into the implant's interior during healing. Conversely, the lyophilized scaffold, although structurally porous, likely experienced a transient delay in rehydration and cell interaction, resulting in slightly reduced early bone deposition in the innermost regions. This observation reinforces the importance of scaffold hydration and matrix continuity in promoting uniform regeneration across large defect volumes. These findings highlight the importance of the incorporation of bioactive scaffolds in achieving deep and uniform bone regeneration throughout the defect volume, potentially overcoming one of the key limitations of traditional metallic implants.

Although several material‐based strategies incorporating cells, osteogenic growth factors, pharmacological agents, or bioceramic fillers have demonstrated strong bone regeneration in similar critical‐size models [39, 66], such systems typically depend on exogenous biological components, multistep fabrication procedures, and precisely controlled release kinetics, all of which complicate clinical translation [32]. In contrast, the current hybrid scaffold achieves substantial and spatially uniform bone regeneration using only endogenous host cells, demonstrating that the supramolecular assembly of FmocFF/HA provides sufficient biochemical and topographical cues to induce osteogenesis without external stimuli. This cell‐free, minimally processed approach therefore represents a distinct and translationally relevant advancement compared with previous complex biomaterial systems.

2.4.3. Histological and Histomorphometric Evaluation of Bone Regeneration

Calvarial specimens were sectioned using a microtome (Figure 6A–C) and subjected to hematoxylin and eosin (H&E) (Figure 6D–I) and Masson's Goldner trichrome (MGT) (Figure 6J–O) staining. The histological findings were consistent with the micro‐CT results, demonstrating the contribution of FmocFF/HA, either in its lyophilized form or as a hydrogel, to the promotion of osteogenesis and bone ingrowth toward the implant core.

FIGURE 6.

Histological analysis of the regenerated bone. (A–C) Overview images of undecalcified bone sections during preparation, prior to laser microtome removal of metallic components. (D–I) H&E staining of tissues adjacent to the implants. (D–F) Low‐magnification images of entire sections; (G–I) Higher magnification views of selected regions of interest (ROIs). In H&E staining, bone tissue appears dark pink (black arrows), whereas connective tissue exhibits a lighter pink coloration. Inflammatory areas appear purple (yellow arrows). The area previously occupied by the implant appears as a white void, indicated in the magnified images. (J–O) MGT staining to evaluate bone formation, osteoid presence (magenta arrows), and active periosteum. (J–L) Low‐magnification images of entire sections; (M–O) Higher magnification images of specific ROIs. In MGT staining, mature bone appears bright to dark turquoise, osteoid is brown‐orange, loose connective tissue appears blue, dense connective tissue appears bright orange (white arrows), bone marrow stains dark purple, and active periosteum is orange. The former implant space is visualized as a white void and labeled in the enlarged images. Images correspond to the following experimental groups: (A, D, G, J, M) inert implant; (B, E, H, K, N) lyophilized FmocFF/HA‐incorporated implant; (C, F, I, L, O) implant combined with FmocFF/HA hydrogel. (P) Quantification of blood vessel density per ROI (vessels/mm²) (n = 4). (Q) Quantification of woven bone formation (n = 4). One‐way ANOVA was used for all quantitative analyses. ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001. Scale bars: (D–F, J–L) 1000 µm; (G–I, M–O) 200 µm.

In the inert implant group, H&E staining revealed moderate inflammatory infiltrate surrounding the implant (Figure 6G, yellow arrows), suggesting a mild foreign body response. Importantly, no such inflammatory response was observed in either FmocFF/HA‐integrated group, indicating that the bioactive scaffold may exert a protective modulatory influence on local tissue responses (Figure 6H, I).

Prolonged retention of HMW‐HA within the hybrid scaffold may further explain the favorable biological response observed in vivo. HMW‐HA is well known to exert anti‐inflammatory and immunomodulatory effects by promoting macrophage polarization toward a pro‐regenerative M2 phenotype and by limiting the release of pro‐inflammatory cytokines [64, 70, 71]. We have previously demonstrated the immunomodulatory effect of FmocFF/HA hydrogels in vivo, where they increased M2 macrophage prevalence and reduced inflammatory cell infiltration in bone tissue [65]. In contrast, fragmented or LMW‐HA generated during uncontrolled enzymatic degradation acts as a danger‐associated molecular pattern (DAMP), stimulating Toll‐like receptor signaling and promoting inflammation, fibrosis, and delayed healing [70]. The reduced enzymatic cleavage observed in the FmocFF/HA systems (Figure 2K) likely preserved the bioactive HMW‐HA fraction for longer durations, maintaining an anti‐inflammatory microenvironment that facilitated tissue remodeling. This mechanism aligns with the histological findings showing the absence of peri‐implant inflammation and active bone formation in the FmocFF/HA groups (Figure 6). Together, these results suggest that the hybrid scaffold not only provides mechanical and structural support through the FmocFF network but also sustains an immunoprotective niche by stabilizing HMW‐HA and preventing the accumulation of pro‐inflammatory degradation fragments.

In both the lyophilized and hydrogel FmocFF/HA groups, the bone marrow appeared healthy and active, even in regions immediately adjacent to the implant surface. In contrast, markedly less bone marrow was evident around the inert titanium implant. Furthermore, significant new bone formation was observed (Figure 6H, I, black arrows), accompanied by pronounced active periosteum and osteoid formation in both groups containing the FmocFF/HA scaffolds (Figure 6N, O, magenta arrows). These findings suggest that FmocFF/HA scaffolds not only support osteoconduction but may also stimulate osteoinductive processes, enhancing bone remodeling activity near the implant. The hydrogel group exhibited a higher level of direct bone‐implant interactions, as evidenced by more continuous bone tissue in close proximity to the lattice structure (Figure 6I, black arrows). This may reflect an improved diffusion and spatial distribution of the hydrogel within the implant's porous architecture.

Although the inert implant did not actively promote bone formation, its bioinert properties allowed for some degree of bone integration surrounded by dense connective tissue at the periphery of the implant (Figure 6M, white arrows). This limited peripheral integration underscores the importance of bioactive modifications, such as FmocFF/HA incorporation, in achieving substantial bone ingrowth and osseointegration throughout the implant structure.

Quantitative histomorphometric analysis further supported these observations, showing increased blood vessel density in both lyophilized scaffold and hydrogel integrated implants, compared to the inert implant. The hydrogel‐integrated implants exhibited the highest vascularization per mm² (Figure 6P). Woven bone formation followed a similar trend, showing markedly increased new bone area in both bioactive implants, particularly in the hydrogel group (Figure 6Q). These results confirm that incorporation of the bioactive scaffold not only enhanced osteogenesis but also promoted neovascularization within the defect, both essential for sustained bone regeneration.

The elevated vascular density observed in regenerated bone around the bioactive implants suggests that the FmocFF/HA scaffolds promote a more conducive environment for angiogenesis, likely through their immunomodulatory and biomimetic properties, that supports macrophage polarization and vessel sprouting. Enhanced woven bone formation, particularly in the hydrogel group, may result from improved nutrient diffusion and cellular infiltration within the hydrated matrix, accelerating the transition from osteoid to mineralized tissue. Together, these findings demonstrate that FmocFF/HA scaffolds facilitate coordinated vascular and bone tissue regeneration, providing a more integrated and functional repair response than inert titanium alone.

3. Conclusion

This study demonstrates the successful integration of a bioactive FmocFF/HA hybrid hydrogel and its lyophilized scaffold into 3D‐printed porous Ti‐6Al‐4V implants, resulting in enhanced bone regeneration in critical‐size calvarial defects over an eight‐week period. The supramolecular nanofibrillar architecture mimics key features of the ECM, improving cell adhesion and enzymatic stability compared to pristine HA. In vivo, both hydrogel‐integrated implants and lyophilized scaffold‐integrated implants significantly increased bone formation throughout the defect relative to the standard inert implant, with the hydrogel yielding the most pronounced regeneration. Notably, the amount of newly formed bone in the inner region of the implant was approximately doubled in the hydrogel group compared to the inert implant. Together, these findings establish a simple, cell‐free, and growth‐factor‐free strategy for fabricating bioactive implants for personalized bone reconstruction, offering a promising pathway for next‐generation regenerative therapies. Nonetheless, important limitations remain, including the need to establish sterilization protocols compatible with scale‐up supramolecular scaffolds and to determine the long‐term stability of the integrated constructs in vivo.

Notably, the incorporation of the FmocFF/HA hybrid scaffold into a 3D‐printed Ti‐6Al‐4 V lattice provides several distinct translational advantages. First, the metallic framework offers patient‐specific mechanical strength and geometric precision, supporting limb‐salvage and joint‐preservation approaches in cases of extensive bone loss. Second, the FmocFF/HA scaffold exerts immunomodulatory activity, previously shown to promote macrophage polarization toward a pro‐regenerative M2 phenotype, thereby allowing a favorable healing microenvironment [65]. Third, the nanofibrillar network formed through FmocFF self‐assembly closely resembles native ECM without requiring chemical cross‐linking, enabling reproducibility, scalability, and enhanced protein adsorption and cell adhesion. Fourth, the hydrogel provides mechanical rigidity within an osteoinductive range (G′ > 60 kPa), sufficient to activate mechanotransductive signaling pathways [68] that support osteogenic differentiation. Finally, the platform offers substantial translational simplicity: the composite implant is entirely cell‐free, cytokine‐free, and drug‐free, relying solely on endogenous host responses.

4. Materials and Methods

4.1. Materials

Fmoc‐FF was purchased from GL Biochem Ltd. (Shanghai, China). LaserForm Ti Gr23 (A) powder was obtained from 3D Systems (Rock Hill, SC, USA), and 3D‐printed implants were supplied by Sharon Tuvia (1982) Ltd. (Ness Ziona, Israel). Hyaluronidase enzyme powder, dimethyl sulfoxide (DMSO), and fluorescein‐conjugated hyaluronic acid (Fluorescein‐HA) were acquired from Sigma‐Aldrich (Rehovot, Israel). Pre‐filled syringes containing 1% (w/v) HA in phosphate‐buffered saline (PBS) were purchased from BTG‐Ferring (Kiryat Malachi, Israel). Minimum Essential Medium Alpha (MEM‐α), fetal bovine serum (FBS), penicillin–streptomycin solution, and non‐essential amino acids were obtained from Biological Industries (Beit HaEmek, Israel).

4.2. Ti‐6Al‐4 V Lattice Implant Design and Fabrication

The lattice implants were designed using 3DXpert software (3D Systems, Rock Hill, South Carolina, USA) and fabricated from Ti‐6Al‐4 V powder (LaserForm Ti Gr23, 3D Systems) by selective laser melting (SLM) using Flex 350 printer (3D Systems, Rock Hill, South Carolina) at Sharon Tuvia (1982) Ltd. (Ness Ziona, Israel). Two cylindrical structures 50 mm in length and 8 mm in diameter with a gyroid lattice architecture (unit cell size of 5 × 5 × 5 mm) were printed. Post‐processing included standard heat treatment for stress relief, mechanical removal of supports, and dry electropolishing (DLyte, DryLyte Technology, Barcelona, Spain). Subsequently, the cylinder was sectioned into discs 1.8 mm in thickness using an IsoMet 1000 Precision Diamond Saw (Buehler, Chicago, USA). The implants were cleaned and sterilized by autoclaving at 135°C prior to incorporation of the bioactive component, which was performed under sterile conditions. In addition, for in vitro experiments aimed at assessing the penetration of the bioactive material into the core of the implant, cylindrical implants 10 mm in height and 15 mm in diameter were fabricated. These implants were designed to be easily split into two semi‐cylindrical halves, enabling visual and microscopic evaluation of material distribution throughout the implant's core. Fabrication and post‐processing of these implants followed the same SLM parameters and procedures described above.

4.3. Integration of the Peptide‐Polysaccharide Hydrogel

The FmocFF/HA hydrogel was prepared using a solvent‐switch method, as described previously [62, 65]. Briefly, lyophilized FmocFF powder (GL Biochem Ltd., Shanghai, China) was dissolved in DMSO at a concentration of 100 mg/mL. HA (BTG Ltd. Israel) was initially supplied as a 1% (w/v) stock solution and diluted with double‐distilled water (DDW) to a final concentration of 0.167% (w/v). The diluted HA solution was mixed on an orbital shaker for 2 h to ensure complete dissolution. The implants were immersed in the diluted HA solution prior to the addition of the peptide solution. To 1 mL of the prepared HA solution, 50 µL of the FmocFF/DMSO stock was added, yielding final concentrations of 0.5% (w/v) FmocFF and 0.167% (w/v) HA. The mixture was briefly vortexed to promote uniform dispersion and then incubated at 4°C for at least 4 h to allow self‐assembly. For lyophilized samples, hydrogel‐embedded implants were first frozen at −20°C for 1 h, then transferred to −80°C for deep freezing. Implants were subsequently lyophilized overnight while still embedded in the hydrogel matrix, resulting in constructs uniformly filled and coated with lyophilized peptide‐polysaccharide gel.

4.4. High Resolution Scanning Electron Microscopy (HRSEM)

Implants integrated with FmocFF/HA in either hydrogel or lyophilized form were prepared as described in Section 4.3. Following complete gelation, hydrogel samples were fixed in 2.5% (v/v) glutaraldehyde at 4°C overnight. The next day, samples were dehydrated through a graded ethanol series (25%, 50%, 75%, 95%, and 100%), with each step lasting 15 min. The samples were then coated with an 8 nm layer of gold (Au) to enhance surface conductivity.

High‐resolution imaging was performed using a Zeiss GeminiSEM 300 scanning electron microscope (Carl Zeiss, Germany). Elemental analysis of selected regions was conducted using energy‐dispersive X‐ray spectroscopy (EDS) integrated into the HRSEM system.

4.5. Rheological Analysis

Rheological measurements were performed using an AR‐G2 controlled‐stress rheometer (TA Instruments, New Castle, DE, USA). 100 µL of pre‐formed FmocFF/HA hydrogel was loaded onto an 8 mm parallel‐plate geometry with a 1 mm gap at 25°C. Oscillatory strain (0.01%–100%) and frequency (0.01–5 Hz) sweeps were first conducted to identify the linear viscoelastic region (LVR) (Figure S2). Time‐sweep oscillatory measurements were then performed at 0.1% strain and 5 Hz within the LVR using freshly prepared 250 µL hydrogel samples placed under a 20 mm parallel‐plate geometry with a 0.8 mm gap. The storage (G′) and loss (G′′) moduli were recorded as a function of time to evaluate gel stability under constant oscillatory conditions.

4.6. Lyophilized Scaffold Rehydration

HA, FmocFF solution, and FmocFF/HA were incorporated into the implants as outlined in section 4.3, followed by lyophilization. Images of the dry scaffold‐incorporated implants were taken, and then 2 mL of 10% FBS MEM‐Alpha medium was added to each of the samples. The samples were incubated inside a cell culture incubator at 37°C and 5% CO₂. Images were taken after 24 h and 7 days of incubation, using a zoom stereomicroscope (Nikon SMZ800N) equipped with a DS‐Fi2 camera (Nikon, Japan).

4.7. In Vitro Enzymatic Degradation by Hyaluronidase

Bovine testicular hyaluronidase (Type IV‐S, lyophilized powder; Sigma‐Aldrich, Rehovot, Israel) was reconstituted in 100 mm sodium acetate buffer supplemented with 2 mm CaCl₂. To enable fluorescence‐based monitoring of HA degradation, Fluorescein‐HA (Sigma‐Aldrich, Rehovot, Israel) was incorporated into the HA solution at a concentration consistent with that of the original unlabeled HA. Lyophilized FmocFF/HA‐integrated implants and FmocFF/HA hydrogel‐integrated implants were incubated in enzyme solution containing 200 U/mL hyaluronidase at 37°C to evaluate enzymatic degradation kinetics. All samples were maintained in the dark throughout the experiment to minimize photobleaching of the fluorescent probe. At predefined time intervals, the enzyme solution was carefully removed and replaced with fresh aliquots to ensure sustained enzymatic activity. The collected supernatants were analyzed for fluorescence intensity at 485 nm excitation and 535 nm emission using a TECAN Spark microplate reader (Tecan, Switzerland). Implants containing pure HA hydrogels served as positive controls for enzymatic degradation.

4.8. In Vitro Hydrogel Mass Loss in the Presence of HyaluronidaseS

The degradation of the hydrogels was evaluated by measuring their mass loss in the presence of hyaluronidase. Pre‐weighed 35 mm culture plates were filled with 1 mL of freshly prepared HA, FmocFF, or FmocFF/HA hydrogels, and the combined weight was recorded. Each sample was then immersed in 2 mL of hyaluronidase buffer and incubated under standard cell culture conditions (37°C, 5% CO₂, humidified atmosphere). At designated time points, the buffer was carefully removed, the remaining hydrogel and plate were weighed, and 2 mL of fresh hyaluronidase buffer was added. This process was repeated daily for 14 days. The percentage of remaining mass was calculated relative to the initial hydrogel weight.

4.9. Examination of the Cell‐Scaffold Interaction

Human osteosarcoma MG‐63 cells (ATCC—CRL‐1427 ) were cultured in MEM‐α medium supplemented with 10% FBS, 1% penicillin–streptomycin, and 1% non‐essential amino acids. Inert, lyophilized‐integrated, and hydrogel‐integrated implants were prepared as described above. Each implant was placed in a well of a 12‐well cell‐repellent culture plate, and 2 × 10⁵ cells were seeded onto each sample.

After 48 h of incubation, the culture medium was removed, and the samples were washed twice with PBS, followed by overnight fixation in 2.5% glutaraldehyde at 4°C. The next day, samples were washed twice with PBS and dehydrated through a graded ethanol series (25%, 50%, 75%, 95%, and 100%; 10 min at each step). Critical point drying was then performed. For imaging, samples were sputter‐coated with an 8 nm gold layer and analyzed using a Zeiss GeminiSEM 300 high‐resolution scanning electron microscope (Carl Zeiss, Germany).

4.10. Protein Adsorption Quantification (BCA Assay)

Protein adsorption on the different implants was quantified using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific, USA) following the manufacturer's instructions. Inert titanium implants, implants integrated with lyophilized FmocFF/HA scaffolds, and implants integrated with FmocFF/HA hydrogels were incubated in standard cell culture medium containing 10% FBS at 37°C for 3 h. Following incubation, samples were gently washed with PBS to remove non‐adherent proteins. Adsorbed proteins were then extracted by incubating each sample in PBS containing 0.5% Triton X‐100 for 1 h at room temperature under gentle agitation. The collected supernatants were analyzed for total protein content using the BCA assay with bovine serum albumin (BSA) as a standard. Absorbance was measured at 562 nm using a microplate reader (Tecan Infinite 200 PRO, Switzerland). Blank controls consisting of untreated implants and scaffolds processed in the same buffer were used to correct for background absorbance.

4.11. Cell Viability and Proliferation

Osteoblast‐like MG63 cells (ATCC, CRL‐1427) were cultured in MEM‐α medium supplemented with 10% FBS, 100 U mL⁻¹ penicillin, and 100 U mL⁻¹ streptomycin (Sartorius, Israel). Inert implants, lyophilized scaffold (either HA, FmocFF or FmocFF/HA) integrated implants, and hydrogel (HA, FmocFF or FmocFF/HA) integrated implants were placed into each well of a 12‐well plate. The implants were sterilized for 30 min by UV exposure, followed by washing with the corresponding culture medium for 2 h at 37°C. MG63 cells were then seeded onto the implants at a density of 50 000 cells per well in 1 mL media. At 24 h, 72 h, and 1 week, the metabolic activity of the cells was assessed by replacing the culture medium with Alamar Blue reagent (Enco, Israel) diluted 1:9 in the medium and incubating for 4 h at 37°C in 5% CO₂. Absorbance measurements were performed using a Tecan Spark plate reader at two wavelengths, 570 and 600 nm. The percentage reduction of Alamar Blue was calculated according to the manufacturer's instructions. Results were normalized by comparing the Alamar Blue reduction values to those obtained from cells cultured on standard tissue culture plastic. The results are presented as the percentage of viable cells relative to the control group, which consisted of cells cultured in a standard 12‐well plate without hydrogels.

4.12. Alkaline Phosphatase (ALP) Quantification

MC3T3‐E1 murine pre‐osteoblasts were seeded on the different implant groups similarly to the MG63 cells described in Section 4.11. After 48 h, osteogenic differentiation was induced by replacing the culture medium with α‐MEM supplemented with 10% FBS, 1% penicillin/streptomycin, 50 µg/mL ascorbic acid, and 10 mm β‐glycerophosphate (differentiation medium). The differentiation medium was refreshed every two days for 8 days. At the end of the incubation period, cells were washed with PBS and lysed in 0.2% Triton X‐100 for 30 min on ice to extract intracellular enzymes. ALP activity was quantified using the fluorogenic substrate 4‐methylumbelliferyl phosphate (4‐MUP; Sigma‐Aldrich) by mixing equal volumes of lysate and substrate solution and incubating for 30 min in the dark at room temperature. Fluorescence was measured using a microplate reader (Ex = 360 nm, Em = 450 nm), and ALP activity was normalized to the number of cells in each well based on cell nuclei staining using DAPI measured via fluorescence (Ex = 360 nm, Em = 461 nm).

4.13. Alizarin Red Staining for Mineral Deposition Detection

Matrix mineralization was assessed using the Alizarin Red staining assay. MC3T3‐E1 pre‐osteoblasts were seeded and differentiated as described in Section 4.12. The differentiation medium was refreshed every two days for 14 days. On day 14, the cells were washed once with cold PBS and fixed in 70% ethanol for 1 h at 4°C. After two washes with ultrapure water, cells were stained with 40 mm Alizarin Red (Sigma, A5533; pH 4.2) for 30 min at room temperature. Excess dye was removed by washing with cold distilled water, and plates were allowed to dry overnight. Bound dye was solubilized in a 1:2 (v/v) mixture of acetic acid and methanol. Absorbance was measured at 405 nm using a microplate reader. Data was normalized to the number of cells in each well based on cell nuclei staining with DAPI measured via fluorescence (Ex = 360 nm, Em = 461 nm). The control group represents cells that were cultured simultaneously without osteogenic media.

4.14. In Vivo Rabbit Calvarial Critical Size Bone Defect Model

Adult female New Zealand White rabbits (2.7–3.3 kg) were subjected to calvarial bone defects. The experimental procedures involving live vertebrates in this study were approved by the institutional committee of animal care and use (IACUC) of Tel Aviv University (no. TAU—MD—IL – 2306 – 143 – 4), and all experiments were strictly conducted in accordance with the approved guidelines. Animals were individually housed in the Central Animal Facility of Tel Aviv University. Animals were fed with Teklad Global Rabbit Diet (Envigo, Madison, Wisconsin), autoclaved hay, and tap water ad libitum.

Surgical procedure was performed under general anesthesia following sedation with 5 mg/kg subcutaneous (SC) xylazine (Sedaxylan Veterinary, Eurovet Animal Health BV, Bladel, The Netherlands) and 35 mg/kg SC ketamine (Clorketam, Vetoquinol, Lure, France). Oxygen was delivered by means of a facemask connected to an anesthetic machine with an oxygen flowmeter (model Mix4; Foures SAS, Bordeaux) at a rate of 2 L/min. Blood oxygen saturation, pulse, and body temperature were monitored using a PhysioSuite monitor (Kent Scientific). Once anesthetized, the surgical area was shaved and disinfected with iodine solution, and local anesthesia was administered with 2% lidocaine hydrochloride and norepinephrine (1:100 000) for the reduction of pain and hemostasis. The rabbit calvarium was exposed via a U‐shaped incision. The periosteum was carefully separated from the bone, and skin flaps were elevated to expose the calvarium. Three circular defects 8 mm in diameter were made in the bone under saline irrigation. To mark the dimensions of the defects, an 8‐mm trephine bur was used, followed by a gentle bone removal process using an inverted conical diamond bur. Attention was given to avoid damaging the dura mater. The three treatment modalities were randomly allocated for all 27 defects, with three defects per animal: 1. Inert 3D‐printed implant. 2. FmocFF/HA “wet” hydrogel‐integrated 3D‐printed implant. 3. FmocFF/HA “dry” lyophilized‐integrated 3D‐printed implant. To assess the non‐unity of the defect, four defects remained unfilled. The conditioned implants were prepared as described in Section 4.3. To keep the implant sterile during preparation, titanium implants were first sterilized by autoclaving at 132°C. After autoclaving, hydrogel or lyophilized scaffold integration was performed under sterile conditions in a biosafety cabinet, followed by UV irradiation of both sides of the implant for 30 min each to ensure surface sterility. The sterilized implants were stored in a closed sterile container until implantation. Representative images of the surgical procedure workflow are shown in Figure S5.

Eight weeks after the implantation, the rabbits were euthanized in accordance with ethical guidelines. The calvaria from each rabbit was harvested and fixed in 10% neutral buffered formalin (Sigma‐Aldrich, Rehovot, Israel) for subsequent micro‐computed tomography (micro‐CT) and histological analyses.

4.15. Micro‐CT Analysis of the Bone and the Implants

All calvaria were scanned using Micro‐CT (XT H 225 ST, Nikon Metrology NV, Leuven, Belgium), equipped with a 225 kV 225 W reflection target. Scans were performed at an isotropic resolution of 25 µm utilizing the following parameters: 180 kVp energy, at 133 µA intensity, a 0.5 mm Tin filter, with 1570 projections utilizing a 354 msec exposure time. Raw scans underwent reconstruction procedure using the Nikon CT Pro 3D software (v. 6.9.1; Nikon Metrology NV, Leuven, Belgium) and subsequent segmentation using various semiautomatic tools based on grayscale thresholds with appropriate manual refinement prior to the analysis in Amira software (v. 6.3, www.fei.com). Each scan was segmented into bone and implant, with the latter clearly visible and morphologically distinguishable from the surrounding bone. Circular regions of interest (ROI) 8 mm in diameter were identified via the surgical borders and subjected to volumetric analysis. Quantitative morphometric analysis included the calculation of the trabecular number and trabecular separation within defined regions. Each 8 mm circular defect was virtually divided into two regions (outer and inner) using two coaxial 3D cylinders of varying radii (modelled to divide the defect area into two identical volumes). The circular ROI was subjected to adjacent intact bones to serve as a reference for a normal bone. Volumes of restored bone were calculated as the ratio of bone volume present within each defect ROI to the bone volume in an adjacent, pristine bone that served as a control. All 3D visualizations present in this work were generated using Blender software (version 4.3).

4.16. Histological Processing and Analysis of the Bone

Histological processing was carried out at Patho‐Logica Laboratories (Ness Ziona, Israel). Samples were dehydrated and embedded in Spurr epoxy resin for non‐decalcified sectioning. Plastic blocks were trimmed and sectioned using a laser microtome (TissueSurgeon, LLS ROWIAK LaserLabSolutions, Hannover, Germany), employing a femtosecond near‐infrared laser for precise, non‐contact cutting of implant‐containing hard tissues with minimal material loss. Serial transverse sections, 12 µm thick, were prepared, mounted on glass slides, and stained using Hematoxylin and Eosin (H&E) for general morphology and with Masson‐Goldner Trichrome (MGT) for connective tissue visualization. Slides were scanned using a KF‐BIO‐40 slide scanner. Quantitative evaluation of the histological sections included assessment of blood vessel density and woven bone formation across representative regions of interest (ROIs).

Quantitative evaluation included measuring blood vessel density and woven bone formation within representative regions of interest (ROIs) located within the defect area. ROIs were delineated on the scanned digital slides using the calibrated scale bar, ensuring consistent and non‐overlapping sampling across groups. Blood vessels were identified morphologically as circular or elongated luminal structures lined by endothelial cells, with or without intraluminal erythrocytes. All vessels within each ROI were manually counted, and vessel density was calculated as the number of vessels per mm².

Woven bone formation was assessed on MGT‐stained sections by a pathologist blinded to the treatment groups. Newly formed woven bone was distinguished from mature lamellar bone and fibrous tissue based on its characteristic trichrome staining pattern and immature trabecular architecture. The area occupied by woven bone within each ROI was outlined and quantified digitally, and values were expressed as the percentage of the defect area filled with newly formed bone.

4.17. Statistical Analysis

Statistical analyses were performed using GraphPad Prism (version 10.4.1, GraphPad Software, San Diego, CA, USA). For comparisons involving more than two groups, one‐way or two‐way analysis of variance (ANOVA) was used as appropriate, followed by Tukey's post hoc test to assess differences between individual groups. Statistical significance was defined as p < 0.05. All quantitative data are presented as mean ± standard deviation (SD). Significance levels are indicated in the figures as follows: ^* p < 0.05, ^** p < 0.01, and corresponding group comparisons are annotated with symbols as defined in the respective figure legends.

Funding

This work was supported by the European Research Council (ERC), under the European Union's Horizon 2020 research and innovation program (grant agreement no. 948 102) (L.A.‐A.), ERC‐2023‐POC under the Horizon Europe (Grant agreement ID: 101123407)(L.A.‐A.), and by the Israel Ministry of Science and Technology (Grant number 3–17971) (L.A.‐A. and S.D.).

Author Contributions

L.A.A., N.R., S.D., and E.G. conceived and designed the experiments. N.R., I.B., and M.H.S. planned and performed the experiments. N.R., R.S., and A.P. analyzed the in vivo results. N.R. and L.A.A. wrote the manuscript. All authors discussed the results, provided intellectual input and critical feedback, and commented on the manuscript.

Conflicts of Interest

The authors declare no conflict of interest.

Supporting information

Supporting file: adhm70721‐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.