Abstract

   Porous calcium phosphate ceramics have attracted widespread attention
   owing to their excellent bioactivity. However, their poor mechanical
   properties severely limit their clinical applications. Significantly
   improving the mechanical strength of porous CaP ceramics while
   maintaining their bioactivity remains a major challenge. To address
   this issue, calcium sulfate is used to regulate the directional growth
   of hydroxyapatite grains during ceramic sintering. The in situ oriented
   grains can not only alleviate the stress concentration but also
   strengthen the bonding force between the ceramic grain boundaries.
   Calcium sulfate improves the release of active calcium ions from
   calcium phosphate ceramics, further enhancing their bioactivity and
   osteoinductivity in vivo. Transcriptome and proteome sequencing reveals
   that the in situ whisker‐reinforced ceramics increase the expression of
   proteins related to calcium ion binding and promote the expression of
   osteogenesis‐related proteins. In the supercritical bone defect repair
   model, repair of the defect is achieved within 3 months, with
   mechanical recovery reaching more than 70% of the autologous bone.

   Keywords: bioceramic, grain growth regulation, mechanical enhancement,
   osteoinduction supercritical bone defect
     __________________________________________________________________

   This study induced the in situ oriented grain growth of 3D‐printed
   porous calcium phosphate ceramic through calcium sulfate, resulting in
   a mechanically enhanced microstructure, which increased the compressive
   strength of the porous calcium phosphate ceramic by 5–10 times.
   Additionally, the release of active calcium ions is enhanced, further
   optimizing the bioactivity for regenerative repair of supercritical
   bone defects.

   graphic file with name ADVS-12-2408459-g011.jpg

1. Introduction

   Calcium phosphate (CaP) ceramics, such as hydroxyapatite (HAP),
   tricalcium phosphate (TCP), and biphasic calcium phosphate (BCP), have
   been employed as synthetic materials for autologous bone owing to their
   stable physicochemical properties and excellent biocompatibility,
   osteoconductivity, and osseointegration.^[ [42]^1 , [43]^2 ^] However,
   the poor mechanical properties of porous CaP ceramics severely restrict
   their clinical applications. The compressive strength of bioactive CaP
   ceramics is often less than 3 MPa, which is far less than that of
   cortical bone (88–164 MPa).^[ [44]^3 , [45]^4 ^] Therefore, bioactive
   CaP ceramics are primarily used to repair non‐load‐bearing bone defects
   or as coating and filling materials.^[ [46]^5 , [47]^6 ^]

   A highly porous structure can ensure the bioactivity of CaP.^[ [48]^7
   ^] However, the mechanical strength of the ceramic decreased
   exponentially with increasing porosity.^[ [49]^8 ^] Additionally, the
   capillary micropores (<20 µm) on the pore wall affect the adsorption of
   growth factors and other proteins that regulate cytological
   behaviors.^[ [50]^9 , [51]^10 ^] In ceramic sintering, it is ideal to
   maintain porous/microporous structures and avoid complete densification
   caused by high temperature and high pressure; however, these
   porous/microporous structures are often stress concentration points.^[
   [52]^11 , [53]^12 ^] Therefore, improving the mechanical properties of
   porous CaP while maintaining its bioactivity remains challenging.

   Recent studies on the mechanical enhancement of bioactive CaP ceramics
   mainly focused on two aspects: reducing stress concentration and
   preventing crack propagation. Combined with finite element mechanical
   simulations, the shape and porous structure of the ceramic can be
   optimized to reduce the structural stress concentration, thus enhancing
   the mechanical strength.^[ [54]^13 , [55]^14 ^] In recent years, with
   the emergence of ceramic 3D printing technologies, such as digital
   light projection (DLP), 3D printed CaP ceramics with customizable
   shapes and internal porous structures have created new possibilities
   for mechanical optimization and precise repair of bone defects.^[
   [56]^15 , [57]^16 , [58]^17 , [59]^18 ^] However, this engineering
   method has certain limitations because of the intrinsic poor mechanical
   characteristics of ceramic materials. Second‐phase reinforcement is the
   most widely researched ceramic enhancement method. Its principle is to
   add a second phase, such as bioinert whiskers, lamellae, fibers, or
   particles, to the ceramic matrix to slow down crack propagation, thus
   improving mechanical properties.^[ [60]^19 ^] Some materials with high
   mechanical strength and biological stability, such as alumina,
   zirconia, titanium dioxide, and magnesium oxide, have been used as
   reinforcement phases for CaP ceramics.^[ [61]^20 , [62]^21 , [63]^22 ^]
   Unfortunately, the externally added second phase does not conglutinate
   in porous ceramics, resulting in limited mechanical improvement. In
   addition, the second phase is generally non‐bioactive, which might
   significantly deteriorate the bioactivity and degradability of CaP
   ceramics.

   Another avenue to enhance the mechanical properties of porous CaP
   ceramics is through regulating the growth of ceramic grains. In a
   hydrothermal environment, the HAP grain growth depends on the external
   environment (pH, temperature, and ions in solution), resulting in
   different morphologies, such as needle‐like, short rod‐like, and
   lamellar grains.^[ [64]^23 , [65]^24 ^] Similarly, in solid states, the
   grain morphology of HAP can be changed by the presence of potassium
   chloride (KCl), sodium chloride (NaCl), potassium sulfate (K[2]SO[4]),
   calcium sulfate (CaSO[4]), and calcium fluoride (CaF[2]).^[ [66]^25 ,
   [67]^26 , [68]^27 , [69]^28 ^] The binding energies between each
   crystal face of HAP and these substances affect the crystal growth
   rate, enabling oriented growth of HAP grains and improving the
   mechanical properties of the ceramic.^[ [70]^29 ^]

   Therefore, this study aimed to explore a novel research scheme that can
   simultaneously improve the mechanical properties and bioactivity of
   CaP. High‐precision porous ceramic green bodies were prepared using a
   bioink composed of CaSO[4]/HAP via DLP printing technology. CaSO[4] was
   chosen because it has been successfully applied in clinical practice,
   and its biosafety has been widely verified.^[ [71]^30 , [72]^31 ^]
   During the sintering process, CaSO[4] regulates the in situ oriented
   growth of the internal HAP crystals, forming a mechanically enhanced
   microstructure to reinforce the porous CaP. In addition, CaSO[4] can
   modulate the release of active calcium ions (Ca^2+) from porous CaP,
   further enhancing the bioactivity of the CaP ceramic.

2. Results

2.1. Fabrication of In Situ Whisker‐Reinforced Ceramics

   First, a high‐precision ceramic green body was fabricated using the
   HAP‐CaSO[4] bioink through DLP technology. In the sintering process,
   CaSO[4] could regulate the oriented growth of HAP grains to form an
   in‐situ whisker structure (Figure [73] 1a). Compared with externally
   added second phases, CaSO[4] has less impact on the printing process
   and printing accuracy. The prepared porous ceramic scaffold is shown in
   Figure [74]1b,c, which has a clear outline and porous structure. To
   observe the in situ oriented growth of the grains, a system composed of
   50 wt% CaSO[4] + 50 wt% HAP (referred to as 50 SH) was first studied.
   The 3D printed 50 SH ceramic green body was maintained at 900 °C for
   different times (2, 6, 8, and 10 h) to induce whisker formation and
   then at 1150 °C for 2 h to complete the sintering, as shown in Figure
   [75]S1 (Supporting Information). After 2 h at 900 °C, whiskers formed
   within the HAP matrix with a length of ≈2 µm. After 6 h at 900 °C, a
   large number of short rods appeared inside the matrix, exhibiting
   distinct whisker characteristics, with some whiskers reaching a length
   of 4 µm. After 8 h, the whiskers grew further. When the holding time
   reached 10 h, the matrix was fully whiskerized, leading to clean
   prismatic whiskers with diameters of ≈1–2 µm and lengths up to 10 µm.
   Compared to pure HAP, no significant changes were observed in the
   matrix morphology, indicating that long‐term heating at 900 °C did not
   alter the morphology of the original material. These phenomena
   illustrate that CaSO[4] can affect the growth of the internal grains at
   high temperatures; therefore, it is promising to regulate the grain
   morphology for mechanical optimization. However, the high CaSO[4]
   content can degrade the stability of the entire material, and the
   entire matrix of 50 SH underwent whiskerization and densification.
   Therefore, it is necessary to optimize the ceramic composition and
   microstructure.

Figure 1.

   Figure 1
   [76]Open in a new tab

   Fabrication of in situ whisker‐reinforced ceramics (IWRC). (a) The
   regulation process to in situ grow‐oriented HAP grains. CaSO[4], an
   active component capable of modulating HAP crystal growth, was
   incorporated into the printing slurry to circumvent the impact of
   external whisker addition on the precision of DLP printing.
   Subsequentially, during the sintering process, the oriented crystals
   could reinforce the HAP structures. (b,c) Photographs and
   microstructure of IWRC. The printed scaffolds retain personalized
   external geometries and high‐precision porous architectures. (d–g)
   Microstructural morphologies of HAP, 10, 20, and 30 SH. The HAP group
   consists of oval‐shaped grains and micropores. With the increasing
   calcium sulfate content, distinct rod‐like whisker structures appear
   within the ceramic matrix (labeled with purple), and the original
   picture is shown in Figure [77]S3 (Supporting Information). (h,i) EDS
   mapping analysis of the 20 SH sample. (j) X‐ray diffraction (XRD)
   pattern of the IWRC with main characteristic peaks corresponding to
   those of CaSO[4] and HAP.

   3D‐printed ceramic green bodies containing 5, 10, 20, and 30 wt%
   CaSO[4]in HAP were prepared. After debinding, the samples were
   maintained at 900 °C for 10 h and subsequently sintered to obtain
   whiskerized ceramic samples (referred to as 5, 10, 20, and 30 SH
   respectively), with their microstructures shown in Figure [78]1d–g. The
   morphology of 5 SH was similar to that of HAP, and no whisker
   structures were observed. In 10 SH, distinct short rod‐like whiskers
   emerged with lengths of ≈3–5 µm and diameters of ≈1 µm. With the
   increase in CaSO[4] content, the whisker features were more pronounced.
   In 30 SH, whiskers grew to 10 µm long and were uniformly distributed in
   the ceramic and closely combined with the matrix. Energy Dispersive
   X‐ray Spectroscopy (EDS) line scans were used to analyze the elemental
   composition of individual whiskers. The whiskers were composed of
   calcium (Ca), phosphorus (P), and oxygen (O), and no sulfur (S) element
   was detected (Figure [79]S2, Supporting Information). The atomic molar
   ratio of Ca to P is ≈1.57, which is close to that of HAP. In the EDS
   mapping, owing to the presence of CaSO[4], S appeared in the ceramic
   matrix with an atomic molar ratio of ≈4.12%, and the atomic molar ratio
   of Ca to P rose to 2.08. All elements were evenly distributed in the
   ceramic matrix (Figure [80]1h,i). In the XRD pattern, the
   characteristic peaks of both CaSO[4] and HAP were present without
   significant shifts, and no other impurity peaks were observed. This
   indicated that the material was composed solely of CaSO[4] and HAP
   crystals (Figure [81]1j). Combined with the EDS analysis, the generated
   whiskers were determined to be HAP.

2.2. Mechanical Properties

   To understand the effect of the in situ developed whisker structure on
   the mechanical properties of the ceramic, solid ceramics and porous
   ceramics with a porosity of 68% were 3D printed for compression testing
   (Figure [82] 2a–d). The stress‐strain curves of the porous ceramics
   showed that all scaffolds exhibited brittle fracture, i.e., the stress
   decreased rapidly after material failure. The in situ developed
   whiskers significantly strengthened HAP. In the solid pieces (the print
   model was a solid cylinder), the compressive strength of HAP was
   8.87 MPa, whereas that of 30 SH increased by ≈10 times to 93.12 MPa. In
   porous ceramics, the compressive strength of HAP was only 2 MPa,
   whereas 30 SH exhibited a strength of 10.3 MPa, which represented an
   approximate five‐fold enhancement. The elastic moduli of the four
   materials were calculated based on the stress‐strain curves, revealing
   that pure HAP,10, 20, and 30 SH had elastic moduli of 273.30 ± 1.7,
   433.17 ± 2.55, 515.71 ± 2.02, and 564.16 ± 2.68 MPa, respectively.
   Owing to the inadequate mechanical strength of porous HAP ceramics,
   they were prone to fracture during surgical implantation and break into
   numerous fragments, leading to implant failure. With the reinforcement
   of in situ grown whiskers, the 20 SH scaffold remained intact when
   subjected to a compressive force of the hemostatic forceps, whereas the
   HAP scaffold easily fractured, as illustrated in Figure [83]2e and
   Movie [84]S1 (Supporting Information). This demonstrates that the
   whisker‐reinforced ceramic scaffolds possess the strength required to
   withstand normal surgical manipulations.

Figure 2.

   Figure 2
   [85]Open in a new tab

   Mechanical properties of IWRC. (a) Stress‐strain curves of the four
   porous scaffolds. (b,c) Compressive strength of solid and porous
   ceramic scaffolds. (d) Elastic modulus of porous ceramic scaffolds. (e)
   Visual comparison of the mechanical reinforcement of the scaffolds
   demonstrating that the enhanced scaffolds can withstand the bite of
   surgical forceps, as shown in Movie [86]S1 (Supporting Information).
   (f,g) Nano‐mechanical testing of HAP, where a probe scanned a 5 × 5 µm
   area to obtain a surface height map. Six points were selected at both
   grain boundaries and within grains (f) to measure individual
   force‐displacement curves (g). The Young's modulus was calculated based
   on these curves (h). (i,k) Nano‐mechanical properties of 20 SH. (l) A
   3D surface height map of 20 SH and the corresponding nano‐mechanical
   map of the entire area (5 × 5 µm) (m), where the modulus within grains
   is significantly higher than at the grain boundaries. (n,o) Microscopic
   morphologies of HAP and 20 SH after fracture, with 20 SH showing
   numerous trans‐granular fractures.

   The nanomechanical properties of the materials were further
   investigated using atomic force microscopy (AFM), in which the
   force‐displacement curves were measured at three distinct points across
   the grains and grain boundaries for both the HAP and 20SH samples
   (Figure [87]2f–m). The Young's modulus within the crystal was
   significantly higher than that at the grain boundaries. The
   nanomechanical mapping of the material confirmed that the forces at the
   ceramic grain sites were greater than those at the grain boundaries.
   Within the HAP sample, the Young's modulus within the grains was
   ≈21.8 GPa, while it was 8.5 GPa at the grain boundaries, yielding a
   ratio of 0.39. In contrast, in the whisker‐reinforced sample, the
   modulus values were increased to 38.5 GPa (within the grains) and
   30.2 GPa (at the grain boundaries), and the ratio was increased to
   0.78. Compared to HAP, the grain boundaries in 20 SH were significantly
   reinforced, possibly because CaSO[4] accelerated CaP grain fusion,
   which also contributes to the mechanical properties of the scaffold.
   The fracture morphology (Figure [88]2n,o) indicated that the
   whisker‐reinforced ceramics exhibited both trans‐granular and
   intergranular fracture modes, whereas pure HAP ceramics only
   demonstrated intergranular fractures. Trans‐granular fractures can
   consume more energy and enhance mechanical strength more effectively.

2.3. Mechanical Enhancement Mechanisms

   To further understand the mechanism of in situ whisker reinforcement,
   we designed a simplified model of a 2D monolayer structure based on the
   microstructural characteristics of the ceramics. The original HAP
   grains were simplified to spherical shapes, and the cylindrical
   whiskers were simplified to columnar shapes (Figure [89] 3a). Using
   static analysis, a compressive stress of 2 MPa was applied to the
   surface. The calculated Mises stress for the structure without whiskers
   was 60.138 MPa, whereas the Mises stress for the structure with
   whiskers decreased to 18.643 MPa. The compressive stress of the
   scaffold without and with whiskers are 30.989 and 9.173 MPa,
   respectively. The presence of whiskers significantly reduced the stress
   concentration under the same conditions. Furthermore, the stress
   contour plots showed that the whisker structure effectively alleviated
   the stress concentration at the grain boundaries, regardless of whether
   the pores were filled. A shear stress of 2 MPa was applied to the upper
   surface of the scaffold. As observed from Figure [90]3c, the presence
   of whiskers significantly improved the deformation resistance of the
   scaffold, and the average Mises stress decreased from 7.38 MPa for the
   scaffold without whiskers to 5.29 MPa for the scaffold with whiskers.
   The stress contour plots showed significant stress concentration in the
   micropores of the scaffold without whiskers.

Figure 3.

   Figure 3
   [91]Open in a new tab

   Finite element analysis (FEA) of mechanical enhancement mechanisms. (a)
   FEA model constructed from Scanning Electron Microscope (SEM) images.
   The original HAP grains were simplified as spherical shapes, and the
   cylindrical whiskers were simplified as columnar shapes. (b)
   Compression simulation of two structures. (c) Shearing simulation of
   two structures (d) stress‐strain curves. (e–g) Failure simulation of
   two structures, stress distribution in compression (e), where the group
   with whiskers had a uniform stress distribution on both sides of the
   micropores, the morphology after compression failure (f), the time of
   failure during compression (g), which was 14.12 s for HAP and 24.6 s
   for the in situ whisker‐reinforced scaffold.

   We employed a transient model to simulate the material failure during
   compression. The scaffold base was fixed, and a continuous deformation
   of 0.5 µm s^−1 was applied at the top with a time step of 1 s
   (Figure [92]3e). In the whisker‐reinforced scaffold, the stress was
   evenly distributed on both sides of the pores, effectively delaying
   material failure. The in situ whisker‐reinforced scaffold started to be
   damaged from the 11th second and completely failed at 24.6 s. However,
   in the scaffold without whiskers, the stress was mainly concentrated on
   one side of the pore, and the damage started in the 8th second, with
   the scaffold completely failing within 14 s. Bioceramics, unlike
   structural ceramics, require abundant micropores to ensure
   biocompatibility. However, in terms of the mechanical structure, the
   edges of these micropores often become stress concentration points to
   initiate cracks, severely compromising the mechanical performance of
   the scaffold. According to our stress simulation results, the columnar
   whisker structure within the ceramic matrix effectively alleviated the
   stress concentration, delaying the occurrence of damage and improving
   the mechanical performance of the scaffold. Additionally, when cracks
   appeared, whiskers effectively impeded crack propagation, thereby
   enhancing the mechanical performance of the ceramics.

2.4. Proliferation and Differentiation

   The composition and degradation products of in situ whisker‐reinforced
   ceramics (IWRC) are vital for regulating its biocompatibility and
   biological functions. HAP is a stable phase of CaP with a slow
   degradation rate in the body. However, CaSO[4] degrades more rapidly
   and does not match new bone growth. The addition of CaSO[4] is expected
   to improve the degradation rate of HAP ceramics. As shown in
   Figure [93] 4a, the degradation rate of the IWRC was proportional to
   the content of CaSO[4]. The mass of HAP remained stable, whereas the
   mass of 30 SH continuously decreased and reached a mass change of ≈10%
   on the 14th day. The dissolution rate of CaSO[4] in water was
   significantly higher than that of HAP, increasing the release of Ca^2+
   from the ceramic. The 5 SH and HAP scaffolds exhibited consistently low
   Ca^2+ release throughout the testing period. The release rates of 20
   and 30 SH were significantly higher than the rest, reaching saturation
   on the 7th day. The saturation concentrations for 20 and 30 SH were
   ≈230 and 320 mg L^−1, respectively (Figure [94]4b). The release pattern
   of S was similar to that of Ca^2+, reaching saturation on the 7th day.
   The saturation concentrations of sulfate ions for 20 and 30 SH were
   ≈190 and 230 mg L^−1, respectively (Figure [95]4c), indicating that the
   initial release of Ca^2+ was mainly attributed to CaSO[4]. However, in
   practical applications, once bone damage occurs, hypoxia and
   inflammatory responses could cause the pH at the defect site to shift
   from weakly alkaline to acidic. Additionally, following the defect,
   osteoclasts became activated and adhered to the damaged area, forming
   resorption cavities. The osteoclasts secreted hydrogen ions (H+) into
   these cavities through proton pumps, leading to an acidic environment
   within the cavities. This acidic microenvironment accelerated the
   degradation of calcium sulfate/hydroxyapatite scaffolds and further
   promoted the release of minerals such as calcium and phosphorus.

Figure 4.

   Figure 4
   [96]Open in a new tab

   In vitro evaluation of 3D‐printed IWRC with BMSCs. (a) In vitro
   degradation rate of IWRC. (b–d) The ion release rate of calcium,
   phosphorus, and sulfur ions. (e) Top‐down view, laser‐scanning confocal
   reconstructions of live (green) and dead (red) stains on days 1, 3, and
   5. (f) Alkaline phosphatase (ALP) staining 10 days after osteoinductive
   culture with BMSCs, where dark blue is positive expression of ALP.
   Alizarin red staining 14 days after osteoinductive culture with BMSCs,
   where red spots were calcium nodules. (g) Immunofluorescence staining
   14 days after osteoinductive culture with BMSCs, nucleus (blue), Runx2
   (red), Col1 (green), OCN (red), and OPN (green). (h) OD values of the
   BMSCs in the culture for 1, 3, and 7 days by the CCK‐8 kit (n = 6). (i)
   Semi‐quantitative analysis of ALP on days 4, 7, and 10 (n = 3) (^* p <
   0.05, ^** p < 0.01, ^*** p < 0.001, and ^**** p < 0.0001, ns, not
   significant).

   In vitro biocompatibility assessments of the whisker‐reinforced and HAP
   scaffolds were conducted using bone marrow stromal cells (BMSCs). The
   CCK‐8 assay revealed significant cell proliferation on all scaffolds.
   The optical density (OD) values for 30, 20, and 10 SH at each testing
   interval were similar to those of the positive blank control and HAP.
   This indicated that the addition of 10%–30 wt.% CaSO[4] did not exert
   cytotoxic effects on the stem cells (Figure [97]4h). The BMSC
   proliferations on the four ceramic scaffolds were further compared
   using laser scanning confocal microscopy (Figure [98]4e). On day 1
   post‐seeding, the cell count was low, and most cells showed spherical
   shapes without spreading. By day 3, the cells had spread, exhibiting
   spindle shapes with elongated pseudopodia. The cell count substantially
   increased across all four materials, and the morphology of the
   scaffolds began to emerge. Sparse dead cells were observed in 30 SH due
   to normal cellular apoptosis. By day 5, the cell numbers were further
   augmented, and cell death was reduced.

   After seeding BMSCs onto the scaffold for osteogenic induction, the
   effect of the IWRC on the osteogenic differentiation of BMSCs was
   monitored. Alkaline phosphatase (ALP), which induces early osteogenic
   differentiation, was assessed on day 10 by ALP staining. For 30 and 20
   SH, dark and uniform ALP staining was observed, indicating strong ALP
   expression (Figure [99]4f). In the coculture within the scaffold, IWRC
   also possessed high ALP expression (Figure [100]S8, Supporting
   Information). At later stages of stem cell osteogenic differentiation,
   calcium ions precipitated as calcium salts to form “bone nodules
   (calcium nodules).” On day 14 of induction, Alizarin red staining was
   used to visualize the calcium nodules within the BMSCs. 30 and 20 SH
   presented deeper colors in larger areas than the other groups,
   suggesting a higher density of calcium nodules, as shown in the
   high‐magnification images (Figure [101]4f). Although the number of
   calcium nodules in 10 SH was lower than that in 20 and 30 SH, it was
   significantly higher than that in HAP. Furthermore, several
   osteogenesis‐related proteins were tested by immunofluorescence
   (Figure [102]4g). Runx2, which served as an essential transcription
   factor for stimulating osteogenesis, showed higher expression in 20 and
   30 SH. Stronger fluorescence intensities were also detected in 20 and
   30 SH for collagen1 (Col1), a marker in pre‐osteoblasts and mature
   osteoblasts during BMSC osteogenic differentiation. Osteocalcin (OCN)
   and osteobridging protein (OPN), produced at the end of osteoblast
   differentiation, were highly expressed in 20 and 30 SH. Because the
   cells were not in direct contact with the scaffolds during these
   experiments, the material composition emerged as a key factor in
   facilitating osteogenic differentiation.

2.5. Transcriptomic Analysis of the IWRC for Osteogenic Differentiation

   To further investigate the molecular mechanism of osteogenic
   differentiation, we performed a transcriptomic analysis of BMSCs
   cultured on HAP and 20 SH for 21 days in the absence of osteogenic
   growth factors. Differential screening was performed based on the
   expression of protein‐coding genes, and 2065 differential genes were
   detected, as shown in Figure [103] 5a. A total of 1373 genes were
   significantly downregulated and labeled in blue, and 692 genes were
   significantly upregulated and labeled in red. Genes without significant
   differences are labeled in grey. The top 20 differentially expressed
   genes (DEGs) that were upregulated or downregulated are shown in the
   clustered heatmap (Figure [104]5b). Compared with those in HAP, the
   expressions of ANGPTL7, KLF2, and LTBP1 in 20 SH were 21.7, 9.73, and
   3.69 folds higher, respectively. Gene Ontology (GO) enrichment analysis
   was performed on the DEGs in terms of molecular functions, biological
   processes, and cellular components (Figure [105]5c). Five genes related
   to bone mineralization, 39 related to cell adhesion, and 41 related to
   calcium ion binding were significantly upregulated. After the
   screening, genes related to calcium ion binding, such as FNN1, LTBP1,
   NPNT, and ENPP1, were significantly upregulated (Figure [106]5g); MGP,
   ENPP1, and OMD, which are related to bone mineralization, were also
   significantly upregulated (Figure [107]5h). These results were
   consistent with the transcriptome data measured by qPCR
   (Figure [108]5i,j). The enrichment analysis and chord diagram
   (Figure [109]5e) illustrated the top 10 categories with the smallest
   p‐values. The Kyoto Encyclopedia of Genes and Genomes (KEGG) database
   was used for pathway analysis of the DEGs (Figure [110]5d), and the
   enrichment score was represented on the horizontal axis. The results
   indicated that the PI3K‐ART, TGF‐beta, and Wnt signaling pathways were
   all effectively activated, with the TGF‐beta signaling pathway
   exhibiting the highest enrichment score, suggesting its predominant
   role in promoting osteogenic differentiation.

Figure 5.

   Figure 5
   [111]Open in a new tab

   Transcriptomic analysis for BMSCs osteogenic differentiation. (a)
   Volcano plot of DEGs, grey represents genes without significant
   differential expression, red represents significantly upregulated
   genes, and blue represents significantly downregulated genes. (b)
   Clustered heatmaps of DEGs showing the top 20 genes that are
   significantly upregulated and downregulated. (c) GO enrichment analysis
   results, filtering three categories of GO terms with corresponding
   PopHits ≥5. The top 10 terms were sorted in descending order based on
   the ‐log10p value. (e) KEGG enrichment analysis results. (d) Top 20
   KEGG pathway enrichment analysis (filtered for pathways with
   corresponding PopHits ≥5, sorted by ‐log10p value in descending order).
   (e) Top 10 GO enrichment chord diagram. (f) Top 10 KEGG enrichment
   chord diagram. (g) Expression of genes associated with calcium ion
   binding in transcriptome sequencing, consistent with the qPCR results
   (i) (n = 3). (h) Expression of genes related to bone mineralization in
   transcriptome sequencing, consistent with the qPCR results (j) (n = 3)
   (^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, and ^**** p < 0.0001, ns,
   not significant).

2.6. Ectopic Osteoinduction by IWRC

   Ectopic ossification is a critical criterion for evaluating the
   osteoinductive potential of materials. To assess the bone repair
   capabilities of the IWRC, we implanted four groups of scaffolds (30 SH,
   20 SH, 10 SH, and HAP) into the paravertebral muscles of beagles for 3
   months to observe the osteogenic phenomena. The relative position of
   the implant within the canine body is depicted in Figure [112] 6a.
   Three months post‐implantation, the materials were retrieved from the
   muscle tissue. The scaffolds had integrated tightly with the
   surrounding musculature, and tissues grew into the scaffold interiors
   without evidence of fibrous encapsulation (Figure [113]6b). Further
   histological analysis with Hematoxylin and Eosin (H&E) staining
   revealed that soft tissues proliferated throughout the porous
   structures of all four scaffolds, with dense, deep red bone tissue
   detected on the scaffold walls. In HAP, new‐bone tissues showed
   elongated shapes along the inner porous walls with a thickness of
   ≈20 µm (Figure [114]6d). In the whisker‐reinforced groups, new‐bone
   tissues grew into sheets with thicknesses exceeding 100 µm in certain
   areas. In particular, contiguous bone tissue was observed within 20 SH
   (Figure [115]6e–g). Semi‐quantitative analysis using the ImageJ
   software showed a significantly greater volume in the
   whisker‐reinforced group than in the HAP (Figure [116]6c). This
   demonstrates that the in situ whisker‐reinforced ceramics retain their
   osteoinductive properties, outperform HAP ceramics, and show potential
   for further applications. Tartrate‐resistant acid phosphatase (Trap)
   staining revealed that osteoclasts emerged within the four groups of
   materials, adhering closely to the surface of the materials. This
   indicated that the induced bone tissue had self‐remodeling functions
   (Figure [117]6h–k). Following a clear observation of osteoinduction, we
   employed polymerase chain reaction (PCR) techniques to analyze the
   expression levels of osteogenesis‐related genes in the four materials
   (Figure [118]6l–o). The Runx2 mRNA levels and OPN expression in 20 SH
   were significantly higher than those in the other three groups. The
   expression of COL‐1α was also the highest in 20 and 30 SH.
   Immunohistochemical staining also indicated that IWRC promoted the
   expression of genes related to osteogenesis and blood vessel
   generation, as shown in Figure [119]S10 (Supporting Information).

Figure 6.

   Figure 6
   [120]Open in a new tab

   Osteoinduction evaluation in vivo with a beagle dog subcutaneous
   implant. (a) Representative radiograph illustrating the bilateral
   placement of the scaffold in the beagle's back muscles. (b) Photograph
   of the muscular tissue containing scaffolds 3 months after
   implantation. The placement of scaffolds (black box) is visible. (c)
   Semi‐quantitative analysis of the new bone tissue area. The
   calculations were based on histological H&E staining. (d–g)
   Histological H&E staining images for the four materials. The far‐left
   column presents an overview of the stained tissue sections. Blue boxes
   indicate the regions that have been magnified. Scaffold material is
   denoted by (M), and newly formed bone tissue (NB) is indicated by
   yellow arrows. (H–K) Histological images of Trap staining for the four
   materials to identify osteoclasts within the new bone tissue. The
   presence of osteoclasts is marked by dark red staining (yellow arrows)
   on the surface of the material (M) (the overall image of Trap staining
   is shown in Figure [121]S11, Supporting Information). (i–o) Gene
   expression levels (Runx2, COL‐1α, BMP2, and OPN) of osteogenic‐relevant
   transcripts in the scaffolds 4 months after implantation (n = 3) (^* p
   < 0.05, ^** p < 0.01, ^*** p < 0.001, and ^**** p < 0.0001).

2.7. Proteomic Quantitative Analysis of Osteoinductivity in Muscle Tissue

   To further elucidate the mechanism underlying the enhanced
   osteoinductivity of IWRC, we conducted quantitative proteomic analysis
   on the subcutaneously implanted scaffolds after 1 month. At this stage,
   no significant bone tissue was observed within the material,
   facilitating the investigation of the mechanisms involved in
   material‐induced osteogenesis. Figure [122] 7a displays the
   differentially expressed proteins in the HAP and 20 SH scaffolds, with
   141 proteins significantly upregulated and 108 proteins significantly
   downregulated. Enrichment analysis of the differential proteins, using
   the GO database, revealed that the 20 SH exhibited significant
   upregulation in the expression of ossification‐related proteins, such
   as IGF2, AHSG, MGP, and SPP1. Moreover, the expression of proteins
   associated with calcium channel regulation was upregulated in 20 SH,
   with the expression levels of F2 and FFAR2 being 2.02‐fold and
   1.53‐fold higher than those in HAP. Figure [123]6c shows the combined
   data from the KEGG database, indicating that 20 SH significantly
   promoted the expression of osteoclast differentiation‐related proteins,
   including G8GUH4, ACP5, CSF1R, and LCP2. The differences in the
   expression of these proteins between the two groups are presented in
   the clustering heat map (Figure [124]7d). The upregulation of these
   proteins can promote the development of bone tissues. By selecting
   related species in the STRING database, we analyzed the interaction
   relationships of differentially expressed proteins and identified the
   top 25 proteins by connectivity. After recalculating their connectivity
   and constructing a network diagram (Figure [125]7e), it was evident
   that IWRC may promote osteoinductivity through the networked expression
   of osteogenesis‐related proteins, such as AHSG, ACP5, IGF2, SPP1, and
   MGP. Among these, calcium ion‐related proteins, such as F2 and FFAR2,
   synergistically regulate the expression of osteogenesis‐related
   proteins.

Figure 7.

   Figure 7
   [126]Open in a new tab

   Proteomic quantitative analysis of osteoinductivity in muscle tissue.
   (a) Volcano plot of differential proteins. The horizontal axis
   represents log[2] (fold change). Grey points represent proteins with a
   p‐value ≥ 0.05. (b)Top 30 GO enrichment analysis bar chart. The
   horizontal axis shows the ‐log10p values, while the vertical axis lists
   GO term names. (c) Top 20 KEGG enrichment analysis bubble chart. The
   horizontal axis represents the enrichment score, while the vertical
   axis shows the information of the top 20 pathways. (d) Cluster heatmap
   of osteogenesis‐related proteins, selected through GO enrichment and
   KEGG enrichment analyses. (e) Protein‐protein interaction network with
   high connectivity, where the proteins related to ossification are
   marked in red.

2.8. IWRC Efficacy in Supercritical Bone Defect Regeneration

   To assess the ability of IWRC to promote bone regeneration in
   preclinical supercritical bone defects, we established a 15 mm
   truncated bone defect model in rabbit femurs. In this study, the
   average diameter of the rabbit femur was ≈7 mm, and the defect created
   in the animal model was twice that size, exceeding the requirement of
   1.5 times the critical bone defect threshold. We implanted 10 SH, 20
   SH, and HAP scaffolds and observed the growth of new bone. The
   implantation process and the images obtained after restoration are
   shown in Figure [127]S13 (Supporting Information). Micro‐computed
   tomography (micro‐CT) was used to semi‐quantitatively observe the
   generation of new bone, including the regeneration of new osteal tissue
   within the implant and the integration of the material with the host
   bone. Figure [128] 8a presents the post‐repair 3D reconstructions,
   single‐layer radiographic images, and quarter‐sectional views to
   compare the overall morphologies of the scaffold and femur at 1‐ and
   3‐month post‐implantation. One month after implantation, a clear
   demarcation between the scaffold and host bone persisted, and a layer
   of osseous tissue (red) was formed on the internal surface of the
   scaffold. However, the porous structure of the scaffold was discernible
   with prominent voids. The proportion of new bone (BV/TV) and density of
   new bone (BMD‐BV) exhibited a trend of 20 SH > 10 SH > HAP, whereas the
   trabecular spacing (Tb.Sp) decreased (Figure [129]8c). Compared to the
   first month, the new bone tissue significantly increased at 3 months.
   It can be clearly observed from the micro‐CT images that, in the 10 and
   20 SH groups, the new bone tissue has merged into continuous structures
   and the scaffold has fused with both ends of the bone defect. In
   conjunction with the semi‐quantitative analysis results, it was found
   that the proportion of new bone (BV/TV) and new bone density (BMD‐BV)
   increased with an increase in trabecular thickness (Tb.Th) and a
   decrease in trabecular spacing (Tb.Sp) compared to the first month. The
   trends in the volume and density of new bone among the three groups at
   3 months were consistent with those observed in the first month, i.e.,
   20SH > 10SH > HAP.

Figure 8.

   Figure 8
   [130]Open in a new tab

   Radiological assessment of bone repair at weight‐bearing sites. (a)
   Computed tomography (CT) evaluation of bone tissue regeneration within
   different materials, including post‐repair 3D reconstructions,
   single‐layer scan images (M: scaffold material, NB: new bone tissue),
   and quarter‐sectional views of the new bone tissue (red represents bone
   tissue, and grey represents scaffold material). (b) X‐ray assessments
   of bone integration status for different scaffolds at 2 weeks and 3
   months post‐implantation (yellow outlines denote the experimental
   sites). (c) MicroCT–rendered parameters of ingrown bone at different
   time points (n = 3) (^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ns, not
   significant).

2.9. Clinical Assessment of Bone Healing via X‐ray Imaging

   X‐ray imaging is a crucial diagnostic tool for the clinical evaluation
   of bone healing at weight‐bearing sites. Therefore, we employed an
   X‐ray to observe the repair site at 2 weeks and 3 months
   post‐implantation. As illustrated in Figure [131]8b, immediately
   following implantation, the distinct porous structure of the scaffold
   was visible, with a clear boundary between the scaffold and the host
   bone. After 3 months, the boundaries began to blur, indicating the
   integration of the scaffold with the host tissue. Furthermore, the
   entire scaffold was gradually covered by new tissues, and the original
   porous structure was nearly imperceptible, suggesting the ingrowth of
   new bone within the material. Additionally, traces of fractures were
   observed in the HAP. During the experiment, we had six control subjects
   without implants. Four of these suffered fractures within 1‐month
   post‐surgery, and the remaining two fractured at ≈2 months post‐surgery
   with typical failure characteristics, as shown in the X‐ray images
   (Figure [132]8b). This further underscores the importance of material
   support in repairing segmental bone defects at weight‐bearing sites.
   Although ceramic materials cannot serve as primary load‐bearing
   structures, they facilitate the maintenance of overall stability.
   Particularly, during the repair process, the ingrowth of new tissue can
   further assist the scaffold in stabilizing the skeletal structure.

2.10. Morphology of New Bone Tissue

   Integration of the material and bone tissue was observed using optical
   microscopy. One month post‐implantation, the porous structure of the
   material was filled with newly formed tissue, and no fibrous
   encapsulation was detected (Figure [133] 9a). At this stage, the tissue
   was transparent with a low‐density, allowing clear visualization of the
   porous architecture of the material. A distinct demarcation between the
   left side of the scaffold and the host bone was observed. Three months
   after implantation, the material and bone tissue were tightly
   integrated, and the boundary between them was completely obscured. The
   color and morphology of the newly formed tissue within the scaffold
   closely resembled those of the host bone (Figure [134]9b). SEM and EDS
   were conducted to further investigate the interface between the
   material and the tissue. The new tissue contained not only carbon but
   also calcium and phosphorus. Combined with the histological sections,
   it can be preliminarily concluded that the newly formed tissue on the
   material surface was osseous (Figure [135]9f). As shown in
   Figure [136]9c–e, the new osteal tissue tightly enveloped the scaffold
   material, and the new tissue proliferated along the surface of the
   ceramic grains, substantiating the biocompatibility and bioactivity of
   the IWRCs.

Figure 9.

   Figure 9
   [137]Open in a new tab

   Morphology of new bone tissue. (a) Images of the scaffold and repair
   site after 1‐month post‐implantation, with the overall view on the left
   and a magnified view of the area framed in white on the right, where
   the light part represents the scaffold material. (b) Images of the
   scaffold and repair site after 3 months post‐implantation (HB: host
   bone, M: scaffold material, with the material‐tissue interface outlined
   by a yellow curvilinear frame). (c–e) Scanning electron microscopy
   (SEM) images of new bone tissue and the scaffold. Overall morphology is
   shown in (c), where new bone tissue and host bone are marked in red.
   The material‐tissue interface (d) and surface (e) were magnified for
   detailed observation. (f) Energy‐dispersive X‐ray spectroscopy (EDS)
   mapping of bone tissue adhered to the material surface, with color
   overlays representing Ca, C, P, O, S, and N. (g–i) Hematoxylin and
   Eosin (H&E) histological staining of 10 SH, 20 SH, and HAP after
   1‐month implantation. The left column shows the overall H&E staining of
   the scaffold‐repaired femoral shaft truncation sites, with selected
   areas 1 and 2 magnified to observe the morphology of the new bone.
   (j–l) H&E histological staining after 3 months implantation. (HB: host
   bone, M: scaffold material, NB: new bone tissue).

   One month after implantation, H&E staining was performed to evaluate
   the formation of new bone within the scaffolds. As shown in
   Figure [138]9g–i, evident new bone ingrowth with a depth of 3 mm was
   observed in HAP at the right end, where the lower right end of the
   scaffold and the bone tissue had fused. No new bone tissue was formed
   at the left end of the scaffold. A large amount of material debris was
   found next to the lower left cortical bone, suggesting that the
   fracture of the material was likely caused by the compression of the
   cortical bone. In contrast, new bone tissue was observed at the right
   end of 10 SH, achieving partial fusion with the host bone. In
   comparison, bone tissues grew from the outer edge of the scaffold
   toward the center in 20 SH, with a length of 6 mm. The newly formed
   bone tissue was dense and toward the host's cortical bone. At this
   time, the fusion with the host bone was distinct at the left end of the
   scaffold and noticeable at the right end. These results indicate that 1
   month after implantation, the defect began to heal gradually, and the
   scaffold had partially fused with the host bone.

   Three months after implantation, the growth of new bone in each group
   is shown in Figure [139]9j–l. Compared to 1 month, the new bone tissue
   in the scaffolds grew significantly, and the bone mass increased. In
   HAP, the new bone tissue was dense and concentrated at the left end of
   the scaffold, where it completely fused with the host cortical bone.
   However, no significant bone tissue was observed on the right side of
   the material, and the right end of the scaffold was not fused to the
   host. Additionally, on the lower‐right side of the scaffold, the
   cortical bone had a crushed layer with the scaffold's pore structure.
   Although the crushed material remained in close contact with the host
   cortical bone, no new bone ingrowth was observed. In 10 SH, new bone
   tissue grew significantly at both ends of the scaffold and spanned the
   entire scaffold, leaving only a 2–3 mm gap unhealed. Furthermore, the
   fusion of the scaffold with the host on the right side was more
   significant than that on the left side, where local fragmentation of
   the scaffold was observed without new bone tissue. In 20 SH, the
   scaffold structure was clear, and both ends were completely fused with
   the host bone. New bone tissue spanned almost the entire scaffold along
   the outer edge, and the direction of the tissue was close to that of
   the cortical bone of the host. The new bone tissue was dense and
   resembled the host's cortical bone, and a large number of Haversian
   canals were found within the new bone, indicating that the bone tissue
   matured.

   Furthermore, in the immunohistochemical staining, the
   osteogenic‐related genes (Runx2, COL‐1α, BMP2, and OPN) and the
   angiogenic gene CD31 in the IWRC group had strong positive expression
   (Figures [140]S14–S16, Supporting Information). In this study,
   experimental animals were intravenously injected with fluorescein and
   tetracycline hydrochloride 3 and 13 days before sample collection. The
   borders of the mineralized bone tissue were marked with green and
   orange fluorescence. A significant spacing between the two fluorescent
   lines indicated that the bone tissue growth was continuous (Figure
   [141]S17, Supporting Information). The overall repair effect followed
   the order of 20SH > 10SH > HAP.

2.11. Mechanical Testing after Repair

   FEA was performed to assess the feasibility of the scaffold. The
   scaffold design model is shown in Figure [142] 10a, and supercritical
   bone defect repair scaffolds were 3D printed using HAP, 10 SH, and 20
   SH according to the design. In the analysis, it was assumed that each
   rabbit weighed ≈3 kg, and when hopping, the legs bore a weight
   significantly heavier than their own. We set it to ten times the weight
   of the rabbit itself, that is, a total force of 300 N, with each side
   bearing 150 N. During the surgery, a steel plate was used for
   unilateral fixation (Figure [143]10a). A force of 150 N was applied to
   the top end of the rabbit femur to analyze the stress on the femur
   shaft, scaffold, and steel plate (Figure [144]10b). The results showed
   that the steel plate bore most of the force, with stress mainly
   distributed around the screw holes on the bone shaft, and the remaining
   pressure was borne by the scaffold, with the calculated compressive
   stress on it being 7.42 MPa. The stress on the scaffold was mainly
   distributed on the side opposite the steel plate, and the scaffold
   could fail from one side if further deformation was applied
   (Figure [145]10b). The compressive strengths of 10 SH and 20 SH
   scaffolds were 8.53 and 10.00 MPa, respectively, higher than the
   strength borne by the scaffold in the simulation experiment. Therefore,
   these materials can meet the usage requirements for large‐segment bone
   repair (Figure [146]10c–f). However, the compressive strength of pure
   HAP ceramics is only 1.5 MPa, which presents a significant risk of
   failure. This simulated result was in accordance with the experimental
   results.

Figure 10.

   Figure 10
   [147]Open in a new tab

   Biomechanical evaluation of bone repair at weight‐bearing sites. (a–c)
   Pre‐implantation finite element mechanical analysis of truncated bone
   animal models. A force of 150 N was applied to the femoral head for
   stress analysis of the repair system (a), with corresponding stress
   contour maps (b) and potential deformation patterns (c) calculated for
   the scaffold. (d–f) Compression tests of the personalized repair
   constructs for rabbit femur shafts, where the custom‐made scaffolds
   matched the external structure of the rabbit femur shafts (d),
   compression strength (e), and elastic modulus (f) were calculated,
   integrating the results with finite element simulations to predict the
   failure risk of the three scaffold groups. (h–l) Mechanical tests 3
   months post‐implantation of 10 SH, 20 SH, and HAP, involving
   compression testing of femur shaft repair site sections containing the
   scaffolds (h), corresponding stress‐strain curves (i), and calculation
   of compressive strength (j) and elastic modulus (k), recording the
   scaffold failure process during compression (l) (0% indicates no
   deformation; 100% indicates complete failure; t denotes the time
   required for the sample to completely fail). (m) X‐ray images of
   natural bone tissue and the three scaffolds after failure (red frames
   mark the locations of bone fractures).

   The recovery status of bone mechanics can provide an intuitive
   evaluation of the bone repair process. Three months post‐implantation,
   we removed the samples and disassembled the internal fixation devices,
   such as the steel plate. Considering that the femur of the lower limb
   is the primary load‐bearing bone, we then cut the femoral shaft
   containing the material along the two screw holes in the middle and
   conducted compression tests. The bones implanted with steel plates were
   used as positive controls. The compression test layout and the
   morphology after failure are shown in Figure [148]10h. The compressed
   stress‐strain curve is shown in Figure [149]10i. For HAP, the scaffold
   material was compressed and deformed during the initial test stages.
   When the material was flattened, the bones at both ends came into
   contact, splitting the bone. It was evident from the photograph that
   the HAP scaffold was squeezed into a lump and protruded outside the
   bone. X‐ray images revealed that the HAP scaffold was completely
   compressed with a significant crack in the bone and a compressive
   strength of 10.27 MPa. In contrast, the bone with 20 SH scaffold first
   split at the bottom during compression and then expanded until complete
   failure. The X‐ray image indicated that after failure, the main body of
   the 20 SH scaffold remained intact, and its compressive strength was
   21.12 MPa (twice that of the HAP ceramics after repair and close to
   71.35% of the control bone). The HAP and 10 SH scaffolds, after 3
   months of repair, could recover 34.7% and 68.11% of the mechanical
   support of the original bone, respectively. The compression processes
   of the control and 10 SH are shown in Figure [150]S18 (Supporting
   Information).

3. Discussion

3.1. Mechanical Reinforcement of 3D‐Printed Porous Calcium Phosphate Ceramics

   A porous structure is an indispensable factor for bioactive bone repair
   materials, which can provide suitable growth space for the tissue and
   ensure substance exchange.^[ [151]^32 , [152]^33 ^] Although porous CaP
   has an outstanding bioactivity, its poor mechanical properties
   significantly limit its clinical applications. Studies show that the
   optimum porosity to promote new bone growth is ≈60%–80%, and the ideal
   pore diameter is 200–800 µm.^[ [153]^34 ^] The sintering temperature of
   porous CaP ceramics is generally lower than 1200 °C to ensure a large
   number of micropores in the matrix.^[ [154]^35 , [155]^36 ^] These
   micropores, as stress concentration points, first crack under stress,
   causing the failure of the entire scaffold. Although the compressive
   strength of the dense HAP sintered at 1300 °C can reach 500 MPa, which
   is much higher than the cortical bone of the human body, such ceramics
   completely lose their bioactivity and have no application value in bone
   regeneration.^[ [156]^37 ^] In this study, a whisker‐reinforced
   structure was formed during the sintering process by regulating the in
   situ oriented growth of HAP grains. The simulation results show that
   the in situ grown whiskers can effectively slow down the stress
   concentration at the interface of the microporous material, which
   effectively alleviates the negative impact of the microporous structure
   on the mechanical properties. Moreover, CaSO[4] promotes the fusion
   between the HAP grains, and the strength at grain boundaries is
   significantly enhanced. Therefore, the fracture modes of the IWRC
   include trans‐granular and intergranular fractures, which can consume
   more fracture energy.

   Compared with the reinforcement by adding external whiskers, the
   whisker structure formed by in situ oriented growth is more closely
   combined with the matrix with lower internal stress. The existing 3D
   printing technology, especially SLA/DLP, is not suitable for HAP
   whiskers because the addition of pre‐prepared whiskers in the bio‐ink
   severely reduces fluidity and printing accuracy.^[ [157]^38 , [158]^39
   , [159]^40 , [160]^41 ^] In our study, CaSO[4] was added to the slurry
   as an inducer to promote whisker growth, which did not change the
   fluidity and ensured the normal progress of printing. Subsequently, a
   rod‐like whisker structure was formed during the sintering process
   through the oriented growth of HAP grains. The growth of whiskers
   depended on the content of CaSO[4]. When the content was less than 5%,
   whisker growth was not evident. In this study, although the regulatory
   mechanisms of in situ preferential grain growth and their effects on
   the mechanical enhancement of hydroxyapatite ceramics were explored,
   the investigation into the mechanisms underlying grain orientation
   growth was lacking. In future research, we aim to explore the kinetic
   factors influencing the oriented growth of hydroxyapatite ceramics
   regulated by calcium sulfate. Molecular dynamics simulations will be
   used to examine the interactions between different crystal facets of
   CaSO[4] and HAP, thereby studying the effects of CaSO[4] on the growth
   characteristics of HAP crystals on various planes.

3.2. IWRC Enhance Osteogenesis

   Porous structure and chemical composition are two fundamental factors
   that affect bioactivity.^[ [161]^42 ^] Porous scaffolds, which serve as
   bridges for tissue growth, should possess multilevel pore structures to
   provide suitable habitats for key factors involved in tissue growth,
   such as proteins, cells, and blood vessels.^[ [162]^43 ^] We used 3D
   printing technology to prepare scaffolds with customized macroporous
   structures. On this basis, by growing whiskers at a lower temperature
   (900 °C) and maintaining the sintering temperature at 1150 °C, we
   retained the capillary micropores on the scaffold surface.

   CaP is advantageous chemically because of its ability to release active
   ions.^[ [163]^44 ^] HAP, as the thermodynamically stable phase of CaP,
   has a low hydrolysis rate (≈6.62 × 10^−126) and degrades very slowly in
   the body. BCP ceramics, owing to the inclusion of fast‐degrading TCP,
   degrade slightly faster than HAP and release more calcium and phosphate
   ions, exhibiting higher biological activity than HAP.^[ [164]^45 ,
   [165]^46 ^] Ca^2+ is not only an important component of the inorganic
   part of the skeleton but also a signaling factor in living organisms
   that can regulate the transcription of genes. Ca^2+ plays an important
   role in cell proliferation and differentiation. Typically, the
   concentration of Ca^2+ in cells is ≈100 nm, which increases to
   500–1000 nm upon activation. An increase in Ca^2+ concentration in the
   extracellular environment can increase the intracellular Ca^2+
   concentration through L‐type and non‐L‐type calcium channels and Ca^2+
   receptors, thereby activating osteogenesis‐related signaling
   pathways.^[ [166]^47 ^] In this study, owing to the introduction of
   CaSO[4], the release rate of Ca^2+ was significantly increased. In an
   in vitro experiment in the absence of osteogenic inducers, IWRC
   upregulated the expression of genes related to calcium ion binding and
   promoted osteogenic differentiation. In addition, within the back
   muscles of Beagle dogs, IWRC significantly promoted the expression of
   calcium ion‐regulated proteins (F2 and FFAR2) and ossification‐related
   proteins (AHSG, ACP5, IGF2, SPP1, and MGP), demonstrating improved
   osteoinductivity and bioactivity. Consequently, in the repair of
   supercritical bone defects, the regenerative ability of IWRC was
   significantly enhanced, and new bone tissue penetrated the entire
   scaffold in just 3 months to repair the defect.

4. Conclusion

   In this study, high‐precision porous ceramic scaffolds were printed
   using a CaSO[4]/HAP ceramic slurry, followed by controlling grain
   orientation growth during subsequent sintering to achieve a
   mechanically reinforced grain arrangement structure. The study
   investigated the mechanisms by which calcium sulfate regulates the in
   situ oriented growth of HAP grains and their associated mechanical
   enhancement effects. It was found that CaSO[4] initially promoted grain
   merging within the ceramic, strengthening the grain boundary binding
   strength. Moreover, the in situ whiskers effectively reduced stress
   concentration within the porous material, resulting in a 5–10 times
   increase in the compressive strength of the porous CaP ceramics.
   Additionally, CaSO[4] was observed to enhance the release of active
   Ca^2+ in IWRC ceramics, significantly improving the osteoinductivity of
   HAP. Transcriptomic and proteomic results indicated that IWRC could
   activate the calcium ion pathway and the expression of
   osteogenesis‐related proteins. In preclinical animal experiments, IWRC
   achieved the repair of supercritical bone defects within just 3 months,
   with mechanical recovery exceeding 70% of autologous bone. This study
   reconciled the contradiction between the bioactivity and mechanical
   strength of porous ceramics, successfully achieving the dual
   optimization of the bioactivity and mechanical strength of CaP
   ceramics.

5. Experimental Section

Scaffold Preparation

   First, 50 g of polyurethane acrylate, 15 g of polyethylene glycol
   diacrylate, and 1 g of BYK‐2155 were thoroughly mixed. Subsequently,
   60% HAp was added to the aforementioned mixture in a planetary ball
   mill, and the mixture was milled at a speed of 240 rpm for 15 h. Next,
   10, 20, and 30% of calcium sulfate dihydrate (CaSO[4]·2H[2]O) based on
   the HAP mass was added, and milling was continued for 2 h. Two grams of
   TPO was then added, and milling was continued for an additional 2 h.
   The slurry and grinding balls were separated, and the debubbling
   process was carried out using a vacuum‐defoaming device to obtain the
   printing slurry.

   After printing, the samples were subjected to ultrasonic cleaning with
   anhydrous ethanol to remove uncured slurry, with precautions taken to
   avoid contact with water during the cleaning process. The samples were
   then placed in a muffle furnace for degreasing and sintering, following
   a temperature profile: ramping at 5 °C min^−1 to 300 °C and holding for
   6 h, further ramping at 5 °C min^−1 to 600 °C and holding for 2 h,
   subsequent ramping at 5 °C min^−1 to 900 °C and holding for 10 h, and
   finally ramping at 5 °C min^−1 to 1150 °C and holding for 2 h, followed
   by natural cooling to room temperature.

Material Microstructure

   The microstructure of the samples was observed and photographed via
   scanning electron microscopy (SEM; JSE‐5900LV, Japan). All ceramic
   samples were sputter‐coated with gold for 70–140 s before the analysis.
   The composition of the samples was analyzed using the EDS extension
   function of the scanning electron microscope.

Ion Release

   Porous materials of the same macro size were placed in a 0.1 m Tris‐HCl
   buffer solution to create a system that was 30 times their own weight.
   This solution was then oscillated at 37 °C for 1, 2, 4, 8, and 16 days.
   Next, the supernatant was extracted and analyzed via inductively
   coupled plasma–optical emission spectrometry (ICP‐OES; 5100 SVDV,
   Agilent, USA).

In Vitro Degradation

   The in vitro degradation rate of the scaffold was tested by using a
   0.1 m Tris‐HCl buffer as the medium. The tested sample was a porous
   circular sheet with a diameter of 2 mm and a diameter of 8 mm, and the
   material porosity was 68%. The porous sample was placed in a 50 mL
   centrifuge tube, Tris‐HCl buffer was added at a ratio of 1 g/30 mL, and
   the tube was placed in a 37 °C shaker for 1, 4, 7, and 14 days. During
   this test, the buffer solution was replaced once every 2 days. At each
   test point, the sample was removed, washed with water three times,
   placed in an oven until a constant weight was achieved, and the
   material mass difference was calculated.

Mechanical Simulation

   The mechanical simulation was carried out using the mechanical
   simulation section of COMSOL Multiphysics (version 5.6). Because the
   investigated material was novel and does not have standard material
   parameters, the Young's modulus of the material was uniformly set to
   281 GPa and the Poisson's ratio was set to 0.25 in the simulation
   experiment. A 2D single‐layer simplified model was used for convenience
   of analysis. According to the SEM results, the inside of the model
   mainly consisted of pores, equiaxed particles, and columnar whiskers.
   The raw HA particles were simplified as spherical and the columnar
   whiskers were simplified as cylindrical systems. Considering that the
   model did not undergo plastic deformation, it was a linear elastic
   model.

Nanomechanics

   The surface nanomechanics of the materials were measured using atomic
   force microscopy (AFM; MFP‐3D‐BIO, Asylum Research, USA). The
   force‐displacement curves in the grains and at the grain boundaries of
   the ceramics were measured to calculate the Young's modulus.

XRD

   The phase structures of the raw materials and ceramic scaffolds were
   analyzed using X‐ray diffraction (XRD; Philips X'Pert 1 X‐ray
   diffractometer, Netherlands) after grinding the ceramic scaffolds into
   powder. Using a Cu target with a voltage of 30 kV and current of 20 mA,
   the samples were scanned at a rate of 0.05 °/s in the range of 10°–70°.

Compression Test

   The compressive strength of the ceramic samples was analyzed using a
   universal testing machine according to the standard ISO13314:2011 (E).
   The test samples were cylinders with dimensions of Φ6.5 × 13 mm that
   were compressed at a rate of 1 mm min^−1 along the axis of the cylinder
   until failure. Each group consisted of at least three parallel samples.
   Stress–strain curves were calculated based on the load and displacement
   during the compression process.

In Vitro Proliferation and Differentiation

   All cell extraction, media change, passaging, and planting operations
   in the experiment were performed on a sterile workbench. All cells were
   cultured in an incubator at 37 °C with 5% CO[2], and the media was
   changed or passaged every other day. A density of 5 × 10^4 BMSCs cells
   was seeded on the surface of the material. After incubation for 15 min,
   1 mL of α‐MEM complete medium per well was used for routine culturing.
   At 1, 3, and 7 days, the CCK‐8 reagent kit was used to detect the
   cytotoxicity of the material and determine the proliferation, and each
   group of materials had three parallel samples. Fluorescein diacetate
   (FDA)/propidium iodide (PI) was used to fluorescently stain the cells,
   and the morphology of the live/dead cells was observed by confocal
   laser scanning microscopy. Similarly, a density of 5 × 10^4 BMSCs cells
   was seeded on the surface of the material for the osteogenic
   differentiation test. When the cells at the bottom of the well plate
   reached 80% confluence, the medium was replaced with an osteogenic
   induction medium, and the culturing was continued for 10 days to
   measure the alkaline phosphatase (ALP) activity using an ALP activity
   kit (Beyotime Co., China). On the 14th day, the calcification nodules
   of the cells were observed using alizarin red S (ARS) staining.
   Further, immunofluorescence staining was employed to label the
   expressions of Runx2, Collagen1 (Col1), Osteocalcin (OCN), and
   Osteobridging protein (OPN).

Transcriptomic Analysis

   The co‐culturing of BMSCs and the scaffold was carried out using a
   six‐hole Transwell culture plate, in which the 20 SH and HAP porous
   scaffolds (n = 3) were placed in the Transwell chamber, and 5 × 10^4
   BMSCs were seeded at the bottom of the well plate. The ordinary culture
   medium α‐MEM without any induction components was used for the routine
   cell culture. After 21 days, the cells were lysed using trizol and RNA
   was extracted for transcriptomic analysis.

Proteomics Analysis

   20 SH and HAP porous ceramics were implanted into the back muscle of
   beagle dogs, and the samples were removed after 1 month for 4D‐DIA
   proteomic analysis, with three samples in each group. The extra tissue
   outside the scaffold was removed, the frozen samples were added to
   liquid nitrogen, fully ground, and the protein was extracted. After
   protease digestion, liquid chromatography‐mass spectrometry was used to
   perform mass spectrometry analysis on the peptide segments. The
   proteins corresponding to the peptide segments were quantified in
   relation to one another by comparing the signal intensity of the
   corresponding peptide segments in different samples.

Ectopic Osteoinduction

   The back muscle model for beagle dogs was used to verify the
   osteoinductive properties of the investigated material. This study was
   approved by the Animal Ethics Committee of Sichuan University (Ethics
   registration number: 20230310045). The beagle dog (male, 8 kg) was
   purchased from Chengdu Dasuo Experimental Animal Co., Ltd., and raised
   in the Experimental Animal Center of the West China Science and
   Technology Park of Sichuan University. Under fasting conditions, the
   animal was fully anesthetized using an intraperitoneal injection of a
   3% pentobarbital sodium solution. After disinfection and sterilization,
   the back skin and muscle were exposed, a small hole ≈2 cm was cut, and
   the four sets of sterilized brackets were placed inside. The sample
   diameter was 6 mm, the height was 10 mm, and the porosity was 68%. Each
   group had at least three samples. After the operation, the experimental
   dogs were raised until 3 months later, and then euthanized. The
   implanted samples were removed for photography, omics staining, and the
   detection of osteogenic gene expression.

Establishment of a Large Bone Defect Animal Model

   This experiment was conducted under the approval and guidance of the
   Animal Care and Use Committee of Sichuan University (Ethics
   registration number: 20230310045). Forty‐eight rabbits were randomly
   divided into four groups: a blank control group and three material
   groups (HAP, 10 SH, and 20 SH), with six rabbits in each group. After
   successful anesthesia, the surgical area was prepared by shaving the
   skin 3 cm below the knee joint and 4 cm below the greater trochanter of
   the femur. A curved incision ≈4–5 cm in length was made along the
   lateral side of the knee joint, taking care to protect the lateral
   collateral ligament of the knee joint and the joint capsule. A cutting
   point 15 mm from the middle of the bone was marked, and the bone was
   cut using a wire saw at the marked position. The material was placed
   and tied behind the steel plate using surgical silk, and the bone and
   steel plate were then reset. Screws with dimensions of Φ 2.4 mm (L14),
   Φ 2.4 mm (L16), and Φ 2.4 mm (L18) were sequentially used to fix the
   bone and the steel plate from the middle to the sides of the steel
   plate. The surgical incision was sutured layer‐by‐layer after rinsing
   with a 0.9% saline and iodine tincture, and the skin was disinfected.
   After 1 and 3 months of implantation, the experimental animals were
   euthanized, and the samples were removed for photography, histological
   staining, and biomechanical analysis.

Compression Testing of the Femur

   After 3 months of implantation, specimens from the healthy femur, HAP,
   10 SH, and 20 SH groups of rabbits were obtained. The proliferative
   tissue around the steel plate was removed, and the screws and internal
   fixation steel plate were removed. The experimental sample was
   extracted along the positions of the two central screws, and the
   remaining bone shaft was cut to the same length as the experimental
   samples and reserved as a control sample. Another segment of the
   healthy femur from the opposite side of the experimental rabbits was
   taken as a positive control group, cut to the same length as the
   experimental samples. Compression testing was conducted using a
   universal mechanical testing machine with a loading speed of
   1 mm min^−1. The test was terminated when the stress value sharply
   decreased or when the sample exhibited obvious fracture.

Statistical Analysis

   All data in this section were analyzed using GraphPad Version 8.0
   (GraphPad Prism). Two‐way analysis of variance was employed to assess
   the statistical significance of differences among the groups, with ^* p
   < 0.05, ^** p < 0.01, ^*** p < 0.001, and ^**** p < 0.0001 indicating
   statistically significant differences.

Conflict of Interest

   The authors declare no conflict of interest.

Ethics Approval Statement

   This animal experiment was conducted under the approval and guidance of
   the Animal Care and Use Committee of Sichuan University (Ethics
   registration number: 20230310045).

Supporting information

   Supporting Information
   [167]ADVS-12-2408459-s002.docx^ (22.9MB, docx)

   Supplemental Movie 1
   [168]Download video file^ (14.5MB, MP4)

Acknowledgements