Tolerant and Rapid Endochondral Bone Regeneration Using Framework-Enhanced 3D Biomineralized Matrix Hydrogels

Abstract

Tissue-engineered bone has emerged as a promising alternative for bone defect repair due to the advantages of regenerative bone healing and physiological functional reconstruction. However, there is very limited breakthrough in achieving favorable bone regeneration due to the harsh osteogenic microenvironment after bone injury, especially the avascular and hypoxic conditions. Inspired by the bone developmental mode of endochondral ossification, a novel strategy is proposed for tolerant and rapid endochondral bone regeneration using framework-enhanced 3D biomineralized matrix hydrogels. First, it is meticulously designed 3D biomimetic hydrogels with both hypoxic and osteoinductive microenvironment, and then integrated 3D-printed polycaprolactone framework to improve their mechanical strength and structural fidelity. The inherent hypoxic 3D matrix microenvironment effectively activates bone marrow mesenchymal stem cells self-regulation for early-stage chondrogenesis via TGFβ/Smad signaling pathway due to the obstacle of aerobic respiration. Meanwhile, the strong biomineralized microenvironment, created by a hybrid formulation of native-constitute osteogenic inorganic salts, can synergistically regulate both bone mineralization and osteoclastic differentiation, and thus accelerate the late-stage bone maturation. Furthermore, both in vivo ectopic osteogenesis and in situ skull defect repair successfully verified the high efficiency and mechanical maintenance of endochondral bone regeneration mode, which offers a promising treatment for craniofacial bone defect repair.

Keywords: biomineralization; bone regeneration; endochondral ossification; hydrogels; hypoxic microenvironment.

Figure 1

Schematic illustration of a tolerant and rapid endochondral bone regeneration strategy using framework‐enhanced 3D biomineralized matrix hydrogels. BMSCs, bone marrow stem cells; NOIS, native‐constituent osteogenic inorganic salt; GH, GelMA/HAMA; sGH, NOIS@GelMA/HAMA; PCL, polycaprolactone; sPCL, NOIS@PCL.

Figure 2

Characterization of framework‐enhanced biomineralized matrix hydrogels. A,B) Photographs (A) and rheological data (B) of GH and sGH hydrogels before and after 365 nm light irradiation. Data are presented as mean ± standard deviations (SD), n = 3, ns: not statistically significant. C–E) The swelling ratio (C) and degradation rate of sGH hydrogels (D) and the sPCL framework (E) with different NOIS contents. Data are presented as mean ± SD, n = 3, ^*: p < 0.05. F) The preparation process of 3D‐printed sPCL‐enhanced sGH hydrogel scaffolds. G) A comparison of sGH and sGH/PCL scaffolds before and after pressing. H) The compressive modulus of GH, sGH, and sGH/PCL scaffolds. Data are presented as mean ± SD, n = 3, ^*: p < 0.05. I,J) Elemental analyses (I) and photographs (J) of the sGH hydrogels and sPCL framework. K) Scanning electron microscopy (SEM) images of the sGH hydrogels and sPCL framework. Red arrows indicate the existence of inorganic salts. NOIS, native‐constituent osteogenic inorganic salt; GH, 5% (w/v) GelMA/1% (w/v) HAMA; sGH, 5% (w/v) NOIS@ 5% (w/v) GelMA/1% (w/v) HAMA; sPCL, 20% (w/v) NOIS@PCL (PCL: NOIS = 4:1); light, 365 nm LED, 20 mW cm⁻².

Figure 3

In vitro cytocompatibility and biological function evaluation. A) Cytotoxicity analyses of GH and sGH hydrogels at 4 h, as well as 1, 4, and 7 days. B,C) DNA content and Live/dead staining of BMSCs in the sGH hydrogels after 1‐ and 7‐days culture in vitro. Data are presented as mean ± SD, n = 3, ns: not statistically significant. D) Immunofluorescence staining of COL2 and OCN in 2D‐CP, 3D‐GH, and 3D‐sGH groups at 14 days. E,F) Quantitative analyses of COL2 (E) and OCN (F) fluorescence intensity. Data are presented as mean ± SD, n = 3, ^*: p < 0.05. G–K) Expression level of cartilage‐related genes (COL2, SOX9), hypoxia‐inducible gene (HIF‐1α), and bone‐related genes (OCN, RUNX2) in 2D‐CP, 3D‐GH, and 3D‐sGH groups after 7‐ and 14‐days culture in vitro. Data are presented as mean ± SD, n = 3, ^*: p < 0.05. The scaffold compositions are the same as shown in Figure 2.

Figure 4

Molecular mechanisms underlying spontaneous chondrogenesis in 3D matrix hydrogels. A) Schematic illustration of stem cell fate in 2D and 3D culture platforms. B,C) Heatmap and volcano plot of differentially expressed genes (DEGs) between the 2D‐CP and 3D‐GH groups. D,E) Gene ontology (GO) enrichment analyses of upregulated (D) and downregulated (E) DEGs for total biological processes. F,G) Heatmaps of biological processes involving “regulation of cartilage development” (F) and “aerobic respiration” (G) (marked in red) in the corresponding GO enrichment. H,I) Gene expression correlation analyses of biological processes involving “regulation of cartilage development” (H) and “aerobic respiration” (I). J,K) Real‐time polymerase chain reaction (RT‐PCR) results of the expression levels of cartilage‐related genes TGF‐β, SMAD3, and SOX5, as well as aerobic respiration‐related genes COX1, SIRT3, MDH, and SDHB. Data are presented as mean ± SD, n = 3, ^*: p < 0.05. L) Weston blot analysis of the expression levels of cartilage‐related proteins HIF‐1α, SOX9, COL2, TGF‐β, and SMAD3. M) Schematic illustration of spontaneous chondrogenesis via the transforming growth factor beta (TGF‐β)/Smad signaling pathway.

Figure 5

Molecular mechanisms underlying efficient osteogenesis in 3D biomineralized matrix hydrogels. A) Schematic illustration of stem cell fate with and without the addition of biomineralized inorganic salts in the 3D culture platform. B,C) Heatmap and volcano plot of DEGs between the GH and Sgh groups. D) GO enrichment analyses of upregulated DEGs for total biological processes. E,F) Heatmaps of biological processes involving “regulation of bone mineralization” and “osteoclast differentiation” (marked in red) in the corresponding GO enrichment. G,H) Gene expression correlation analyses of biological processes involving “regulation of bone mineralization“ (G) and ”osteoclast differentiation" (H). I,J) RT‐PCR results of the expression levels of osteogenic genes BMP2/6, OMD, and BGLAP (I), as well as osteoclastic genes CALCR, OSTM1, OSCAR, and OCSTAMP (J). Data are presented as mean ± SD, n = 3, ^*: p < 0.05. K) Weston blot analysis of the expression levels of bone‐related proteins RNX2, OCN, and BMP2/6. L) Schematic illustration of efficient osteogenesis achieved via striking a balance between bone formation and remodeling processes.

Figure 6

Evaluation of regenerated bone at four weeks post‐implantation in nude mice. A) Schematic illustration of the 3D‐printed framework‐enhanced biomineralized matrix hydrogels. B) The gross views and micro‐CT images of the GH, sGH, and sGH/PCL groups. C–F) Quantitative analyses of bone volume (BV) (C), bone volume fraction (BV/TV) (D), bone mineral density (BMD) (E), and the number of bone trabeculae (Tb.N) (F). Data are presented as mean ± SD, n = 3, ^*: p < 0.05. G) Histological examinations of hematoxylin and eosin (H&E), safranin‐O/fast green (SO/FG), Masson, Type II collagen (COL2), osteocalcin (OCN), CD31, and tartrate‐resistant acid phophatase (TRAP) staining in the GH, sGH, and sGH/PCL groups. The red arrows represent new vessels. The scaffold compositions are the same as those shown in Figure 2.

Figure 7

Shape maintenance and quantitative analyses of regenerated bone in nude mice. A,B) Round and pentacle shape maintenance of the sGH and sGH/PCL groups four weeks after implantation in nude mice. C–F) Expression levels of cartilage‐related genes COL2 and SOX9, as well as bone‐related genes OCN and RUNX2, in the GH, sGH, and sGH/PCL groups four weeks after implantation. Data are presented as mean ± SD, n = 3, ^*: p < 0.05. G–J) Quantitative analyses of the DNA content (G), GAG content (H), total collagen (I), and Yang's modulus (J) in the GH, sGH, and sGH/PCL groups four weeks after implantation. Data are presented as mean ± SD, n = 3, ^*: p < 0.05. The scaffold compositions are the same as those shown in Figure 2. ACR, area change rate.

Figure 8

Evaluation of rabbit skull defect repair at 12 weeks post‐surgery. A,B) Gross view (A) and 3D reconstruction microcomputed tomography scan images (B) of the bone marrow mesenchymal stem cells (BMSCs)‐loaded sGH/PCL, BMSCs‐free sGH/PCL, and blank groups. C–G) Quantitative analyses of bone volume (BV) (C), bone volume fraction (BV/TV) (D), bone mineral density (BMD) (E), number of bone trabeculae (Tb.N) (F), and the Yang's modulus (G) in the BMSCs‐loaded sGH/PCL, BMSCs‐free sGH/PCL, and blank groups. Data are presented as mean ± SD, n = 3, ^*: p < 0.05. H) Elemental analyses of regenerated skull defects in the BMSCs‐loaded sGH/PCL groups. I) Histological examinations of H&E, SO/FG, Masson, OCN, CD31, and TRAP staining in the BMSCs‐loaded sGH/PCL, BMSCs‐free sGH/PCL, and blank groups. The scaffold compositions are the same as those shown in Figure 2.