A Novel Deer Antler-Inspired Bone Graft Triggers Rapid Bone Regeneration

Abstract

Adult mammals are unable to regenerate bulky bone tissues, making large bone defects clinically challenging. Deer antler represents an exception to this rule, exhibiting the fastest bony growth in mammals, offering a unique opportunity to explore novel strategies for rapid bone regeneration. Here, a bone graft exploiting the biochemical, biophysical, and structural characteristics of antlers is constructed. It is decellularized antler cancellous bone (antler-DCB) to obtain a bone scaffold. Then, an antler-based bone graft is constructed by integrating antler-DCB with antler-derived biological signals, delivered by extracellular vesicles (EVs) from antler blastema progenitor cells (ABPCs), a novel stem cells responsible for antlerogenesis is discovered. The antler-based bone graft transformed bone marrow stromal cells into cells with an ABPC-like phenotype and transcriptomic signature. In vivo, the antler-based graft triggered rapid bone formation in a rat model, with doubled volume of newly formed bones than commercial DCBs. In addition, the antler-based graft orchestrated a coordinated process of vascularization, neurogenesis, and immunomodulation during osteogenesis, partially imitating early antlerogenesis. These findings provide practical insights to develop a therapeutic intervention for treating severe bone defects.

Keywords: angiogenesis; antler‐based bone graft; bioinspiration; bone regeneration; neurogenesis.

Scheme 1

Overview of antler‐DCB@EVs^ABPCs triggers rapid bone regeneration. A) Antler‐DCB@EVs^ABPCs mimics the biological characteristics of antlers, consisted of EVs^ABPCs, HAMA, and antler‐DCB. B) Antler‐DCB@EVs^ABPCs triggers a rapid bone growth process resembling antlerogenesis, couples with osteogenesis with angiogenesis, neurogenesis, and immunomodulation.

Figure 1

Fabrication and characterization of DCB from multiple species sources. A) Schematic depiction of the fabrication process and characterization techniques for decellularized marrow bone. B) Naturally shed Sika deer antlers. C) DAPI staining of the antler before and after decellularization. D) Collagen I and GAGs measurement in the antler before and after decellularization (n = 4). E) Scanning electron microscope (SEM) images of DCB surface. F) Specific surface area (SSA) measurement in DCB (n = 4). G) Measurement of the roughness (Ra) of the DCBs using 3D atomic force microscope (AFM). H) Water contact angle (WCA) measurement in DCBs (n = 4). I) Compressive stress−strain curve of DCBs. J,K) Quantitative analysis of collagen I and GAGs content in DCB using assay kits. L) X‐Ray Diffraction (XRD) patterns of the DCBs. M) Fourier‐transform infrared (FT‐IR) spectra of the DCBs. N–P) Quantitative analysis of the Mg (N), S (O), and Si (P) content in each group (n = 4). Q) Hematoxylin–eosin (H&E) and Masson staining images demonstrating the formation of new bone at 12 weeks post‐surgery (NB, newly formed bone tissue). R) 3D reconstruction and 2D micro‐CT images showing regenerated bone around defects. S–U) Quantitative analysis of microstructural parameters of regenerated bone tissues, including BMD (S), BV/TV (T), and Tb.Th (U) (n = 5). All statistical data are represented as mean ± SD. Statistical analyses were performed using one‐way ANOVA with Bonferroni's post‐hoc test. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 2

Physical and biological properties and functions of EVs. A) Schematic diagram of characterization techniques for EVs. B–D), Characterization of EVs by (B) transmission electron microscopy (TEM), (C) nanoparticle tracking analysis (NTA), and (D) Western blotting. E) Heatmap showing the distribution of different expressed genes (DEGs) between EVs^ABPCs and EVs^BMSCs, and enriched pathways of these DEGs (right). F) Volcano map of DEGs in EVs^ABPCs versus EVs^BMSCs. Red dots indicate significantly upregulated in EVs^ABPCs; the blue dots indicate significantly upregulated EVs^BMSCs; and the gray dots indicate no significance. G) Volcano map of different expressed proteins (DEPs) of EVs^ABPCs versus EVs^BMSCs. H,I) Bar plot showing enriched Gene Ontology (GO) terms for upregulated genes (H), and downregulated genes (I) in EVs^ABPCs versus EVs^BMSCs. J) Workflow for evaluating the effects of EVs on BMSCs. K,M) Wound healing rate of BMSCs with each treatment over 24 h (n = 4). L,N) Analysis of BMSC migration behavior using transwell assay (n = 4). O) Multiple volcanic maps showing DEGs of BMSCs treated with PBS, EVs^ABPCs, and EVs^BMSCs. P) Principal component analysis illustrating the variance in BMSCs treated with PBS, EVs^BMSCs, and EVs^ABPCs. Q) Network diagram showing enriched GO terms and pathways for upregulated genes in EVs^ABPCs group versus the EVs^BMSCs group. All statistical data are represented as mean ± SD. Statistical analyses were performed using one‐way ANOVA with Bonferroni's post‐hoc test. ***p < 0.001.

Figure 3

In vitro osteogenic capacity of EVs^ABPCs. A) Workflow for evaluating the effects of EVs on BMSCs and ABPCs. B–D) The proliferation of BMSCs and ABPCs in different groups was analyzed using 5‐ethynyl‐2′deoxyuridine (EdU) B,C) and CCK‐8 assays D) (n = 4). E) Microscopic images of ALP and alizarin red S (ARS) staining of BMSCs and ABPCs in different groups after osteo‐induction. F,G) Quantitative analyses of all positively stained areas in ALP F) and ARS G) staining. H) Relative expression of ABPC‐related genes in BMSCs, validated by RT‐qPCR (n = 4). I) Relative expression of BMSC‐related genes in BMSCs, validated by RT‐qPCR (n = 4). J) Schematic illustration showing the RNA‐seq of BMSCs and ABPCs in different groups. K) Multiple volcanic maps showing DEGs between BMSCs treated with EVs^ABPCs and EVs^BMSCs and ABPCs treated with PBS. L) Correlation scatter plot comparing the global gene expression between EVs^ABPCs‐treated BMSCs and ABPCs. M) Correlation scatter plot comparing the global gene expression between EVs^BMSCs‐treated BMSCs and ABPCs. N) Heatmaps showing the expression of stemness and proliferation genes in BMSCs treated with EVs^BMSCs and EVs^ABPCs and in ABPCs treated with PBS. O) Workflow showing the protein sequence of EVs^ABPCs and EVs^BMSCs. P) Heatmap showing the DEGs in EVs^ABPCs and EVs^BMSCs overlapped with the epigenetic regulation‐associated genes from the GO database. Q) Network plot analysis showing DEPs in EVs^ABPCs versus EVs^BMSCs related to osteogenesis pathway. Node colors from purple to blue indicate log₂(fold change) from low to high. R) KMT2A levels in BMSCs, as detected by Western blot (n = 3). S) KMT2A levels in BMSCs with EVs^ABPCs and EVs^{ABPCs/KMT2A‐} treatment (n = 3). T) Relative expression of ABPC‐related genes in BMSCs, validated RT‐qPCR (n = 4). U) Relative expression of BMSC‐related genes was validated in BMSCs by RT‐qPCR (n = 4). All statistical data are represented as mean ± SD. Statistical analyses were performed using one‐way ANOVA with Bonferroni's post‐hoc test. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 4

Construction of an antler‐inspired bone graft, antler‐DCB@EVs^ABPCs. A) Schematic diagram of the interaction between HAMA hydrogel and EVs. B) EVs labeled with PKH26 (red) in HAMA hydrogel examined by z‐stack scanning using confocal laser scanning microscopy. Fluorescent intensity results (left) are shown as 3D views of surface remodeling presented as angled (middle) and front views (right). C–F) Root mean square deviation (RMSD) (C), root mean square fluctuation (RMSF) (D), radius of gyration (Rg) (E), and solvent accessible surface area (SASA) (F) of VCAN and VCAN/HAMA. G) Free energy landscape of VCAN and VCAN/HAMA hydrogels based on principal coordinates analysis. H) VCAN and HAMA hydrogel compound docking simulation. I) Schematic illustration of the fabrication and characterization of the antler‐DCB@EVs. J) Turface potential of the antler‐DCB and HAMA hydrogel was quantitatively analyzed by kelvin probe force microscopy. K) Pore size distribution of antler‐DCB@EVs^ABPCs. L) Characterization of antler‐DCB@EVs^ABPCs by TEM. M,N) Degradation rate of antler‐DCB@EVs^ABPCs and HAMA hydrogel (n = 3). O) EV accumulative release using a bicinchoninic acid assay at days 1, 3, 6, 9, 12, 15, and 18 (n = 3). P) WCA measurement of antler‐DCB@EVs^ABPCs. Q) CCK‐8 analysis of BMSCs cultured on various scaffolds for 1, 3, and 7 d (n = 3). All statistical data are represented as mean ± SD. Statistical analyses were performed using one‐way ANOVA with Bonferroni's post‐hoc test. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 5

Antler‐DCB@EVs^ABPCs triggers rapid bone formation in vivo. A) Schematic diagram for exploring the osteogenic capacity of antler‐DCB@EVs^ABPCs. B) H&E and Masson staining images demonstrating the formation of new bone at 8 weeks post‐surgery (M, material; NB, newly‐formed bone tissue). C) 2D micro‐CT (up) and 3D reconstruction (down) images showing regenerated bone around defects. D) Quantitative analysis of microstructural parameters of regenerated bone tissues, including BV/TV, BMD, Tb.Th., and Tb.N. (n = 5). E,F) Representative images (E) and summarized quantitative data (F) showing the newly formed trabecular bone using calcein AM (green fluorescence) and alizarin red (red fluorescence) labeling at 7 and 8 weeks after scaffolds insertion (n = 5). G) Representative Western blot images showing ALP, RUNX2, and OCN levels in newly formed bone tissue with various scaffolds. H,I) The recruitment of various scaffolds on BMSCs in vivo was detected using flow cytometry (n = 5). J) Clustering analysis of DEGs in different group and GO analysis results identifying key distinct features and biological significance. K) Schematic diagram of the shared mechanism of deer antler growth and antler‐DCB@EVs^ABPCs induced bone growth. All statistical data are represented as mean ± SD. Statistical analyses were performed using one‐way ANOVA with Bonferroni's post‐hoc test. **p < 0.01 and ***p < 0.001.

Figure 6

Antler‐DCB@ EVs^ABPCs stimulates angiogenesis. A–C) Protein sequencing analyses of DEPs related to the regulation of angiogenesis in EVs^ABPCs versus EVs^BMSCs. D,E) Cellular uptake assays showing PKH26‐labeled EVs internalized and distributed in HUVECs. HUVECs were stained with phalloidin (green) and EVs with PKH26 (red). F,H) Analysis of HUVECs migration behavior via transwell assay (n = 3). G,I) Tube formation ability of HUVECs with each treatment (n = 3). J) Schematic diagram of exploring the angiogenic capacity of antler‐DCB@EVs^ABPCs. K,L) 3D reconstruction images and summarized quantitative data showing newly formed blood vessels at 4 and 8 weeks after implantation in different groups (n = 5). M) Representative immunofluorescence images of CD31 (green), EMCN (green), and cell nucleus (blue) staining at eight weeks. N,O) Quantitative analysis of all positively stained areas. All statistical data are represented as mean ± SD. Statistical analyses were performed using one‐way ANOVA with Bonferroni's post‐hoc test. **p < 0.01, ***p < 0.001.

Figure 7

Antler‐DCB@EVs^ABPCs facilitates neurogenesis. A–C) Protein sequencing analyses of DEPs related to neurogenesis regulation in EVs^ABPCs versus EVs^BMSCs. D,E) Cellular uptake assays showing PKH26‐labeled EVs internalized and distributed in neurons. Neurons were stained with Tuj1 (green), the nucleus with DAPI (blue), and EVs with PKH26 (red). F–H) DRG neurons and explants co‐cultured with different groups of EVs for 72 h. I) Axonal lengths were quantified for DRGs (n = 4). J) Ratio of the total area of neurites to the total area of explant body (n = 4). K) Schematic diagram showing Antler‐DCB@EVs^ABPCs facilitates neurogenesis and the transition of neuron fibers from sympathetic to cholinergic. L,M) Immunofluorescence staining showing the distribution of CGRP⁺ nerves in the bone regeneration region after different treatments (n = 5). N) Representative immunofluorescence images of cholinergic marker vesicular acetylcholine transporter (VAChT) (green), sympathetic marker tyrosine hydroxylase (TH) (green), and cell nucleus (blue) stained at the site of bone formation in rats. O,P) Quantitative analysis of all positively stained areas of cholinergic and sympathetic nerves (n = 5). Q) Acetylcholine content in different groups (n = 5). All statistical data are represented as mean ± SD. Statistical analyses were performed using one‐way ANOVA with Bonferroni's post‐hoc test. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 8

Antler‐DCB@EVs^ABPCs modulates inflammatory response during osteogenesis. A–C) Protein sequencing analyses of DEPs related to the regulation of inflammation in EVs^ABPCs versus EVs^BMSCs. D) Quantification of the rate of iNOS⁺F4/80⁺ cells. E) Quantification of the rate of Arg1⁺F4/80⁺ cells. F) Assessment of proinflammatory TNF‐𝛼, PGE2, IL‐10 and IL‐1β in macrophages by enzyme‐linked immunosorbent assay (ELISA). G) Schematic diagram of the evaluation of immunomodulation by antler‐DCB@EVs^ABPCs in vivo over 7 d after implantation. H) Representative immunofluorescence images of iNOS (red), CD206 (green), and cell nucleus (blue) staining. I) Quantitative analysis of all positively stained areas (n = 4). J,K) Western blot analysis and summarized quantitative data for TNF‐𝛼, COX2, and Arg1. All statistical data are represented as mean ± SD. Statistical analyses were performed using one‐way ANOVA with Bonferroni's post‐hoc test. *p < 0.05, **p < 0.01, and ***p < 0.001.