Alveolar ridge preservation with lyophilized microspheres of human dental pulp stem cells following 3D dynamic osteogenic induction
Yangyang Li, Jethro Zih-Shuo Wang, Zihan Yang, Fawen Wang, Haoyu Li, Chi Yang, Wenwen Yu, Zhiyuan Zhang
TL;DR
This study explores using human dental pulp stem cells in a new way to improve alveolar ridge preservation after tooth extraction.
Contribution
A novel bioactive material using lyophilized hDPSC microspheres preconditioned with 3D dynamic osteogenic induction is introduced.
Findings
Lyophilized hDPSC microspheres showed enhanced alveolar bone preservation in preclinical models.
The microspheres promoted superior new bone formation and improved bone microarchitecture.
Results outperformed traditional artificial bone powder and control groups.
Abstract
The clinical use of bone graft materials for alveolar ridge preservation following tooth extraction has become a standard procedure to facilitate subsequent implant restoration and prosthetic rehabilitation. However, the therapeutic efficacy of these materials is substantially limited by their bio-inertness, lack of cellular activity, and unpredictable resorption rates. The development of bioactive osteogenic materials capable of host integration and active promotion of bone regeneration would represent a significant advancement over current clinical protocols. To address this challenge, our research has focused on developing bioactive biomaterials using human dental pulp stem cells (hDPSCs). This study proposes a novel strategy for alveolar ridge repair utilizing lyophilized hDPSC microspheres preconditioned through a three-dimensional (3D) dynamic osteogenic induction system. We…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6| Gene Name | Forward (5‘-3’) | Reverse(5‘-3’) |
|---|---|---|
|
| GGAAGCTTGTCATCAATGGAAATC | TGATGACCCTTTTGGCTCCC |
|
| CTCCTCGGAAGACACTCTGACC | CTGCGCCTGGTAGTTGTTGTG |
|
| CCCCTGGAAAGAATGGAGATG | AGCTGTTCCGGGCAATCCT |
|
| CTACTATGGCACTTCGTCAGGAT | ATCAGCGTCAACACCATCATT |
|
| CCAGGCAACACTCCTACTCCA | GCCTTGGGTTTATAGACATCTTGG |
|
| GACCTGCCAGCAACCGAAGT | GGTACTGGATGTCAGGTCTGCG |
|
| AGCCACCGAGACACCATGAG | GCTCCCAGCCATTGATACAGG |
|
| TATCGCAGGCACTCAGGTCAG | GGGTTGTTTTCCCACTCGTTTC |
| Primary antibodies | Host | Dilution | Supplier | Country |
|---|---|---|---|---|
|
| Mouse | 1:200 | BD | America |
|
| Mouse | 1:100 | BD | America |
|
| Mouse | 1:50 | BioLegend | America |
|
| Mouse | 1:40 | BioLegend | America |
|
| Mouse | 1:40 | BioLegend | America |
|
| Mouse | 1:60 | BD | America |
|
| Rabbit | 1:100 | CST | America |
|
| Rabbit | 1:100/1000 | CST | America |
|
| Rabbit | 1:100 | CST | America |
|
| Rabbit | 1:100 | CST | America |
|
| Rabbit | 1:100/1000 | CST | America |
|
| Rabbit | 1:1000 | CST | America |
- —Shanghai’s Top Priority Research Center
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsBone Tissue Engineering Materials · Mesenchymal stem cell research · Tissue Engineering and Regenerative Medicine
Introduction
Insufficient alveolar ridge bone volume represents a critical factor limiting the success of dental implant placement and other prosthetic rehabilitations.1 The application of commercial artificial bone powder to reconstruct alveolar ridge defects post–extraction has emerged as a conventional approach to maintain jaw morphology for subsequent restoration.2 Allogeneic bone, xenogeneic bone, and various synthetic biomaterials serve primarily as physical barriers and space maintainers, with their regenerative capacity largely dependent on osteoconductive properties.3 However, these materials are fundamentally constrained by inherent bio-inertness and unpredictable degradation kinetics,4 while the absence of bioactive factors frequently leads to compromised vascularization and inadequate osseointegration.5 Clinical evidence indicates that contemporary alveolar ridge preservation techniques centered on commercial bone substitutes carry considerable risks of incomplete bone repair and contour collapse.6
Tissue engineering technologies have opened new avenues for bone regeneration. Successful implementation relies on the synergistic integration of 3 fundamental components: functional seed cells, appropriate biological scaffolds, and biomimetic induction microenvironments.7 Human dental pulp stem cells (hDPSCs), characterized by their abundant availability (from discarded wisdom teeth and orthodontic extractions), robust proliferative capacity, and well-defined multi–lineage differentiation potential, have recently emerged as an ideal cell source for bone tissue engineering applications.8 Substantial evidence demonstrates that hDPSCs undergoing osteogenic induction exhibit enhanced alkaline phosphatase activity and increased calcium nodule formation.9
Nevertheless, conventional two-dimensional (2D) culture systems present limitations including rapid functional deterioration and loss of morphological polarity, thereby restricting the development of dental pulp stem cells as functional seed cells.10 Research indicates that three-dimensional (3D) culture platforms can significantly augment stem cell paracrine activity and viability through restoration of cell-cell and cell-matrix interactions.11 Specifically, microsphere structures fabricated via hanging drop or dynamic culture systems effectively mimic pre-osteoblast condensation processes, promoting upregulation of osteogenesis-related gene expression.12
Beyond 3D architectural support, biomechanical stimulation plays a pivotal role in directing stem cell fate. Dynamic culture environments not only optimize nutrient-waste exchange but also activate mechanosensitive ion channels and Rho/ROCK signaling pathways through fluid shear stress.13 Studies confirm that cyclic mechanical stimulation induces cytoskeletal reorganization and enhanced integrin clustering, subsequently promoting osteogenic differentiation through Hippo-YAP pathway activation.14
Notably, cell-based therapeutic products face critical clinical translation challenges including limited shelf stability, manufacturing scalability issues, and stringent transportation requirements.15 Lyophilization technology addresses these limitations through sublimation-based dehydration at low temperatures, maximizing preservation of biomolecular activity while enabling ambient-temperature storage of tissue-engineered constructs.16 Recent advances demonstrate that optimized lyophilization protocols maintain over 85% viability in stem cell microspheres post–rehydration without compromising osteogenic differentiation capacity.17
Building upon this scientific foundation, we have innovatively developed lyophilized hDPSC microspheres bioengineered through an advanced 3D dynamic osteogenic induction system. Through comprehensive evaluation of their osteogenic efficacy in extraction socket repair, we aim to establish a novel alveolar ridge restoration strategy integrating biological functionality with clinical practicality, thereby providing both theoretical rationale and practical solutions to overcome limitations of conventional bone graft materials.
Materials and methods
Isolation, culture and characterization of hDPSCs
Cell isolation and primary culture
Human dental pulp tissues were obtained from clinically extracted healthy impacted third molars and premolars extracted for orthodontic reasons (with patient informed consent and ethics committee approval, the approval number of the Ethics Committee of Shanghai Ninth People’s Hospital is SH9H-2019-T167-6). Cells were isolated using an explant-enzymatic digestion method: pulp tissues were minced into approximately 1 mm³ fragments and digested with 3 mg/mL collagenase XI and 4 mg/mL dispase at 37 °C for 30 minutes. After digestion termination, tissue fragments were collected by centrifugation at 1000 rpm for 5 minutes resuspended in α-MEM complete medium supplemented with 20% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 μg/mL streptomycin, and cultured in dishes at 37 °C with 5% CO_2_. Medium was changed every 3 days, and cells were passaged at 1:2 ratio upon reaching 80%-90% confluence. Cells from passages 3-5 were used for all experiments.
Cytoskeleton staining
Cells grown on coverslips were rinsed twice with pre-warmed PBS and fixed with 4% paraformaldehyde for 15 minutes at room temperature. Subsequent experiments were conducted according to the kit’s instructions. Imaging was performed using a high-resolution fluorescence microscopy.
Immunophenotypic characterization
hDPSCs surface markers were identified by flow cytometry. Cells (1 × 10^6^) were incubated with anti–human CD73-FITC, CD90-PE, CD105-APC, CD34-FITC, and CD45-PE antibodies at 4 °C for 30 minutes protected from light. After treatment with permeabilization buffer containing 0.1% saponin, cells were analyzed using BD FACS Canto II flow cytometer, with data processed by FlowJo software. All antibody information is shown in Table 2.
Multi–lineage differentiation potential assessment
Osteogenic differentiation: At 60%-70% confluence, medium was replaced with osteogenic induction medium containing 10 mM β-glycerophosphate, 50 μg/mL ascorbic acid, and 100 nM dexamethasone, with medium changes every 3 days. After 21 days of induction, cells were fixed with 4% paraformaldehyde and stained with 2% Alizarin Red S (pH 4.2) to visualize mineralized nodule formation.
Adipogenic differentiation: After reaching full confluence, cells were induced with adipogenic medium containing 1 μM dexamethasone, 0.5 mM IBMX, 10 μg/mL insulin, and 200 μM indomethacin for 21 days. Following 4% paraformaldehyde fixation, lipid droplets were stained with 0.5% Oil Red O isopropanol solution.
Chondrogenic differentiation: Cell pellets (2.5 × 10^5^ cells) were formed by centrifugation and induced with chondrogenic medium containing 10 ng/mL TGF-β3, 100 nM dexamethasone, 50 μg/mL ascorbic acid, and 1% ITS for 21 days. Sulfated glycosaminoglycans in paraffin sections were detected by Alcian Blue staining.
3D microsphere construction and osteogenic induction
P3 hDPSCs were prepared as single-cell suspensions in osteogenic induction medium at a density of 1 × 10^5^ cells/mL. Cell suspensions were inoculated as 20 μL droplets (approximately 2000 cells/droplet) onto ultra-low attachment dishes and cultured for 2 days, with additional medium supplied to each droplet to prevent drying. The initially formed microspheres were transferred to 15 mL centrifuge tubes and subjected to centrifugal enhancement treatment (100 rpm, 3 minutes each, 3 times daily) on Days 1, 3, and 5. Subsequently, microspheres were transferred to a 3D rotating cell culture system and cultured for a total of 30 days, with induction medium refreshed every 2 days.
In vitro osteogenic efficacy evaluation
Alkaline phosphatase staining
Microspheres cultured for 30 days were fixed with 4% paraformaldehyde and incubated with BCIP/NBT chromogenic kit for 30 minutes protected from light to detect ALP activity.
Immunofluorescence staining
Paraffin sections of microspheres underwent antigen retrieval with sodium citrate buffer and were incubated overnight at 4 °C with primary antibodies against OCN (1:200), VEGF (1:100), Runx2 (1:150), COL I (1:300), and OSX (1:250). Alexa Fluor 594-conjugated secondary antibodies (1:500) were applied and incubated at room temperature for 1 hour protected from light, followed by DAPI counterstaining and fluorescence microscopy observation. All antibody information is shown in Table 2.
RT-qPCR analysis
Total RNA was extracted from hDPSCs, 5-day 3D dynamic-induced microspheres, and 30-day 3D dynamic-induced microspheres using TRIzol method. Subsequent experiments were completed following the manufacturer’s protocol. Gene and primer information are shown in Table 1.
Western blot analysis
Total protein was extracted with RIPA lysis buffer and quantified by BCA method. Subsequent experiments were completed following the manufacturer’s protocol.
Lyophilization processing and characterization
Mature microspheres were washed with PBS, pre-frozen at −80 °C for 4 hours, and transferred to a freeze-dryer for primary drying (−40 °C, 24 hours) and secondary drying (25 °C, 12 hours). The surface ultrastructure of lyophilized microspheres was examined using SU8010 field emission scanning electron microscope.
Animal experiments
Animal model establishment
All animal experiments were approved by the Animal Care & Welfare Committee, the approval number of the Shanghai Jiagan Biotechnology Co., Ltd. Ethics Committee is JGLL-20250819-01. Five-week-old male SD rats and 3-month-old male New Zealand white rabbits were randomly divided into 5 groups (n = 3): (1) Blank control group; (2) non–induced hDPSCs group; (3) 2D osteogenic-induced group; (4) artificial bone powder group; (5) lyophilized microsphere group. Under general anesthesia, mandibular first molars were minimally invasively extracted, and corresponding materials were implanted according to grouping followed by suturing.
Micro-CT analysis
Animals were sacrificed 4 weeks post–operation, and mandibular specimens were collected. Scanning was performed using Skyscan 1276 system.
Histological analysis
Specimens were decalcified with 10% EDTA for 4 weeks, embedded in paraffin, and sectioned at 5 μm thickness. Sections were stained with H&E, Masson’s trichrome, and COL I immunohistochemistry, then observed and imaged under optical microscope.
Transcriptome sequencing and analysis
Total RNA was extracted from hDPSCs, 5-day-induced microspheres, and 30-day-induced microspheres. Subsequent experiments were completed following the manufacturer’s protocol.
Statistical analysis
Statistical analysis was performed using GraphPad Prism 9.0 software. All data are presented as mean ± standard deviation. Multiple group comparisons were conducted using one-way ANOVA, with pairwise comparisons using LSD-t test. P <0.05 was considered statistically significant.
Results
Isolation of hDPSCs, establishment of 3D dynamic osteogenic induction system, and formation of hDPSC microspheres
Cells used in this study were derived from clinically extracted wisdom teeth and orthodontically extracted teeth. As described in the method, the tissue was then cultured in α-MEM medium containing 20% FBS (Figure 1A). After approximately 1 week, spindle-shaped cells were observed migrating from the tissue explants. Following several passages, uniformly dense populations of morphologically consistent spindle-shaped cells were observed under microscopy (Figure 1B). Phalloidin staining of P3 cells revealed clearly organized stress fibers (red) traversing the cytoplasm, with these fiber bundles aligned parallel to the long axis of the cells and exhibiting uniform fluorescence intensity (Figure 1C). Flow cytometric analysis demonstrated negative expression of hematopoietic stem/progenitor cell markers CD34 and CD45, while showing strong positivity for mesenchymal stem cell markers CD73, CD90, and CD105, with weak CD146 expression (Figure 1D). Trilineage differentiation assays revealed scattered lipid droplets with Oil Red O staining (Figure 1E1); extensive deep red staining areas indicating calcium deposition with Alizarin Red staining (Figure 1E2); and oval cells with light red nuclei and blue extracellular matrix with Alcian Blue staining and Nuclear Fast Red counterstaining (Figure 1E3).
Isolation, culture and characterization of human dental pulp stem cells (hDPSCs). (A) Schematic diagram of hDPSC isolation procedure. Freshly extracted intact third molars and orthodontically extracted teeth were collected. Dental pulp tissues were aseptically harvested, minced, enzymatically digested, and cultured in α-MEM medium containing 20% fetal bovine serum (FBS). (B) Culture and subculturing of hDPSCs. (B1) Explant culture of dental pulp tissue fragments at Day 7. (B2) Cells after 3 passages. (C) Fluorescence phalloidin staining of hDPSCs. (D) Immunophenotypic analysis. hDPSCs were characterized using mesenchymal stem cell markers (CD73, CD90, CD105, and CD146) and hematopoietic stem cell markers (CD34 and CD45). Histograms show overlay of isotype control and specific antibody staining. (E) Trilineage differentiation potential of hDPSCs. (E1) Adipogenic differentiation: Neutral triglycerides and lipids stained with Oil Red O. (E2) Osteogenic differentiation: Calcium deposits visualized by Alizarin Red S staining. (E3) Chondrogenic differentiation: Sections stained with Nuclear Fast Red and Alcian Blue (pH 2.5). Representative images from 3 different donors are shown (n = 3). Scale bar = 100 μm.
Three-dimensional dynamic osteogenic induction and HE staining of human dental pulp stem cells (hDPSCs)-microspheres. (A) Schematic of hDPSC osteogenic induction culture system. (B) Formation process of hDPSC microspheres. (B1) Day 1 of osteogenic differentiation. (B2) Day 3. (B3) Day 10. (B4) Day 30. (C) Macroscopic view of hDPSC microspheres placed on blue surgical drape, with microspheres circled in black. (D) HE staining: (D1) Day 5 of induction. (D2) Day 30 of induction. Scale bar = 200 μm.
Bone development-related gene and matrix protein expression in human dental pulp stem cell (hDPSC) microspheres and post–lyophilization status. (A, A1) COL I immunohistochemical staining at Day 7 of osteogenic differentiation. (A1') Higher magnification of boxed area in A1. (A2) COL I immunohistochemical staining at Day 30. (A3) Alizarin Red staining at Day 30. (A4) TRAP staining at Day 30. (A5) ALP staining at Day 30. (B) Tissue immunofluorescence staining: (B1) OCN, (B2) VEGF, (B3) COL I, (B4) OSX, (B5) Runx2. Bottom row shows higher magnification of boxed areas. (C) Western blot analysis of osteogenic-related proteins (ALP, Runx2, OCN) in hDPSCs, Day 10, and Day 30 osteogenically induced samples. (D) PCR analysis of osteogenic-related genes (ALP, Runx2, OCN, OPN, OSX, COL1A1, and BMP-2) in hDPSCs, Day 5, and Day 30 osteogenically induced samples. (E) Macroscopic view of lyophilized hDPSC microspheres. (F) (F1-F2) Scanning electron micrographs: left panel shows surface microstructure of lyophilized hDPSC microspheres, right panel shows artificial bone powder surface microstructure. (F3) Schematic diagram of collection and encapsulation of lyophilized hDPSC microspheres. Scale bar = 50 μm. Significant differences at P <.05 (), P <.01 (), and P <.001 (*, ***).
Based on previous literature and preliminary experiments, we established a 3D dynamic osteogenic differentiation system. Through suspension and separation culture, initial microsphere aggregation was achieved, followed by transfer to 15 mL centrifuge tubes for further maturation. Centrifugation facilitated microsphere compaction, and final dynamic osteogenic induction was performed using a 3D rotating cell culture system (Figure 2A). Microscopic documentation of hDPSC microspheres at Days 1 (Figure 2B1), 3 (Figure 2B2), 10 (Figure 2B3), and 30 (Figure 2B4) of 3D dynamic osteogenic induction revealed progressively darker pigmentation and stabilized morphology. Macroscopically, hDPSC microspheres appeared gray-white, semi-transparent, with glossy surfaces and moderate elasticity (Figure 2C). H&E staining of microspheres at Day 7 (Figure 2D1) and Day 30 (Figure 2D2) of osteogenic induction demonstrated looser cellular organization at Day 7 compared to the compact spherical structure observed at Day 30.
RNA sequencing analysis of human dental pulp stem cell (hDPSC) microspheres after sequential induction. (A) Schematic diagram of sequential 3D dynamic osteogenic induction of hDPSCs. (B) Differentially expressed transcripts in hDPSCs at Day 5 and Day 30 of 3D dynamic osteogenic induction compared to non–induced hDPSCs. (C) Venn diagram showing common differentially expressed genes among non–induced hDPSCs, Day 5, and Day 30 induced hDPSCs. (D) KEGG analysis of Day 5 induced versus non–induced hDPSCs. (E) GO analysis of Day 5 induced versus non–induced hDPSCs. (F) KEGG analysis of Day 30 induced versus non–induced hDPSCs. (G) GO analysis of Day 30 induced versus non–induced hDPSCs. (H) Heatmap of representative upregulated genes related to bone development.
In vitro validation of bone Development-Related gene and protein expression during hDPSC microsphere induction and transcriptomic analysis
Following successful establishment of the 3D dynamic osteogenic induction system and stable acquisition of dental pulp stem cell microspheres, we analyzed bone development-related gene and protein expression along with transcriptomic profiling. COL I immunohistochemical staining of Day-7 microspheres revealed numerous cell nuclei with minimal matrix deposition (Figure 3A1), while OCN immunohistochemical staining at day 30 showed prominent matrix protein staining separated by unstained areas, resembling multiple “ossification centers” (Figure 3A2). Alizarin Red staining (Figure 3A3), osteoclast staining (Figure 3A4), and ALP staining (Figure 3A5) of Day-30 microspheres demonstrated intense calcium deposition, particularly in central regions, while TRAP and ALP showed minimal staining.
Evaluation of lyophilized human dental pulp stem cell (hDPSC) microspheres on SD rat extraction socket healing. (A) Extraction and transplantation procedure of rat mandibular first molars. (A1) Implantation of centrifuged cell pellets into socket. (A2) Implantation of artificial bone powder. (A3) Implantation of lyophilized hDPSC microspheres. (A4) Post–suturing. (B) Bone recovery in extraction sockets at 4 weeks post–operation. (B1) Blank control. (B2) Non–induced hDPSCs. (B3) 2D osteogenically induced hDPSCs. (B4) Artificial bone powder. (B5) Lyophilized 3D dynamic osteogenically induced hDPSC microspheres. (C) 3D CT images of rat mandibles at 4 weeks. Groups correspond to B1-B5. (D-F) Micro-CT images of extraction sockets in sagittal (D), coronal (E), and transverse (F) planes at 4 weeks post–operation. (D1, E1, F1) Pre-extraction images. (D2, E2, F2) Blank control. (D3, E3, F3) non–induced hDPSCs. (D4, E4, F4) 2D osteogenically induced hDPSCs. (D5, E5, F5) Artificial bone powder. (D6, E6, F6) Lyophilized 3D dynamic osteogenically induced hDPSC microspheres. (G) Quantitative micro-CT analysis of bone regeneration (BV/TV, BMD, Tb.Th) in rat extraction sockets. Significant differences at P <.05 (), P <.01 (), and P <.001 (,**). (H, I) Coronal sections of extraction sockets stained with H&E (H) and Masson’s trichrome (I). Groups H1-I1 to H5-I5 correspond to B1-B5. (J) COL I immunohistochemical staining of artificial bone powder group (J1) and hDPSC microsphere group (J3). J2 and J4 are magnified images corresponding to the dashed areas in J1 and J3, respectively. The specific scale is shown in the figure.
Tissue immunofluorescence staining of Day-30 microspheres for OCN (Figure 3B1), VEGF (Figure 3B2), COL I (Figure 3B3), OSX (Figure 3B4), and Runx2 (Figure 3B5) revealed distinct spatial distribution patterns. OCN and VEGF demonstrated uniform distribution throughout microspheres, while COL I staining, though generally uniform, showed separation by strongly stained regions resembling “ossification centers,” with these boundary areas reminiscent of type I collagen-rich “periosteum.” Both OSX and Runx2 exhibited peripheral-to-central staining gradients, with OSX demonstrating gradual fading from edges toward the center, while Runx2 was predominantly localized to microsphere peripheries with minimal central expression.
RT-qPCR analysis of osteogenesis-related genes (COL-1, RUNX2, OSX, OPN, OCN, BMP-2, and ALP) during induction revealed dynamic expression patterns (Figure 3D). ALP expression initially increased then decreased; RUNX2 showed early upregulation followed by downregulation; OSX and BMP-2 demonstrated gradual upregulation; late osteogenic matrix proteins COL-1, OCN, and OPN exhibited significant upregulation, reaching levels 15-20 times higher than controls. Western blot analysis (Figure 3C) confirmed these trends, showing initial increase followed by disappearance of ALP protein, while Runx2 demonstrated progressive increase, and OCN expression gradually elevated.
To enhance clinical applicability, 3D dynamic osteogenically induced hDPSC microspheres underwent lyophilization (Figure 3E). Scanning electron microscopy revealed homogeneous porous microstructure in artificial bone powder (Figure 3F2), while lyophilized microspheres exhibited stereoscopic irregular porous structures (Figure 3F1). Figure 3F3 schematically illustrates the encapsulation process of lyophilized microspheres.
Our results demonstrate that 3D dynamic osteogenic induction mimicking in vivo bone development crucially influences hDPSC microspheres, with particularly pronounced effects at Day 30. To investigate dynamic changes during induction, we performed transcriptomic analysis of non–induced hDPSCs, Day 5, and Day 30 induced samples. The cells were adherent dental pulp stem cells on Day 0, underwent osteogenic induction to form spheroids in a 15 mL centrifuge tube by Day 5, and were in a state of dynamic osteogenesis by Day 30 (Figure 4A). Induction at day 30 resulted in significant alterations in a substantial proportion of transcripts, with 22% upregulation and 17.1% downregulation of total transcripts, compared to only 7.4% upregulation and 9.7% downregulation at Day 5 (Figure 4B). Venn diagram analysis identified 710 genes consistently changing throughout induction (Figure 4C).
GO enrichment analysis of upregulated genes at Day 5 revealed significant enrichment in biological processes including mesenchymal stem cell osteogenic commitment and osteogenic differentiation initiation; cellular components such as collagen matrix and mechanical anchoring; and molecular functions including osteogenic differentiation signaling pathway activation and intracellular structure remodeling (Figure 4D). Day 30 upregulated genes showed enrichment in MAPK pathway, collagen matrix support, and bone mineralization biological processes; extracellular collagen matrix and mechano-chemical signal transduction cellular components; and bone matrix mineralization and osteogenic differentiation signaling regulation molecular functions (Figure 4E). KEGG analysis further demonstrated that Day 5 microspheres were enriched in osteogenic initiation pathways, including PI3K-AKT signaling (providing the fundamental osteogenic environment) and mTOR signaling (serving as the osteogenic differentiation engine) (Figure 4F). Day 30 microspheres showed enrichment in pathways regulating stem cell pluripotency, terminal osteogenic differentiation initiation, and mechano-chemical signal transduction, including Rap1 signaling, stem cell pluripotency regulation, cell cycle, and TGF-β signaling pathways (Figure 4G). Gene enrichment analysis validated these findings, highlighting upregulation of bone development-related transcription factors, growth factors, signaling pathways, and bone collagen matrix proteins during induction (Figure 4H).
In vivo evaluation of lyophilized hDPSC microspheres on animal extraction socket healing
To validate the in vivo osteogenic efficacy of lyophilized hDPSC microspheres, we established a rat first molar extraction socket model with 5 experimental groups. Figure 5A presents representative images of the rat tooth extraction sockets following implantation with different materials, with panels A1, A2, A3, and A4 depicting the centrifuged cell pellet, artificial bone powder, lyophilized microspheres, and the sutured wound, respectively. Postoperative evaluation at 4 weeks revealed gradually diminishing bone depression from blank control group to lyophilized microsphere group, with lyophilized microsphere (Figure 5B5) and bone powder (Figure 5B4) groups showing essentially no residual depression, while the other 3 groups exhibited varying degrees of socket depression: Blank control group (Figure 5B1) > non–induced hDPSCs group(Figure 5B2) > 2D osteogenic-induced group(Figure 5B3). Although the bone powder group showed no significant surface depression, granular bone powder particles were visibly protruding.
Micro-CT analysis of rat mandibular specimens at 4 weeks revealed no significant low-density areas in extraction sockets of bone powder (Figure 5C4) and lyophilized microsphere (Figure 5C5) groups, while 2D-induced (Figure 5C3), non–induced (Figure 5C2), and blank (Figure 5C1) groups showed progressively increasing low-density areas. Figure 5D1 shows the mandibular first molar and its surrounding alveolar bone before tooth extraction. Multi–planar analysis demonstrated flatter mesiodistal alveolar ridge contours in bone powder (Figure 5D5) and microsphere (Figure 5D6) groups compared to concave profiles in other(Figure 5D2-4)groups, with numerous scattered low-density shadows observed in non–induced (Figure 5D3)and blank groups(Figure 5D2).Coronal views showed similar patterns, with bone powder and microsphere groups maintaining flat buccolingual contours while other groups exhibited concave profiles and scattered low-density shadows (Figure 5E). Transverse sections revealed scattered low-density shadows following the pattern: bone powder (Figure 5F5) ≈ lyophilized microsphere (Figure 5F6) < 2D-induced (Figure 5F4) < non–induced (Figure 5F3) < blank (Figure 5F2) groups. Quantitative micro-CT analysis (BV/TV, BMD, and trabecular thickness [Tb.Th]) confirmed these observations (Figure 5G). While blank, non–induced, and 2D-induced groups showed no significant differences among themselves, all 3 demonstrated significant differences (P <.001) compared to artificial bone powder and lyophilized microsphere groups. No significant difference in BV/TV was observed between artificial bone powder and lyophilized microsphere groups, though significant differences were detected in BMD (P <.001) and Tb.Th (P <.01). Histological analysis supported imaging findings. H&E staining revealed nearly complete socket repair in lyophilized microsphere (Figure 5H5) and bone powder (Figure 5H4) groups, with convex and flat buccolingual contours respectively. In contrast, 2D-induced (Figure 5H3), non–induced (Figure 5H2), and blank (Figure 5H1) groups all exhibited concave buccolingual profiles, with blank and non–induced groups showing significant buccal bone plate loss resembling natural human socket healing, and 2D-induced group developing “knife-edge” alveolar ridges. Masson’s trichrome staining showed similar blue staining intensity in bone powder (Figure 5I4) and lyophilized microsphere (Figure 5I5) groups, superior to other groups. Non–induced (Figure 5I2) and 2D-induced groups (Figure 5I3) displayed typical “knife-edge” ridges indicating post–extraction alveolar width reduction, while blank group (Figure 5I1) showed additional alveolar height loss. COL I immunohistochemistry revealed residual unabsorbed bone powder particles in the bone powder group (Figure 5J2), These residual particles are marked with black arrows while lyophilized microsphere group sockets consisted primarily of bone cells and matrix (Figure 5J4).
Evaluation of lyophilized human dental pulp stem cell (hDPSC) microspheres on rabbit extraction socket healing. (A) Extraction and transplantation procedure of rabbit mandibular first molars. (A1) Extraction socket. (A2) Implantation of lyophilized hDPSC microspheres. (A3) Implantation of artificial bone powder. (A4) post–suturing. (B) Bone recovery in extraction sockets at 4 weeks post–operation. Groups B1-B5 correspond to Figure 5B. (C) 3D CT images of rabbit mandibles at 4 weeks. Groups C1-C5 correspond to B1-B5. (D-F) Micro-CT images of extraction sockets in coronal (D), sagittal (E), and transverse (F) planes at 1 month post–operation. Groups D2-F2 to D6-F6 correspond to B1-B5. (G) Quantitative micro-CT analysis of bone regeneration (BV/TV, BMD, and Tb.Th) in rabbit extraction sockets. Significant differences at P <.05 (), P <.01 (), and P <.001 (,**). (H) H&E staining of coronal sections of extraction sockets. Groups H1-H5 correspond to B1-B5. The specific scale is shown in the figure.
Due to the relatively shallow sockets and strong inherent regenerative capacity in rats, which might minimize intergroup differences, we employed a rabbit mandibular first molar model with substantially longer roots relative to crowns. Group assignments mirrored the rat study. Figure 6A shows representative intraoperative scenes following material implantation into the rabbit extraction socket. The panels include the empty socket post–extraction (A1), the lyophilized DPSC microspheres (A2), the artificial bone powder (A3), and the sutured wound (A4). At 4 weeks post–operation, lyophilized microsphere (Figure 6B5) and bone powder (Figure 6B4) groups showed essentially complete resolution of socket depression, while the remaining 3 groups exhibited significant depression, most pronounced in blank (Figure 6B1), followed by non–induced (Figure 6B2), and 2D-induced (Figure 6B3) groups.
Micro-CT 3D reconstruction revealed persistent wide low-density areas in blank group sockets (Figure 6C1), slightly reduced but deeper defects in non–induced group (Figure 6C2), while 2D-induced (Figure 6C3), artificial bone powder (Figure 6C4), and lyophilized microsphere (Figure 6C5) groups showed better repair, though 2D-induced group displayed more prominent shadowing. Figure 6D1, E1, and F1 represents 3D reconstructed views from different orientations, depicting the spatial relationship between the tooth and the alveolar bone prior to extraction. Multi–planar analysis demonstrated more evident differences: coronal views showed significant radiolucent areas in blank (Figure 6D2) and non–induced (Figure 6D3) groups; 2D-induced group (Figure 6D4) showed no obvious depression but numerous low-density shadows; while artificial bone powder (Figure 6D5) and lyophilized microsphere (Figure 6D6) groups essentially restored original socket morphology, with the latter unexpectedly exceeding original height (possibly due to minor scanning orientation variations). Sagittal (Figure 6E) and transverse (Figure 6F) views confirmed these patterns, with blank (Figure 6E2, andF2) and non–induced groups (Figure 6E3 and F3) showing significant radiolucent voids, while artificial bone powder (Figure 6E5 and F5)and lyophilized microsphere (Figure 6E6 and F6) groups exhibited similar, reduced shadow numbers compared to 2D-induced group (Figure 6E4 and F4). Quantitative analysis (BV/TV, BMD, and Tb.Th) further validated these results (Figure 6G). While blank, non–induced, and 2D-induced groups showed similar patterns to rat results, significant differences in BV/TV were observed between blank (P <.01) and non–induced (P <.05) groups compared to 2D-induced group. Both blank and non–induced groups showed significant differences (P <.001) in BV/TV and BMD compared to artificial bone powder and lyophilized microsphere groups. The 2D-induced group demonstrated significant differences in BV/TV (P <.01), BMD (P <.001), and Tb.Th (P <.001 vs artificial bone powder; P <.01 vs lyophilized microspheres) compared to artificial bone powder and lyophilized microsphere groups. The lyophilized microsphere group and the artificial bone powder group showed comparable bone volume fractions (BV/TV), the artificial bone powder group exhibited significantly bone mineral density (BMD) and Tb.Th compared to the lyophilized microsphere group (P <.05). The observed disparity in BMD and Tb.Th, is likely attributable to the residual, non–integrated bone powder particles within the defect site. H&E staining correlated well with Micro-CT findings, showing essentially complete repair in bone powder (Figure 6H4) and lyophilized microsphere (Figure 6H5) groups, while blank (Figure 6H1), non–induced (Figure 6H2), and 2D-induced (Figure 6H3) groups exhibited obvious unrepaired defect area with lower density repair tissue in socket areas.
Discussion
The utilization of bone graft materials for reconstructing alveolar ridge defects following tooth extraction represents a standard clinical protocol for maintaining jaw morphology and facilitating subsequent prosthetic rehabilitation.18 However, the inherent bio-inertness of allogeneic bone, xenogeneic bone, and synthetic biomaterials limits their osteogenic potential, while their degradation kinetics often fail to synchronize with new bone formation, potentially leading to long-term bone volume deficiency.19 We propose that combining osteogenically competent live cells with microenvironments conducive to cellular function represents a viable strategy for achieving physiological regeneration and functional reconstruction of the alveolar ridge through bioactive tissue-engineered microtissues.20 This study successfully developed lyophilized hDPSC microspheres preconditioned via a 3D dynamic osteogenic induction system, demonstrating their remarkable capacity for promoting bone regeneration in both rat and rabbit extraction socket models.
Our research initially established and optimized a 3D dynamic osteogenic induction system for hDPSCs. Compared to conventional 2D culture, 3D microsphere culture better simulates in vivo cell-cell interactions and mechanical microenvironments.21 Our findings reveal significant extracellular matrix deposition and the formation of multiple “ossification center”-like structures within microspheres following 30 days of dynamic induction. Transcriptomic analysis further confirmed that late-stage induced microspheres were significantly enriched in biological processes including MAPK signaling pathway, collagen matrix assembly, and bone mineralization,22 closely resembling later stages of natural bone development.23 Particularly noteworthy, lyophilization processing successfully transformed these active 3D bone microtissues into room-temperature stable “off-the-shelf” preparations, addressing critical clinical translation bottlenecks in production scale-up, storage, and transportation of viable tissue-engineered products.24
In animal model validation, our lyophilized microspheres demonstrated substantial osteogenic advantages. Both micro-CT and histological analyses consistently showed that in rat and rabbit models, the lyophilized microsphere group significantly outperformed blank control, non–induced cell, and 2D-induced groups in terms of extraction socket bone repair extent, BMD, and trabecular thickness. Their efficacy was comparable to clinically widely used artificial bone powder, while exhibiting superior bone matrix maturity and integration.25 Notably, the lyophilized microsphere group maintained full alveolar ridge contour after repair, effectively preventing the common post–extraction “knife-edge” ridge atrophy,26 which is crucial for subsequent implant placement and other prosthetic treatments.27 We attribute their excellent repair efficacy to multiple factors: the 3D microstructure provides physical support for cell survival and functional expression;28 dynamic induction initiates robust osteogenic differentiation programs at earlier stages;29 while the lyophilization process preserves bioactivity while potentially conferring an ideal microstructure for guiding host cell migration and vascularization, as evidenced by the irregular porous structure observed in scanning electron micrographs.30
Despite these encouraging results, several limitations warrant consideration. First, the precise cellular fate of lyophilized microspheres in vivo remains incompletely elucidated.31 Although we attempted cell tracking, signal attenuation occurred during long-term in vivo experiments. Future studies should employ more stable genetic labeling methods for precise cell fate tracing.32 Second, although significant osteogenic effects were observed in both rat and rabbit models, the observation period was relatively short. Before advancing to clinical translation, validation in large animal models (eg, minipigs) is essential to evaluate repair efficacy and long-term stability under mechanical environments more closely resembling human jaw biomechanics.33
Conclusion
This study established a serialized 3D dynamic osteogenic induction culture system and successfully generated lyophilized hDPSC microspheres. In vitro experiments confirmed that this system effectively promotes extracellular matrix deposition, upregulates osteogenesis-related gene expression, and forms bone-like microstructures containing multiple “ossification centers.” In rat and rabbit extraction socket models, these lyophilized microspheres significantly enhanced bone regeneration, outperforming traditional 2D-induced cells in BMD, trabecular thickness, and alveolar ridge morphology preservation, while matching the performance of artificial bone powder and demonstrating unique advantages in tissue integration and maturity. In summary, these lyophilized microspheres successfully integrate readily available multipotent dental pulp stem cells, 3D dynamic induction culture methods simulating in vivo osteogenesis, and clinically convenient lyophilization technology, providing solid experimental evidence and promising solutions for developing a new generation of “off-the-shelf” active bone regeneration materials to overcome limitations of current alveolar ridge repair protocols.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Mac Beth Neil D , Donos Nikolaos, Mardas Nikos. Alveolar ridge preservation with guided bone regeneration or socket seal technique. A randomised, single-blind controlled clinical trial. Clin Oral Implants Res.2022;33(7):681-699. doi:10.1111/clr.1393335488477 PMC 9541021 · doi ↗ · pubmed ↗
- 2Jeong Seungho , Strauss Franz J, Shin Hae Jee, Park Jin-Young, Cha Jae-Kook, Lee Jung-Seok. Efficacy of collagenated bone substitutes for bone regeneration in two-wall-damaged extraction sockets without barrier membranes. Clin Oral Investig. 2025;29(4):201. doi:10.1007/s 00784-025-06281-w 40119996 · doi ↗ · pubmed ↗
- 3Robin Marc , Mouloungui Elodie, Castillo Dali Gabriel, et al Mineralized collagen plywood contributes to bone autograft performance. Nature. 2024;636(8041):100-107. doi:10.1038/s 41586-024-08208-z 39567697 PMC 11618095 · doi ↗ · pubmed ↗
- 4Wang H , Zhang Y, Zhang Y et al Activating Macrophage Continual Efferocytosis via Microenvironment Biomimetic Short Fibers for Reversing Inflammation in Bone Repair. Adv Mater. 2024;36(30):e 2402968. doi:10.1002/adma.20240296838706203 · doi ↗ · pubmed ↗
- 5Lv Zehui , Hu Tingting, Bian Yixin, et al A Mg Fe-LDH Nanosheet-Incorporated Smart Thermo-Responsive Hydrogel with Controllable Growth Factor Releasing Capability for Bone Regeneration. Adv Mater. 2023;35(5):e 2206545. doi:10.1002/adma.20220654536426823 · doi ↗ · pubmed ↗
- 6Keranmu Dilinuer , Nuermuhanmode Nijiati, Ainiwaer Ailimaierdan, et al Clinical application of concentrate growth factors combined with bone substitute in Alveolar ridge preservation of anterior teeth. BMC Oral Health. 2022;22(1):54. doi:10.1186/s 12903-022-02091-835241047 PMC 8895668 · doi ↗ · pubmed ↗
- 7Avanzi Ingrid Regina , Parisi Julia Risso, Souza Amanda, et al 3D-printed hydroxyapatite scaffolds for bone tissue engineering: A systematic review in experimental animal studies. J Biomed Mater Res B Appl Biomater. 2023;111(1):203-219. doi:10.1002/jbm.b.3513435906778 · doi ↗ · pubmed ↗
- 8Man Kenny , Joukhdar Habib, Manz Xue D, et al Bone tissue engineering using 3D silk scaffolds and human dental pulp stromal cells epigenetic reprogrammed with the selective histone deacetylase inhibitor MI 192. Cell Tissue Res. 2022;388(3):565-581. doi:10.1007/s 00441-022-03613-035362831 PMC 9110470 · doi ↗ · pubmed ↗