ABSTRACT
Background
In elderly patients, bone regeneration is impeded by age-related shifts in mesenchymal stem cell differentiation propensity toward adipogenesis over osteogenesis. We investigated whether DNA demethylation by 5‑aza‑2′‑deoxycytidine (5azaC) synergizes with Wnt Family Member 3A (Wnt3a) signaling to induce osteogenic potential in 3T3‑L1 pre-adipocytes, generating osteoblast-like cells.
Methods
3T3‑L1 pre-adipocytes were treated with 5azaC and/or Wnt3a. Osteogenic differentiation was assessed via ALP activity, mineralization assays, and marker expression. Transcriptomic and epigenomic profiling were performed and compared with MC3T3-E1 cells. Functional relevance of candidate genes was examined using siRNA knockdown.
Results
Transcriptomic and epigenomic profiling revealed that 5azaC and Wnt3a co-treatment induced broader gene expression and methylation changes than either treatment alone, closely resembling the osteogenic profile of MC3T3-E1 pre-osteoblasts. Among the overlapping differentially methylated and steadily expressed genes, Tyrosine 3-Monooxygenase/Tryptophan 5-Monooxygenase Activation Protein Eta (Ywhah) and Tyrosine 3-Monooxygenase/Tryptophan 5-Monooxygenase Activation Protein Epsilon (Ywhae) emerged as key regulators, whose knockdown notably enhanced Alpl expression even without 5azaC.
Conclusions
Combining 5azaC-induced demethylation with Wnt3a is a potent strategy to redirect pre-adipocytes toward osteogenesis. Identification of key targets like Ywhah and Ywhae provides mechanistic insight into trans-differentiation and suggests therapeutic potential for bone regeneration, particularly in elderly periodontal patients.
KEYWORDS: Trans-differentiation, 5-Aza-2’-deoxycytidine, Wnt3a, osteoblast, Ywhah, Ywhae
Plain Language Summary
When people get older, their bones don’t heal as well. This is partly because the special cells in bone marrow that can make bone tend to make more fat instead. Our study looked for a way to help these fat-making cells change into bone-making cells instead. We tested two treatments in the lab: one that changes chemical marks on DNA (like flipping switches on or off) and one that sends signals to cells (like a message). We treated lab-grown fat precursor cells with these substances and looked closely at their genes and DNA tags to see how they changed. We found that using both treatments together worked better than using either one alone. It caused big changes, making the cells look and act more like bone-making cells. We also found two important “controller genes,” called Ywhah and Ywhae, that play key roles in this change. When we turned down the activity of these control genes, it helped the cells make bone markers. This research shows a powerful way to potentially guide cells that would normally make fat toward making bone instead, using a combination of treatments. Finding the specific control genes gives us new ideas for developing more precise treatments to help bones heal better in the future, which could be especially helpful for older adults.
1. Introduction
Adipocytes adversely affect the outcomes of periodontal regenerative therapy for elderly patients by increasing marrow adiposity, reducing osteogenesis [1], promoting inflammatory responses through the secretion of pro-inflammatory cytokines, and inhibiting angiogenesis, which impede bone regeneration [2]. To avoid these effects of adipocytes, research is ongoing to induce mesenchymal stem cell (MSC) differentiation into osteoblasts while suppressing adipogenic differentiation using growth factors such as BMP2 and the activation of the Wnt/β-catenin signaling pathway [3]. Efforts are also being made to either reduce inflammation using anti-inflammatory agents and natural compounds or inhibit adipocyte maturation and free fatty acid release via PPARG inhibitors [4]. Furthermore, biomaterials and tissue engineering technologies are being developed to mitigate the impact of adipocytes and enhance bone regeneration [2,5].
MSCs are multipotent progenitor cells capable of differentiating into various mesenchymal tissues, including adipocytes and osteoblasts [6]. MSC lineage commitment is tightly regulated by the microenvironment, with adipogenic factors inhibiting osteogenesis and osteo-inductive factors suppressing adipogenesis, reflecting a reciprocal relationship. Wnt signaling pathways play a critical role in MSC differentiation, with the canonical Wnt/β-catenin pathway acting as a key regulator that inhibits adipogenesis while promoting osteogenesis, chondrogenesis, and myogenesis. Non-canonical Wnt signaling complements this role by supporting embryonic osteoblast migration, skeletal homeostasis, and bone remodeling. Wnt3a, a pivotal ligand, facilitates MSC differentiation into osteoblasts via both canonical and non-canonical pathways, underscoring its importance in skeletal tissue regeneration and homeostasis [7].
Our research has demonstrated the potential for trans-differentiation of pre-adipocyte cells (3T3-L1) into osteoblasts without inducing pluripotency [8–10]. The effect of Wnt3a on MSC differentiation into osteoblasts is restricted to cells with intrinsic osteogenic potential. Wnt3a induces osteogenesis by regulating the expression of osteogenic marker genes [11,12], such as Bmp2 and Alpl, whose expression is epigenetically suppressed in non-osteogenic cells. Treatment with the CpG demethylating agent 5-aza-2′-deoxycytidine (5azaC) facilitates the activation of the promoter regions of Bmp2 and Alpl, thereby restoring their responsiveness to Wnt3a [13,14]. This process enables osteogenic priming of adipocytes or fibroblasts, yielding cells with osteoblast‑like features. This process enables direct trans-differentiation of adipocytes or fibroblasts into osteoblasts.
DNA methylation is a critical epigenetic mechanism involved in gene expression, chromatin remodeling, and stem cell pluripotency. Histone demethylases promote osteogenic differentiation of MSCs by regulating key genes [15–17]. and DNA methyltransferase inhibitors such as 5azaC also show promise in inducing osteogenic differentiation by facilitating genome-wide DNA demethylation. However, unlike their individual roles, the combined impact of Wnt3a and 5azaC on MSC osteogenesis remains unclear. Furthermore, despite the potential of epigenetic modifiers like 5azaC in inducing differentiation, identifying their specific epigenetic targets, particularly during complex processes like trans-differentiation from preadipocytes into osteoblasts, poses a significant challenge. Identification of such targets holds significant potential for enabling targeted osteo-inductive therapies. To address these gaps and challenges, this study aimed to identify key regulatory targets involved in Wnt3a and 5azaC-mediated trans-differentiation of 3T3-L1 cells into osteoblasts. We employed integrated transcriptomic (RNA-seq) and epigenomic (methyl-seq) analyses using next-generation sequencing (NGS) technologies as a means to identify these targets and understand their role. Specifically, we examined regions of significant methylation changes induced by the treatment, analyzed their associated regulatory genes, and correlated the findings with gene expression patterns. Herein, we propose critical regulatory targets that may underpin this trans-differentiation process, offering insights for developing precise osteo-inductive therapeutic strategies.
2. Materials and methods
2.1. Cell culture
Mouse MC3T3-E1 pre-osteoblasts were cultured in alpha minimal essential medium as described previously [18]. 3T3-L1 pre-adipocytes were cultured in Dulbecco’s modified Eagle’s medium (DMEM, Logan, UT). All the media contained 10% fetal bovine serum or bovine calf serum with 1% penicillin/streptomycin. All the cell lines were purchased from ATCC (US, CRL-2593). An osteogenic medium containing 5 mm β-glycerophosphate and 50 μg/mL ascorbic acid was prepared.
2.2. Materials
Bioactive recombinant mouse Wnt3a proteins were purchased from R&D Systems (US, 1324-WN). The 5azaC was purchased from Sigma Aldrich (US, A2385).
2.3. RNA-seq data processing
Paired-end RNA-seq data obtained from three biological replicates of 3T3-L1 cells were analyzed under the following conditions: control, 5azaC treatment, Wnt3a treatment, and a combination of 5azaC and Wnt3a (hereinafter referred to as 5azaC+Wnt3a) treatment. MC3T3-E1 data from two biological replicates were analyzed. The average total number of reads for the replicates was 75.21 M, 78.49 M, 88.41 M, 67.36 M, and 63.53 M for the control, 5azaC, Wnt3a, 5azaC+Wnt3a, and MC3T3-E1 conditions, respectively. The median read length was 151 bp. The reads were aligned to the mouse genome (mm10) using Bowtie2 (v2.2.5), Samtools (v1.13), Bamtools (v2.5.1), Bedtools (v2.30.0), Biobambam (v2.0.183), Cutadapt (v4.0), and Sambamba (v0.8.2). Expression calling was performed using Salmon (v1.7.0) and Kallisto (v0.46.2).
2.4. Methyl-seq data processing
Paired-end methyl-seq data were obtained under the same conditions and with the same number of biological replicates as those for the RNA-seq data. The average total number of reads for the replicates was 114.97 M, 124.33 M, 129.81 M, 125.73 M, and 127.41 M for the control, 5azaC, Wnt3a, 5azaC+Wnt3a, and MC3T3-E1 conditions, respectively. The median read length was 151 bp. The reads were aligned to the mouse genome (mm10) using Bowtie2 (v2.2.5), Bismark (v0.22.3), Samtools (v1.13), and Cutadapt (v4.0).
2.5. Differential expression analysis (DEA), steady state expression analysis (SEA), and differential methylation analysis (DMA)
DEA, SEA, and DMA were performed using DESeq2 (v1.32.0). For differential expression within MSCs, each condition was compared against the control. A p-value cutoff of < 0.05 and a 1.5-fold change were used. A p-value cutoff of < 0.05 and a 1.2-fold change were used to compare steady state expression in MC3T3-E1 cells and those treated with 5azaC+Wnt3a. For differential methylation within the MSCs, each condition was compared against the control. A p-value cutoff of < 0.1 and a 1.2-fold change were used.
2.6. Gene set enrichment analysis (GSEA)
GSEA and over-representation enrichment analysis (ORA) were performed using Gene Ontology (GO) terms to identify biological pathways. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis was performed using the KEGG database (accessed January 2021) to identify signaling pathways. The org.Mm.eg.db (v3.17.0) and msigdbr (v7.5.1) databases were used for annotation. EnrichGO and gofilter of clusterProfiler (v4.8.1) were used for ORA, and GOSemSim (v2.26.0) and fgsea (v1.26.0) were used for GSEA. TxDb.Mmusculus.UCSC.mm10.knownGene and enrichKEGG were used for KEGG.
2.7. Protein–protein interaction (PPI) analysis
The overlaps between differentially expressed genes (DEGs) and differentially methylated genes (DMGs) were investigated to analyze the effects of 5azaC+Wnt3a treatment. Specifically, 283 genes were differentially expressed between MC3T3-E1 cells and those treated with 5azaC+Wnt3a, and 3252 genes were differentially methylated between control cells and those treated with the same combination. Overlapping these two gene sets yielded 42 genes for further analysis. Mouse-specific PPI analysis was performed using the Mouse Integrated Protein-Protein Interaction Reference (MIPPIE) database to expand the 42 genes, gain deeper insights, and identify a more robust set of significant genes. This analysis incorporated an additional 51 genes, contributing to a more comprehensive and reproducible gene set.
2.8. Reverse transcription-PCR and Quantitative real-time PCR
RNA was isolated using QIAzol lysis reagent (Qiagen). A reverse transcription kit was purchased from TAKARA Bio. Quantitative real-time PCR for murine Bmp2 and Alpl was performed with previously used primers [19] using Takara SYBR premix Ex Taq (Takara Bio) on an Applied Biosystems 7500 Real Time PCR system. PCR primers were synthesized by Integrated DNA Technologies (Coralville, IA). All samples were run in duplicates, and the relative mRNA expression levels were normalized to those of glyceraldehyde-3-phosphate dehydrogenase mRNA. The primer sets for real-time PCR are published elsewhere [19].
2.9. Alkaline phosphatase (ALP) staining
A standardized kit for ALP staining was purchased from Sigma Aldrich. Cells were washed twice with phosphate-buffered saline and stained for ALP according to the manufacturer’s instructions [20].
2.10. Statistical analysis
Quantitative data were expressed as means ± SDs. Each experiment was performed at least three times, and the results of one representative experiment are shown herein. Significant differences were analyzed using Student’s t-test. A p-value < 0.05 was considered statistically significant.
3. Results
3.1. Synergistic effects of 5azaC and Wnt3a on osteogenic differentiation and transcriptomic shifts in 3T3-L1 cells
Pre-adipocyte (3T3-L1) cells were treated with 5azaC for 1 day under previously established conditions to inhibit DNA methylation, after which Wnt3a was administered concurrently with osteogenesis induction (Figure 1(A)). Under these conditions, non-osteogenic 3T3-L1 cells successfully expressed the osteogenic marker gene Alpl in response to treatment with 5azaC, Wnt3a, and an osteogenic medium (Figure 1(B)). Consistent with earlier findings, neither 5azaC nor Wnt3a individually induced Alpl expression in 3T3-L1 cells; clear Alpl induction was observed only under the 5azaC+Wnt3a treatment condition.
Figure 1.
Synergistic 5azaC-Wnt3a induction of transdifferentiation in 3T3-L1 cells.
(A) Cells were treated with 5′-aza-dC (10 μM) for 24 h individually, and then Wnt3a was added sequentially and cultured in the osteogenic media including β-glycerophosphate (5 mM) and ascorbic acid (50 μg/ml) for 3 days. Day 0 means initiation of differentiation. (B) Alp staining of 3T3-L1 cells. ALP activity was determined by cytochemical staining. (C – D) Principal Component Analysis (PCA) was performed on mRNA expression data obtained from cells treated with 5azaC, Wnt3a, a combination of 5azaC and Wnt3a, and a control group. Each dot in the plot represents an individual biological replicate. (E) A volcano plot was generated to visualize the differential gene expression profiles between the control group and groups treated with 5azaC, Wnt3a, or a combination of 5azaC and Wnt3a. Genes with significant differential expression, defined as a 5% p-value threshold and a 1.5-fold change, are demarcated by solid lines. Up-regulated genes are colored red, while down-regulated genes are colored blue.
Based on the model described above, RNA-seq was performed on each experimental group (3T3-L1 cells (control) and the 5azaC, Wnt3a, and 5azaC+Wnt3a groups), and their transcriptomes were compared to that of MC3T3-E1 pre-osteoblasts cultured under the same osteogenesis induction conditions (Figure 1(C)). Principal component analysis (PCA) of the RNA-seq data revealed a marked distinction between 3T3-L1 and MC3T3-E1 cells, whereas the 5azaC, Wnt3a, and 5azaC+Wnt3a groups clustered between these two cell types. Notably, the 5azaC group displayed greater within-group variability compared with those of the other treatment groups. This pattern was even more pronounced when examining the subset of GO-defined ossification-related genes (Figure 1D). The 5azaC+Wnt3a group exhibited the most pronounced shift toward the transcriptomic profile of MC3T3-E1 cells.
DEA indicated that 166 genes were upregulated, and 57 genes were downregulated in the 5azaC group. In contrast, the Wnt3a group displayed 300 upregulated and 289 downregulated genes, indicating that Wnt3a had a more pronounced impact on DEGs than 5azaC. The 5azaC+Wnt3a group exhibited the greatest number of DEGs, with 503 upregulated and 476 downregulated genes, suggesting that this dual treatment exerts a more substantial effect on gene expression than either individual treatment (Figure 1E).
3.2. Comprehensive gene expression modulation induced by synergistic combination of 5azaC and Wnt3a
Genes that were uniquely dysregulated after each treatment were selected from the DEGs. Considering the direction of expression change, these genes were categorized into six distinct gene sets (Figure 2A). Analysis of these gene sets revealed that those specifically upregulated and downregulated by the 5azaC+Wnt3a combination were the largest, consistent with the observed DEG distribution pattern. ORA was performed to investigate the biological processes associated with these gene sets (Figure 2(B–F)). Genes specifically upregulated by 5azaC were associated with chemotaxis and migration, whereas downregulated genes were not significantly enriched. Genes specifically upregulated by Wnt3a were associated with ROS and ATP, whereas downregulated genes were related to cell motility. Finally, genes specifically upregulated by the 5azaC+Wnt3a combination were associated with peptide hormones, whereas downregulated genes were linked to cilium organization. Thus, mRNA-seq-based transcriptome analysis indicated that the 5azaC+Wnt3a combination elicited the most pronounced gene expression changes compared to those of the control, exceeding the effects observed with 5azaC and Wnt3a alone.
Figure 2.
Extensive gene expression modulation by combined 5azaC and Wnt3a.
(A) Genes exhibiting condition-specific expression patterns compared to the control were categorized into six distinct sets. The expression levels of genes within each set were visualized using a heatmap. Up-regulated genes are colored red, while down-regulated genes are colored blue. (B-F) Overrepresentation enrichment analysis (ORA) was performed using GO terms on five gene sets expressed in Figure 2(A).
3.3. Epigenomic profiling reveals distinct demethylation patterns induced by 5azaC and Wnt3a
Demethylation is a critical factor in the osteogenic differentiation of 3T3-L1 cells. Hence, methyl-seq-based epigenomic analysis was performed to investigate the demethylation effects of each treatment condition. The results of PCA were similar at the methylation and transcriptomic levels. Although biological replicates within each condition clustered closely together, MSCs subjected to different treatments showed clear distinctions when compared with MC3T3-E1 cells, both across the entire mm10 reference genome and specifically within an ossification-related gene set (Figure 3(A) and (B)). DMGs were then identified by comparing each treatment group to the control (Figure 3(C)). 5azaC-treated cells showed a higher number of genes with reduced methylation compared to that in the control, yielding a downregulated-to-upregulated gene ratio of 1.67 (128 upregulated and 214 downregulated), consistent with the genome-wide demethylation property of 5azaC. A similar trend emerged when this analysis was extended to all treatment conditions, with a 1.54-fold increase in genes exhibiting reduced methylation compared to those showing increased methylation (381 upregulated and 589 downregulated). Further examination revealed that 5azaC treatment produced the highest number of both upregulated and downregulated DMGs, and ORA of these DMGs indicated their involvement in diverse biological processes, including autophagy, vesicle fusion, and hormone-related functions (Figure 3(D) and supplementary Figure S1). Minimal overlap of DMGs was observed across the different treatment groups, suggesting that demethylation triggered by 5azaC alone or in combination with Wnt3a could regulate distinct biological pathways. Finally, GSEA corroborated that each condition affected different biological processes (Figure 3E).
Figure 3.
Distinct 5azaC- and Wnt3a-induced demethylation revealed by methyl-seq analysis.
(A-B) Principal Component Analysis (PCA) was performed on methylation level data obtained from cells treated with 5azaC, Wnt3a, a combination of 5azaC and Wnt3a, and a control group. Gene sets were analyzed using either all genes in the mm10 reference genome or a subset of genes related to ossification. Each dot in the plot represents an individual biological replicate. (C) The scatter plot illustrates the relationship between genes exhibiting significant differential methylation induced by 5azaC and Wnt3a. Each dot represents a gene with a significant change in methylation status between the compared conditions C-E, defined by a p-value threshold of 10% and a 1.2-fold change. (D) An upset plot was generated to illustrate the distribution of differentially methylated genes in each condition, categorized as up regulated and down regulated. (E) Gene Set Enrichment Analysis (GSEA) was performed using GO term annotations to identify enriched functional categories among differentially methylated genes in each treatment condition.
3.4. Integrated epigenomic and transcriptomic analyses reveal key epigenetic targets of 5azaC+Wnt3a-mediated trans-differentiation
The demethylation process induced by 5azaC+Wnt3a treatment may generate distinct effects compared to those exerted by 5azaC alone. Given the established role of Wnt3a in promoting osteogenic differentiation, 5azaC and Wnt3a were postulated to act synergistically to enhance this differentiation process. To identify specific markers associated with osteogenesis, an integrated analysis that combined mRNA-seq – based transcriptome profiling with methyl-seq – based epigenomic profiling was conducted.
In particular, the DMGs of the control and 5azaC+Wnt3a groups, as well as the steadily expressed genes (SEGs) shared by the MC3T3-E1 and 5azaC+Wnt3a groups (Supplementary Figure S2), were utilized to investigate the significant methylation changes and the resemblance of expression to that of MC3T3-E1 cells after combined 5azaC and Wnt3a treatment (relative to the control). The overlapping DMGs and SEGs yielded a significant gene set with methylation levels markedly altered compared to those of the control but with expression levels closely mirroring those of MC3T3-E1 cells (Figure 4(A)). Integration of gene expression profiles with methylation levels yielded two distinct gene subsets: one comprising 23 genes with expression patterns in regions of increased methylation resembling those observed in the positive control group, and another consisting of 19 genes with expression being associated with regions displaying decreased methylation (Figure 4(B)). PPI analysis was performed to elucidate the intracellular signaling pathways associated with the genes identified using MIPPIE (Figure 4(B)), and thereby refine and expand this significant gene set for more reproducible insights into how 5azaC- and Wnt3a-induced epigenetic changes contribute to osteogenic differentiation (Figure 4(C)).
Figure 4.
Integrated methyl-seq and RNA-seq analysis of 5azaC-Wnt3a mediated transdifferentiation. (A) line plots were generated to visualize the methylation and expression level for the significant genes identified in A. Genes with increased methylation levels are depicted in red, while those with decreased methylation levels are shown in blue. (B) a heatmap was generated to visualize the methylation and expression levels of the significant genes identified in integrative analysis. (C) the network plot shows a protein-protein interaction network constructed using MIPPIE. The blue circles represent the significant genes identified in A. while the gray circles represent their interacting protein partners.
3.5. Synergistic epigenetic regulation by 5azaC and Wnt3a identifies Ywhah and Ywhae as key mediators of 3T3-L1 trans-differentiation
Building on the integrated transcriptomic and epigenomic findings—5azaC+Wnt3a-induced epigenetic changes led to the identification of a significant gene set with methylation and expression patterns closely mirroring those of MC3T3-E1 cells – KEGG pathway analysis was conducted to uncover key signaling cascades involved in osteogenesis (Supplementary Figure S3). Among the enriched pathways identified were the PI3K – Akt and Hippo pathways, both known to play pivotal roles in guiding MSCs toward the osteoblast lineage. Within this significant gene set, Ywhah and Ywhae emerged as shared components of both the PI3K – Akt and Hippo pathways, displaying expression levels in the 5azaC+Wnt3a treatment group that closely paralleled those observed in MC3T3-E1 cells (Figure 5(A)).
Figure 5.
Ywhah and Ywhae validated as crucial regulators in 5azaC-Wnt3a mediated transdifferentiation. (A) the bar plot displays the expression patterns of Ywhah and Ywhae, two genes commonly involved in the PI3K-Akt and Hippo signaling pathways as identified in E, under control, a combined 5azaC and Wnt3a treatment and MC3T3-E1. (B) quantitative gene expression Ywhah and Ywhae of each group. Data were confirmed by triplicate determinations from each mRNA and at least three biological replicates. Values are the means ± S.D. of a representative triplicate determination. A value of p < 0.05 was considered statistically significant marked asteroid. (C) Alp activity was determined by cytochemical staining.
These findings were verified by performing qPCR to measure the expression of Ywhah and Ywhae in the 5azaC+Wnt3a group, which yielded results consistent with the patterns identified by RNA-seq. Notably, both genes, which are relatively highly expressed in 3T3-L1 cells, showed lower expression levels – comparable to those in MC3T3-E1 cells – following combined drug treatment (Figure 5(B)). These findings were verified using trans-differentiation assays using 3T3-L1 cells. Although treatment with Wnt3a alone did not induce osteogenic differentiation, knockdown of Ywhah and Ywhae markedly increased Alpl expression. Moreover, each gene appeared to function independently in regulating this process (Figure 5(C)).
Therefore, Ywhah and Ywhae, potentially activated via the synergistic demethylation and signaling interplay induced by 5azaC and Wnt3a, may coordinate the PI3K – Akt and Hippo pathways to foster osteogenic differentiation of MSCs.
4. Discussion
Short-term 5azaC pretreatment, followed by Wnt3a administration under osteogenic conditions, synergistically enhances the osteogenic differentiation of 3T3-L1 cells. Notably, only 5azaC+Wnt3a treatment induces Alpl expression, whereas each factor alone fails to do so, underscoring the importance of both epigenetic priming and osteo-inductive signaling. This synergy is further corroborated by the larger number of DEGs in the 5azaC+Wnt3a group and its transcriptomic shift toward MC3T3-E1 cells. Overall, these findings suggest that 5azaC-induced global demethylation, when coupled with Wnt3a signaling, reprograms 3T3-L1 cells toward an osteogenic lineage more effectively than either treatment alone, providing a promising strategy for bone regenerative applications.
Co-treatment with 5azaC and Wnt3a orchestrates a broader and more pronounced shift in gene expression than that induced by either agent alone, indicating a synergistic interaction. Categorization of genes uniquely dysregulated under each treatment, and examination of the associated biological processes, indicate that 5azaC and Wnt3a modulate distinct pathways. Nevertheless, their combination elicits additional transcriptional changes, particularly in peptide hormone regulation and cilium organization, that exceed the sum of their individual effects. Such a cooperative mechanism underscores the potential of integrating epigenetic modulators with specific signaling cues to achieve more robust osteogenic or tissue-specific outcomes.
The synergy of 5azaC and Wnt3a not only induces a demethylation pattern different from that mediated by 5azaC alone but also highlights a subset of genes with altered methylation and expression profiles mirroring those of MC3T3-E1 cells. Integrated methyl-seq and RNA-seq analysis pinpointed specific gene groups associated with both increased and decreased methylation, suggesting that selectively modulating these epigenetic markers is central to enhancing osteogenic differentiation. Moreover, PPI analysis of these genes indicated potential signaling networks through which 5azaC+Wnt3a co-treatment drives trans-differentiation, offering an insight into how synchronized epigenetic and osteogenic signals can more effectively promote the osteoblastic phenotype.
Ywhah and Ywhae act as crucial nodes at the intersection of the PI3K – Akt and Hippo pathways, highlighting their pivotal roles in orchestrating osteogenic trans-differentiation under 5azaC+Wnt3a treatment. The PI3K – Akt pathway is widely recognized for its roles in cell survival [21], proliferation [22], and metabolism [23]. In the context of osteogenic differentiation, Akt activation can increase the activity of key transcription factors such as Runx2, which is pivotal for osteoblast lineage commitment [24]. Enhanced PI3K – Akt signaling often supports osteoblast maturation by promoting anabolic processes and inhibiting apoptosis, thereby creating a cellular environment conducive to bone matrix production and mineralization. The Hippo pathway, primarily known for its regulation of organ size and cell proliferation, exerts its effects through core components and downstream effectors such as yes-associated protein (YAP) and transcriptional coactivator with PDZ-binding motif (TAZ). When the Hippo pathway is modulated in favor of nuclear YAP/TAZ accumulation, these effectors can interact with osteogenic transcription factors (including Runx2) to enhance the expression of bone-specific genes [25,26]. Consequently, appropriate Hippo signaling activity is crucial for balancing cell growth and differentiation, and thereby promoting osteoblast formation and function.
Together, the PI3K – Akt and Hippo pathways intersect at various nodes of intracellular signaling, and their coordinated regulation can significantly augment osteogenic potential. Dysregulation of either pathway can hamper normal bone formation, whereas their targeted activation, potentially through epigenetic modulation or specific signaling cues, may yield synergistic benefits for bone regeneration and tissue engineering.
Ywhah and Ywhae encode members of the 14–3–3 protein family, which bind to phosphorylated serine/threonine residues on target proteins, thereby influencing their stability, localization, and interaction with other signaling components [27,28]. In the PI3K – Akt pathway, 14–3–3 proteins can sequester or stabilize phosphorylated targets (such as certain transcription factors or downstream effectors) [29], modulating cell survival and proliferation signals essential for osteoblast differentiation. In the Hippo pathway, 14–3–3 proteins often bind to phosphorylated YAP/TAZ, retaining them in the cytoplasm and preventing their nuclear translocation [30]. This interaction fine-tunes the balance between cell growth and differentiation, so that YAP/TAZ can appropriately activate osteogenic gene expression only under the right conditions. Consequently, Ywhah and Ywhae serve as pivotal “molecular adapters” in both pathways, helping to orchestrate the phosphorylation-dependent switches that guide cells toward an osteoblastic fate. The combined demethylation and signaling interplay via 5azaC and Wnt3a may activate distinct regulatory networks, positioning Ywhah and Ywhae as key mediators in the transition of 3T3-L1 cells from an adipogenic to an osteogenic fate.
This study has thus investigated how the combination of 5azaC-induced demethylation and Wnt3a signaling synergistically drives 3T3-L1 pre-adipocytes toward an osteogenic fate, as shown by transcriptomic and epigenomic profiling, which identified that Ywhah and Ywhae are the pivotal regulatory nodes in this process.
5. Conclusion
Collectively, our data establish that the synergistic application of 5‑aza‑2′‑deoxycytidine – mediated DNA demethylation and Wnt3a signaling enhances osteogenic potential in pre‑adipocytes, promoting an osteoblast‑like phenotype. Multi‑omic integration, reinforced by functional assays, positions the 14‑3‑3 family members Ywhah and Ywhae as critical nodes in this lineage reprogramming, thereby clarifying key mechanistic underpinnings. From a translational perspective, this dual‑modality approach highlights a promising therapeutic avenue for augmenting bone repair in clinical scenarios characterized by compromised osteogenesis, such as age‑related skeletal fragility or focal bone defects. Moreover, the identification of Ywhah and Ywhae as actionable regulators provides a rational foundation for the development of more precise osteo‑inductive interventions.
Supplementary Material
Funding Statement
This manuscript was funded by the Bio & Medical Technology Development Program of the National Research Foundation (NRF) of the Korean government (MSIT) (No. RS-2022-NR067350, RS-2023-00207971, RS-2024-00349549) and Creative-Pioneering Researchers Program through Seoul National University. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Author contributions
Young-Dan Cho was mainly responsible for the conception, design, and methodology. Sungtae Kim was mainly responsible for the interpretation and extraction of data. Seung Gwa Park was mainly responsible for the methodology. Ki-Tae Kim. was mainly responsible for the conception and data interpretation. Seung Gwa Park was mainly responsible for the acquisition, analysis and interpretation of data. Ki-Tae Kim and Woo-Jin Kim drafted the manuscript. Young-Dan Cho and Sungtae Kim critically revised the manuscript. All authors gave final approval of the version to be published. Each author agrees to be accountable for all aspects of the accuracy or integrity of the work
Disclosure statement
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
Medical writing support was provided from editage (job code, GDHOZ_63) and was funded by NRF of the Korean government (MSIT).
Reviewer disclosures
Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.
Article highlights
Combining 5azaC‑induced demethylation with Wnt3a signaling is a powerful strategy for enhancing osteogenic potential in pre‑adipocytes.
5azaC and Wnt3a co-treatment synergistically induces extensive transcriptomic and epigenomic changes that mirror the osteogenic profile of osteoblasts.
Integrated multi-omics analysis identified Ywhah and Ywhae as pivotal regulatory targets in this trans-differentiation process.
Ywhah and Ywhae function as crucial nodes linking PI3K-Akt and Hippo pathways, orchestrating osteogenic fate.
Functional validation shows that knockdown of Ywhah or Ywhae enhances osteogenic differentiation, highlighting their regulatory role.
Identification of Ywhah and Ywhae offers novel mechanistic insights and suggests potential targets for bone regenerative therapies.
Data availability statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Supplementary material
Supplemental data for this article can be accessed online at https://doi.org/10.1080/17501911.2025.2544513
References
Papers of special note have been highlighted as either of interest (•) or of considerable interest (••) to readers.
- 1.Ledesma-Martinez E, Mendoza-Nunez VM, Santiago-Osorio E, et al. Mesenchymal stem cells for periodontal tissue regeneration in elderly patients. J Gerontol A Biol Sci Med Sci. 2019. Aug 16;74(9):1351–1358. doi: 10.1093/gerona/gly227 [DOI] [PubMed] [Google Scholar]; • This study provides a broad interpretation of MSC regeneration.
- 2.Dos Santos DM, Moon JI, Kim DS, et al. Hierarchical chitin Nanocrystal-based 3D printed dual-layer membranes hydrogels: a dual drug delivery Nano-platform for periodontal tissue regeneration. ACS Nano. 2024. Sep 3;18(35):24182–24203. doi: 10.1021/acsnano.4c05558 [DOI] [PubMed] [Google Scholar]
- 3.Zhou S, Chen S, Jiang Q, et al. Determinants of stem cell lineage differentiation toward chondrogenesis versus adipogenesis. Cell Mol Life Sci. 2019. May;76(9):1653–1680. doi: 10.1007/s00018-019-03017-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Santini E, Fallahi P, Ferrari SM, et al. Effect of PPAR-gamma activation and inhibition on glucose-stimulated insulin release in INS-1e cells. Diabetes. 2004. Dec;Suppl 53(suppl_3):S79–83. doi: 10.2337/diabetes.53.suppl_3.S79 [DOI] [PubMed] [Google Scholar]
- 5.Yoon H, Park SG, Kim HJ, et al. Nicotinamide enhances osteoblast differentiation through activation of the mitochondrial antioxidant defense system. Exp Mol Med. 2023. Jul;55(7):1531–1543. doi: 10.1038/s12276-023-01041-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Chen Q, Shou P, Zheng C, et al. Fate decision of mesenchymal stem cells: adipocytes or osteoblasts? Cell Death Differ. 2016. Jul;23(7):1128–1139. doi: 10.1038/cdd.2015.168 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Cho YD, Kim BS, Kim WJ, et al. Histone acetylation together with DNA demethylation empowers higher plasticity in adipocytes to differentiate into osteoblasts. Gene. 2020. Apr 5;733:144274. doi: 10.1016/j.gene.2019.144274 [DOI] [PubMed] [Google Scholar]; • This study provides a comprehensive analysis of the relationship between adipocytes and osteoblasts in relation to methylation.
- 8.Yoon H, Park SG, Shin H-R, et al. Unraveling the dynamics of osteoblast differentiation in MC3T3-E1 cells: transcriptomic insights into matrix mineralization and cell proliferation. Bone. 2025. May 01;194:117442. doi: 10.1016/j.bone.2025.117442 [DOI] [PubMed] [Google Scholar]
- 9.Cho YD, Yoon WJ, Kim WJ, et al. Epigenetic modifications and canonical wingless/int-1 class (WNT) signaling enable trans-differentiation of nonosteogenic cells into osteoblasts. J Biol Chem. 2014. Jul 18;289(29):20120–20128. doi: 10.1074/jbc.M114.558064 [DOI] [PMC free article] [PubMed] [Google Scholar]; • This paper provides a broad interpretation of osteoblast differentiation induced by Wnt signaling.
- 10.Cho Y, Kim B, Bae H, et al. Direct gingival Fibroblast/osteoblast transdifferentiation via epigenetics. J Dent Res. 2017. May;96(5):555–561. doi: 10.1177/0022034516686745 [DOI] [PubMed] [Google Scholar]; • This study provides extensive insights into transdifferentiation influenced by epigenetic changes.
- 11.Qi H-Z, Ye Y-L, Suo Y, et al. Wnt/β-catenin signaling mediates the abnormal osteogenic and adipogenic capabilities of bone marrow mesenchymal stem cells from chronic graft-versus-host disease patients. Cell Death Dis. 2021. Mar 23;12(4):308. doi: 10.1038/s41419-021-03570-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Cho YD, Kim WJ, Yoon WJ, et al. Wnt3a stimulates mepe, matrix extracellular phosphoglycoprotein, expression directly by the activation of the canonical Wnt signaling pathway and indirectly through the stimulation of autocrine bmp-2 expression. J Cell Physiol. 2012. Jun;227(6):2287–2296. doi: 10.1002/jcp.24038 [DOI] [PubMed] [Google Scholar]; • This study provides a comprehensive analysis the canonical Wnt signaling pathway on bone markers.
- 13.Okamoto M, Udagawa N, Uehara S, et al. Noncanonical Wnt5a enhances Wnt/β-catenin signaling during osteoblastogenesis. Sci Rep. 2014. Mar 27;4(1):4493. doi: 10.1038/srep04493 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Zhou GS, Zhang XL, Wu JP, et al. 5-azacytidine facilitates osteogenic gene expression and differentiation of mesenchymal stem cells by alteration in DNA methylation. Cytotechnology. 2009. Jul;60(1–3):11. doi: 10.1007/s10616-009-9203-2 [DOI] [PMC free article] [PubMed] [Google Scholar]; • Fundamental study showing 5-Azacytidine promotes osteogenic differentiation of mesenchymal stem cells via DNA methylation alteration.
- 15.Shan L, Yang X, Liao X, et al. Histone demethylase KDM7A regulates bone homeostasis through balancing osteoblast and osteoclast differentiation. Cell Death Dis. 2024. Feb 12;15(2):136. doi: 10.1038/s41419-024-06521-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Deng P, Chen Q-M, Hong C, et al. Histone methyltransferases and demethylases: regulators in balancing osteogenic and adipogenic differentiation of mesenchymal stem cells. Int J Oral Sci. 2015. Dec 01;7(4):197–204. doi: 10.1038/ijos.2015.41 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Lai Z, Shu Q, Song Y, et al. Effect of DNA methylation on the osteogenic differentiation of mesenchymal stem cells: concise review. Front Genet. 2024;15:1429844. doi: 10.3389/fgene.2024.1429844 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Cho YD, Yoon WJ, Woo KM, et al. Molecular regulation of matrix extracellular phosphoglycoprotein expression by bone morphogenetic protein-2. J Biol Chem. 2009. Sep 11;284(37):25230–25240. doi: 10.1074/jbc.M109.008391 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kim WJ, Kim BS, Kim HJ, et al. Intratesticular Peptidyl prolyl isomerase 1 protein delivery using cationic lipid-coated fibroin nanoparticle complexes rescues male infertility in mice. ACS Nano. 2020. Oct 27;14(10):13217–13231. doi: 10.1021/acsnano.0c04936 [DOI] [PubMed] [Google Scholar]
- 20.Kim WJ, Ryu JH, Kim JW, et al. Bone-targeted lipoplex-loaded three-dimensional bioprinting bilayer scaffold enhanced bone regeneration. Regen Biomater. 2024;11:rbae055. doi: 10.1093/rb/rbae055 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Datta SR, Brunet A, Greenberg ME.. Cellular survival: a play in three akts. Genes Dev. 1999. Nov 15;13(22):2905–2927. doi: 10.1101/gad.13.22.2905 [DOI] [PubMed] [Google Scholar]
- 22.Lau MT, Leung PC.. The PI3K/Akt/mTOR signaling pathway mediates insulin-like growth factor 1-induced E-cadherin down-regulation and cell proliferation in ovarian cancer cells. Cancer Lett. 2012. Dec 30;326(2):191–198. doi: 10.1016/j.canlet.2012.08.016 [DOI] [PubMed] [Google Scholar]
- 23.Lien EC, Lyssiotis CA, Cantley LC. Metabolic reprogramming by the PI3K-Akt-mTOR pathway in cancer. Recent Results Cancer Res. 2016;207:39–72. [DOI] [PubMed] [Google Scholar]
- 24.Tong Y, Feng W, Wu Y, et al. Mechano-growth factor accelerates the proliferation and osteogenic differentiation of rabbit mesenchymal stem cells through the PI3K/AKT pathway. BMC Biochem. 2015. Jan 15;16(1):1. doi: 10.1186/s12858-015-0031-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Pan C, Hao X, Deng X, et al. The roles of Hippo/YAP signaling pathway in physical therapy. Cell Death Discov. 2024. Apr 26;10(1):197. doi: 10.1038/s41420-024-01972-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Wang H, Yu H, Huang T, et al. Hippo-YAP/TAZ signaling in osteogenesis and macrophage polarization: therapeutic implications in bone defect repair. Genes Dis. 2023. Nov;10(6):2528–2539. doi: 10.1016/j.gendis.2022.12.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Qing Y, Sun L, Yang C, et al. Dysregulated 14–3–3 family in peripheral blood leukocytes of patients with schizophrenia. Sci Rep. 2016. Mar 31;6(1):23791. doi: 10.1038/srep23791 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Xu Y, Ren J, He X, et al. YWHA/14–3–3 proteins recognize phosphorylated TFEB by a noncanonical mode for controlling TFEB cytoplasmic localization. Autophagy. 2019. Jun;15(6):1017–1030. doi: 10.1080/15548627.2019.1569928 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.He J, Zeng F, Jin XI, et al. YWHAH activates the HMGA1/PI3K/AKT/mTOR signaling pathway by positively regulating fra-1 to affect the proliferation of gastric cancer cells. Oncol Res. 2023;31(4):615–630. doi: 10.32604/or.2023.029698 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Ou WB, Lundberg MZ, Zhu S, et al. YWHAE-NUTM2 oncoprotein regulates proliferation and cyclin D1 via RAF/MAPK and Hippo pathways. Oncogenesis. 2021. May 4;10(5):37. doi: 10.1038/s41389-021-00327-w [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.