Epigenetic regulation of dental-derived stem cells and their application in pulp and periodontal regeneration

Why this review is needed and who it is intended for

Endodontic and periodontal diseases are common and frequently-occurring diseases of the oral cavity. Traditional treatments for endodontics include root canal therapy and apical surgery. These treatments provide good relief of symptoms but cannot avoid tooth discoloration and pulp inactivation and necrosis. Periodontitis can cause destruction of the alveolar bone and also tooth loss. Traditional periodontal treatment may meet the needs of most patients with periodontitis, but fail to achieve periodontal tissue regeneration. In recent years, with the development of tissue engineering and the discovery of dental-derived stem cells, pulp and periodontal regeneration have become a potential treatment for these two diseases. The discovery of the mechanism of epigenetic modification of dental-derived stem cells has led to the potential use of epigenetic regulation in dental pulp and periodontal regeneration. However, no article targeting the same subject has been published to provide a clear guidance for clinic trials.

This article aims to provide perspectives for dentists and researchers investigating dental pulp and periodontal regeneration using epigenetic regulated dental-derived stem cells, in order to reduce failures and improve the prognosis for patients. It is our hope that these techniques can develop more secure and effective treatment approaches.

Properties of dental-derived stem cells

Stem cells have significant proliferative capacity and multi-directional differentiation potential. The development of regenerative medicine has led to the use of stem cells in the repair of damaged cells, tissues, and organs, which have low self-healing abilities, with excellent safety and efficacy. Dental-derived stem cells mediate the process of tooth regeneration by upregulating odontogenic and angiogenic capacity in the form of secreted exosomes (Exo) (Mai et al., 2021). Dental-derived stem cells have been used in relevant clinical trials in the fields of periodontal tissue, maxillofacial bone tissue repair, and apical pulp disease treatment (Feng et al., 2010; Giuliani et al., 2013; Nakashima & Iohara, 2014). However, the preclinical models for the use of stem cells in nerve regeneration, diabetes, and autoimmune diseases have only been preliminarily validated (Kanafi et al., 2013; Mead et al., 2017; Shimojima et al., 2016).

Properties and potential of DPSCs

Dental pulp stem cells have a relatively high proliferation rate, a low cellular senescence, multi-directional differentiation potential, and immunomodulatory properties (Ma et al., 2019; Mortada & Mortada, 2018). Damage to dental pulp causes dental pulp stem cells to induce the formation of various cellular components, including odontoblasts, to replenish damaged cells. In addition, in vitro studies have shown that DPSCs can differentiate into neural-like cells, osteoblasts, chondrocytes, adipocytes, muscle cells, endothelial cells, hepatocytes, and renal pericytes, etc. (Barros et al., 2015; Gandia et al., 2008; Saito et al., 2015). DPSCs can effectively promote pulp and periodontal regeneration. Guo et al. (2021) combined decellularized tooth matrix (DTM) with human dental pulp stem cells and successfully achieved the regeneration of dental pulp and periodontal tissue.

Properties and potential of SHEDs

SHEDs are isolated from the pulp tissue of exfoliated deciduous teeth, whose expression level of osteocalcin and alkaline phosphatase activity are higher compared to DPSCs (Koyama et al., 2009). SHEDs can express osteocalcin and RUNX-2 markers, resulting in the differentiation potential of osteoblasts and odontoblasts (Miura et al., 2003). SHEDs can induce the migration of naive bone marrow mesenchymal stromal cells (BMSCs) by secreting extracellular vesicles (EVs) containing various cytokines, thereby promoting the bone healing process (Luo, Avery & Waddington, 2021). SHEDs can also induce pulp and periodontal regeneration. Yang et al. (2019) combined SHEDs cell sheets and DFSCs cell sheets with dentin matrix (TDM) and implanted them into the orthotopic jawbone of nude mice. The results indicated that SHED/TDM successfully achieved periodontal tissue regeneration with better migration ability and neurogenic differentiation potential (Yang et al., 2019).

Properties and potential of SCAPs

SCAPs exist in the apex of the developing tooth before tooth eruption and differentiate to odontoblasts, which mainly secrete apical dentin. SCAPs are less resistant to immune cell-mediated toxicity compared with other dental-derived stem cells, but can induce high levels of pro-inflammatory cytokine secretion (Whiting et al., 2018). Since SCAPs are derived from developing odontogenic tissues, they are more widely used in the field of tissue regenerative. For example, in dental pulp engineering, the regeneration process of dental pulp can be achieved by inducing endogenous stem cells to move to the regeneration site (Rombouts et al., 2017). Wei, Sun & Hou (2021) successfully used the silk fibroin-RGD-stem cell factor scaffold (the RGD peptide was arginine-glycine-aspartic acid polypeptide) to promote the migration and proliferation of SCAPs. This approach is promising for the further use of cell homing in dental pulp regeneration.

Properties and potential of DFPCs

The dental follicle contains a large number of undifferentiated precursor cells. In 2005, DFPCs were first isolated from the dental follicles of human third molars (Morsczeck et al., 2005). DFPCs are derived from neural crest and are direct precursor cells of periodontal tissue (Zhang et al., 2019a). DFPCs can promote pulp regeneration through the paracrine pathway. Hong et al. (2020) found that DFPCs could effectively enhance the proliferation, migration, and odontogenic differentiation of inflammatory dental follicle cells (DPCs) in vitro and their ectopic dentinogenesis in vivo.

Properties and potential of PDLSCs

Periodontitis often leads to the destruction of periodontal tissue and may even lead to tooth loss. After root canal treatment, the periodontal tissue is the only source of nutrition for the root canals. Multiple preclinical studies have demonstrated the effectiveness of PDLSCs to restore damaged periodontal tissue in periodontal regenerative therapy (Li et al., 2020a). PDLSCs are mainly isolated from the periodontal tissue of permanent teeth and can be further differentiated into osteoblasts, chondrocytes, and adipocytes under appropriate conditions (Deng et al., 2018a). Studies have shown that PDLSCs have the strongest osteogenic ability, followed by DPSCs, and DFPCs (Qu et al., 2021).

Epigenetic regulations of dental-derived stem cells

Epigenetics refers to changing the expression level and function of a gene without changing its sequence. Its regulatory processes are regulated by signaling molecules whose interactions with neighboring cells induce appropriate transcriptional and epigenetic responses (Surani, Hayashi & Hajkova, 2007). The way in which epigenetic mechanisms regulate gene expression related to environmental factors plays an important role in the development of various diseases, such as tumors and inflammation (Yuan, Dong & Shen, 2022; Zarzour, Kim & Weintraub, 2019). In addition, epigenetic regulation may also affect the tooth number, size, and shape (Fernández et al., 2020). The common epigenetic regulations include DNA methylation, histone modification, and non-coding RNA regulations. Tables 1–3 summarize the regulatory mechanisms and potential applications of DNA methylation, histone modifications, and ncRNAs in dental-derived stem cells.

DNA methylation

DNA methylation is one of the important epigenetic modifications and it is common in most eukaryotes (Lin et al., 2018). The formation of 5-methylcytosine (5mC) from cytosine-phosphorothioate-guanine (CpG) dinucleotides by DNA methyltransferase (DNMT) leads to the silencing of gene expression (Radhakrishnan, Kabekkodu & Satyamoorthy, 2011). The level of methylated CpG is regulated by DNMT and DNA demethylase ten-eleven translocation (TET) (Ren, Gao & Song, 2018; Li et al., 2015).

DNA methylation levels are associated with stemness and the differentiation potential of dental-derived stem cells. For instance, DPSCs exhibits low DNA methylation levels and repressive mark H3K9Me2 enrichment, which is mediated by increased DNMT3B and G9a expression, respectively. This leads to decreased AKT phosphorylation and promotes osteogenesis (Shen et al., 2019). The decreased expression of the serine metabolism-related enzyme phosphoserine aminotransferase 1 (PSAT1) provides less methyl donor S-adenosylmethionine (SAM) for the methylation of the aging marker p16 (CDNK2A), resulting in the reduced stemness and osteogenic differentiation capacity of DPSCs (Yang et al., 2021). Wnt can effectively regulate the epigenetic mechanism of DPSCs. The short-term activation of Wnt signaling by Wnt-3A can cause a decrease in the content of 5-methylcytosine (5-mC) in DPSCs, which reduces the ability of DPSCs to differentiate into osteoblasts (Uribe-Etxebarria et al., 2020).

Different odontogenic stem cell genes have varied methylation levels and differentiation potentials. In DPSCs, PDLSCs, and DFPCs, the methylation of genes CD109 and SMAD3 are significantly different. At the transcriptional level, PDLSCs showed significantly higher expression levels of CD109, SMAD3, ALP, and RUNX2, which were identical to the differences in their DNA methylation profiles. The transcription levels of osteogenic differentiation-related factors and their osteogenic differentiation potential are higher in PDLSCs (Ai et al., 2018). The osteogenic differentiation process can be altered by modulating the methylation levels of specific genes in PDLSCs. Advanced glycation end-products (AGE) can increase the expression of DNMT1 and inhibit the methylation activation of calcitonin-related polypeptide α (CALCA) promoter, which inhibits the osteogenic differentiation of PDLSCs (Wang et al., 2022). Additionally, periostin (POSTN) can reduce the level of AGE receptors and DNA methylation of the CALCA promoter, thereby attenuating the inhibitory effect of AGE induction (Wang et al., 2022).

Histone modifications

Histones are located in the nucleus of eukaryotic cells and can form nucleosomes, the basic structure of chromatin, when bound to DNA. Modifications of amino acid residues in histone tails can cause structural changes in histones, which provide sites that can be recognized by specific proteins (Strahl & Allis, 2000). The regulation of specific genes can also be achieved through the binding of specific proteins to sites. Many studies have been conducted on the modification of histones, specifically on methylation and acetylation.

Noncoding RNA

Non-coding RNAs (ncRNA) are transcripts with no or low coding potential, including ribosomal RNA (rRNA), transfer RNA (tRNA), and microRNA (miRNA) (Ren & Wang, 2021). miRNA is the most studied in the field of epigenetics (Ren & Wang, 2021). miRNAs directly interact with partially complementary target sites located in the 3′ untranslated region of target mRNAs and repress their expression (Hombach & Kretz, 2016). Endogenous competing RNAs (ceRNAs) mainly regulate gene expression by competitively binding to miRNA (Qi et al., 2015). ceRNAs typically include long noncoding RNAs (lncRNAs) and circular RNAs (circRNAs).

circRNA

In PDLSCs, circRNAs can indirectly regulate osteogenic differentiation by binding to miRNAs (Gu et al., 2017). circRNA cerebellar degeneration-related protein 1 transcript (CDR1as) and miR-7 can regulate the osteogenic differentiation and stemness of PDLSCs. CDR1as may promote the upregulation of GDF5 and the phosphorylation of Smad1/5/8 and p38 MAPK by inhibiting the expression of miR-7, inducing the differentiation of PDLSCs to osteoblasts. In addition, the interaction of CDR1as with miR-7 can also upregulate the expression of KLF4 to maintain the stemness of PDLSCs, while RNA-binding protein hnRNPM regulates its expression in PDLSCs by interacting with CDR1as (Gu et al., 2021). During the osteogenic differentiation of SCAPs, the expression profiles of circRNAs are significantly altered, and circRNAs mainly function as ceRNAs (Li et al., 2019a). circ SIPA1L1 can promote the expression of the gene ALPL (alkaline phosphatase alkaline phosphatase) by binding to miR-204-5p, which causes the osteogenic differentiation of SCAPs (Li et al., 2020b).

Epigenetic regulatory network

In the epigenetic regulation of dental-derived stem cells, there are multiple links between histone modifications, DNA methylation, and ncRNA, which interact with each other and participate in genetic regulation together. ncRNAs participate in the regulation of gene expression in stem cells by regulating DNA methylation. For example, lncRNA H19 can inhibit the activity of DNMT3B, which reduces the methylation of the distal-less homeobox (DLX3) of the gene, thereby promoting the odontogenic differentiation of DPSCs (Zeng et al., 2018a). Similarly, miR-675 can also promote the odontogenic differentiation of DPSCs by inhibiting DNMT3B (Zeng et al., 2018b). The lncRNA HOTAIRM1 inhibits the expression and enrichment of DNMT1 on the HOXA2 promoter and mechanically binds to the CpG island in the HOXA2 promoter region, leading to hypomethylation and the induction of HOXA2 and DFSC differentiation into osteoblasts (Chen et al., 2020b).

ncRNAs can also play a role in histone modifications, including histone methylation, histone acetylation, and histone deacetylation. miR-153-3p inhibits the transcription of ALP, Runx2, and OPN by targeting KDM6A, which results in the attenuated osteogenic differentiation of PDLSCs (Jiang & Jia, 2021). miRNAs are involved in the aging and differentiation process of dental-derived stem cells by regulating the expression of HAT or HDAC. The upregulation of miR-152 represses HAT sirtuinc 7 (SIRT7) expression and affects the degree of histone acetylation, which accelerates the aging process of DPSCs (Gu et al., 2016). The upregulation of miRNA-383-5p can promote the down-regulation of the HDAC9 mRNA level, which leads to increased alkaline phosphatase activity, mineral node formation, and the expressions of RUNX2, osteocalcin, and Smad4 in PDLSCs and other osteogenic markers (Ma & Wu, 2021). Similarly, miR-22 can inhibit HDAC6 expression and promote the osteogenic differentiation of PDLSCs (Yan et al., 2017).

Therapeutic application of dental-derived stem cells in dental pulp and periodontal regeneration

In 1971, Nygaard-Ostby & Hjortdal (1971) proposed the concept of pulp tissue regeneration. Pulp regeneration refers to the formation of new pulp tissue through tissue engineering to replace the infected or necrotic pulp tissue, thereby restoring the structure and function of the pulp-dentin complex under physiological conditions. Conventional apexogenesis may result in the thinning of the dentin wall and underdevelopment of the root, which greatly increases the risk of long-term root fracture. However, pulp regeneration can effectively form healthy pulp tissue and promote the formation of dentin. Periodontitis, a chronic inflammation of the periodontal tissue caused by dental plaque, can cause the destruction and absorption of the alveolar bone and tooth loss. Traditional periodontal treatments, such as scaling, focuses on controlling the occurrence of inflammation, but fails to restore the structure and function of periodontal tissue entirely. Periodontal tissue regeneration reconstructs periodontal tissue damaged by periodontitis and restores its structure and function by means of tissue engineering (Chen et al., 2010). The key elements of tissue regeneration are stem cells, scaffolds, and signaling molecules. The biological behavior of stem cells is regulated by epigenetics. Further research will likely lead to the discovery of an increasing number of new factors. These factors may regulate the development of stem cells towards odontogenic differentiation, angiogenesis, neurogenesis, and osteogenic differentiation through epigenetic mechanisms, and may facilitate the application of pulp regeneration and periodontal regeneration. Figure 1 demonstrates the epigenetic regulation of dental-derived stem cell differentiation and its application in pulp regeneration and periodontal regeneration.

Nerve regeneration

CHD8 can affect neural progenitor cells and neurons, and also plays a role in maintaining the active transcription of neural-specific genes (Wilkinson et al., 2015). Meanwhile, CHD8 can also alter neurogenesis and cortical development by regulating the Wnt/β-catenin signaling pathway (Platt et al., 2017). In SHEDs, upregulated miR-221 can bind to CHD8 and activate the Wnt/β-catenin pathway (Wen et al., 2020). Wen et al. (2020) divided the SHEDs in the third-generation logarithmic growth phase into six groups: blank group, NC group (transfected with miR-221 negative sequence), miR-221 mimic group (transfected with miR-221 mimic), miR-221 inhibitor group (transfected with miR-221 inhibitor), siRNA-CHD8 group (transfected with siRNA into CHD8 vector), and miR-221 inhibitor + siRNA-CHD8 group (co-transfected with miR-221 inhibitor and siRNA-CHD8). The results of Western blot analysis revealed that the expressions of NSE, NESTIN, MAP-2, NF-M, and TH in the miR-221 inhibitor group were significantly lower than those in the NC group, while the miR-221 mimic group and siRNA-CHD8 group were both lower than those in the NC group. Immunofluorescence examination showed that the expressions of NSE and MAP-2 in the miR-221 inhibitor + siRNA-CHD8 group were higher than that in the miR-221 inhibitor group (Wen et al., 2020). Among them, neuron-specific enolase (NSE) was a highly specific marker of neurons and peripheral neuroendocrine cells, NESTIN was a key early neural progenitor cell marker, NF-M and microtubule-associated protein 2 (MAP2) was a neuron-associated marker, and TH was the rate-limiting enzyme in dopamine neurotransmitter biosynthesis. These results suggest that miR-221 can promote SHED differentiation into neurons by inhibiting CHD8.

BMP2 is a neurotrophic factor that induces the growth of brain dopaminergic (DA) neurons in vitro and in vivo, whose induction depends on the Smad signaling pathway (Hegarty, Sullivan & O’Keeffe, 2013). In SHEDs, lncRNA C21 or f121 competitively binds to miR-140-5p and upregulates BMP2 expression (Liu et al., 2018a). The results of bioinformatics analysis conducted by Liu et al. (2018a) showed that there was a targeting relationship between the second spliceosome of lncRNA C21 or f121 and miR-140-5p, the same as miR-140-5p and BMP2. This suggested that lncRNA C21 or f121 competed with BMP2 to bind to miR-140-5p. Liu et al. (2018a) grouped and experimented with SHEDs in the third-generation logarithmic growth phase. The results showed that the protein expressions of both Nestin and βIII-tubulin decreased, but increased in the transfected miR-140-5p inhibitor group compared with the NC group (transfected with lncRNA C21 or f121 negative sequence), the BMP2 and MAP2 in the si-C21 or f121 group, the miR-140-5p group, and the si-C21 or f121+miR-140-5p group. Further experiments showed that the up-regulation of lncRNA C21 or f121 or down-regulation of miR-140-5P increased the frequency of social behavior in rats and decreased the cumulative time of repetitive stereotyped movements in young rats (Liu et al., 2018a). All of the abovementioned studies show that lncRNA C21 or f121 can effectively promote the neurogenic differentiation of SHEDs.

Finally, some HDACi have the effect of inducing neurogenic differentiation of cells. For instance, Okubo et al. (2016) found that the total number of mRNAs of mature neuronal markers, neurofilament medium polypeptide (NeFM), and microtubule-associated protein 2 (MAP2) significantly increased to approximately 80% in VPA-treated rats compared with untreated rats (Okubo et al., 2016). Other studies have shown that the neurite number on the cells increased and branched processes were elongated after treating MSCs with combinations of MS-275 or NaB (a kind of HDACi). The cells were visualized by immunofluorescence staining of the neuronal markers (Jang et al., 2019). The above studies demonstrate the role of HDACi in inducing neurogenic differentiation. Further research may reveal whether dental-derived stem cells can be inducted to differentiate into neural cells.

Angiogenesis

TGF-β/SMAD2 signaling can promote angiogenesis and the secretion of vascular endothelial growth factor (Ji et al., 2014). Wu et al. (2021) discovered that the expression of miR-26a was up-regulated in SHED-secreted exosomes (SA-Exo), and miR-26a could promote the expression of TGF-β/SMAD2/3 signaling. The expression of angiogenesis-related proteins (VEGF, angiopoietin 2 and PDGF) of SHEDs was up-regulated, and the endothelial differentiation potential was increased after being treated with SA-Exo. SA-Exo treatment also increased the expression levels of angiogenesis-related proteins in HUVECs. Wu et al. (2021) implanted SHED aggregates into immunodeficient mice and performed histological analysis, which revealed the formation of a new, continuous dentin layer and blood vessels, and the regeneration of the dentin-pulp complex. In addition, dentin and blood vessel formation were enhanced by the combined implantation of SHED aggregates and SA-Exo. The expression level of the angiogenic marker CD31 was also higher. The inhibition of SA-Exo repressed dentin-pulp complex regeneration; however, supplementation with exogenous SA-Exo could restores this process. The results of qRT-PCR confirmed that the expression of miR-26a was significantly increased in SA-Exo, and the inhibition of miR-26a in SA-Exo could not cause the endothelial differentiation of SHED and HUVECs. Western blot analysis revealed that the overexpression of miR-26a upregulates TGF-β/SMAD2/3 signaling, and the inhibition of these two pathways led to reduced endothelial differentiation in SHEDs and HUVECs (Wu et al., 2021). These studies confirm that miR-26a in SA-Exo promote angiogenesis in SHEDs through the TGF-β/SMAD2/3 signaling pathway.

ncRNAs have a strong ability to regulate endothelial cell migration, proliferation, and differentiation. miR-30a-3p targets the epigenetic factor methyl-CpG-binding protein 2 (MeCP2), and the overexpression of MeCP2 damages important genes involved in the regulation of endothelial function such as sirtuin1 (Volkmann et al., 2013). Volkmann et al. (2013) transfected endothelial cells with miR-30a-3p precursors, which significantly reduced MeCP2 protein levels and increased the migratory ability of endothelial cells. This suggests that miR-30a-3p has the ability to regulate endothelial cells. In addition, lncRNAs also have a role in regulating endothelial cells. Neumann et al. (2018) discovered that lncRNA GATA6 could inhibit the action of the epigenetic regulator, LOXL2, reduce the endothelial-mesenchymal transition in vitro, and promote the formation of blood vessels in mice. These two studies demonstrated the potential of ncRNAs in promoting angiogenesis. Future studies may reveal whether they can regulate dental-derived stem cells for angiogenesis.

Osteogenic differentiation

Different dental-derived stem cells have a diverse range of DNA methylation levels and unique osteogenic differentiation potentials. Compared with DPSCs and DFPCs, PDLSCs have lower methylation levels of genes related to osteogenesis, higher expression levels of factors such as SMAD3, ALP, OCN, and RUNX2, and a higher osteogenic differentiation potential (Ai et al., 2018). After culturing PDLSCs, DFPCs, and DPSCs14 in osteoinductive medium, Ai et al. (2018) used Alizarin Red S positive staining and found that the relative intensity of staining in PDLSCs was significantly higher than that in DPFCs and DPSCs. The simultaneous subcutaneous transplantation of cell deposits mixed with hydroxyapatite onto the dorsal surface of immunocompromised male mice found that PDLSCs formed more osteoid. This study demonstrated that DNA methylation can regulate the osteogenic differentiation potential of dental-derived stem cells by affecting the expression of related genes.

Insulin-like growth factor (IGF) and its binding proteins play an important role in promoting bone formation (Nguyen et al., 2013). KDM6B can promote IGFBP5 transcription by reducing histone K27 methylation (Han et al., 2017). Han et al. (2017) found that by administering a local injection of rhIGFBP5 into the periodontitis area of a piglet model, they could significantly promote the regeneration of periodontal tissues such as alveolar bone and gingiva after 12 weeks.

The Wnt/β-catenin pathway can promote/inhibit the osteogenic differentiation of cells under various conditions (Wagner et al., 2011). HAT GCN5 inhibits the Wnt/β-catenin signaling pathway by increasing the levels of H3K9ac and H3K14ac in the DKK1 promoter region (Li et al., 2016). Li et al. (2016) found that more active osteogenic differentiation was presented in cell populations with higher GCN5 expression, and GCN5 downregulation may lead to defective osteogenic differentiation of PDLSCs. GCN5 knockdown resulted in increased expression of β-catenin and decreased expression of genes and proteins related to osteogenic differentiation, such as RUNX2 and ALP. ChIP assays indicated that GCN5 binds to the promoter region of DKK1. Alveolar bone loss in the first and second maxillary molars was significantly reduced with increased GCN5 expression in periodontital rats (Li et al., 2016). Therefore, HAT GCN5 can promote the osteogenic differentiation of PDLSCs to regenerate alveolar bone by inhibiting the Wnt/β-catenin signaling pathway.

Both the MAPK and TGF-β/Smad signaling pathways may be involved in BMP-mediated osteogenesis (Kim, Park & Choung, 2018; Zhu et al., 2018). CDR1as is an inhibitor of miR-7 that can cause the upregulation of TGF-β family member GDF5, and the phosphorylation of p38 MAPK (Li et al., 2018c). In PDLSCs, the knockdown of CDR1as or the overexpression of miR-7 significantly suppressed the mRNA and protein levels of GDF5. In contrast, a lower expression of GDF5 resulted in a decrease in the osteogenic markers ALP and RUNX2, as well as phosphorylated p38 MAPK. Li et al. (2018c) loaded CDR1as siRNA and negative control siRNA-treated PDLSCs onto scaffold material and implanted this into the calvarial defect area of nude mice. The results showed that the CDR1as knockdown group had less bone formation and significantly lower new bone formation than the control group. In the control group, bone tissue was generated at the edges of the bone defects, but in the CDR1as knockout group, little new bone was observed (Li et al., 2018c). This study strongly demonstrates that CDR1as can promote the osteogenic differentiation of PDLSCs.