Dental pulp stem cells (DPSCs) was first isolated and named by Gronthos[11]. It can be defined as “undifferentiated mesenchymal cells in the cell-rich zone at the central of the dental pulp”, they are the precursors of odontoblasts[12]. The odontoblastic differentiation of hDPSCs is a crucial factor in reparative dentin generation and dental tissue self-repair in inflammatory microenvironments[13]. It is well documented that hDPSCs produce a variety of extracellular matrix proteins, including dentin sialophosphoprotein (DSPP) and dentin matrix protein (DMP-1), forming dentin layers to block the stimulating factors and protect the pulp tissue from further damage[14]. Accroding to results of flow cytometry assay, the dental pulp-derived cells used in this study express MSC marker genes and negative for hematopoietic cell markers. Meanwhile, the cells reveal form colonies capacity and osteogenic differentiation potential in our study, which perform characteristics of stem cells.
LncRNAs are involved in a variety of physiological and pathological activities of cells, including proliferation, differentiation, migration and apoptosis[15–17]. Recent studies have indicated that lncRNAs also participate in biological mineralization[18–21]. As one type of biological mineralization, odontoblastic differentiation shares many similarities with other types of biological mineralization, including the synthesis of matrix proteins. These similarities have led to the speculation that lncRNAs are involved in the differentiation of hDPSCs into odontoblasts. However, few studies have focused on the role of lncRNAs in this process, and only DANCR was found to suppress the odontoblastic differentiation of hDPSCs by acting as an inhibitor of the Wnt pathway[22]. To verify the regulatory effect of lncRNAs on the odontoblastic differentiation of hDPSCs, a lncRNA microarray analysis using mineralized solution-induced and un-induced cells was performed. According to the microarray expression profiles, the expression levels of 1106 lncRNAs were significantly changed by more than 2.0-fold; 617 of these were up-regulated, and 489 were down-regulated. Among those lncRNAs, MALAT1, MIR31HG, H19 and WNT2 were selected for further identification, based on reports that concentrated on lncRNAs regulating biological mineralization[23, 24]. Due to the qRT-PCR results, H19 was chosen as a candidate lncRNA because the expression level of H19 was most consistent with the lncRNA microarray results, and its expression increased remarkably at day 7 after mineralization induction.
H19 plays a critical role in the differentiation of many vital organs[25]. H19 may regulate the occurrence and development of tumors, such as breast carcinoma, bladder carcinoma, glioma and colorectal carcinoma[26–28]. Significantly, H19 can promote the differentiation of numerous types of cells, such as human mesenchymal stem cells (hMSCs)[29, 30], skeletal muscle satellite cells[31] and myoblasts[32]. Accumulating evidence indicates that H19 plays pivotal roles in the differentiation of stem-like cells. A. Keniry et al.[33] found that H19 can enhance the expression level of RUNX2 to promote bone regeneration. Zhou et al.[34] revealed that H19 can help in bone tissue reparation by inhibiting p53. As a biomineralization promoter, H19 can promote the osteogenic differentiation of mesenchymal stem cells (MSCs) through the Notch signaling pathway[30]. It can also suppress miR-141 and miR-22 as a miRNA sponge to derepress the WNT/β-catenin signaling pathway inhibited by these two miRNAs[35]. Recently, Li et al.[36] revealed that H19 enhanced osteo/odontogenesis of SCAPs via the miR-141/SPAG9 pathway.
Although the gene expression between osteogenesis and odontogenesis are somehow similar, it's worth noting the difference between the BMSCs and DPSCs. With the mineralization-inducing environment, BMSCs could formed lamellar bone containing osteocytes and osteoblasts, with regeneration of surrounding fibrous vascular tissue[11]. By contrast, DPSCs could generate dentin/pulp-like structure after odontoblastic induction culture but the bone matrix protein and bone sialoprotein were absent[37]. Furthermore, DSPP and DMP-1 are odontoblast-specific markers[38]. In this study, the specific effect and mechanism of H19 in mediating the odontogenic differentiation of hDPSCs were investigated. We initially applied transfection technology to regulate the expression of H19 and analyzed the biological effects of H19 on hDPSCs. After transfection with lentiviruses or plasmids, the expression level of H19 was significantly altered according to the qRT-PCR results. Consistent with our assumption, the odontoblastic differentiation of hDPSCs was inhibited after interfering with H19 expression. In contrast, overexpressing H19 could promote the odontoblastic differentiation of hDPSCs. In addition, an ectopic odontogenesis nude mice model was used to further confirm the role of H19 in hDPSCs in vivo. Both micro-CT analysis and histological examination revealed significantly increased levels of dentin-like structures formation and greater abundance of odontogenic specific markers DMP-1 in H19-overexpression hDPSCs loaded group, which were consistent with the in-vitro experiments. Similarly, Li et al.[39] reported that H19 enhanced the osteo/dentinogenesis of SCAPs in vivo, and our research exhibited the same trend. These results suggested that H19 exerted important effects on the differentiation of hDPSCs into odontoblasts.
LncRNAs have been reported to regulate gene expression via the following mechanisms: exerting post-transcriptional effects, interacting with proteins and operating as molecular scaffolds[40–42]. Numerous studies have demonstrated that lncRNAs can act as miRNA sponges through direct binding to miRNAs[43–45]. Several studies have indicated that H19 plays various regulatory functions by acting as a miRNA sponge[46, 47]. Here, we hypothesized that H19 could promote the odontoblastic differentiation of hDPSCs via sponging miRNAs. Six mineralization-related miRNAs (miR-17-5p, miR-93-5p, miR-103a-3p, miR-106b-5p, miR-140-5p and miR-148-5p) that may have potential binding sites with H19 were predicted by bioinformatics databases. They were the intersection of the predicted results of three database, including Starbase, DIANA-LncBase and RegRNA.
The expression levels of these candidate miRNAs were evaluated after 14 days of induction. Compared with the non-induction group, the expression levels of miR-140-5p were remarkably decreased in the induction group. To affirm if there is direct binding between H19 and miR-140-5p, a dual-luciferase assay was performed. MiR-140-5p significantly reduced the luciferase activity of H19-wild compared to that of H19-mut, indicating that H19 might function as a sponge of miR-140-5p. Hence, miR-140-5p was selected as the research target and the influence of miR-140-5p on odontogenic differentiation of hDPSCs was further explored.
Previous studies have indicated that miR-140-5p play essential roles in regulating the osteogenic differentiation of stem cells. Zheng L et. al.[48] revealed that lncRNA MEG3 promoted the osteogenesis of hADSCs via sponging miR-140-5p, indicating the inhibitory role of miR-140-5p in osteogenic differentiation. Hwang[49] found that miR-140-5p suppressed BMP2-mediated osteogenesis in undifferentiated human mesenchymal stem cells. To date, related studies of miR-140-5p in odontogenic differentiation are rare. Interestingly, so far, no studies have proven the interaction and regulation between H19 and miR-140-5p. Hence, it aroused our interest in demonstrating the relationship of them. In our study, we confirmed the inhibitory effect of miR-140-5p on odontoblastic differentiation by transfecting miRNA mimics into hDPSCs. The expression levels of odontoblast-related genes and matrix mineralization levels were obviously decreased by overexpression of miR-140-5p, compared with miR-NC mimics.
Furthermore, rescue experiments indicated that H19 could partially abrogate the inhibitory effects on odontoblastic differentiation induced by miR-140-5p in vitro, which implied that H19 is involved in the ceRNA regulatory network and acts as a miR-140-5p sponge. These results showed that H19 promoted the odontoblastic differentiation of hDPSCs by interacting with miR-140-5p. However, the potential targets of miR-140-5p, which are involved in the H19-mediated odontoblastic differentiation of hDPSCs, remain to be further elucidated.
To reveal the downstream molecular mechanism of H19/miR-140-5p pathway regulating the odontoblastic differentiation of hDPSCs, miRDB and Targetscan were applied to search for potential targets. BMP-2 and FGF9 were predicted as the candidate target genes since the 3’-UTR region of BMP-2 and FGF9 contained potential binding sites with miR-140-5p respectively. Dual-luciferase reporter assays were conducted to further confirmed the results. Overexpression of miR-140-5p significantly reduced the luciferase activity in BMP-2-wild and FGF9-wild groups, compared with HEK293T cells co-transfected with miR-140-5p mimics and the BMP-2-mut/FGF9-mut 3’UTR vector groups separately. Moreover, overexpression of H19 increased the expression level of BMP-2 and FGF9, whereas overexpression of miR-140-5p decreased BMP-2/FGF9 mRNA and protein expression. The expression level of BMP-2/FGF9 were partly increased in hDPSCs co-transfected with H19 and miR-140-5p mimics compared to hDPSCs transfected with miR-140-5p mimics along. These results showed that H19 regulated BMP-2/FGF9 expression by inhibiting the effect of miR-140-5p.
In our study, the biological role of BMP-2 and FGF9 in regulating the odontogenic differentiation of hDPSCs have been verified. hDPSCs cultured in osteogenic induction medium enriched with BMP-2 or FGF9 separately, were demonstrated to encourages osteogenic differentiation compared to the normal osteogenic induction group. Furthermore, the results showed that the odontogenic capacity of BMP-2 is better than that of FGF9. It is well recognized that BMP-2 is a critical growth factor and important biomarker involved in osteo/odontoblastic differentiation and bone formation[50–52]. Our results revealed the similar trend that BMP-2 offered a strong signal for differentiation and mineralization of hDPSCs. On the other hand, as a member of the fibroblast growth factor family, certain studies have focused on the biological role of FGF9 in bone formation. Behr et al.[53] indicated that endogenous FGF9 protein promoted angiogenesis and played an important role in long bone repair of loss of function (Fgf-9−/−) mice. Wang et al.[54] demonstrated that FGF9 secreted by mature osteoblasts could regulate bone formation in adult male mice. Wallner et al.[55] reported that FGF9 facilitated osteogenesis by stimulating the expression of several osteogenic related genes. The above researches have demonstrated that FGF9 is essential for osteogenesis, supporting our results.
In summary, the results indicated that H19/miR-140-5p pathway promoted the odontoblastic differentiation of hDPSCs partially through the regulation of BMP-2 and FGF9.