ABSTRACT Genetic studies have revealed a critical role of Distal-homeobox (Dlx) genes in bone formation, and our previous study showed that _Dlx2_ overexpressing in neural crest cells leads

to profound abnormalities of the craniofacial tissues. The aim of this study was to investigate the role and the underlying molecular mechanisms of _Dlx2_ in osteogenic differentiation of

mouse bone marrow stromal cells (BMSCs) and pre-osteoblast MC3T3-E1 cells. Initially, we observed upregulation of Dlx2 during the early osteogenesis in BMSCs and MC3T3-E1 cells. Moreover,

_Dlx2_ overexpression enhanced alkaline phosphatase (ALP) activity and extracellular matrix mineralization in BMSCs and MC3T3-E1 cell line. In addition, micro-CT of implanted tissues in nude

maturation. Importantly, luciferase analysis showed that _Dlx2_ overexpression stimulated both _OCN_ and _Alp_ promoter activity. Through chromatin-immunoprecipitation assay and

site-directed mutagenesis analysis, we provide molecular evidence that Dlx2 transactivates _OCN_ and _Alp_ expression by directly binding to the Dlx2-response _cis_-acting elements in the

promoter of the two genes. Based on these findings, we demonstrate that _Dlx2_ overexpression enhances osteogenic differentiation in vitro and accelerates bone formation in vivo via direct

upregulation of the _OCN_ and _Alp_ gene, suggesting that Dlx2 plays a crucial role in osteogenic differentiation and bone formation. SIMILAR CONTENT BEING VIEWED BY OTHERS THE GUANINE

NUCLEOTIDE EXCHANGE FACTOR DOCK5 NEGATIVELY REGULATES OSTEOBLAST DIFFERENTIATION AND BMP2-INDUCED BONE REGENERATION VIA THE MKK3/6 AND P38 SIGNALING PATHWAYS Article Open access 01 January

2025 CIRCULAR RNA AFF4 MODULATES OSTEOGENIC DIFFERENTIATION IN BM-MSCS BY ACTIVATING SMAD1/5 PATHWAY THROUGH MIR-135A-5P/FNDC5/IRISIN AXIS Article Open access 18 June 2021 BMP-INDUCED ATOH8

ATTENUATES OSTEOCLASTOGENESIS BY SUPPRESSING RUNX2 TRANSCRIPTIONAL ACTIVITY AND REDUCING THE RANKL/OPG EXPRESSION RATIO IN OSTEOBLASTS Article Open access 02 September 2020 INTRODUCTION The

distal-less homeobox (Dlx) gene family consists of six members (_Dlx1_, _Dlx2_, _Dlx3_, _Dlx5_, _Dlx6_, and _Dlx7_); these members are expressed in a complex pattern in the first and second

branchial arch region.1 Dlx1/2 regulates the development of the upper jaw, while Dlx5/6 regulates lower jaw development.2 Together with other homeobox proteins, the Dlx family regulates

osteoblast differentiation. As one of the key transcription factors regulating osteogenic differentiation, Dlx5 stimulates two other key transcription factors, Runx2 and Osterix (Osx), which

sequentially induce expression of bone markers such as _Osteocalcin_ (_OCN_) and _Alkaline_ _Phosphatase_ (_ALP_). The expression of _Dlx5_ is induced by bone morphogenetic protein-2

(BMP-2).3 Msx2, another homeobox gene and a key regulator of osteogenic differentiation, represses the expression of _Alp_ by directly binding to its promoter, while Dlx5 activates its

expression by interfering with the ability of Msx2.4 Thus, Dlx5 coordinates with Msx2 to regulate osteogenic differentiation due to their reciprocal ability to compete with each other.

Sharing strong sequence similarity with Dlx5, Dlx2 has been shown to play a crucial role in craniofacial skeletal development.5 _Dlx2_ is upregulated in the central area of the first

branchial arch during days 9.5 and 10.5 of embryonic development in mice. This upregulation of _Dlx2_ is important for the differentiation and development of the primordium, as it leads to

the development of the maxillofacial skeletal pattern.6 Given that Dlx5 controls osteogenic differentiation,7 it is reasonable to speculate that Dlx2 might be involved in this process. So

far, only a few studies have reported that _Dlx2_ overexpression increases the osteogenic differentiation potential of pre-osteoblast cells.8 However, how Dlx2 regulates osteogenic

differentiation and the underlying cellular and molecular mechanisms remain unknown. In a previous study, we found that elevated _Dlx2_ expression led to midfacial development defects, nasal

deformities, premaxillary bony deficiency, and spine deformities.9 Thus, it is crucial to examine how _Dlx2_ overexpression leads to abnormal bone formation both in vitro and in vivo. To

investigate the role of Dlx2 during osteogenic differentiation both in vitro and in vivo, we used mouse bone marrow stromal cells (BMSCs) in our study, as the ability of BMSCs to

differentiate toward adipogenic, chondrogenic, and osteogenic cell lineages has been characterized extensively in vivo and in vitro by various researchers.10 Osteogenic differentiation of

BMSCs can be assayed in vitro by ALP and Alizarin red staining and in vivo by transplantation assays.11,12 Therefore, mouse BMSCs are suitable for investigating the effect of _Dlx2_

overexpression on osteogenesis both in vitro and in vivo. Murine osteoblastic cell line MC3T3-E1 cells were also chosen to verify the effect of _Dlx2_ overexpression on osteogenesis in

vitro. Initially, we observed the upregulation of _Dlx2_ in both mouse BMSCs and MC3T3-E1 cells during osteogenic differentiation. Moreover, forced overexpression of _Dlx2_ led to enhanced

osteogenic differentiation potential of both BMSCs and MC3T3-E1 cells in vitro, and accelerated bone formation in vivo. These findings prompted us to explore the underlying mechanisms. To

our surprise, we found that _Dlx2_ overexpression had no significant effect on the expression levels of _Dlx5_, _Msx2_, _Runx2_, and _Osx_, but led to upregulation of _Alp_ and _OCN_ in

BMSCs and MC3T3-E1 cells. Considering the fact that Alp promotes the early stage of osteogenic differentiation and OCN accelerates the late stage, we next analyzed the promoter of _OCN_ and

_Alp_ through luciferase-reporter assay and chromatin-immunoprecipitation (ChIP) analysis, and found that _Dlx2_ transcriptionally regulated _OCN_ and _Alp_ expression by directly binding to

their promoters. Taken together, our data demonstrates for the first time that _Dlx2_ overexpression enhances the early stage of osteogenic differentiation via direct upregulation of _Alp_,

and promotes the late stage of osteogenic differentiation via direct upregulation of OCN. RESULTS ENDOGENOUS DLX2 EXPRESSION IN BMSCS AND MC3T3-E1 CELLS DURING OSTEOGENESIS First, we

examined the levels of _Dlx2_ expression upon osteogenic induction in mouse BMSCs and MC3T3-E1 cells. Quantitative reverse transcription polymerase chain reaction (RT-qPCR) results showed

that when BMSCs were exposed to osteogenic-inducing medium (OIM), _Dlx2_ expression was upregulated within 0.5 and 3 h after induction (Fig. 1a). However, after 7- or 14-day culture in OIM,

these cells express similar mRNA level of _Dlx2_ with the cells cultured in normal culture medium (data not shown). In addition, western blot analysis with an anti-Dlx2 antibody detected

only a very weak signal of Dlx2 protein in BMSCs cultured both in normal medium and OIM for 3 h. This could be explained by the low protein level of endogenous Dlx2 in BMSCs. Supporting this

notion is the finding that the endogenous protein level of other Dlx proteins, such as Dlx5, is also quite low in BMSCs.13 The expression pattern of _Dlx2_ in MC3T3-E1 cells was similar

with that in BMSCs. The mRNA level of _Dlx2_ in MC3T3-E1 cells was upregulated within and the first 6 h after osteogenic induction (Fig. 1b), but showed no differences with that in the cells

cultured in normal culture medium after culture for 7 or 14 days (data not shown). Consistently, previous study in stem cells from apical papilla (SCAP) also showed a similar expression

pattern that Dlx2 was upregulated within the first few hours upon osteogenic induction, and its expression then showed no significant difference with that in cells grown in normal cell

culture medium. This reduction of _Dlx2_ expression in the induced group could be explained by the regulation of mRNA stability by microRNAs (miRNAs). Latest study reveals that Dlx2 is a

target gene of the miRNA, miR-185-5p, and its expression is negatively regulated by Dlx2.14 The initial upregulation of Dlx2 may stimulate the expression of downstream target genes, which

induces miRNA expression and sequentially exert a feedback on Dlx2 expression. In addition, western blot analysis detects only weak expression of Dlx2 in MC3T3-E1 cells treated either with

OIM or normal culture medium, consistent with the endogenous protein level of Dlx2 in BMSCs. Together, these results demonstrate that _Dlx2_ was upregulated during early osteogenesis in

mouse BMSCs and MC3T3-E1 cells. FORCED OVEREXPRESSION OF _DLX2_ IN BMSCS AND MC3T3-E1 CELLS Next, to investigate the effect of _Dlx2_ on osteogenesis, we established _Dlx2_-overexpressing

BMSCs and MC3T3-E1 cells. Cultured BMSCs or MC3T3-E1 cells were transduced with Lenti-DLX2 OE lentivirus to allow stable _Dlx2_ overexpression. As a control, parallel BMSCs or MC3T3-E1 cells

were transduced with Lenti-CTRL lentivirus as mock control. Dlx2 expression was then evaluated by RT-qPCR and western blot analysis. Lenti-DLX2 OE-transduced BMSCs (over) displayed much

more Dlx2 transcripts (Fig. 1c) and correspondingly increased protein production (Fig. 1d) than wildtype or Lenti-CTRL-transduced BMSCs (control) did. Similarly, Lenti-DLX2 OE transduction

in MC3T3-E1 cells also led to increased mRNA and protein levels of Dlx2 (Fig. 1c, d). These observations indicate that _Dlx2_ was successfully overexpressed in both cell lines. DLX2

OVEREXPRESSION ENHANCES THE OSTEOGENIC DIFFERENTIATION POTENTIAL OF BMSCS AND MC3T3-E1 CELLS To investigate whether the overexpression of _Dlx2_ effects on osteogenic differentiation in

vitro, we carried out ALP staining and Alizarin staining assays. Lentivirus-transduced BMSCs or MC3T3-E1 cells were cultured in OIM for 7, 14, or 21 days to assess their osteogenic

differentiation potential.14 Interestingly, ALP staining was significantly enhanced at day 7 and 14 in the _Dlx2_-overexpressing BMSCs compared with that in control BMSCs (Fig. 2a, c).

Similarly, higher ALP activity levels were observed in _Dlx2_-overexpressing BMSCs at day 14 after osteogenic induction (Fig. 2b). Furthermore, Alizarin red staining revealed that

mineralization was markedly enhanced in _Dlx2_-overexpressing BMSCs during the entire culture period, especially at day 21 (Fig. 2a, c). Consistently, we also observed enhanced ALP and

Alizarin red staining in _Dlx2_-overexpressing MC3T3-E1 cells (Fig. S1). Given that ALP activity is involved in the early stage of osteogenic differentiation while mineralization is involved

in the late stage, we therefore proposed that Dlx2 overexpression accelerates the early stage of osteogenesis by increasing ALP activity and the late stage of osteogenesis by enhancing

mineralization. FORCED OVEREXPRESSION OF DLX2 IN BMSCS ACCELERATED BONE FORMATION IN VIVO To investigate whether _Dlx2_ overexpression could affect osteogenesis in vivo, we subcutaneously

implanted the BMSCs/β-tricalcium phosphate (β-TCP) constructs into nude mice. The whole implanted tissues were then analyzed with microscopic computed tomography (micro-CT) examination 6- or

8-weeks after implantation (Fig. 3a). Micro-CT revealed new bone formation in both control and _Dlx2_ overexpression groups (Fig. 3b, c). However, quantitative morphometric analysis showed

that bone volume/tissue volume (BV/TV) in the _Dlx2_-overexpressing group (17.81% ± 1.23% and 23.64% ± 1.71% at week 6 and week 8, respectively) was significantly higher than that in the

control group (12.91% ± 1.16% and 17.04% ± 1.62% at week 6 and week 8, respectively) (_P_ < 0.05) (Fig. 3d). Similarly, the _Dlx2_-overexpressing group showed higher bone mineral density

(BMD) of the newly formed bone ((1.477 ± 0.097) g· cm−3 and (1.550 ± 0.121) g· cm−3 at week 6 and week 8, respectively) than the control group ((1.055 ± 0.072) g· cm−3 and (1.107 ± 0.098) g·

 cm−3 at week 6 and week 8, respectively) (Fig. 3e). Besides, trabecular number (Tb.N) in the _Dlx2_-overexpressing group ((1.362 ± 0.110) g· cm−3 and (1.713 ± 0.129) g· cm−3 at week 6 and

8, respectively) was significantly higher than that in the control group ((0.979 ± 0.086) g· cm−3 and (1.232 ± 0.097) g· cm−3 at week 6 and 8, respectively) (Fig. 3f). Additionally, the

_Dlx2_-overexpressing group exhibited decreased trabecular space (Tb.Sp) ((0.634 ± 0.059) mm and (0.482 ± 0.033) mm at week 6 and 8, respectively) when compared to the control group ((0.885 

± 0.079) mm and (0.672 ± 0.044) mm at week 6 and 8, respectively) (Fig. 3g). Taken together, these results demonstrate that overexpression of _Dlx2_ in BMSCs substantially triggers

osteogenic differentiation and improves bone formation in vivo, defining a pivotal role of Dlx2 in osteogenic differentiation. EFFECT OF FORCED OVEREXPRESSION OF DLX2 ON THE EXPRESSION OF

OSTEOGENIC GENES IN BMSCS To gain an insight into the mechanism of _Dlx2_-induced osteogenesis, we examined the expression of a panel of osteogenesis-associated genes by RT-qPCR in

_Dlx2_-overexpressing BMSCs and control BMSCs cultured in OIM. As shown in Fig. 4a, RT-qPCR analysis revealed that transduction of Lenti-DLX2 OE into BMSCs resulted in a 526.2-fold and

481.3-fold increase in _Dlx2_ expression at days 14 and 21 of osteogenic differentiation, respectively. Unexpectedly, there was no difference in _Runx2_, _Dlx5_, _Msx2_, and _Osx_ expression

between _Dlx2_-overexpressed BMSCs and mock-transduced BMSCs at 14 and 21 days after osteogenic induction (Fig. 4b–e) nor at 1, 2, and 7 days after osteogenic induction (data not shown),

indicating that these genes might not be involved in the _Dlx2_-induced promotion of osteogenesis. Importantly, forced overexpression of _Dlx2_ led to upregulation of _OCN_ at days 14 and 21

after osteogenic induction and upregulation of _Alp_ at day 14 (Fig. 4f, g), consistent with the ALP staining and Alizarin red staining results (Fig. 2a–c). In summary, we found, for the

first time, that forced overexpression of _Dlx2_ in BMSCs leads to upregulation of _Alp_ and _OCN_, instead of _Runx2_, _Dlx5_, _Msx2_, and _Osx_. DLX2 OVEREXPRESSION LEADS TO INCREASED

EXPRESSION OF _OCN_ AND _ALP_ IN MC3T3-E1 CELLS We next tested whether _Dlx2_ overexpression could also lead to upregulation of _OCN_ and _Alp_ in other osteoblast precursor cell lines.

Compared with mock-transduced cells, we observed enhanced mRNA and protein levels of _OCN_ in _Dlx2_-overexpressing MC3T3-E1 cells after osteogenic induction (Fig. 4h, i).

_Dlx2_-overexpressing MC3T3-E1 cells also exhibited higher _Alp_ transcription (Fig. 4h) and enhanced ALP activity (Fig. S1A). Moreover, as in BMSCs, we observed no significant difference in

_Msx2_, _Dlx5_, _Runx2_, and _Osx_ expression after Lenti-DLX2 OE transduction in MC3T3-E1 cells after osteogenic induction, as shown in the Figure S2. Taken together, these data

demonstrate that forced overexpression of _Dlx2_ induces upregulation of _OCN_ and _Alp_ upon osteogenic induction both in BMSCs and MC3T3-E1 cell lines. Given that the expression levels of

_Runx2_, _Dlx5_, _Msx2_, and _Osx_ remained unchanged upon _Dlx2_ overexpression, we speculate that _OCN_ and _Alp_ are the direct target genes of Dlx2. CHARACTERIZATION OF THE MOUSE OCN

PROMOTER AND IDENTIFICATION OF ITS DLX2-RESPONSE ELEMENT Previous genetic studies have shown that through direct binding to the promoter of _OCN_, homeodomain (HD) proteins Msx2, Dlx3, and

Dlx5 regulate the expression of _OCN_ in osteogenic cells. Dlx3 binds the _OCN_ promoter to stimulate its expression, while the binding of Msx2 and the recruitment of Dlx5 represses _OCN_

expression. Moreover, Dlx5 and Msx2 regulate _Alp_ expression by directly binding to its promoter.15,16 Considering that Dlx2, an HD protein, shares strong sequence similarity with Dlx5 and

Dlx3, we speculated that Dlx2 might also be involved in the regulation of _OCN_ and _Alp_ transcription in osteogenic cells by binding to their promoters. We first determined whether _OCN_

is directly regulated by Dlx2, we analyzed its promoter in MC3T3-E1 cells. We inserted the whole promoter region of _OCN_ into the pGL3-basic plasmid, generating pGL3-OCN, and transferred

this vector into MC3T3-E1 cells along with pCMV-Dlx2-FLAG to allow the overexpression of _Dlx2_. As shown in Fig. 5a, we observed an ~3.2-fold increase in the transcriptional activity of

pGL3-OCN after introduction of pCMV-Dlx2-FLAG into MC3T3-E1 cells, suggesting that the _OCN_ promoter contains at least one Dlx2-response element (RE). To further identify the Dlx2-RE(s) in

the _OCN_ promoter, we performed ChIP analysis in MC3T3-E1 cells transfected with pCMV-Dlx2-FLAG. Semi-quantitative PCR analysis showed one strong signal in the E6 region (−1 311 bp to −1 

175 bp) matching the predicted ~175 bp size of the E6 PCR product (Fig. 5b). Considering the low sensitivity of semi-quantitative PCR, we next carried out RT-qPCR. Consistent with the above

findings, qPCR also revealed that Dlx2 was highly enriched at region E6 (−1 311 to −1 175 bp) and at region E9 (−1 073 bp to – 932 bp) (Fig. 5c). The qPCR results of region E10 to region E16

are not shown since the signals at these regions were as weak as the negative control in semi-quantitative PCR results. In addition, the occupancy of Dlx2 in the _OCN_ promoter was

correlated with increased transcription represented by elevated occupancy of RNA polymerase II (Pol II) (Fig. 5c), while the negative control (cells introduced with pCMV-FLAG) showed very

weak signal (data not shown). These results indicate that there is a Dlx2-RE in the _OCN_ promoter in region E6 (in primer set 6) and E9 (in primer set 9). Consistent with this,

bioinformatics analysis (JASPAR database) indicated that the promoter region of _OCN_ contains potential Dlx2-REs (−1 447 bp to −1 444 bp and −1 113 bp to −1 110 bp), both of which have the

ATTA sequence. We next checked whether Dlx2 can directly bind to the two predicted Dlx2-REs in the _OCN_ promoter. To do this, we inserted mutated _OCN_ promoter into the pGL3-basic plasmid

in which the ATTA sequence was partly mutated, generating pGL3-mut1 (upstream RE mutant) and pGL3-mut2 (downstream RE mutant) (Fig. 5d). As shown in Fig. 5e, we found that mutation of either

the upstream RE or downstream RE abrogated the ability of Dlx2 to regulate its activity. These observations demonstrate that in MC3T3-E1 cells, Dlx2 directly binds to the upstream RE and

downstream RE in the _OCN_ promoter to positively regulate its transcription. DLX2 UPREGULATES _ALP_ EXPRESSION BY DIRECTLY BINDING TO ITS PROMOTER Next, we tried to investigate whether

_Alp_ was regulated by Dlx2 in a similar way by binding to its promoter. To verify this transcriptional regulation, we first cloned and inserted the whole promoter region of _Alp_ into the

pGL3-basic plasmid, generating pGL3-ALP. This vector was introduced into MC3T3-E1 cells along with pCMV-Dlx2-FLAG to allow overexpression of Dlx2. As shown in Fig. 6a, we observed an

~6.7-fold increase in the transcriptional activity of pGL3-ALP after introduction of pCMV-Dlx2-FLAG into MC3T3-E1 cells, indicating that there is at least one Dlx2-RE in the _Alp_ promoter.

Supporting this notion is that bioinformatics analysis (JASPAR database) indicated that the promoter region of _Alp_ contains potential Dlx2-RE (−1 194 bp to −1 187 bp, region A1). Next, we

checked whether Dlx2 can bind directly to the predicted Dlx2-REs (A1) in the _Alp_ promoter. ChIP analysis was performed in MC3T3-E1 cells transfected with pCMV-Dlx2-FLAG or pCMV-FLAG, and

primer set was designed to amplify the region A1. As shown in Fig. 6b, c, both qPCR and semi-quantitative PCR revealed that Dlx2 was highly enriched in region A1. In addition, elevated

occupancy of Pol II in the _Alp_ promoter was associated with increased occupancy of Dlx2 in this region, while the negative control (MC3T3-E1 cells introduced with pCMV-FLAG) showed very

weak signal (data not shown). To determine whether A1 is the Dlx2-binding site in the _Alp_ promoter, we carried out site-directed mutagenesis in A1 region. We inserted mutated _Alp_

promoter into the pGL3-basic plasmid, generating pGL3-mutationA1 (Fig. 6d). As shown in Fig. 6e, we found that mutation of A1 significantly attenuated the ability of Dlx2 to regulate its

activity. These findings demonstrated that Dlx2 directly binds to region A1 in the _Alp_ promoter to positively regulate its expression (Fig. 6e). Taken together, all these data proved that

Dlx2 promotes _OCN_ and _Alp_ expression by directly binding to their promoters, and then regulates osteogenic differentiation in vitro and accelerates bone formation in vivo. DISCUSSION Dlx

gene family plays a critical role in osteogenesis. Previous study showed that _Dlx2_, _Dlx5_, and _Dlx6_ are upregulated in immature osteoblasts, while the expression of _Dlx3_ is elevated

in differentiated osteoblasts and osteocytes. Dlx3 and Dlx5 potently transactivates osteoblastic marker genes15; Dlx6 also has stimulatory effects on osteogenic differentiation.17 These

findings indicate that as a member of Dlx gene family, _Dlx2_ may be also involved in the osteogenic differentiation. Dlx2, a member of vertebrate Dlx gene family, is expressed in the

epithelium and mesenchyme of the mandible and maxilla.18 Previous studies have shown that newborn _Dlx2_−/− mice die immediately after birth and have abnormal craniofacial bones originating

from the first branchial arch maxillary process,9 while _Dlx2_ overexpression in cranial neural crest cell (CNCC) leads to premaxillary hypoplasia and spinal deformities in mice.9 Although

_Dlx2_ knockout and overexpression models demonstrate a crucial role of Dlx2 in promoting skeleton formation, the molecular and cellular mechanisms underlying the regulation of osteogenic

differentiation by Dlx2 still remain unclear. Here, we present evidence that forced overexpression of _Dlx2_ enhances the osteogenic differentiation potential of mouse BMSCs and MC3T3-E1 in

vitro and accelerates bone formation in vivo by directly regulating _OCN_ and _Alp_. Our finding is consistent with the previous study in SCAPs that overexpression of Dlx2 also enhanced

osteogenic differentiation.19 SCAPs are mesenchymal-like stem cells that are able to differentiate into multiple linages, including odontoblastic and osteoblastic lineage, and do not undergo

adipogenic differentiation, while BMSCs are able to undergo osteogenic, chondrogenic, and adipogenic differentiation.20 Relative to BMSCs, SCAPs display elevated secretion of proteins

involved in metabolic processes, chemokines, and neutrophins, whereas BMSCs secret much more proangiogenic factors and ECM proteins. Therefore, overexpression of _Dlx2_ in SCAPs and BMSCs

could be used in dentin regeneration and bone formation, respectively.21 The rodent _Dlx5_ and _Dlx2_ HD transcription factors are critical for bone development. During osteogenesis, Dlx5

upregulates _Alp_ expression while suppresses _OCN_ transcription.22 On the contrary, we found that Dlx2 transactivates _OCN_ and _ALP_ by directly binding to their promoters. Moreover, we

identified two Dlx2-REs in _OCN_ promoter, and one Dlx2-RE in _Alp_ promoter involved in Dlx2-induced _Alp_ expression. These findings demonstrate that Dlx2 is a crucial regulator regulating

the osteogenic differentiation potential of both mesenchymal stem cells and osteogenic cells. BMP-2 is one of the most important cytokines promoting differentiation of mesenchymal cells

into osteoblasts.23 Stimulated by BMP-2, the transcription factors Dlx5, Msx2, and Runx2 work coordinately to regulate osteogenic differentiation.24 Both Runx2 and Dlx3 positively regulate

_OCN_ expression while Dlx5 represses its expression; Dlx5 stimulates ALP expression, whereas Msx2 depresses its expression.25 At the onset of osteogenic differentiation, Msx2 is released

from the promoter of OCN, while Dlx3, Dlx5, and Runx2 are recruited. The released Msx2 then binds to the _Alp_ promoter to upregulate its expression. At a later stage of osteogenic

differentiation, during matrix mineralization, Dlx5 replaces Dlx3 to regulate OCN expression.26 Runx2 and Osx are another key transcription factors that are necessary for osteogenesis.27

After differentiating into pre-osteoblasts, Runx2 and Osx promote the cells to produce bone matrix. Dlx gene family is involved in the regulation of _Runx2_ and OSX transcription. Dlx5

induces expression of Runx2 and Osx, which work sequentially to stimulate the expression of _OCN_ and _Alp_. Dlx3 also contributes to the activation of Runx2 during osteogenic

differentiation.28 These findings indicate that Dlx gene family plays a crucial role in expression of osteogenic-associated genes. The results of the present study showed that overexpression

of Dlx2 showed no significant effects on _Runx2_, _Msx2_, and _Dlx5_ expression upon osteogenic induction, but stimulated _OCN_ and _Alp_ expression, indicating that Runx2, Msx2, and Dlx5

may not participate in Dlx2-induced osteogenesis; Dlx2 may directly upregulate _OCN_ and _Alp_ to promote osteogenic differentiation. _Alp_ and _OCN_ are two key marker genes of osteoblastic

cells. ALP plays a critical role in early osteogenesis and hydrolyzes various types of phosphates to promote cell maturation and calcification, while OCN promotes the later stage of

osteogenesis through combining with minerals.29,30 Both Dlx3 and Dlx5 directly upregulate _Alp_ expression, while OCN is activated by Dlx3 but suppressed by Dlx5.16,31 Moreover, forced

overexpression of _Dlx5_ in BMSCs led to a reduction in the mineralized matrix deposition, and impaired the ability of these cells to develop to the final stages of osteogenesis, and

severely affected in vivo bone formation in immunodeficient mice.13 Although Dlx2 shares a strong sequence similarity with Dlx5, we found that Dlx2 positively regulates both _Alp_ and _OCN_

expression in BMSCs and MC3T3-E1 cells. A previous study showed that loss of Dlx1/2−/− leads to abnormal bone formation of the upper jaw, while Dlx5/6−/− deficient mice exhibit profound

abnormalities of the lower jaw tissue.22 However, piles of studies revealed that Dlx5 is the master regulator of osteogenic differentiation, since it directly controls the transcription of

multiple osteogenic-associated genes, including _Alp_, _OCN_, _Runx2_, _OSX_, and _Smads_ family, affecting the whole process of bone formation.32,33 Therefore, Dlx5 is also involved in the

development of the upper jaw, but may be not as important as Dlx2. On the other hand, the maxilla is only formed by intramembranous ossification of the craniofacial mesenchyme, while the

mandible can be formed by both intramembranous and endochondral ossification.34 Our previous study has shown that Dlx2 is involved in endochondral ossification.9,35 Therefore, Dlx2 is also

involved in the bone formation of lower jaw, but is not as important as Dlx5. Together, both Dlx2 and Dlx5 are involved in the development of mandible and maxilla. Considering the fact that

a variety of homeobox genes work coordinately during the bone formation,36 further investigations are required to find out how Dlx2 is involved during the bone formation in mandible and

maxilla. In conclusion, our data demonstrated for the first time that forced overexpression of _Dlx2_ enhances the osteogenic differentiation potential of BMSCs and MC3T3-E1 cells by

directly upregulating _OCN_ and _Alp_ (Fig. 6f). We also presented evidence that there are Dlx2-REs in mouse _OCN_ and _Alp_ promoter that mediate the regulation of Dlx2 on _OCN_ or _Alp_

expression. This study may present a promising future strategy for the treatment of bone defects with _Dlx2_-overexpressing BMSCs. MATERIAL AND METHODS ISOLATION AND CULTURE OF MOUSE BMSCS

All animal experiments were performed according to guidelines of the Institutional Animal Care and Use Committees of the Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University

School of Medicine. All experimental protocols were reviewed and approved by the Institutional Animal Care and Use Committees of the Shanghai Ninth People’s Hospital, Shanghai Jiao Tong

University School of Medicine, Shanghai, China. BMSCs were isolated from the tibias of 6-week-old male C57/BL6 mice ((10 ± 0.5) g) and cultured according to previous study.37 An MC3T3-E1

cell line was obtained from the Cell Bank of the Chinese Academy of Science (Shanghai, China), C3H10 T1/2 cell line from American Type Culture Collection (Rockville, MD, USA), and a Human

embryonic kidney 293T (HEK 293T) cell line from American Type Culture Collection (Rockville, MD, USA). The MC3T3-E1, C3H10 T1/2 cell line, and HEK 293T cells were cultured as described

previously.38,39 OIM contained 50 mg· L−1 ascorbic acid, 1 × 10−7 mol· L−1 dexamethasone, and 50 mg· L−1 β-glycerophosphate plus α-MEM (Sigma-Aldrich Corp. (St. Louis, MO, USA)). HEK 293T

cell line was utilized for packaging viral constructs. LENTIVIRAL CONSTRUCTION AND TRANSDUCTION The lentiviral expression system overexpressing _Dlx2_ was termed as Lenti-Dlx2 OE. The open

reading frame of mice _Dlx2_ (NM_010054) was synthesized and cloned into pL/IRES/GFP plasmid (Novobio, Shanghai, China) for generating pL/IRES/GFP-DLX2. The empty lentiviral expression

system without insertion was termed as Lenti-CTRL and used as the control. 293T cells were then transfected with plasmids listed above. The transfection and lentiviral transduction was done

as described previously.40,41 ALP, ALIZARIN RED STAINING, AND SEMI-QUANTITATIVE ANALYSIS Transduced BMSCs or MC3T3-E1 cells were first cultured in OIM for 14 or 21 days. ALP and Alizarin red

staining were carried out as described previously.40 ALP staining was carried out with BCIP/NBT Alkaline Phosphatase Color Development Kit (Beyotime Institute of Biotechnology, China), and

semi-quantitative analysis of ALP activity was performed using _p_-nitrophenyl phosphate (_p_-NPP) (Sigma-Aldrich) as the substrate. For Alizarin red staining, cells were first fixed with

70% ethanol. Afterward, the fixed cells were stained with 2% Alizarin Red (Sigma-Aldrich), according to the previous study.19 SEMI-QUANTITATIVE RT-PCR, RT-QPCR, AND WESTERN BLOT ANALYSIS

RT-PCR, RT-qPCR, and western blot analysis were performed as described previously.40 Total RNA was extracted from cultured cells using TRIzol RNA Isolation reagent (Takara, Tokyo, Japan),

according to the manufacturer’s instruction. Three independent cultures were used for RNA preparations. First-strand cDNA was generated with High-Capacity cDNA Reverse Transcription Kit,

(Applied Biosystems, San Diego, CA), and one microliter of each RT reaction mixture was amplified with Ex Taq DNA polymerase (Takara, Tokyo, Japan). As for RT-qPCR, cDNA was amplified using

premix SYBR Green Ex Taq reagent kit (DRR820A, Takara) with a STEP ONE PLUS real-time PCR system (Applied Biosystems, Forster City, CA), according to the manufacturer’s instruction. All the

primers used in this study are listed in Table 1. As for western blotting, anti-Dlx2 (1:800; ab85995, Abcam, Cambridge, UK), anti-OCN (1:1 000; ab93876, Abcam), and anti-β-actin (1:3 000;

EPR16769, Abcam, Cambridge, UK) were used for the detection of Dlx2, OCN, and β-actin, respectively. The secondary antibodies used this study were bought from Sigma-Aldrich and conjugated to

horseradish peroxidase (anti-rabbit (1:5 000, A0545) or anti-mouse (1:5 000, SAB3701214)). IN VIVO OSTEOGENIC DIFFERENTIATION The osteogenic differentiation potential of transduced BMSCs

was evaluated with in vivo ectopic bone formation analysis, as described previously.13 Briefly, osteogenic-induced cells of passage 3 were injected into the β-TCP (Shanghai Rebone

Biomaterials Co., Ltd., Shanghai, China) with a syringe, and appropriate volume of OIM was added to cover the BMSCs/ β-TCP constructs. After 7 days of culturing in vitro, the constructs were

implanted subcutaneously in the back of immunocompromised female nude mice (CD-1 Nu/Nu, 10-week-old, Charles River). At each time point (6 and 8 weeks after implantation), six mice for each

group were sacrificed with an overdose of pentobarbital (210 mg/kg intraperitoneally) to retrieve the BMSCs/β-TCP constructs. Micro-CT images of each cellular construct were taken with a

micro-CT system (SMX-100CT-SV3; Shimadzu, Japan). Radiological density of each cellular construct was measured in Hounsfield density units. LUCIFERASE ASSAY The whole mouse _OCN_ promoter

(2000 bp) was subcloned into the pGL3 basic vector (Promega), generating pGL3-OCN. To produce constructs that contain mutation in E6 region, we carried out mutagenic PCR with Mt1 primer and

Mt2 primer, generating pGL3-mut1. Likewise, Mt3 and Mt4 primers were used for generating pGL3-mut2 (Table 1). The open reading frame of mice _Dlx2_ (NM_010054) was synthesized and cloned

into pCMV-FLAG, generating pCMV-Dlx2-FLAG for _Dlx2_ overexpressing. Transfection assay was performed as described previously.31 As for each transfection assay, 0.5 μg of the _Dlx2_

overexpression vector (pCMV-Dlx2-FLAG) or pCMV-FLAG and 0.5 μg of the luciferase reporter vector were transfected into MC3T3-E1 cells. Similarly, we inserted the whole _Alp_ promoter into

the pGL3-basic vectors, generating pGL3-ALP, and produced a vector that contains mutated _Alp_ promoter, generating pGL3-mutationA1. Afterward, pCMV-Dlx2-FLAG or pCMV-FLAG was introduced

into MC3T3-E1 cells along with pGL3-ALP or pGL3-mutationA1. CHIP ANALYSIS ChIP analysis was carried out according to the standard protocol.15 MC3T3-E1 cells transfected with pCMV-Dlx2-FLAG

or pCMV-FLAG were fixed with 1% formaldehyde and this cross-linking was quenched with glycine (0.125 mol· L−1 for final concentration). After the cells were lysed and homogenized with a

Dounce homogenizer, the nuclei were collected by centrifuging. The nuclei pellet was then resuspended in sonication buffer followed by sonication. Chromatin were incubated with anti-FLAG

antibodies (F3165, Sigma-Aldrich), mouse IgG and anti-RNA-polymerase II (PLA0292, Sigma-Aldrich), and this immune complex were incubated with protein G sepharose (17-0618-01, GE Amersham).

The DNA fragments in this immune complex were purified with phenol–chloroform extraction. ChIP samples were then analyzed with RT-PCR and RT-qPCR. Pol II was used as a positive control, as

described in a previous study.15 The primers used in ChIP analysis are listed in Table 1. REFERENCES * Ulsamer, A. et al. BMP-2 induces Osterix expression through up-regulation of Dlx5 and

its phosphorylation by p38. _J. Biol. Chem._ 283, 3816–3826 (2008). Article  Google Scholar  * Stock, D. W. et al. The evolution of the vertebrate Dlx gene family. _Proc. Natl Acad. Sci.

USA_ 93, 10858–10863 (1996). Article  Google Scholar  * Lee, M.-H., Kwon, T.-G., Park, H.-S., Wozney, J. M. & Ryoo, H.-M. BMP-2-induced Osterix expression is mediated by Dlx5 but is

independent of Runx2. _Biochem. Biophys. Res. Commun._ 309, 689–694 (2003). Article  Google Scholar  * Shirakabe, K., Terasawa, K., Miyama, K., Shibuya, H. & Nishida, E. Regulation of

the activity of the transcription factor Runx2 by two homeobox proteins, Msx2 and Dlx5. _Genes Cells_ 6, 851–856 (2001). Article  Google Scholar  * Depew, M. J., Lufkin, T. & Rubenstein,

J. L. Specification of jaw subdivisions by Dlx genes. _Science_ 298, 381–385 (2002). Article  Google Scholar  * Ferguson, C. A., Tucker, A. S. & Sharpe, P. T. Temporospatial cell

interactions regulating mandibular and maxillary arch patterning. _Development_ 127, 403–412 (2000). PubMed  Google Scholar  * Merlo, G. R. et al. Multiple functions of Dlx genes. _Int. J.

Dev. Biol._ 44, 619–626 (2004). Google Scholar  * Sun, H., Liu, Z., Li, B., Dai, J. & Wang, X. Effects of DLX2 overexpression on the osteogenic differentiation of MC3T3‑E1 cells. _Exp.

Ther. Med._ 9, 2173–2179 (2015). Article  Google Scholar  * Dai, J. et al. The effect of overexpression of Dlx2 on the migration, proliferation and osteogenic differentiation of cranial

neural crest stem cells. _Biomaterials_ 34, 1898–1910 (2013). Article  Google Scholar  * Pittenger, M. F. et al. Multilineage potential of adult human mesenchymal stem cells. _Science_ 284,

143–147 (1999). Article  Google Scholar  * Kuznetsov, S. A. et al. Single‐colony derived strains of human marrow stromal fibroblasts form bone after transplantation in vivo. _J. Bone

Mineral. Res._ 12, 1335–1347 (1997). Article  Google Scholar  * Krebsbach, P. H. et al. Bone formation in vivo: comparison of osteogenesis by transplanted mouse and human marrow stromal

fibroblasts. _Transplantation_ 63, 1059–1069 (1997). Article  Google Scholar  * Muraglia, A. et al. DLX5 overexpression impairs osteogenic differentiation of human bone marrow stromal cells.

_Eur. J. Cell Biol._ 87, 751–761 (2008). Article  Google Scholar  * Dalby, M. J. et al. The control of human mesenchymal cell differentiation using nanoscale symmetry and disorder. _Nat.

Mater._ 6, 997 (2007). Article  Google Scholar  * Hassan, M. Q. et al. Dlx3 transcriptional regulation of osteoblast differentiation: temporal recruitment of Msx2, Dlx3, and Dlx5 homeodomain

proteins to chromatin of the osteocalcin gene. _Mol. Cell. Biol._ 24, 9248–9261 (2004). Article  Google Scholar  * Ryoo, H.-M. et al. Stage-specific expression of Dlx-5 during osteoblast

differentiation: involvement in regulation of osteocalcin gene expression. _Mol. Endocrinol._ 11, 1681–1694 (1997). Article  Google Scholar  * Pfeffer, U. et al. The coding region of the

human DLX6 gene contains a polymorphic CAG/CCG repeat. _Int. J. Oncol._ 18, 1293–1297 (2001). PubMed  Google Scholar  * Depew, M. J., Simpson, C. A., Morasso, M. & Rubenstein, J. L. R.

Reassessing the Dlx code: the genetic regulation of branchial arch skeletal pattern and development. _J. Anat._ 207, 501–561 (2005). Article  Google Scholar  * Qu, B. et al. Distal-less

homeobox 2 promotes the osteogenic differentiation potential of stem cells from apical papilla. _Cell Tissue Res._ 357, 133–143 (2014). Article  Google Scholar  * Huang, G. T. J. et al.

Stem/progenitor cell–mediated de novo regeneration of dental pulp with newly deposited continuous layer of dentin in an in vivo model. _Tissue Eng. Part A_ 16, 605–615 (2009). Article 

Google Scholar  * Yu, S., Zhao, Y., Ma, Y. & Ge, L. Profiling the secretome of human stem cells from dental apical papilla. _Stem Cells Dev._ 25, 499–508 (2016). Article  Google Scholar

  * Samee, N., De Vernejoul, M.-C. & Levi, G. Role of DLX regulatory proteins in osteogenesis and chondrogenesis. _Crit. Rev. Eukaryot. Gene Expr._ 17, 173–186 (2007). * Hanai, J.-i et

al. Interaction and functional cooperation of PEBP2/CBF with smads synergistic induction of the immunoglobulin germline Cα promoter. _J. Biol. Chem._ 274, 31577–31582 (1999). Article  Google

Scholar  * Nishimura, R., Hata, K., Harris, S., Ikeda, F. & Yoneda, T. Core-binding factor α1 (Cbfa1) induces osteoblastic differentiation of C2C12 cells without interactions with Smad1

and Smad5. _Bone_ 31, 303–312 (2002). Article  Google Scholar  * Javed, A. et al. runt homology domain transcription factors (Runx, Cbfa, and AML) mediate repression of the bone

sialoprotein promoter: evidence for promoter context-dependent activity of Cbfa proteins. _Mol. Cell. Biol._ 21, 2891–2905 (2001). Article  Google Scholar  * Langenbach, F. & Handschel,

J. Effects of dexamethasone, ascorbic acid and β-glycerophosphate on the osteogenic differentiation of stem cells in vitro. _Stem Cell Res. Ther._ 4, 117 (2013). Article  Google Scholar  *

Zhang, C. Transcriptional regulation of bone formation by the osteoblast-specific transcription factor Osx. _J. Orthop. Surg. Res._ 5, 37 (2010). Article  Google Scholar  * Hassan, M. Q. et

al. BMP2 commitment to the osteogenic lineage involves activation of Runx2 by DLX3 and a homeodomain transcriptional network. _J. Biol. Chem._ 281, 40515–40526 (2006). * Glowacki, J., Rey,

C., Glimcher, M., Cox, K. & Lian, J. A role for osteocalcin in osteoclast differentiation. _J. Cell. Biochem._ 45, 292–302 (1991). Article  Google Scholar  * Beck, G. R., Zerler, B.

& Moran, E. Phosphate is a specific signal for induction of osteopontin gene expression. _Proc. Natl Acad. Sci. USA_ 97, 8352–8357 (2000). Article  Google Scholar  * Kim, Y.-J., Lee,

M.-H., Wozney, J. M., Cho, J.-Y. & Ryoo, H.-M. Bone morphogenetic protein-2-induced alkaline phosphatase expression is stimulated by Dlx5 and repressed by Msx2. _J. Biol. Chem._ 279,

50773–50780 (2004). Article  Google Scholar  * Lee, M.-H. et al. BMP-2-induced Runx2 expression is mediated by Dlx5, and TGF-β1 opposes the BMP-2-induced osteoblast differentiation by

suppression of Dlx5 expression. _J. Biol. Chem._ 278, 34387–34394 (2003). Article  Google Scholar  * Robledo, R. F., Rajan, L., Li, X. & Lufkin, T. The Dlx5 and Dlx6 homeobox genes are

essential for craniofacial, axial, and appendicular skeletal development. _Genes Dev._ 16, 1089–1101 (2002). Article  Google Scholar  * Tortelli, F., Tasso, R., Loiacono, F. & Cancedda,

R. The development of tissue-engineered bone of different origin through endochondral and intramembranous ossification following the implantation of mesenchymal stem cells and osteoblasts in

a murine model. _Biomaterials_ 31, 242–249 (2010). Article  Google Scholar  * Zhang, J., Zhang, W., Shi, J., Dai, J. & Shen, S. G. Dlx2 overexpression enhanced accumulation of type II

collagen and aggrecan by inhibiting MMP13 expression in mice chondrocytes. _Biochem. Biophys. Res. Commun._ 503, 528–535 (2018). Article  Google Scholar  * Thesleff, I. Homeobox genes and

growth factors in regulation of craniofacial and tooth morphogenesis. _Acta Odontol. Scand._ 53, 129–134 (1995). Article  Google Scholar  * Russo, F. P. et al. The bone marrow functionally

contributes to liver fibrosis. _Gastroenterology_ 130, 1807–1821 (2006). Article  Google Scholar  * Yamamoto, K., Arakawa, T., Ueda, N. & Yamamoto, S. Transcriptional roles of nuclear

factor B and nuclear factor-interleukin-6 in the tumor necrosis factor-dependent induction of cyclooxygenase-2 in MC3T3-E1 cells. _J. Biol. Chem._ 270, 31315–31320 (1995). Article  Google

Scholar  * Heim, M. et al. Phytanic acid, a natural peroxisome proliferator-activated receptor (PPAR) agonist, regulates glucose metabolism in rat primary hepatocytes. _FASEB J._ 16, 718–720

(2002). Article  Google Scholar  * Xie, Q. et al. The role of miR-135-modified adipose-derived mesenchymal stem cells in bone regeneration. _Biomaterials_ 75, 279–294 (2016). Article 

Google Scholar  * Cheng, S. L. et al. MSX2 promotes osteogenesis and suppresses adipogenic differentiation of multipotent mesenchymal progenitors[J]. _J. Biol. Chem._ 278, 45969–45977

(2003). Article  Google Scholar  Download references ACKNOWLEDGEMENTS The authors thank Dr. Wenbin Zhang, Dr. Jianjun Zhang, Dr. Ming Yan and Dr. Xu Wang from the Shanghai Ninth People’s

Hospital for helping us with the technique. We would like to thank the members of the Shen- and Chen-labs for helpful advice and discussion. We also thank Native EE for its linguistic

assistance during the preparation of this manuscript. This work is supported by grant (81771036) from National Natural Science Foundation of China (to S.G.S.), grant (81741028) from National

Natural Science Foundation of China (to J.D.) and grant (17410710500) Shanghai International Scientific and Technological Cooperation Projects Laser Micro-machine and Vascularization of

TCP/PCL Scaffolds (to W.Z.). AUTHOR INFORMATION AUTHORS AND AFFILIATIONS * Department of Oral & Cranio-maxillofacial Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong

University School of Medicine, Shanghai Key Laboratory of Stomatology, No. 639 Zhizaoju Road, Shanghai, China Jianfei Zhang, Wenbin Zhang, Jiewen Dai, Xudong Wang & Steve Guofang Shen

Authors * Jianfei Zhang View author publications You can also search for this author inPubMed Google Scholar * Wenbin Zhang View author publications You can also search for this author

inPubMed Google Scholar * Jiewen Dai View author publications You can also search for this author inPubMed Google Scholar * Xudong Wang View author publications You can also search for this

author inPubMed Google Scholar * Steve Guofang Shen View author publications You can also search for this author inPubMed Google Scholar CORRESPONDING AUTHORS Correspondence to Xudong Wang

or Steve Guofang Shen. ETHICS DECLARATIONS COMPETING INTERESTS The authors declare no competing interests. SUPPLEMENTARY INFORMATION SUPPLEMENTAL FIGURE LEGENDS RIGHTS AND PERMISSIONS OPEN

ACCESS This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format,

as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third

party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the

article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright

holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Reprints and permissions ABOUT THIS ARTICLE CITE THIS ARTICLE Zhang, J., Zhang, W., Dai, J. _et

al._ Overexpression of _Dlx2_ enhances osteogenic differentiation of BMSCs and MC3T3-E1 cells via direct upregulation of _Osteocalcin_ and _Alp_. _Int J Oral Sci_ 11, 12 (2019).

https://doi.org/10.1038/s41368-019-0046-1 Download citation * Received: 07 November 2018 * Revised: 02 January 2019 * Accepted: 03 January 2019 * Published: 18 March 2019 * DOI:

https://doi.org/10.1038/s41368-019-0046-1 SHARE THIS ARTICLE Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable link is not

currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative